2
With the rapid growth of the use of synthetic chemicals and the exceedance of the
planetary boundary on novel entities, new tools and multi-disciplinary strategies are
needed to understand the major drivers of human exposure to synthetic chemicals of
concern.
In this thesis, I summarize and link 20 studies that together provide a multi-faceted
approach to improving our understanding of human exposure to chemical pollutants. The
articles cover the development and improvement of sampling techniques for human
exposure, the integration of multiple data sources to improve our understanding of
exposure, and the integration of elements of policy, human behaviour and environmental
impacts to provide a framework for human exposure evaluation. Results from original
research highlight the development and optimization of sampling tools necessary to
collect data in indoor and urban environments, and ensure comparability across studies.
Integration of exposure estimation techniques allows us to contextualize and prioritize the
broad set of chemical exposure data that is available from indoor and outdoor monitoring.
The included studies also incorporate evaluation of environmental policy and regulation;
this is critical in evaluating the implementation and effectiveness of policy actions to ensure
a science-backed strategy for chemical management, with the ultimate goal of reducing
our exposures to hazardous chemicals.
3
First, I would like to express my deepest thanks to my husband, Ondřej Sáňka, and my
children, Alexandra and Ester. None of this would be possible without Ondra’s support and
patience. Big thanks also go to my family in Canada and all the friends that have supported
me over the years.
I would like to express my gratitude to Prof. Jana Klánová for opening the door for me at
RECETOX 12 years ago, and for supporting me in innumerable ways over the years. I am
also very grateful for the opportunity for the continued support of Prof. Miriam Diamond
and other colleagues in Canada.
I have been lucky to work with many great colleagues and collaborators, and I thank all of
them for the pleasure of working together. A particular thanks goes to my RECETOX
colleagues and my research group for your hard work, new ideas and support, and
especially to Rozárka Jílková, who makes sure everything works and everyone is supported.
4
BDE-209 decabromodiphenyl ether
BFR brominated flame retardant
c-pentaBDE commercial pentabromodiphenyl ether mixture
DBDPE decabromodiphenyl ethane
DDT dichlorodiphenyltrichloroethane
DDX DDT and related metabolites/degradation products
DEHP bis(2-ethylhexyl)phthalate
DPHP diphenyl phosphate
FR flame retardant
GAPS Global Atmospheric Passive Sampling network
HBM4EU Human Biomonitoring for Europe
HCH hexachlorocyclohexane
MONET Monitoring Network
OPE organophosphate ester
PAH polycyclic aromatic hydrocarbon
PAS passive air sampler
PBDE polybrominated diphenyl ether
PCB polychlorinated biphenyl
PCP personal care product
PFAS per- and polyfluoroalkyl substances
POP persistent organic pollutant
PUF polyurethane foam
PVC polyvinyl chloride
SVOCs semi-volatile organic compound
TB-117 Technical Bulletin 117
TCIPP tris(2-chloroisopropyl)phosphate
TDCIPP tris(1,3-dichloro-2-propyl)phosphate
TPHP triphenyl phosphate
UNEP United Nations Environment Program
WHO World Health Organization
XAD polystyrene-divinyl benzene-based resins
5
1 Introduction..............................................................................................................................................6
2 A framework for understanding chemical exposure .................................................................7
2.1 Chemical Policy as a driver of human exposure to chemicals......................................8
2.2 Lifestyle impacts on chemical exposure .............................................................................14
2.3 Environment as a determinant of chemical exposure...................................................16
3 Tools and techniques for understanding chemical exposures ............................................21
3.1 Air sampling techniques...........................................................................................................21
3.2 Dust sampling techniques .......................................................................................................22
3.3 Analytical considerations..........................................................................................................23
4 Conclusions.............................................................................................................................................24
5 List of Appendices................................................................................................................................25
6 References...............................................................................................................................................28
6
Human exposure to chemicals has been known to impact health since the 1700s – chimney
sweep's cancer (later identified as scrotal cancer) was linked to exposure to chimney soot,
in a paradigm-shifting insight by British doctor Percivall Pott (Dronsfield, 2006). While it
was not until the 1920s that the implicated chemical was identified as benzo[a]pyrene, a
polycyclic aromatic hydrocarbon (PAH) (Dronsfield, 2006), the identification of
environmental exposure to chemicals as a trigger for health impacts was one of the first
steps in exposure science, linking environmental causes with health outcomes. Over the
past centuries, many links between chemicals and health impacts have been established,
e.g., ingestion exposure to lead identified in Australian children in 1892 (Needleman, 2009),
Minamata disease caused the discharge and biomagnification of mercury in the Japanese
area of Minamata Bay in the 1950s (Budnik and Casteleyn, 2019), and the field of human
exposure to chemicals continues to grow in complexity.
Synthetic chemicals, a basis for so much development in modern society, pose one of the
biggest challenges to exposure science today. Around 350 000 chemicals are being
marketed globally (Fenner and Scheringer, 2021) and the diversity and quantity of synthetic
chemicals have been increasing at rates surpassing the drivers of global environmental
change (Bernhardt et al., 2017). Conventional strategies are challenged by this rapid
increase in complexity: the constantly increasing number of chemicals in commerce and
under development requires advanced tools and strategies for a comprehensive overview
of human exposure to chemicals. The planetary boundary of Novel Entities (Steffen et al.,
2015) is estimated to have been passed as the increasing rate of production and releases
of novel entities exceed society’s ability to conduct safety-related assessments and
monitoring (Persson et al., 2022).
This thesis presents a collection of 20 peer-reviewed papers published between 2014 and
2023 related to human exposure to chemicals. The papers are grouped according to a
focus on the development and improvement of sampling techniques for human exposure
(Articles 17-20), integration of multiple data sources to improve our understanding of
exposure (Articles 9-16), and synthesis, integrating elements of policy, human behaviour
and environmental impacts to provide a framework for human exposure evaluation that
can support the challenge of increasing chemical complexity (Articles 1-8).
Currently, I have published 88 papers, including 78 articles and 10 conference proceedings.
My contribution to the 20 articles selected for this habilitation thesis is summarized in the
tables in the attached commentary document, giving my percentage contribution to
different elements of the work.
7
Chemical exposure is typically categorized according to exposure route – for most
chemicals this is distributed between inhalation, ingestion (including both diet and nondietary
ingestion), and dermal uptake. However, a broader framework is necessary to
understand the relevance of specific exposure pathways for individual substances or
groups of chemicals, what factors lead to within-population differences in exposures, and
more broadly, identify actions which can be taken to reduce exposure to harmful chemicals.
Human chemical exposure sits at the junction of several key factors. To provide a
framework for guiding research on chemical exposures, we consider a framework
presenting the “lifecycle” of a chemical exposure, from product use involving direct
exposures (also called “near-field” exposures in some exposure models), and indirect
exposures from the wider environment (also called “far-field” exposures). A conceptual
overview of this framework is presented in Figure 1
Each element of this exposure framework can be influenced by three major types of
determinants of exposure:
• Chemical use, influenced by policy and regulations permitting/restricting chemicals,
and industrial developments in the synthesis and production of new materials and
products;
• Human behaviour, and shifts towards certain lifestyle patterns, professions, dietary
patterns and consumer choices;
• Environment, combining the influence of near and far-field exposures on humans.
While there are many overlaps and interlinkages between these factors, this thesis uses
these three concepts as themes to highlight key scientific understanding in each area and
show my contribution to the study of human exposure to synthetic chemicals, from the
development of sampling techniques through to data synthesis.
8
Figure 1 A conceptual representation of key factors driving human exposure to chemicals
At the fundamental level, for human exposure to a synthetic chemical substance to occur,
there must be sufficient use of the chemical in some application that could lead either to
its direct exposure to humans, or to its release into the environment and subsequent
exposure. In practice, both these aspects often occur simultaneously.
Therefore, a framework for understanding human exposure to synthetic chemicals must
address the element of chemical use, through consideration of policy and regulatory
aspects that lead to the use or restriction of a chemical, regulatory changes altering
chemical use patterns over time, and broad societal level trends impacting exposures.
A clear potential policy impact on chemical exposures is when a restriction of a chemical is
introduced; restrictions to reduce risks of hazardous chemicals are introduced with the aim
that the use of the restricted substance, and consequently exposure, should decrease over
time. We see clear examples of this in environmental monitoring of persistent organic
pollutants (POPs), which have been restricted through the Stockholm Convention on POPs
(UNEP, 2017), entered into force in 2004 and now with 186 parties. For example, air
concentrations of POP chemicals such as hexachlorocyclohexane, polychlorinated
biphenyls (PCBs), dichlorodiphenyltrichloroethane (DDT) and polybrominated diphenyl
ethers (PBDEs) showed continuous declining trends at Arctic monitoring stations from
1992-2018, reflecting global decreases in chemical stocks and emissions (Wong et al.,
2021), and similar responses to restrictions and reductions of POP use have been seen in
Presence of synthetic chemicals in the environment
Environmental
transport of
synthetic
chemicals/
uptake by biota
Direct exposure
Influencing factors
• Chemical distribution,
transport and fate
• Behavioural patterns
and lifestyle
Presence of synthetic
chemicals in products
and materials
Influencing factors
• Policy (chemical
use/restrictions)
• Product use/societal patterns
• Manufacturing and marketing
of products, chemicals
Dietary exposure
Influencing factors
• Dietary patterns
• Environmental
contamination of food
chain
Human body burdens
of chemicals
Influencing factors
• Exposure pathway
(inhalation, ingestion, dermal)
• Biological half-lives,
distribution in the body
Outdoor
env.
exposure
9
other air (Kalina et al., 2018; Venier and Hites, 2010), sediments (Bogdal et al., 2008), and
biota monitoring (Crimmins et al., 2012; Rigét et al., 2019).
While the time trends generated from routine monitoring are indicative of overall
reductions in environmental emissions, and provide a suggestion of human exposure
trends, more direct evaluation of the effectiveness of restrictions on reducing human
exposures can be achieved through human biomonitoring (Magulova and Priceputu, 2016).
Many POPs are persistent, lipophilic compounds which are best monitored through lipidheavy
matrices, e.g., blood serum and breast milk (Vorkamp et al., 2021). Human milk has
been selected as a core medium of the Global Monitoring Plan of the Stockholm
Convention, and has routinely been monitored through a strategic partnership between
the World Health Organization (WHO) and United Nations Environment Programme
(UNEP) to track patterns and trends of human exposure to chemicals (Magulova and
Priceputu, 2016).
A meta-analysis of human biomonitoring data reporting levels of PBDEs was one of the
first analyses to generate alarm regarding PBDE exposures: the analysis found a doublingtime
of levels of PBDEs in human tissues of five years, translating to rapidly increasing levels
in human populations (Hites, 2004). PBDEs are flame retardants - chemicals added to
consumer products and building materials to reduce their flammability. They have been
widely used in consumer products and building materials on a global basis (Abbasi et al.,
2015), and the concerns raised about PBDEs’ environmental persistence and toxicity
(Darnerud, 2003), and specifically neurodevelopmental impacts related to children’s
exposure (Hudson-Hanley et al., 2018), have led to a transition over the past 20 years away
from PBDE flame retardants to alternative synthetic chemicals. Our recent meta-analysis of
PBDEs identified clear evidence of a breakpoint in increasing time trends of PBDEs in
human tissues, evidence of a positive impact of regulations on reducing human
exposure to chemicals (Figure 2, van der Schyff et al., 2023a, Appendix 1). However, this
decreasing time trend does not translate universally to all restricted chemicals; interfering
factors can moderate the impact of restrictions on human exposures. In the case of PBDEs,
we see evidence that differing chemical half-lives in the body delay the impact of
regulations (notably seen in the lack of decrease in PBDE 153, which has a longer half-life
in the human body), but additionally, we capture the lag-time between restriction of a
chemical and the removal of the items containing that chemical from use. Most restrictions
on a chemical eliminate the new production and sale of a chemical, but, for practical
reasons, typically exempt existing uses. Without the effort to explicitly remove existing uses
of a chemical, there will be a substantial lag between the restrictions on new use, and the
elimination of chemical use as a source to humans and the environment. In the case of
flame-retarded items, the product lifetimes vary widely, from smartphones (2-6 years) to
building insulation (30-50 years), suggesting that responses to regulations will be
10
significantly delayed for flame retardants predominantly used in insulation (van der Schyff
et al., 2023a).
a) BDE 47 b) BDE 99
c) BDE 153 d) BDE 209
Figure 2 Results of breakpoint analysis for PBDEs in human milk for Europe for individual PBDE congeners:
BDE-47, BDE-99, BDE-153 and BDE-209. The shaded area indicates 95th percent confidence interval. The
dotted blue line indicates when regulation was implemented by the Stockholm Convention (2009: BDE-47,
-99, and -153; 2019: BDE-209). From van der Schyff et al. (2023a).
The differing impact of regulations restricting production vs. required removal from use
are seen in a global evaluation of the impact of policy of the management of PCB stocks.
PCBs were widely used in electrical equipment, and as plasticizers and flame retardants
from 1930s to 1980s (Breivik et al., 2002). Due to concerns about their persistence and
toxicity, restrictions were introduced in many jurisdictions over the 1970s and 1980s,
eventually eliminating the intentional production of PCBs (Breivik et al., 2002; Melymuk et
al., 2022). However, many existing uses remained, as early laws restricted only production,
and did not include a requirement to remove existing PCB stocks from use. PCBs are
included in the Stockholm Convention which includes a 2025 deadline for removal all PCBs
from use. Therefore, parties to the Stockholm Convention are required to take active
measures to remove PCBs from use, which in most regions has resulted in decreasing
stocks of PCBs (Melymuk et al., 2022, Appendix 2). However, the USA, notably with the
highest historical PCB use per capita, is not a party to the Stockholm Convention, and as
such, has not enacted laws requiring removal of PCBs from use. A comparative analysis of
the implications of this regulatory difference examined PCB stocks over the past 15 years
in Canada and Czechia (parties to the Stockholm Convention) compared with the USA (as
11
a non-party). Our analysis identified that the active removal of PCBs resulted in a >90 %
reduction in stocks of PCBs in Canada and Czechia, attributed to the active measures
undertaken to ensure compliance with the Stockholm deadlines, while USA has had only a
3 % decline, showing that without active policy measures to require chemical
elimination, lag times between initial restrictions and eliminations of exposures can
be very substantial (Melymuk et al., 2022).
It is not only chemicals policy that directly impacts chemical use and therefore exposure,
but other policies that trigger chemical use. The incorporation of flame retardants in
products is primarily driven by regulations: the flammability standards to which an item
must conform in order to be sold on a given market. Such flammability standards exist for
building materials, vehicles, and in some jurisdictions, for upholstered furniture (Stapleton
et al., 2012), and we can see evidence of regional differences in these flammability
standards on human chemical exposure.
An early decision introduced by the state of California in the 1970s has had widespread
implications for human exposure to flame retardants. California introduced a standard
called TB-117 which required a high resistance to flame in upholstered furniture (Stapleton
et al., 2012). Because of shared markets, this flammability standard indirectly impacted the
whole USA, and to some extent, Canada, as furniture manufacturers tend to conform to
the highest standards (rather than manufacturing dedicated items for California). The
introduction of this flammability standard resulted in the high use of synthetic organic
flame retardants in upholstered furniture to meet this standard (Stapleton et al., 2012). For
the period of 1970s-2000, this flammability standard was primarily met by the
incorporation of a commercial mixture of polybrominated diphenyl ethers (PBDEs),
specifically c-pentaBDE, a commercial mixture of tetra, penta- and hexabromodiphenyl
ether isomers. The majority of the global use of c-pentaBDE was in the USA, in large part
as additives to upholstered furniture (Abbasi et al., 2019).
12
Figure 3 . Box-and-whiskers (horizontal lines are medians, 95% confidence intervals, minima, and maxima)
of individual PBDE congeners: A) BDE-47, B) BDE-99, C) BDE-153, and D) BDE-209 concentrations in
different regions (CSA= Central- and South America and the Caribbean), in breast milk data collected after
year 2000. Non-parametric ANOVA tests (Kruskal-Wallis with Dunn’s post-tests) were conducted to
determine significant differences. From van der Schyff et al. (2023a).
We see clear impacts of this flammability standard and the high use of PBDEs in USA when
comparing measurements of PBDEs in environmental and human samples from USA vs.
other countries. Our meta-analysis of existing biomonitoring data for PBDEs in human
breast milk clearly identified that North American women had levels of PBDEs in their
breast milk 1-2 orders of magnitude higher than other regions (Figure 3), attributed to the
higher use of PBDEs in USA, resulting in higher direct and environmental exposures.
Moreover, we noted similar variations in a comparison of American vs. Canadian and Czech
homes: concentrations of PBDEs measured in surface dust from USA found levels that were
consistently 1-2 orders of magnitude higher than in Czechia, and similar geographic
differences were seen in other indicators of indoor contamination – surface wipes and
indoor air (Figure 4, Venier et al., 2016, Appendix 3).
13
Figure 4 Boxplots of concentrations for selected PBDEs congeners. The plots share the same y-axis
concentration scales although the units are different for each matrix: dust is ng/g, window film is ng/m2,
and air is pg/m3
. Within each matrix, boxes that share the same letter are not statistically significantly
different at a 5% level in an ANOVA analysis using the Tukey's test. From Venier et al. (2016).
The similarity in exposure profiles in many populations further emphasizes the importance
of regulatory policy and other broad societal factors in driving exposures. In the recent
HBM4EU (Human Biomonitoring for Europe) aligned studies, indicators of exposure to
many chemicals were similar across populations from different regions (Govarts et al.,
2023). We found that a urinary metabolite of triphenyl phosphate (TPHP), an
organophosphate ester (OPE) flame retardant and plasticizer widely used in consumer
products (van der Veen and de Boer, 2012), was detected in >99% of the European
children measured, with medians across all nine countries within the range of 1.43 to 2.43
μg/g creatinine (Figure 5, van der Schyff et al., 2023b, Appendix 4). We noted similar
homogeneity in a study of flame retardant and plasticizers in the homes and dermal wipes
14
of 51 Canadian women: a consistent similarity in the profiles of both chemical groups was
noted in wipes from the hands of 51 unrelated women, demonstrating the ubiquity of
the target chemicals in modern environments, regardless of the participants’ home
contents, activities, or other personal characteristics (Diamond et al., 2021, Appendix
5).
Figure 5 Box plot of creatinine-adjusted concentrations of diphenyl phosphate (DPHP), a metabolite of
TPHP, for Belgium (BE), Czech Republic (CZ), Germany (DE), Denmark, France (FR), Norway (NO), Slovenia
(SI), and Slovakia (SK). The box indicates median, 25% and 75% percentiles. Whiskers indicate 5% and 95%
percentiles. From van der Schyff et al. (2023b). The table on the right groups the countries according to
where significant differences exist between urinary concentrations, e.g., DK is lower than all other countries,
while DE and NO are similar concentrations.
While policy and regulations have broader societal implications on chemical exposures,
many of the differences in human exposure to chemicals are driven by individual variations
in lifestyle and behaviour. For example, while our HBM4EU study found only a 50%
difference in population medians of diphenyl phosphate (DPHP; metabolite of TPHP)
across countries, the variation within individual populations far exceeds that, covering a
10-fold range in most regions (Figure 5), suggesting the importance of individual factors
in controlling exposures.
Variations in the use of personal care products (PCPs) within a population is a clear source
of within-population variations in exposures. Use of personal care products has been linked
with exposures to synthetic chemicals used as additives in the products or their packaging.
This includes exposure to TPHP from nail polish use (Mendelsohn et al., 2016), exposure to
synthetic fragrance compounds from the use of perfumes (Nakata et al., 2015), and
exposure to UV filters from sunscreen use (Krause et al., 2017). In Czech teenagers and
young adults, we found that higher use of PCPs is associated with higher levels of
phthalate ester metabolites in urine (Stuchlík Fišerová et al., 2022, Appendix 6), as
phthalate esters are used as carriers and plasticizers in PCP formulations and
packaging, and this link has been shown in other analyses of biomonitoring data for both
men and women (Fruh et al., 2022; Nassan et al., 2017).
15
The case of PCPs is also an example of how shifts in personal behaviour can alter chemical
impacts. In many markets, including the cosmetics market, there is an increasing share of
“green” products and a societal shift towards what are perceived to be “safer” products can
also translate into shifts in chemical exposures (Chin et al., 2018). Studies evaluating the
impact of a shift to products “free-from” certain chemicals have shown impacts: when
consumers intentionally replace products with those with lower levels of parabens, UV
filters and biocides, indicators of exposure, in this case urinary metabolites, have been
shown to decrease (Harley et al., 2016). However, there remains many elements of
uncertainty in the shift towards “safer” products; one important one is the uncertainty
associated with various green labels and other marketing elements suggesting green/safe
products. The use of key phrases associated with “safer” products (e.g. eco, bio, free-from)
can in some cases constitute “greenwashing”. In a set of 50 products available on the Czech
market, we observed no significant difference in concentrations of parabens and triclosan
between conventional PCPs and those with a “green”-indicating label, suggesting a
disconnect between a societal shift towards interest in safer products, and actual human
exposure (van der Schyff et al., 2022, Appendix 7). Thus, in some cases, shifts in personal
behaviour may not have the expected or intended impact, and it is crucial that
recommended actions to reduce exposure to hazardous chemicals have a clear
scientific backing.
Other individual/household behaviours can also impact chemical exposures, in the case of
multiple flame retardants (TPHP, tris(1,3-dichloro-2-propyl)phosphate (TDCIPP), and BDE-
47), biomarkers of exposure in children were higher those whose homes had a lower
cleaning frequency (van der Schyff et al., 2023b). Given that indoor dust is a known
reservoir of flame retardants (Jílková et al., 2018; Venier et al., 2016; Vykoukalová et al.,
2017) removing dust from living spaces is a good way to limit exposure, and has been
linked to lower levels of flame retardants in indoor environments (Sugeng et al., 2018).
16
Occupation is known to be a key determinant of
exposure for many occupations associated with high
chemical releases, for example, e-waste processing
(Balasch et al., 2022; Julander et al., 2014; Nguyen et al.,
2019) and construction (Estill et al., 2020; Jarvholm,
2006; Wingfors et al., 2006). However, with the ubiquity
of synthetic materials containing additive chemicals in
all indoor environments, it is not only industrial
occupations that are associated with chemical
exposure, but rather all time spent indoors. In
developed countries, we spend an average of 90% of
our time in indoor spaces (Figure 6), largely distributed
between home and work/school, and consequently,
these environments are key determinants of our
chemical exposures (Matz et al., 2014; Schweizer et al.,
2007).
When spending time indoors, we inhale indoor air, including indoor aerosols, and through
hand-to-mouth behaviours, accidentally ingest settled dust particles, which can be a
particularly large reservoir of lipophilic compounds (Weschler and Nazaroff, 2010). We
additionally have direct contact with consumer products and dust particles, leading to
potential dermal exposure (Weschler and Nazaroff, 2012). While initial concerns about
indoor environments focussed largely on inhalation exposure, and this is often dominant
in occupational exposure settings, non-dietary ingestion and dermal contact are important
exposure pathways in residential and non-industrial settings (Lioy et al., 2002; Salthammer
et al., 2018). Young children are particularly susceptible to high exposures to house dust
given their increased frequency of hand-to-mouth behaviour and mouthing of objects
(Moya and Phillips, 2014); ingestion of indoor dust is the dominant pathway for young
children’s exposure to some flame retardants (Demirtepe et al., 2019; Johnson-Restrepo
and Kannan, 2009). Dermal exposure, although receiving limited attention, also appears to
be an important pathway. Comparisons of hand wipe profiles of Canadian women with the
chemical profiles on the surfaces of their electronics found similarities in the profiles
between the flame retardants in hand-held electronics and their hand surfaces (Diamond
et al., 2021), although substantial uncertainty remains regarding dermal penetration.
Estimates of the contribution of inhalation, ingestion and dermal exposure pathways in
Slovak children found that dermal contact contributed ~30% of total exposures for
PBDEs and halogenated OPEs (Demirtepe et al., 2019, Appendix 8). There is a large set of
55%31%
5%
9%
Home indoors
Work indoors
Other indoors (shop, vehicle,
restaurant, etc)
Outdoors
Figure 6 Typical distribution
of time for adult, Prague. Data
from Schweizer et al. (2007).
17
studies suggesting that the concentrations of air and dust in indoor spaces are associated
with human exposure to a range of chemicals, notably for flame retardants, PCBs and
phthalate esters (Coakley et al., 2013; Fromme et al., 2014; Herrick et al., 2011), and thus
determinants of indoor environmental quality will also be determinants of human exposure
to selected chemicals.
The age of a building is frequently cited as a primary environmental factor impacting
chemical levels in indoor air and dust. There are two layers to this age impact. One is related
to chemical regulations as discussed in section 2.1. Older buildings can contain elevated
residues of chemicals which are now restricted, particularly those that were built at the
time of peak use of some legacy POPs of concern, e.g., PCBs and DDT. In some cases, this
is related to documented past use of the chemicals indoors. For example, a wood
preservative called pentalidol (2% DDT; 5% pentachlorophenol and 0.1% γhexachlorocyclohexane
(HCH)) was applied in the Baroque theater of Český Krumlov Castle
in the 1970s and 1980s to address dry rot, and this resulted in acute health impacts in the
1990s, before remediation removed 5 t of contaminated material (Holt et al., 2017,
Appendix 9). However, even after this removal exceedances of exposure guidance
values remained for DDT and HCHs, and indoor levels were orders of magnitude
above background levels (Holt et al., 2017), indicating the persistence of these
chemicals indoors, as well as their emission from secondary indoor sources. But in
many cases, surveys of indoor air or dust find an association between levels of legacy POPs
in indoor air and dust without documented sources, indicating past general use of the
chemicals. For example, higher levels of OCPs were detected in dust and air from
Slovak homes built before restrictions on the chemical use (Figure 7, Demirtepe et al.,
2019, Appendix 8). This is in line with what is the known about past use of both PCBs and
DDT, as uses were very diverse, but were not well-documented.
18
Figure 7 Differences in indoor dust and air concentrations of ∑DDX (a and b) and ∑HCH (c and d) measured
in homes built before (n = 18 for ∑DDX, n = 22 for ∑HCH) and after (n = 14 for ∑DDX, n = 10 for ∑HCH)
OCPs controls were implemented in Czechoslovakia. Excerpted from Demirtepe et al. (2019).
The properties of indoor environments are ideal for the long-term persistence of chemical
residues. The potential for degradation is much lower than outdoors due to lower solar
radiation, lower levels of atmospheric oxidants, high surface area-to-volume ratios, small
variations in temperature, and low air exchange (Abbatt and Wang, 2020). As a result,
chemical half-lives indoors can be substantially longer than what is estimated for outdoor
environments, and these are poorly characterized (Abbatt and Wang, 2020).
However, on top of differences seen due to regulations, in newer buildings we observe the
impact of a transition to indoor environments based heavily on synthetic polymer-based
materials. The past 70 years have seen a substantial increase in the amount of synthetic
materials used in indoor spaces (Weschler 2009), and in conjunction, an increase in the
amount of chemical additives found in these materials. Typical building materials and
consumer products today have a heavy reliance on plastics (e.g., building materials), and
consequently, higher use of chemical additives such as plasticizers, flame retardants,
stabilizers and antioxidants in indoor spaces. Many of these chemicals are semi-volatile,
and the exposure pathways are then multifaceted, as these semi-volatile organic
compounds (SVOCs) are distributed through multiple phases in indoor environments,
19
leading to exposure through air, dust and surface contact (Weschler and Nazaroff, 2008).
This wide range of synthetic chemicals used as additives in building materials and
furnishings are now consistently detected in most indoor spaces, the most notable being
phthalate esters, organophosphate flame retardants, chlorinated paraffins, and siloxanes
(Lucattini et al., 2018).
Building materials and consumer products have very large ranges in the presence of
additive chemicals. In an examination of the chemical content of 126 Czech consumer
products and building materials, we identified order of magnitude ranges even within
individual product groups. For example, levels of perfluoroalkyl acids ranged from
0.0068-34.3 µg/kg across 16 insulation materials (Bečanová et al., 2016, Appendix 10), while
BDE-209 ranged from 1.77-626000 µg/kg (Vojta et al., 2017a, Appendix 11). It is clear that
the chemical content of consumer products and building materials has a direct impact on
indoor environmental levels and human exposures, however the larger variability in
product groups creates a challenge in generalizing sources. Through the subsequent
introduction of new materials and use of those materials in new buildings, we see an
increase in the indoor environmental burdens of chemicals. During the construction phases
of a university lecture room, the levels of flame retardants increased in steps with the
addition of carpet, furnishings, and most substantially, with turning on of computers,
suggesting the specific contribution of these items to the indoor environmental burden of
chemicals (Vojta et al., 2017b, Appendix 12). We see in new indoor environments a
transition from the older brominated flame retardants (Venier et al., 2016, Appendix 3)
to the replacement chemicals, which today are found at levels 1-3 orders of
magnitude higher than older BFRs (Vykoukalová et al., 2017, Appendix 13).
Ventilation rates are another key factor impacting chemical levels indoors: for most
synthetic chemicals, higher ventilation rates will lead to reductions in indoor chemical
levels, as the primary chemical sources are in the indoor space. In urban areas, the impact
of ventilation as a source to outdoors is substantial enough to one of the main factors
driving outdoor air concentrations of synthetic organic chemicals (Björklund et al., 2012;
Melymuk et al., 2012). In Czech homes, lower levels of novel flame retardants were
observed in summer compared with winter, attributed to higher summer air exchange
rates due to open windows (Melymuk et al., 2016, Appendix 14).
However, many more indoor determinants are chemical-specific. The presence of PVC
flooring in homes has been associated with higher levels of phthalates (Bornehag et al.,
2005), while the presence of TVs and electronics in general has been associated with higher
levels of flame retardants (Harrad et al., 2009; Muenhor and Harrad, 2012; Yang et al., 2020).
Reductions in indoor settled dust levels of bis(2-ethylhexyl)phthalate (DEHP), a phthalate
ester that has now been restricted in many applications, suggest that stocks and indoor
uses of DEHP are being rapidly removed from indoor environments, and provide evidence
20
of the effectiveness of restrictions in reducing individual chemical exposures
(Demirtepe et al., 2021, Appendix 15).
While for many chemicals the indoor environment is the dominant exposure determinant,
outdoor exposures can be important for certain chemicals and lifestyle patterns. Of the
exposure pathways outside, air inhalation is most often the focus, although soil ingestion
can be substantial for young children (Özkaynak et al., 2023). Established air monitoring
networks exist to track synthetic chemicals in outdoor environments. For SVOCs, these are
notably the MONET networks (White et al., 2023) and the GAPS network (Pozo et al., 2006;
Saini et al., 2020), both of which rely on passive air sampling to increase the spatial
resolution of available data. Passive air sampling (PAS) relies on passive diffusion of
chemicals from air to a sorbent (Shoeib and Harner, 2002; Wania and Shunthirasingham,
2020), in the case of MONET and GAPS these sorbents are based on polyurethane foam
(Harner et al., 2013; Melymuk et al., 2011), XAD resin (Armitage et al., 2013), or a
combination of both (Schuster et al., 2012). They have the advantage in outdoor
environments of not requiring electricity, being low-cost and easy to deploy, which has
greatly increased the spatial resolution of outdoor air monitoring. However, even with the
greater spatial resolution possible because of passive sampling, measurements are limited
to point locations, which may not reflect the small-scale variations observed in populated
areas (Melymuk et al., 2013, 2012). To supplement these air monitoring networks, data
modeling techniques can be incorporated to provide better spatial resolution of air
concentration estimates that can be incorporated into exposure assessment (Hoek et al.,
2008; Mikeš et al., 2023). Land use regression has proved to be a useful technique to
translate coverage of passive air sampling networks into concentration maps providing
better spatial resolution. For Czechia, outdoor ambient air concentrations of PCBs were
found to be dependent on soil concentrations and topography, while PAHs were
related to fuel consumption and industrial sources, as well as topography (White et
al., 2021, Appendix 16); the identification of such relationships allows point passive sampler
measurements to be extended to provide broader spatial coverage. The regression
relationships generated by this analysis could be used to produce maps highlight regional
variations in outdoor concentrations (Figure 8). Similar techniques have been applied for
other air pollutants (Mikeš et al., 2023) and SVOCs in different locations (Melymuk et al.,
2013).
21
Figure 8 Application of the LUR models across all of Czechia, indicating the estimated gradient of low to
high atmospheric levels of Σ7PCB, (R2
= 0.62). Cities with population > 100,000 are indicated on the map.
From White et al. (2021)
A comprehensive understanding of human exposure requires a diverse set of sampling,
analytical, and data-handling tools. One fundamental step in characterizing human
exposure pathways is selection of representative and unbiased environmental samples. For
characterization of indoor environments, a set of sampling techniques unique to indoor
spaces is often used.
Passive air samplers, particularly polyurethane foam-based passive air samplers (PUF-PAS)
have been a major innovation in indoor environments, with the significant advantages of
being non-intrusive indoors and providing integrated concentrations, typically over one
month (Bohlin et al., 2014; Vojta et al., 2024). While some uncertainty remains in the
conversion of masses sampled to quantified air concentrations due to uncertainty in air
uptake to the sampler (Wania and Shunthirasingham, 2020) indoor environments are
actually ideal for PAS, as they are not subject to the large variations in temperature and
wind speed, which are known to vary sampling rates outdoors (Bohlin-Nizzetto et al., 2020;
22
Chaemfa et al., 2009), and lower levels of atmospheric oxidants, which should minimize the
within-sampler degradation that has been noted in outdoor PUF-PAS (Melymuk et al.,
2017). The applicability of PUF-PAS is broad, however some limitations exist. As a sorbent,
PUF is best suited to lipophilic compounds of intermediate volatility – more volatile
compounds may be susceptible to non-linear uptake or breakthrough (Melymuk et al.,
2014, Appendix 17), while polar compounds are not well-captured by the sorbent. This
presents a particular limitation for their application to per- and polyfluoroalkyl substances
(PFAS), as PUF-PAS have not been found to have consistent uptake of many PFAS over
a typical deployment period, which precludes the conversion of masses sorbed to air
concentrations (Karásková et al., 2018, Appendix 18).
PUF-PAS were originally developed for use outdoors (Bidleman and Melymuk, 2019;
Harner et al., 2006), and this remains the most substantial use (e.g., GAPS and MONET, as
mentioned in above). The low cost and ease of use has led to many regional variations in
the samplers with slight differences in sampler geometry, sorbent mass and density
(Melymuk et al., 2021, Appendix 19). However, the sampler performance appears
substantially robust that such small variations do not lead to large impacts on
chemical uptake to the PUF; variations caused by sampler geometry are within the same
ranges as uncertainties due to wind and temperature impacts on the samplers (BohlinNizzetto
et al., 2020; Chaemfa et al., 2009; Melymuk et al., 2021).
Many broad surveys of chemical pollution in indoor spaces use settled indoor dust as an
indicator of indoor levels and human exposure (Lioy et al., 2002; Melymuk et al., 2020).
Dust has the advantage of being easy to collect, a relatively stable matrix that can be easily
transported, and typically has high levels of SVOC chemicals, making it a good screening
matrix for a broad set of compounds. In addition, unlike PAS, it presents a reliable matrix
for the quantification of indoor levels of PFAS (Karásková et al., 2016), which are of
particular interest given the EU universal PFAS restriction proposal of 2023 (ECHA, 2024).
However, in order to collect a dust sample representative of the average conditions of an
indoor space, care must be taken to collect an aggregated sample: within an indoor space.
Substantial variations in the chemical levels in indoor dust exist between different surfaces
(Jílková et al., 2018, Appendix 20); samples from individual surfaces can be strongly
influenced by individual products and not representative of the whole room
conditions. For example, in floor dust from classrooms was substantially elevated in two
flame retardants, TCIPP and DBDPE, relative to dust on surfaces, whereas in a flat, surface
dust was elevated in the same compounds (Figure 9, Jílková et al., 2018). Given the smallscale
spatial heterogeneity in dust concentrations, composite or aggregated samples are
needed to reflect average exposures; localized samples can show substantial bias.
23
Figure 9 Area density of TCIPP and DBDPE in a flat and a classroom floor dust and wipe samples.
Horizontal lines show mean and whiskers show standard deviation. Extracted from Jilkova et al. (2018). SR
refers to Seminar room and CR refers to Computer Room.
While uncertainties in sampler configurations and implementation can impact data quality
and comparability across studies, the greater source of uncertainty is consistently due to
aspects of the laboratory analysis. In the global intercomparison of PAS configurations
discussed in Section 3.1, while sampler configurations accounted for 50% variations
between concentrations, laboratory analysis and quantification were responsible for
up to four orders of magnitude variations in concentrations of some POPs (Melymuk
et al., 2021, Appendix 19). Similarly, for FRs the variations in elements of the analytical
method across laboratories lead to substantial variations in reported chemical
concentrations and present challenges when comparing results across studies (Melymuk
et al., 2018).
Given the diversity of chemicals of interest in indoor environments, interest is growing in
non-target and suspect screening analysis of chemicals. While these are starting to be
more widely used on indoor dust in particular, a challenge still exists in the comparability
and the ability to interpret this data at a sufficient quality necessary to understand
exposures (Caballero-Casero et al., 2021; Hollender et al., 2023; Rostkowski et al., 2019).
This will be one of the major future challenges as we develop the field of indoor exposure
assessment.
Moreover, interpretation of exposure data can rarely be effectively done in the absence of
other supporting information, taking the form of questionnaires, chemical use databases,
product content, and time-activity patterns. It is crucial that these datasets be given equal
TCIPP DBDPE
24
consideration as the chemical datasets, as evidence of exposures in the absence of source
information has limited value.
Our framework for understanding the drivers of chemical exposure provides a structure for
the identification of the most substantial drivers. The goal of exposure assessment is to
contribute to an overall risk evaluation and reduction process to identify chemical risks,
and provide the data needed to take actions to reduce such risks.
These actions can take many forms, be it regulatory, behavioural or technological, however,
the necessary fundament to selecting the correct actions is a sufficient understanding of
the sources and key factors driving the exposure. Thus, our evaluation of exposures in the
context of regulation, environment and personal behaviours provides a valuable resource
for both evaluation exposure, and where needed supporting actions to reduce risks.
Regulatory actions have been shown to have a clear impact on either increasing or
decreasing exposures, and ongoing effectiveness evaluation is crucial to support the most
effective implementation of chemical policy, and other regulatory actions that impact
chemical exposures (Chapter 2.1). Behaviour and lifestyle provide an additional, more
individual layer to understanding exposures, with the potential to lead to wide variations
in exposures within similar populations (Chapter 2.2). Finally, our environment, particularly
the indoor spaces where we spend the majority of our time, is a major driver of exposure
to selected chemicals (Chapter 2.3), and the impact of the environment and, in particular,
actionable solutions to improve environmental quality to reduce exposures, has not been
fully exploited in the current scientific framework.
We have developed a reliable set of sampling and analytical tools that can enable us to
explore the links between environment, behaviour, policy and exposure (Chapter 3). Many
of these tools have the advantage of broad applicability for chemicals of interest, and ease
of use enabling widespread data collection. However, better interlinkages and integrations
of techniques are needed to allow more comprehensive determinations and data gap-
filling.
25
1. van der Schyff, Veronica, Jiří Kalina, Annalisa Abballe, Anna Laura Iamiceli, Eva
Govarts, and Lisa Melymuk. 2023. "Has Regulatory Action Reduced Human
Exposure to Flame Retardants?" Environmental Science & Technology 57 (48):
19106-24. https://doi.org/10.1021/acs.est.3c02896
2. Melymuk, Lisa, Jonathan Blumenthal, Ondřej Sáňka, Adriana Shu-Yin, Veena Singla,
Kateřina Sebková, Kristi Pullen Fedinick, and Miriam L Diamond. 2022. "Persistent
Problem: Global Challenges to Managing PCBs." Environmental Science &
Technology 2022: 9040. https://doi.org/10.1021/acs.est.2c01204
3. Venier, Marta, Ondřej Audy, Šimon Vojta, Jitka Bečanová, Kevin Romanak, Lisa
Melymuk, Martina Krátká, Petr Kukucka, Joseph Okeme, Amandeep Saini, Miriam L.
Diamond, Jana Klanova. 2016. "Brominated Flame Retardants in Indoor
Environment - Comparative Study of Indoor Contamination from Three Countries."
Environment International 94: 150-60. https://doi.org/10.1016/j.envint.2016.04.029
4. van der Schyff, Veronica, Jiři Kalina, Eva Govarts, Liese Gilles, Greet Schoeters,
Argelia Castaño, Marta Esteban-López, Jiři Kohoutek, Petr Kukučka, Adrian Covaci,
Gudrun Koppen, Lenka Andrýsková, Pavel Piler, Jana Klánová, Tina Kold Jensen, Loic
Rambaud, Margaux Riou, Marja Lamoree, Marike Kolossa-Gehring, Nina Vogel, Till
Weber, Thomas Göen, Catherine Gabriel, Dimosthenis A. Sarigiannis, Amrit Kaur
Sakhi, Line Småstuen Haug, Lubica Palkovicova Murinova, Lucia Fabelova, Janja Snoj
Tratnik, Darja Mazej, Lisa Melymuk. 2023. "Exposure to Flame Retardants in
European Children - Results from the HBM4EU Aligned Studies." International
Journal of Hygiene and Environmental Health 247: 114070.
https://doi.org/10.1016/j.ijheh.2022.114070
5. Diamond, Miriam L., Joseph O. Okeme, and Lisa Melymuk. 2021. "Hands as Agents
of Chemical Transport in the Indoor Environment." Environmental Science &
Technology Letters 8 (4): 326-32. https://doi.org/10.1021/acs.estlett.0c01006
6. Stuchlík Fišerová, Petra, Lisa Melymuk, Klára Komprdová, Elena DomínguezRomero,
Martin Scheringer, Jiří Kohoutek, Petra Přibylová, Lenka Andrýsková, Pavel
Piler, Holger M. Koch, Martin Zvonař, Marta Esteban-López, Argelia Castaño, Jana
Klánová. 2022. "Personal Care Product Use and Lifestyle Affect Phthalate and DINCH
Metabolite Levels in Teenagers and Young Adults." Environmental Research 213:
113675. https://doi.org/10.1016/j.envres.2022.113675
26
7. Schyff, Veronica van der, Lenka Suchánková, Katerina Kademoglou, Lisa Melymuk,
and Jana Klánová. 2022. "Parabens and Antimicrobial Compounds in Conventional
and 'Green' Personal Care Products." Chemosphere 297: 1-8.
https://doi.org/10.1016/j.Chemosphere.2022.134019
8. Demirtepe, Hale, Lisa Melymuk, Miriam L. Diamond, Lola Bajard, Šimon Vojta,
Roman Prokeš, Ondřej Sáňka, Jana Klánová, Ľubica Palkovičová Murínová, Denisa
Richterová, Vladimíra Rašplová, Tomáš Trnovec. 2019. "Linking Past Uses of Legacy
SVOCs with Today's Indoor Levels and Human Exposure." Environment International
127: 653-63. https://doi.org/10.1016/j.envint.2019.04.001
9. Holt, Eva, Ondřej Audy, Petra Booij, Lisa Melymuk, Roman Prokes, and Jana Klánová.
2017. "Organochlorine Pesticides in the Indoor Air of a Theatre and Museum in the
Czech Republic: Inhalation Exposure and Cancer Risk." Science of The Total
Environment 609: 598-606. https://doi.org/10.1016/j.scitotenv.2017.07.203
10. Bečanová, Jitka, Lisa Melymuk, Šimon Vojta, Klára Komprdová, and Jana Klánová.
2016. "Screening for Perfluoroalkyl Acids in Consumer Products, Building Materials
and Wastes." Chemosphere 164: 322-29.
https://doi.org/10.1016/j.Chemosphere.2016.08.112
11. Vojta, Šimon, Jitka Bečanová, Lisa Melymuk, Klára Komprdová, Jiří Kohoutek, Petr
Kukučka, and Jana Klánová. 2017. "Screening for Halogenated Flame Retardants in
European Consumer Products, Building Materials and Wastes." Chemosphere 168:
457-66. https://doi.org/10.1016/j.Chemosphere.2016.11.032
12. Vojta, Šimon, Lisa Melymuk, and Jana Klánová. 2017. "Changes in Flame Retardant
and Legacy Contaminant Concentrations in Indoor Air during Building Construction,
Furnishing, and Use." Environmental Science & Technology 51 (20): 11891-99.
https://doi.org/10.1021/acs.est.7b03245
13. Vykoukalová, Martina, Marta Venier, Šimon Vojta, Lisa Melymuk, Jitka Bečanová,
Kevin Romanak, Roman Prokeš, Joseph O. Okeme, Amandeep Saini, Miriam L.
Diamond, Jana Klanova. 2017. "Organophosphate Esters Flame Retardants in the
Indoor Environment." Environment International 106: 97-104.
https://doi.org/10.1016/j.envint.2017.05.020
27
14. Melymuk, Lisa, Pernilla Bohlin-Nizzetto, Petr Kukučka, Šimon Vojta, Jiří Kalina, Pavel
Cupr, Jana Klánová, Pavel Čupr, and Jana Klánová. 2016. "Seasonality and
Indoor/Outdoor Relationships of Flame Retardants and PCBs in Residential Air."
Environmental Pollution 218: 392-401. https://doi.org/10.1016/j.envpol.2016.07.018
15. Demirtepe, Hale, Lisa Melymuk, Garry Codling, Ľubica Palkovičová Murínová, Denisa
Richterová, Vladimíra Rašplová, Tomáš Trnovec, and Jana Klánová. 2021. "Targeted
and Suspect Screening of Plasticizers in House Dust to Assess Cumulative Human
Exposure Risk." Science of The Total Environment 781: 146667.
https://doi.org/10.1016/j.scitotenv.2021.146667
16. White, Kevin B., Ondřej Sáňka, Lisa Melymuk, Petra Přibylová, and Jana Klánová.
2021. "Application of Land Use Regression Modelling to Describe Atmospheric
Levels of Semivolatile Organic Compounds on a National Scale." Science of The Total
Environment 793. https://doi.org/10.1016/j.scitotenv.2021.148520
17. Melymuk, Lisa, Pernilla Bohlin-Nizzetto, Ondřej Sáňka, Karla Pozo, and Jana Klánová.
2014. "Current Challenges in Air Sampling of Semivolatile Organic Contaminants:
Sampling Artifacts and Their Influence on Data Comparability." Environmental
Science & Technology 48 (24): 14077-91. https://doi.org/10.1021/es502164r
18. Karásková, Pavlína, Garry P. Codling, Lisa Melymuk, and Jana Klánová. 2018. "A
Critical Assessment of Passive Air Samplers for Per- and Polyfluoroalkyl Substances."
Atmospheric Environment 185: 186-95.
https://doi.org/10.1016/j.atmosenv.2018.05.030
19. Melymuk, Lisa, Pernilla Bohlin Nizzetto, Tom Harner, Kevin B. White, Xianyu Wang,
Maria Yumiko Tominaga, Jun He, Jun Li, Jianmin Ma, Wan-Li Ma, Beatriz H.
Aristizábal, Annekatrin Dreyer, Begoña Jiménez, Juan Muñoz-Arnanz, Mustafa
Odabasi, Yetkin Dumanoglu, Baris Yaman, Carola Graf, Andrew Sweetman, Jana
Klánová. 2021. "Global Intercomparison of Polyurethane Foam Passive Air Samplers
Evaluating Sources of Variability in SVOC Measurements." Environmental Science
and Policy 125: 1-9. https://doi.org/10.1016/j.envsci.2021.08.003
20. Jílková, Simona, Lisa Melymuk, Šimon Vojta, Martina Vykoukalová, Pernilla BohlinNizzetto,
and Jana Klánová. 2018. "Small-Scale Spatial Variability of Flame
Retardants in Indoor Dust and Implications for Dust Sampling." Chemosphere 206:
132-41. https://doi.org/10.1016/J.CHEMOSPHERE.2018.04.146
28
Abbasi, G., Li, L., Breivik, K., 2019. Global Historical Stocks and Emissions of PBDEs.
Environmental Science & Technology 53, 6330–6340.
https://doi.org/10.1021/acs.est.8b07032
Abbasi, G., M. Buser, A., Soehl, A., W. Murray, M., L. Diamond, M., 2015. Stocks and flows of
PBDEs in products from use to waste in the U.S. and Canada from 1970 to 2020.
Environmental science & technology 49, 1521–1528. https://doi.org/10.1021/es504007v
Abbatt, J.P.D., Wang, C., 2020. The atmospheric chemistry of indoor environments.
Environmental Science: Processes and Impacts 22, 25–48.
https://doi.org/10.1039/c9em00386j
Armitage, J.M., Hayward, S.J., Wania, F., 2013. Modeling the uptake of neutral organic
chemicals on XAD passive air samplers under variable temperatures, external wind speeds
and ambient air concentrations (PAS-SIM). Environmental science & technology 47, 13546–
54. https://doi.org/10.1021/es402978a
Balasch, A., López, M., Reche, C., Viana, M., Moreno, T., Eljarrat, E., 2022. Exposure of ewaste
dismantlers from a formal recycling facility in Spain to inhalable organophosphate
and halogenated flame retardants. Chemosphere 294.
https://doi.org/10.1016/j.chemosphere.2022.133775
Bečanová, J., Melymuk, L., Vojta, Š., Komprdová, K., Klánová, J., 2016. Screening for
perfluoroalkyl acids in consumer products, building materials and wastes. Chemosphere
164, 322–329. https://doi.org/10.1016/j.chemosphere.2016.08.112
Bernhardt, E.S., Rosi, E.J., Gessner, M.O., 2017. Synthetic chemicals as agents of global
change. Frontiers in Ecology and the Environment 15, 84–90.
https://doi.org/10.1002/fee.1450
Bidleman, T.F., Melymuk, L., 2019. Forty-five Years of Foam: A Retrospective on Air
Sampling with Polyurethane Foam. Bulletin of Environmental Contamination and
Toxicology 102, 447–449. https://doi.org/10.1007/s00128-019-02591-4
Björklund, J.A., Thuresson, K., Cousins, A.P., Sellström, U., Emenius, G., de Wit, C.A., 2012.
Indoor air is a significant source of tri-decabrominated diphenyl ethers to outdoor air via
ventilation systems. Environmental science & technology 46, 5876–84.
https://doi.org/10.1021/es204122v
Bogdal, C., Schmid, P., Kohler, M., Müller, C.E., Iozza, S., Bucheli, T.D., Scheringer, M.,
Hungerbühler, K., 2008. Sediment Record and Atmospheric Deposition of Brominated
Flame Retardants and Organochlorine Compounds in Lake Thun, Switzerland: Lessons from
29
the Past and Evaluation of the Present. Environmental Science & Technology 42, 6817–
6822. https://doi.org/10.1021/es800964z
Bohlin, P., Audy, O., Škrdlíková, L., Kukučka, P., Vojta, Š., Přibylová, P., Prokeš, R., Čupr, P.,
Klánová, J., 2014. Evaluation and guidelines for using polyurethane foam (PUF) passive air
samplers in double-dome chambers to assess semi-volatile organic compounds (SVOCs)
in non-industrial indoor environments. Environmental Science: Processes & Impacts 16,
2617–2626. https://doi.org/10.1039/C4EM00305E
Bohlin-Nizzetto, P., Melymuk, L., White, K.B., Kalina, J., Madadi, V.O., Adu-Kumi, S., Prokeš,
R., Přibylová, P., Klánová, J., 2020. Field- and model-based calibration of polyurethane foam
passive air samplers in different climate regions highlights differences in sampler uptake
performance. Atmospheric Environment 238.
https://doi.org/10.1016/j.atmosenv.2020.117742
Bornehag, C.G., Lundgren, B., Weschler, C.J., Sigsgaard, T., Hagerhed-Engman, L., Sundell,
J., 2005. Phthalates in indoor dust and their association with building characteristics.
Environmental Health Perspectives 113, 1399–1404. https://doi.org/10.1289/ehp.7809
Breivik, K., Sweetman, A., Pacyna, J.M., Jones, K.C., 2002. Towards a global historical
emission inventory for selected PCB congeners - A mass balance approach: 2. Emissions.
Science of the Total Environment 290, 199–224. https://doi.org/10.1016/S0048-
9697(01)01076-2
Budnik, L.T., Casteleyn, L., 2019. Mercury pollution in modern times and its socio-medical
consequences. Science of The Total Environment 654, 720–734.
https://doi.org/10.1016/j.scitotenv.2018.10.408
Caballero-Casero, N., Belova, L., Vervliet, P., Antignac, J.-P., Castaño, A., Debrauwer, L.,
López, M.E., Huber, C., Klanova, J., Krauss, M., Lommen, A., Mol, H.G.J., Oberacher, H., Pardo,
O., Price, E.J., Reinstadler, V., Vitale, C.M., van Nuijs, A.L.N., Covaci, A., 2021. Towards
harmonised criteria in quality assurance and quality control of suspect and non-target LCHRMS
analytical workflows for screening of emerging contaminants in human
biomonitoring. TrAC Trends in Analytical Chemistry 136, 116201.
https://doi.org/10.1016/j.trac.2021.116201
Chaemfa, C., Wild, E., Davison, B., Barber, J.L., Jones, K.C., 2009. A study of aerosol
entrapment and the influence of wind speed, chamber design and foam density on
polyurethane foam passive air samplers used for persistent organic pollutants. Journal of
environmental monitoring 11, 1135–9. https://doi.org/10.1039/B823016A
Chin, J., Jiang, B., Mufidah, I., Persada, S., Noer, B., 2018. The Investigation of Consumers’
Behavior Intention in Using Green Skincare Products: A Pro-Environmental Behavior Model
Approach. Sustainability 10, 3922. https://doi.org/10.3390/su10113922
30
Coakley, J.D., Harrad, S.J., Goosey, E.R., Ali, N., Dirtu, A.C., Van den Eede, N., Covaci, A.,
Douwes, J., Mannetje, A. t., 2013. Concentrations of polybrominated diphenyl ethers in
matched samples of indoor dust and breast milk in New Zealand. Environment
International 59, 255–261. https://doi.org/10.1016/j.envint.2013.06.020
Crimmins, B.S., Pagano, J.J., Xia, X., Hopke, P.K., Milligan, M.S., Holsen, T.M., 2012.
Polybrominated diphenyl ethers (PBDEs): Turning the corner in Great Lakes trout 1980-
2009. Environmental Science and Technology 46, 9890–9897.
https://doi.org/10.1021/es302415z
Darnerud, P.O., 2003. Toxic effects of brominated flame retardants in man and in wildlife.
Environment international 29, 841–53. https://doi.org/10.1016/S0160-4120(03)00107-7
Demirtepe, H., Melymuk, L., Codling, G., Murínová, Ľ.P., Richterová, D., Rašplová, V., Trnovec,
T., Klánová, J., 2021. Targeted and suspect screening of plasticizers in house dust to assess
cumulative human exposure risk. Science of The Total Environment 781, 146667.
https://doi.org/10.1016/j.scitotenv.2021.146667
Demirtepe, H., Melymuk, L., Diamond, M.L., Bajard, L., Vojta, Š., Prokeš, R., Sáňka, O.,
Klánová, J., Palkovičová Murínová, Ľ., Richterová, D., Rašplová, V., Trnovec, T., 2019. Linking
past uses of legacy SVOCs with today’s indoor levels and human exposure. Environment
International 127, 653–663. https://doi.org/10.1016/j.envint.2019.04.001
Diamond, M.L., Okeme, J.O., Melymuk, L., 2021. Hands as Agents of Chemical Transport in
the Indoor Environment. Environmental Science & Technology Letters 8, 326–332.
https://doi.org/10.1021/acs.estlett.0c01006
Dronsfield, A., 2006. Percivall Pott, chimney sweeps and cancer. RSC Education.
https://edu.rsc.org/feature/percivall-pott-chimney-sweeps-and-cancer/2020205.article
ECHA, 2024. Per- and polyfluoroalkyl substances (PFAS). https://echa.europa.eu/hot-
topics/perfluoroalkyl-chemicals-pfas
Estill, C.F., Slone, J., Mayer, A., Chen, I.-C., La Guardia, M.J., 2020. Worker exposure to flame
retardants in manufacturing, construction and service industries. Environment International
135, 105349. https://doi.org/10.1016/j.envint.2019.105349
Fenner, K., Scheringer, M., 2021. The Need for Chemical Simplification As a Logical
Consequence of Ever-Increasing Chemical Pollution. Environmental Science and
Technology 55, 14470–14472. https://doi.org/10.1021/acs.est.1c04903
Fromme, H., Lahrz, T., Kraft, M., Fembacher, L., Mach, C., Dietrich, S., Burkardt, R., Völkel, W.,
Göen, T., 2014. Organophosphate flame retardants and plasticizers in the air and dust in
German daycare centers and human biomonitoring in visiting children (LUPE 3).
Environment International 71, 158–163. https://doi.org/10.1016/j.envint.2014.06.016
31
Fruh, V., Preston, E.V., Quinn, M.R., Hacker, M.R., Wylie, B.J., O’Brien, K., Hauser, R., JamesTodd,
T., Mahalingaiah, S., 2022. Urinary phthalate metabolite concentrations and personal
care product use during pregnancy – Results of a pilot study. Science of The Total
Environment 835, 155439. https://doi.org/10.1016/j.scitotenv.2022.155439
Govarts, E., Gilles, L., Rodriguez Martin, L., Santonen, T., Apel, P., Alvito, P., Anastasi, E.,
Andersen, H.R., Andersson, A.-M., Andryskova, L., Antignac, J.-P., Appenzeller, B., Barbone,
F., Barnett-Itzhaki, Z., Barouki, R., Berman, T., Bil, W., Borges, T., Buekers, J., Cañas-Portilla,
A., Covaci, A., Csako, Z., Den Hond, E., Dvorakova, D., Fabelova, L., Fletcher, T., Frederiksen,
H., Gabriel, C., Ganzleben, C., Göen, T., Halldorsson, T.I., Haug, L.S., Horvat, M., Huuskonen,
P., Imboden, M., Jagodic Hudobivnik, M., Janasik, B., Janev Holcer, N., Karakitsios, S.,
Katsonouri, A., Klanova, J., Kokaraki, V., Kold Jensen, T., Koponen, J., Laeremans, M., Laguzzi,
F., Lange, R., Lemke, N., Lignell, S., Lindroos, A.K., Lobo Vicente, J., Luijten, M., Makris, K.C.,
Mazej, D., Melymuk, L., Meslin, M., Mol, H., Montazeri, P., Murawski, A., Namorado, S.,
Niemann, L., Nübler, S., Nunes, B., Olafsdottir, K., Palkovicova Murinova, L., Papaioannou,
N., Pedraza-Diaz, S., Piler, P., Plichta, V., Poteser, M., Probst-Hensch, N., Rambaud, L.,
Rauscher-Gabernig, E., Rausova, K., Remy, S., Riou, M., Rosolen, V., Rousselle, C., Rüther, M.,
Sarigiannis, D., Silva, M.J., Šlejkovec, Z., Snoj Tratnik, J., Stajnko, A., Szigeti, T., Tarazona, J.V.,
Thomsen, C., Tkalec, Ž., Tolonen, H., Trnovec, T., Uhl, M., Van Nieuwenhuyse, A., Vasco, E.,
Verheyen, V.J., Viegas, S., Vinggaard, A.M., Vogel, N., Vorkamp, K., Wasowicz, W., Weber, T.,
Wimmerova, S., Woutersen, M., Zimmermann, P., Zvonar, M., Koch, H., Kolossa-Gehring, M.,
Esteban López, M., Castaño, A., Stewart, L., Sepai, O., Schoeters, G., 2023. Harmonized
human biomonitoring in European children, teenagers and adults: EU-wide exposure data
of 11 chemical substance groups from the HBM4EU Aligned Studies (2014–2021).
International Journal of Hygiene and Environmental Health 249, 114119.
https://doi.org/10.1016/j.ijheh.2023.114119
Harley, K.G., Kogut, K., Madrigal, D.S., Cardenas, M., Vera, I.A., Meza-Alfaro, G., She, J., Gavin,
Q., Zahedi, R., Bradman, A., Eskenazi, B., Parra, K.L., 2016. Reducing Phthalate, Paraben, and
Phenol Exposure from Personal Care Products in Adolescent Girls: Findings from the
HERMOSA Intervention Study. Environmental Health Perspectives 124, 1600–1607.
https://doi.org/10.1289/ehp.1510514
Harner, T., Pozo, K., Gouin, T., Macdonald, A.-M., Hung, H., Cainey, J., Peters, A., 2006. Global
pilot study for persistent organic pollutants (POPs) using PUF disk passive air samplers.
Environmental Pollution 144, 445–452. https://doi.org/10.1016/j.envpol.2005.12.053
Harner, T., Su, K., Genualdi, S., Karpowicz, J., Ahrens, L., Mihele, C., Schuster, J.K., Charland,
J.-P., Narayan, J., 2013. Calibration and application of PUF disk passive air samplers for
tracking polycyclic aromatic compounds (PACs). Atmospheric Environment 75, 123–128.
https://doi.org/10.1016/j.atmosenv.2013.04.012
32
Harrad, S.J., Abdallah, M.A.-E., Covaci, A., 2009. Causes of variability in concentrations and
diastereomer patterns of hexabromocyclododecanes in indoor dust. Environment
International 35, 573–579. https://doi.org/10.1016/j.envint.2008.10.005
Herrick, R.F., Meeker, J.D., Altshul, L., 2011. Serum PCB levels and congener profiles among
teachers in PCB-containing schools: a pilot study. Environmental Health 10, 56.
https://doi.org/10.1186/1476-069X-10-56
Hites, R.A., 2004. Polybrominated diphenyl ethers in the environment and in people: a
meta-analysis of concentrations. Environmental science & technology 38, 945–956.
https://doi.org/10.1021/es035082g
Hoek, G., Beelen, R., de Hoogh, K., Vienneau, D., Gulliver, J., Fischer, P., Briggs, D.J., 2008. A
review of land-use regression models to assess spatial variation of outdoor air pollution.
Atmospheric Environment 42, 7561–7578. https://doi.org/10.1016/j.atmosenv.2008.05.057
Hollender, J., Schymanski, E.L., Ahrens, L., Alygizakis, N., Béen, F., Bijlsma, L., Brunner, A.M.,
Celma, A., Fildier, A., Fu, Q., Gago-Ferrero, P., Gil-Solsona, R., Haglund, P., Hansen, M.,
Kaserzon, S., Kruve, A., Lamoree, M., Margoum, C., Meijer, J., Merel, S., Rauert, C.,
Rostkowski, P., Samanipour, S., Schulze, B., Schulze, T., Singh, R.R., Slobodnik, J., SteiningerMairinger,
T., Thomaidis, N.S., Togola, A., Vorkamp, K., Vulliet, E., Zhu, L., Krauss, M., 2023.
NORMAN guidance on suspect and non-target screening in environmental monitoring.
Environmental Sciences Europe 35, 75. https://doi.org/10.1186/s12302-023-00779-4
Holt, E., Audy, O., Booij, P., Melymuk, L., Prokes, R., Klánová, J., 2017. Organochlorine
pesticides in the indoor air of a theatre and museum in the Czech Republic: Inhalation
exposure and cancer risk. Science of The Total Environment 609, 598–606.
https://doi.org/10.1016/j.scitotenv.2017.07.203
Hudson-Hanley, B., Irvin, V., Flay, B., MacDonald, M., Kile, M.L., 2018. Prenatal PBDE
Exposure and Neurodevelopment in Children 7 Years Old or Younger: a Systematic Review
and Meta-analysis. Current Epidemiology Reports 5, 46–59.
https://doi.org/10.1007/s40471-018-0137-0
Jarvholm, B., 2006. Carcinogens in the Construction Industry. Annals of the New York
Academy of Sciences 1076, 421–428. https://doi.org/10.1196/annals.1371.055
Jílková, S., Melymuk, L., Vojta, Š., Vykoukalová, M., Bohlin-Nizzetto, P., Klánová, J., 2018.
Small-scale spatial variability of flame retardants in indoor dust and implications for dust
sampling. Chemosphere 206, 132–141.
https://doi.org/10.1016/J.CHEMOSPHERE.2018.04.146
33
Johnson-Restrepo, B., Kannan, K., 2009. An assessment of sources and pathways of human
exposure to polybrominated diphenyl ethers in the United States. Chemosphere 76, 542–
8. https://doi.org/10.1016/j.chemosphere.2009.02.068
Julander, A., Lundgren, L., Skare, L., Grandér, M., Palm, B., Vahter, M., Lidén, C., 2014. Formal
recycling of e-waste leads to increased exposure to toxic metals: An occupational exposure
study from Sweden. Environment International 73, 243–251.
https://doi.org/10.1016/j.envint.2014.07.006
Kalina, J., Scheringer, M., Borůvková, J., Kukucka, P., Přibylová, P., Sáňka, O., Melymuk, L.,
Vana, M., Klánová, J., 2018. Characterizing spatial diversity of passive sampling sites for
measuring levels and trends of semivolatile organic chemicals. Environmental Science &
Technology 52, 10599–10608. https://doi.org/10.1021/acs.est.8b03414
Karásková, P., Codling, G.P., Melymuk, L., Klánová, J., 2018. A critical assessment of passive
air samplers for per- and polyfluoroalkyl substances. Atmospheric Environment 185, 186–
195. https://doi.org/10.1016/j.atmosenv.2018.05.030
Karásková, P., Venier, M., Melymuk, L., Bečanová, J., Vojta, Š., Prokeš, R., Diamond, M.L.,
Klánová, J., 2016. Perfluorinated alkyl substances (PFASs) in household dust in Central
Europe and North America. Environment international 94, 315–324.
https://doi.org/10.1016/j.envint.2016.05.031
Krause, M., Andersson, A.-M., Skakkebaek, N.E., Frederiksen, H., 2017. Exposure to UV filters
during summer and winter in Danish kindergarten children. Environment International 99,
177–184. https://doi.org/10.1016/j.envint.2016.11.011
Lioy, P.J., Freeman, N.C.G., Millette, J.R., 2002. Dust: A Metric for Use in Residential and
Building Exposure Assessment and Source Characterization. Environmental Health
Perspectives 110, 969–983. https://doi.org/10.1289/ehp.02110969
Lucattini, L., Poma, G., Covaci, A., de Boer, J., Lamoree, M.H., Leonards, P.E.G., 2018. A review
of semi-volatile organic compounds (SVOCs) in the indoor environment: occurrence in
consumer products, indoor air and dust. Chemosphere 201, 466–482.
https://doi.org/10.1016/j.chemosphere.2018.02.161
Magulova, K., Priceputu, A., 2016. Global monitoring plan for persistent organic pollutants
(POPs) under the Stockholm Convention: Triggering, streamlining and catalyzing global
POPs monitoring. Environmental Pollution 217, 82–84.
https://doi.org/10.1016/j.envpol.2016.01.022
Matz, C., Stieb, D., Davis, K., Egyed, M., Rose, A., Chou, B., Brion, O., 2014. Effects of Age,
Season, Gender and Urban-Rural Status on Time-Activity: Canadian Human Activity Pattern
34
Survey 2 (CHAPS 2). International Journal of Environmental Research and Public Health 11,
2108–2124. https://doi.org/10.3390/ijerph110202108
Melymuk, L., Blumenthal, J., Sáňka, O., Shu-Yin, A., Singla, V., Sebková, K., Fedinick, K.P.,
Diamond, M.L., 2022. Persistent Problem: Global Challenges to Managing PCBs. Environ.
Sci. Technol 2022, 9040. https://doi.org/10.1021/acs.est.2c01204
Melymuk, L., Bohlin-Nizzetto, P., Kukučka, P., Vojta, Š., Kalina, J., Cupr, P., Klánová, J., Čupr,
P., Klánová, J., 2016. Seasonality and indoor/outdoor relationships of flame retardants and
PCBs in residential air. Environmental Pollution 218, 392–401.
https://doi.org/10.1016/j.envpol.2016.07.018
Melymuk, L., Bohlin-Nizzetto, P., Prokeš, R., Kukučka, P., Přibylová, P., Vojta, Š., Kohoutek,
J., Lammel, G., Klánová, J., 2017. Uncertainties in monitoring of SVOCs in air caused by
within-sampler degradation during active and passive air sampling. Atmospheric
Environment 167, 553–565. https://doi.org/10.1016/j.atmosenv.2017.08.038
Melymuk, L., Bohlin-Nizzetto, P., Sáňka, O., Pozo, K., Klánová, J., 2014. Current challenges
in air sampling of semivolatile organic contaminants: Sampling artifacts and their influence
on data comparability. Environmental science & technology 48, 14077–14091.
https://doi.org/10.1021/es502164r
Melymuk, L., Demirtepe, H., Jílková, S.R., 2020. Indoor dust and associated chemical
exposures. Current Opinion in Environmental Science & Health 15, 1–6.
https://doi.org/10.1016/j.coesh.2020.01.005
Melymuk, L., Diamond, M.L., Riddell, N., Wan, Y., Vojta, Š., Chittim, B., 2018. Challenges in
the analysis of novel flame retardants in indoor dust: results of the INTERFLAB 2
interlaboratory evaluation. Environmental Science & Technology 52, 9295–9303.
https://doi.org/10.1021/acs.est.8b02715
Melymuk, L., Nizzetto, P.B., Harner, T., White, K.B., Wang, X., Tominaga, M.Y., He, J., Li, J.,
Ma, J., Ma, W.L., Aristizábal, B.H., Dryer, A., Jiménez, B., Muñoz-Arnanz, J., Odabasi, M.,
Dumanoglu, Y., Yaman, B., Graf, C., Sweetman, A., Klánová, J., 2021. Global intercomparison
of polyurethane foam passive air samplers evaluating sources of variability in SVOC
measurements. Environmental Science and Policy 125, 1–9.
https://doi.org/10.1016/j.envsci.2021.08.003
Melymuk, L., Robson, M., Helm, P.A., Diamond, M.L., 2013. Application of Land Use
Regression to Identify Sources and Assess Spatial Variation in Urban SVOC Concentrations.
Environmental Science & Technology 47, 1887–1895. https://doi.org/10.1021/es3043609
Melymuk, L., Robson, M., Helm, P.A., Diamond, M.L., 2012. PCBs, PBDEs, and PAHs in
Toronto air: spatial and seasonal trends and implications for contaminant transport.
35
Science of the Total Environment 429, 272–80.
https://doi.org/10.1016/j.scitotenv.2012.04.022
Melymuk, L., Robson, M., Helm, P.A., Diamond, M.L., 2011. Evaluation of passive air sampler
calibrations: Selection of sampling rates and implications for the measurement of
persistent organic pollutants in air. Atmospheric Environment 45, 1867–1875.
https://doi.org/10.1016/j.atmosenv.2011.01.011
Mendelsohn, E., Hagopian, A., Hoffman, K., Butt, C.M., Lorenzo, A., Congleton, J., Webster,
T.F., Stapleton, H.M., 2016. Nail polish as a source of exposure to triphenyl phosphate.
Environment International 86, 45–51. https://doi.org/10.1016/j.envint.2015.10.005
Mikeš, O., Sáňka, O., Rafajová, A., Vlaanderen, J., Chen, J., Hoek, G., Klánová, J., Čupr, P.,
2023. Development of historic monthly land use regression models of SO2, NOx and
suspended particulate matter for birth cohort ELSPAC. Atmospheric Environment 301,
119688. https://doi.org/10.1016/j.atmosenv.2023.119688
Moya, J., Phillips, L., 2014. A review of soil and dust ingestion studies for children. Journal
of exposure science & environmental epidemiology 24, 545–554.
https://doi.org/10.1038/jes.2014.17
Muenhor, D., Harrad, S.J., 2012. Within-room and within-building temporal and spatial
variations in concentrations of polybrominated diphenyl ethers (PBDEs) in indoor dust.
Environment international 47, 23–7. https://doi.org/10.1016/j.envint.2012.06.001
Nakata, H., Hinosaka, M., Yanagimoto, H., 2015. Macrocyclic-, polycyclic-, and nitro musks
in cosmetics, household commodities and indoor dusts collected from Japan: Implications
for their human exposure. Ecotoxicology and Environmental Safety 111, 248–255.
https://doi.org/10.1016/j.ecoenv.2014.09.032
Nassan, F.L., Coull, B.A., Gaskins, A.J., Williams, M.A., Skakkebaek, N.E., Ford, J.B., Ye, X.,
Calafat, A.M., Braun, J.M., Hauser, R., 2017. Personal care product use in men and urinary
concentrations of select phthalate metabolites and parabens: Results from the
environment and reproductive health (EARTH) study. Environmental Health Perspectives
125, 1–10. https://doi.org/10.1289/EHP1374
Needleman, H., 2009. Low Level Lead Exposure: History and Discovery. Annals of
Epidemiology 19, 235–238. https://doi.org/10.1016/j.annepidem.2009.01.022
Nguyen, L.V., Diamond, M.L., Venier, M., Stubbings, W.A., Romanak, K., Bajard, L., Melymuk,
L., Jantunen, L.M., Arrandale, V.H., 2019. Exposure of Canadian electronic waste dismantlers
to flame retardants. Environment International 129, 95–104.
https://doi.org/10.1016/j.envint.2019.04.056
36
Özkaynak, H., Cohen, J., Hubbard, H., Thomas, K., Phillips, L., Tulve, N., 2023. Advancing
Methodologies Used in Trace Element-Based Mass Balance Studies to Separately Estimate
Soil and Dust Ingestion Rates for Children. Environment International 178, 107983.
https://doi.org/10.1016/j.envint.2023.107983
Persson, L., Carney Almroth, B.M., Collins, C.D., Cornell, S., de Wit, C.A., Diamond, M.L.,
Fantke, P., Hassellöv, M., MacLeod, M., Ryberg, M.W., Søgaard Jørgensen, P., VillarrubiaGómez,
P., Wang, Z., Hauschild, M.Z., 2022. Outside the Safe Operating Space of the
Planetary Boundary for Novel Entities. Environmental Science & Technology 56, 1510–
1521. https://doi.org/10.1021/acs.est.1c04158
Pozo, K., Harner, T., Wania, F., Muir, D.C.G., Jones, K.C., Barrie, L.A., 2006. Toward a global
network for persistent organic pollutants in air: results from the GAPS study. Environmental
science & technology 40, 4867–73. https://doi.org/10.1021/es060447t
Rigét, F., Bignert, A., Braune, B., Dam, M., Dietz, R., Evans, M., Green, N., Gunnlaugsdóttir,
H., Hoydal, K.S., Kucklick, J., Letcher, R., Muir, D., Schuur, S., Sonne, C., Stern, G., Tomy, G.,
Vorkamp, K., Wilson, S., 2019. Temporal trends of persistent organic pollutants in Arctic
marine and freshwater biota. Science of The Total Environment 649, 99–110.
https://doi.org/10.1016/j.scitotenv.2018.08.268
Rostkowski, P., Haglund, P., Aalizadeh, R., Alygizakis, N., Thomaidis, N., Arandes, J.B.,
Nizzetto, P.B., Booij, P., Budzinski, H., Brunswick, P., Covaci, A., Gallampois, C., Grosse, S.,
Hindle, R., Ipolyi, I., Jobst, K., Kaserzon, S.L., Leonards, P., Lestremau, F., Letzel, T., Magnér,
J., Matsukami, H., Moschet, C., Oswald, P., Plassmann, M., Slobodnik, J., Yang, C., 2019. The
strength in numbers: comprehensive characterization of house dust using complementary
mass spectrometric techniques. Analytical and Bioanalytical Chemistry 411, 1957–1977.
https://doi.org/10.1007/s00216-019-01615-6
Saini, A., Harner, T., Chinnadhurai, S., Schuster, J.K., Yates, A., Sweetman, A., AristizabalZuluaga,
B.H., Jiménez, B., Manzano, C.A., Gaga, E.O., Stevenson, G., Falandysz, J., Ma, J.,
Miglioranza, K.S.B., Kannan, K., Tominaga, M., Jariyasopit, N., Rojas, N.Y., Amador-Muñoz,
O., Sinha, R., Alani, R., Suresh, R., Nishino, T., Shoeib, T., 2020. GAPS-megacities: A new
global platform for investigating persistent organic pollutants and chemicals of emerging
concern in urban air. Environmental Pollution 267, 115416.
https://doi.org/10.1016/j.envpol.2020.115416
Salthammer, T., Zhang, Y., Mo, J., Koch, H.M., Weschler, C.J., 2018. Assessing Human
Exposure to Organic Pollutants in the Indoor Environment. Angewandte Chemie International
Edition 57. https://doi.org/10.1002/anie.201711023
37
Schuster, J.K., Gioia, R., Harner, T., Lee, S.C., Breivik, K., Jones, K.C., 2012. Assessment of
sorbent impregnated PUF disks (SIPs) for long-term sampling of legacy POPs. Journal of
environmental monitoring 14, 71–8. https://doi.org/10.1039/C1EM10697J
Schweizer, C., Edwards, R.D., Bayer-Oglesby, L., Gauderman, W.J., Ilacqua, V., Jantunen, M.J.,
Lai, H.K., Nieuwenhuijsen, M.J., Künzli, N., 2007. Indoor time-microenvironment-activity
patterns in seven regions of Europe. Journal of exposure science & environmental
epidemiology 17, 170–81.
Shoeib, M., Harner, T., 2002. Characterization and comparison of three passive air samplers
for persistent organic pollutants. Environmental science & technology 36, 4142–51.
https://doi.org/10.1021/es020635t
Stapleton, H.M., Sharma, S., Getzinger, G., Ferguson, P.L., Gabriel, M., Webster, T.F., Blum,
A., 2012. Novel and high volume use flame retardants in US couches reflective of the 2005
PentaBDE phase out. Environmental Science & Technology 46, 13432–13439.
https://doi.org/10.1021/es303471d
Steffen, W., Richardson, K., Rockström, J., Cornell, S., Fetzer, I., Bennett, E., Biggs, R.,
Carpenter, S.R., de Wit, C.A., Folke, C., Mace, G., Persson, L.M., Veerabhadran, R., Reyers, B.,
Sörlin, S., 2015. Planetary Boundaries: Guiding human development on a changing planet.
Science 347. https://doi.org/10.1126/science.1259855
Stuchlík Fišerová, P., Melymuk, L., Komprdová, K., Domínguez-Romero, E., Scheringer, M.,
Kohoutek, J., Přibylová, P., Andrýsková, L., Piler, P., Koch, H.M., Zvonař, M., Esteban-López,
M., Castaño, A., Klánová, J., 2022. Personal care product use and lifestyle affect phthalate
and DINCH metabolite levels in teenagers and young adults. Environmental Research 213,
113675. https://doi.org/10.1016/j.envres.2022.113675
Sugeng, E.J., de Cock, M., Leonards, P.E.G., van de Bor, M., 2018. Electronics, interior
decoration and cleaning patterns affect flame retardant levels in the dust from Dutch
residences. Science of the Total Environment 645, 1144–1152.
https://doi.org/10.1016/j.scitotenv.2018.07.127
UNEP, 2017. Stockholm Convention on Persistent Organic Pollutants (POPs).
https://www.pops.int/
van der Schyff, V., Kalina, J., Abballe, A., Iamiceli, A.L., Govarts, E., Melymuk, L., 2023a. Has
Regulatory Action Reduced Human Exposure to Flame Retardants? Environ. Sci. Technol.
57, 19106–19124. https://doi.org/10.1021/acs.est.3c02896
van der Schyff, V., Kalina, J., Govarts, E., Gilles, L., Schoeters, G., Castaño, A., Esteban-López,
M., Kohoutek, J., Kukučka, P., Covaci, A., Koppen, G., Andrýsková, L., Piler, P., Klánová, J.,
Jensen, T.K., Rambaud, L., Riou, M., Lamoree, M., Kolossa-Gehring, M., Vogel, N., Weber, T.,
38
Göen, T., Gabriel, C., Sarigiannis, D.A., Sakhi, A.K., Haug, L.S., Murinova, L.P., Fabelova, L.,
Tratnik, J.S., Mazej, D., Melymuk, L., 2023b. Exposure to flame retardants in European
children — Results from the HBM4EU aligned studies. International Journal of Hygiene and
Environmental Health 247, 114070. https://doi.org/10.1016/j.ijheh.2022.114070
van der Schyff, V., Suchánková, L., Kademoglou, K., Melymuk, L., Klánová, J., 2022. Parabens
and antimicrobial compounds in conventional and “green” personal care products.
Chemosphere 297, 1–8. https://doi.org/10.1016/j.chemosphere.2022.134019
van der Veen, I., de Boer, J., 2012. Phosphorus flame retardants: properties, production,
environmental occurrence, toxicity and analysis. Chemosphere 88, 1119–53.
https://doi.org/10.1016/j.chemosphere.2012.03.067
Venier, M., Audy, O., Vojta, Š., Bečanová, J., Romanak, K., Melymuk, L., Krátká, M., Kukučka,
P., Okeme, J.O., Saini, A., Diamond, M.L., Klánová, J., 2016. Brominated flame retardants in
indoor environment - comparative study of indoor contamination from three countries.
Environment international 94, 150–160. https://doi.org/10.1016/j.envint.2016.04.029
Venier, M., Hites, R.A., 2010. Time trend analysis of atmospheric POPs concentrations in the
Great Lakes region since 1990. Environmental science & technology 44, 8050–5.
https://doi.org/10.1021/es101656u
Vojta, Š., Bečanová, J., Melymuk, L., Komprdová, K., Kohoutek, J., Kukučka, P., Klánová, J.,
2017a. Screening for halogenated flame retardants in European consumer products,
building materials and wastes. Chemosphere 168, 457–466.
https://doi.org/10.1016/j.chemosphere.2016.11.032
Vojta, Š., Melymuk, L., Klánová, J., 2017b. Changes in Flame Retardant and Legacy
Contaminant Concentrations in Indoor Air during Building Construction, Furnishing, and
Use. Environmental Science & Technology 51, 11891–11899.
https://doi.org/10.1021/acs.est.7b03245
Vojta, Š., Melymuk, L., Prokeš, R., Klánová, J., 2024. Assessment of the impact of sampler
housing on indoor passive air sampler measurements of SVOCs. Journal of Environmental
Exposure Assessment 3, 6. https://doi.org/10.20517/jeea.2023.39
Vorkamp, K., Castaño, A., Antignac, J.-P., Boada, L.D., Cequier, E., Covaci, A., Esteban López,
M., Haug, L.S., Kasper-Sonnenberg, M., Koch, H.M., Pérez Luzardo, O., Osīte, A., Rambaud,
L., Pinorini, M.-T., Sabbioni, G., Thomsen, C., 2021. Biomarkers, matrices and analytical
methods targeting human exposure to chemicals selected for a European human
biomonitoring initiative. Environment International 146, 106082.
https://doi.org/10.1016/j.envint.2020.106082
39
Vykoukalová, M., Venier, M., Vojta, Š., Melymuk, L., Bečanová, J., Romanak, K., Prokeš, R.,
Okeme, J.O., Saini, A., Diamond, M.L., Klánová, J., 2017. Organophosphate esters flame
retardants in the indoor environment. Environment International 106, 97–104.
https://doi.org/10.1016/j.envint.2017.05.020
Wania, F., Shunthirasingham, C., 2020. Passive air sampling for semi-volatile organic
chemicals. Environmental Science: Processes & Impacts 22, 1925–2002.
https://doi.org/10.1039/D0EM00194E
Weschler, C.J., Nazaroff, W.W., 2012. SVOC exposure indoors: fresh look at dermal
pathways. Indoor air 22, 356–377.
Weschler, C.J., Nazaroff, W.W., 2010. SVOC partitioning between the gas phase and settled
dust indoors. Atmospheric Environment 44, 3609–3620.
Weschler, C.J., Nazaroff, W.W., 2008. Semivolatile organic compounds in indoor
environments. Atmospheric Environment 42, 9018–9040.
White, K.B., Kalina, J., Scheringer, M., Přibylová, P., Kukučka, P., Kohoutek, J., Prokeš, R.,
Klánová, J., 2023. Spatial and Temporal Trends of Persistent Organic Pollutants across
Europe after 15 Years of MONET Passive Air Sampling. Environ. Sci. Technol. 57, 11583–
11594. https://doi.org/10.1021/acs.est.3c00796
White, K.B., Sáňka, O., Melymuk, L., Přibylová, P., Klánová, J., 2021. Application of land use
regression modelling to describe atmospheric levels of semivolatile organic compounds
on a national scale. Science of the Total Environment 793.
https://doi.org/10.1016/j.scitotenv.2021.148520
Wingfors, H., Seldén, A.I., Nilsson, C., Haglund, P., 2006. Identification of markers for PCB
exposure in plasma from Swedish construction workers removing old elastic sealants. The
Annals of occupational hygiene 50, 65–73. https://doi.org/10.1093/annhyg/mei063
Wong, F., Hung, H., Dryfhout-Clark, H., Aas, W., Bohlin-Nizzetto, P., Breivik, K.,
Mastromonaco, M.N., Lundén, E.B., Ólafsdóttir, K., Sigurðsson, Á., Vorkamp, K., Bossi, R.,
Skov, H., Hakola, H., Barresi, E., Sverko, E., Fellin, P., Li, H., Vlasenko, A., Zapevalov, M.,
Samsonov, D., Wilson, S., 2021. Time trends of persistent organic pollutants (POPs) and
Chemicals of Emerging Arctic Concern (CEAC) in Arctic air from 25 years of monitoring.
Science of the Total Environment 775. https://doi.org/10.1016/j.scitotenv.2021.145109
Yang, C., Jílková, S.R., Melymuk, L., Harris, S.A., Jantunen, L.M., Pertili, J., Winn, L., Diamond,
M.L., 2020. Are We Exposed to Halogenated Flame Retardants from both Primary and
Secondary Sources? Environmental Science & Technology Letters 7, 585–593.
https://doi.org/10.1021/acs.estlett.0c00268
van der Schyff, Veronica, Jiří Kalina, Annalisa Abballe, Anna Laura Iamiceli, Eva Govarts,
and Lisa Melymuk. 2023. "Has Regulatory Action Reduced Human Exposure to Flame
Retardants?" Environmental Science & Technology 57 (48): 19106-24.
https://doi.org/10.1021/acs.est.3c02896
Has Regulatory Action Reduced Human Exposure to Flame
Retardants?
Veronica van der Schyff, Jiří Kalina, Annalisa Abballe, Anna Laura Iamiceli, Eva Govarts,
and Lisa Melymuk*
Cite This: Environ. Sci. Technol. 2023, 57, 19106−19124 Read Online
ACCESS Metrics & More Article Recommendations *sı Supporting Information
ABSTRACT: Flame retardant (FR) exposure has been linked to
several environmental and human health effects. Because of this,
the production and use of several FRs are regulated globally. We
reviewed the available records of polybrominated diphenyl ethers
(PBDEs) and hexabromocyclododecanes (HBCDDs) in human
breast milk from literature to evaluate the efficacy of regulation to
reduce the exposure of FRs to humans. Two-hundred and seven
studies were used for analyses to determine the spatial and
temporal trends of FR exposure. North America consistently had
the highest concentrations of PBDEs, while Asia and Oceania
dominated HBCDD exposure. BDE-49 and -99 indicated
decreasing temporal trends in most regions. BDE-153, with a
longer half-life than the aforementioned isomers, typically exhibited
a plateau in breast milk levels. No conclusive trend could be established for HBCDD, and insufficient information was available to
determine a temporal trend for BDE-209. Breakpoint analyses indicated a significant decrease in BDE-47 and -99 in Europe around
the time that regulation has been implemented, suggesting a positive effect of regulation on FR exposure. However, very few studies
have been conducted globally (specifically in North America) after 2013, during the time when the most recent regulations have
been implemented. This meta-analysis provides insight into global trends in human exposure to PBDEs and HBCDD, but the
remaining uncertainty highlights the need for ongoing evaluation and monitoring, even after a compound group is regulated.
KEYWORDS: flame retardant, polybrominated diphenyl ether, hexabromocyclododecane, breast milk, biomonitoring, temporal trends,
effectiveness evaluation
■ INTRODUCTION
Flame retardants are added to a wide range of consumer
products and materials in order to reduce ignition or
flammability of a material or fulfill fire safety requirements.1
However, past efforts to reduce flammability through the
addition of synthetic organic flame retardants have led to
negative impacts on human and environmental health due to
exposure to harmful chemicals.2
Polybrominated diphenyl ethers (PBDEs) and hexabromocyclododecanes
(HBCDDs) were among the dominant FRs
used for decades.3,4
The three technical mixtures of PBDEs
(penta-, octa-, and deca-BDE) had multiple uses, including
polyurethane foam, electrical and electronic equipment,
building materials, and vehicle parts.5
Technical HBCDD (a
mixture of the stereoisomers, α-, β-, and γ-HBCDD, with γHBCDD
being the most abundant) was primarily used in
electronics, textiles, and especially in expanded (EPS) and
extruded polystyrene (XPS) applied as construction and
packing materials.6,7
PBDEs and HBCDDs are known to be persistent,
bioaccumulative, and subject to long-range transport in the
environment and are ubiquitous across environmental
systems.8,9
PBDEs have been reported in human blood,10−12
adipose tissues,13−15
and milk16−19
since the early 1990s, and
evidence of human exposure to HBCDDs arose shortly
thereafter.20−23
PBDE exposure has been associated with numerous adverse
health outcomes, including alterations to thyroid function,
reproductive systems, and breast cancer,24−26
with strong
evidence for neurodevelopmental impacts, including lower IQ
and ADHD.27,28
Elevated levels of PBDEs in breast milk have
specifically been associated with neurodevelopmental effects
and alterations to the gut microbiome in young children.29,30
Similar adverse effects are associated with elevated HBCDD
Received: May 4, 2023
Revised: August 10, 2023
Accepted: September 29, 2023
Published: November 22, 2023
Policy Analysispubs.acs.org/est
© 2023 The Authors. Published by
American Chemical Society
19106
https://doi.org/10.1021/acs.est.3c02896
Environ. Sci. Technol. 2023, 57, 19106−19124
This article is licensed under CC-BY 4.0
DownloadedviaMASARYKUNIVonOctober8,2024at10:00:18(UTC).
Seehttps://pubs.acs.org/sharingguidelinesforoptionsonhowtolegitimatelysharepublishedarticles.
concentrations, including endocrine disruption, specifically
thyroid, neurobehavioral, and developmental disorders.26,31
In response to concerns regarding the environmental and
human health impacts of certain FRs, actions were taken to
reduce production.32
In 2004, the European Union stated that
“in order to protect health and the environment the placing on
the market and the use of pentaBDE and octaBDE and the
placing on the market of articles containing one or both of
these substances should be prohibited”,33
and in the same year
these mixtures were voluntarily withdrawn from the U.S.
marketplace by their manufacturers.34
The lower brominated
PBDE congeners, tetra- and penta-BDE (main components of
commercial penta-BDE435
) and hexa- and hepta-BDEs (main
components of commercial octa-BDE4
)36
were listed in the
Stockholm Convention on Persistent Organic Pollutants
(POPs) in 2009, requiring parties to eliminate the production
and use of the compounds. Deca-BDE, the fully brominated
PBDE molecule and main component of the decaBDE
commercial product,37
was similarly listed in 2017.38
In
2008, HBCDDs were recognized as substances of very high
concern (SVHC) in the EU due to environmental and human
health risks39,40
and were added to the Stockholm Convention
in 2013.41
In specific cases, individual countries were permitted
continued production of HBCDDs until 2024.7
The evaluation of temporal patterns of a chemical’s
concentration in a predetermined medium is an effective tool
to determine the efficiency of policy in mitigating chemical
exposure.42
Recent studies have identified declines in
components of the penta- and octa-BDE technical mixtures,9
in air (1993−2018),43
soil (1998−2008),44
sediment (2002−
2012),45
sewage sludge (2004−2010),46
and fish (1980−
2009),47
while BDE-209 has been stable or increasing in many
matrices.46,48,49
In humans, declines in levels of less
brominated PBDEs and a more recent plateau have been
identified in several countries, however, this is not uniform
across regions or matrices:9,22,50,51
there is a lack of
understanding of how generalizable these regional trends are.
For HBCDDs, there is even less evidence of a global trend.
Although time trends of HBCDD concentrations in multiple
environmental matrices52−54
have been determined, analysis of
temporal patterns of HBCDDs has been limited to regional
scales. No consensus or clear global time trend of HBCDD
concentrations in humans has been identified,9
with some
studies reporting an increase of HBCDDs,54−56
others a
decrease57
or no trend.58,59
Table 1. Countries Used in This Study Grouped According to the United Nations Geoscheme
region countries included PBDEs HBCDDs
Africa
Congo South Africa South Africa
7 studies79−85
5 studies79−83
Cote
d’Ivoire
Mauritius Tanzania
Djibouti Morocco Togo
Egypt Niger Tunisia
Ethiopia Nigeria Uganda
Ghana Senegal Zambia
Kenya
Asia
China Japan Syria
53 studies18,81,86−136
15 studies81,89,95,132−135,137−144
Georgia Macao Taiwan
India Philippines Tajikistan
Indonesia Russiaa
Vietnam
Israel South Korea
Central and South
America and
Caribbean
Antigua and
Barbuda
Chile Peru
3 studies73,81,145
2 studies73,81
Barbados Haiti Suriname
Brazil Jamaica Uruguay
Europe
Belgium Greece Romania
61 studies1,19,29,30,59,81,146−200
25 studies23,59,76,81,146−151,153−159,167,185,189,193,201−204
Bulgaria Hungary Russiaa
Croatia Ireland Slovakia
Cyprus Italy Spain
Czechia Lithuania Sweden
Denmark Luxembourg Switzerland
Faroe
Islands
Moldova Turkey
Finland Netherlands UK
France Norway Ukraine
Germany Poland
North America Canada Mexico USA 25 studies17,51,70,77,81,205−222
5 studies51,77,78,81,83
Oceania
Australia Kiribati Tonga
7 studies223−228
2 studies81,226
Fiji New Zealand Tuvalu
a
General samples from Russia were included within the European category. When a study specified a geographic region that was within the Asian
part of Russia, it was included in Asia.
Environmental Science & Technology pubs.acs.org/est Policy Analysis
https://doi.org/10.1021/acs.est.3c02896
Environ. Sci. Technol. 2023, 57, 19106−19124
19107
Studying the effectiveness of policies concerning FRs
through time trend analysis of biomonitoring data presents
inherent complexities. Different FRs, each with distinct
chemical structures, are incorporated into diverse applications
such as furniture, electronics, and building insulation, which
are associated with different emission and exposure routes.4
After restriction, continued presence of FR-containing
products further complicates this, as different product types
have very different replacement rates (e.g., smartphones, 2−6
years60
vs building insulation, 30−50 yrs61
). The environmental
persistence of compounds is generally longer in-
doors62,63
and indoor levels are sustained until active removal
of sources.64,65
Exposure to FRs is further influenced by
regional factors like building and cleaning practices and dietary
patterns.66
Moreover, the differing persistence of FRs within
the body complicates our understanding of exposure,67
with
some FRs possessing longer half-lives in human tissues68
and
partitioning within body tissues varying by compound/
congener.69,70
Concentrations of POPs in human tissues
typically reflect long-term exposures, for example, variations
of PCB concentrations in breast milk levels can be explained by
differences in early life exposures rather than current dietary
exposures.71
Despite the above-mentioned complexities, our
insight into the effectiveness of restrictions and the remaining
risks to human populations from legacy FRs can be improved
by multistudy analyses to understand and interpret the global
time patterns of PBDEs and HBCDDs in humans.
In this analysis, we first review available records of PBDE
and HBCDD in human milk to determine time patterns of
global human exposure to FRs. Second, we evaluate the impact
of regulations that were introduced over the past 20 years on
exposure to these legacy FRs. Finally, we investigate the
regional differences in exposure to FRs in relation to use. We
supplement this with a review of past studies evaluating
temporal patterns of PBDEs and HBCDDs in human matrices,
to provide a comprehensive review of trends in global
exposure.
■ METHODS
Rationalization of Study Matrix. It is impossible to select
a single biological matrix that encompasses the global
population, as well as all target compounds. Due to the
lipophilic nature of PBDEs and HBCDDs, they are best
evaluated through matrices such as blood serum or breast/
maternal milk.72
Breast milk has a high lipid content and can
be collected noninvasively, making it a reliable and accessible
matrix for assessing body burdens of PBDEs and HBCDDs, as
well as many other POPs. The Stockholm Convention, in
cooperation with the World Health Organization, has
identified human milk as a core matrix of its Global Monitoring
Plan and supported with routine quantification in pooled milk
samples.70,73
In addition to its importance in routine monitoring
programs, maternal milk is an ideal matrix for meta-analyses
of biomonitoring data because of the relative homogeneity of
the study population: all female, with an age range generally
spanning 18−45 years. Moreover, breast milk is not only an
indicator of human exposure but also represents a direct
exposure route to infants, and breast milk is typically the most
important exposure pathway of young children to POPs,
including PBDEs.74,75
Search Strategy and Selection Criteria−Human Milk
Meta-Analysis. A literature search of peer-reviewed studies
and reports produced by regulatory bodies (e.g., UNEP,
German Federal Environment Agency) was conducted using
ISI Web of Science and Google Scholar. The search was not
limited by years or language of publication. The search was
initially conducted in March 2020 and updated in September
2022. The following search terms were used.
For HBCDDs. All fields: [hexabromocyclododecane* OR
HBCD*] AND [(human milk) or (breast milk)]. This search
produced 168 results, which were evaluated for their
appropriateness. The criteria for inclusion were that the
studies provided lipid-standardized HBCDD levels in human
milk (either isomer-specific HBCDD or ∑HBCDD) and
included basic information on the study population (country of
residence, sampling year). Of the initial 168 studies identified,
49 met the criteria and were used for further analysis (Figure
S1, data sources listed in Table 1).
For PBDEs. All fields: [polybrominated diphenyl ether* OR
PBDE*] AND [(human milk) or (breast milk)]. This search
identified 1204 results, which were then evaluated for their
appropriateness. Only studies reporting individual PBDE
congeners were included, and four congeners were selected
as indicators due to their prevalence in literature: BDE-47,
BDE-99, BDE-153, and BDE-209. Studies also had to include
lipid-standardized concentrations for at least one of these
congeners and include information on the study population
(country of residence, sampling year). Of the initial 1204
studies, 158 met the criteria and were used for further analysis
(Figure S1, data sources listed in Table 1).
All available data (either primary data reporting individual
concentrations or all summary statistics) were extracted from
the articles to a spreadsheet database. Data reported from
pooled samples were treated as mean values.
Data Set Standardization. Statistical evaluation was
carried out by R software (version R 4.1.2). The data sources
were separated by compound, sampling location, and/or date
of sample collection. These records were aggregated by
country and date of sampling, characterized either by mean
or median value, minimum, 5th, 10th, 25th, 75th, 90th, and
95th quantile and maximum or any combination of these
descriptive statistics. If all primary data/summary statistics
within an aggregate record were taken during one year, the
aggregated record was assigned to that year. In the other cases,
the aggregated record was assigned to the middle point
between the years of the oldest and the newest primary data/
summary statistic.
For HBCDDs, 41 aggregated records included both αHBCDD
and the sum of α, β, and γ-isomers. These 41 records
were used for estimating the contribution of α-HBCDD to the
sums. This contribution was 94.0%, showing clearly that αHBCDD
dominates over the β and γ isomers. This 94.0% was
then used to extrapolate α-HBCDD from records where only
∑HBCDDs were reported for the original data set, resulting in
260 aggregated records for α-HBCDD.
Primary FR data from five locations76−78
were used to
establish that α-HBCDD and PBDE concentrations in breast
milk have a general log-normal distribution of primary data
within an aggregate data record. On the basis of the log-normal
distribution assumption, the maximum likelihood estimation
was used to apply a log-normal distribution for each aggregated
sample and thus estimate its median value in cases where only
other summary statistics were reported. In a few cases, the
maximum likelihood estimate was not directly applicable since
the only descriptive statistic characterizing the aggregated
Environmental Science & Technology pubs.acs.org/est Policy Analysis
https://doi.org/10.1021/acs.est.3c02896
Environ. Sci. Technol. 2023, 57, 19106−19124
19108
sample was an arithmetic mean. In such cases, a median
standard deviation based on the rest of the samples was used
(0.62 ng/g lipid weight; lw for α-HBCDD, 0.32 ng/g lw for
BDE-47, 0.23 ng/g lw for BDE-99, 0.35 ng/g lw for BDE 153,
and 50.15 ng/g lw for BDE-209) to derive quantiles describing
the expected log-normal distribution. Finally, since there are no
theoretical minimal and maximal values for the log-normal
statistical distribution, the min and max values were considered
as (1/n)th and (1−1/n)th quantiles for aggregated samples
with known n; for aggregated samples with an unknown
number of primary samples, n = 40 was used as it was the
median value of studies where n was specified (Figure S2).
Temporal Pattern Meta-Analysis. Data were grouped
regionally following the United Nations geoscheme (Table 1).
With the use of the aggregate data by region and sampling
year, a weighted Theil-Sen trend analysis was conducted,
assigning each aggregated sample a weight in the range of 0.1
to 1.0 for 10 to 100 primary samples and 1.0 for more than 100
primary samples. A weighted Mann-Kendall test was then used
for assessing the trends’ significance.
Breakpoint Analysis. An additional type of temporal
pattern analysis was applied to European and Asian data sets,
as these continents had the most complete records for both
PBDEs and HBCDDs and are of particular interest given the
early introduction of FR regulations. To evaluate whether this
early introduction of restrictions led to a change in the FR time
patterns over time, breakpoint analysis was applied to find a
time point when the slope of the trend breaks,229
i.e.,
suggesting a change in the rate of change of a given FR
concentration in human milk. The breakpoint analysis
identifies the breakpoint by aiming for normally distributed
residuals of both linear trends before and after the breakpoint,
which indicates an optimal fit. The method searches for all
possible breakpoints (in this case in increments of whole years
only) and selects the breakpoint with the smallest sum of
squares of residuals. Only significant results according to the
difference of halving times before and after the breakpoint are
considered.229
Analysis of Geographic Patterns. Additional comparative
statistical analyses were conducted using Graphpad Prism
8.0.2 using all studies after the year 2000. Data were grouped
according to geographic region (Table 1). The concentrations
from the different regions were compared using Kruskall-Wallis
nonparametric ANOVA tests, and individual regions were
compared with all others using Dunn’s multiple comparison
tests. For geographic patterns, significance was set at p < 0.05.
Limitations. The quality of the analytical work performed
in individual studies was not evaluated, in favor of allowing for
a greater breadth of data to be incorporated in the metaanalysis.
All studies were published in peer-reviewed journals
or as reports available from reputable national/international
organizations, leading to the assumption of an acceptable level
of data quality. Some reports (notably the UNEP/WHO data
included in the Stockholm Convention GMP reports) do not
include analytical information, although data are produced by
recognized, accredited laboratories, and exclusion of this data
would lead to a substantial loss in geographic coverage.
Additionally, older studies reflecting very early analyses of
PBDEs and HBCDDs may have more generous allowances in
terms of QA/QC; however, it was important for the temporal
analyses that these could be included. However, a consequence
of this is that not all studies will meet the most stringent QA/
QC standards.
While breastfeeding mothers present a relatively homogeneous
population with respect to age and sex, some additional
factors can impact breast milk concentrations of FRs, and these
could not be incorporated into our meta-analysis, primarily
because of inconsistent data across studies.
Representativeness. Data from breastmilk only reflects the
chemical burden in the breast-feeding female population. The
prevalence of breastfeeding mothers also varies by socioeconomic
status and cultural background.230
Parity. Some studies have shown that parity is related to
differences in levels of persistent compounds in breast milk.
Primiparous mothers have been shown to have higher
HBCDD concentrations than multiparous mothers,132
while
studies focused on PBDEs have not found a relation-
ship.84,187,207,231
As information on parity was not consistently
recorded for all data sources, we did not include this
confounder in our analyses and used all available data,
regardless of parity.
Duration of Lactation and Sample Collection. The timing
of breast milk collection within the lactation period varied
widely, from 1 week to 10 months after birth, although the
most typical was 3−8 weeks after birth, following the guidance
of the WHO/UNEP breast milk surveys.232
Whether this
impacts levels of PBDEs and HBCDDs in milk is unclear.
Some studies have reported variability166,207
or significant
decreases233
in PBDE levels in breast milk up to a year
postpartum; however, Harrad and Abdallah151
reported no
change in HBCDDs in milk over 12 months of lactation.
Maternal Age. This is often identified as a determinant of
breast milk levels; however, this is directly related to the
understanding of the persistence of these FRs in the body and
temporal changes in exposures within a country.234
Older
maternal age has been associated with higher PBDE levels in
some studies in breast milk129,204
and sera,235
while others
have found no association187
or an inverse association of lower
levels in breast milk from older mothers.84,207
Similarly for
HBCDDs, Fujii et al. identified age-dependency of γ-HBCDD
in milk,132
but not other HBCDD isomers, while Drage et
al.235
found no age-dependency in HBCDDs in sera.
Selection of Breast Milk as Biomonitoring Matrix. While
breast milk has the advantage of being noninvasive and widely
monitored, BDE-209 preferentially partitions to serum lipids
rather than milk lipids,69
leading to proportionally lower levels
in breast milk compared with exposures. However, while
milk:serum partitioning can vary by congener/compound,69
the relative geographic patterns and temporal trends (1227
samples from 1973 to 2019) should be appropriately captured
by either matrix, as milk and serum concentrations are typically
well-correlated.215
The limited number of spatial and temporal
studies conducted on certain compounds, such as BDE-209
and HBCDD in human breast milk is also a limitation for this
study.
Uncertainties Regarding Partitioning and Half-Lives of
PBDEs in Human Breast Milk. Very few studies have been
conducted on the human biological distributions and half-lives
of PBDE and HBCDDs. The existing evidence suggests higher
persistence of BDE-153 in the human body68,236,237
and
decreased partitioning to milk for higher molecular weight
FRs.69
As a result, comparisons of concentrations across
congeners would not necessarily reflect exposure trends;
however, the bulk of our analysis relies on trends built
individually for each congener and thus should not be
impacted by the uncertainty in partitioning and half-lives.
Environmental Science & Technology pubs.acs.org/est Policy Analysis
https://doi.org/10.1021/acs.est.3c02896
Environ. Sci. Technol. 2023, 57, 19106−19124
19109
Temporal Patterns: Search Strategy and Selection
Criteria. We supplemented our meta-analysis of temporal
patterns of PBDE and HBCDD exposure with a comprehensive
review of published time trends. A literature search was
conducted using ISI Web of Science and Google Scholar,
initially in March 2020, updated in September 2022. Search
terms for the temporal trend studies were a combination of
chemical-related terms (polybrominated diphenyl ether*,
PBDE*, hexabromocyclododecane*, HBCD*), matrix-related
(human, blood, serum, plasma, milk), and trend-related (timetrend*
or temporal*).
For the temporal trend analysis of PBDEs, 268 data sources
were identified, and the data were further examined to identify
only studies that reported any of the four indicator PBDEs
(−47, −99, −153, or −209), reflected the general population
(not occupational exposure), reported basic biomonitoring
parameters (e.g., geographic region, matrix, year of sample
collection, number of samples), and had at least two time
points with harmonized analyses (e.g., by the same laboratory).
This resulted in 24 studies which were included in the
overview of temporal trends (Table S2). For the temporal
trend analysis of HBCDDs, the literature search identified 88
data sources, and the inclusion criteria (reported either α- or
Figure 1. Box-and-whiskers (horizontal lines are medians, 95% confidence intervals, minima, and maxima) of (A) BDE-47, (B) BDE-99, (C) BDE-
153, (D) BDE-209, and (E) α-HBCDD concentrations in different regions (CSA = Central and South America and the Caribbean), in breast milk
data collected after year 2000. Nonparametric ANOVA tests (Kruskal−Wallis with Dunn’s post-tests) were conducted to determine significant
differences.
Environmental Science & Technology pubs.acs.org/est Policy Analysis
https://doi.org/10.1021/acs.est.3c02896
Environ. Sci. Technol. 2023, 57, 19106−19124
19110
∑HBCDD, not occupational exposure, reported basic
biomonitoring parameters) led to the inclusion of 20 studies
(Table S3). The trends from these studies were extracted,
using the interpretation of the study authors to determine
whether a trend is classified as increasing or decreasing over
time, or whether no time trend is apparent.
■ RESULTS
Spatial Patterns. Available data were not equally
distributed across all geographic regions. Most studies were
conducted in Europe and Asia, while Central and South
America, North America, and Oceania had only limited data
(Table 1). Very few studies on human biomonitoring of
HBCDDs have been conducted in North America, which is
surprising, considering that the region is known to have had
stringent flame retardant regulations and historically high use
of BFRs.3
Oceania had the highest median α-HBCDD concentration
in breast milk (2.7 ng/g lipid), followed by Asia (1.5 ng/g
lipid) and Europe (0.9 ng/g lipid) (Figure 1). The elevated
concentrations of α-HBCDD in breast milk from Oceania were
unexpected but may reflect a common market with many
products from Asian manufacturers and Oceania implementing
HBCDD regulations years later than Europe.238
Breast milk
from Asia had significantly higher concentrations of αHBCDD
than milk from Africa, and Europe had significantly
higher concentrations than Central and South America (Figure
1).
The concentrations of BDE-47, -99, and -153 in breast milk
from North America were significantly higher than those from
Europe, Africa, Asia, and Central- and South America: BDE-
47, and -99 concentrations in North America were 39 and 65
times higher, respectively, than concentrations in Asia (Figure
1, Table S1). For the higher brominated compounds, North
America had 50 and 138 times higher concentrations than
Africa for BDE-153 and -209, respectively (Figure 1, Table
S1). Europe had significantly higher concentrations of BDE-99
than Africa, and Africa had significantly lower concentrations
of BDE-153 than all the other regions. For BDE-209, there
were fewer data records which limited the comparison. The
fact that BDE-209 is notoriously difficult to quantify due to
high molecular mass and chemical instability likely contributes
to the lack of data on levels in breast milk.239,240
For BDE-209,
concentrations in milk from Asia were substantially lower than
in North America, and no other regions had significant
differences (Figure 1).
Temporal Patterns. Africa, Oceania, and Central and
South America had limited or no data on BDE-209
concentrations in breast milk, and temporal patterns could
not be fully evaluated (Table 1; Figures S5, S7, and S8). The
most prominent differences in trends are seen between Europe
Figure 2. Weighted temporal trends of BDE-47, BDE-153, and α-HBCDD concentrations (ng/g lipid weight, lw) in breast milk from Europe and
North America from literature. Shaded area indicates 95% confidence interval.
Environmental Science & Technology pubs.acs.org/est Policy Analysis
https://doi.org/10.1021/acs.est.3c02896
Environ. Sci. Technol. 2023, 57, 19106−19124
19111
Figure 3. Results of breakpoint analysis for PBDEs in human milk for Asia and Europe for BDE-47, BDE-99, BDE-153, BDE-209, and α-HBCDD.
The shaded area indicates the 95th percent confidence interval. The dotted blue line indicates when regulation was implemented by the Stockholm
Convention (2009: BDE-47, -99, and -153; 2013: HBCDD; 2019: BDE-209). Please note that the Stockholm Convention date does not directly
indicate the introduction of restrictions in each country; these may be earlier due to national/regional initiatives, or later, as Stockholm Convention
parties enact regulations to implement the Convention. Thus, they are only indicative of the general timing of global restrictions.
Environmental Science & Technology pubs.acs.org/est Policy Analysis
https://doi.org/10.1021/acs.est.3c02896
Environ. Sci. Technol. 2023, 57, 19106−19124
19112
and North America for BDE-47, -153, and α-HBCDD. The
temporal trends of BDE-99 and -209 are depicted elsewhere
(Figures S3 and S4).
In Europe, BDE-47 and -99 decreased significantly (p =
0.0001; annual change of −9.3% and −10.1% respectively),
while BDE-153 and -209 had no change over time (Figures 2
and S3). α-HBCDD concentrations increased significantly in
the European population (p < 0.0001; 13.9% per annum)
(Figure 2).
No significant changes (p < 0.05) were observed for any
compound in North America (Figures 2 and S4), although we
note that North American data was generally very sparse,
limiting the ability to distinguish temporal trends. Notably,
North America was the region with the least available αHBCDD
data points in breastmilk. The western hemisphere
regions were the only regions where BDE-47 and -99 did not
show a decreasing trend (Figures 2 and S4). The temporal
patterns of all other geographical regions are presented in the
Figures S5−S8.
Africa had a decrease in BDE-47 (p = 0.022; −7.8% per
annum). Africa was the only continent where a significant
decrease of α-HBCDD was observed (p < 0.0001; −33.6% per
annum) (Figure S5). A similar pattern to Europe was seen in
Asia. BDE-47 decreased (p = 0.002; −9.7 per annum), while
BDE-153 and BDE-209 both increased, with BDE-209
increasing by 12.8% per annum (p = 0.05). α-HBCDD
concentrations increased at a rate of 7.9% per annum (p =
0.006) (Figure S6).
Figure 4. Time trends reported in human matrices in literature for (a) BDE-47, (b) BDE-99, (c) BDE-153, (d) BDE-209, and (e) HBCDD. Red
bars indicate an increasing trend, yellow bars indicate no trend or a plateau, and green bars indicate a decreasing trend. Trends are classified based
on the interpretations of the authors of each study. References for all studies can be found in Tables S2 and S3.
Environmental Science & Technology pubs.acs.org/est Policy Analysis
https://doi.org/10.1021/acs.est.3c02896
Environ. Sci. Technol. 2023, 57, 19106−19124
19113
In Central and South America and the Caribbean, BDE-153
increased significantly (p = 0.08) with an annual change of
14.8% (Figure S7). Oceania was the only region with a
consistent decrease in all PBDE congeners. BDE-47 and -99
decreased significantly with an annual change of −13.1% and
−15.5%, respectively. BDE-153 decreased at a rate of −6% per
annum (p = 0.054) (Figure S8).
Breakpoint Analysis. Only Asia and Europe had sufficient
data for breakpoint analyses (Figure 3). Breakpoints were
calculated for other regions to identify the timing of
concentration peaks (Figures S9−S12), but the trends are of
limited value due to scarce data and are not discussed in detail.
In Europe, an increase in α-HBCDD of 13.9% per annum
was observed from 1980 to 2010. In 2010, there was a change
(breakpoint) in the temporal trends of α-HBCDD (Figure 3).
The post-2010 decrease is not statistically significant due to
limited data collected since 2010; thus, we cannot determine if
recent concentrations are stable or declining. Although the
breakpoint is not statistically significant, it represents the best
fit for the available data and suggests a shift in exposure post-
2010. In Asia, the α-HBCDD increase was even sharper
(+31.6% per year; p = 0.007 Figure 3) and the breakpoint was
identified earlier (1998), suggesting Asian concentrations
reached a plateau at this point. Like Europe, the modeled
decrease since the breakpoint in Asia is not statistically
significant.
In both Asian and European breast milk data, the breakpoint
in concentrations for BDE-47 and 99 was substantially earlier
than for HBCDDs, close to 1990 for BDE-47 and between
1995 and 2000 for BDE-99. In all cases, concentrations
decreased after the breakpoint for both congeners, but this
postbreakpoint decrease is only significant for BDE-47 and
BDE-99 in Europe (p > 0.0001).
However, there is a clear contrast between the breakpoints
and before/after trends for BDE-153. The breakpoint for BDE-
153 reflects only a change in the rate of increase of BDE-153
(Asia) or plateau (Europe) with no evidence of declining
breast milk levels. Temporal patterns for BDE-209 in Asia were
not significant, indicating no clear time trends. The European
breakpoint for BDE-209 indicated a shift from the significant
increase before 2004 to a current plateau or declining phase.
Comparison with Other Reported Time Trends. For
the PBDEs, clear differences in the time trends by congener
and by study timing are seen. For BDE-47 and -99 (Figure 4),
there is a clear shift from early increasing time trends to more
recent plateaus or decreasing time trends. The same pattern of
early increasing trends and more recent reports of decreasing
time trends is visible for BDE-153 (Figure 4), but the first
decreases are not reported until much more recently. Most
time trends for BDE-153 indicate a plateau. Relatively few
studies report time trend analysis for HBCDD and BDE-209.
For BDE-209 and HBCDD, no discernible time trend can be
derived from literature for human matrices; trends were
variable and not generalizable by region or duration of the time
trend (Figure 4).
■ DISCUSSION AND IMPLICATIONS
Temporal Patterns. The analysis of published studies
suggests that BDE-47 and -99 have a global decreasing trend.
Decreasing temporal patterns were found in Europe, Asia, and
Oceania. BDE-47 and -99 have human elimination half-lives of
approximately 0.37−3 and 0.77−8 years, respectively.68,236,237
In the time since the penta- and octa-PBDEs were included in
the Stockholm Convention in 2004, the population would have
been exposed to lower concentrations of the compounds, and
the existing compounds would have been eliminated from their
bodies. The United States is a signatory to the Stockholm
Convention but has yet to ratify or implement the convention
in national legislation.241
The North American region was the
only region with a continuous increase in BDE-47 and -99
(Figure 2).
While BDE-153 was also added to the Stockholm
Convention in 2004, it does not share the same decreasing
trends as the lower brominated congeners. All regions
exhibited either a plateau or an increase in BDE-153
concentrations. Beyond the reduction in exposure due to the
introduction of chemical restrictions, congener-specific differences
in metabolism and storage in the body also affect breast
milk trends differently. The human elimination half-life of
BDE-153 is between 3.5 and 11.7 years.68,236,237
When
considering worst case scenario, a half-life of up to 11.7
years would result in a 4-fold reduction (in the absence of
continued exposure) of concentrations measured around 23
years ago, at the time of the implementation of the Stockholm
Convention. Therefore, a full elimination would not be
expected by the time of writing. This, coupled with continued
exposure from existing products, could explain the lack of a
decrease of BDE-153 in breast milk (Figures 2 and 4).
Global restrictions on penta- and octa-BDE technical
mixtures, which are dominated by lower brominated congeners
were generally between 2004 and 2013,33,242,243
whereas decaBDE/BDE-209
was restricted only in 2017.244
Conclusions on
BDE-209 are limited because of the lack of data and likely
because the temporal changes are not yet significant enough to
be identified in the generally short-time trend analyses that
have been performed (Figure 4). The short half-life of BDE-
209 in the body (e.g., 15 days in blood245
), combined with the
lack of a visible decline in BDE-209 levels in milk, suggests
ongoing consistent BDE-209 exposures, despite recent
restrictions in production, particularly in Asia, where the
temporal trend (Figure S6) and the inverse breakpoint (Figure
3) indicated increasing concentrations in BDE-209 in breast
milk. This agrees with our understanding of the recent high use
of BDE-209 in consumer products and building materials on a
global scale and the lag time between chemical restrictions and
product replacements: emission of BDE-209 from in-use and
waste stocks is estimated to continue until 2050.3
It is concerning how few studies investigated HBCDDs in
residents from the Americas (n = 7), Africa (n = 5), and
Oceania (n = 2) (Table 1). A similar problem was observed
with PBDEs, where Africa (n = 7), Oceania (n = 7), and
Central- and South America (n = 3) had limited studies, while
Asia and Europe had more studies (n = 53 and 61,
respectively) (Table 1). Without the proper information on
the state of contamination in the Americas, particularly the
highly developed and industrialized North America (n = 25
studies on PBDE), it is impossible to determine a global
perspective on the state of human exposure.
Breakpoint analyses are useful tools to determine whether
the accumulation trend of a compound has increased,
plateaued, or decreased over time, highlighting the approximate
time when the change occurred. In Asia and Europe, the
broad time trends of FR concentrations in breast milk are
closely tied to the timing of chemical restrictions (Figure 3).
The European trend indicated a breakpoint in ∼2010 for αHBCDD
(Figure 3), which coincides with the increase in
Environmental Science & Technology pubs.acs.org/est Policy Analysis
https://doi.org/10.1021/acs.est.3c02896
Environ. Sci. Technol. 2023, 57, 19106−19124
19114
restrictions in Europe (identified as SVHC in 2009 and listed
in Annex XIV of REACH in 2011)40
and provides evidence
that the restrictions impacted HBCDD use and thus exposures
in Europe. After 2010, it is unclear whether α-HBCDDs are in
a plateau phase or whether we begin seeing evidence of a
decrease in Europe and Asia (Figure 3), but it suggests
ongoing human exposure at levels close to the European peak.
It is important to highlight that in Figure 3, none of the
breakpoints observed for Europe or Asia align with the timing
of the Stockholm Convention’s implementation (represented
by the blue dotted line) but rather occurred earlier. This
observation suggests that regional restrictions likely had a more
pronounced impact than global restrictions. However, it should
be acknowledged that the number of studies conducted after
the implementation of the Stockholm Convention is limited
compared to those conducted before its implementation,
making it challenging to precisely assess the Convention’s
effectiveness, especially concerning HBCDD and deca-BDE.
Spatial Patterns. The strong contrast between PBDE
concentrations in North America, particularly the USA, and
most other regions is directly related to differences in
flammability standards and PBDE use. BFRs have been
quantified at higher concentrations in North America than in
Europe in multiple matrices, including human tissue,26
house
dust,246
and bird eggs.247
The USA has historically had higher
concentrations of flame retardants in its consumer products
compared with other regions due to stricter flammability
standards.248−250
The fact that PBDE congeners still show an
increasing time trend in North America (Figure 2) is likely
linked to the large past use of PBDEs, combined with the later
introduction of regulations. The United States has not ratified
the Stockholm Convention and does not have any federal
regulations on FR in existing uses, although 13 states have
internal, state-wide concentration limits on FRs in selected
products.251
Significant New Use Rules (SNURs), implemented
by the United States Environmental Protection Agency
in 2012, aim to ensure that any new uses of specific flame
retardant chemicals undergo a thorough review and approval
process prior to manufacturing or processing. The US EPA
implemented SNURs for Penta and OctaBDE in 2006,
ensuring no new use or manufacturing of these compounds.252
All manufacture, import, processing and distribution of
decaBDE was under the US TSCA in 2021; however,
significant exemptions remain, e.g., in motor vehicle parts
until 2036.253
Large numbers of products containing PBDEs
are likely still in use or circulation, which leads to continuous
exposure and a higher body burden.
Surprisingly, breast milk from the Oceania region was also
significantly higher than most other global regions, save North
America (Figure 1). While most of the world reduced PBDE
use in 2004, whether through regulation voluntary action,
Australia only began implementing regulation on PBDE use,
manufacturing, and import in 2007.238
Even though the import
of PBDEs is banned, no regulation exists for the import of
products that potentially could contain PBDEs, such as
automobile parts, textiles, or electronic products.254
BDE-47 and -99, both primary components of penta-BDE,
displayed similar spatial patterns (Figure 1), attributed to
patterns in the use of technical penta-BDE formula worldwide.
General Observations. A substantial lag-time exists
between cessation of production and cessation of use of FRs
because of the long half-lives of the compounds and the
lifespan of products that they are used in. Significant
reductions in production have a slower effect on use and
emissions because of the large stock of PBDE-containing
materials in use. Abbasi et al.3
estimated that the peak in PBDE
use occurred in 2003; however, thousands of tons of PBDEs
will remain in use in consumer products for decades. Plastic,
textile, and electronic products containing FRs are still in use,
to say nothing of buildings’ thermal insulation containing EPS
or XPS, which accounts for more than 97% of the global
HBCDD volume used.255
The ongoing human exposure to
HBCDDs will be further mediated by HBCDD exposure
through the renovation and demolition of buildings.
Demolition activity can release significant amounts of building
material-associated chemicals and demolition waste, including
EPS or XPS panels, which are estimated to stay in place for
∼50 years before renovation takes place.61
Thus, direct
exposures to HBCDDs in indoor spaces and environmental
release will continue for many decades,256
effectively slowing
the decreasing concentration levels through constant primary
exposures. These can either contribute directly to either
occupational or local population exposures, as well as increase
the burden of secondary environmental exposures through
landfill disposal and subsequent leaching of HBCDDs from the
products.31
Furthermore, as we move toward a circular economy, there
is significant potential for FRs to be incorporated into new
consumer products made from recycled materials.257,258
Abbasi
et al.3
estimated that 45000 t of PBDEs may reappear in new
products made from recycled materials, such as plastics, food
contact materials,259,260
and children’s toys.261,262
Due to the
persistence of FRs, all environmental releases can contribute to
secondary FR contamination in the surrounding air, soil, and
water sources and lead to human exposure via dietary
sources.263
The lack of recent biomonitoring studies on PBDEs and
HBCDDs limits the evaluation of current population exposures
and the impact of regulation on time trends. Of the human
biomonitoring studies (Table 1 and Figure 4), less than 10% of
the records cover the period after 2013, limiting our ability to
evaluate the effectiveness of legislation on the global exposure
trend of FRs. This may be due to the perception that once
chemicals have been regulated, the problem has been dealt
with, and it can be difficult to maintain interest and/or
financial support for chemicals that are perceived to have been
already addressed.110
Regulation has a quantifiable effect on the concentrations of
FRs in human breast milk. Regions such as Asia and Europe
where earlier regional regulations, prior to the Stockholm
Convention, have been implemented had the clearest decline
of most of the compounds that were considered in this study.
Australia, where regulations were implemented later still shows
elevated concentrations in human matrices, as does North
America, which had the historically highest FR use globally. On
top of use and regulation, the chemical characteristics of
specific PBDE congeners also affect the response to regulation
and can impact the time needed to evaluate the efficacy of
policy actions. Although PBDEs and HBCDD are regulated
through a ban on production, regulation will have different
outcomes for different individual compounds based on their
dominant use, the volume of historical use, and biological halflives.
However, it is encouraging to see that existing regulation
and policy, when implemented early and comprehensively, had
a positive impact on the decreasing human body burden of
lower brominated flame retardants.
Environmental Science & Technology pubs.acs.org/est Policy Analysis
https://doi.org/10.1021/acs.est.3c02896
Environ. Sci. Technol. 2023, 57, 19106−19124
19115
■ ASSOCIATED CONTENT
*sı Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.est.3c02896.
Figures showing temporal trend and breakpoint analysis
for all flame retardants from all regions and tables
detailing references used in literature review and metaanalysis
(PDF)
■ AUTHOR INFORMATION
Corresponding Author
Lisa Melymuk − RECETOX, Faculty of Science, Masaryk
University, 61137 Brno, Czech Republic; orcid.org/0000-
0001-6042-7688; Email: lisa.melymuk@recetox.muni.cz
Authors
Veronica van der Schyff − RECETOX, Faculty of Science,
Masaryk University, 61137 Brno, Czech Republic;
orcid.org/0000-0002-5345-4183
Jiří Kalina − RECETOX, Faculty of Science, Masaryk
University, 61137 Brno, Czech Republic
Annalisa Abballe − Department of Environment and Health,
Italian National Institute for Health, 00161 Rome, Italy
Anna Laura Iamiceli − Department of Environment and
Health, Italian National Institute for Health, 00161 Rome,
Italy
Eva Govarts − VITO Health, Flemish Institute for
Technological Research (VITO), 2400 Mol, Belgium
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.est.3c02896
Notes
The authors declare no competing financial interest.
■ ACKNOWLEDGMENTS
This study has received funding from the European Union’s
Horizon 2020 research and innovation programme under
Grant no. 733032 (HBM4EU). The authors also thank the
RECETOX Research Infrastructure (LM2023069) financed by
the Ministry of Education, Youth and Sports, and Operational
Programme Research, Development and Innovation - project
CETOCOEN EXCELLENCE (CZ.02.1.01/0.0/0.0/17_043/
0009632) for supportive background. The work was also
supported by the European Union’s Horizon 2020 research
and innovation programme under Grant no. 857560. This
publication reflects only the authors’ views, and the European
Commission is not responsible for any use that may be made
of the information it contains.
■ REFERENCES
(1) Wemken, N.; Drage, D. S.; Cellarius, C.; Cleere, K.; Morrison, J.
J.; Daly, S.; Abdallah, M. A. E.; Tlustos, C.; Harrad, S.; Coggins, M. A.
Emerging and Legacy Brominated Flame Retardants in the Breast
Milk of First Time Irish Mothers Suggest Positive Response to
Restrictions on Use of HBCDD and Penta- and Octa-BDE
Formulations. Environ. Res. 2020, 180, No. 108805.
(2) Montalbano, A. M.; Albano, G. D.; Anzalone, G.; Moscato, M.;
Gagliardo, R.; Di Sano, C.; Bonanno, A.; Ruggieri, S.; Cibella, F.;
Profita, M. Cytotoxic and Genotoxic Effects of the Flame Retardants
(PBDE-47, PBDE-99 and PBDE-209) in Human Bronchial Epithelial
Cells. Chemosphere 2020, 245, No. 125600.
(3) Abbasi, G.; Li, L.; Breivik, K. Global Historical Stocks and
Emissions of PBDEs. Environ. Sci. Technol. 2019, 53 (11), 6330−
6340.
(4) De Wit, C. A. An Overview of Brominated Flame Retardants in
the Environment. Chemosphere 2002, 46 (5), 583−624.
(5) Abbasi, G.; Buser, A. M.; Soehl, A.; Murray, M. W.; Diamond,
M. L. Stocks and Flows of PBDEs in Products from Use to Waste in
the U.S. and Canada from 1970 to 2020. Environ. Sci. Technol. 2015,
49 (3), 1521−1528.
(6) Covaci, A.; Gerecke, A. C.; Law, R. J.; Voorspoels, S.; Kohler,
M.; Heeb, N. V.; Leslie, H.; Allchin, C. R.; de Boer, J.
Hexabromocyclododecanes (HBCDs) in the Environment and
Humans: A Review. Environ. Sci. Technol. 2006, 40 (12), 3679−3688.
(7) Koch, C.; Schmidt-Kötters, T.; Rupp, R.; Sures, B. Review of
Hexabromocyclododecane (HBCD) with a Focus on Legislation and
Recent Publications Concerning Toxicokinetics and -Dynamics.
Environ. Pollut. 2015, 199, 26−34.
(8) de Wit, C. A.; Herzke, D.; Vorkamp, K. Brominated Flame
Retardants in the Arctic Environment - Trends and New Candidates.
Sci. Total Environ. 2010, 408 (15), 2885−2918.
(9) Law, R. J.; Covaci, A.; Harrad, S. J.; Herzke, D.; Abdallah, M. A.E.;
Fernie, K.; Toms, L.-M. L.; Takigami, H. Levels and Trends of
PBDEs and HBCDs in the Global Environment: Status at the End of
2012. Environ. Int. 2014, 65, 147−158.
(10) Sjodin, A; Hagmar, L; Klasson-Wehler, E; Kronholm-Diab, K;
Jakobsson, E; Bergman, A Flame Retardant Exposure: Polybrominated
Diphenyl Ethers in Blood from Swedish Workers. Environ.
Health Perspect 1999, 107 (8), 643−648.
(11) Klasson-Wehler, E.; Hovander, L.; Bergman, A. New Organohalogens
in Human Plasma - Identification and Quantification.
Organohalogen Compd. 1997, 33, 420−425.
(12) Schroter-Kermani, C.; Helm, D.; Herrmann, T.; Papke, O. The
German Environmental Specimen Bank-Application in Trend
Monitoring of Polybrominated Diphenyl Ethers in Human Blood.
Organohalogen Compd. 2000, 47, 49−52.
(13) Stanley, J. S.; Cramer, P. H.; Thornburg, K. R.; Remmers, J. C.;
Breen, J. J.; Schwemberger, J. Mass Spectral Confirmation of
Chlorinated and Brominated Diphenylethers in Human Adipose
Tissues. Chemosphere 1991, 23 (8−10), 1185−1195.
(14) Haglund, P. S.; Zook, D. R.; Buser, H. R.; Hu, J. Identification
and Quantification of Polybrominated Diphenyl Ethers and MethoxyPolybrominated
Diphenyl Ethers in Baltic Biota. Environ. Sci. Technol.
1997, 31 (11), 3281−3287.
(15) Meneses, M.; Wingfors, H.; Schuhmacher, M.; Domingo, J.L.;
Lindstrom, G.; Bavel, B.v. Polybrominated Diphenyl Ethers Detected
in Human Adipose Tissue from Spain. Chemosphere 1999, 39 (13),
2271−2278.
(16) Norén, K.; Meironyté, D. Certain Organochlorine and
Organobromine Contaminants in Swedish Human Milk in
Perspective of Past 20−30 Years. Chemosphere 2000, 40 (9−11),
1111−1123.
(17) Siddiqi, M. A.; Laessig, R. H.; Reed, K. D. Polybrominated
Diphenyl Ethers (PBDEs): New Pollutants-Old Diseases. Clin Med.
Res. 2003, 1, 281−290.
(18) Inoue, K.; Harada, K.; Takenaka, K.; Uehara, S.; Kono, M.;
Shimizu, T.; Takasuga, T.; Senthilkumar, K.; Yamashita, F.; Koizumi,
A. Levels and Concentration Ratios of Polychlorinated Biphenyls and
Polybrominated Diphenyl Ethers in Serum and Breast Milk in
Japanese Mothers. Environ. Health Perspect 2006, 114 (8), 1179−
1185.
(19) Lind, Y.; Darnerud, P. O.; Atuma, S.; Aune, M.; Becker, W.;
Bjerselius, R.; Cnattingius, S.; Glynn, A. Polybrominated Diphenyl
Ethers in Breast Milk from Uppsala County, Sweden. Environ. Res.
2003, 93 (2), 186−194.
(20) Bjermo, H.; Aune, M.; Cantillana, T.; Glynn, A.; Lind, P. M.;
Ridefelt, P.; Darnerud, P. O. Serum Levels of Brominated Flame
Retardants (BFRs: PBDE, HBCD) and Influence of Dietary Factors
in a Population-Based Study on Swedish Adults. Chemosphere 2017,
167, 485−491.
Environmental Science & Technology pubs.acs.org/est Policy Analysis
https://doi.org/10.1021/acs.est.3c02896
Environ. Sci. Technol. 2023, 57, 19106−19124
19116
(21) Barghi, M.; Shin, E. su; Choi, S. D.; Dahmardeh Behrooz, R.;
Chang, Y. S. HBCD and TBBPA in Human Scalp Hair: Evidence of
Internal Exposure. Chemosphere 2018, 207, 70−77.
(22) Drage, D. S.; Mueller, J. F.; Hobson, P.; Harden, F. A.; Toms,
L. M. L. Demographic and Temporal Trends of Hexabromocyclododecanes
(HBCDD) in an Australian Population. Environ. Res. 2017,
152, 192−198.
(23) Inthavong, C.; Hommet, F.; Bordet, F.; Rigourd, V.; Guérin, T.;
Dragacci, S. Simultaneous Liquid Chromatography−Tandem Mass
Spectrometry Analysis of Brominated Flame Retardants (Tetrabromobisphenol
A and Hexabromocyclododecane Diastereoisomers) in
French Breast Milk. Chemosphere 2017, 186, 762−769.
(24) Dorman, D. C.; Chiu, W.; Hales, B. F.; Hauser, R.; Johnson, K.
J.; Mantus, E.; Martel, S.; Robinson, K. A.; Rooney, A. A.; Rudel, R.;
Sathyanarayana, S.; Schantz, S. L.; Waters, K. M. Polybrominated
Diphenyl Ether (PBDE) Neurotoxicity: A Systematic Review and
Meta-Analysis of Animal Evidence. Journal of Toxicology and
Environmental Health, Part B 2018, 21 (4), 269−289.
(25) Wan, M. L. Y.; Co, V. A.; El-Nezami, H. Endocrine Disrupting
Chemicals and Breast Cancer: A Systematic Review of Epidemiological
Studies. Crit Rev. Food Sci. Nutr 2022, 62 (24), 6549−6576.
(26) Lyche, J. L.; Rosseland, C.; Berge, G.; Polder, A. Human Health
Risk Associated with Brominated Flame-Retardants (BFRs). Environment
International. Elsevier Ltd, 2015; pp 170−180, DOI: 10.1016/
j.envint.2014.09.006.
(27) Lam, J.; Lanphear, B. P.; Bellinger, D.; Axelrad, D. A.;
McPartland, J.; Sutton, P.; Davidson, L.; Daniels, N.; Sen, S.;
Woodruff, T. J. Developmental PBDE Exposure and IQ/ADHD in
Childhood: A Systematic Review and Meta-Analysis. Environ. Health
Perspect 2017, 125 (8), DOI: 10.1289/EHP1632.
(28) Moore, S.; Paalanen, L.; Melymuk, L.; Katsonouri, A.; KolossaGehring,
M.; Tolonen, H. The Association between ADHD and
Environmental Chemicals�A Scoping Review. Int. J. Environ. Res.
Public Health 2022, 19 (5), 2849.
(29) Iszatt, N.; Janssen, S.; Lenters, V.; Dahl, C.; Stigum, H.; Knight,
R.; Mandal, S.; Peddada, S.; González, A.; Midtvedt, T.; Eggesbø, M.
Environmental Toxicants in Breast Milk of Norwegian Mothers and
Gut Bacteria Composition and Metabolites in Their Infants at 1
Month. Microbiome 2019, 7, 34.
(30) Gascon, M.; Fort, M.; Martinez, D.; Carsin, A.-E.; Forns, J.;
Grimalt, J. O.; Santa Marina, L.; Lertxundi, N.; Sunyer, J.; Vrijheid, M.
Polybrominated Diphenyl Ethers (PBDEs) in Breast Milk and
Neuropsychological Development in Infants. Environ. Health Perspect
2012, 120 (12), 1760−1765.
(31) Feiteiro, J.; Mariana, M.; Cairrão, E. Health Toxicity Effects of
Brominated Flame Retardants: From Environmental to Human
Exposure. Environ. Pollut. 2021, 285, No. 117475.
(32) Brown, P.; Cordner, A. Lessons Learned from Flame Retardant
Use and Regulation Could Enhance Future Control of Potentially
Hazardous Chemicals. Health Aff 2011, 30 (5), 906−914.
(33) European Commission. Directive 2003/11/EC of the European
Parliament and of the Council of February 6 2003 Amending for the
24th Time Council Directive 76/669/EEC Relating to Restrictions on
the Marketing and Use of Certain Dangerous Substances and
Preparations (Pentabromodip. Off J. Eur. Union OJ. L 2003, 16 (1),
45−46.
(34) ATSDR. Toxic Substances Portal. ToxFAQs. https://wwwn.
cdc.gov/TSP/ToxFAQs/ToxFAQsDetails.aspx?faqid=1462&toxid=
183 (accessed 2023-01-28).
(35) Stockholm Convention; TetraBDE and pentaBDE. https://
chm.pops.int/Implementation/Alternatives/AlternativestoPOPs/
ChemicalslistedinAnnexA/TetraBDEandpentaBDE/tabid/5868/
Default.aspx (accessed 2023-07-24).
(36) Stockholm Convention: HexaBDE and heptaBDE. https://
chm.pops.int/Implementation/Alternatives/AlternativestoPOPs/
ChemicalslistedinAnnexA/HexaBDEandheptaBDE/tabid/5862/
Default.aspx (accessed 2023-07-24).
(37) Stockholm Convention: Decabromodiphenyl ether (commercial
mixture, c-decaBDE). https://chm.pops.int/Implementation/
Alternatives/AlternativestoPOPs/ChemicalslistedinAnnexA/
cdecaBDE/tabid/5985/Default.aspx (accessed 2023-07-24).
(38) UNEP. Stockholm Convention: United Nations Environment
Programme (UNEP). www.pops.int (accessed 2023-02-13).
(39) Risk Assessment: Hexabromocyclododecane, Final report, May
2008. https://echa.europa.eu/documents/10162/661bff17-dc0a-
4475-9758-40bdd6198f82. (accessed 2023-01-28).
(40) Kemmlein, S.; Herzke, D.; Law, R. J. Brominated Flame
Retardants in the European Chemicals Policy of REACH-Regulation
and Determination in Materials. Journal of Chromatography A 2009,
1216 (3), 320−333.
(41) UNEP. Stockholm Convention Global Monitoring Plan for
Persistent Organic Pollutants: First Global Monitoring Report;
Geneva, 2009. https://chm.pops.int/implementation/
globalmonitoringplan/monitoringreports/tabid/525/default.aspx.
(accessed 2023-03-30).
(42) Sharma, B. M.; Kalina, J.; Whaley, P.; Scheringer, M. Towards
Guidelines for Time-Trend Reviews Examining Temporal Variability
in Human Biomonitoring Data of Pollutants. Environ. Int. 2021, 151,
No. 106437.
(43) Wong, F.; Hung, H.; Dryfhout-Clark, H.; Aas, W.; BohlinNizzetto,
P.; Breivik, K.; Mastromonaco, M. N.; Lundén, E. B.;
Ólafsdóttir, K.; Sigurosson, Á.; Vorkamp, K.; Bossi, R.; Skov, H.;
Hakola, H.; Barresi, E.; Sverko, E.; Fellin, P.; Li, H.; Vlasenko, A.;
Zapevalov, M.; Samsonov, D.; Wilson, S. Time Trends of Persistent
Organic Pollutants (POPs) and Chemicals of Emerging Arctic
Concern (CEAC) in Arctic Air from 25 Years of Monitoring. Sci.
Total Environ. 2021, 775, No. 145109.
(44) Schuster, J. K.; Gioia, R.; Moeckel, C.; Agarwal, T.; Bucheli, T.
D.; Breivik, K.; Steinnes, E.; Jones, K. C. Has the Burden and
Distribution of PCBs and PBDEs Changed in European Background
Soils between 1998 and 2008? Implications for Sources and Processes.
Environ. Sci. Technol. 2011, 45 (17), 7291−7297.
(45) Sutton, R.; Sedlak, M. D.; Yee, D.; Davis, J. A.; Crane, D.;
Grace, R.; Arsem, N. Declines in Polybrominated Diphenyl Ether
Contamination of San Francisco Bay Following Production PhaseOuts
and Bans. Environ. Sci. Technol. 2015, 49 (2), 777−784.
(46) Olofsson, U.; Bignert, A.; Haglund, P. Time-Trends of Metals
and Organic Contaminants in Sewage Sludge. Water Res. 2012, 46
(15), 4841−4851.
(47) Crimmins, B. S.; Pagano, J. J.; Xia, X.; Hopke, P. K.; Milligan,
M. S.; Holsen, T. M. Polybrominated Diphenyl Ethers (PBDEs):
Turning the Corner in Great Lakes Trout 1980−2009. Environ. Sci.
Technol. 2012, 46 (18), 9890−9897.
(48) Hale, R. C.; La Guardia, M. J.; Harvey, E.; Chen, D.; Mainor, T.
M.; Luellen, D. R.; Hundal, L. S. Polybrominated Diphenyl Ethers in
U.S. Sewage Sludges and Biosolids: Temporal and Geographical
Trends and Uptake by Corn Following Land Application. Environ. Sci.
Technol. 2012, 46 (4), 2055−2063.
(49) Leslie, H. A.; Brandsma, S. H.; Barber, J. L.; Gabrielsen, G. W.;
Bersuder, P.; Barry, J.; Shore, R. F.; Walker, L. A.; de Boer, J.
Decabromodiphenylether Trends in the European Environment: Bird
Eggs, Sewage Sludge and Surficial Sediments. Sci. Total Environ. 2021,
774, No. 145174.
(50) Turyk, M. E.; Anderson, H. A.; Steenport, D.; Buelow, C.; Imm,
P.; Knobeloch, L. Longitudinal Biomonitoring for Polybrominated
Diphenyl Ethers (PBDEs) in Residents of the Great Lakes Basin.
Chemosphere 2010, 81 (4), 517−522.
(51) Schecter, A.; Papke, O.; Tung, K. C.; Joseph, J.; Harris, T R.;
Dahlgren, J. Polybrominated Diphenyl Ether Flame Retardants in the
U.S. Population: Current Levels, Temporal Trends, and Comparison
with Dioxins, Dibenzofurans, and Polychlorinated Biphenyls. J. Occup
Environ. Med. 2005, 47 (3), 199−211.
(52) Law, R. J.; Barry, J.; Barber, J. L.; Bersuder, P.; Deaville, R.;
Reid, R. J.; Brownlow, A.; Penrose, R.; Barnett, J.; Loveridge, J.;
Smith, B.; Jepson, P. D. Contaminants in Cetaceans from UK Waters:
Status as Assessed within the Cetacean Strandings Investigation
Programme from 1990 to 2008. Mar. Pollut. Bull. 2012, 64 (7), 1485−
1494.
Environmental Science & Technology pubs.acs.org/est Policy Analysis
https://doi.org/10.1021/acs.est.3c02896
Environ. Sci. Technol. 2023, 57, 19106−19124
19117
(53) Munschy, C.; Marchand, P.; Venisseau, A.; Veyrand, B.;
Zendong, Z. Levels and Trends of the Emerging Contaminants
HBCDs (Hexabromocyclododecanes) and PFCs (Perfluorinated
Compounds) in Marine Shellfish along French Coasts. Chemosphere
2013, 91 (2), 233−240.
(54) Smythe, T. A.; Loseto, L. L.; Bignert, A.; Rosenberg, B.;
Budakowski, W.; Halldorson, T.; Pleskach, K.; Tomy, G. T. Temporal
Trends of Brominated and Fluorinated Contaminants in Canadian
Arctic Beluga Delphinapterus Leucas. Arct Sci. 2018, 4 (3), 388−404.
(55) Johansson, A.-K.; Sellström, U.; Lindberg, P.; Bignert, A.; de
Wit, C. A. Temporal Trends of Polybrominated Diphenyl Ethers and
Hexabromocyclododecane in Swedish Peregrine Falcon (Falco
Peregrinus Peregrinus) Eggs. Environ. Int. 2011, 37 (4), 678−686.
(56) Vorkamp, K.; Falk, K.; Møller, S.; Rigét, F. F.; Sørensen, P. B.
Regulated and Unregulated Halogenated Flame Retardants in
Peregrine Falcon Eggs from Greenland. Environ. Sci. Technol. 2018,
52 (2), 474−483.
(57) Rüdel, H.; Müller, J.; Nowak, J.; Ricking, M.; Klein, R.;
Kotthoff, M. Hexabromocyclododecane Diastereomers in Fish and
Suspended Particulate Matter from Selected European Waters�
Trend Monitoring and Environmental Quality Standard Compliance.
Environmental Science and Pollution Research 2017, 24 (22), 18048−
18062.
(58) Aznar-Alemany, Ò.; Sala, B.; Jobst, K. J.; Reiner, E. J.; Borrell,
A.; Aguilar, À.; Eljarrat, E. Temporal Trends of Halogenated and
Organophosphate Contaminants in Striped Dolphins from the
Mediterranean Sea. Sci. Total Environ. 2021, 753, No. 142205.
(59) Tao, F.; Abou-Elwafa Abdallah, M.; Ashworth, D. C.; Douglas,
P.; Toledano, M. B.; Harrad, S. Emerging and Legacy Flame
Retardants in UK Human Milk and Food Suggest Slow Response
to Restrictions on Use of PBDEs and HBCDD. Environ. Int. 2017,
105, 95−104.
(60) Magnier, L.; Mugge, R. Replaced Too Soon? An Exploration of
Western European Consumers’ Replacement of Electronic Products.
Resour Conserv Recycl 2022, 185, No. 106448.
(61) Charbonnet, J. A.; Weber, R.; Blum, A. Flammability Standards
for Furniture, Building Insulation and Electronics: Benefit and Risk.
Emerg. Contam. 2020, 6, 432−441.
(62) Abbatt, J. P. D.; Wang, C. The Atmospheric Chemistry of
Indoor Environments. Environ. Sci.: Processes Impacts 2020, 22, 25−
48.
(63) Melymuk, L.; Demirtepe, H.; Jílková, S. R. Indoor Dust and
Associated Chemical Exposures. Current Opinion in Environmental
Science and Health 2020, 15, 1−6.
(64) Rodgers, K. M.; Bennett, D.; Moran, R.; Knox, K.; Stoiber, T.;
Gill, R.; Young, T. M.; Blum, A.; Dodson, R. E. Do Flame Retardant
Concentrations Change in Dust after Older Upholstered Furniture Is
Replaced? Environ. Int. 2021, 153, No. 106513.
(65) Bennett, D. H.; Moran, R. E.; Wu, X. M.; Tulve, N. S.; Clifton,
M. S.; Colon, M.; Weathers, W.; Sjodin, A.; Jones, R.; Hertz-Picciotto,
I. Polybrominated Diphenyl Ether (PBDE) Concentrations and
Resulting Exposure in Homes in California: Relationships among
Passive Air, Surface Wipe and Dust Concentrations, and Temporal
Variability. Indoor Air 2015, 25 (2), 220−229.
(66) van der Schyff, V.; Kalina, J.; Govarts, E.; Gilles, L.; Schoeters,
G.; Castaño, A.; Esteban-López, M.; Kohoutek, J.; Kukučka, P.;
Covaci, A.; Koppen, G.; Andrýsková, L.; Piler, P.; Klánová, J.; Jensen,
T. K.; Rambaud, L.; Riou, M.; Lamoree, M.; Kolossa-Gehring, M.;
Vogel, N.; Weber, T.; Göen, T.; Gabriel, C.; Sarigiannis, D. A.; Sakhi,
A. K.; Haug, L. S.; Murinova, L. P.; Fabelova, L.; Tratnik, J. S.; Mazej,
D.; Melymuk, L. Exposure to Flame Retardants in European Children
� Results from the HBM4 EU Aligned Studies. International Journal
of Hygiene and Environmental Health 2023, 247, No. 114070.
(67) Shin, H. M.; McKone, T. E.; Tulve, N. S.; Clifton, M. S.;
Bennett, D. H. Indoor Residence Times of Semivolatile Organic
Compounds: Model Estimation and Field Evaluation. Environ. Sci.
Technol. 2013, 47 (2), 859−867.
(68) Geyer, H. J.; Schramm, K.-W.; Darnerud, O.; Aune, M.; Feicht,
E. A.; Fried, K. W.; Henkelmann, B.; Lenoir, D.; Schmid, P.;
Mcdonald, T. A. Terminal Elimination Half-Lives of the Brominated
Flame Retardants TBBPA, HBCD, and Lower Brominated PBDEs in
Humans. Organohalogen Compd. 2004, 66, 3820−3825.
(69) t’Mannetje, A.; Coakley, J.; Mueller, J. F.; Harden, F.; Toms, L.M.;
Douwes, J. Partitioning of Persistent Organic Pollutants (POPs)
between Human Serum and Breast Milk: A Literature Review.
Chemosphere 2012, 89 (8), 911−918.
(70) Schecter, A.; Colacino, J.; Sjödin, A.; Needham, L.; Birnbaum,
L. Partitioning of Polybrominated Diphenyl Ethers (PBDEs) in
Serum and Milk from the Same Mothers. Chemosphere 2010, 78 (10),
1279−1284.
(71) Komprda, J.; Komprdová, K.; Domínguez-Romero, E.; Mikeš,
O.; Řiháčková, K.; Čupr, P.; Černá, M.; Scheringer, M. Dynamics of
PCB Exposure in the Past 50 Years and Recent High Concentrations
in Human Breast Milk: Analysis of Influencing Factors Using a
Physiologically Based Pharmacokinetic Model. Sci. Total Environ.
2019, 690, 388−399.
(72) Hajeb, P.; Castaño, A.; Cequier, E.; Covaci, A.; López, M. E.;
Antuña, A. G.; Haug, L. S.; Henríquez-Hernández, L. A.; Melymuk,
L.; Pérez Luzardo, O.; Thomsen, C.; Vorkamp, K. Critical Review of
Analytical Methods for the Determination of Flame Retardants in
Human Matrices. Anal. Chim. Acta 2022, 1193, No. 338828.
(73) UNEP. Global Monitoring Plan for Persistent Organic
Pollutants under the Stockholm Convention Article 16 on
Effectiveness Evaluation - Third Regional Monitoring Report Latin
America and the Caribbean; 2021. https://chm.pops.int/
implementation/globalmonitoringplan/monitoringreports/tabid/
525/default.aspx. (accessed 2023-03-30).
(74) Johnson-Restrepo, B.; Kannan, K. An Assessment of Sources
and Pathways of Human Exposure to Polybrominated Diphenyl
Ethers in the United States. Chemosphere 2009, 76 (4), 542−548.
(75) Sahlström, L. M. O.; Sellström, U.; de Wit, C. A.; Lignell, S.;
Darnerud, P. O. Brominated Flame Retardants in Matched Serum
Samples from Swedish First-Time Mothers and Their Toddlers.
Environ. Sci. Technol. 2014, 48 (13), 7584−7592.
(76) Eljarrat, E.; Guerra, P.; Martínez, E.; Farré, M.; Alvarez, J. G.;
López-Teijón, M.; Barceló, D. Hexabromocyclododecane in Human
Breast Milk: Levels and Enantiomeric Patterns. Environ. Sci. Technol.
2009, 43 (6), 1940−1946.
(77) Ryan, J. J.; Rawn, D. F. K. The Brominated Flame Retardants,
PBDEs and HBCD, in Canadian Human Milk Samples Collected
from 1992 to 2005; Concentrations and Trends. Environ. Int. 2014,
70, 1−8.
(78) Carignan, C. C.; Abdallah, M. A. E.; Wu, N.; Heiger-Bernays,
W.; McClean, M. D.; Harrad, S.; Webster, T. F. Predictors of
Tetrabromobisphenol-A (TBBP-A) and Hexabromocyclododecanes
(HBCD) in Milk from Boston Mothers. Environ. Sci. Technol. 2012,
46 (21), 12146−12153.
(79) Asante, K. A.; Adu-Kumi, S.; Nakahiro, K.; Takahashi, S.; Isobe,
T.; Sudaryanto, A.; Devanathan, G.; Clarke, E.; Ansa-Asare, O. D.;
Dapaah-Siakwan, S.; Tanabe, S. Human Exposure to PCBs, PBDEs
and HBCDs in Ghana: Temporal Variation, Sources of Exposure and
Estimation of Daily Intakes by Infants. Environ. Int. 2011, 37 (5),
921−928.
(80) Darnerud, P. O.; Aune, M.; Larsson, L.; Lignell, S.; Mutshatshi,
T.; Okonkwo, J.; Botha, B.; Agyei, N. Levels of Brominated Flame
Retardants and Other Pesistent Organic Pollutants in Breast Milk
Samples from Limpopo Province, South Africa. Sci. Total Environ.
2011, 409 (19), 4048−4053.
(81) Hůlek, R.; Borůvková, J.; Gregor, J.; Kalina, J.; Bednářová, Z.;
Šebková, K.; Melkes, O.; Šalko, M.; Novák, R.; Jarkovský, J.; Dušek,
L.; Klánová, J. Global Monitoring Plan of the Stockholm Convention on
Persistent Organic Pollutants: visualisation and on-line analysis of global
levels of chemicals in air, water, breast milk and blood [online]. Masaryk
University, 2014, http://visualization.pops-gmp.org/2014/ (accessed
2022-07-14).
(82) Müller, M. H. B.; Polder, A.; Brynildsrud, O. B.; Lie, E.; Løken,
K. B.; Manyilizu, W. B.; Mdegela, R. H.; Mokiti, F.; Murtadha, M.;
Nonga, H. E.; Skaare, J. U.; Lyche, J. L. Brominated Flame Retardants
Environmental Science & Technology pubs.acs.org/est Policy Analysis
https://doi.org/10.1021/acs.est.3c02896
Environ. Sci. Technol. 2023, 57, 19106−19124
19118
(BFRs) in Breast Milk and Associated Health Risks to Nursing Infants
in Northern Tanzania. Environ. Int. 2016, 89−90, 38−47.
(83) UNEP. Global Monitoring Plan for Persistent Organic
Pollutants under the Stockholm Convention Article 16 on
Effectiveness Evaluation - Third Regional Monitoring Report - Africa
Region; 2021. https://chm.pops.int/implementation/
globalmonitoringplan/monitoringreports/tabid/525/default.aspx.
(accessed 2023-03-30).
(84) Hassine, S. B.; Ameur, W. B.; Gandoura, N.; Driss, M. R.
Determination of Chlorinated Pesticides, Polychlorinated Biphenyls,
and Polybrominated Diphenyl Ethers in Human Milk from Bizerte
(Tunisia) in 2010. Chemosphere 2012, 89 (4), 369−377.
(85) Matovu, H.; Sillanpää, M.; Ssebugere, P. Polybrominated
Diphenyl Ethers in Mothers’ Breast Milk and Associated Health Risk
to Nursing Infants in Uganda. Science of The Total Environment 2019,
692, 1106−1115.
(86) Horng, C.; Chao, H.; Chang, C.; Govindasamy, A.; Tzeng, T.;
Shieh, P. Polybrominated Diphenyl Ethers in Human Milk in Taiwan.
Asian J. Chem. 2010, 22 (4), 2869−2878.
(87) Zhang, J. G.; Sun, X. W.; Ai, H. Levels and Congener Profiles of
Polybrominated Diphenyl Ethers (PBDEs) in Primipara Breast Milk
from Shenzhen and Exposure Risk for Breast-Fed Infants. Journal of
Environmental Monitoring 2012, 14, 893−900.
(88) Sudaryanto, A.; Kajiwara, N.; Takahashi, S.; Muawanah;
Tanabe, S. Geographical Distribution and Accumulation Features of
PBDEs in Human Breast Milk from Indonesia. Environ. Pollut. 2008,
151 (1), 130−138.
(89) UNEP. Global monitoring plan for Persistent Organic
Pollutants. Third regional monitoring report Asia-Pacific region.
http://chm.pops.int/implementation/globalmonitoringplan/
monitoringreports/tabid/525/default.aspx (accessed 2023-03-30).
(90) Zhu, L.; Ma, B.; Li, J.; Wu, Y.; Gong, J. Distribution of
Polybrominated Diphenyl Ethers in Breast Milk from North China:
Implication of Exposure Pathways. Chemosphere 2009, 74, 1429−
1434.
(91) Zhang, X.; Zhang, K.; Yang, D.; Ma, L.; Lei, B.; Zhang, X.;
Zhou, J.; Fang, X.; Yu, Y. Polybrominated Biphenyl Ethers in Breast
Milk and Infant Formula from Shanghai, China: Temporal Trends,
Daily Intake, and Risk Assessment. Sci. Total Environ. 2014, 497,
508−515.
(92) Yin, S.; Guo, F.; Aamir, M.; Liu, Y.; Tang, M.; Liu, W.
Multicenter Biomonitoring of Polybrominated Diphenyl Ethers
(PBDEs) in Colostrum from China: Body Burden Profile and Risk
Assessment. Environ. Res. 2019, 179, No. 108828.
(93) Yang, L.; Lu, Y.; Wang, L.; Chang, F.; Zhang, J.; Liu, Y. Levels
and Profiles of Polybrominated Diphenyl Ethers in Breast Milk
During Different Nursing Durations. Bull. Environ. Contam. Toxicol.
2016, 97, 510−516.
(94) Wang, Y.-F.; Wang, S.-L.; Chen, F.-A.; Chao, H. A.; Tsou, T.C.;
Shy, C.-G.; Päpke, O.; Kuo, Y.-M.; Chao, H.-R. Associations of
Polybrominated Diphenyl Ethers (PBDEs) in Breast Milk and Dietary
Habits and Demographic Factors in Taiwan. Food Chem. Toxicol.
2008, 46, 1925−1932.
(95) Tue, N. M.; Sudaryanto, A.; Minh, T. B.; Isobe, T.; Takahashi,
S.; Viet, P. H.; Tanabe, S. Accumulation of Polychlorinated Biphenyls
and Brominated Flame Retardants in Breast Milk from Women Living
in Vietnamese E-Waste Recycling Sites. Science of the Total
Environment, The 2010, 408, 2155−2162.
(96) Tsydenova, O. V.; Sudaryanto, A.; Kajiwara, N.; Kunisue, T.;
Batoev, V. B.; Tanabe, S. Organohalogen Compounds in Human
Breast Milk from Republic of Buryatia, Russia. Environ. Pollut. 2007,
146, 225−232.
(97) Sun, S.; Zhao, J.; Leng, J.; Wang, P.; Wang, Y.; Fukatsu, H.; Liu,
D.; Liu, X.; Kayama, F. Levels of Dioxins and Polybrominated
Diphenyl Ethers in Human Milk from Three Regions of Northern
China and Potential Dietary Risk Factors. Chemosphere 2010, 80,
1151−1159.
(98) Sudaryanto, A.; Kajiwara, N.; Tsydenova, O. V.; Isobe, T.; Yu,
H.; Takahashi, S.; Tanabe, S. Levels and Congener Specific Profiles of
PBDEs in Human Breast Milk from China: Implication on Exposure
Sources and Pathways. Chemosphere 2008, 73 (10), 1661−1668.
(99) Shy, C.-G.; Huang, H.-L.; Chao, H.-R.; Chang-Chien, G.-P.
Cord Blood Levels of Thyroid Hormones and IGF-1 Weakly
Correlate with Breast Milk Levels of PBDEs in Taiwan. Int. J. Hyg
Environ. Health 2012, 215, 345−351.
(100) Shi, Z.; Wang, Y.; Niu, P.; Wang, J.; Sun, Z.; Zhang, S.; Wu, Y.
Concurrent Extraction, Clean-up, and Analysis of Polybrominated
Diphenyl Ethers, Hexabromocyclododecane Isomers, and Tetrabromobisphenol
A in Human Milk and Serum. J. Sep Sci. 2013, 36 (20),
3402−3410.
(101) Shi, Z.; Jiao, Y.; Hu, Y.; Sun, Z.; Zhou, X.; Feng, J.; Li, J.; Wu,
Y. Levels of Tetrabromobisphenol A, Hexabromocyclododecanes and
Polybrominated Diphenyl Ethers in Human Milk from the General
Population in Beijing, China. Sci. Total Environ. 2013, 452−453, 10−
18.
(102) Shi, Z.; Zhang, L.; Zhao, Y.; Sun, Z.; Zhou, X.; Li, J.; Wu, Y. A
National Survey of Tetrabromobisphenol-A, Hexabromocyclododecane
and Decabrominated Diphenyl Ether in Human Milk from
China: Occurrence and Exposure Assessment. Sci. Total Environ.
2017, 599−600, 237−245.
(103) Shen, H.; Ding, G.; Wu, Y.; Pan, G.; Zhou, X.; Han, J.; Li, J.;
Wen, S. Polychlorinated Dibenzo-p-Dioxins/Furans (PCDD/Fs),
Polychlorinated Biphenyls (PCBs), and Polybrominated Diphenyl
Ethers (PBDEs) in Breast Milk from Zhejiang, China. Environ. Int.
2012, 42, 84−90.
(104) Ohta, S.; Ishizuka, D.; Nishimura, H.; Nakao, T.; Aozasa, O.;
Shimidzu, Y.; Ochiai, F.; Kida, T.; Nishi, M.; Miyata, H. Comparison
of Polybrominated Diphenyl Ethers in Fish, Vegetables, and Meats
and Levels in Human Milk of Nursing Women in Japan. Chemosphere
2002, 46, 689−696.
(105) Ma, S.; Yu, Z.; Zhang, X.; Ren, G.; Peng, P.; Sheng, G.; Fu, J.
Levels and Congener Profiles of Polybrominated Diphenyl Ethers
(PBDEs) in Breast Milk from Shanghai: Implication for Exposure
Route of Higher Brominated BDEs. Environ. Int. 2012, 42, 72−77.
(106) Lin, Y.; Feng, C.; Xu, Q.; Lu, D.; Qiu, X.; Jin, Y.; Wang, G.;
Wang, D.; She, J.; Zhou, Z. A Validated Method for Rapid
Determination of Dibenzo-p-Dioxins/Furans (PCDD/Fs), Polybrominated
Diphenyl Ethers (PBDEs) and Polychlorinated Biphenyls
(PCBs) in Human Milk: Focus on Utility of Tandem Solid Phase
Extraction (SPE) Cleanup. Anal Bioanal Chem. 2016, 408, 4897−
4906.
(107) Li, X.; Tian, Y.; Zhang, Y.; Ben, Y.; Lv, Q. Accumulation of
Polybrominated Diphenyl Ethers in Breast Milk of Women from an EWaste
Recycling Center in China. Journal of Environmental Sciences
2017, 52, 305−313.
(108) Li, J.; Yu, H.; Zhao, Y.; Zhang, G.; Wu, Y. Levels of
Polybrominated Diphenyl Ethers (PBDEs) in Breast Milk from
Beijing, China. Chemosphere 2008, 73, 182−186.
(109) Li, J.; Ma, W.; Zhao, Y.; Jin, Y.; Xiao, X.; Ge, W.; Shi, H.;
Zhang, Y. Lactational Exposure of Polybrominated Diphenyl Ethers
and Its Association with Infant Developmental Measurements. J.
Hazard Mater. 2020, 388, No. 122031.
(110) Leung, A. O. W.; Chan, J. K. Y.; Xing, G. H.; Xu, Y.; Wu, S.
C.; Wong, C. K. C.; Leung, C. K. M.; Wong, M. H. Body Burdens of
Polybrominated Diphenyl Ethers in Childbearing-Aged Women at an
Intensive Electronic-Waste Recycling Site in China. Environmental
Science and Pollution Research 2010, 17, 1300−1313.
(111) Lee, S.; Kim, S.; Kim, E.; Lee, I.-S.; Choi, G.; Kim, H.-J.; Park,
J.; Jae Lee, J.; Choi, S.; Young Kim, S.; Kim, S.; Kim, S.; Choi, K.;
Moon, H.-B. Polybrominated Diphenyl Ethers (PBDEs) in Breast
Milk of Korea in 2011: Current Contamination, Time Course
Variation, Influencing Factors and Health Risks. Environ. Res. 2013,
126, 76−83.
(112) Koh, T.-W.; Chih-Cheng Chen, S.; Chang-Chien, G.-P.; Lin,
D.-Y.; Chen, F.-A.; Chao, H.-R. Breast-Milk Levels of Polybrominated
Diphenyl Ether Flame Retardants in Relation to Women’s Age and
Pre-Pregnant Body Mass Index. Int. J. Hyg Environ. Health 2010, 213,
59−65.
Environmental Science & Technology pubs.acs.org/est Policy Analysis
https://doi.org/10.1021/acs.est.3c02896
Environ. Sci. Technol. 2023, 57, 19106−19124
19119
(113) Kim, U.-J.; Lee, I.-S.; Kim, H. S.; Oh, J.-E. Monitoring of
PBDEs Concentration in Umbilical Cord Blood and Breast Milk from
Korean Population and Estimating the Effects of Various Parameters
on Accumulation in Humans. Chemosphere 2011, 85, 487−493.
(114) Kim, T. H.; Bang, D. Y.; Lim, H. J.; Jin Won, A.; Ahn, M. Y.;
Patra, N.; Chung, K. K.; Kwack, S. J.; Park, K. L.; Han, S. Y.; Choi, W.
S.; Han, J. Y.; Lee, B. M.; Oh, J. E.; Yoon, J. H.; Lee, J.; Kim, H. S.
Comparisons of Polybrominated Diphenyl Ethers Levels in Paired
South Korean Cord Blood, Maternal Blood, and Breast Milk Samples.
Chemosphere 2012, 87 (1), 97−104.
(115) KAWASHIRO, Y.; FUKATA, H.; OMORI-INOUE, M.;
KUBONOYA, K.; JOTAKI, T.; TAKIGAM, H.; SAKAI, S.-i.; MORI,
C. Perinatal Exposure to Brominated Flame Retardants and
Polychlorinated Biphenyls in Japan. Endocr J. 2008, 55 (6), 1071−
1084.
(116) Kao, C.-C.; Chen, C.-C.; Avelino, J. L.; Cortez, M. P.; Tayo, L.
L.; Lin, Y.-H.; Tsai, M.-H.; Lin, C.-W.; Hsu, Y.-C.; Hsieh, L.-T.; Lin,
C.; Wang, L.-C.; Yu, K.-L. J.; Chao, H.-R. Infants’ Neurodevelopmental
Effects of PM2.5 and Persistent Organohalogen
Pollutants Exposure in Southern Taiwan. Aerosol Air Qual Res.
2019, 19, 2793−2803.
(117) Kang, C. S.; Lee, J.-H.; Kim, S.-K.; Lee, K.-T.; Lee, J. S.; Park,
S.; Yun, H.; Kannan, K.; Yoo, Y. W.; Ha, Y.; Lee, S. W.
Polybrominated Diphenyl Ethers and Synthetic Musks in Umbilical
Cord Serum, Maternal Serum, and Breast Milk from Seoul, South
Korea. Chemosphere 2010, 80, 116−122.
(118) Jin, J.; Wang, Y.; Yang, C.; Hu, J.; Liu, W.; Cui, J.; Tang, X.
Polybrominated Diphenyl Ethers in the Serum and Breast Milk of the
Resident Population from Production Area, China. Environ. Int. 2009,
35, 1048−1052.
(119) Hedley, A. J.; Hui, L. L.; Kypke, K.; Malisch, R.; van Leeuwen,
F. X. R.; Moy, G.; Wong, T. W.; Nelson, E. A. S. Residues of
Persistent Organic Pollutants (POPs) in Human Milk in Hong Kong.
Chemosphere 2010, 79 (3), 259−265.
(120) Haraguchi, K.; Koizumi, A.; Inoue, K.; Harada, K. H.; Hitomi,
T.; Minata, M.; Tanabe, M.; Kato, Y.; Nishimura, E.; Yamamoto, Y.;
Watanabe, T.; Takenaka, K.; Uehara, S.; Yang, H. R.; Kim, M. Y.;
Moon, C. S.; Kim, H. S.; Wang, P.; Liu, A.; Hung, N. N. Levels and
Regional Trends of Persistent Organochlorines and Polybrominated
Diphenyl Ethers in Asian Breast Milk Demonstrate POPs Signatures
Unique to Individual Countries. Environ. Int. 2009, 35 (7), 1072−
1079.
(121) Fujii, Y.; Ito, Y.; Harada, K. H.; Hitomi, T.; Koizumi, A.;
Haraguchi, K. Regional Variation and Possible Sources of Brominated
Contaminants in Breast Milk from Japan. Environ. Pollut. 2012, 162,
269−274.
(122) Eslami, B.; Koizumi, A.; Ohta, S.; Inoue, K.; Aozasa, O.;
Harada, K.; Yoshinaga, T.; Date, C.; Fujii, S.; Fujimine, Y.; Hachiya,
N.; Hirosawa, I.; Koda, S.; Kusaka, Y.; Murata, K.; Nakatsuka, H.;
Omae, K.; Saito, N.; Shimbo, S.; Takenaka, K.; Takeshita, T.;
Todoriki, H.; Wada, Y.; Watanabe, T.; Ikeda, M. Large-Scale
Evaluation of the Current Level of Polybrominated Diphenyl Ethers
(PBDEs) in Breast Milk from 13 Regions of Japan. Chemosphere 2006,
63 (4), 554−561.
(123) Cui, C.; Tian, Y.; Zhang, L.; Gao, Y.; Jin, J.; Wang, P.; Ding,
W.; Wang, X.; Shi, R.; Wang, Y. Polybrominated Diphenyl Ethers
Exposure in Breast Milk in Shanghai, China: Levels, Influencing
Factors and Potential Health Risk for Infants. Science of The Total
Environment 2012, 433, 331−335.
(124) Chen, Z. J.; Liu, H. Y.; Cheng, Z.; Man, Y. B.; Zhang, K. S.;
Wei, W.; Du, J.; Wong, M. H.; Wang, H. S. Polybrominated Diphenyl
Ethers (PBDEs) in Human Samples of Mother-Newborn Pairs in
South China and Their Placental Transfer Characteristics. Environ.
Int. 2014, 73, 77−84.
(125) Chen, T.; Huang, M.; Li, J.; Li, J.; Shi, Z. Polybrominated
Diphenyl Ethers and Novel Brominated Flame Retardants in Human
Milk from the General Population in Beijing, China: Occurrence,
Temporal Trends, Nursing Infants’ Exposure and Risk Assessment.
Science of The Total Environment 2019, 689, 278−286.
(126) Chen, M.-W.; Castillo, B. A. A.; Lin, D.-Y.; Chao, H.-R.; Tayo,
L. L.; Gou, Y.-Y.; Chen, F.-A.; Huang, K.-L. Levels of PCDD/Fs,
PBDEs, and PBDD/Fs in Breast Milk from Southern Taiwan. Bull.
Environ. Contam. Toxicol. 2018, 100, 369−375.
(127) Chao, H. R.; Wang, S. L.; Lee, W. J.; Wang, Y. F.; Päpke, O.
Levels of Polybrominated Diphenyl Ethers (PBDEs) in Breast Milk
from Central Taiwan and Their Relation to Infant Birth Outcome and
Maternal Menstruation Effects. Environ. Int. 2007, 33 (2), 239−245.
(128) Chao, H.-R.; Tsou, T.-C.; Huang, H.-L.; Chang-Chien, G.-P.
Levels of Breast Milk PBDEs From Southern Taiwan and Their
Potential Impact on Neurodevelopment. Pediatr. Res. 2011, 70 (6),
596−600.
(129) Chao, H. A.; Chen, S. C. C.; Chang, C. M.; Koh, T. W.;
Chang-Chien, G. P.; Ouyang, E.; Lin, S. L.; Shy, C. G.; Chen, F. A.;
Chao, H. R. Concentrations of Polybrominated Diphenyl Ethers in
Breast Milk Correlated to Maternal Age, Education Level, and
Occupational Exposure. J. Hazard Mater. 2010, 175 (1−3), 492−500.
(130) Bi, X.; Qu, W.; Sheng, G.; Zhang, W.; Mai, B.; Chen, D.; Yu,
L.; Fu, J. Polybrominated Diphenyl Ethers in South China Maternal
and Fetal Blood and Breast Milk. Environ. Pollut. 2006, 144 (3),
1024−1030.
(131) Akutsu, K.; Kitagawa, M.; Nakazawa, H.; Makino, T.; Iwazaki,
K.; Oda, H.; Hori, S. Time-Trend (1973−2000) of Polybrominated
Diphenyl Ethers in Japanese Mother’s Milk. Chemosphere 2003, 53
(6), 645−654.
(132) Fujii, Y.; Kato, Y.; Masuda, N.; Harada, K. H.; Koizumi, A.;
Haraguchi, K. Contamination Trends and Factors Affecting the
Transfer of Hexabromocyclododecane Diastereomers, Tetrabromobisphenol
A, and 2,4,6-Tribromophenol to Breast Milk in Japan.
Environ. Pollut. 2018, 237, 936−943.
(133) Devanathan, G.; Subramanian, A.; Sudaryanto, A.; Takahashi,
S.; Isobe, T.; Tanabe, S. Brominated Flame Retardants and
Polychlorinated Biphenyls in Human Breast Milk from Several
Locations in India: Potential Contaminant Sources in a Municipal
Dumping Site. Environ. Int. 2012, 39 (1), 87−95.
(134) Malarvannan, G.; Isobe, T.; Covaci, A.; Prudente, M.; Tanabe,
S. Accumulation of Brominated Flame Retardants and Polychlorinated
Biphenyls in Human Breast Milk and Scalp Hair from the
Philippines: Levels, Distribution and Profiles. Sci. Total Environ. 2013,
442, 366−379.
(135) Malarvannan, G.; Kunisue, T.; Isobe, T.; Sudaryanto, A.;
Takahashi, S.; Prudente, M.; Subramanian, A.; Tanabe, S. Organohalogen
Compounds in Human Breast Milk from Mothers Living in
Payatas and Malate, the Philippines: Levels, Accumulation Kinetics
and Infant Health Risk. Environ. Pollut. 2009, 157 (6), 1924−1932.
(136) Tsai, M. H.; Chao, H. R.; Hsu, W. L.; Tsai, C. C.; Lin, C. W.;
Chen, C. H. Analysis of Polybrominated Diphenyl Ethers and Lipid
Composition in Human Breast Milk and Their Correlation with
Infant Neurodevelopment. Int. J. Environ. Res. Public Health 2021, 18
(21), 11501.
(137) Huang, M.; Li, J.; Xiao, Z.; Shi, Z. Tetrabromobisphenol A
and Hexabromocyclododecane Isomers in Breast Milk from the
General Population in Beijing, China: Contamination Levels,
Temporal Trends, Nursing Infant’s Daily Intake, and Risk Assessment.
Chemosphere 2020, 244, No. 125524.
(138) Kakimoto, K.; Akutsu, K.; Konishi, Y.; Tanaka, Y. Time Trend
of Hexabromocyclododecane in the Breast Milk of Japanese Women.
Chemosphere 2008, 71 (6), 1110−1114.
(139) Kakimoto, K.; Akutsu, K.; Konishi, Y.; Toriba, A.; Hayakawa,
K. Hexabromocyclododecane Levels in Foodstuff and Human Breast
Milk in Osaka, Japan. Interdisciplinary Studies on Environmental
Chemistry 2012, 6, 253−259, DOI: 10.1021/bk-2016-1244.ch006.
(140) Lu, S.; Tan, Z.; Jiang, Y.; Wu, D.; Zhang, J.; Zhou, J.; Lin, X.
Hexabromocyclododecanes in Breast Milk from Residents in
Shenzhen, China: Implications for Infant Exposure. Sci. Total Environ.
2018, 622−623, 1090−1097.
(141) Shi, Z.; Wang, Y.; Niu, P.; Wang, J.; Sun, Z.; Zhang, S.; Wu, Y.
Concurrent Extraction, Clean-up, and Analysis of Polybrominated
Diphenyl Ethers, Hexabromocyclododecane Isomers, and TetrabroEnvironmental
Science & Technology pubs.acs.org/est Policy Analysis
https://doi.org/10.1021/acs.est.3c02896
Environ. Sci. Technol. 2023, 57, 19106−19124
19120
mobisphenol A in Human Milk and Serum. J. Sep. Sci. 2013, 36,
3402−3410.
(142) Shi, Z.; Zhang, L.; Zhao, Y.; Sun, Z.; Zhou, X.; Li, J.; Wu, Y. A
National Survey of Tetrabromobisphenol-A, Hexabromocyclododecane
and Decabrominated Diphenyl Ether in Human Milk from
China: Occurrence and Exposure Assessment. Sci. Total Environ.
2017, 599−600, 237−245.
(143) Shi, Z. X.; Wu, Y. N.; Li, J. G.; Zhao, Y. F.; Feng, J. F. Dietary
Exposure Assessment of Chinese Adults and Nursing Infants to
Tetrabromobisphenol-A and Hexabromocyclododecanes: Occurrence
Measurements in Foods and Human Milk. Environ. Sci. Technol. 2009,
43 (12), 4314−4319.
(144) Shi, Z.; Jiao, Y.; Hu, Y.; Sun, Z.; Zhou, X.; Feng, J.; Li, J.; Wu,
Y. Levels of Tetrabromobisphenol A, Hexabromocyclododecanes and
Polybrominated Diphenyl Ethers in Human Milk from the General
Population in Beijing, China. Sci. Total Environ. 2013, 452−453, 10−
18.
(145) Souza, M. C. O.; Devóz, P. P.; Ximenez, J. P. B.; Bocato, M.
Z.; Rocha, B. A.; Barbosa, F. Potential Health Risk to Brazilian Infants
by Polybrominated Diphenyl Ethers Exposure via Breast Milk Intake.
Int. J. Environ. Res. Public Health 2022, 19 (17), 11138.
(146) Antignac, J. P.; Main, K. M.; Virtanen, H. E.; Boquien, C. Y.;
Marchand, P.; Venisseau, A.; Guiffard, I.; Bichon, E.; Wohlfahrt-Veje,
C.; Legrand, A.; Boscher, C.; Skakkebæk, N. E.; Toppari, J.; Le Bizec,
B. Country-Specific Chemical Signatures of Persistent Organic
Pollutants (POPs) in Breast Milk of French, Danish and Finnish
Women. Environ. Pollut. 2016, 218, 728−738.
(147) Aune, M.; Barregård, L.; Claesson, A.; Darnerud, P. O.
Resultatrapport till Miljöövervakningen: Organiska Miljögifter i
Bröstmjölk Från Göteborg 2001; Livsmedelsverket: 2002.
(148) Darnerud, P. O.; Lignell, S.; Aune, M.; Isaksson, M.;
Cantillana, T.; Redeby, J.; Glynn, A. Time Trends of Polybrominated
Diphenylether (PBDE) Congeners in Serum of Swedish Mothers and
Comparisons to Breast Milk Data. Environ. Res. 2015, 138, 352−360.
(149) Eggesbø, M.; Thomsen, C.; Jørgensen, J. V.; Becher, G.;
Øyvind Odland, J.; Longnecker, M. P. Associations between
Brominated Flame Retardants in Human Milk and ThyroidStimulating
Hormone (TSH) in Neonates. Environ. Res. 2011, 111
(6), 737−743.
(150) Glynn, A.; Lignell, S.; Darnerud, P. O.; Aune, M.; Halldin
Ankarberg, E.; Bergdahl, I. A.; Barregård, L.; Bensryd, I. Regional
Differences in Levels of Chlorinated and Brominated Pollutants in
Mother’s Milk from Primiparous Women in Sweden. Environ. Int.
2011, 37 (1), 71−79.
(151) Harrad, S.; Abdallah, M. A. E. Concentrations of
Polybrominated Diphenyl Ethers, Hexabromocyclododecanes and
Tetrabromobisphenol-A in Breast Milk from United Kingdom
Women Do Not Decrease over Twelve Months of Lactation. Environ.
Sci. Technol. 2015, 49 (23), 13899−13903.
(152) Hoopmann, M.; Albrecht, U. V.; Gierden, E.; Huppmann, R.;
Suchenwirth, R. Time Trends and Individual Characteristics
Associated with Polybrominated Diphenyl Ethers in Breast Milk
Samples 2006−2009 in Lower Saxony, Germany. Int. J. Hyg Environ.
Health 2012, 215 (3), 352−359.
(153) Lignell, S.; Aune, M.; Darnerud, P. O.; Cnattingius, S.; Glynn,
A. Persistent Organochlorine and Organobromine Compounds in
Mother’s Milk from Sweden 1996−2006: Compound-Specific
Temporal Trends. Environ. Res. 2009, 109 (6), 760−767.
(154) Polder, A.; Thomsen, C.; Lindström, G.; Løken, K. B.; Skaare,
J. U. Levels and Temporal Trends of Chlorinated Pesticides,
Polychlorinated Biphenyls and Brominated Flame Retardants in
Individual Human Breast Milk Samples from Northern and Southern
Norway. Chemosphere 2008, 73 (1), 14−23.
(155) Polder, A.; Gabrielsen, G. W.; Odland, J.; Savinova, T. N.;
Tkachev, A.; Løken, K. B.; Skaare, J. U. Spatial and Temporal
Changes of Chlorinated Pesticides, PCBs, Dioxins (PCDDs/PCDFs)
and Brominated Flame Retardants in Human Breast Milk from
Northern Russia. Sci. Total Environ. 2008, 391 (1), 41−54.
(156) Pratt, I.; Anderson, W.; Crowley, D.; Daly, S.; Evans, R.;
Fernandes, A.; Fitzgerald, M.; Geary, M.; Keane, D.; Morrison, J. J.;
Reilly, A.; Tlustos, C. Brominated and Fluorinated Organic Pollutants
in the Breast Milk of First-Time Irish Mothers: Is There a
Relationship to Levels in Food? Food Addit Contam Part A Chem.
Anal Control Expo Risk Assess 2013, 30 (10), 1788−1798.
(157) Thomsen, C.; Liane, V. H.; Becher, G. Automated Solid-Phase
Extraction for the Determination of Polybrominated Diphenyl Ethers
and Polychlorinated Biphenyls in Serum-Application on Archived
Norwegian Samples from 1977 to 2003. J. Chromatogr B Analyt
Technol. Biomed Life Sci. 2007, 846 (1−2), 252−263.
(158) UNEP. Global monitoring report for Persistent Organic
Pollutants. Third regional monitoring report Western Europe and
Others Group (WEOG) region. https://chm.pops.int/
implementation/globalmonitoringplan/monitoringreports/tabid/
525/default.aspx (accessed 2023-03-30).
(159) UNEP. Global monitoring report for Persistent Organic
Pollutants. Third regional monitoring report Central and Eastern
European region. 2021. http://chm.pops.int/implementation/
globalmonitoringplan/monitoringreports/tabid/525/default.aspx (accessed
2023-03-30).
(160) Čechová, E.; Vojta, Š.; Kukučka, P.; Kočan, A.; Trnovec, T.;
Murínová, L. P.; de Cock, M.; van de Bor, M.; Askevold, J.; Eggesbø,
M.; Scheringer, M. Legacy and Alternative Halogenated Flame
Retardants in Human Milk in Europe: Implications for Children’s
Health. Environ. Int. 2017, 108, 137−145.
(161) Fürst, P. Dioxins, Polychlorinated Biphenyls and Other
Organohalogen Compounds in Human Milk. Mol. Nutr Food Res.
2006, 50 (10), 922−933.
(162) Strandman, T.; Koistinen, J.; Vartiainen, T. Polybrominated
Flame Retardants (PBDEs) in Placenta and Human Milk. Organohalogen
Compd. 2000, 47, 61−64.
(163) AMAP Assessment 2002 : The Influence of Global Change on
Contaminant Pathways to within, and from the Arctic; Arctic
Monitoring and Assessment Programme, 2003.
(164) Zeilmaker, M.; Moermond, C.; Brandon, E.; Hoogerhuis, P.;
Razenberg, L.; Janssen, M. Persistent Organic Pollutants in Human Milk
RIVM Letter Report 2020−0025; Rijksinstituut voor Volksgezondheid
en Milieu RIVM, 2020, DOI: 10.21945/RIVM-2020-0025.
(165) Weber, H.; Heseker, H. Analysis of Polybrominated Diphenyl
Ethers in Breast Milk of German Mothers - Results of a Pilot Study.
Fresenius Environmental Bulletin and Advances in Food Sciences 2004,
13 (4), 356−360.
(166) Thomsen, C.; Haug, L. S.; Stigum, H.; Frøshaug, M.;
Broadwell, S. L.; Becher, G. Changes in Concentrations of
Perfluorinated Compounds, Polybrominated Diphenyl Ethers, and
Polychlorinated Biphenyls in Norwegian Breast-Milk during Twelve
Months of Lactation. Environ. Sci. Technol. 2010, 44, 9550−9556.
(167) Sahlström, L. M. O.; Sellström, U.; de Wit, C. A.; Lignell, S.;
Darnerud, P. O. Estimated Intakes of Brominated Flame Retardants
via Diet and Dust Compared to Internal Concentrations in a Swedish
Mother-Toddler Cohort. Int. J. Hyg Environ. Health 2015, 218 (4),
422−432.
(168) Raab, U.; Preiss, U.; Albrecht, M.; Shahin, N.; Parlar, H.;
Fromme, H. Concentrations of Polybrominated Diphenyl Ethers,
Organochlorine Compounds and Nitro Musks in Mother’s Milk from
Germany (Bavaria). Chemosphere 2008, 72 (1), 87−94.
(169) Ozcan, S.; Tor, A.; Aydin, M. E. Levels of Organohalogenated
Pollutants in Human Milk Samples from Konya City, Turkey. Clean
(Weinheim) 2011, 39 (10), 978−983.
(170) Meironyté Guvenius, D.; Aronsson, A.; Ekman-Ordeberg, G.;
Bergman, Å.; Norén, K. Human Prenatal and Postnatal Exposure to
Polybrominated Diphenyl Ethers, Polychlorinated Biphenyls, Polychlorobiphenylols,
and Pentachlorophenol. Environ. Health Perspect
2003, 111 (9), 1235−1241.
(171) Main, K. M.; Kiviranta, H.; Virtanen, H. E.; Sundqvist, E.;
Tuomisto, J. T.; Tuomisto, J.; Vartiainen, T.; Skakkebæk, N. E.;
Toppari, J. Flame Retardants in Placenta and Breast Milk and
Environmental Science & Technology pubs.acs.org/est Policy Analysis
https://doi.org/10.1021/acs.est.3c02896
Environ. Sci. Technol. 2023, 57, 19106−19124
19121
Cryptorchildism in Newborn Boys. Environ. Health Perspect 2007, 115
(10), 1519−1526.
(172) Lignell, S.; Aune, M.; Darnerud, P. O.; Stridsberg, M.;
Hanberg, A.; Larsson, S. C; Glynn, A. Maternal Body Burdens of
PCDD/Fs and PBDEs Are Associated with Maternal Serum Levels of
Thyroid Hormones in Early Pregnancy: A Cross-Sectional Study.
Environmental Health 2016, 15, 1−11.
(173) Li, Z.-M.; Albrecht, M.; Fromme, H.; Schramm, K.-W.; De
Angelis, M. Persistent Organic Pollutants in Human Breast Milk and
Associations with Maternal Thyroid Hormone Homeostasis. Environ.
Sci. Technol. 2020, 54, 1111−1119.
(174) Lacorte, S.; Ikonomou, M. G. Occurrence and Congener
Specific Profiles of Polybrominated Diphenyl Ethers and Their
Hydroxylated and Methoxylated Derivatives in Breast Milk from
Catalonia. Chemosphere 2009, 74 (3), 412−420.
(175) Kazda, R.; Hajšlová, J.; Poustka, J.; Čajka, T. Determination of
Polybrominated Diphenyl Ethers in Human Milk Samples in the
Czech Republic: Comparative Study of Negative Chemical Ionisation
Mass Spectrometry and Time-of-Flight High-Resolution Mass
Spectrometry. Anal. Chim. Acta 2004, 520 (1−2), 237−243.
(176) Kalantzi, O. I.; Martin, F. L.; Thomas, G. O.; Alcock, R. E.;
Tang, H. R.; Drury, S. C.; Carmichael, P. L.; Nicholson, J. K.; Jones,
K. C. Different Levels of Polybrominated Diphenyl Ethers (PBDEs)
and Chlorinated Compounds in Breast Milk from Two U.K. Regions.
Environ. Health Perspect 2004, 112 (10), 1085−1091.
(177) Jaraczewska, K.; Lulek, J.; Covaci, A.; Voorspoels, S.; KalubaSkotarczak,
A.; Drews, K.; Schepens, P. Distribution of Polychlorinated
Biphenyls, Organochlorine Pesticides and Polybrominated
Diphenyl Ethers in Human Umbilical Cord Serum, Maternal Serum
and Milk from Wielkopolska Region, Poland. Sci. Total Environ. 2006,
372 (1), 20−31.
(178) Jakobsson, K., Athanasiadou, M., Christiansson, A.,
Athanassiadis, I., Bergman, Å, Hagmar, L. Polybrominated diphenyl
ethers (PBDEs) in serum from Swedish men 1988−2002 : A longitudinal
study. Lunds universitet, 2005, https://www.diva-portal.org/smash/
record.jsf?pid=diva2%3A657721&dswid=-8348.
(179) Ingelido, A. M.; Ballard, T.; Dellatte, E.; di Domenico, A.;
Ferri, F.; Fulgenzi, A. R.; Herrmann, T.; Iacovella, N.; Miniero, R.;
Päpke, O.; Porpora, M. G.; Felip, E. De. Polychlorinated Biphenyls
(PCBs) and Polybrominated Diphenyl Ethers (PBDEs) in Milk from
Italian Women Living in Rome and Venice. Chemosphere 2007, 67
(9), S301−S306.
(180) Hernik, A.; Struciński, P.; Buckley, B. T.; Góralczyk, K.; Czaja,
K.; Korcz, W.; Matuszak, M.; Łyczewska, M.; Minorczyk, M.;
Liszewska, M.; Ludwicki, J. K. Relationship between Paired Cord
Blood and Milk POPs Levels as a Tool for Assessing Perinatal
Exposure, a Pilot Study. Human and Ecological Risk Assessment: An
International Journal 2016, 22 (7), 1456−1468, DOI: 10.1080/
10807039.2016.1185688.
(181) Hernik, A.; Goralczyk, K.; Strucinski, P.; Czaja, K.; Kucharska,
A.; Korcz, W.; Snopczynski, T.; Ludwicki, J. K. Polybrominated
Diphenyl Ethers, Polychlorinated Biphenyls and Organochlorine
Pesticides in Human Milok as Markers of Environmental Exposure
to These Compounds. Ann. Agric. Environ. Med. 2011, 18 (1), 113−
118.
(182) Gómara, B.; Herrero, L.; Pacepavicius, G.; Ohta, S.; Alaee, M.;
González, M. J. Occurrence of Co-Planar Polybrominated/Chlorinated
Biphenyls (PXBs), Polybrominated Diphenyl Ethers (PBDEs)
and Polychlorinated Biphenyls (PCBs) in Breast Milk of Women
from Spain. Chemosphere 2011, 83 (6), 799−805.
(183) Gómara, B.; Herrero, L.; Ramos, J. J.; Mateo, J. R.; Fernandez,
M. A.; Garcia, J. F.; Gonzalez, M. J. Distribution of Polybrominated
Diphenyl Ethers in Human Umbilical Cord Serum, Paternal Serum,
Maternal Serum, Placentas, and Breast Milk from Madrid Population,
Spain. Environ. Sci. Technol. 2007, 41 (20), 6961−6968.
(184) Fängström, B.; Strid, A.; Grandjean, P.; Weihe, P.; Bergman,
Å. Environmental Health: A Global Access Science Source A
Retrospective Study of PBDEs and PCBs in Human Milk from the
Faroe Islands. Environmental health 2005, 4, 1−9.
(185) Fängström, B.; Athanassiadis, I.; Odsjö, T.; Norøn, K.;
Bergman, Š. Temporal Trends of Polybrominated Diphenyl Ethers
and Hexabromocyclododecane in Milk from Stockholm Mothers,
1980−2004. Mol. Nutr. Food Res. 2008, 52, 187−193.
(186) Erdoǧrul, Ö.; Covaci, A.; Kurtul, N.; Schepens, P. Levels of
Organohalogenated Persistent Pollutants in Human Milk from
Kahramanmaraş Region, Turkey. Environ. Int. 2004, 30 (5), 659−666.
(187) Dimitriadou, L.; Malarvannan, G.; Covaci, A.; Iossifidou, E.;
Tzafettas, J.; Zournatzi-Koiou, V.; Kalantzi, O. I. Levels and Profiles of
Brominated and Chlorinated Contaminants in Human Breast Milk
from Thessaloniki, Greece. Sci. Total Environ. 2016, 539, 350−358.
(188) Croes, K.; Colles, A.; Koppen, G.; Govarts, E.; Bruckers, L.;
Van de Mieroop, E.; Nelen, V.; Covaci, A.; Dirtu, A. C.; Thomsen, C.;
Haug, L. S.; Becher, G.; Mampaey, M.; Schoeters, G.; Van Larebeke,
N.; Baeyens, W. Persistent Organic Pollutants (POPs) in Human
Milk: A Biomonitoring Study in Rural Areas of Flanders (Belgium).
Chemosphere 2012, 89 (8), 988−994.
(189) Colles, A.; Koppen, G.; Hanot, V.; Nelen, V.; Dewolf, M. C.;
Noël, E.; Malisch, R.; Kotz, A.; Kypke, K.; Biot, P.; Vinkx, C.;
Schoeters, G. Fourth WHO-Coordinated Survey of Human Milk for
Persistent Organic Pollutants (POPs): Belgian Results. Chemosphere
2008, 73 (6), 907−914.
(190) Chovancová, J.; Čonka, K.; Kočan, A.; Sejáková, Z. S. PCDD,
PCDF, PCB and PBDE Concentrations in Breast Milk of Mothers
Residing in Selected Areas of Slovakia. Chemosphere 2011, 83 (10),
1383−1390.
(191) Bramwell, L.; Fernandes, A.; Rose, M.; Harrad, S.; PlessMulloli,
T. PBDEs and PBBs in Human Serum and Breast Milk from
Cohabiting UK Couples. Chemosphere 2014, 116, 67−74.
(192) Bordajandi, L. R.; Abad, E.; González, M. J. Occurrence of
PCBs, PCDD/Fs, PBDEs and DDTs in Spanish Breast Milk:
Enantiomeric Fraction of Chiral PCBs. Chemosphere 2008, 70 (4),
567−575.
(193) Björklund, J. A.; Sellström, U.; de Wit, C. A.; Aune, M.;
Lignell, S.; Darnerud, P. O. Comparisons of Polybrominated Diphenyl
Ether and Hexabromocyclododecane Concentrations in Dust
Collected with Two Sampling Methods and Matched Breast Milk
Samples. Indoor Air 2012, 22 (4), 279−288.
(194) Bergman, Å.; Hovander, L.; Sundström, M.; Athanassiadis, I.;
Athanasiadou, M.; Sällsten, G.; Bignert, A.; Nyberg, E. Insamling Och
Kemisk Analys Av Miljöföroreningar i Svensk Modersmjölk; Resultat från
2008−2010. Rapport till Naturvårdsverket 2010−11−24 2008, 20.
(195) Athanasiadou, M.; Bergman, Å. Insamling Av Bröstmjölksprover
Från Stockholm, Göteborg, Lund Och Umeå Samt Analyser Av Insamlade
Bröstmjölksprover Resultat Från 2007 Års Arbete; Department of
Environmental Chemistry, Stockholm University: 2008.
(196) Alivernini, S.; Battistelli, C. L.; Turrio-Baldassarri, L. Human
Milk as a Vector and an Indicator of Exposure to PCBs and PBDEs:
Temporal Trend of Samples Collected in Rome. Bull. Environ.
Contam. Toxicol. 2011, 87, 21−25.
(197) Abdallah, M. A. E.; Harrad, S. Polybrominated Diphenyl
Ethers in UK Human Milk: Implications for Infant Exposure and
Relationship to External Exposure. Environ. Int. 2014, 63, 130−136.
(198) Desalegn, A. A.; Collet, B.; Iszatt, N.; Stigum, H.; Jensen, T.
K.; Jonker, L.; Besselink, H.; van der Burg, B.; Eggesbø, M. Aryl
Hydrocarbon Receptor Activity in Human Breast Milk and
Cryptorchidism: A Case-Control Study within the Prospective
Norwegian HUMIS Cohort. Environ. Res. 2022, 214, No. 113861.
(199) Fromme, H.; Fuchs, V.; Albrecht, M.; Aschenbrenner, B.;
Röhl, C.; Janitzki, N.; Herber-Jonat, S.; Wöckner, M.; Völkel, W.;
Flemmer, A. W.; Schober, W. Polychlorinated Dioxins and
Dibenzofurans (PCDD/F), Polybrominated Dioxins and Dibenzofurans
(PBDD/F), Polychlorinated Biphenyls (PCB), Polybrominated
Diphenyl Ethers (PBDE), and per- and Polyfluoroalkyl Substances
(PFAS) in German Breast Milk Samples (LUPE 8). Sci. Total Environ.
2022, 825, No. 154066.
(200) Rovira, J.; Martínez, M. Á.; Mari, M.; Cunha, S. C.; Fernandes,
J. O.; Marmelo, I.; Marques, A.; Haug, L. S.; Thomsen, C.; Nadal, M.;
Domingo, J. L.; Schuhmacher, M. Mixture of Environmental
Environmental Science & Technology pubs.acs.org/est Policy Analysis
https://doi.org/10.1021/acs.est.3c02896
Environ. Sci. Technol. 2023, 57, 19106−19124
19122
Pollutants in Breast Milk from a Spanish Cohort of Nursing Mothers.
Environ. Int. 2022, 166, No. 107375.
(201) Roosens, L.; D’Hollander, W.; Bervoets, L.; Reynders, H.; Van
Campenhout, K.; Cornelis, C.; Van Den Heuvel, R.; Koppen, G.;
Covaci, A. Brominated Flame Retardants and Perfluorinated
Chemicals, Two Groups of Persistent Contaminants in Belgian
Human Blood and Milk. Environ. Pollut. 2010, 158 (8), 2546−2552.
(202) Stoffmonographie Für 1,2,5,6,9,10-Hexabromcyclododecan
(HBCDD) - HBM-Werte Für HBCDD Im Fettanteil Der
Muttermilch Oder Des Blutplasmas. Bundesgesundheitsblatt Gesundheitsforschung
Gesundheitsschutz. 2015, 58, 889−907, .
(203) Lankova, D.; Lacina, O.; Pulkrabova, J.; Hajslova, J. The
Determination of Perfluoroalkyl Substances, Brominated Flame
Retardants and Their Metabolites in Human Breast Milk and Infant
Formula. Talanta 2013, 117, 318−325.
(204) Thomsen, C.; Stigum, H.; Frøshaug, M.; Broadwell, S. L.;
Becher, G.; Eggesbø, M. Determinants of Brominated Flame
Retardants in Breast Milk from a Large Scale Norwegian Study.
Environ. Int. 2010, 36 (1), 68−74.
(205) Butryn, D. M.; Gross, M. S.; Chi, L. H.; Schecter, A.; Olson, J.
R.; Aga, D. S. One-Shot” Analysis of Polybrominated Diphenyl Ethers
and Their Hydroxylated and Methoxylated Analogs in Human Breast
Milk and Serum Using Gas Chromatography-Tandem Mass
Spectrometry. Anal. Chim. Acta 2015, 892, 140−147.
(206) Butryn, D. M.; Chi, L. H.; Gross, M. S.; McGarrigle, B.;
Schecter, A.; Olson, J. R.; Aga, D. S. Retention of Polybrominated
Diphenyl Ethers and Hydroxylated Metabolites in Paired Human
Serum and Milk in Relation to CYP2B6 Genotype. J. Hazard Mater.
2020, 386, No. 121904.
(207) Daniels, J. L.; Pan, I. J.; Jones, R.; Anderson, S.; Patterson, D.
G.; Needham, L. L.; Sjödin, A. Individual Characteristics Associated
with PBDE Levels in U.S. Human Milk Samples. Environ. Health
Perspect 2010, 118 (1), 155−160.
(208) Dunn, R. L.; Huwe, J. K.; Carey, G. B. Biomonitoring
Polybrominated Diphenyl Ethers in Human Milk as a Function of
Environment, Dietary Intake, and Demographics in New Hampshire.
Chemosphere 2010, 80 (10), 1175−1182.
(209) Guo, W.; Holden, A.; Smith, S. C.; Gephart, R.; Petreas, M.;
Park, J. S. PBDE Levels in Breast Milk Are Decreasing in California.
Chemosphere 2016, 150, 505−513.
(210) Hartle, J. C.; Cohen, R. S.; Sakamoto, P.; Barr, D. B.;
Carmichael, S. L. Chemical Contaminants in Raw and Pasteurized
Human Milk. Journal of Human Lactation 2018, 34 (2), 340−349.
(211) Hoffman, K.; Adgent, M.; Goldman, B. D.; Sjödin, A.; Daniels,
J. L. Lactational Exposure to Polybrominated Diphenyl Ethers and Its
Relation to Social and Emotional Development among Toddlers.
Environ. Health Perspect 2012, 120 (10), 1438−1442.
(212) Hoffman, K.; Mendez, M.; Siega-Riz, A. M.; Herring, A. H.;
Sjödin, A.; Daniels, J. L. Lactational Exposure to Polybrominated
Diphenyl Ethers and Its Relation to Early Childhood Anthropometric
Measurements. Environ. Health Perspect 2016, 124 (10), 1656−1661.
(213) Johnson-Restrepo, B.; Addink, R.; Wong, C.; Arcaro, K.;
Kannan, K. Polybrominated Diphenyl Ethers and Organochlorine
Pesticides in Human Breast Milk from Massachusetts, USA. Journal of
Environmental Monitoring 2007, 9, 1205−1212.
(214) LaKind, J. S.; Berlin, C. M.; Sjödin, A.; Turner, W.; Wang, R.
Y.; Needham, L. L.; Paul, I. M.; Stokes, J. L.; Naiman, D. Q.;
Patterson, D. G. Do Human Milk Concentrations of Persistent
Organic Chemicals Really Decline during Lactation? Chemical
Concentrations during Lactation and Milk/Serum Partitioning.
Environ. Health Perspect 2009, 117 (10), 1625−1631.
(215) Marchitti, S. A.; Fenton, S. E.; Mendola, P.; Kenneke, J. F.;
Hines, E. P. Polybrominated Diphenyl Ethers in Human Milk and
Serum from the U.S. EPA MAMA Study: Modeled Predictions of
Infant Exposure and Considerations for Risk Assessment. Environ.
Health Perspect 2017, 125 (4), 706−713.
(216) Schecter, A.; Pavuk, M.; Päpke, O.; Ryan, J. J.; Birnbaum, L.;
Rosen, R. Polybrominated Diphenyl Ethers (PBDEs) in U.S. Mothers’
Milk. Environ. Health Perspect 2003, 111 (14), 1723−1729.
(217) She, J.; Holden, A.; Sharp, M.; Tanner, M.; Williams-Derry,
C.; Hooper, K. Polybrominated Diphenyl Ethers (PBDEs) and
Polychlorinated Biphenyls (PCBs) in Breast Milk from the Pacific
Northwest. Chemosphere 2007, 67 (9), S307.
(218) Sjödin, A.; Jones, R. S.; Focant, J. F.; Lapeza, C.; Wang, R. Y.;
McGahee, E. E.; Zhang, Y.; Turner, W. E.; Slazyk, B.; Needham, L. L.;
Patterson, D. G. Retrospective Time-Trend Study of Polybrominated
Diphenyl Ether and Polybrominated and Polychlorinated Biphenyl
Levels in Human Serum from the United States. Environ. Health
Perspect 2004, 112 (6), 654−658.
(219) Zhou, S. N.; Buchar, A.; Siddique, S.; Takser, L.;
Abdelouahab, N.; Zhu, J. Measurements of Selected Brominated
Flame Retardants in Nursing Women: Implications for Human
Exposure. Environ. Sci. Technol. 2014, 48, 8873−8880.
(220) Wu, N.; Herrmann, T.; Paepke, O.; Tickner, J.; Hale, R.;
Harvey, E.; La Guardia, M.; McClean, M. D.; Webster, T. F. Human
Exposure to PBDEs: Associations of PBDE Body Burdens with Food
Consumption and House Dust Concentrations. Environ. Sci. Technol.
2007, 41 (5), 1584−1589.
(221) Park, J.-S.; She, J.; Holden, A.; Sharp, M.; Gephart, R.;
Souders-Mason, G.; Zhang, V.; Chow, J.; Leslie, B.; Hooper, K. High
Postnatal Exposures to Polybrominated Diphenyl Ethers (PBDEs)
and Polychlorinated Biphenyls (PCBs) via Breast Milk in California:
Does BDE-209 Transfer to Breast Milk? Environ. Sci. Technol. 2011,
45, 4579−4585.
(222) Martínez, C.; Martínez Arroyo, A.; Barrientos Alemán, D.;
Gavilán García, A.; Caba, M.; Calderón Garcidueñas, A. L.; Mora, A.;
Zenteno, E. Persistent Organic Compounds in Human Milk and
Evaluation of the Effectiveness of the Stockholm Convention in
Mexico. Environmental Advances 2022, 8, No. 100190.
(223) Mannetje, A.; Coakley, J.; Bridgen, P.; Brooks, C.; Harrad, S.;
Smith, A. H.; Pearce, N.; Douwes, J. Current Concentrations,
Temporal Trends and Determinants of Persistent Organic Pollutants
in Breast Milk of New Zealand Women. Sci. Total Environ. 2013,
458−460, 399−407.
(224) Coakley, J. D.; Harrad, S. J.; Goosey, E.; Ali, N.; Dirtu, A. C.;
Van den Eede, N.; Covaci, A.; Douwes, J.; Mannetje, A.
Concentrations of Polybrominated Diphenyl Ethers in Matched
Samples of Indoor Dust and Breast Milk in New Zealand. Environ. Int.
2013, 59, 255−261.
(225) Chen, Y.; Wang, X.; Li, Y.; Toms, L. M. L.; Gallen, M.; Hearn,
L.; Aylward, L. L.; McLachlan, M. S.; Sly, P. D.; Mueller, J. F.
Persistent Organic Pollutants in Matched Breast Milk and Infant
Faeces Samples. Chemosphere 2015, 118 (1), 309−314.
(226) Toms, L. M. L.; Guerra, P.; Eljarrat, E.; Barceló, D.; Harden,
F. A.; Hobson, P.; Sjodin, A.; Ryan, E.; Mueller, J. F. Brominated
Flame Retardants in the Australian Population: 1993−2009. Chemosphere
2012, 89 (4), 398−403.
(227) Toms, L. M. L.; Hearn, L.; Kennedy, K.; Harden, F.; Bartkow,
M.; Temme, C.; Mueller, J. F. Concentrations of Polybrominated
Diphenyl Ethers (PBDEs) in Matched Samples of Human Milk, Dust
and Indoor Air. Environ. Int. 2009, 35 (6), 864−869.
(228) Toms, L. M. L.; Harden, F. A.; Symons, R. K.; Burniston, D.;
Fürst, P.; Müller, J. F. Polybrominated Diphenyl Ethers (PBDEs) in
Human Milk from Australia. Chemosphere 2007, 68 (5), 797−803.
(229) Hites, R. A. Break Point Analyses of Human or Environmental
Temporal Trends of POPs. Science of The Total Environment 2019,
664, 518−521.
(230) Heck, K. E.; Braveman, P.; Cubbin, C.; Chávez, G. F.; Kiely, J.
L. Socioeconomic Status and Breastfeeding Initiation among
California Mothers. Public Health Reports 2006, 121 (1), 51−59.
(231) Chao, H. A.; Chen, S. C. C.; Chang, C. M.; Koh, T. W.;
Chang-Chien, G. P.; Ouyang, E.; Lin, S. L.; Shy, C. G.; Chen, F. A.;
Chao, H. R. Concentrations of Polybrominated Diphenyl Ethers in
Breast Milk Correlated to Maternal Age, Education Level, and
Occupational Exposure. J. Hazard Mater. 2010, 175 (1−3), 492−500.
(232) Malisch, K.; Malisch, R.; Fiedler, H. Guidelines for
Organization, Sampling and Analysis of Human Milk on Persistent
Organic Pollutants. Chemicals Branch Division of Technology, Industry
Environmental Science & Technology pubs.acs.org/est Policy Analysis
https://doi.org/10.1021/acs.est.3c02896
Environ. Sci. Technol. 2023, 57, 19106−19124
19123
and Economics (DTIE) United Nations Environment Programme
(UNEP) 2017, 39.
(233) Hooper, K.; She, J.; Sharp, M.; Chow, J.; Jewell, N.; Gephart,
R.; Holden, A. Depuration of Polybrominated Diphenyl Ethers
(PBDEs) and Polychlorinated Biphenyls (PCBs) in Breast Milk from
California First-Time Mothers (Primiparae). Environ. Health Perspect
2007, 115 (9), 1271−1275.
(234) Li, L.; Hoang, C.; Arnot, J. A.; Wania, F. Clarifying Temporal
Trend Variability in Human Biomonitoring of Polybrominated
Diphenyl Ethers through Mechanistic Modeling. Environ. Sci. Technol.
2020, 54, 166−175.
(235) Drage, D. S.; Harden, F. A.; Jeffery, T.; Mueller, J. F.; Hobson,
P.; Toms, L. M. L. Human Biomonitoring in Australian Children:
Brominated Flame Retardants Decrease from 2006 to 2015. Environ.
Int. 2019, 122, 363−368.
(236) Wong, F.; Cousins, I. T.; MacLeod, M. Bounding
Uncertainties in Intrinsic Human Elimination Half-Lives and Intake
of Polybrominated Diphenyl Ethers in the North American
Population. Environ. Int. 2013, 59, 168−174.
(237) Trudel, D.; Scheringer, M.; von Goetz, N.; Hungerbuhler, K.
Total Consumer Exposure to Polybrominated Diphenyl Ethers in
North America and Europe. Environ. Sci. Technol. 2011, 45, 2391−
2397.
(238) McGrath, T. J.; Morrison, P. D.; Ball, A. S.; Clarke, B. O.
Concentrations of Legacy and Novel Brominated Flame Retardants in
Indoor Dust in Melbourne, Australia: An Assessment of Human
Exposure. Environ. Int. 2018, 113, 191−201.
(239) Pietroń, W. J.; Małagocki, P. Quantification of Polybrominated
Diphenyl Ethers (PBDEs) in Food. A Review. Talanta 2017,
167, 411−427.
(240) Stapleton, H. M. Instrumental Methods and Challenges in
Quantifying Polybrominated Diphenyl Ethers in Environmental
Extracts: A Review. Anal. Bioanal. Chem. 2006, 386, 807−817.
(241) United States Department of State. Stockholm Convention on
Persistent Organic Pollutants. https://www.state.gov/key-topics-
office-of-environmental-quality-and-transboundary-issues/stockholmconvention-on-persistent-organic-pollutants/
(accessed 2022-12-05).
(242) Undeman, E.; Johansson, J. Polybrominated diphenyl ethers
(PBDEs) in the Baltic Sea − Sources, transport routes and trends.
Helcom Baltic Sea Environment Proceedings 2020, 172, 1−22.
(243) The 16 New POPs. In Stockholm Convention on Persistent
Organic Pollutants (POPs), UN Environment; Stockholm Convention
Secretariat United Nations Environment, 2017; pp 1−25.
(244) UNEP. Eighth Meeting of the Conference of the Parties to the
Stockholm Convention. http://chm.pops.int/theconvention/
conferenceoftheparties/meetings/cop8/tabid/5309/default.aspx (accessed
2023-01-06).
(245) Thuresson, K.; Höglund, P.; Hagmar, L.; Sjödin, A.; Bergman,
Å.; Jakobsson, K. Apparent Half-Lives of Hepta- to Decabrominated
Diphenyl Ethers in Human Serum as Determined in Occupationally
Exposed Workers. Environ. Health Perspect 2006, 114 (2), 176−181.
(246) Kalachova, K.; Hradkova, P.; Lankova, D.; Hajslova, J.;
Pulkrabova, J. Occurrence of Brominated Flame Retardants in
Household and Car Dust from the Czech Republic. Sci. Total
Environ. 2012, 441, 182−193.
(247) Guerra, P.; Alaee, M.; Jiménez, B.; Pacepavicius, G.; Marvin,
C.; MacInnis, G.; Eljarrat, E.; Barceló, D.; Champoux, L.; Fernie, K.
Emerging and Historical Brominated Flame Retardants in Peregrine
Falcon (Falco Peregrinus) Eggs from Canada and Spain. Environ. Int.
2012, 40 (1), 179−186.
(248) Dodson, R. E.; Van Den Eede, N.; Covaci, A.; Perovich, L. J.;
Brody, J. G.; Rudel, R. A. Urinary Biomonitoring of Phosphate Flame
Retardants: Levels in California Adults and Recommendations for
Future Studies. Environ. Sci. Technol. 2014, 48 (23), 13625−13633.
(249) Wu, X.; Bennett, D. H.; Moran, R. E.; Sjödin, A.; Jones, R. S.;
Tancredi, D. J.; Tulve, N. S.; Clifton, M. S.; Colón, M.; Weathers, W.;
Hertz-Picciotto, I. Polybrominated Diphenyl Ether Serum Concentrations
in a Californian Population of Children, Their Parents, and
Older Adults: An Exposure Assessment Study. Environ. Health 2015,
14 (1), DOI: 10.1186/s12940-015-0002-2.
(250) Parry, E.; Zota, A. R.; Park, J. S.; Woodruff, T. J.
Polybrominated Diphenyl Ethers (PBDEs) and Hydroxylated PBDE
Metabolites (OH-PBDEs): A Six-Year Temporal Trend in Northern
California Pregnant Women. Chemosphere 2018, 195, 777−783.
(251) Sharkey, M.; Harrad, S.; Abou-Elwafa Abdallah, M.; Drage, D.
S.; Berresheim, H. Phasing-out of Legacy Brominated Flame
Retardants: The UNEP Stockholm Convention and Other Legislative
Action Worldwide. Environ. Int. 2020, 144, No. 106041.
(252) Environmental Protection Agency. Significant New Use and
Test Rules: Certain Polybrominated Diphenylethers. Federal Register
2012, 77 (63), 19862−19899.
(253) Environmental Protection Agency. Decabromodiphenyl Ether
(DecaBDE); Regulation of Persistent, Bioaccumulative, and Toxic
Chemicals Under TSCA Section 6(h). Federal Register 2021, 86 (3),
880−894.
(254) National Industrial Chemicals Notification and Assessment
Scheme. Interim Public Health Risk Assessment of Certain PBDE
Congeners 2007, 62.
(255) Li, L.; Wania, F. Elucidating the Variability in the
Hexabromocyclododecane Diastereomer Profile in the Global
Environment. Environ. Sci. Technol. 2018, 52 (18), 10532−10542.
(256) Ali, N.; Dirtu, A. C.; Eede, N. V. d.; Goosey, E.; Harrad, S.;
Neels, H.; Mannetje, A.; Coakley, J.; Douwes, J.; Covaci, A.
Occurrence of Alternative Flame Retardants in Indoor Dust from
New Zealand: Indoor Sources and Human Exposure Assessment.
Chemosphere 2012, 88 (11), 1276−1282.
(257) Hennebert, P. Concentrations of Brominated Flame
Retardants in Plastics of Electrical and Electronic Equipment,
Vehicles, Construction, Textiles and Non-Food Packaging: A Review
of Occurrence and Management. Detritus 2020, 12, 34−50.
(258) Drage, D. S.; Sharkey, M.; Abdallah, M. A.-E.; Berresheim, H.;
Harrad, S. Brominated Flame Retardants in Irish Waste Polymers:
Concentrations, Legislative Compliance, and Treatment Options. Sci.
Total Environ. 2018, 625, 1535−1543.
(259) Samsonek, J.; Puype, F. Occurrence of Brominated Flame
Retardants in Black Thermo Cups and Selected Kitchen Utensils
Purchased on the European Market. Food Additives & Contaminants:
Part A 2013, 30 (11), 1976−1986.
(260) Kuang, J.; Abdallah, M. A.-E.; Harrad, S. Brominated Flame
Retardants in Black Plastic Kitchen Utensils: Concentrations and
Human Exposure Implications. Sci. Total Environ. 2018, 610−611,
1138−1146.
(261) Ionas, A. C.; Dirtu, A. C.; Anthonissen, T.; Neels, H.; Covaci,
A. Downsides of the Recycling Process: Harmful Organic Chemicals
in Children’s Toys. Environ. Int. 2014, 65, 54−62.
(262) Fatunsin, O. T.; Oluseyi, T. O.; Drage, D.; Abdallah, M. A.-E.;
Turner, A.; Harrad, S. Children’s Exposure to Hazardous Brominated
Flame Retardants in Plastic Toys. Sci. Total Environ. 2020, 720,
No. 137623.
(263) Zhang, B.; Lu, S.; Huang, M.; Zhou, M.; Zhou, Z.; Zheng, H.;
Jiang, Y.; Bai, X.; Zhang, T. Urinary Metabolites of Organophosphate
Flame Retardants in 0−5-Year-Old Children: Potential Exposure Risk
for Inpatients and Home-Stay Infants. Environ. Pollut. 2018, 243,
318−325.
Environmental Science & Technology pubs.acs.org/est Policy Analysis
https://doi.org/10.1021/acs.est.3c02896
Environ. Sci. Technol. 2023, 57, 19106−19124
19124
Melymuk, Lisa, Jonathan Blumenthal, Ondřej Sáňka, Adriana Shu-Yin, Veena Singla,
Kateřina Sebková, Kristi Pullen Fedinick, and Miriam L Diamond. 2022. "Persistent
Problem: Global Challenges to Managing PCBs." Environmental Science & Technology
2022: 9040. https://doi.org/10.1021/acs.est.2c01204
Persistent Problem: Global Challenges to Managing PCBs
Lisa Melymuk, Jonathan Blumenthal, Ondřej Sáňka, Adriana Shu-Yin, Veena Singla, Kateřina Šebková,
Kristi Pullen Fedinick, and Miriam L. Diamond*
Cite This: Environ. Sci. Technol. 2022, 56, 9029−9040 Read Online
ACCESS Metrics & More Article Recommendations *sı Supporting Information
ABSTRACT: Polychlorinated biphenyls (PCBs), “famous” as
persistent organic pollutants (POPs), have been managed
nationally since the 1970s and globally under the Stockholm
Convention on POPs since 2004, requiring environmentally sound
management (ESM) of PCBs by 2028. At most, 30% of countries
are on track to achieve ESM by 2028. Globally over 10 million
tonnes of PCB-containing materials remain, mostly in countries
lacking the ability to manage PCB waste. Canada (Ontario) and
Czechia, both parties to the Stockholm Convention, are close to
achieving the 2028 goal, having reduced their stocks of pure PCBs
by 99% in the past 10 years. In contrast, the USA, not a party to
the Stockholm Convention, continues to have a substantial but
poorly inventoried stock of PCBs and only ∼3% decrease in mass
of PCBs since 2006. PCB management, which depends on Stockholm Convention support and national compliance, portends major
challenges for POP management. The failure to manage global PCB stocks >30 years after the end of production highlights the
urgent need to prioritize reducing production and use of newer, more widely distributed POPs such as chlorinated paraffins and perand
polyfluorinated alkyl substances, as these management challenges are unlikely to be resolved in the coming decades.
KEYWORDS: polychlorinated biphenyls, Stockholm Convention, chemicals management, persistent organic pollutants, PCB stocks,
environmentally sound management, Canada, Czechia, USA
■ INTRODUCTION
Polychlorinated biphenyls (PCBs) are the epitome of a
persistent organic pollutant (POP) because of their persistence,
bioaccumulative potential, and toxicity. Owing to their
environmental mobility and persistence, they are distributed
globally, from the high Arctic and Antarctic to the Mariana
Trench in the deep Pacific Ocean.1
PCBs pose risks to
ecosystems as they potently bioaccumulate through the food
web to reach levels of concern among top trophic level animals.
In utero exposures are associated with neurodevelopmental
toxicity, manifesting as learning, behavioral, or intellectual
impairment in children.2−6
PCB exposures are also associated
with impaired immunological function, auditory deficits, and
central nervous system disorders such as Parkinson-like
symptoms.7−9
PCBs were introduced for use in dielectric fluids to reduce
the risk of explosion in capacitors and transformers and saw
widespread use as plasticizers and flame retardants in products
such as building materials and paints.10
Breivik et al.11
estimated that more than 1.3 million tonnes of pure PCBs
were manufactured between 1930 and 1993 in at least 10
countries, primarily in the USA, followed by West Germany,
the USSR, and France (Figure S1). However, PCBs were
widely exported from manufacturing countries, resulting in use
in at least 114 countries.11
The ∼1.3 million tonnes of pure
PCBs, through dilution for use and subsequent poor
management, expanded to 17 million tonnes of PCBcontaminated
materials and waste (Figure 1), with an
estimated 20−35% of PCBs already released in the environ-
ment.17
Monsanto, in the USA, produced more than 50% of
global PCBs and recognized the toxicity of PCBs shortly after
the start of large-scale commercial production in the 1930s.13
According to Monsanto’s documentation, the company argued
that PCBs were of minimal risk in closed systems such as
capacitors and transformers, but the company intended to use
PCBs in a very wide array of products that would result in
environmental release and human exposure.12
Clear evidence
of their widespread environmental distribution came in the
1960s and 1970s, first in Baltic Sea biota,13
along with
indications of their toxicity.14
Global production of PCBs decreased with the introduction
of restrictions in the 1970s in Western countries (Figures 1 and
Received: February 17, 2022
Revised: May 4, 2022
Accepted: May 4, 2022
Published: June 1, 2022
Articlepubs.acs.org/est
© 2022 The Authors. Published by
American Chemical Society
9029
https://doi.org/10.1021/acs.est.2c01204
Environ. Sci. Technol. 2022, 56, 9029−9040
DownloadedviaMASARYKUNIVonOctober8,2024at10:01:20(UTC).
Seehttps://pubs.acs.org/sharingguidelinesforoptionsonhowtolegitimatelysharepublishedarticles.
S2). These were followed by restrictions implemented through
international agreements, notably the Stockholm Convention
on POPs, which entered into force in 2004 and currently has
185 parties (184 countries plus the European Union,
chm.pops.int). The Stockholm Convention bans the production
of PCBs and aims to phase out in-use PCBs by 2025 and
ensure environmentally sound management (ESM) of
materials with >0.005% (50 mg/kg) PCB content by 2028.
ESM largely constitutes chemical destruction by high-temperature
combustion methods where the PCB content of the
waste is destroyed, with exceptions for low PCB content wastes
with large volumes, in which case specially engineered landfills
or permanent storage in underground mines/rock formations
can serve as reasonable substitutes (Table S1).15
Each party
implements the Stockholm Convention through the enactment
of binding legislation. Parties have devoted considerable efforts
over decades to eliminating PCB stocks and controlling further
primary and major secondary releases. Much has been
accomplished. However, as of 2016, UNEP identified that
only 17% of PCB-containing materials have been eliminated, at
the rate of about 200,000 t/y since 2000. Addressing the
remaining 83% would require the elimination of ∼1 million
tonnes of PCB-containing oils and contaminated equipment
per year to reach the 2028 target.16
Today, despite restrictions, primary PCB emissions continue
from on-going use in products and materials: closed
applications (e.g., transformers, capacitors, electric motors,
and light ballasts), partially open applications (e.g., hydraulic
fluid, heat transfer fluid, switches, electrical cables, and vacuum
pumps), and open applications (e.g., paints, sealants, inks,
lubricants, flame retardants, insulation, dyes, and pesticides)
(see Table S2).21,22
This diversity of uses, combined with their
poor documentation, creates a global challenge for managing
primary PCB sources.
Although the greatest mass of PCB use has been in closed
applications, open applications of PCBs have received
increasing attention, particularly in relation to sensitive indoor
environments such as schools.23−25
Open applications have
been shown to result in direct exposure, particularly in schools,
and secondary exposure from emissions to the surrounding
environment; open applications were estimated to be the
primary contributors to global emissions up to 1980.29
Building materials have received the most attention, notably
joint sealants and paints, but >15 types of open applications
have been identified by UNEP,26
and previously undocumented
open source uses (e.g., floor waxes,27
book bindings28
)
continue to be identified. Open applications were the first to
receive international regulatory attention through OECD
restriction on open applications in 1973.30
While reports
frequently state that approximately 21−26% of PCBs were
used in open applications,26,29
this is a rough global average,
and the type and amount of open PCB use varied substantially
by region: in Japan, the majority of open use was in carbonless
copy paper, while in Western Europe, it was in building
sealants.26,29
Open applications of PCBs present a unique
challenge as they are not typically included in national
inventories and are frequently not even recognized as PCB
wastes.26
Countries that have estimated stocks of PCBs in
open applications (e.g., Germany, Finland, Norway, Sweden,
and Switzerland) have identified amounts from hundreds to
Figure 1. Timeline of major policy actions on PCBs in the USA, Canada, and Czechia and in the Stockholm Convention. Rate of PCB production
and associated increase in stocks is estimated based on the study by Breivik et al.11
The Stockholm Convention estimates of remaining stocks are
obtained from refs 17−181920. The total stock of PCB-containing materials and waste (up to 20 million t) reflects how pure PCBs are diluted to
create this mass and how mismanagement spreads pure PCBs to create a larger contaminated mass. The variation in the estimated stock over
1994−2016 reflects how reported global inventories changed over time due to uncertainties in reported PCB stocks. Additional details on the
timeline of policies impacting PCBs are given in Figure S2.
Environmental Science & Technology pubs.acs.org/est Article
https://doi.org/10.1021/acs.est.2c01204
Environ. Sci. Technol. 2022, 56, 9029−9040
9030
thousands of tonnes.26
For example, Germany had an
estimated 12,000 t remaining in open applications as of
2013, contributing 7−12 t of PCBs to the environment
annually.31
Global PCB contamination has many facets, from the legacy
use of PCBs in industrial and consumer products, to
unintentional production and releases from industries and
combustion32,33
and to emissions of unintentionally produced
nonlegacy PCB congeners from modern materials such as
paints and cabinetry.34,35
Here, we focus on the challenge of
managing primary legacy PCB stocks through a review and
analysis of current PCB management status globally. We assess
progress toward ESM of PCBs through a two-part analysis: (1)
challenges of PCB management on a global scale and (2) a
detailed case study comparison of the current status of PCB
management in three economically developed countries
(Canada, Czechia, and USA) with differing histories of
Stockholm Convention participation, PCB production and
use, and management capabilities.
The motivation for this analysis is two-fold. First, it is critical
to understand the scale of the future threat posed by PCBs to
human and ecosystem health. After countries enacted controls
in the 1970s and early 1980s, concentrations in air, water, and
relatively short-lived biota dropped rapidly.36
However,
decreases have slowed in recent years.37,38
Today, 40 years
after major production ceased, PCBs may cause the demise of
over 50% of the world’s killer whale populations.39
As a
neurotoxicant, PCBs contribute to the significant global
burden of disease attributable to widespread human exposure
to hazardous pollutants.40
Second, the analysis of successes and
failures in managing PCBs provides a clear cautionary lesson
on the long-term impacts of producing and widely using
persistent compounds and the inability of even wealthy
countries to manage and eliminate their on-going use.
■ METHODS
Global PCB Management. Historical PCB consumption
was adapted from Breivik et al.11
to reflect the total PCB mass
used (see Text S1). The status of PCB use and management of
all UN-registered countries was classified based on the most
recent information (Table S3). In most cases, this information
was the most recent Stockholm Convention status document
(National Implementation Plan or Conference of the Parties
update) and the responses to a 2018 questionnaire given to
Stockholm Convention parties on PCB management. Where
available, other reports were also used, particularly Global
Environment Facility (GEF) project reports (see references in
Table S3). Based on this information, we placed countries into
eight categories according to their PCB management, ranging
from no existing PCB management plan or inventory, to full
ESM. We also included countries that are not parties to the
Stockholm Convention.
Case Study: Canada. Data from Ontario, containing
nearly 40% of the Canadian population,41
were used as an
indicator of Canadian performance. Details of the Ontario
PCB inventory are provided in the Supporting Information
(Text S2). Briefly, the PCB stock in Ontario was estimated by
combining data from the Canadian federal “ePCB” database
and a provincial-level PCB Waste database maintained by the
Ontario Ministry of the Environment, Conservation and Parks.
The federal ePCB reporting system lists locations of PCB
holdings in-use and in-storage with concentrations >50 mg/kg.
Data as of December 31, 2016 for closed sources were used.
The Ontario database separately lists PCB waste storage sites
in Ontario; data used were as of 2013. Combined, the
databases reported a total of 270 unique sites with PCBs in
use, stored, or classified as wastes not yet subject to ESM.
Any values given in units of volume were converted to mass
based on assumptions of the density of the PCB-containing
material. The masses of PCB-containing materials were then
converted to estimates of pure PCB mass assuming average
concentrations per category of application.42
Askarel fluid was
assumed to have 600,000 mg/kg PCBs (range 400,000−
800,000 mg/kg) and mineral oil 250 mg/kg PCBs (range 50−
500 mg/kg).43,44
Concentrations in waste categories were
estimated based on their classification as either low- or highlevel
waste. Assumptions are detailed in Table S4.
We also compared these 2013−2016 data with an inventory
of closed sources from 2006 before the enactment of revised
Canadian Federal PCB regulations in 2008. This comparison
was restricted to the Toronto (largest city in Canada) area
because of data availability.42
Case Study: Czechia. Data for Czechia were compiled
from the current inventory of PCB-containing products and
materials, maintained by the CENIAthe Czech Environmental
Information Agency. Since the itemized database is not
publicly available, our inventory relied on the totals of
individual application categories, which have been reported
to the Stockholm Convention45
and the European Union.46
Where available, exact reported masses of PCB-contaminated
material were used, and when missing, the mass was
estimated based on the number of items and the median mass
of PCB-containing fluid per item in the category (Table S5).
The masses of PCB-containing fluids were converted to
estimates of pure PCB mass using assumptions specific to
Czech/EU regulations (Table S5). Czechia complies with the
European Council Directive 96/59/EC, which required
elimination of all materials containing >500 mg/kg PCBs by
2010; thus, all remaining large PCB equipment (>5 dm3
)
should be <500 ppm PCBs. However, there is some ambiguity
in the reported information; thus, we have used a 10,000 ppm
upper threshold, based on Czech reporting to UNEP,45
to
account for a worst-case scenario of instances of non-
compliance.47
To evaluate progress since the ratification of the Stockholm
Convention, we compared the most recent inventory with
totals from the initial Czech National Implementation Plan
reflecting 2002−2004.48
As the original inventory preceded EU
legislation, we assumed higher concentrations of PCBs in
materials using the values of 20,000 mg/kg (range 10,000−
30,000 mg/kg).47
Case Study: USA. To estimate the current stock of PCBs
in use and waste in USA, we utilized publicly available
information from the US Environmental Protection Agency
(US EPA), specifically the PCB transformer registration
database49
(Table S6, as of Jan 2020) and the PCB Cleanup
and Disposal Program50
(up to 2020). The US EPA tracks
transformers and regulated PCB waste pursuant to regulations
under the Toxic Substances Control Act (TSCA). Owners of
PCB transformers must register details on transformer
location, ownership, and mass in the PCB transformer
registration database. PCB transformers that are removed
from use may be optionally deregistered from the database. We
extracted information on the number, location, and mass of
transformers to estimate the stock of PCBs currently held in inuse
and stored transformers. Incomplete records were assigned
Environmental Science & Technology pubs.acs.org/est Article
https://doi.org/10.1021/acs.est.2c01204
Environ. Sci. Technol. 2022, 56, 9029−9040
9031
the median value of the complete records (median two
transformers per site; median mass 782 kg/transformer). A
transformer was considered to be 30% fluid by mass;51
this
factor was used to convert total transformer masses to masses
of PCB-containing fluids. Fluid masses were converted to
estimates of pure PCB mass using assumptions about the
concentrations in a typical transformer, as for the Canadian
inventory. We assumed that before Jan 1, 2000, all transformers
reported were askarel transformers, and after that date,
all were mineral oil, based on the assumption that transformer
owners would be aware of high-level PCB content and report
in compliance with the TSCA regulations. However, PCBcontaminated
mineral oil may be less thoroughly documented,
leading to delayed reporting.52
Average concentrations used for
askarel transformers were 600,000 mg/kg PCBs (range
400,000−800,000 mg/kg) and those for mineral oil transformers
were 1000 mg/kg PCBs (range 500−5000 mg/
kg).43,44,52
To evaluate the completeness of the transformer
registration database, we compared the total number and mass
of transformers reported to the PCB Cleanup and Disposal
Figure 2. Global PCB use and management. (A) Total PCB consumption by country throughout the period 1930−2000 based on data from
Breivik et al.11
extrapolated to total PCB mass consumption according to Text S1. Figure S6 presents the same data presented as per capita
consumption. (B) Current status of PCB management according to the latest reported status for each country, compiled from Stockholm
Convention reporting and other sources. Sources used are given in Table S3. NIP indicates Stockholm Convention National Implementation Plan.
A color-blind accessible version of this figure can be found in Figure S5.
Environmental Science & Technology pubs.acs.org/est Article
https://doi.org/10.1021/acs.est.2c01204
Environ. Sci. Technol. 2022, 56, 9029−9040
9032
Program with the total number of transformers in the
registration database and the mass of deregistered transformers
from 1998 to 2018.
To evaluate changes in the USA stock over time, we
compared the current remaining stock of transformers to
records of PCB transformer deregistrations to calculate the
number of transformers and mass of PCBs that would have
remained in use/stored in 2006.
Uncertainty Analyses. Monte Carlo analysis was
performed for the Canadian PCB databases, which identified
that concentration levels (i.e., mg/kg PCBs per equipment/
waste category), particularly the concentration selected for the
high-level equipment/wastes, had the greatest impact on the
estimate of the total stock (Text S3, Figures S3, S4). Density
and other assumptions made to complete missing data had
negligible impact. Therefore, we addressed the uncertainties in
the conversion of database entries to pure PCB masses through
the inclusion of upper and lower concentration thresholds for
individual PCB equipment and waste categories according to
the regulations and database thresholds for each country, as
described above.
■ RESULTS
Global Management of PCB Stocks. Global use of PCBs
varies widely, and the current PCB management capacities of
countries also vary widely (Figure 2B, Table S3). One country
(Democratic People’s Republic of Korea) continues to
produce PCBs.17
185 parties (184 countries plus the European
Union) have ratified the Stockholm Convention, while 13
countries have not, notably the USA, Italy, Malaysia, Haiti,
Israel, and Turkmenistan; all other nonratifying countries have
<1 million population or are recently established (e.g., South
Sudan). We note that while Italy has not ratified the
Stockholm Convention, the European Union, of which Italy
is a member, is a party to the Convention, and the EU has
stricter regulations on PCB management than the Stockholm
Convention. Of the 184 ratifying parties, 10 have not
submitted any implementation plan. Greece and Malta, as
EU member states, should also follow EU PCB management
regulations, despite not yet submitting documentation to the
Stockholm Convention.
For the 174 Parties that have submitted reports, our analysis
highlights that 72 national PCB inventories (42%) are partial
or preliminary. Many inventories are limited to transformers
and/or only to the public electricity sector, which may capture
only half of the uses of PCBs (considering 48% of the PCBs
produced were used in transformers22
). An additional 23
countries (13%) reported complete PCB inventories but no
capacity to achieve ESM, while 11 countries had inventories
and capacity to manage PCBs but had made no significant
progress toward ESM. The number of countries achieving or
progressing toward ESM was small; 34 countries (18%) are
progressing toward ESM through removal from use and
destruction of PCB materials. Only 23 countries (13%) have
achieved ESM of PCBs. With three exceptions (Nepal, Kenya,
and Micronesia), all of the 23 countries that have achieved
ESM are classified as “very high” in the UNDP Human
Development Index, or “high income” under by the World
Bank (Table S3). Only three countries classified as “low
development”/“low income” are currently making substantial
progress toward ESM of PCBs: Benin, Rwanda, and Uganda.
National reporting to the Stockholm Convention contains a
wealth of information about PCB management, but the quality
and quantity of information provided by individual countries
vary widely. A questionnaire from the Convention was
distributed to 182 parties to evaluate progress toward ESM
of PCBs. Fewer than 60 parties provided responses.53
Based on
these responses and additional Stockholm Convention
information sources, estimates of current PCB stocks could
be determined for 52 countries (see references in Table S3).
Many countries lacked recent information, with some countries
not submitting documentation since 2004. Therefore, our
analysis is uncertain for two reasons. First, we may be
presenting a worst-case scenario since countries may have
made unreported progress toward PCB elimination. Second,
and conversely, many countries reported incomplete inventories
(e.g., only inventories of transformers owned by a national
electricity provider), and thus most recent reports underrecord
true stocks.
Many countries are challenged by weak institutions,
corruption, and mismanagement, making tracking PCB stocks
and limiting their misuse extremely difficult. There are multiple
reports of transformers being improperly recycled. For
example, Sri Lanka identified that PCB-containing transformers
were transferred to informal recyclers and that spilled
PCB-containing oils were cleaned up with sawdust of which
some were disposed of through burning.54
Nauru reported that
a transformer confirmed to contain PCBs was slated to be
shipped to Australia but was instead collected by a scrap metal
recycling company with an undetermined fate.55
In other
instances, transformer owners were reported to have actively
drained and disposed of PCB contents to avoid responsibility
for PCB materials. For example, in Malawi, numerous pieces of
equipment suspected to contain PCBs had their contents
poured directly onto the ground before they could be tested.56
The Dominican Republic reported that owners of transformer
shops, to avoid PCB disposal regulations, diluted the PCB
concentration in the fluids by continually removing them and
adding more mineral oils and sold the removed PCB oils to
illegal foundries.57
Ghana reported the use of PCB oils to
create beauty creams and to lubricate domestic sewing
machines,58
while in Montenegro, factory workers were
reported to have used PCB oils for handwashing and to heat
homes.59
Several initiatives, such as those funded by the Global
Environment Fund (GEF) and implemented by UNEP, have
made progress in addressing some of the challenges lowincome
countries face in managing PCBs. For example, a
project, harmonizing efforts in Southern Africa to centralize
dismantling, draining, and accumulation of PCB oils/equipment
for disposal,60
was among more than 40 GEF-funded
projects on ESM of PCBs. Together, these projects have
eliminated 23,000 tonnes of PCBs.17
However, even these
specific projects can be hindered by unreliable national reports,
delays in laboratory analysis of suspected PCB materials, and,
most crucially, incomplete inventories. The technological and
financial capacities required to eliminate PCBs are not available
in many regions. GEF-funded projects on PCBs have received
∼$450 million USD to support the elimination of 88,000
tonnes of PCB-containing materials and waste (23,000 tonnes
eliminated and 65,000 tonnes planned), averaging USD 5,000
per tonne of PCB waste eliminated. It is noted that these
project costs include items not directly related to elimination
(i.e., capacity building and education).17
Currently, the cost
burden of managing PCBs lies with national governments or
international agencies (e.g., national environment agencies,
Environmental Science & Technology pubs.acs.org/est Article
https://doi.org/10.1021/acs.est.2c01204
Environ. Sci. Technol. 2022, 56, 9029−9040
9033
UNEP, and GEF). Producer financial responsibility to date has
only been in the form of legal settlements. Funding PCB
elimination is clearly a problem, with the estimated need to
eliminate 1 million tonnes of PCB-containing materials and
waste per year to achieve Stockholm Convention compliance,
mostly in countries with minimal financial and/or technical
capacity. Yet despite the costs of elimination, there is also a
clear public health cost of inaction: toxic chemicals, including
PCBs, are neurotoxicants contributing to the “pandemic of
developmental neurotoxicity,” placing a significant burden on
societal resources.6,10
At most, 30% of countries are on track to achieve ESM by
2028, and the lack of capacities for PCB management is a
barrier to achieving this goal. With the specific examples of
Canada, Czechia, and the USA, we demonstrate successes and
barriers to ESM even in high-income/highly developed
countries.
Case Studies. Inventory Results: Canada. Based on the
combined provincial and federal databases as of 2013 and
2016, we identified a total of 12,200 tonnes of material in
Ontario with PCB content >50 mg/kg. These materials were
estimated to contain 32 t of pure PCBs (Figure 3, Table S11)
with a range of 21−44 t. The PCB sites were widely distributed
across heavily populated Southern Ontario and within or close
to urban areas. The stock of PCB materials in Ontario was
dominated by PCB-contaminated soil/gravel held at two sites,
making up 97% of the bulk mass of PCB materials but <10% of
the pure PCBs, which was largely held in transformers (77%)
and capacitors (12%) (Figure 3G).
According to a 2006 PCB inventory,42
Toronto had 455
sites containing 850 t of PCB-containing equipment and
materials, equivalent to 424 t (range 282−565 t) of pure PCBs.
A large proportion of this was located in Toronto’s central
business district in electrical transformers in large office
Figure 3. Inventories of PCBs for USA, Czechia, and Ontario, Canada. (A−C) Estimated stocks as of mid 2010s by (A) number of items, (B) bulk
mass of PCB-containing materials, and (C) pure PCBs, with error bars indicating the uncertainty of the estimates of mass. (D−F) Change over
time in PCB-contaminated items, bulk mass of PCBs, and pure PCBs. (G,H) Distribution of current stocks of pure PCBs in Ontario and Czechia
according to item categories. USA is not shown because only transformers are included in the inventory.
Environmental Science & Technology pubs.acs.org/est Article
https://doi.org/10.1021/acs.est.2c01204
Environ. Sci. Technol. 2022, 56, 9029−9040
9034
towers.42
In 2016, Toronto had only 7 sites holding 31
separate PCB items, all located outside of the city center, of
which 3 sites had PCBs remaining in use (all others were
stored for disposal), constituting only ∼40 g of pure PCBs.
The remaining four sites stored 2.9 t of bulk PCB material,
equivalent to 0.8 t (range 0.6−1.0 t) of pure PCBs. Thus,
>99% of the 2006 stock of PCBs in Toronto was removed in
the 10 years between inventories and after 2008 legislation
mandating ESM, clearly showing progress toward elimination
of PCB stocks in Toronto (Figure 3F). This agrees with
Canada’s reporting to the Stockholm Convention61
which
states that 0.02% of the 2008 PCB stock remains today.
Inventory Results: Czechia. Based on the most recent
Czech records, we estimated a total stock of 6092 tonnes of
material with PCB content >50 mg/kg in 2016. This bulk mass
was estimated to contain 2.84 t (0.30−62 t) of pure PCBs
(Table S11). Most of the mass of PCB materials (74% of bulk
material and 79% of pure PCBs) was in transformers held by a
large electricity production and distribution company responsible
for more than 14,000 pieces of PCB-containing materials,
including close to 10,000 transformers (Figure 3A). However,
despite the large number of transformers, Czechia has
prioritized the removal of equipment containing high
concentrations of PCBs in compliance with European Union
and Stockholm Convention regulations. This has resulted in a
relatively low level of pure PCBs (∼2 tonnes) in these ∼10,000
transformers.
In 2005, 25,000 items contained ∼3000 t of PCBcontaminated
fluid/materials,48
equivalent to an estimated
460 tonnes of pure PCBs. An inventory in the intervening
years (from 2009) reported 9193 t of known PCB materials
and 3228 t of possible PCB materials.62
Later inventories
(2009, 2016) were more comprehensive, which resulted in a
higher reported bulk mass of PCBs (Figure 3E). The
prioritized removal of high concentration items has led to
the large decrease of pure PCBs since the 2005 inventory was
compiled (Figure 3F).48
Inventory Results: USA. The USA inventory was based only
on the US EPA transformer registration database49
(Table S6)
and, consequently, was incomplete in two respects. First, it
contained only records of transformersno inventories exist
in the USA for any other PCB-containing materials. Data from
the PCB Cleanup and Disposal program50
(Tables S7 and S8)
indicate the disposal of millions of kilograms of large low- and
high-voltage capacitors and bulk waste from 1998 to 2018,
none of which has been included in any inventory. Second, the
transformer registration database likely does not include all
PCB transformers. From 1998 to 2018, the registration
database listed 20,130 total transformers, but the PCB disposal
program data indicated that over 180,000 transformers were
disposed of over the same period, strongly suggesting that the
registration database did not, and likely still does not, include
all PCB-containing transformers. Therefore, our stock
estimates in Table S11 are a clear underestimate of closedsource
PCBs in USA.
As of 2020, the USA transformer database contained records
of 11,577 transformers, estimated to contain 13,755 tonnes of
PCB material with 776 (517−1040) tonnes of pure PCBs. To
provide a temporal comparison similar to Canada and Czechia,
the database was re-evaluated considering the stock of
transformers existing in 2006. In 2006, the USA had 14,457
transformers containing 47,500 tonnes of PCB material and
770 tonnes of pure PCBs. This suggests a 20% reduction in the
number of transformers and a 71% reduction in the mass of
bulk PCB material, but only a 3% reduction in pure PCBs over
∼15 years (Figure 3D−F).
■ DISCUSSION
Canada. PCBs were never manufactured in Canada, but an
estimated 40,000 t of PCBs were imported up to 1980,44
leading to the second-highest per capita use in the world (1.2
kg/person), behind only the USA (Figure S5). New uses of
PCBs were banned in 1977. Canada signed the Stockholm
Convention in 2001 and enacted legislation to comply with the
Convention in 2008. The purpose of the 2008 Canadian PCB
regulations was to accelerate the elimination of PCBs in
concentrations greater than 50 mg/kg by 2025 by stipulating
end-of-use deadlines, especially for PCBs near sensitive sites
(schools, daycares, hospitals, etc.).63
Of the three profiled countries, Canada has most successfully
managed PCBs. The mass of pure PCBs in Toronto, Canada’s
largest city, decreased by 3 orders of magnitude within 10
years, indicating that regulations were successful in phasing out
PCBs in Toronto. PCB material has been removed from all
transformers in large skyscrapers built in the 1960s and 1970s
in the downtown core and from sensitive sites such as
schools.42
The enforcement of PCB regulations in Canada
includes compliance strategies and environmental officers that
inspect PCB facilities. For example, in 2015, inspections were
conducted at 44 companies that were set to remove and
destroy their PCB equipment, finding 89% compliance.64
Some inventory data indicated a lack of compliance in a small
number of cases; three entries in Ontario databases reported
concentrations over 500 mg/kg for equipment that should
have been removed by 2009.
Canada’s database has only limited inclusion of open sources
of PCBs, although these were used in Ontario, for example, as
joint sealants in buildings constructed from the 1950s to
1970s.65
Past inventories estimated that the contributions of
PCB-containing building sealants were low relative to the total
amount of PCBs held in transformers,42,65
but with the
prioritized removal of closed PCB equipment and little
attention given to open sources, their relative importance
may now be greater. In countries that consider open sources in
their PCB inventories, they typically account for more than a
third of the remaining PCB bulk mass, for example, 37% in
Germany66
and 39% in Switzerland.67
Committing to national and international PCB agreements
has helped expedite Canada’s progress on phasing out PCBs.
The 2008 regulation was proposed to meet the targets and
commitments of the Stockholm Convention.68
If the current
trend continues across Canada, we suspect that Canada will be
well on track to meet the 2025 target set by the Stockholm
Convention.
Czechia. At the time of PCB production, Czechia was
Czechoslovakia, and PCBs were produced by the Chemko
Strážske factory in the east of the country (now Slovakia).
Chemko produced 21,481 tonnes of PCBs from 1959 to 1984,
of which 11,613 tonnes were used in Czechoslovakia, and the
rest was exported to other Eastern Bloc countries, primarily
East Germany.47
Per capita use in Czechoslovakia was 580 g/
person, the 20th highest in the world (Figure S6). It is
estimated that 7000−8500 t of pure PCBs were used in the
area that is now Czechia.48
Most PCB use was by three state
companies manufacturing PCB-containing paints and coatings,
electrical capacitors, and electrical equipment for transport and
Environmental Science & Technology pubs.acs.org/est Article
https://doi.org/10.1021/acs.est.2c01204
Environ. Sci. Technol. 2022, 56, 9029−9040
9035
industry.46
These materials were then distributed to over 200
other companies/state agencies. PCB manufacturing ceased in
1984, and use restrictions were introduced in stages over the
1990s as Czechia moved toward compliance with EU
regulations with the goal of eventually entering the EU.
These late restrictions are the reason why the breast milk of
Czech and Slovak mothers contains, on average, the highest
concentration of PCBs among industrialized countries.69,70
The Czech PCB stockpile and waste management plan
covering 2003 to 2013 dictates that all PCB materials
exceeding 50 ppm be managed according to ESM provisions
of the Stockholm Convention. Czechia, as a member of the
EU, additionally follows the EU PCB regulations, which are
stricter than those of the Stockholm Convention for large
equipment, requiring that equipment with PCB volumes >5
dm3
was decontaminated or disposed of by 2010.
As with Canada, open sources were not considered in the
Czech inventory. PCB-containing building sealants had limited
use in Czechia and other former Eastern Bloc countries, unlike
in Western Europe and North America, due to their higher
cost. The major open use of PCBs was in paints, particularly on
bridges and in military applications, constituting approximately
21% of total use.45,47
PCBs released from paints have caused
significant environmental contamination, including of Elbe
River sediments with up to 6 mg/kg PCBs due to renovations
to a railway bridge that had PCB-containing paint.71
The biggest challenge in Czechia is not the documented
inventory of PCBs but rather the numerous abandoned
industrial/contaminated sites. This number is larger than
typical in Western countries due to the country’s transition
from a communist economy in the early 1990s, dissolution of
state-owned companies, and subsequent bankruptcies and
abandonment of these sites. Numerous facilities now remain
without responsible ownership and contain abandoned
industrial infrastructure, with possible PCB contamination, as
well as contaminated soils and other materials. The database of
contaminated sites maintained by the Czech Ministry of
Environment listed 387 sites with PCB contamination as of
2016; however, 88% remain only as suspected contamination,
still lacking a proper site survey.45
Since the early 2000s, the Czech PCB inventory appears to
have grown because of the “discovery” of many of these sites
and reporting from companies that had not initially disclosed
their stocks. The 2002−2005 database reported 25,000
contaminated or potentially contaminated items, with ∼3000
t of bulk PCB fluid/materials.48
By 2016, there were slightly
fewer PCB-containing items (21,300) but a much higher bulk
mass of PCBs (6093 t) compared to 2005. This discrepancy
arose because of the inclusion of small PCB items that were
not included in the original inventory. Most importantly, the
mass of pure PCBs has dramatically decreased: only 1% (0.1−
17%) of the 2005 stock of pure PCBs was estimated to remain
in 2016 (Figure 3). This large decrease in pure PCBs is
attributed to progress toward compliance with EU regulations,
with priority given to ESM of high-level and large-volume PCB
equipment.
Czechia has sufficient capacity for ESM of PCBs (e.g.,
annual hazardous/POP waste incineration capacity of >21,000
t46
); however, the country requires action from both private
entities and the state, in case of the abandoned stock, and a
significant effort to remove PCBs from use if it is to achieve
ESM by 2028.
USA. The USA was the world’s largest producer and
consumer of PCB products with an estimated use of 500,000
tonnes of PCBs, or 1.9 kg per capita, the highest in the world
(Figures 2A and S5). The USA passed regulations in 1979
under the TSCA to prohibit the manufacture, processing,
distribution in commerce, and new use of PCBs. However, the
USA has not ratified the Stockholm Convention and does not
have national legislation that sets deadlines for PCB
elimination.72
This lack of national legislation is also reflected
in its fragmented and incomplete PCB data. While it is clear
from the PCB Cleanup and Disposal program50
that the USA
has removed a large stock of PCBs from use, the incomplete
current inventories prohibit the evaluation of the remaining
burden of PCBs in the USA. Our analysis strongly suggested
that the transformer registration database is missing a
substantial number of transformers, and no inventory exists
for other PCB materials (capacitors, ballasts, other electrical
equipment, and contaminated soil). In Canada, these materials
accounted for ∼25% of the total mass of pure PCBs and 99%
of the mass of bulk PCB materials; we expect similar
proportions in the USA. The US transformer database requires
responsible parties to self-report information, with limited
enforcement, and the concentration of PCBs is not reported.
The incomplete or erroneous information greatly limits the
accuracy of the inventory.
Our assessment of the US records as incomplete contrasts
with a recent UNEP report suggesting that US PCB records
are more comprehensive than those for other countries.17
While this may be the case in comparison to developing
countries, we argue that the fragmented and inconsistent
nature of reporting, limited mainly to transformers, coupled
with the highest global PCB manufacturing and use, presents
the USA as a worst-case scenario for PCB management in
countries with a capacity to do so.
Even based on incomplete information, the stock of PCBs in
the USA remains large compared to Canada and Czechia.
While the USA has removed 100,000 t of PCBs from use, the
impact of this removal on total PCB stocks is highly uncertain
because of poor record-keeping. Moreover, the USA did not
show a significant decrease in the mass of pure PCBs between
2005 and 2020 as was seen for Canada and Czechia (Figure
3F), which have prioritized removal of high-level PCB
materials (e.g., askarel transformers). Further, PCBs removed
from use in the USA are legally allowed to be disposed of by
methods not considered ESM by the Stockholm Convention,
such as landfilling. This is a major concern as landfills may act
as secondary PCB sources by contaminating surrounding
ecosystems with resulting ongoing human and environmental
exposures.73
The environmental and societal burden due to the
ongoing use and non-ESM disposal of PCBs in the USA is a
clear concern, given the country’s history as the world’s largest
producer and user of PCBs.
The per capita mass of pure PCBs in the USA inventory
(2.40 g/person for 2020) is comparable to Ontario, Canada
(2.41 g/person for 2013−2016), while Czechia is lower, at
0.27 g/person for 2016. However, of all three countries,
Canada has the most complete inventory, which includes large
masses of contaminated soil and gravel that are not included in
inventories in either Czechia or USA. The per capita stock in
the USA would be significantly higher if the inventory included
the additional categories of capacitors, other electrical
equipment, and contaminated waste materials that have been
significant portions of the Canadian stock (Figure 3G). It is
Environmental Science & Technology pubs.acs.org/est Article
https://doi.org/10.1021/acs.est.2c01204
Environ. Sci. Technol. 2022, 56, 9029−9040
9036
estimated that USA has at least 26 million cubic meters of soils
contaminated with PCBs17
and 350 “Superfund” sites with
reported PCB contamination.
■ IMPLICATIONS AND LESSONS LEARNED FROM
PCB FAILURES
In 2016, UNEP completed an assessment of global efforts to
eliminate PCBs, with an update in 2019.17,53
The report
highlighted similar challenges as identified here: incomplete
inventories of PCBs in many regions and large uncertainties in
current stocks and extent of ESM. The report identified that
“the majority of countries (with some notable exceptions) are
currently not on track to achieve the ESM of PCBs by 2028.”
Our analysis of the progress, available infrastructure, and the
challenges within individual countries in managing PCBs
confirm the implausibility of achieving this goal without a rapid
change in actions. UNEP also reported that the PCBs
eliminated so far were likely “low-hanging fruit,” and further
elimination will present more logistical and technical problems,
challenging our ability to achieve the rates of elimination
needed to reach Stockholm Convention goals.17
One substantial challenge is PCBs remaining in open
applications, which are poorly documented even in countries
that report ESM. To date, Sweden has given the most
comprehensive attention to open applications through a
program of identification and decontamination of buildings
with PCBs in building materials.74
Also, a small number of
other countries have developed general estimates and
addressed contaminated buildings on a case-by-case basis.
The difficulty in managing open applications of PCBs is the
lack of documentation of use combined with their importance
to human exposure25,75
as well as the lack of coherent
strategies to manage PCB-containing/PCB-contaminated
building materials without increasing emissions of PCBs.
In addition to the challenge of ESM for the remaining global
stock of PCBs, we are faced with the challenge that some large
fraction of PCBs is no longer “manageable.” These
“unmanageable” stocks have been released to the environment,
landfilled without documentation, or, in arguably worst cases,
are “lost” due to a lack of labelling and documentation and
have entered the commercial sector as oils without identified
PCB content. The longer the use and improper storage of
PCBs persists, the greater the potential for environmental
releases and human exposures, particularly considering the
aging infrastructure that houses PCBs.
Strong regulation combined with financial and technological
capacity and enforcement, as demonstrated in Canada, can
successfully reduce stocks of PCBs and advance ESM. Many
countries face substantial challenges due to historical structures
(e.g., as in Czechia), and even with sufficient financial and
technological capacity, their progress toward achieving the
Stockholm Convention deadlines is hindered by legacies of
poor record keeping, environmental practices, and site
ownership. However, most troubling is the prevalent lack of
capacity to manage PCB stocks. The Stockholm Convention
plays an important role in education and capacity building
toward PCB elimination, motivating countries, particularly
those with the ability to implement the Convention, to
inventory and dispose of PCBs. Nonetheless, the inventory
quality is generally poor in many countries, compounding the
challenge due to the lack of resources and competing national
pressures.
The USA is absent from the Stockholm Convention and
lacks effective federal policies to remove and safely dispose of
PCBs despite being the largest producer, user, and likely holder
of the largest stock of PCBs. This is a clear challenge to the
global goal of achieving ESM of PCBs because, by their
properties of persistence and long-range transport, PCBs are a
global threat.
The global distribution of PCB use and stocks is not
uniform. The USA, Canada, Soviet Union, Japan, and Western
European countries dominated the use of PCBs (Figure 2A).
In contrast, today that legacy is shifting globally due to
transboundary transport of PCB-containing equipment and
wastes. While such movement of hazardous substances is
under the purview of the Basel and Rotterdam Conventions,
compliance remains a challenge. This global shift is particularly
problematic considering the lack of PCB management
capacities in lower income/lower development countries.
Global inequality is a major challenge in the implementation
of the Stockholm Convention objective of achieving ESM of
PCBs and POPs in general.
The Stockholm Convention set a deadline to phase out
PCBs some 40 years after cessation of production and more
than 50 years after many highly developed countries banned
their manufacturing, import, and new use. Yet, this deadline
appears unachievable due to the resources (financial and
technical) and the political will required to address the
problem. As the effort required to eliminate PCBs has been
seriously underestimated,17
we question the feasibility of
removing other newer POPs that have entered widespread use.
In the context of POPs, it could be argued that PCBs are one
of the simpler problems: most use was in large, closed items,
global trade in PCB-containing materials was limited, and most
PCBs were held by large industries which were compelled to
inventory and report them. This is a sharp contrast to two
classes of POPs which have received recent attention
chlorinated paraffins and per- and polyfluoroalkyl substances,
such as PFOS and PFOA. The major use of these chemicals is
fundamentally different from PCBs; they are primarily used in
open sources and held by millions of individuals, which stymies
efforts to inventory and manage removal, with major
consequences for the environmental and human health of
future generations.76
Chlorinated paraffins are a clear case of
regrettable substitution as the short-chained chlorinated
paraffins replaced PCBs in many open applications26
and,
since 2018, are also restricted under the Stockholm
Convention. This highlights the critical need and urgency to
curtail production and use of chemicals with POP characteristics
as complex management challenges will not soon be
solved, and the consequences are likely to fall disproportionately
on lower-income/lower-development countries.
■ ASSOCIATED CONTENT
*sı Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.est.2c01204.
Additional methods details, sensitivity analysis, and
additional figures and tables (PDF)
Complete data sources for global analysis (XLSX)
Complete data sources for USA analysis (XLSX)
Environmental Science & Technology pubs.acs.org/est Article
https://doi.org/10.1021/acs.est.2c01204
Environ. Sci. Technol. 2022, 56, 9029−9040
9037
■ AUTHOR INFORMATION
Corresponding Author
Miriam L. Diamond − Department of Earth Sciences,
University of Toronto, Toronto, Ontario M5S 3B1, Canada;
School of the Environment, University of Toronto, Toronto,
Ontario M5T 1P5, Canada; orcid.org/0000-0001-6296-
6431; Email: miriam.diamond@utoronto.ca
Authors
Lisa Melymuk − Faculty of Science, Masaryk University, Brno
611 37, Czech Republic; Department of Earth Sciences,
University of Toronto, Toronto, Ontario M5S 3B1, Canada
Jonathan Blumenthal − Department of Earth Sciences,
University of Toronto, Toronto, Ontario M5S 3B1, Canada;
orcid.org/0000-0003-4399-0422
Ondřej Sáňka − Faculty of Science, Masaryk University, Brno
611 37, Czech Republic
Adriana Shu-Yin − Department of Earth Sciences, University
of Toronto, Toronto, Ontario M5S 3B1, Canada; Present
Address: Canadian Urban Transit Research and
Innovation Consortium (CUTRIC), Toronto, Ontario,
Canada
Veena Singla − Healthy People & Thriving Communities
Program, Natural Resources Defense Council, San Francisco,
California 94104, United States; orcid.org/0000-0002-
5107-0717
Kateřina Šebková − Faculty of Science, Masaryk University,
Brno 611 37, Czech Republic
Kristi Pullen Fedinick − Science Office, Natural Resources
Defense Council, Washington, D.C. 20005, United States
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.est.2c01204
Notes
The authors declare no competing financial interest.
■ ACKNOWLEDGMENTS
Thanks to Karen Swetland-Johnson at US EPA for
consultations on USA PCB records. Thanks to Ana Miralles
Marco, Olga Popova, and Lola Bajard for help with the
translation of NIPs. This project has received funding from the
European Union’s Horizon 2020 research and innovation
programme under Marie Sklodowska-Curie grant agreement
no. 734522. The authors acknowledge the support of the
Research Infrastructure RECETOX RI (no. LM2018121 and
CZ.02.1.01/0.0/0.0/17_043/0009632) financed by the Ministry
of Education, Youth and Sports, support from Environment
and Climate Change Canada, Great Lakes Harmful
Pollutants Section (no. 3000659886), and the Natural Sciences
and Engineering Research Council of Canada (RGPIN-2017-
06654). This project was supported from the European
Union’s Horizon 2020 research and innovation programme
under grant agreement no. 857560. This publication reflects
only the author’s view and the European Commission is not
responsible for any use that may be made of the information it
contains.
■ REFERENCES
(1) Jamieson, A. J.; Malkocs, T.; Piertney, S. B.; Fujii, T.; Zhang, Z.
Bioaccumulation of Persistent Organic Pollutants in the Deepest
Ocean Fauna. Nat. Ecol. Evol. 2017, 1, 0051.
(2) Roze, E.; Meijer, L.; Bakker, A.; Van Braeckel, K. N. J. A.; Sauer,
P. J. J.; Bos, A. F. Prenatal Exposure to Organohalogens, Including
Brominated Flame Retardants, Influences Motor, Cognitive, and
Behavioral Performance at School Age. Environ. Health Perspect. 2009,
117, 1953−1958.
(3) Berghuis, S. A.; Bos, A. F.; Sauer, P. J. J.; Roze, E. Developmental
Neurotoxicity of Persistent Organic Pollutants: An Update on
Childhood Outcome. Arch. Toxicol. 2015, 89, 687−709.
(4) Boucher, O.; Jacobson, S. W.; Plusquellec, P.; Dewailly, É.;
Ayotte, P.; Forget-Dubois, N.; Jacobson, J. L.; Muckle, G. Prenatal
Methylmercury, Postnatal Lead Exposure, and Evidence of Attention
Deficit/Hyperactivity Disorder among Inuit Children in Arctic
Québec. Environ. Health Perspect. 2012, 120, 1456−1461.
(5) Bennett, D.; Bellinger, D. C.; Birnbaum, L. S.; Bradman, A.;
Chen, A.; Cory-Slechta, D. A.; Engel, S. M.; Fallin, M. D.; Halladay,
A.; Hauser, R.; Hertz-Picciotto, I.; Kwiatkowski, C. F.; Lanphear, B.
P.; Marquez, E.; Marty, M.; McPartland, J.; Newschaffer, C. J.; PayneSturges,
D.; Patisaul, H. B.; Perera, F. P.; Ritz, B.; Sass, J.; Schantz, S.
L.; Webster, T. F.; Whyatt, R. M.; Woodruff, T. J.; Zoeller, R. T.;
Anderko, L.; Campbell, C.; Conry, J. A.; DeNicola, N.; Gould, R. M.;
Hirtz, D.; Huffling, K.; Landrigan, P. J.; Lavin, A.; Miller, M.; Mitchell,
M. A.; Rubin, L.; Schettler, T.; Tran, H. L.; Acosta, A.; Brody, C.;
Miller, E.; Miller, P.; Swanson, M.; Witherspoon, N. O. Project
TENDR: Targeting Environmental Neuro-Developmental Risks The
TENDR Consensus Statement. Environ. Health Perspect. 2016, 124,
118−122.
(6) Grandjean, P.; Landrigan, P. Developmental Neurotoxicity of
Industrial Chemicals. Lancet 2006, 368, 2167−2178.
(7) Jusko, T. A.; Sisto, R.; Iosif, A.-M.; Moleti, A.; Wimmerová, S.;
Lancz, K.; Tihányi, J.; Šovčiková, E.; Drobná, B.; Palkovičová, L. u.;
Jurečková, D.; Thevenet-Morrison, K.; Verner, M.-A.; Sonneborn, D.;
Hertz-Picciotto, I.; Trnovec, T. Prenatal and Postnatal Serum PCB
Concentrations and Cochlear Function in Children at 45 Months of
Age. Environ. Health Perspect. 2014, 122, 1246−1252.
(8) Poon, E.; Bandara, S. B.; Allen, J. B.; Sadowski, R. N.; Schantz, S.
L. Developmental PCB Exposure Increases Susceptibility to
Audiogenic Seizures in Adulthood. Neurotoxicology 2015, 46, 117−
124.
(9) Jones, D. C.; Miller, G. W. The Effects of Environmental
Neurotoxicants on the Dopaminergic System: A Possible Role in
Drug Addiction. Biochem. Pharmacol. 2008, 76, 569−581.
(10) Diamond, M. L. Toxic Chemicals as Enablers and Poisoners of
the Technosphere. Anthr. Rev. 2017, 4, 72−80.
(11) Breivik, K.; Sweetman, A. J.; Pacyna, J. M.; Jones, K. C.
Towards a Global Historical Emission Inventory for Selected PCB
Congeners–a Mass Balance Approach. 1. Global Production and
Consumption. Sci. Total Environ. 2002, 290, 181−198.
(12) Markowitz, G.; Rosner, D. Monsanto, PCBs, and the Creation
of a “World-wide Ecological Problem”. J Public Health Pol. 2018, 39,
463.
(13) Jensen, S. The PCB Story. Ambio 1972, 1, 123−131.
(14) Koppe, J. G.; Keys, J. PCBs and the Precautionary Principle.
The Precautionary Principle in the 20th Century: Late Lessons from Early
Warnings; Earthscan Publications, 2002.
(15) UNEP. Listing of the Basel Convention Technical Guidelines for
Environmentally Sound Management (ESM) of Persistent Organic
Pollutants (POPs) Wastes, 2005.
(16) UNEP. Towards Elimination of PCBs. https://www.unep.org/
explore-topics/chemicals-waste/what-we-do/persistent-organicpollutants/toward-elimination-pcb
(accessed on May 21, 2021).
(17) UNEP. Consolidated Assessment of Efforts Made towards the
Elimination of Polychlorinated Biphenyls: Geneva, 2016.
(18) Petrlík, J. When the Polluter Does Not Pay. PEN Mag 2010, 1,
12−13.
(19) UNEP. PCBs Elimination Network (PEN)2028Environmentally
Sound Management of PCBs: Geneva, 2011.
(20) Warmuth, A.; Ohno, K. The PCBs Elimination Network: The
Information Exchange Platform Created for the Risk Reduction of
Polychlorinated Biphenyls (PCBs). J. Epidemiol. Community Health
2013, 67, 4.
Environmental Science & Technology pubs.acs.org/est Article
https://doi.org/10.1021/acs.est.2c01204
Environ. Sci. Technol. 2022, 56, 9029−9040
9038
(21) UNEP. Guidelines for the Identification of PCBs and Materials
Containing PCBs; United Nations Environment Programme, 1999; p
40.
(22) Wagner, U. K. Inventories of PCBs: An Expert’s Point of View.
PEN Magazine, 2010; No. 1.
(23) Herrick, R. F.; Stewart, J. H.; Allen, J. G. Review of PCBs in US
Schools: A Brief History, an Estimate of the Number of Impacted
Schools, and an Approach for Evaluating Indoor Air Samples. Environ.
Sci. Pollut. Res. 2016, 23, 1975−1985.
(24) Thomas, K.; Xue, J.; Williams, R. W.; Jones, P. D.; Whitaker, D.
Polychlorinated Biphenyls (PCBs) in School Buildings: Sources; Environmental
Levels, and Exposures, 2012.
(25) Marek, R. F.; Thorne, P. S.; Herkert, N. J.; Awad, A. M.;
Hornbuckle, K. C. Airborne PCBs and OH-PCBs Inside and Outside
Urban and Rural U.S. Schools. Environ. Sci. Technol. 2017, 51, 7853−
7860.
(26) UNEP. Consolidated Guidance on PCB in Open Applications,
2019.
(27) Rudel, R. A.; Seryak, L. M.; Brody, J. G. PCB-Containing Wood
Floor Finish Is a Likely Source of Elevated PCBs in Residents’ Blood,
Household Air and Dust: A Case Study of Exposure. Environ. Health
2008, 7, 2.
(28) Parker, J. S.; Mayer-Blackwell, K. Open Book, Open Source:
PCB Usage in Mass-Market Paperback Book Adhesives. Environ. Sci.
Technol. Lett. 2019, 6, 565.
(29) Breivik, K.; Sweetman, A.; Pacyna, J. M.; Jones, K. C. Towards
a Global Historical Emission Inventory for Selected PCB Congeners A
Mass Balance Approach: 2. Emissions. Sci. Total Environ. 2002, 290,
199−224.
(30) OECD. Decision of the Council on Protection of the Environment
by Control of Polychlorinated BiphenylsOECD/LEGAL/0108
C(73)1(Final); Organisation for Economic Co-operation and
Development, 1973.
(31) Weber, R.; Herold, C.; Hollert, H.; Kamphues, J.; Ungemach,
L.; Blepp, M.; Ballschmiter, K. Life Cycle of PCBs and Contamination
of the Environment and of Food Products from Animal Origin.
Environ. Sci. Pollut. Res. 2018, 25, 16325−16343.
(32) Cui, S.; Qi, H.; Liu, L.-Y.; Song, W.-W.; Ma, W.-L.; Jia, H.-L.;
Ding, Y.-S.; Li, Y.-F. Emission of Unintentionally Produced
Polychlorinated Biphenyls (UP-PCBs) in China: Has This Become
the Major Source of PCBs in Chinese Air? Atmos. Environ. 2013, 67,
73−79.
(33) Fan, Y.; Ren, M.; Zhang, H.; Geng, N.; Li, Y.; Zhang, N.; Zhao,
L.; Gao, Y.; Chen, J. Levels and Fingerprints of Chlorinated Aromatic
Hydrocarbons in Fly Ashes from the Typical Industrial Thermal
Processes: Implication for the Co-Formation Mechanism. Chemosphere
2019, 224, 298−305.
(34) Hu, D.; Hornbuckle, K. C. Inadvertent Polychlorinated
Biphenyls in Commercial Paint Pigments. Environ. Sci. Technol.
2010, 44, 2822−2827.
(35) Herkert, N. J.; Jahnke, J. C.; Hornbuckle, K. C. Emissions of
Tetrachlorobiphenyls (PCBs 47, 51, and 68) from Polymer Resin on
Kitchen Cabinets as a Non-Aroclor Source to Residential Air. Environ.
Sci. Technol. 2018, 52, 5154.
(36) Schneider, A. R.; Stapleton, H. M.; Cornwell, J.; Baker, J. E.
Recent Declines in PAH, PCB, and Toxaphene Levels in the
Northern Great Lakes as Determined from High Resolution Sediment
Cores. Environ. Sci. Technol. 2001, 35, 3809−3815.
(37) Hung, H.; Katsoyiannis, A. A.; Brorström-Lundén, E.;
Olafsdottir, K.; Aas, W.; Breivik, K.; Bohlin-Nizzetto, P.; Sigurdsson,
A.; Hakola, H.; Bossi, R.; Skov, H.; Sverko, E.; Barresi, E.; Fellin, P.;
Wilson, S. Temporal Trends of Persistent Organic Pollutants (POPs)
in Arctic Air: 20 Years of Monitoring under the Arctic Monitoring and
Assessment Programme (AMAP). Environ. Pollut. 2016, 217, 52−61.
(38) Salamova, A.; Pagano, J. J.; Holsen, T. M.; Hites, R. A. Post-
1990 Temporal Trends of PCBs and Organochlorine Pesticides in the
Atmosphere and in Fish from Lakes Erie, Michigan, and Superior.
Environ. Sci. Technol. 2013, 47, 9109−9114.
(39) Desforges, J.-P.; Hall, A.; McConnell, B.; Rosing-Asvid, A.;
Barber, J. L.; Brownlow, A.; De Guise, S.; Eulaers, I.; Jepson, P. D.;
Letcher, R. J.; Levin, M.; Ross, P. S.; Samarra, F.; Víkingson, G.;
Sonne, C.; Dietz, R. Predicting Global Killer Whale Population
Collapse from PCB Pollution. Science 2018, 361, 1373−1376.
(40) Bellinger, D. C. Applying Methods of the Global Burden of
Diseases, Injuries, and Risk Factors Study to Developmental
Neurotoxicants: A Commentary. Environ. Health 2018, 17, 1−6.
(41) Statistics Canada. Population estimates, quarterly. https://
www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=1710000901 (accessed
on Feb 16, 2022).
(42) Diamond, M. L.; Melymuk, L.; Csiszar, S. A.; Robson, M.
Estimation of PCB Stocks, Emissions, and Urban Fate: Will Our
Policies Reduce Concentrations and Exposure? Environ. Sci. Technol.
2010, 44, 2777−2783.
(43) US EPA. PCB Inspection Manual: Washington D.C., 2004.
(44) Canadian Council of Ministers of the Environment. PCB
Transformer Decontamination, CCME EPC-HW-105E: Ottawa, 1995; p
35.
(45) Bláha, K.; Borůvková, J.; Budňáková, M.; Čupr, P.; Eyblová, D.;
Gadas, P.; Gruntorád, J.; Hlavatý, E.; Holoubek, I.; Hůlek, R.; Jarolím,
O.; Klánová, J.; Kolář, J.; Krech, L.; Krejčová, V.; Prokeš, R.;
Přibylová, P.; Ptáčková, Ž.; Puklová, V.; Straková, H.; Šebková, K.;
Špringar, Z.; Štěpánek, P.; Udatný, M.; Vávra, T.; Veselý, I.;
Vytopilová, M. Updated National Implementation Plan for Implementation
of Stockholm Convention on Persistent Organic Pollutants in the
Czech Republic for the Period 2018-2023: Prague, 2017.
(46) Ministry of the Environment of the Czech Republic.
Questionnaire on PCB Wastes: Prague, 2016.
(47) Electronic Reporting System of the Stockholm Convention (Third
Reporting)Czech Republic; Stockholm Convention, reported by
Czech Republic: Geneva, 2018.
(48) Holoubek, I.; Adamec, V.; Bartoš, M.; Bláha, K.; Bláha, L.;
Budňáková, M.; Černá, M.; Čupr, P.; Demnerová, K.; Drápal, J.;
Hajšlová, J.; Hanzálková, M.; Holoubková, I.; Hrabětová, S.; Jech, L.;
Klánová, J.; Kohoutek, J.; Kužílek, V.; Machálek, P.; Matějů, V.;
Matoušek, J.; Matoušek, M.; Mestřík, V.; Novák, J.; Ocelka, T.;
Pekárek, V.; Petrlík, J.; Petira, O.; Provazník, K.; Punčochář, M.;
Rieder, M.; Ruprich, J.; Sáňka, M.; Tomaniová, M.; Vácha, R.; Volka,
K.; Zbíral, J. The National Implementation Plan for Implementation of
the Stockholm Convention in the Czech Republic: Brno, 2006.
(49) U.S. Environmental Protection Agency (EPA). Registering
Transformers Containing Polychlorinated Biphenyls (PCBs). https://
www.epa.gov/pcbs/registering-transformers-containingpolychlorinated-biphenyls-pcbs
(accessed on Nov 1, 2021).
(50) US Environmental Protection Agency (EPA). Managing
Remediation Waste From Polychlorinated Biphenyls (PCBs)
Cleanups. https://www.epa.gov/pcbs/managing-remediation-wastepolychlorinated-biphenyls-pcbs-cleanups
(accessed on Nov 1, 2021).
(51) Shanahan, C. E.; Spak, S. N.; Martinez, A.; Hornbuckle, K. C.
Inventory of PCBs in Chicago and Opportunities for Reduction in
Airborne Emissions and Human Exposure. Environ. Sci. Technol. 2015,
49, 13878−13888.
(52) Gimlin, P. Environmental Protection Specialist. Personal
Communication [September 2020]; US EPA: Washington, DC, 2020.
(53) UNEP. Report on Progress towards the Elimination of
Polychlorinated Biphenyls. UNEP/POPS/COP.9/INF/10, 2019.
(54) Sri Lanka Ministry of Environment. National Implementation
Plan under the Stockholm Convention on POPs for Sri Lanka; Persistent
Organic Pollutants (POPs) Project, Ministry of Environment:
Colombo, Sri Lanka, 2004.
(55) Republic of Nauru. National Implementation Plan for Persistent
Organic Pollutants; POPs, 2012.
(56) Environmental Affairs Department. National Implementation
Plan (NIP) for the Management of Persistent Organic Pollutants; Malawi
Ministry of Mines, Natural Resources, and Environment: Lilongwe,
Malawi, 2005.
(57) Undersecretariat for Environmental Management. National
Implementation Plan of the Stockholm Convention in the Dominican
Environmental Science & Technology pubs.acs.org/est Article
https://doi.org/10.1021/acs.est.2c01204
Environ. Sci. Technol. 2022, 56, 9029−9040
9039
Republic; Ministry of the Environment and Natural Resources
(SEMARENA), United Nations Development Programme
(UNDP): Santo Domingo, Dominican Republic, 2008.
(58) Environmental Protection Agency. National Implementation
Plan of the Stockholm Convention on Persistent Organic Pollutants:
Accra: Ghana, 2007.
(59) Ministry of Sustainable Development and Tourism. National
Implementation Plan for the Stockholm Convention 2014−2021:
Podgorica, Montenegro, 2013.
(60) Global Environment Facility. Disposal of PCB Oils Contained
in Transformers and Disposal of Capacitors Containing PCB in
Southern Africa. https://www.thegef.org/project/disposal-pcb-oils-
contained-transformers-and-disposal-capacitors-containing-pcbsouthern
(accessed on May 20, 2021).
(61) Environment and Climate Change Canada. Questionnaire for
the review of progress towards the elimination of PCB in accordance
with paragraph (h) of part II of Annex A. http://www.pops.int/
Countries/CountryProfiles/tabid/4501/Default.aspx (accessed on
May 20, 2021).
(62) Manhart, J. POPs and PCBs Management in the Czech Republic;
Ministry of the Environment, Environmental Damage Department:
Prague, 2009.
(63) PCB Regulations. PCB Regulations (SOR/2008-273), 2018.
(64) Treasury Board of Canada Secretariat. Chemicals Management
Plan, 2022.
(65) Robson, M.; Melymuk, L.; Csiszar, S. A.; Giang, A.; Diamond,
M. L.; Helm, P. A. Continuing Sources of PCBs: The Significance of
Building Sealants. Environ. Int. 2010, 36, 506−513.
(66) Federal Ministry for the Environment Nature Conservation
Building and Nuclear Safety. Questionnaire for the review of progress
towards the elimination of PCB in accordance with paragraph (h) of
part II of Annex A. http://www.pops.int/Countries/CountryProfiles/
tabid/4501/Default.aspx (accessed on May 20, 2021).
(67) Federal Office for the Environment. Questionnaire for the
review of progress towards the elimination of PCB in accordance with
paragraph (h) of part II of Annex A. http://www.pops.int/Countries/
CountryProfiles/tabid/4501/Default.aspx (accessed on May 20,
2021).
(68) Canada Gazette. Canada Gazette: Ottawa November, 2006; pp
1−65.
(69) van den Berg, M.; Kypke, K.; Kotz, A.; Tritscher, A.; Lee, S. Y.;
Magulova, K.; Fiedler, H.; Malisch, R. WHO/UNEP global surveys of
PCDDs, PCDFs, PCBs and DDTs in human milk and benefit-risk
evaluation of breastfeeding. Arch. Toxicol. 2017, 91, 83−96.
(70) Komprda, J.; Komprdová, K.; Domínguez-Romero, E.; Mikeš,
O.; Řiháčková, K.; Čupr, P.; Černá, M.; Scheringer, M. Dynamics of
PCB Exposure in the Past 50 Years and Recent High Concentrations
in Human Breast Milk: Analysis of Influencing Factors Using a
Physiologically Based Pharmacokinetic Model. Sci. Total Environ.
2019, 690, 388−399.
(71) ELSA. PCB in Der Elbe, Eigenschaften, Vorkommen Und Trends
Sowie Ursachen Und Folgen Der Erhöhten Freisetzung Im Jahr 2015,
2016.
(72) U.S. EPA. PART 761Polychlorinated Biphenyls (PCBs)
Manufacturing, Processing, Distribution in Commerce and Use
Prohibitions; TSCA, 2018.
(73) Weber, R.; Watson, A.; Forter, M.; Oliaei, F. Review Article:
Persistent organic pollutants and landfills - a review of past
experiences and future challenges. Waste Manag. Res. 2011, 29,
107−121.
(74) Bernevi Rex, G. Inventory and Clearance of PCBs in Buildings
and Facilities. REPORT 688; Swedish Environmental Protection
Agency: Stockholm, 2019.
(75) Herrick, R. F.; McClean, M. D.; Meeker, J. D.; Baxter, L. K.;
Weymouth, G. A. An Unrecognized Source of PCB Contamination in
Schools and Other Buildings. Environ. Health Perspect. 2004, 112,
1051−1053.
(76) Harrad, S.; Diamond, M. New Directions: Exposure to
Polybrominated Diphenyl Ethers (PBDEs) and Polychlorinated
Biphenyls (PCBs): Current and Future Scenarios. Atmos. Environ.
2006, 40, 1187−1188.
Environmental Science & Technology pubs.acs.org/est Article
https://doi.org/10.1021/acs.est.2c01204
Environ. Sci. Technol. 2022, 56, 9029−9040
9040
Recommended by ACS
Codeployment of Passive Samplers and Mussels Reveals
Major Source of Ongoing PCB Inputs to the Anacostia River
in Washington, DC
Nathalie J. Lombard, Upal Ghosh, et al.
JANUARY 09, 2023
ENVIRONMENTAL SCIENCE & TECHNOLOGY READ
Disentangling Source Profiles and Time Trends of
Halogenated Flame Retardants in the Great Lakes
Staci L. Capozzi, Marta Venier, et al.
JANUARY 09, 2023
ENVIRONMENTAL SCIENCE & TECHNOLOGY READ
Does Using Corsi–Rosenthal Boxes to Mitigate COVID-19
Transmission Also Reduce Indoor Air Concentrations of
PFAS and Phthalates?
Robin E. Dodson, Joseph M. Braun, et al.
DECEMBER 23, 2022
ENVIRONMENTAL SCIENCE & TECHNOLOGY READ
Regulatory Significance of Plastic Manufacturing Air
Pollution Discharged into Terrestrial Environments and
Real-Time Sensing Challenges
Yoorae Noh, Andrew J. Whelton, et al.
JANUARY 20, 2023
ENVIRONMENTAL SCIENCE & TECHNOLOGY LETTERS READ
Get More Suggestions >
Venier, Marta, Ondřej Audy, Šimon Vojta, Jitka Bečanová, Kevin Romanak, Lisa Melymuk,
Martina Krátká, Petr Kukucka, Joseph Okeme, Amandeep Saini, Miriam L. Diamond, Jana
Klanova. 2016. "Brominated Flame Retardants in Indoor Environment - Comparative Study
of Indoor Contamination from Three Countries." Environment International 94: 150-60.
https://doi.org/10.1016/j.envint.2016.04.029
Brominated flame retardants in the indoor environment — Comparative
study of indoor contamination from three countries
Marta Venier a,
⁎, Ondřej Audy b
, Šimon Vojta b
, Jitka Bečanová b
, Kevin Romanak a
, Lisa Melymuk b
,
Martina Krátká b
, Petr Kukučka b
, Joseph Okeme c
, Amandeep Saini c
, Miriam L. Diamond d,c
, Jana Klánová b
a
RECETOX, Masaryk University, Kamenice 753/5, pavilion A29, 62500 Brno, Czech Republic
b
School of Public and Environmental Affairs, Indiana University, 702 Walnut Grove Avenue, Bloomington, IN 47405, United States
c
Department of Physical and Environmental Sciences, University of Toronto Scarborough, 1265 Military Trail, Toronto, ON, Canada, M1C 1A4
d
Department of Earth Sciences, University of Toronto, 22 Russell Street, Toronto, Canada M5S 3B1
a b s t r a c ta r t i c l e i n f o
Article history:
Received 8 January 2016
Received in revised form 15 April 2016
Accepted 16 April 2016
Available online 29 May 2016
Concentrations of more than 20 brominated flame retardants (FRs), including polybrominated diphenyl ethers
(PBDEs) and emerging FRs, were measured in air, dust and window wipes from 63 homes in Canada, the
Czech Republic and the United States in the spring and summer of 2013. Among the PBDEs, the highest concentrations
were generally BDE-209 in all three matrices, followed by Penta-BDEs. Among alternative FRs, EHTBB
and BEHTBP were detected at the highest concentrations. DBDPE was also a major alternative FR detected in
dust and air. Bromobenzenes were detected at lower levels than PBDEs and other alternative FRs; among the
bromobenzenes, HBB and PBEB were the most abundant compounds. In general, FR levels were highest in the
US and lowest in the Czech Republic — a geographic trend that reflects the flame retardants' market. No statistically
significant differences were detected between bedroom and living room FR concentrations in the same
house (n = 10), suggesting that sources of FRs are widespread indoors and mixing between rooms. The concentrations
of FRs in air, dust, and window film were significantly correlated, especially for PBDEs. We found a significant
relationship between the concentrations in dust and window film and in the gas phase for FRs with
log KOA values b14, suggesting that equilibrium was reached for these but not compounds with log KOA values
N14. This hypothesis was confirmed by a large discrepancy between values predicted using a partitioning
model and the measured values for FRs with log KOA values N14.
© 2016 Elsevier Ltd. All rights reserved.
Keywords:
Brominated flame retardants
Indoor
Air
Dust
Window film
1. Introduction
Because the indoor environment is an important human exposure
route for semi-volatile organic compounds (SVOCs), such as brominated
flame retardants (BFRs), knowledge of their levels and partitioning
between different indoor matrices is essential for evaluating their impact
on human health. Most studies on BFRs in the indoor environment
have focused on their presence in dust because of its putative contribution
to exposure [e.g., Harrad et al. (2008b); Wilford et al. (2004)]. Although,
some studies have documented air concentrations indoors
(Abdallah et al., 2008; Dodson et al., 2015), significantly fewer studies
have dealt with accumulation of BFRs on surfaces such as window
films (Bennett et al., 2015; Butt et al., 2004; Cetin and Odabasi, 2011).
Of these studies, only Bennett et al. (2015) compared indoor film concentrations
with those of other indoor matrices, i.e. air and dust.
Indoor air can be sampled using passive or active techniques, and
each has its advantages and disadvantages. Passive samplers are easy
to deploy and are unobtrusive, which is important in an indoor setting.
These samplers do not require electricity, but they need to be deployed
for several weeks, providing an integrated measurement over this time
period. In comparison, active samplers are bulky and noisy, which is
particularly problematic indoors. They require trained personnel to be
deployed, but they can be left at the site for shorter periods. The most
common passive air sampling design uses a polyurethane foam (PUF)
disk enclosed in a stainless steel bowl (Shoeib and Harner, 2002).
With knowledge of sampling rates, one can calculate time-integrated
air concentrations for compounds mainly present in the gas phase
(Bohlin et al., 2014a, 2014b; Saini et al., 2015). Recent studies have
shown that these samplers can also provide reliable results for higher
molecular weight compounds that are found mainly in the particulate
phase (Bohlin et al., 2014a; Harner et al., 2013; Harrad and Abdallah,
2008a; Peverly et al., 2015).
Indoor dust is a complicated, heterogeneous matrix for which different
sampling approaches have been used. The most common technique
is to collect floor dust, although in some circumstances undisturbed settled
dust on other surfaces can be used (Björklund et al., 2012; de Wit
et al., 2012; Lioy et al., 2002). Comparisons between different studies
Environment International 94 (2016) 150–160
⁎ Corresponding author.
http://dx.doi.org/10.1016/j.envint.2016.04.029
0160-4120/© 2016 Elsevier Ltd. All rights reserved.
Contents lists available at ScienceDirect
Environment International
journal homepage: www.elsevier.com/locate/envint
can be confounded by differences in sampling methods, within room
spatial variability, and by the presence in the room of specific products
(e.g. electronics) (Harrad et al., 2009; Muenhor and Harrad, 2012). Recently,
Li et al. (2015) found a strong positive correlation between the
power consumption of electronics and PBDE levels in air and dust in a
large room, which they attributed to heat generated from in-use electronics
promoting the release of these compounds. The least common
medium studied in the indoor setting is the film formed on hard
surfaces by condensation and deposition of gas-phase chemicals and
particles respectively (Diamond et al., 2000; Gingrich et al., 2001). Windows
are convenient to use for sampling the film because the glass is
inert, because of the lack of contamination that could arise from other
surfaces, and because the film can be removed quantitatively from this
surface. The most common approach to windows film sampling employs
pre-cleaned wipes (Butt et al., 2004).
With the control and subsequent decline and cessation of production
of polybrominated diphenyl ethers (PBDEs) in the last decade in
many countries, other brominated flame retardants (BFRs) have
risen in importance. Even though some of these alternative flame retardants
have been produced for a long time, most of them have only
came to the attention of the public and the global scientific
community in the past few years. While levels of PBDEs in the environment
are generally stable or decreasing (Crimmins et al., 2012; Ma
et al., 2013), concentrations of “new” brominated flame retardants, notably
2-ethylhexyl tetrabromobenzoate (EHTBB) and bis(2ethylhexyl)tetrabromophthalate
(BEHTBP), are increasing as more
products containing these compounds are introduced to the market to
replace products containing PBDEs (Dodson et al., 2012; Ma et al.,
2012; Stapleton et al., 2011). In spite of this change in the flame retardant
market, data regarding the presence of these replacements of
alternative compounds in the environment are limited.
In this paper, we report the concentrations of newer and legacy
flame retardants in the indoor environments in three different countries
(United States, Canada, and the Czech Republic). We collected air, dust,
and windows films in 63 private homes, and we measured about 20
brominated flame retardants and Dechlorane Plus, a chlorinated flame
retardant. In this paper, we compare concentrations in these three
countries, and we put them in the context of their usage in North
America and Europe. We also look at differences between rooms in
the same home to elucidate possible sources. Finally, we evaluate how
these compounds partition between phases (air, dust, and window
film) and evaluate which sampling media provide the most comprehensive
characterization of indoor levels.
2. Materials and methods
2.1. Study population
Samples were collected in three different locations: Bloomington,
Indiana, United States, Toronto, Canada, and Brno, Czech Republic in
May–August 2013. Air, dust, and window film samples were collected
from a total of 63 houses and apartments: 20 homes each from the
Czech Republic and the U.S. and 23 from Canada. At least one room
was sampled in each home (i.e. the main bedroom), and a second
room was sampled in 10 houses per country (i.e. the living room). Participation
in the campaign was voluntary and did not include any
compensation.
On day 1, passive samplers were deployed, and selected windows
were cleaned with Kimwipes moistened with 2-propanol until no dirt
was visible on the Kimwipes. Participants were asked not to vacuum
the room where the sampler was located until completion of the
campaign, if at all possible. Participants were interviewed by a field
technician to gather information about the house and the household
(e.g. electronic equipment and furniture in the sampled rooms, number
of occupants, and cleaning and ventilation habits).
2.2. Sample collection
Before sampling, all matrices (PUF disks, nylon vacuum socks, and
Kimwipes) were pre-cleaned by Soxhlet extraction (8 h in acetone,
then 8 h in toluene), dried, wrapped in aluminum foil, and transported
to the site. PUF disks for passive air sampling were exposed to indoor air
for 28 days using a single (U.S. and Canada) or double-bowl shaped
housing (Czech Republic) (see Fig. S1). Sampling rates for each sampler
configuration were calculated in a separate experiment by simultaneously
deploying single bowl and double bowl samplers (see
Supporting Information for details and Fig. S2). For this study, we used
a sampling rate of 1.6 m3
/day for the double bowl sampler and
2.9 m3
/day for the single bowl sampler. These values are consistent
with previously reported sampling rates indoors (Zhang et al., 2011).
Window film samples were collected after 28 days using pre-cleaned
Kimwipes moistened with 2-propanol. Windows were wiped with a
succession of Kimwipes until no dirt was visible on the Kimwipes, and
all Kimwipes from one window were composited. The sampled area averaged
at 0.32 m2
for Canada, 0.93 m2
for the U.S., and 1.8 m2
for the
Czech Republic. Floor dust samples from each room were taken using
pre-cleaned polyester socks inserted on a vacuum cleaner hose attachment,
by vacuuming the largest possible area and recording it. All collected
samples were wrapped in clean aluminum foil, sealed, labeled,
and subsequently stored at −20 °C until analysis. Pre-cleaned PUF,
Kimwipes, and polyester socks, which had been exposed by unsealing
the aluminum foil wrap during sample retrievals, were treated as field
blanks.
2.3. Target compounds
In this paper, we have focused on the following compounds:
polybrominated diphenyl ethers (congeners 28, 47, 66, 85, 99,
100, 153, 154, 183, and 209), hexabromobenzene (HBB), ptetrabromoxylene
(p-TBX), pentabromobenzene (PBBz), 2-ethylhexyl-
2,3,4,5-tetrabromobenzoate (EHTBB), bis(2-ethylhexyl)tetrabromophthalate
(BEHTBP), 1,2-bis(2,4,6-tribromophenoxy)ethane (BTBPE),
decabromodiphenylethane (DBDPE), and Dechlorane Plus (syn and
anti). All the analytical standards [except BDE-118, which was purchased
from AccuStandard (AccuStandard, New Haven, CT)], were purchased
from Wellington Laboratories, Guelph, Canada.
2.4. Sample analysis
The U.S. samples were analyzed in Bloomington, Indiana, U.S.
(Indiana University, IU), and the Canadian and Czech samples were
analyzed in Brno, Czech Republic (RECETOX). Details of the analytical
protocols are given in the SI and are summarized here. Before
extraction, all samples were spiked with known amounts of recovery
standards. Socks with dust were weighed, the dust was sieved to
b500 μm, approximately 100 mg were weighed, and the excess dust
was stored in an aluminum foil packet for future use. The sock was
rinsed with solvent (30 mL hexane in acetone, 1:1), and the solvent
was combined with weighed dust. Dust was sonicated in 30 mL of acetone
in hexane 1:1 (v:v); left to settle for 30 min, and the supernatant
was decanted. The procedure was repeated 2 additional times with
10 mL of solvent, and the extracts were combined. At IU, the extract
was rotary evaporated to 2 mL and then fractionated on a silica column
(3.5% water deactivated) using 25 mL of hexane, 25 mL hexane in dichloromethane
1:1 (v:v), and 25 mL of dichloromethane in acetone
3:7 (v:v) as eluting solvents. At RECETOX, the volume of the combined
extracts were reduced under a N2 stream and separated by weight to
two aliquots. The first aliquot was 70% of the extract, and it was treated
with sulfuric acid-modified silica. The remaining 30% of the extract was
cleaned using a standard non-modified silica column. PUF and
Kimwipes samples were Soxhlet extracted with 400 mL of acetone in
hexane 1:1 (v:v) for 24 h at IU (Peverly et al., 2015) and with 250 mL
151M. Venier et al. / Environment International 94 (2016) 150–160
of dichloromethane using automated warm Soxhlet extraction (Büchi
B-811, Switzerland) at RECETOX. Extracts were fractionated on silica
columns as per the dust procedure and eluted with 25 mL of hexane,
and 25 mL hexane: dichloromethane 1:1 (v:v) at IU and with 20 mL of
dichloromethane at RECETOX. Extracts were analyzed using GC/MS at
IU and GC/HRMS at RECETOX, using previously published methods
(Lohmann et al., 2013; Ma et al., 2013). More analytical details are reported
in the Supporting Information.
2.4.1. QA/QC
Several measures were taken to ensure data comparability between
the different laboratories and the accuracy and reliability of the
measurements (see Fig. S3 and Table S1 in the Supporting Information
and the related detailed discussion). For the non-PBDE halogenated
flame retardants, both the U.S. and Czech laboratories participated
in an interlaboratory comparison study [called INTERFLAB Phase I
(Melymuk et al. (2015)]. PBDE data from the Czech Republic and
Canadian homes were obtained using an isotope dilution method, and
the results were recovery corrected. Results were not recovery
corrected at IU. Average surrogate recoveries were mostly within the
50–150% range. Solvent blanks were used to evaluate contamination
from the laboratory procedures. Blank levels were low, and no correction
was necessary. Three field blanks per matrix per country were collected
and analyzed. The concentrations of the target compounds in
each sample were then compared to the average concentrations in the
field blanks (on a country and matrix specific basis — see Table S2 in
the Supporting Information for more details) and treated as follows: If
the blank level was b10% of the measured level, there was no correction.
If the blank level was 10–35% of the measured level, the blank level was
subtracted from the measured level. If the blank level was N35% of the
measured level, the value was reported as “non-detect”
2.5. Data analysis
Basic and descriptive statistics were calculated using Minitab, and
Microsoft Excel software. Plots were generated using SigmaPlot 13
(Systat Software Inc.).
3. Results and discussion
A summary of the results for air, dust, and window films at the three
locations is reported in Tables 1–3, respectively. Boxplots for selected
compounds are shown in Figs. 1–3.
3.1. PBDEs
BDE-209 was generally the most abundant PBDE congener,
followed by the sum of the congeners representing the Penta-BDE
commercial mixture (the sum of congeners 28, 47, 66, 85, 99, 100,
153, and 154) — see Fig. 1. The only exceptions to this trend were
window film samples from the Czech Republic (CZ), where the levels
of BDE-209 were lower than those of Penta-BDEs.
The flame retardant concentrations in the U.S. samples were significantly
higher than those measured in samples from the Czech Republic,
with Canada (CAN) typically being in between (see Fig. 1). We observed
the geographic trend US N CAN N CZ for BDE-209 in dust and window
film and for total PBDEs in dust and air (p b 0.05). However, the trend
was US ≈ CAN N CZ for Penta-BDE in air and dust and for total PBDEs
in window film. These geographic trends can be linked to flame retardant
use patterns in different regions of the world. North America was
the largest global user of both Penta-BDE and Deca-BDE, whereas
Europe's usage was far lower (Abbasi et al., 2015). The fact that DecaBDE
has been produced in the largest volumes of any commercial PBDEs
accounts for the high relative contribution of BDE-209 in all the samples
(Earnshaw et al., 2015).
BDE-209 was the most abundant PBDE congener in dust and window
film, and BDE-47 and Penta-BDE congeners were the most abundant
in air. This distribution is consistent with differences in their
vapor pressures and log KOA; the vapor pressure of BDE-209 is
6.2 × 10−10
Pa at 25 ° C, but that of BDE-47 is 3.2 × 10−5
Pa at 25 °C
and their log KOA values are 10.686 and 18.423, respectively (Episuite
(2012).
An overview of available literature data for indoor studies is given in
the Supporting Information (see Table S3). In general, for all matrices,
the flame retardant concentrations for the U.S. and Canadian samples
Table 1
Summary results for air samples (pg/m3
) including mean with standard error, median, minimum, maximum, and number of detects. The ANOVA results (calculated on log transformed
data) are given as a–c letters in each line; concentrations for compounds that share a letter are not significantly different from one another at p b 0.05.
Air (pg/m3
)
US CAN CZ
Mean ± SE Median Range N Mean ± SE Median Range N Mean ± SE Median Range N
BDE-28 8.5 ± 5.1 2.2 ND–145 28 b 18 ± 5.4 7.6 1.0–166 34 a 1.2 ± 0.3 0.58 0.10–9.2 29 c
BDE-47 128 ± 32 52 4.5–818 30 a 99 ± 27 39 5.4–759 34 a 3.0 ± 0.7 1.6 0.56–16 29 b
BDE-100 19 ± 10 4.2 ND–272 29 a 4.7 ± 1.0 1.8 0.38–22 34 b 0.19 ± 0.04 0.11 0.05–1.0 29 c
BDE-99 95 ± 51 15 ND–1291 26 a 15 ± 3.3 5.3 1.3–73 34 b 0.46 ± 0.07 0.29 0.16–1.4 29 c
BDE-154 7.6 ± 4.7 0.62 ND–122 27 a 0.80 ± 0.22 0.36 ND–4.4 22 b 0.045 ± 0.003 0.045 ND–0.056 9 c
BDE-153 11 ± 6 5.0 ND–71 12 a 0.60 ± 0.17 0.26 0.06–5.7 34 b 0.072 ± 0.008 0.069 ND–0.10 6 c
BDE-183 2.8 ± 0.7 2.5 ND–5 4 a 1.3 ± 0.32 1.3 ND–1.6 2 b 0.12 ± 0.01 0.12 ND–0.23 28 c
BDE-209 836 ± 475 257 ND–5461 11 a 83 ± 31 49 ND–217 6 ab 11 ± 2 9.4 ND–15 3 b
Penta-BDE 254 ± 85 79 4.9–2291 30 a 140 ± 36 59 8.3–1024 34 ab 5.0 ± 1.1 2.9 0.94–28 29 b
TotBDE 561 ± 202 148 4.9–5756 30 a 155 ± 40 60 8.3–1149 34 b 6.3 ± 1.2 3.0 1.1–28 29 b
pTBX 1.8 ± 0.3 1.9 ND–2.4 6 a 0.66 ± 0.19 0.38 ND–3.4 19 b 0.31 ± 0.03 0.29 ND–0.71 26 b
PBBZ 3.8 ± 1.1 2.5 ND–12 11 b 8.8 ± 2.5 4.8 1.3–74 34 a 4.9 ± 0.8 3.4 0.76–17 29 ab
PBEB 9.0 ± 4.4 1.9 ND–92 21 a 4.0 ± 0.30 3.9 0.95–7.4 34 a 0.66 ± 0.1 0.49 ND–1.4 8 b
HBB 9.6 ± 2.7 4.0 ND–58 26 a 9.9 ± 2.2 5.8 0.85–60 34 a 7.2 ± 1.7 4.6 ND–31 19 a
Tot bromobenzenes 17.6 ± 4.1 9.1 ND–98 28 b 23 ± 3.5 16 5.65–91 34 a 10 ± 1.9 7.6 0.8–41 29 b
EHTBB 23 ± 6 9.2 1.3–142 30 ab 32 ± 10 12 3.1–291 34 a 6.6 ± 0.7 5.5 ND–15 28 b
BEHTBP 16 ± 7 6.0 ND–109 17 a 6.8 ± 2.2 3.1 ND–43 22 ab 3.0 ± 0.3 3.2 ND–5.4 15 b
EHTBB + BEHTBP 32 ± 9 14 1.3–212 30 ab 37 ± 11 14 3.1–293 34 a 8.2 ± 0.8 7.2 ND–17 28 b
BTBPE 0.18 ± 0.03 0.083 ND–0.74 27 c 1.3 ± 0.39 1.0 ND–2.4 4 a 0.49 ± 0.12 0.38 ND–1.4 9 b
syn-DP 0.70 ± 0.26 0.37 ND–4.0 14 b 28 ± 4.9 23 ND–76 22 a
anti-DP 8.2 ± 3.9 4.1 ND–23 5 b 43 ± 13 25 ND–243 18 ab 65± 65 ND–65 1 a
DPsum 3.4 ± 1.8 0.37 ND–27 15 b 61 ± 15 38 ND–316 23 a 65± 65 ND–65 1 a
DBDPE 42 ± 28 42 ND–71 2 a 13 ± 2.6 9.2 ND–74 29 b
fanti 0.87 ± 0.01 0.86 0.85–0.91 4 a 0.50 ± 0.03 0.46 0.3–0.77 17 b
fEHTBB 0.64 ± 0.05 0.63 0.16–0.95 17 b 0.80 ± 0.04 0.83 0.3–0.99 22 b 0.69 ± 0.05 0.72 0.27–0.89 15 ab
152 M. Venier et al. / Environment International 94 (2016) 150–160
are within the same range of values previously reported, but the concentrations
for the Czech Republic are lower than those found in the literature
for other European countries. This anomaly may be a result of an
over-representation of the United Kingdom (UK) in the European measurements.
In the UK, the levels for FRs have been historically higher
than the rest of Europe because of stricter flammability standards compared
to continental or central Europe (Harrad et al. (2008b). An alternative
explanation is that many of these UK measurements were
obtained about 10 years ago, when the flame retardant market was
still heavily dominated by PBDEs.
3.2. EHTBB and BEHTBP
EHTBB and BEHTBP are the two main brominated components of
two commercial mixtures, Firemaster 550 and Firemaster BZ-54,
which are now considered major replacements of the withdrawn
Table 2
Summary results for dust (ng/g) including mean with standard error, median, minimum, maximum, and number of detects. The ANOVA results (calculated on log transformed data) are
given as a–c letters in each line; concentrations for compounds that share a letter are not significantly different from one another at p b 0.05.
DUST (ng/g)
US CAN CZ
Mean ± SE Median Range N Mean ± SE Median Range N Mean ± SE Median Range N
BDE-28 12 ± 2.2 6.9 0.38–50 30 a 9.0 ± 3.0 3.9 0.21–93 35 b 0.34 ± 0.08 0.23 ND–1.2 18 c
BDE-47 365 ± 60 271 20–1260 30 a 480 ± 127 230 19–3580 35 a 6.3 ± 2.4 3.8 ND–65 26 b
BDE-100 86 ± 12 65 3.2–270 30 a 115 ± 32 42 3.9–901 35 a 1.3 ± 0.59 0.53 ND–17 28 b
BDE-99 634 ± 142 336 20–2800 30 a 534 ± 139 221 23–3830 35 a 7.0 ± 3.1 2.5 ND–84 28 b
BDE-154 35 ± 5.4 30 ND–128 28 a 46 ± 12 18 1.8–344 35 a 0.75 ± 0.29 0.33 ND–6.0 23 b
BDE-153 48 ± 7.6 32 1.7–178 30 a 61 ± 16 25 2.6–479 35 a 1.3 ± 0.41 0.77 ND–8.9 26 b
BDE-183 12 ± 1.5 11 ND–37 29 a 31 ± 8.4 13 2.3–255 35 a 3.4 ± 0.68 1.9 ND–15 24 b
BDE-209 2780 ± 365 2220 75–7450 30 a 1210 ± 206 713 223–4860 35 b 223 ± 39 139 16–788 30 c
Penta-BDE 1210 ± 220 735 47–4410 30 a 1310 ± 338 514 53–8360 35 a 16 ± 6.5 8.0 0.05–193 30 b
TotBDE 4000 ± 446 3650 122–9730 30 a 2550 ± 397 1770 284–9610 35 b 241 ± 40 163 18–797 30 c
pTBX 0.32 ± 0.11 0.21 ND–1.3 11 a 0.39 ± 0.12 0.50 ND–0.53 3 a
PBBZ 2.6 ± 0.68 1.6 ND–15 22 a 3.3 ± 0.92 3.2 ND–5.6 4 a 0.44 ± 0.10 0.29 ND–2.8 28 b
PBEB 1.4 ± 0.40 0.60 ND–8.4 26 a 1.2 ± 0.12 1.2 ND–1.7 8 a
HBB 13 ± 2.4 7.2 0.92–46 30 a 6.4 ± 1.3 6.1 ND–11 6 a 2.1 ± 0.45 1.4 ND–6.4 13 b
Tot bromobenzenes 16 ± 2.3 13 1.2–48 30 a 3.9 ± 1.0 1.9 ND–13 16 b 1.4 ± 0.37 0.7 ND–9.2 28 c
EHTBB 918 ± 520 240 ND–15,400 29 b 2410 ± 613 966 121–15,300 35 a 17 ± 5.4 7.8 ND–150 28 c
BEHTBP 2540 ± 900 624 112–22,800 30 a 2650 ± 1170 431 69–34,500 35 a 60 ± 13 42 ND–373 29 b
EHTBB + BEHTBP 3423 ± 1065 797 155–23,648 30 a 5058 ± 1427 1570 194–36,047 35 a 77 ± 14 55 ND–395 29 b
BTBPE 22 ± 7.7 8.5 ND–204 29 a 27 ± 6.8 12 ND–157 31 a 5.8 ± 1.1 3.9 ND–29 28 b
syn-P 5.0 ± 0.80 3.3 ND–16 29 b 8.8 ± 3.1 4.6 ND–99 31 b 24 ± 10 15 ND–62 5 a
anti-DP 45 ± 15 14 ND–311 28 a 35 ± 18 15 ND–634 34 a 45 ± 15 20 ND–215 14 a
DPsum 48 ± 15 18 ND–322 29 a 43 ± 21 22 ND–732 34 a 53 ± 19 20 ND–277 14 a
DBDPE 367 ± 119 148 ND–3140 29 a 95 ± 60 15 ND–2060 34 b 20 ± 7 4.7 ND–114 23 c
fanti 0.81 ± 0.02 0.8161 0.57–0.97 28 a 0.71 ± 0.03 0.71 0.35–0.92 31 b 0.81 ± 0.01 0.81 0.78–0.86 5 ab
fEHTBB 0.30 ± 0.036 0.28 0.0075–0.81 29 b 0.6 ± 0.041 0.71 0.044–0.91 35 a 0.24 ± 0.043 0.14 0.054–0.68 28 b
Table 3
Summary results for window film (ng/m2
) including mean with standard error, median, minimum, maximum, and number of detects. The ANOVA results (calculated on log transformed
data) are given as a–c letters in each line; concentrations for compounds that share a letter are not significantly different from one another at p b 0.05.
Window film (ng/m2
)
US CAN CZ
Mean ± SE Median Range N Mean ± SE Median Range N Mean ± SE Median Range N
BDE-28 0.14 ± 0.04 0.091 ND–0.75 21 a 0.13 ± 0.029 0.1 ND–0.65 27 a 0.0053 ± 0.0006 0.0049 ND–0.011 14 b
BDE-47 5.3 ± 1.4 2.7 ND–28 21 a 5.8 ± 3.5 3.5 ND–13 3 a 0.46 ± 0.069 0.39 ND–1.1 14 b
BDE-100 1.8 ± 0.5 0.71 ND–7.6 20 a 0.69 ± 0.38 0.69 ND–1.1 2 ab 0.14 ± 0.025 0.10 ND–0.4 14 b
BDE-99 5.1 ± 1.0 3.0 ND–20 20 a 2.5 ± 1.2 2.5 ND–3.7 2 ab 0.74 ± 0.13 0.52 ND–2.0 14 b
BDE-154 0.33 ± 0.06 0.23 ND–1.3 22 a 0.23 ± 0.05 0.27 ND–0.35 6 a 0.058 ± 0.01 0.042 ND–0.17 19 b
BDE-153 0.74 ± 0.27 0.30 ND–4.1 19 a 0.33 ± 0.06 0.30 ND–0.55 7 a 0.077 ± 0.02 0.044 ND–0.27 23 b
BDE-183 0.35 ± 0.08 0.28 ND–1.1 13 a 0.50 ± 0.14 0.17 ND–3.9 33 a 0.014 ± 0.001 0.013 0.003–0.028 29 b
BDE-209 45 ± 9 53 ND–54 3 a 12 ± 2.9 5.5 ND–69 33 b 0.61 ± 0.14 0.40 ND–3.1 21 c
Penta-BDE 9.9 ± 2.4 6.1 ND–60 29 a 1.1 ± 0.7 0.14 ND–19 28 b 0.98 ± 0.22 0.86 ND–4.1 23 b
TotBDE 14 ± 4 7.0 0.02–78 30 a 13 ± 3 6.5 1.1–73 34 a 1.2 ± 0.24 0.98 0.005–5.1 29 b
pTBX 0.053 ± 0.026 0.044 ND–0.12 4
PBBZ 0.031 ± 0.006 0.027 ND–0.074 10 b 0.098 ± 0.017 0.067 0.015–0.27 25 a 0.0051 ± 0.0008 0.0042 0.001–0.02 28 c
PBEB 0.028 ± 0.006 0.016 ND–0.1 22 b −
HBB 0.60 ± 0.23 0.12 ND–2.9 17 a 0.58 ± 0.27 0.24 ND–8.1 29 a 0.036 ± 0.003 0.038 ND–0.056 19 b
Tot Bromobenzenes 0.48 ± 0.17 0.11 ND–2.9 24 a 0.57 ± 0.27 0.27 ND–8.1 32 a 0.0285 ± 0.004 0.032 0.001–0.069 29 b
EHTBB 6.0 ± 1.6 2.0 ND–34 28 a 5.0 ± 1.3 2.4 0.59–35 34 a 0.047 ± 0.007 0.041 ND–0.099 14 b
BEHTBP 13 ± 3 6.4 ND–46 18 a 2.0 ± 0.9 0.83 ND–20 22 b 0.30 ± 0.07 0.11 ND–1.2 26 c
EHTBB + BEHTBP 18 ± 3 11 ND–68 29 a 6.3 ± 1.8 3.2 0.61–54 34 b 0.31 ± 0.07 0.11 ND–1.3 26 c
BTBPE 0.50 ± 0.22 0.16 ND–3.7 19 a 0.56 ± 0.22 0.15 ND–5.7 28 a 0.032 ± 0.009 0.017 ND–0.23 28 b
synP 0.92 ± 0.40 0.40 ND–6.5 16 c 12 ± 5 6.7 ND–28 5 a 2.8 ± 1.5 0.97 ND–12 7 b
antiDP 1.9 ± 0.7 1.1 ND–8.8 12 b 9.4 ± 2.5 7.2 ND–21 7 a 4.5 ± 2.5 0.99 ND–22 10 b
DPsum 2.2 ± 0.9 1.5 ND–15 17 b 16 ± 5.0 8.4 ND–48 8 a 6.5 ± 3.6 1.4 ND–34 10 ab
DBDPE 9.7 ± 3.4 5.4 ND–56 15 a 1.6 ± 0.8 0.60 ND–6.8 8 b 0.15 ± 0.06 0.068 ND–0.84 13 c
fanti 0.65 ± 0.06 0.7108 0.21–0.859 11 a 0.462 ± 0.0 0.45 0.39–0.57 4 a 0.488 ± 0.07 0.45 0.31–0.84 7 ab
fEHTBB 0.32 ± 0.04 0.32 0.026–0.86 25 b 0.74 ± 0.03 0.77 0.28–0.95 22 a 0.14 ± 0.02 0.15 0.026–0.25 13 c
153M. Venier et al. / Environment International 94 (2016) 150–160
Penta-BDE commercial mixture (Stapleton et al., 2008). BEHTBP is also
sold and used separately as a flame retardant in the commercial mixture
DP-45. Recently, Abbasi et al. (2016) reported finding both these compounds
in electronic equipment such as personal computers and televisions.
Among the alternative flame retardants measured in this study,
EHTBB and BEHTBP were the major non-PBDE components in dust
and air but not in window film. Concentrations of EHTBB + BEHTBP
were generally higher in the U.S. and Canada than in the Czech
Republic, an observation which reflects the higher usage of FRs in
North America than in Europe (see Fig. 2).
The EHTBB fraction was calculated from the concentrations of EHTBB
and BEHTBP as follows: fEHTBB = EHTBB / (EHTBB + BEHTBP), and these
values are plotted in Fig. 2, rightmost boxes. The median ranges were
0.15–0.77 in window film, 0.14–0.71 in dust, and 0.63–0.83 in air.
Fig. 1. Boxplots of concentrations for selected PBDEs congeners. The plots share the same y-axis concentration scales although the units are different for each matrix: dust is ng/g, window
film is ng/m2
, and air is pg/m3
. Within each matrix, boxes that share the same letter are not statistically significantly different at a 5% level in an ANOVA analysis using the Tukey's test.
Fig. 2. Boxplots of concentrations for EHTBB, BEHTBP and the fEHTBB ratio. The leftmost three plots share the same left y-axis concentration scale, and the rightmost three plots share the
same right y-axis scale. The units are different for each matrix: dust is ng/g, window film is ng/m2
, and air is pg/m3
. Within each matrix, boxes that share the same letter are not statistically
significantly different at a 5% level in an ANOVA analysis using the Tukey's test.
154 M. Venier et al. / Environment International 94 (2016) 150–160
High ratios indicate the predominance of EHTBB and low ratios the predominance
of BEHTBP. Low ratios were also recently observed in indoor
samples from Norway (Cequier et al., 2014). In this study, the fEHTBB
ratio was generally highest in samples from Canada and lowest in
those from the Czech Republic. Previous studies have reported values
of the fEHTBB ratio of 0.69 in outdoor air in Toronto (Shoeib et al.,
2014), 0.70 in outdoor air in Chicago (Peverly et al., 2015), 0.26–0.54
in outdoor air from the Great Lakes area (Ma et al., 2012), and 0.59–
0.81 in indoor dust (Stapleton et al., 2008). Cao et al. (2014) reported
variable fEHTBB ratios with different types of dust. In commercial mixtures,
the fEHTBB ratio was 0.77 ± 0.03 in Firemaster 550 (Ma et al.,
2012) and 0.70 in Firemaster BZ-54.
The fEHTBB ratios measured in this study for air are similar to those reported
in previous studies, but the ratios for window wipes and dust are
lower.
A higher fEHTBB ratio is expected in air than in window films and dust
because of the different volatilities of these two compounds: EHTBB is
expected to have a higher vapor pressure than BEHTBP; hence, EHTBB
is likely to be relatively more concentrated in air than BEHTBP. Another
explanation for the lower ratios observed for window films and dust
could be more photodegradation of EHTBB relative to BEHTBP (Davis
and Stapleton, 2009).
The differences in the fEHTBB ratio among countries (high in Canada
and low in the Czech Republic) can be explained by looking at sources:
lower fEHTBB ratios could be due to BEHTBP-specific sources. BEHTBP
was used as flame retardant in neoprene and polyvinylchloride
(Andersson et al., 2006), and in the commercial mixture DP-45,
BEHTBP was the sole component. However, the concentrations of
EHTBB and BEHTBP were significantly correlated (Pearson correlation
coefficient = 0.939, p b 0.0001; see Fig. S4), suggesting that they have
a similar source. Previous studies have shown that the concentrations
of EHTBB and BEHTBP were significantly correlated in dust from
Belgium and from UK homes (Ali et al., 2011a), but other studies
found no such correlations (Ali et al., 2011b; Shoeib et al., 2012). More
data are needed to reconcile the seemingly contradictory information
provided by the lower than expected ratios and yet significant
correlation.
3.3. Dechlorane Plus
Dechlorane Plus (DP) is a chlorinated flame retardant used mainly in
electronics and electrical cable coatings as an alternative to Deca-BDE
(Dodson et al., 2012). Boxplots of the concentrations of DP in all the
samples are shown in Fig. 3. DP was abundant in both the U.S. and
Canadian air samples, with median total DP concentrations of 0.37 and
38 pg/m3
, respectively. DP was detected in only one Czech air sample.
In this case, only the anti isomer was detected at a relatively high
concentration of 65 pg/m3
. Interestingly, in the U.S. and Canadian samples,
the syn isomer was detected more often than the anti isomer but at
lower concentrations. In the window wipe samples, the highest levels of
DP were found in Canada, with a median level of 8.4 ng/m2
. However,
the Canadian samples also had the lowest DP detection frequency at
only 24%. In U.S. and Czech window wipes samples, DP was found in
57 and 33% of samples, respectively, at similar levels (medians: 1.5
and 1.4 ng/m2
, respectively). DP was detected in most of the U.S. and
Canadian dust samples, but only in half of the Czech samples. Nevertheless,
levels in dust between countries are statistically indistinguishable
(p N 0.05), with medians of 18, 22, and 20 ng/g in the U.S., Canadian,
and Czech samples, respectively.
There are only a few studies that have measured DP levels in indoor
samples and none that measured it in windows wipes (see Table S3).
Cequier et al. (2014) reported DP levels of b10 pg/m3
in indoor air in
Norway using active samplers, levels which are similar to the results reported
here. Concentrations of DP in dust varied largely based on where
they were collected. Median DP values in the U.S. dust were 4.5–10 ng/g
(Dodson et al., 2012; Johnson et al., 2013). In China, median DP levels
were 4.0–65 ng/g in rural and urban areas, but they reached values as
high as 3500 ng/g in the vicinity of an e-waste recycling facility (Chen
et al., 2014; Li et al., 2015; Zheng et al., 2010, 2015). Newton et al.
(2015) measured DP concentrations of b10 ng/g in dust samples in
Sweden. In Canada, Abbasi et al. (2016) reported DP concentrations of
up to 283 ng/g (with a geomean value of 8.6 ng/g), values which are
comparable to those reported by Shoeib et al. (2012). In general, our
levels for dust are similar to those previously reported for the U.S. and
for rural and urban China, but our levels are much lower than those
around the Chinese e-waste recycling facility.
The fanti ratios (the ratio of the anti-DP concentration to the total DP
concentration) in these samples are shown in Fig. 3, rightmost boxes.
For reference, fanti = 0.75 for the U.S. commercial product and 0.60 for
the Chinese product (Sverko et al., 2011). The medians for the fanti ratios
in dust samples were 0.82 in the U.S. samples and 0.71 in the Canadian
samples, values that were significantly different from one another
(p b 0.05). The medians for the fanti ratios in window wipes samples
ranged between 0.45 for the Canadian and Czech Republic samples to
0.71 for the U.S. samples, but they were statistically indistinguishable
(p N 0.05). The median fanti ratios in the air samples were significantly
higher in the U.S. samples than in the Canadian samples (0.86 vs. 0.46;
p b 0.001). The values of the fanti ratio in these samples suggest that
the source of DP in the U.S. and Canada are products containing DP
manufactured in North America rather than imported from China.
Cequier et al. (2014) reported fanti in the range 0.65–0.75. Newton
et al. (2015) reported fanti as low as 0.20 in a dust sample collected
from a Stockholm home in an area around electrical cables. Abbasi
et al. (2016) found fanti of 0.69 in dust samples from 35 Toronto
Fig. 3. Boxplots of concentrations for syn-DP, anti-DP and the fanti ratio. The leftmost three plots share the same left y-axis concentration scale, and the rightmost three plots share the same
right y-axis scale. The units are different for each matrix: dust is ng/g, window film is ng/m2
, and air is pg/m3
. Within each matrix, boxes that share the same letter are not statistically
significantly different at a 5% level in an ANOVA analysis using the Tukey's test.
155M. Venier et al. / Environment International 94 (2016) 150–160
homes. Cao et al. (2014) reported fanti of 0.8 in indoor and outdoor dust
samples in Beijing, China. Several fanti values lower than in the commercial
mixtures were also reported in environmental compartments by
Sverko et al. (i.e. outdoor air), who suggested preferential degradation
of the anti isomer (Sverko et al., 2011).
3.4. Bromobenzenes
This group includes hexabromobenzene (HBB), pentabromoethylbenzene
(PBEB), pentabromobenzene (PBBz) and tetrabromop-xylene
(pTBX). Information about current and past uses of
bromobenzenes is scarce, making it difficult to investigate their sources
and fates. HBB is used in paper, wood, textiles, electronics, and plastic
goods (Covaci et al., 2011). PBEB is an additive flame retardant used in
thermoset polyester resins for applications such as circuit boards, textiles,
adhesives, and wire and cable coatings (Covaci et al., 2011).
Abbasi et al. (2016) reported PBBz and PBEB in about 30 and b10% of
surface wipes taken from electronic products sampled in 35 and 10 Toronto
homes and offices, respectively. Interestingly, they found PBBz in
the casings of four cathode ray tube (CRTs) televisions that were
N10 years old. In our study, the median sum of bromobenzene concentrations
ranged between 7.6 pg/m3
for Czech and 15.7 pg/m3
in Canadian
air samples, between 0.032 ng/m2
for Czech and 11.5 ng/m2
for
Canadian window film samples, and between 0.70 ng/g for Czech and
15.7 ng/g for U.S. in dust (see Tables 1–3 for details). In general, the concentrations
of total bromobenzenes were significantly lower in the
Czech samples compared to U.S. and Canadian samples (p b 0.001). In
air, bromobenzene concentrations in samples from Canada were significantly
higher than those from the U.S. and from the Czech Republic
(p b 0.001).
Among the bromobenzenes, PBEB was not detected in the Czech
dust samples or in the window wipes from Canada or the Czech
Republic. In air, HBB and PBEB concentrations in the U.S. and Canadian
samples were similar to each other, and they were significantly higher
than those in the Czech Republic (p b 0.05). Both the levels and the
spatial trends for bromobenzenes confirm their relatively low current
or past usage in the flame retardant market compared to PBDEs and
other alternative compounds such as EHTBB or BEHTBP, as well as
their relatively higher usage in North America compared to Europe.
3.5. Decabromodiphenylethane (DBDPE)
DBDPE was introduced in the market as a replacement for BDE-209
in the early 1990s, and it was first identified in the environment in 2003
(Ricklund et al., 2008). Environmental data suggest that BDE-209 and
DBDPE have been used independently and that the substitution
of DBDPE for BDE-209 has not progressed very far, especially in the
U.S. -which could be due to the lack of regulations (Ricklund et al.,
2008). Based on DBDPE's low vapor pressure (Dungey and Akintoye,
2007), we expected to find DBDPE mostly in the dust and window
wipes samples. Rather, DBDPE was detected in 85% of the Canadian air
samples, in 2 of the U.S. samples, and in none of the Czech samples
(see Tables 1-3). The median level for the Canadian air samples was significantly
lower than for the U.S. samples (9.2 pg/m3
vs. 42 pg/m3
). In
the dust and window film samples, the concentration of DBDPE followed
the spatial trend: US N CAN N CZ (p b 0.01). Median concentrations in
dust ranged from 148 ng/g in the U.S. to 4.73 ng/g in the Czech Republic
samples and in window film from 5.40 ng/m2
in the U.S. to 0.068 ng/m2
in the Czech Republic. The relatively high variability in the measurement
of this compound might be contributing to some of these geographic
patterns (Melymuk et al., 2015).
Concentrations of DBDPE in dust in this study are consistent with
previous studies that reported median concentrations of 51–138 ng/g
in the U.S. (Dodson et al., 2012; Stapleton et al., 2008), a mean and geometric
mean of 221 and 0.5 (range: bdetection-5500) ng/g in Canada
(Abbasi et al., 2016), and 10–200 ng/g in Europe (Ali et al., 2011a;
Newton et al., 2015) (see Table S3 for a more extensive list of literature
values). DBDPE literature data in air are scarce: Cequier et al. (2014) reported
a median of 8.30 pg/m3
in classrooms in Norway, and Newton
et al. (2015) reported a median of b90 pg/m3
in indoor air in Sweden.
Results from this study are comparable. No literature data exist to date
for DBDPE in window wipes.
Fig. S5 shows a strong positive relationship between the concentrations
of DBDPE and BDE-209, which is the opposite of what would be
expected if DBDPE was entering the flame retardant market primarily
as a BDE-209 replacement. These data confirm that both these compounds
have been used simultaneously both in North America and in
Europe.
3.6. 1.2-Bis(2,4,6-tribromophenoxy)ethane (BTBPE)
BTBPE is an additive flame retardant marketed as FF-680 by the
Great Lakes Chemical Corporation, now part of Chemtura (Covaci
et al., 2011). BTBPE has been used since the 1980s as a replacement
for the Octa-BDE commercial mixture.
Fig. 4. Bar plot showing the median concentration of the main flame retardants targeted in
this study. The values used for this plot are those reported in Tables 1-3. Asterisks indicate
compounds detected with low detection frequency in the samples.
156 M. Venier et al. / Environment International 94 (2016) 150–160
BTBPE was detected with high frequency in all three matrices in all
three countries (with the exception of air in Canada and Czech
Republic), suggesting that this compound is ubiquitous in indoor environments.
BTBPE levels were higher in the U.S. and Canadian samples
than in the Czech samples for both dust and window film (p b 0.01).
In air, the highest levels were observed in the Canadian samples and
the lowest in the Czech Republic samples (p b 0.01), albeit with low
detection frequencies in both cases. Levels measured in this study are
similar to those reported in previous studies for dust (see Table S3).
No comparable data for air or window wipes currently exist. These
results indicate that BTBPE was more heavily used in North America
than in Europe. Concentrations of BTBPE were on the same order as
those of BDE-183, for which BTBPE is a replacement.
3.7. Summary of all flame retardants
An overview of the most abundant FRs in the three phases and
three countries is shown in Fig. 4, which depicts both geographical
trends and compound patterns. When considering all FRs together,
the median concentrations in the U.S. are higher than those in
Canada and in the Czech Republic for all three phases. As mentioned
above, this geographic trend reflects market and usage patterns. Historically,
the U.S. has been one of the major producers and consumers
of flame retardants (Abbasi et al., 2015). In Europe, in
general, the United Kingdom was the largest market for FRs,
explaining the relatively low levels observed for the Czech samples
in this study (Abbasi et al., 2015).
Among individual compounds, Penta-BDE, BDE 209, EHTBB and
BEHTBP generally showed the highest median concentrations (see
Fig. 4). It is not surprising that Penta-BDE and BDE 209 still play a major
role in the FR arena: Abbasi et al. (2015) estimated that total consumption
of Penta-BDEs and BDE 209 from 1970 to 2020 was about 46,000 and
about 380,000 t, respectively. No specific market and usage data are
available for EHTBB and BEHTBP; however, because they are the major
replacement for Penta-BDE, a large production volume is likely.
Other compounds were measured in these samples at relatively high
median concentrations (e.g. DBDPE and DP), but it should be noted that
their detection frequencies were low in some phases; thus, the plot in
Fig. 4 for these compounds should be interpreted with caution.
Fig. 5. Dependence of the dust/gas phase partition coefficients and of the window wipes/gas phase partition coefficients on the octanol/air partition coefficients (log KOA). In both graphs,
each circle represents the average, and the bars the standard error.
157M. Venier et al. / Environment International 94 (2016) 150–160
3.8. Within house distribution
In each country, samples from both the main bedroom and the living
room area were collected in a subset of houses (n = 10 in each country).
A side-by-side bar plot showing median concentrations in the bedrooms
and living rooms for the three phases for PBDEs and for alternative
flame retardants is showed in Fig. S6. While the bar plots suggest
that bedrooms had higher levels of flame retardants than living rooms
(especially for dust), a paired t-test for each matrix individually showed
that the levels of FRs in bedrooms and living rooms were not significantly
different (p N 0.05). Muenhor and Harrad (2012) and Harrad et al.
(2009) reported some room-to-room differences, albeit not statistically
significant, and they suggested that the dust concentrations were related
to the number of specific items (such as electronic products) in each
room. Coakley et al. (2013) reported an association between living
room floor dust and mattress dust for BDE-183, BDE-206 and BDE-
209. Ali et al. (2012) found significant differences between floor and
mattress dust for BEHTBP and DBDPE but not for EHTBB or BTBPE. To
our knowledge, no other study did a systematic room to room comparison
on a relatively large number of samples (n = 30).
3.9. Correlation between air, dust, and window film concentrations
Spearman rank analysis gave good correlations between flame retardant
concentrations in air, dust, and window films, especially for PBDEs
(see SI Tables S4-S6). The correlation between concentrations in air and
window film was the least powerful, showing significant relationships
only for PBDEs, including BDE-209. The other FR concentrations were
either not correlated, or their correlation was only marginally
significant (0.05 b p b 0.09). The correlation between dust and air and
dust and window film concentrations was generally stronger and
encompassed a larger range of compounds. In fact, only four compounds
were not significantly correlated (pTBX, PBEB, PBBz, and DPs), probably
due to their relatively low detection frequencies (and hence low number
of pairs for the correlation analysis). It is not surprising to find significant
correlations between gas phase and window film concentrations
because the thin layer of film allows for rapid equilibration between
these two media (Csiszar et al., 2012). The correlation between air and
dust concentrations was generally significant for lower brominated
BDEs and low molecular weight compounds.
Under equilibrium conditions, the partitioning between dust and air
and between window film and air is expected to be proportional to log
KOA, which has been used to explain the sorption of chemicals to organic
matter in dust (Weschler and Nazaroff, 2010). Fig. 5 shows the dependence
on log KOA of the partition coefficients between dust and air concentrations
(top panel) and of the partition coefficients between
window wipe and air concentrations (bottom panel); see Table S7 for
log KOA values used. Fig. 5 shows that the partitioning between dust/
air and window film/air increases with the log KOA for compounds
with log KOA up to 14, where it reaches a maximum, and it then decreases
for compounds with log KOA higher than 14. Zhang et al.
(2011) reported a similar relationship for PCBs and PBDEs; however,
their “break point” was log KOA of 11. The decreasing trend of the partition
coefficients with log KOA for compounds with log KOA N14 can be
due to several factors: (a) failure to reach equilibrium partitioning between
air, dust, and window film; (b) overestimation of the gas-phase
concentrations due to trapping of particles in the passive samplers; or
(c) higher molecular weight compounds in dust as a result of physical
abrasion of materials containing these chemicals, making them unavailable
to exchange with gas-phase air (Zhang et al., 2011). As pointed out
by Zhang et al. (2011), the latter explanation seems less likely because
one would expect the ratios to be higher than the log KOA rather than
lower, as we have observed.
We hypothesize that in these samples, the equilibrium between air
and dust and window film and dust was not reached at least for some
compounds with log KOA. To test this hypothesis, we used the Weschler
and Nazaroff partitioning model (Weschler and Nazaroff, 2010). We
first estimated air concentrations from dust and from window films,
and then we compared these estimates with the measured ones. For
this exercise, data from the three countries were aggregated to achieve
higher statistical power. Gas-phase concentrations were calculated
from dust using the following relationship:
Cg ¼
Xdρd
KOA f OM
ð1Þ
where Xd is the measured dust concentration in ng/g, ρd is the density of
dust (2.0 × 106
g/m3
as per Weschler and Nazaroff, 2010), and fOM is the
fraction of organic matter in the dust [0.3, as suggested by Bennett and
Furtaw (2004)]. Gas-phase air concentrations were calculated from
window film levels with the following relationship:
Cg ¼
Sf
KOA f OMt
ð2Þ
where Sf is the measured window wipe concentrations (in ng/m2
), and t
is the thickness of the window film (5.0 × 10−8
m) (see the Supporting
Information for more details on the parameters used in calculations).
Fig. 6 shows the relationship between the median measured gasphase
concentrations (pg/m3
) and the median gas-phase concentrations
(pg/m3
) predicted from window film and from dust. For window
film, the predicted values for compounds with log KOA b14 correlated
with those measured (r2
= 0.48) with a slope close to unity (1.60 ±
0.52) and an intercept of 0.73. For dust, the predicted values for compounds
with log KOA b14 correlated with those measured (r2
= 0.44)
Fig. 6. Dependence of the median measured air concentrations versus the predicted
concentrations using window wipes (upper) and dust (lower). Each dot represents
the logarithm of the median concentration. The equation for the linear regressions
are y = 0.73 + 1.60 × (r2
= 0.48, p = 0.0121) for window film and y =
0.094 + 1.13 × (r2
= 0.44, p = 0.0131) for dust.
158 M. Venier et al. / Environment International 94 (2016) 150–160
with a slope close to unity (1.13 ± 0.38) and an intercept of 0.094. Both
slopes were not significantly different from unity (p N 0.05), and the intercepts
indicate that, on average, the predicted mass fractions in window
films and dust were 8.3 and 1.3 times larger (100.73
= 5.4 and
100.094
= 1.2) than the measured ones. The relatively large discrepancy
between predicted and measured air concentrations from window films
could be related to assuming too large a value for the fraction of organic
matter (fOM) or for the thickness of film layer (t). Both for dust and window
films, the measured and predicted values for compounds with log
KOA N14 did not correlate, supporting the hypothesis of Weschler and
Nazaroff (2010) and Zhang et al. (2011), who suggested that the mass
fraction in settled dust and window films may not have sufficient time
to equilibrate with the gas phase.
On the basis of these findings, we conclude that compounds with log
KOA b14 have reached equilibrium and they can be measured in any of
the three matrices. Conversely, compounds with log KOA N14 have not
reached equilibrium and hence they ought to be measured in each matrix
separately because their values can't be predicted from one matrix
alone. Sampling window films captures both gas- and particle-phase
compounds, but lower concentrations lead to greater uncertainty. In
these cases, window films do not provide a comprehensive picture of
compound concentrations in either phase, making this a complementary
technique to air and dust. In addition, all media are relevant to estimating
exposure, either through inhalation (air and fine dust),
ingestion (larger dust particles), and through hand contact followed
by hand-to-mouth behavior (dust and window films).
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.envint.2016.04.029.
Acknowledgements
This work was supported by the Czech-American Scientific Cooperation
Program (AMVIS/KONTAKT II, LH12074), and U.S. Environmental
Protection Agencys Great Lakes National Program Office (Grant GL
00E01422, Todd Nettesheim, project officer). Support in Canada was
provided by Allergy, Genes and Environment Network (AllerGen NCE;
grant no, 12AS13) and the Natural Sciences and Engineering Research
Council of Canada (NSERC; grant nos. RGPAS 429679-12 and RGPIN
124042-12). RECETOX research infrastructure was supported by the
Czech Ministry of Education (LM2011028, LO1214). We thank all
volunteers in Bloomington, IN, USA, Toronto, Canada and Brno, Czech
Republic for their participation. We also thank Ronald A. Hites for helpful
discussion and for his comments on the text.
References
Abbasi, G., Buser, A.M., Soehl, A., Murray, M.W., Diamond, M.L., 2015. Stocks and flows of
PBDEs in products from use to waste in the U.S. And Canada from 1970 to 2020.
Environ. Sci. Technol. 49, 1521–1528.
Abbasi, G., Saini, A., Goosey, E., Diamond, M.L., 2016. Product screening for sources of halogenated
flame retardants in Canadian house and office dust. Sci. Total Environ.
545–546, 299–307.
Abdallah, M.A.-E., Harrad, S., Covaci, A., 2008. Hexabromocyclododecanes and
tetrabromobisphenol-a in indoor air and dust in Birmingham, Uk: implications for
human exposure. Environ. Sci. Technol. 42, 6855–6861.
Ali, N., Harrad, S., Goosey, E., Neels, H., Covaci, A., 2011a. “Novel” brominated flame retardants
in Belgian and Uk indoor dust: implications for human exposure. Chemosphere
83, 1360–1365.
Ali, N., Harrad, S., Muenhor, D., Neels, H., Covaci, A., 2011b. Analytical characteristics and
determination of major novel brominated flame retardants (Nbfrs) in indoor dust.
Anal. Bioanal. Chem. 400, 3073–3083.
Ali, N., Dirtu, A.C., Eede, N.V.d., Goosey, E., Harrad, S., Neels, H., ’t Mannetje, A., Coakley, J.,
Douwes, J., Covaci, A., 2012. Occurrence of alternative flame retardants in indoor dust
from New Zealand: indoor sources and human exposure assessment. Chemosphere
88, 1276–1282.
Andersson, P.L., Öberg, K., Örn, U., 2006. Chemical characterization of brominated flame
retardants and identification of structurally representative compounds. Environ.
Toxicol. Chem. 25, 1275–1282.
Bennett, D.H., Furtaw, E.J., 2004. Fugacity-based indoor residential pesticide fate model.
Environ. Sci. Technol. 38, 2142–2152.
Bennett, D.H., Moran, R.E., Wu, X., Tulve, N.S., Clifton, M.S., Colón, M., Weathers, W.,
Sjödin, A., Jones, R., Hertz-Picciotto, I., 2015. Polybrominated diphenyl ether (PBDE)
concentrations and resulting exposure in homes in California: relationships among
passive air, surface wipe and dust concentrations, and temporal variability. Indoor
Air 25, 220–229.
Björklund, J.A., Sellström, U., de Wit, C.A., Aune, M., Lignell, S., Darnerud, P.O., 2012. Comparisons
of polybrominated diphenyl ether and hexabromocyclododecane concentrations
in dust collected with two sampling methods and matched breast milk
samples. Indoor Air 22, 279–288.
Bohlin, P., Audy, O., Skrdlikova, L., Kukucka, P., Vojta, S., Pribylova, P., Prokes, R., Cupr, P.,
Klanova, J., 2014a. Evaluation and guidelines for using polyurethane foam (PUF)
passive air samplers in double-dome chambers to assess semi-volatile organic
compounds (Svocs) in non-industrial indoor environments. Environ. Sci.: Processes
Impacts 16, 2617–2626.
Bohlin, P., Audy, O., Škrdlίková, L., Kukučka, P., Přibylová, P., Prokeš, R., Vojta, Š., Klánová,
J., 2014b. Outdoor passive air monitoring of semi volatile organic compounds
(Svocs): a critical evalutation of performance and limitations of polyurethane foam
(PUF) disks. Environ. Sci.: Processes Impacts 16, 433–444.
Butt, C.M., Diamond, M.L., Truong, J., Ikonomou, M.G., ter Schure, A.F.H., 2004. Spatial
distribution of polybrominated diphenyl ethers in southern Ontario as measured in
indoor and outdoor window organic films. Environ. Sci. Technol. 38, 724–731.
Cao, Z., Xu, F., Covaci, A., Wu, M., Wang, H., Yu, G., Wang, B., Deng, S., Huang, J., Wang, X.,
2014. Distribution patterns of brominated, chlorinated, and phosphorus flame
retardants with particle size in indoor and outdoor dust and implications for
human exposure. Environ. Sci. Technol. 48, 8839–8846.
Cequier, E., Ionas, A.C., Covaci, A., Marcé, R.M., Becher, G., Thomsen, C., 2014. Occurrence of
a broad range of legacy and emerging flame retardants in indoor environments in
Norway. Environ. Sci. Technol. 48, 6827–6835.
Cetin, B., Odabasi, M., 2011. Polybrominated diphenyl ethers (Pbdes) in indoor and
outdoor window organic films in Izmir, Turkey. J. Hazard. Mater. 185, 784–791.
Chen, S.-J., Ding, N., Zhu, Z.-C., Tian, M., Luo, X.-J., Mai, B.-X., 2014. Sources of halogenated
brominated retardants in house dust in an industrial city in southern China and
associated human exposure. Environ. Res. 135, 190–195.
Coakley, J.D., Harrad, S.J., Goosey, E., Ali, N., Dirtu, A.-C., Van den Eede, N., Covaci, A.,
Douwes, J., Mannetje, A.t., 2013. Concentrations of polybrominated diphenyl ethers
in matched samples of indoor dust and breast milk in New Zealand. Environ. Int.
59, 255–261.
Covaci, A., Harrad, S., Abdallah, M.A.E., Ali, N., Law, R.J., Herzke, D., de Wit, C.A., 2011.
Novel brominated flame retardants: a review of their analysis, environmental fate
and behaviour. Environ. Int. 37, 532–556.
Crimmins, B.S., Pagano, J.J., Xia, X., Hopke, P.K., Milligan, M.S., Holsen, T.M., 2012.
Polybrominated diphenyl ethers (PBDEs): turning the corner in Great Lakes Trout
1980–2009. Environ. Sci. Technol. 46, 9890–9897.
Csiszar, S.A., Diamond, M.L., Thibodeaux, L.J., 2012. Modeling urban films using a dynamic
multimedia fugacity model. Chemosphere 87, 1024–1031.
Davis, E.F., Stapleton, H.M., 2009. Photodegradation pathways of nonabrominated diphenyl
ethers, 2-ethylhexyltetrabromobenzoate and di(2-ethylhexyl)tetrabromophthalate:
identifying potential markers of photodegradation. Environ. Sci. Technol. 43, 5739–5746.
de Wit, C.A., Björklund, J.A., Thuresson, K., 2012. Tri-decabrominated diphenyl ethers and
hexabromocyclododecane in indoor air and dust from Stockholm microenvironments
2: indoor sources and human exposure. Environ. Int. 39, 141–147.
Diamond, M.L., Gingrich, S.E., Fertuck, K., McCarry, B.E., Stern, G.A., Billeck, B., Grift, B.,
Brooker, D., Yager, T.D., 2000. Evidence for organic film on an impervious urban
surface: characterization and potential teratogenic effects. Environ. Sci. Technol. 34,
2900–2908.
Dodson, R.E., Perovich, L.J., Covaci, A., Van den Eede, N., Ionas, A.C., Dirtu, A.C., Brody, J.G.,
Rudel, R.A., 2012. After the PBDE phase-out: a broad suite of flame retardants in repeat
house dust samples from California. Environ. Sci. Technol. 46, 13056–13066.
Dodson, R.E., Camann, D.E., Morello-Frosch, R., Brody, J.G., Rudel, R.A., 2015. Semivolatile organic
compounds in homes: strategies for efficient and systematic exposure measurement
based on empirical and theoretical factors. Environ. Sci. Technol. 49, 113–122.
Dungey, S.M., Akintoye, L., 2007. Environmental risk evaluation report: 1,1′-(ethane-1,2diyl)bis[pentabromobenzene].
Environment Agency for England and Wales.
Earnshaw, M.R., Jones, K.C., Sweetman, A.J., 2015. A First European Scale Multimedia Fate
Modelling of Bde-209 from 1970 to 2020. Environ. Int. 74, 71–81.
Episuite. Us Epa, 2012. [Estimation Programs Interface Suite™ for Microsoft® Windows, V
4.11 or Insert Version Used]. United States Environmental Protection Agency,
Washington, Dc, USA.
Gingrich, S.E., Diamond, M.L., Stern, G.A., McCarry, B.E., 2001. Atmospherically derived organic
surface films along an urban-rural gradient. Environ. Sci. Technol. 35, 4031–4037.
Harner, T., Su, K., Genualdi, S., Karpowicz, J., Ahrens, L., Mihele, C., Schuster, J., Charland, J.-P.,
Narayan, J., 2013. Calibration and application of PUF disk passive air samplers for tracking
polycyclic aromatic compounds (Pacs). Atmos. Environ. 75, 123–128.
Harrad, S., Abdallah, M.A.-E., 2008a. Calibration of two passive air sampler configurations
for monitoring concentrations of hexabromocyclododecanes in indoor air. J. Environ.
Monit. 10, 527–531.
Harrad, S., Ibarra, C., Diamond, M., Melymuk, L., Robson, M., Douwes, J., Roosens, L., Dirtu,
A.C., Covaci, A., 2008b. Polybrominated diphenyl ethers in domestic indoor dust from
Canada, New Zealand, United Kingdom and United States. Environ. Int. 34, 232–238.
Harrad, S., Abdallah, M.A.-E., Covaci, A., 2009. Causes of variability in concentrations and
diastereomer patterns of hexabromocyclododecanes in indoor dust. Environ. Int.
35, 573–579.
Johnson, P.I., Stapleton, H.M., Mukherjee, B., Hauser, R., Meeker, J.D., 2013. Associations
between brominated flame retardants in house dust and hormone levels in men.
Sci. Total Environ. 445–446, 177–184.
Li, W.-L., Qi, H., Ma, W.-L., Liu, L.-Y., Zhang, Z., Zhu, N.-Z., Mohammed, M.O.A., Li, Y.-F.,
2015. Occurrence, behavior and human health risk assessment of dechlorane plus
and related compounds in indoor dust of China. Chemosphere 134, 166–171.
159M. Venier et al. / Environment International 94 (2016) 150–160
Lioy, P.J., Freeman, N.C.G., Millette, J.R., 2002. Dust: a metric for use in residential and
building exposure assessment and source characterization. Environ. Health Perspect.
110, 969–983.
Lohmann, R., Klanova, J., Kukucka, P., Yonis, S., Bollinger, K., 2013. Concentration, fluxes,
and residence time of PBDEs across the tropical Atlantic Ocean. Environ. Sci. Technol.
47, 13967–13975.
Ma, Y., Venier, M., Hites, R.A., 2012. 2-Ethylhexyl tetrabromobenzoate and bis(2ethylhexyl)
tetrabromophthalate flame retardants in the Great Lakes atmosphere.
Environ. Sci. Technol. 46, 204–208.
Ma, Y., Salamova, A., Venier, M., Hites, R.A., 2013. Has the phase-out of PBDEs affected
their atmospheric levels? Trends of PBDEs and their replacements in the Great
Lakes atmosphere. Environ. Sci. Technol. 47, 11457–11464.
Melymuk, L., Goosey, E., Riddell, N., Diamond, M., 2015. Interlaboratory study of novel
halogenated flame retardants: interflab. Anal. Bioanal. Chem. 407, 6759–6769.
Muenhor, D., Harrad, S., 2012. Within-room and within-building temporal and spatial
variations in concentrations of polybrominated diphenyl ethers (PBDEs) in indoor
dust. Environ. Int. 47, 23–27.
Newton, S., Sellström, U., de Wit, C.A., 2015. Emerging flame retardants, PBDEs, and
Hbcdds in indoor and outdoor media in Stockholm, Sweden. Environ. Sci. Technol.
49, 2912–2920.
Peverly, A.A., Ma, Y., Venier, M., Rodenburg, Z., Spak, S.N., Hornbuckle, K.C., Hites, R.A.,
2015. Variations of flame retardant, polycyclic aromatic hydrocarbon, and pesticide
concentrations in Chicago's atmosphere measured using passive sampling. Environ.
Sci. Technol. 49, 5371–5379.
Ricklund, N., Kierkegaard, A., McLachlan, M.S., 2008. An international survey of
decabromodiphenyl ethane (Debdethane) and decabromodiphenyl ether (Decabde)
in sewage sludge samples. Chemosphere 73, 1799–1804.
Saini, A., Okeme, J.O., Goosey, E., Diamond, M.L., 2015. Calibration of two passive air samplers
for monitoring phthalates and brominated flame-retardants in indoor air.
Chemosphere 137, 166–173.
Shoeib, M., Harner, T., 2002. Characterization and comparison of three passive air samplers
for persistent organic pollutants. Environ. Sci. Technol. 36, 4142–4151.
Shoeib, M., Harner, T., Webster, G.M., Sverko, E., Cheng, Y., 2012. Legacy and current-use
flame retardants in house dust from Vancouver, Canada. Environ. Pollut. 169,
175–182.
Shoeib, M., Ahrens, L., Jantunen, L., Harner, T., 2014. Concentrations in air of
organobromine, organochlorine and organophosphatee flame retardants in Toronto,
Canada. Atmos. Environ. 99, 140–147.
Stapleton, H.M., Allen, J.G., Kelly, S.M., Konstantinov, A., Klosterhaus, S., Watkins, D.,
McClean, M.D., Webster, T.F., 2008. Alternate and new brominated flame retardants
detected in U.S. house dust. Environ. Sci. Technol. 42, 6910–6916.
Stapleton, H.M., Klosterhaus, S., Keller, A., Ferguson, P.L., van Bergen, S., Cooper, E.,
Webster, T.F., Blum, A., 2011. Identification of flame retardants in polyurethane
foam collected from baby products. Environ. Sci. Technol. 45, 5323–5331.
Sverko, E., Tomy, G.T., Reiner, E.J., Li, Y.-F., McCarry, B.E., Arnot, J.A., Law, R.J., Hites, R.A.,
2011. Dechlorane plus and related compounds in the environment: a review. Environ.
Sci. Technol. 45, 5088–5098.
Weschler, C.J., Nazaroff, W.W., 2010. Svoc partitioning between the gas phase and settled
dust indoors. Atmos. Environ. 44, 3609–3620.
Wilford, B.H., Harner, T., Zhu, J., Shoeib, M., Jones, K.C., 2004. Passive sampling survey of
polybrominated diphenyl ether flame retardants in indoor and outdoor air in Ottawa,
Canada: implications for sources and exposure. Environ. Sci. Technol. 38, 5312–5318.
Zhang, X., Diamond, M.L., Robson, M., Harrad, S., 2011. Sources, emissions, and fate of
polybrominated diphenyl ethers and polychlorinated biphenyls indoors in Toronto,
Canada. Environ. Sci. Technol. 45, 3268–3274.
Zheng, J., Wang, J., Luo, X.-J., Tian, M., He, L.-Y., Yuan, J.-G., Mai, B.-X., Yang, Z.-Y., 2010.
Dechlorane plus in human hair from an E-waste recycling area in South China: comparison
with dust. Environ. Sci. Technol. 44, 9298–9303.
Zheng, X., Xu, F., Chen, K., Zeng, Y., Luo, X., Chen, S., Mai, B., Covaci, A., 2015. Flame retardants
and organochlorines in indoor dust from several E-waste recycling sites in
South China: composition variations and implications for human exposure. Environ.
Int. 78, 1–7.
160 M. Venier et al. / Environment International 94 (2016) 150–160
van der Schyff, Veronica, Jiři Kalina, Eva Govarts, Liese Gilles, Greet Schoeters, Argelia
Castaño, Marta Esteban-López, Jiři Kohoutek, Petr Kukučka, Adrian Covaci, Gudrun Koppen,
Lenka Andrýsková, Pavel Piler, Jana Klánová, Tina Kold Jensen, Loic Rambaud, Margaux
Riou, Marja Lamoree, Marike Kolossa-Gehring, Nina Vogel, Till Weber, Thomas Göen,
Catherine Gabriel, Dimosthenis A. Sarigiannis, Amrit Kaur Sakhi, Line Småstuen Haug,
Lubica Palkovicova Murinova, Lucia Fabelova, Janja Snoj Tratnik, Darja Mazej, Lisa
Melymuk. 2023. "Exposure to Flame Retardants in European Children - Results from the
HBM4EU Aligned Studies." International Journal of Hygiene and Environmental Health 247:
114070. https://doi.org/10.1016/j.ijheh.2022.114070
International Journal of Hygiene and Environmental Health 247 (2023) 114070
Available online 25 November 2022
1438-4639/© 2022 The Authors. Published by Elsevier GmbH. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Exposure to flame retardants in European children — Results from the
HBM4EU aligned studies
Veronica van der Schyff a
, Jiˇri Kalina a
, Eva Govarts b
, Liese Gilles b
, Greet Schoeters b,c
,
Argelia Casta˜no d
, Marta Esteban-L´opez d
, Jiˇri Kohoutek a
, Petr Kukuˇcka a
, Adrian Covaci e
,
Gudrun Koppen b
, Lenka Andrýskov´a a
, Pavel Piler a
, Jana Kl´anov´a a
, Tina Kold Jensen f
,
Loic Rambaud g
, Margaux Riou g
, Marja Lamoree h
, Marike Kolossa-Gehring i
, Nina Vogel i
,
Till Weber i
, Thomas G¨oen j
, Catherine Gabriel k,l
, Dimosthenis A. Sarigiannis k,l,m
,
Amrit Kaur Sakhi n
, Line Småstuen Haug n
, Lubica Palkovicova Murinova o
, Lucia Fabelova o
,
Janja Snoj Tratnik p
, Darja Mazej p
, Lisa Melymuk a,*
a
RECETOX, Faculty of Science, Masaryk University, Kotlarska 2, Brno, Czech Republic
b
VITO Health, Flemish Institute for Technological Research (VITO), Mol, 2400, Belgium
c
Department of Biomedical Sciences, University of Antwerp, 2020, Antwerp, Belgium
d
National Centre for Environmental Health, Instituto de Salud Carlos III, Majadahonda, Madrid, Spain
e
Toxicological Center, University of Antwerp, 2610 Wilrijk, Belgium
f
Department of Environmental Medicine, Institute of Public Health, University of Southern Denmark, Odense, 5000, Denmark
g
Sant´e Publique France, French Public Health Agency (ANSP), Saint-Maurice, 94415, France
h
Vrije Universiteit, Amsterdam Institute for Life and Environment, Section Chemistry for Environment & Health, De Boelelaan 1108, 1081 HZ, Amsterdam, Netherlands
i
German Environment Agency (UBA), 06844 Dessau-Roßlau, Germany
j
IPASUM - Institute and Outpatient Clinic of Occupational, Social and Environmental Medicine, Henkestrasse 9-11, 91054, Erlangen, Germany
k
Environmental Engineering Laboratory, Department of Chemical Engineering, Aristotle University of Thessaloniki, 54124, Thessaloniki, Greece
l
HERACLES Research Center on the Exposome and Health, Center for Interdisciplinary Research and Innovation, Balkan Center, Bldg. B, 10th km Thessaloniki-Thermi
Road, 57001, Greece
m
Environmental Health Engineering, Institute of Advanced Study, Palazzo del Broletto, Piazza Della Vittoria 15, 27100, Pavia, Italy
n
Department of Environmental Health, Norwegian Institute of Public Health, Oslo, Norway
o
Faculty of Public Health, Slovak Medical University, Bratislava, 833 03, Slovakia
p
Department of Environmental Sciences, Joˇzef Stefan Institute, Ljubljana, 1000, Slovenia
A R T I C L E I N F O
Keywords:
Polybrominated diphenyl ethers
Organophosphate flame retardants
Children
Europe
HBM4EU
Human biomonitoring
A B S T R A C T
Many legacy and emerging flame retardants (FRs) have adverse human and environmental health effects. This
study reports legacy and emerging FRs in children from nine European countries from the HBM4EU aligned
studies. Studies from Belgium, Czech Republic, Germany, Denmark, France, Greece, Slovenia, Slovakia, and
Norway conducted between 2014 and 2021 provided data on FRs in blood and urine from 2136 children. All
samples were collected and analyzed in alignment with the HBM4EU protocols. Ten halogenated FRs were
quantified in blood, and four organophosphate flame retardants (OPFR) metabolites quantified in urine. Hexabromocyclododecane
(HBCDD) and decabromodiphenyl ethane (DBDPE) were infrequently detected (<16% of
samples). BDE-47 was quantified in blood from Greece, France, and Norway, with France (0.36 ng/g lipid)
having the highest concentrations. BDE-153 and -209 were detected in <40% of samples. Dechlorane Plus (DP)
was quantified in blood from four countries, with notably high median concentrations of 16 ng/g lipid in
Slovenian children. OPFR metabolites had a higher detection frequency than other halogenated FRs. Diphenyl
phosphate (DPHP) was quantified in 99% of samples across 8 countries at levels ~5 times higher than other
OPFR metabolites (highest median in Slovenia of 2.43 ng/g lipid). FR concentrations were associated with
lifestyle factors such as cleaning frequency, employment status of the father of the household, and renovation
status of the house, among others. The concentrations of BDE-47 in children from this study were similar to or
* Corresponding author.
E-mail address: lisa.melymuk@recetox.muni.cz (L. Melymuk).
Contents lists available at ScienceDirect
International Journal of Hygiene and Environmental Health
journal homepage: www.elsevier.com/locate/ijheh
https://doi.org/10.1016/j.ijheh.2022.114070
Received 7 July 2022; Received in revised form 29 September 2022; Accepted 3 November 2022
International Journal of Hygiene and Environmental Health 247 (2023) 114070
2
lower than FRs found in adult matrices in previous studies, suggesting lower recent exposure and effectiveness of
PBDE restrictions.
1. Introduction
Human biomonitoring (HBM) is the analysis of substances and/or
their respective metabolites in human matrices. HBM is a crucial tool to
evaluate and monitor internal chemical exposure in specific population
samples, identify chemicals of concern, study determinants of exposure,
or evaluate the efficacy of regulations in place to mitigate chemical
exposure (Angerer et al., 2007, 2011).
Children often have elevated exposure to many chemicals (Gibson
et al., 2019; Koppen et al., 2019), including industrial compounds such
as flame retardants (FRs), due to an increased inhalation rate, different
breathing zone, increased hand-to-mouth activity, and faster metabolism
(van den Eede et al., 2015). Several FRs are known endocrine
disruptors that disrupt thyroid function in children, resulting in adverse
effects on the individual’s long-term mental health, cognitive ability,
metabolism, and reproduction (Preston et al., 2017). Since children are
the most vulnerable age group to chemical exposure, they should be a
priority for human and environmental health policy (Au, 2002).
The European Human Biomonitoring Initiative (HBM4EU) is a largescale
HBM project co-funded under the European Commission’s
Research and Innovation Program Horizon 2020 that includes 30 European
countries and the European Environment Agency. One of its
major aims is to enhance the body of evidence of European citizens’
internal exposure to chemicals (David et al., 2020; Lange et al., 2021).
Flame retardants were identified by European Union (EU) institutions
and HBM4EU partner countries as priority substances to be studied as
knowledge gaps with an impact on regulation still exist (Louro et al.,
2019). Several detrimental ecological and human health issues have
been associated with elevated FR concentrations, including environmental
persistence, long-range transport, bioaccumulation, and endocrine
and neurologically disruptive effects on organisms, including
humans (Aznar-Alemany et al., 2019; Bajard et al., 2019, 2021; Stieger
et al., 2014; Sverko et al., 2011)
Flame retardants have been widely used since the 1970s in textiles,
furnishing, plastic, and electronic equipment (Horrocks, 2011; McGrath
et al., 2018; Morel et al., 2022). Brominated flame retardants (BFRs),
including polybrominated diphenyl ethers (PBDEs) and hexabromocyclododecanes
(HBCDDs), were the primary FRs for more than
thirty years (Pantelaki and Voutsa, 2019). PBDEs and HBCDDs have
been strictly regulated by several international bodies, including the
Registration, Evaluation, Authorization, and Restriction of Chemicals
(REACH), the United States Environmental Protection Agency (USEPA),
and the Stockholm Convention (Guo et al., 2016; Kemmlein et al., 2009;
Sharkey et al., 2020) since 2004. These compounds are often referred to
as legacy FRs given their limited new use. Since the restrictions on the
use and production of legacy FRs came into force, the production of
alternative FRs, such as organophosphate flame retardants (OPFRs) and
novel halogenated flame retardants (NFRs), has increased (Covaci et al.,
2011; He et al., 2021). Some commonly used halogenated NFRs are
decabromodiphenyl ethane (DBDPE), bis(2-ethylhexyl) tetrabromophthalate
(BEH-TEBP), tetrabromobisphenol A (TBBPA) and Dechlorane
Plus (DP). Collectively, NFRs and OPFRs are referred to as emerging FRs.
Because OPFRs are less persistent than legacy FRs and have shorter
elimination half-lives, it was assumed that these compounds are less
harmful to human and environmental health (Blum et al., 2019).
However, this assumption has recently been questioned, as several
OPFRs are associated with endocrine disruptive effects at the same levels
as PBDEs (Behl et al., 2016).
Since the twin projects COPHES (Consortium to Perform Human
Biomonitoring on a European Scale) (Joas et al., 2012) and the feasibility
study DEMOCOPHES (DEMOnstration of a study to COordinate
and Perform Human Biomonitoring on a European Scale) from 2009 to
2012 (Den Hond et al., 2015), no large-scale multi-country HBM project
has been conducted to determine chemical compound concentrations in
the biological matrices of children or quantified emerging FRs in children.
Our study aims to quantify legacy and emerging FRs in children
from nine European countries through information obtained by the
HBM4EU aligned studies. This activity will enable concentrations of
different countries to be comparable to each other. To understand the
potential sources of flame retardant exposure, several lifestyle factors
were investigated based on ancillary data gathered by questionnaires.
2. Materials and methods
2.1. Study alignment and participation
One of the aims of the HBM4EU aligned studies is to harmonize
chemical and data analyses of human biological samples throughout
Europe (Ganzleben et al., 2017; Gilles et al., 2021). The project builds on
existing capacities within the EU by aligning already existing studies
targeting different populations or regions (Gilles et al., 2021). Eight
countries and six HBM4EU qualified laboratories were involved in the
study of FRs in children (Table 1). The CELSPAC cohort from the Czech
Republic was not part of the HBM4EU study. However, the collection,
analyses, and QA/QC protocols were aligned with the HBM4EU study as
other CELSPAC age groups were used for HBM4EU aligned studies, thus
the data collected from the CELSPAC study can be compared with the
cohorts participating in the HBM4EU study.
The HBM4EU defined “current exposure” as samples collected between
2014 and 2020 (Gilles et al., 2021), and as such, urine and blood
samples from 2136 children aged 6–13 between 2014 and 2021 were
used to evaluate current FR exposure in European children (Gilles et al.,
2021). Blood and urine (both spot and morning void) samples collected
before 2017 were acquired from the biobank of the specific study, while
samples from 2017 to 2020 were new collection. Table S1 presents a
complete description of the studies, matrices, and institutes involved.
2.2. Sample analysis
OPFRs have half-lives of hours to days, as opposed to most halogenated
FRs with half-lives of weeks to years (Dvorakova et al., 2021). Due
to the rapid rate of metabolization and elimination of OPFRs, it is most
appropriate to quantify metabolites in urine instead of the parent OPFR
compounds. OPFR metabolites were quantified in children’s urine,
while the persistent brominated and chlorinated flame retardants were
quantified in children’s blood samples (Table 2).
Eighty-six laboratories from 26 countries were invited to participate
in HBM4EU proficiency tests. Seventy-four of these laboratories successfully
quantified at least one biomarker and were selected to participate
in three rounds of interlaboratory comparison investigations
(Esteban L´opez et al., 2021). In this study, all laboratories responsible
for the analysis of FR biomarkers in urine and/or blood successfully
completed the interlaboratory comparison investigations and external
quality assurance schemes, described in detail by Dvorakova et al.
(2021).
Six HBM4EU qualified laboratories were involved in the FR analyses.
The institutes that participated in the analyses are: the University of
Chemistry and Technology, Prague (Czech Republic), Department of
Environment and Health, Vrije Universiteit Amsterdam (Netherlands),
RECETOX, Masaryk University (Czech Republic), Norwegian Institute of
Public Health (Norway), Toxicological Center, University of Antwerp
(Belgium), and Friedrich-Alexander-Universit¨at Erlangen-Nürnberg
V. van der Schyff et al.
International Journal of Hygiene and Environmental Health 247 (2023) 114070
3
(Germany). The analytical procedures used by the laboratories are
presented in the Supplementary Information (Supplementary Table S1
and Text S1). All participating studies in the HBM4EU survey adhered to
national and European ethics regulations (Gilles et al., 2021). The
ethical information of all participating studies is provided in Table S2
(Gilles et al., 2022). Three laboratories analyzed brominated FRs and
DPs using gas chromatography-tandem mass spectrometry
(GC-MS/MS). Four laboratories analyzed OPFR metabolites using liquid
chromatography-tandem mass spectrometry (LC-MS/MS), and one laboratory
used GC-MS. Limits of quantification (LOQs) are given in
Tables S3 and S4. With the stringent interlaboratory comparison
described above, the concentrations of FRs were deemed comparable,
despite differences in analytical methods across laboratories.
2.3. Data analyses
In cases where less than 20% of individuals were represented by leftcensored
data (below limits of detection/quantification), imputation
was performed to replace censored data. Imputation can provide more
realistic distributions of data than those relying on simple substitution of
a fixed value, e.g., 0.5*LOQ (Bernhardt et al., 2015), and was selected
for use in the aligned studies of HBM4EU (Govarts et al., 2021). The
maximum-likelihood estimation (MLE) method was used to find a distribution
best fitting the non-censored data (R Core Team, 2019). From
an infinite set of all theoretical log-normal distributions, the MLE
method identifies the one for which the probability that given set of
values comes from that distribution is maximal. This is done by maximizing
a likelihood function of that distribution by computing first derivative
of that function. Based on that distribution, random values were
generated for the intervals from 0 to LOD (limit of detection) and from
LOD to LOQ (limit of quantification).
After imputation, the urinary data were standardized for creatinine,
by dividing metabolite concentrations by the known creatinine concentration,
while blood serum/plasma data were standardized for lipid
content according to the harmonized level of lipid enzymes calculated as
Total lipids (mg/dL) = 2.27* (Total cholesterol) + triglycerides + 62.3
mg/dL (Bernert et al., 2007)
A set of univariate analyses were conducted to check the distribution
of both the continuous and discrete (categorical) exposure determinants,
typically lifestyle factors. If an exposure determinant was not available
for more than 40% of the individuals with a known concentration of a
compound, the determinant was excluded from further analyses.
Discrete exposure determinants with only one factor level, e.g. when all
respondents answered “no” to a particular question, or exposure determinants
linearly dependent on other determinants were also excluded
from the subsequent steps.
Differences between countries were tested using Kruskal-Wallis oneway
analysis of variance on the log-transformed concentration data. Due
to the sparse exposure determinants matrix, data were only adjusted for
sex and age of the participants using the mean value as the reference,
and the investigation of exposure determinants was reserved for withincountry
analyses only.
Finally, a single effect linear regression was conducted on logtransformed
data taking all non-excluded exposure determinants using
least squares linear regression (Table S7) based on both the variance
inflation factor (VIF) and the determinant p-value (Govarts et al., 2021).
Exposure determinants with VIF ≤10 or p-value ≤ 0.05 were considered
as significantly associated with the flame retardant concentrations.
Continuous exposure determinants were investigated using Spearman
correlation tests and visualized using scatterplots with 95% confidence
level.
2.4. Study limitations
The sampling and analytical procedures were aligned according to
the HBM4EU standards. However, the aligned study designs were not
identical. Each country had different sample sizes (studies ranged from
55 to 413 participants) and prioritized different compounds, as well as
different timing of sample collection. Although the sampling years
slightly differed, most overlapped and all were within a five-year range
(Table 1). Six of the studies provided newly collected samples, while the
remainded relied on biobanked samples that had been stored at stable
conditions (− 80 ◦
C). The difference in storage time is not expected to
impact comparability, as storage time has not been found to impact
levels of the reported OPFRs in urine (Carignan et al., 2017).
Table 1
Descriptive statistics of the participants of the study and the compounds analyzed in each country.
COUNTRY
(COUNTRY
CODE)a
STUDY TOTAL
PARTICIPANTS
FEMALE MALE AGE RANGE
(YEARS)
SAMPLING
YEARS
COMPOUNDS ANALYZED STUDY REFERENCE,
WHERE AVAILABLE
BELGIUM (BE) 3xG 133 67 66 7–8 2019–2020 BDCIPP, DPHP, BCIPP Govarts et al. (2020)
CZECH
REPUBLIC (CZ)
CELSPAC 195 106 89 11–12 2019 BCEP, BCIPP. BDCIPP, DPHP Not available
GERMANY (DE) GerES V 300 150 150 6–12 2015–2016 BCEP, BCIPP, BDCIPP, DPHP European Commission
(2021a)
DENMARK (DK) OCC 291 130 161 7 2018–2019 BDCIPP, DPHP, BCIPP Kyhl et al. (2015)
GREECE (EL) CROME 55 31 24 6–11 2020–2021 PBDE-47, -153, − 209, DP, HBCDD,
DBDPE
(EnvE Lab, 2022; Gilles
et al., 2022)
FRANCE (FR) ESTEBAN 413 191 222 7–13 2014–2016 PBDE-47, -153, − 209, DP, HBCDD,
DBDPE, BEH-TEBP, TBBPA, BCIPP,
BDCIPP, DPHP
European Commission
(2021b)
NORWAY (NO) NEB II 300 140 160 8–12 2016–2017 PBDE-47, -153, DP, BDCIPP, DPHP Caspersen et al. (2019)
SLOVENIA (SI) SLO CRP 149 82 67 7–10 2018 PBDE-47, -153, − 209, DP, HBCDD,
DBDPE, BDCIPP, DPHP, BCIPP, BCEP
Stajnko et al. (2020)
SLOVAKIA (SK) PCB
cohort
300 167 133 10–13 2014–2017 BDCIPP, DPHP, BCEP, BCIPP Hertz-Picciotto et al.
(2003)
a
Not all studies are nationally representative.
Table 2
Flame retardants and metabolites analyzed.
Persistent FRs (in blood) Metabolized FRs (in urine) (Parent compound/
metabolite)
PBDE-47, 153, − 209 synDP,
anti-DP
α-HBCDD, γ-HBCDD
DBDPE
BEH-TEBP
TBBPA
tris(2-chloroethyl) phosphate (TCEP)/Bis(2chloroethyl)
phosphate (BCEP)
tris(1,3-dichloroisopropyl) phosphate (TDCIPP)/Bis
(1,3-dicholoro-2-propyl) phosphate (BDCIPP)
tris(1-chloro-2-propyl) phosphate (TCIPP)/Bis(1chloro-2-propyl)
phosphate (BCIPP)
triphenyl phosphate (TPHP)/Diphenyl phosphate
(DPHP)a
a
DPHP can be a product by itself, a primary metabolite of TPHP, or a secondary
metabolite of several OPFRs (Liu et al., 2021).
V. van der Schyff et al.
International Journal of Hygiene and Environmental Health 247 (2023) 114070
4
All countries except FR and DE had 100% participants with a European
country of birth; DE had 99.3% and FR 95% European birth
country representation. In FR, 18 participants were not born in Europe.
However, the results of the study are still deemed to be indicative of the
European population. The age range of children participating in the
study differed (Table 1), and some studies (e.g., CELSPAC (CZ), PCB
cohort (SK)) had children on the upper end of the age bracket (10–13
years), while others (e.g., OCC (DK), 3xG (BE)) covered only the lower
end of the age bracket (7–8 years). Urine was collected as both morning
void and spot samples. In the NEB II study (NO) analyzed BFRs and DP
were analyzed in blood plasma, while in the other cohorts they were
analyzed blood serum (Table S1). The questionnaire regarding lifestyle
factors also differed between countries. Because most lifestyle factors
were reported by all countries, or were incompletely reported, multivariate
analyses could not be conducted across the full dataset. Although
the laboratories participated in a complete QA/QC programme to ensure
the quality and comparability of the analytical results (Dvorakova et al.,
2021), they had different LODs and LOQs (Tables S3 and S4). However,
all laboratories passed the two-stage evaluation process (Dvorakova
et al., 2021; Esteban L´opez et al., 2021) and were declared proficient by
the HBM4EU project.
3. Results and discussions
3.1. FR concentrations in children’s blood from different European
countries
The concentrations of halogenated FRs in children’s blood are presented
in Table 3. The detection frequencies of all halogenated FR
quantified in children’s blood are presented in Table S5. Only compounds
with detection frequency >40% were considered for statistical
analyses. α-HBCDD, γ-HBCDD, and DBDPE were analyzed in blood
samples from France, Greece, and Slovenia, but were not present at more
than 40% detection frequency (Table S5). BEH-TEBP and TBBPA were
only investigated in blood from France but were not present above
quantifiable concentrations in any sample (Table S5).
PBDE-47 was present at quantifiable concentrations in France,
Greece, Norway, and Slovenia. Slovenia had a 15% detection frequency
of BDE-47 and is not included in the inter-country comparison, although
we note that the 95th percentile is similar to that of Greece, the other
country from the Southern European region (Table 3). When comparing
levels across the three countries, Greece had significantly lower concentrations
of BDE-47 than either France or Norway (Kruskal-Wallis test,
p < 0.001, Fig. 1), in both the unadjusted data as well as in data adjusted
for age and sex. BDE-153 was detected in more than 40% of the blood
samples from Greece and Slovenia (Fig. 2). The compound was analyzed
in blood from France and Norway as well but was detected in only 35%
and 19% of the samples, respectively. Greece had significantly lower
concentrations of BDE-153 than Slovenia (p < 0.001), both in the
adjusted and unadjusted data. Lipophilic compounds such as halogenated
FRs are typically associated with dietary exposure and breastfeeding
(Pan et al., 2020; Poma et al., 2017). Norway is one of the
countries with the highest breastfeeding rate for the first six months of
infancy (Theurich et al., 2019), while Greece is known to have low
exclusive breastfeeding (Pechlivani et al., 2005), which could be reflected
in the concentration differences of PBDEs (Figs. 1 and 2).
BDE-209 was only present above 40% detection frequency in France,
although it was analyzed in blood from Greece and Slovenia as well.
Quantification limits for BDE-209 for Greece and Slovenia were comparable
to or lower than those for France (Table S3), suggesting that the
differences in detection are due to population exposures rather than
analytical differences. BDE-209 was found at the highest concentration
for a single compound in blood compared with all quantified FRs in our
study (8.73 ng/g lipid), as expected based on the timing of restrictions
and relatively higher recent use of BDE-209 (Abbasi et al., 2019).
BDE-47 and BDE-153 were banned from production and use by the
Stockholm Convention in 2004, while BDE-209 was only included in the
Stockholm Convention in 2017 (Sharkey et al., 2020).
BDE-209 is a compound that can be difficult to quantify due to its
high molecular mass (Pietro´n and Małagocki, 2017; Stapleton, 2006)
and often has low detection frequencies in other studies, typically
because of high limits of detections and issues with blank contamination
(Crosse et al., 2013; Darrow et al., 2017; Johnson et al., 2010; Wu et al.,
2007). In our study, three countries analyzed BDE-209, but the compound
was only quantifiable above detection limits in blood from
France. Although Dvorakova et al. (2021) stated that Europe has ample
laboratories with the capacity to quantify BDE-209, the limitation may
be a disconnect between detection and quantification limits and concentrations
in biological matrices. DBDPE, which is often considered a
replacement for BDE-209, suffers from even more serious analytical
challenges due to poorer instrument sensitivity and higher LOQs
Table 3
Median concentrations and 95th percentiles (in parentheses) of halogenated
flame retardants in children’s blood (ng/g lipid).
Country PBDE-47 PBDE-153 PBDE-209 ƩDP
France 0.36 (2.81) 35% DF
(1.50)
8.73 (159) 34% DF
(1.29)
Greece 0.03 (0.43) 0.09 (0.82) 0% DF 2% DF
Slovenia a
15% DF
(0.46)
0.29 (1.02) 8% DF
(10.10)
16.0 (24.8)
Norway 0.34 (0.93) 19% DF (0.1) b
- 3.17 (33.1)
All
countries
c
0.32 (1.62) 40% DF
(0.99)
38% DF
(115)
2.13 (23.6)
a
Where a compound had <40% detection frequency (DF), the DF and 95th
percentiles are reported.
b
A dash (-) indicates that the compound was not analyzed in blood from that
country.
c
Values in bold indicate the median and 95th percentile of all data.
Fig. 1. Box plot of lipid-adjusted concentrations of BDE-47 for France (FR),
Greece (EL) and Norway (NO). The box indicates median, 25% and 75% percentiles.
Whiskers indicate 5% and 95% percentiles.
Fig. 2. Box plot of lipid-adjusted concentrations of BDE-153 in blood of children
from Greece (EL) and Slovenia (SI). The box indicates median, 25% and
75% percentiles. Whiskers indicate 5% and 95% percentiles.
V. van der Schyff et al.
International Journal of Hygiene and Environmental Health 247 (2023) 114070
5
(Melymuk et al., 2015).
Both syn- and anti-DP congeners were analyzed in blood from France,
Slovenia, Greece, and Norway. Blood from France had detection frequencies
below 40%. Only one blood sample from Greece had detectable
concentrations of syn-DP and no anti-DP was detected. Slovenia had a
median ƩDP concentration of 16 ng/g lipid, which was the highest
concentration quantified in our study (Fig. 3). Anti-DP was present in
41% of the Slovenian blood samples, and syn-DP in only 12%. Norway
had 100% and 99% detection frequencies for syn- and anti-DP, respectively.
The median concentration of ƩDP was 3.17 ng/g lipid in blood
from Norway, which was significantly lower than concentrations in
children’s blood from Slovenia (p < 0.001), both for adjusted and unadjusted
data. Other countries may also have elevated DP concentrations,
but it is difficult to determine due to high LODs (Table S3).
The French ESTEBAN study presented a unique profile in BDE-209
and DP compared with other studies. While detection frequencies are
limited for both of these compounds, the comparison of medians, (where
available) and 95th percentiles suggest higher exposure to BDE-209 in
France and lower exposure to DP compared with Slovenia and Norway.
DP has been suggested to be used as a replacement for BDE-209 (Bar´on
et al., 2016) and the dominance of BDE-209 in France compared to DP in
other regions may suggest countries are at different points in the transition
away from BDE-209. We also note that the French data were
among the earliest collected out of all studies, which may be an additional
reason for the dominance of BDE-209.
3.2. OPFR metabolite concentrations in children’s urine from different
European countries
The detection frequencies of OPFR metabolites in children’s urine
are presented in Table S6. The concentrations of OPFR metabolites in
children’s urine are presented in Table 4.
BCEP was analyzed in four countries, but was only present above the
LOQ in >40% of urine samples from Germany and Czech Republic
(Table 4). Czech samples had significantly higher concentrations of
BCEP than German (Kruskal-Wallis; p < 0.001; Fig. 4A), both in unadjusted
data and data adjusted for age and sex. BCIPP was analyzed in
seven countries, but, as with BCEP, only samples from Germany and
Czech Republic had more than 40% >LOQ. In contrast to BCEP, samples
from Germany had higher levels of BCIPP than those from the Czech
Republic (Kruskal-Wallis; p = 0.019, Fig. 4B), both in unadjusted and
adjusted data.
BDCIPP was quantified in Belgium, France, Germany, Slovenia,
Czech Republic, Norway, Slovakia, and Denmark (Table 4), but was
below LOQ in >60% of samples from Slovakia and Norway. A nonparametric
Mann-Whitney pairwise comparison revealed three groupings
within the countries (Fig. 5). Czech samples had lower concentrations
(p < 0.001; Mann-Whitney) than all other countries, while France,
Germany, and Slovenia had significantly higher concentrations than the
other countries (Fig. 5b). The differences between the country groupings
were significant both in the unadjusted data and data adjusted for age
and sex. When grouping countries by geographic region (e.g., north/
south/east/west according to the UN Geoscheme for Europe) we did not
see significant regional differences, suggesting that differences between
countries are not due to broad geographic trends (e.g., east-west differences
within Europe).
BCEP, BCIPP, and BDCIPP are direct metabolites of chlorinated
OPFRs (Liu et al., 2021). The primary compounds (TCEP, TCIPP, and
TDCIPP) are usually present in consumer products such as furniture,
products containing PUF foam, and electronics (Hoffman et al., 2017).
These compounds have been frequently reported in matrices of direct
relevance for human exposure, particularly indoor settled dust (Liu and
Folk, 2021), indoor air (Yadav et al., 2017), and food products (Blum
et al., 2019). While regional differences have been identified in OPFR
levels in other matrices, the differences are typically pronounced between
Europe, Asia, and North America (He et al., 2015). All countries
involved in our study are members of the European Economic Area,
which has a common market system enabling the free movement of
goods and generally harmonized chemical regulations. Thus, only small
differences between countries are to be expected, and these are more
likely to be due to small differences in the country’s study populations
and lifestyle differences between European regions (Section 3.3). The
ubiquity of OPFR metabolite contamination can be ascribed to the broad
use of chlorinated OPFRs, particularly after the ban on legacy FRs (Percy
et al., 2020).
DPHP was the most widely detected FR, present in 99% of urine
samples. Mann-Whitney pairwise comparisons revealed five overlapping
groups of countries (Fig. 6b). The country groups were identical
in both unadjusted data and adjusted data. Denmark had lower DPHP
concentrations than all other countries, while Belgium and Slovenia had
the highest (Fig. 6a and b). DPHP was also the OPFR metabolite found at
the highest concentration in all countries. In some cases, DPHP concentration
in children’s urine was an order of magnitude higher than
other OPFR metabolites (Table 4). DPHP from Slovenian samples had
the highest median concentration of all the OPFR metabolites (2.43 μg/g
creatinine). Other studies have also reported DPHP to be ubiquitous in
human populations (Butt et al., 2014; Dodson et al., 2014; Hoffman
et al., 2017). Unlike the other three quantified OPFR metabolites, DPHP
is a non-specific metabolite. It is a metabolite of TPHP, a compound used
as a flame retardant and plasticizer, and that is present in a large variety
of products including PUF, hydraulic fluid, polyvinyl chloride (PVC),
and cosmetic products such as nail polish (Preston et al., 2017). DPHP is
also a metabolite of, among others, 2-ethylhexyldiphenyl phosphate
Fig. 3. Box plot of lipid adjusted concentrations of ƩDechlorane Plus (DP) in
blood of children from Norway (NO) and Slovenia (SI). The box indicates median,
25% and 75% percentiles. Whiskers indicate 5% and 95% percentiles.
Table 4
Median concentrations and 95th percentiles (in parentheses) of OPFR metabolites
in children’s urine (μg/g creatinine).
Country BCEP BCIPP BDCIPP DPHP
Belgium -a
14% DF
(1.02)b
0.38 (3.11) 2.41 (7.63)
Czech
Republic
0.22 (5.60) 0.10 (0.96) 0.26 (1.95) 2.06 (5.97)
Germany 0.14 (0.86) 0.12 (0.61) 0.64 (2.49) 1.66 (6.21)
Denmark – 7% DF (1.35) 0.49 (3.71) 1.43 (5.16)
France – 31% DF (5.04) 0.58 (4.55) 1.92 (9.70)
Slovenia 20% DF
(1.04)
18% DF (0.25) 0.64 (1.59) 2.43 (5.47)
Slovakia 20% DF
(0.98)
29% DF (0.28) 37% DF
(1.16)
2.21 (6.48)
Norway – – 17% DF
(0.56)
1.78 (7.29)
All countries c
0.23 (1.32) 0.19 (1.59) 0.38 (2.49) 1.91
(6.87)
a
A dash (-) indicates that the compound was not analyzed in urine from that
country.
b
Where a compound had <40% detection frequency (DF), the DF and 95th
percentiles are reported.
c
Values in bold indicate the median and 95th percentile of all data.
V. van der Schyff et al.
International Journal of Hygiene and Environmental Health 247 (2023) 114070
6
(EHDPHP) and resorcinol bis(diphenyl) phosphate (RDP) (Hou et al.,
2016). Beyond being a FR, DPHP itself is also produced as a plasticizer,
and used in certain chemical reagents and intermediate of pesticides,
medicines, and organic material, is a polymerization catalyst and is an
additive in paints and coatings (Liu et al., 2021). Considering the multiple
exposure sources, it is not surprising that DPHP was the compound
found most frequently at the highest detection frequencies in most
countries.
3.3. Comparisons with other studies
Only a limited number of studies have investigated FRs in children
between the ages of 6 and 12 years old. Many studies have quantified
FRs in infants or toddlers (often in conjunction with samples from the
mother) (Caspersen et al., 2016; He et al., 2018a; Morello-Frosch et al.,
2016) or in adults (Brasseur et al., 2014; Lenters et al., 2013; Sochorov´a
et al., 2017; Wang et al., 2019). However, since Moya et al. (2004) noted
significant differences in both behavior and physiology between
different age groups, we do not expect similar concentrations of FRs
across age groups. Typically, when a study includes both adults and
children, children have higher concentrations of BFRs than adults
(Fischer et al., 2006; Toms et al., 2009; van den Eede et al., 2015). One
study that quantified PBDEs in four members of the same household
found that the children had 2- to 5-fold higher concentrations than the
adults despite identical living conditions and diets (Fischer et al., 2006).
BDE-47 and 153 in children’s blood from our study were both
comparable, if slightly lower than studies from China (Table S8). The
HELIX study analyzed chemicals, including BDE-47 and 153, in children
and mothers from various European countries, providing much overlap
with our study (Haug et al., 2018). The concentrations found in the
HELIX study were comparable to our study. Only BDE-47 from France
(0.36 ng/g lipid) was slightly higher in our study than in the HELIX
study (0.27 ng/g lipid); the other concentrations from Greece and
Norway were slightly higher in the HELIX study (Haug et al., 2018).
Studies from the US and Canada reported concentrations of BDE-47 and
-153 that were two orders of magnitude higher than what was found in
our study (Table S8). PBDEs are typically significantly higher in studies
from the US, and Canada to a lesser extent, when compared with other
global regions (Frederiksen et al., 2009; Siddiqi et al., 2003; Sj¨odin et al.,
2008) due to higher past use of PBDEs. Serum from Norway (Thomsen
et al., 2007) had higher concentrations than what was found in Norway
by the NEB II study. However, it should be kept in mind that the samples
were collected in 1998, and it is known that the use of, and consequently
exposure to, PBDEs has decreased since the 1990s. It was unexpected to
Fig. 4. Box plot of creatinine-adjusted concentrations (μg/g creatinine) of (A) BCEP and (B) BCIPP for Czech Republic (CZ) and Germany (DE). The box indicates
median, 25% and 75% percentiles. Whiskers indicate 5% and 95% percentiles.
Fig. 5. a. Box plot of creatinine adjusted concentrations (μg/g creatinine) of BDCIPP for Belgium (BE), Czech Republic (CZ), Germany (DE), Denmark (DK), France
(FR), and Slovenia (SI). The box indicates median, 25% and 75% percentiles. Whiskers indicate 5% and 95% percentiles.
5b. Mann-Whitney pairwise comparisons between countries. A, B, and C indicate groups of countries with different BDCIPP levels.
Fig. 6. a. Box plot of creatinine-adjusted concentrations of DPHP for Belgium (BE), Czech Republic (CZ), Germany (DE), Denmark, France (FR), Norway (NO),
Slovenia (SI), and Slovakia (SK). The box indicates median, 25% and 75% percentiles. Whiskers indicate 5% and 95% percentiles.
6b. Mann-Whitney pairwise comparisons between countries. Letters groupings indicate groups of countries with different DPHP levels. Countries that do not overlap
can be considered to be different from each other.
V. van der Schyff et al.
International Journal of Hygiene and Environmental Health 247 (2023) 114070
7
see that BDE-209 from France in our study (8.73 ng/g lipid) was the
second-highest concentration of this comparison, after one study from
China (95 ng/g lipid (Guo et al., 2016)).
Previous studies have analyzed BFRs in adults from European
countries. In 2015, BFRs were quantified in serum from 300 Czech
adults. PBDEs were only detected in 9% of the samples above detection
limits (Sochorov´a et al., 2017). Other studies quantified PBDEs in serum
from France (Brasseur et al., 2014), Poland, Ukraine (Lenters et al.,
2013), and other European countries (Garí and Grimalt, 2013). Almost
all concentrations from prior studies in adults were higher than the
concentrations found in our study, which we attribute to the more recent
sample collection. All children that were involved in our study were
born between 2001 and 2015. The European Union restricted the use of
Penta- and Octa-BDE commercial products from 2004, and the lower
brominated congeners of PBDEs (tetra-, penta-, hexa-, and heptabromodiphenyl
ethers) were listed for elimination under Annex A of the
Stockholm Convention on Persistent Organic Pollutants in 2009
(Stockholm Convention, 2019). Though these literature-based comparisons
should be interpreted with caution, the fact that PBDE concentrations
in children’s blood from our study are at comparable and lower
concentrations than adults from studies in 1990s and early 2000s can be
an indication that regulation restricting PBDE production and use are
effective at reducing exposure.
While BFRs are reported in lipid-standardized units (e.g., ng/g lipid
mass) enabling comparison across studies, studies that quantified OPFRs
in urine do not have a standardized unit of quantification; data are
presented in unadjusted units (e.g, μg/L), specific gravity (SG) normalized,
or creatinine-standardized (μg/g crt), as is the case for our data.
Given the limited data on OPFR metabolites in children, we have
included studies using unadjusted of SG normalized urine concentrations
in our comparison, and done an average-based recalculation based
on ratios of creatinine to specific gravity for this age group of children
(Table S8). Given the use of this recalculation, the comparisons with
other studies should be interpreted with caution.
All BCEP and BCIPP concentrations from our study and previous
studies were within the same order of magnitude. BDCIPP concentrations
in urine were an order of magnitude higher in US compared with
China; and European values from our study fell between these two levels
(Table S8). DPHP concentrations were similar in Europe and the US, but
China and Japan had lower concentrations of the compound in children’s
urine (Table S8).
3.4. Lifestyle factors associated with flame retardant exposure
Forty lifestyle factors were examined and the responses recorded by
country, presented in Table S9.
A significant inverse correlation between age and FR exposure was
observed for both BDCIPP and DPHP in France (p < 0.001 and p =
0.005) and Germany (p = 0.002 and p = 0.006) (Table S9). This agrees
with the findings from several other studies, where OPFR concentrations
decreased with children’s age (Chen et al., 2018; He et al., 2018a; Sun
et al., 2018; van den Eede et al., 2015; Zhang et al., 2018), and is likely
related to more hand-to-mouth activity in younger children, resulting in
increased exposure to dust-containing FRs (Cequier et al., 2015; Ionas
et al., 2016; Moya et al., 2004).
In addition to age, our study found associations between FR concentrations
and the physical structure of the indoor environment,
cleaning habits, socioeconomic determinants, and diet (Table S9).
Cequier et al. (2015) noted that the residential environment might be a
more important exposure pathway of FRs than food, which agrees with
the findings of our study. Renovation status, PVC floor, and vacuum and
cleaning frequency per week were associated with FR concentrations in
children from several countries (Table S9). In the Czech Republic, BCIPP
(p = 0.012) was higher in houses that have been renovated within two
years before the study, compared with unrenovated homes (Table S9).
BCIPP was also higher in newer homes in France (p = 0.007), where
home age correlated negatively with BCIPP concentrations. BDCIPP was
lower in Czech children from homes with PVC floors (p = 0.013).
Similarly, lower BDE-153 was quantified in children’s blood from
Slovenian households with PVC walls than in the blood from homes
without (p = 0.036). The reason for this is unknown.
BDCIPP and DPHP were both higher in Czech households that
“never” or “rarely” clean (p = 0.003 and p < 0.001). Similarly, BDE-47
was higher in Greek households that “rarely” vacuum (p = 0.009) and
BDE-153 in Slovenian households that “never” or “rarely” clean (p =
0.027) (Table S9). Since FRs are present in indoor house dust (Ali et al.,
2012; Fromme et al., 2014; Jílkov´a et al., 2018), removing dust from
living spaces is a good way to limit inhalation and dermal exposure, and
has been linked to lower levels of FRs in indoor dust (Sugeng et al.,
2018). However, BDE-47 was higher in French households that “often”
vacuum (p = 0.009). The reason for this disparity is unknown.
The only significant association between measured FR levels and
consumption of food in our study was observed for DPHP in seafood
from Belgium (p = 0.021). Dietary exposure to OPFRs has been indicated
in prior studies (He et al., 2018b; Xu et al., 2017), but was not very
prominent in our study Similarly, the effect of passive smoking was
almost negligible in this study; only BDCIPP in Slovenia was significantly
higher in children from smoking households (p = 0.047).
In the French and Belgium studies, BDCIPP concentrations were
positively correlated with time spent in a car (ρ = 0.189, p = 0.001 for
France and Belgium ρ = 0.195, p = 0.024). TDCIPP, the parent compound
of BDCIPP, is known to be present in car upholstery, where
elevated temperatures can lead to an increase in volatilization (Phillips
et al., 2018). There is evidence from the USA that longer commutes lead
to increased TDCIPP exposure (Reddam et al., 2020).
An interesting feature was observed when studying socioeconomic
factors. BCIPP (p = 0.006), and DPHP (p = 0.002) were higher in
German households where the father was employed, versus households
where the father was unemployed. Another socioeconomic association
was found with household education in Germany, where households
with the lowest levels of education had the lowest levels of DPHP in
children’s urine (p = 0.005). Conversely, the legacy FR BDE-47, was
prominent in households where the father was unemployed (p = 0.016),
as seen in blood from Norway (Table S9). This suggests a link between
socioeconomic status and furniture or product replacement rates. We
hypothesize that the higher levels of BDE-47 in houses where the father
is unemployed are linked with lower income and related to the older
furnishing or appliances containing legacy FRs (e.g., purchased before
restrictions on PBDEs). In contrast, employment is linked to higherincome
households that have higher purchasing power and are more
likely to replace products and furnishings, leading to more products
containing OPFRs instead of the now-banned PBDEs. Previous studies
have linked BCIPP exposure with the number of electrical appliances
and electronics in a home (He et al., 2018a; Sun et al., 2018), which can
also be an indication of socioeconomic position.
Gender appeared to impact concentrations in a few instances.
Slovenian boys had significantly higher BDCIPP and BDE-153 than girls
(p = 0.021 and p = 0.01). Similarly, boys had significantly higher
concentrations of BCIPP in Czechia (p = 0.019) and BDE-47 in Greece (p
= 0.024). However, since this trend was not consistent throughout the
study, the effect of gender on FR exposures should be interpreted with
caution.
Most associations between FR concentrations and lifestyle factors
were found for BDCIPP and DPHP. These compounds had the highest
detection frequencies of the studied compounds. The information gained
should be interpreted with caution. A more in-depth study of the lifestyle
factors correlating with FR and other chemical compounds is
recommended.
4. Conclusions
Halogenated FRs were quantified in children’s blood from four
V. van der Schyff et al.
International Journal of Hygiene and Environmental Health 247 (2023) 114070
8
countries and OPFRs in urine from children from eight countries. This
was the largest aligned study across multiple European countries to
quantify FRs in children. This was also the first large-scale comparative
study of OPFRs between different countries, providing valuable data to
both researchers and policymakers. OPFR metabolites, particularly
BDCIPP and DPHP, have ubiquitous distribution in Europe, with limited
differences between countries, perhaps due to the open market conditions.
OPFR concentrations should be critically evaluated by regulatory
institutions due to their high prevalence and indications of endocrinedisrupting
effects. The concentrations of BDE-47 in children’s blood
collected recently were comparable and lower than BDE-47 in adult
samples from several years ago, suggesting that the regulation of PBDEs
does mitigate the exposure of the compounds to humans. This study has
highlighted the need to further build capacity to enable more laboratories
to analyze OPFRs, BDE-209 and other halogenated alternative
FRs, such as DP, DBDPE, TBBPA, and DBDPE. While it was difficult to
ascribe specific lifestyle factors to flame retardant concentrations, factors
concerning cleaning, socioeconomic status, and physical properties
of the residence had the most significant correlations. It is recommended
that future studies further investigate these and other lifestyle factors to
better understand FR exposure to children.
Funding
The HBM4EU project has received funding from the European
Union’s Horizon 2020 research and innovation program under grant
agreement No 733032 and received co-funding from the author’s organizations.
Co-funding by German Ministry for the Environment, Nature
Conservation, Nuclear Safety and Consumer protection is gratefully
acknowledged. The Norwegian Institute of Public Health (NIPH) has
contributed to funding of the Norwegian Environmental Biobank (NEB)
and laboratory measurements have partly been funded by the Research
Council of Norway through research projects (275903 and 268465).
Support was also provided by the Research Infrastructure RECETOX RI
(No. LM2018121) and CETOCOEN EXCELLENCE (CZ.02.1.01/0.0/0.0/
17_043/0009632), the Operational Programme Research, Development
and Innovation – project Cetocoen Plus (CZ.02.1.01/0.0/0.0/15_003/
0000469) and the European Union’s Horizon 2020 research and innovation
programme under grant agreement No. 857560. The Slovenian
SLO-CRP study was co-financed by the Joˇzef Stefan Institute program
P1- 0143, and a national project “Exposure of children and adolescents
to selected chemicals through their habitat environment” (grant agreement
No. C2715-16-634802). The PCB cohort was supported by Ministry
of Health of the Slovak Republic, grant no. 2012/47-SZU-11 and by
the Slovak Research and Development Agency, grant no. APVV-0571-
12. The Cross-Mediterranean Environment and Health Network
(CROME) study has been co-funded by the European Commission
research funds of Horizon 2020.
Declaration of competing interest
The authors declare that they have no conflict of interest.
Acknowledgements
We thank all collaborating field workers, laboratory and administrative
personnel, and especially the cohort participants who invested
their time and provided samples and information for this study. This
study reflects only the authors’ view and the European Commission is
not responsible for any use that may be made of the information it
contains.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.ijheh.2022.114070.
References
Abbasi, G., Li, L., Breivik, K., 2019. Global historical stocks and emissions of PBDEs.
Environ. Sci. Technol. 53, 6330–6340. https://doi.org/10.1021/acs.est.8b07032.
Ali, N., Dirtu, A.C., van den Eede, N., Goosey, E., Harrad, S., Neels, H., ’t Mannetje, A.,
Coakley, J., Douwes, J., Covaci, A., 2012. Occurrence of alternative flame retardants
in indoor dust from New Zealand: indoor sources and human exposure assessment.
Chemosphere 88, 1276–1282. https://doi.org/10.1016/j.
chemosphere.2012.03.100.
Angerer, J., Ewers, U., Wilhelm, M., 2007. Human biomonitoring: state of the art. Int. J.
Hyg Environ. Health. https://doi.org/10.1016/j.ijheh.2007.01.024.
Angerer, J., Aylward, L.L., Hays, S.M., Heinzow, B., Wilhelm, M., 2011. Human
biomonitoring assessment values: approaches and data requirements. Int. J. Hyg
Environ. Health 214, 348–360. https://doi.org/10.1016/j.ijheh.2011.06.002.
Au, W.W., 2002. International journal of hygiene and environmental health commentary
susceptibility of children to environmental toxic substances. Int. J. Hyg Environ.
Health.
Aznar-Alemany, `O., Yang, X., Alonso, M.B., Costa, E.S., Torres, J.P.M., Malm, O.,
Barcel´o, D., Eljarrat, E., 2019. Preliminary study of long-range transport of
halogenated flame retardants using Antarctic marine mammals. Sci. Total Environ.
650, 1889–1897. https://doi.org/10.1016/j.scitotenv.2018.09.287.
Bajard, L., Melymuk, L., Blaha, L., 2019. Prioritization of hazards of novel flame
retardants using the mechanistic toxicology information from ToxCast and Adverse
Outcome Pathways. Environ. Sci. Eur. https://doi.org/10.1186/s12302-019-0195-z.
Bajard, L., Negi, C.K., Mustieles, V., Melymuk, L., Jomini, S., Barthelemy-Berneron, J.,
Fernandez, M.F., Blaha, L., 2021. Endocrine disrupting potential of replacement
flame retardants – review of current knowledge for nuclear receptors associated with
reproductive outcomes. Environ. Int. https://doi.org/10.1016/j.
envint.2021.106550.
Bar´on, E., Dissanayake, A., Vil`a-Cano, J., Crowther, C., Readman, J.W., Jha, A.N.,
Eljarrat, E., Barcel´o, D., 2016. Evaluation of the genotoxic and physiological effects
of decabromodiphenyl ether (BDE-209) and dechlorane plus (DP) flame retardants in
marine mussels (Mytilus galloprovincialis). Environ. Sci. Technol. 50, 2700–2708.
https://doi.org/10.1021/acs.est.5b05814.
Behl, M., Rice, J.R., Smith, M.v., Co, C.A., Bridge, M.F., Hsieh, J.H., Freedman, J.H.,
Boyd, W.A., 2016. Comparative toxicity of organophosphate flame retardants and
polybrominated diphenyl ethers to Caenorhabditis elegans. Toxicol. Sci. 154,
241–252. https://doi.org/10.1093/toxsci/kfw162.
Bernert, J.T., Turner, W.E., Patterson, D.G., Needham, L.L., 2007. Calculation of serum
“total lipid” concentrations for the adjustment of persistent organohalogen toxicant
measurements in human samples. Chemosphere 68, 824–831. https://doi.org/
10.1016/J.CHEMOSPHERE.2007.02.043.
Bernhardt, P.W., Wang, H.J., Zhang, D., 2015. Statistical methods for generalized linear
models with covariates subject to detection limits. Statistics in biosciences 7, 68–89.
https://doi.org/10.1007/s12561-013-9099-4.
Blum, A., Behl, M., Birnbaum, L.S., Diamond, M.L., Phillips, A., Singla, V., Sipes, N.S.,
Stapleton, H.M., Venier, M., 2019. Organophosphate ester flame retardants: are they
a regrettable substitution for polybrominated diphenyl ethers? Environ. Sci. Technol.
Lett. https://doi.org/10.1021/acs.estlett.9b00582.
Brasseur, C., Pirard, C., Scholl, G., de Pauw, E., Viel, J.F., Shen, L., Reiner, E.J., Focant, J.
F., 2014. Levels of dechloranes and polybrominated diphenyl ethers (PBDEs) in
human serum from France. Environ. Int. 65, 33–40. https://doi.org/10.1016/j.
envint.2013.12.014.
Butt, C.M., Congleton, J., Hoffman, K., Fang, M., Stapleton, H.M., 2014. Metabolites of
organophosphate flame retardants and 2-ethylhexyl tetrabromobenzoate in urine
from paired mothers and toddlers. Environ. Sci. Technol. 48, 10432–10438. https://
doi.org/10.1021/es5025299.
Carignan, C.C., Butt, C.M., Stapleton, H.M., Meeker, J.D., Minguez-Alarc´on, L.,
Williams, P.L., Hauser, R., 2017. Influence of storage vial material on measurement
of organophosphate flame retardant metabolites in urine. Chemosphere 181,
440–446. https://doi.org/10.1016/j.chemosphere.2017.04.083.
Caspersen, I.H., Kvalem, H.E., Haugen, M., Brantsæter, A.L., Meltzer, H.M., Alexander, J.,
Thomsen, C., Frøshaug, M., Bremnes, N.M.B., Broadwell, S.L., Granum, B.,
Kogevinas, M., Knutsen, H.K., 2016. Determinants of plasma PCB, brominated flame
retardants, and organochlorine pesticides in pregnant women and 3 year old
children in the Norwegian Mother and Child Cohort Study. Environ. Res. 146,
136–144. https://doi.org/10.1016/j.envres.2015.12.020.
Caspersen, I.H., Thomsen, C., Haug, L.S., Knutsen, H.K., Brantsæter, A.L.,
Papadopoulou, E., Erlund, I., Lundh, T., Alexander, J., Meltzer, H.M., 2019. Patterns
and dietary determinants of essential and toxic elements in blood measured in midpregnancy:
the Norwegian Environmental Biobank. Sci. Total Environ. 671,
299–308. https://doi.org/10.1016/J.SCITOTENV.2019.03.291.
Cequier, E., Sakhi, A.K., Marc´e, R.M., Becher, G., Thomsen, C., 2015. Human exposure
pathways to organophosphate triesters - a biomonitoring study of mother-child pairs.
Environ. Int. 75, 159–165. https://doi.org/10.1016/j.envint.2014.11.009.
Chen, Y., Fang, J., Ren, L., Fan, R., Zhang, J., Liu, G., Zhou, L., Chen, D., Yu, Y., Lu, S.,
2018. Urinary metabolites of organophosphate esters in children in South China:
concentrations, profiles and estimated daily intake. Environ. Pollut. 235, 358–364.
https://doi.org/10.1016/j.envpol.2017.12.092.
Convention, Stockholm, 2019. Listing of POPs in the Stockholm Convention [WWW
Document], 6.8.22. http://www.pops.int/TheConvention/ThePOPs/AllPOPs/tabid/
2509/Default.aspx.
Covaci, A., Harrad, S., Abdallah, M.A.E., Ali, N., Law, R.J., Herzke, D., de Wit, C.A.,
2011. Novel brominated flame retardants: a review of their analysis, environmental
fate and behaviour. Environ. Int. https://doi.org/10.1016/j.envint.2010.11.007.
V. van der Schyff et al.
International Journal of Hygiene and Environmental Health 247 (2023) 114070
9
Crosse, J.D., Shore, R.F., Jones, K.C., Pereira, M.G., 2013. Key factors affecting liver
PBDE concentrations in sparrowhawks (Accipiter nisus). Environ. Pollut. 177,
171–176. https://doi.org/10.1016/j.envpol.2013.02.006.
Darrow, L.A., Jacobson, M.H., Preston, E.v., Lee, G.E., Panuwet, P., Hunter, R.E.,
Marder, M.E., Marcus, M., Barr, D.B., 2017. Predictors of serum polybrominated
diphenyl ether (PBDE) concentrations among children aged 1-5 years. Environ. Sci.
Technol. 51, 645–654. https://doi.org/10.1021/acs.est.6b04696.
David, M., Schwedler, G., Reiber, L., Tolonen, H., Andersson, A.M., Esteban L´opez, M.,
Joas, A., Sch¨opel, M., Polcher, A., Kolossa-Gehring, M., 2020. Learning from
previous work and finding synergies in the domains of public and environmental
health: EU-funded projects BRIDGE Health and HBM4EU. Arch. Publ. Health 78.
https://doi.org/10.1186/s13690-020-00460-9.
Dodson, R.E., van den Eede, N., Covaci, A., Perovich, L.J., Brody, J.G., Rudel, R.A., 2014.
Urinary biomonitoring of phosphate flame retardants: levels in California adults and
recommendations for future studies. Environ. Sci. Technol. 48, 13625–13633.
https://doi.org/10.1021/es503445c.
Dvorakova, D., Pulkrabova, J., Gramblicka, T., Polachova, A., Buresova, M., L´opez, M.E.,
Casta˜no, A., Nübler, S., Haji-Abbas-Zarrabi, K., Klausner, N., G¨oen, T., Mol, H.,
Koch, H.M., Vaccher, V., Antignac, J.P., Haug, L.S., Vorkamp, K., Hajslova, J., 2021.
Interlaboratory comparison investigations (ICIs) and external quality assurance
schemes (EQUASs) for flame retardant analysis in biological matrices: results from
the HBM4EU project. Environ. Res. 202 https://doi.org/10.1016/j.
envres.2021.111705.
Esteban L´opez, M., G¨oen, T., Mol, H., Nübler, S., Haji-Abbas-Zarrabi, K., Koch, H.M.,
Kasper-Sonnenberg, M., Dvorakova, D., Hajslova, J., Antignac, J.P., Vaccher, V.,
Elbers, I., Thomsen, C., Vorkamp, K., Pedraza – Díaz, S., Kolossa-Gehring, M.,
Casta˜no, A., 2021. The European human biomonitoring platform - design and
implementation of a laboratory quality assurance/quality control (QA/QC)
programme for selected priority chemicals. Int. J. Hyg Environ. Health 234. https://
doi.org/10.1016/j.ijheh.2021.113740.
European Commission, 2021a. GerES V - German environmental survey 2014-2017
[WWW document]. https://doi.org/10.1007/S10654-019-00500-X.
European Commission, 2021b. ESTEBAN - environment, health, biomonitoring, physical
activity. Nutrition : Esteban [WWW Document], 6.8.22. https://ipchem.jrc.ec.
europa.eu/#showmetadata/ESTEBAN.
Fischer, D., Hooper, K., Athanasiadou, M., Athanassiadis, I., Bergman, Å., 2006. Children
show highest levels of polybrominated diphenyl ethers in a California family of four:
a case study. Environ. Health Perspect. 114, 1581–1584. https://doi.org/10.1289/
ehp.8554.
Frederiksen, M., Vorkamp, K., Thomsen, M., Knudsen, L.E., 2009. Human internal and
external exposure to PBDEs - a review of levels and sources. Int. J. Hyg Environ.
Health. https://doi.org/10.1016/j.ijheh.2008.04.005.
Fromme, H., Lahrz, T., Kraft, M., Fembacher, L., Mach, C., Dietrich, S., Burkardt, R.,
V¨olkel, W., G¨oen, T., 2014. Organophosphate flame retardants and plasticizers in the
air and dust in German daycare centers and human biomonitoring in visiting
children (LUPE 3). Environ. Int. 71, 158–163. https://doi.org/10.1016/j.
envint.2014.06.016.
Ganzleben, C., Antignac, J.P., Barouki, R., Casta˜no, A., Fiddicke, U., Kl´anov´a, J.,
Lebret, E., Olea, N., Sarigiannis, D., Schoeters, G.R., Sepai, O., Tolonen, H., KolossaGehring,
M., 2017. Human biomonitoring as a tool to support chemicals regulation
in the European Union. Int. J. Hyg Environ. Health. https://doi.org/10.1016/j.
ijheh.2017.01.007.
Garí, M., Grimalt, J.O., 2013. Inverse age-dependent accumulation of
decabromodiphenyl ether and other PBDEs in serum from a general adult
population. Environ. Int. 54, 119–127. https://doi.org/10.1016/j.
envint.2013.01.012.
Gibson, E.A., Stapleton, H.M., Calero, L., Holmes, D., Burke, K., Martinez, R., Cortes, B.,
Nematollahi, A., Evans, D., Anderson, K.A., Herbstman, J.B., 2019. Differential
exposure to organophosphate flame retardants in mother-child pairs. Chemosphere
219, 567–573. https://doi.org/10.1016/j.chemosphere.2018.12.008.
Gilles, L., Govarts, E., Rambaud, L., Vogel, N., Casta˜no, A., Esteban L´opez, M., Rodriguez
Martin, L., Koppen, G., Remy, S., Vrijheid, M., Montazeri, P., Birks, L., Sepai, O.,
Stewart, L., Fiddicke, U., Loots, I., Knudsen, L.E., Kolossa-Gehring, M., Schoeters, G.,
2021. HBM4EU combines and harmonises human biomonitoring data across the EU,
building on existing capacity – the HBM4EU survey. Int. J. Hyg Environ. Health 237.
https://doi.org/10.1016/j.ijheh.2021.113809.
Gilles, L., Govarts, E., Rodriguez Martin, L., Andersson, A.-M., Appenzeller, B.M.R.,
Barbone, F., Casta˜no, A., Coertjens, D., den Hond, E., Dzhedzheia, V., Erˇzen, I.,
Esteban L´opez, M., F´abelov´a, L., Fillol, C., Franken, C., Frederiksen, H., Gabriel, C.,
Huag, L.S., Horvat, M., Halld´orsson, T.I., Janasik, B., Holcer, N.J., Kakucs, R.,
Karakitsios, S., Katsonouri, A., Kl´anov´a, J., Kold-Jensen, T., Kolossa-Gehring, M.,
Konstantinou, C., Koponen, J., Lignell, S., Lindroos, A.K., Makris, K.C., Mazej, D.,
Morrens, B., Murínov´a, L.P., Namorado, S., Pedraza-Diaz, S., Peisker, J., ProbstHensch,
N., Rambaud, L., Rosolen, V., Rucic, E., Rüther, M., Sarigiannis, D., Snoj
Tratnik, J., Standaert, A., Stewart, L., Szigeti, T., Thomsen, C., Tolonen, H.,
Eiríksd´ottir, ´A., van Nieuwenhuyse, A., Verheyen, V.J., Vlaanderen, J., Vogel, N.,
Wasowicz, W., Weber, T., Zock, J.-P., Sepai, O., Schoeters, G., 2022. Harmonization
of human biomonitoring studies in europe:characteristics of the HBM4EU-aligned
studies participants. Int. J. Environ. Res. Publ. Health 19, 6787. https://doi.org/
10.3390/ijerph19116787.
Govarts, E., Portengen, L., Lambrechts, N., Bruckers, L., den Hond, E., Covaci, A.,
Nelen, V., Nawrot, T.S., Loots, I., Sioen, I., Baeyens, W., Morrens, B., Schoeters, G.,
Vermeulen, R., 2020. Early-life exposure to multiple persistent organic pollutants
and metals and birth weight: pooled analysis in four Flemish birth cohorts. Environ.
Int. 145, 106149 https://doi.org/10.1016/J.ENVINT.2020.106149.
Govarts, E., Gilles, L., Rambaud, L., Vogel, N., Montazeri, P., Berglund, M., Santonen, T.,
Katsonouri, A., 2021. Deliverable 10.12 Update statistical analysis plan for the cofunded
studies of WP8 - HBM4EU - science and policy for a healthy future [WWW
Document], 6.8.22. https://www.hbm4eu.eu/work-packages/deliverable-10-12-u
pdate-statistical-analysis-plan-for-the-co-funded-studies-of-wp8/.
Guo, W., Holden, A., Smith, S.C., Gephart, R., Petreas, M., Park, J.S., 2016. PBDE levels
in breast milk are decreasing in California. Chemosphere 150, 505–513. https://doi.
org/10.1016/j.chemosphere.2015.11.032.
Haug, L.S., Sakhi, A.K., Cequier, E., Casas, M., Maitre, L., Basagana, X., Andrusaityte, S.,
Chalkiadaki, G., Chatzi, L., Coen, M., de Bont, J., Dedele, A., Ferrand, J.,
Grazuleviciene, R., Gonzalez, J.R., Gutzkow, K.B., Keun, H., McEachan, R.,
Meltzer, H.M., Petraviciene, I., Robinson, O., Saulnier, P.J., Slama, R., Sunyer, J.,
Urquiza, J., Vafeiadi, M., Wright, J., Vrijheid, M., Thomsen, C., 2018. In-utero and
childhood chemical exposome in six European mother-child cohorts. Environ. Int.
121, 751–763. https://doi.org/10.1016/j.envint.2018.09.056.
He, C.T., Zheng, J., Qiao, L., Chen, S.J., Yang, J.Z., Yuan, J.G., Yang, Z.Y., Mai, B.X.,
2015. Occurrence of organophosphorus flame retardants in indoor dust in multiple
microenvironments of southern China and implications for human exposure.
Chemosphere 133, 47–52. https://doi.org/10.1016/j.chemosphere.2015.03.043.
He, C., English, K., Baduel, C., Thai, P., Jagals, P., Ware, R.S., Li, Y., Wang, X., Sly, P.D.,
Mueller, J.F., 2018a. Concentrations of organophosphate flame retardants and
plasticizers in urine from young children in Queensland, Australia and associations
with environmental and behavioural factors. Environ. Res. 164, 262–270. https://
doi.org/10.1016/j.envres.2018.02.040.
He, C., Wang, X., Tang, S., Thai, P., Li, Z., Baduel, C., Mueller, J.F., 2018b.
Concentrations of organophosphate esters and their specific metabolites in food in
southeast Queensland, Australia: is dietary exposure an important pathway of
organophosphate esters and their metabolites? Environ. Sci. Technol. 52,
12765–12773. https://doi.org/10.1021/acs.est.8b03043.
He, J., Wang, Z., Zhao, L., Ma, H., Huang, J., Li, H., Mao, X., Huang, T., Gao, H., Ma, J.,
2021. Gridded emission inventory of organophosphorus flame retardants in China
and inventory validation. Environ. Pollut. 290 https://doi.org/10.1016/j.
envpol.2021.118071.
Hertz-Picciotto, I., Trnovec, T., Koˇcan, A., Charles, M.J., ˇCiˇznar, P., Langer, P.,
Sovˇcikova, E., James, R., 2003. PCBs and early childhood development in Slovakia:
study design and background. Fresenius Environ. Bull. 12, 208–214.
Hoffman, K., Butt, C.M., Webster, T.F., Preston, E.v., Hammel, S.C., Makey, C.,
Lorenzo, A.M., Cooper, E.M., Carignan, C., Meeker, J.D., Hauser, R., Soubry, A.,
Murphy, S.K., Price, T.M., Hoyo, C., Mendelsohn, E., Congleton, J., Daniels, J.L.,
Stapleton, H.M., 2017. Temporal trends in exposure to organophosphate flame
retardants in the United States. Environ. Sci. Technol. Lett. 4, 112–118. https://doi.
org/10.1021/acs.estlett.6b00475.
Horrocks, A.R., 2011. Flame retardant challenges for textiles and fibres: new chemistry
versus innovatory solutions. Polym. Degrad. Stabil. 96, 377–392. https://doi.org/
10.1016/j.polymdegradstab.2010.03.036.
Hou, R., Xu, Y., Wang, Z., 2016. Review of OPFRs in animals and humans: absorption,
bioaccumulation, metabolism, and internal exposure research. Chemosphere.
https://doi.org/10.1016/j.chemosphere.2016.03.003.
Ionas, A.C., Ulevicus, J., G´omez, A.B., Brandsma, S.H., Leonards, P.E.G., van de Bor, M.,
Covaci, A., 2016. Children’s exposure to polybrominated diphenyl ethers (PBDEs)
through mouthing toys. Environ. Int. 87, 101–107. https://doi.org/10.1016/j.
envint.2015.11.018.
Jílkov´a, S., Melymuk, L., Vojta, ˇS., Vykoukalov´a, M., Bohlin-Nizzetto, P., Kl´anov´a, J.,
2018. Small-scale spatial variability of flame retardants in indoor dust and
implications for dust sampling. Chemosphere 206, 132–141. https://doi.org/
10.1016/j.chemosphere.2018.04.146.
Joas, R., Casteleyn, L., Biot, P., Kolossa-Gehring, M., Castano, A., Angerer, J.,
Schoeters, G., Sepai, O., Knudsen, L.E., Joas, A., Horvat, M., Bloemen, L., 2012.
Harmonised human biomonitoring in Europe: activities towards an EU HBM
framework. Int. J. Hyg Environ. Health 215, 172–175. https://doi.org/10.1016/j.
ijheh.2011.08.010.
Johnson, P.I., Stapleton, H.M., Sjodin, A., Meeker, J.D., 2010. Relationships between
polybrominated diphenyl ether concentrations in house dust and serum. Environ.
Sci. Technol. 44, 5627–5632. https://doi.org/10.1021/es100697q.
Kemmlein, S., Herzke, D., Law, R.J., 2009. Brominated flame retardants in the European
chemicals policy of REACH-Regulation and determination in materials.
J. Chromatogr. A. https://doi.org/10.1016/j.chroma.2008.05.085.
Koppen, G., Govarts, E., Vanermen, G., Voorspoels, S., Govindan, M., Dewolf, M.C., den
Hond, E., Biot, P., Casteleyn, L., Kolossa-Gehring, M., Schwedler, G., Angerer, J.,
Koch, H.M., Schindler, B.K., Casta˜no, A., L´opez, M.E., Sepai, O., Exley, K.,
Bloemen, L., Knudsen, L.E., Joas, R., Joas, A., Schoeters, G., Covaci, A., 2019.
Mothers and children are related, even in exposure to chemicals present in common
consumer products. Environ. Res. 175, 297–307. https://doi.org/10.1016/j.
envres.2019.05.023.
Kyhl, H.B., Jensen, T.K., Barington, T., Buhl, S., Norberg, L.A., Jørgensen, J.S., Jensen, D.
F.G., Christesen, H.T., Lamont, R.F., Husby, S., 2015. The odense child cohort: aims,
design, and cohort profile. Paediatr. Perinat. Epidemiol. 29, 250–258. https://doi.
org/10.1111/PPE.12183.
EnvE Lab, 2022. CROME – life. Cross-mediterranean environment and health Network
[WWW document], 6.20.22. https://www.enve-lab.eu/index.php/work/cr
ome-life/.
Lange, R., Apel, P., Rousselle, C., Charles, S., Sissoko, F., Kolossa-Gehring, M., Ougier, E.,
2021. The European Human Biomonitoring Initiative (HBM4EU): human
biomonitoring guidance values for selected phthalates and a substitute plasticizer.
Int. J. Hyg Environ. Health 234. https://doi.org/10.1016/j.ijheh.2021.113722.
V. van der Schyff et al.
International Journal of Hygiene and Environmental Health 247 (2023) 114070
10
Lenters, V., Thomsen, C., Smit, L.A.M., J¨onsson, B.A.G., Pedersen, H.S., Ludwicki, J.K.,
Zviezdai, V., Piersma, A.H., Toft, G., Bonde, J.P., Becher, G., Vermeulen, R.,
Heederik, D., 2013. Serum concentrations of polybrominated diphenyl ethers
(PBDEs) and a polybrominated biphenyl (PBB) in men from Greenland, Poland and
Ukraine. Environ. Int. 61, 8–16. https://doi.org/10.1016/j.envint.2013.09.001.
Liu, X., Folk, E., 2021. Sorption and migration of organophosphate flame retardants
between sources and settled dust. Chemosphere 278. https://doi.org/10.1016/j.
chemosphere.2021.130415.
Liu, Y., Gong, S., Ye, L., Li, J., Liu, C., Chen, D., Fang, M., Letcher, R.J., Su, G., 2021.
Organophosphate (OP) diesters and a review of sources, chemical properties,
environmental occurrence, adverse effects, and future directions. Environ. Int.
https://doi.org/10.1016/j.envint.2021.106691.
Louro, H., Hein¨al¨a, M., Bessems, J., Buekers, J., Vermeire, T., Woutersen, M., van
Engelen, J., Borges, T., Rousselle, C., Ougier, E., Alvito, P., Martins, C., Assunç˜ao, R.,
Silva, M.J., Pronk, A., Schaddelee-Scholten, B., del Carmen Gonzalez, M., de
Alba, M., Casta˜no, A., Viegas, S., Humar-Juric, T., Kononenko, L., Lampen, A.,
Vinggaard, A.M., Schoeters, G., Kolossa-Gehring, M., Santonen, T., 2019. Human
biomonitoring in health risk assessment in Europe: current practices and
recommendations for the future. Int. J. Hyg Environ. Health. https://doi.org/
10.1016/j.ijheh.2019.05.009.
McGrath, T.J., Morrison, P.D., Ball, A.S., Clarke, B.O., 2018. Spatial distribution of novel
and legacy brominated flame retardants in soils surrounding two Australian
electronic waste recycling facilities. Environ. Sci. Technol. 52, 8194–8204. https://
doi.org/10.1021/acs.est.8b02469.
Melymuk, L., Goosey, E., Riddell, N., Diamond, M.L., 2015. Interlaboratory study of
novel halogenated flame retardants: INTERFLAB. Anal. Bioanal. Chem. 407,
6759–6769. https://doi.org/10.1007/S00216-015-8843-7/FIGURES/7.
Morel, C., Schroeder, H., Emond, C., Turner, J.D., Lichtfouse, E., Grova, N., 2022.
Brominated flame retardants, a cornelian dilemma. Environ. Chem. Lett. https://doi.
org/10.1007/s10311-022-01392-2.
Morello-Frosch, R., Cushing, L.J., Jesdale, B.M., Schwartz, J.M., Guo, W., Guo, T.,
Wang, M., Harwani, S., Petropoulou, S.S.E., Duong, W., Park, J.S., Petreas, M.,
Gajek, R., Alvaran, J., She, J., Dobraca, D., Das, R., Woodruff, T.J., 2016.
Environmental chemicals in an urban population of pregnant women and their
newborns from San Francisco. Environ. Sci. Technol. 50, 12464–12472. https://doi.
org/10.1021/acs.est.6b03492.
Moya, J., Bearer, C.F., Etzel, R.A., 2004. Children’s Behavior and Physiology and How it
Affects Exposure to Environmental Contaminants.
Pan, H.Y., Li, J.F.T., Li, X.H., Yang, Y.L., Qin, Z.F., Li, J.B., Li, Y.Y., 2020. Transfer of
dechlorane plus between human breast milk and adipose tissue and comparison with
legacy lipophilic compounds. Environ. Pollut. 265, 115096 https://doi.org/
10.1016/J.ENVPOL.2020.115096.
Pantelaki, I., Voutsa, D., 2019. Organophosphate flame retardants (OPFRs): a review on
analytical methods and occurrence in wastewater and aquatic environment. Sci.
Total Environ. https://doi.org/10.1016/j.scitotenv.2018.08.286.
Pechlivani, F., Vassilakou, T., Sarafidou, J., Zachou, T., Anastasiou, C.A., Sidossis, L.S.,
2005. Prevalence and determinants of exclusive breastfeeding during hospital stay in
the area of Athens, Greece. Acta Paediatrica, International Journal of Paediatrics 94,
928–934. https://doi.org/10.1080/08035250410023746.
Percy, Z., la Guardia, M.J., Xu, Y., Hale, R.C., Dietrich, K.N., Lanphear, B.P., Yolton, K.,
Vuong, A.M., Cecil, K.M., Braun, J.M., Xie, C., Chen, A., 2020. Concentrations and
loadings of organophosphate and replacement brominated flame retardants in house
dust from the home study during the PBDE phase-out. Chemosphere 239. https://
doi.org/10.1016/j.chemosphere.2019.124701.
Phillips, A.L., Hammel, S.C., Hoffman, K., Lorenzo, A.M., Chen, A., Webster, T.F.,
Stapleton, H.M., 2018. Children’s residential exposure to organophosphate ester
flame retardants and plasticizers: investigating exposure pathways in the TESIE
study. Environ. Int. 116, 176–185. https://doi.org/10.1016/j.envint.2018.04.013.
Pietro´n, W.J., Małagocki, P., 2017. Quantification of polybrominated diphenyl ethers
(PBDEs) in food. A review. Talanta. https://doi.org/10.1016/j.talanta.2017.02.043.
Poma, G., Glynn, A., Malarvannan, G., Covaci, A., Darnerud, P.O., 2017. Dietary intake
of phosphorus flame retardants (PFRs) using Swedish food market basket
estimations. Food Chem. Toxicol. 100, 1–7. https://doi.org/10.1016/J.
FCT.2016.12.011.
Preston, E.v., McClean, M.D., Claus Henn, B., Stapleton, H.M., Braverman, L.E.,
Pearce, E.N., Makey, C.M., Webster, T.F., 2017. Associations between urinary
diphenyl phosphate and thyroid function. Environ. Int. 101, 158–164. https://doi.
org/10.1016/j.envint.2017.01.020.
R Core Team, 2019. R: A Language and Environment for Statistical Computing. R
Foundation for Statistical Computing.
Reddam, A., Tait, G., Herkert, N., Hammel, S.C., Stapleton, H.M., Volz, D.C., 2020.
Longer commutes are associated with increased human exposure to tris(1,3-dichloro-
2-propyl) phosphate. Environ. Int. 136 https://doi.org/10.1016/j.
envint.2020.105499.
Sharkey, M., Harrad, S., Abou-Elwafa Abdallah, M., Drage, D.S., Berresheim, H., 2020.
Phasing-out of legacy brominated flame retardants: the UNEP Stockholm Convention
and other legislative action worldwide. Environ. Int. https://doi.org/10.1016/j.
envint.2020.106041.
Siddiqi, M.A., Laessig, R.H., Reed, K.D., 2003. Polybrominated diphenyl ethers (PBDEs):
new pollutants-old diseases. Clin. Med. Res.
Sj¨odin, A., P¨apke, O., McGahee, E., Focant, J.F., Jones, R.S., Pless-Mulloli, T., Toms, L.M.
L., Herrmann, T., Müller, J., Needham, L.L., Patterson, D.G., 2008. Concentration of
polybrominated diphenyl ethers (PBDEs) in household dust from various countries.
Chemosphere 73. https://doi.org/10.1016/j.chemosphere.2007.08.075.
Sochorov´a, L., Hanzlíkov´a, L., ˇCern´a, M., Drg´aˇcov´a, A., Fialov´a, A., ˇSvarcov´a, A.,
Grambliˇcka, T., Pulkrabov´a, J., 2017. Perfluorinated alkylated substances and
brominated flame retardants in serum of the Czech adult population. Int. J. Hyg
Environ. Health 220, 235–243. https://doi.org/10.1016/j.ijheh.2016.09.003.
Stajnko, A., Snoj Tratnik, J., Kosjek, T., Mazej, D., Jagodic, M., Erˇzen, I., Horvat, M.,
2020. Seasonal glyphosate and AMPA levels in urine of children and adolescents
living in rural regions of Northeastern Slovenia. Environ. Int. 143 https://doi.org/
10.1016/j.envint.2020.105985.
Stapleton, H.M., 2006. Instrumental methods and challenges in quantifying
polybrominated diphenyl ethers in environmental extracts: a review. Anal. Bioanal.
Chem. https://doi.org/10.1007/s00216-006-0400-y.
Stieger, G., Scheringer, M., Ng, C.A., Hungerbühler, K., 2014. Assessing the persistence,
bioaccumulation potential and toxicity of brominated flame retardants: data
availability and quality for 36 alternative brominated flame retardants.
Chemosphere 116, 118–123. https://doi.org/10.1016/j.chemosphere.2014.01.083.
Sugeng, E.J., de Cock, M., Leonards, P.E.G., van de Bor, M., 2018. Electronics, interior
decoration and cleaning patterns affect flame retardant levels in the dust from Dutch
residences. Sci. Total Environ. 645, 1144–1152. https://doi.org/10.1016/j.
scitotenv.2018.07.127.
Sun, Y., Gong, X., Lin, W., Liu, Y., Wang, Y., Wu, M., Kannan, K., Ma, J., 2018.
Metabolites of organophosphate ester flame retardants in urine from Shanghai,
China. Environ. Res. 164, 507–515. https://doi.org/10.1016/j.envres.2018.03.031.
Sverko, E., Tomy, G.T., Reiner, E.J., Li, Y.F., McCarry, B.E., Arnot, J.A., Law, R.J.,
Hites, R.A., 2011. Dechlorane plus and related compounds in the environment: a
review. Environ. Sci. Technol. 45, 5088–5098. https://doi.org/10.1021/es2003028.
Theurich, M.A., Davanzo, R., Busck-Rasmussen, M., Díaz-G´omez, N.M., Brennan, C.,
Kylberg, E., Bærug, A., McHugh, L., Weikert, C., Abraham, K., Koletzko, B., 2019.
Breastfeeding rates and programs in europe: a survey of 11 national breastfeeding
committees and representatives. J. Pediatr. Gastroenterol. Nutr. 68, 400–407.
https://doi.org/10.1097/MPG.0000000000002234.
Thomsen, C., Liane, V.H., Becher, G., 2007. Automated solid-phase extraction for the
determination of polybrominated diphenyl ethers and polychlorinated biphenyls in
serum-application on archived Norwegian samples from 1977 to 2003.
J. Chromatogr., B: Anal. Technol. Biomed. Life Sci. 846, 252–263. https://doi.org/
10.1016/j.jchromb.2006.09.011.
Toms, L.M.L., Sj¨odin, A., Harden, F., Hobson, P., Jones, R., Edenfield, E., Mueller, J.F.,
2009. Serum polybrominated diphenyl ether (PBDE) levels are higher in children
(2–5 Years of age) than in infants and adults. Environ. Health Perspect. 117,
1461–1465. https://doi.org/10.1289/EHP.0900596.
van den Eede, N., Heffernan, A.L., Aylward, L.L., Hobson, P., Neels, H., Mueller, J.F.,
Covaci, A., 2015. Age as a determinant of phosphate flame retardant exposure of the
Australian population and identification of novel urinary PFR metabolites. Environ.
Int. 74, 1–8. https://doi.org/10.1016/j.envint.2014.09.005.
Wang, Y., Li, W., Martínez-Moral, M.P., Sun, H., Kannan, K., 2019. Metabolites of
organophosphate esters in urine from the United States: concentrations, temporal
variability, and exposure assessment. Environ. Int. 122, 213–221. https://doi.org/
10.1016/j.envint.2018.11.007.
Wu, N., Herrmann, T., Paepke, O., Tickner, J., Hale, R., Harvey, E., la Guardia, M.,
McClean, M.D., Webster, T.F., 2007. Human exposure to PBDEs: associations of
PBDE body burdens with food consumption and house dust concentrations. Environ.
Sci. Technol. 41, 1584–1589. https://doi.org/10.1021/es0620282.
Xu, F., Tay, J.H., Covaci, A., Padilla-S´anchez, J.A., Papadopoulou, E., Haug, L.S.,
Neels, H., Sellstr¨om, U., de Wit, C.A., 2017. Assessment of dietary exposure to
organohalogen contaminants, legacy and emerging flame retardants in a Norwegian
cohort. Environ. Int. 102, 236–243. https://doi.org/10.1016/j.envint.2017.03.009.
Yadav, I.C., Devi, N.L., Zhong, G., Li, J., Zhang, G., Covaci, A., 2017. Occurrence and fate
of organophosphate ester flame retardants and plasticizers in indoor air and dust of
Nepal: implication for human exposure. Environ. Pollut. 229, 668–678. https://doi.
org/10.1016/j.envpol.2017.06.089.
Zhang, B., Lu, S., Huang, M., Zhou, M., Zhou, Z., Zheng, H., Jiang, Y., Bai, X., Zhang, T.,
2018. Urinary metabolites of organophosphate flame retardants in 0–5-year-old
children: potential exposure risk for inpatients and home-stay infants. Environ.
Pollut. 243, 318–325. https://doi.org/10.1016/j.envpol.2018.08.051.
Elly Den Hond, Eva Govarts, Hanny Willems, Roel Smolders, Ludwine Casteleyn, Marike
Kolossa-Gehring, Gerda Schwedler, Margarete Seiwert, Ulrike Fiddicke, Argelia
Casta˜no, Marta Esteban, Jürgen Angerer, Holger M. Koch, Birgit K. Schindler, Ovnair
Sepai, Karen Exley, Louis Bloemen, Milena Horvat, Lisbeth E. Knudsen, Anke Joas,
Reinhard Joas, Pierre Biot, Dominique Aerts, Gudrun Koppen, Andromachi
Katsonouri, Adamos Hadjipanayis, Andrea Krskova, Marek Maly, Thit A. Mørck,
Peter Rudnai, Szilvia Kozepesy, Maurice Mulcahy, Rory Mannion, Arno C. Gutleb,
Marc E. Fischer, Danuta Ligocka, Marek Jakubowski, M. F´atima Reis, S´onia
Namorado, Anca Elena Gurzau, Ioana-Rodica Lupsa, Katarina Halzlova, Michal
Jajcaj, Darja Mazej, Janja Snoj Tratnik, Ana L´opez, Estrella Lopez, Marika Berglund,
Kristin Larsson, Andrea Lehmann, Pierre Crettaz, and Greet Schoeters, 2015, First
Steps toward Harmonized Human Biomonitoring in Europe: Demonstration Project
to Perform Human Biomonitoring on a European Scale. Environ. Health Perspect.,
123:3 ,10.1289/ehp.1408616.
V. van der Schyff et al.
Diamond, Miriam L., Joseph O. Okeme, and Lisa Melymuk. 2021. "Hands as Agents of
Chemical Transport in the Indoor Environment." Environmental Science & Technology
Letters 8 (4): 326-32. https://doi.org/10.1021/acs.estlett.0c01006
Hands as Agents of Chemical Transport in the Indoor Environment
Miriam L. Diamond,* Joseph O. Okeme,* and Lisa Melymuk
Cite This: Environ. Sci. Technol. Lett. 2021, 8, 326−332 Read Online
ACCESS Metrics & More Article Recommendations *sı Supporting Information
ABSTRACT: Indoor environments are important sources of exposure to chemicals
intentionally added to consumer products, building materials, etc. Previous work has
shown that semivolatile organic compounds (SVOCs) migrate from product/material
sources to partition to indoor surfaces, including skin and hands, and that SVOCs on
hands reasonably indicate nondietary exposure to indoor SVOCs. We hypothesize that
the hands of indoor occupants, which contact numerous products and surfaces,
transport SVOCs in the indoor environment to an extent comparable to that of
fugacity-driven and advective transport. This process of “hand-based” chemical
transport is analogous to that of fomite transmission of pathogens. We explore this
hypothesis using a data set of halogenated flame retardants, organophosphate esters,
and phthalate esters in indoor air, floor dust, and wipes of hands and surfaces of
electronic devices of 51 participants. Cluster analysis shows the similarity of the SVOC
profiles on all participants’ hands relative to those of all device surfaces, demonstrating
the ubiquity of these SVOCs. Network analysis consistently shows the centrality of
hands, followed by air, dust, and laptops, indicating that hands are most correlated with all sample types. The significance of this
hypothesis lies in the ability of hands to rapidly transfer SVOCs among surfaces indoors, with implications for exposure.
■ INTRODUCTION
It is painfully clear that people are potent agents of
environmental change at planetary to local scales.1,2
People
often retreat to the indoor environment for comfort, respite,
and safety. In developed countries, people typically spend
>90% of their time indoors and nearly 70% in indoor
residential environments.3,4
By virtue of the amount of time
we spend indoors, these environments affect our health and
well-being, both positively and negatively.4−7
Modern indoor environments contain a wide variety of
chemical contaminants in air and dust that provide pathways
for exposure to occupants. Although only a relatively small
fraction of the complex chemical mixture that occurs indoors
has been identified,8,9
some chemicals, such as lead, formaldehyde,
and some phthalate esters, have been associated with
adverse health effects.10−12
Among the chemicals released indoors, semivolatile organic
compounds (SVOCs) undergo partitioning and transport
processes that result in their distribution among indoor
media such as air, dust, thin organic films that accumulate
on surfaces, skin, clothing and other textiles, and polyurethane
foam in upholstered furniture.13−15
The indoor environment,
especially as energy-saving measures tighten building envelopes,
presents fewer opportunities for chemical loss than
outdoor environments,14,16
leading to increased concentrations
per unit emission relative to outdoors and relatively high
chemical persistence.17−19
In addition to indoor products and materials, people
influence indoor chemical and particle concentrations by
virtue of their presence and activities.20−22
For example, the
exhaled breath of indoor occupants increases levels of carbon
dioxide.23
Adults and particularly young children resuspend
dust when walking and crawling, respectively, which increases
particle concentrations in air and hence inhalation exposure to
particles in dust and particle-phase chemicals.24
One more
example here is that each person is surrounded by her or his
personal cloud of increased particle concentrations that
increase as a function of activity.25−27
Hands can be particularly effective in indoor transport.
Hands play an essential role in transferring the personal
microbiome to indoor surfaces.28
Hands are very efficient in
transporting infectious disease agents, notably the norovirus
and rhinovirus, to create fomites (surfaces or objects
contaminated with infectious agents) such as doorknobs,
handrails, keyboards, and telephones.29−32
Transmission of
infectious agents among fomites by hands through an indoor
environment can occur rapidly.33,34
The surfaces of hands are known to reflect our chemical
environment and to be an indicator of internal body burdens of
a wide range of chemicals. Bouslimani et al.35
identified a
plethora of chemicals on hands, including pesticides and
Received: December 23, 2020
Revised: March 7, 2021
Accepted: March 8, 2021
Published: March 23, 2021
Letterpubs.acs.org/journal/estlcu
© 2021 American Chemical Society
326
https://doi.org/10.1021/acs.estlett.0c01006
Environ. Sci. Technol. Lett. 2021, 8, 326−332
DownloadedviaMASARYKUNIVonOctober8,2024at10:03:55(UTC).
Seehttps://pubs.acs.org/sharingguidelinesforoptionsonhowtolegitimatelysharepublishedarticles.
insecticides, plasticizers, chemicals from a person’s diet, and
pharmaceuticals. Significant correlations have been found
between several flame retardant and phthalate levels on hand
wipes of adults and children and their internal body levels.36−40
Hands are presumed to gather these chemicals by direct
contact with products containing these SVOCs, contact with
contaminated dust, and by air−skin partitioning.41−43
While hands clearly transport some infectious disease agents
and a person’s microbiome community, hands have not been
seen as agents of transport of indoor contaminants such as
SVOCs. Rather, SVOC transport is typically considered to
occur by means of diffusion according to fugacity (or chemical
activity) gradients between media, particle transport processes
(i.e., deposition and resuspension), air advection from
ventilation, heating and cooling systems, and cleaning
processes that remove chemicals associated with dust and
surfaces.18,44,45
We hypothesize that humans are agents of SVOC transport
indoors and that hands play a key role in that transport
process. We explored this hypothesis using a data set of
SVOCs in residential air and dust and surface wipes of
electronic devices and hands of 51 participants obtained
through the OEHS-Homes Study (Ontario Environmental
Health-Homes Study). The following SVOCs were measured:
six phthalate esters,13
22 halogenated flame retardants (HFRs),
including 14 congeners of polybrominated diphenyl ethers
(PBDEs),46
and 23 organophosphate esters (OPEs).39
This
paper is the fifth in a series published from the OEHS-Homes
Study; all data along with detailed explorations of associations
and explanations have been previously published as noted
above.
■ MATERIALS AND METHODS
Chemical Concentration Data. Full details of the study
design, sampling, and chemical analysis used in this analysis
have been provided by Okeme and Yang et al.,47
Yang and
Jı́lková et al.,13,39
and Yang et al.46
These publications also
include detailed information about household characteristics
such as the air exchange rate, house volume, activities, and
information provided in questionnaires by participants where
such factors were significantly related to chemical concentrations
and profiles. Briefly, 51 women were recruited from
the Toronto and Ottawa areas in Canada, aged 18−44 and of
mid-socio-economic status. Ethics approvals were obtained
from the Research Ethics Boards of University of Toronto and
Health Canada. The following samples for chemical analysis
were taken from the bedroom (n = 51) and a subset of
participants’ most used room (n = 26) from February to July
2015: (1) vacuumed floor dust (n = 77), (2) air using
polyurethane passive foam samplers (PUF) deployed for 3
weeks (n = 77), (3) surface wipes from hand-held (cell or
mobile phones, home phones, tablets, and laptops) and nonhand-held
[desktop computers, televisions (TVs), and stereos]
electronic devices (n = 199), and (4) wipes of each
participant’s hand palms and backs (right and left were
composited) taken twice, 3 weeks apart (n = 204 total). Note
that we wiped the surface of the hand-held devices that came
into contact with each participant’s hands, which could have
included cases and not the device itself. Field blanks were
taken for each sample type.
The reported analytes are listed in Table S1. Analytical
methods, including QA/QC procedures (e.g., analysis of SRM
2585, methods for blank and recovery correction, and
deployment of duplicate PUF samplers), have been explained
elsewhere.13,39,46,47
Briefly, after extraction, all samples were
analyzed by gas chromatography with a mass selective detector
(Agilent 6890 GC-5975 MS). Data were blank corrected
according to the criteria of Saini et al.48
Surrogate standard
recoveries were between 74% and 131%; no recovery
correction was applied.
Statistical Analysis. The data set analyzed consisted of
concentrations of six HFRs, 11 OPEs, and six phthalates with
>40% detection frequency in dust, air, and surface wipes from
hands and electronic devices. In cases in which more than one
sample was collected from a home (e.g., hand wipes taken 3
weeks apart and air and dust measurements from the bedroom
and the most used room), an arithmetic mean was calculated
to minimize overrepresentation of any sample type in the
analysis. Values of half of the method detection limit (MDL)
were imputed for values < MDL. The final data set included
concentrations in air (n = 51), dust (n = 51), hands (n = 51),
and surface wipes from hand-held [cell phones (n = 31), home
phones (n = 10), tablets (n = 27), and laptops (n = 32)] and
non-hand-held [desktop computers (n = 13), TVs (n = 68),
and stereos (n = 7)] electronic devices.
Cluster Analysis. We conducted cluster analysis to assess
the similarities between wipe samples (hands and electronic
device surfaces). Data are listed in Table S2. We normalized
the wipe data to percentage contributions to reduce betweensample
variations. Data were log-transformed, and the log
concentration of each chemical was centered by its mean and
scaled by its standard deviation to improve the distribution of
the data and to minimize heteroscedasticity. We performed
hierarchical clustering in MetaboAnalyst version 3.0 using
Ward’s method for Euclidean distances between pairs of
samples. Samples were not reorganized so as to maintain
natural contrast among sample types.49
Network Analyses. We conducted network analyses to
assess relationships between sample types. We used the nonnormalized
final data set [air, dust, and wipes of hands and
electronic device surfaces (Table S3)]. First, we calculated
Spearman’s correlation coefficients (ρ) between pairs of
sample types, i.e., samples from the same home. Next, we
used the data set to create igraph objects of arithmeticweighted
mean ρ for statistically significant correlations. We
used a weighted ρ of >0.4 to generate undirected networks
using the igraph package in RStudio version 3.6.1. The edges
(interconnections) of the networks are proportional to the
weighted ρ values, and node sizes (sample types) are
proportional to the number of edges. We used the PageRank
function (global, unsupervised ranking algorithm50
) in Rstudio
to calculate the centrality or importance of each node. The
total PageRank scores from all nodes of a network amount to
1. Several networks were built: all chemicals grouped according
to compound class [HFRs, OPEs, and phthalates (Table S4)],
grouped according to log KOA values of <11 or ≥11 (Table
S5), and all sample types except for hands.
■ RESULTS AND DISCUSSION
Summary of Previous Results. Dust and surface wipes of
most hand-held and non-hand-held electronic devices had high
detection frequencies of most SVOCs analyzed (Figure
S1).13,39,46
This finding was contrary to our expectation that
surface wipes of electronic devices would reveal one or very
few dominant plasticizer(s) and/or flame retardant(s) that had
been intentionally added to that product. The exception was
Environmental Science & Technology Letters pubs.acs.org/journal/estlcu Letter
https://doi.org/10.1021/acs.estlett.0c01006
Environ. Sci. Technol. Lett. 2021, 8, 326−332
327
the surface wipes of TVs that had increased concentrations of
DBDPE and BDE-209 that, we believe, were intentionally
added at high concentrations to those casings.46
We also found
significantly higher concentrations of plasticizers and flame
retardants in hand-held than non-hand-held devices, with the
exceptions of DBDPE and BDE-209. We postulated that the
discovery of multiple plasticizers and flame retardants in the
surface wipes of all electronic devices was due to within-home
partitioning of these SVOCs emitted from primary sources
(e.g., TV casings, wiring, and upholstered furniture) to indoor
surfaces that included other electronic devices. In addition, we
postulated that the significantly higher concentrations of
SVOCs (except DBDPE and BDE-209) in wipes of handheld
than non-hand-held devices were due to hands transferring
these chemicals from sources to hand-held devices that
are infrequently washed relative to hands. This hypothesis was
supported by significant correlations between chemical profiles
on hands and hand-held devices.13,39,46
We also found that
chemical profiles from wipes of hand backs and palms were
significantly correlated, although concentrations were higher
on palms and concentrations were consistent in samples taken
3 weeks apart.13,39,46
Similarity of Chemical Profiles. Although we found
significant correlations between paired chemical profiles on
hands and hand-held devices, the cluster analysis showed that
all hand profiles clustered together, along with some hand-held
devices (e.g., cell phones and tablets) (Figure 1 and Figure
S1). As expected, TVs mostly clustered together because of
their unusual flame retardant profiles.46
Chemical profiles from
device surfaces from individual homes did not cluster together,
which was expected on the basis of within-home partition-
ing.13,14,51
The similarity of chemical profiles on hand wipes of all 51
unrelated women is consistent with the findings of Lax et al.,28
who reported greater similarity of microbial communities on
the hand samples of 18 participants versus greater differentiation
of microbial communities on each participant’s home
surfaces (e.g., doorknobs and kitchen counters) and other area
Figure 1. Hierarchical cluster analysis based on Euclidean distance for all surfaces sampled, namely, hands and hand-held and non-hand-held
electronic devices (top or x-axis clustering and color bar). Samples were not reorganized so as to maintain natural contrast among sample types;
25% features were ranked by t test to show optimal contrasting patterns as described by Xia and Wishart.49
The heat map shows clustering of the
relative concentrations of chemicals (y-axis) in the different samples (x-axis). The red-to-blue color scale represents the relative concentrations of
the individual phthalates, HFRs, OPEs, and phthalates as shown by the scale.
Environmental Science & Technology Letters pubs.acs.org/journal/estlcu Letter
https://doi.org/10.1021/acs.estlett.0c01006
Environ. Sci. Technol. Lett. 2021, 8, 326−332
328
samples such as each participant’s nose and feet. They
concluded that microbial populations on hands reflect the
numerous inputs obtained from touching a variety of surfaces,
with disruption due to washing. We suggest that the same
explanation pertains here: chemical profiles reflect numerous
inputs from home and other environments as hands touch
numerous surfaces, sorb chemicals from air, and gather dust,
with disruption from washing.
Networks of Chemical Profiles. Next, we examined the
relationships among chemical profiles in all media (air, dust,
and hand and device wipes) using network analysis. Here, the
purpose was to understand the correlations among all sample
types when seen as a connected system. We note that these
networks express the central tendency of the data set, within
which we expect variability due to differences in households
such as differences in individual activity patterns, numbers of
residents, the presence of pets, etc.
SVOC indoor distributions typically consider air, dust, and
emission sources, with transport by diffusion according to
fugacity gradients between air and condensed phases, and
particle or dust transport; hands are not typically considered
relevant vectors for transport of indoor contaminants.14,52,53
Figure S2 shows the network of all flame retardants and
plasticizers (with edges shown distinctly for HFRs, OPEs, and
phthalates) based on this paradigm, excluding the hands as a
possible agent of chemical transport. Air, followed by dust, was
most central or important in the network as indicated by
PageRank values of 0.128 and 0.127, respectively, conforming
to this conventional understanding.
Figure 2 illustrates the network for the same set of chemicals
with the addition of hands as a node. Hands were most central
with the highest PageRank score of 0.116: hands had the
greatest number and strength of significant correlations with
other nodes (i.e., correlations of chemical profiles). The
centrality of hands is consistent with chemical partitioning
and/or chemical transfer between hands and all indoor media.
Air was the next most central node (PageRank score of 0.113),
and dust was fourth (PageRank score of 0.103), which
conforms to the conventional understanding of the distribution
of SVOCs discussed above.
The greater centrality of hand-held laptops and tablets than
non-hand-held desktops, TVs, and stereos is consistent with
the transfer of SVOCs from hands to hand-held devices that,
we have hypothesized, augments chemical abundance attributable
to only partitioning from air and accumulation from
dust.13,40
The low centrality of home phones followed by cell
phones in the networks with and without hands was not
expected and is not consistent with the previous explanation,
especially given the frequent handling of cell phones. One
explanation for the low centrality of cell phones is that their
Figure 2. Network analysis based on significant Spearman ρ correlation coefficients (>0.4) between sample types. Edges (lines connecting nodes)
are green for halogenated flame retardants, pink for organophosphate esters, and purple for phthalates. The table shows the degree of centrality,
expressed as PageRank, between nodes in the network analysis, arranged from most to least central.
Figure 3. Network analysis based on significant Spearman ρ correlation coefficients (>0.4) between sample types for (A) more volatile compounds
with log KOA values of <11 and (B) less volatile compounds with KOA values of ≥11. See Table S6 for PageRank values of each node.
Environmental Science & Technology Letters pubs.acs.org/journal/estlcu Letter
https://doi.org/10.1021/acs.estlett.0c01006
Environ. Sci. Technol. Lett. 2021, 8, 326−332
329
chemical profiles reflect the numerous surfaces with which they
come into contact (e.g., purses and pockets) and upon which
they are placed, including different environments outside the
home. In addition, some people wash the exterior of their cell
phones (we did not collect this information).
Finally, we examined networks produced when the flame
retardants and plasticizers were separated into more volatile
(with log KOA < 11) and less volatile (log KOA ≥ 11) chemicals
(Figure 3 and Table S6). We chose the physicochemical
property of vapor pressure, expressed as KOA, because of its
importance in governing indoor chemical partitioning.13,14,45
Again, hands were most central among more volatile chemicals,
followed by air and dust (PageRank values of 0.142, 0.136, and
0.119, respectively). The centrality of air was expected on the
basis of air-condensed phase partitioning. In the network of
less volatile chemicals, hands were the second-most central,
followed by dust (PageRank values of 0.131 and 0.123,
respectively). The centrality of dust was expected as less
volatile chemicals partition to this condensed phase that is
distributed throughout the indoor environment. Laptops had
the greatest centrality (PageRank value of 0.140), which could
be due to their frequent touching as well as presenting a
condensed phase for partitioning. Air had limited centrality
(PageRank value of 0.083), as expected given the tendency of
these lower-volatility chemicals to partition to condensed
phases. Differences in volatility, in part, explain the difference
in networks according to chemical class; i.e., the phthalates
considered here as a group are more volatile than HFRs and
OPEs (Figure S2 and Table S7).
Implications. Overall, hands were either most (all
compound classes and chemicals with log KOA values of
<11) or second most (chemicals with log KOA values of ≥11)
central in the networks constructed, illustrating the high degree
of correlation between hands and other media in the same
residence, even those not frequently touched (e.g., tops of
stereos and TVs). The centrality of hands is consistent with
but not a test of the hypothesis that hands transport chemicals,
where hands can gather chemicals from contacting multiple
surfaces directly, transport adhered dust, and transport
chemicals that have partitioned from air. The process of
hand transport was posited to explain the higher concentrations
of numerous flame retardants and plasticizers on handheld
than non-hand-held devices.13,40,47
The similarity of flame
retardant and plasticizer profiles on the hands of 51 unrelated
women, as shown in the cluster analysis, is also consistent with
hands coming into contact with multiple surfaces both inside
and outside the home. Tangentially, these results demonstrate
the ubiquity of these chemicals in modern environments,
regardless of the participants’ home contents, activities, or
other personal characteristics.
If hands are agents of chemical transport, as hypothesized
here, then we expect that this mechanism is faster than
diffusive or particle-based transport mechanisms. For example,
hands can rapidly transport chemicals within a home and
between environments such as offices, schools, entertainment
venues, etc. This hypothesis of rapid hand-based transport
within and between locations is in line with the current
understanding of community spread of fomites and the
microbiome.29−35
Finally, these results have implications for human exposure.
Numerous studies have shown that chemical profiles on hands
are proportional to internal exposure.37,39,54
This study
advances our understanding of the ubiquity of these largeproduction
volume chemicals with which hands come into
contact. We suggest that hands likely perpetuate exposure by
transporting chemicals to multiple surfaces that, unlike hands,
are less frequently washed and thus can “re-contaminate”
hands, in the same way that microbial and viral transfer occurs.
Our results provide yet another good reason to wash your
hands frequently!
■ ASSOCIATED CONTENT
*sı Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.estlett.0c01006.
Full names and CAS registry numbers of target
compounds, data sets used to generate networks,
hierarchical cluster and heat map of chemical profiles
and concentrations in samples, network diagram
excluding the hands node, and concentrations of
compounds in samples (PDF)
■ AUTHOR INFORMATION
Corresponding Authors
Miriam L. Diamond − Department of Earth Sciences,
University of Toronto, Toronto, Ontario, Canada M5S 3B1;
School of the Environment and Dalla Lana School of Public
Health, University of Toronto, Toronto, Ontario, Canada
M5S 3E8; orcid.org/0000-0001-6296-6431; Phone: 1
(416) 978-1586; Email: miriam.diamond@utoronto.ca
Joseph O. Okeme − Occupational Cancer Research Centre,
Ontario Health, Toronto, Ontario, Canada M5G 1X3;
orcid.org/0000-0001-7604-0736; Phone: 1 (647)-797-
2759; Email: joseph.okeme@ontariohealth.ca
Author
Lisa Melymuk − Research Centre for Toxic Compounds in the
Environment (RECETOX), 625 00 Brno, Czech Republic;
orcid.org/0000-0001-6042-7688
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.estlett.0c01006
Notes
The authors declare no competing financial interest.
■ ACKNOWLEDGMENTS
Funding was provided by Health Canada’s Clean Air
Regulatory Agenda (CARA), Canada’s Chemicals Management
Plan (CMP), the Natural Sciences and Engineering
Research Council of Canada (to M.L.D., NSERC, RGPAS
429679-12 and RGPIN-2017-06654), the European Union’s
Horizon 2020 research and innovation program under Marie
Skłodowska-Curie Grant Agreement 734522 (INTERWASTE),
and the Czech Ministry of Education, Youth and
Sports (RECETOX RI, LM2018121). The authors thank Jiří
Kalina (RECETOX) and Ashton Anderson (University of
Toronto) for consulting on statistical and network analysis
methods and Congqiao Yang and Shelley A. Harris (University
of Toronto), Liisa Jantunen and Bruce Fraser (Environment
and Climate Change Canada), Lidija Latovic and Dina Tsirlin
(formerly of Cancer Care Ontario), and Regina de la Campa,
Melissa St.-Jean, Ryan Kukla, and Hongyu You (Exposure
Assessment Division of Health Canada). J.O.O. thanks Victoria
Arrandale (Cancer Care Ontario) for guidance.
Environmental Science & Technology Letters pubs.acs.org/journal/estlcu Letter
https://doi.org/10.1021/acs.estlett.0c01006
Environ. Sci. Technol. Lett. 2021, 8, 326−332
330
■ REFERENCES
(1) Steffen, W.; Crutzen, P. J.; McNeill, J. R. The Anthropocene: Are
Humans Now Overwhelming the Great Forces of Nature. Ambio
2007, 36, 614−621.
(2) Steffen, W.; Rockström, J.; Richardson, K.; Lenton, T. M.; Folke,
C.; Liverman, D.; Summerhayes, C. P.; Barnosky, A. D.; Cornell, S. E.;
Crucifix, M.; Donges, J. F.; Fetzer, I.; Lade, S. J.; Scheffer, M.;
Winkelmann, R.; Schellnhuber, H. J. Trajectories of the Earth System
in the Anthropocene. Proc. Natl. Acad. Sci. U. S. A. 2018, 115, 8252−
8259.
(3) Matz, C.; Stieb, D.; Davis, K.; Egyed, M.; Rose, A.; Chou, B.;
Brion, O. Effects of Age, Season, Gender and Urban-Rural Status on
Time-Activity: Canadian Human Activity Pattern Survey 2 (CHAPS
2). Int. J. Environ. Res. Public Health 2014, 11, 2108−2124.
(4) Allen, J. G.; MacNaughton, P.; Satish, U.; Santanam, S.;
Vallarino, J.; Spengler, J. D. Associations of Cognitive Function Scores
with Carbon Dioxide, Ventilation, and Volatile Organic Compound
Exposures in Office Workers: A Controlled Exposure Study of Green
and Conventional Office Environments. Environ. Health Perspect.
2016, 124, 805−812.
(5) Schmeer, K. K.; Yoon, A. J. Home Sweet Home? Home Physical
Environment and Inflammation in Children. Soc. Sci. Res. 2016, 60,
236−248.
(6) Cedeño-Laurent, J. G.; Williams, A.; MacNaughton, P.; Cao, X.;
Eitland, E.; Spengler, J.; Allen, J. G. Building Evidence for Health:
Green Buildings, Current Science, and Future Challenges. Annu. Rev.
Public Health 2018, 39, 291−308.
(7) Logue, J. M.; Price, P. N.; Sherman, M. H.; Singer, B. C. A
Method to Estimate the Chronic Health Impact of Air Pollutants in
U.S. Residences. Environ. Health Perspect. 2012, 120, 216−222.
(8) Peng, H.; Saunders, D. M. V.; Jones, P. D.; Giesy, J. P.
Mutagenic Azo Dyes, Rather than Flame Retardants, Are the
Predominant Brominated Compounds in House Dust Response.
Environ. Sci. Technol. 2017, 51, 3591−3592.
(9) Rostkowski, P.; Haglund, P.; Aalizadeh, R.; Alygizakis, N.;
Thomaidis, N.; Beltran Arandes, J.; Nizzetto, P. B.; Booij, P.;
Budzinski, H.; Brunswick, P.; Covaci, A.; Gallampois, C.; Grosse, S.;
Hindle, R.; Ipolyi, I.; Jobst, K.; Kaserzon, S. L.; Leonards, P.;
Lestremau, F.; Letzel, T.; Magner, J.; Matsukami, H.; Moschet, C.;
Oswald, P.; Plassmann, M.; Slobodnik, J.; Yang, C. The Strength in
Numbers: Comprehensive Characterization of House Dust Using
Complementary Mass Spectrometric Techniques. Anal. Bioanal.
Chem. 2019, 411, 1957−1977.
(10) Logue, J. M.; Mckone, T. E.; Sherman, M. H.; Singer, B. C.
Hazard Assessment of Chemical Air Contaminants Measured in
Residences. Indoor Air 2011, 21, 92−109.
(11) Dixon, S. L.; Gaitens, J. M.; Jacobs, D. E.; Strauss, W.; Nagaraja,
J.; Pivetz, T.; Wilson, J. W.; Ashley, P. J. Exposure of U.S. Children to
Residential Dust Lead, 1999−2004: II. The Contribution of LeadContaminated
Dust to Children’s Blood Lead Levels. Environ. Health
Perspect. 2009, 117, 468−474.
(12) Radke, E. G.; Braun, J. M.; Meeker, J. D.; Cooper, G. S.
Phthalate Exposure and Male Reproductive Outcomes: A Systematic
Review of the Human Epidemiological Evidence. Environ. Int. 2018,
121, 764−793.
(13) Yang, C.; Harris, S. A.; Jantunen, L. M.; Kvasnicka, J.; Nguyen,
L. V.; Diamond, M. L. Phthalates: Relationships between Air, Dust,
Electronic Devices, and Hands with Implications for Exposure.
Environ. Sci. Technol. 2020, 54, 8186−8197.
(14) Weschler, C. J.; Nazaroff, W. W. Semivolatile Organic
Compounds in Indoor Environments. Atmos. Environ. 2008, 42,
9018−9040.
(15) Salthammer, T.; Zhang, Y.; Mo, J.; Koch, H. M.; Weschler, C. J.
Assessing Human Exposure to Organic Pollutants in the Indoor
Environment. Angew. Chem., Int. Ed. 2018, 57, 12228−12263.
(16) Zhang, X.; Diamond, M. L.; Robson, M.; Harrad, S. J. Sources,
Emissions, and Fate of Polybrominated Diphenyl Ethers and
Polychlorinated Biphenyls Indoors in Toronto, Canada. Environ. Sci.
Technol. 2011, 45, 3268−3274.
(17) Rudel, R. A.; Perovich, L. J. Endocrine Disrupting Chemicals in
Indoor and Outdoor Air. Atmos. Environ. 2009, 43, 170−181.
(18) Shin, H.-M.; McKone, T. E.; Tulve, N. S.; Clifton, M. S.;
Bennett, D. H. Indoor Residence Times of Semivolatile Organic
Compounds: Model Estimation and Field Evaluation. Environ. Sci.
Technol. 2013, 47, 859−867.
(19) Audy, O.; Melymuk, L.; Venier, M.; Vojta, Š.; Bečanová, J.;
Romanak, K.; Vykoukalová, M.; Prokeš, R.; Kukucka, P.; Diamond,
M. L.; Klánová, J. PCBs and Organochlorine Pesticides in Indoor
Environments - A Comparison of Indoor Contamination in Canada
and Czech Republic. Chemosphere 2018, 206, 622−631.
(20) Weschler, C. J. Roles of the Human Occupant in Indoor
Chemistry. Indoor Air 2016, 26, 6−24.
(21) Lunderberg, D. M.; Kristensen, K.; Liu, Y.; Misztal, P. K.; Tian,
Y.; Arata, C.; Wernis, R.; Kreisberg, N.; Nazaroff, W. W.; Goldstein,
A. H. Characterizing Airborne Phthalate Concentrations and
Dynamics in a Normally Occupied Residence. Environ. Sci. Technol.
2019, 53, 7337−7346.
(22) Kvasnicka, J.; Cohen Hubal, E.; Ladan, J.; Zhang, X.; Diamond,
M. L. Transient Multimedia Model for Investigating the Influence of
Indoor Human Activities on Exposure to SVOCs. Environ. Sci.
Technol. 2020, 54, 10772−10782.
(23) Stönner, C.; Edtbauer, A.; Williams, J. Real-World Volatile
Organic Compound Emission Rates from Seated Adults and Children
for Use in Indoor Air Studies. Indoor Air 2018, 28, 164−172.
(24) Wu, T.; Täubel, M.; Holopainen, R.; Viitanen, A.-K.;
Vainiotalo, S.; Tuomi, T.; Keskinen, J.; Hyvärinen, A.; Hämeri, K.;
Saari, S. E.; Boor, B. E. Infant and Adult Inhalation Exposure to
Resuspended Biological Particulate Matter. Environ. Sci. Technol.
2018, 52, 237−247.
(25) Allen, J. G.; McClean, M. D.; Stapleton, H. M.; Nelson, J. W.;
Webster, T. F. Personal Exposure to Polybrominated Diphenyl Ethers
(PBDEs) in Residential Indoor Air. Environ. Sci. Technol. 2007, 41,
4574−4579.
(26) Rodes, C. E.; Kamens, R. M.; Wiener, R. W. The Significance
and Characteristics of the Personal Activity Cloud on Exposure
Assessment Measurements for Indoor Contaminants. Indoor Air
1991, 1, 123−145.
(27) You, R.; Cui, W.; Chen, C.; Zhao, B. Measuring the ShortTerm
Emission Rates of Particles in the “Personal Cloud” with
Different Clothes and Activity Intensities in a Sealed Chamber.
Aerosol Air Qual. Res. 2013, 13, 911−921.
(28) Lax, S.; Smith, D. P.; Hampton-Marcell, J.; Owens, S. M.;
Handley, K. M.; Scott, N. M.; Gibbons, S. M.; Larsen, P.; Shogan, B.
D.; Weiss, S.; Metcalf, J. L.; Ursell, L. K.; Vazquez-Baeza, Y.; Van
Treuren, W.; Hasan, N. A.; Gibson, M. K.; Colwell, R.; Dantas, G.;
Knight, R.; Gilbert, J. A. Longitudinal Analysis of Microbial
Interaction between Humans and the Indoor Environment. Science
2014, 345, 1048−1052.
(29) Bloomfield, S. F.; Aiello, A. E.; Cookson, B.; O’Boyle, C.;
Larson, E. L. The Effectiveness of Hand Hygiene Procedures in
Reducing the Risks of Infections in Home and Community Settings
Including Handwashing and Alcohol-Based Hand Sanitizers. Am. J.
Infect. Control 2007, 35, S27−S64.
(30) Boone, S. A.; Gerba, C. P. Significance of Fomites in the Spread
of Respiratory and Enteric Viral Disease. Appl. Environ. Microbiol.
2007, 73, 1687−1696.
(31) Zhang, N.; Miao, R.; Huang, H.; Chan, E. Y. Y. Contact
Infection of Infectious Disease Onboard a Cruise Ship. Sci. Rep. 2016,
6, 38790.
(32) Kraay, A. N. M.; Hayashi, M. A. L.; Hernandez-Ceron, N.;
Spicknall, I. H.; Eisenberg, M. C.; Meza, R.; Eisenberg, J. N. S.
Fomite-Mediated Transmission as a Sufficient Pathway: A Comparative
Analysis across Three Viral Pathogens 11 Medical and Health
Sciences 1117 Public Health and Health Services. BMC Infect. Dis.
2018, 18, 540.
(33) Zhang, N.; Li, Y.; Huang, H. Surface Touch and Its Network
Growth in a Graduate Student Office. Indoor Air 2018, 28, 963−972.
Environmental Science & Technology Letters pubs.acs.org/journal/estlcu Letter
https://doi.org/10.1021/acs.estlett.0c01006
Environ. Sci. Technol. Lett. 2021, 8, 326−332
331
(34) Lei, H.; Li, Y.; Xiao, S.; Yang, X.; Lin, C.; Norris, S. L.; Wei, D.;
Hu, Z.; Ji, S. Logistic Growth of a Surface Contamination Network
and Its Role in Disease Spread. Sci. Rep. 2017, 7, 14826.
(35) Bouslimani, A.; Melnik, A. V.; Xu, Z.; Amir, A.; da Silva, R. R.;
Wang, M.; Bandeira, N.; Alexandrov, T.; Knight, R.; Dorrestein, P. C.
Lifestyle Chemistries from Phones for Individual Profiling. Proc. Natl.
Acad. Sci. U. S. A. 2016, 113, E7645−E7654.
(36) Stapleton, H. M.; Kelly, S. M.; Allen, J. G.; Mcclean, M. D.;
Webster, T. F. Measurement of Polybrominated Diphenyl Ethers on
Hand Wipes: Estimating Exposure from Hand-to-Mouth Contact.
Environ. Sci. Technol. 2008, 42, 3329−3334.
(37) Watkins, D. J.; McClean, M. D.; Fraser, A. J.; Weinberg, J.;
Stapleton, H. M.; Sjödin, A.; Webster, T. F. Exposure to PBDEs in the
Office Environment: Evaluating the Relationships Between Dust,
Handwipes, and Serum. Environ. Health Perspect. 2011, 119, 1247−
1252.
(38) Hoffman, K.; Garantziotis, S.; Birnbaum, L. S.; Stapleton, H. M.
Monitoring Indoor Exposure to Organophosphate Flame Retardants:
Hand Wipes and House Dust. Environ. Health Perspect. 2015, 123,
160−165.
(39) Yang, C.; Harris, S. A.; Jantunen, L. M.; Siddique, S.; Kubwabo,
C.; Tsirlin, D.; Latifovic, L.; Fraser, B.; St-Jean, M.; De La Campa, R.;
You, H.; Kulka, R.; Diamond, M. L. Are Cell Phones an Indicator of
Personal Exposure to Organophosphate Flame Retardants and
Plasticizers? Environ. Int. 2019, 122, 104−116.
(40) Gong, M.; Weschler, C. J.; Liu, L.; Shen, H.; Huang, L.;
Sundell, J.; Zhang, Y. Phthalate Metabolites in Urine Samples from
Beijing Children and Correlations with Phthalate Levels in Their
Handwipes. Indoor Air 2015, 25, 572−581.
(41) Liu, X.; Cao, Z.; Yu, G.; Wu, M.; Li, X.; Zhang, Y.; Wang, B.;
Huang, J. Estimation of Exposure to Organic Flame Retardants via
Hand Wipe, Surface Wipe, and Dust: Comparability of Different
Assessment Strategies. Environ. Sci. Technol. 2018, 52, 9946−9953.
(42) Wu, C.-C.; Bao, L.-J.; Tao, S.; Zeng, E. Y. Dermal Uptake from
Airborne Organics as an Important Route of Human Exposure to EWaste
Combustion Fumes. Environ. Sci. Technol. 2016, 50, 6599−
6605.
(43) Shi, W.; Guo, J.; Zhou, Y.; Deng, D.; Han, Z.; Zhang, X.; Yu,
H.; Giesy, J. P. Phthalate Esters on Hands of Office Workers:
Estimating the Influence of Touching Surfaces. Environ. Sci. Technol.
Lett. 2017, 4, 1−5.
(44) Bennett, D. H.; Furtaw, E. J. Fugacity-Based Indoor Residential
Pesticide Fate Model. Environ. Sci. Technol. 2004, 38, 2142−2152.
(45) Zhang, X.; Diamond, M. L.; Ibarra, C.; Harrad, S. J. Multimedia
Modeling of Polybrominated Diphenyl Ether Emissions and Fate
Indoors. Environ. Sci. Technol. 2009, 43, 2845−2850.
(46) Yang, C.; Rozárka Jílková, S.; Melymuk, L.; Harris, S. A.;
Jantunen, L. M.; Pertili, J.; Winn, L.; Diamond, M. L. Are We Exposed
to Halogenated Flame Retardants from Both Primary and Secondary
Sources? Environ. Sci. Technol. Lett. 2020, 7, 585−593.
(47) Okeme, J. O.; Yang, C.; Abdollahi, A.; Dhal, S.; Harris, S. A.;
Jantunen, L. M.; Tsirlin, D.; Diamond, M. L. Passive Air Sampling of
Flame Retardants and Plasticizers in Canadian Homes Using PDMS,
XAD-Coated PDMS and PUF Samplers. Environ. Pollut. 2018, 239,
109−117.
(48) Saini, A.; Okeme, J. O.; Goosey, E. R.; Diamond, M. L.
Calibration of Two Passive Air Samplers for Monitoring Phthalates
and Brominated Flame-Retardants in Indoor Air. Chemosphere 2015,
137, 166−173.
(49) Xia, J.; Wishart, D. S. Using MetaboAnalyst 3.0 for
Comprehensive Metabolomics Data Analysis. Curr. Protoc. Bioinf.
2016, 55, 14.10.1−14.10.91.
(50) Sun, Y.; Han, J. Ranking Methods for Networks. In Encyclopedia
of Social Network Analysis and Mining; Alhajj, R., Rokne, J., Eds.;
Springer: New York, 2014; pp 1488−1497.
(51) Xu, Y.; Cohen Hubal, E. A.; Clausen, P. A.; Little, J. C.
Predicting Residential Exposure to Phthalate Plasticizer Emitted from
Vinyl Flooring: A Mechanistic Analysis. Environ. Sci. Technol. 2009,
43, 2374−2380.
(52) Rauert, C. B.; Harrad, S. J. Mass Transfer of PBDEs from
Plastic TV Casing to Indoor Dust via Three Migration Pathways - A
Test Chamber Investigation. Sci. Total Environ. 2015, 536, 568−574.
(53) Eichler, C. M. A.; Hubal, E. A. C.; Xu, Y.; Cao, J.; Bi, C.;
Weschler, C. J.; Salthammer, T.; Morrison, G. C.; Koivisto, A. J.;
Zhang, Y.; Mandin, C.; Wei, W.; Blondeau, P.; Poppendieck, D.; Liu,
X.; Delmaar, C. J. E.; Fantke, P.; Jolliet, O.; Shin, H.-M.; Diamond, M.
L.; Shiraiwa, M.; Zuend, A.; Hopke, P. K.; von Goetz, N.; Kulmala,
M.; Little, J. C. Assessing Human Exposure to Chemicals in Materials,
Products and Articles: A Modular Mechanistic Framework. Environ.
Sci. Technol. 2021, 55, 25−43.
(54) Hammel, S. C.; Hoffman, K.; Phillips, A. L.; Levasseur, J. L.;
Lorenzo, A. M.; Webster, T. F.; Stapleton, H. M. Comparing the Use
of Silicone Wristbands, Hand Wipes, And Dust to Evaluate Children’s
Exposure to Flame Retardants and Plasticizers. Environ. Sci. Technol.
2020, 54, 4484−4494.
Environmental Science & Technology Letters pubs.acs.org/journal/estlcu Letter
https://doi.org/10.1021/acs.estlett.0c01006
Environ. Sci. Technol. Lett. 2021, 8, 326−332
332
Stuchlík Fišerová, Petra, Lisa Melymuk, Klára Komprdová, Elena Domínguez-Romero,
Martin Scheringer, Jiří Kohoutek, Petra Přibylová, Lenka Andrýsková, Pavel Piler, Holger M.
Koch, Martin Zvonař, Marta Esteban-López, Argelia Castaño, Jana Klánová. 2022. "Personal
Care Product Use and Lifestyle Affect Phthalate and DINCH Metabolite Levels in Teenagers
and Young Adults." Environmental Research 213: 113675.
https://doi.org/10.1016/j.envres.2022.113675
Environmental Research 213 (2022) 113675
Available online 11 June 2022
0013-9351/© 2022 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Personal care product use and lifestyle affect phthalate and DINCH
metabolite levels in teenagers and young adults
Petra Stuchlík Fiˇserov´a a
, Lisa Melymuk a,*
, Kl´ara Komprdov´a a
, Elena Domínguez-Romero a
,
Martin Scheringer a
, Jiˇrí Kohoutek a
, Petra Pˇribylov´a a
, Lenka Andrýskov´a a
, Pavel Piler a
,
Holger M. Koch b
, Martin Zvonaˇr a,c
, Marta Esteban-L´opez d
, Argelia Casta˜no d
, Jana Kl´anov´a a
a
RECETOX, Faculty of Science, Masaryk University, Kotlarska 2, Brno, Czech Republic
b
Institute for Prevention and Occupational Medicine of the German Social Accident Insurance - Institute of the Ruhr-University Bochum (IPA), Bürkle-de-la-Camp-Platz
1, Bochum, Germany
c
Faculty of Sports, Masaryk University, Kamenice, Brno, Czech Republic
d
National Centre for Environmental Health, Instituto de Salud Carlos III, Majadahonda, Madrid, Spain
A R T I C L E I N F O
Keywords:
Urine
Phthalates
Human biomonitoring
Plasticizers
Human exposure
A B S T R A C T
Humans are widely exposed to phthalates and their novel substitutes, and considering the negative health effects
associated with some phthalates, it is crucial to understand population levels and exposure determinants. This
study is focused on 300 urine samples from teenagers (aged 12–17) and 300 from young adults (aged 18–37)
living in Czechia collected in 2019 and 2020 to assess 17 plasticizer metabolites as biomarkers of exposure. We
identified widespread phthalate exposure in the study population. The diethyl phthalate metabolite monoethyl
phthalate (MEP) and three di (2-ethylhexyl) phthalate metabolites were detected in the urine of >99% of study
participants. The highest median concentrations were found for metabolites of low-molecular-weight (LMW)
phthalates: mono-n-butyl phthalate (MnBP), monoisobutyl phthalate (MiBP) and MEP (60.7; 52.6 and 17.6 μg/L
in young adults). 1,2-cyclohexanedicarboxylic acid diisononyl ester (DINCH) metabolites were present in 68.2%
of the samples with a median of 1.24 μg/L for both cohorts. Concentrations of MnBP and MiBP were similar to
other European populations, but 5–6 times higher than in populations in North America. We also observed large
variability in phthalate exposures within the study population, with 2–3 orders of magnitude differences in
urinary metabolites between high and low exposed individuals. The concentrations varied with season, gender,
age, and lifestyle factors. A relationship was found between high levels of MEP and high overall use of personal
care products (PCPs). Cluster analysis suggested that phthalate exposures depend on season and multiple lifestyle
factors, like time spent indoors and use of PCPs, which combine to lead to the observed widespread presence of
phthalate metabolites in both study populations. Participants who spent more time indoors, particularly
noticeably during colder months, had higher levels of high-molecular weight phthalate metabolites, whereas
participants with higher PCP use, particularly women, tended to have higher concentration of LMW phthalate
metabolites.
Abbreviations: 5cx-MEPP, mono(2-ethyl-5-carboxy-pentyl) phthalate; 5OH-MEHP, mono(2-ethyl-5-hydroxy-hexyl) phthalate; 5oxo-MEHP, mono(2-ethyl-5-oxohexyl)
phthalate; BzBP, benzyl butyl phthalate; BMI, body mass index; CELSPAC, Central European Longitudinal Studies of Parents and Children; cx-MINCH,
cyclohexane-1,2-dicarboxylic acid-mono(carboxy-isooctyl) ester; cx-MiNP, 7-carboxy-(monomethyl-heptyl) phthalate; DEHP, di(2-ethylhexyl) phthalate; DEP,
diethyl phthalate; DCHP, dicyclohexyl phthalate; DiBP, diisobutyl phthalate; DiDP, diisodecyl phthalate; DINCH, 1,2-cyclohexanedicarboxylic acid diisononyl ester;
DiNP, diisononyl phthalate; DMP, Dimethyl phthalate; DnBP, di-n-butyl phthalate; DnOP, di-n-octyl phthalate; DPHP, bis(2-propylheptyl) phthalate; ELSPAC, European
Longitudinal Study of Pregnancy and Childhood; EU, European Union; HBM, human biomonitoring; HMW, high-molecular weight; IQ, intelligence quotient;
LMW, low-molecular weight; MBzP, monobenzyl phthalate; MEHP, mono(2-ethylhexyl) phthalate; MEP, monoethyl phthalate; MCHP, monocyclohexyl phthalate;
MiBP, monoisobutyl phthalate; MiNP, monoisononyl phthalate; MMP, monomethyl phthalate; MnBP, mono-n-butyl phthalate; MnOP, mono-n-octyl phthalate; OHMiDP,
mono-hydroxy-isodecyl phthalate; OH-MINCH, cyclohexane-1,2-dicarboxylic acid-mono(hydroxyl-isononyl) ester; OH-MiNP, 7-hydroxy-(monomethyl-octyl)
phthalate; oxo-MiDP, mono-oxo-isodecyl phthalate; oxo-MiNP, 7-oxo-(monomethyl-octyl) phthalate; PCPs, personal care products; PVC, polyvinylchloride.
* Corresponding author.
E-mail address: lisa.melymuk@recetox.muni.cz (L. Melymuk).
Contents lists available at ScienceDirect
Environmental Research
journal homepage: www.elsevier.com/locate/envres
https://doi.org/10.1016/j.envres.2022.113675
Received 27 December 2021; Received in revised form 8 June 2022; Accepted 10 June 2022
Environmental Research 213 (2022) 113675
2
1. Introduction
Phthalate esters, commonly referred to as phthalates, are heavily
used as plasticizers and solvents because of their durability and stability.
Phthalates are high production volume chemicals (Heudorf et al., 2007)
with worldwide production of around 5.5 million tonnes per year (Fr´ery
et al., 2020). Phthalates with low molecular weight (LMW) and shorter
chains (e.g. DMP, DEP, DBP and DiBP) are often used as solvents in
personal care products (PCPs), cosmetics, and pharmaceuticals, and as
plasticizers in non-polyvinyl chloride (PVC) products, including textiles,
paints, adhesives, and food packaging. Phthalates with high molecular
weight (HMW) and longer chains (e.g. DEHP, DiNP and DPHP) are
mainly used as plasticizers in PVC products, including in medical devices
and children’s toys (Fr´ery et al., 2020; Koch and Angerer, 2012;
Silano et al., 2019). Phthalates are not covalently bonded to plastic,
therefore they can be emitted from consumer products and become
widely distributed in humans and the environment (Husøy et al., 2019).
In humans, several phthalates are considered endocrine disruptors
(WHO, 2013) associated with potential negative health effects. The effects
on developmental and reproductive disorders, including neurobehavioral
disorders and low IQ, respiratory problems or asthma and
other allergic disorders have been reviewed elsewhere (Hlisníkov´a et al.,
2020; Katsikantami et al., 2016; Wang et al., 2019). Due to these adverse
health effects, some phthalates have been legislatively banned in the EU
(EC, 2009; REACH, 2006) and alternative non-aromatic, non-­
ortho-phthalate plasticizers are seeing increasing use (Lemke et al.,
2021; Frederiksen et al., 2020; Lessmann et al., 2019). One of these
plasticizers is DINCH (1,2-cyclohexanedicarboxylic acid diisononyl
ester), which is used as a substitute for HMW phthalates, mainly DEHP
and DiNP, particularly in sensitive products such as toys, food contact
material and medical devices (Koch et al., 2013; Correia-S´a et al., 2017;
Kasper-Sonnenberg et al., 2019). DINCH production is high (>10,000
t/year production and import to EU, 300,000 t/y global production),
and DINCH is among the alternative plasticizers seeing substantial increases
in use and exposure (Bui et al., 2016; Kasper-Sonnenberg et al.,
2019). DINCH intake thresholds and hazard-based limit values suggest a
lower hazard than for phthalates, however, there is particular concern
about DINCH exposure in children, given its use as a replacement for
HMW phthalates in toys, and evidence of exposure close to the tolerable
daily intake (Bui et al., 2016).
Inhalation and dermal uptake is a relevant uptake route for the more
volatile LMW phthalates, like DMP, DEP and DBP (Giovanoulis et al.,
2018; Wormuth et al., 2006; Janjua et al., 2008; Lorber et al., 2017).
Oral routes play the most important roles in total exposure to the HMW
phthalates like DEHP and DiNP (Koch et al., 2013; Martínez et al., 2018;
Correia-S´a et al., 2018). After exposure, phthalates in humans are
rapidly metabolized by phase I and phase II reactions and excreted in
both conjugated and free forms via urine (Silva et al., 2003), and
partially also in feces (Domínguez-Romero and Scheringer, 2019). Thus,
human biomonitoring typically focuses on phthalate metabolites in
urine (Koch and Calafat, 2009; Calafat et al., 2013). More polar and
LMW phthalates are hydrolysed to monoester forms and are eliminated
mostly in free form and glucuronidated conjugates. In contrast, longer
chain HMW phthalates (and DINCH) are further metabolized to secondary,
alkyl chain oxidised metabolites (Fr´ery et al., 2020; Wittassek
et al., 2011; Schütze et al., 2017).
Biomonitoring suggests that nearly all the European population is
exposed to phthalates (Den Hond et al., 2015; Koppen et al., 2019). A
recent study on phthalate metabolites in 140 Norwegian adults found 10
metabolites in 100% of the samples and all remaining phthalate and
DINCH metabolites in 88–97% of the samples (Husøy et al., 2019). In a
Slovenian study of 387 people, detection rates were 97–100% for all
phthalate metabolites (Runkel et al., 2020). Several recent studies also
point out the rather rapid changes in exposures to phthalates and their
substitutes due to market changes and regulatory measures (Frederiksen
et al., 2020; Apel et al., 2020; Lemke et al., 2021; Gyllenhammar et al.,
2017). Human biomonitoring (HBM) is a useful and necessary tool to
assess total exposure to phthalates in a timely and rapid manner,
regardless of the route of exposure. In combination with other accompanying
data (e.g. questionnaire data) HBM provides valuable information
to help us better understand important exposure sources, as well
as the health effects related to chemical exposure. HBM4EU is a joint EU
project co-funded under Horizon 2020 to coordinate and advance
human biomonitoring, involving 30 countries, the European Environment
Agency and the European Commission. HBM under HBM4EU
provides better evidence and understanding of the exposure of European
citizens to chemicals, and the possible impact of chemical exposure on
human health. In this study, we focus on plasticizer metabolites in urine
samples from teenagers and young adults from the Czech Republic,
collected as a part of the HBM4EU project.
2. Materials and methods
2.1. Sample information
Three hundred urine samples were obtained from 12 to 17-year-old
teenagers residing in the South Moravian region of Czechia, identified as
the “CELSPAC: Teenagers cohort” (TAC) (Table 1). The CELSPAC:
Teenagers study was approved by the Research Ethics Committee of
Masaryk University, Czech Republic (Ref. No: EKV-2019-046, dated May
27, 2019). Most urine was collected in 2019, between October and
December; one urine sample was collected in January 2020. An additional
300 urine samples were obtained from participants from 18 to 37
years of age, also from South Moravia, Czechia, identified as the “CELSPAC:
Young Adults cohort” (YAC) (Table 1). CELSPAC: Young Adults
represents a follow-up study of longitudinal ELSPAC study in the Brno
region of the Czech Republic (Piler et al., 2017) and follows ELSPAC
children born in 1991 and 1992, their siblings, and spouses (the study is
ongoing). The CELSPAC: Young Adults study was approved by the
ELSPAC Ethics Committee (Ref. No: ELSPAC/EK/2/2019, dated March
13, 2019). YAC urine samples were collected in 2019, between March
and December. All urine samples from the TAC and YAC cohorts were
spot samples.
All study participants completed a questionnaire at the time of urine
sample collection gathering information including age, gender, number
of siblings, education and household income, time spent indoors and
various parameters of the home environment, dietary information,
smoking (active and passive), PCP usage and chronic illnesses. Details
about the cohorts (age distribution, education and income, lifestyle
variables) are given in the Supplementary Material, Figures S1-S5,
Table S1.
Urine was analysed for creatinine concentration as well as for specific
gravity (SG) and a wide set of chemical biomarkers, including
phthalates and DINCH (presented here), as well as bisphenols, hydroxylated
polycyclic aromatic hydrocarbons and metals (in preparation).
2.2. Chemicals and reagents
High purity (>97%) standards of 15 phthalate metabolites (MEP,
MiBP, MBzP, MnBP, MCHP, MEHP (AccuStandard, USA), 5OH-MEHP,
5oxo-MEHP, 5cx-MEPP, (Toronto Research Chemicals, Canada),
MnOP, OH-MiNP, cx-MiNP, oxo-MiNP, OH-MiDP and oxo-MiDP (Cambridge
Isotope Laboratories, USA)), along with a creatinine standard
solution, were purchased. OH-MINCH, cx-MINCH, OH-MINCH-d4 and
cx-MINCH-d2 standards were provided by the Institute for Prevention
and Occupational Medicine of the German Social Accident Insurance Institute
of the Ruhr-University Bochum (IPA, Germany). High purity
(>98%) standards of 15 isotopically-labelled phthalate metabolites
(MEP-13
C4, MiBP-13
C4, MBzP-13
C4, MnBP–13
C4, MCHP-13
C4, 5OH-
MEHP-13
C4, 5oxo-MEHP-13
C4, 5cx-MEPP-13
C2, OH-MiNP-13
C2, cx-
MiNP-13
C4, oxo-MiNP-13
C2, OH-MiDP-13
C2, oxo-MiDP-13
C2, (Cambridge
Isotope Laboratories, USA), MEHP-d4 and MnOP-d4 (Chiron, Norway))
P. Stuchlík Fiˇserov´a et al.
Environmental Research 213 (2022) 113675
3
were purchased.
The following chemicals were used: a solution of β-glucuronidase
(E. coli, Type IX-A 1–5MU/g), acetic acid, ammonium fluoride, ammonium
acetate and Milli-Q water (Sigma-Aldrich, USA). The following
solvents were purchased: acetonitrile and methanol for LC-MS (Biosolve
Chimie, France), dimethylsulfoxide (DMSO, Sigma-Aldrich, USA).
2.3. Sample preparation procedure
Frozen aliquots were allowed to reach laboratory temperature, homogenized
with a Wizard Advanced IR vortex mixer (Velp Scientifica,
USA) and 500 μL of each sample were pipetted into a 96-well sample
collection plate with 2 mL square wells (Waters, Prague). Then, 500 μL
of a mixture containing β-glucuronidase solution (1000 units/mL) and
isotopically-labelled standards (c = 10 ng/mL) in 0.1 M acetate buffer
were added to the samples. The 96-well plate was covered with foil and
samples were incubated using a sample concentrator (Miu Lab, China)
for 120 min at 55 ◦
C. After incubation, 50 μL of 1% acetic acid was
added. Samples were then precleaned using 96-well plate SPE (Oasis
HLB; 3 mL, 60 mg; Waters, Ireland), which was previously conditioned
with 1 mL of methanol and activated with 0.1% acetic acid. Samples
were passed through the SPE plate, then the wells were washed with
0.1% acetic acid and dried using a Laboport vacuum pump (KNF, Germany)
for at least 30 s. Finally, the samples were eluted with 1.5 mL of
acetone and collected into a new 96-well plate containing 10 μL of
DMSO. After the clean-up step, samples were concentrated to 10 μL.
Finally, 500 μL of 50% methanol were added, samples were covered
with a foil, homogenized and kept in the fridge (4 ◦
C) until analysis.
2.4. Liquid chromatography-mass spectrometry (LC-MS) conditions
The analytical method for phthalate metabolite determination was
developed on the Agilent 1290 Infinity II HPLC system (Agilent Technologies,
Germany) for separation. An Acquity UPLC BEH C18 (100 ×
2.1 mm, 1.7 μm, Waters, Czechia) was used as an analytical column. The
mobile phases were A: 0.1 mM ammonium fluoride in Milli-Q water and
B: 0.1 mM ammonium fluoride in methanol. The gradient elution was set
up as follows: mobile phase A (10%) for 7 min, followed by an immediate
increase to 50% for the rest of the analysis. The injection volume
was 5 μL and the flow rate was set to 250 μL/min.
Agilent QQQ 6495 (Agilent Technologies, Germany) was used for
analyte detection, with an electrospray ionisation (ESI) source, in
negative mode. Optimal parameters for phthalate metabolites are shown
in Table S2. The isotopic dilution method was used for data quantification.
A corresponding isotopically-labelled standard was used for each
compound as the internal standard. MassHunter software was used to
process the data from instrumental analysis.
2.5. Specific gravity
For the SG measurement, a “Pocket” Urine Specific Gravity Refractometer
PAL-10 S (Atago, Japan) was used. First, 300 μl of Milli-Q water
was pipetted to the refractometer to reach a value of 1.000. Then, the
refractometer was wiped with a paper towel. The urine sample was
vortexed and 300 μL were pipetted onto the refractometer. When the
temperature stabilised, the value was deducted from the refractometer.
The procedure was repeated for all the samples.
2.6. Quality assurance/quality control (QA/QC)
The methods used were developed and validated under the HBM4EU
framework through the successful completion of four rounds of proficiency
testing (Esteban L´opez et al., 2021). We report 10 phthalate and 2
DINCH biomarkers that are quality assured under HBM4EU (both
external and internal quality assurance) and 4 biomarkers under additional
internal quality assurance.
External quality assurance: For external quality assurance we participated
in all four rounds of the HBM4EU Proficiency testing (Esteban
L´opez et al., 2021) and received certificates for the following biomarkers:
MEP, MBzP, MiBP, MnBP, MEHP, 5OH-MEHP, 5oxo-MEHP,
5cx-MEPP, OH-MiNP, cx-MiNP (Elbers and Mol, 2019a), OH-MINCH
and cx-MINCH (Elbers and Mol, 2019b).
Internal quality assurance: Procedural blanks were prepared with
Milli-Q water, approximately one blank for every ten urine samples. The
samples were blank subtracted based on the average mass of metabolite
found in the blanks (Table S3). LODs were defined as three times the
standard deviation of the blank samples (Table S3). A calibration curve
was prepared in the range of 0.01–100 ng/mL, containing 10 concentration
points and showing linearity. QC samples were analysed
approximately every ten samples. The average recoveries of QC samples
ranged from 83 to 110% (except for MnBP with 124.1%) with RSD
ranging from 5.1 to 16.5 (except for MEP and MnBP with 38.1 and
40.5% respectively).
2.7. Data adjustment
SG adjustment was applied to urine samples, since creatinine
correction has a higher dependence on age (Carrieri et al., 2000), time of
urine sampling, sex, activity, diet (Miller et al., 2004), muscularity, BMI,
disease status (Wang et al., 2015) and seasonal variations (Pearson et al.,
2009). Moreover, degradation studies suggest that after only 10 days,
approximately 20% of creatinine is degraded via freezing (storage at
− 20 ◦
C), while SG is not (Schneider et al., 2002). Considering all this, SG
correction was deemed to be the more reliable correction for the TAC
and YAC samples.
2.8. Data analysis
For the correlation analyses, only values > LOD were used. For the
rest of the statistical analyses, values below LOD were substituted with
LOD/√2. Nonparametric Spearman’s correlation was used to determine
the relationship among phthalate metabolites. To investigate seasonal
variations, YAC samples were divided into two groups according to the
sampling season. The cold season covered seven months from October to
April, and the warm season covered five months from May to September.
Basic statistical characteristics (detection frequency, median, minimal
Table 1
Profile of studied cohorts. YAC represents Young Adult cohort, TAC represents Teenage cohort, Nmis represents number of samples where we lack corresponding data
from questionnaires.
Young Adults (YAC) (n = 300) Teenagers (TAC) (n = 300)
Median 10th90th Minmax Nmis Median 10th90th Minmax
Age (years) 27 26–28 18–37 13 12–15 12–17
Weight (kg) 72.6 56.5–92.5 47.3–117.3 2 Not collected
Height (cm) 174.1 163.2–188.3 155.1–202.3 2 Not collected
BMI (kg m− 2
) 23.3 19.9–28.7 17.3–39.2 2 Not collected
Gender female 155; 52% 124; 42%
male 145; 48% 176; 58%
P. Stuchlík Fiˇserov´a et al.
Environmental Research 213 (2022) 113675
4
and maximal value with 5th and 95th percentiles) are shown in Table S4
for YAC and in Table S5 for TAC. Table S6 refers to number of samples
and medians for categories of the exposure determinants used for nonparametric
testing.
Linear regression was used to explore the relationship between
phthalate metabolites and multiple exposure factors (gender, age, season
of sample collection, time indoors, and sum of all PCP products
used). Only metabolites with more than 75% of samples above the LOD
for each cohort were used for regression analysis (MEP, MiBP, MBzP,
MnBP, 5OH-MEHP, 5cx-MEPP, OH-MiNP). The regression analysis used
gender and season as categorical predictors; age, time spent indoors, and
sum of PCPs were used as continuous predictors. Use of individual PCPs
could not be incorporated as predictors in the regression models, due to
the small sample sizes in some groups stratified by e.g., cohort/age,
gender and season. Metabolite concentrations were log-transformed,
and outliers were excluded prior to analysis. The normal distribution
of the residuals was checked by using histograms and the KolmogorovSmirnov
test. Regressions for each exposure factor and metabolite were
adjusted for other factors to explain the main effects of each factor.
Another regression model for each metabolite was computed with all
factors and their interactions to determine the maximum explained
variability (coefficient of determination, R2
) in phthalate concentrations
by the given factors. Factor interactions were added to the model
because we assume that one factor can influence the effect of another
factor (e.g. indoors and season) on the dependent variable (phthalate
metabolite concentrations).
To supplement the key factors identified by the regression analysis,
individual exposure factors were also examined, both individually, and
stratified by age, season, and gender. The non-parametric Mann-Whitney
U test was used to test for differences in concentrations of phthalate
metabolites with questionnaire parameters with two categories such as
gender, season (cold vs. warm), smoking (YES/NO), redecoration and
renovation made at home in the last two years (YES/NO), drinking
beverages from plastic bottles (YES/NO) and for the use of some PCPs
(Table S7). Tests were also performed for each combination of factors.
Due to inhomogeneity of the variances and non-equal sample sizes in
some categories of PCPs, nonparametric Kruskal-Wallis ANOVA was
used to test for differences in concentrations of phthalate metabolites
between multiple categories of frequencies of PCP use. To remove the
influence of other factors, participants were first stratified by age (TAC
vs YAC), gender, and season before testing of PCPs. For some frequencies
of PCP use stratified by gender, age and season, the number of samples
for testing was insufficient, due to the nature of the PCP use (e.g. eye
make-up or nail polish is used mainly by women). The PCP use categories
“never” and “sometimes” for shampoo and deodorant were
combined because very few participants did not use these products at all.
Tests were performed when the number of samples in each category was
greater than 20 (Table S7) and only for metabolites with more than 75%
samples above the LOD for each cohort (MEP, MiBP, MBzP, MnBP, 5OHMEHP,
5cx-MEPP, OH-MiNP).
Finally, two cluster analyses were completed as an explanatory
technique for cumulative phthalate exposure and cumulative PCP use:
i) First, Ward’s hierarchical clustering method with the squared
Euclidean distance was used as an explanatory technique to create
clusters of participants with similar phthalate exposure. Metabolite
concentrations were log-transformed and standardized (z-score)
prior to cluster analysis. The cluster analysis provides insight into the
total cumulative exposure of all metabolites. Specifically, participants
with similar concentrations of phthalate metabolites (e.g.
participants with high/low levels of all metabolites) were clustered.
Then it was determined how the exposure factors (age, gender,
season, time spent indoors, PCPs) were distributed among these
clusters. Differences in cumulative concentrations of LMW, HMW
phthalate metabolites and DINCH between clusters were tested for
by the Kruskal-Wallis ANOVA.
ii) In the second case, cluster analysis was used to create clusters of
participants with similar use of personal care products. Frequencies
of use of individual PCPs were expressed as number of days per week
and clustered using Ward’s hierarchical clustering method with the
squared Euclidean distance. In contrast to the Kruskal-Wallis
ANOVA, which tests differences for each PCP separately, the cluster
analysis shows the common use of all PCPs and complements the
results from the regression analysis where individual PCPs and their
interactions could not be included. This analysis was conducted for
the YAC and TAC cohorts separately. The distribution of gender and
season was determined for each cluster and differences in the concentrations
of each metabolite in the clusters were tested using
Kruskal-Wallis ANOVA.
3. Results and discussion
Twelve of the 17 targeted phthalate metabolites were broadly
detected in TAC and YAC. The most frequently detected compounds
were MEP, 5oxo-MEHP and 5cx-MEPP, detected in >99% of the YAC
samples and in all TAC samples, and OH-MINP and 5OH-MEHP were
often detected (>98%) in both cohorts. The detection frequencies, medians,
geometric means, and ranges are shown in Table 2. Other statistical
characteristics (minimum, maximum, 5th and 95th percentiles) for
both cohorts are shown in Tables S4 and S5.
In the YAC, the highest median concentration was found for MnBP,
followed by MiBP and MEP (60.6; 52.6 and 17.6 μg/L, respectively). All
DEHP metabolites, OH-MiNP, oxo-MiNP and OH-MINCH had medians
between 1.34 and 7.92 μg/L, while the other metabolites were detected
with median concentrations under 1.01 μg/L (Table 2). In the TAC, the
highest median concentrations were also found for MnBP, followed by
MEP and MiBP (45.5; 38.1 and 28.8 μg/L, respectively). MBzP, all DEHP
metabolites, the secondary DiNP metabolites and OH-MINCH were
detected with the medians in the range of 1.71–13.9 μg/L. Despite the
use of DINCH as a replacement for DEHP in sensitive applications (Bui
et al., 2016), the DINCH metabolites are detected at lower levels and
with lower detection frequency than DEHP metabolites. The other metabolites
were detected in median concentrations below 1 μg/L
(Table 2). MEP, MiBP, MnBP, MBzP, 5OH-MEHP, 5cx-MEPP, OH-MiNP,
cx-MiNP, OH-MiDP and OH-MINCH (marked with a
in Table 2) were
selected as priority metabolites for further statistical analysis based on
their appropriateness as biomarkers of the parent compounds and
detection frequency. Correlations between individual phthalate metabolites
were calculated to examine associations between specific phthalate
metabolites and common sources. The strongest correlations were
found between metabolites of the same parent compound (e.g., DEHP
metabolites 5OH-MEHP, 5oxo-MEHP and 5cx-MEPP; and DINCH metabolites
OH-MINCH, cx-MINCH) (Figure S6).
3.1. Determinants of phthalate exposure
The regression analysis on all data (YAC + TAC, N = 600) for MiBP,
MEP, MBzP, MnBP, 5cx-MEPP, 5OH-MEHP, OH-MINP show the contributions
of both individual factors and their interactions to metabolite
concentration (Table S8a,b).
Regression analysis highlighted the differences between exposures to
low and high molecular weight phthalates. Gender, sampling season,
and age were significant factors for HMW phthalate metabolites 5cxMEPP,
5OH-MEHP, OH-MINP, with explained variability around 10%.
The only significant interaction for HMW metabolites were gender and
age for 5cx-MEPP, time indoors, and age for 5OH-MEHP and time indoors
and season for OH-MINP, but these interactions did not significantly
increase R2
(Table S8). For the LMW phthalate metabolites MiBP,
MEP, MBzP, the main significant factors were similar – gender and age –
with an explained variability between 10% and 16%, but the significant
interaction of age and gender with total PCP use increased the explained
variability of MiBP and MEP to 18% and 25%, respectively. This
P. Stuchlík Fiˇserov´a et al.
Environmental Research 213 (2022) 113675
5
suggests the greater importance of PCPs use for LMW phthalate exposure
(Table S8). We found no relationship between the factors and MnBP
concentration. However, an important outcome of this is the limited
explanatory power of these selected factors for phthalate exposure.
Thus, the majority of the variance in phthalate metabolite levels remains
unexplained by these factors.
We evaluated the impact of key determinants in greater detail: age,
season, gender, time indoors and PCP use on phthalate metabolite levels
to identify key determinants of phthalate exposure in the teenager and
young adult cohorts.
3.1.1. Age
A limited relationship between age as a continuous variable and
HMW metabolites was observed (Table S8), but when age was treated as
a categorical variable (YAC vs. TAC) a few clear differences were
apparent (Fig. 1, Table S9). Notably, MiBP was significantly higher in
YAC (p < 0.05), with median concentrations two times higher than in
TAC. MnBP was also detected in higher concentrations (p < 0.1) in YAC
(60.65 μg/L compared to 45.47 μg/L in TAC). On the other hand, statistically
significant differences were found for MEP, MBzP and 5OHMEHP
with median concentrations approximately two times higher in
TAC; and the rest of the secondary DEHP metabolites were also detected
in higher concentrations in TAC (Table 2), but these differences were not
statistically significant. These differences may be due to lifestyle differences
between teenagers and young adults, and are discussed further
below.
3.1.2. Season
Analysis of the season of sample collection was stratified by cohort
and gender, and only performed on the YAC, as all TAC samples were
collected in the cold season. The results of the non-parametric tests
showed that in YAC, statistically significant differences were found for
5OH-MEHP and OH-MiNP with 20% and 100% higher concentrations,
respectively, in urine samples collected in the warm season (p < 0.05),
and concentrations of cx-MiNP were also 50% higher (p < 0.1)
(Tables S6 and S9). This may indicate a link between levels of phthalates
in the ambient environment and human exposures (Pilka et al., 2015).
Phthalates are found in outdoor air in higher concentrations in warmer
months (Pilka et al., 2015; Puklov´a et al., 2019), attributed to increases
in both primary and secondary emissions at higher temperatures
Table 2
LODs, detection frequencies and median concentrations (μg/L, SG adjusted) with range (min-max) and geometric means of phthalate metabolites for both cohorts.
YAC TAC
Parent
phthalate
Phthalate
metabolite
LOD (μg/
L)
Detection frequency
(%)
Median
(minmax)
Geo.
Mean
Detection frequency
(%)
Median
(minmax)
Geo.
Mean
DEP MEP a
0.20 99.0 17.5 (0.02–860) 22.0 100 38.1 (3.04–704) 42.2
DiBP/DnBP MiBP a
0.07 99.6 52.6 (0.07–446) 64.0 97.3 28.8 (0.03–486) 26.1
BzBP MBzP a
0.10 75.6 0.69 (0.10–25.3) 0.74 94.6 1.74 (0.07–112) 1.85
DnBP/BzBP MnBP a
0.18 96.3 60.6 (0.18–664) 60.5 91.0 45.5 (0.08–303) 28.8
DCHP MCHP 0.20 0 0.20 (0.20–0.20) 0.18 0.3 0.14 (0.08–1.07) 0.15
DEHP MEHP 0.18 93.6 2.29 (0.18–40.1) 2.57 93.6 2.14 (0.11–63.6) 2.09
DEHP 5OH-MEHP a
0.07 99.6 7.85 (0.07–150) 9.30 99.3 13.9 (0.03–679) 14.2
DEHP 5oxo-MEHP 0.07 99.3 3.65 (0.07–113) 4.43 100 5.56 (1.33–74.5) 6.00
DEHP 5cx-MEPP a
0.07 99.6 5.04 (0.07–108) 6.02 100 7.52 (1.74–128) 8.32
DnOP MnOP 0.20 0.3 0.20 (0.20–2.95) 0.18 3.3 0.14 (0.08–6.63) 0.16
DiNP OH-MiNP a
0.30 98.3 7.92 (0.30–1649) 11.1 98.3 11.4 (0.16–169) 12.1
DiNP cx-MiNP a
0.20 49.6 0.20 (0.20–105) 0.48 100 5.34 (0.96–503) 6.19
DiNP oxo-MiNP 0.20 92.0 1.99 (0.20–111) 2.59 97.3 2.75 (0.16–228) 3.01
DiDP OH-MiDP a
0.20 31.6 0.20 (0.20–102) 0.37 52.3 0.55 (0.09–13.4) 0.48
DiDP oxo-MiDP 0.10 41.6 0.10 (0.10–60.4) 0.26 55.0 0.39 (0.05–14.2) 0.30
DINCH OH-MINCH a
0.30 61.3 1.34 (0.30–952) 1.67 75.0 1.71 (0.15–494) 1.76
DINCH cx-MINCH 0.25 39.6 0.25 (0.25–277) 0.74 51.0 0.76 (0.10–204) 0.80
a
Indicates a metabolite was selected as priority biomarker for further analysis.
Fig. 1. Median concentration of phthalate metabolites in both cohorts. M represents males, F represents females.
P. Stuchlík Fiˇserov´a et al.
Environmental Research 213 (2022) 113675
6
(Vasiljevic et al., 2021). Higher emissions, and resulting higher environmental
concentrations suggest the potential for greater human
exposure in warm seasons, reflected in urinary metabolites (Puklov´a
et al., 2019).
3.1.3. Gender
The results of the non-parametric tests show statistically significant
gender differences in three phthalate metabolites (MEP, 5oxo-MEHP and
5cx-MEPP; Figure S7) in YAC, and no gender differences for any
metabolite in TAC (Table S9). There are greater gender differences in
lifestyles in the YAC cohort compared to the TAC: teenagers typically
spend weekdays in the same location (school) regardless of gender,
while there is greater differentiation in adult occupations by gender, and
PCP use in teenagers is similar across males and females, but differs
significantly by gender for young adults (Figures S4 and S5).
In YAC, MEP had significantly higher concentrations in females, and
this difference was more significant in the cold season (p < 0.05 in cold
season, p < 0.1 in whole year) (Table S9). Two DEHP metabolites (5oxoMEHP
and 5cx-MEPP) also showed significantly higher concentrations
in females (Table S9). Gender-dependent differences in oxidative DEHP
metabolism have been identified (Tait et al., 2020), which could lead to
differences between male and female concentrations of urinary metabolites
(Koch et al., 2017). Higher MEP levels in women have been noted
in other studies (Runkel et al., 2020; Wormuth et al., 2006) attributed to
gender differences in PCP use. This suggests an association between
higher MEP levels and frequency and amount of PCP use, which the
gender differences are indirectly indicating.
3.2. Lifestyle determinants of phthalates exposure
The seasonal and gender-related differences, while in some part
attributable to differences in environmental levels and physiological
factors, are largely due to the lifestyle differences that manifest across
seasons and between different populations. The key lifestyle aspects
which are hypothesized to be related to phthalate exposure are time
spent indoors, indoor activities such as cleaning, use of personal care
products containing phthalates, and diet. We examined the individual
influence of individual exposure determinants, excluding diet, as well as
possible combined effects. Diet was excluded because we had limited
information on dietary patterns that would be directly relevant to
phthalate exposure. We only examined a link between phthalate
metabolite levels and frequency of drinking from plastic bottles, and for
this variable no relationships were found.
3.2.1. Exposures via the indoor environment
When comparing urinary metabolite concentrations with selfreported
time spent indoors, we did not find a statistically significant
relationship between the time spent indoors and individual phthalate
metabolite for either YAC or TAC. Higher MEP, MBzP, 5oxo-MEHP,
5OH-MEHP, 5cx-MEPP, OH-MiNP and cx-MiNP concentrations were
observed in TAC samples, indicating higher exposures to DEP, DEHP and
DiNP. This may be related to longer times spent indoors by teenagers
(with a median of 14 h, compared to 12 h for YAC, Figure S3). Moreover,
this is similar to the effect of inclusion of time spent indoors in the
regression model, where interaction of time spent indoors with age for
5OH-MEHP and with season for OH-MINP was statistically significant
but did not lead to a significant increase in explained variability
(Table S8). Urinary levels of MBzP and MEP have been positively
correlated with indoor levels of BzBP and DEP in several previous
studies, suggesting exposure via indoor air and dust as important routes
of exposure to these phthalates (Adibi et al., 2008; Bek¨o et al., 2013;
Langer et al., 2014); however, we note that these studies did not identify
indoor sources as important for HMW phthalates.
No statistically significant differences were found between any metabolites
and renovations/redecorations done in the past 2 years.
3.2.2. Exposures via PCP use
Participants reported frequency of PCP use per week in the four
weeks prior to sample collection. The specific PCPs identified in questionnaires
were shampoo, scented products (deodorants and perfumes),
lotions, nail polish, make-up/foundation creams, lip balms and eye
make-up. We tested differences in phthalate metabolite concentrations
for different frequencies separately stratified by age (YAC vs TAC),
gender and season to remove the influence of these factors. In TAC, we
found statistically significantly higher levels of MEP for both genders in
participants reporting using deodorant compared with those “never” or
“once a week“ using them (Table S7), as well as an increasing trend with
increasing frequency of use (Fig. 2), which agrees with reports of high
levels of DEP in perfumes (Guo and Kannan, 2013) and a previous study
comparing urinary metabolites with PCP use (Nassan et al., 2017). In the
YAC cohort, no differences in MEP concentrations were found when
scented products were used because there is likely overlap with the use
of other PCPs (see below).
Associations with nail polish use were observed in YAC in both the
stratified and overall analyses. Women reporting use of nail polish had
statistically significant higher levels of 5cx-MEPP and OH-MINP in the
cold season than those reporting no use (Table S7), and when including
all YAC (male and female), participants reporting “never” using nail
polish had lower levels of 5cx-MEPP compared with those using nail
polish four to six times per week (Figure S8). However, due to the low
number of participants who used nail polish at a high frequency (7 times
per week, n = 3), the significance of the highest use category could not
be statistically tested. Nail polish use presents a documented pathway
for dermal uptake of chemicals, as noted by Mendelsohn et al. (2016),
and while exposure to DEHP is not typically associated with PCP use,
some nail polishes have been also identified to contain DEHP (Guo and
Kannan, 2013; Young et al., 2018). Moreover, nail polish is a long-term
PCP exposure route, as nail polishes typically stay on the nails for days to
weeks, meaning that the spot samples collected for this study may better
capture this exposure source, in contrast to rinse-off PCP products. In the
TAC cohort, no difference was observed in females who reported use vs.
no use of nail polish (Table S7), likely because almost all participants
used nail polish with a very low frequency (Figure S5A).
While the analysis comparing the use of individual products was able
to identify some relationships (notably with deodorant and nail polish
use), many of the product categories represent relatively small potential
exposures (e.g., lip balm, eye shadow), which independently may not
lead to substantial increases in phthalate exposures. Moreover, we do
not know which brands of products were used by study participants, and
there may be high variability in the phthalate content of products on the
market, e.g., only some nail polishes contain phthalates (Guo and
Kannan, 2013; Young et al., 2018). However, we hypothesized that by
clustering the study participants according to patterns of PCP use, we
could discern the influence of overall PCP use on human exposure to
Fig. 2. Difference in MEP concentration for different frequency (times per
week) of deodorant use in the TAC cohort (teenagers).
P. Stuchlík Fiˇserov´a et al.
Environmental Research 213 (2022) 113675
7
phthalates. Regression analysis shows that added interaction with the
total PCP use increased the explained variability of MEP concentrations
by 9%. In the case of TAC, the clusters do not provide any additional
insight into PCP use patterns, mainly due to the similarity of behaviour
among teenagers and low use of make-up products. However, for YAC,
clear patterns of PCP use and associated exposures are visible
(Figure S9). YAC data were separated into seven clusters according to
PCP use, with cluster A associated with the least overall PCP use, and
cluster D-G associated with frequent PCP use across most categories.
Clusters A, B, and C were dominated by use of only shampoo and scented
products, and included mainly male subjects, while clusters D-G
included mainly female subjects and had notably higher uses of additional
PCPs like lotion and decorative cosmetic products. This pattern of
higher overall use of PCPs, mainly decorative cosmetic products, is
connected with higher urinary concentrations of MEP (Figure S9) and
presents a possible explanation for the higher levels of MEP noted in
females (Table S7). However, there is no statistically significant difference
in MEP concentrations between clusters, mainly due to the high
variance in concentrations.
3.2.3. Patterns in overall phthalate exposure
In view of the complex factors leading to phthalate exposures, we
combined all phthalate metabolite data (YAC + TAC, N = 600) to
evaluate the overall pattern of phthalate metabolites and exposure determinants.
We hypothesize that the cumulative effect of different lifestyles
and behaviours drives phthalate exposures, rather than individual
products or activities.
Cluster analysis was selected as an additional multivariate technique
to provide insight into patterns of exposure and exposure determinants.
Cluster analysis was used to group participants who had similar
metabolite concentrations from the lowest (A) to the highest (H)
according to the median metabolite concentrations in the cluster
(Fig. 3). These clusters had significant differences in metabolites associated
with LMW phthalates, HMW phthalates and DINCH, with Cluster
H notably higher in HMW phthalate and DINCH metabolites, and Cluster
G having substantially higher levels of LMW phthalate metabolites
(Figure S10; Table S10). The lifestyle variables associated with each
cluster were then summarized. However, it is important to note that it is
not possible to strictly divide phthalate metabolites according to their
source, in this case, either from PCPs or from building materials/indoor
products. We note that the cluster with the highest phthalate metabolite
levels (Cluster H in Fig. 3) contains mainly cold season samples from
men with an average age of 19 years with high PCP use, especially of
products like shampoo and lotion which have larger application quantities,
and among the highest time spent indoors. Additionally, this is the
only cluster where a higher concentration of the alternative plasticizer
metabolite was observed: Cluster H has a much higher contribution of
OH-MINCH to the total plasticizer metabolites than any other cluster.
Use of DINCH as a DEHP and DINP replacement is primarily in sensitive
applications such as food contact materials, medical devices and toys
(Bui et al., 2016). The reasons for the predominance of OH-MINCH in
Cluster H are not fully clear; it may be due to behaviours associated with
increased time indoors, as well as food consumption habits which were
not well captured by the determinants of exposures available.
In contrast, a low exposure cluster (Cluster C) has low use of the PCPs
with large application quantities, and the lowest time spent indoors.
Similarly, clusters C and E, where participants declare lower time spent
indoors, have smaller contributions of HMW phthalate metabolites.
Additionally, clusters A and B have low levels and were largely all
collected in cold seasons, which corresponds with the identification of
low environmental levels and resulting low exposures during colder
months. In some cases, the relation between the metabolite levels and
Fig. 3. Cluster analysis of metabolite concentrations
in relation to lifestyle factors relating to phthalate
exposure. For the metabolites, median concentrations
in μg/L are given for each cluster. LMW metabolites
which are often connected with PCP use are shown in
tones of purple and green, HMW metabolites which
are often considered “indoor exposure” biomarkers
from consumer products and building materials are
shown in tones of red and pink. Alternative plasticizers
metabolites are shown in tones of grey. For
lifestyle factors, the values indicate average PCP
weekly use frequency (two-level scale – more
frequently used in tones of purple, less frequently used
in tones of pink), average hours per day spent indoors,
and average age. (For interpretation of the references
to colour in this figure legend, the reader is referred to
the Web version of this article.)
P. Stuchlík Fiˇserov´a et al.
Environmental Research 213 (2022) 113675
8
the lifestyle factors is unclear, e.g., phthalate metabolites are high in
cluster G, but that cluster has relatively low median use of PCPs, and low
time spent indoors.
The results of our cluster analysis support our hypothesis that
phthalate metabolite concentrations relate to a combination of separate
lifestyle factors affecting phthalate exposure, rather than one dominant
exposure source. Some part of this disconnect may be because we are
omitting a major phthalate exposure pathway: dietary ingestion. We
tested whether there was a relationship between phthalate metabolites
and consumption of drinks in plastic bottles, and found no relationship,
but beyond this, no factors related to dietary exposure were tested.
However, dietary ingestion is known to be important for many phthalates,
particularly the HMW phthalates (Giovanoulis et al., 2018; Wormuth
et al., 2006). Nevertheless, cluster analysis presents a useful
technique to profile the complexity of exposure factors and interrelations
between variables such as age, gender, and season with PCP
use and behaviours.
3.3. Comparison with other studies
We compared the median urinary concentrations of phthalates in
teenagers and young adults in our study with values reported in the
literature for these age groups. Specifically, median phthalate metabolite
concentrations from our study were compared with data from Germany
(GerES V study, 2015–2017) (Schwedler et al., 2020a, 2020b),
Canada (Canadian Health Measures Survey, cycle 5, 2016–2017)
(Health Canada, 2019) and USA (NHANES, 2015–2016) (CDC, 2019).
Marked differences between countries were found for MnBP and
MiBP concentrations, which were around 5–6 times higher in young
adults in our study than the values reported in adults from Canada and
USA. In teenagers, median MnBP and MiBP concentrations were also
generally higher in our study than the values reported in Canada, USA,
and Germany (Fig. 4, Table S11). However, data from other studies from
Czechia (Puklov´a et al., 2019) or other European countries (Slovakia
(Pilka et al., 2015) and Denmark (Søeborg et al., 2012)) show median
MnBP and MiBP concentrations more similar to those from Czechia than
to those from North America (Frederiksen et al., 2010). OH-MiNP concentrations
in young adults and teenagers in our study were around 15
times higher than the values reported for the same age groups in Canada,
with smaller differences (2-fold) in OH-MiNP levels between Czech and
German teenagers. Conversely, the median MBzP concentration in
Czech young adults was around 5–6 times lower than values reported for
this metabolite in Canada and the USA. DINCH metabolites are reported
in all regions at similar levels.
With the general similarities in development, climate and lifestyles
between Europe and North America, we attribute the differences in
specific metabolites to differences in food composition; in Europe, the
major sources of DEHP, DBP, DiBP and BzBP in adults is food or diet in
general (Wormuth et al., 2006). Moreover, differences in PCP use,
chemical legislation and building practices between North America and
the EU, as well as within the EU, can also lead to differences in phthalates
exposure of the general population between countries (Runkel
et al., 2020). Some smaller contribution may come from differences in
the identified cohort populations, e.g., teenagers are defined as 11–17
years old in Czechia, 14–17 years old in Germany, and 12–19 years old
in Canada and USA.
3.4. Study limitations
Samples for both cohorts were taken in different seasons. In some
cases, there are indications that self-reported data may not be correct, e.
g., implausible time spent indoors. A few participants declared an
average time spent indoors less than 8 h (sleep included), which seems,
particularly in the case of teenagers in cold/winter months, not probable.
Therefore, these values were excluded from the statistical analysis.
Moreover, questionnaires asked about products used in the previous
four weeks, which may not correspond directly to metabolites detected
in urine samples. Diet was excluded because we had limited information
on dietary patterns that would be directly relevant to phthalate exposure.
Another limitation is that urine samples were spot samples
collected at different times of day, with insufficient information about
the timing of sample collection to allow further grouping. Values of SG
were measured after one freeze-thaw cycle; the ideal measurement
would be on fresh urine samples (Pearson et al., 2009). Additionally, we
cannot distinguish the origin of some metabolites, particularly if the
parent phthalate is DiDP or DPHP (Gries et al., 2012; Koch et al., 2017).
4. Conclusions
We identified widespread exposure to phthalate esters in the Czech
population based on the quantification of phthalate metabolites in spot
urine samples from teenagers and young adults. While some relationships
were identified between individual exposure determinants (use of
certain PCPs, season, gender) and specific phthalate metabolites, in
general, the individual factors did not show a strong relationship with
phthalate metabolites. We hypothesize that phthalate exposures relate
Fig. 4. Comparison of urinary concentrations of phthalate metabolites in teenagers and young adults from Czechia, Germany, Canada and USA, logarithmic scale.
P. Stuchlík Fiˇserov´a et al.
Environmental Research 213 (2022) 113675
9
to a combination of separate lifestyle factors, rather than one dominant
exposure source. We tested this through cluster analyses, identifying the
profiles of study participants with higher levels of phthalate metabolites.
Participants who spent more time indoors, particularly noticeable during
colder months, had higher levels of HMW phthalate metabolites,
whereas participants with higher PCP use, particularly women, tended
to have higher concentrations of LMW phthalate metabolites. We
conclude that while phthalate exposure is ubiquitous, it is also very
variable, according to age, gender, time spent indoors, food, and lifestyle
in general.
Credit author statement
Petra Stuchlík Fiˇserov´a; – Methodology, Formal analysis, Investigation,
Data curation, Writing – original draft, Writing – review &
editing; Lisa Melymuk – Investigation, Writing – review & editing,
Supervision; Kl´ara Komprdov´a; – Data curation, Visualization; Writing
– review & editing; Elena Domínguez-Romero – Investigation, Writing
– review & editing; Martin Scheringer – Writing – review & editing; Jiˇrí
Kohoutek – Methodology, Formal analysis; Petra Pˇribylov´a; – Resources,
Supervision; Lenka Andrýskov´a; – Resources, Supervision;
Pavel Piler – Resources, Data curation; Holger M. Koch – Validation,
Writing – review & editing; Martin Zvonaˇr; – Supervision, Funding
acquisition, Resources; Marta Esteban-L´opez – Supervision, Validation,
Writing – review & editing; Argelia Casta˜no – Supervision, Validation,
Writing – review & editing; Jana Kl´anov´a; – Resources, Funding
acquisition, Supervision, Writing – review & editing.
Funding
Authors thank the Research Infrastructure RECETOX RI (No.
LM2018121) and CETOCOEN EXCELLENCE (CZ.02.1.01/0.0/0.0/
17_043/0009632) for a supportive background. The work was supported
by the Operational Programme Research, Development and
Innovation – project Cetocoen Plus (CZ.02.1.01/0.0/0.0/15_003/
0000469) and the European Union’s Horizon 2020 research and innovation
programme under grant agreement No. 857560. This study has
received funding from the European Union’s Horizon 2020 research and
innovation programme under grant agreement No. 733032. We thank all
collaborating field workers, laboratory and administrative personnel,
and especially the cohort participants who invested their time and
provided samples and information for this study. This study reflects only
the authors’ view and the European Commission is not responsible for
any use that may be made of the information it contains.
Declaration of competing interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.envres.2022.113675.
References
Adibi, J.J., Whyatt, R.M., Williams, P.L., Calafat, A.M., Camann, D., Herrick, R.,
Nelson, H., Bhat, H.K., Perera, F.P., Silva, M.J., Hauser, R., 2008. Characterization of
phthalate exposure among pregnant women assessed by repeat air and urine
samples. Environ. Health Perspect. 116, 467–473. https://doi.org/10.1289/
ehp.10749.
Apel, P., Kortenkamp, A., Koch, H.M., Vogel, N., Rüther, M., Kasper-Sonnenberg, M.,
Conrad, A., Brüning, T., Kolossa-Gehring, M., 2020. Time course of phthalate
cumulative risks to male developmental health over a 27-year period: biomonitoring
samples of the German Environmental Specimen Bank. Environ. Int. 137, 105467
https://doi.org/10.1016/j.envint.2020.105467.
Bek¨o, G., Weschler, C.J., Langer, S., Callesen, M., Toftum, J., Clausen, G., 2013.
Children’s phthalate intakes and resultant cumulative exposures estimated from
urine compared with estimates from dust ingestion, inhalation and dermal
absorption in their homes and daycare centers. PLoS One 8. https://doi.org/
10.1371/journal.pone.0062442.
Bui, T.T., Giovanoulis, G., Cousins, A.P., Magn´er, J., Cousins, I.T., de Wit, C.A., 2016.
Human exposure, hazard and risk of alternative plasticizers to phthalate esters. Sci.
Total Environ. 541, 451–467. https://doi.org/10.1016/j.scitotenv.2015.09.036.
Calafat, A.M., Koch, H.M., Swan, S.H., Hauser, R., Goldman, L.R., Lanphear, B.P.,
Longnecker, M.P., Rudel, R.A., Teitelbaum, S.L., Whyatt, R.M., Wolff, M.S., 2013.
Misuse of blood serum to assess exposure to bisphenol A and phthalates. Breast
Cancer Res. 15, 403. https://doi.org/10.1186/bcr3494.
Carrieri, M., Trevisan, A., Bartolucci, G.B., 2000. Adjustment to concentration-dilution of
spot urine samples: correlation between specific gravity and creatinine. Int. Arch.
Occup. Environ. Health 74, 63–67. https://doi.org/10.1007/s004200000190.
CDC, 2019. Fourth national report on human exposure to environmental chemicals.
Updated Tables, january 2019, volume one. Fourth Natl. Rep. Hum. Expo. to
Environ. Chem. 1–529.
Correia-S´a, L., Kasper-Sonnenberg, M., P¨almke, C., Schütze, A., Norberto, S., Calhau, C.,
Domingues, V.F., Koch, H.M., 2018. Obesity or diet? Levels and determinants of
phthalate body burden - a case study on Portuguese children. Int. J. Hyg Environ.
Health 221, 519–530. https://doi.org/10.1016/j.ijheh.2018.02.001.
Correia-S´a, L., Schütze, A., Norberto, S., Calhau, C., Domingues, V.F., Koch, H.M., 2017.
Exposure of Portuguese children to the novel non-phthalate plasticizer di-(isononyl)-cyclohexane-1,2-dicarboxylate
(DINCH). Environ. Int. 102, 79–86. https://
doi.org/10.1016/j.envint.2017.02.001.
Den Hond, E., Govarts, E., Willems, H., Smolders, R., Casteleyn, L., Kolossa-Gehring, M.,
Schwedler, G., Seiwert, M., Fiddicke, U., Casta˜no, A., Esteban, M., Angerer, J.,
Koch, H.M., Schindler, B.K., Sepai, O., Exley, K., Bloemen, L., Horvat, M.,
Knudsen, L.E., Joas, A., Joas, R., Biot, P., Aerts, D., Koppen, G., Katsonouri, A.,
Hadjipanayis, A., Krskova, A., Maly, M., Mørck, T.A., Rudnai, P., Kozepesy, S.,
Mulcahy, M., Mannion, R., Gutleb, A.C., Fischer, M.E., Ligocka, D., Jakubowski, M.,
Reis, M.F., Namorado, S., Gurzau, A.E., Lupsa, I.-R., Halzlova, K., Jajcaj, M.,
Mazej, D., Tratnik, J.S., L´opez, A., Lopez, E., Berglund, M., Larsson, K., Lehmann, A.,
Crettaz, P., Schoeters, G., 2015. First steps toward harmonized human biomonitoring
in Europe: demonstration project to perform human biomonitoring on a European
scale. Environ. Health Perspect. 123, 255–263. https://doi.org/10.1289/
ehp.1408616.
Domínguez-Romero, E., Scheringer, M., 2019. A review of phthalate pharmacokinetics in
human and rat: what factors drive phthalate distribution and partitioning? Drug
Metab. Rev. 51, 314–329. https://doi.org/10.1080/03602532.2019.1620762.
Elbers, I., Mol, H., 2019a. ICI/EQUAS REPORT; Phthalates/round_02 (2018)- Phthalate
Biomarkers in Urine.
Elbers, I., Mol, H., 2019b. ICI/EQUAS REPORT; DINCH/round_02 (2018)- DINCH
Biomarkers in Urine.
Esteban L´opez, M., G¨oen, T., Mol, H., Nübler, S., Haji-Abbas-Zarrabi, K., Koch, H.M.,
Kasper-Sonnenberg, M., Dvorakova, D., Hajslova, J., Antignac, J.-P.P., Vaccher, V.,
Elbers, I., Thomsen, C., Vorkamp, K., Pedraza – Díaz, S., Kolossa-Gehring, M.,
Casta˜no, A., 2021. The European human biomonitoring platform - design and
implementation of a laboratory quality assurance/quality control (QA/QC)
programme for selected priority chemicals. Int. J. Hyg Environ. Health 234, 113740.
https://doi.org/10.1016/j.ijheh.2021.113740.
Frederiksen, H., Jorgensen, N., Andersson, A.-M., 2010. Correlations between phthalate
metabolites in urine, serum, and seminal plasma from young Danish men determined
by Isotope dilution liquid chromatography tandem mass spectrometry. J. Anal.
Toxicol. 34, 400–410. https://doi.org/10.1093/jat/34.7.400.
Frederiksen, H., Nielsen, O., Koch, H.M., Skakkebaek, N.E., Juul, A., Jørgensen, N.,
Andersson, A.-M., 2020. Changes in urinary excretion of phthalates, phthalate
substitutes, bisphenols and other polychlorinated and phenolic substances in young
Danish men; 2009–2017. Int. J. Hyg Environ. Health 223, 93–105. https://doi.org/
10.1016/j.ijheh.2019.10.002.
Fr´ery, N., Santonen, T., Porras, S.P., Fucic, A., Leso, V., Bousoumah, R., Duca, R.C., El
Yamani, M., Kolossa-Gehring, M., Ndaw, S., Viegas, S., Iavicoli, I., 2020.
Biomonitoring of occupational exposure to phthalates: a systematic review. Int. J.
Hyg Environ. Health 229, 113548. https://doi.org/10.1016/j.ijheh.2020.113548.
Giovanoulis, G., Bui, T., Xu, F., Papadopoulou, E., Padilla-S´anchez, J.A., Covaci, A.,
Haug, L.S., Cousins, A.P., Magn´er, J., Cousins, I.T., de Wit, C.A., 2018. Multipathway
human exposure assessment of phthalate esters and DINCH. Environ. Int.
112, 115–126. https://doi.org/10.1016/j.envint.2017.12.016.
Gries, W., Ellrich, D., Küpper, K., Ladermann, B., Leng, G., 2012. Analytical method for
the sensitive determination of major di-(2-propylheptyl)-phthalate metabolites in
human urine. J. Chromatogr. B 908, 128–136. https://doi.org/10.1016/j.
jchromb.2012.09.019.
Guo, Y., Kannan, K., 2013. A Survey of phthalates and parabens in personal care products
from the United States and its implications for human exposure. Environ. Sci.
Technol. Just Accep.
Gyllenhammar, I., Glynn, A., J¨onsson, B.A.G., Lindh, C.H., Darnerud, P.O., Svensson, K.,
Lignell, S., 2017. Diverging temporal trends of human exposure to bisphenols and
plastizisers, such as phthalates, caused by substitution of legacy EDCs? Environ. Res.
153, 48–54. https://doi.org/10.1016/j.envres.2016.11.012.
Health Canada, 2019. Fifth Report on Human Biomonitoring of Environmentla
Chemicals in Canada.
Heudorf, U., Mersch-Sundermann, V., Angerer, J., 2007. Phthalates: toxicology and
exposure. Int. J. Hyg Environ. Health 210, 623–634. https://doi.org/10.1016/j.
ijheh.2007.07.011.
P. Stuchlík Fiˇserov´a et al.
Environmental Research 213 (2022) 113675
10
Hlisníkov´a, H., Petroviˇcov´a, I., Kolena, B., ˇSidlovsk´a, M., Sirotkin, A., 2020. Effects and
mechanisms of phthalates’ action on reproductive processes and reproductive
health: a literature review. Int. J. Environ. Res. Publ. Health 17, 6811. https://doi.
org/10.3390/ijerph17186811.
Husøy, T., Andreassen, M., Hjertholm, H., Carlsen, M.H., Norberg, N., Sprong, C.,
Papadopoulou, E., Sakhi, A.K., Sabaredzovic, A., Dirven, H.A.A.M., 2019. The
Norwegian biomonitoring study from the EU project EuroMix: levels of phenols and
phthalates in 24-hour urine samples and exposure sources from food and personal
care products. Environ. Int. 132, 105103 https://doi.org/10.1016/j.
envint.2019.105103.
Janjua, N.R., Frederiksen, H., Skakkebæk, N.E., Wulf, H.C., Andersson, A.-M., 2008.
Urinary excretion of phthalates and paraben after repeated whole-body topical
application in humans. Int. J. Androl. 118–129. https://doi.org/10.1111/j.1365-
2605.2007.00841.x.
Kasper-Sonnenberg, M., Koch, H.M., Apel, P., Rüther, M., P¨almke, C., Brüning, T.,
Kolossa-Gehring, M., 2019. Time trend of exposure to the phthalate plasticizer
substitute DINCH in Germany from 1999 to 2017: biomonitoring data on young
adults from the Environmental Specimen Bank (ESB). Int. J. Hyg Environ. Health
222, 1084–1092. https://doi.org/10.1016/j.ijheh.2019.07.011.
Katsikantami, I., Sifakis, S., Tzatzarakis, M.N., Vakonaki, E., Kalantzi, O.-I., Tsatsakis, A.
M., Rizos, A.K., 2016. A global assessment of phthalates burden and related links to
health effects. Environ. Int. 97, 212–236. https://doi.org/10.1016/j.
envint.2016.09.013.
Koch, H.M., Angerer, J., 2012. Chapter 3A phthalates: biomarkers and human
biomonitoring. In: Biomarkers and Human Biomonitoring, vol. 1. The Royal Society
of Chemistry, pp. 179–233. https://doi.org/10.1039/9781849733373-00179.
Koch, H.M., Calafat, A.M., 2009. Human body burdens of chemicals used in plastic
manufacture. Philos. Trans. R. Soc. B Biol. Sci. 364, 2063–2078. https://doi.org/
10.1098/rstb.2008.0208.
Koch, H.M., Rüther, M., Schütze, A., Conrad, A., P¨almke, C., Apel, P., Brüning, T.,
Kolossa-Gehring, M., 2017. Phthalate metabolites in 24-h urine samples of the
German Environmental Specimen Bank (ESB) from 1988 to 2015 and a comparison
with US NHANES data from 1999 to 2012. Int. J. Hyg Environ. Health 220, 130–141.
https://doi.org/10.1016/j.ijheh.2016.11.003.
Koch, H.M., Schütze, A., P¨almke, C., Angerer, J., Brüning, T., 2013. Metabolism of the
plasticizer and phthalate substitute diisononyl-cyclohexane-1,2-dicarboxylate
(DINCH®) in humans after single oral doses. Arch. Toxicol. 87, 799–806. https://
doi.org/10.1007/s00204-012-0990-4.
Koppen, G., Govarts, E., Vanermen, G., Voorspoels, S., Govindan, M., Dewolf, M.-C., Den
Hond, E., Biot, P., Casteleyn, L., Kolossa-Gehring, M., Schwedler, G., Angerer, J.,
Koch, H.M., Schindler, B.K., Casta˜no, A., L´opez, M.E., Sepai, O., Exley, K.,
Bloemen, L., Knudsen, L.E., Joas, R., Joas, A., Schoeters, G., Covaci, A., 2019.
Mothers and children are related, even in exposure to chemicals present in common
consumer products. Environ. Res. 175, 297–307. https://doi.org/10.1016/j.
envres.2019.05.023.
Langer, S., Bek¨o, G., Weschler, C.J., Brive, L.M., Toftum, J., Callesen, M., Clausen, G.,
2014. Phthalate metabolites in urine samples from Danish children and correlations
with phthalates in dust samples from their homes and daycare centers. Int. J. Hyg
Environ. Health 217, 78–87. https://doi.org/10.1016/j.ijheh.2013.03.014.
Lemke, N., Murawski, A., Lange, R., Weber, T., Apel, P., Dębiak, M., Koch, H.M., KolossaGehring,
M., 2021. Substitutes mimic the exposure behaviour of REACH regulated
phthalates – a review of the German HBM system on the example of plasticizers. Int.
J. Hyg Environ. Health 236, 113780. https://doi.org/10.1016/j.ijheh.2021.113780.
Lessmann, F., Kolossa-Gehring, M., Apel, P., Rüther, M., P¨almke, C., Harth, V.,
Brüning, T., Koch, H.M., 2019. German Environmental Specimen Bank: 24-hour
urine samples from 1999 to 2017 reveal rapid increase in exposure to the paraphthalate
plasticizer di(2-ethylhexyl) terephthalate (DEHTP). Environ. Int. 132,
105102 https://doi.org/10.1016/j.envint.2019.105102.
Lorber, M., Weschler, C.J., Morrison, G., Bek¨o, G., Gong, M., Koch, H.M., Salthammer, T.,
Schripp, T., Toftum, J., Clausen, G., 2017. Linking a dermal permeation and an
inhalation model to a simple pharmacokinetic model to study airborne exposure to
di(n-butyl) phthalate. J. Expo. Sci. Environ. Epidemiol. 27, 601–609. https://doi.
org/10.1038/jes.2016.48.
Martínez, M.A., Rovira, J., Prasad Sharma, R., Nadal, M., Schuhmacher, M., Kumar, V.,
2018. Comparing dietary and non-dietary source contribution of BPA and DEHP to
prenatal exposure: a Catalonia (Spain) case study. Environ. Res. 166, 25–34. https://
doi.org/10.1016/j.envres.2018.05.008.
Mendelsohn, E., Hagopian, A., Hoffman, K., Butt, C.M., Lorenzo, A., Congleton, J.,
Webster, T.F., Stapleton, H.M., 2016. Nail polish as a source of exposure to triphenyl
phosphate. Environ. Int. 86, 45–51. https://doi.org/10.1016/j.envint.2015.10.005.
Miller, R.C., Brindle, E., Holman, D.J., Shofer, J., Klein, N.A., Soules, M.R., O’Connor, K.
A., 2004. Comparison of specific gravity and creatinine for normalizing urinary
reproductive hormone concentrations. Clin. Chem. 50, 924–932. https://doi.org/
10.1373/clinchem.2004.032292.
Nassan, F.L., Coull, B.A., Gaskins, A.J., Williams, M.A., Skakkebaek, N.E., Ford, J.B.,
Ye, X., Calafat, A.M., Braun, J.M., Hauser, R., 2017. Personal care product use in men
and urinary concentrations of select phthalate metabolites and parabens: results
from the environment and reproductive health (EARTH) study. Environ. Health
Perspect. 125, 1–10. https://doi.org/10.1289/EHP1374.
Pearson, M.A., Lu, C., Schmotzer, B.J., Waller, L.A., Riederer, A.M., 2009. Evaluation of
physiological measures for correcting variation in urinary output: implications for
assessing environmental chemical exposure in children. J. Expo. Sci. Environ.
Epidemiol. 19, 336–342. https://doi.org/10.1038/jes.2008.48.
Piler, P., Kandrnal, V., Kukla, L., Andrýskov´a, L., ˇSvancara, J., Jarkovský, J., Duˇsek, L.,
Pikhart, H., Bob´ak, M., Kl´anov´a, J., 2017. Cohort profile: the European longitudinal
study of pregnancy and childhood (ELSPAC) in the Czech republic. Int. J. Epidemiol.
46 https://doi.org/10.1093/ije/dyw091, 1379-1379F.
Pilka, T., Petrovicova, I., Kolena, B., Zatko, T., Trnovec, T., 2015. Relationship between
variation of seasonal temperature and extent of occupational exposure to phthalates.
Environ. Sci. Pollut. Res. Int. 22, 434–440. https://doi.org/10.1007/s11356-014-
3385-7.
Puklov´a, V., Janoˇs, T., Sochorov´a, L., Vavrouˇs, A., Vrbík, K., Fialov´a, A., Hanzlíkov´a, L.,
ˇCern´a, M., 2019. Exposure to mixed phthalates in Czech preschool and school
children. Arch. Environ. Contam. Toxicol. 77, 471–479. https://doi.org/10.1007/
s00244-019-00645-6.
REACH, 2006. REGULATION (EC) No 1907/2006 OF THE EUROPEAN PARLIAMENT
AND OF THE COUNCIL of 18 December 2006 [WWW Document].
Runkel, A.A., Snoj-Tratnik, J., Mazej, D., Horvat, M., 2020. Urinary phthalate
concentrations in the slovenian population: an attempt to exposure assessment of
family units. Environ. Res. 186, 109548 https://doi.org/10.1016/j.
envres.2020.109548.
Schneider, U., Schober, E.A., Streich, N.A., Breusch, S.J., 2002. Urinary creatinine
instability falsely increases the deoxypyridinoline/creatinine quotient. Clin. Chim.
Acta 324, 81–88. https://doi.org/10.1016/S0009-8981(02)00209-7.
Schütze, A., Otter, R., Modick, H., Langsch, A., Brüning, T., Koch, H.M., 2017. Additional
oxidized and alkyl chain breakdown metabolites of the plasticizer DINCH in urine
after oral dosage to human volunteers. Arch. Toxicol. 91, 179–188. https://doi.org/
10.1007/s00204-016-1688-9.
Schwedler, G., Conrad, A., Rucic, E., Koch, H.M., Leng, G., Schulz, C., SchmiedTobies,
M.I.H., Kolossa-Gehring, M., 2020a. Hexamoll® DINCH and DPHP
metabolites in urine of children and adolescents in Germany. Human biomonitoring
results of the German Environmental Survey GerES V, 2014–2017. Int. J. Hyg
Environ. Health 229, 113397. https://doi.org/10.1016/j.ijheh.2019.09.004.
Schwedler, G., Rucic, E., Lange, R., Conrad, A., Koch, H.M., P¨almke, C., Brüning, T.,
Schulz, C., Schmied-Tobies, M.I.H., Daniels, A., Kolossa-Gehring, M., 2020b.
Phthalate metabolites in urine of children and adolescents in Germany. Human
biomonitoring results of the German Environmental Survey GerES V, 2014–2017.
Int. J. Hyg Environ. Health 225, 113444. https://doi.org/10.1016/j.
ijheh.2019.113444.
Silano, V., Barat Baviera, J.M., Bolognesi, C., Chesson, A., Cocconcelli, P.S., Crebelli, R.,
Gott, D.M., Grob, K., Lampi, E., Mortensen, A., Rivi`ere, G., Steffensen, I.L.,
Tlustos, C., Van Loveren, H., Vernis, L., Zorn, H., Cravedi, J.P., Fortes, C., Tavares
Poças, M. de F., Waalkens-Berendsen, I., W¨olfle, D., Arcella, D., Cascio, C.,
Castoldi, A.F., Volk, K., Castle, L., 2019. Update of the risk assessment of dibutylphthalate
(DBP), butyl-benzyl-phthalate (BBP), bis(2-ethylhexyl)phthalate
(DEHP), di-isononylphthalate (DINP) and di-isodecylphthalate (DIDP) for use in food
contact materials. EFSA J. 17 https://doi.org/10.2903/j.efsa.2019.5838.
Silva, M.J., Barr, D.B., Reidy, J.A., Kato, K., Malek, N.A., Hodge, C.C., Hurtz, D.,
Calafat, A.M., Needham, L.L., Brock, J.W., 2003. Glucuronidation patterns of
common urinary and serum monoester phthalate metabolites. Arch. Toxicol. 77,
561–567. https://doi.org/10.1007/s00204-003-0486-3.
Søeborg, T., Frederiksen, H., Andersson, A.-M., 2012. Cumulative risk assessment of
phthalate exposure of Danish children and adolescents using the hazard index
approach. Int. J. Androl. 35, 245–252. https://doi.org/10.1111/j.1365-
2605.2011.01240.x.
Tait, S., Carli, F., Busani, L., Buzzigoli, E., Della Latta, V., Deodati, A., Fabbrizi, E.,
Gaggini, M., Maranghi, F., Tassinari, R., Toffol, G., Cianfarani, S., Gastaldelli, A., La
Rocca, C., 2020. Biomonitoring of Bis(2-ethylhexyl)phthalate (DEHP) in Italian
children and adolescents: data from LIFE PERSUADED project. Environ. Res. 185,
109428 https://doi.org/10.1016/j.envres.2020.109428.
The European parliament, The Council of the European Union, 2009. REGULATION (EC)
No 1223/2009 of the EUROPEAN PARLIAMENT and of the COUNCIL of 30
November 2009 on Cosmetic Products.
Vasiljevic, T., Su, K., Harner, T., 2021. A first look at atmospheric concentrations and
temporal trends of phthalates in distinct urban sectors of the Greater Toronto Area.
Atmos. Pollut. Res. 12, 173–182. https://doi.org/10.1016/j.apr.2020.10.019.
Wang, B., Tang, C., Wang, H., Zhou, W., Chen, Y., Zhou, Y., Jiang, Q., 2015. Influence of
body mass index status on urinary creatinine and specific gravity for epidemiological
study of children. Eur. J. Pediatr. 174, 1481–1489. https://doi.org/10.1007/s00431-
015-2558-9.
Wang, Y., Zhu, H., Kannan, K., 2019. A review of biomonitoring of phthalate exposures.
Toxics 7, 1–28. https://doi.org/10.3390/TOXICS7020021.
Who, 2013. Endocrine Disrupting Chemicals 2012. Summary for Decision-Makers.
https://doi.org/10.1002/9781118346747.ch1.
Wittassek, M., Koch, H.M., Angerer, J., Brüning, T., 2011. Assessing exposure to
phthalates - the human biomonitoring approach. Mol. Nutr. Food Res. 55, 7–31.
https://doi.org/10.1002/mnfr.201000121.
Wormuth, M., Scheringer, M., Vollenweider, M., Hungerbühler, K., 2006. What are the
sources of exposure to eight frequently used phthalic acid esters in Europeans? Risk
Anal. 26, 803–824. https://doi.org/10.1111/j.1539-6924.2006.00770.x.
Young, A.S., Allen, J.G., Kim, U.J., Seller, S., Webster, T.F., Kannan, K., Ceballos, D.M.,
2018. Phthalate and organophosphate plasticizers in nail polish: evaluation of labels
and ingredients. Environ. Sci. Technol. 52, 12841–12850. https://doi.org/10.1021/
acs.est.8b04495.
P. Stuchlík Fiˇserov´a et al.
Schyff, Veronica van der, Lenka Suchánková, Katerina Kademoglou, Lisa Melymuk, and Jana
Klánová. 2022. "Parabens and Antimicrobial Compounds in Conventional and 'Green'
Personal Care Products." Chemosphere 297: 1-8.
https://doi.org/10.1016/j.Chemosphere.2022.134019
Chemosphere 297 (2022) 134019
Available online 17 February 2022
0045-6535/© 2022 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Parabens and antimicrobial compounds in conventional and “green”
personal care products
Veronica van der Schyff, Lenka Such´ankov´a, Katerina Kademoglou 1,2
, Lisa Melymuk *
,
Jana Kl´anov´a
RECETOX, Faculty of Science, Masaryk University, Kotlarska 2, Brno, Czech Republic
H I G H L I G H T S G R A P H I C A L A B S T R A C T
• Parabens and antimicrobials were
detected in 83% of personal care
products.
• No difference in paraben levels between
green and conventional products.
• Methylparaben was typically present at
the highest concentrations.
• Estimated exclusive use of green cosmetics
lowered cumulative paraben
exposure.
A R T I C L E I N F O
Handling Editor: Myrto Petreas
Keywords:
Cosmetics
Human health
Greenwashing
Endocrine disruptors
Parabens
Triclosan
A B S T R A C T
The personal care product (PCP) industry is a worldwide multi-billion-dollar industry. Several synthetic compounds
like parabens and antimicrobial agents triclosan (TCS) and triclocarban (TCC) are ingredients in many
PCPs. Due to growing public awareness of potential risks associated with parabens and other synthetic compounds,
more PCPs are being marketed as “green,” “alternative,” or “natural.” We analyzed 19 green and 34
conventional PCP products obtained from a European store for seven parabens, TCC, and TCS. We found no
statistically significant difference in the concentrations between green and conventional products. Only four
products mentioned parabens in the list of ingredients; however, parabens were detected in 43 products, and at
μg/g levels in seven PCPs. Methylparaben was typically present at the highest concentration, and one mascara
exceeded the European legal concentration limit of methylparaben. Low concentrations of isopropyl-, isobutyl-,
and benzylparabens, which are banned in the EU, were detected in 70% of PCPs. The cumulative estimated daily
intake of parabens is an order of magnitude higher for people using only conventional products than those using
green products exclusively. We propose that legislation be developed with more explicit rules on when a product
can be advertised as “green” to aid consumers’ choices.
* Corresponding author. RECETOX, Faculty of Science, Masaryk University, Kotlarska 2, Brno, Czech Republic.
E-mail address: lisa.melymuk@recetox.muni.cz (L. Melymuk).
1
Current address: Department of Chemistry, Aristotle University of Thessaloniki, 541 24, Thessaloniki, Greece.
2
Current address: Biomic_AUTh, Center for Interdisciplinary Research and Innovation (CIRI-AUTH), Balkan Center, B1.4, Thessaloniki, 10th km ThessalonikiThermi
Rd, P.O. Box 8318, GR 57001.
Contents lists available at ScienceDirect
Chemosphere
journal homepage: www.elsevier.com/locate/chemosphere
https://doi.org/10.1016/j.chemosphere.2022.134019
Received 10 December 2021; Received in revised form 14 February 2022; Accepted 15 February 2022
Chemosphere 297 (2022) 134019
2
1. Introduction
Personal care products (PCPs) have been used since ancient times for
cosmetic and hygienic purposes. The global value of PCPs is estimated to
reach $716.6 billion by 2025 (Grand View Research, 2018). Consumer
awareness of potentially hazardous compounds in PCPs has increased
(Zollo et al., 2021). As a result, global market demand for products that
promote wellness, healthy aging, and environmental safety has also
increased (Ghazali et al., 2017). Certain PCPs are advertised as “green”
products, supposedly containing fewer and less harmful synthetic
compounds than conventional products. By using these products, consumers
are allegedly exposed to fewer harmful compounds (Harley
et al., 2016). The most effective way for manufacturers to communicate
the green status to potential consumers is by indicating it on the label.
Words like “natural,” “organic,” or “without artificial preservatives” are
used on product labels to persuade buyers to purchase the product.
However, these terms are largely unregulated, with few parameters and
no standardization (Dodson et al., 2012; Rubin and Brod, 2019). While
some producers are truthful, others have employed green labels as a
marketing strategy without significantly reducing the concentrations of
harmful compounds in their products (Jog and Singhal, 2019). This
strategy is known as “greenwashing” (Urba´nski & ul Haque, 2020). One
group of compounds that are often prioritized for elimination or
reduction in green products are parabens.
Parabens, esters of p-hydroxybenzoic acid, are low-cost, broadspectrum
antimicrobial, and antiseptic preservatives (Guo and Kannan,
2013; Li et al., 2021). They have been used widely in PCPs since the
1920s (Ana and Paula, 2016). Although parabens typically have short
half-lives and rapid excretion rates in mammalian systems (Janjua et al.,
2007), they have been quantified in human bodily fluids such urine,
breast milk, and plasma in ng/ml concentration levels (Honda et al.,
2018; Dualde et al., 2020; Sosvorova et al., 2017). The continuous
presence of parabens in various bodily fluids suggests that these compounds
are pseudo-persistent in the human body. Studies have found
that the most prevalent route of human exposure to parabens is dermal
exposure (Guo and Kannan, 2013; Liao et al., 2013). Parabens are
known endocrine disrupting compounds with androgenic and estrogenic
effects on humans (Darbre and Harvey, 2008). Long-chained parabens,
such as butylparaben (BuP) and propylparaben (PrP), have higher
endocrine disruptive potency than short-chained compounds such as
methylparaben (MeP) and ethylparaben (EtP) (Oishi, 2001). Parabens
with a branched structure, such as isobutyl- and isopropylparabens
(iso-BuP and iso-PrP), are the most estrogenic paraben congeners
(Darbre et al., 2002, 2003). Studies have linked topical application of
PCPs containing parabens with the onset of breast cancer, obesity,
gestational diabetes mellitus, and reduced sperm quality (Anderson,
1995; Darbre et al., 2004; Kolatorova et al., 2018; Li et al., 2019).
Therefore, the European Union (EU) limited short-chained parabens to
0.4% w/w of a total product and long-chained parabens to 0.19% w/w
in PCPs (Commission regulation (EU) no. 1004/2014). Isopropyl-, isobutyl-,
phenyl-, benzyl-, and pentylparabens have been banned from use
in PCPs in Europe since 2014 (Commission regulation (EU) no.
1004/2014)).
Triclosan (TCS) and triclocarban (TCC) are also popular additives to
PCP formulations, particularly to products that hold antimicrobial and
antibacterial properties (Nowak et al., 2021). TCC and TCS are known to
suppress thyroid hormone production in rats and are bioaccumulative
and persistent in the environment (Halden et al., 2017). More than 200
scientists and medical experts have signed the Florence Statement on
Triclosan and Triclocarban in 2016, stating that potential health and
environmental hazards caused by these compounds outweigh their
benefits thereof (Halden et al., 2017).
The endocrine disrupting properties of PCP compounds such as
parabens and TCS have been receiving increased attention. They are
among the compounds included in the target list of several projects of
the Horizon 2020 European Cluster to Improve Identification of
Endocrine Disruptors (EURION cluster: https://eurion-cluster.eu/).
Their possible associations with metabolic diseases (Kolatorova et al.,
2018; Han et al., 2021; Reimann et al., 2021), for instance, have
received attention in the OBERON project; a collaborative
inter-European research project examining the exposures and effects of
endocrine disrupting chemicals associated with metabolic disorders
(https://oberon-4eu.com/). To support the understanding of adverse
health effects of these compounds, current knowledge on human exposures
is needed, incorporating current levels of chemical use in products
as well as the consumer trends driving exposures.
The definition of a “cosmetic product” is made on a case-by-case
assessment, but all products used in this study are accepted as cosmetics
by Regulation (EC) no. 1223/2009 of the European Parliament
and of the Council (Regulation (EC) no. 1223/2009). However,
following the accepted scientific terminology, we will refer to the
products as “personal care products” throughout the manuscript.
This study compares the concentrations of parabens and antimicrobial
agents in PCP products marketed as a) green products and b) conventional
PCPs purchased from a popular European retail store. We
hypothesize that green products contain lower concentrations of parabens
and other antimicrobial compounds than conventional products.
We also present a determination of cumulative estimated daily intake
(EDI) of parabens that consumers may be exposed to by using multiple
green and conventional products.
2. Materials and methods
2.1. Sample collection
Fifty-three individual PCPs were purchased from a popular drug
store (part of a multinational chain) in Brno, Czech Republic, in August
2019 (Table 1). All items are commonly used types of PCPs according to
product use surveys (Wu et al., 2010). For the sake of anonymity, the
store and products are not mentioned by name or brand. Green-labeled
brands were identified by one of the specific marketing terms (Table 2)
printed on the product packaging. After purchasing, all products were
stored at room temperature until sample preparation. Samples were
opened directly before sample processing.
No permits or ethical requirements were necessary to collect or
analyze the samples.
2.2. Extraction and cleanup
Depending on the sample matrix, two different extraction methods
were used due to the inherent variability in the composition of different
PCPs. Liquid and semi-liquid products that were expected to contain
low-lipid content, namely nail polish, liquid deodorant, micellar water,
Table 1
Classification and number of PCPs used for analyses.
Product Type Green products Conventional products
Shampoo Rinse-off 3 3
Conditioner Rinse-off 2 2
Sunscreen Leave-on 1 3
Toothpaste Rinse-off 1 3
Shower gel Rinse-off 3 3
Hand soap Rinse-off 0 2
Hand cream Leave-on 1 2
Face cream Leave-on 0 2
Body lotion Leave-on 2 1
Lipstick Leave-on 1 3
Deodorant Leave-on 2 2
Face wash Rinse-off 1 2
Micellar water Rinse-off 0 1
Nail polish Leave-on 0 4
Mascara Leave-on 1 2
TOTAL 18 35
V. van der Schyff et al.
Chemosphere 297 (2022) 134019
3
and mascara, were analyzed using an adaptation of a sample preparation
method described by Young et al. (2018). The original extraction
method used by Young et al. (2018) consisted of liquid-liquid extraction
of phthalates and organophosphates with methanol and acetone:ethyl
acetate (1:1 v/v); we optimized it for our purposes to liquid-liquid
extraction only with methanol. Samples that were highly viscous,
solid, or semi-solid matrix and a medium to high-lipid content were
prepared for analyses using a modified quick, easy, cheap, effective,
rugged, and safe (QuEChERS) method. All samples and blanks were
spiked with 100 μl of internal standard (IS; Supplementary Table S1) mix
containing 13
C12- mass-labeled parabens (1 μg/ml) and with 100 μl of
mass-labeled triclosan-13
C12 (1 μg/ml in isooctane) prior to extraction.
A complete summary of both extraction methods is presented in the
Supplementary Material.
Before instrumental analysis, samples were diluted with methanol to
ensure the levels were within the calibration curve range. Low-lipid
samples were diluted at 1:2; medium-lipid and turbid samples were
diluted at 1:10, and high-lipid samples were diluted at 1:20. Although
nail polish and mascara were considered low-lipid products, some
samples exhibited turbidity until the end of the extraction and cleanup
phase and were also diluted at 1:20.
2.3. Instrumental analyses
Samples were quantified for seven parabens (MeP, EtP, PrP, iso-PrP,
BuP, iso-BuP, and BenzylP) and two antimicrobials (TCS and TCC).
Analysis was conducted on a liquid chromatograph (Agilent 1290 Infinity
II; Santa Clara, CA, USA) equipped with a vacuum degasser, binary
pump, autosampler and column thermostat coupled to a mass spectrometer
(ESI/QqQ Triple Quad 610; Santa Clara, CA, USA), and using
Mass Hunter software. A Phenomenex Synergi Fusion C-18 end-capped
(3 μm) 100 × 2.1 mm i.d. column was equipped with a Phenomenex
SecureGuard C18 guard column (Phenomenex, Torrance, CA, USA). The
mobile phase consisted of 5 mM ammonium acetate in water (A) and 5
mM ammonium acetate in methanol (B). The binary pump gradient
increased from 30% to 40% B at 3 min, 50% B at 10 min, and finally to
100% B at 13 min, holding to 35 min, with 1 min column equilibration at
initial conditions (30% B) with a flow rate of 0.12 mL/min 10 μL of the
individual sample was injected for the analyses. Compounds were
ionized with electrospray ionization. Ions were detected in the negative
mode, and the ionization parameters were as follows: gas temperature
280 ◦
C, gas flow 10 l/min, nebulizer 45 psi, capillary voltage 3.5 kV.
Masses were quantified according to the relative response of corresponding
internal standards.
2.4. Quality assurance and quality control
The analytical method was evaluated using replicates of six samples:
a methanol solvent as a non-matrix blank sample, and matrix-spiked
shampoo and face cream to evaluate the QuEChERS method, and a
methanol solvent, and matrix-spiked mascara and face cream to evaluate
the Young et al. (2018) extraction method. The spike-recovery
samples were fortified at two levels (spike low and spike high) with
native standards (Table S2) and analyzed as per the PCP samples. The
relative recoveries for the native standards in the non-matrix spiked
blanks were average 100% (range 92–115%) for parabens and 89%
(64–112%) for TCS and TCC (Supplemental Figure S1). The low concentration
matrix spiked sample recoveries were average 140% (range
83–406%) for parabens, and 126% (90–225%) for TCS and TCC, with
the high recoveries from the spiked mascara sample (Figure S2). The
high concentration matrix spiked sample recoveries were average 109%
(range 95–138%) for parabens and 98% (73–139%) for TCS and TCC
(Supplemental Figure S2).
Instrument calibration curves ranged from 0.1 ng/ml – 250 ng/ml
with R2
> 0.99 for all compounds. The researchers conducting laboratory
analyses did not use any PCPs on the days that analyses or extractions
were performed to limit external contamination and procedural
blanks, consisting of methanol solvents spiked with IS were used to track
contamination throughout the laboratory analyses. Two blanks were
prepared for each extraction method and analyzed in parallel with each
batch of samples. Ten blanks were processed with the QuEChERs
method and two blanks with the Young et al. (2018) method. The concentrations
of target analytes in blanks are presented in Table S3.
Procedural blanks were used to determine the method limit of
detection (LOD). The LOD was set to equal the average of the blanks +3
times the standard deviation of the blanks. If a compound was not
detected in any procedural blank, the LOD was set to equal the instrumental
detection limit (Table 3). Sample values above LOD were
adjusted to account for blank contamination by subtracting the average
of the procedural blanks.
2.5. Statistical analyses and grouping
Graphpad Prism 8.0.2 (www.graphpad.com) was used for summary
statistics. The products were grouped according to “green” or “conventional”
classification, based on the marketing on the label. According to
their mode of application, these groups were subdivided into “leave-on”
and “rinse-off” products. We used Mann-Whitney U-tests to compare
differences between the sample groups. Significance was set as p < 0.05.
For the statistical analysis, values < LOD were substituted with LOD/2.
2.6. Estimated daily intake and hazard quotient
To contextualize the concentrations of parabens that humans are
theoretically exposed to on a daily basis, the following formula of estimated
daily intake (EDI) as used by Nakata et al. (2015) was adapted to
determine estimated paraben exposure from PCPs in a European
country.
Σi
Ci⋅Ei⋅Ni⋅A⋅Fi
BW
Ci Concentration in product (μg/g);
Ei application quantities (g/time);
Ni application frequency (time per day applied);
A absorption factor through the skin;
Fi retention factor;
BW bodyweight.
Li et al. (2021) determined that the dermal absorption factor of
parabens is 0.4. The retention factor for leave-on products is 1, and for
rinse-off products, it is 0.01 (McGinty et al., 2011). The average adult
body mass in the Czech Republic is 91.9 kg for men and 74.2 kg for
women (WorldData, 2021). The application quantities of the selected
PCPs are taken from Bremmer et al. (2006), and the application frequency
of several PCPs is taken from questionnaire data from 300 Czech
adults as a part of the CELSPAC: Young Adults study (Supplementary
Figure S3; Fiˇserov´a et al. submitted). CELSPAC: Young Adults is an
on-going follow-up study of the longitudinal ELSPAC study in the Brno
region of the Czech Republic (Piler et al., 2017) and follows ELSPAC
children born in 1991 and 1992, their siblings, and spouses. The CELSPAC:
Young Adults study was approved by the ELSPAC Ethics Committee
(Ref. No: ELSPAC/EK/2/2019, dated 13.03.2019). This cohort’s
Table 2
Marketing terms on PCP labels that qualified the product as “green” for selection
purposes.
Inclusion criteria
Paraben free/free from parabens Natural cosmetic
Silicone free/without silicone Without artificial preservatives
Without artificial colorants Without artificial scents
Bio Certified natural cosmetics
100/90% natural origin Without phthalates
V. van der Schyff et al.
Chemosphere 297 (2022) 134019
4
Table 3
Concentrations (μg/g) of parabens, triclosan, and triclocarban in green and conventional personal care products from a European store.
LOD values are blank-based values, with the exception of BuP and iso-BuP, which are instrument detection limits. Where the quantified
concentrations were lower than LOD values, the cell is marked as “<LOD”, and the LOD/2 value used to calculate Ʃparabens is given in
the bottom row.
V. van der Schyff et al.
Chemosphere 297 (2022) 134019
5
minimum and maximum application frequency were used to estimate
two paraben exposure scenarios for both green and conventional
products.
3. Results and discussion
3.1. Difference between green and conventional PCPs and product labels
We tested 35 conventional and 18 green products for nine compounds
of interest. The concentrations of the target compounds in the
selected PCPs ranged several orders of magnitude — from <0.0001 μg/g
to >1000 μg/g (Table 3). Most of the quantifiable concentrations of
parabens were in the 0.01–0.1 μg/g range. TCS was present at five times
higher concentrations than TCC.
Parabens were detected in 89% of green products and 77% of conventional
products, while antimicrobial compounds had less frequent
detection (66% of green products and 69% of conventional products).
One green and eight conventional products did not contain parabens or
antimicrobial compounds above the LOD. Methylparaben was found at
the highest concentrations in PCPs (Table 3), with a maximum measured
concentration of 4522 μg/g MeP in mascara. Similar to what was found
in other studies (Dodson et al., 2012; Matwiejczuk et al., 2020), MeP had
the highest detection frequency (38 of 53 products; 72%), followed by
EtP (36 of 53 products; 67%). Detectable concentrations of EtP were
found in a green shower gel that was labeled as “free from parabens”.
It was unexpected that conventional rinse-off products had the
lowest detection frequency of parabens (Table 4); however this is primarily
because no parabens were present above detectable concentrations
in any of the four conventional toothpaste brands (Table 3).
In all product categories, for both parabens and TCS, concentrations
varied several orders of magnitude, even within the same specific group
of products, e.g., conditioners (Table 3). The measured concentrations
did not support our hypothesis that green products should have lower
levels of parabens, TCS and TCC; we found no significant difference
between the concentrations in green PCPs and the corresponding conventional
PCPs for any of the target compounds. The clearer differences
were not between green/conventional products but rather between the
leave-on and rinse-off products across green and conventional product
categories. The conventional leave-on products had significantly higher
median concentrations of Σ7parabens than conventional rinse-off
products (Mann-Whitney U test, p = 0.025, Fig. 1). Similarly, for TCS,
the conventional leave-on products had statistically significantly higher
TCS concentrations than the conventional rinse-off products (p =
0.0072, Fig. 1). However, due to the small sample size employed by this
study, the results should be interpreted with caution.
3.2. Legislation regarding product composition and labelling
Current legislation states that the maximum concentration of parabens
in PCP products is to be 0.8% of the ready for use preparation
(Commission regulation (EU) no. 1004/2014). A distinction is made
between short- and long-chained parabens because of the higher endocrine
disruptive potential of long-chained parabens (Oishi, 2001). One
conventional mascara sample exceeded these guidelines, where 0.452%
(4521 μg/g) of the ready for use preparation consisted of MeP. However,
waterproof mascara is a “borderline” or “extreme cosmetic” product.
Borderline products are regulated on a case-by-case basis according to
the Manual on the Scope of Application of the Cosmetic Regulation, which
can suggest whether or not a product falls within the scope of the
Regulation (EC) no. 1223/2009 (Lores et al., 2018). 80% of the extreme
cosmetic products analyzed by Lores et al. (2018) exceeded concentrations
of various compounds as prescribed by the Regulation (EC) no.
1223/2009.
Although the EU bans isopropyl-, isobutyl-, and benzylparabens
(Commission regulation (EU) no. 1004/2014), quantifiable concentrations
of these compounds were found in several products, including
green and conventional mascaras, and conventional face wash gel. The
highest concentration of a banned paraben was 81 μg/g isopropylparaben
in the green toothpaste sample, labeled as “herbal gel for
gums.” The maximum concentration of TCS in a PCP is 0.3% of the total
ready for use preparation and 1.5% for TCC (Regulation (EC) no.
1223/2009). No products had concentrations that were close to 0.3% or
1.5% of the product composition for either compound.
Even though parabens were quantified in 81% of the PCP samples in
this study, only four products included parabens as an ingredient on the
label: three conventional products (conditioner, face cream, and
mascara) and one green toothpaste. Most of the products that included
parabens in the ingredient list contained more than 400 μg/g of a
compound. MeP was included as an ingredient in all four products, PrP
in two, and EtP and BuP in one product each. One green conditioner
contained 525 μg/g MeP without including MeP in the ingredient list.
Neither TCC nor TCS was ever included in the ingredient lists, even in
antimicrobial soap. According to the European Union cosmetics regulation
(no. 1223/2009), compounds need not to be included in the
ingredient list if they are present as impurities in raw materials or a
subsidiary of technical materials used in the mixture but are not present
in the final product (Regulation (EC) no. 1223/2009). Because most of
the concentrations of parabens and other antimicrobial compounds were
below 0.1 μg/g, they might have been considered impurities by the
production companies or incidental ingredients from the processing and
packaging procedure. Cosmetic packaging is not as strictly regulated as
food packaging, which increases the chances of accidental inclusion of
harmful substances from packaging materials in personal care products
(Regulation (EC) no. 1272/2008).
3.3. Estimated daily intake and health risk of parabens through green and
conventional PCPs
The daily use of PCPs is not restricted to a single product. Multiple
PCPs are used daily to serve different cosmetic and hygienic purposes
(Fisher et al., 2017). The application frequency of PCPs also differs between
different groups of people. According to Czech questionnaire data
(see section 2.6; Figure S3), the response group with the highest application
frequency of PCPs consisted exclusively of female respondents.
The group with the lowest frequency was exclusively male; as a result,
we separate our exposure assessments by gender (Figure S3). EDIs were
calculated for male and female exposure to parabens using only the
green and conventional products sampled during this study. The cumulative
EDI was calculated to determine the cumulative effect of using
multiple products (see section 2.6; Table 5). No information on toothpaste,
conditioner, and shower gel was included in the questionnaire.
These products (marked with an asterisk in Table 5) were included in the
cumulative EDI calculation under the assumption that toothpaste and
shower gel is used once a day (Bremmer et al., 2006) and that conditioner
is used at the same application frequency as shampoo.
Even though only a snapshot of products were included in the study,
a clear pattern could be seen with exposures differing by gender and
selection of green vs. conventional products due to women’s higher use
of PCPs (Figure S3 and Table 5). The cumulative EDI of green products
was an order of magnitude lower than conventional products for both
men and women (Table 4). We also note that EDI is substantially higher
for women than men due to their higher use of PCPs and lower body
weight.
Table 4
The detection frequencies (df) of Ʃ7parabens and TCS in different groups of
PCPs. Different product groups are green rinse-off (GRO), green leave-on (GLO),
conventional rinse-off (CRO), and conventional leave-on (CLO).
GRO (n = 10) GLO (n = 8) CRO (n = 16) CLO (n = 19)
Ʃ7Parabens df 90% 88% 56% 95%
TCS df 30% 88% 38% 89%
V. van der Schyff et al.
Chemosphere 297 (2022) 134019
6
According to the theoretical calculation, a reduction of 82% in paraben
EDI can be expected if only green products are used in lieu of
conventional products. This contrasts with the statistical comparisons
between green and conventional products (Section 3.1), where there
were no statistically significant differences in paraben concentrations
between the product groups. It suggests that even if the concentrations
of parabens in some individual green products do not significantly differ
from conventional products, the cumulative effect of reduced paraben
exposure may be significant if a consumer commits to a lifestyle of using
green products exclusively.
Even though the cumulative EDIs for this study were found to be
relatively low and no health risks were expected, it should be noted that
the cumulative EDI only accounted for seven products. Typically, people
use more PCPs on a regular basis, and will have other exposure sources
such as diet and pharmaceuticals (Błędzka et al., 2014; Liao et al., 2013).
Fig. 1. Box-and-whisker plot (horizontal lines are medians, 95% confidence intervals, minima, and maxima) of the sum of parabens detected in all product groups.
Different product groups are green rinse-off (GRO), green leave-on (GLO), conventional rinse-off (CRO), and conventional leave-on (CLO). The p-values of significant
differences between product groups are indicated.
Table 5
Estimated daily intake (EDI) of parabens through conventional personal care products for men and women in the Czech Republic. An asterisk (*) indicates products
that are included in the EDI calculation, but not included in the questionnaire of Figure S3.
Product Ʃparaben median Application quantitya
Application frequencyb
EDI
(μg/g) (g/application) (application/day) (μg/kg-bw/day)
Green products
Female use
Shampoo 0.012 12 0.7 0.000005
Conditioner* 263 14 0.7 0.14
Lotion 0.25 10.5 0.9 0.013
Deodorant 0.17 0.5 1 0.00046
Mascara 0.12 0.025 0.8 0.00001
Toothpaste* 679 0.08 1 0.03
Shower gel* 0.014 8.7 1 0.00007
Cumulative EDI 0.18
Conventional products
Female use
Shampoo 0.19 12 0.7 0.00009
Conditioner* 1335 14 0.7 0.71
Lotion 0.27 10.5 0.9 0.0014
Deodorant 0.3 0.5 1 0.0008
Mascara 2994 0.025 0.8 0.32
Toothpaste* 0.0089 0.08 1 0.00000004
Shower gel* 0.022 8.7 1 0.00001
Cumulative EDI 1.03
Green products
Male use
Shampoo 0.012 12 0.3 0.000002
Conditioner* 263 14 0.3 0.048
Lotion 0.25 10.5 0.1 0.0011
Deodorant 0.17 0.5 0.5 0.00018
Mascara 0.12 0.025 0 0
Toothpaste* 679 0.08 1 0.0024
Shower gel* 0.014 8.7 1 0.000005
Cumulative EDI 0.052
Conventional products
Male use
Shampoo 0.19 12 0.3 0.00003
Conditioner* 1335 14 0.3 0.24
Lotion 0.27 10.5 0.1 0.00013
Deodorant 0.3 0.5 0.5 0.00033
Mascara 2994 0.025 0 0
Toothpaste* 0.0089 0.08 1 0.00000003
Shower gel* 0.022 8.7 1 0.000008
Cumulative EDI 0.24
The cumulative EDI values were determined by summing the individual EDIs from each category. The EDIs of parabens from the individual PCPs are very low (<1 μg/
kg-bw/day). The highest EDI was 0.71 μg/kg-bw/day for conventional conditioners used by women.
V. van der Schyff et al.
Chemosphere 297 (2022) 134019
7
3.4. Consumer attitude to green products
Due to increased public awareness of certain chemicals compounds’
potential health and environmental effects, many consumers gravitate
towards alternative or green products (Dodson et al., 2020). In 2019, the
sale of natural cosmetics brands accounted for $ 1.6 billion in sales in the
United States (NPD Group, 2018). Hence, many companies produce and
promote green cosmetic products to retain a competitive advantage in
the market (Luo et al., 2020).
The practice of paraben avoidance is proven to be implementable
and effective. A study by Dodson et al. (2020) revealed that cosmetic
users are willing to avoid potentially hazardous compounds if they have
prior knowledge. The participants that actively avoided products with
harmful compounds by studying product ingredient lists had lower
urinary concentrations of parabens than people who did not practice
avoidance (Dodson et al., 2020). Similarly, a study by Harley et al.
(2016) revealed that MeP and PrP in the urine of adolescent girls
decreased by 43.9% after using products labeled as “paraben-free” for
three days (researchers screened these products beforehand to confirm
that they were paraben-free).
However, since most consumers are not familiar with technical terms
or chemical nomenclature, ingredient lists are not ideal communicators,
even if the producers are transparent (Mar´c and Martyn, 2019). Many
consumers rely on social media to choose PCPs, leading to skewed
perspectives (Luo et al., 2020). Consumers also fear “greenwashed”
products, where producers intentionally label a product as green, but no
effort has been put into making the product green (Urba´nski and ul
Haque, 2020). The greatest danger of green products is the fact that the
term “green,” “bio-," “natural,” and other similar labels are haphazardly
communicated and currently unregulated, with no standards in place to
determine ingredient content or concentrations of chemical compounds
(Dodson et al., 2012; Rubin and Brod, 2019). The label “paraben-free” is
not officially registered in the EU, and producers are under no legal
obligation to remove all parabens from the products (Nowak et al.,
2021).
4. Conclusion
The PCP industry is a big international business. Many synthetic
compounds are included in the production of PCPs, including potential
endocrine disrupting compounds such as parabens, TCS, and TCC. Due
to growing public awareness of potential health risks of parabens and
other synthetic compounds, more PCPs are being produced and marketed
as “green,” “alternative,” or “natural.” However, due to vague
regulations regarding green products, it is difficult to determine whether
products marketed as green contain lower concentrations of harmful
chemical. After analyzing various green and conventional PCP products
from a European store for seven parabens, TCC, and TCS, we found that
there was no significant difference in the paraben concentrations between
the product types, thus rejecting our initial proposed hypothesis.
Methylparaben was the compound present at the highest concentration
and detection frequency among all target analytes. Only four products
included parabens in the list of ingredients. Except for one mascara,
which can be considered a borderline or extreme cosmetic product, the
concentrations of all compounds were below the maximum concentrations
set by the EU Regulations. It was concerning that four banned
paraben compounds were quantifiable in the majority of the PCP samples.
The cumulative EDI for parabens was an order of magnitude higher
for conventional products than for green PCPs. Even though the EDIs for
individual products were very low, the cumulative effect indicates that
consumers could be exposed to fewer parabens using predominantly
green PCPs in their daily routine. The potential health risks of cumulated
PCP use warrant further investigation. The lack of set parameters on
product labels limits the ability of consumers to make informed decisions
on product purchases and may pose a risk for the consumer”.
Credit author statement
Veronica van der Schyff: drafting the manuscript, data interpretation;
Lenka Such´ankov´a: Acquisition of data, data interpretation,
manuscript revision; Katerina Kademoglou: Acquisition of data, data
interpretation, manuscript revision, study design; Lisa Melymuk: Study
design, data interpretation, manuscript revision, Supervision; Jana
Kl´anov´a: Supervision, Funding acquisition, manuscript revision.
Declaration of competing interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
Acknowledgements
The authors acknowledge the support of the Research Infrastructure
RECETOX RI (No LM2018121) financed by the Ministry of Education,
Youth and Sports of the Czech Republic (MEYS), and the European
Union’s Horizon 2020 Research and Innovation programme under Grant
Agreement No. 857560 and MEYS (Czech Operational Programme
Research, Development and Education (OP RDE) of European Structural
and Investment Funds No CZ.02.1.01/0.0/0.0/17_043/0009632). This
work has received funding from the Horizon 2020 Research and Innovation
programme under Grant Agreement No 825712, and OP RDE
project Postdoc@MUNI (CZ.02.2.69/0.0/0.0/16_027/0008360). This
publication reflects only the author’s view and the European Commission
is not responsible for any use that may be made of the information it
contains.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.chemosphere.2022.134019.
References
Ana, F., Paula, F.A., 2016. Parabens paradoxes in cosmetic formulations: a review. Int. J.
Med. Res. Pharmaceut. Sci. 8, 1–11.
Anderson, F.A., 1995. Final report on the safety assessment of isobutylparaben and
isopropylparaben. J. Am. Coll. Toxicol. 14, 364–372.
Błędzka, D., Gromadzi´nska, J., Wąsowicz, W., 2014. Parabens. From environmental
studies to human health. Environ. Int. 67, 27–42.
Bremmer, H.J., Prud’Homme de Lodder, L.C.H., Van Engelen, J.G.M., 2006. Cosmetics
Fact Sheet. To Assess the Risks for the Consumer. Updated version for ConsExpo 4.
RIVM rapport 320104001. https://www.rivm.nl/bibliotheek/rapporten/320104001.
pdf. (Accessed 16 April 2021). Date of access.
Commission Regulation (EU) no. 1004/2014 of 18 September 2014 amending Annex V
to Regulation (EC) No 1223/2009 of the European Parliament and of the Council on
cosmetic products. Off. J. Eur. Union 282/5. https://eur-lex.europa.eu/legal-content
/EN/TXT/PDF/?uri=CELEX:32014R1004&from=EN. Date accessed: 28 April 2021.
Darbre, P.D., Harvey, P.W., 2008. Paraben esters: review of recent studies of endocrine
toxicity, absorption, esterase and human exposure, and discussion of potential
human health risks. J. Appl. Toxicol. 28, 561–578.
Darbre, P.D., Aljarrah, A., Miller, W.R., Coldham, N.G., Sauer, M.J., Pope, G.S., 2004.
Concentrations of parabens in human breast tumours. J. Appl. Toxicol.: Int. J. 24,
5–13.
Darbre, P.D., Byford, J.R., Shaw, L.E., Hall, S., Coldham, N.G., Pope, G.S., Sauer, M.J.,
2003. Oestrogenic activity of benzylparaben. J. Appl. Toxicol.: Int. J. 23, 43–51.
Darbre, P.D., Byford, J.R., Shaw, L.E., Horton, R.A., Pope, G.S., Sauer, M.J., 2002.
Oestrogenic activity of isobutylparaben in vitro and in vivo. J. Appl. Toxicol.: Int. J.
22, 219–226.
Dodson, R.E., Boronow, K.E., Susmann, H., Udesky, J.O., Rodgers, K.M., Weller, D.,
Woudneh, M., Brody, J.G., Rudel, R.A., 2020. Consumer behavior and exposure to
parabens, bisphenols, triclosan, dichlorophenols, and benzophenone-3: results from
a crowdsourced biomonitoring study. Int. J. Hyg Environ. Health 230, 113624.
Dodson, R.E., Nishioka, M., Standley, L.J., Perovich, L.J., Brody, J.G., Rudel, R.A., 2012.
Endocrine disruptors and asthma-associated chemicals in consumer products.
Environ. Health Perspect. 120, 935–943.
Dualde, P., Pardo, O., Corpas-Burgos, F., Kuligowski, J., Gormaz, M., Vento, M.,
Pastor, A., Yus`a, V., 2020. Biomonitoring of parabens in human milk and estimated
daily intake for breastfed infants. Chemosphere 240, 124829.
V. van der Schyff et al.
Chemosphere 297 (2022) 134019
8
Fiˇserov´a, P.S., Melymuk, L., Komprdov´a, K., Domínguez-Romero, E., Scheringer, M.,
Kohoutek, J., Pˇribylov´a, P., Andrýskov´a, L., Piler P. & Kl´anov´a, J. PCP Usage and
Lifestyle Affects Phthalate and DINCH Metabolites Levels in Teenagers and Young
Adults” Submitted to Environmental Research.
Fisher, M., MacPherson, S., Braun, J.M., Hauser, R., Walker, M., Feeley, M., Mallick, R.,
B´erub´e, R., Arbuckle, T.E., 2017. Paraben concentrations in maternal urine and
breast milk and its association with personal care product use. Environ. Sci. Technol.
51, 4009–4017.
Ghazali, E., Soon, P.C., Mutum, D.S., Nguyen, B., 2017. Health and cosmetics:
investigating consumers’ values for buying organic personal care products.
J. Retailing Consum. Serv. 39, 154–163.
Grand View Research, 2018. Beauty & Personal Care Products Market Worth $716.6
Billion by 2025. Date of access. https://www.grandviewresearch.com/press-release
/global-beauty-personal-care-products-market. (Accessed 15 March 2021).
Guo, Y., Kannan, K., 2013. A survey of phthalates and parabens in personal care products
from the United States and its implication for human exposure. Environ. Sci.
Technol. 47, 14442–14449.
Halden, R.U., Lindeman, A.E., Aiello, A.E., Andrews, D., Arnold, W.A., Fair, P., Fuoco, R.
E., Geer, L.A., Johnson, P.I., Lohmann, R., McNeill, K., 2017. The Florence statement
on triclosan and triclocarban. Environ. Health Perspect. 125, 064501.
Han, M., Wang, Y., Tang, C., Fang, H., Yang, D., Wu, J., Wang, H., Chen, Y., Jiang, Q.,
2021. Association of triclosan and triclocarban in urine with obesity risk in Chinese
school children. Environ. Int. 157, 106846.
Harley, K.G., Kogut, K., Madrigal, D.S., Cardenas, M., Vera, I.A., Meza-Alfaro, G., She, J.,
Gavin, Q., Zahedi, R., Bradman, A., Eskenazi, B., 2016. Reducing phthalate, paraben,
and phenol exposure from personal care products in adolescent girls: findings from
the HERMOSA intervention study. Environ. Health Perspect. 124, 1600–1607.
Honda, M., Robinson, M., Kannan, K., 2018. Parabens in human urine from several Asian
countries, Greece, and the United States. Chemosphere 201, 13–19.
Janjua, N.R., Mortensen, G.K., Andersson, A.M., Kongshoj, B., Skakkebaek, N.E., Wulf, H.
C., 2007. Systemic uptake of diethyl phthalate, dibutyl phthalate, and butyl paraben
following whole-body topical application and reproductive and thyroid hormone
levels in humans. Environ. Sci. Technol. 41, 5564–5570.
Jog, D., Singhal, D., 2019. Pseudo green players and their greenwashing practices: a
differentiating strategy for real green firms of personal care category. Strat. Dir. 35,
4–7.
Kolatorova, L., Sramkova, M., Vitku, J., Vcelak, J., Lischkova, O., Starka, L., Duskova, M.,
2018. Parabens and their relation to obesity. Physiol. Res. 67, S465–S472.
Li, C., Zhao, Y., Liu, S., Yang, D., Ma, H., Zhu, Z., Kang, L., Lu, S., 2021. Exposure of
Chinese adult females to parabens from personal care products: estimation of intake
via dermal contact and health risks. Environ. Pollut. 272, 116043.
Li, Y., Xu, S., Li, Y., Zhang, B., Huo, W., Zhu, Y., Wan, Y., Zheng, T., Zhou, A., Chen, Z.,
Huang, Z., 2019. Association between urinary parabens and gestational diabetes
mellitus across prepregnancy body mass index categories. Environ. Res. 170,
151–159.
Liao, C., Liu, F., Kannan, K., 2013. Occurrence of and dietary exposure to parabens in
foodstuffs from the United States. Environ. Sci. Technol. 47, 3918–3925.
Lores, M., Celeiro, M., Rubio, L., Llompart, M., Garcia-Jares, C., 2018. Extreme cosmetics
and borderline products: an analytical-based survey of European regulation
compliance. Anal. Bioanal. Chem. 410, 7085–7102.
Luo, B., Sun, Y., Shen, J., Xia, L., 2020. How does green advertising skepticism on social
media affect consumer intention to purchase green products? J. Consum. Behav. 19,
371–381.
Mar´c, M.M., Martyn, J., 2019. Factors determining the level of knowledge about
parabens in cosmetics. Eur. J. Clin. Exper. Med. 17, 236–241.
Matwiejczuk, N., Galicka, A., Brz´oska, M.M., 2020. Review of the safety of application of
cosmetic products containing parabens. J. Appl. Toxicol. 40, 176–210.
McGinty, D., Letizia, C.S., Api, A.M., 2011. Fragrance material review on ethylene
brassylate. Food Chem. Toxicol. 49, S174–S182.
Nakata, H., Hinosaka, M., Yanagimoto, H., 2015. Macrocyclic-, polycyclic-, and nitro
musks in cosmetics, household commodities and indoor dusts collected from Japan:
implications for their human exposure. Ecotoxicol. Environ. Saf. 111, 248–255.
National Purchase Diary Panel (NPD Group), 2018. U.S. Prestige Beauty Industry Sales
Grow 6 Percent in 2018, Reports the NPD Group. Date of access. https://www.npd.
com/wps/portal/npd/us/news/press-releases/2019/u-s–prestige-beauty-industry-s
ales-grow-6-percent-in-2018–reports-the-npd-group/. (Accessed 15 April 2021).
Nowak, K., Jabło´nska, E., Ratajczak-Wrona, W., 2021. Controversy around parabens:
alternative strategies for preservative use in cosmetics and personal care products.
Environ. Res. 198, 110488.
Oishi, S., 2001. Effects of butylparaben on the male reproductive system in rats. Toxicol.
Ind. Health 17, 31–39.
Piler, P., Kandrnal, V., Kukla, L., Andrýskov´a, L., ˇSvancara, J., Jarkovský, J., Duˇsek, L.,
Pikhart, H., Bob´ak, M., Kl´anov´a, J., 2017. Cohort profile: the european longitudinal
study of pregnancy and childhood (ELSPAC) in the Czech Republic. Int. J. Epidemiol.
46, 1379, 1379f.
Regulation (EC) no. 1223/2009 of the European parliament and of the council of 30
November 2009 on cosmetic products. https://eur-lex.europa.eu/legal-content/EN/
TXT/HTML/?uri=CELEX:32009R1223&from=EN. Date of access: 1 July 2021.
Regulation (EC) no. 1272/2008 — classification, labelling and packaging of substances
and mixtures (CLP). https://osha.europa.eu/en/legislation/directives/regulation-ec
-no-1272-2008-classification-labelling-and-packaging-of-substances-and-mixtures.
Date of access: 22 July 2021.
Reimann, B., Vrijens, K., Roels, H.A., Wang, C., Cosemans, C., Van Overmeire, I.,
Nawrot, T.S., Plusquin, M., 2021. In utero exposure to parabens and early childhood
BMI z-scores – associations between placental ethyl paraben, longitudinal BMI
trajectories and cord blood metabolic biomarkers. Environ. Int. 157, 106845.
Rubin, C.B., Brod, B., 2019. Natural does not mean safe—the dirt on clean beauty
products. JAMA Dermatol. 155, 1344–1345.
Sosvorova, L.K., Chlupacova, T., Vitku, J., Vlk, M., Heracek, J., Starka, L., Saman, D.,
Simkova, M., Hampl, R., 2017. Determination of selected bisphenols, parabens and
estrogens in human plasma using LC-MS/MS. Talanta 174, 21–28.
Urba´nski, M., ul Haque, A., 2020. Are you environmentally conscious enough to
differentiate between greenwashed and sustainable items? A global consumers
perspective. Sustainability 12, 1786.
WorldData, 2021. Average Sizes of Men and Women. Date of access. https://www.
worlddata.info/average-bodyheight.php#by-area. (Accessed 10 May 2021).
Wu, X., Bennett, D.H., Ritz, B., Cassady, D.L., Lee, K., Hertz-Picciotto, I., 2010. Usage
pattern of personal care products in California households. Food Chem. Toxicol. 48,
3109–3119.
Young, A.S., Allen, J.G., Kim, U.J., Seller, S., Webster, T.F., Kannan, K., Ceballos, D.M.,
2018. Phthalate and organophosphate plasticizers in nail polish: evaluation of labels
and ingredients. Environ. Sci. Technol. 52, 12841–12850.
Zollo, L., Carranza, R., Faraoni, M., Díaz, E., Martín-Consuegra, D., 2021. What
influences consumers’ intention to purchase organic personal care products? The
role of social reassurance. J. Retailing Consum. Serv. 60, 102432.
V. van der Schyff et al.
Demirtepe, Hale, Lisa Melymuk, Miriam L. Diamond, Lola Bajard, Šimon Vojta, Roman
Prokeš, Ondřej Sáňka, Jana Klánová, Ľubica Palkovičová Murínová, Denisa Richterová,
Vladimíra Rašplová, Tomáš Trnovec. 2019. "Linking Past Uses of Legacy SVOCs with Today's
Indoor Levels and Human Exposure." Environment International 127: 653-63.
https://doi.org/10.1016/j.envint.2019.04.001
Contents lists available at ScienceDirect
Environment International
journal homepage: www.elsevier.com/locate/envint
Linking past uses of legacy SVOCs with today's indoor levels and human
exposure
Hale Demirtepea
, Lisa Melymuka,⁎
, Miriam L. Diamondb
, Lola Bajarda
, Šimon Vojtaa,1
,
Roman Prokeša
, Ondřej Sáňkaa
, Jana Klánováa
, Ľubica Palkovičová Murínovác
,
Denisa Richterovác
, Vladimíra Rašplovác
, Tomáš Trnovecc
a
RECETOX, Masaryk University, Kamenice 753/5, pavilion A29, 625 00 Brno, Czech Republic
b
Department of Earth Sciences, and Dalla Lana School of Public Health, University of Toronto, Toronto, Ontario, Canada
c
Department of Environmental Medicine, Faculty of Public Health, Slovak Medical University, Limbová 12, 83303 Bratislava, Slovakia
A R T I C L E I N F O
Handling Editor: Heather Stapleton
Keywords:
Human intake
Indoor environment
Flame retardants
Legacy POPs
PAHs
Risk prioritization
A B S T R A C T
Semivolatile organic compounds (SVOCs) emitted from consumer products, building materials, and indoor and
outdoor activities can be highly persistent in indoor environments. Human exposure to and environmental
contamination with polychlorinated biphenyls (PCBs) was previously reported in a region near a former PCB
production facility in Slovakia. However, we found that the indoor residential PCB levels did not correlate with
the distance from the facility. Rather, indoor levels in this region and those reported in the literature were
related to the historic PCB use on a national scale and the inferred presence of primary sources of PCBs in the
homes. Other SVOCs had levels linked with either the activities in the home, e.g., polycyclic aromatic hydrocarbons
(PAHs) with wood heating; or outdoor activities, e.g., organochlorine pesticides (OCPs) with agricultural
land use and building age. We propose a classification framework to prioritize SVOCs for monitoring in
indoor environments and to evaluate risks from indoor SVOC exposures. Application of this framework to 88
measured SVOCs identified several PCB congeners (CB-11, -28, -52), hexachlorobenzene (HCB), benzo(a)pyrene,
and γ-HCH as priority compounds based on high exposure and toxicity assessed by means of toxicity reference
values (TRVs). Application of the framework to many emerging compounds such as novel flame retardants was
not possible because of either no or outdated TRVs. Concurrent identification of seven SVOC groups in indoor
environments provided information on their comparative levels and distributions, their sources, and informed
our assessment of associated risks.
1. Introduction
Semivolatile organic compounds (SVOCs) found in indoor environments
include flame retardants (FRs), plasticizers, pesticides, combustion
by-products, those added to personal care products, and their degradation
products (Lucattini et al., 2018). Concentrations in indoor
dust can vary over orders of magnitude from pg/g to μg/g levels (Venier
et al., 2016; Vykoukalová et al., 2017; Weschler and Nazaroff, 2008).
Indoor SVOC levels are affected by emission rates from consumer products,
building materials and indoor activities, rates of SVOC transport
from outdoor to indoor, removal rates by ventilation and cleaning, and
rates of depletion/degradation indoors (Weschler, 2009; Weschler and
Nazaroff, 2008; Zhang et al., 2009). Since loss rates relative to outdoor
environments are low, SVOCs persist indoors more than outdoors (Shin
et al., 2013; Weschler, 2009). Combined with > 90% of time spent in
indoor environments, the indoor environment becomes an important
source of human exposure to SVOCs.
In the environmental literature, SVOCs first gained notoriety with
compounds now identified as legacy persistent organic pollutants
(POPs). Specifically polychlorinated biphenyls (PCBs), dichlorodiphenyltrichloroethane
(DDT), and hexachlorobenzene (HCB)
were among the original “Dirty Dozen” chemicals listed on the
Stockholm Convention on POPs, which entered into force in 2004.
Pentachlorobenzene (PeCB) and hexachlorocyclohexane isomers
(HCHs) were added in 2009, followed by commercial mixtures of
polybrominated diphenyl ethers (PBDEs) in 2009 and 2017, and hexabromocyclododecane
(HBCDD) in 2013 (Stockholm Convention,
2018). Intense efforts to enact controls of Stockholm POPs followed
https://doi.org/10.1016/j.envint.2019.04.001
Received 26 November 2018; Received in revised form 29 March 2019; Accepted 1 April 2019
⁎
Corresponding author.
E-mail address: melymuk@recetox.muni.cz (L. Melymuk).
1
Current address: Graduate School of Oceanography, University of Rhode Island, Narragansett, Rhode Island 02882-1197, USA.
Environment International 127 (2019) 653–663
Available online 13 April 2019
0160-4120/ © 2019 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/BY/4.0/).
T
clear evidence of their ubiquity in the environment and biota, including
humans. Many studies have documented their outdoor concentrations,
in part to gauge the success of controls on new uses of legacy POPs
implemented through the Stockholm Convention (Gioia et al., 2006;
Harner et al., 2004; Hung et al., 2016; Wang et al., 2007). Fewer studies
have documented indoor contaminant levels, and generally they are
limited to PBDEs, HBCDD and PCBs (Ali et al., 2012; Harrad et al.,
2009b; Marek et al., 2017; Melymuk et al., 2016b; Tao et al., 2016).
Indoor levels of legacy POPs are controlled by primary indoor
sources and continuing emissions (Audy et al., 2018; Frederiksen et al.,
2012; Zhang et al., 2011). SVOCs such as PCBs and FRs were used in
floor coverings, electrical equipment, upholstered furnishing, building
materials and carpets, while organochloride pesticides (OCPs) were
used for pest control indoors and outdoors (Lucattini et al., 2018).
Polycyclic aromatic hydrocarbons (PAHs) are produced unintentionally
during indoor combustion processes (e.g., cooking, heating, smoking)
(Maertens et al., 2008). The presence of these compounds in the indoor
environment results in increased human exposure. For example, inhalation
of PCBs from indoor environments was found to be a significant
exposure route in school children (Ampleman et al., 2015;
Marek et al., 2017). Additionally, organophosphate FRs in house dust
were shown to be significantly correlated with altered hormone levels
and decreased semen quality in men (Meeker and Stapleton, 2010). To
reduce exposures, we need to better understand indoor sources of
SVOCs.
This study was motivated by high levels of PCBs observed in breast
milk (Drobná et al., 2011) and serum samples (Hovander et al., 2006;
Jursa et al., 2006; Petrik et al., 2006; Strémy et al., 2019; Wimmerová
et al., 2015) from people living in and around Michalovce, Slovakia.
The observed serum levels were two to three times higher than that of
the general population of Slovakia (Hovander et al., 2006; Jursa et al.,
2006; Petrik et al., 2006). These exposures have been attributed to a
nearby PCB production facility (Strémy et al., 2019). The Chemko
Strážske factory, which operated between 1959 and 1984, produced
21,000 t of PCBs – 1.6% of the total global PCB volume (Breivik et al.,
2007). Production resulted in the contamination of surrounding soils,
sediments, ambient air and surface waters (Hiller et al., 2011; Kocan
et al., 2001). For example, 53,000 ppm PCB was observed in a soil
sample at asphalt-gravel mix plant in the factory complex and ambient
air levels around the factory were 10 to 20 times higher than control
areas (Kocan et al., 2001).
Despite high PCB levels outdoors, indoor contamination in the region
had not yet been investigated. We hypothesized that levels of PCBs
in homes in the region would be similarly elevated due to contamination
from the factory and outdoor-to-indoor transfer of pollution. Here
we report PCB levels in air and dust in 60 homes in the region, and
compare them with other legacy POPs such as OCPs and PBDEs, as well
as other SVOCs such as current use FRs and PAHs. We also present a
framework that enables evaluation of the potential risk posed by indoor
levels of SVOCs in terms of both compound toxicity and estimated
human exposure.
2. Materials and methods
2.1. Sampling
Samples were collected in 60 homes in Eastern Slovakia from
March–April 2015 (Fig. S1). Homes were a subset of those participating
in the PCB cohort (Hertz-Picciotto et al., 2003) established in the region
in 2003. Homes were distributed evenly between urban and rural areas
within 40 km of the Chemko Strážske factory, covering a 750 km2
region.
The construction years of the homes ranged from 1930 to 2000,
with the mean and median of 1971. One air sample and one floor dust
sample was collected per home in the child's bedroom.
Single-bowl polyurethane foam passive air samplers (PUF-PAS)
were deployed for 32–35 days to measure indoor ambient air
concentrations. PUFs were pre-cleaned for 8 h in acetone and 8 h in
dichloromethane (DCM). Indoor PUF-PAS were previously calibrated
for this set of compounds (Audy et al., 2018; Venier et al., 2016;
Vykoukalová et al., 2017). A generic sampling rate of 1.6 m3
/day was
applied, leading to each sampler capturing approximately 50 m3
of air
during the deployment period. After deployment, PUFs were removed
from the PAS housing, packed in two layers of aluminum foil and stored
in a portable freezer at −18 °C for transport to the laboratory.
Dust samples were collected at the end of the PUF-PAS deployment
in the same room using a household vacuum cleaner with polyester
sock inserts. Socks were precleaned via Soxhlet extraction in DCM for
8 h, and before sampling and between samples the vacuum nozzle and
tube were cleaned with propan-2-ol. To collect each sample, a polyester
sock was inserted into the front of the vacuum tube and held in place by
the vacuum nozzle, and the accessible floor surface was vacuumed
(typically between 1 and 3 m2
). The sock was removed from the vacuum
cleaner, packed in two layers of aluminum foil, labeled and put
into a zip-lock bag, stored in a portable freezer at −18 °C for transport
to the laboratory.
2.2. Analysis
The target analytes were 10 PBDE congeners (28, 47, 66, 85, 99,
100, 153, 154, 183 and 209), 3 HBCDD isomers, 22 “novel” halogenated
flame retardants (NFRs) (TBP-AE, α-DBE-DBCH, β-DBE-DBCH,
TBX, TBP-BAE, α-TBCO, β-TBCO, PBBZ, TBCT, DDC-CO-MA, PBT,
PBEB, TBP-DBPE, HBB, PBBA, DBHCTD, EH-TBB, BTBPE, TDBPTAZTO,
s-DDC-CO, a-DDC-CO, BEH-TEBP), 18 organophosphate esters
(OPEs) (TPrP, TIBP, TNBP, TCEP, TCIPP, DBPP, TPeP, BDPP, TDCIPP,
TBOEP, TPHP, EHDPP, TEHP, TOTP, TMTP, TPTP, TIPPP, TDMPP), 9
PCB congeners (9, 11, 28, 52, 101, 118, 153, 138, 180), 12 OCPs (PeCB,
HCB, α-HCH, β-HCH, γ-HCH, δ-HCH, o,p'-DDE, p,p'-DDE, o,p'-DDD,
p,p'-DDD, o,p'-DDT, p,p'-DDT) and 27 PAHs (ACY, ACE, FLU, PHE, ANT,
FLA, PYR, BAA, CHR, BBF, BKF, BAP, IcdP, DBA, BGP, RET, benzo[b]
fluorene, benzo‑naptho‑thiophene, benzo[ghi]fluoranthene, cyclopenta
[cd]pyrene, triphenylene, benzo[j]fluoranthene, benzo[e]pyrene,
PERY, dibenzo[ac]anthracene, antanthrene, coronene). Full compound
names, CAS numbers, and further details are given in Table S1; abbreviations
of FRs are adopted from Bergman et al. (2012).
All the analytical procedures used here were published previously
(Jílková et al., 2018; Venier et al., 2016; Vojta et al., 2017;
Vykoukalová et al., 2017) and are described in detail in the SI. PUFs
were extracted with DCM using a Büchi B-811 automated warm Soxhlet
extraction system. Before extraction, dusts were sieved with a 500 μm
sieve to remove coarse particles (e.g., hair, large fibres). Approximately
100 mg of the sieved dust was used for extraction. Dust samples were
extracted via three times repeated sonication in 1:1 v/v hexane:acetone.
13
C-labeled or deuterated internal standards (for PBDEs, HBCDD, PAHs,
OPEs) and non-environmental PCB congeners (for PCBs and OCPs)
(Wellington Laboratories Inc., Canada) were added before extraction.
Dust and PUF extracts were split 70:30. The 70% aliquot was purified
with a H2SO4-modified silica gel column and analyzed for PBDEs,
HBCDD, PCBs and OCPs. The 30% aliquot was cleaned and fractionated
using a 5 g activated silica column eluted with DCM, followed by 7:3 v/
v acetone:DCM. The first fraction was used for analysis of PAHs and
NFRs, and the second fraction for OPEs.
PBDEs and NFRs were analyzed using gas chromatography coupled
to high resolution mass spectrometry (HRGC-MS), while PCBs, OCPs
and OPEs were analyzed by gas chromatography-tandem mass spectrometry.
PAHs were analyzed using gas chromatography–mass spectrometry.
HBCDDs were analyzed after exchanging solvent to acetonitrile
by liquid chromatography electrospray ionization mass
spectrometry. See SI for further details on instrumental analysis.
H. Demirtepe, et al. Environment International 127 (2019) 653–663
654
2.3. QA/QC
The recoveries of individual compounds were determined using a
set of pre-cleaned PUFs (n = 8) spiked with the native analytes prior to
extraction. The recoveries were 70–120%. Sample masses were adjusted
for recoveries based on the internal standards (for PCBs, OCPs,
PAHs, and OPEs) or through the isotopic dilution instrumental method
applied (PBDEs, NFRs, HBCDDs).
Field blanks (5 PUFs and 4 vacuum bags) were transported to the
sampling site, manipulated as per the samples, and then treated as per
the samples for the analytical process. Method detection limits (MDLs)
were calculated as the average of the field blank plus three times the
standard deviation of the blanks. The instrument detection limit (IDL)
was used as the MDL for compounds that were not detected in the
blanks (Tables S2-S7). The average of the blank of the corresponding
matrix was subtracted from sample values that were > MDL, and
values < MDL were recorded as such.
2.4. Data analysis
Statistical analyses were conducted using IBM SPSS 25. For statistical
analysis, values below detection were substituted by √2/2∙MDL.
Non-parametric tests were applied for correlations (Spearman correlation,
rs) and comparison of differences between compounds (Mann
Whitney U test) since the Shapiro-Wilk normality test showed nonnormal
distributions (p < 0.001) for dust and air data. Correlations
and differences were considered significant if p < 0.05. Principal
component analysis (PCA) and analysis of variance (ANOVA) were
conducted after log-transforming the data.
3. Results and discussion
3.1. Compound concentrations and geographical comparison
The concentrations of all compounds and compound groups found
in indoor dust and air from Slovakia are provided in Table S8. Median
Fig. 1. Comparison of indoor concentrations for (A) air and (B) dust from Slovakia (green boxplots), and literature values from low-, middle-, and high-income
countries classified based on criteria from the World Bank (2018a). Each dot represents a central tendency value (median or mean) for one country (see Tables
S10–S25). For the Slovakia data, the boxes represent interquartile ranges, error bars below and above the box represent the 10th and 90th percentiles, respectively.
(For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
H. Demirtepe, et al. Environment International 127 (2019) 653–663
655
concentrations in dust ranged from 0.13 ng/g for CB-9 to 2450 ng/g for
TBOEP. Generally, OPEs had the highest concentrations in dust, followed
by PAHs. Median air concentrations ranged from 0.03 pg/m3
for
TBCT to 16,790 pg/m3
for phenanthrene. PAHs, followed by OPEs, had
the highest air concentrations.
We examined whether common profiles or sources for compound
groups could be identified using PCA. PCA was able to explain 61.7% of
the variability within the dust concentration dataset with three components.
Component 1 included ∑7PCBs, ∑HCH, ∑DDX and ∑27PAH,
component 2 involved ∑9PBDEs, ∑14NFRs, ∑14OPEs and BDE-209, while
CB-11, PeCB and HCB formed component 3. Hence, we can infer that
for dust concentrations, PCBs and pesticides have similar profiles, i.e.
many houses have high OCPs and high PCBs, while all flame retardants
and plasticizers have similar profiles, suggesting similar sources. This
clustering is consistent with differences in building age and equipment,
with homes from before the restriction of PCBs and OCPs having higher
levels of these compounds (Audy et al., 2018), further supported in our
data analysis below (Section 3.3). Correlations among the compound
groups in component 1 were also significant and strong (Table S9).
Significant but relatively weaker correlations were found in component
2 between BDE-209 and ∑9PBDEs (rs = 0.48), ∑14NFRs and ∑9PBDEs
(rs = 0.43) and BDE-209 (rs = 0.27), ∑14OPEs and ∑9PBDEs (rs = 0.47)
and BDE-209 (rs = 0.28). These compounds all have similar uses as
flame retardants. Correlations between compounds in component 3
were also significant: PeCB and HCB (rs = 0.73), CB-11 and PeCB
(rs = 0.58), CB-11 and HCB (rs = 0.60). For indoor air concentrations,
PCA was able to explain 68.7% of the variability with five components,
among which only component 1 constituted a meaningful cluster,
containing ∑HCH, ∑DDX, HCB and ∑7PCBs. Additionally, air concentrations
of compound groups did not have significant and strong
correlations with each other, except for a few pesticides, such as ∑HCH
and ∑DDX (rs = 0.71), HCB and ∑DDX (rs = 0.57). Again, these
groupings are consistent with the explanation related to building age.
We observed differences in indoor air and dust concentrations when
qualitatively comparing our data with previously published indoor levels
from other geographic regions (Fig. 1), supporting the findings of
others (Dirtu et al., 2012; Harrad et al., 2009b; Venier et al., 2016;
Vykoukalová et al., 2017). To understand the reasons for such differences,
we selected the studies reporting residential indoor SVOC levels
after the year 2000 regardless of the room type. For PCBs, we used the
sum of 7 PCB indicator congeners (CB-28, -52, -101, -118, -153, -138,
-180) and for PAHs the sum of 16 EPA-PAHs. Data for this comprehensive
comparison is listed in Tables S10-S25. Slovakian levels fell
within the range of those previously measured in other countries for
most compounds. However, the Slovakia samples had notably higher
levels of HCB in dust and air, and lower levels of PBDEs, DBE-DBCH and
HBB in air compared to median levels from other countries (Fig. 1).
Next, we tested the hypothesis that country-level indoor concentrations
of these legacy and current use SVOCs were related that
country's economic status (categorized according to World Bank,
2018a). Typically, low-income countries (LIC) had lower concentrations
of SVOCs than middle and high income countries (MIC and HIC)
(Fig. 1). For example, the highest concentrations of FRs in dust were
reported in HICs. In contrast, the higher levels of OCPs were reported in
dust of MICs, especially when compared to HICs for which agriculture
value had the lowest contribution to the gross domestic product (GDP)
(World Bank, 2018b). The highest PAH levels in indoor dust and air
were reported in China, attributed to residential coal combustion for
cooking and heating (Lv et al., 2009; Wang et al., 2013).
For PCBs, we further hypothesized that indoor air and dust concentrations
were related to historical PCB use per capita in a specific
country. We explored this hypothesis by calculating PCB use per capita
per country using the consumption estimates of Breivik et al. (2002).
Seven PCB congeners were summed for the period of 1930 to 1984 for
each country and divided by its population in 1970, reflecting the time
of peak PCB use (Table S26; for Slovakia and Czech Republic, total use
for Czechoslovakia and total population of two countries were used).
When more than one median concentration for dust and air was reported
for a country, we included all values as separate data points. We
found a significant correlation between PCB use per capita and median
PCB dust and air concentrations for 10 countries (rs = 0.59 and 0.91,
for dust and air, respectively, p < 0.05). We concluded that broad,
national-level differences in current indoor residential PCB concentrations
are related to the past national use of PCBs. It should be noted that
the data from these studies are based on samples of convenience that
were often of limited sample size, i.e. the number of homes varying
between five and 60, which might limit the representativeness of the
data for an entire country. Another limitation to this conclusion was
that indoor air PCB concentrations were reported for only five countries
of which most were HICs.
3.2. Polychlorinated biphenyls
The median concentration of ∑9PCBs in dust from Slovakian homes
was 58 ng/g (17.4–465 ng/g), and dust was dominated by CB-11 (30
%), -153 (20%) and -180 (13%). The median concentration of ∑9PCBs in
air was 1.09 ng/m3
(0.28–6.92 ng/m3
), CB-11, -28 and -52 dominated
indoor air. While CB-28, -52, -101, -118, -153, -138, -180 were the
dominant legacy PCB congeners that were present in technical mixtures
(e.g., Aroclor mixtures manufactured in USA and Delor mixtures manufactured
at Chemko Strážske and used in the former Czechoslovakia),
CB-11 was not present in any Delor mixtures (Taniyasu et al., 2003).
Rather, CB-11 is produced unintentionally during the manufacturing of
pigments, used in paints, paper products and textiles (Hu and
Hornbuckle, 2010; Rodenburg et al., 2010; Vorkamp, 2016), and can be
emitted from floor and cabinet surfaces (Herkert et al., 2018). Hence,
the predominance of this non-legacy PCB congener in Slovakian indoor
dust and air suggests the importance of modern indoor sources of PCBs,
and suggests that attention should be given to a broader set of PCB
congeners beyond the standard seven indicator PCBs representing only
legacy PCB uses.
Correlations between PCB congeners in dust revealed that CB-9 and
CB-11 correlated strongly only with CB-28, CB-52 and with each other
(rs > 0.60, p < 0.001), while the seven indicator PCBs correlated
significantly with each other (p < 0.05), possibly related to their vapor
pressure. On the other hand, indoor air concentrations of CB-9 and
seven indicator PCBs correlated significantly (p < 0.01), whereas CB-
11 did not correlate with any other PCB congeners, supporting the
hypothesis of unique modern sources of CB-11 in indoor air (Hu and
Hornbuckle, 2010; Rodenburg et al., 2010).
Next we tested the hypothesis that the main source of legacy PCB
levels (Σ7PCBs) to Slovakian homes was the nearby Chemko Strážske
PCB production facility, where production ceased in 1984. We compared
indoor levels with the distance of homes (between 1.7 and 41 km)
from the Chemko Strážske PCB production facility. However, we found
no significant correlation between legacy PCB levels in indoor dust or
air and distance from the facility, suggesting that Chemko Strážske is
not the major influence on current indoor PCB contamination in the
homes. A previous study in the region, conducted 15 years after the
cessation of PCB production, found that ambient outdoor air concentrations
in residential areas < 5 km from the factory were up to
1700 pg/m3
and levels decreased to regional background within 15 km
(Kocan et al., 2001). Additionally, a recent study found that the elevated
levels of PCBs in human serum samples from the region were
associated with the factory, i.e. higher levels were detected in people
living downwind from the factory when compared to other regions
(Strémy et al., 2019). In our study, the home showing highest legacy
PCB concentration was 6 km from the factory. However, we only
sampled seven homes within 5 km from the factory, which may have
limited our ability to see an effect from proximity to the factory. Rather,
we found that indoor levels in this region were comparable to indoor
levels in similar locations regardless of distance from the Chemko
H. Demirtepe, et al. Environment International 127 (2019) 653–663
656
Strážske plant, e.g., within the range of indoor levels measured in other
HICs (Fig. 1).
In addition to the “macro-level” factor of country-wide PCB use,
indoor legacy PCB levels can be influenced by building age and materials
used (Hazrati and Harrad, 2006; Marek et al., 2017; Melymuk
et al., 2016a). The homes sampled in this study were mainly made of
brick (n = 26) and concrete (n = 19). The legacy PCB concentrations in
indoor dust were significantly higher in brick than concrete homes
(one-way ANOVA, F = 5.67, df = 3, p < 0.05, Fig. S2). This was
contrary to our expectations given the association of PCBs with building
sealants used in concrete buildings (Kohler et al., 2005; Liu et al.,
2016). Other primary indoor PCB sources include paints, adhesives,
wire and cable insulation manufactured before 1984, i.e. sources that
cannot be easily removed and may be still present in homes. Thus,
higher PCB concentrations in brick homes could be attributed to these
PCB-containing materials and subsequent absorption into bricks (U.S.
EPA, 2012), and the presence of materials serving as secondary sources
of PCBs, such as window frames and wallpaper (Weis et al., 2003).
We also compared indoor PCB levels in homes built before and after
the PCB ban in 1984 in Czechoslovakia (Holoubek et al., 2006). No
significant difference was found in legacy PCB concentrations before
and after the ban, however, there were only four homes built after
1984, limiting the power of this analysis.
3.3. Organochlorine pesticides
Eleven of 12 OCPs had > 50% detection frequency either in dust or
in air. The median concentration of ∑11OCPs in dust was 59.4 ng/g
(12.2–634 ng/g) and dusts were dominated by p,p’-DDE (34.2%), p,p’DDT
(20.8%), HCB (15.1%) and γ-HCH (10%). Indoor air concentrations
of ∑11OCPs ranged from 336 to 7110 pg/m3
, with a median of
945 pg/m3
. The dominant compounds in air were HCB, γ-HCH, and
p,p’-DDE. Dust concentrations of all eleven OCPs correlated significantly
with each other (p < 0.05), while air concentrations of all
but PeCB correlated strongly with each other (rs > 0.50, p < 0.01),
suggesting either similar sources of OCPs or from homes of similar age.
We hypothesized that the major sources of OCPs in the homes were
either outdoor secondary sources from nearby agricultural lands with
past use of OCPs, and/or past indoor uses for indoor insect control
(Audy et al., 2018; Booij et al., 2016; Holt et al., 2017). We tested this
hypothesis by considering: (1) percentage of agricultural land in the
vicinity of each home, and (2) building age (as a proxy for past indoor
chemical use). The percentage of agricultural land within a 1 km radius
of each home was determined using a geographic information system
(GIS) and the Corine Land Cover database (European Environment
Agency, 2012).
∑HCH and ∑DDX concentrations in both dust and air were significantly
correlated with the % agricultural land within a 1 km radius
of the homes (rs > 0.31, p < 0.01). Except for α-HCH in air, individual
isomers of HCH and DDX also correlated significantly with %
agricultural land (rs > 0.29, p < 0.03). Additionally, homes having
≥65% agricultural land within 1 km radius had significantly higher
∑HCH and ∑DDX concentrations than homes with lower % agricultural
land (one-way ANOVA, p < 0.05, Fig. S3). This relationship suggests
that past OCP use in agricultural lands in the vicinity of homes significantly
contributes to indoor OCP levels.
If past direct use of OCPs indoors is a significant source, we hypothesized
that higher OCP levels would be expected in older homes
(Audy et al., 2018). ∑DDX concentrations in dust and air were significantly
higher in homes built before than after the 1973 ban (MannWhitney
test U = 61.0, z = −2.47, p < 0.05 and U = 47.0,
z = −3.00 p < 0.05; Fig. 2a and b). Concentrations of individual
isomers and ∑HCH were also significantly higher in homes built before
the ban on HCH in 1977 (U = 31.0, z = −3.21, p < 0.05 for dust and
U = 35.0, z = −3.05, p < 0.05 for air; Fig. 2c and d). This suggests
that use of DDX and HCH in homes > 40 years ago remains a significant
contributor to present-day indoor levels. Indoor use of OCPs can result
in extremely high levels indoors at the time of use, and elevated levels
many years after use (Booij et al., 2016; Holt et al., 2017). Moreover,
SVOCs associated with indoor dust do not readily degrade due to limited
microbial and abiotic degradation processes (Hwang et al., 2008).
Hence, OCPs used in the indoor environment may persist. No significant
difference was found in HCB concentrations according to building age.
Furthermore, we tested the relation between the two variables affecting
OCP concentrations, which were % agricultural land and building age.
We found no significant correlations between these two variables
(p > 0.05), which suggested that they independently affect indoor
OCP levels.
3.4. Flame retardants
The median ∑9PBDEs concentration was 12.3 ng/g (3.2–197 ng/g)
in indoor dust. The median concentration of BDE-209 was 193 ng/g
(24.8 to 2965 ng/g). BDE-209 was the predominant congener, accounting
for 38–99.6% of ∑10PBDEs. Other dominant congeners in dust
were BDE-99 and -47. The median concentration of ∑9PBDEs in air was
4.3 pg/m3
(0.70–113 pg/m3
). As expected, BDE-209 had a low detection
frequency in air (45%) but concentrations ranged
from < MDL–420 pg/m3
. The dominant congeners in air were BDE-47,
-28 and -99.
The median concentration of ∑HBCDD in dust was 153 ng/g
(20.2–1900 ng/g) with the greatest contribution from γ-HBCDD in 47
homes, whereas α-HBCDD dominated in 13 homes. The γ-isomer constitutes
> 70% of the technical HBCDD mixture, hence its dominance in
dust samples was expected. However, many studies reported higher
indoor levels of α-isomer, possibly due to photolytic isomerization in
dust (Harrad et al., 2009a) and/or interconversion between isomers
during manufacturing of flame-retarded products (Abdallah et al.,
2016; Dirtu et al., 2012). Median ∑HBCDD in air was 5.4 pg/m3
(0.46 to
239 pg/m3
). Similar to the dust concentrations, air concentrations were
dominated by γ-HBCDD in all but seven homes that were dominated by
the α-isomer, three of which were the same as homes where α-isomer
dominated in dust.
Fourteen out of 22 NFRs had > 50% detection frequency either in
dust or in air. The median dust concentration of ∑14NFRs was 122 ng/g
(29.6–798 ng/g), dominated by BEH-TEBP (61%) and α- and β-DDC-CO
(15%). The median concentration of ∑14NFRs in air was 40.8 pg/m3
(11.2–407 pg/m3
). PBT, PBBZ, HBB and α- and β-DBE-DBCH made up
71% of the total NFR air levels.
Among the 18 OPEs analyzed, 14 OPEs had detection frequencies
> 50% either in dust or air. The concentrations of median
∑14OPEs in dust was 12,400 ng/g (3080–312,000 ng/g). These levels
were two orders of magnitude higher than other flame retardants (i.e.,
BDE-209, HBCDD and NFRs), which is consistent with other reports (Ali
et al., 2013; Dirtu et al., 2012; Okeme et al., 2018). This greater
abundance of OPEs levels has been ascribed to the use of some compounds
as plasticizers (e.g., non-chlorinated OPEs) (Marklund et al.,
2003), the need to use higher concentrations than brominated flame
retardants to achieve the same flame retardancy, and the higher vapor
pressures of some of the OPEs relative to brominated compounds
(Sühring et al., 2016; Zhang et al., 2016). The abundant OPE species in
dust in the Slovakian homes were TBOEP, TDCIPP and TPHP, constituting
nearly 70% of all OPEs. OPE distributions in Slovakia were
similar to those observed in other European countries (Dirtu et al.,
2012; García et al., 2007; Van Den Eede et al., 2012), and Canada (Yang
et al., 2019) in that TBOEP was the greatest contributor in dust samples.
This profile differs from typical profiles in Asia where TDCIPP or TCIPP
were the highest contributors (Tan et al., 2017; Zheng et al., 2017).
Indoor air concentrations of ∑14OPEs were a median of 13.7 ng/m3
(5.25–74.7 ng/m3
). These levels were three orders of magnitude higher
than NFRs. The main contributors of OPEs in indoor air were TNBP,
TIBP, TCEP and TCIPP.
H. Demirtepe, et al. Environment International 127 (2019) 653–663
657
All PBDE congeners in dust correlated well with each other, indicating
common emission sources, while air concentrations showed
correlations among lower brominated and among higher brominated
congeners separately, due to the different volatility of congeners.
HBCDD isomers in dust correlated with each other but not with other
FRs in dust, while air concentrations correlated with lower brominated
PBDE congeners in air. Strong correlations were found between BDE-
183 and BTBPE in dust and air (dust rs = 0.71, air rs = 0.61,
p < 0.001); and most of the other individual FRs correlated with each
other, but these were weak correlations.
Additionally, correlations between FRs showed an unexpected difference
between urban and rural homes. ∑9PBDEs in dust correlated
significantly with ∑14NFRs (rs = 0.61, p < 0.05) and ∑14OPEs
(rs = 0.48, p < 0.05) for urban homes, while no correlations were
found between these groups for rural homes. The reason for this difference
between urban and rural homes is unclear, but may suggest
differences in the timing of product replacement and new product
purchase between urban and rural homes.
As mentioned above, indoor FR levels are expected to be influenced
by indoor sources and building characteristics. We found no significant
correlations between FRs in air or dust and the prevalence of PVC
flooring, wall-to-wall carpeting or new furniture in the home. However,
we were able to identify regional differences in replacement FRs used
by comparing the Slovak indoor compound distributions with those
from North American indoor environments. FireMaster-550 (FM-550)
was a common substitute for penta-BDE mixture after its restriction in
North America (Stapleton et al., 2008). The ratio of the brominated
compounds in FM-550, i.e. ƒEHTBB = EH-TBB/EH-TBB + BEH-TEBP, can
indicate whether these compounds are emitted from a material containing
FM-550 (Venier et al., 2016). Before calculating the ratio, air
concentrations were corrected for the volatilities of compounds by
using PUF-air partition coefficient, KPUF-air, according to Francisco et al.
(2017) and Okeme et al. (2017). The median of ƒEHTBB was 0.09
(0.02–0.54) and 0.001 (0.00–0.02) for dust and air concentrations, respectively.
These ratios were both much lower than that of FM-550,
which is 0.77 (Ma et al., 2012). Lower ƒEHTBB ratios have been typically
observed in European indoor samples (Cequier et al., 2014; Venier
et al., 2016) than in North American samples (Dodson et al., 2012;
Venier et al., 2016). One explanation is the use of BEH-TEBP individually
as a flame retardant and plasticizer in Europe, compared
with greater use of FM-550 in North America (Venier et al., 2016). For
example, another commercial mixture, DP-45 which is 100% BEHTEBP,
is listed as a flame retardant and plasticizer for PVC applications,
e.g. wire and cable insulation, coating fabrics, and sheeting and flexible
polyurethane (Lanxess, 2016). Both FM-550 and DP-45 are high production
volume chemicals in USA, whereas in Europe FM-550 is not
registered but DP-45 is manufactured and/or imported at volumes similar
to USA (ECHA, 2018). The wide application of BEH-TEBP in
consumer products in Europe likely explains its abundance relative to
EH-TBB (medians = 69.5 and 6.8 ng/g in dust, respectively). In fact,
EH-TBB does not have an EC number and it has no REACH record.
Hence, direct use of EH-TBB in commercial products in Europe still
remains unidentified. Although ƒEHTBB suggested that EH-TBB and BEHTEBP
had different sources, the two compounds were significantly
correlated (rs = 0.46, p < 0.001). Therefore, we speculate that the low
amount of EH-TBB detected in homes likely originates from the limited
use of FM-550 in products in Europe, e.g., imported as finished pro-
ducts.
3.5. Polycyclic aromatic hydrocarbons
∑27PAHs had a median dust concentration of 2008 ng/g (505 to
Fig. 2. Differences in indoor dust and air concentrations of ∑DDX (a and b) and ∑HCH (c and d) measured in homes built before (n = 18 for ∑DDX, n = 22 for ∑HCH)
and after (n = 14 for ∑DDX, n = 10 for ∑HCH) OCPs controls were implemented in Czechoslovakia.
H. Demirtepe, et al. Environment International 127 (2019) 653–663
658
37,018 ng/g). The dust concentrations were dominated by phenanthrene,
fluoranthene and pyrene. The median of ∑27PAHs in indoor air
was 43.1 ng/m3
(17.2 to 1923 ng/m3
), dominated by acenaphthene,
fluorene and phenanthrene. All PAH compounds in dust correlated well
with each other, indicating similar sources for all, while air concentrations
were not correlated for some PAHs (notably between acenaphthene
and PAHs with > 4 rings), possibly due to differences in
volatility.
Indoor PAH contamination is typically attributed to outdoor sources
such as vehicle or industrial emissions, or indoor sources such as
smoking or domestic biomass burning (Fromme et al., 2004; Whitehead
et al., 2013; Yadav et al., 2018). The nearest industrial PAH emission
source to the Slovak homes was a steel mill also located at the Chemko
Strážske site. Nevertheless, no meaningful correlation was found between
indoor PAH levels and distance to the steel mill. We also tested
the influence of vehicle emissions using the total length of roads within
a 1 km radius of each home derived using GIS. However, road length
was also not useful for explaining indoor PAH levels.
Rather, we identified domestic heating as the likely major indoor
source of PAH. ∑27PAHs in dust, but not air, was significantly higher in
homes heated with wood (median = 3100 ng/g) than those using central
or gas heating (median = 1679 ng/g) (Mann-Whitney test U = 170,
z = −2.69, p < 0.05, Fig. S4). Additionally, levels in wood-burning
homes were significantly higher for 18 individual PAHs in dust, seven
of which were identified as having indoor emissions due to wood
burning (Gustafson et al., 2008). We also considered whether cigarette
smoking (n = 27) affected indoor PAH levels, however, no correlation
between PAH concentrations and indoor smoking was found. Hence,
indoor wood burning was the dominant factor controlling indoor PAH
concentrations.
3.6. Human exposure to SVOCs and associated risk
A challenge in evaluating implications of compounds detected in
indoor environments and prioritizing monitoring is the wide range of
concentrations in different indoor matrices, which are associated with
different human exposure pathways, in addition to the wide range of
toxicities of the various compounds. Compounds are often highlighted
due to elevated levels in indoor environments (e.g., OPEs); however,
this type of prioritization does not address the risk, i.e., exposure
combined with toxicity. Several attempts have been made to address
the risk associated with residential indoor exposure (Bonvallot et al.,
2010; Mitro et al., 2016; Pelletier et al., 2018; Shin et al., 2012; Smith
et al., 2016). We propose a classification framework to merge information
on human exposure and toxicity to evaluate risk for indoorrelevant
compounds. Further, we have applied it to prioritize risk from
indoor exposures from the levels reported in Slovakian homes.
To address the combined influence of indoor air and dust on human
exposure, human intake of SVOCs was calculated as the sum of exposure
doses via air inhalation, dust ingestion, and dermal contact with
dust. The daily exposure dose formulas and assumptions are provided in
Table S27, and Fig. S5 summarizes the % distribution of exposure
routes for individual compounds. Intakes were calculated for median
and high (95th percentile) scenarios for a child between ages six and
eleven, as the samples were taken in children's bedrooms. We also
calculated intake for an adult female for comparison. Total intakes of
SVOCs for both scenarios are given in Tables S28 and S29.
The total intake for individual compounds ranged between
7.50 ∗ 10−5
ng/kg/d and 2.13 ∗ 10−5
ng/kg/d for TBX to 4.9 ng/kg/d
and 1.7 ng/kg/d for phenanthrene according to the median intake
scenario, for a child and an adult, respectively. The highest exposures
were observed for PAHs > OPEs > PCBs. Estimated exposure for the
high intake scenario for higher chlorinated PCB congeners, DDTs, BDE-
99, PBT, HBB, acenaphthene and cyclopenta(cd)pyrene were more than
one order of magnitude higher than their median intake scenario rates.
Lower chlorinated PCBs, OCPs except DDT, PBBZ, PBT, TIBP, TNBP and
lower molecular weight PAHs had higher exposure doses via inhalation
than dust ingestion and dermal dust contact due to their higher concentration
in indoor air (Figs. S5 and S6). As expected, the child had
higher estimated intakes of SVOCs than the adult female.
To evaluate the relative toxicities of the target compounds, we
compiled non-carcinogenic oral chronic (or sub-chronic if the chronic
value was not available) toxicity reference values (TRVs) from the literature
and regulatory sources. TRVs are defined as a maximum dose of
a compound that can be ingested daily without risk for people's health.
We selected those for which the details on the calculation method were
Fig. 3. Total human intake for child age
6–11 vs toxicity reference values (TRVs) of
indoor SVOCs. Vertical lines represent the
range of toxicity values from literature and
regulatory agencies, horizontal lines represent
the high intake scenario, and points
represent median exposure scenario and
median values of TRVs. Compounds having
both high intake and high toxicity (i.e.
highest priority) are labeled. The box below
the graph shows compounds with no available
TRVs.
H. Demirtepe, et al. Environment International 127 (2019) 653–663
659
accessible for further assessment (see supplementary information for
methods and Table S29).
Finally, intake estimates and TRVs were visualized on a scatterplot
to allow prioritization of compounds for monitoring and evaluating the
risk in the indoor environment for a child (Fig. 3). When several selected
TRVs were available for one compound, the median and the full
range of values were shown. The high intake scenario is reflected as half
error bars. For comparison, intake scenarios for an adult female are
presented in Fig. S7. Priority compounds are those in the upper right
corner that have the highest intake and highest toxicity (lowest TRVs).
The highest priority compound in our study was CB-11. It does not have
an individual TRV; however, it was assigned the tolerable daily intake
value for dioxins and planar PCBs (Baars et al., 2001) since it is a planar
PCB congener. It has not been frequently monitored in indoor environments
since CB-11 is a non-legacy PCB congener. However, its
emission indoors from paints and indoor materials has been reported
(see Section 3.2). This linking of intake and TRVs emphasizes that CB-
11 should be prioritized for indoor monitoring and evaluation of associated
risk.
Other high priority SVOCs were HCB, γ-HCH, CB-28 and-52, and
benzo[a]pyrene. It is important to note that new uses of HCB, γ-HCH
and PCBs were restricted decades ago, yet they still exhibit high human
intake due to their past usage. Individual PCB congeners have no specific
TRV. As such, we assigned the values reported for non-planar PCBs
and/or for Aroclors to individual PCB congeners except for CB-11.
Benzo[a]pyrene is one of the many compounds emitted from indoor
wood burning (Gustafson et al., 2008) and has a higher TRV than other
PAHs which had higher intake. The risk from PAHs is probably even
higher, as we are considering only non-carcinogenic TRVs, but note that
benzo[a]pyrene is classified as carcinogenic to humans (Group 1) according
to the International Agency for Research on Cancer (IARC)
(IARC, 2010). Bonvallot et al. (2010) ranked compounds present in
indoor dust depending on the selected TRVs and concentrations in dust
and similarly highlighted PCBs and γ-HCH (Bonvallot et al., 2010).
Some SVOCs did not appear to present a high risk because they had
low human intake in this study. However, some of these chemicals had
low TRVs, e.g., BDE-99, -47, and -153, PCBs, β-HCH and DDT, suggesting
that these compounds require attention in indoor environments
with higher levels.
Prioritization was not possible for 60% of the 88 compounds evaluated,
and these are shown in Fig. 3 according to only their estimated
intake. TCIPP, TPHP, TIBP, acenaphthylene, EHDPP, retene, TPrP and
chrysene had high intake but no reliable TRVs. More effort should be
given to toxicological studies of these compounds, especially those for
which the current literature suggests high toxicity. For example, Behl
et al. (2016) found that TPHP was the second most active reproductive
toxicant for C. elegans after penta-BDE commercial mixture DE-71.
Additionally, four OPEs had the highest intake but current TRVs suggested
low risk. However, the TRVs of these compounds are highly
uncertain because of insufficient and/or outdated testing. For example,
a cursory view of TDCIPP and TCEP might suggest minimal risk,
however their TRVs were derived from studies conducted almost
30 years ago (ATSDR, 2012), while much of the research on their
toxicity was done in the last 10 years but TRVs have not yet been determined.
We also noted that different calculation methods of TRVs
could produce values differing by up to several orders of magnitude, as
seen for BDE-99 and -209 (ATSDR, 2018; National Research Council,
2000; USEPA, 2018). For BDE-99, RIVM (Baars et al., 2001) provides a
maximal allowed intake that is 4 orders of magnitude lower than the
reference dose from the US EPA (IRIS), based on a different experimental
value and a different calculation method that takes into account
known bioaccumulation properties of the chemical in human as compared
to rodents. It is also surprising that TRVs for OPEs, such as
TDCIPP, are several orders of magnitude higher than for PBDEs such as
BDE-47 or -99, although they were shown to have comparable levels of
toxicity when tested in parallel in experimental systems (Behl et al.,
2015; Jarema et al., 2015; Noyes et al., 2015). Finally, epidemiology
studies indicate that several OPEs may represent a risk for people's
health. For example, two epidemiological studies reported a statistical
association between levels of TNBP (one of the most abundant compounds
in air and dust) and the risk of asthma, mucosal symptoms of
SHS (sick house syndrome) or allergic rhinitis (Araki et al., 2014;
Kanazawa et al., 2010). Other epidemiological studies indicate associations
between TCEP or TPHP exposure with neurobehavioral defects
in children and thyroid cancer, and between TDCIPP or TPHP exposure
and decrease in male fertility (Carignan et al., 2018; Castorina et al.,
2017; Hoffman et al., 2017; Hutter et al., 2013; Meeker et al., 2013;
Meeker and Stapleton, 2010). However, these outcomes are not captured
in the available TRVs.
The use of non-carcinogenic oral TRVs was the most appropriate to
estimate the toxicity for most of the compounds analyzed, but, in some
cases, it could underestimate the risk, for example, for chemicals that
are classified as carcinogenic to humans (Group 1) according to IARC,
such as benzo[a]pyrene or PCBs (IARC, 2010, 2016). Also, the compounds
having higher exposure via inhalation than ingestion (see Fig.
S5) should be investigated further in terms of their reference dose for
inhalation (if available) to have a better understanding of their toxicity
via this exposure route. Among the compounds with higher inhalation
intake, only γ-HCH has an inhalation reference dose (0.1 μg/kg/d) (US
Environmental Protection Agency, 2018). This value is within the range
of TRVs used in our study for γ-HCH, and hence does not change the
conclusion on the risk associated with this compound.
In our framework to prioritize and evaluate the risk of individual
compounds, we did not distinguish the main health effects associated
with each chemical, as done by others (Mitro et al., 2016). Future
iterations of this framework could be adapted to consider this aspect,
which could be especially useful for identifying potential mixture ef-
fects.
4. Conclusions and implications
Indoor air and dust levels of 88 SVOCs from seven compound classes
in Slovakia appeared to be related to a country's income level (e.g.,
PCBs), outdoor uses (OCPs), and/or indoor activities (e.g., PAHs). In
particular, a comparison by country with published indoor levels of
PCBs and FRs showed higher levels in high-income countries. Indoor
PCBs concentrations in Slovakian homes nearby a former PCB production
facility fell within the range of previously reported values from
other countries, and were not uniquely elevated due to proximity to the
PCB factory. Indoor OCP levels were correlated with the percentage of
agricultural lands in the vicinity, and with the age of the sampled
homes, both taken as surrogates for past outdoor and indoor OCP use,
respectively. FR concentrations and distributions were consistent with
the findings of previous European studies. Indoor PAH levels were associated
with wood burning for home heating and not smoking.
The highest indoor exposures for a child between ages 6–11 via air
inhalation, dust ingestion and dermal contact were for
PAHs > OPEs > PCBs. We presented a classification framework to
prioritize the risk of these compounds by considering human intake and
toxicity where toxicity was expressed as the highest recommended intake
per day (TRV). According to this classification, CB-11, HCB, γHCH,
CB-28, CB-52 and benzo[a]pyrene should be prioritized for indoor
monitoring and evaluation of the associated risks. However, 53
compounds (60% of the total number of compounds considered) could
not be evaluated due to a lack of reliable toxicological information.
Identification of a wide range of compounds in the indoor environment
enabled comparing not only the levels of various compound
groups, but also the associated risks. Overall, we concluded that exposure
via indoor routes can continue long after home construction and
the enactment of restrictions on new uses of the compounds, e.g., exposure
to legacy PCBs, OCPs. We also concluded that other current
indoor sources could be targeted for reduction, e.g., PAH from home
H. Demirtepe, et al. Environment International 127 (2019) 653–663
660
heating by wood burning. Further, industrial manufacturing processes
leading to inadvertent production of PCB, especially CB-11, should be
adapted and improved to reduce PCB emissions from new commercial
products. Finally, attention should be given to SVOCs that had high
human intake but had no reliable TRVs, e.g., TCIPP, TPHP, TIBP, acenaphthylene,
EHDPP, retene, TPrP, and chrysene, or had TRVs that may
underestimate their toxicity, e.g. TDCIPP and TCEP. A “next step”
would be to consider the mixture effects of the wide range of SVOCs
with significant human exposure in residential indoor environments.
Acknowledgements
This research was supported by the RECETOX Research
Infrastructure (LM2015051 and CZ.02.1.01/0.0/0.0/16_013/
0001761), the Czech Ministry of Education, Youth and Sports (NPU
RECETOX – LO1214), the European Union's Horizon 2020 research and
innovation programme under the Marie Skłodowska-Curie GA No
734522 (INTERWASTE), the Slovak Ministry of Health (2012/47-SZU-
11) and in part by Center of Excellence of Environmental Health, ITMS
No. 26240120033, based on the supporting operational Research and
Development Program financed from the European Regional
Development Fund. The research is associated with the European
Union's Horizon 2020 research and innovation programme grant
agreement GA No 733032 (HBM4EU). We thank all participants for
permitting sampling in their residences.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.envint.2019.04.001.
References
Abdallah, M.A.E., Bressi, M., Oluseyi, T., Harrad, S., 2016. Hexabromocyclododecane and
tetrabromobisphenol-A in indoor dust from France, Kazakhstan and Nigeria: implications
for human exposure. Emerg. Contam. 2, 73–79. https://doi.org/10.1016/j.
emcon.2016.03.006.
Ali, N., Van Den Eede, N., Dirtu, A.C., Neels, H., Covaci, A., 2012. Assessment of human
exposure to indoor organic contaminants via dust ingestion in Pakistan. Indoor Air
22, 200–211. https://doi.org/10.1111/j.1600-0668.2011.00757.x.
Ali, N., Ali, L., Mehdi, T., Dirtu, A.C., Al-Shammari, F., Neels, H., Covaci, A., 2013. Levels
and profiles of organochlorines and flame retardants in car and house dust from
Kuwait and Pakistan: implication for human exposure via dust ingestion. Environ. Int.
55, 62–70. https://doi.org/10.1016/j.envint.2013.02.001.
Ampleman, M.D., Martinez, A., DeWall, J., Rawn, D.F.K., Hornbuckle, K.C., Thorne, P.S.,
2015. Inhalation and dietary exposure to PCBs in urban and rural cohorts via congener-specific
measurements. Environ. Sci. Technol. 49, 1156–1164. https://doi.org/
10.1021/es5048039.
Araki, A., Saito, I., Kanazawa, A., Morimoto, K., Nakayama, K., Shibata, E., Tanaka, M.,
Takigawa, T., Yoshimura, T., Chikara, H., Saijo, Y., Kishi, R., 2014. Phosphorus flame
retardants in indoor dust and their relation to asthma and allergies of inhabitants.
Indoor Air 24, 3–15. https://doi.org/10.1111/ina.12054.
ATSDR, 2012. Toxicological Profile for Phosphate Ester Flame Retardants. https://doi.
org/10.1007/s00244-006-0094-3.
ATSDR, 2018. Agency for Toxic Substances & Disease Registry Minimal Risk Levels
(MRLs) for Hazardous Substances [WWW Document]. URL. https://www.atsdr.cdc.
gov/mrls/mrllist.asp, Accessed date: 27 June 2018.
Audy, O., Melymuk, L., Venier, M., Vojta, S., Becanova, J., Romanak, K., Vykoukalova,
M., Prokes, R., Kukucka, P., Diamond, M.L., Klanova, J., 2018. PCBs and organochlorine
pesticides in indoor environments - a comparison of indoor contamination in
Canada and Czech Republic. Chemosphere 206, 622–631. https://doi.org/10.1016/j.
chemosphere.2018.05.016.
Baars, A.J., Theelen, R.M.C., Janssen, P.J.C.M., Hesse, J.M., van Apeldoorn, M.E.,
Meijerink, M.C.M., Verdam, L., Zeilmaker, M.J., 2001. Re-evaluation of human-toxicological
maximum permissible risk levels, RIVM report 711701 025. doi:https://
doi.org/10.1051/vetres:2008046.
Behl, M., Hsieh, J.H., Shafer, T.J., Mundy, W.R., Rice, J.R., Boyd, W.A., Freedman, J.H.,
Hunter, E.S., Jarema, K.A., Padilla, S., Tice, R.R., 2015. Use of alternative assays to
identify and prioritize organophosphorus flame retardants for potential developmental
and neurotoxicity. Neurotoxicol. Teratol. 52, 181–193. https://doi.org/10.
1016/j.ntt.2015.09.003.
Behl, M., Rice, J.R., Smith, M.V., Co, C.A., Bridge, M.F., Hsieh, J.-H., Freedman, J.H.,
Boyd, W.A., 2016. Comparative toxicity of organophosphate flame retardants and
polybrominated diphenyl ethers to Caenorhabditis elegans. Toxicol. Sci. 154 (2),
241–252.
Bergman, A., Rydén, A., Law, R.J., de Boer, J., Covaci, A., Alaee, M., Birnbaum, L.S.,
Petreas, M.X., Rose, M., Sakai, S., Van den Eede, N., van der Veen, I., 2012. A novel
abbreviation standard for organobromine, organochlorine and organophosphorus
flame retardants and some characteristics of the chemicals. Environ. Int. 49, 57–82.
Bonvallot, N., Mandin, C., Mercier, F., Le Bot, B., Glorennec, P., 2010. Health ranking of
ingested semi-volatile organic compounds in house dust: an application to France.
Indoor Air 20, 458–472. https://doi.org/10.1111/j.1600-0668.2010.00667.x.
Booij, P., Holoubek, I., Klánová, J., Kohoutek, J., Dvorská, A., Magulová, K., Al-Zadjali, S.,
Čupr, P., 2016. Current implications of past DDT indoor spraying in Oman. Sci. Total
Environ. 550, 231–240. https://doi.org/10.1016/j.scitotenv.2015.12.044.
Breivik, K., Sweetman, A., Pacyna, J.M., Jones, K.C., 2002. Towards a global historical
emission inventory for selected PCB congeners - a mass balance approach: 1. Global
production and consumption. Sci. Total Environ. 290, 181–198. https://doi.org/10.
1016/S0048-9697(01)01075-0.
Breivik, K., Sweetman, A., Pacyna, J.M., Jones, K.C., 2007. Towards a global historical
emission inventory for selected PCB congeners–a mass balance approach 3. An update.
Sci. Total Environ. 377, 296–307. https://doi.org/10.1016/j.scitotenv.2007.02.
026.
Carignan, C.C., Mínguez-Alarcón, L., Williams, P.L., Meeker, J.D., Stapleton, H.M., Butt,
C.M., Toth, T.L., Ford, J.B., Hauser, R., 2018. Paternal urinary concentrations of
organophosphate flame retardant metabolites, fertility measures, and pregnancy
outcomes among couples undergoing in vitro fertilization. Environ. Int. 111,
232–238. https://doi.org/10.1016/j.envint.2017.12.005.
Castorina, R., Bradman, A., Stapleton, H.M., Butt, C.M., Avery, D., Harley, K.G., Gunier,
R.B., Holland, N., Eskenazi, B., 2017. Current-use flame retardants: maternal exposure
and neurodevelopment in children of the CHAMACOS cohort. Chemosphere
189, 574–580. https://doi.org/10.1016/j.chemosphere.2017.09.037.
Cequier, E., Ionas, A.C., Covaci, A., Marce, R.M., Becher, G., Thomsen, C., 2014.
Occurrence of a broad range of legacy and emerging flame retardants in indoor environments
in Norway. Environ. Sci. Technol. 48, 6827–6835. https://doi.org/10.
1021/es500516u.
Dirtu, A.C., Ali, N., Van den Eede, N., Neels, H., Covaci, A., 2012. Country specific
comparison for profile of chlorinated, brominated and phosphate organic contaminants
in indoor dust. Case study for Eastern Romania, 2010. Environ. Int. 49,
1–8. https://doi.org/10.1016/j.envint.2012.08.002.
Dodson, R.E., Perovich, L.J., Covaci, A., Van Den Eede, N., Ionas, A.C., Dirtu, A.C., Brody,
J.G., Rudel, R.A., 2012. After the PBDE phase-out: a broad suite of flame retardants in
repeat house dust samples from California. Environ. Sci. Technol. 46, 13056–13066.
https://doi.org/10.1021/es303879n.
Drobná, B., Fabišiková, A., Čonka, K., Chovancová, J., Dömötörová, M., Wimmerová, S.,
Šovčíková, E., Kočan, A., 2011. Differences between dioxin-like PCB, non-dioxin-like
PCB, polychlorinated dibenzo-p-dioxin and dibenzofuran intake from human milk
and infant milk formula by infants in the Michalovce district (Slovakia). J. Food Nutr.
Res. 50, 106–117.
ECHA, 2018. European Chemicals Agency Information on Chemicals. [WWW Document].
URL. https://echa.europa.eu/information-on-chemicals, Accessed date: 29 June
2018.
European Environment Agency, 2012. Corine Land Cover 2012 [WWW Document].
Francisco, A.P., Harner, T., Eng, A., 2017. Measurement of polyurethane foam – air
partition coefficients for semivolatile organic compounds as a function of temperature:
application to passive air sampler monitoring. Chemosphere 174, 638–642.
https://doi.org/10.1016/j.chemosphere.2017.01.135.
Frederiksen, M., Meyer, H.W., Ebbehøj, N.E., Gunnarsen, L., 2012. Polychlorinated biphenyls
(PCBs) in indoor air originating from sealants in contaminated and uncontaminated
apartments within the same housing estate. Chemosphere 89, 473–479.
Fromme, H., Lahrz, T., Piloty, M., Gebhardt, H., Oddoy, A., Rüden, H., 2004. Polycyclic
aromatic hydrocarbons inside and outside of apartments in an urban area. Sci. Total
Environ. 326, 143–149. https://doi.org/10.1016/j.scitotenv.2004.02.002.
García, M., Rodríguez, I., Cela, R., 2007. Microwave-assisted extraction of organophosphate
flame retardants and plasticizers from indoor dust samples. J. Chromatogr. A
1152, 280–286. https://doi.org/10.1016/j.chroma.2006.11.046.
Gioia, R., Steinnes, E., Thomas, G.O., Mejier, S.N., Jones, K.C., 2006. Persistent organic
pollutants in European background air: derivation of temporal and latitudinal trends.
J. Environ. Monit. 8, 700–710. https://doi.org/10.1039/b604821h.
Gustafson, P., Östman, C., Sällsten, G., 2008. Indoor levels of polycyclic aromatic hydrocarbons
in homes with or without wood burning for heating. Environ. Sci.
Technol. 42, 5074–5080. https://doi.org/10.1021/es800304y.
Harner, T., Shoeib, M., Diamond, M., Stern, G., Rosenberg, B., 2004. Using passive air
samplers to assess urban-rural trends for persistent organic pollutants. 1.
Polychlorinated biphenyls and organochlorine pesticides. Environ. Sci. Technol. 38,
4474–4483. https://doi.org/10.1021/es040302r.
Harrad, S., Abdallah, M.A.E., Covaci, A., 2009a. Causes of variability in concentrations
and diastereomer patterns of hexabromocyclododecanes in indoor dust. Environ. Int.
35, 573–579. https://doi.org/10.1016/j.envint.2008.10.005.
Harrad, S., Ibarra, C., Robson, M., Melymuk, L., Zhang, X., Diamond, M., Douwes, J.,
2009b. Polychlorinated biphenyls in domestic dust from Canada, New Zealand,
United Kingdom and United States: implications for human exposure. Chemosphere
76, 232–238. https://doi.org/10.1016/j.chemosphere.2009.03.020.
Hazrati, S., Harrad, S.J., 2006. Causes of variability in concentrations of polychlorinated
biphenyls and polybrominated diphenyl ethers in indoor air. Environ. Sci. Technol.
40, 7584–7589.
Herkert, N.J., Jahnke, J.C., Hornbuckle, K.C., 2018. Emissions of Tetrachlorobiphenyls
(PCBs 47, 51, and 68) from polymer resin on kitchen cabinets as a non-Aroclor source
to residential air. Environ. Sci. Technol. 52, 5154–5160. https://doi.org/10.1021/
acs.est.8b00966.
Hertz-Picciotto, I., Trnovec, T., Kočan, A., Charles, M.J., Čižnar, P., Langer, P., Sovčikova,
H. Demirtepe, et al. Environment International 127 (2019) 653–663
661
E., James, R., 2003. PCBs and early childhood development in Slovakia: study design
and background. Fresenius Environ. Bull. 12, 208–214.
Hiller, E., Zemanová, L., Sirotiak, M., Jurkovič, L., 2011. Concentrations, distributions,
and sources of polychlorinated biphenyls and polycyclic aromatic hydrocarbons in
bed sediments of the water reservoirs in Slovakia. Environ. Monit. Assess. 173,
883–897. https://doi.org/10.1007/s10661-010-1431-6.
Hoffman, K., Lorenzo, A., Butt, C.M., Hammel, S.C., Henderson, B.B., Roman, S.A., Scheri,
R.P., Stapleton, H.M., Sosa, J.A., 2017. Exposure to flame retardant chemicals and
occurrence and severity of papillary thyroid cancer: a case-control study. Environ.
Int. 107, 235–242. https://doi.org/10.1016/j.envint.2017.06.021.
Holoubek, I., Adamec, V., Bartoš, M., Bláha, K., Bláha, L., Budňáková, M., Černa, M.,
Čupr, P., Demnerová, K., Drápal, J., Hajšlová, J., Hanzálková, M., Holoubková, I.,
Hrabětová, S., Jech, L., Klánová, J., Kohoutek, J., Kužílek, V., Machálek, P., Matějů,
V., Matoušek, J., Matoušek, M., Mestřík, V., Novák, J., Ocelka, T., Pekárek, V.,
Petrlík, J., Petira, O., Provazník, K., Punčochář, M., Rieder, M., Ruprich, J., Sáňka,
M., Tomaniová, M., Vácha, R., Volka, K., Zbíral, J., 2006. The National
Implementation Plan for Implementation of the Stockholm Convention in the Czech
Republic [WWW Document].
Holt, E., Audy, O., Booij, P., Melymuk, L., Prokes, R., Klánová, J., 2017. Organochlorine
pesticides in the indoor air of a theatre and museum in the Czech Republic: inhalation
exposure and cancer risk. Sci. Total Environ. 609, 598–606. https://doi.org/10.1016/
j.scitotenv.2017.07.203.
Hovander, L., Linderholm, L., Athanasiadou, M., Athanassiadis, I., Bignert, A., Fängström,
B., Kocan, A., Petrik, J., Trnovec, T., Bergman, A., 2006. Levels of PCBs and their
metabolites in the serum of residents of a highly contaminated area in Eastern
Slovakia. Environ. Sci. Technol. 40, 3696–3703. https://doi.org/10.1021/
es0525657.
Hu, D., Hornbuckle, K.C., 2010. Inadvertent Polychlorinated Biphenyls in Commercial
Paint Pigments 44. pp. 2822–2827. https://doi.org/10.1021/es902413k.
Hung, H., Katsoyiannis, A.A., Brorström-Lundén, E., Olafsdottir, K., Aas, W., Breivik, K.,
Bohlin-Nizzetto, P., Sigurdsson, A., Hakola, H., Bossi, R., Skov, H., Sverko, E., Barresi,
E., Fellin, P., Wilson, S., 2016. Temporal trends of persistent organic pollutants
(POPs) in Arctic air: 20 years of monitoring under the Arctic monitoring and assessment
Programme (AMAP). Environ. Pollut. 217, 52–61. https://doi.org/10.1016/
j.envpol.2016.01.079.
Hutter, H.-P., Haluza, D., Piegler, K., Hohenblum, P., Fröhlich, M., Scharf, S., Uhl, M.,
Damberger, B., Tappler, P., Kundi, M., Wallner, P., Moshammer, H., 2013.
Semivolatile compounds in schools and their influence on cognitive performance of
children. Int. J. Occup. Med. Environ. Health 26, 628–635. https://doi.org/10.2478/
s13382-013-0125-z.
Hwang, H.M., Park, E.K., Young, T.M., Hammock, B.D., 2008. Occurrence of endocrinedisrupting
chemicals in indoor dust. Sci. Total Environ. 404, 26–35. https://doi.org/
10.1016/j.scitotenv.2008.05.031.
IARC, 2010. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans: Some
Non-heterocyclic Polycyclic Aromatic Hydrocarbons and some Related Exposures,
IARC Monographs on the Evaluation of Carcinogenic Risks to Humans. (Lyon,
France).
IARC, 2016. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans:
Polychlorinated Biphenyls and Polybrominated Biphenyls. Lyon, France.
Jarema, K.A., Hunter, D.L., Shaffer, R.M., Behl, M., Padilla, S., 2015. Acute and developmental
behavioral effects of flame retardants and related chemicals in zebrafish.
Neurotoxicol. Teratol. 52, 194–209. https://doi.org/10.1016/j.ntt.2015.08.010.
Jílková, S., Melymuk, L., Vojta, Š., Vykoukalová, M., Bohlin-Nizzetto, P., Klánová, J.,
2018. Small scale spatial variability of flame retardants in indoor dust and implications
for dust sampling. Chemosph 206, 132–141. https://doi.org/10.1016/j.
chemosphere.2018.04.146.
Jursa, S., Chovancová, J., Petrík, J., Lokša, J., 2006. Dioxin-like and non-dioxin-like PCBs
in human serum of Slovak population. Chemosphere 64, 686–691. https://doi.org/
10.1016/j.chemosphere.2005.10.048.
Kanazawa, A., Saito, I., Araki, A., Takeda, M., Ma, M., Saijo, Y., Kishi, R., 2010.
Association between indoor exposure to semi-volatile organic compounds and
building-related symptoms among the occupants of residential dwellings. Indoor Air
20, 72–84. https://doi.org/10.1111/j.1600-0668.2009.00629.x.
Kocan, A., Petrik, J., Jursa, S., Chovancova, J., Drobna, B., 2001. Environmental contamination
with polychlorinated biphenyls in the area of their former manufacture in
Slovakia. Chemosphere 43, 595–600. https://doi.org/10.1016/S0045-6535(00)
00411-2.
Kohler, M., Tremp, J., Zennegg, M., Seiler, C., Minder-Kohler, S., Beck, M., Lienemann, P.,
Wegmann, L., Schmid, P., 2005. Joint sealants: an overlooked diffuse source of
polychlorinated biphenyls in buildings. Environ. Sci. Technol. 39, 1967–1973.
https://doi.org/10.1021/es048632z.
Lanxess, 2016. DP-45 - Halogenated Flame Retardant [WWW Document].
Liu, X., Guo, Z., Krebs, K.A., Greenwell, D.J., Roache, N.F., Stinson, R.A., Nardin, J.A.,
Pope, R.H., 2016. Laboratory study of PCB transport from primary sources to building
materials. Indoor Built Environ 25, 635–650. https://doi.org/10.1177/
1420326X15623355.
Lucattini, L., Poma, G., Covaci, A., de Boer, J., Lamoree, M.H., Leonards, P.E.G., 2018. A
review of semi-volatile organic compounds (SVOCs) in the indoor environment: occurrence
in consumer products, indoor air and dust. Chemosphere 201, 466–482.
https://doi.org/10.1016/j.chemosphere.2018.02.161.
Lv, J., Xu, R., Wu, G., Zhang, Q., Li, Y., Wang, P., Liao, C., Liu, J., Jiang, G., Wei, F., 2009.
Indoor and outdoor air pollution of polycyclic aromatic hydrocarbons (PAHs) in
Xuanwei and Fuyuan, China. J. Environ. Monit. 11, 1368. https://doi.org/10.1039/
b900382g.
Ma, Y., Venier, M., Hites, R.A., 2012. 2-Ethylhexyl tetrabromobenzoate and Bis(2-ethylhexyl)
tetrabromophthalate flame retardants in the great lakes atmosphere. Environ.
Sci. Technol. 46, 204–208. https://doi.org/10.1021/es203251f.
Maertens, R.M., Yang, X., Zhu, J., Gagne, R.W., Douglas, G.R., White, P.A., 2008.
Mutagenic and carcinogenic hazards of settled house dust I: polycyclic aromatic
hydrocarbon content and excess lifetime cancer risk from preschool exposure.
Environ. Sci. Technol. 42, 1747–1753. https://doi.org/10.1021/es702449c.
Marek, R.F., Thorne, P.S., Herkert, N.J., Awad, A.M., Hornbuckle, K.C., 2017. Airborne
PCBs and OH-PCBs inside and outside urban and rural U.S. schools. Environ. Sci.
Technol. 51, 7853–7860. https://doi.org/10.1021/acs.est.7b01910.
Marklund, A., Andersson, B., Haglund, P., 2003. Screening of organophosphorus compounds
and their distribution in various indoor environments. Chemosphere 53,
1137–1146. https://doi.org/10.1016/S0045-6535(03)00666-0.
Meeker, J.D., Stapleton, H.M., 2010. House dust concentrations of organophosphate
flame retardants in relation to hormone levels and semen quality parameters.
Environ. Health Perspect. 118, 318–323. https://doi.org/10.1289/ehp.0901332.
Meeker, J.D., Cooper, E.M., Stapleton, H.M., Hauser, R., 2013. Exploratory analysis of
urinary metabolites of phosphorus-containing flame retardants in relation to markers
of male reproductive health. Endocr. Disruptors 1, e26306. https://doi.org/10.4161/
endo.26306.
Melymuk, L., Bohlin-Nizzetto, P., Kukučka, P., Vojta, Š., Kalina, J., Čupr, P., Klánová, J.,
2016a. Seasonality and indoor/outdoor relationships of flame retardants and PCBs in
residential air. Environ. Pollut. 218, 392–401. https://doi.org/10.1016/J.ENVPOL.
2016.07.018.
Melymuk, L., Bohlin-Nizzetto, P., Vojta, Š., Krátká, M., Kukučka, P., Audy, O., Přibylová,
P., Klánová, J., 2016b. Distribution of legacy and emerging semivolatile organic
compounds in five indoor matrices in a residential environment. Chemosphere 153,
179–186. https://doi.org/10.1016/j.chemosphere.2016.03.012.
Mitro, S.D., Dodson, R.E., Singla, V., Adamkiewicz, G., Elmi, A.F., Tilly, M.K., Zota, A.R.,
2016. Consumer product chemicals in indoor dust: a quantitative meta-analysis of
U.S. studies. Environ. Sci. Technol. 50, 10661–10672. https://doi.org/10.1021/acs.
est.6b02023.
National Research Council, 2000. Toxicological Risks of Selected Flame-Retardant
Chemicals. National Academy Press, Washington, DC.
Noyes, P.D., Haggard, D.E., Gonnerman, G.D., Tanguay, R.L., 2015. Advanced morphological
- behavioral test platform reveals neurodevelopmental defects in embryonic
zebrafish exposed to comprehensive suite of halogenated and organophosphate flame
retardants. Toxicol. Sci. 145, 177–195. https://doi.org/10.1093/toxsci/kfv044.
Okeme, J.O., Webster, E.M., Parnis, J.M., Diamond, M.L., 2017. Approaches for estimating
PUF-air partitions coefficient for semi-volatile organic compounds: a critical
comparison. Chemosphere 168, 199–204. https://doi.org/10.1016/j.chemosphere.
2016.10.001.
Okeme, J.O., Yang, C., Abdollahi, A., Dhal, S., Harris, S.A., Jantunen, L.M., Tsirlin, D.,
Diamond, M.L., 2018. Passive air sampling of flame retardants and plasticizers in
Canadian homes using PDMS, XAD-coated PDMS and PUF samplers. Environ. Pollut.
239, 109–117. https://doi.org/10.1016/j.envpol.2018.03.103.
Pelletier, M., Glorennec, P., Mandin, C., Le Bot, B., Ramalho, O., Mercier, F., Bonvallot,
N., 2018. Chemical-by-chemical and cumulative risk assessment of residential indoor
exposure to semivolatile organic compounds in France. Environ. Int. 117, 22–32.
https://doi.org/10.1016/j.envint.2018.04.024.
Petrik, J., Drobna, B., Pavuk, M., Jursa, S., Wimmerova, S., Chovancova, J., 2006. Serum
PCBs and organochlorine pesticides in Slovakia: age, gender, and residence as determinants
of organochlorine concentrations. Chemosphere 65, 410–418. https://doi.
org/10.1016/j.chemosphere.2006.02.002.
Rodenburg, L.A., Guo, J., Du, S., Cavallo, G.J., 2010. Evidence for unique and ubiquitous
environmental sources of 3,3′-dichlorobiphenyl (PCB 11). Environ. Sci. Technol. 44,
2816–2821. https://doi.org/10.1021/es901155h.
Shin, H.M., McKone, T.E., Bennett, D.H., 2012. Intake fraction for the indoor environment:
a tool for prioritizing indoor chemical sources. Environ. Sci. Technol. 46,
10063–10072. https://doi.org/10.1021/es3018286.
Shin, H.-M., McKone, T.E., Tulve, N.S., Clifton, M.S., Bennett, D.H., 2013. Indoor residence
times of semivolatile organic compounds: model estimation and field evaluation.
Environ. Sci. Technol. 47, 859–867.
Smith, M.N., Grice, J., Cullen, A., Faustman, E.M., 2016. A toxicological framework for
the prioritization of children's safe product act data. Int. J. Environ. Res. Public
Health 13. https://doi.org/10.3390/ijerph13040431.
Stapleton, H.M., Allen, J.A., Kelly, S.M., Konstantinov, A., Klosterhaus, S., Watkins, D.,
Mcclean, M.D., Webster, T.F., 2008. Alternate and new brominated flame retardants
detected in U. S. House dust. Environ. Sci. Technol. 42, 6910–6916. https://doi.org/
10.1021/es801070p.
Stockholm Convention, 2018. Stockholm Convention on Persistent Organic Pollutants
[WWW Document]. URL. http://chm.pops.int/, Accessed date: 10 January 2018.
Strémy, M., Šutová, Z., Murínová, Ľ.P., Richterová, D., Wimmerová, S., Čonka, K.,
Drobná, B., Fábelová, L., Jurečková, D., Jusko, T.A., Tihányi, J., Trnovec, T., 2019.
The spatial distribution of congener-specific human PCB concentrations in a PCBpolluted
region. Sci. Total Environ. 651, 2292–2303. https://doi.org/10.1016/j.
scitotenv.2018.10.123.
Sühring, R., Diamond, M.L., Scheringer, M., Wong, F., Pućko, M., Stern, G., Burt, A.,
Hung, H., Fellin, P., Li, H., Jantunen, L.M., 2016. Organophosphate esters in
Canadian Arctic air: occurrence, levels and trends. Environ. Sci. Technol. 50,
7409–7415. https://doi.org/10.1021/acs.est.6b00365.
Tan, H., Peng, C., Guo, Y., Wang, X., Wu, Y., Chen, D., 2017. Organophosphate flame
retardants in house dust from South China and related human exposure risks. Bull.
Environ. Contam. Toxicol. 99, 344–349. https://doi.org/10.1007/s00128-017-
2120-8.
Taniyasu, S., Kannan, K., Holoubek, I., Ansorgova, A., Horii, Y., Hanari, N., Yamashita, N.,
Aldous, K.M., 2003. Isomer-specific analysis of chlorinated biphenyls, naphthalenes
and dibenzofurans in Delor: polychlorinated biphenyl preparations from the former
H. Demirtepe, et al. Environment International 127 (2019) 653–663
662
Czechoslovakia. Environ. Pollut. 126, 169–178. https://doi.org/10.1016/S0269-
7491(03)00207-0.
Tao, F., Abdallah, M.A.-E., Harrad, S., 2016. Emerging and legacy flame retardants in UK
indoor air and dust: evidence for replacement of PBDEs by emerging flame retardants?
Environ. Sci. Technol. 50, 13052–13061. https://doi.org/10.1021/acs.est.
6b02816.
U.S. EPA, 2012. Literature Review of Remediation Methods for PCBs in Buildings.
US Environmental Protection Agency, 2018. US Environmental Protection Agency
Chemistry Dashboard. (WWW Document).
U.S. EPA, 2018. US Environmental Protection Agency Chemistry Dashboard, [WWW
Document]. [WWW Document]. https://comptox.epa.gov/dashboard, Accessed
date: 6 January 2018.
Van Den Eede, N., Dirtu, A.C., Ali, N., Neels, H., Covaci, A., 2012. Multi-residue method
for the determination of brominated and organophosphate flame retardants in indoor
dust. Talanta 89, 292–300. https://doi.org/10.1016/j.talanta.2011.12.031.
Venier, M., Audy, O., Vojta, Š., Bečanová, J., Romanak, K., Melymuk, L., Krátká, M.,
Kukučka, P., Okeme, J., Saini, A., Diamond, M.L., Klánová, J., 2016. Brominated
flame retardants in the indoor environment - comparative study of indoor contamination
from three countries. Environ. Int. 94, 150–160. https://doi.org/10.
1016/j.envint.2016.04.029.
Vojta, Š., Melymuk, L., Klánová, J., 2017. Changes in flame retardant and legacy contaminant
concentrations in indoor air during building construction, furnishing, and
use. Environ. Sci. Technol. 51, 11891–11899. https://doi.org/10.1021/acs.est.
7b03245.
Vorkamp, K., 2016. An overlooked environmental issue? A review of the inadvertent
formation of PCB-11 and other PCB congeners and their occurrence in consumer
products and in the environment. Sci. Total Environ. 541, 1463–1476. https://doi.
org/10.1016/j.scitotenv.2015.10.019.
Vykoukalová, M., Venier, M., Vojta, Š., Melymuk, L., Bečanová, J., Romanak, K., Prokeš,
R., Okeme, J.O., Saini, A., Diamond, M.L., Klánová, J., 2017. Organophosphate esters
flame retardants in the indoor environment. Environ. Int. 106, 97–104. https://doi.
org/10.1016/j.envint.2017.05.020.
Wang, J., Guo, L., Li, J., Zhang, G., Lee, C.S.L., Li, X., Jones, K.C., Xiang, Y., Zhong, L.,
2007. Passive air sampling of DDT, chlordane and HCB in the Pearl River Delta, South
China: implications to regional sources. J. Environ. Monit. 9, 582–588. https://doi.
org/10.1039/b700798a.
Wang, W., Wu, F., Zheng, J., Wong, M.H., 2013. Risk assessments of PAHs and Hg exposure
via settled house dust and street dust, linking with their correlations in human
hair. J. Hazard. Mater. 263, 627–637. https://doi.org/10.1016/j.jhazmat.2013.10.
023.
Weis, N., Kohler, M., Zorn, C., 2003. Highly PCB-contaminated schools due to PCB-containing
roughcast. In: Health Buildings 2003, Proceedings of the 7th International
Conference on Healthy Buildings, International Society of Indoor Air Quality and
Climate (ISIAQ). Singapore, December 7–11 2003, pp. 283–288.
Weschler, C.J., 2009. Changes in indoor pollutants since the 1950s. Atmos. Environ. 43,
153–169. https://doi.org/10.1016/j.atmosenv.2008.09.044.
Weschler, C.J., Nazaroff, W.W., 2008. Semivolatile organic compounds in indoor environments.
Atmos. Environ. 42, 9018–9040. https://doi.org/10.1016/j.atmosenv.
2008.09.052.
Whitehead, T.P., Metayer, C., Petreas, M., Does, M., Buffler, P.A., Rappaport, S.M., 2013.
Polycyclic aromatic hydrocarbons in residential dust: sources of variability. Environ.
Health Perspect. 121, 543–550. https://doi.org/10.1289/ehp.1205821.
Wimmerová, S., Watson, A., Drobná, B., Šovčíková, E., Weber, R., Lancz, K., Patayová, H.,
Richterová, D., Koštiaková, V., Jurečková, D., Závacký, P., Strémy, M., Jusko, T.A.,
Palkovičová Murínová, Ľ., Hertz-Picciotto, I., Trnovec, T., 2015. The spatial distribution
of human exposure to PCBs around a former production site in Slovakia.
Environ. Sci. Pollut. Res. 22, 14405–14415. https://doi.org/10.1007/s11356-015-
5047-9.
World Bank, 2018a. World Bank Country and Lending Groups [WWW Document]. URL.
https://datahelpdesk.worldbank.org/knowledgebase/articles/906519-world-bankcountry-and-lending-groups,
Accessed date: 1 May 2018.
World Bank, 2018b. Agricultural Value Added (% of GDP) [WWW Document]. URL.
https://data.worldbank.org/indicator/NV.AGR.TOTL.ZS, Accessed date: 1 May
2018.
Yadav, I.C., Devi, N.L., Li, J., Zhang, G., 2018. Polycyclic aromatic hydrocarbons in house
dust and surface soil in major urban regions of Nepal: implication on source apportionment
and toxicological effect. Sci. Total Environ. 616–617, 223–235. https://doi.
org/10.1016/j.scitotenv.2017.10.313.
Yang, C., Harris, S.A., Jantunen, L.M., Siddique, S., Kubwabo, C., Tsirlin, D., Latifovic, L.,
Fraser, B., St-Jean, M., De La Campa, R., You, H., Kulka, R., Diamond, M.L., 2019. Are
cell phones an indicator of personal exposure to organophosphate flame retardants
and plasticizers? Environ. Int. 122, 104–116. https://doi.org/10.1016/j.envint.2018.
10.021.
Zhang, X., Diamond, M.L., Ibarra, C., Harrad, S., 2009. Multimedia modeling of polybrominated
diphenyl ether emissions and fate indoors. Environ. Sci. Technol. 43,
2845–2850. https://doi.org/10.1021/es802172a.
Zhang, X., Diamond, M.L., Robson, M., Harrad, S., 2011. Sources, emissions, and fate of
polybrominated diphenyl ethers and polychlorinated biphenyls indoors in Toronto,
Canada. Environ. Sci. Technol. 45, 3268–3274. https://doi.org/10.1021/es102767g.
Zhang, X., Sühring, R., Serodio, D., Bonnell, M., Sundin, N., Diamond, M.L., 2016. Novel
flame retardants: estimating the physical-chemical properties and environmental fate
of 94 halogenated and organophosphate PBDE replacements. Chemosphere 144,
2401–2407. https://doi.org/10.1016/j.chemosphere.2015.11.017.
Zheng, X., Qiao, L., Covaci, A., Sun, R., Guo, H., Zheng, J., Luo, X., Xie, Q., Mai, B., 2017.
Brominated and phosphate flame retardants (FRs) in indoor dust from different microenvironments:
implications for human exposure via dust ingestion and dermal
contact. Chemosphere 184, 185–191. https://doi.org/10.1016/j.chemosphere.2017.
05.167.
H. Demirtepe, et al. Environment International 127 (2019) 653–663
663
Holt, Eva, Ondřej Audy, Petra Booij, Lisa Melymuk, Roman Prokes, and Jana Klánová. 2017.
"Organochlorine Pesticides in the Indoor Air of a Theatre and Museum in the Czech
Republic: Inhalation Exposure and Cancer Risk." Science of The Total Environment 609: 598-
606. https://doi.org/10.1016/j.scitotenv.2017.07.203
Organochlorine pesticides in the indoor air of a theatre and museum in
the Czech Republic: Inhalation exposure and cancer risk
Eva Holt ⁎, Ondřej Audy, Petra Booij, Lisa Melymuk, Roman Prokes, Jana Klánová
Research Centre for Toxic Compounds in the Environment (RECETOX), Kamenice 753/5, 62500 Brno, Czech Republic
H I G H L I G H T S
• Indoor air in buildings of cultural and historical
importance had high OCP levels
• Cancer risk (CR) from inhalation exposure
to OCPs determined for workers
• CR from DDE exposure reduced by orders
of magnitude after theatre restoration
• CR for museum workers from γ-HCH
exposure were significant as to require
mitigation
• Museum worker CR was reduced, but
still moderate where γ-HCH granules
were removed
G R A P H I C A L A B S T R A C T
a b s t r a c ta r t i c l e i n f o
Article history:
Received 18 April 2017
Received in revised form 22 July 2017
Accepted 23 July 2017
Available online 27 July 2017
Editor: Adrian Covaci
Organochlorine pesticides (OCPs) have been used to preserve the integrity of historical buildings or to protect
collections of artefacts at potentially large volumes and often without detailed application records. Previous research
has focused on the efficiency of remediation at contaminated sites (where identified), as well as improvement
of preservation techniques and workplace health and safety. Few studies have assessed the human health
risks from occupational exposure to OCPs in buildings of cultural and historical importance. Thus, potential risks
may remain unidentified. In the present study, OCPs in indoor air were measured in a baroque theatre and a natural
history museum in the Czech Republic, both of which had suspected past indoor application. In the theatre
attic p,p′-dichlorodiphenyldichloroethylene (p,p′-DDE) levels in air were up to 190 ng m−3
, confirming past indoor
use of p,p′-dichlorodiphenyltrichloroethane (p,p′-DDT). There was also evidence of γ-hexachlorocyclohexane
(γ-HCH) use in the theatre (max γ-HCH in air of 56 ng m−3
). Yet, the cancer risk (CR) from occupational
exposure via inhalation (Expi) to OCPs in the theatre was low (CR b 4.0 × 10−6
). γ-HCH was found at elevated
levels in air of the museum (max γ-HCH in air of 15,000 ng m−3
). CR from Expi in the museum was moderate
to high (N1 × 10−4
). Our results show the CR through Expi to OCPs in buildings, such as museums can still be significant
enough to warrant mitigation measures, e.g., remediation.
© 2017 Elsevier B.V. All rights reserved.
Keywords:
Organochlorine pesticide
Inhalation exposure
Human health (cancer) risk
Remediation
1. Introduction
Inorganic and organic pesticides are regularly used as a protectant
for wood used in building materials, e.g., impregnated into wood or
used as a coating for wooden building materials (Unger et al., 2001).
In the past, these wood protectant pesticides included organochlorine
pesticides (OCPs) (e.g., γ-hexachlorocyclohexane (γ-HCH)), pentachlorophenol,
dichlorodiphenyltrichloroethane (DDT) (Unger et al., 2001),
most of which have since been replaced by other pesticides (e.g., synthetic
pyrethroids) (Unger et al., 2001). The use of pesticides was also
common practice amongst collectors of artefacts, biological specimens,
Science of the Total Environment 609 (2017) 598–606
⁎ Corresponding author.
E-mail address: holt@recetox.muni.cz (E. Holt).
http://dx.doi.org/10.1016/j.scitotenv.2017.07.203
0048-9697/© 2017 Elsevier B.V. All rights reserved.
Contents lists available at ScienceDirect
Science of the Total Environment
journal homepage: www.elsevier.com/locate/scitotenv
etc. to prevent damage from pests, such as insects or fungi (Palmer et al.,
2003; Unger et al., 2001). A range of pesticides has been detected in artefacts
or significant artistic, cultural, historical, or scientific objects, as
well as in the buildings (e.g., museums, herbariums) that store such
items (Fellowes et al., 2011; Goldberg, 1996; Marcotte et al., 2014;
Musshoff et al., 2010; Palmer et al., 2003; Schieweck et al., 2007). However,
records of past pesticide application to protect collections from insects
and other pests are often limited (Goldberg, 1996; Schieweck et
al., 2007). For example, in Germany use of organic and inorganic pesticides
in museum collections was largely undocumented or poorly documented
prior to the 1970s (Schieweck et al., 2007). Some of the
knowledge of the pesticides used is incidental, obtained through
word-of-mouth; and the type of pesticides used varied according to
the type of collection, efficacy of the pesticide, as well as human health
and safety (Goldberg, 1996; Schieweck et al., 2007). Determination of
pesticide use in artefacts or for the treatment of wooden building materials
is therefore difficult.
Levels of pesticides in museum objects can be high due to repeated
application (Wörle et al., 2012). In one case, objects shown in a museum
were so heavily treated with DDT they emitted a strong odour, and decontamination
methods for the artefacts was sought so they were safer
to display (Wörle et al., 2012). Yet, evaluation of health outcomes from
exposure to pesticides through indoor air in buildings housing important
artefacts is rare, despite the potential for serious health effects. In the past,
pesticides were applied by museum staff and hence exposure to the pesticides
may have occurred during application, leading to acute effects such
as nausea (Musshoff et al., 2010). Museum staff may also be exposed to
the pesticides during their everyday work, which may include handling
or cleaning contaminated objects (Glastrup, 2000; Marcotte et al., 2014;
Musshoff et al., 2010; Schieweck et al., 2007).
Dermal uptake and ingestion through handling contaminated objects
or contact with contaminated dust, as well as inhalation of particle-associated
or volatile compounds are expected exposure routes in
indoor environments such as museums (Covaci et al., 2006; Glastrup,
2000; Purewal, 2000; Schieweck et al., 2007). Exposure via dermal uptake
or inhalation of particles or gas-phase compounds may also occur
from museum artefacts used for ceremonial purposes (Palmer et al.,
2003). Marcotte et al. (2014) assessed the daily inhalation and ingestion
exposure in The Natural History Museum of Rouen, and neither exposure
pathway was found to be significant in the short-term (i.e., inhalation
and ingestion exposures were orders of magnitude less than
occupational guidelines and permissible uptake limits, respectively).
However, to date no other studies have assessed multiple exposure
pathways (dermal uptake, inhalation and ingestion) to OCPs for
workers or visitors in museums, or in historical buildings where people
may work or visit, such as theatres or castles. In addition, long-term exposure
in these indoor environments has been neglected. Therefore, the
relative importance of these different exposure pathways in such environments
is unknown.
The main objective of the present study was to assess potential risks
related to pesticide exposure via inhalation exposure in a largely unexplored
occupational setting in the Czech Republic – buildings of cultural
and historical importance, specifically a theatre and a museum. In the
Czech Republic, OCPs were used extensively in agriculture in the past
and their volatilization from secondary sources such as contaminated
soils into the ambient air has been well documented (Beránek and
Havel, 2006; Beránek and Petrlík, 2005; Dvorska et al., 2008;
Holoubek et al., 2007). Technical HCH (mixture of HCH isomers) and
γ-HCH (also known as lindane) were also used for wood treatment
(Beránek and Havel, 2006), which may be a source of this compound
in indoor air (Kohoutek et al., 2005). Formulations such as Pentalidol
(containing 2% DDT but also 5% pentachlorophenol (PCP) and 0.1% γHCH)
were produced and used in the former Czechoslovakia for the
treatment of different kinds of wood, construction material, furniture,
floors and roofs against insects, fungi and mold in the late 1970s and
early 1980s (Beránek and Havel, 2006; Kohoutek et al., 2005). Yet,
there is no information available in the published literature on the use
of OCPs in historical buildings of cultural significance or in museum collections
in the Czech Republic. This is despite evidence of OCP use and
residues in museum air and artefacts in other regions (Fellowes et al.,
2011; Goldberg, 1996; Marcotte et al., 2014; Musshoff et al., 2010;
Palmer et al., 2003; Schieweck et al., 2007).
We have investigated the levels of selected organochlorine pesticides
(p,p′-DDT and its metabolites p,p′-dichlorodiphenyldichloroethylene
(p,p′-DDE) and p,p′-dichlorodiphenyltrichloroethane (p,p′-DDD) and
HCH isomers γ-, α- β- and δ-HCH), as well as industrial chemicals (hexachlorobenzene
(HCB) and pentachlorobenzene (PeCB)) in a baroque theatre
and a natural history museum. In the theatre, OCPs were used for
preservation of the wooden structure of the building itself, and in the museum
they were applied to protect historical collections. Air quality data
were used to estimate occupational inhalation exposure and risk to the
health (i.e., cancer risk (CR)) of workers at the baroque theatre and at
the museum.
2. Methods
2.1. Sampling sites
2.1.1. Baroque theatre, Český Krumlov
Air sampling sites were located in the stage area and attic of the theatre
(see Fig. S1). The site history is described in detail in Kohoutek et al.
(2006, 2005). In brief, in the 1950s, wooden materials in the theatre
were affected by dry rot and between 1966 and 1997, the theatre was
closed to public due to restorations. During this time various fungicides,
insecticides and conservation treatments were applied, but the type of
pesticide used, the application volume, frequency, and technique was
not well documented. Pentalidol (2% DDT; 5% PCP and 0.1% γ-HCH)
use was documented since 1980, but this pesticide was most likely
used earlier at the site; and Pentalidol use was restricted in 1983. It is
suspected that pesticide application occurred on dusty surfaces in the
theatre thus creating a large volume of heavily polluted mobile particles,
which resulted in contamination of the theatre with a complex mixture
of chemicals. In the early 1990s, there were reports of eye, membrane
and skin irritation from people who had spent time in the theatre. Consequently,
~5 t of contaminated material, including contaminated dust,
was removed from the theatre but acute health issues were still being
reported when visitors returned. A sampling campaign occurred in
summer 2001 to isolate contamination sources. Consequently, wooden
floors, tarred cardboard and insulation were removed from the attic between
2002 and 2003, and the attic floor was covered with plastic to
prevent particles from penetrating into the auditorium of the theatre.
In summer 2003, wooden beams in the attic were tested to determine
whether the white surface layers required removal. Trials carried out
in spring 2006 showed the most appropriate procedure to remove the
whitish surface layer of the timber, and reduce contamination, was
dry planing and subsequent sanding of corners and other hard-to-access
places. Planing and sanding was coupled with a high quality exhaust
system to capture and collect dust. Dry planing allowed for the
preservation of the authentic shape of the original material. However,
this remediation was never completed. Results of building material
and air sampling discussed below span from 2001, when acute health
issues from past pesticide use were apparent, to 2006, a few years
after site restoration and remediation trials. No further analysis of air
or building materials occurred in the theatre since 2006. Given the persistent
nature of the OCPs analyzed in the present study, as well as the
fact that no further building works or site remediation have occurred
in the theatre, it is expected that levels of OCPs are still high in indoor
air.
2.1.2. Museum of South Bohemia, České Budějovice
The Museum of South Bohemia, located in České Budějovice, is an
important historical building, part of which had been subject to interior
599E. Holt et al. / Science of the Total Environment 609 (2017) 598–606
renovations at the time of sampling. As with the theatre, records of past
pesticide use in the museum are poor. Rooms in two different buildings
(A and B), were sampled using the methods described below in Section
2.2.2. In the botanical and zoological sciences storage areas (storage
rooms 1, 2 and 3) of building A some plant and animals specimens
were treated with γ-HCH (Prokeš et al., 2015). Granulated γ-HCH was
applied in drawers of specimens up to 40 years ago and visible signs
of this application remained up to and including at the time of sampling
in at least one of the storage rooms (storage room 2) (Prokeš et al.,
2015). γ-HCH applied in storage rooms 1 and 3, building A had been removed
prior to sampling, but residues of past application of this pesticide
were expected to remain (Prokeš et al., 2015). γ-HCH was also
applied to the floor of an office in the same building (A) in approximately
2011 or 2012, although there are no specific details on the application
process and rate (Prokeš et al., 2015). There may also have been some
past application of γ-HCH in exhibition room 1 of building B but there
are no details or of when this occurred or the application rate and
process. There was no history of pesticide use in exhibition room 2 of
building B. For further details of the site rooms refer to Table S1 and
Fig. S2 in the Supplementary data (SD).
2.2. Sampling
2.2.1. Baroque theatre, Český Krumlov
The original aim of indoor air monitoring in this theatre was to investigate
the source of OCPs (2001 sampling). In 2002 to 2003, air and
building material were monitored to inform restoration works and to
deal with the acute health issues of theatre visitors. Air monitoring in
2004 to 2006 was to assess how OCP levels changed after restoration
and determine whether remediation was required for OCP embedded
in building materials (trial remediation occurred in spring of 2005). Indoor
temperature and humidity were recorded alongside collection of
air samples. Samples of building materials were also collected to determine
the potential source of pesticides in indoor air.
Air sampling was carried out based on internationally recognized
methods (Kohoutek et al., 2005). In brief, air was sampled using the
US EPA test method TO-13 (US EPA, 1988) and a high volume active
air sampler (HV AAS (Fig. S2, SD); PS-1 (Thermo, USA)) (Kohoutek et
al., 2005). The particle phase was captured on quartz fiber filters (QFF)
(Whatman, USA), and the gas phase was collected on polyurethane
foam (PUF) plugs. Sampling was carried out in parallel in the stage,
backstage and auditorium area, as well as in the attic in July 2001, August
2003 and 2004, September 2005 and August 2006. Each AAS sampling
campaign consisted of one, two or three consecutive periods of
approximately 12 h, each time collecting volumes of approximately
200 m3
. Details of air samples (room and year) are included in Table
2. In 2001, there were two consecutive samples taken in the stage
area, and one air sample each was taken in the auditorium, upper
ropes and attic areas of the theatre. One air sample was taken per year
in the stage and attic areas of the theatre in 2003 and 2004. In 2005,
two consecutive air samples were taken in the stage area and attic,
and in 2006, three consecutive air samples were taken in the stage
area and attic of the theatre. The PUF plugs and quartz filters used in
air sampling were collected, wrapped in two layers of aluminum foil, labeled,
sealed in polythene bags and transported to the laboratory at a
temperature of approximately 5 °C. In the laboratory, the samples
were stored at −18 °C before processing.
Temperature and humidity are key factors likely to influence the intensity
of secondary emissions of target compounds from any potentially
contaminated building materials in the baroque theatre and their
redistribution through air between rooms in gaseous form. Therefore,
all sampling was carried out in the same season, summer, when there
were days with longer periods of sunshine and higher temperatures of
the air inside and outside the theatre than at other times of the year.
Twenty-six grab samples of building materials such as wood (shaved
or scraped), plaster and insulation were taken from the attic, stage area
and auditorium of the theatre from 2001 to 2006. Samples were taken
with the assistance of theatre staff to help determine the source of compounds
identified in indoor air. Solid building material samples were
taken from the surface layer (scraped off to a depth of 1 mm). All samples
were wrapped in foil, placed in Ziploc bags and transported to the
laboratory at a temperature of approximately 5 °C. In the laboratory,
the samples were stored at −18 °C before processing.
2.2.2. Museum of South Bohemia, České Budějovice
The main aim of the sampling campaign was to measure the levels of
γ-HCH and a number of other OCPs in indoor air. At the museum, air
sampling was carried out using a medium volume (MV) AAS (Leckel
MVS6, Sven Leckel, Germany), over a two-day period at six sites within
the museum offices, collection storage area and public access areas/exhibition
areas. Sampling times for AAS were based on the volumes of individual
rooms, i.e., 40 h for a smaller room and 50 h for a larger room.
Sampling was carried out in September 2014. No additional air sampling
has been carried out in the museum since 2014. Yet, given the persistent
nature of the OCPs sampled, and continued presence of the
lindane granules, it is not expected that levels of these compounds in
the museum air have changed significantly in the past three years.
In total there were six samples taken (one per room) in the museum.
Active sampling was carried out according to the method outlined in
Czech technical standards 16000-01, 16000-12 and 16000-32. The sampler
collected particles b10 μm (PM10). The AAS sampler was fitted
with a QFF (Whatman, QMA, United Kingdom) for collecting the particle-bound
compounds and two PUF plugs for sampling gas-phase compounds
(Molitan a.s., Czech Republic). The flow rate of the sampler was
2.3 m3
h−1
. PUF plugs were pre-cleaned by hot Soxhlet extraction for
8 h in acetone followed by 8 h in dichloromethane.
Samples were collected, transported to the laboratory, and stored
prior to analysis in the same manner as those from the baroque theatre.
2.3. Sample processing, clean-up and analysis
2.3.1. Baroque theatre, Český Krumlov and Museum of South Bohemia,
České Budějovice
Analyses of target OCPs ((a) α-, β-, γ-, δ- and ε-HCH, (b) o,p′-DDT
and p,p′-DDT, o,p′-DDE, p,p′-DDE, o,p′-DDD and p,p′-DDD and (c) HCB
and PeCB) was carried out in a similar manner to analysis described in
Lohmann et al. (2012). In brief, all sample media (QFFs, PUF plugs, and
5 g of solid matrices) were extracted with 120 mL of dichloromethane
in a Büchi System B-811 automatic extractor (Büchi Labortechnik AG,
Switzerland) for 40 min followed by a 20-minute thimble wash and
condensation. AAS (QFF and PUF plug) were spiked with recovery standards,
PCB 30 and 185 (Wellington Laboratories Inc., Canada) prior to
extraction. Sample extract volumes were reduced under a gentle nitrogen
stream at ambient temperature. The extracts were subject to cleanup
on a H2SO4 modified silica gel column (44% w/w) (eluted with 30 mL
of 1:1 hexane/dichloromethane). Analytes were eluted from the column
with a mixture of n-hexane and dichloromethane (1:1). Extracts were
concentrated under a gentle stream of nitrogen, transferred to nonane
and an internal standard of PCB 121 (Wellington Laboratories Inc., Canada)
was added.
Baroque theatre solid and air samples were analyzed using a gas
chromatograph (HP 6890) coupled with a mass spectrometer (HP
5975) in selected ion monitoring (SIM) mode. The separation took
place on a J&W Scientific fused silica column (DB-5MS; 60 m ×
0.25 mm × 0.25 μm) with helium carrier gas at a flow of
1.5 mL min−1
. Sample volumes of 1 μL were injected in splitless mode
at 280 °C−1
. The GC temperature program was as follows: 80 °C (hold
for 1 min), 15 °C min−1
to 180 °C, 5 °C min−1
to 310 °C (10 min). Analysis
for the Museum of South Bohemia air samples was carried out using
gas chromatography with tandem mass spectroscopy (GC–MS/MS;
Agilent GC 7890/MS-MS Triple Quadrupole 7000B) in multiple reaction
monitoring (MRM) mode. The separation took place on a SGE Analytical
600 E. Holt et al. / Science of the Total Environment 609 (2017) 598–606
Science HT-8 column (60 m × 0.25 mm internal diameter × 0.25 μm; 8%
phenyl polycarbonate-siloxane) with helium as a carrier gas at a flow
rate of 1.5 mL min−1
. The GC temperature program was as follows: 80 °C
(hold for 1.5 min), 40 °C min−1
to 200 °C (hold for 18 min), 5 °C min−1
to 305 °C. Splitless injection method was executed at 280 °C.
2.4. Quality assurance/quality control (QA/QC)
2.4.1. Baroque theatre, Český Krumlov and Museum of South Bohemia,
České Budějovice
Compounds were identified and quantified with an eight point calibration
of all target compounds (i.e., at concentrations of 1, 4, 10, 40,
100, 400, 1000 and 4000 ng mL−1
). Linearity was achieved for the entire
concentration range. The OCP analytical method for AAS has been previously
evaluated using a certified reference material (ASLAB soil standard,
Czech Republic) (Lohmann et al., 2012). Recovery of all native
OCPs from the reference material ranged from 75 to 98% at the time of
sample analysis for the theatre and the museum.
Recoveries for the baroque theatre air and solid samples, based on
PCB 30 and 185 added prior to extraction, were N71%. For the Museum
of South Bohemia the recoveries ranged from 47 to 84% for PCB 30 and
76 to 104% for PCB 185. Recovery factors were not applied to any of the
data from either the theatre or the museum, i.e., no correction was made
to the results for recovery.
One trip blank and one solvent blank were analyzed with each batch
of air samples (2 to 6 samples per batch) from the theatre and the museum;
and one solvent blank was analyzed for each batch of solid samples
(2 to 6 samples per batch) from the theatre. Field blanks were
extracted and analyzed in the same way as the samples, and the levels
of OCPs in field blanks never exceeded 1% of the quantities detected in
samples from either the theatre or museum for any OCPs, indicating
minimal contamination during the transport, storage, and analysis. As
a result, no blank corrections were performed on the data. Limits of
quantification (LOQ) were determined using the lowest external calibration
point. The LOQ was defined as the level at which signal-tonoise
(S/N) ratio was 10:1 (Peak-to-Peak).
2.5. Human health risk assessment
2.5.1. Cancer risk assessment
Cancer risk (CR) was calculated for workers potentially exposed to
α-HCH, γ-HCH, DDT, DDD, DDE1
and HCB in the museum and theatre
through inhalation. Occupational exposure concentration was calculated
using the following equation (US EPA, 2009),
EC ¼ CA Â ET Â EF Â EDð Þ=AT ð1Þ
Where, EC is the exposure concentration (μg m−3
), CA is contaminant
concentration in air (μg m−3
), ET is exposure time (hours day−1
), EF
is exposure frequency (days year−1
), ED is exposure duration (year),
and AT is averaging time (hours). The values for each of these parameters
are included in Table 1.
The estimated occupational exposure via inhalation was multiplied
by the appropriate toxicity factor - Inhalation Unit Risk Factor (URFi)
or Inhalation Unit Risk (IUR; (μg m−3
)−1
) in order to derive an estimate
of the potential CR associated with exposure,
CR ¼ EC Â URFi or IUR ð2Þ
IUR for α-HCH and DDT were obtained from the Integrated Risk Information
System (IRIS) database (US EPA, 2003) (in Table S2, SD).
The IUR for γ-HCH was derived by the California (2005). The URFi
values for DDD, DDE and HCB originated from the ACCESS database supplied
by the US EPA as part of the Human Health Risk Assessment Protocol
for Hazardous Waste Combustion Facilities (US EPA, 2005b). The
procedures used to calculate the URFi or IUR values are described in
California (2005); US EPA (2003) and US EPA (2005b).
According to the United States Environmental Protection Agency
(US EPA) a CR of b1 × 10−6
(or 1 in one million) is considered to be
very low or negligible (US EPA, 1991). A CR of N1 × 10−4
(or 1 in ten
thousand) is considered large enough to warrant risk management actions
such as mitigation or reduction measures (e.g., remediation) (US
EPA, 1991). CR within the range of 1 × 10−4
and 1 × 10−6
, where risk
is low to moderate, may not be considered large enough to warrant
risk management actions but should be evaluated on a case-by-case
basis (US EPA, 1991).
3. Results and discussion
3.1. Levels and patterns of pesticides in air and building materials
3.1.1. Baroque theatre, Český Krumlov
Levels of p,p′-DDE and p,p′-DDT in air were higher in the attic than in
the theatre (the stage area and auditorium) (Table 2). For example, p,p′DDE
ranged from 45 ng m−3
to 190 ng m−3
in the attic and 5.5 ng m−3
to 26 ng m−3
in the stage and auditorium areas, respectively. The source
of the p,p′-DDE and p,p′-DDT in air of the attic was suspected to be the
contaminated materials (e.g., insulation) in the attic (Table 3). Indoor
air levels of p,p′-DDE, p,p′-DDD and p,p′-DDT decreased in the attic
once insulation, etc. was removed from the attic between 2002 and
2003 (Table 2). There was also a significant negative linear relationship
between the air concentrations of these pollutants and year sampled
(p b 0.05) (Fig. S3). However, the wooden beams coated in white
powder and shown to contain high levels of p,p′-DDE (untreated
beam = 1600 to 27,000 ng g−1
; Table 3) and p,p′-DDT (untreated
beam = 260 to 8600 ng g−1
; Table 3) remained in the attic. As a result,
relatively high levels of p,p′-DDE in particular were still detected in attic
air from 2004 to 2006 (DDE: 8.0 to 100 ng g−1
; Table 2), particularly in
2005. The use of plastic sheeting in 2003 to prevent particle-associated
DDT and DDE entering the theatre does not appear to have reduced
1
The p,p′-DDT, p,p′-DDE and p,p′-DDD isomers were measured in air at the baroque
theatre and o,p′- and p,p′-DDT, DDE and DDD isomers were measured in air at the
museum.
Table 1
Input values for the inhalation exposure calculation for workers in the theatre and museum.
Parameter Value Description
CA See Table 2 and Table 4 Lower bound values (i.e., 0) where air concentrations were bLOQ
ET 10 h day−1
It was estimated people spend 8 to 10 h a day at work and the more conservative value was used for the calculation.
Percentage of the day worked is 42%.
EF 241 days year−1
It was assumed 4 weeks holiday were taken by workers and work was undertaken from Monday to Friday (i.e., 5 days a week).
ED 30 years The exposure duration was defined as 30 years, which is the reasonable maximum exposure for an adult resident and is
based on an established hazardous waste facility with a useful lifetime of 30 years. 30 years was assumed to represent a
reasonable maximum period of employment and time spent by a worker in the theatre or museum (US EPA, 2005a).
AT 613,200 h The averaging time (AT) is the period over which exposure is averaged. AT was estimated to be lifetime in years × 365
days/year × 24 h/day, where lifetime is the average lifetime for chemicals of potential concern for carcinogenicity of 70
years (US EPA, 2009).
601E. Holt et al. / Science of the Total Environment 609 (2017) 598–606
p,p′-DDE and p,p′-DDT levels in theatre air and the relationship between
air concentrations of these compounds and year sampled was not significant
(p N 0.05) (Fig. S3).
HCH concentrations in air were generally higher in the stage and auditorium
area (5.2 to 37 ng m−3
) than the attic (2.6 to 17 ng m−3
)
(Table 2). The source of the γ-HCH in the indoor air around the stage
area was most likely previous indoor application of the pesticide. This
was supported by analysis of wood scrapings from the backstage area
around the lower machinery and stage areas of the theatre, which had
much higher γ-HCH levels than in other materials tested in the building
(Table 3). The linear relationship between the air concentrations and
time was negative and significant (p b 0.05) suggesting that concentrations
of γ-HCH in the attic decreased after the removal of building materials
(between 2002 and 2003) and laying of plastic sheeting on the
attic floor in 2003 (Fig. S3).
PeCB and HCB concentrations in air were similar in the attic, stage
and auditorium areas of the theatre. The most likely source of the chlorobenzenes
in the indoor air is outdoor air. Most of the HCB present in
outdoor air of the Czech Republic is thought to have re-volatilized
from past soil contamination (Dvorska et al., 2008). Other potential
sources of HCB in the environment of the Czech Republic are the
manufacture of chlorinated solvents and pesticides, as well as the application
of HCB-contaminated pesticides and inadequate incineration of
chlorine-containing wastes (Dvorska et al., 2008).
As expected from past application of DDT, the p,p′-DDT metabolites
(p,p′-DDE and p,p′-DDD) dominated relative to p,p′-DDT in air and
building materials. The average p,p′-DDT metabolite contribution to
∑p,p′-DDT, p,p′-DDE and p,p′-DDD in air was 0.84 and was largely unchanged
in air over the sampling period from 2001 to 2006 (ranging
from 0.68 to 0.98). The HCH fingerprint (contribution of each isomer
to the ∑4HCH, i.e., α-, β-, γ- and δ-HCH) was dominated by γ-HCH in
the air and the solid building materials in both the stage area and in
the theatre attic, with only a few exception where the α- or β-HCH isomers
dominated. This suggests that the source of the HCH in the theatre
is a γ-HCH formulation rather than technical HCH.
Annual fluctuations in the temperature and humidity of the theatre
were expected to affect the air concentrations of OCPs, but this was not
evaluated because all sampling occurred at approximately the same
time of the year, i.e., summer. At the time of sampling, temperature
and humidity in the theatre ranged from 17 °C to 18 °C and 63% to
67%, respectively. Throughout the year in the theatre the temperature
and humidity ranged from −0.50 °C to 21 °C and 49% to 82%,
Table 2
Levels (ng m−3
) of selected OCPs in bulk indoor air in the baroque theatre.
Year 2001a
2003 2004 2005 2006
Sample
location
Stage
1
Stage
2 Auditorium
Upper
ropes Attic Stage Attic Stage Attic
Stage
1
Stage
2
Attic
1
Attic
2
Stage
1
Stage
2
Stage
3
Attic
1
Attic
2
Attic
3
α-HCH NA NA NA NA NA 0.41 0.16 0.55 0.27 1.1 1.1 1.5 1.1 0.26 0.21 0.32 0.10 0.18 0.13
β-HCH NA NA NA NA NA 0.047 0.016 0.032 0.067 0.29 0.23 0.56 0.36 0.036 0.026 0.035 0.012 0.015 0.012
γ-HCH 5.2 25 14 28 17 24 11 12 4.5 37 36 5.2 4.4 6.2 5.5 8.0 2.6 5.2 3.2
δ-HCH NA NA NA NA NA 0.038 0.047 bLOQ bLOQ 0.077 0.065 0.14 0.097 bLOQ bLOQ bLOQ bLOQ bLOQ bLOQ
p,p′-DDD NA NA NA NA NA 3.1 14 0.55 0.44 2.6 2.8 5.0 4.6 0.31 0.34 0.40 0.22 0.60 0.26
p,p′-DDE 5.5 22 8.5 18 190 20 110 8.9 58 26 12 92 100 6.8 4.7 9.3 8.0 14 9.4
p,p′-DDT 1.1 3.4 1.9 3.1 21 4.6 15 1.2 1.8 2.3 2.6 2.9 2.7 1.6 1.5 2.3 1.2 3.1 1.5
HCB NA NA NA NA NA 1.1 0.59 1.8 0.72 1.5 1.8 0.88 0.96 1.2 0.94 1.6 0.68 1.2 0.67
PeCB NA NA NA NA NA 0.094 0.022 0.11 0.031 0.47 0.72 0.27 0.28 0.39 0.39 0.43 0.31 0.48 0.20
NA = not available. LOQ = 0.00050 ng m−3
.
a
All values as reported in Kohoutek et al. (2006).
Table 3
Levels (ng g−1
) of selected OCPs in solid materials from the baroque theatrea
.
Year Material Location α HCH β-HCH γ HCH δ-HCH p,p′-DE p,p′-DDD p,p′-DDT HCB PeCB
2001 Cardboard insulation A NA NA NA NA 250 NA 2900 NA NA
Fiberglass insulation NA NA NA NA 100 NA 41 NA NA
Wood NA NA NA NA 580 NA 2700 NA NA
Crushed stone NA NA NA NA 86 NA 380 NA NA
2003 Non-treated beam, scraping A 96 6.4 170 3.1 21,000 1100 8600 7.1 NQ
11 NQ 38 1.5 1600 51 260 1 0.14
Floor, wood shavings 0.38 NQ 5.0 0.18 290 12 66 0.26 0.13
2004 Lower machinery, scraping LM 8.4 2 400 NQ 4300 300 1100 11 6.8
Stage, scraping S 13 7.9 450 NQ 2900 190 590 14 40
Upper rope, scraping UR 16 NQ NQ NQ 3600 510 1900 29 65
Non-treated beam, scraping A NQ NQ 54 NQ 8700 280 930 NQ NQ
2005 Lower machinery, scraping LM 32 23 490 NQ 3800 690 6400 1.6 NQ
Stage, scraping S 27 34 320 NQ 2300 310 1800 0.59 NQ
Upper rope, scraping UR 26 94 260 NQ 4100 350 2300 1.1 NQ
Upper rope, plaster UR 2.7 18 12 NQ 0.63 34 0.44 NQ NQ
Non-treated beam, scraping A 360 14 150 NQ 18,000 920 1700 13 NQ
A 590 42 240 NQ 27,000 1200 2400 21 23
Ceiling boards, scraping A 0.49 NQ 41 NQ 3200 46 70 2.8 NQ
A 170 120 130 NQ 5600 100 190 40 20
A NQ NQ 140 NQ 6000 90 170 33 17
A 12 14 28 NQ 4200 84 590 0.20 NQ
Plaster A 4.1 30 11 NQ 610 59 260 0.10 NQ
Clean beam, scaping A NQ NQ 29 NQ 730 70 160 NQ NQ
A NQ NQ 23 NQ 590 100 330 NQ NQ
2006 Clean beam A NQ 0.11 4.1 NQ 130 20 81 NQ 0.31
A NQ NQ 11 NQ 180 35 140 NQ 0.15
NA = not available. A = attic, S = stage, LM = lower machinery and UR = upper rope. NQ = not quantified due to interference in analysis.
a
All values as reported in Kohoutek et al. (2006).
602 E. Holt et al. / Science of the Total Environment 609 (2017) 598–606
respectively. As the target OCPs are fairly volatile, the levels of p,p′-DDE,
p,p′-DDT and γ-HCH measured in the indoor air of the theatre are therefore
likely to represent the highest annual levels, i.e., a worst case scenario
for the year. Rapid heating during the daytime and high
temperatures throughout the day would have allowed volatilization of
p,p′-DDT from treated attic beams thus increasing concentrations in air.
3.1.2. Museum of South Bohemia, České Budějovice
Organochlorine pesticides were detected in air samples in all rooms
of the Museum of South Bohemia (Table 4). Levels of ∑5HCH (α-, β-, γ-,
δ- and ε-HCH) in air varied markedly between rooms (average of
3400 ng m−3
and range of 8.0 to 15,000 ng m−3
) and most of the
∑5HCH was γ-HCH. Levels were highest in the storage rooms where
ɣ-HCH was applied in the past as a granulated formulation for the preservation
of the museums botanical and zoological collections, and a significant
residue of the applied formulation remains to the present day.
γ-HCH levels were also at least 3 times higher in the air in the office
compared to the exhibition areas, most likely due to the application of
a γ-HCH formulation in the office a few years prior to the air sampling.
Average and range of ∑6DDT (o,p′- and p,p′-DDT, DDE and DDD) levels
in air (0.13 ng m−3
and 0.066–0.35 ng m−3
) were orders of magnitude
lower than ∑5HCH. There is no record of DDT use in the museum. DDT
(sum of o,p′- and p,p′-DDT isomers) was N3 times higher in one of the
rooms used to house the museum's collection of zoological specimens
(storage room 3) than other rooms of the museum, suggesting that
this pesticide may have been used here in the past. However, in general,
the source of DDT and its metabolites in the indoor air of the museum is
likely to be outdoor air, considering that the levels in most of the rooms
are similar. ∑6DDT was dominated by p,p′-DDE in the indoor air of the
museum because DDT has not been used for several decades in the
Czech Republic. HCB levels were similar in all rooms, indicating that outdoor
air was also the source of the indoor contamination of this
compound.
3.1.3. Comparison to OCPs in indoor air of other museums and historical
buildings
There are limited studies on OCP levels in indoor air of museums (air,
dust) and artefacts (Table 5). To the authors' knowledge, there is no information
on the levels of OCPs in air of historical buildings such as the
baroque theatre (e.g., castles, churches, etc.). In general, levels of DDT
and γ-HCH were higher in artefacts or museum air in other studies
than in the present study. p,p′-DDT in all materials from the theatre included
in the present study were in the range of 41 to 29,000 ng g−1
with a median of 590 ng g−1
. These p,p′-DDT levels are much lower
than DDT levels in fur- and skin-based artefacts from the National Museum
of Denmark, where the median levels were 870,000 ng g−1
and
370,000 ng g−1
before and after cleaning, respectively (Glastrup,
2000) or in Native American artefacts (DDT levels detected up to
Table 5
Levels of OCPs in museum air and dust from museums, and artefacts.
Sample description Results Reference
Air Air measured with AAS in invertebrate, bird and mammal galleries, 2
stockrooms and a laboratory in Natural History Museum, Rouen.
DDT + DDD: b0.5. 0.9, 9.0, b0.5, b0.5 and b0.5 μg m−3
(Marcotte et al., 2014)b
γ-HCH: 0.5, 4.8, 32, 1.8, b0.3 and b0.3 μg m−3
Air measured in 2-year period on persons in preservation (PS),
maintenance (MS) and inventory (IS) services at the Natural History
Museum of Rouen.
DDT + DDD: 546 (PS 1st year), b0.5 (PS 2nd year), 46
(MS 1st year), b0.5 (MS 2nd year) and 10 (IS) μg m−3
γ-HCH: 2.8 (PS 1st year), b0.3 (PS 2nd year), 2.1
(MS 1st year), b0.3 (MS 2nd year) and b0.3 (IS) μg m−3
Dust Dust samples from the invertebrate, bird and mammal galleries, as well as a
stockroom in the Natural History Museum of Rouen.
DDT + DDD: 968, 403, b0.5 and 159 μg g−1
(Marcotte et al., 2014)
γ-HCH: 1.5, 0.7, b0.3 and 0.4 μg g−1
Settled dust of German museum rooms, where wooden sculptures are
displayed. Collected by manual wiping.
DDT: b1–100 μg g−1
(Schieweck et al., 2007)a
γ-HCH: 5–50 μg g−1
Dust, dirt off exhibits and storage areas were collected from several
locations in the Reiss-Engelhorn-Museen, Mannheim in 2006.
DDT, DDE and DDD detected. Levels not reported (Musshoff et al., 2010)b
Artefact/object Fur/skin scraped or cut off 16 artefacts (1983) kept at the National Museum
of Denmark. Artefacts cleaned with compressed air.
DDT range and median: b39–5100 and 870 μg g−1
(before clean). b39–7100 and 370 μg g−1
(after clean)
(Glastrup, 2000)b
Biocide concentrations in the diorama (soil) of the Lower Saxony State
Museum, Hanover.
DDT: b1–10 μg g−1
γ-HCH: b1–4 μg g−1
(Schieweck et al., 2007)b
Artefacts from (a) Lower Saxony State Museum, Hanover, (b) an unknown
museum and (c) a museum in southwest Germany
DDT: not detected – 21 μg g−1
γ-HCH: b1–38 μg g−1
Museum objects repatriated to the Hupa Tribe of California DDT: not detected – 130 μg g−1
(Palmer et al., 2003)b
γ-HCH: not detected – 30 μg g−1
Extract of unknown museum object. DDT detected. Not quantified (Palmer, 2000)b
Stores and collection dust and object samples in 1987. DDT detected in all samples. Levels not reported (Schmidt, 2000)b
Artefact - bear paw pouch DDT detected. Levels not reported (Sirois, 2000)b
Artefact - fly catcher γ-HCH detected. Levels not reported
Wooden printing blocks from the Plantin-Moretus Museum, Antwerp.
Objects were treated circa 1950.
Detected in all 10 samples. Levels not reported (Covaci et al., 2006)b
Not detected in all 10 samples
Wooden sculptures and masks from the Ethnographic Museum, Antwerp.
Objects were treated circa 1950.
DDT not detected in all 7 samples
γ-HCH not detected in all 7 samples
Walnut veneered spruce cabinet from Swiss National Museum. DDT crystals observed. DDT levels not quantified. (Wörle et al., 2012)
Washing/abrasion from 2 artefacts from the Reiss-Engelhorn-Museen,
Mannheim in 2006.
DDT, DDE and DDD detected. Levels not reported (Musshoff et al., 2010)b
a
The DDT values represent the sum of isomers and derivatives (o,p′-DDD, p,p′-DDD, o,p′-DDE, p,p′-DDE, o,p′-DDT, and p,p′-DDT).
b
Note that the isomers of DDT, DDE or DDD analyzed were not identified for these studies.
Table 4
Levels of selected OCPs (ng m−3
) in bulk indoor air in the Museum of South Bohemia.
Compound Exhibition
area 1
Exhibition
area 2
Office Storage
room 1
Storage
room 2
Storage
room 3
α-HCH 1.5 0.078 2.6 7.4 23 18
β-HCH b0.0013 b0.0013 b0.0016 b0.0013 b0.0016 b0.0013
γ-HCH 200 7.9 630 2800 15,000 1900
δ-HCH 0.018 0.0051 0.033 0.11 0.65 0.23
ε-HCH b0.0010 b0.0010 0.022 0.058 0.17 0.10
o,p′-DDE 0.0094 0.010 0.011 0.0062 0.0072 0.013
p,p′-DDE 0.041 0.056 0.049 0.026 0.026 0.11
o,p′-DDD b0.00068 0.0016 b0.00085 0.0020 0.0014 0.0081
p,p′-DDD b0.00059 b0.00060 b0.00075 b0.00059 b0.00074 b0.00059
o,p′-DDT 0.013 0.016 0.033 0.032 0.033 0.15
p,p′-DDT 0.0031 0.0058 0.017 0.0085 0.0064 0.068
HCB 0.31 0.28 0.29 0.49 0.48 0.32
603E. Holt et al. / Science of the Total Environment 609 (2017) 598–606
13,000,000 ng g−1
) (Palmer et al., 2003). Yet, p,p′-DDT in theatre attic
beam scrapings (this study) are similar to soil from inside a diorama
and artefacts from German museums, where levels of DDT ranged
from b1000 to 38,000 ng g−1
(Schieweck et al., 2007). Levels of γHCH,
p,p′-DDT and p,p′-DDD in the theatre air measured in the present
study ranged from 3.0 to 56 ng m−3
, 1.0 to 25 ng m−3
and 3.0 to
14 ng m−3
, respectively. Therefore, the OCPs in the baroque theatre
were well below those measured in air of the Natural History Museum
of Rouen, where levels of γ-HCH were up to 32,000 ng m−3
and levels
of DDT + DDD were up to 9000 ng m−3
(Marcotte et al., 2014). Levels
of γ-HCH in the storage rooms of the Museum of South Bohemia (median
= 1300 ng m−3
; this study) and the galleries of the Natural History
Museum of Rouen (median = 3300 ng m−3
(Marcotte et al., 2014))
were similar.
3.1.4. Comparison to OCPs in non-impacted outdoor air in the Czech
Republic
Indoor air from the present study was compared with OCPs measured
in outdoor air at the Czech European Monitoring and Evaluation
Program (EMEP) site, the Košetice Observatory, situated in a largely
rural area in central Czech Republic (Table 6) (Dvorská et al., 2009;
Dvorska et al., 2008; Holoubek et al., 2007; Holt et al., 2017). Levels of
some OCPs in the indoor air of the theatre and museum were orders
of magnitude higher than the outdoor air at the background site in the
Czech Republic. Average p,p′-DDT, p,p′-DDE, p,p′-DDD in the air of the
stage area and attic were at least 3 orders of magnitude higher than
the average background outdoor air levels in the Czech Republic, even
after building restoration and dust removal to reduce indoor air contamination.
γ-HCH was also significantly higher (2 orders of magnitude)
in the air of the theatre than in outdoor air. γ-HCH in the indoor
air of the museum was between 100 and 1 × 105
fold the outdoor air
levels.
There were, however, some similarities between indoor air levels of
OCPs in the theatre and museum, and background outdoor air in the
Czech Republic, supporting the hypothesis of outdoor air as a potential
source of some OCPs in indoor air. PeCB levels in the museum air and
HCB levels in both the theatre and museum air were generally within
the same range or slightly above the levels in background outdoor air.
β-HCH in the theatre and museum also likely originated from outdoor
air. In contrast with the theatre, levels of p,p′-DDT and its metabolites
in the museum air were also in the range of background outdoor air.
This suggests that for the OCPs not used in indoor preservation, outdoor
air is the primary source of indoor air contamination.
3.2. Inhalation cancer risk
3.2.1. Baroque theatre, Český Krumlov
The CR from exposure through inhalation to OCPs, particularly p,p′DDE
from past treatment of wood in the theatre, was estimated
(Fig. 1). The CR for employees was negligible (b1 in 1000,000) to low
(between 1 in 100,000 and 1 in 1000,000) for p,p′-DDE prior to theatre
restoration (2002 to 2003), with CR depending on the location of the
building, i.e., CR was higher in the attic. CR decreased to negligible in
the attic after site restoration, which involved the removal of contaminated
building materials including insulation. γ-HCH and HCB were
also estimated to pose a low risk to theatre workers. γ-HCH was also
lower after theatre restoration despite the fact that higher levels were
found in wood from the stage area rather than materials that were
eventually removed from the attic.
3.2.2. Museum of South Bohemia, České Budějovice
The CR from occupational inhalation exposure to γ- and α-HCH, p,p′DDE
and HCB (Fig. 2) was estimated using air monitoring data from the
Museum of South Bohemia. CR from γ-HCH in the storage rooms (CR N1
× 10−4
) is substantial enough to warrant some form of action to reduce
or eliminate risks (e.g., remediation), according to the US EPA classification
for contaminated sites (US EPA, 1991). There also appears to be
higher CR from HCB (Fig. 2E), which is from outdoor rather than indoor
Fig. 1. Estimated CR for employees from exposure to (a) γ-HCH, (b) α-HCH, (c) p,p′-DDE,
(d) p,p′-DDT and (e) HCB compounds in indoor air of the baroque theatre, Český Krumlov.
y-axes are on a logarithmic scale. Below the red line (1 × 10−6
) risk is considered very low
or negligible and above the red line the risk is considered moderate and should be subject
to case specific evaluation (US EPA, 1991). Missing values from 2001 are where air
concentrations were not available.
Table 6
Levels of OCPs in air (ng m−3
; measured using AAS), Košetice observatory from 2003 to
2012 (Data provided by the RECETOX laboratory upon request. Košetice air monitoring data
have previously been published in Dvorská et al. (2009); Dvorska et al. (2008);
Holoubek et al. (2007); Holt et al. (2017)).
Compound Min Max Mean Median
α-HCH bLOQa
0.12 0.015 0.011
β-HCH bLOQa
0.096 0.0086 0.0058
γ-HCH bLOQa
0.13 0.022 0.017
p,p′-DDE 0.0014 0.064 0.018 0.016
p,p′-DDD bLOQa
0.026 0.0023 0.0012
p,p′-DDT bLOQa
0.045 0.0050 0.0033
PeCB bLOQa
0.078 0.011 0.0086
HCB bLOQa
0.50 0.093 0.067
a
From 2003 to 2010 average LOQ for all compounds (0.00050 ng m−3
) was reported by
the RECETOX laboratory. From 2011 laboratory reporting changed to a range of LOQ values
for each compound: α-HCH (0.000030–0.0010 ng m−3
), β-HCH (0.000030–
0.0017 ng m−3
), γ-HCH (0.000030–0.0022 ng m−3
), p,p′-DDD (0.000030–
0.0015 ng m−3
), p,p′-DDE (0.000030–0.00067 ng m−3
), p,p′-DDT (0.000030–
0.0028 ng m−3
), PeCB (0.000030–0.00050 ng m−3
) and HCB (0.000030–0.00062 ng m−3
).
604 E. Holt et al. / Science of the Total Environment 609 (2017) 598–606
sources. The CR from HCB is ~5 × 10−4
in all rooms where air sampling
occurred, and is high enough to prompt actions to reduce or eliminate
potential risks. CR for α-HCH ranges between 1 × 10−4
and 1 × 10−6
(moderate risk) in the storage rooms. Therefore CR for α-HCH may
not be large enough to warrant risk management actions but should
be evaluated on a case-by-case basis (US EPA, 1991). CR for p,p′-DDE
and p,p′-DDT was negligible (CR b b1 × 10−6
).
3.2.3. Human health risk (cancer risk) assessment results in context
Few other studies have investigated exposure to OCPs in similar indoor
environments. Daily (short-term) inhalation exposure to γ-HCH,
DDT and DDD was determined for Natural History Museum of Rouen
staff and was orders of magnitude lower than occupational exposure
guidelines (Marcotte et al., 2014). Yet, no studies have investigated
potential long-term exposure and quantitatively estimated long-term
risk in museums or historical buildings such as the baroque theatre.
Therefore, direct comparisons of long-term health outcomes were not
possible.
However, while short-term exposure to high levels of OCPs may be
insignificant or low risk for human health, prolonged exposure to
OCPs, even at lower levels can be a significant risk for human health.
The exposure for staff at the Natural History Museum of Rouen was predicted
to be insignificant despite the fact that levels of γ-HCH and DDT
and DDD were high in some rooms (up to 32,000 ng m−3
and
9000 ng m−3
, respectively) (Marcotte et al., 2014). In contrast, at the
Museum of South Bohemia (present study) long-term health risks
from inhalation exposure to γ-HCH (15,000 ng m−3
) was high.
While the present study quantified levels of OCPs in indoor air, and
thus focused on inhalation exposure, occupational exposure via ingestion
of dust particles and dermal contact with contaminated surfaces
could also be significant in both the theatre and museum, and would
have to be separately estimated to determine overall occupational
exposure.
4. Conclusions
While OCPs such as DDT, γ-HCH or technical HCH are no longer applied
in many countries, apart from specifically controlled uses, the legacy
of previous application of these compounds is that their release in
indoor environments can continue due to the large volumes used in
the past and their persistence in the environment. Levels of OCPs in
air, dust, and building materials, as well as on or embedded in artefacts
are poorly characterized for settings such as buildings of cultural and
historical significance. Such buildings can to be hotspots for OCP contamination
but the extent of the problem is largely unknown. To date
there is very little information on human occupational exposure and
the potential human health risk from past use of OCPs in museum collections
or in historical buildings. The present study provides evidence
of high levels of OCPs in indoor air of important historical and cultural
buildings in the Czech Republic, such as the baroque theatre in Český
Krumlov and the Museum of South Bohemia, from pesticide use that occurred
decades ago. OCP contamination in indoor air may lead to an increased
CR to exposed workers to the point where remediation may be
required, as is the case with the Museum of South Bohemia.
Acknowledgements
This work was supported by the Ministry of Education, Youth and
Sports of the Czech Republic (RECETOX Research Infrastructure
(LM2015051 and CZ.02.1.01/0.0/0.0/16_013/0001761) and the
CETOCOEN UP project (CZ.1.05/2.1.00/19.0382)).
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.scitotenv.2017.07.203.
References
Beránek, M., Havel, M., 2006. Lindane in the Czech Republic. International Persistent Organic
Pollutants Elimination Network (IPEN), pp. 1–31.
Beránek, M., Petrlík, J., 2005. POPs Pesticides in the Czech Republic. International Persistent
Organic Pollutant Elimination Network (IPEN), pp. 1–21.
California, E.P.A., 2005. Air Toxics Hot Spots Program Risk Assessment Guidelines: part II
Technical Support Document for Describing Available Cancer Potency Factors. California
Environmental Protection Agency.
Covaci, A., Kawaki, P., Indekeu, C., Schepens, P., Neels, H., 2006. Highly chlorinated toxic
contaminants in pesticide-treated wooden art objects. Arch. Environ. Occup. Health
61, 25–248.
Dvorska, A., Lammel, G., Klanova, J., Holoubek, I., 2008. Kosetice, Czech Republic–ten years
of air pollution monitoring and four years of evaluating the origin of persistent organic
pollutants. Environ. Pollut. 156, 403–408.
Fig. 2. Estimated CR through inhalation exposure to (a) γ-HCH, (b) α-HCH, (c) p,p′-DDE
and (d) HCB in indoor air of the Museum of South Bohemia. y-axis is on a logarithmic
scale. Below the red line (1 × 10−6
) risk is considered very low or negligible and above
blue line (1 × 10−4
) risk may require mitigation measures to reduce risk; between red
and blue lines the risk is considered moderate and should be subject to case-specific
evaluation (US EPA, 1991).
605E. Holt et al. / Science of the Total Environment 609 (2017) 598–606
Dvorská, A., Lammel, G., Holoubek, I., 2009. Recent trends of persistent organic pollutants
in air in central Europe - air monitoring in combination with air mass trajectory statistics
as a tool to study the effectivity of regional chemical policy. Atmos. Environ. 43,
1280–1287.
Fellowes, J.W., Pattrick, R.A.D., Green, D.I., Dent, A., Lloyd, J.R., Pearce, C.I., 2011. Use of biogenic
and abiotic elemental selenium nanospheres to sequester elemental mercury released
from mercury contaminated museum specimens. J. Hazard. Mater. 189, 660–669.
Glastrup, J., 2000. The effectiveness of compressed air in removal of pesticides from ethnographic
objects. In: Caldararo, N., Davis, L., Palmer, P., Waddington, J. (Eds.), The
Contamination of Museum Materials and the Repatriation Process for Native California:
Proceedings of a Working Conference at the San Francisco State University. vol.
16. Collection Fourm, San Francisco State University, pp. 19–22.
Goldberg, L., 1996. A history of pest control measures in the anthropology collections, national
museum of natural history, Smithsonian Institution. J. Am. Inst. Conserv. 35,
23–43.
Holoubek, I., Klanova, J., Jarkovsky, J., Kohoutek, J., 2007. Trends in background levels of
persistent organic pollutants at Kosetice observatory, Czech Republic. Part I. Ambient
air and wet deposition 1996–2005. J. Environ. Monit. 9, 557–563.
Holt, E., Bohlin Nizzetto, P., Borůvková, J., Harner, T., Kalina, J., Melymuk, L., et al., 2017.
Using long-term air monitoring of semi-volatile organic compounds to evaluate the
uncertainty in polyurethane-disk passive sampler-derived air concentrations. Environ.
Pollut. 220 (Part B), 1100–1111.
Kohoutek, J., Holoubek, I., Klánová, J., Zachař, P., Tříska, J., 2005. Vývoj Znečištění Vnitřních
Prostor Barokního Divadla V Českém Krumlově Chemickými Látkami Ochrana
Ovzduší (in Czech). 5 pp. 1–7.
Kohoutek, J., Holoubek, I., Klánová, J., Zachař, P., 2006. Tocoen Report No. 312: Vývoj
znečitění vnitřních prostor barokního divadla v Českém Krumlově v letech 2001–
2006 (in Czech). TOCOEN s.r.o., Brno, Czech Republic.
Lohmann, R., Klanova, J., Kukucka, P., Yonis, S., Bollinger, K., 2012. PCBs and OCPs on a
east-to-west transect: the importance of major currents and net volatilization for
PCBs in the Atlantic Ocean. Environ. Sci. Technol. 46, 10471–10479.
Marcotte, S., Estel, L., Leboucher, S., Minchin, S., 2014. Occurrence of organic biocides in
the air and dust at the Natural History Museum of Rouen, France. J. Cult. Herit. 15,
68–72.
Musshoff, F., Gottsmann, S., Mitschke, S., Rosendahl, W., Madea, B., 2010. Potential occupational
exposures in the Reiss-Engelhorn-Museen Mannheim/Germany. Bull. Environ.
Contam. Toxicol. 85, 638–641.
Palmer, P., 2000. A review of analytical methods for the determination of mercury, arsenic,
and pesticide residues on museum objects. In: Caldararo, N., Davis, L., Palmer, P.,
Waddington, J. (Eds.), The Contamination of Museum Materials and the Repatriation
Process for Native California: Proceedings of a Working Conference. vol. 16. Collection
Fourm, San Francisco State University, pp. 25–39.
Palmer, P.T., Martin, M., Wentworth, G., Caldararo, N., Davis, L., Kane, S., et al., 2003. Analysis
of pesticide residues on museum objects repatriated to the Hupa Tribe of California.
Environ. Sci. Technol. 37, 1083–1088.
Prokeš, R., Čupr, P., Vaňková, L., Přibylová, P., Audy, O., Stanovení, Klánová J., 2015.
vybraných organických polutantů ve vnitřním ovzduší Jihočeského muzea v Českých
Budějovicích, 2014 (in Czech). Centre for Toxic Compounds in the Environment
(RECETOX), Brno, Czech Republic, pp. 1–33.
Purewal, V., 2000. The identification of four persistent and hazardous residues present on
historic plant collections housed within the National Museum and Galleries of Wales.
In: Caldararo, N., Davis, L., Palmer, P., Waddington, J. (Eds.), The Contamination of
Museum Materials and the Repatriation Process for Native California: Proceedings
of a Working Conference. vol. 16. Collection Fourm, San Francisco State University,
pp. 77–86.
Schieweck, A., Delius, W., Siwinski, N., Vogtenrath, W., Genning, C., Salthammer, T., 2007.
Occurrence of organic and inorganic biocides in the museum environment. Atmos.
Environ. 41, 3266–3275.
Schmidt, O., 2000. Insecticide contamination at the National Museum of Denmark: a case
study. In: Caldararo, N., Davis, L., Palmer, P., Waddington, J. (Eds.), The Contamination
of Museum Materials and the Repatriation Process for Native California: Proceedings
of a Working Conference. vol. 16. Collection Fourm, San Francisco State University,
pp. 92–95.
Sirois, P., 2000. The analysis of museum objects for the presence of arsenic and mercury:
non-destructive analysis and sample analysis. In: Caldararo, N., Davis, L., Palmer, P.,
Waddington, J. (Eds.), The Contamination of Museum Materials and the Repatriation
Process for Native California: Proceedings of a Working Conference. vol. 16. Collection
Fourm, San Francisco State University, pp. 65–75.
Unger, A., Schniewind, A., Unger, W., 2001. Conservation of Wood Artifacts. SringerVerlag,
Berlin.
US EPA, 1988. TO-13: Polycyclic Aromatic Hydrocarbons (PAH).
US EPA, 1991. OSWER Directive 9355.0-30: Role of the Baseline Risk Assessment in
Superfund Remedy Selection Decisions. United States Environmental Protection
Agency.
US EPA, 2003. Integrated risk information system. http://www.epa.gov/iris.
US EPA, 2005a. Human Health Risk Assessment Protocol for Hazardous Waste Combustion
Facilities (Final). pp. 1–810.
US EPA, 2005b. Human Health Risk Assessment Protocol for Hazardous Waste Combustion
Facilities (Final) - The Hazardous Waste Companion Database (ACCESS). 2016.
United States Environmental Protection Agency.
US EPA, 2009. Risk Assessment Guidance for Superfund Volume I: Human Health Evaluation
Manual (Part F, Supplemental Guidance for Inhalation Risk Assessment). United
States Environmental Protection Agency, Washington, D.C., pp. 1–68.
Wörle, M., Hubert, Vera, Hildbrand, E., Hunger, K., Lehmann, E., Mayer, I., et al., 2012. Evaluation
of decontamination methods of pesticide contaminated wooden objects in
museum collections: efficiency of the treatments and influence on the wooden structure.
J. Cult. Herit. 135, S209–S215.
606 E. Holt et al. / Science of the Total Environment 609 (2017) 598–606
Bečanová, Jitka, Lisa Melymuk, Šimon Vojta, Klára Komprdová, and Jana Klánová. 2016.
"Screening for Perfluoroalkyl Acids in Consumer Products, Building Materials and Wastes."
Chemosphere 164: 322-29. https://doi.org/10.1016/j.Chemosphere.2016.08.112
Screening for perfluoroalkyl acids in consumer products, building
materials and wastes
Jitka Becanova, Lisa Melymuk*
, Simon Vojta, Klara Komprdova, Jana Klanova
Research Centre for Toxic Compounds in the Environment (RECETOX), Masaryk University, Kamenice 753/5, 625 00 Brno, Czech Republic
h i g h l i g h t s
 15 target PFAAs were analyzed in 126 samples of consumer products and building materials.
 In 88% of samples at least one PFAA was detected; dominated by PFOS.
 The highest S15PFAA concentration (77.61 mg kgÀ1
) was found in a textile.
 Several materials contained high levels of unregulated short-chain PFAAs.
 C5-C8-chain PFCAs in wood-based building materials were identified for the first time.
a r t i c l e i n f o
Article history:
Received 3 March 2016
Received in revised form
22 August 2016
Accepted 23 August 2016
Available online 2 September 2016
Handling Editor: I. Cousins
Keywords:
Building materials
Consumer products
Indoor
PFASs
PFOS
a b s t r a c t
Perfluoroalkyl and polyfluoroalkyl substances (PFASs) are a large group of important chemical compounds
with unique and useful physico-chemical properties, widely produced and used in many applications.
However, due to the toxicity, bioaccumulation and long-range transport potential of certain
PFASs, they are of significant concern to scientists and policy makers. To assess human exposure to PFASs,
it is necessary to understand the concentrations of these emerging contaminants in our environment,
and particularly environments where urban population spend most of their time, i.e. buildings and
vehicles.
A total of 126 samples of building materials, consumer products, car interior materials and wastes
were therefore analyzed for their content of key PFASs - 15 perfluoroalkyl acids (PFAAs). At least one of
the target PFAAs was detected in 88% of all samples. The highest concentration of S15PFAAs was found in
textile materials (77.61 mg kgÀ1
), as expected, since specific PFAAs are known to be used for textile
treatment during processing. Surprisingly, PFAAs were also detected in all analyzed composite wood
building materials, which were dominated by perfluoroalkyl carboxylic acids with 5e8 carbons in the
chain (S4PFCAs up to 32.9 mg kgÀ1
). These materials are currently widely used for building refurbishment,
and this is the first study to find evidence of the presence of specific PFASs in composite wood
materials. Thus, in addition to consumer products treated with PFASs, materials used in the construction
of houses, schools and office buildings may also play an important role in human exposure to PFASs.
© 2016 Elsevier Ltd. All rights reserved.
1. Introduction
Perfluoroalkyl and polyfluoroalkyl substances (PFASs) are a
diverse and large group of chemicals which consists of various
polymers and nonpolymer substances (Buck et al., 2011). PFASs are
applied to a wide range of commercial and industrial materials in
order to change their physico-chemical properties (Banks, 1994;
Kissa, 2001) resulting in lowered interfacial or/and surface
tension and enhanced resistance to water, stains, oil, and fire. Major
producers of PFASs are 3M, DuPont, Clariant, Daikin and Asahi Glass
and applied in large volumes on a global scale: e.g. the total production
of perfluorooctane sulfonyl fluoride from 1970 to 2002 was
estimated to be about 100 000 metric tons (Paul et al., 2009). PFASs
are specifically used in applications such as surface treatment of
textiles (e.g. carpets, furniture materials and clothing), leather
products, paper and packaging, coating additives, cleaning agents
and firefighting foams (Paul et al., 2009; Prevedouros et al., 2006;
Wang et al., 2014). In addition, specific PFASs are also used in the
photographic industry, photolithography and manufacturing of* Corresponding author.
E-mail address: melymuk@recetox.muni.cz (L. Melymuk).
Contents lists available at ScienceDirect
Chemosphere
journal homepage: www.elsevier.com/locate/chemosphere
http://dx.doi.org/10.1016/j.chemosphere.2016.08.112
0045-6535/© 2016 Elsevier Ltd. All rights reserved.
Chemosphere 164 (2016) 322e329
semiconductors (OECD, 2002; EFSA, 2008). Since PFASs are typically
used to treat surface layers they can be directly released from
final products into the environment during their lifecycle. Wang
et al. (2014) estimated global emissions of perfluoroalkyl carboxylic
acids (PFCAs) (C4 to C14) from direct and indirect sources between
1951 and 2015 of up to 21 400 metric tons.
The fate of PFASs in the environment as well as their biological
activity are determined by their physico-chemical properties,
which vary greatly with molecular structure (e.g. functional group
and carbon chain length) (Kissa, 2001). In general, neutral PFASs
(e.g. fluorotelomer alcohols, perfluorinated sulfonamides and their
derivatives) may undergo long range atmospheric transport (Lai
et al., 2016) and may be partially degraded into perfluoroalkyl
acids (PFAAs) in the presence of OH radicals (Ellis et al., 2004).
These PFAAs, such as PFCAs and perfluoroalkane sulfonates (PFSAs)
have been shown to persist in the environment (Scheringer et al.,
2014) and have no known significant natural degradation mechanism
(Sulbaek Andersen et al., 2003.). Typically they can undergo
long-range transport and therefore they are widespread in a range
of environmental compartments around the globe (Butt et al.,
2007; Giesy and Kannan, 2001). Humans are exposed to PFASs
through various pathways (e.g. dust ingestion, drinking water, diet
and inhalation) (Enault et al., 2015; Haug et al., 2011; MirallesMarco
and Harrad, 2015). Similarly to the legacy persistent
organic pollutants (POPs, e.g. polychlorinated biphenyls, organochlorine
pesticides), the neutral PFASs accumulate in lipid-rich
tissues (Chu et al., 2016). In contrast, the ionic PFASs, particularly
the longer chain PFAS, bind to blood proteins and accumulate in the
liver, kidney and bile secretions (Ng and Hungerbühler, 2014).
Almost all human blood samples collected around the world were
found to contain ionic PFASs at levels of ng mLÀ1
(Gebbink et al.,
2015; Miralles-Marco and Harrad, 2015; Wang et al., 2015b). Information
about the effects on humans are primarily available for
PFOA and PFOS and are related to early menopause, changes in a
sperm quality and increased risk of cancer (Bonefeld-Jorgensen
et al., 2011; Joensen et al., 2009).
In 2009, PFOS and related compounds were listed under Annex
B of the Stockholm Convention on Persistent Organic Pollutants,
which restricts manufacturing and use to a few specific applications
(Stockholm Convention, 2009). Within the European Union, the use
of PFOS is regulated by European Commission Regulation No. 122/
2006 (EU, 2006a), 552/2009 (EU, 2009) and 757/2010 (EU, 2010)
Intermediate products or articles with concentrations of PFOS
greater than 0.1% by weight are banned since 2007. For textiles or
other coated materials, the amount of PFOS must be lower than
1 mg mÀ2
. Moreover, PFOA, other C9-C14 PFCAs and their salts and
precursors have been identified by the European Chemicals Agency
as a Substance of Very High Concern due to their irreversible effects
on the environment and human health (ECHA, 2015) and presence
of PFOA in articles on the European market at levels higher than
0.1% must be reported (EU Regulation No. 1907/2006 (EU, 2006b)).
The group of regulated PFAAs includes also several other identified
precursors and intermediates (OECD, 2007), which are currently
being replaced by the manufacturers with alternative PFASs; e.g.
PFAAs with shorter (C4 to C6) fluorinated chains or PFASs with
hetero atoms (N, S, O) in their carbon chains (EPA, 2012; Wang et al.,
2013). Use volumes and the environmental and human impact of
these alternative PFASs are currently under scientific focus (Blum
et al., 2015).
Both regulated and non-regulated PFASs are frequently found to
be present in consumer products used on a daily basis in homes,
schools and workplaces. While intensive research has focused on
food packaging materials (Begley et al., 2005; Gallart-Ayala et al.,
2013; Herzke et al., 2012; Martanez-Moral and Tena, 2012; Poothong
et al., 2012; Sinclair et al., 2007; Trier et al., 2011a, 2011b;
Vestergren et al., 2008; Zafeiraki et al., 2014), information regarding
levels found in other daily-use consumer products is limited
(Herzke et al., 2012; Liu et al., 2014; Trier et al., 2011a; Vestergren
et al., 2015; Ye et al., 2015) and no information regarding levels in
building materials is currently available. However, the indoor
environment is a potential source of human PFAS exposure, especially
in view of the fact that people in urban areas spend more than
20 h per day in indoor spaces (Klepeis et al., 2001).
Quantifying PFAS exposure is a complex and challenging process,
not least due to a lack of knowledge concerning the exact PFAS
composition in materials. Thus, a necessary component of quantifying
PFAS exposure is obtaining information on their concentrations
in broad range of materials types, since these materials
release PFASs and contribute to their elevated levels in indoor air
and dust. In this study we analyzed 126 individual samples of
building materials, consumer products, car interior materials and
wastes in order to provide insight into the distribution and
amounts of PFASs added to indoor materials and to identify the
potential sources of specific PFASs to the indoor environment.
2. Experimental
2.1. Sample collection
The aim of this study was to obtain PFAA levels in a representative
selection of different types of materials. Both new and used
materials were included to cover the widest range of materials used
in the construction of buildings, household equipment, interior of
cars and wastes. These included mainly construction materials used
in the past three decades, new and used electrical devices, flooring,
fabric, upholstery and other daily-use materials. Samples of recycled
materials were also included, as their contamination from
primary materials remains unclear. New materials were purchased
while older and used materials were collected from various sources.
These materials were also analyzed for flame retardants, presented
elsewhere (Vojta et al., in prep).
A total of 126 samples were split into four categories according
to use and composition. The first category included household
equipment; i.e. (1A) textiles, (1B) floor coverings, (1C) electrical &
electronic equipment (EEE), and (1D) plastics. The second category
included building materials; i.e. (2A) oriented strand board (OSB),
other composite wood and wood, (2B) insulation materials, (2C)
mounting and sealant foam, (2D) facade materials, (2E) polystyrene
and (2F) air conditioner components. Some building materials were
supplied by a company dealing with the construction of low-energy
houses. In the third category were car interior materials. The final
group consisted of wastes of electrical & electronic equipment
(WEEE) collected at a sorting plant. See Supplementary Materials
Table SI.2 for a detailed list of sample categorization.
2.2. Sample preparation and extraction
Samples of solid materials were crushed, chopped or ground
while samples of foam and fabrics were cut into small pieces. After
grinding or cutting, 5 g of each sample was extracted with methanol
with the addition of ammonium acetate (final concentration
5 mM) using warm Soxhlet extraction (60 min warm Soxhlet followed
by 30 min of solvent rinsing) in a B-811 extraction unit
(Büchi, Switzerland). Following extraction, samples were cleaned
up according to the procedures for PFAAs analysis described in
detail elsewhere (Karaskova et al., 2016). Briefly, concentrated extracts
were filtered using a syringe filter (nylon membrane, 13 mm
diameter and 0.45 mm pore size). The filtrate was concentrated
using a stream of nitrogen in a TurboVap II (Caliper LifeSciences,
USA) concentrator unit to 500 mL and transferred to a minivial.
J. Becanova et al. / Chemosphere 164 (2016) 322e329 323
Finally, samples were diluted using a solution of ammonium acetate
in water (concentration 5 mM) up to a final volume (50/50,
ammonium acetate in water/ammonium acetate in methanol, v/v).
Prior to final analysis, syringe standards (13
C4 PFOA, 13
C4 PFOS)
were added to all samples.
2.3. Quality assurance and quality control
The extraction efficiency of individual PFAAs from analyzed
materials was tested by repeated warm Soxhlet extraction cycles.
Since no PFAAs were detected in repeat extracts, we assumed that
the sample preparation method described above was sufficient to
remove all extractable amounts of PFAAs from materials and
therefore to obtain realistic levels of PFAAs present in samples.
The accuracy of method was evaluated using a set of spiked solid
blank materials (polyurethane foam (n ¼ 6) and sand matrix blank
(n ¼ 10)). Obtained recoveries varied between 70% and 120% for
individual compounds with relative standard deviation up to 21%
for PFCAs and 8.3% for PFSAs (extraction method details are provided
in Supplementary Materials and results for individual compounds
are in Table SI.1b). Laboratory procedural blanks were also
analyzed (n ¼ 10) using empty extraction cartridges and no laboratory
contamination was detected for any of the analyzed compounds.
For quantification of target compounds, set of calibration
standards solution (CS) were analyzed and the instrumental limit of
quantification (ILQ) was calculated from the signal to noise ratio (S/
N ¼ 9) for the lower calibration point (Table SI.1a). Method quantitation
limits (MQL) were calculated from spiked solid blank materials
as a lowest quantifiable concentration and were recalculated
to the sample weights. Precision of measurement was determined
by repetitive injection of QA/QC standard solution as described in
Becanova et al. (2016).
2.4. Instrumental analysis
The separation and detection of 15 PFAAs: short-chain (C5-C7)
PFCAs, long-chain (C8-C14) PFCAs and C4, C6, C7, C8, C10 PFSAs
(listed in full in Table SI.1a) were performed by liquid chromatography
using an Agilent 1100 (Agilent Technologies, Palo, Alto, CA)
equipped with a SYNERGI 4m Fusion RP 80€A 50 mm  2 mm column
(Phenomenex, USA) coupled to a mass spectrometer QTRAP 5500
(ABSciex, CA, USA) interfaced with an electrospray ionization
source (HPLC-ESI-MS/MS). The mass spectrometer was operated in
negative ion mode (EI-) using two MRM transitions for each compound.
For detailed mass spectrometer conditions see Supplementary
Materials Tables SI.3 and SI.4. Aliquots of 10 mL were
injected on the column and eluted using a gradient of 200 mL minÀ1
methanol and water (with 5 mM ammonium acetate). The initial
gradient was set at 45/55 methanol/water, with methanol content
increased to 80% over two minutes (then held for 10 min) and
further increased to 100% methanol (with 5 mM ammonium acetate)
and held for 3 min. The column was equilibrated using the
initial content of the mobile phase for 5 min between injections.
2.5. Statistical analysis
Prior to data treatment, an anomalous sample of a 7-year-old
polyurethane foam with 2 times higher concentrations of target
chemicals compared to the second highest concentrated one was
excluded from the dataset due to detectable levels of all of the
target compounds at unusually high levels (see Table SI.2).
The Kruskal-Wallis test and multiple comparisons of mean
ranks were used for statistical differences of sum PFAAs among
material categories according to use and composition. To identify
PFAA concentration patterns in different types of materials, cluster
analysis was performed. Only PFAAs with more than 10% of values
above detection limits were used for this analysis. The concentrations
were standardized to three levels. Values below or close to
limits of quantification were assigned zeros, values with normal
distribution range were marked as 1 and outlier and extreme values
were marked as 2. This type of transformation allowed direct
comparison of different levels of PFAAs in materials. The samples
without any detected concentrations (with all zeros, in total 33)
were excluded from clustering. The Ward's method with Euclidean
distance was used as the clustering method. All statistical analyses
were performed in STATISTICA (version 12, StatSoft, Inc.).
3. Results and discussion
3.1. PFAA detection frequency and general composition profiles
Two classes of PFAAs (PFSAs and PFCAs) were quantified in 126
samples. PFAAs were found in all but eleven samples (88%). As
expected, the most frequently detected compound was PFOS, with
a detection frequency of 64% (Figure SI.1). The detection frequency
of the remaining PFSAs varied between 27 and 45%. The detection
frequency of the short-chain PFCAs (C5eC7) and PFOA ranged from
21 to 44%; by contrast, the detection frequency of PFCAs with 9e14
carbons in their chain was only 2e5%. Although the low frequency
of PFCAs with carbon chain !10 was predictable due to their recent
limited production and usage in consumer products (Wang et al.,
2014), their high bioaccumulation potential (Martin et al., 2004)
warrants their continued monitoring in consumer products.
As a result of the relation between their median concentration
and detection frequency, PFAAs were split into two major groups
(Fig. 1).
The first group (I.) comprises PFAAs found at low concentrations
with a low detection frequency, and contains long-chain PFCAs
with 11e14 carbons in their chain, as expected considering there is
no reported intensive use of long-chain PFCAs in consumer products
or building materials. The second group (II.) contains compounds
with detection frequencies between 25 and 45% and
median concentrations of up to 0.6 mg kgÀ1
of material. This group
includes both PFCAs and PFSAs with short carbon chains. Some of
these compounds are used as alternatives to banned C8 compounds
(EPA, 2012); moreover, they were identified as impurities occurring
during C8 compound production (Buck et al., 2011). The remaining
ungrouped substances are PFOS, PFOA and PFHpA. PFOS (median
concentration 0.12 mg kgÀ1
and a detection frequency of 64%)
suggests broad usage in materials, as previously reported (Herzke
et al., 2012; Washburn et al., 2005). The C7 and C8 PFCAs have a
similar detection frequency (20 and 14%, respectively) but different
median concentrations (0.88 and 0.37 mg kgÀ1
). Similar results for
short-chain PFCAs have been reported in various applications, i.e.
firefighting foams (Herzke et al., 2012), textiles and food packaging
materials (Begley et al., 2005; Vestergren et al., 2015; Ye et al.,
2015).
3.2. Levels of PFAA in consumer product groups
Total PFAA concentrations in consumer products with detectable
levels ranged from 0.01 to 77.6 mg kgÀ1
. All materials were split
into product use groups (as described in Section 2.1, and Table SI.1)
to generalize the measured concentrations (Table 1).
In the first category (household equipment) there were no statistically
significant differences between individual groups (1A, 1B,
1C, and 1D). The textile group (1A) showed high variability due to
the relatively high number of extremes and outliers, and contained
the highest concentration of
P
15PFAAs (77.6 mg kgÀ1
). Floor coverings
(1B) also covered large ranges, with
P
15PFAAs from 0.3 to
J. Becanova et al. / Chemosphere 164 (2016) 322e329324
38.4 mg kgÀ1
. Within the second category (building materials),
statistically significant differences were found between the OSB
and wood group (2A) and the other three (2C, 2E, 2F) groups. Group
2A exhibited the highest median S15PFAA concentration
(4.87 mg kgÀ1
) with a narrow distribution and 100% detection frequency.
This uniformity could be caused by similarities in the
production for all materials within this group. In the third category
(car interior materials) although all samples were from car interiors,
the materials themselves were very diverse (e.g. plastics,
foams and textiles), and this is reflected in the wide range (from
0.03 to 35.5 mg kgÀ1
) of S15PFAA concentrations. Similar ranges
were also found for groups 1B and 2B, which we hypothesize, is due
to differences in materials type, ages and production technologies
within these groups as well. In contrast, the last category (WEEE)
had a smaller range of S15PFAA concentrations, attributed to similar
type of material (mainly plastics) within this group and the homogenization
that occurs as products enter the waste sorting
stream.
3.3. PFAA composition of specific samples in individual consumer
products categories
As discussed further in section 3.4, for most materials it is not
possible to identify whether the PFAAs originate from production of
the materials or from the product lifetime. However, some
materials exhibited either considerable concentrations of one
specific PFAA or a compound-specific pattern was observed for a
particular group of materials. In the following section the materials
with patterns indicating the intentional addition of PFAAs or their
precursors during production are investigated in detail.
3.3.1. Household equipment
Of the new textile materials, the sample with the highest concentration
of PFAAs was stain resistant upholstery material produced
in 2010, with a predominant contribution of PFHpS
(73.8 mg kgÀ1
) at levels 23 times higher than PFOS (3.2 mg kgÀ1
). No
toxicological characteristics, data on associated human risk or
regulated levels of PFHpS are currently available. Additionally two
carpet samples produced in 2006 and listed as used in Table SI.2
contained high concentrations of
P
15PFAA (8.54 and
38.4 mg kgÀ1
) with a majority of PFOS (6.08 and 19.6 mg kgÀ1
).
Following a recalculation based on the typical face weight of carpets
used in Europe (500e2000 g mÀ2
; Fung, 2002), the surface
concentrations in the carpets were determined to be 16e44 mg mÀ2
and 4.8e13.2 mg mÀ2
, respectively. These levels exceeded the PFOS
limit (1 mg per m2
of coated fabrics) stipulated by the EU directive
by more than 5 times (EU, 2006a). However, since both carpets
were used prior to the analysis, measured concentrations may
originate from both treatment of carpeting with PFAA-containing
stain-proofing sprays/washes during carpet use or intentional
Fig. 1. Median concentration vs. detection frequency of individual PFAAs in 125 samples.
Table 1
Comparison of S15PFAA concentrations within four categories of materials.
Category Group ntotal
a
(npositive) min - max (median)b
mg kgÀ1
Intra-category Kruskal-Wallis test c
Household equipment Textile 1A 23 (19) <MQLd
- 77.6 (1.15) n.s.
Floor covering 1B 9 (9) 0.310e38.4 (2.09) n.s.
EEE 1C 18 (16) <MQL - 11.7 (0.384) n.s.
Plastics 1D 4 (3) <MQL - 0.384 (0.029) n.s.
Building materials OSB and wood 2A 14 (14) 1.39e18.3 (4.87) * (2C, 2E, 2F)
Insulation materials 2B 16 (16) 0.068e34.3 (3.55) *(2E)
Mounting and sealing foam 2C 6 (4) <MQL - 1.28 (0.230) *(2A)
Facade materials 2D 8 (6) 0e24.5 (0.618) n.s.
Polystyrene 2E 5 (1) <MQL - 0.18 (0.00) *(2A, 2B)
Air conditioning 2F 5 (5) 0.057e0.295 (0.135) *(2A)
Car interior materials e 3 10 (10) 0.033e35.5 (1.34) n.a.
WEEE e 4 7 (7) 0.046e2.20 (1.42) n.a.
a
ntotal - total number of samples, npositive - number of positive detected samples within each group.
b
of S15PFAAs.
c
*p < 0.05; n.s. - not statistically significant; n.a. - not applicable; (p values in Table SI.5).
d
MQL e method quantification limit (specified in Table SI.1).
J. Becanova et al. / Chemosphere 164 (2016) 322e329 325
addition during carpet manufacturing. Presence and levels of PFOS
and its alternatives in various textile materials (especially in floor
covering and upholstery materials) are of interest, since direct
contact with these materials is an important exposure pathway,
particularly for small children.
The last group of household equipment (1D - plastics) included
plastics made from recycled materials which might be potential
sources of indoor contamination. However, there is a lack of data
regarding the potential content of PFAAs in these recycled materials.
In our study, no PFAAs were detected in recycled plastics.
Thus, while recycled plastics may be potential sources of compounds
such as brominated flame retardants to the indoor environment
(Hirai and Sakai, 2007), our analysis suggests that they
may not be a significant source of PFAAs.
3.3.2. Building materials
All 14 samples in group 2A (OSB and wood) contained primarily
short-chain PFCAs and PFOA at concentrations ranging from 1.38 to
13.9 mg kgÀ1
for S4PFCAs. The pattern and composition of those
PFCAs were similar in all samples within this group. Although
wood-based material technology has undergone significant
development in many countries over the past 50 years (Buehlmann
et al., 2000) and composition of additives used in composite wood
manufacturing depends on the producer, the most commonly used
resins are based on urea- or phenol-formaldehyde due to their good
performance and low cost (Tang et al., 2009). However, no information
about PFAS concentrations in these resins is available from
their producers. The only information about the usage of PFOSrelated
substances in analogous application is their addition in
sealants and adhesive products (POPRC, 2011, 2012). Therefore, we
considered the addition of these adhesives or other additives to
wooden strips or plates during the production of OSB and woodbased
materials to be a possible source of PFASs. A detailed investigation
of these composite wood materials may be useful in the
future, especially when worldwide production has been growing
steadily since 2008 (FAO, 2009) (74 million m3
per year in 2012 in
Europe and over 117 million m3
in China (FAO, 2013)), as well as due
to the recycling potential of these materials.
In group 2B, two samples of wood fibre insulation (produced in
2010) contained high amounts of PFHpA (20.6 and 28.4 mg kgÀ1
)
and other 5- to 8-carbon chain PFCAs (12.3 and 5.8 mg kgÀ1
). Both
insulations were produced by a company manufacturing
environmentally-friendly materials and are regularly used for the
insulation of floors and walls in modern buildings and especially in
low-energy houses. Following a recalculation using insulation area
and material density, the PFHpA concentration was found to be
61.5 mg mÀ2
(wall insulation) and 181.8 mg mÀ2
(floor insulation).
Although these shortechain PFOS alternatives are currently manufactured
by major producers, there are no current regulations on
short-chain PFCAs in indoor materials in the European Union.
However, at the same time there is scientific pressure on manufacturers
to inform consumers about the usage of these alternatives
(Cousins et al., 2015).
Materials with high concentrations of PFOS were also found in
group 2B. A sample of phenolic insulation foam produced in 2010,
used as an insulating layer on interior walls, contained 22.8 mg kgÀ1
of PFOS, which corresponds to a concentration of 54.7 mg mÀ2
after
recalculation for area. There is no limit for PFOS in building materials,
but the EU legislative limit for carpets (1 mg mÀ2
) as a floor
covering was thus exceeded 50 times in this case. While insulation
is not a strictly equivalent surface layer to carpet, the comparison
with the EU limits for PFOS for carpets as a surface covering suggests
that the levels of PFOS in insulation may be a problematic
contamination source in the indoor environment, especially if used
in open applications.
3.3.3. Car interior materials
We investigated ten samples of plastics, textile and upholstery
materials used for the interiors of two cars: Skoda and Hyundai,
each from ~1995 (Table SI.1). We found PFAAs in all samples in the
same range as in facade and insulation materials (Table 1). One of
the foams used as upholstery material contained 35.5 mg kgÀ1
of
PFOS. As no other PFAAs were present in this sample, we suspect
that PFOS was added during the production of this foam.
3.3.4. Wastes of electrical & electronic equipment (WEEE)
Within this group all materials contained PFOS in the range of
0.07e0.43 mg kgÀ1
and also other PFASs and PFCAs at detectable
levels. Since the materials are homogenized in the waste sorting
systems and therefore the origin and age of individual materials is
not known, the origin of PFAAs in WEEE cannot be identified.
3.4. Possible sources of PFAAs in consumer products
Generally, in a majority of materials (Table SI.2), the origin of
detected PFAAs is not clear and they may originate either from
production of selected materials or from transport and usage. We
can further divide this into four types of PFAA sources: (1) intentional
addition during production to achieve properties associated
with the use of PFAAs; (2) contamination with by-products/
Fig. 2. Detection frequency of PFAAs in new vs. previously used materials.
J. Becanova et al. / Chemosphere 164 (2016) 322e329326
Fig. 3. Dendrogram of material samples (according to raw material and final product application) based on selected PFAA concentrations. The gray scale on the left reflects three
concentration levels (darker grey indicates higher concentrations). Color coding indicates similar materials, and numbers 1 to 5 denote clusters of different composition described
further in the text.
J. Becanova et al. / Chemosphere 164 (2016) 322e329 327
impurities during production or product lifetime; (3) intentional
addition during product use, such as application of stain-proofing
sprays to carpets, and (4) contamination from the surrounding
environment during the product lifetime. In a few cases, products
can be easily linked with one of the PFAA source types, for example,
the car interior foam mentioned above, containing only PFOS,
suggests intentional addition during manufacturing; while the
presence of PFHpA, which is predominantly a breakdown product
of coatings on carpets, textiles and food packaging (Wang et al.,
2015a) suggests breakdown of coatings, potentially beginning
rapidly after product manufacture.
However, in most cases it is very challenging to identify the
exact source of the detected PFAAs. A comparison of new and used
materials offers some insight (Fig. 2). The higher detection frequency
of short-chain PFSAs in older materials suggests possible
uptake of PFSAs during the product lifetime. However, further
interpretation of the differences between new and previously used
products is challenging as the composition of the new and used
product groups was not identical so these differences may also be
influenced by the presence of different product types in new vs.
used groupings (Figure SI.2).
3.5. PFAA patterns in material based on composition
Due to the uncertainties in the origin of PFAAs reported in many
materials (see above), statistical analyses of materials based on
composition of materials was utilized. The goal of this approach
was to link the pattern of PFAAs with the type of manufacturing of
the materials. Firstly, PFAA concentrations were grouped according
to three concentration levels, and secondly a clustering technique
(Ward's method with Euclidean distance) was utilized. Samples
were hierarchically clustered depending on their level of PFAA
concentrations; the resulting dendrogram thus reflects patterns of
materials with similar PFAA content (Fig. 3).
In the dendrogram, the top cluster (marked as number 1) contains
mainly composite wood materials previously grouped in
different categories, i.e. primarily OSB but also wooden floor
covering and wood-based insulation. These samples exhibited the
highest concentrations of short-chain PFCAs, whose likely sources
were discussed in section 3.2. The lower cluster of the dendrogram
contains two main subgroups. The first subgroup (cluster 2) includes
primarily plastics and textiles, all of which contained mainly
C4 and C6 PFASs, i.e. replacements for PFOS. This contrasts with
cluster 4, which contains mainly plastics produced before the 2009
PFOS ban and which thus have a high PFOS content. Cluster 3
contains textile floor coverings (from both cars and buildings) and
foam insulation. The presence of a broad range of PFAS in this group
suggests that these compounds were intentionally added in order
to change textile and foam surface properties. Although the sorption
of PFASs during use may contribute to their concentration, the
presence of both new and used materials in this group thus suggests
intentional addition during manufacturing. The last group
(cluster 5) consists predominantly of plastics and textiles with C4 to
C9 PFSAs. Despite the fact that some of these materials were produced
after 2009, samples in this group also contained PFOS.
However, the levels of short-chain PFSAs were found to be similar
or higher than levels of PFOS, which further supports abovediscussed
changes in PFAS production. All above-mentioned findings
indicate that the PFAA content is associated with material to a
greater extent than with usage.
4. Conclusions
We analyzed perfluoroalkyl substances in 126 individual samples
of building materials, consumer products, car interior
materials and wastes which potentially affect indoor environments
where people spend most of their time. A WEEE category was also
included in the study to facilitate the monitoring of PFAA concentrations
throughout the lifespan of consumer products. The concentrations
of PFAAs in the majority of studied materials suggested
that the presence of these compounds was not caused by the
intentional addition of PFAAs or their precursors to the materials
during the manufacturing, as the levels were typically low and
product groups contained multiple assorted and unrelated PFAAs.
However, regardless of the origin of PFAAs in specific consumer
products (e.g. impurities or degradation of precursors) knowledge
about their levels is important when assessing human exposure.
However in a few particular materials high concentrations of
banned PFOS and unregulated short-chain PFCAs and PFSAs were
found. The concentrations of some individual samples exceeded the
EU limit for PFOS content in consumer products. The highest concentrations
of PFASs were found in construction and textile materials.
As construction materials are normally installed during
building construction or renovation, they may not be under the
control of the occupant. Moreover, some materials, like woodbased
building materials, are not typically considered as potential
sources of PFAAs. Therefore, better consumer awareness of the use
of perfluorinated additives in these materials is desirable. Better
information and awareness is also useful for household equipment,
where the choice and use depends on a resident and potential
unknown indoor exposure may thus be controlled to a greater
extent. Detailed monitoring of indoor environments along with
research regarding new materials and potential testing of PFOS
alternatives is needed. Increased attention should be paid especially
to new polymers and nonpolymer substances introduced by
key PFAS producers, e.g. Novec™, Scotchguard™ and FORAFAC™,
which are currently unregulated.
Acknowledgments
The authors thank all volunteers who provided us with the
household equipment and building materials as well as students
(Vera Bacova, Jana Benesova and Hanka Liskova) and visiting colleagues
from the University of Belgrade (Vesna and Malisa Antic)
who helped with sample collection and preparation. This research
was supported by the Czech Ministry of Education, Youth and
Sports (LO1214 and LM2015051).
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.chemosphere.2016.08.112.
References
Banks, R.E., 1994. Organofluorine Chemistry: Principles and Commercial Applications.
Plenum Press.
Becanova, J., Komprdova, K., Vrana, B., Klanova, J., 2016. Annual dynamics of perfluorinated
compounds in sediment: a case study in the Morava River in Zlín
district, Czech Republic. Chemosphere 151, 225e233.
Begley, T.H., White, K., Honigfort, P., Twaroski, M.L., Neches, R., Walker, R.A., 2005.
Perfluorochemicals: potential sources of and migration from food packaging.
Food Addit. Contam. 22, 1023e1031.
Blum, A., Balan, S.A., Scheringer, M., Trier, X., Goldenman, G., Cousins, I.T.,
Diamond, M., Fletcher, T., Higgins, C., Lindeman, A.E., Peaslee, G., de Voogt, P.,
Wang, Z.Y., Weber, R., 2015. The Madrid statement on poly- and perfluoroalkyl
substances (PFASs). Environ. Health Perspect. 123, A107eA111.
Bonefeld-Jorgensen, E.C., Long, M., Bossi, R., Ayotte, P., Asmund, G., Kruger, T.,
Ghisari, M., Mulvad, G., Kern, P., Nzulumiki, P., Dewailly, E., 2011. Perfluorinated
compounds are related to breast cancer risk in Greenlandic Inuit: a case control
study. Environ. Health 10.
Buck, R.C., Franklin, J., Berger, U., Conder, J.M., Cousins, I.T., de Voogt, P., Jensen, A.A.,
Kannan, K., Mabury, S.A., van Leeuwen, S.P.J., 2011. Perfluoroalkyl and polyfluoroalkyl
substances in the environment: terminology, classification, and
J. Becanova et al. / Chemosphere 164 (2016) 322e329328
origins. Integr. Environ. Assess. Manag. 7, 513e541.
Buehlmann, U., Ragsdale, C.T., Gfeller, B., 2000. A spreadsheet-based decision support
system for wood panel manufacturing. Decis. Support Syst. 29, 207e227.
Butt, C.M., Mabury, S.A., Muir, D.C.G., Braune, B.M., 2007. Prevalence of long-chained
perfluorinated carboxylates in seabirds from the Canadian Arctic between 1975
and 2004. Environ. Sci. Technol. 41, 3521e3528.
Chu, S., Letcher, R.J., McGoldrick, D.J., Backus, S.M., 2016. A new fluorinated surfactant
contaminant in biota: perfluorobutane sulfonamide in several fish
species. Environ. Sci. Technol. 50, 669e675.
Cousins, I.T., Balan, S.A., Scheringer, M., Weber, R., Wang, Z.Y., Blum, A., Diamond, M.,
Fletcher, T., Goldenman, G., Higgins, C., Lindeman, A.E., Peaslee, G., Trier, X., De
Voogt, P., 2015. Comment on “Fluorotechnology is critical to modern life: the
fluoro council counterpoint to the Madrid statement”. Environ. Health Perspect.
123, A170.
ECHA, 2015. ANNEX XV Proposal for a Restriction e Perfluorooctanoic Acid (PFOA),
PFOA Salts and PFOA-related Substances Committee for Risk Assessment and
Committee for Socio-economic Analysis.
EFSA, 2008. Perfluorooctane sulfonate (PFOS), perfluorooctanoic acid (PFOA) and
their salts. EFSA J. 6, 1e131.
Ellis, D.A., Martin, J.W., De Silva, A.O., Mabury, S.A., Hurley, M.D., Andersen, M.P.S.,
Wallington, T.J., 2004. Degradation of fluorotelomer alcohols: a likely atmospheric
source of perfluorinated carboxylic acids. Environ. Sci. Technol. 38,
3316e3321.
Enault, J., Robert, S., Schlosser, O., de The, C., Loret, J.-F., 2015. Drinking water, diet,
indoor air: comparison of the contribution to environmental micropollutants
exposure. Int. J. Hyg. Environ. Health 218, 723e730.
EPA, 2012. Emerging Contaminants e Perfluorooctane Sulfonate (PFOS) and Perfluorooctanoic
Acid (PFOA). United States Environmental Protection Agency.
EU, 2006a. European Union Directive 2006/122/EC of the European Parliament and
of the Council.
EU, 2006b. Regulation (EC) No 1907/2006 of the European Parliament and of the
Council.
EU, 2009. Commission Regulation (EC) No 552/2009.
EU, 2010. Commission Regulation (EU) No. 757/2010.
FAO, 2009. Global Demand for Wood Products. State World's Forestry. 2009, p. 144.
FAO, 2013. Global Forest Products Facts and Figures. State of the World's Forests
2009. FAO Forestry Department, p. 144.
Fung, W., 2002. Coated and Laminated Textiles, first ed. Woodhead Publishing.
Gallart-Ayala, H., Nunez, O., Lucci, P., 2013. Recent advances in LC-MS analysis of
food-packaging contaminants. TrAC Trends Anal. Chem. 42, 99e124.
Gebbink, W.A., Glynn, A., Berger, U., 2015. Temporal changes (1997-2012) of perfluoroalkyl
acids and selected precursors (including isomers) in Swedish human
serum. Environ. Pollut. 199, 166e173.
Giesy, J.P., Kannan, K., 2001. Global distribution of perfluorooctane sulfonate in
wildlife. Environ. Sci. Technol. 35, 1339e1342.
Haug, L.S., Huber, S., Becher, G., Thomsen, C., 2011. Characterisation of human
exposure pathways to perfluorinated compounds - comparing exposure estimates
with biomarkers of exposure. Environ. Int. 37, 687e693.
Herzke, D., Olsson, E., Posner, S., 2012. Perfluoroalkyl and polyfluoroalkyl substances
(PFASs) in consumer products in Norway - a pilot study. Chemosphere
88, 980e987.
Hirai, Y., Sakai, S.-i, 2007. Brominated Flame Retardants in Recycled Plastic Products.
BFR, Amsterdam, The Netherlands.
Joensen, U.N., Bossi, R., Leffers, H., Jensen, A.A., Skakkebæk, N.E., Jørgensen, N., 2009.
Do Perfluoroalkyl compounds impair human semen quality? Environ. Health
Perspect. 117, 923e927.
Karaskova, P., Venier, M., Melymuk, L., Becanova, J., Vojta, S., Prokes, R.,
Diamond, M.L., Klanova, J., 2016. Perfluorinated alkyl substances (PFASs) in
household dust in Central Europe and North America. Environ. Int. 94, 315e324.
Kissa, E., 2001. Fluorinated Surfactants and Repellents. Marcel Dekker Inc., New
York.
Klepeis, N.E., Nelson, W.C., Ott, W.R., Robinson, J.P., Tsang, A.M., Switzer, P.,
Behar, J.V., Hern, S.C., Engelmann, W.H., 2001. The National Human Activity
Pattern Survey (NHAPS): a resource for assessing exposure to environmental
pollutants. J. Expo. Anal. Environ. Epidemiol. 11, 231e252.
Lai, S., Song, J., Song, T., Huang, Z., Zhang, Y., Zhao, Y., Liu, G., Zheng, J., Mi, W.,
Tang, J., Zou, S., Ebinghaus, R., Xie, Z., 2016. Neutral polyfluoroalkyl substances
in the atmosphere over the northern South China Sea. Environ. Pollut. 214,
449e455.
Liu, X.Y., Guo, Z.S., Krebs, K.A., Pope, R.H., Roache, N.F., 2014. Concentrations and
trends of perfluorinated chemicals in potential indoor sources from 2007
through 2011 in the US. Chemosphere 98, 51e57.
Martanez-Moral, M.P., Tena, M.T., 2012. Determination of perfluorocompounds in
popcorn packaging by pressurised liquid extraction and ultra-performance
liquid chromatography-tandem mass spectrometry. Talanta 101, 104e109.
Martin, J.W., Smithwick, M.M., Braune, B.M., Hoekstra, P.F., Muir, D.C.G.,
Mabury, S.A., 2004. Identification of long-chain perfluorinated acids in biota
from the Canadian Arctic. Environ. Sci. Technol. 38, 373e380.
Miralles-Marco, A., Harrad, S., 2015. Perfluorooctane sulfonate: a review of human
exposure, biomonitoring and the environmental forensics utility of its chirality
and isomer distribution. Environ. Int. 77, 148e159.
Ng, C.A., Hungerbühler, K., 2014. Bioaccumulation of perfluorinated alkyl acids:
observations and models. Environ. Sci. Technol. 48, 4637e4648.
OECD, 2002. Hazard assessment of perfluorooctane sulfonate (PFOS) and its salts.
Co-operation on existing chemicals. Organ. Econ. Co-oper. Dev. 362.
OECD, 2007. Report of an OECD Workshop on Perfluorocarboxylic Acids (PFCAs) and
Precursors. Organisation for Economic Co-operation and Development, p. 82.
Paul, A.G., Jones, K.C., Sweetman, A.J., 2009. A first global production, emission, and
environmental inventory for perfluorooctane sulfonate. Environ. Sci. Technol.
43, 386e392.
Poothong, S., Boontanon, S.K., Boontanon, N., 2012. Determination of perfluorooctane
sulfonate and perfluorooctanoic acid in food packaging using
liquid chromatography coupled with tandem mass spectrometry. J. Hazard.
Mater. 205e206, 139e143.
POPRC, 2011. Guidance on Alternatives to Perfluorooctane Sulfonic Acid and its
Derivatives. Report of the Persistent Organic Pollutants Review Committee on
the Work of its Sixth Meeting. Stockholm Convention on Persistent Organic
Pollutants, Geneva.
POPRC, 2012. Technical Paper on the Identification and Assessment of Alternatives
to the Use of Perfluorooctane Sulfonic Acid, its Salts, Perfluorooctane Sulfonyl
Fluoride and Their Related Chemicals in Open Applications. Stockholm
Convention, Geneva.
Prevedouros, K., Cousins, I.T., Buck, R.C., Korzeniowski, S.H., 2006. Sources, fate and
transport of perfluorocarboxylates. Environ. Sci. Technol. 40, 32e44.
Scheringer, M., Trier, X., Cousins, I.T., de Voogt, P., Fletcher, T., Wang, Z., Webster, T.F.,
2014. Helsingør Statement on poly- and perfluorinated alkyl substances
(PFASs). Chemosphere 114, 337e339.
Sinclair, E., Kim, S.K., Akinleye, H.B., Kannan, K., 2007. Quantitation of gas-phase
perfluoroalkyl surfactants and fluorotelomer alcohols released from nonstick
cookware and microwave popcorn bags. Environ. Sci. Technol. 41, 1180e1185.
Stockholm Convention. In: Parties, C.o.t. (Ed.), 2009. Stockholm convention on
persistent organic pollutants (POPs), p. 64.
Sulbaek Andersen, M.P., Hurley, M.D., Wallington, T.J., Ball, J.C., Martin, J.W.,
Ellis, D.A., Mabury, S.A., 2003. Atmospheric chemistry of C2F5CHO: mechanism
of the C2F5C(O)O2 þ HO2 reaction. Chem. Phys. Lett. 381, 14e21.
Tang, X.J., Bai, Y., Duong, A., Smith, M.T., Li, L.Y., Zhang, L.P., 2009. Formaldehyde in
China: production, consumption, exposure levels, and health effects. Environ.
Int. 35, 1210e1224.
Trier, X., Granby, K., Christensen, J.H., 2011a. Polyfluorinated surfactants (PFS) in
paper and board coatings for food packaging. Environ. Sci. Pollut. Res. 18,
1108e1120.
Trier, X., Nielsen, N.J., Christensen, J.H., 2011b. Structural isomers of polyfluorinated
di- and tri-alkylated phosphate ester surfactants present in industrial blends
and in microwave popcorn bags. Environ. Sci. Pollut. Res. 18, 1422e1432.
Vestergren, R., Cousins, I.T., Trudel, D., Wormuth, M., Scheringer, M., 2008. Estimating
the contribution of precursor compounds in consumer exposure to
PFOS and PFOA. Chemosphere 73, 1617e1624.
Vestergren, R., Herzke, D., Wang, T., Cousins, I.T., 2015. Are imported consumer
products an important diffuse source of PFASs to the Norwegian environment?
Environ. Pollut. 198, 223e230.
Vojta, S., Melymuk, L., Becanova, J., Kukucka, P., Kohoutek, J., Klanova, J., 2016.
Screening for halogenated flame retardants in european consumer products.
Build. Mater. Wastes (in. prep).
Wang, Z., Cousins, I.T., Scheringer, M., Hungerbühler, K., 2013. Fluorinated alternatives
to long-chain perfluoroalkyl carboxylic acids (PFCAs), perfluoroalkane
sulfonic acids (PFSAs) and their potential precursors. Environ. Int. 60, 242e248.
Wang, Z.Y., Cousins, I.T., Scheringer, M., Buck, R.C., Hungerbuhler, K., 2014. Global
emission inventories for C-4-C-14 perfluoroalkyl carboxylic acid (PFCA) homologues
from 1951 to 2030, Part I: production and emissions from quantifiable
sources. Environ. Int. 70, 62e75.
Wang, F., Shih, K., Ma, R., Li, X.-y, 2015a. Influence of cations on the partition
behavior of perfluoroheptanoate (PFHpA) and perfluorohexanesulfonate
(PFHxS) on wastewater sludge. Chemosphere 131, 178e183.
Wang, T.Y., Wang, P., Meng, J., Liu, S.J., Lu, Y.L., Khim, J.S., Giesy, J.P., 2015b. A review
of sources, multimedia distribution and health risks of perfluoroalkyl acids
(PFAAs) in China. Chemosphere 129, 87e99.
Washburn, S.T., Bingman, T.S., Braithwaite, S.K., Buck, R.C., Buxton, L.W., Clewell, H.J.,
Haroun, L.A., Kester, J.E., Rickard, R.W., Shipp, A.M., 2005. Exposure assessment
and risk characterization for perfluorooctanoate in selected consumer articles.
Environ. Sci. Technol. 39, 3904e3910.
Ye, F., Zushi, Y., Masunaga, S., 2015. Survey of perfluoroalkyl acids (PFAAs) and their
precursors present in Japanese consumer products. Chemosphere 127,
262e268.
Zafeiraki, E., Costopoulou, D., Vassiliadou, I., Bakeas, E., Leondiadis, L., 2014.
Determination of perfluorinated compounds (PFCs) in various foodstuff packaging
materials used in the Greek market. Chemosphere 94, 169e176.
J. Becanova et al. / Chemosphere 164 (2016) 322e329 329
Vojta, Šimon, Jitka Bečanová, Lisa Melymuk, Klára Komprdová, Jiří Kohoutek, Petr Kukučka,
and Jana Klánová. 2017. "Screening for Halogenated Flame Retardants in European
Consumer Products, Building Materials and Wastes." Chemosphere 168: 457-66.
https://doi.org/10.1016/j.Chemosphere.2016.11.032
Screening for halogenated flame retardants in European consumer
products, building materials and wastes
Simon Vojta, Jitka Becanova, Lisa Melymuk*
, Klara Komprdova, Jirí Kohoutek,
Petr Kukucka, Jana Klanova
Research Centre for Toxic Compounds in the Environment (RECETOX), Masaryk University, Kamenice 753/5, 625 00, Brno, Czechia
h i g h l i g h t s
 PBDE, HBCDD and NFR concentrations were determined in 137 consumer products.
 Differences in HFR content between product groups, esp. textiles, plastic and EEE.
 Differences in HFRs between recycled and virgin plastics.
 Recycled plastics contained low levels of a wide range of HFRs.
a r t i c l e i n f o
Article history:
Received 8 September 2016
Received in revised form
5 November 2016
Accepted 7 November 2016
Available online 14 November 2016
Handling Editor: Myrto Petreas
Keywords:
Building material
Consumer product
Indoor environment
Halogenated flame retardant
Recycled plastic
e-waste
a b s t r a c t
To fulfill national and international fire safety standards, flame retardants (FRs) are being added to a wide
range of consumer products and building materials consisting of flammable materials like plastic, wood
and textiles. While the FR composition of some products and materials has been identified in recent
years, the limited global coverage of the data and the large diversity in consumer products necessitates
more information for an overall picture of the FR composition in common products/materials.
To address this issue, 137 individual samples of various consumer products, building materials and
wastes were collected. To identify and characterize potential sources of FRs in indoor environment, all
samples were analyzed for content of polybrominated diphenyl ethers (PBDEs), hexabromocyclododecanes
(HBCDDs) and novel flame retardants (NFRs).
The most frequently detected were HBCDDs (85%), with the highest median concentration of
S4HBCDDs of 300 mg kgÀ1
in polystyrenes. The highest median concentration of S10PBDEs was found in
recycled plastic materials, reaching 4 mg kgÀ1
. The lowest concentrations were observed for NFRs, where
the median of S12NFRs reached 0.4 mg kgÀ1
in the group of electrical & electronic equipment wastes.
This suggests that for consumer products and building materials that are currently in-use, legacy compounds
still contribute to the overall burden of FRs. Additionally, contrasting patterns of FR composition
in recycled and virgin plastics, revealed using principle component analysis (PCA), suggest that legacy
flame retardants are reentering the market through recycled products, perpetuating the potential for
emissions to indoor environments and thus for human exposure.
© 2016 Elsevier Ltd. All rights reserved.
1. Introduction
Halogenated flame retardants (HFRs) are chlorine or bromine
based compounds added to a broad range of commercial products
to increase their fire resistance. They are used to meet fire safety
regulations in flammable petroleum-based materials such as
polymers e which have increased in production in the past several
decades and which surround us in the form of clothing, furniture,
electronics, cars and computers as well as in the form of combustible
materials such as wood, paper and textiles.
Brominated compounds are frequently used organohalogenated
flame retardants due to their higher trapping efficiency for free
radicals and lower decomposition temperature in comparison with
chlorinated compounds. Brominated flame retardants (BFRs) make
up 22% of the global FR market (Posner et al., 2011), and despite
recent shifts towards non-halogenated flame retardants such as
* Corresponding author.
E-mail address: melymuk@recetox.muni.cz (L. Melymuk).
Contents lists available at ScienceDirect
Chemosphere
journal homepage: www.elsevier.com/locate/chemosphere
http://dx.doi.org/10.1016/j.chemosphere.2016.11.032
0045-6535/© 2016 Elsevier Ltd. All rights reserved.
Chemosphere 168 (2017) 457e466
organophosphate esters (OPEs), BFRs still dominate (>40 000
metric tons/year consumed globally in 2008, compared to ~ 20 000
tons/year for OPEs; (Posner et al., 2011)). Throughout the 1990s and
early 2000s, the BFR market was dominated by polybrominated
diphenyl ethers (PBDEs), hexabromocyclododecane (HBCDD) and
tetrabromobisphenol A (TBBPA). However, due to environmental
and human health concerns, two commercial mixtures of PBDEs
were added to the annexes of Stockholm Convention on Persistent
Organic Pollutants in 2009 (UNEP, 2009), while HBCDD was added
in 2013 (UNEP, 2013). National bans on PBDEs (EU, 2004, 2010;
Wager et al., 2012), Stockholm Convention restrictions and voluntary
manufacturer phase-outs have led to a shift in current BFR
usage and the consequent introduction of alternative flame retardants
(FRs) to the market. The usage of HBCDD is projected to
continue to 2020 (ECHA, 2014) while TBBPA and structurallyrelated
compounds remain high-use BFRs. However, a diverse set
of alternative HFRs, often structurally similar to the restricted/legacy
compounds, is currently seeing wider use. These “novel” flame
retardants (NFRs) include a wide variety of halogenated formulas
that are new to the market or newly observed in the environment
and often serve as replacements for banned PBDE formulations.
Currently, there is a large number of different aromatic and cycloaliphatic
HFRs with varying physico-chemical properties designed
to achieve compatibility with polymers and stability during product
lifetimes. Some of these substances (like TBBPA) can be covalently
bound to materials during the production process, but most
legacy FRs and NFRs are additive compounds, i.e. they are not
bound to the material and thus have a greater potential for release
from final products during their lifecycle. Many of these compounds
are persistent, lipophilic and have been shown to accumulate
in the environment (Besis and Samara, 2012; Covaci et al.,
2009; de Wit et al., 2010; Law et al., 2003; Papachlimitzou et al.,
2012; Venier et al., 2010).
HFRs are used in materials typically found in indoor environments
(homes, vehicles and workplaces) where we typically spend
over 20 h per day (Schweizer et al., 2007). The presence of indoor
sources leads to elevated indoor concentrations of FRs, exacerbated
by lower removal and degradation rates in the indoor environments
due to low advective air movement, reduced photolysis and
biodegradation and controlled climate. In addition to acting as a
source of such compounds to outdoor environment (Bjorklund
et al., 2012; Melymuk et al., 2016), the indoor environments also
represent a significant potential exposure route for humans. In
view of the toxicity of HFRs (Bruchajzer et al., 2011; Darnerud,
2003, 2008; Lyche et al., 2015; van der Ven et al., 2009), understanding
indoor exposure is a key research concern. Indoor air and
dust are frequently analyzed (Blanchard et al., 2014; Brommer et al.,
2012; Brown et al., 2014; Cequier et al., 2014; Dodson et al., 2012;
Fromme et al., 2014; Harrad et al., 2010; Shoeib et al., 2012;
Sj€odin et al., 2008; Takeuchi et al., 2014; Thuresson et al., 2012;
Venier et al., 2016; Whitehead et al., 2015) to cover the main
exposure routes: air inhalation and dust ingestion. Nevertheless,
while the link between consumer product sources and measured
indoor concentrations has been extensively investigated to define
the mechanisms of this transfer (Rauert et al., 2015, 2014a, 2014b;
Stubbings and Harrad, 2014), a comprehensive characterization of
the levels in these sources is a missing piece. Although some work
focusing on selected specific consumer products (Ionas et al., 2015;
Kajiwara and Takigami, 2013; Peaslee et al., 2014; Rani et al., 2014;
Stapleton et al., 2011, 2012) has been carried out, only a limited
number of studies deal with a full range of products/materials
commonly present in the indoor environment (Abbasi et al., 2016;
Kumari et al., 2014). Knowledge of the concentrations of these
chemicals in their presumed sources such as building materials and
consumer products allows us to evaluate their contribution to
concentrations of HFRs in indoor air and dust, and is thus a
necessary component of human exposure assessment.
To determine the typical composition and levels of HFRs in indoor
materials in the Czech Republic, 137 individual samples of
construction materials, electrical and electronic devices, flooring,
fabric, upholstery and other daily use materials including wastes
and recycled products were investigated in order to identify and
characterize potential sources of HFRs in indoor environments. All
materials were analyzed for their content of PBDEs, HBCDDs and
NFRs to provide a broad profile of the composition of HFRs in
materials with different intended applications.
2. Materials and methods
2.1. Chemicals
The target analytes included 10 PBDEs (congeners 28, 47, 66, 85,
99, 100, 153, 154, 183 and 209), 4 HBCDD isomers and 12 NFRs (TBX,
TBP-BAE, PBT, PBEB, TBP-DBPE, HBB, BTBPE, s-DDC-CO, a-DDC-CO,
DBE-DBCH, TBCO, DBHCTD). See Table S 1 for further details. 13
C
labeled internal standards of PBDEs and HBCDDs were added
before extraction. All the analytical standards (native and 13
C
labeled standards) were purchased from Wellington Laboratories
Inc. (Canada). Pesticide residue grade solvents were obtained from
Lab-scan (Poland) and Silica Gel 60 (70e230 mesh) from Merck
(Germany).
2.2. Sample collection
Both new and used materials were included to cover the widest
range of materials used in building construction, household
equipment and cars interiors. These included mainly construction
materials from the past three decades (including some supplied
from a company specializing in “green” low-energy buildings), new
and used electrical and electronic devices, flooring, fabric, upholstery
and other daily-use materials. Samples of recycled materials
were also included, as their contamination from primary materials
remains unclear. New materials were purchased while older and
used materials were collected from various sources. A subset of 126
of these materials was also analyzed for perfluoroalkyl substances
(PFASs), presented elsewhere (Becanova et al., 2016).
A total of 137 samples were split into four categories according
to use and composition. See Supplementary Material and Table S 2
for categorization and sample details.
2.3. Sample preparation, extraction and instrumental analysis
Samples of solid materials were crushed, chopped or ground
while samples of foam and fabrics were cut into small pieces about
1e5 mm in diameter. After grinding or cutting, 5 g of each sample
was extracted with dichloromethane (DCM) using warm Soxhlet
extraction (60 min warm Soxhlet followed by 30 min of solvent
rinsing) in a B-811 extraction unit (Büchi, Switzerland). Concentrated
extracts were cleaned up using H2SO4 modified silica column
chromatography (elution with 40 mL DCM:n-hexane mixture 1:1).
The eluate was concentrated using a stream of nitrogen in a TurboVap
II (Caliper LifeSciences, USA) concentrator unit and transferred
to a GC vial insert. Solvent was exchanged to nonane for gas
chromatography/mass spectrometry (GC/MS) analysis and consequently
to acetonitrile for liquid chromatography/mass spectrometry
(LC/MS) analysis. Prior to instrumental analysis, injection
standards (13
C BDE-77, 138) were added to all samples.
PBDEs and NFRs were analyzed by GC-MS/MS using a 7890A GC
(Agilent, USA) equipped with a 15 m  0.25 mm x 0.10 mm DB5
column (Agilent J&W, USA) coupled to a Quattro Micro MS (Waters,
S. Vojta et al. / Chemosphere 168 (2017) 457e466458
Micromass, UK). Separation and detection of HBCDDs was performed
by HPLC-ESI-MS/MS using an Agilent 1100 (Agilent, USA)
equipped with a LUNA C-18 endcapped (3 mm) 100 mm  2 mm
column (Phenomenex, Torrance, USA) coupled to a tandem mass
spectrometer QTRAP 5500 (ABSciex, USA). See SM for more details.
2.4. Quality assurance and quality control
The efficiency of the extraction method was tested by repeated
warm Soxhlet extraction cycles of selected materials. No target
analytes were detected in subsequent extracts and thus we
assumed that all extractable compounds were extracted in the first
cycle.
The recoveries of individual compounds were determined using
a set of pre-cleaned (8 h in acetone and 8 h in dichloromethane)
polyurethane foam samples (n ¼ 8) spiked with the native analytes
prior to extraction. The recoveries of the 26 target compounds
varied between 70% and 120% (Table S 2).
To calculate limits of detection (LODs) of individual compounds,
eight laboratory procedural blanks were processed using empty
extraction cartridges. An average value plus three times the standard
deviation of the blanks was used as LOD, and the average of
the procedural blanks was subtracted from the samples above LOD.
When a compound was not detected in the blanks, instrumental
LODs were used.
2.5. Statistical analysis
For statistical analysis, relevant compounds were summed according
to their usage and properties, resulting in three groups, i.e.
S10PBDEs, S4HBCDDs and S12NFRs. Additionally, S10PBDEs was
subdivided either into the three PBDE commercial mixtures or
considered as BDE-209 vs. S9PBDEs. The Kruskal-Wallis test and
multiple comparisons of mean ranks were used to identify statistical
differences in BDE-209, S9PBDEs, S4HBCDDs and S12NFRs
among material categories according to use and composition.
Principle component analysis (PCA) based on a correlation matrix
was performed for visualization of relevant sums of FRs in relation
to material type and usage. The sums were logarithmically transformed
before analysis due to their log-normal distribution. As a
first step, the PCA with all variables was computed and the most
important compounds or their sums (estimates by eigenvalues)
were selected. The final PCA was performed on selected sums of FRs
(penta-BDE, octa-BDE, deca-BDE, S4HBCDDs and S12NFRs) and the
others (S9PBDEs, S10PBDEs) were visualized in the same ordination
space for interpretation purposes. Additionally, histograms created
along the most important axes were used for description of variability
and levels of FRs in individual consumer products. For PCA,
all the non-detects were substituted by LOD/2. All statistical analyses
were performed with STATISTICA (version 12, StatSoft, Inc.).
3. Results and discussion
3.1. Overall distribution of FRs
To address the overall compound distribution among all material
samples, median concentrations and detection frequency were
compared (Fig. 1). Six main clusters were identified based on the
combination of concentration, detection frequency and compound
structure.
The most frequently detected compounds (~80% of samples) and
with considerably higher concentrations were the three main diastereoisomers
of HBCDD (Formation I, Fig. 1). The fourth HBCDD
diastereoisomer (d-HBCDD) was detected only in 6.5% of samples,
mostly alongside exceptionally high concentrations of other
diastereoisomers, corresponding with its presence in the commercial
HBCDD mixture only as a trace impurity (Arsenault et al.,
2007). Generally, HBCDDs had the highest maximum concentrations
of all the compounds, reaching 4.4 g kgÀ1
of aÀHBCDD in a
sample of sealant foam. Isomer profiles of HBCDDs in these consumer
products are discussed in greater detail in Okonski et al.
(Submitted). The two isomers of Dechlorane Plus were detected
in ~30% of samples at concentrations up to 40 mg kgÀ1
(Formation
II). The PBDE congeners formed two formations. Formation III is
formed by the main components of the penta-BDE commercial
mixture (BDEs 47, 99 and 100), with a detection frequency of about
20e30%, other penta-BDE congeners (BDEs 28, 66 and 85) and BDE-
153 and 154, which are present in both penta and octa-BDE commercial
mixtures. Together with BDE-183, BDE-153 and 154 formed
a sub-set of octa-BDE components (Formation IV) with a detection
frequency around 10%. The last two formations belong to the NFRs.
Bromobenzenes (Formation V, Fig. 1), including HBB, PBT, PBEB and
TBX, had low concentrations and low detection frequencies. Formation
VI contained the tribromophenoxy and related compounds:
TBP-DBPE, TBP-BAE and BTBPE. TBP-DBPE is a flame retardant
produced in Germany in 1980s and TBP-BAE is a transformation
product of TBP-DBPE (Ma et al., 2012). TBP-DBPE and TBP-BAE were
infrequently detected (<5%), while considerable amounts of BTBPE
were found in electronics-related categories.
The remaining three structurally exceptional NFRs were found
only in samples containing recycled plastic materials or mixed
waste of small electronic devices, which means materials containing
broad spectra of compounds. Although DBHCTD exhibits
relatively high median concentrations, it was detected in only four
samples. Considering the lack of information about this compound
containing both bromine as well as chlorine atoms, its rare occurrence
was expected. A similar scenario applies for DBE-DBCH and
TBCO. Finally, BDE-209, the main component of the deca-BDE
commercial mixture, was found in a relatively small number of
samples, in contrast to its presumed use in broad range of polymers
(Alaee et al., 2003; Stubbings and Harrad, 2014). However, the
maximum amounts were the second highest (after HBCDDs) of all
the compounds, reaching 0.6 g kgÀ1
in a sample of pipe insulation.
3.2. FR concentrations in individual product categories
All samples were categorized according to their usage or
composition (see Table S 1), as described in section 2.2. Of all
analyzed materials, only two (linoleum floor covering and a polystyrene
disposable food box) did not contain any of the target
chemicals. An additional ten samples had only one detectable FR.
On the other hand, four samples contained more than 18 out of 26
(70%) target compounds. These materials included recycled plastics
and heavily used polyurethane foam (further discussed in section
3.4).
An overview of the basic descriptive statistics for individual
product categories is displayed in Table 1. Generally, among all
categories, S10PBDEs ranged from 0.03 mg kgÀ1
to 600 000 mg kgÀ1
,
while S4HBCDDs ranged from 0.7 mg kgÀ1
to 5 800 000 mg kgÀ1
and
S12NFRs reached up to 65 000 mg kgÀ1
.
3.3. Household equipment
In textiles (Group 1A, containing 24 samples), the only notable
FRs were penta-BDE congeners and HBCDDs, both with detection
frequencies over 80%. Interestingly, penta-BDEs were dominated by
BDE-28, found in 88% of the samples, while two main components
of the commercial mixture (BDE-47, 99) were detected only in 29
and 13%, respectively. This may indicate sorption of more volatile
congeners to fabrics during their use period, as suggested by Saini
S. Vojta et al. / Chemosphere 168 (2017) 457e466 459
et al. (2016). Although frequently detected, the concentrations of
PBDEs were low, with a median of S9PBDEs of 0.616 mg kgÀ1
only,
which contrasts with findings in North America (Stapleton et al.,
2012), potentially reflecting market differences. Surprisingly, we
found very low amounts of BDE-209 and HBCDDs, major HFRs used
in textiles in USA and Japan (Shaw et al., 2014). The highest concentration
of S4HBCDDs (544 mg kgÀ1
) was in a curtain sample,
while BDE-209 was only detected in one textile sample at
1.66 mg kgÀ1
.
Although the concentrations were very low, a more consistent
pattern of penta-BDE was observed in 13 floor covering samples
(group 1B), dominated by BDE-47, 99 and 100, while no octa- and
deca-BDEs were detected. Significantly higher (p < 0.05) were the
amounts of S4HBCDDs, especially in three samples of carpets,
reaching a maximum of 1140 mg kgÀ1
. NFR concentrations were
negligible; only DDC-CO was detected.
In addition to penta-BDEs, similar concentrations of octa-BDEs
and a few detections of BDE-209 were found in electric and electronic
equipment (EEE, Group 1C, 19 samples), reflecting the higher
portion of plastic materials in this group, e.g., samples of keyboards,
screens, TV sets and printed circuit boards. The presence of octaBDE
in EEE reflects its typical use for flame retarding of thermoplastics
like acrylonitrile butadiene styrene (ABS) resins, widely
used in such applications (Stubbings and Harrad, 2014). Additionally,
BTBPE was also detected in EEE, reaching a maximum of
317 mg kgÀ1
which corresponds with the use of BTBPE as a
replacement for octa-BDE in ABS, high impact polystyrene (HIPS)
and other thermoplastics (Covaci et al., 2011). Lower, but still
notable, were the amounts of DDC-COs, while concentrations and
detection frequency of HBCDDs were less than in floor coverings or
textiles.
Finally, in the plastics group (1D), two out of the six samples had
substantial concentrations of a wide variety of target chemicals (18
and 22 out of 26 analytes), typically at higher levels (medians of
hundreds to thousands mg.kgÀ1
) than in the other groups of
household equipment.
3.4. Building materials
Building materials were dominated by HBCDDs and BDE-209;
no other flame retardants had any significant levels. The concentrations
of HBCDDs far exceeded the levels in household plastics
(Groups 1C and 1D) and the only other non-recycled materials with
comparable levels of HBCDDs were four carpet samples from the
floor covering group (1B), which could also be considered to
overlap with the building material category.
Only four out of 14 samples of oriented strand board (OSB) and
wood (Group 2A) exhibited elevated concentrations of S4HBCDDs,
with a maximum of 483 mg kgÀ1
. Such a low abundance of FRs in
OSB was quite unexpected, considering their use as building materials
and the content of additive adhesives and resins.
Insulation materials (Group 2B, 18 samples) were dominated by
BDE-209, exceeding 1000 mg kgÀ1
in three samples of glass fiber,
foam glass and phenolic foam insulation and 500 mg kgÀ1
in a
sample of tube insulation foam, discussed in section 3.4. The only
other notable sample from this group was stone-fiber insulation
containing 8470 mg kgÀ1
of S4HBCDDs.
The highest levels of HBCDDs were found in mounting and
sealing foams (Group 2C, 8 samples). With one exception, discussed
in section 3.4., maxima of hundreds of mg kgÀ1
for S4HBCDDs were
common in this group, as well as in facade materials (2D), polystyrene
(2E) and heating, ventilation and air conditioning (HVAC)
components (2F), These four groups had the highest levels of
HBCDDs of all the products. The highest median of 299 mg kgÀ1
was
found in polystyrene (group 2E, 5 samples) which agrees with
typical use of HBCDDs (Alaee et al., 2003; Marvin et al., 2011;
Stubbings and Harrad, 2014) as well as with previously published
data (Rani et al., 2014). Additionally, HVAC components (Group 2F,
5 samples) contained considerable amounts (350e5890 mg kgÀ1
) of
BDE-209.
3.5. Car interior materials and WEEE
These two heterogeneous groups had no BDE-209 and low
concentrations of HBCDDs, contrasting with frequent detection and
relatively higher levels of NFRs and other PBDEs. This pattern was
Fig. 1. Median concentration vs. detection frequency of individual HFRs in all analyzed samples.
S. Vojta et al. / Chemosphere 168 (2017) 457e466460
Table 1
Summary of results for four FR groups in individual product categories.
Category Group BDE-209 S9PBDEs S4HBCDDs S12NFRs
ntotal
(ndetect)
Min - max
(median) mg
kgÀ1
Median
mg kgÀ1
Detection
frequency
%
ntotal
(ndetect)
Min - max
(median) mg
kgÀ1
Median
mg kgÀ1
Detection
frequency
%
ntotal
(ndetect)
Min - max
(median) mg
kgÀ1
Median
mg kgÀ1
Detection
frequency
%
ntotal
(ndetect)
Min - max
(median) mg
kgÀ1
Median
mg kgÀ1
Detection
frequency
%
Household equipment Textiles 1A 24 (1) 1.66e1.66 10.5 14.5 24 (21) 0.119e155 0.720 71.0 24 (20) 4.90e544 19.1 85.5 24 (8) 0.588e3910 4.49 50.0
(1.66) (0.616) (21.2) (3.53)
Floor
coverings
1B 13 (2) 3.77e10.5 13 (6) 0.377e6.73 13 (12) 2.85e1140 13 (6) 0.541e23.7
(7.12) (1.60) (42.5) (1.43)
EEE 1C 19 (3) 2.20e37.8 19 (15) 0.069e178 19 (17) 0.654e288 19 (13) 0.115e359
(2.47) (0.721) (4.65) (40.7)
Plastics 1D 6 (3) 17.6e17 500 6 (2) 708-3910 6 (4) 12-1160 6 (4) 0.138e18
600
(3360) (2310) (367) (497)
Building materials OSB and
wood
2A 14 (0) ND 80.0 32.8 14 (2) 0.044e0.132 0.324 19.0 14 (14) 9.18e483 108 94.8 14 (8) 0.307e36.1 4.05 48.3
(N/A) (0.088) (34.0) (5.58)
Insulation
materials
2B 18 (13) 1.77
e626 000
18 (5) 0.324e20.8 18 (16) 2.47e8470 18 (7) 1.28e22.0
(73.4) (1.02) (63.2) (2.10)
Mounting
and sealing
foam
2C 8 (0) ND 8 (0) ND 8 (8) 31.4e5
810 000
8 (6) 0.526e64.8
(N/A) (N/A) (1340) (5.71)
Facade
materials
2D 8 (3) 2.42e79.9 8 (1) 0.184e0.184 8 (8) 6.47
e233 000
8 (2) 3.96e5.38
(70.2) (0.184) (179) (4.67)
Polystyrene 2E 5 (0) ND 5 (1) 0.085e0.085 5 (4) 33.4
e468 000
5 (2) 0.769e45.5
(N/A) (0.085) (299 000) (23.1)
HVAC
components
2F 5 (3) 350-5890 5 (2) 0.217e18.9 5 (5) 21.3
e531 000
5 (3) 0.845e1090
(394) (9.58) (16,300) (9.46)
Car interior materials e 3 10 (0) ND (N/A) 0.00 10 (8) 0.034e54.0 1.22 80.0 10 (10) 0.709e134 21.7 100 10 (9) 2.13e121 11.3 90.0
(N/A) (1.22) (21.7) (11.3)
WEEE e 4 8 (0) ND (N/A) 0.00 8 (8) 0.068e32
700
144 100 8 (8) 1.49e485 12.9 100 8 (7) 5.74e64 600 426 87.5
(N/A) (144) (12.9) (426)
S.Vojtaetal./Chemosphere168(2017)457e466461
especially clear in waste electrical and electronic equipment
(WEEE), consisting of homogenized mixed electronic waste samples.
Both penta and octa-BDE congeners were frequently detected,
with a median concentration of S9PBDEs of 1.22 mg kgÀ1
for car
interior materials and 144 mg kgÀ1
for WEEE. Beside the main use of
penta-BDE formulations in polyurethane foams, textile and automotive
applications (European Chemicals Bureau, 2000), considerable
amounts detected in electronic waste also suggest their use
in plastics. As car interior materials were considered something of a
crossover category as it was composed of ten samples of plastics,
textile and upholstery materials from two car interiors, relatively
low amounts of target flame retardants, comparable to household
equipment category (Table 1) were expected. The lack of detection
of BDE-209 in contrast with frequent detection of penta- and octaBDE
congeners may indicate differences in the timeline of decaBDE
usage, suggesting that the main stocks of deca-BDE did not
reach the waste disposal/recycling point of the product lifecycle at
the time of sampling.
Together with plastics (1D) and EEE (1C), the highest concentrations
of NFRs (Table 1) were observed in the WEEE category.
NFRs were dominated by DDC-COs and BTBPE, which may reflect
wide use of DDC-CO as a flame retardant in electrical hard plastic
connectors in televisions and computer monitors, wire coatings,
and furniture (Betts, 2006; Hoh et al., 2006; Sverko et al., 2011) and
use of BTBPE as a replacement for octa-BDE in ABS (Hoh et al., 2005;
WHO, 1997). Low levels of bromobenzenes were also detected in
plastics, EEE and WEEE, corresponding with the fact that these
compounds were produced in relatively low amounts and used
locally in specific applications. HBB was used mainly in Japan as an
additive flame retardant to paper, wood, textiles, electronic and
plastic goods (Covaci et al., 2011). In our samples, HBB was found
mainly in WEEE, one sample of air conditioning heat exchanger and
the abovementioned heavily used polyurethane foam. PBT, produced
in Israel, USA and China and used in polyethylene, polypropylene,
polystyrene, latex, textile, rubber and ABS (Covaci et al.,
2011) was found in very low amounts across all categories, but
mainly in plastics, polystyrene, car interior materials and WEEE.
Only one sample of WEEE contained PBEB, which agrees with the
fact that PBEB was produced mainly in 1970s and 1980s in Israel
and USA and used in circuit boards, textiles, adhesives, wire and
cable coatings and polyurethane foam (Covaci et al., 2011; Hoh
et al., 2005; WHO, 1997).
3.6. Identification of FR patterns by PCA
When PCA was used to investigate the relationship and overall
distribution of sums of FRs among all product samples, two main
gradients distributed along two PCA axes explaining a high amount
of variability (70%) were distinguished and identified with corresponding
product classification. The visualization of the distribution
of products between these axes according to the categorization
is displayed in Fig. 2a. The directions of compound sums are plotted
together with ellipses indicating 95% of the samples associated
with the corresponding category. The gradient distributed along
the first axis (Factor I) reflects increasing concentrations of sums of
penta- and octa-BDEs, NFRs and S9PBDEs. Increasing concentration
gradients of these compounds were found in WEEE and also in
household equipment, with both having similar patterns of relatively
low concentrations of a wide variety of these chemicals.
Although less distinct due to even lower concentrations, a similar
pattern was observed for car interior materials. The second axis
(Factor II) represents the correlation with the gradient of the
S4HBCDDs and BDE-209. A contrasting pattern of high concentrations
of only these two compounds was found in building materials.
Moreover, while an even stronger relationship was identified
between the subgroup of insulation (2B) and BDE-209, samples of
household equipment were also distributed along the gradient of
S10PBDEs, reflecting the subgroup of plastics (1D) where recycled
materials containing both lower and fully brominated PBDEs were
present.
With respect to the individual samples, the PCA revealed two
distinct patterns. We hypothesize that this corresponds to two
types of plastics: recycled and virgin (non-recycled) plastics. When
applying these findings to the same PCA ordination space grouped
by material composition (textile, wood, plastics, paper) rather than
product use, these two different plastics can be clearly identified
(Fig. 2b).
Use of penta- and octa-BDEs declined since the early 2000s
(Stapleton et al., 2011) and thus their presence in recycled materials
more likely comes from original materials or usage period than
from intended use in the sampled consumer product/material.
However, recently restricted BDE-209 and HBCDDs (Stubbings and
Harrad, 2014) are more likely to be found in primary use. The low
levels of NFRs suggest that no consumer product with direct
application of NFRs was included in this study; they follow the
pattern of penta- and octa-BDEs reflecting their low production
volumes and rather local or limited use in specific applications.
Generally the lower concentrations of all compounds occurred in
non-plastic materials like textile, paper and wood (Fig. 2b).
Overall, the PCA suggests three main material types according to
flame retardant content and composition (Fig. 2b). Plastic materials
can be divided into (1) virgin plastics with relatively higher
amounts of one or a few particular FRs, added intentionally to
provide a flame retarding effect and (2) recycled plastics with
relatively lower amounts of a wide spectra of FRs not added
intentionally, but rather coming from the original source materials.
The last type (3) contains all remaining non-plastic materials with
negligible target FRs.
The distribution of FRs according to product age was also
considered; the highest concentrations in insulation (2B) and
household equipment were found mainly in new samples (produced
after 2010), however middle and low concentrations were
independent of product age.
For a more comprehensive description of FR content, the histograms
of individual groups were visualized for PCA axes (Fig. 3)
and supplemented by a Kruskal-Wallis test for identification of
intra-category statistical significance (Table S 3). In accordance with
concentration gradients displayed in Fig. 2, the individual factors in
Fig. 3 show the frequency of specific concentrations of the samples
from each particular category for the compound gradient correlating
with the corresponding factor. Thus, the broad histograms
along the Factor I axis indicate large concentration ranges in
products with higher sums of penta and octa-BDE, NFRs and
S9PBDEs, especially in WEEE (4), plastics (1D) and the fraction of
EEE (1C) made of recycled plastics (extracted as a subgroup from
EEE and identified as EEE (1C)R in Fig. 3). Similarly, the contribution
of recycled plastics to the product groups with higher concentrations
on the Factor I axis is clearly revealed in combination with
Fig. 2b. The remaining product groups, especially from the category
of building materials (2) have rather lower concentrations of these
compounds.
Histograms along the second axis, representing the distribution
of concentrations of BDE-209 and HBCDDs, indicate non-recycled
plastics (Fig. 3) and clearly reveal of the dominance of building
materials. Five main groups of HVAC system components (2F),
insulation materials (2B), mounting and sealant foams (2C), facade
materials (2D) and polystyrene (2F) exhibited a typical nonrecycled
plastics pattern.
Due to the high variability of results, no significant intracategory
differences between individual material groups were
S. Vojta et al. / Chemosphere 168 (2017) 457e466462
Fig. 2. Results of PCA showing (a) the relationships and main directions of FR sums and distribution variability among main categories of usage and (b) distribution variability
among different material types in the same ordination space. Ellipses indicate 95% of samples from different categories. Solid lines indicate the direction of increasing compound
concentrations. Dashed lines show supplementary variables (only for visualization, not PCA calculations). The red arrow highlights the sample of highly used polyurethane foam,
discussed in section 3.4. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
found (Kruskal-Wallis, p < 0.05, Table S 3), except for the case of
BDE-209, where the exceptionally high concentrations in insulation
materials (2B) significantly differed from OSB and wood (2A),
mounting and sealant foam (2C) and polystyrene (2E), which had
no positive detections. Similarly, S9PBDEs was below detection in
mounting and sealant foam (2C), significantly differing from other
groups.
3.7. FRs concentrations in individual samples
As suggested by the PCA, samples of household equipment, car
interior materials and WEEE frequently consisted of recycled plastic
materials containing a wide variety of FRs. The two samples most
representative of this pattern were plastics which contained 70 and
85% of target analytes. The first sample was a DVD case and the
second one was a poorly engineered recycled plastic part with
visible chunks of original materials. Although the concentrations of
PBDEs in these samples were the highest of all household equipment,
even the highest level of 21 400 mg kgÀ1
of S10PBDEs in the
recycled plastic part represents only 0.002% of the material weight,
which is much lower than amounts typically applied (3e18% for
penta-BDE and 10e18% for octa-BDE) to the materials when PBDEs
are intentionally used as FRs (Alaee et al., 2003; Gallen et al., 2014;
Stapleton et al., 2011; Stubbings and Harrad, 2014). Similar amounts
of PBDEs were previously reported in electronics components and
other plastics (Chen et al., 2010; Kajiwara et al., 2011; Kumari et al.,
2014). On the other hand, consistently higher amounts were found
in Australian consumer products (Gallen et al., 2014), North
American and Chinese baby products (Chen et al., 2009; Stapleton
et al., 2011) and Japanese electronics (Kajiwara et al., 2011).
Another sample with many FRs was a heavily used seven-yearold
polyurethane foam, containing 21 out of the 26 (81%) target
chemicals. Although this resembles a typical pattern of recycled
plastics (Fig. 2b), we assume that this contamination is more likely
originating from the usage period of the material rather than from
its source materials, and demonstrates the typical passive sampling
behavior of foam, with a higher contribution of lower molecular
weight compounds: the sum of penta-BDE congeners was
129 mg kgÀ1
, the sum of octa-BDEs was 25.3 mg kgÀ1
and BDE-209
was only 1.66 mg kgÀ1
. Since these levels are far below penta-BDE
amounts typically used in polyurethane foam (Stubbings and
Harrad, 2014), we assume this sample was not originally pentaBDE
treated. Furthermore, there is the possibility that PUF and
similar materials could change from acting as a chemical sink to a
secondary source when equilibrium is affected by changes of surrounding
conditions, thus emitting legacy FRs to the indoor environment
at a later stage (Zhang et al., 2009).
As described previously, building materials were dominated by
Fig. 3. Histograms of samples of consumer products along first and second PCA axes (same as Fig. 2). Arrows indicate the direction of increasing compound concentrations (right to
left for Factor I and left to right for Factor II) and r signifies correlation of compound with axis.
S. Vojta et al. / Chemosphere 168 (2017) 457e466464
HBCDDs and BDE-209, typically present in non-recycled plastic
materials. Of all the FRs in this study, the highest amount was found
for S4HBCDDs, which reached 5.81 g kgÀ1
(0.6% w/w) in a sample of
sealing foam. This is close to the minimum concentration added to
expanded or extruded polystyrene (0.8e4.0%; (Alaee et al., 2003;
Marvin et al., 2011; Stubbings and Harrad, 2014). The highest
concentration of BDE-209 from all samples (626 mg kgÀ1
), found in
tube insulation foam, represents 0.06% of the material mass, which
is much less than the typically added levels of deca-BDE of 10e25%
(Weil and Levchik, 2008). However, similar amounts have been
reported in samples of textiles (Ionas et al., 2015), where the flame
retarding function was obviously also neglected.
4. Conclusions
Low levels of FRs were found in most consumer products and
building materials, usually at levels of units of mg kgÀ1
, which are far
below levels which would ensure a flame retarding function. This
finding is consistent with previous studies (Abbasi et al., 2016;
Ionas et al., 2015). The highest mass fraction was found for
HBCDDs (0.6% w/w) in a sealing foam sample, and the highest level
of BDE-209 (0.06% w/w) was found in a tube insulation sample. A
maximum of 0.003% w/w of both penta and octa-BDEs was found in
a sample of mixed electronic waste from a waste sorting system.
Similarly, the NFR with the highest observed concentrations,
Dechlorane Plus, typically added in amounts of 10e35%, was found
at only 0.006% of material mass in a sample also originating from a
waste sorting system.
Based on revealed FR composition patterns, two main types of
polymeric materials were identified. The first was virgin plastic
containing elevated levels of either one or a few FRs, typically BDE-
209 and HBCDDs, which we hypothesize had been added intentionally
to provide a flame retarding effect. This pattern was predominantly
observed in insulation, mounting and facade materials,
polystyrene and air conditioning system components. The second
type was recycled plastic containing lower amounts of a wide variety
of FRs. This pattern was most pronounced in the case of EEE,
WEEE and car interior materials. Since FR concentrations in these
types of products were far below levels capable of providing any
flame retarding effect, this suggests unintended accumulation of
FRs due to recycling of originally flame retarded materials or uptake
from the environment during product use. Although the overall
concentrations of FRs in consumer products and materials was
lower than expected, this suggests both that recycled and virgin
plastics may both act as sources to indoor environments, possibly
releasing a wide suite of FRs, including those that are not in current
use. Thus, recycled plastics may be inadvertently and unintentionally
contaminating the indoor environment with unwanted and
in some cases even currently banned synthetic chemicals with no
desired function. This presents a challenge, as recycling of plastics is
generally a desirable activity, but re-contamination of indoor environments
with currently banned FRs may be a negative sideeffect,
and this requires further attention. Moreover, recycled
plastics can contribute a mixture of FRs (and potentially other
plastic additives) to indoor environments, potentially necessitating
consideration of mixture effects rather than only individual compounds.
Due to their reintroduction from existing stocks (Abbasi
et al., 2014) throughout the recycling phase of the original materials,
these chemicals may be observed not only long after legislative
restrictions have been imposed, but also in places where they
were never employed for any primary purpose. Gearhart and Miller
(2016) indicated that the quality of the recycling process might also
influence the amount and availability of FRs in final recycled
product, suggesting bigger concerns in case of cheaper plastics.
Overall, there is a lack of knowledge regarding the recycling
process, the extent of this issue and potential for indoor
contamination.
Acknowledgments
The authors thank all volunteers who provided us with the
household equipment and building materials and also students
(Vera Bacova, Jana Benesova and Hana Liskova) and visiting colleagues
from the University of Belgrade (Vesna Antic and Malisa
Antic) who helped with sample collection and preparation. This
research was supported by the Czech Ministry of Education, Youth
and Sports (LO1214 and LM2015051).
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.chemosphere.2016.11.032.
References
Abbasi, G., Buser, A.M., Soehl, A., Murray, M.W., Diamond, M.L., 2014. Stocks and
flows of PBDEs in products from use to waste in the U.S. And Canada from 1970
to 2020. Environ. Sci. Technol. 49, 1521e1528.
Abbasi, G., Saini, A., Goosey, E., Diamond, M.L., 2016. Product screening for sources
of halogenated flame retardants in Canadian house and office dust. Sci. Total
Environ. 545, 299e307.
Alaee, M., Arias, P., Sjodin, A., Bergman, A., 2003. An overview of commercially used
brominated flame retardants, their applications, their use patterns in different
countries/regions and possible modes of release. Environ. Int. 29, 683e689.
Arsenault, G., Konstantinov, A., Marvin, C.H., MacInnis, G., McAlees, A.,
McCrindle, R., Riddell, N., Tomy, G.T., Yeo, B., 2007. Synthesis of the two minor
isomers, delta- and epsilon-1,2,5,6,9,10-hexabromocyclododecane, present in
commercial hexabromocyclododecane. Chemosphere 68, 887e892.
Becanova, J., Melymuk, L., Vojta, S., Komprdova, K., Klanova, J., 2016. Screening for
perfluoroalkyl acids in consumer products, building materials and wastes.
Chemosphere 164, 322e329.
Besis, A., Samara, C., 2012. Polybrominated diphenyl ethers (PBDEs) in the indoor
and outdoor environments-A review on occurrence and human exposure. Environ.
Pollut. 169, 217e229.
Betts, K.S., 2006. A new flame retardant in the air. Environ. Sci. Technol. 40,
1090e1091.
Bjorklund, J.A., Thuresson, K., Cousins, A.P., Sellstrom, U., Emenius, G., de Wit, C.A.,
2012. Indoor air is a significant source of tri-decabrominated diphenyl ethers to
outdoor air via ventilation systems. Environ. Sci. Technol. 46, 5876e5884.
Blanchard, O., Glorennec, P., Mercier, F., Bonvallot, N., Chevrier, C., Ramalho, O.,
Mandin, C., Le Bot, B., 2014. Semivolatile organic compounds in indoor air and
settled dust in 30 french dwellings. Environ. Sci. Technol. 48, 3959e3969.
Brommer, S., Harrad, S., Van den Eede, N., Covaci, A., 2012. Concentrations of
organophosphate esters and brominated flame retardants in German indoor
dust samples. J. Environ. Monit. 14, 2482e2487.
Brown, F.R., Whitehead, T.P., Park, J.-S., Metayer, C., Petreas, M.X., 2014. Levels of
non-polybrominated diphenyl ether brominated flame retardants in residential
house dust samples and fire station dust samples in California. Environ. Res.
135, 9e14.
Bruchajzer, E., Frydrych, B., Sporny, S., Szymanska, J.A., 2011. The effect of short-term
intoxication of rats with pentabromodiphenyl ether (in mixture mimic commercial
products). Hum. Exp. Toxicol. 30, 363e378.
Cequier, E., Ionas, A.C., Covaci, A., Maria Marce, R., Becher, G., Thomsen, C., 2014.
Occurrence of a broad range of legacy and emerging flame retardants in indoor
environments in Norway. Environ. Sci. Technol. 48, 6827e6835.
Chen, S.-J., Ma, Y.-J., Wang, J., Tian, M., Luo, X.-J., Chen, D., Mai, B.-X., 2010. Measurement
and human exposure assessment of brominated flame retardants in
household products from South China. J. Hazard. Mater. 176, 979e984.
Chen, S.J., Ma, Y.J., Wang, J., Chen, D., Luo, X.J., Mai, B.X., 2009. Brominated flame
retardants in Children's toys: concentration, composition, and Children's
exposure and risk assessment. Environ. Sci. Technol. 43, 4200e4206.
Covaci, A., Harrad, S., Abdallah, M.A.E., Ali, N., Law, R.J., Herzke, D., de Wit, C.A., 2011.
Novel brominated flame retardants: a review of their analysis, environmental
fate and behaviour. Environ. Int. 37, 532e556.
Covaci, A., Voorspoels, S., Abdallah, M.A.E., Geens, T., Harrad, S., Law, R.J., 2009.
Analytical and environmental aspects of the flame retardant
tetrabromobisphenol-A and its derivatives. J. Chromatogr. A 1216, 346e363.
Darnerud, P.O., 2003. Toxic effects of brominated flame retardants in man and in
wildlife. Environ. Int. (The State-of-Science and Trends of BFRs in the Environment)
29, 841e853.
Darnerud, P.O., 2008. Brominated flame retardants as possible endocrine disrupters.
Int. J. Androl 31, 152e160.
de Wit, C.A., Herzke, D., Vorkamp, K., 2010. Brominated flame retardants in the
Arctic environment - trends and new candidates. Science of the Total
S. Vojta et al. / Chemosphere 168 (2017) 457e466 465
Environment 408, 2885e2918.
Dodson, R.E., Perovich, L.J., Covaci, A., Van den Eede, N., Ionas, A.C., Dirtu, A.C.,
Brody, J.G., Rudel, R.A., 2012. After the PBDE phase-out: a broad suite of flame
retardants in repeat house dust samples from California. Environ. Sci. Technol
46, 13056e13066.
ECHA, 2014. Template for Third Party Submission of Information on Alternatives for
Applications for Authorisation HBCD Use in EPS for Building Applications. ICL-IP
Europe BV.
EU, 2004. Directive 850/2004 of the European Parliament and of the Council.
EU, 2010. Directive 757/2010 of the European Parliament and of the Council.
European Chemicals Bureau, 2000. European Union Risk Assessment Report,
Diphenyl Ether, Pentabromo Derivative (Pentabromodiphenylether).
Fromme, H., Hilger, B., Kopp, E., Miserok, M., Volkel, W., 2014. Polybrominated
diphenyl ethers (PBDEs), hexabromocyclododecane (HBCD) and “novel”
brominated flame retardants in house dust in Germany. Environ. Int 64, 61e68.
Gallen, C., Banks, A., Brandsma, S., Baduel, C., Thai, P., Eaglesham, G., Heffernan, A.,
Leonards, P., Bainton, P., Mueller, J.F., 2014. Towards development of a rapid and
effective non-destructive testing strategy to identify brominated flame retardants
in the plastics of consumer products. Science of the Total Environment
491, 255e265.
Gearhart, J., Miller, G., 2016. Evidence of Electronic Waste, Including Flame Retardants
Chemicals, Recycled into New Consumer Products. BFR2016.
Harrad, S., de Wit, C.A., Abdallah, M.A.E., Bergh, C., Bjorklund, J.A., Covaci, A.,
Darnerud, P.O., de Boer, J., Diamond, M., Huber, S., Leonards, P., Mandalakis, M.,
Oestman, C., Haug, L.S., Thomsen, C., Webster, T.F., 2010. Indoor contamination
with hexabromocyclododecanes, polybrominated diphenyl ethers, and perfluoroalkyl
compounds: an important exposure pathway for people? Environ.
Sci. Technol 44, 3221e3231.
Hoh, E., Zhu, L.Y., Hites, R.A., 2005. Novel flame retardants, 1,2-bis(2,4,6tribromophenoxy)ethane
and 2,3,4,5,6-pentabromoethylbenzene, in United
States' environmental samples. Environ. Sci. Technol 39, 2472e2477.
Hoh, E., Zhu, L.Y., Hites, R.A., 2006. Dechlorane plus, a chlorinated flame retardant,
in the Great Lakes. Environ. Sci. Technol 40, 1184e1189.
Ionas, A.C., Gomez, A.B., Uchida, N., Suzuki, G., Kajiwara, N., Takata, K., Takigami, H.,
Leonards, P.E.G., Covaci, A., 2015. Comprehensive characterisation of flame retardants
in textile furnishings by ambient high resolution mass spectrometry,
gas chromatography-mass spectrometry and environmental forensic microscopy.
Environmental Research 142, 712e719.
Kajiwara, N., Noma, Y., Takigami, H., 2011. Brominated and organophosphate flame
retardants in selected consumer products on the Japanese market in 2008.
Journal of Hazardous Materials 192, 1250e1259.
Kajiwara, N., Takigami, H., 2013. Emission behavior of hexabromocyclododecanes
and polybrominated diphenyl ethers from flame-retardant-treated textiles.
Environ. Sci.-Process Impacts 15, 1957e1963.
Kumari, K., Sharma, J.K., Kanade, G.S., Kashyap, S.M., Juwarkar, A.A., Wate, S.R., 2014.
Investigation of polybrominated diphenyl ethers in old consumer products in
India. Environ. Monit. Assess 186, 3001e3009.
Law, R.J., Alaee, M., Allchin, C.R., Boon, J.P., Lebeuf, M., Lepom, P., Stern, G.A., 2003.
Levels and trends of polybrominated diphenylethers and other brominated
flame retardants in wildlife. Environment International (The State-of-Science
and Trends of BFRs in the Environment) 29, 757e770.
Lyche, J.L., Rosseland, C., Berge, G., Polder, A., 2015. Human health risk associated
with brominated flame-retardants (BFRs). Environ. Int 74, 170e180.
Ma, Y.N., Venier, M., Hites, R.A., 2012. Tribromophenoxy flame retardants in the
great lakes atmosphere. Environ. Sci. Technol 46, 13112e13117.
Marvin, C.H., Tomy, G.T., Armitage, J.M., Arnot, J.A., McCarty, L., Covaci, A., Palace, V.,
2011. Hexabromocyclododecane: current understanding of chemistry, environmental
fate and toxicology and implications for global management. Environ.
Sci. Technol 45, 8613e8623.
Melymuk, L., Bohlin-Nizzetto, P., Kukucka, P., Vojta, S., Kalina, J., Cupr, P., Klanova, J.,
2016. Seasonality and indoor/outdoor relationships of flame retardants and
PCBs in residential air. Environmental Pollution 218, 392e401.
Okonski K., Melymuk L., Kohoutek J. and Klanova J., Hexabromocyclododecane:
concentrations and isomer profiles from sources to environmental sinks,
Submitted
Papachlimitzou, A., Barber, J.L., Losada, S., Bersuder, P., Law, R.J., 2012. A review of
the analysis of novel brominated flame retardants. Journal of Chromatography
A 1219, 15e28.
Peaslee, G., Savagian, L., Maley, A., Greve, E., O'Connell, H., Gearhart, J., Pena, K.,
Roberts, B., 2014. A Study of Halogenated Flame Retardants in Automotive Seats.
BFR2014.
Posner, S., Roos, S., Olsson, E., 2011. Exploration of Management Options for HBCD.
Swerea IVF Project Report 10/11.
Rani, M., Shim, W.J., Han, G.M., Jang, M., Song, Y.K., Hong, S.H., 2014. Hexabromocyclododecane
in polystyrene based consumer products: an evidence of
unregulated use. Chemosphere 110, 111e119.
Rauert, C., Harrad, S., Stranger, M., Lazarov, B., 2015. Test chamber investigation of
the volatilization from source materials of brominated flame retardants and
their subsequent deposition to indoor dust. Indoor Air 25, 393e404.
Rauert, C., Harrad, S., Suzuki, G., Takigami, H., Uchida, N., Takata, K., 2014a. Test
chamber and forensic microscopy investigation of the transfer of brominated
flame retardants into indoor dust via abrasion of source materials. Science of
the Total Environment 493, 639e648.
Rauert, C., Lazarov, B., Harrad, S., Covaci, A., Stranger, M., 2014b. A review of
chamber experiments for determining specific emission rates and investigating
migration pathways of flame retardants. Atmospheric Environment 82, 44e55.
Saini, A., Rauert, C., Simpson, M.J., Harrad, S., Diamond, M.L., 2016. Characterizing
the sorption of polybrominated diphenyl ethers (PBDEs) to cotton and polyester
fabrics under controlled conditions. Science of the Total Environment 563,
99e107.
Shaw, S.D.H., Jennifer, H., Berger, Michelle L., Subedi, B., Kannan, K., 2014. Brominated
flame retardants and their replacements in food packaging and household
products: uses, human exposure, and health effects. In: Snedeker, S.M.
(Ed.), Toxicants in Food Packaging and Household Plastics. Springer-Verlag,
London.
Shoeib, M., Harner, T., Webster, G.M., Sverko, E., Cheng, Y., 2012. Legacy and currentuse
flame retardants in house dust from Vancouver, Canada. Environmental
Pollution 169, 175e182.
Schweizer, C., Edwards, R.D., Bayer-Oglesby, L., Gauderman, W.J., Ilacqua, V.,
Jantunen, M.J., Lai, H.K., Nieuwenhuijsen, M., Kunzli, N., 2007. Indoor timemicroenvironment-activity
patterns in seven regions of Europe. Journal of
Exposure Science and Environmental Epidemiology 17, 170e181.
Sj€odin, A., P€apke, O., McGahee, E., Focant, J.-F., Jones, R.S., Pless-Mulloli, T., Toms, L.M.L.,
Herrmann, T., Müller, J., Needham, L.L., Patterson Jr., D.G., 2008. Concentration
of polybrominated diphenyl ethers (PBDEs) in household dust from
various countries. Chemosphere e halogenated persistent organic pollutants
dioxin 2005-Selected papers from the 25th international symposium on halogenated
environmental organic pollutants and POPs held in toronto. Canada 73,
S131eS136. August 2005.
Stapleton, H.M., Klosterhaus, S., Keller, A., Ferguson, P.L., van Bergen, S., Cooper, E.,
Webster, T.F., Blum, A., 2011. Identification of flame retardants in polyurethane
foam collected from baby products. Environ. Sci. Technol 45, 5323e5331.
Stapleton, H.M., Sharma, S., Getzinger, G., Ferguson, P.L., Gabriel, M., Webster, T.F.,
Blum, A., 2012. Novel and high volume use flame retardants in US couches
reflective of the 2005 PentaBDE phase out. Environ. Sci. Technol 46,
13432e13439.
Stubbings, W.A., Harrad, S., 2014. Extent and mechanisms of brominated flame
retardant emissions from waste soft furnishings and fabrics: a critical review.
Environ. Int 71, 164e175.
Sverko, E., Tomy, G.T., Reiner, E.J., Li, Y.F., McCarry, B.E., Arnot, J.A., Law, R.J.,
Hites, R.A., 2011. Dechlorane plus and related compounds in the environment: a
review. Environ. Sci. Technol 45, 5088e5098.
Takeuchi, S., Kojima, H., Saito, I., Jin, K., Kobayashi, S., Tanaka-Kagawa, T., Jinno, H.,
2014. Detection of 34 plasticizers and 25 flame retardants in indoor air from
houses in Sapporo, Japan. Science of the Total Environment 491, 28e33.
Thuresson, K., Bjorklund, J.A., de Wit, C.A., 2012. Tri-decabrominated diphenyl
ethers and hexabromocyclododecane in indoor air and dust from Stockholm
microenvironments 1: levels and profiles. Science of the Total Environment 414,
713e721.
UNEP, 2009. Stockholm Convention on Persistent Organic Pollutants (POPs).
Amendments to Annexes A, B and C. Secretariat of the Stockholm Convention.
UNEP, 2013. Stockholm Convention on Persistent Organic Pollutants (POPs).
Amendments to Annex A. Secretariat of the Stockholm Convention.
van der Ven, L.T.M., van de Kuil, T., Leonards, P.E.G., Slob, W., Lilienthal, H., Litens, S.,
Herlin, M., Hakansson, H., Canton, R.F., van den Berg, M., Visser, T.J., van
Loveren, H., Vos, J.G., Piersma, A.H., 2009. Endocrine effects of hexabromocyclododecane
(HBCD) in a one-generation reproduction study in
Wistar rats. Toxicology Letters 185, 51e62.
Venier, M., Audy, O., Vojta, S., Becanova, J., Romanak, K., Melymuk, L., Kratka, M.,
Kukucka, P., Okeme, J., Saini, A., Diamond, M.L., Klanova, J., 2016. Brominated
flame retardants in the indoor environment d comparative study of indoor
contamination from three countries. Environ. Int 94, 150e160.
Venier, M., Wierda, M., Bowerman, W.W., Hites, R.A., 2010. Flame retardants and
organochlorine pollutants in bald eagle plasma from the Great Lakes region.
Chemosphere 80, 1234e1240.
Wager, P.A., Schluep, M., Muller, E., Gloor, R., 2012. RoHS regulated substances in
mixed plastics from waste electrical and electronic equipment. Environ. Sci.
Technol 46, 628e635.
Weil, E.A., Levchik, S.V., 2008. Flame retardants in commercial use or development
for polyolefins. Journal of Fire Sciences 26, 5e43.
Whitehead, T.P., Smith, S.C., Park, J.-S., Petreas, M.X., Rappaport, S.M., Metayer, C.,
2015. Concentrations of persistent organic pollutants in California women's
serum and residential dust. Environmental Research 136, 57e66.
WHO, 1997. Flame Retardants: a General Introduction. 192. Environmental Health
Criteria.
Zhang, X.M., Diamond, M.L., Ibarra, C., Harrad, S., 2009. Multimedia modeling of
polybrominated diphenyl ether emissions and fate indoors. Environ. Sci.
Technol 43, 2845e2850.
S. Vojta et al. / Chemosphere 168 (2017) 457e466466
Vojta, Šimon, Lisa Melymuk, and Jana Klánová. 2017. "Changes in Flame Retardant and
Legacy Contaminant Concentrations in Indoor Air during Building Construction, Furnishing,
and Use." Environmental Science & Technology 51 (20): 11891-99.
https://doi.org/10.1021/acs.est.7b03245
Changes in Flame Retardant and Legacy Contaminant
Concentrations in Indoor Air during Building Construction,
Furnishing, and Use
Šimon Vojta, Lisa Melymuk,* and Jana Klánová
Research Centre for Toxic Compounds in the Environment (RECETOX), Kamenice 753/5, 62500 Brno, Czech Republic
*S Supporting Information
ABSTRACT: A newly constructed university building was
selected for targeted assessment of changes in the levels of
flame retardants and legacy contaminants during the installation
of building equipment, furniture, electronics, and first year of
building use. Indoor air samples were collected during several
periods of intensive equipment installation to determine a
relationship between newly introduced equipment and changes
in the concentrations and profiles of contaminants in indoor air.
Samples were analyzed for polybrominated diphenyl ethers
(PBDEs), hexabromocyclododecanes (HBCDDs), and new types
of flame retardants: brominated (BFRs) and organophosphate
esters (OPEs). Additionally, typical outdoor contaminants such
as polychlorinated biphenyls (PCBs) and organochlorine pesticides (OCPs) were also analyzed for comparison. From the set of
90 compounds analyzed here, hexabromobenzene (HBB) and tris(2-chloroisopropyl)phosphate (TCIPP) showed a significant
concentration increase in indoor air concentrations during computer installation and operation, suggesting emission by operating
computers, while an order of magnitude concentration increase in tris(1,3-dichloro-2-propyl)phosphate (TDCIPP) and tri-mcresyl
phosphate (TMTP) was observed after the furniture and carpet was introduced to the computer room, suggesting
furniture or carpet as a source. However, the majority of compounds had no systematic change in concentrations during
equipment installation, indicating that no sources of target compounds were introduced or, that source introduction was not
reflected in indoor air concentrations. Generally, low levels of legacy flame retardants compared to their novel alternatives were
observed.
■ INTRODUCTION
Flame retardants (FRs) are synthetic compounds added to a
wide variety of building materials and consumer products1
in
order to reduce flammability, prevent combustion, and delay
the spread of fire after ignition. They are also frequently
required by many national and international fire safety codes,
standards and regulations. Most FRs are typically added to a
material after polymer manufacture, which considerably lowers
their affinity to the final product and allows them to be released
during the entire lifecycle of the product. Therefore, they have
become widespread,2−13
especially in indoor microenvironments
such as homes, offices, cars, or schools.14
Brominated flame retardants (BFRs) are historically the most
widely used group of organic FRs due to their high efficiency
and low cost. Although there are concerns about the
persistence, bioaccumulation and potential health effects of
many BFRs,15−17
only polybrominated diphenyl ethers
(PBDEs) and hexabromocyclododecane (HBCDD) are
currently legislatively restricted. Penta and octaBDE technical
mixtures and HBCDD are currently banned in North America
and the European Union, and they have been listed by the
Stockholm Convention on Persistent Organic Pollutants.18,19
DecaBDE, which consists mainly of decabromodiphenyl ether
(BDE-209), was phased out of use in the EU in 200820
and
North America in 2013,21
and it was added to the Stockholm
Convention in 2017.22
To replace restricted FRs while continuing to fulfill existing
international fire safety standards, worldwide production has
turned to alternative flame retarding compounds, namely novel
(alternative) halogenated flame retardants (NFRs)15,23
and
chemicals based on organophosphate esters (OPEs).24
For
example, the pentaBDE formulation has been replaced by
Firemaster commercial mixtures, containing 2-ethylhexyl-
2,3,4,5-tetrabromobenzoate (EH-TBB), bis(2-ethylhexyl)-tetrabromophthalate
(BEH-TEBP) and triphenyl phosphate
(TPHP), while octaBDE has been replaced by, for example,
1,2-bis(2,4,6-tribromophenoxy)ethane (BTBPE), commercially
known as FF680, and decaBDE has been replaced by
decabromodiphenyl ethane (DBDPE).10,11,15,25
Halogenated
OPEs are usually used as flame retardants, whereas the
nonhalogenated OPEs are primarily used as plasticizers. In
Received: June 26, 2017
Revised: September 8, 2017
Accepted: September 14, 2017
Published: September 14, 2017
Article
pubs.acs.org/est
© 2017 American Chemical Society 11891 DOI: 10.1021/acs.est.7b03245
Environ. Sci. Technol. 2017, 51, 11891−11899
Cite This: Environ. Sci. Technol. 2017, 51, 11891-11899
DownloadedviaMASARYKUNIVonOctober8,2024at10:09:58(UTC).
Seehttps://pubs.acs.org/sharingguidelinesforoptionsonhowtolegitimatelysharepublishedarticles.
both cases, OPEs are being added to a wide range of various
materials resulting in elevated concentrations detected in the
environment.24,26
Although they are considered as less
persistent than above-mentioned halogenated FRs, they can
still undergo long-range atmospheric transport.27,28
Moreover,
some OPEs were identified as carcinogenic or neurotoxic29
and
due to their higher hydrophilicity, they are reported to be more
accessible via inhalation exposure.30
Although these alternatives
have been chosen as substitutes for banned chemicals, many
have very similar structures and physicochemical properties and
raise similar environmental and health concerns.15,28,29,31−33
Since FRs are found in many consumer products, their
indoor concentrations are often higher than outdoors,3,34
leading to indoor exposure to FRs via inhalation of indoor air
and ingestion of indoor dust;35,36
furthermore, the indoor
environment can act as a source of FRs to the outdoor
environment.37
Indoor air levels of FRs and other semivolatile organic
compounds (SVOCs) are typically measured in in-use
buildings.14,38,39
In these studies, indoor levels are often
attributed to building materials or indoor equipment, however
without measurements of emissions from the materials
themselves, these links between single time-point measurements
and sources are often circumstantial. Although elevated
concentrations of FRs are generally considered to be connected
to the presence of electronics or furniture,35,40−42
correlations
between indoor levels and potential sources are difficult to
identify35,43
and links have only been made in a few studies
with repeated indoor measurements during changes in room
equipment or furnishings.44,45
Stronger links have been made
between FR sources and indoor concentrations, for example,
presence of a television and dust concentrations of HBCDD46
and presence of a computer and levels of PBDEs in air,47,48
however these are limited to legacy FRs. Recently, Dodson et
al.49
identified differences in FR profiles in response to different
flammability standards, suggesting both regulations and equipment
affect the profile of FRs indoors.
Useful information concerning source identification can be
obtained from chamber experiments, which represent a
powerful tool for determining source emission characteristics,
but they are difficult to apply to real indoor conditions.35
Evaluating levels of FRs and other semivolatile organic
compounds (SVOCs) during building construction and
furnishing provides a unique opportunity to identify compound
sources as either building materials, furnishings, consumer
products, or some other aspect of room use. In this study,
indoor air samples were collected during several periods of
extensive equipment installation in a newly constructed empty
building in order to identify potential sources of FRs and other
SVOCs.
We hypothesized that initial concentrations of FRs in the
building reflect FRs associated with the building materials
themselves, whereas any increases in FRs during installation of
building equipment and furnishing indicate emissions from the
newly added products. In contrast, SVOCs with no expected
indoor sources, that is, polychlorinated biphenyls (PCBs),
organochlorine pesticides (OCPs), and polycyclic aromatic
hydrocarbons (PAHs) should show no changes in indoor air
concentrations in response to equipping of the building. The
strategy of sampling during the stages of construction has
previously been applied to measurements of volatile organic
compounds,50−52
but not yet to SVOCs.
■ MATERIALS AND METHODS
Sampling Design. Indoor air was measured in four rooms
of a newly constructed university building (in Brno, Czech
Figure 1. Sampling scheme for computer room (A) and laboratory (B). The asterisk indicates the events when the lecture room was also sampled.
The sample timing for the laboratory also applies for the office.
Environmental Science & Technology Article
DOI: 10.1021/acs.est.7b03245
Environ. Sci. Technol. 2017, 51, 11891−11899
11892
Republic) during equipment and furniture installation. The
rooms were chosen according to their properties and functions
to represent various indoor environments and identify potential
differences in air composition: a computer room (3rd floor, 400
m3
), lecture room (2nd floor, 400 m3
), office (4th floor, 62 m3
),
and laboratory (1st basement, 65 m3
). Indoor air in the two
larger rooms was collected using high-volume active air
samplers (HVAAS) while the smaller office and laboratory
were sampled using low-volume air samplers (LVAAS). The
HVAAS was a Digitel DH-77 sampler (Digitel Elektronik,
Austria) with a flow rate of 30 m3
/h equipped with one glass
fiber filter (47 mm GFF, Whatman) and two polyurethane
foam (PUF) plugs (110 mm diameter, 50 mm length, 0.030 g
cm−3
density, T-3037 Molitan a.s., Czech Republic) to collect
both particulate matter and the gas phase. The LVAAS was a
LVS3 (Sven Leckel Ingenieurburo GmbH, Germany) with a
flow rate of 2.3 m3
/h, equipped with 1 GFF and 2 PUFs (55
mm diameter, 50 mm length).
In the computer room, sampling was repeated eight times
(Figure 1). The first sampling was carried out in the empty
room (event one), followed by subsequent sample collection as
carpet and furniture were added (event two), after computers
were installed (events three to six including room cleaning
episode at event five) and after computers were switched on
(events seven and eight). The room was equipped with 56 allin-one
computers, 56 wood composite chairs, and 12 long
desks. The lecture room was considered as a control without
electronic devices, as it is an identical room to the computer
room. Sample collection occurred after the room was equipped
with carpet and furniture (80 chairs, 60 desks). In the office and
laboratory, air samples were collected twice: before and after
furniture installation. The indoor temperature over the
sampling period was stable, varying by less than ±2 °C. See
Supporting Information (SI) Table S1 for further sampling
details.
All sampling media were precleaned by Soxhlet extraction in
acetone and dichloromethane, 8 h each. After sampling,
exposed media were wrapped in aluminum foil and sealed in
plastic bags for transport to the laboratory, where they were
stored at −18 °C until analysis. Detailed methods for PUF/
GFF preparation and deployment have been described by
Bohlin et al.53,54
Analysis. The target analytes were 10 PBDEs (congeners
28, 47, 66, 85, 99, 100, 153, 154, 183, and 209), 3 HBCDD
isomers, 23 NFRs (TBP-AE, TBX, TBP-BAE, PBT, PBBZ,
PBEB, TBCT, DDC−CO-MA, PBB-Acr, TBP-DBPE, HBB,
BTBPE, EH-TBB, TDBP-TAZTO, s-DDC−CO, a-DDC−CO,
α-DBE-DBCH, β-DBE-DBCH, α-TBCO, β-TBCO, BEHTEBP,
DBHCTD, DBDPE), 15 non-chlorinated OPEs (nonCl-OPEs:
TPrP, TiBP, TnBP, DBPP, TPeP, BDPP, TBOEP,
TPHP, EHDPP, TEHP, TOTP, TMTP, TPTP, TIPPP,
TDMPP), 3 chlorinated OPEs (Cl-OPEs: TCEP, TCIPP,
TDCIPP), 9 PCBs (congeners 9, 11, 28, 52, 101, 118, 138, 153,
180), 12 OCPs (PeCB, HCB, α-HCH, β-HCH, γ-HCH, δHCH,
o,p′-DDE, p,p′-DDE, o,p′-DDD, p,p′-DDD, o,p′-DDT,
p,p′-DDT), and 15 PAHs (ACY, ACE, FLU, PHE, ANT, FLA,
PYR, BAA, CHR, BBF, BKF, BAP, IDP, DBA, BGP). See SI
Table S2 for full compound names and further details;
abbreviations of FRs are adopted from Bergman et al.55 13
Clabeled
or deuterated internal standards were added before
extraction. All the analytical standards were purchased from
Wellington Laboratories Inc. (Canada). Pesticide residue grade
solvents were purchased from Lab-scan (Poland) and Silica Gel
60 (70−230 mesh) from Merck (Germany).
The exposed PUFs and GFFs were extracted with dichloromethane
using an automated warm Soxhlet extraction system
and the extracts were purified using column chromatography.
PBDEs and NFRs were analyzed using gas chromatography
coupled to high resolution mass spectrometry, whereas PCBs,
OCPs, and OPEs were analyzed by gas chromatographytandem
mass spectrometry systems. HBCDDs were analyzed
after exchanging solvent to acetonitrile by liquid chromatography
electrospray ionization mass spectrometry. All the
analytical procedures used here were published previ-
ously39,56−59
and are described in detail in the SI.
QA/QC. The recoveries of individual compounds were
determined using a set of pre-cleaned (8 h in acetone and 8 h in
dichloromethane) PUFs (n = 8) spiked with the native analytes
prior to extraction. The recoveries of individual compounds are
given in SI Table S2.
To calculate method detection limits (MDLs) of individual
compounds for each matrix, eight laboratory procedural blanks
were processed using precleaned sampling media. An average
value plus three times the standard deviation of the blanks was
used as MDL, and the average of the procedural blanks was
subtracted from the samples above MDL. When a compound
was not detected in the blanks, instrumental detection limits
were used (SI Table S2).
For statistical data analysis, values below detection were
substituted by √2/2·MDL.60
The Kruskal−Wallis test and
multiple comparisons of mean ranks were used to investigate
statistical differences. All statistical analyses were performed
with STATISTICA (version 13, StatSoft, Inc.) or Microsoft
Excel.
■ RESULTS AND DISCUSSION
Overview of Indoor Air Concentrations. Order of
magnitude differences were observed between the different
compound groups, reflecting differences in use and sources
(Figure 2; summary statistics for individual compounds are
given in SI Tables S4−S7). The lowest concentrations were
found for PBDEs; in all four rooms, the ∑10PBDEs in bulk (gas
+particle phase) air ranged from 0.803 pg m−3
to 9.83 pg m−3
with a median of 2.28 pg m−3
, and the dominant congeners
were BDE 47 and 99. These indoor air levels were very low
compared to North America (∑PBDEs = 1260 pg m−3
in a
U.S. institutional building,61
148 pg m−3
in U.S. apartments, 60
pg m−3
in Canadian apartments39
) and other European
homes62
(∑PBDEs = 223 pg m−3
), which was expected
considering that the building was newly constructed and
equipped after the restrictions on PBDEs. On the other hand,
similar or even lower amounts of PBDEs have been reported in
family houses and apartments in France63
and Czech
Republic.39
A similar result was observed for HBCDDs, where
consistently low concentrations were found. The ∑4HBCDDs
was dominated by α-HBCDD and bulk air concentrations
ranged from 0.0988 pg m−3
to 13.6 pg m−3
with a median of
5.63 pg m−3
. These relatively low levels are in agreement with
the building documentation, specifying that no polystyrene was
used for building insulation.
Twelve of the 23 target NFRs were consistently detected in
indoor matrices: DBE-DBCH, TBP-BAE, PBBZ, PBT, PBEB,
TBP-DBPE, HBB, PBB-Acr, EH-TBB, BEH-TEBP, BTBPE,
and DDC−CO. In contrast to the restricted PBDEs and
Environmental Science & Technology Article
DOI: 10.1021/acs.est.7b03245
Environ. Sci. Technol. 2017, 51, 11891−11899
11893
HBCDD, elevated levels of many replacement NFRs were
observed, in particular for five compounds: HBB, PBBZ, PBT,
TBP-DBPE, and DBE-DBCH. The concentrations of ∑23NFRs
ranged from 12.3 pg m−3
to 1820 pg m−3
, with a median of 158
pg m−3
. The highest median concentration of an individual
NFR was for HBB, at 97.0 pg m−3
(range 1.29 to 1400 pg m−3
).
Concentrations of NFRs in this study are in general agreement
with those published previously,25,39,62,64
with the exception of
two compounds. The range of concentrations and median of
HBB is larger than those reported by Cequier et al.62
(median:
4.11 pg m−3
in Norway), Newton et al.25
(median: 3.6 pg m−3
in Sweden), Tao et al.64
(median 53 pg m−3
in UK) and Venier
et al.39
(medians of 4.0, 5.8, and 4.6 pg m−3
in the U.S., Canada,
and Czech Republic, respectively). The median concentration
of DBE-DBCH of 9.28 pg m−3
from our study is lower than
that found by Cequier et al.62
(median: 77.9 pg m−3
), Newton
et al.25
(median: 55 pg m−3
) and Tao et al.64
(median 280 pg
m−3
).
The indoor concentrations of banned PCBs and OCPs were
of a similar order of magnitude to the in-use NFRs. The median
concentration of ∑9PCBs was 147 pg m−3
(range 14.1 pg m−3
to 452 pg m−3
), dominated by congeners 11 and 28; while the
median concentration of ∑12OCPs was 59.8 pg m−3
(range
38.4 pg m−3
to 118 pg m−3
), and this was dominated by HCB,
α-HCH, γ-HCH, and p,p′-DDE.
Two compound classes were orders of magnitude higher: the
OPEs and PAHs (Figure 2). Twelve out of the 18 OPEs were
consistently detected: TIBP, TNBP, TCEP, TCIPP, DBPP,
TDCIPP, TBOEP, TPHP, EHDPP, TEHP, TMTP, and
TIPPP. The ∑18OPEs ranged from 12 900 pg m−3
to 63 300
pg m−3
with a median of 30 700 pg m−3
, with the major
contribution from TCIPP (75% on average). Similar levels of
TCIPP, the highest OPE contributor in our study (median 22
500 pg m−3
) were reported in Cequier et al.62
(median 42 300
pg m−3
), but Cequier et al. observed 20 times higher
concentrations of TBOEP in school classrooms (median
12 900 pg m−3
) than in residential living rooms (median 598
pg m−3
), attributed to floor polish. Our findings, with a median
of 1230 pg m−3
for TBOEP, suggest no floor polish during the
sampling period, as expected considering that the rooms are
carpeted. The amounts of TCEP found in our study were 1−2
orders of magnitude lower compared to available literature
data,62,65,66
corresponding with its gradual substitution with
TCIPP prior to the building construction.24,27,30
The median of ∑15PAHs was 7200 pg m−3
(range 5200 pg
m−3
to 10 420 pg m−3
) with the highest contribution from
phenanthrene. These order of magnitude higher concentrations
for OPEs and PAHs are typical for these compound
classes,62,63,67
and emphasize that the strongest control on
overall indoor levels is the magnitude of use and/or emission.
Further, a comparison with outdoor levels from a concurrent
study at the same location54
provides an additional indication of
compound sources. Concentrations of PBDEs, OCPs, and
PCBs were similar in indoor and outdoor air, excepting for
PCBs 28 and 52, which were 10× higher indoors, whereas
concentrations of NFRs were consistently higher indoors, and
concentrations of PAHs were 10× higher outdoors. This
suggests the outdoor air as a source to indoors for PAHs, while
indoor environments are a source for in-use NFRs, in keeping
with the hypothesis of the study.
The majority of the target compounds were found
predominantly (>75%) in the gaseous phase (i.e., sorbed to
the PUF in the active air samplers). Only HBCDDs and all
OPEs had significant particle phase contributions (see SI
Figure 2. Indoor air concentrations (gas + particle phase) of PBDEs,
HBCDDs, OCPs, PCBs, NFRs, OPEs, and PAHs. The central lines
indicate the median, the boxes show the interquartile range (25th/
75th percentiles), and the whiskers indicate the maximum and
minimum concentrations.
Figure 3. Concentration changes of sums of target compounds during the equipment installation. The event numbers correspond to the sampling
scheme given in Figure 1. PAH and OPE data for laboratory and office are not available.
Environmental Science & Technology Article
DOI: 10.1021/acs.est.7b03245
Environ. Sci. Technol. 2017, 51, 11891−11899
11894
Tables S4−S7). There were no significant spatial or temporal
differences in the gas-particle partitioning.
Temporal Changes in Concentrations. In keeping with
the hypothesis of outdoor sources of the legacy compounds and
PAHs, and indoor sources of the emerging FRs, we expected
temporal changes over the course of equipment installation for
NFRs and OPEs, and no such changes for PCBs, OCPs, PAHs,
and PBDEs. The measurements in the computer room at
different time-points support this hypothesis. No significant
change in concentrations of PBDEs was observed during
equipment installation except for event five in the computer
room where increased activity in the room (due to room
cleaning) is reflected in slightly higher concentrations. We
assume that no direct sources of PBDEs to the air were present
in building materials or introduced by the furniture or
equipment. Similarly, no temporal trend was observed for
HBCDDs, PAHs, PCBs, or OCPs (Figure 3).
Six NFRs had notable temporal trends, and these are shown
in more detail in Figure 4. An order of magnitude increase in
the concentration of HBB in bulk air was observed at events
seven and eight, that is, after computers were switched on. The
same pattern was observed also for PBBZ and TBP-DBPE,
although at lower concentrations. A similar pattern was also
identified for TBP-DBPE-related tribromophenoxy compounds:
TBP-AE and TBP-BAE, and for PBB-Acr, another
bromobenzene compound, at much lower concentrations
(median about 0.1 pg m−3
). Concurrently, another concentration
increase was observed between event one and two for
PBT and DBE-DBCH (Figure 4). No significant temporal
changes were observed for PBEB, EH-TBB, BEH-TEBP,
BTBPE, and DDC−CO. This suggests that three bromobenzene
compounds, dominated by HBB, and three tribromophenoxy
compounds, dominated by TBP-DBPE, were emitted by
operating computers. This is supported by concentrations from
the laboratory and office, where no electronics were present,
and only background concentrations of these compounds were
found (Figure 3). Moreover, intermediate concentrations
(median of 338 pg m−3
for HBB) were found in the lecture
room, which only contained one computer and overhead
projector (Figure 3).
Different OPEs had different temporal trends. Similarly to
PBT and DBE-DBCH, an order of magnitude concentration
increase in four OPEs (TDCIPP, TCIPP, TEHP, and TMTP)
was observed after the furniture and carpet was introduced to
the computer room (Figure 5), which directly suggests
furniture, carpet (or associated products used in carpet
installation) as a source of these OPEs and NFRs. TIPPP
and DBPP exhibited high concentrations in the initial sample,
followed by a gradual decrease over time until Event 7
(computers switched on), where another substantial increase
occurred (Figure 5). TCIPP also had a distinct increase in
concentrations at sample 7. Thus, a computer-related source is
suggested for TIPPP, DBPP, and TCIPP. TCEP and EHDPP
also had concentration peaks at Event 7. TBOEP decreased
gradually over time. No temporal trends were observed for
TiBP, TnBP and TPHP.
Spatial Differences between Rooms. As mentioned
above for HBB, the comparison of rooms within the same
building should further allow us to distinguish between sources
due to building materials, which are common to all rooms, vs
sources from room equipment and furnishings, which are
unique to individual rooms. No striking difference between the
room types was observed in the case of PBDEs (medians of
∑10PBDEs from 1.41 pg m−3
to 6.61 pg m−3
), HBCDDs
(medians of ∑3HBCDDs from 0.7841 pg m−3
to 6.54 pg m−3
),
OCPs (medians of ∑12OCPs from 55.6 pg m−3
to 82.6 pg
m−3
), or PAHs (medians of ∑15PAHs from 6931 pg m−3
to
7270 pg m−3
) (Figure 3). A slightly higher concentration of
∑9PCBs was noted in the lecture room, due to high
concentrations of PCB 28 and 52, as mentioned above. The
median of ∑9PCBs in the lecture room was 420 pg m−3
compared to 155 pg m−3
in the computer room and 35.4 pg
m−3
the office and laboratory. The higher concentrations of
lighter PCBs in the lecture room may be due to emission from
nearby excavation and construction, but no source can be
concretely identified.
Concentrations of NFRs in the office and laboratory were
similar to the concentrations in the computer room before the
addition of the computers (Figure 3), supporting the result of
computer operation as a source of HBB, TBP-AE, TBP-BAE,
PBB-Acr, PBBZ, and TBP-DBPE, rather than any source from
building materials or flooring. For example, when electronics
were switched on in the computer room in our study,
Figure 4. Changes in air concentrations (pg m−3
) of NFRs over time
in computer room. Sample numbers (x-axis) correspond to the
sampling events identified in Figure 1.
Environmental Science & Technology Article
DOI: 10.1021/acs.est.7b03245
Environ. Sci. Technol. 2017, 51, 11891−11899
11895
concentrations of HBB jumped from hundreds to thousands of
pg m−3
, while the median concentration of HBB in nonelectronics
equipped rooms was only 6.71 pg m−3
. Similarly, a
maximum concentration of 297 pg m−3
and median of 4.11 pg
m−3
from 37 dwellings in Cequier et al.62
suggest varying
numbers of electronics in the study environments. The median
concentration of ∑18OPEs from the lecture room (26 300 pg
m−3
) agrees with the median (26 900 pg m−3
) of events 1 to 6
(before the computers were turned on) from the computer
room, likely reflecting the same technology used in both rooms.
Identification of Potential FRs Sources. Although the
levels of flame retardants in indoor environments are often
considered to be connected to presence of electronics or
upholstered furniture in a particular microenvironment,
evidence is usually indirect.33,35
The identification of potential
sources of contamination is often hindered by factors including
sampling techniques,3
source misclassification,35
or experimental
conditions,43
such as temperature, a key parameter for the
emission of FRs from operating electronics.68
Moreover, the
life-cycle stage of the product plays an important role in the
emission rates of the compounds, especially in case of more
volatile flame retardants.35,45,68,69
This also applies for another
important group of consumer products often considered a
source of FRs, that is, building materials.1,70,71
Materials initially
have a higher emission for volatilization emissions of
SVOCs,16,44
and thus sampling in a recently constructed
building with the presence of new materials provides a stronger
opportunity to identify emissions from building materials and
furnishings.
Operating computers were identified as the most notable
source of FRs to the indoor air, primarily for HBB and TCIPP.
HBB emissions from computers agree with its suggested use in
electronic and plastic goods,15
but possible emission of TCIPP
from computers is quite surprising, considering that TCIPP is
mainly used as FR in insulation and foams.24,71
Additionally,
temporal trends suggest computers as a source of TBP-DBPE
and related tribromophenoxy compounds (TBP-AE and TBPBAE),
PBB-Acr and PBBZ. TBP-DBPE is used as an FR in
polypropylene72
and as an additive FR to acrylonitrilebutadiene-styrene
(ABS) plastics.73
Although TBP-AE is used
in expanded polystyrene, considering the low concentration
levels detected, together with TBP-BAE, is a possible
transformation product or impurity10,74
of TBP-DBPE.10,74
PBB-Acr is a reactive FR which little is known about,75
whereas
PBBZ may be a reactive debromination product of HBB,75
and
has been identified in 30% of electronic product wipes in a
study in Canada.76
Thus, it could be that elevated operating
temperatures within computers could be resulting in the
formation of some brominated products from the additive FRs,
and the concurrent volatilization losses of both the parent FRs
and products. However, more specific studies (e.g., emission
chambers) are necessary to confirm this hypothesis. Finally,
elevated concentrations of DBPP, TIPPP, TCEP, and EHDPP
may also (in part) be connected to emissions from operation of
computers. These compounds are used as FR in PVC, polyester
resins and coatings (for TCEP), PVC and thermoplastics (for
TIPPP) or as plasticizers (EHDPP).
Moreover, we emphasize that the increase in emissions
comes from operation of the computers, not simple presence of
the computers; no significant increase in concentrations was
observed in the room with the introduction of 56 brand new
idle computers (Event 6); rather, the jump in air concentrations
occurred when computers were switched on (Event 7). We also
note that compound physicochemical properties are expected
to have an influence on the patterns and time trends of
emissions. The two main chemicals detected in the indoor air,
HBB and TCIPP might be emitted from the source primarily in
the initiation phase due to their high volatility and other less
volatile chemicals might be released as time goes on, gradually
increasing their total mass emitted.
Two NFRs (DBE-DBCH and PBT) and four OPEs
(TDCIPP, TCIPP, TEHP, and TMTP) are suggested as
having sources in either furniture or carpet. The room did not
contain any significant foam-containing furniture, which
contrasts with typically reported uses of TCIPP in insulation
and foam. TDCIPP, TEHP, and TMTP are used in plastic,
textile and PUF (TDCIPP), PVC, cellulose, paints and
coatings, rubber, and PUF (TEHP), and hydraulic fluids,
cellulose, coatings, polystyrene, plastics, and thermoplastics
(TMTP);24
thus, use in carpet textile, carpet backing or
furniture coatings may account for their increase in emissions in
conjunction with installation of furniture and carpet. Sources of
DBE-DBCH and PBT are less clear. DBE-DBCH is an additive
FR in polystyrene and polyurethane, electronic cable coating
Figure 5. Changes in air concentrations (pg m−3
) of OPEs over time
in computer room. Sample numbers (x-axis) correspond to the
sampling events identified in Figure 1.
Environmental Science & Technology Article
DOI: 10.1021/acs.est.7b03245
Environ. Sci. Technol. 2017, 51, 11891−11899
11896
and construction materials, whereas PBT is used in various
polymers, latex, textiles, rubber, and ABS.75
Overall, operation of computers increased total air
concentrations of NFRs by 886%, from 167 pg m−3
(average
before) to 1480 pg m−3
(average after), and installation of
flooring and furniture increased total air concentrations of
OPEs by 310%, from 12 900 pg m−3
(before) to 40 000 pg m−3
(after). While the elevated concentrations may not persist in
the long term, this suggests that the indoor burden of FRs is
strongly linked to the products and materials used indoors, and
consequently, FR burdens could be reduced with appropriate
selection of products that use methods other than synthetic
organic flame retardants to meet flammability standards.
However, even in this targeted study with samples in
conjunction with changes in room equipment, definitive
identification of sources is difficult, emphasizing the general
challenges in correlating between putative sources and FRs
concentrations in air.35,43
Besides the main issue with source
classification, the main condition to achieve a good correlation
is that the presence of the FR must be clearly reflected in the air
concentration. Better correlations with potential sources are
frequently reported for indoor dust,35
as elevated concentrations
and higher detection frequencies are usually obtained
due to the presence of the entire material particles introduced
by the abrasion migration pathway.77
Analysis of such source
material particles provides a good link with a potential source,
but also has the disadvantage of spatial variability driven by the
distance from the source.3,67,78,79
Thus, regardless of the
correlation with the sources, indoor air and dust sampling
provides complementary information to characterize the indoor
environment, covering various real-world factors (sinks, mass
dynamics), which are difficult to simulate in chamber emission
experiments. While chamber experiments are a useful tool for
characterizing sources by determining specific emission rates,
environmental information is more relevant for human
exposure. Moreover, material samples are usually investigated
by chamber emission experiments, rather than complex
dynamic systems such as operating computers.
A source introduction causing even a marginal increase in air
concentrations of chemicals is still relevant from the viewpoint
of long-term human exposure. For example, the scenario
demonstrated in this study typically applies to open-concept
office environments and other computer rooms, where people
often spend substantial time. For example, the daily inhalation
exposure to HBB for a 70 kg adult, spending 8 h per day in the
computer room from our study will increase from 205 pg kg−1
day−1
to 2094 pg kg−1
day−1
(see SI for the details) when the
computers are switched on.
Moreover, while both brominated and organophosphate FRs
were identified as having in-room sources, there are large
concentration differences between the two compound groups;
the median concentration of TCIPP found in this study was
more than 200 times higher than that of HBB, emphasizing the
importance of OPEs as a potential concern in indoor
environments, as well as the necessity of indoor air monitoring,
since inhalation is considered as an important exposure route
for chlorinated OPEs.30
On the other hand, 20 and 50 times
lower concentrations of banned HBCDDs and PBDEs,
respectively, compared to NFRs, show the decline of legacy
FRs in the indoor environments.
■ ASSOCIATED CONTENT
*S Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.est.7b03245.
Analytical method details, QA/QC details, summary
statistics, exposure assessment calculations (PDF)
■ AUTHOR INFORMATION
Corresponding Author
*E-mail: melymuk@recetox.muni.cz.
Notes
The authors declare no competing financial interest.
■ ACKNOWLEDGMENTS
This work was supported by the National Sustainability
Programme of the Czech Ministry of Education, Youth and
Sports (LO1214) and the RECETOX research infrastructure
(Ministry of Education, Youth and Sports of the Czech
Republic, LM2015051). We thank Petra Přibylová, Jiří
Kohoutek, and Ondřej Audy for instrumental analysis of legacy
POPs, Roman Prokeš for sample collection and Jiří Kalina for
statistics consultations.
■ REFERENCES
(1) Vojta, Š.; Bečanová, J.; Melymuk, L.; Komprdová, K.; Kohoutek,
J.; Kukučka, P.; Klánová, J. Screening for halogenated flame retardants
in European consumer products, building materials and wastes.
Chemosphere 2017, 168, 457−466.
(2) Harrad, S.; Ibarra, C.; Diamond, M.; Melymuk, L.; Robson, M.;
Douwes, J.; Roosens, L.; Dirtu, A. C.; Covaci, A. Polybrominated
diphenyl ethers in domestic indoor dust from Canada, New Zealand,
United Kingdom and United States. Environ. Int. 2008, 34 (2), 232−
238.
(3) Harrad, S.; de Wit, C. A.; Abdallah, M. A. E.; Bergh, C.;
Bjorklund, J. A.; Covaci, A.; Darnerud, P. O.; de Boer, J.; Diamond,
M.; Huber, S.; Leonards, P.; Mandalakis, M.; Oestman, C.; Haug, L. S.;
Thomsen, C.; Webster, T. F. Indoor Contamination with
Hexabromocyclododecanes, Polybrominated Diphenyl Ethers, and
Perfluoroalkyl Compounds: An Important Exposure Pathway for
People? Environ. Sci. Technol. 2010, 44 (9), 3221−3231.
(4) Law, R. J.; Barry, J.; Bersuder, P.; Barber, J. L.; Deaville, R.; Reid,
R. J.; Jepson, P. D. Levels and Trends of Brominated Diphenyl Ethers
in Blubber of Harbor Porpoises (Phocoena phocoena) Stranded or
Bycaught in the UK: An Update to 2006. Environ. Sci. Technol. 2010,
44 (12), 4447−4451.
(5) Law, R. J.; Allchin, C. R.; de Boer, J.; Covaci, A.; Herzke, D.;
Lepom, P.; Morris, S.; Tronczynski, J.; de Wit, C. A. Levels and trends
of brominated flame retardants in the European environment.
Chemosphere 2006, 64 (2), 187−208.
(6) Law, R. J.; Covaci, A.; Harrad, S.; Herzke, D.; Abdallah, M. A. E.;
Femie, K.; Toms, L.-M. L.; Takigami, H. Levels and trends of PBDEs
and HBCDs in the global environment: Status at the end of 2012.
Environ. Int. 2014, 65, 147−158.
(7) Muir, D. C. G.; de Wit, C. A. Trends of legacy and new persistent
organic pollutants in the circumpolar arctic: Overview, conclusions,
and recommendations. Sci. Total Environ. 2010, 408 (15), 3044−3051.
(8) Salamova, A.; Hermanson, M. H.; Hites, R. A. Organophosphate
and Halogenated Flame Retardants in Atmospheric Particles from a
European Arctic Site. Environ. Sci. Technol. 2014, 48 (11), 6133−6140.
(9) Venier, M.; Ma, Y. N.; Hites, R. A. Bromobenzene Flame
Retardants in the Great Lakes Atmosphere. Environ. Sci. Technol. 2012,
46 (16), 8653−8660.
(10) Ma, Y. N.; Venier, M.; Hites, R. A. Tribromophenoxy Flame
Retardants in the Great Lakes Atmosphere. Environ. Sci. Technol. 2012,
46 (24), 13112−13117.
Environmental Science & Technology Article
DOI: 10.1021/acs.est.7b03245
Environ. Sci. Technol. 2017, 51, 11891−11899
11897
(11) Ma, Y. N.; Venier, M.; Hites, R. A. 2-Ethylhexyl
Tetrabromobenzoate and Bis(2-ethylhexyl) Tetrabromophthalate
Flame Retardants in the Great Lakes Atmosphere. Environ. Sci.
Technol. 2012, 46 (1), 204−208.
(12) Shoeib, M.; Ahrens, L.; Jantunen, L.; Harner, T. Concentrations
in air of organobromine, organochlorine and organophosphate flame
retardants in Toronto, Canada. Atmos. Environ. 2014, 99, 140−147.
(13) Möller, A.; Xie, Z. Y.; Caba, A.; Sturm, R.; Ebinghaus, R.
Occurrence and air-seawater exchange of brominated flame retardants
and Dechlorane Plus in the North Sea. Atmos. Environ. 2012, 46, 346−
353.
(14) Harrad, S.; Goosey, E.; Desborough, J.; Abdallah, M. A. E.;
Roosens, L.; Covaci, A. Dust from UK Primary School Classrooms and
Daycare Centers: The Significance of Dust As a Pathway of Exposure
of Young UK Children to Brominated Flame Retardants and
Polychlorinated Biphenyls. Environ. Sci. Technol. 2010, 44 (11),
4198−4202.
(15) Covaci, A.; Harrad, S.; Abdallah, M. A. E.; Ali, N.; Law, R. J.;
Herzke, D.; de Wit, C. A. Novel brominated flame retardants: A review
of their analysis, environmental fate and behaviour. Environ. Int. 2011,
37 (2), 532−556.
(16) Stubbings, W. A.; Harrad, S. Extent and mechanisms of
brominated flame retardant emissions from waste soft furnishings and
fabrics: A critical review. Environ. Int. 2014, 71, 164−175.
(17) van der Ven, L. T. M.; van de Kuil, T.; Leonards, P. E. G.; Slob,
W.; Lilienthal, H.; Litens, S.; Herlin, M.; Hakansson, H.; Canton, R. F.;
van den Berg, M.; Visser, T. J.; van Loveren, H.; Vos, J. G.; Piersma, A.
H. Endocrine effects of hexabromocyclododecane (HBCD) in a onegeneration
reproduction study in Wistar rats. Toxicol. Lett. 2009, 185
(1), 51−62.
(18) UNEP. Stockholm Convention on Persistent Organic Pollutants
(POPs). Amendments to Annexes A, B and C, 2009.
(19) UNEP. Stockholm Convention on Persistent Organic Pollutants
(POPs). Amendments to Annex A. Secretariat of the Stockholm
Convention, 2013.
(20) EU. Directive 2002/95/EC and Commission Decision 2005/
717/EC; 2008. In European Court of Justice: Cases C-14/06 and C-
295/06, Judgement of the Court, 2008.
(21) USEPA. DecaBDE phase-out initiative, 2010.
(22) UNEP. Draft Decision SC-8/[ ]: Decabromodiphenyl ether
(commercial mixture, c-decaBDE), 2017.
(23) Sverko, E.; Tomy, G. T.; Reiner, E. J.; Li, Y. F.; McCarry, B. E.;
Arnot, J. A.; Law, R. J.; Hites, R. A. Dechlorane Plus and Related
Compounds in the Environment: A Review. Environ. Sci. Technol.
2011, 45 (12), 5088−5098.
(24) van der Veen, I.; de Boer, J. Phosphorus flame retardants:
Properties, production, environmental occurrence, toxicity and
analysis. Chemosphere 2012, 88 (10), 1119−1153.
(25) Newton, S.; Sellström, U.; de Wit, C. A. Emerging Flame
Retardants, PBDEs, and HBCDDs in Indoor and Outdoor Media in
Stockholm, Sweden. Environ. Sci. Technol. 2015, 49 (5), 2912−2920.
(26) Schreder, E. D.; La Guardia, M. J. Flame Retardant Transfers
from U.S. Households (Dust and Laundry Wastewater) to the Aquatic
Environment. Environ. Sci. Technol. 2014, 48 (19), 11575−11583.
(27) Sühring, R.; Diamond, M. L.; Scheringer, M.; Wong, F.; Pućko,
M.; Stern, G.; Burt, A.; Hung, H.; Fellin, P.; Li, H.; Jantunen, L. M.
Organophosphate Esters in Canadian Arctic Air: Occurrence, Levels
and Trends. Environ. Sci. Technol. 2016, 50 (14), 7409−7415.
(28) Zhang, X. M.; Suhring, R.; Serodio, D.; Bonnell, M.; Sundin, N.;
Diamond, M. L. Novel flame retardants: Estimating the physicalchemical
properties and environmental fate of 94 halogenated and
organophosphate PBDE replacements. Chemosphere 2016, 144, 2401−
2407.
(29) Wei, G. L.; Li, D. Q.; Zhuo, M. N.; Liao, Y. S.; Xie, Z. Y.; Guo,
T. L.; Li, J. J.; Zhang, S. Y.; Liang, Z. Q. Organophosphorus flame
retardants and plasticizers: Sources, occurrence, toxicity and human
exposure. Environ. Pollut. 2015, 196, 29−46.
(30) Schreder, E. D.; Uding, N.; La Guardia, M. J. Inhalation a
significant exposure route for chlorinated organophosphate flame
retardants. Chemosphere 2016, 150, 499−504.
(31) Darnerud, P. O. Brominated flame retardants as possible
endocrine disrupters. Int. J. Androl. 2008, 31 (2), 152−160.
(32) Kuramochi, H.; Takigami, H.; Scheringer, M.; Sakai, S. I.
Estimation of physicochemical properties of 52 non-PBDE brominated
flame retardants and evaluation of their overall persistence and longrange
transport potential. Sci. Total Environ. 2014, 491, 108−117.
(33) Liagkouridis, L.; Cousins, A. P.; Cousins, I. T. Physical-chemical
properties and evaluative fate modelling of ’emerging’ and ’novel’
brominated and organophosphorus flame retardants in the indoor and
outdoor environment. Sci. Total Environ. 2015, 524, 416−426.
(34) Rudel, R. A.; Dodson, R. E.; Perovich, L. J.; Morello-Frosch, R.;
Camann, D. E.; Zuniga, M. M.; Yau, A. Y.; Just, A. C.; Brody, J. G.
Semivolatile Endocrine-Disrupting Compounds in Paired Indoor and
Outdoor Air in Two Northern California Communities. Environ. Sci.
Technol. 2010, 44 (17), 6583−6590.
(35) Rauert, C.; Lazarov, B.; Harrad, S.; Covaci, A.; Stranger, M. A
review of chamber experiments for determining specific emission rates
and investigating migration pathways of flame retardants. Atmos.
Environ. 2014, 82, 44−55.
(36) Webster, T. F.; Stapleton, H. M.; McClean, M. D. Exposure to
Polybrominated Diphenyl Ethers in the Indoor Environment. Fire
Technol. 2015, 51 (1), 85−95.
(37) Bjorklund, J. A.; Thuresson, K.; Cousins, A. P.; Sellstrom, U.;
Emenius, G.; de Wit, C. A. Indoor Air Is a Significant Source of Tridecabrominated
Diphenyl Ethers to Outdoor Air via Ventilation
Systems. Environ. Sci. Technol. 2012, 46 (11), 5876−5884.
(38) Melymuk, L.; Bohlin-Nizzetto, P.; Kukučka, P.; Vojta, Š.; Kalina,
J.; Čupr, P.; Klánová, J. Seasonality and indoor/outdoor relationships
of flame retardants and PCBs in residential air. Environ. Pollut. 2016,
218, 392−401.
(39) Venier, M.; Audy, O.; Vojta, Š.; Bečanová, J.; Romanak, K.;
Melymuk, L.; Krátká, M.; Kukučka, P.; Okeme, J.; Saini, A.; Diamond,
M. L.; Klánová, J. Brominated flame retardants in the indoor
environment  Comparative study of indoor contamination from
three countries. Environ. Int. 2016, 94, 150−160.
(40) D’Hollander, W.; Roosens, L.; Covaci, A.; Cornelis, C.;
Reynders, H.; Campenhout, K. V.; Voogt, P. d.; Bervoets, L.
Brominated flame retardants and perfluorinated compounds in indoor
dust from homes and offices in Flanders, Belgium. Chemosphere 2010,
81 (4), 478−487.
(41) Mandalakis, M.; Atsarou, V.; Stephanou, E. G. Airborne PBDEs
in specialized occupational settings, houses and outdoor urban areas in
Greece. Environ. Pollut. 2008, 155 (2), 375−382.
(42) Watkins, D. J.; McClean, M. D.; Fraser, A. J.; Weinberg, J.;
Stapleton, H. M.; Sjodin, A.; Webster, T. F. Exposure to PBDEs in the
Office Environment: Evaluating the Relationships Between Dust,
Handwipes, and Serum. Environ. Health Perspect. 2011, 119 (9),
1247−1252.
(43) Liagkouridis, I.; Cousins, I. T.; Cousins, A. P. Emissions and fate
of brominated flame retardants in the indoor environment: A critical
review of modelling approaches. Sci. Total Environ. 2014, 491, 87−99.
(44) Kemmlein, S.; Hahn, O.; Jann, O. Emissions of organophosphate
and brominated flame retardants from selected consumer
products and building materials. Atmos. Environ. 2003, 37 (39−40),
5485−5493.
(45) Bako-Biro, Z.; Wargocki, P.; Weschler, C. J.; Fanger, P. O.
Effects of pollution from personal computers on perceived air quality,
SBS symptoms and productivity in offices. Indoor Air 2004, 14 (3),
178−187.
(46) Harrad, S.; Abdallah, M. A. E.; Covaci, A. Causes of variability in
concentrations and diastereomer patterns of hexabromocyclododecanes
in indoor dust. Environ. Int. 2009, 35 (3), 573−579.
(47) de Wit, C. A.; Bjorklund, J. A.; Thuresson, K. Tridecabrominated
diphenyl ethers and hexabromocyclododecane in
indoor air and dust from Stockholm microenvironments 2: Indoor
sources and human exposure. Environ. Int. 2012, 39 (1), 141−147.
Environmental Science & Technology Article
DOI: 10.1021/acs.est.7b03245
Environ. Sci. Technol. 2017, 51, 11891−11899
11898
(48) Zhang, X. M.; Diamond, M. L.; Ibarra, C.; Harrad, S.
Multimedia Modeling of Polybrominated Diphenyl Ether Emissions
and Fate Indoors. Environ. Sci. Technol. 2009, 43 (8), 2845−2850.
(49) Dodson, R. E.; Rodgers, K. M.; Carey, G.; Cedeno Laurent, J.
G.; Covaci, A.; Poma, G.; Malarvannan, G.; Spengler, J. D.; Rudel, R.
A.; Allen, J. G. Flame Retardant Chemicals in College Dormitories:
Flammability Standards Influence Dust Concentrations. Environ. Sci.
Technol. 2017, 51 (9), 4860−4869.
(50) Liang, W.; Wang, C.; Yang, C.; Yang, X. Volatile organic
compounds in different interior construction stages of an apartment.
Building and Environment 2014, 81, 380−387.
(51) Park, J. C.; Kwon, Y. C.; Do Jun, H. A Study on the
Improvement of Indoor Air Quality of Newly-Built Apartment Houses
Using Low Emission Building Materials. J. Asian Archit. Build. Eng.
2011, 10 (1), 235−240.
(52) Pang, S. K.; Cho, H.; Sohn, J. Y.; Song, K. D. Assessment of the
emission characteristics of VOCs from interior furniture materials
during the construction process. Indoor Built Environ. 2007, 16 (5),
444−455.
(53) Bohlin, P.; Audy, O.; Skrdlikova, L.; Kukucka, P.; Vojta, S.;
Pribylova, P.; Prokes, R.; Cupr, P.; Klanova, J. Evaluation and
guidelines for using polyurethane foam (PUF) passive air samplers in
double-dome chambers to assess semi-volatile organic compounds
(SVOCs) in non-industrial indoor environments. Environ. Sci.-Process
Impacts 2014, 16 (11), 2617−2626.
(54) Bohlin, P.; Audy, O.; Skrdlikova, L.; Kukucka, P.; Pribylova, P.;
Prokes, R.; Vojta, S.; Klanova, J. Outdoor passive air monitoring of
semi volatile organic compounds (SVOCs): a critical evaluation of
performance and limitations of polyurethane foam (PUF) disks.
Environ. Sci.-Process Impacts 2014, 16 (3), 433−444.
(55) Bergman, A.; Ryden, A.; Law, R. J.; de Boer, J.; Covaci, A.;
Alaee, M.; Birnbaum, L.; Petreas, M.; Rose, M.; Sakai, S.; Van den
Eede, N.; van der Veen, I. A novel abbreviation standard for
organobromine, organochlorine and organophosphorus flame retardants
and some characteristics of the chemicals. Environ. Int. 2012, 49,
57−82.
(56) Lohmann, R.; Klanova, J.; Pribylova, P.; Liskova, H.; Yonis, S.;
Bollinger, K. PAHs on a West-to-East Transect Across the Tropical
Atlantic Ocean. Environ. Sci. Technol. 2013, 47 (6), 2570−2578.
(57) Lohmann, R.; Klanova, J.; Kukucka, P.; Yonis, S.; Bollinger, K.
Concentrations, Fluxes, and Residence Time of PBDEs Across the
Tropical Atlantic Ocean. Environ. Sci. Technol. 2013, 47 (24), 13967−
13975.
(58) Lohmann, R.; Klanova, J.; Kukucka, P.; Yonis, S.; Bollinger, K.
PCBs and OCPs on a East-to-West Transect: The Importance of
Major Currents and Net Volatilization for PCBs in the Atlantic Ocean.
Environ. Sci. Technol. 2012, 46 (19), 10471−10479.
(59) Vykoukalova, M.; Venier, M.; Vojta, S.; Melymuk, L.; Becanova,
J.; Romanak, K.; Prokes, R.; Okeme, J. O.; Saini, A.; Diamond, M. L.;
Klanova, J. Organophosphate esters flame retardants in the indoor
environment. Environ. Int. 2017, 106, 97−104.
(60) Antweiler, R. C. Evaluation of Statistical Treatments of LeftCensored
Environmental Data Using Coincident Uncensored Data
Sets. II. Group Comparisons. Environ. Sci. Technol. 2015, 49 (22),
13439−13446.
(61) Batterman, S.; Godwin, C.; Chernyak, S.; Jia, C.; Charles, S.
Brominated flame retardants in offices in Michigan, U.S.A. Environ. Int.
2010, 36 (6), 548−556.
(62) Cequier, E.; Ionas, A. C.; Covaci, A.; Maria Marce, R.; Becher,
G.; Thomsen, C. Occurrence of a Broad Range of Legacy and
Emerging Flame Retardants in Indoor Environments in Norway.
Environ. Sci. Technol. 2014, 48 (12), 6827−6835.
(63) Blanchard, O.; Glorennec, P.; Mercier, F.; Bonvallot, N.;
Chevrier, C.; Ramalho, O.; Mandin, C.; Le Bot, B. Semivolatile
Organic Compounds in Indoor Air and Settled Dust in 30 French
Dwellings. Environ. Sci. Technol. 2014, 48 (7), 3959−3969.
(64) Tao, F.; Abdallah, M. A. E.; Harrad, S. Emerging and Legacy
Flame Retardants in UK Indoor Air and Dust: Evidence for
Replacement of PBDEs by Emerging Flame Retardants? Environ. Sci.
Technol. 2016, 50 (23), 13052−13061.
(65) Bergh, C.; Torgrip, R.; Emenius, G.; Ostman, C. Organophosphate
and phthalate esters in air and settled dust - a multi-location
indoor study. Indoor Air 2011, 21 (1), 67−76.
(66) Hartmann, P. C.; Burgi, D.; Giger, W. Organophosphate flame
retardants and plasticizers in indoor air. Chemosphere 2004, 57 (8),
781−787.
(67) Melymuk, L.; Bohlin-Nizzetto, P.; Vojta, S.; Kratka, M.;
Kukucka, P.; Audy, O.; Pribylova, P.; Klanova, J. Distribution of
legacy and emerging semivolatile organic compounds in five indoor
matrices in a residential environment. Chemosphere 2016, 153, 179−
186.
(68) Carlsson, H.; Nilsson, U.; Ostman, C. Video display units: An
emission source of the contact allergenic flame retardant triphenyl
phosphate in the indoor environment. Environ. Sci. Technol. 2000, 34
(18), 3885−3889.
(69) Ni, Y.; Kumagai, K.; Yanagisawa, Y. Measuring emissions of
organophosphate flame retardants using a passive flux sampler. Atmos.
Environ. 2007, 41 (15), 3235−3240.
(70) Duan, H.; Yu, D. F.; Zuo, J.; Yang, B.; Zhang, Y. K.; Niu, Y. N.
Characterization of brominated flame retardants in construction and
demolition waste components: HBCD and PBDEs. Sci. Total Environ.
2016, 572, 77−85.
(71) Babrauskas, V.; Lucas, D.; Eisenberg, D.; Singla, V.; Dedeo, M.;
Blum, A. Flame retardants in building insulation: a case for reevaluating
building codes. Build. Res. Informat. 2012, 40 (6), 738−755.
(72) CECBP Brominated and Chlorinated Organic Chemical
Compounds Used as Flame Retardants; Meeting of the California
Environmental Contaminant Biomonitoring Program (CECBP)
Scientific Guidance Panel (SGP): 2009.
(73) von der Recke, R.; Vetter, W. Synthesis and characterization of
2,3-dibromopropyl-2,4,6-tribromophenyl ether (DPTE) and structurally
related compounds evidenced in seal blubber and brain. Environ.
Sci. Technol. 2007, 41 (5), 1590−1595.
(74) Harju, M. Emerging ″New″ Brominated Flame Retardants in
Flame Retarded Products and the Environment; The Norwegian
Pollution Control Authority: Oslo, 2009.
(75) EFSA. EFSA Panel on Contaminants in the Food Chain
(CONTAM). Scientific Opinion on Emerging and Novel Brominated
Flame Retardants (BFRs) in Food. EFSA Journal 2012, 10 (10), 2908,
133 pp.
(76) Abbasi, G.; Saini, A.; Goosey, E.; Diamond, M. L. Product
screening for sources of halogenated flame retardants in Canadian
house and office dust. Sci. Total Environ. 2016, 545, 299−307.
(77) Rauert, C.; Harrad, S.; Suzuki, G.; Takigami, H.; Uchida, N.;
Takata, K. Test chamber and forensic microscopy investigation of the
transfer of brominated flame retardants into indoor dust via abrasion
of source materials. Sci. Total Environ. 2014, 493, 639−648.
(78) Muenhor, D.; Harrad, S. Within-room and within-building
temporal and spatial variations in concentrations of polybrominated
diphenyl ethers (PBDEs) in indoor dust. Environ. Int. 2012, 47, 23−
27.
(79) Allen, J. G.; McClean, M. D.; Stapleton, H. M.; Nelson, J. W.;
Webster, T. F. Personal exposure to Polybrominated Diphenyl Ethers
(PBDEs) in residential indoor air. Environ. Sci. Technol. 2007, 41 (13),
4574−4579.
Environmental Science & Technology Article
DOI: 10.1021/acs.est.7b03245
Environ. Sci. Technol. 2017, 51, 11891−11899
11899
Vykoukalová, Martina, Marta Venier, Šimon Vojta, Lisa Melymuk, Jitka Bečanová, Kevin
Romanak, Roman Prokeš, Joseph O. Okeme, Amandeep Saini, Miriam L. Diamond, Jana
Klanova. 2017. "Organophosphate Esters Flame Retardants in the Indoor Environment."
Environment International 106: 97-104. https://doi.org/10.1016/j.envint.2017.05.020
Contents lists available at ScienceDirect
Environment International
journal homepage: www.elsevier.com/locate/envint
Organophosphate esters flame retardants in the indoor environment
Martina Vykoukalováa
, Marta Venierb,⁎
, Šimon Vojtaa
, Lisa Melymuka
, Jitka Bečanováa
,
Kevin Romanakb
, Roman Prokeša
, Joseph O. Okemec
, Amandeep Sainic,1
, Miriam L. Diamondd,c
,
Jana Klánováa
a
Research Centre for Toxic Compounds in the Environment (RECETOX), Masaryk University, Kamenice 753/5, pavilion A29, 625 00 Brno, Czech Republic
b
School of Public and Environmental Affairs, Indiana University, 702 N. Walnut Grove Ave., Bloomington, 47405 Indiana, USA
c
Department of Physical & Environmental Sciences, University of Toronto Scarborough, 1265 Military Trail, Toronto, Ontario M1C 1A4, Canada
d
Department of Earth Sciences, University of Toronto, 22 Russell Street, Toronto, Ontario M5S 3B1, Canada
A R T I C L E I N F O
Keywords:
Organophosphate esters (OPEs)
Indoor Environment
Human Exposure
Air Concentration
Indoor Dust
Window Film
A B S T R A C T
Concentrations of 13 organophosphate ester flame retardants (OPEs) were measured in air, dust and window
wipes from 63 homes in Canada, the Czech Republic and the United States in the spring and summer of 2013 to
look for abundances, differences among regions, and partitioning behavior. In general, we observed the highest
concentrations for halogenated OPEs, particularly TCEP, TCIPP and TDCIPP, and also non-halogenated TPHP.
Differences between regions strongly depended on the matrix. The concentrations of OPEs in dust were significantly
higher in the US than in Canada (CAN) and Czech Republic (CZ). CZ had the highest concentrations in
window film and CAN in air. ΣOPE concentrations were 2–3 and 1–2 orders of magnitude greater than ΣBFRs in
air, and dust and window films, respectively. We found a significant relationship between the concentrations in
dust and air, and between the concentrations in window film and air for OPEs with log KOA values < 12, suggesting
that equilibrium was reached for these compounds but not for those with log KOA > 12. This hypothesis
was confirmed by a large discrepancy between values predicted using a partitioning model and the measured
values for OPEs with log KOA values > 12.
1. Introduction
Polybrominated diphenyl ethers (PBDEs) are the most thoroughly
studied class of flame retardants (FRs). They were widely used in numerous
household products until the early 2000s when two of the
commercial mixtures, Penta and OctaBDE, were withdrawn from the
market in the US due to mounting evidence of adverse health effects
and widespread environmental presence. This resulted in an increased
use of alternative compounds, mostly either brominated compounds
(e.g. 2-Ethylhexyl-2,3,4,5-tetrabromobenzoate (EHTBB), Bis(2-ethyl-1hexyl)
tetrabromophthalate (BEHTBP) or organophosphate esters
(OPEs).
OPEs are phosphoric acid esters used as FRs, plasticizers and antifoaming
agents (Rauert and Harrad, 2015). Their application in products
ranges from textiles, polyurethane foam (PUF) upholstered furniture,
and electronics to construction materials (e.g., building insulation)
and vehicles (Marklund et al., 2003; van der Veen and de Boer,
2012; Wei et al., 2015). While chlorinated and brominated OPEs see
widespread use mostly as FRs, the non-halogenated OPEs are used also
as plasticizers, lubricants and pore size regulators (Andresen et al.,
2004). Since OPEs are typically used as additive chemicals and are
therefore not covalently bound to polymeric materials, they can easily
migrate from products into the environment by means of volatilization,
leaching and abrasion, and direct transfer to dust (Marklund et al.,
2003; van der Veen and de Boer, 2012; Wei et al., 2015).
Due to their physical and chemical characteristics, OPEs are ubiquitous
in various environmental compartments worldwide and have
been detected in abiotic matrices such as sediment (Cao et al., 2012),
surface and groundwater water (Regnery et al., 2010; Regnery et al.,
2011; Venier et al., 2014), outdoor air including remote locations
(Salamova et al., 2016; Sühring et al., 2016a), indoor air (Marklund
et al., 2005; van der Veen and de Boer, 2012), and house dust (Dodson
et al., 2012; Stapleton et al., 2009). They are also found in biota (van
der Veen and de Boer, 2012) and human breast milk (Sundkvist et al.,
2010), indicating that these compounds are bioavailable and might
bioaccumulate (Greaves et al., 2016). Studies on the toxicity of OPEs
are still limited but some OPEs have been reported to be mutagenic,
carcinogenic, and neurotoxic, as well as potential developmental and
http://dx.doi.org/10.1016/j.envint.2017.05.020
Received 31 January 2017; Received in revised form 23 May 2017; Accepted 25 May 2017
⁎
Corresponding author.
1
Current address: Air Quality Processes Research Section, Environment and Climate Change Canada, 4905 Dufferin St., Toronto, Ontario, Canada M3H 5T4.
E-mail address: mvenier@indiana.edu (M. Venier).
Environment International 106 (2017) 97–104
0160-4120/ © 2017 Elsevier Ltd. All rights reserved.
MARK
reproductive toxins (Behl et al., 2015; Hendriks and Westerink, 2015;
Schweizer et al., 2007; van der Veen and de Boer, 2012).
Due to its persistence, bioaccumulative potential and toxicity, Tris
(2-chloroethyl) phosphate (TCEP) has been restricted from use in concentrations
greater than 5 mg/kg in toys for children up to 3 years of
age and for any toys intended for mouthing and has been designated as
a substance of very high concern in the European Union (Toy Safety
Directive, 2009/48/EC, European Chemicals Agency, 2015). As such, it
is no longer produced in Europe (Green et al., 2008; Schreder et al.,
2016; Sühring et al., 2016a). Some reports suggest that it has been
replaced by other FRs, primarily Tris(1-chloro-2-propyl) phosphate
(TCIPP). In 2001, this structurally similar and relatively cheap replacement
for TCEP, represented approximately 80% of the chlorinated
OPEs used in Europe (Leisewitz et al., 2001). In 2014 Canada prohibited
new use of TCEP in products containing PUF intended for
children under 3 years of age (Canada Gazette, 2014). TCEP was restricted
for new uses in several US states starting in 2011 (Safer States)
and is listed as a carcinogen under California's Proposition 65. TCIPP
has been recommended for designation as “toxic” under the Canadian
Environmental Protection Act (Environment and Climate Change
Canada, 2016). The proposed risk management measures are limiting
its use to < 0.1% in mattresses and upholstered furniture.
Tris(1,3-dichloro-2-propyl) phosphate (TDCIPP) appears to be used
as a substitute for brominated FRs. TDCIPP is primarily used in foams in
the automotive industry and to some extent also in upholstered furniture
(Cooper et al., 2016). Tris(2,3-dibromopropyl) phosphate (TDBPP),
a brominated analog for TDCIPP, was banned in children's sleepwear in
the US in 1977 due to carcinogenicity concerns after mutagenic metabolites
were detected in children's urine (Gold et al., 1978) and it is
listed as a carcinogen in California's Proposition 65. TDCIPP is also
listed as a carcinogen in California's Proposition 65 but has not been
recommended for designation as “toxic” under the Canadian Environmental
Protection Act (Environment and Climate Change Canada,
2016).
Triphenyl phosphate (TPHP), one of the most effective FRs used in
polymers (van der Veen and de Boer, 2012), is used also in hydraulic
fluids (Andresen et al., 2004). TPHP is a component in the Firemaster
(FM) 550 mixture, introduced in 2004 as a replacement for the PentaBDE
commercial mixture used in upholstered furniture. It is also used
as a plasticizer in, for example, nail polish (Mendelsohn et al., 2016)
and electronic components such as televisions and monitor screens
(Kajiwara et al., 2011). EHDPP is mainly used as a flame retardant/
plasticizer in flexible PVC (e.g., wire and cable insulators, connectors)
and in certain food packaging in the US (Brooke et al., 2009).
In this study, we measured the concentrations of 13 OPEs - four
halogenated (TCEP, TCIPP, TDCIPP and TDBPP) and nine non-halogenated
(TPHP, EHDPP, TEHP, TNBP, o-TMPP, p-TMPP, TIPPP, TDMPP
and TBPP) in air, dust and window film from homes in three different
countries (Czech Republic, Canada and USA) during the spring and
summer of 2013. These three matrices were chosen as the most relevant
for human exposure, as well as being novel and convenient. The goals of
this project were to evaluate the partitioning behavior of OPEs in the
indoor environment, to compare within-house differences, and to examine
regional differences between Central Europe and North America.
Two companion papers reported on levels of brominated flame retardants
(Venier et al., 2016) and perfluorinated alkyl substances
(Karásková et al., 2016) in the same homes.
2. Experimental section
2.1. Sample collection
Air, dust and window film samples were collected from a total of 63
houses and apartments (20 homes each in Brno, Czech Republic and
Bloomington, IN, US and 23 in Toronto, ON, Canada) during a sampling
period of 28 days in May–August 2013. Samples from one room, usually
the bedroom, were collected in each home while a second room, usually
the living room, was sampled in at least nine randomly chosen homes in
each country. Participants were recruited as a “sample of convenience”
among colleagues, friends, relatives and acquaintances.
Sampling involved deploying polyurethane foam (PUF) passive air
samplers to estimate air concentrations, collecting settled floor dust
into nylon socks by vacuuming, and collecting interior window films as
representative of surface films on interior surfaces using Kimwipes.
Details of sample collection can be found elsewhere (Venier et al.,
2016) and only a brief description is provided here. Before sampling, all
matrices (PUF disks, Kimwipes and nylon vacuum socks) were precleaned
in a Soxhlet extractor (8 h in acetone, then 8 h in toluene),
dried and packed in aluminum foil and transported to the sites. Field
blanks were collected by exposing pre-cleaned matrices during sample
retrieval.
On day 1, floors were vacuumed and windows were cleaned with
Kimwipes moistened with 2-propanol until no dirt was visible
(Kimwipes were not saved at this time), PUF passive air samplers were
deployed, and participants were asked not to vacuum rooms or wash
the windows where the samplers were located until the end of the
campaign. PUF disks were exposed to indoor air using a single-bowl in
US and Canada or double-bowl in the Czech Republic (see Fig. S1)
passive sampler housing for 28 days. Sampling rates for each passive air
sampler configuration were calculated in a separate experiment by simultaneously
deploying single- and double-bowl samplers along with
active air samplers (Venier et al., 2016). More details on the calibration
of the passive air samplers are reported in the Supporting Information.
For this study, we used a sampling rate of 1.6 m3
/day for the singlebowl
sampler and 0.82 m3
/day for the double-bowl sampler. Given the
difficulty of calculating accurate and meaningful sampling rates, this
approach seems reasonable.
On day 28, PUF discs were retrieved and window wipe samples
were collected using a pre-cleaned Kimwipe moistened with 2-propanol
(Kimwipes were saved at this time). Windows were wiped until no dirt
was visible on the Kimwipe. The sampled area averaged at 0.32 m2
for
Canada, 0.93 m2
for US, and 1.79 m2
for the Czech Republic. Floor dust
samples were collected using a pre-cleaned nylon sampling sock inserted
into the tube of a conventional household vacuum cleaner, vacuuming
the largest possible floor area and recording the area. After
collection, all samples were wrapped in aluminum foil and sealed in
plastic bags for transport to the laboratory. Samples were stored at
−18 °C until extraction and analysis.
2.2. Target compounds and chemicals
The list of target compounds is reported in Table 1. The following
OPEs standards were purchased from Wellington Laboratories (Guelph,
ON, Canada): Tri-n-butylphosphate (TnBP), Tris(2-chloroethyl) phosphate
(TCEP), Tris(1-chloro-2-propyl) phosphate (TCIPP), Tris(1,3-dichloro-2-propyl)
phosphate (TDCIPP), Triphenyl phosphate (TPHP), 2Ethylhexyl-diphenyl
phosphate (EHDPP), Tris(2-ethylhexyl) phosphate
(TEHP), Tri-o-tolyl phosphate (o-TMPP), Tri-p-tolyl phosphate (pTMPP),
Tris(2-isopropylphenyl) phosphate (TIPPP), and Tris(3,5-dimethylphenyl)
phosphate (TDMPP). Tris(2,3-dibromopropyl) phosphate
(TDBPP) was purchased from AccuStandard (New Haven, CT).
Tris(4-tert-butylphenyl) phosphate (TBPP) was purchased from SigmaAldrich
(St. Louis, MO). d12-Tris(2-chloroethyl) phosphate (d12-TCEP),
13
C18-triphenyl phosphate (MTPP) and 13
C12-BDE-77 were purchased
from Wellington. The internal quantitation standards, d10-anthracene,
d12-benz[a]anthracene, and d12-perylene, were obtained from Chem
Service (West Chester, PA). All solvents were HPLC or Optima grade.
Silica gel (100–200 mesh, 75–150 μm, Grade 644) and granular anhydrous
sodium sulfate (Na2SO4) were purchased from Fisher Scientific
(Pittsburgh, PA).
M. Vykoukalová et al. Environment International 106 (2017) 97–104
98
2.3. Extraction
The US samples were analyzed in Bloomington, IN at Indiana
University (IU) and the Canadian and Czech samples were analyzed in
Brno, Czech Republic at RECETOX. Details of the analytical protocols
are given in the SI and are summarized here.
Before extraction, all samples were spiked with known amounts of
recovery standards [13
C18-triphenyl phosphate (MTPP) at RECETOX
and 13
C18-triphenyl phosphate (MTPP) and d12-Tris(2-chloroethyl)
phosphate (d12-TCEP) at IU]. PUF and Kimwipes were extracted in
250 mL dichloromethane (DCM) using automated warm Soxhlet extraction
(three cycles: 40 min warm Soxhlet followed by 20 min of
solvent rinsing and concentration of the extract) at RECETOX and with
400 mL of acetone: n-hexane (1:1, v/v) using Soxhlet extraction for 24 h
at IU. In both labs, socks with dust were weighed, the dust was sieved
to < 500 μm, approximately 100 mg were weighed and the remaining
dust was wrapped in foil for future use. The sock rinsate (30 mL hexane:acetone,
1:1) and the sieved fraction of dust were sonicated with
30 mL acetone: n-hexane (1:1, v/v). The dust was allowed to settle, the
liquid extract was removed, and this procedure was repeated twice.
2.4. Sample fractionation and clean-up
At RECETOX, 30% of the extract for each sample (measured by
weighing; the rest of the sample was used for the analysis of other
compounds) underwent non-destructive clean-up on a 5 g non-modified
activated silica column topped with 1 cm of Na2SO4 using 20 mL DCM
for the first fraction and 20 mL acetone:DCM (7:3, v/v) for the second
fraction. OPEs eluted in the second fraction. The OPE fraction was
concentrated under N2 and transferred into a 1 mL vial, solvent exchanged
to nonane, further concentrated to approximately 0.5 mL and
spiked with the injection standard 13
C12-BDE-77. At IU, extracts were
reduced in volume to about 2 mL by rotary evaporation. The solvent
was exchanged to hexane and the extracts were fractionated on a
column containing 3.5% water deactivated silica gel. The column was
eluted with 25 mL of 7:3 acetone in DCM. After N2 blow down to about
1 mL, the samples were spiked with the quantitation internal standards
(d10-anthracene, d12-benz[a]anthracene and d12-perylene).
2.5. Instrumental analysis
At RECETOX, OPEs were analyzed using an Agilent 7890A gas
chromatograph (GC) equipped with a 15 m Restek RTX 1614 column
(with 60 cm retention gap) coupled to a Waters APGC XEVO TQ-S
tandem mass spectrometer (MS/MS). The mass spectrometer was operated
in positive atmospheric pressure ionization mode (APGC+)
using multiple reaction monitoring mode (MRM). At least two transitions
were recorded for each compound. The following transitions were
used for quantification (confirmation): 267 > 99 (267 > 211) for
TNBP, 287 > 99 (249 > 99), for TCEP, 327 > 99 (329 > 99) for
TCIPP, 431 > 99 (431 > 209) for TDCIPP, 327 > 215 (327 > 152)
for TPHP, 362 > 251 (362 > 94) for EHDPP, 435 > 99
(435 > 323) for TEHP, 368 > 277 (369 > 278) for o-TMPP,
368 > 261 (369 > 165) for m-TMPP and p-TMPP, 453 > 118
(453 > 251) for TIPPP, 411 > 194 (411 > 179) for TDMPP,
618 > 137 (698 > 99) for TDBPP, 495 > 327 (494 > 211) for
TBPP, 345 > 277 (345 > 164) for 13
C12-TPHP and 498 > 338
(496 > 336) for 13
C12-BDE-77. Additional parameters were set as
follows: source temperature 150 °C, cone gas flow 160 L/h and auxillary
gas flow 220 L/Hr. Injection was splitless 1 μL at 250 °C, with He
as the carrier gas at 1.2 mL min−1
. The GC temperature program was
80 °C with a 1 min hold, then 30 °C min− 1
to 140 °C, followed by 4 °C
min−1
to 175 °C, then 8 °C min− 1
to 270 °C and 15 °C min−1
to 325 °C
with a 7 min hold. Makeup gas pressure was 60 psi and transfer line
temperature was set to 300 °C.
At IU, an Agilent 6890 series GC coupled to an Agilent 5973 MS was
used to quantify OPEs. The MS was operated in the electron impact
mode. The GC resolution was achieved with a 30 m × 250 μm
i.d. × 0.25 μm film thickness DB-5MS Ultra Inert capillary column
(Agilent Technologies, Santa Clara, CA). One μL of the sample was injected
in the pulsed splitless mode at 280 °C. The GC–MS interface was
kept at 300 °C. Temperatures of the ion source and quadrupole were set
at 230 °C and 150 °C, respectively. High purity helium (99.999%;
Liquid Carbonic, Chicago) was used as the carrier gas. The GC oven
temperature was held at 90 °C for 1 min, increased to 170 °C at 10 °C/
min, held for 3 min, then increased to 230 °C at 10 °C/min, held for
4 min, then increased to 260 °C at 5 °C/min, finally increased to 300 °C
at 10 °C/min, and held for 4 min. TCIPP and TEHP were monitored
using m/z 99 and 125. Monitoring ions for the other OPEs were: m/z
155 and 99 for TNBP, m/z 249 and 63 for TCEP, m/z 75 and 191 for
TDCIPP, m/a 326 and 325 for TPHP, m/z 251 and 250 for EHDPP, m/z
410 and 411 for TDMPP, m/z 368 and 367 for o-TMPP and p-TMPP, m/
z 368 and 165 for m-TMPP, m/z 201 and 119 for TDBPP, m/z 494 and
479 for TBPP. For the surrogate and internal standards, the following
ions were monitored: MTPP, m/z 343; d12-TCEP, m/z 261; d10-anthracene,
m/z 188; d12-benz[a]anthracene, m/z 240, and d12-perylene,
m/z 264. OPE data were quantitated using the internal standard
method at both labs.
2.6. Quality control and quality assurance
Several measures were taken to ensure data comparability between
Table 1
OPEs names, abbreviations (alternate name between parentheses), CAS numbers and log KOA values (EPI Suite log KOA values were used in this paper; the others are provided for
reference only).
Name Abbreviations CAS number log KOA (Sühring et al., 2016b)
EPI Suite SPARC Absolv
Tris(2-chloroethyl) phosphate TCEP 115–96-8 7.6 7.0 8.9
Tris(1-chloro-2-propyl) phosphate TCIPP (TCPP) 13674–84-5 8.5 7.6 10.0
Tris(1,3-dichloro-2-propyl) phosphate TDCIPP (TDCPP) 13674–87-8 10.6 10.3 12.0
Tris(2,3-dibromopropyl) phosphate TDBPP 126–72-7 14.1 12.9 16.3
Tri-phenylphosphate TPHP (TPP) 115–86-6 10.5 10.3 14.0
2-ethylhexyldiphenylphosphate EHDPP 1241–94-7 11.3 10.6 14.1
Tris(2-ethylhexyl) phosphate TEHP 78–42-2 11.9 12.0 14.5
Tri n-butyl phosphate TNBP (TBP) 126–73-8 7.7 7.0 9.5
Tri-o-tolylphosphate o-TMPP (TOTP) 78–30-8 12.0 11.6 15.1
Tri-p-tolylphosphate p-TMPP (TPTP) 78–32-0 12.0 11.8 15.1
Tris(2-isopropylphenyl) phosphate TIPPP (T2IPPP) 64532–95-2 14.0 13.5 17.5
Tris(3,5-dimethylphenyl) phosphate TDMPP (T35DMPP) 25653–16-1 13.5 12.9 16.2
Tris(4-tert-butylphenyl) phosphate TBPP 78–33-1 15.0 14.7 18.1
M. Vykoukalová et al. Environment International 106 (2017) 97–104
99
the different laboratories and the accuracy and reliability of the measurements.
For dust samples, spiked (n = 3) and non-spiked (n = 3)
house dust SRM 2585 were analyzed and compared with literature data
(Table SI2). The results were also used to ensure the data comparability
between the laboratories.
The same blank treatment procedure and method limit of detection
(LOD) calculation was used as described elsewhere (Karásková et al.,
2016; Venier et al., 2016). Solvent blanks were used to evaluate contamination
from the laboratory procedures. Solvent blank levels were
low and no correction was necessary. Three field blanks per matrix per
country were collected and analyzed. The mass of the target compounds
in each sample were then compared to the average mass in the field
blanks (on a country and matrix specific basis – see Table S1 in the
Supporting Information for more details) and treated as follows: If the
blank level was < 10% of the measured level, there was no correction.
If the blank level was 10–35% of the measured level, the blank level
was subtracted from the measured level. If the blank level was > 35%
of the measured level, the value was reported as “non-detect.”
For air and window film samples, the recoveries of individual
compounds were determined using a set of pre-cleaned (8 h in acetone
and 8 h in DCM) sampling media (n = 8) spiked with the native analytes
prior to extraction. The recoveries of target compounds in matrix
spike experiments varied between 70% and 120%. Average surrogate
recoveries were mostly within the 50–150% range, which is considered
acceptable. Final values were not recovery corrected.
For statistical data analysis, values below detection were replaced
by √2/2*LOD (Antweiler, 2015). Compounds with low detection
(generally < 60% detection frequency in a given matrix/room) were
not used for statistical analysis.
2.7. Data analysis
Basic calculations, descriptive statistics and graphs were made using
Microsoft Excel software and Statistica 12. Central trends were described
with median concentrations because the data were not normally
distributed. Differences between sample populations were determined
using a nonparametric Kruskal-Wallis, Mann-Whitney U and Wilcoxon
tests. Differences were considered significant at p < 0.05.
3. Results and discussion
A summary of the results for dust, window film and air from all
three countries is reported in Table 2. An overview of the concentrations
of the main targeted OPEs is shown in Fig. S2, which depicts both
regional trends and compound patterns. Box plots for selected compounds
are shown in Fig. S3. In general, we observed the highest
concentrations for TCEP, TCIPP, TDCIPP and TPHP. Interestingly, all
three matrices in all the three countries were dominated by halogenated
OPEs.
Regional patterns of OPE concentrations followed those seen for
PBDEs and novel BFRs (Harrad et al., 2008; Venier et al., 2016). For
OPEs, we note that the comparison of differences country-to-country
must be interpreted with caution due to the challenges associated with
OPE measurement. We found that the regional trends of OPEs strongly
depended on the matrix. The concentrations of OPEs in dust were significantly
higher in the US than in Canada (CAN) and Czech Republic
(CZ) (see Table 2). CZ had the highest concentrations in window film
and CAN in air.
In particular, dust samples were dominated by TDBPP in the US
(median 4530 ng/g, range ND - 11,000 ng/g), TPHP in CAN (median
2350 ng/g, range 377–31,900 ng/g) and TCIPP in CZ (median 1860 ng/
g, range 163–26,700 ng/g). Although TCIPP was also prevalent in CZ
window film samples (median 566 ng/m2
, range 15.0–6670 ng/m2
), US
and CAN window film samples primarily had TDCIPP (medians
83.2 ng/m2
and 82.3 ng/m2
, ranges ND – 542 ng/m2
and
20.8–3590 ng/m2
, respectively). In all three countries, air
concentrations were dominated by TCIPP, with the highest levels found
in CAN (median 73.6 ng/m3
, range 7.68–4190 ng/m3
), followed by the
US (median 26.3 ng/m3
, range 0.280–226 ng/m3
) and the CZ (median
16.4 ng/m3
, range 3.62–139 ng/m3
). See Fig. S2 and Table 2 for details.
The abundance of TCIPP in air is consistent with its common use and
relatively high vapor pressure (log Pl −2.1 EPI Suite V4.1; − 1.3 Pa,
ABSOLV; Zhang et al., 2016).
3.1. Dust
The highest concentrations of halogenated OPEs were found in US
dust, with medians ranging from 1440 ng/g (TCEP) to 4530 ng/g
(TDBPP), followed by CAN and CZ. With the exception of TCIPP, the
differences between US and the other two countries were significant
(p < 0.05). TDCIPP was significantly lower in CZ than in CAN and US
while TDBPP was not detected in CZ at all.
Among non-halogenated OPEs, TPHP was generally most abundant.
TPHP was significantly lower in CZ (median 811 ng/g) than in the US
(median 3040 ng/g) and in CAN (median 2350 ng/g). In contrast,
EHDPP did not follow this trend, with similar concentrations in all
countries. For TNBP, TBPP and TIPPP, levels were significantly higher
in US than in the other two countries. Both isomers of TMPP were
significantly higher in CZ than in CAN. In the US, p-TMPP showed similar
concentrations to CZ, while o-TMPP was not detected in the US.
TDMPP was only detected in a single US dust sample.
Concentrations of OPEs in dust samples measured here were comparable
to those reported in the literature, with a few exceptions (see
Table S3 for a comprehensive list of literature data). Remarkably higher
levels of TCEP and TDCIPP were found in Sweden by Bergh et al.
(2011). TCEP concentrations in Spain (García et al., 2007) were higher
than those in CZ homes reported here but sampling occurred at a much
earlier year, before any restrictions were put in place. Dust measured in
California by Dodson et al. (2012) was higher than our data from Indiana,
which could reflect the legacy of more stringent flammability
standards in California. TDBPP in dust was only reported by Dodson
et al. (2012) for California and were lower than those measured here.
Somewhat higher TNBP concentrations than here were reported by
several European studies (Bergh et al., 2011; García et al., 2007;
Marklund et al., 2003; Van den Eede et al., 2011) and higher TPHP
values were found in the US by Stapleton et al. (2009) compared with
our results. Conversely, lower TPHP values were found in Germany by
Brommer et al. (2012) and Sweden (Marklund et al., 2003) relative to
results from all our locations. While TEHP values measured in dust were
higher in our study compared to the literature, o-TMPP were lower than
those reported by Bergh et al. (2011) for Sweden.
3.2. Window film
The highest median concentration (566 ng/m2
, TCIPP) among halogenated
OPEs was found in CZ window film samples. TDBPP was only
detected in a single CAN sample. While TDCIPP values were found to be
significantly higher in the US and CAN than in CZ, a reverse trend was
apparent in the case of TCIPP. Finally, TCEP was found to be significantly
lower in CAN than in the US and CZ.
For non-halogenated OPEs, the highest median concentrations in
window film samples were measured for TNBP in CAN and CZ (29.6
and 72.6 ng/m2
, respectively) and EHDPP in the US (48.1 ng/m2
). The
remaining non-halogenated OPEs (o-TMPP, p-TMPP, TIPPP, TDMPP
and TBPP) were measured at low concentration levels (range: 0.134 to
70.5 ng/m2
) and with lower detection frequencies. Most of the nonhalogenated
OPEs in window film were significantly higher in North
America than in CZ, with the exception of TNBP, for which concentrations
were significantly higher in CAN and CZ. To our knowledge,
no literature data exist for OPEs in window film.
M. Vykoukalová et al. Environment International 106 (2017) 97–104
100
3.3. Air
As mentioned above, the highest median concentrations among
halogenated OPEs in air were measured for TCIPP, with CAN values
(73.6 ng/m3
) significantly higher than those of the US (26.3 ng/m3
)
and CZ (16.4 ng/m3
). TCEP concentrations in CAN and the US were
significantly higher than in CZ which is consistent with regulatory actions
taken against TCEP in Europe and Canada. No significant differences
between countries were found for TDCIPP. While TDBPP was not
detected in CAN or CZ, the US detection frequency was up to 90% and
the median concentration was 23.6 ng/m3
.
We observed higher concentrations in the US and CAN compared to
the CZ for TNBP (6.20 ng/m3
in CAN), and EHDPP (1.71 ng/m3
in
CAN), while TEHP was found to be significantly higher in US only
(0.376 ng/m3
). No significant country trend was observed for TPHP
with median concentrations ranging from 0.592 ng/m3
in CZ to
0.799 ng/m3
in the US. Similarly to window film samples, the remaining
non-halogenated OPEs were detected rarely or at concentrations
just above the limit of quantification. The only exception was
TBPP, which was frequently detected in US, reaching a median concentration
of 0.439 ng/m3
.
Similarly to previous studies, TCEP air concentrations were substantially
lower than TCIPP concentrations (see Table SI3), which may
be explained by the recent restrictions on the uses of TCEP and the
associated rising usage of TCIPP as discussed above (Leisewitz et al.,
2001). Lower TCIPP than TCEP concentrations were reported for dust
only in Sweden (Bergh et al., 2011) and in the US (Dodson et al., 2012).
TCIPP was the most frequently measured halogenated OPE in our
study. This reflects its use as a major FR globally. To the authors'
knowledge, no data exist for TDBPP in indoor home air. TDCIPP concentrations
below the detection limit were reported in previous studies
(Bergh et al., 2011; Bergh et al., 2010; Marklund et al., 2005), which is
consistent with the very low concentrations found in our study and its
low vapor pressure (log Pl of −4.40, EPI Suite V4.1; −4.44 ABSOLV;
Zhang et al., 2016). US, CAN and CZ air concentrations of TCEP and
TCIPP were similar or slightly higher than those reported by previous
studies from both continents (Bergh et al., 2011; Bergh et al., 2010;
Marklund et al., 2005; Staaf and Ostman, 2005). No comparable data
for all measured non-halogenated OPEs in indoor home air currently
exist, with the exception of TNBP. Concentrations of TNBP in the literature
are comparable to those reported here (see Table S3).
3.4. Correlations among concentrations
The results of the Spearman correlation analysis between the log
transformed concentrations of all the OPEs from this study and of all the
brominated flame retardants (BFRs) reported by Venier et al. (2016) are
included in the Supporting Information (see Table S4). Correlations
Table 2
Summary results for dust (ng/g), window film (ng/m2
) and air (ng/m3
) samples including median, minimum, maximum, percentage of samples above the detection limit (%), and
Kruskal-Wallis test results. The Kruskal-Wallis results are to be read across countries for each compound; compound concentrations from countries that have different letter are
significantly different from one another (p < 0.05). ND indicates that a compound was not detected in the samples.
US CAN CZ
Median Range % Median Range % Median Range %
Dust [ng/g] TCEP 1440 123 – 9080 100 a 181 73.7 – 6750 100 b 155 41.8 – 1680 100 b
TCIPP 2790 177 – 26,500 100 a 1470 270 – 39,300 100 a 1860 163 – 26,700 100 a
TDCIPP 3680 ND – 8940 97 a 917 206 – 9530 100 b 183 56 – 1220 100 c
TDBPP 4530 ND – 11,000 60 a 0.193 ND – 41.4 26 b ND
TPHP 3040 22.5 – 24,800 100 a 2350 377 – 31,900 100 a 811 72.8 – 11,000 100 b
EHDPP 889 ND – 4760 97 a 754 150 – 8400 100 a 836 144 – 8230 100 a
TEHP 1360 ND – 18,600 97 a 101 18.6 – 611 100 b 153 37.2 – 10,300 100 b
TNBP 114 81.6 – 4270 100 a 63 25.5 – 288 100 b 51.6 6.81 – 160 100 b
o-TMPP ND 0.712 ND – 7.02 71 a 3.34 ND – 249 b
p-TMPP 82.5 ND – 453 47 a 6.18 0.871 – 116 100 b 201 14 – 2500 100 a
TIPPP 52.4 ND – 317 77 a 20.4 4.26 – 74.3 100 b 11.9 1.64 – 508 100 b
TDMPP 35.2 ND – 146 3 a ND ND
TBPP 41.8 ND – 500 73 a 4.85 0.959 – 57.7 100 b 6.49 0.254 – 39.6 100 b
Window film [ng/m2
] TCEP 56.5 8.17 – 267 100 a 12.5 5.09 – 258 97 b 46.1 2.35 – 1630 97 a
TCIPP 41.7 4.42 – 959 100 b 57.9 24.6 – 3620 97 b 566 15 – 6670 97 a
TDCIPP 83.2 ND – 542 90 a 82.3 20.8 – 3590 97 a 13.6 1.01 – 181 97 b
TDBPP ND 0.060* ND – 1.31 3 a ND 0
TPHP 39.6 4.2 – 1050 100 a 20.2 8.82 – 426 97 a 4.56 0.961 – 25.3 97 b
EHDPP 48.1 3.66 – 656 100 a 27.1 9.68 – 491 97 a 9.88 ND – 103 93 b
TEHP 23.2 ND – 283 63 a 3.72 0.978 – 49.6 97 b 6.85 0.749 – 91.9 97 b
TNBP 8.65 3.14 – 123 100 b 29.8 12.1 – 1530 97 a 72.6 2.93 – 538 97 a
o-TMPP 7.45 ND – 51.2 7 a 0.354 ND – 7.76 51 b 0.006 ND – 1.17 c
p-TMPP 12.7 ND – 22.1 7 a 0.18 ND – 13 43 b 0.02 ND – 1.57 47 b
TIPPP 6.36 ND – 70.5 7 a 0.34 0.134 – 29.2 97 b 0.084 ND – 0.223 90 c
TDMPP ND ND 0.401 ND – 0.814 3 a
TBPP 6.33 ND – 39.8 7 a 0.089 ND – 1.89 63 b 0.009 ND – 39.8 10 c
Air [ng/m3
] TCEP 6.81 ND – 28.9 97 a 6.35 1.34 – 145 97 a 2.96 0.781 – 36.4 97 b
TCIPP 26.3 0.28 – 226 100 b 73.6 7.68 – 4190 97 a 16.4 3.62 – 139 97 b
TDCIPP 0.372 ND – 6.46 80 a 0.525 0.045 – 5.5 97 a 0.311 0.228 – 0.572 97 a
TDBPP 23.6 ND – 45 a ND 0
TPHP 0.799 ND – 3.31 97 a 0.723 0.21 – 2.99 97 a 0.592 0.304 – 0.91 97 a
EHDPP 0.739 ND – 9.58 97 b 1.71 0.369 – 66.5 97 a 0.375 0.193 – 1.09 97 c
TEHP 0.376 ND – 24.7 53 a 0.042 0.014 – 2.38 97 b 0.037 0.02 – 0.163 97 b
TNBP 6.07 ND – 57 97 ab 6.2 2.61 – 461 97 a 2.34 0.745 – 33.1 97 b
o-TMPP 0.142 ND – 0.142 3 a 0.005 ND – 0.05 77 b 0.001 ND – 0.025 c
p-TMPP 0.256 ND – 0.256 7 a 0.002 ND – 0.023 71 b 0.002 ND – 0.016 63 b
TIPPP 0.121 ND – 0.121 7 a 0.007 0.002 – 0.034 97 b 0.005 0.003 – 0.023 97 c
TDMPP 0.121 ND – 0.121 a ND ND
TBPP 0.439 ND – 1.1 97 a ND 0.001 ND – 0.001 3 b
M. Vykoukalová et al. Environment International 106 (2017) 97–104
101
were done for each matrix separately and only one room per house was
included (i.e. the living room, since not all the houses had samples from
the bedroom).
The strongest correlations between OPE concentrations and between
OPE and BFR concentrations were measured in dust, which was
probably due to higher concentrations and higher detection frequency
in dust than in air or window film. Concentrations of most of the OPEs
were correlated with one another, with a few exceptions. For example,
EHDPP in dust correlated only with the other non-haolgenated OPEs
TPHP, TIPPP and TBPP. EHDPP was not correlated with any of the
PBDE congeners and with very few of the alternative FRs (namely
Pentabromoetheyl benzene (PBEB) and Dechlorane Plus). TCIPP in dust
and window film were not correlated with any of the PBDEs, except for
BDE209 and a few non PBDEs FRs. The opposite trend was evident in
air concentrations. Conversely, TDCIPP was significantly correlated
with PBDEs in dust and window film but not in air.
TPHP was significantly correlated with both EHTBB and BEHTBP in
dust and window film (r = 0.32, p = 0.02 for EHTBB in dust; r = 0.28,
p = 0.006 for BEHTBP in dust; r = 0.70 p < 0.0001 for EHTBB in
window film; r = 0.58, p < 0.0001 for BEHTBP in window film),
suggesting that in these samples these three chemicals could share the
same source. Differing from our study, Brandsma et al. (2014) did not
find a relationship between TPHP and the two brominated components
of FM 550 in house and car dust in Europe. They speculated that TPHP
had sources other than FM 550 (e.g., electronics). This result seems to
suggest that FM 550 was also used in Europe, since the correlation was
significant also for CZ samples alone (although only for window film
samples). TPHP in dust and window film was also significantly
correlated with BDE 209 and DBDPE. TPHP is a by-product in the
technical mixture of resorcinol bis and bisphenol A bis, both alternatives
to BDE 209, which was also used in conjunction with DBDPE
(Ballesteros-Gómez et al., 2016; Brandsma et al., 2014).
A comparison of OPE concentrations with BFR concentrations, using
data reported by Venier et al. (2016), revealed that OPEs were generally
higher than BFRs in these houses. In particular, the sum of all
OPEs was 1 to 2 orders of magnitude higher than the sum of all BFRs for
all countries for window film and dust, and 2 to 3 orders of magnitude
higher than the sum of all BFRs for all countries for air.
3.5. Comparison of compound distributions between rooms
In order to assess variability of FR concentrations between rooms
within one home, results for sample pairs collected from both the main
bedroom and the living room area were compared for at least nine
homes in each country. Differences between rooms were tested for each
matrix and country separately using the Mann-Whitney U test (for
pooled samples) and the Wilcoxon matched pairs test (for room pairs).
The Mann-Whitney U test results showed no significant differences
(p > 0.05) between individual OPEs with the exception of TDCIPP
(p = 0.014) in CZ dust samples (see Table S4), where concentrations
were systematically higher in bedrooms. The Wilcoxon matched pairs
test, which is more powerful when evaluating differences within a
single home, showed an additional six significant cases: TCIPP, TDBPP
and TBPP in US dust samples, TDCIPP and TIPPP in CAN air samples,
and TCIPP in CZ air samples (p < 0.05, see Table SI). While the concentrations
of TCIPP, TDBPP and TBPP in US dust samples and TDCIPP
in CAN air samples were higher in bedrooms, concentrations of TIPPP
in CAN air samples and TCIPP in CZ air samples were higher in living
rooms.
Overall, these results suggest that differences between individual
rooms are negligible, similarly to what was previously reported by
Venier et al. (2016) for BFR in these samples. The only other study on
this topic (Muenhor and Harrad, 2012), reported some room-to-room
differences, albeit not statistically significant. They suggested that the
dust concentrations were related to the number of specific items (such
as electronic products) in each room. To our knowledge, no other study
has done a systematic room-to-room comparison on a relatively large
number of samples (n = 30).
3.6. Correlations among air, dust and window film
Next, we considered the relationship among chemical concentrations
in air, dust and window film for two reasons. First, this comparison
tested the hypothesis for OPEs that under equilibrium conditions,
the partitioning between dust and air and between window film and air
is expected to be proportional to a chemical's KOA. This relationship has
been used for characterizing the sorption of chemicals to organic matter
in dust and window films (Butt et al., 2004; Weschler and Nazaroff,
2010). Second, if equilibrium has been achieved, then the concentrations
of a chemical can be assessed from measurements in one rather
than more media, as a way to reduce sampling effort.
Fig. 1 shows the dependence on log KOA of the partition coefficients
between dust and air (top) and between window film and air (bottom).
The log KOA values were those from the US EPA's EPI Suite (see
Table 1). Fig. 1 shows that the partition coefficients between dust and
air, and window film and air increased for compounds with log KOA
between 7 and 12, beyond which the coefficients decreased. Similar
partitioning behavior was observed for PBDEs and BFRs with a “break
point” of KOA > 14 (Venier et al., 2016) and for PBDEs and PCBs with
a “break point” of KOA > 11 (Zhang et al., 2011). Surprisingly, the
quadratic curve used to describe this partitioning behavior for OPEs
was quite different from that of BFRs. For example, for OPEs, the
maximum value of the log dust/air partition coefficient was 3.4 at log
KOA 12 whereas the value for BFRs from Venier et al. (2016) was 7.6 at
Fig. 1. Dependence of the dust/air partition coefficients and of window film/air partition
coefficients on the octanol/air partition coefficient (log KOA). In both graphs each circle
represents the average and the bars the standard error. The equations of the regression
curves were: Y = −11.59 + 2.59 X − 0.1099 × 2
(top) and Y = −5.60 + 1.24
X − 0.0511 × 2
(bottom).
M. Vykoukalová et al. Environment International 106 (2017) 97–104
102
log KOA 14 (see Fig. 5 in Venier et al., 2016). Taking into account the
logarithmic scale, these differences translate to about 4 orders of
magnitude difference in the value of the partition coefficient at a 2
order of magnitude offset in log KOA. This large difference in partition
coefficients suggests that OPEs have a significantly stronger tendency
than BFRs to partition into the air than in dust, assuming that values of
KOA are accurate and that OPEs follow expected gas-particle partitioning
behavior, as reflected in PUF passive sampler collection. A similar
difference was observed also for window film/air, with the
logarithms of the partition coefficient reaching a maximum of 1.9 for
OPEs at log KOA 12 compared with 5.64 for BFRs at log KOA 14. Again,
this result is surprising since one would expect similar partitioning
behavior of OPEs and BFRs at the same values of KOA. The difference
seen in partitioning behavior is consistent with the lower sampling rates
for OPEs than for BFRs for the PUF passive air samplers used (0.82 m3
/
day for double bowl and 1.6 m3
/day for single bowl for OPEs versus
1.6 m3
/day and 2.9 m3
/day, respectively, for BFRs).
Based on the analysis described above, we hypothesize that equilibrium
between air and dust and between window film and dust was not
reached for compounds with log KOA > 12. Alternatively, since the
total concentration on PUF was used in the calculations, air concentrations
could have over-estimated the gas phase by capturing also
particle-phase compounds (Abdollahi et al., 2017). OPEs have been
found in high percentages in the particle phase (Sühring et al., 2016b).
The data presented here do not allow for distinguishing between these
two hypotheses.
Using the same approach described in our companion paper (Venier
et al., 2016), we estimated air concentrations using dust and window
film for compounds with log KOA < 12 (see the Supporting Information
for details on the calculations). Fig. 2 shows the relationship between
the median measured air concentrations and the median gasphase
concentrations predicted from dust (top) and window film
(bottom) for compounds that appeared to reach equilibrium. For both
window film and dust, the predicted and measured values were correlated
(r2
= 0.84 and 0.68, respectively). For both the window film and
dust, the slopes of the linear regression were close to unity
(1.91 ± 0.31 and 1.18 ± 0.31, respectively). The negative values of
the intercepts (−0.60 for window film and −0.28 for dust), albeit not
statistically significant, indicated that, in general, the air concentrations
predicted from either window film or dust are lower than the measured
air concentrations (10–0.60
= 0.25 and 10–0.28
= 0.52). This discrepancy
between measured and predicted air concentration might be
due to assuming too large value for the organic matter fraction in both
media or for the thickness of the film layer. However, based on these
results, we recommend that for OPEs with log KOA < 12 that either air,
dust or window films can be sampled, but not all are necessary.
The partitioning analysis suggests that several OPEs compounds (i.e.
those with log KOA > 12) did not reach equilibrium between air, dust
and window film and, differently from BFRs, the three matrices used
here are not equivalent. The partitioning behavior suggests that OPEs
tend to accumulate more in air than dust or window film compared to
PBDEs and BFRs.
3.7. Limitations
OPEs are challenging compounds to measure, as reflected in the
large variability in the measurements of a standard reference material
(see Table S2 and discussion in the Supporting Information). While
different analytical methods between IU and RECETOX could have
contributed to differences or lack thereof, the regional trends are in
agreement with the usage history of FRs (US > CAN > CZ). If differences
due to analytical methods would be at play here, they would
not affect the partitioning and prediction modeling nor the correlations
or comparisons with BFRs. It should be noted that, although samples in
the three countries were collected in a relatively short time window
(May–August), we acknowledge that seasonal variability might play a
role in the differences we have observed. Hoffman et al. (2017) reported
higher exposure concentrations and hence exposures for TDCIPP
and TPHP in the summer than in winter.
Acknowledgments
This work was supported by the Czech-American Scientific
Cooperation Program (AMVIS/KONTAKT II, LH12074) and U. S.
Environmental Protection Agency's Great Lakes National Program
Office (Grant GL 00E01422, Todd Nettesheim, project officer).
RECETOX research infrastructure was supported by the Czech Ministry
of Education (LM2015051LM2011028, LO1214). Support in Canada
was provided by Allergy, Genes and Environment Network
(AllerGenNCE) and the Natural Sciences and Engineering Research
Council of Canada (NSERC). We thank all volunteers in Bloomington
(IN, USA), Toronto (Canada) and Brno (Czech Republic) for their par-
ticipation.
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.envint.2017.05.020.
References
Abdollahi, A., Eng, A., Jantunen, L.M., Ahrens, L., Shoeib, M., Parnis, J.M., Harner, T.,
2017. Characterization of polyurethane foam (PUF) and sorbent impregnated PUF
(SIP) disk passive air samplers for measuring organophosphate flame retardants.
Fig. 2. Dependence of the average measured air concentrations versus the concentrations
predicted using window wipes (top) and dust (bottom). Each dot represents the logarithm
of the mean concentration. Only compounds with log KOA > 12 are included. The
equations for the linear regressions are Y = −0.60 + 1.91 X (r2
= 0.84, p = 0.0005) for
window film and Y = −0.28 + 1.18 X (r2
= 0.68, p = 0.0065) for dust.
M. Vykoukalová et al. Environment International 106 (2017) 97–104
103
Chemosphere 167, 212–219.
Andresen, J.A., Grundmann, A., Bester, K., 2004. Organophosphorus flame retardants and
plasticisers in surface waters. Sci. Total Environ. 332, 155–166.
Antweiler, R.C., 2015. Evaluation of statistical treatments of left-censored environmental
data using coincident uncensored data sets: II. Group Comparisons. Environ. Sci.
Technol. 49, 13439–13446.
Ballesteros-Gómez, A., Aragón, Á., Van den Eede, N., de Boer, J., Covaci, A., 2016.
Impurities of resorcinol Bis(diphenyl phosphate) in plastics and dust collected on
electric/electronic material. Environ. Sci. Technol. 50, 1934–1940.
Behl, M., Hsieh, J.-H., Shafer, T.J., Mundy, W.R., Rice, J.R., Boyd, W.A., Freedman, J.H.,
Hunter III, E.S., Jarema, K.A., Padilla, S., Tice, R.R., 2015. Use of alternative assays to
identify and prioritize organophosphorus flame retardants for potential developmental
and neurotoxicity. Neurotoxicol. Teratol. 52, 181–193.
Bergh, C., Torgrip, R., Östman, C., 2010. Simultaneous selective detection of organophosphate
and phthalate esters using gas chromatography with positive ion chemical
ionization tandem mass spectrometry and its application to indoor air and dust.
Rapid Commun. Mass Spectrom. 24, 2859–2867.
Bergh, C., Torgrip, R., Emenius, G., Östman, C., 2011. Organophosphate and phthalate
esters in air and settled dust – a multi-location indoor study. Indoor Air 21, 67–76.
Brandsma, S.H., de Boer, J., van Velzen, M.J.M., Leonards, P.E.G., 2014.
Organophosphorus flame retardants (PFRs) and plasticizers in house and car dust and
the influence of electronic equipment. Chemosphere 116, 3–9.
Brommer, S., Harrad, S., Van den Eede, N., Covaci, A., 2012. Concentrations of organophosphate
esters and brominated flame retardants in German indoor dust samples. J.
Environ. Monit. 14, 2482–2487.
Brooke, D.N., Crookes, M.J., Quarterman, P., Burns, J., 2009. Environmental risk evaluation
report: 2-Ethylhexyl diphenyl phosphate (CAS no. 1241-94-7) SCHO0809BQTY-E-P.
UK Environment Agency, Bristol, UK.
Butt, C.M., Diamond, M.L., Truong, J., Ikonomou, M.G., ter Schure, A.F.H., 2004. Spatial
distribution of Polybrominated diphenyl ethers in southern Ontario as measured in
indoor and outdoor window organic films. Environ. Sci. Technol. 38, 724–731.
Cao, S.X., Zeng, X.Y., Song, H., Li, H.R., Yu, Z.Q., Sheng, G.Y., Fu, J.M., 2012. Levels and
distributions of organophosphate flame retardants and plasticizers in sediment from
Taihu Lake, China. Environ. Toxicol. Chem. 31, 1478–1484.
Cooper, E.M., Kroeger, G., Davis, K., Clark, C.R., Ferguson, P.L., Stapleton, H.M., 2016.
Results from screening polyurethane foam based consumer products for flame retardant
chemicals: assessing impacts on the change in the furniture flammability
standards. Environ. Sci. Technol. 50, 10653–10660.
Dodson, R.E., Perovich, L.J., Covaci, A., Van den Eede, N., Ionas, A.C., Dirtu, A.C., Brody,
J.G., Rudel, R.A., 2012. After the PBDE phase-out: a broad suite of flame retardants in
repeat house dust samples from California. Environ. Sci. Technol. 46, 13056–13066.
Environment and Climate Change Canada, 2016. List of Toxic Substances Managed Under
CEPA (Schedule 1).
European Chemicals Agency, 2015. List of Substances Included in Annex XIV of REACH
("Authorisation List").
García, M., Rodríguez, I., Cela, R., 2007. Microwave-assisted extraction of organophosphate
flame retardants and plasticizers from indoor dust samples. J. Chromatogr. A
1152, 280–286.
Gazette, Canada, 2014. Regulations amending schedule 2 to the Canada consumer product
safety act (TCEP). http://www.gazette.gc.ca/rp-pr/p2/2014/2014-04-23/html/
sor-dors79-eng.php (accessed on 5/12/2017).
Gold, M., Blum, A., Ames, B., 1978. Another flame retardant, tris-(1,3-dichloro-2-propyl)phosphate,
and its expected metabolites are mutagens. Science 200, 785–787.
Greaves, A.K., Letcher, R.J., Chen, D., McGoldrick, D.J., Gauthier, L.T., Backus, S.M.,
2016. Retrospective analysis of organophosphate flame retardants in herring gull
eggs and relation to the aquatic food web in the Laurentian Great Lakes of North
America. Environ. Res. 150, 255–263.
Green, N.S., Bakke, M., Brevik, T., Dye, E.M., Herzke, C., Huber, D., Plosz, S., Remberger,
B., Schoyen, M., Uggerud, H.T., Vogelsang, C., 2008. Screening of selected metals and
new organic contaminants 2007. Norwegian Pollution Control Agency.
Harrad, S., Ibarra, C., Diamond, M., Melymuk, L., Robson, M., Douwes, J., Roosens, L.,
Dirtu, A.C., Covaci, A., 2008. Polybrominated diphenyl ethers in domestic indoor
dust from Canada, New Zealand, United Kingdom and United States. Environ. Int. 34,
232–238.
Hendriks, H.S., Westerink, R.H.S., 2015. Neurotoxicity and risk assessment of brominated
and alternative flame retardants. Neurotoxicol. Teratol. 52 (Part B), 248–269.
Hoffman, K., Butt, C.M., Webster, T.F., Preston, E.V., Hammel, S.C., Makey, C., Lorenzo,
A.M., Cooper, E.M., Carignan, C., Meeker, J.D., Hauser, R., Soubry, A., Murphy, S.K.,
Price, T.M., Hoyo, C., Mendelsohn, E., Congleton, J., Daniels, J.L., Stapleton, H.M.,
2017. Temporal trends in exposure to organophosphate flame retardants in the
United States. Environ. Sci. Technol. Lett. 4 (3), 112–118.
Kajiwara, N., Noma, Y., Takigami, H., 2011. Brominated and organophosphate flame
retardants in selected consumer products on the Japanese market in 2008. J. Hazard.
Mater. 192, 1250–1259.
Karásková, P., Venier, M., Melymuk, L., Bečanová, J., Vojta, Š., Prokeš, R., Diamond,
M.L., Klánová, J., 2016. Perfluorinated alkyl substances (PFASs) in household dust in
Central Europe and North America. Environ. Int. 94, 315–324.
Leisewitz, A., Kruse, H., Schramm, E., 2001. Substituting Environmentally Relevant
Flame Retardants: Assessment Fundamentals (Report No. UBA-FB 000171/1).
Federal Environmental Agency (Umweltbundesamt).
Marklund, A., Andersson, B., Haglund, P., 2003. Screening of organophosphorus compounds
and their distribution in various indoor environments. Chemosphere 53,
1137–1146.
Marklund, A., Andersson, B., Haglund, P., 2005. Traffic as a source of organophosphorus
flame retardants and plasticizers in snow. Environ. Sci. Technol. 39, 3555–3562.
Mendelsohn, E., Hagopian, A., Hoffman, K., Butt, C.M., Lorenzo, A., Congleton, J.,
Webster, T.F., Stapleton, H.M., 2016. Nail polish as a source of exposure to triphenyl
phosphate. Environ. Int. 86, 45–51.
Muenhor, D., Harrad, S., 2012. Within-room and within-building temporal and spatial
variations in concentrations of polybrominated diphenyl ethers (PBDEs) in indoor
dust. Environ. Int. 47, 23–27.
Rauert, C., Harrad, S., 2015. Mass transfer of PBDEs from plastic TV casing to indoor dust
via three migration pathways — a test chamber investigation. Sci. Total Environ. 536,
568–574.
Regnery, J., Puttmann, W., Merz, C., Berthold, G., 2011. Occurrence and distribution of
organophosphorus flame retardants and plasticizers in anthropogenically affected
groundwater. J. Environ. Monit. 13, 347–354.
Salamova, A., Peverly, A.A., Venier, M., Hites, R.A., 2016. Spatial and temporal trends of
particle phase organophosphate ester concentrations in the atmosphere of the Great
Lakes. Environ. Sci. Technol.
Schreder, E.D., Uding, N., La Guardia, M.J., 2016. Inhalation a significant exposure route
for chlorinated organophosphate flame retardants. Chemosphere 150, 499–504.
Schweizer, C., Edwards, R.D., Bayer-Oglesby, L., Gauderman, W.J., Ilacqua, V., Jantunen,
M.J., Lai, H.K., Nieuwenhuijsen, M., Kunzli, N., 2007. Indoor time-microenvironment-activity
patterns in seven regions of Europe. J. Expo. Sci. Environ. Epidemiol.
17, 170–181.
Staaf, T., Ostman, C., 2005. Indoor air sampling of organophosphate triesters using solid
phase extraction (SPE) adsorbents. J. Environ. Monit. 7, 344–348.
Stapleton, H.M., Klosterhaus, S., Eagle, S., Fuh, J., Meeker, J.D., Blum, A., Webster, T.F.,
2009. Detection of organophosphate flame retardants in furniture foam and US house
dust. Environ. Sci. Technol. 43, 7490–7495.
Sühring, R., Diamond, M.L., Scheringer, M., Wong, F., Pućko, M., Stern, G., Burt, A.,
Hung, H., Fellin, P., Li, H., Jantunen, L.M., 2016a. Organophosphate esters in
Canadian Arctic air: occurrence levels and trends. Environ. Sci. Technol. 50,
7409–7415.
Sühring, R., Wolschke, H., Diamond, M.L., Jantunen, L.M., Scheringer, M., 2016b.
Distribution of organophosphate esters between the gas and particle phase–model
predictions vs measured data. Environmental Science & Technology. 50, 6644–6651.
Sundkvist, A.M., Olofsson, U., Haglund, P., 2010. Organophosphorus flame retardants
and plasticizers in marine and fresh water biota and in human milk. J. Environ.
Monit. 12, 943–951.
Toy Safety Directive, 2009. Directive 2009/48/EC of the European Parliament and of the
Council of 18 June 2009 on the safety of toys. /48/EC. http://eur-lex.europa.eu/
LexUriServ/LexUriServ.do?uri=OJ:L:2009:170:0001:0037:en:PDF (accessed on 5/
12/2017).
Van den Eede, N., Dirtu, A.C., Neels, H., Covaci, A., 2011. Analytical developments and
preliminary assessment of human exposure to organophosphate flame retardants
from indoor dust. Environ. Int. 37, 454–461.
van der Veen, I., de Boer, J., 2012. Phosphorus flame retardants: properties, production,
environmental occurrence, toxicity and analysis. Chemosphere 88, 1119–1153.
Venier, M., Dove, A., Romanak, K., Backus, S., Hites, R., 2014. Flame retardants and
legacy Chemicals in Great Lakes' water. Environ. Sci. Technol. 48, 9563–9572.
Venier, M., Audy, O., Vojta, Š., Bečanová, J., Romanak, K., Melymuk, L., Krátká, M.,
Kukučka, P., Okeme, J., Saini, A., 2016. Brominated flame retardants in the indoor
environment—comparative study of indoor contamination from three countries.
Environ. Int. 94, 150–160.
Wei, G.L., Li, D.Q., Zhuo, M.N., Liao, Y.S., Xie, Z.Y., Guo, T.L., Li, J.J., Zhang, S.Y., Liang,
Z.Q., 2015. Organophosphorus flame retardants and plasticizers: sources, occurrence,
toxicity and human exposure. Environ. Pollut. 196, 29–46.
Weschler, C.J., Nazaroff, W.W., 2010. SVOC partitioning between the gas phase and
settled dust indoors. Atmos. Environ. 44, 3609–3620.
Zhang, X., Diamond, M.L., Robson, M., Harrad, S., 2011. Sources, emissions, and fate of
Polybrominated diphenyl ethers and polychlorinated biphenyls indoors in Toronto,
Canada. Environ. Sci. Technol. 45, 3268–3274.
Zhang, X., Sühring, R., Serodio, D., Bonnell, M., Sundin, N., Diamond, M.L., 2016. Novel
flame retardants: estimating the physical-chemical properties and environmental fate
of 94 halogenated and organophosphate PBDE replacements. Chemosphere 144,
2401–2407.
M. Vykoukalová et al. Environment International 106 (2017) 97–104
104
Melymuk, Lisa, Pernilla Bohlin-Nizzetto, Petr Kukučka, Šimon Vojta, Jiří Kalina, Pavel Cupr,
Jana Klánová, Pavel Čupr, and Jana Klánová. 2016. "Seasonality and Indoor/Outdoor
Relationships of Flame Retardants and PCBs in Residential Air." Environmental Pollution
218: 392-401. https://doi.org/10.1016/j.envpol.2016.07.018
Seasonality and indoor/outdoor relationships of flame retardants and
PCBs in residential air*
Lisa Melymuk a, *
, Pernilla Bohlin-Nizzetto b, **
, Petr Kukucka a
, Simon Vojta a
, Jirí Kalina a
,
Pavel Cupr a
, Jana Klanova a
a
Research Centre for Toxic Compounds in the Environment (RECETOX), Kamenice 753/5, 62500, Brno, Czech Republic
b
NILU e Norwegian Institute for Air Research, Instituttveien 18, PO Box 100, NO-2017, Kjeller, Norway
a r t i c l e i n f o
Article history:
Received 1 March 2016
Received in revised form
13 June 2016
Accepted 5 July 2016
Available online 16 July 2016
Keywords:
Indoor air
Flame retardants
PCBs
Indoor sources
Seasonal trends
a b s t r a c t
This study is a systematic assessment of different houses and apartments, their ages and renovation
status, indoors and outdoors, and in summer vs. winter, with a goal of bringing some insight into the
major sources of semivolatile organic compounds (SVOCs) and their variability. Indoor and outdoor air
concentrations of polychlorinated biphenyls (PCBs), polybrominated diphenyl ethers (PBDEs) and novel
flame retardants (NFRs) were determined at 17e20 homes in Czech Republic in winter and summer.
Indoor concentrations were consistently higher than outdoor concentrations for all compounds; indoor/outdoor
ratios ranged from 2e20, with larger differences for the current use NFRs than for legacy
PCBs. Seasonal trends differed according to the use status of the compounds: the PCBs had higher
summer concentrations both indoors and outdoors, suggesting volatilization as a source of PCBs to air.
PBDEs had no seasonal trends indoors, but higher summer concentrations outdoors. Several NFRs (TBX,
PBT, PBEB) had higher indoor concentrations in winter relative to summer. The seasonal trends in the
flame retardants suggest differences in air exchange rates due to lower building ventilation in winter
could be driving the concentration differences.
Weak relationships were found with building age for PCBs, with higher concentrations indoors in
buildings built before 1984, and with the number of electronics for PBDEs, with higher concentrations in
rooms with three or more electronic items. Indoor environments are the primary contributor to human
inhalation exposure to these SVOCs, due to the high percentage of time spent indoors (>90%) combined
with the higher indoors levels for all the studied compounds. Exposure via the indoor environment
contributed ~96% of the total chronic daily intake via inhalation in summer and ~98% in winter.
© 2016 Elsevier Ltd. All rights reserved.
1. Introduction
The quality of air in a non-industrial indoor environment (e.g., a
home, school, office, public building) is affected by a wide range of
factors, including pollutants from sources inside the buildings as
well as from outdoor air. Knowledge on the fate and distribution of
indoor air pollutants is crucial since 80e90% of our total life time is
spent in different indoor environments. Moreover, people who are
most susceptible to the adverse effects of air pollutants (e.g. infants,
elderly, etc.) spend even more time indoors, and thus inhalation of
indoor air can significantly contribute to the total human exposure
to pollutants. Until recently, the focus on indoor air quality has been
on volatile organic compounds (VOCs), inorganics, radon and particulate
matter, and the World Health Organization (WHO)
Regional Office for Europe has published indoor air quality guidelines
values for these pollutants (WHO European Centre for
Environment and Health, 2010).
A more recent focus is the semivolatile organic compounds
(SVOCs), which also have a plethora of indoor sources. Many
products containing SVOCs are used predominantly in indoor environments
and many indoor sources remain active despite bans
and regulations (e.g., Stockholm Convention). Indoor environments
also contain hydrophobic, sorptive materials that may act as reservoirs
and thereby sources of secondary emissions, even after the
removal of primary sources. This may be enhanced by increased
*
This paper has been recommended for acceptance by David Carpenter.
* Corresponding author.
** Corresponding author.
E-mail addresses: melymuk@recetox.muni.cz (L. Melymuk), pbn@nilu.no
(P. Bohlin-Nizzetto).
Contents lists available at ScienceDirect
Environmental Pollution
journal homepage: www.elsevier.com/locate/envpol
http://dx.doi.org/10.1016/j.envpol.2016.07.018
0269-7491/© 2016 Elsevier Ltd. All rights reserved.
Environmental Pollution 218 (2016) 392e401
persistence of these compounds indoors due to limited exposure to
direct sunlight, OH radicals and bacterial activity. In addition,
physical insulation and reduced ventilation in many indoor environments
may also enhance indoor concentrations. SVOCs are
therefore prone to accumulate and persist in indoor environments,
contributing to long term exposure. Indoor concentrations of many
SVOCs are several times higher than outdoor concentrations
(Menichini et al., 2007; Rudel et al., 2010; Zhang et al., 2011),
emphasizing the importance of the indoor environment as an
exposure route for these compounds. Furthermore, as the composition
of chemicals on the market changes due to regulation, e.g., a
shift from polybrominated diphenyl ethers (PBDEs) to replacement
“novel” halogenated flame retardants (NFRs), the chemicals to
which we are exposed indoors also shifts, and thus may lead to
differences between older and newer dwellings.
With this study we analyse: (i) the relationship between indoor
and outdoor air concentrations, (ii) seasonal variability, and (iii) the
relationship between indoor air concentrations and house characteristics
(i.e. year of construction, renovation status, and type of
house) and house contents (e.g., floor type, furniture, electronics).
Moreover, indoor data from Czech Republic increases the knowledge
on the global spatial distribution of indoor air measurements
in the less frequently studied area of Central and Eastern Europe.
We focus on SVOCs with links to building materials and/or consumer
products that cover a range of legacy and current sources
(Table 1): polychlorinated biphenyls (PCBs), PBDEs and NFRs.
2. Materials and methods
2.1. Sampling site
Air sampling was conducted in residential areas around Brno,
Czech Republic (metropolitan area with a population of 600,000).
The sampling sites were private dwellings located both in the urban
area and nearby semi-rural areas (maximum 40 km from city
centre). The residents were informed about the study and invited to
participate on a voluntary basis.
2.2. Experimental design
Twenty private houses were selected to cover a range of specific
characteristics: (i) house age, (ii) house status, (i.e. original
construction or renovated), and (iii) type of house (i.e. apartment/
flat in a multi-unit building, row house, or detached house). Polyurethane
foam passive air samplers (PUF-PAS) were deployed
concurrently inside and outside the homes for 28 days. The indoor
samplers were deployed in the main living area of the house at
~2 m height and the outdoor samplers on the adjacent balcony or
the garden of the house, with a distance of 0.5e10 m from the
building wall. In three homes, samplers were deployed both in
living rooms and in bedrooms to address spatial variability within
the residence. Two sampling campaigns were carried out to address
seasonal variability: one in summer (JulyeAugust 2010, avg. outdoor
air temperature 19 C) and one in winter (FebruaryeMarch
2011, avg. outdoor air temperature 0 C).
A questionnaire, completed by the residents at the end of the
sampling, provided additional information on type and age of
building materials (floors, walls, windows, insulation, etc.), type
and age of products in the home (furniture, carpets, electronics,
heating systems, water boiler, fireplaces, etc.), ventilation systems
and frequency of aeration (how often windows are kept open), and
living habits (cleaning, cooking, smoking, etc.). Details of the
buildings and questionnaires are presented in Supplementary Information
(SI).
2.3. Passive air sampling
The PUF-PAS disks were 15 cm diameter, 1.5 cm thickness,
424 cm2
total surface area, 0.030 g/cm3
density (type T-3037,
Molitan a.s., Czech Republic), and were deployed in protective
chambers consisting of two stainless steel bowls (upper 30 cm
diameter and lower 24 cm diameter). Methods for PUF-PAS preparation
and deployment are described by Bohlin et al. (2014a,b).
Two generic PUF-PAS sampling rates were used to convert sampled
masses to air concentrations: 1.4 m3
/day for indoor samplers
(Bohlin et al., 2014b) and 3.5 m3
/day for outdoor samplers (Bohlin
et al., 2014a).
Field validations of PUF-PAS have shown good performance for
gas-phase compounds such as PCBs both indoors and outdoors, yet
they have been shown to be a reliable sampling tool for only a
limited number of PBDE congeners (Bohlin et al., 2014a,b; Hazrati
and Harrad, 2007). Consistent results can be obtained for BDE 28,
47, 66, 85, 99, and 100, but the heavier, largely particle-bound BDEs
may not have a consistent sampling rate with the PUF-PAS,
Table 1
List of compounds.
Compound
class
Compound/congener Primary/possible uses in Czech Republic
PCBs CB-28, 52, 101, 118, 138, 153, 180 Electrical and industrial equipment, paints, pre-1984 (Holoubek et al., 2006)
PBDEs BDE-28, 47, 66, 85, 99, 100, 153, 154,
183, and 209
Flame retardant (FR) in polyurethane foams, electrical and electronic equipment, textiles, carpets, draperies, plastic
furniture, roofing, insulation, piping/ducting/hoses, cables (European Chemicals Agency, 2014; Prevedouros et al.,
2004)
NFRs 2,3-dibromopropyl 2,4,6tribromophenyl
ether (TBP-DBPE)
FR in polypropylene (CECBP, 2009), manufactured in Germany, production ceased in 1980s (Ma et al., 2012)
2-bromoallyl 2,4,6-tribromophenyl
ether (TBP-BAE)
Transformation product of TBP-DBPE (Ma et al., 2012)
tetrabromo-p-xylene (TBX) Additive FR in polymers and resins (Venier et al., 2012)
pentabromoethylbenzene (PBEB) FR in polyurethane foam, textiles, adhesives, wire and cable coatings, polyester resins, thermoset polyester resins
(Harju et al., 2009; Salamova and Hites, 2011)
pentabromotoluene (PBT) FR in bulk moulded polyesters (Harju et al., 2009)
hexabromobenzene (HBB) FR in polymers, wood, textiles, electronics (Salamova and Hites, 2011; Schlabach et al., 2011)
1,2-bis(2,4,6-tribromophenoxy)ethane
(BTBPE)
FR in thermoplastics, replacement for octaBDE (Schlabach et al., 2011)
Hexachlorocyclopentadienyldibromocyclooctane
(DBHCTD)
FR in styrenic polymers (Covaci et al., 2011; Riddell et al., 2008)
syn- and anti-Dechlorane Plus (s-DDCCO,
a-DDC-CO)
FR in electrical wires, cables and polymer products (Schlabach et al., 2011)
1,2-dibromo-4-(1,2-dibromoethyl)
cyclohexane (DBE-DBCH)
FR in polystyrene, polyethylene, building insulation (Schlabach et al., 2011)
L. Melymuk et al. / Environmental Pollution 218 (2016) 392e401 393
particularly in indoor environments with limited air flow, and thus
may have higher uncertainty, notably for BDE-209. Thus, when
discussing seasonal and site-related trends in PBDEs, we exclude
any detected levels of BDE-209, although we note that BDE-209
was detected at levels above the limit of detection (LOD) in 16%
of samples (Table S3).
2.4. Analysis
Laboratory and analytical methods are given in full in the SI and
have been previously published (Bohlin et al., 2014b). The target
compounds are listed in Table 1. Briefly, PUF-PAS were extracted
with dichloromethane (DCM) using automated warm Soxhlet
extraction, cleaned-up via a H2SO4 modified silica column eluted
with 1:1 DCM:n-hexane. Samples were analysed for PCBs using a
gas chromatograph (GC)-mass spectrometry (MS) system (Agilent
6890N GC coupled to Micromass Quattro Micro GC tandem quadrupole
MS) with a 60 m DB5-MS column. Samples were analysed
for PBDEs and NFRs using a GC-high resolution MS system (Agilent
7890A GC coupled to Micromass AutoSpec Premier sector MS) with
a 15 m DB5 column.
2.5. QA/QC
Two field blanks consisting of pre-cleaned PUFs were collected
for each summer and winter campaign (4 total). These field blanks,
along with laboratory and solvent blanks, were analysed as per the
samples. The LOD was calculated as the average field blanks plus
twice the standard deviation of the blanks. The samples above the
LOD were blank subtracted based on the average of the field blanks.
For compounds that were not detected in the blanks, instrumental
LODs were used.
Further information on QA/QC is given in Table S4.
2.6. Statistical analyses
Statistical Analyses were performed in GraphPad Prism (version
5.00), Microsoft Excel (Microsoft Office 2010), R studio (version
0.99.491 based on R version 3.2.3) and Maplesoft Maple (version
16). Concentrations were not normally distributed, thus nonparametric
tests were used (e.g., Mann-Whitney U-test) at
a ¼ 0.05. For the statistical analyses, values < LOD were substituted
by 0.5*LOD.
3. Results and discussion
3.1. PCBs
3.1.1. PCB air concentrations
PCBs were detected in all but one indoor air sample. Concentrations
of S7PCB (sum of PCBs 28, 52, 101, 118, 138, 153, and 180)
are presented in Table 2. The variability of the air concentrations
among different indoor sites was relatively small for PCBs (3e10
fold). Mean indoor air concentrations of S7PCB were 89 pg/m3
in
summer (median: 82 pg/m3
) and 61 pg/m3
in winter (median:
61 pg/m3
). These levels are 4e10 Â lower than PCB air concentrations
in indoor environments in Canada (Zhang et al., 2011),
Denmark (Frederiksen et al., 2012), France (Alliot et al., 2014), UK,
and Sweden (Bohlin et al., 2008), while comparable to indoor levels
in Mexico (Bohlin et al., 2008) and offices in Czech Republic (Bohlin
et al., 2014b) (Fig. 1).
Mean outdoor air concentrations of S7PCB were 38 pg/m3
in
summer (median: 34 pg/m3
) and 17 pg/m3
in winter (median:
19 pg/m3
). The winter concentrations were consistent with outdoor
air concentrations from urban and rural sites in the Czech Republic
(Bohlin et al., 2014a; Holoubek et al., 2007), Ontario, Canada
(Melymuk et al., 2012), Sweden, UK and Mexico (Bohlin et al.,
2008).
The low indoor air concentrations of PCBs in this study relative
to those measured in regions with known use of PCBs in building
materials (e.g., in building sealants, Frederiksen et al., 2012; Kohler
et al., 2005) supports the hypothesis of limited indoor primary
sources of PCBs in Czech Republic. Moreover, while evidence from
Western Europe and North America suggests PCB-containing
building sealants were typically used in larger concrete buildings
(Klosterhaus et al., 2014; Robson et al., 2010), air concentrations
here do not suggest any difference between the larger prefabricated
concrete residential buildings and single family homes. However,
we caution that with a small sample size, further work is necessary
to confirm this result. In agreement with this, no reports exist of the
use of PCBs in non-military building applications in Czech Republic.
However, although the indoor air concentrations of PCBs in this
study are lower than many other countries, the concentrations are
higher than the indoor air concentrations of PBDEs and NFRs, which
have known indoor use. Moreover, the measured outdoor air concentrations
of PCBs are similar to those in urban areas in other
countries. We hypothesize while there may be fewer direct sources
to residential indoor environments in this region, some do exist,
and there also exists a similar level of sources to outdoors
compared to other places, e.g., release of PCBs from electrical and
industrial equipment, of which there are documented existing
stocks (Holoubek et al., 2006), and volatilization from a range of
secondary sources. However, it is apparent that the indoor
contamination from PCBs in Czech Republic is not fully characterized,
and further work to identify possible indoor (undocumented)
sources is needed, especially in non-residential buildings.
3.1.2. PCB congener profiles
In individual houses, the congener profiles did not vary significantly
between summer and winter (Fig. S1), however the
congener profile differed between houses, ranging from those
dominated by tri-tetra PCBs (e.g., houses 11, 16 and 20) and those
dominated by hepta-hexa PCBs (e.g., houses 3, 4, and 13). The
congener profiles were more homogeneous outdoors (Fig. S1),
dominated by PCBs 28 and 153.
This range of profiles indoors is similar to the Delor technical
mixtures, the PCB technical mixtures produced in the former Czechoslovakia
(Holoubek et al., 2006). Thus, linear optimization was
used to estimate which Delor technical mixtures could be the primary
contributors to the individual indoor air samples by predicting
the composition of the indoor air samples based on the
known composition of the four Delor mixtures (Taniyasu et al.,
2003), adjusted for vapour pressure, and thereby identifying the
fraction contribution of each Delor to the indoor air samples (details
in SI).
Delors 103 and 106 best approximated the composition of
congeners in air in all in differing combinations (Fig. S2). Houses 3,
4, 13, 14, and 17 had estimated contributions of >80% from Delor
106, while Houses 5e11, and 15 had contributions of 20e40% Delor
103, as well as significant contributions from Delors 105 and 106.
Only two homes had estimated contributions from Delor 104
(~10%) (mainly due to relatively high level of PCB 52 relative to PCB
28). Outdoor air profiles were most similar to Delor 103 and 106
profiles (Fig. S2).
Delor 103 was used as a dielectric fluid in capacitors and as an
insulator, Delor 105 in transformer fluid and Delor 106 primarily as
a paint additive (Table S6). Thus, the dominance of Delors 103 and
106 suggests paints, adhesives, sealants and capacitors as the primary
sources, which is plausible for residential indoor environments.
Delor 104 had minimal production and use as a capacitor
L. Melymuk et al. / Environmental Pollution 218 (2016) 392e401394
fluid (<0.2% of total PCBs produced in former Czechoslovakia,
Holoubek et al., 2006), thus its limited contribution as a source is
expected. No relationships could be found between the Delor
compositions and building parameters such as age, dwelling type,
renovation status or location. This may be due to the limitation of
using only seven congeners to estimate the Delor compositions, or
because the factors contributing to the differences in sources between
the buildings were not captured by the building survey and
questionnaire data. This emphasizes that the sources of PCBs to
indoor environments in the Czech Republic are still not well understood,
despite this being the highest concentration SVOC of the
targeted compounds in these 20 homes, and further work is needed
to determine the plausibility of open indoor sources (paints, sealants)
in residential environments.
3.1.3. Indoor-outdoor PCB relationships
Indoor air concentrations of PCBs (both individual congeners
and sum) were significantly higher than outdoor PCB concentrations
in both seasons (Mann-Whitney U test, p < 0.05), except in
one home. The higher indoor concentrations resulted in positive
indoor to outdoor ratios (I/O) in both summer and winter (Table 3).
Winter I/O ratios were significantly higher than summer I/O ratios
(Mann-Whitney U test, p < 0.0001), reflecting temperature differences
and the influence of volatilization, and also supporting the
Table 2
Air concentrations (pg/m3
) estimated from PUF-PAS in the studied sites (n ¼ 17 in summer and n ¼ 20 in winter). DBE-DBCH is the sum of the a, b, g, and d isomers, and DDC-CO
is the sum of the syn and anti isomers.
Compound Parameter Indoor summer Indoor winter Outdoor summer Outdoor winter
S7PCBs Mean ± SD 89 ± 33 61 ± 23 38 ± 13 17 ± 5.9
Median 82 61 34 19
Range 44e160 <LOD-110 26e82 7.5e29
S9PBDEs Mean ± SD 9.1 ± 10 8.5 ± 11 1.8 ± 2.2 0.49 ± 0.21
Median 4.5 5.8 1.3 0.42
Range 1.4e41 0.79e49 0.50e9.7 <LOD-1.1
TBX Mean ± SD 0.076 ± 0.094 0.16 ± 0.11 0.023 ± 0.009 0.028 ± 0.013
Median 0.046 0.10 0.023 0.029
Range 0.020e0.41 0.051e0.36 <LOD-0.036 <LOD-0.055
PBT Mean ± SD 2.1 ± 3.1 3.5 ± 3.1 0.21 ± 0.10 0.11 ± 0.058
Median 1.4 3.0 0.20 0.11
Range 0.22e14 0.32e13 0.065e0.39 0.045e0.21
PBEB Mean ± SD 0.064 ± 0.083 0.10 ± 0.079 0.026 ± 0.070 0.013 ± 0.010
Median 0.041 0.075 0.0082 0.0092
Range 0.007e0.35 0.031e0.37 0.003e0.30 0.004e0.035
HBB Mean ± SD 1.4 ± 1.2 1.7 ± 1.7 0.13 ± 0.072 0.082 ± 0.051
Median 0.94 1.1 0.14 <LOD
Range 0.17e4.4 0.33e5.7 0.041e0.33 <LOD-0.21
TBP-DBPE Mean ± SD 1.5 ± 1.2 1.3 ± 1.2 0.58 ± 0.50 0.25 ± 0.12
Median 1.4 <LOD 0.47 <LOD
Range <LOD-5.7 <LOD-4.8 <LOD-2.2 <LOD-0.75
BTBPE Mean ± SD 0.29 ± 0.39 0.22 ± 0.37 0.062 ± 0.099 0.065 ± 0.13
Median 0.23 <LOD <LOD <LOD
Range <LOD-1.6 <LOD-1.2 <LOD-0.27 <LOD-0.49
DBE-DBCH Mean ± SD 150 ± 450 65 ± 120 96 ± 360 2.2 ± 2.6
Median 32 20 7.8 0.85
Range 6.5e1900 7.7e530 1.7e1500 0.85e12
SDDC-CO Mean ± SD 1800 ± 6900 66 ± 110 53 ± 120 30 ± 52
Median 54 <LOD 15 6.0
Range 13e29,000 <LOD-440 5.1e490 6.0e170
Fig. 1. Comparison between medians and range of indoor air concentrations of PCBs and PBDEs measured in this study with previously published concentrations from Canada
(Zhang et al., 2011), USA (Allen et al., 2007), Mexico (Bohlin et al., 2008), Sweden (Bjorklund et al., 2012; Bohlin et al., 2008), Norway (Cequier et al., 2014), Denmark (Frederiksen
et al., 2012; Vorkamp et al., 2011), France (Alliot et al., 2014) and UK (Bohlin et al., 2008; Harrad et al., 2006).
L. Melymuk et al. / Environmental Pollution 218 (2016) 392e401 395
hypothesis that ventilation differences in indoor environments
(e.g., lower air exchange in winter) influence SVOC concentrations
indoors (MacIntosh et al., 2012; Zhang et al., 2009). These ratios fall
in the lower end of the range of I/O ¼ 2e50 previously reported for
PCBs in Europe and North America (Bohlin et al., 2008; Menichini
et al., 2007; Rudel and Perovich, 2009; Schulz, 2012), suggesting
that although some building-related sources of PCBs likely exist,
the relative contribution of these compared to the sources emitting
directly to outdoor air is lower than in many other regions, as also
suggested by the concentrations and congener profiles.
3.1.4. Seasonal variability of PCB concentrations
The indoor concentrations of S7PCB and of all individual congeners
were significantly higher in summer than in winter (MannWhitney
U test, p < 0.01 for S7PCB, Fig. 2a). A similar pattern was
observed in outdoor air; summer outdoor concentrations were
consistently higher than winter for all congeners and S7PCB (MannWhitney
U test, p < 0.0001 for S7PCB, Fig. 2a). However, the summer/winter
ratios (S/W) were higher and more consistent in outdoor
air than indoors (Table 3), indicating stronger seasonal
influences in outdoor air. Temperature-dependent outdoor seasonality
is typical for PCBs (Halsall et al., 1995; Holoubek et al.,
2007; Klanova et al., 2006) due to enhanced volatilization from
both primary and secondary PCB sources at warmer temperatures;
we assume the same influence here as the primary driver of the
seasonal difference in concentrations. It is interesting that this
seasonality is mimicked in indoor air, although clearly seasonal
variations indoors are less marked than in outdoor air. This suggests
either an influence of outdoor air on indoor environments, possibly
enhanced in summer due to more ventilation, and seasonal temperature
influences acting on indoor environments. The variation in
temperature between summer and winter seasons is much smaller
indoors (differences of 5e10 C) than outdoors (20e40 C), but this
may contribute.
3.1.5. Influence of house characteristics on PCB concentrations
Despite the small range of air concentrations observed across
houses, there was a difference between houses built before and
after 1984, the end of PCB use in Czech Republic (Holoubek et al.,
2006). Concentrations of S7PCBs in indoor air in homes built
before 1984 were ~35% higher than in those built after 1984, which
may translate into differences in PCB exposure (Egsmose et al.,
2016), however on a seasonal basis this difference was only statistically
significant in summer (Fig. 3, Mann-Whitney U test,
p < 0.05). This again suggests that while the buildings themselves
may be a source of PCBs, perhaps the PCBs are in materials from the
outside of the buildings rather than directly within rooms, and thus
act as a stronger source at higher temperatures. But the consistently
higher indoor concentrations also suggest that the indoor environment
may concentrate/accumulate the PCBs. We note that indoor
concentrations of S7PCBs were consistently higher than
outdoor concentrations even for new buildings which we presume
to have no sources of PCBs. In fact, the I/O ratios had no relationship
with building construction date. The status of the house (i.e. in
original form or renovated) also did not affect PCB concentrations;
the year of the original construction was the strongest influence on
the air concentrations. The differences in outdoor air near pre- and
post-1984 buildings were not significant compared to the seasonal
differences in outdoor air (Fig. 3).
3.2. PBDEs
3.2.1. PBDE air concentrations
PBDEs were detected in all indoor samples, although the higher
molecular weight PBDEs were rarely detected. Concentrations and
ranges of S9PBDE (sum of BDE 28, 47, 66, 85, 99, 100, 153, 154 and
183) are presented in Table 2. BDEs 28, 47, 99, and 100 were most
frequently detected while BDEs 153, 154 and 183 were only
detected in a limited number of samples (Table S4). Air concentrations
of PBDEs varied by 20e100 Â among different houses; this
range was much higher than that observed for PCBs. Mean indoor
air concentration of S9PBDEs was 9.1 pg/m3
in summer (median:
4.5 pg/m3
) and 8.5 pg/m3
in winter (median: 5.8 pg/m3
). These
levels are in the lower range of what has been typically reported for
indoor residential environments in North America (Allen et al.,
2007; Zhang et al., 2011) and some European sites (Cequier et al.,
2014; Harrad et al., 2006; Vorkamp et al., 2011) but comparable
to those measured elsewhere in Europe (Alliot et al., 2014;
Bjorklund et al., 2012; Bohlin et al., 2008) as well as in an indoor
office in Czech Republic (Bohlin et al., 2014b) (Fig. 1).
Mean outdoor air concentrations of S9PBDE were 1.8 pg/m3
in
summer (median: 1.3 pg/m3
) and 0.49 pg/m3
in winter (median
0.42 pg/m3
). The concentrations were consistent with outdoor air
concentrations in Czech Republic (Bohlin et al., 2014a; Jarkovský
et al., 2015), Sweden and UK (Bohlin et al., 2008), while lower
than in Canada (Melymuk et al., 2012; Shoeib et al., 2014), USA
(Salamova and Hites, 2011) and Mexico (Bohlin et al., 2008).
Congener profiles of PBDEs were difficult to evaluate because of
the limited detection of the higher molecular weight PBDEs.
However, the data shows some variability between individual
homes, with higher contributions of penta-to hepta-BDEs in four
homes. The average profiles were similar in summer, both indoor
and outdoor, and winter indoor. Only winter outdoor air had a
different congener profile, with a statistically significant shift
Table 3
Indoor/outdoor (I/O) and summer/winter (S/W) ratios. Ratios were calculated from sites where the compound in question was >LOD in both of the samples in question.
Compound Parameter I/O in summer I/O in winter S/W indoors S/W outdoors
S7PCBs Median 2.4 3.9 1.3 2.1
Range 1.0e3.9 0.41e8.5 0.77e7.8 1.2e3.1
S9PBDEs Median 4.9 13 1.4 2.4
Range 1.7e23 0.23e98 0.32e2.7 0.35e8.8
TBX Median 2.9 4.6 0.58 0.7
Range 0.55e18 1.4e43 0.19e1.3 0.23e1.6
PBT Median 6.7 20 0.60 1.3
Range 1.1e74 2.4e81 0.076e13 0.55e4.1
PBEB Median 6.7 5.0 0.51 0.55
Range 0.078e46 1.1e24 0.099e2.3 0.21e10
HBB Median 6.7 7.3 0.71 0.91
Range 1.8e49 3.5e9.4 0.30e10 0.65e1.2
DBE-DBCH Median 3.1 9.5 1.4 5.2
Range 1.1e29 2.6e46 0.31e5.8 3.4e128
SDDC-CO Median 5.8 6.1 1.7 0.84
Range 1.9e58 0.76e48 0.072e65 0.096e53
L. Melymuk et al. / Environmental Pollution 218 (2016) 392e401396
towards heavier congeners (BDE-99 and -100; Fig. S3), attributed to
lower outdoor winter temperatures.
3.2.2. Indoor-outdoor PBDE relationships
The indoor air concentrations of S9PBDEs were significantly
higher than the corresponding outdoor air concentrations in both
summer and winter (Mann-Whitney U test, p < 0.0001), with
positive I/O ratios in both summer and winter (Table 3). As with
PCBs, I/O ratios were significantly higher in winter, suggesting a
stronger indoor-outdoor gradient in winter than in summer, presumably
caused by a combination of increased building ventilation
rates (due to open windows) and increased volatilization outdoors
in summer. Indoor and outdoor concentrations at each site were
significantly correlated (Spearman r ¼ 0.52 in summer, and r ¼ 0.47
Fig. 2. Concentration ranges for (a) PCBs, (b) PBDEs, and selected NFRs: (c) TBX, (d) PBT, (e) PBEB, (f) HBB, (g) DBE-DBCH, and (h) DDC-CO. Boxes represent medians and 25th/75th
percentiles, and whiskers are 10th and 90th percentiles. Outlying measurements are indicated by points. The x-axis labels indicate the season and location of samples (S ¼ summer,
W ¼ winter, I ¼ indoors, O ¼ outdoors).
L. Melymuk et al. / Environmental Pollution 218 (2016) 392e401 397
in winter, p < 0.05) suggesting indoor environments acting as a
source to local outdoor environments, particularly in summer.
3.2.3. Seasonal variability of PBDE concentrations
Indoors, there was no statistically significant difference between
air concentrations in winter and summer. The summer and winter
indoor concentrations were also highly correlated for individual
homes (Spearman r ¼ 0.81, p < 0.0001) suggesting concentrations
are more strongly linked to the specific individual environment
(and presumably the indoor products and materials, discussed
below), and that seasonal influences on PBDE concentrations indoors
are minimal.
In contrast to the indoor concentrations, summer outdoor concentrations
were 2.7 Â higher than winter outdoor (Mann-Whitney
U test, p < 0.0001). However, by site the outdoor summer and
winter concentrations were again highly correlated (Spearman
r ¼ 0.76, p < 0.0005), suggesting that outdoor concentrations are
driven by the magnitude of emissions from the local sources (e.g.,
higher emissions from indoor-to-outdoor transfer or from volatilization
in summer), but that the distribution of sources stays the
same. The importance of indoor-to-outdoor transfer is also suggested
by the I/O ratios; median I/O ratios in summer (4.9) were
significantly lower than in winter (13) suggesting that in summer,
indoor environments are acting as a source and air exchange decreases
indoor concentrations and elevates outdoor concentrations,
while in winter the reduced air exchange between indoor and
outdoor environments leads to a greater difference between indoor
and outdoor concentrations.
3.2.4. Influence of house characteristics on PBDE concentrations
Although the variability of indoor PBDE concentrations was high
between individual houses, there were no correlations between
PBDE concentrations and building parameters (age, type, renovation
status, window material, furniture, flooring, etc.), While the
indoor/outdoor and summer/winter ratios clearly suggest indoor
environments as a major source to outdoors, the amount of
complexity in the materials and products in indoor environments
makes this difficult to quantitatively relate to the presence or
absence of certain products. For example, concentrations of SPBDEs
in building insulation purchased between 2005 and 2010 had a
10,000-fold range, from below detection to >600 mg/g (Vojta et al.,
in prep), suggesting that simple information on the type and age of
building materials or consumer products is insufficient to link to
indoor concentrations. However, elevated concentrations were
found in homes with higher numbers of electronics, based on a
general count of the number of electronic items (e.g., TVs,
computers, stereos). Median concentrations were 3.1 pg/m3
, 6.5 pg/
m3
and 8.8 pg/m3
for rooms with one, two, and three or more
electronic items (Fig. 4), however there was only a statistically
significant difference between rooms with one and three or more
electronic items.
3.3. NFRs
3.3.1. NFR air concentrations
Of the 10 targeted NFRs, two (TBP-BAE and DBHCTD) were not
detected in any sample. Concentrations of the detected NFRs are
presented in Table 2. TBP-DBPE and BTBPE were infrequently
detected so we do not discuss their trends in further detail. Synand
anti-DDC-CO were quantified separately, but no systematic
variation was observed in the isomer ratios (~37% anti) in relation
to either season or location, so hereafter we discuss DDC-CO as the
sum of both isomers.
As with the PBDEs, large variability existed between different
houses (generally 7e150 Â differences between highest and lowest
concentration houses). DDC-CO had one very high outlier observation:
summer indoor concentrations in one home were
500 Â higher than the median indoor summer DDC-CO concentration.
Indoors in winter, median concentrations of DDC-CO were
lower, but concentrations in the same home remained anomalously
high: 30 Â higher than the winter indoor median. DBE-DBCH (sum
Fig. 3. Differences in PCB concentrations in indoor and outdoor air between buildings constructed before and after the ban on PCB use.
Fig. 4. Relationship between the indoor PBDE concentrations and number of electronics
per room.
L. Melymuk et al. / Environmental Pollution 218 (2016) 392e401398
of a, b, g, and d isomers) and DDC-CO were detected at the highest
levels; 10e1000 Â higher than the other NFRs, and neither
exhibited a significant difference in indoor concentrations between
summer and winter. Mean indoor air concentrations of DDC-CO
were 840 pg/m3
(median: 26 pg/m3
) and DBE-DBCH was 100 pg/
m3
(median: 28 pg/m3
). TBX, PBEB, PBT, and HBB were at much
lower levels, with concentrations of 0.1e10 pg/m3
. NFR levels are
comparable to those in an office in Czech Republic (Bohlin et al.,
2014b) and indoor residential levels in Norway (Cequier et al.,
2014), with the exception of DDC-CO, which was much higher in
the Czech indoor samples, and TBX, which was much higher in
Norwegian samples (Cequier et al., 2014). The levels of individual
NFRs indoors were generally much lower than the banned PCBs,
suggesting indoor “persistence” of legacy SVOCs through continuous
primary or secondary sources.
Large spatial variations in NFR concentrations also existed outdoors,
with differences of 5e900 Â between sites in summer and
4e50 Â in winter. The largest spatial differences were in the
compounds found at the highest concentrations: DBE-DBCH and
DDC-CO. Mean outdoor air concentrations of DBE-DBCH were
96 pg/m3
in summer (median: 7.8 pg/m3
) and 2.2 pg/m3
in winter
(median: 0.85 pg/m3
). Mean outdoor air concentrations of DDC-CO
were 53 pg/m3
in summer (median: 15 pg/m3
) and 30 pg/m3
in
winter (median: 6.0 pg/m3
). As indoors, the concentrations of TBX,
PBEB, PBT, and HBB were much lower, with concentrations of
0.01e0.5 pg/m3
. The concentrations were generally consistent with
other outdoor air concentrations (Bohlin et al., 2014a; Shoeib et al.,
2014; Venier and Hites, 2008).
3.3.2. Indoor-outdoor NFR relationships
The indoor levels of all NFRs were significantly higher than
outdoor levels in both summer and winter (Mann-Whitney U test,
p < 0.05). Average I/O ratios for NFRs were 2.9e6.7 in summer and
4.6e20 in winter. As with the PBDEs, the I/O ratios suggest that
outdoor concentrations strongly depend on indoor sources: I/O
ratios in summer were significantly lower than in winter, suggesting
indoor environments in summer are acting as a source to
outdoors; in winter reduced air exchange between indoor and
outdoor environments led to a greater difference between indoor
and outdoor concentrations.
3.3.3. Seasonal variability of NFR concentrations
As mentioned above, DBE-DBCH and DDC-CO had similar indoor
concentrations in winter and summer. HBB also had no statistically
significant difference. In contrast, TBX, PBEB, and PBT had higher
indoor concentrations in winter than in summer (Mann-Whitney U
test, p < 0.05), with winter concentrations two to three times
higher than summer.
Indoors, summer and winter concentrations were generally
correlated (Spearman r > 0.63, p < 0.01 for DBE-DBCH, TBX, PBEB,
and HBB) suggesting that, as with the PBDEs, concentrations are
strongly linked to the specific individual environment (and presumably
the indoor products and materials, discussed below).
Correlations for PBT and DDC-CO were weaker and not statistically
significant (p ¼ 0.1) due to a higher incidence of non-detects for
these compounds.
Outdoors, the seasonal trends contrasted with those indoors.
Summer outdoor concentrations of DBE-DBCH, PBT and HBB were
significantly higher than winter outdoor concentrations (MannWhitney
U test, p < 0.01). For TBX, PBEB, and DDC-CO, summer
outdoor concentrations were also higher than winter outdoors, but
the differences were not statistically significant. By site, outdoor
summer and winter concentrations were correlated for the high
concentration NFRs: DDC-CO (Spearman r ¼ 0.51, p < 0.05) and
DBE-DBCH (Spearman r ¼ 0.80, p < 0.0001). This suggests that
outdoor levels were driven by the magnitude of emissions from the
local sources but that the distribution of sources stayed the same, as
observed for PBDEs. The lack of correlation for the lower concentration
compounds may be due to the higher number of sites with
levels below limits of detection, particularly outdoors in winter.
3.3.4. Influence of house characteristics on NFR concentrations
As with PBDEs, although the variability in NFR concentrations
was high between houses, there were no apparent correlations
between NFR levels and the house parameters (age, type, reconstruction
status, window material, furniture, flooring, etc.). The
most notable range was in DDC-CO and DBE-DBCH; DDC-CO concentrations
in one home were ~100 Â higher than in the home with
the second highest levels, and DBE-DBCH concentrations in a
different home were 10 Â higher than the second highest levels.
DDC-CO is an additive flame retardant in many polymer products,
most notably in electrical wires and cables and plastic building
materials, and can be 10e35% of a product composition (Schlabach
et al., 2011; Xian et al., 2011). DBE-DBCH is an additive flame
retardant commonly used in polystyrene building insulation at ~1%
of composition (Covaci et al., 2011; Schlabach et al., 2011). However
information on electrical products and building materials in the
homes gave no insight into the uniquely elevated levels of these
flame retardants in these two homes. This emphasizes the
complexity and challenges in linking indoor concentrations of FRs
and other SVOCs with specific building materials and/or consumer
products. For example, a 10,000-fold range was observed in concentrations
of BFRs in televisions from the same country and
similar time period (Gallen et al., 2014). The wide range of flame
retardants in use and the lack of distinctive patterns of use
currently prevent us from expanding beyond broad generalizations
in linking indoor levels with consumer product levels.
3.4. Within-house variability
In order to assess the variability within individual homes, bedrooms
in three homes were sampled in both winter and summer, in
addition to the living room samples that form the core of this study.
In all cases the within-home variability was less than the betweenhome
variability, e.g., the typical range of concentrations observed
within a home was smaller than the typical range of concentrations
observed when comparing different homes. We evaluated the
following hypothesis: that variability between rooms is greater for
compounds with consumer products (electronics, furniture) as
sources rather than building materials or outdoor air.
While these results are based on only three homes, they suggest
that for PCBs, the concentrations were relatively uniform within a
home (Fig. S4), with generally <50% difference between S7PCB
concentrations within the same home. Conversely for PBDEs, while
concentrations in two homes had <50% difference in S9PBDEs between
rooms, in one home concentrations were more than two
times higher in the living room than in the bedroom (Fig. S4). The
differences for NFRs were even more striking. While some homes
had relatively uniform concentrations (e.g., HBB in House 9, Fig. S4),
for other compounds/homes the air concentrations of NFRs differed
more than 10-fold between rooms in the same house (e.g., PBT and
HBB in House 11, Fig. S4). This indicates that compounds with
outdoor or building material sources (in this case PCBs) may have
more uniform concentrations within the house, while compounds
with sources unique to individual rooms (flame retardants) may
have more heterogeneous distributions. Moreover, this suggests
greater spatial heterogeneity in compounds with significant current
use (NFRs), as has also been observed within an individual
room (Melymuk et al., 2016).
There was no consistent pattern related to room type or season.
L. Melymuk et al. / Environmental Pollution 218 (2016) 392e401 399
In some cases the variability in summer was greater and in other
cases winter, and in some cases the living room had higher concentrations
while in some cases bedrooms.
4. Implications for human exposure
The seasonal differences in the indoor environment are expected
to also lead to seasonal differences in human exposure. To
demonstrate these differences, we estimated seasonal chronic daily
intake (CDI) via inhalation to PCBs and FRs in winter and summer.
CDIi is the time-weighted average exposure to a chemical, for
inhalation intake only. We considered both the differences in
concentrations between summer and winter, both indoor and
outdoor, and the difference in human activity patterns between
winter and summer (Leech et al., 2002; Schweizer et al., 2007;
further described in SI), and average inhalation and body mass
parameters for adults (31e61 year old) and children (3e6 year old)
(US EPA, 2011). Calculations are described further in the SI. Exposure
via the indoor environment contributed ~96% of the total CDIi
in summer and ~98% in winter (Fig. 5), emphasizing the importance
of the indoor air for human exposure. This is related to high percentage
of time spent indoors (>90%) combined with the higher
indoors levels for all the studied compounds. Moreover, the differences
between median exposure and high-level exposure (in
this case 90th percentile) were much greater for the current use
compounds than legacy compounds; for example, there were 100x
differences in median and 90th percentile exposure for NFRs, while
PCBs had only 50e60% differences between median and 90th
percentile exposure.
5. Conclusions
Elevated indoor concentrations of PCBs in Czech homes suggest
a diverse set of sources, including indoor and outdoor, primary and
secondary. Indoor sources, while apparently minor compared to
those observed in North America and Western Europe, can
contribute to differences in air concentrations between older and
newer homes. Both indoor and outdoor air concentrations have
seasonal influences; summer concentrations are higher due to
enhanced volatilization at warmer temperatures. The sources may
be linked to the age of buildings, and with more sources in areas
where buildings were constructed before 1984, leading to higher
indoor air concentrations of PCBs. The elevated indoor air concentrations
relative to outdoor may suggest some use of PCBs indoors
in Czech Republic, which has previously not been identified, in
addition to the presence of secondary indoor sources. This requires
further investigation in both residential and non-residential
buildings.
PBDE sources are primarily in the indoor environment, leading
to higher indoor concentrations. Indoor concentrations are generally
constant throughout the year. The distributions suggest that
indoor environments can be a source to outdoors due to building
ventilation; when windows are open and there is greater air exchange
between indoor and outdoor environments, the indooroutdoor
concentration gradient is lower, presumably because indoor
environments are releasing PBDEs to the local outdoor air,
increasing outdoor PBDE air concentrations.
NFRs behave similarly to PBDEs in terms of indoor/outdoor
distributions and seasonal trends, although with even more sourcespecific
heterogeneity between homes due to their on-going use.
It is challenging to identify the specific sources of any of the
compounds from this analysis, however some insight is provided. A
key gap in our understanding of PCB sources relates to the relative
contribution of building vs. electrical/industrial sources. While the
majority of the PCB inventory in Czech Republic is in electrical/industrial
sources, there remains a lack of information about any use
indoors, although the elevated indoor concentrations in pre-1984
buildings suggest this may be worthy of further investigation.
PBDEs and NFRs are known to be related to specific materials, but
the complexity of the range of materials with FRs, and range of
types and levels of FRs used in consumer products make it very
challenging to identify specific products as drivers of indoor levels,
or make generalizations based on building parameters. Linking
indoor concentrations with FR distributions in consumer products,
compound regulations and global market distribution remains a
key gap in our understanding of FRs in indoor environments.
Acknowledgements
This research was supported by the RECETOX research infrastructure
(the Czech Ministry of Education LO1214 and
LM2015051), AMVIS/KONTAKT II project (LH12074), and “Employment
of Best Young Scientists for International Cooperation
Empowerment” (CZ.1.07/2.3.00/30.0037). We thank Alice Dvorska
for assistance with the questionnaires.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.envpol.2016.07.018.
Fig. 5. Seasonal differences in estimated inhalation exposure for children and adults. Boxes show median ± standard error, whiskers show 10th and 90th percentiles, and outliers
are shown by points. For summer n ¼ 17 and for winter n ¼ 20 (details in Table S7).
L. Melymuk et al. / Environmental Pollution 218 (2016) 392e401400
References
Allen, J.G., McClean, M.D., Stapleton, H.M., Nelson, J.W., Webster, T.F., 2007. Personal
exposure to polybrominated diphenyl ethers (PBDEs) in residential indoor air.
Environ. Sci. Technol. 41, 4574e4579.
Alliot, F., Moreau-Guigon, E., Bourges, C., Desportes, A., Teil, M.-J., Blanchard, M.,
Chevreuil, M., 2014. A multi-residue method for characterization of endocrine
disruptors in gaseous and particulate phases of ambient air. Atmos. Environ. 92,
1e8.
Bjorklund, J.A., Thuresson, K., Cousins, A.P., Sellstr€om, U., Emenius, G., de Wit, C.A.,
2012. Indoor air is a significant source of tri-decabrominated diphenyl ethers to
outdoor air via ventilation systems. Environ. Sci. Technol. 46, 5876e5884.
Bohlin, P., Audy, O., Skrdlíkova, L., Kukucka, P., Pribylova, P., Prokes, R., Vojta, S.,
Klanova, J., 2014a. Outdoor passive air monitoring of semi volatile organic
compounds (SVOCs): a critical evaluation of performance and limitations of
polyurethane foam (PUF) disks. Environ. Sci. Process Impacts 16, 433e444.
Bohlin, P., Audy, O., Skrdlíkova, L., Kukucka, P., Vojta, S., Pribylova, P., Prokes, R.,
Cupr, P., Klanova, J., 2014b. Evaluation and guidelines for using polyurethane
foam (PUF) passive air samplers in double-dome chambers to assess semivolatile
organic compounds (SVOCs) in non-industrial indoor environments.
Environ. Sci. Process. Impacts 16, 2617e2626.
Bohlin, P., Jones, K.C., Tovalin, H., Strandberg, B., 2008. Observations on persistent
organic pollutants in indoor and outdoor air using passive polyurethane foam
samplers. Atmos. Environ. 42, 7234e7241.
CECBP, 2009. Brominated and Chlorinated Organic Chemical Compounds Used as
Flame Retardants. Sacremento, California.
Cequier, E., Ionas, A.C., Covaci, A., Marce, R.M., Becher, G., Thomsen, C., 2014.
Occurrence of a broad range of legacy and emerging flame retardants in indoor
environments in Norway. Environ. Sci. Technol. 48, 6827e6835.
Covaci, A., Harrad, S.J., Abdallah, M.A.E., Ali, N., Law, R.J., Herzke, D., de Wit, C.A.,
2011. Novel brominated flame retardants: a review of their analysis, environmental
fate and behaviour. Environ. Int. 37, 532e556.
Egsmose, E.L., Br€auner, E.V., Frederiksen, M., Mørck, T.A., Siersma, V.D., Hansen, P.W.,
Nielsen, F., Grandjean, P., Knudsen, L.E., 2016. Associations between plasma
concentrations of PCB 28 and possible indoor exposure sources in Danish
school children and mothers. Environ. Int. 87, 13e19. http://dx.doi.org/10.1016/
j.envint.2015.11.005.
European Chemicals Agency, 2014. No Title [WWW Document]. URL http://echa.
europa.eu/web/guest (accessed 02.24.15.).
Frederiksen, M., Meyer, H.W., Ebbehøj, N.E., Gunnarsen, L., 2012. Polychlorinated
biphenyls (PCBs) in indoor air originating from sealants in contaminated and
uncontaminated apartments within the same housing estate. Chemosphere 89,
473e479.
Gallen, C., Banks, A., Brandsma, S., Baduel, C., Thai, P., Eaglesham, G., Heffernan, A.,
Leonards, P., Bainton, P., Mueller, J.F., 2014. Towards development of a rapid and
effective non-destructive testing strategy to identify brominated flame retardants
in the plastics of consumer products. Sci. Total Environ. 491e492,
255e265.
Halsall, C.J., Lee, R.G.M., Coleman, P.J., Burnett, V., Harding-Jones, P., Jones, K.C., 1995.
PCBs in U.K. urban air. Environ. Sci. Technol. 29, 2368e2376.
Harju, M.T., Heimstad, E.S., Herzke, D., Sandanger, T., Posner, S., Wania, F., 2009.
Current State of Knowledge and Monitoring Requirements: Emerging “new”
Brominated Flame Retardants in Flame Retarded Products and the Environment
(TA-2462/2008). Norwegian Pollution Control Authority SFT.
Harrad, S.J., Hazrati, S., Ibarra, C., 2006. Concentrations of polychlorinated biphenyls
in indoor air and polybrominated diphenyl ethers in indoor air and dust in
Birmingham, United Kingdom: implications for human exposure. Environ. Sci.
Technol. 40, 4633e4638.
Hazrati, S., Harrad, S.J., 2007. Calibration of polyurethane foam (PUF) disk passive air
samplers for quantitative measurement of polychlorinated biphenyls (PCBs)
and polybrominated diphenyl ethers (PBDEs): factors influencing sampling
rates. Chemosphere 67, 448e455.
Holoubek, I., Adamec, V., Bartos, M., Blaha, K., Blaha, L., Budnakova, M., Cerna, M.,
Cupr, P., Demnerova, K., Drapal, J., Hajslova, J., Hanzalkova, M., Holoubkova, I.,
Hrabetova, S., Jech, L., Klanova, J., Kohoutek, J., Kuzílek, V., Machalek, P.,
Mateju, V., Matousek, J., Matousek, M., Mestrík, V., Novak, J., Ocelka, T.,
Pekarek, V., Petrlík, J., Petira, O., Provazník, K., Puncochar, M., Rieder, M.,
Ruprich, J., Sanka, M., Tomaniova, M., Vacha, R., Volka, K., Zbíral, J., 2006. The
National Implementation Plan for Implementation of the Stockholm Convention
in the Czech Republic. Brno.
Holoubek, I., Klanova, J., Jarkovský, J., Kohoutek, J., 2007. Trends in background
levels of persistent organic pollutants at Kosetice observatory, Czech Republic.1)
Part I. Ambient air and wet deposition 1996-2005. J. Environ. Monit. 9,
557e563.
Jarkovský, J., Dusek, L., Klanova, J., Hulek, R., Sebkova, K., Boruvkova, J., Kalina, J.,
Gregor, J., Bednarova, Z., Novak, R., Salko, M., Hrebícek, J., Holoubek, I., 2015.
Multi-matrix Online Data Browser for Environmental Analysis and Assessment
[online] [WWW Document]. Masaryk Univ.
Klanova, J., Kohoutek, J., Hamplova, L., Urbanova, P., Holoubek, I., 2006. Passive air
sampler as a tool for long-term air pollution monitoring: Part 1. Performance
assessment for seasonal and spatial variations. Environ. Pollut. 144, 393e405.
Klosterhaus, S.L., McKee, L.J., Yee, D., Kass, J.M., Wong, A., 2014. Polychlorinated
biphenyls in the exterior caulk of San Francisco Bay area buildings, California,
USA. Environ. Int. 66, 38e43.
Kohler, M., Tremp, J., Zennegg, M., Seiler, C., Minder-Kohler, S., Beck, M.,
Lienemann, P., Wegmann, L., Schmid, P., 2005. Joint sealants: an overlooked
diffuse source of polychlorinated biphenyls in buildings. Environ. Sci. Technol.
39, 1967e1973.
Leech, J.A., Nelson, W.C., Burnett, R.T., Aaron, S., Raizenne, M.E., 2002. It’s about
time: a comparison of Canadian and American time-activity patterns. J. Expo.
Anal. Environ. Epidemiol. 12, 427e432.
Ma, Y., Venier, M., Hites, R.A., 2012. Tribromophenoxy flame retardants in the Great
Lakes atmosphere. Environ. Sci. Technol. 46, 13112e13117.
MacIntosh, D.L., Minegishi, T., Fragala, M.A., Allen, J.G., Coghlan, K.M., Stewart, J.H.,
McCarthy, J.F., 2012. Mitigation of building-related polychlorinated biphenyls in
indoor air of a school. Environ. Health 11, 24.
Melymuk, L., Bohlin-Nizzetto, P., Vojta, S., Kratka, M., Kukucka, P., Audy, O.,
Pribylova, P., Klanova, J., 2016. Distribution of legacy and emerging semivolatile
organic contaminants in five indoor matrices in a residential environment.
Chemosphere 153, 179e186.
Melymuk, L., Robson, M., Helm, P.A., Diamond, M.L., 2012. PCBs, PBDEs, and PAHs in
Toronto air: spatial and seasonal trends and implications for contaminant
transport. Sci. Total Environ. 429, 272e280.
Menichini, E., Iacovella, N., Monfredini, F., Turrio-Baldassarri, L., 2007. Relationships
between indoor and outdoor air pollution by carcinogenic PAHs and PCBs.
Atmos. Environ. 41, 9518e9529.
Prevedouros, K., Jones, K.C., Sweetman, A.J., 2004. Estimation of the production,
consumption, and atmospheric emissions of pentabrominated diphenyl ether
in Europe between 1970 and 2000. Environ. Sci. Technol. 38, 3224e3231.
Riddell, N., Arsenault, G., Lough, A., McAlees, A., McCrindle, R., Meissner, J.,
Robertson, V., 2008. The three-dimensional structural characterization of
hexachlorocyclopentenyl-dibromocyclooctane (HCDBCO). Chemosphere 73,
479e483.
Robson, M., Melymuk, L., Csiszar, S.A., Giang, A., Diamond, M.L., Helm, P.A., 2010.
Continuing sources of PCBs: the significance of building sealants. Environ. Int.
36, 506e513.
Rudel, R.A., Dodson, R.E., Perovich, L.J., Morello-Frosch, R., Camann, D.E.,
Zuniga, M.M., Yau, A.Y., Just, A.C., Brody, J.G., 2010. Semivolatile endocrinedisrupting
compounds in paired indoor and outdoor air in two northern California
communities. Environ. Sci. Technol. 44, 6583e6590.
Rudel, R.A., Perovich, L.J., 2009. Endocrine disrupting chemicals in indoor and
outdoor air. Atmos. Environ. 43, 170e181.
Salamova, A., Hites, R.A., 2011. Discontinued and alternative brominated flame retardants
in the atmosphere and precipitation from the Great Lakes basin. Environ.
Sci. Technol. 45, 8698e8706.
Schlabach, M., Remberger, M., Kaj, L., Herzke, D., Borgen, A., Harju, M.T., 2011.
Brominated Flame Retardants (BFR) in the Nordic Environment. Nordic Council
of Ministers, Copenhagen.
Schulz, T.J., 2012. Comparison of PCBs in East Chicago, Indiana and Columbus
Junction, Iowa in Indoor and Outdoor Air. University of Iowa.
Schweizer, C., Edwards, R.D., Bayer-Oglesby, L., Gauderman, W.J., Ilacqua, V.,
Jantunen, M.J., Lai, H.K., Nieuwenhuijsen, M., Künzli, N., 2007. Indoor timemicroenvironment-activity
patterns in seven regions of Europe. J. Expo. Sci.
Environ. Epidemiol. 17, 170e181.
Shoeib, M., Ahrens, L., Jantunen, L., Harner, T., 2014. Concentrations in air of organobromine,
organochlorine and organophosphate flame retardants in Toronto,
Canada. Atmos. Environ. 99, 140e147.
Taniyasu, S., Kannan, K., Holoubek, I., Ansorgova, A., Horii, Y., Hanari, N.,
Yamashita, N., Aldous, K.M., 2003. Isomer-specific analysis of chlorinated biphenyls,
naphthalenes and dibenzofurans in Delor: polychlorinated biphenyl
preparations from the former Czechoslovakia. Environ. Pollut. 126, 169e178.
US EPA, 2011. Exposure Factors Handbook: 2011 Edition. U.S. Environ. Prot. Agency
EPA/600/R-, 1e1466 doi:EPA/600/R-090/052F.
Venier, M., Hites, R.A., 2008. Flame retardants in the atmosphere near the Great
Lakes. Environ. Sci. Technol. 42, 4745e4751.
Venier, M., Ma, Y., Hites, R.A., 2012. Bromobenzene flame retardants in the Great
Lakes atmosphere. Environ. Sci. Technol. 46, 8653e8660.
Vojta, S., Melymuk, L., Becanova, J., Kukucka, P., Kohoutek, J., Klanova, J., n.d.
Halogenated flame retardants in building materials, consumer products and ewaste
on the European market. in prep.
Vorkamp, K., Thomsen, M., Frederiksen, M., Pedersen, M., Knudsen, L.E., 2011. Polybrominated
diphenyl ethers (PBDEs) in the indoor environment and associations
with prenatal exposure. Environ. Int. 37, 1e10.
WHO European Centre for Environment and Health, 2010. Guidelines for Indoor Air
Quality e Selected Pollutants. World Health Organization Regional Office for
Europe, Copenhagen.
Xian, Q., Siddique, S., Li, T., Feng, Y.-L., Takser, L., Zhu, J., 2011. Sources and environmental
behavior of dechlorane plusea review. Environ. Int. 37, 1273e1284.
Zhang, X., Diamond, M.L., Ibarra, C., Harrad, S.J., 2009. Multimedia modeling of
polybrominated diphenyl ether emissions and fate indoors. Environ. Sci.
Technol. 43, 2845e2850.
Zhang, X., Diamond, M.L., Robson, M., Harrad, S.J., 2011. Sources, emissions, and fate
of polybrominated diphenyl ethers and polychlorinated biphenyls indoors in
Toronto, Canada. Environ. Sci. Technol. 45, 3268e3274.
L. Melymuk et al. / Environmental Pollution 218 (2016) 392e401 401
Demirtepe, Hale, Lisa Melymuk, Garry Codling, Ľubica Palkovičová Murínová, Denisa
Richterová, Vladimíra Rašplová, Tomáš Trnovec, and Jana Klánová. 2021. "Targeted and
Suspect Screening of Plasticizers in House Dust to Assess Cumulative Human Exposure
Risk." Science of The Total Environment 781: 146667.
https://doi.org/10.1016/j.scitotenv.2021.146667
Targeted and suspect screening of plasticizers in house dust to assess
cumulative human exposure risk
Hale Demirtepea,b
, Lisa Melymuk a,
⁎, Garry Codling a
, Ľubica Palkovičová Murínová c
, Denisa Richterovác
,
Vladimíra Rašplová c
, Tomáš Trnovec c
, Jana Klánová a
a
RECETOX, Masaryk University, Kamenice 753/5, pavilion D29, 625 00 Brno, Czech Republic
b
İzmir Institute of Technology, Faculty of Engineering, Department of Environmental Engineering, 35430, Urla, İzmir, Turkey
c
Department of Environmental Medicine, Faculty of Public Health, Slovak Medical University, Limbová 12, 83303 Bratislava, Slovakia
H I G H L I G H T S
• DEHP & DEHT dominated phthalate &
alternative plasticizer dust levels, re-
spectively.
• Acceptable risks estimated for antiandrogenic
effects of PEs via dust inges-
tion.
• Uncertainty in toxicity data and other
exposure routes may alter the risk.
• Non-target analysis suggests >50 PEs in
dust; many are ignored in risk
assessment.
G R A P H I C A L A B S T R A C T
a b s t r a c ta r t i c l e i n f o
Article history:
Received 8 January 2021
Received in revised form 5 March 2021
Accepted 17 March 2021
Available online 23 March 2021
Editor: Jay Gan
Keywords:
Alternative plasticizer
Cumulative risk assessment
Indoor dust
Phthalate ester
Suspect screening
Indoor dust is an important exposure route to anthropogenic chemicals used in consumer products. Plasticizers
are common product additives and can easily leach out of the product and partition to dust. Investigations of plasticizers
typically focus on a subset of phthalate esters (PEs), but there are many more PEs in use, and alternative
plasticizers (APs) are seeing greater use after recognition of adverse health effects of PEs. In this study we use full
scan high resolution mass spectrometry for targeted and suspect screening of PEs and APs in house dust and to
assess the potential risk of human exposure. House dust samples from Eastern Slovakia were investigated and
concentrations of ∑12PEs and ∑5APs ranged 12–2765 μg/g and 45–13,260 μg/g, respectively. APs were at similar
levels to PEs, indicating common usage of these compounds in products in homes.
Evaluation of individual compound toxicity combined with human intake via dust ingestion suggested PEs are of
lower priority compared to semivolatile organic compounds such as polychlorinated biphenyls due to their lower
toxicity. However, cumulative risk assessment (CRA) is a more appropriate evaluation of risk, considering the
presences of many PEs in dust and their similar toxic mode of action. CRA based on median toxicity reference
values (TRVs) suggested acceptable risks for dust ingestion, however, the wide range of literature-derived
TRVs is a large uncertainty, especially for the APs. Use of newer TRVs suggest risk from dust ingestion alone,
i.e. not even considering diet, inhalation, and dermal contact. Additionally, screening of full-scan instrumental
spectra identified a further 40 suspect PE compounds, suggesting the CRA based on the 12 target PEs underestimates
the risk.
© 2021 Elsevier B.V. All rights reserved.
Science of the Total Environment 781 (2021) 146667
⁎ Corresponding author.
E-mail address: lisa.melymuk@recetox.muni.cz (L. Melymuk).
https://doi.org/10.1016/j.scitotenv.2021.146667
0048-9697/© 2021 Elsevier B.V. All rights reserved.
Contents lists available at ScienceDirect
Science of the Total Environment
journal homepage: www.elsevier.com/locate/scitotenv
1. Introduction
Due to human and industrial needs, synthetic materials have become
ubiquitous in our lives, and most notably plastic products. Global
production of plastics reached almost 370 Mtons in 2019 with extensive
usage in consumer products (Plastics Europe, 2020). Plastic materials
contain a wide range of additives to impart the properties needed in
the final products, and this frequently includes plasticizers, forming up
to 55% by weight of the material (Narvaez Rincon and Suarez Palacios,
2015). Phthalate esters (PEs) are the most commonly used plasticizers,
used to improve flexibility and durability in many consumer products
(Kutz, 2017; Stanley et al., 2003). PEs are found in polyvinyl chloride
(PVC) products, toys, flooring, carpets, wall coverings, food packaging,
medical products, glue and paint (ATSDR, 1997, 2001, 2019; Kutz,
2017; Zhang et al., 2020), as well as cosmetics and insecticides, mainly
containing low molecular weight (LMW) PEs such as diethyl phthalate
(DEP) and dimethyl phthalate (DMP) (ATSDR, 1995). Many PEs are considered
high production volume (HPV) compounds: yearly production
and import of bis(2-ethylhexyl)phthalate (DEHP) in the European economic
area (EEA) is between 104
and 105
t, while that of di-n-butyl
phthalate (DnBP), DEP and DMP is between 103
and 104
t (ECHA,
2018) (Table S1).
Since plasticizers are applied as additives to a product, they can easily
leach out of the product and be emitted to the environment where
the products are used or when they are disposed as wastes, hence PEs
are broadly detected in the global environment (Net et al., 2015). Due
to widespread usage in indoor-related products and physicochemical
properties allowing for significant partitioning to particulates, PEs can
be found ubiquitously in indoor dust, which is an important human exposure
route to PEs, especially to high molecular weight compounds
considering their higher abundance in dust than in indoor air
(Giovanoulis et al., 2018) and more specifically for infants and toddlers
(Wormuth et al., 2006).
Exposure to PEs has been associated with negative effects on reproductive
and developmental systems, endocrine disruption (Hauser and
Calafat, 2005; Meeker et al., 2009; Swan et al., 2005), allergies and
asthma in children (Ait Bamai et al., 2016; Bekö et al., 2015; Bertelsen
et al., 2013), obesity and cardiometabolic risk factors in children and adolescents
(Amin et al., 2018). Due to the evidence of adverse health effects,
especially in children, the use of DEHP, DnBP, butylbenzyl
phthalate (BBP), di-isobutyl phthalate (DiBP), di-isononyl phthalate
(DINP), di-isodecyl phthalate (DIDP) and di-n-octyl phthalate (DnOP)
has been restricted by the European Commission to not exceed 0.1%
by weight of plasticized material in toys and childcare products
(REACH, 2018). The same restriction was applied in the USA, but also included
di-n-pentyl phthalate (DPP), di-n-hexyl phthalate (DHP) and dicyclohexyl
phthalate (DCHP), but not DIDP and DnOP (CPSC, 2017).
The restrictions on the use of legacy PEs has led to the introduction of
alternatives with lower toxic potential. Among these alternative
plasticizers (APs), bis(2-ethylhexyl)terephthalate (DEHT) and bis(2propylheptyl)phthalate
(DPHP) have yearly production and import volume
up to 106
t to the EEA, while acetyl tributyl citrate (ATBC), bis(2ethylhexyl)adipate
(DEHA) and 1,2-cyclohexane dicarboxylic acid diisononyl
ester (DINCH) are individually produced and imported in the
range from 104
to 105
t (ECHA, 2018). The larger production volumes
for APs compared with legacy PEs suggest similar abundance in indoor
environments to the legacy PEs, especially in indoor dust due to their
partitioning coefficients similar to PEs (Bui et al., 2016). Therefore,
their exposure and potential toxicity should be evaluated carefully.
While the APs considered for target analysis in this study have been examined
for developmental and reproductive toxicity, there is limited information
compared to the PEs. ECHA reports no observed adverse
effect levels (NOAELs) based on developmental and reproductive toxicity
for DPHP and ATBC and derived no-effect levels (DNEL) based on repeated
dose toxicity for all APs, except for DINCH which is based on
carcinogenicity (ECHA, 2018). However, despite the perception of
lower hazard for the APs, ATBC was found to have endocrine disrupting
potential and neurotoxicity (Bui et al., 2016).
Evaluation of the risk associated with indoor exposure is crucial, especially
considering the range of semi-volatile organic compounds
(SVOCs) typically present indoors, with wide variations in concentrations
and toxicity. We have previously proposed a framework combining
human intake and toxicity reference values (TRVs) to prioritize
SVOCs for risk assessment (Demirtepe et al., 2019). The framework,
relying on median TRVs reported in literature and by regulatory agencies,
enabled a comparative evaluation of risks of individual SVOCs.
However, this framework ignores the possibility of mixture effects
based on chemicals having similar toxicological endpoints. PEs are
known to have common developmental and reproductive health effects
(Howdeshell et al., 2008), particularly anti-androgenic effects (Pelletier
et al., 2018; Radke et al., 2019). Anti-androgenic effects include decreased
fetal testosterone, reduced anogenital distance, reduced reproductive
organ weights, retained nipples, decreased sperm production
and Leydig cell adenomas (Gray et al., 2000; Kortenkamp and Faust,
2010). Multiple PEs are typically present indoors; therefore cumulative
risk assessment (CRA) can more accurately estimate the human exposure
risk (Kortenkamp and Faust, 2010; Pelletier et al., 2018).
This study employs ultra-high resolution mass spectrometry
coupled with gas chromatography operating in full scan with the aim
of (i) identifying the concentrations of target PEs and APs in Slovakian
indoor dust, considered representative of European SVOC levels
(Demirtepe et al., 2019), and investigating their associations with
home characteristics, (ii) assessing human exposure to PEs and APs
via dust ingestion, and cumulative anti-androgenic risk of PEs, and
(iii) using non-target screening for the evaluation of unquantified/unknown
compounds with PE structure.
2. Methods
2.1. Sample collection and preparation
Indoor dust samples from 60 homes in Eastern Slovakia were collected
in March–April 2015. Details on sampling location are presented in a previous
study (Demirtepe et al., 2019). The dust samples were collected using a
household vacuum cleaner with polyester sock inserted in the front of the
vacuum tube and vacuuming 1 to 3 m2
floor surface. The sock was removed
from vacuum cleaner, packed in aluminum foil, put into a zip-lock bag,
stored in freezer at −18 °C for transport to the laboratory.
The analytical procedure used was published previously (Demirtepe
et al., 2019; Jílková et al., 2018; Venier et al., 2016; Vojta et al., 2017;
Vykoukalová et al., 2017), and is briefly described here. Dusts were
sieved with a 500 μm sieve to remove course particles, and ~ 100 mg
were taken for extraction. Dust samples were sonicated three times in
1:1 v/v hexane:acetone, and extracts were split 70:30. The 30% aliquot
was cleaned and fractionated using activated silica column eluted
with DCM (1st fraction), followed by 7:3 v/v acetone:DCM (2nd
fraction). The fractionated extracts were exchanged to nonane and
stored at −18 °C. Concentrations of polychlorinated biphenyls (PCBs),
organochlorine pesticides (OCPs), polycyclic aromatic hydrocarbons
(PAHs), organophosphate esters (OPEs) and halogenated flame retardants
(FRs) in these dust samples have been previously reported
(Demirtepe et al., 2019). In this study, 55 of the second fraction extracts
were available for analysis of plasticizers, i.e. 13 PEs and five APs, listed
in Table S1 with CAS numbers and acronyms.
2.2. Instrumental analysis
The extracts were fortified with benzo(e)pyrene-d12 (Wellington
Laboratories Inc., Canada) and PEs and APs were analyzed with a Trace
1300 series gas chromatograph (Thermo Scientific) equipped with 30
m × 0.25 mm × 0.25 μm Rxi5-SIL-MS column coupled to a Q-Exactive
GC Orbitrap (Thermo Scientific). The operation was full-scan
H. Demirtepe, L. Melymuk, G. Codling et al. Science of the Total Environment 781 (2021) 146667
2
(70–1000 amu) in EI mode with a mass resolution of 60,000. Injection
was splitless 1 μL at 280 °C. The GC temperature program was 80 °C
(3 min hold), then 7 °C min−1
to 320 °C (23 min hold). Samples were
quantified based on a 15-point internal calibration curve, with R2
greater than 0.99 for all compounds. Mass accuracy was <2 ppm and
the instrument was checked for drift in mass every 12 h during instrumental
run. The calibration range covered 4 orders of magnitude in
order to include the concentration range expected for some PEs.
2.3. QA/QC
These extracts were not initially extracted for PEs; therefore, the potential
for elevated blank contamination was high, especially with the
ubiquitous contamination in a typical laboratory. However, field blanks
(4 vacuum socks) were treated as per the samples. Method detection
limits (MDLs) were calculated as the average of the field blank samples
plus three times the standard deviation of the blanks (ng/sample).
MDLs were converted to μg/g dust by dividing by a nominal sample
amount of 0.1 g. The instrument detection limits (IDLs) were derived
from the lowest concentration of the analyte detected in the calibration
curve. The IDL was used as the MDL for compounds that were not detected
in the blanks (Table S2). For compounds having >50% detection
frequency, average blank contamination corresponded to 1.2 ± 0.5%
of mean concentration of each analyte. The average of the blanks was
subtracted from sample values that were > MDL, and values <MDL
were recorded as such. Additionally, indicative method quantification
limits (MQLs) were calculated as the average of the field blank samples
plus ten times the standard deviation of the blanks. Blank levels, MDLs
and MQLs are reported in Table S2.
The 1st fraction extracts were also checked for PEs & APs, but the target
peak areas were lower than 1% of that in 2nd fraction, hence they
were not included in further analysis. DiBP contamination in the calibration
curve was high and calibration for this compound cannot be accurately
performed; thus, it is not discussed further in this study.
We analyzed NIST standard reference material (SRM) 2585 for PEs &
APs and compared the measured concentrations with previously reported
values in seven studies since SRM 2585 is not certified for plasticizers
(Table S3). PEs were within the range of SRM values previously
reported, although we measured consistently lower DEHP concentration
than previous studies except for one (Christia et al., 2019). For
APs, only two studies have reported SRM concentrations of DEHT, and
one study ATBC and DEHA. Our values for DEHA and DEHT were consistent
with those of Kademoglou et al. (2018) and Christia et al. (2019),
but for ATBC were significantly greater than those of Christia et al.
(Table S3). The SRM samples were spiked with recovery standards before
extraction and the average recoveries were 95.2%, 80.2% and
81.7% (n = 3) for DMP-d4, DnBP-d4, and DEHP-d4, respectively.
2.4. Data analysis
Statistical analyses were conducted using IBM SPSS 25. For statistical
analysis, values below detection were substituted by √2/2∙MDL.
Shapiro-Wilk normality test showed non-normal distribution for indoor
dust data (p < 0.001) hence non-parametric tests, i.e. Spearman correlation
(rs) and Mann Whitney U test, were applied. Analysis of variance
(ANOVA) was conducted after log-transforming the data. p < 0.05 was
considered the threshold for statistical significance.
2.4.1. Cumulative risk assessment
Human intake of PEs and APs via dust ingestion was calculated using
the daily exposure dose (DED) formula and exposure parameters given
in the SI based on the US EPA Exposure Factors Handbook (U.S. EPA,
2011). Intakes were calculated for median and high (95th percentile)
intake scenarios for a child aged between six and eleven, as dust was
collected from children's bedrooms, and for an adult male. For compounds
whose median is <MDL, we used substituted values as
defined above for median scenario. Toxicity reference values (TRVs),
defined as the maximum dose of a compound that a person can be exposed
daily without a health risk, were collected as non-carcinogenic
oral doses from literature and regulatory sources (Table S4). Then, for
cumulative risk assessment of PEs, relative potency factors (RPFs) and
reference doses (RfD) were compiled from literature for individual PEs
(Table 1). Since DEHP is the common compound for studies presenting
RPFs, it was selected as the index compound and median of its TRVs was
used as RfDDEHP in cumulative hazard quotient (HQ) calculation, given
below (Pelletier et al., 2018):
Cumulative HQ ¼
Pn
i¼1 DEDi  RPFi
RfDDEHP
ð1Þ
where DED and RfD were given in μg/kg/d, and RPF was unitless.
2.4.2. Screening for suspect PEs
Full scan MS spectra of indoor dust samples, standard mixtures and
blanks were transferred to MS-DIAL software for deconvolution, alignment
and peak identification (Tsugawa et al., 2015). The details of
MS-DIAL settings and parameters are provided in the SI.
Data from MS-DIAL were processed using Python Pandas software. A
total of 5412 features were found in samples, standards, and blanks,
though some features are likely to be the same compound separated
by retention time drift. To remove contaminant features found in the
laboratory blanks, the MDL (mean plus 3 times the standard deviation)
of nonane instrumental blanks (n = 5) was applied to all samples. Features
below the MDL were excluded from further study. Each feature
detected by MS-DIAL included a comprehensive list of ions; this was reduced
to include just those within 10% of the most abundant ion and
then the top 10 of these were retained for preliminary investigation.
Where two ions were detected with less than 0.0005 amu mass difference,
the most abundant was retained in any feature.
The primary structure of PEs is the 1,2-benzenedicarboxylic acid
with multiple possible alkyl side chains (Fig. S1A). Therefore, the
mass 149.0226 (+/− 0.0003 amu), due to protonated phthalic anhydride
structure (Fig. S1B), is a characteristic feature and often the
most abundant, with the exception of dimethyl phthalates and the
Table 1
Cumulative hazard quotients calculated with various RPFs. RPFs in bold indicates the reference compound used by the corresponding study.
References
RPF
ΣHQ
Child Adult
median high median highDEHP BBP DEP DnBP DPP
Benson (2009) 1.00 0.21 0.64 1.26 1.1 × 10−2
0.64 × 10−1
1.73 × 10−3
1.00 × 10−2
German Federal Environment Agency (2011) 1.00 1.00 0 1.00 3.00 1.1 × 10−2
0.70 × 10−1
1.79 × 10−3
1.10 × 10−2
Hannas et al. (2011) 0.11 1.00 1.2 × 10−3
0.62 × 10−2
1.82 × 10−4
0.97 × 10−3
Fournier et al. (2016)a
1.00 0.088 21 1.2 × 10−2
0.75 × 10−1
1.81 × 10−3
1.17 × 10−2
Howdeshell et al. (2008) 1.00 0.83 0.96 2.93 1.1 × 10−2
0.69 × 10−1
1.78 × 10−3
1.08 × 10−2
Gray et al. (2000) 1.00 1.00 0 0.5 1.1 × 10−2
0.64 × 10−1
1.73 × 10−3
1.00 × 10−2
Varshavsky et al. (2016) 0.61 0.26 0.024 1.00 0.73 × 10−2
0.47 × 10−1
1.14 × 10−3
0.73 × 10−2
a
Calculated with DEHP as reference since cypermethrin was used as reference in Fournier et al. (2016).
H. Demirtepe, L. Melymuk, G. Codling et al. Science of the Total Environment 781 (2021) 146667
3
diphenyl-isophthalates. For dimethyl phthalates, two CH3 groups are
on the alkyl side chain, and the diphenyl isophthalates have two benzene
rings, giving characteristic fragments of 163.0382 and
225.0552, respectively. After filtering the ion list for these masses
and eliminating the features corresponding to the targeted PEs, the
remaining features were considered “suspect PEs”.
3. Results and discussion
3.1. Targeted PEs and APs in indoor dust
Nine out of 12 PEs and three out of five APs had >50% detection frequency
in dust samples from Slovakian homes. The median concentration
of ∑12PEs was 376 μg/g, with the greatest contribution from
DEHP (80%), followed by DnBP and DnOP. ∑5APs had a median concentration
of 200 μg/g. DEHT was the dominant AP, with a mean contribution
of 37%. The median concentrations of DEHT, ATBC and DEHA
were higher than that of six frequently detected PEs, indicating abundant
use of APs in homes. Descriptive statistics for PEs and APs are presented
in Table 2.
When compared to the SVOCs identified in our previous study
(Demirtepe et al., 2019), the median concentrations of ∑12PEs (376
μg/g) and ∑5APs (200 μg/g) were an order of magnitude greater than
that of ∑14OPEs (12.4 μg/g) and two orders of magnitude greater
than that of ∑27PAHs (2.0 μg/g). This was consistent with previous
publications where PEs were identified at one to two orders of magnitude
greater concentration than OPEs (Bergh et al., 2011; He et al.,
2016; Luongo and Östman, 2016; Yang et al., 2020, 2019).
The median PE concentrations in indoor dust in this study were
comparable to the concentrations reported in existing publications on
PEs in house dust (Table S5). In most studies DEHP was the most frequently
detected compound (>70% of samples) and found at the
highest concentration. DEHP concentrations from literature are compared
in Fig. 1 grouped by continent and in Fig. S2 by country. In
Europe, the median concentration across all studies investigated here
was 270 μg/g, with the lowest median observed in Belgium (62 μg/g)
and the greatest in Bulgaria (1050 μg/g). The European median for
DEHP is similar to the median in this study (319 μg/g), and within the
range for other regions (137 μg/g for North America, nNAmer = 6 and
435.5 μg/g for Asia, nAsia = 14; Table S5). Thailand, with the second
highest median concentration of 1739.3 μg/g (Promtes et al., 2019),
has to date no DEHP restrictions (Sedtasiriphokin et al., 2017) though
a new toy safety proposal on DEHP has been proposed in 2020.
An important observation was that DEHP concentrations reported in
indoor dust show a significant decrease over time (n = 33, rs = −0.44,
p < 0.05) (Fig. 1). The regulatory agencies in the EU and US prohibited
DEHP use in children's products in 2005 and listed it in the candidate
list for substance of very high concern in 2008 (CPSC, 2017; ECHA,
2020). More recent DEHP concentrations reported in dust from
Europe and North America are generally lower than those in Asia, and
the decrease in reported dust concentrations of DEHP is much stronger
when considering only European and North American records (n = 17,
rs = −0.88, p < 0.001; Fig. S3). This suggests that stocks and indoor uses
of DEHP are being rapidly removed from indoor environments and that
restrictions in use can be effective in reducing DEHP exposure. Nevertheless,
a more in-depth literature survey would be required to make
a definite conclusion on trends observed.
So far, few studies have identified APs in house dust, although we
found AP concentrations to be comparable to PEs (Table S5). Median indoor
DEHT, ATBC and DEHA concentrations from Slovakia were greater
than those from Norway (Kademoglou et al., 2018), Ireland, Belgium
and Netherlands, except for DEHT from Ireland (Christia et al., 2019).
Median DEHT concentrations from Slovakia were also higher than that
from Germany (Nagorka et al., 2011), and within the range of the studies
reporting DEHT in USA (Hammel et al., 2019; Shin et al., 2020). DEHA
and ATBC concentrations from the US (Subedi et al., 2017) were higher
than those from Slovakia, while DEHA levels from Japan (Kishi et al.,
2018) and Qatar (Nayef et al., 2019) were similar to that from
Slovakia. Finally, median ATBC concentration from Slovakia was higher
than that from China (Tang et al., 2020) and the US (Shin et al., 2020).
3.2. Possible sources of PEs and APs
Spearman's Rank correlations between PEs and APs suggested similar
indoor sources (Table S6). For example, strong correlations were observed
between DEHP, DnOP and DEHA, all of which are known to be
used in wires and cables (ATSDR, 1997, 2019; Lowell Center for
Sustainable Production, 2011) and DEHP and DnBP, which are commonly
used in furniture (Lowell Center for Sustainable Production,
2011). We also explored the correlations between PEs, APs and OPEs
measured in our previous study (Demirtepe et al., 2019). We found
weak significant correlations for some compounds, such as tri-n-butyl
phosphate correlated with DEP (rs = 0.28, p < 0.05) and DMP (rs =
0.40, p < 0.01) which have common uses as plasticizers in cellulose lacquers,
plastics, and vinyl resins (Lyche, 2017; PubChem, 2020).
Table 2
Descriptive statistics for PEs and APs for Slovakian homes (n = 55, μg/g).
Median Geomean Mean SD Min Max % Frequency
Phthalate esters (PEs)
DEEP <MDL 0.012 0.027 0.064 <MDL 0.43 31
DEHP 319 245 470 524 <MDL 2615 98
DMEP <MDL 0.057 0.072 0.066 <MDL 0.36 18
BBP 2.94 2.81 9.14 16.7 <MDL 87.9 93
DCHP 0.98 1.01 2.34 3.74 0.038 19.4 100
DEP 1.42 1.53 7.36 27.3 <MDL 201 80
DHP 0.19 0.23 0.85 2.20 <MDL 12.3 95
DMP 0.071 0.11 0.37 0.82 <MDL 5.44 96
DnBP 24.3 24.6 78.7 196 <MDL 1160 84
DnOP 13.2 12.6 20.2 20.3 <MDL 119 95
DNP 0.94 0.79 1.52 1.82 <MDL 8.16 96
DPP <MDL <MDL 0.039 0.11 <MDL 0.64 36
Total ∑12PEs 376 309 590 655 11.8 2765
Alternative plasticizers (APs)
ATBC 13.4 14.1 41.5 65.8 <MDL 307 96
DEHA 7.50 6.91 21.2 48.6 <MDL 274 93
DEHT 71.3 66.0 186 629 <MDL 4713 95
DPHP <MDL 33.2 49.5 64.8 <MDL 382 38
DINCH <MDL 45.7 300 1770 <MDL 13,200 33
Total ∑5APs 200 227 597 1870 45.4 13,260
Fig. 1. Median DEHP concentrations in indoor dust samples from various continents with
respect to sampling year. The studies reporting these levels are listed in Table S5.
H. Demirtepe, L. Melymuk, G. Codling et al. Science of the Total Environment 781 (2021) 146667
4
A questionnaire provided ancillary information on characteristics of
the Slovakian homes. We hypothesized that presence of PVC flooring
would result in higher PE concentrations in dust, as has been previously
noted (Bornehag et al., 2005). However, no significant difference in individual
PEs and total PE concentrations was found between homes
with (n = 17) and without (n = 33) PVC flooring. A possible reason
can be that the contribution of PVC flooring to PEs found in the dust
may be limited compared to other PE sources in homes, such as furniture,
carpets, wires and cables, packaging materials, etc. (ATSDR, 2019,
2001; Lowell Center for Sustainable Production, 2011). On the other
hand, we found that DnBP, DnOP, DEHP, ∑PEs, ATBC and DEHT had significantly
greater concentrations in homes with carpeting (n = 8) than
homes without carpeting (n = 42) (Mann-Whitney U test, p < 0.05);
these compounds are used in carpets (ATSDR, 2001, 1997; US EPA,
2019), which may be acting as a source to indoor dust. Lastly, we explored
the correlations with building age. Only DMP (rs = 0.44, p <
0.05) and DEP (rs = 0.35, p < 0.05) correlated with building age, having
greater concentrations in older homes. Additionally, DMP concentrations
were significantly higher in homes more than 45 years old than
homes less than 45 years old (one-way ANOVA, F = 2.96, p < 0.05),
which we attribute to past use of DMP as a solvent in insecticides
(Lyche, 2017).
3.3. Exposure and toxicity assessment of PEs and APs
PEs are known or suspected to have a range of health effects including
altered immune system, developmental and reproductive effects,
endocrine disruption, liver and kidney toxicity (ATSDR, 2019; Mitro
et al., 2016), among which developmental and reproductive effects
have been the most studied on laboratory animals and also in humans
(Johnson et al., 2012; Kay et al., 2014; Wang et al., 2019). On the other
hand, the toxicological information for APs is relatively scarce. Studies
have examined their developmental and reproductive toxicity potentials
and carcinogenicity, and APs have DNELs (ECHA, 2018). We first estimated
exposure to PEs and APs and compared them with respect to
their individual toxicity; we then evaluated the mixture effects of PEs
via a cumulative risk assessment (CRA).
The available toxicity reference values (TRVs) for PEs and APs are
given in Table S4. Comparison of median TRVs for PEs revealed the
highest toxicity for DEHP, followed by DnOP, DnBP, DCHP, BBP, DEP,
DMP, while APs have higher TRVs than PEs except for DEP and DMP
(Table S4). DEHP has a large range of TRVs reported, which might be
due to different endpoints used, e.g. ATSDR provided an MRL of 0.1
μg/kg/d for a developmental endpoint and IRIS provided RfD of 20
μg/kg/d for increase in liver weight. Additionally, structure-activity relationship
(SAR) studies showed that straight-chain PEs with four to
seven carbons have higher potency for developmental and reproductive
toxicity, and branching of the side chain and unsaturation of the side
chain increases the potency, while cyclic side chain does not (Li et al.,
2019). Accordingly, DHP, DPP, DnBP should have higher toxic potency
among the straight chain PEs, while DiBP and DEHP should have higher
potency among branched chain PEs (Li et al., 2019). BBP, having a mixed
carbon chain in the structure, is equipotent to DnBP and DEHP
(Howdeshell et al., 2008). On the other hand, DnOP does not have developmental
and reproductive toxicity at the highest doses tested (Li
et al., 2019), although has the second lowest median TRV. Hence,
usage of multiple endpoints in different toxicity assessments creates uncertainty
in evaluating human risk.
In our previous study, we used TRVs derived from literature and regulatory
sources to evaluate relative toxicities of SVOCs and merged this
information with indoor exposure estimates for prioritization of compound
risks (Demirtepe et al., 2019). Using this previously developed
framework, we added PEs and APs into the evaluation from Demirtepe
et al. (2019) (Fig. 2) to compare exposure and toxicity of PEs and APs
to other SVOCs via dust ingestion. The estimated intakes for median
and high intake scenarios and for a child and an adult male are provided
in Table S7 for individual compounds. The human intake via dust ingestion
for a child ranged between <0.01 ng/kg/d for DPP and 211 ng/kg/d
for DEHP according to the median intake scenario. DEHP had a two
order of magnitude higher intake than tris(2-butoxyethyl) phosphate,
which had the highest intake via dust ingestion (1.62 ng/kg/d) among
previously reported SVOCs from these same samples (which included
OPEs, PCBs, PAHs, OCPs and halogenated FRs) (Demirtepe et al., 2019).
PEs and APs have higher TRVs (indicating lower toxicity) than PCBs,
Fig. 2. Human intake via dust ingestion vs toxicity reference values of indoor SVOCs, PEs and APs. Vertical lines represent the range of TRVs reported in the literature, horizontal lines
represent the high intake scenario, and points represent median exposure scenario and median TRVs. The box below the graph shows compounds with no available TRVs. Modified
from Demirtepe et al. (2019).
H. Demirtepe, L. Melymuk, G. Codling et al. Science of the Total Environment 781 (2021) 146667
5
OCPs and PBDEs, but similar TRVs to many PAHs, OPEs and NFRs (Fig. 2).
Hence, although PEs and APs had the highest human intake among
SVOCs, this assessment did not place them among the high priority
compounds for indoor risk assessment, although, as with the OPEs, the
uncertainty regarding the TRVs combined with their high exposures
suggests potential for concern.
Moreover, co-occurrence of PEs in the indoor environment leads to
human exposure to multiple PEs with possible common health effects.
Several PEs affect the reproductive development of fetal male rat via a
common mode of toxicity (Howdeshell et al., 2008), and studies on
the mixture effects of PEs showed that the effect was best predicted
by dose addition (Hannas et al., 2011; Howdeshell et al., 2008). Hence,
CRA was employed to understand the risks associated with the total
PE contamination of indoor dust. We focused on additive antiandrogenic
effects, and calculated the cumulative hazard quotients
(ΣHQ) given in Eq. (1) by using the Tier 2 approach presented by
Pelletier et al. (2018).
ΣHQs were calculated for a child and an adult male for median and
high intake scenarios using the RPFs available for five PEs (Table 1).
ΣHQs were < 1 for each set of RPFs for a child and an adult male for
both scenarios. ΣHQ values calculated by using RPFs from different studies
given in Table 1 were on the same order of magnitude, except for
Hannas et al. (2011) in which only two RPFs were available for PEs. Percent
contribution of PEs to ΣHQs was highest for DEHP, i.e. more than
89% for all set of RPFs, followed by DnBP (Fig. 3). This was expected
since DEHP is the dominant PE and has higher RPF than other PEs. Overall,
the CRA identified no unacceptable risks for children and male adults
regarding anti-androgenic effects due to indoor exposure via dust
ingestion.
Some uncertainties and limitations are associated with use of the
CRA. First, the CRA results represented only indoor dust ingestion exposure
pathway, while dermal contact with dust and inhalation of indoor
air might also contribute to total human exposure to PEs. Additionally,
other SVOCs having anti-androgenic effects, such as p,p’-DDE, BDE-99
(Kortenkamp and Faust, 2010), and benzo(a)pyrene (Fournier et al.,
2016) were identified in the same dust samples but were not included
in this CRA since the first two have no RPFs available. Yet these compounds
should also contribute to the ΣHQs. Another important contribution
might come from DiBP, which has RPFs ranging 0.15–1.00
(Benson, 2009; German Federal Environment Agency, 2011; Hannas
et al., 2011; Howdeshell et al., 2008). However, DiBP was not quantified
in this study due to high background contamination.
Second, the RPFs derived from different studies may be incomparable
since the experimental parameters and uncertainty factors used to
derive reference and benchmark doses might be different (Benson,
2009; Fournier et al., 2016). Subsequently, the differences in specific
endpoints such as the decrease in fetal testosterone vs. reduced reproductive
organ weights might create uncertainty in comparability of
data. However, the variation in RPFs did not lead to large variations
among calculated HQs since daily exposure dose to DEHP was one to
four orders of magnitude higher than other PEs, i.e. HQs were dominated
by DEHP intake. It is also important to note that the median of
available TRVs for DEHP was used as RfDDEHP in the calculation of HQs
(Eq. (1)). If we had used the minimum of reported TRVs, which is 0.1
μg/kg/d, and is also the most recent reported value (ATSDR, 2019), we
would have obtained HQs > 1. Therefore, uncertainty in the TRVs is an
important factor impacting HQ calculation and the CRA. Data available
for many compounds were inconsistent, e.g. DEP was found 21 times
more potent than DEHP according to Fournier et al. (2016) but zero potent
according to most of the studies in Table 1. Positive correlations
were found between DEP exposure and effects in human studies, but
no associations were found in most of the animal studies, which might
be explained by co-exposure to other PEs in human studies (NRC, 2008).
This highlights the importance of placing indoor exposure measurements
into context by acknowledging the differing toxicities and toxic
endpoints of individual compounds, at minimum through a combined
assessment of exposures and TRVs (e.g., Fig. 2), but ideally incorporating
the importance of mixture effects through use of HQs or a similar technique.
Yet in all aspects, the biggest limitation at present remains the
quality and comparability of TRVs, which are lacking for many compounds,
or, when present, often have order-of-magnitude ranges, leading
to similar uncertainty in the risk evaluations for which they are used.
3.4. Screening for other suspect PEs
Approximately 40 PEs are considered important for monitoring due
to their use and HPV (Staples, 2003), however fewer are regularly monitored.
This may be due to the limitations on standard mixtures, instrumental
detection and sample contamination such as observed for DiBP
in this study. All sample data in this study was collected on a GCOrbitrap
operating in full scan, allowing for screening of other PEs not
included in the target analysis. MS-DIAL software was used to identify
peaks, deconvolute and group the samples providing the opportunity
to assess the number of peaks which are suspect PEs according to common
PE feature masses. In total 5411 features were identified by MSDIAL.
After removing features reflecting background contamination,
and selecting only the 10 most abundant ions, the features were filtered
for target feature masses (149.0226, 163.0382 and 225.0552, +/−
0.0003 amu), which resulted in 63 features. Of the three masses, only
the 149.0226 mass had any detectable peaks. Removing those features
with low frequency of detection, i.e. < 10% of samples, 48 suspect compounds
remained. Based on ion masses and retention times, seven of
Fig. 3. Percent contribution of PEs to ΣHQs calculated by using RPFs from studies given in the y-axis. * GFEA refers to German Federal Environment Agency.
H. Demirtepe, L. Melymuk, G. Codling et al. Science of the Total Environment 781 (2021) 146667
6
these were found to correspond to the target PEs already included in the
study, leaving 41 suspect PEs with the protonated phthalic anhydride
structure.
A list of 65 candidate PEs (Table S8) was generated from literature
including the compounds detected by target analysis (n = 13 target
compounds; Table S1). Many more PEs have been published but were
excluded if no published retention time index (RTI) or retention time
(RT) were available and if no GC spectra had been published. From the
list of PEs, 54 had published n-alkane scale RTIs and for the majority
also multiple published RTs were available. Additionally, the methods
using a similar GC column of the current study (30 m with 5% diphenyl)
were given priority. For the remaining compounds, a linear regression
analysis between the RT of corresponding compounds in our target
method and the method in the literature was used to estimate an RTI
value and RT. For most of the remaining compounds more than two
published RTs were used in estimation of RTI.
Using the estimated RT, 12 additional PEs were tentatively identified
(1,2-benzenedicarboxylic acid monopentyl ester (MPeP), 2-{[(4methylpentyl)oxy]carbonyl}benzoate,
bis(2-ethylhexyl) isophthalate,
bis(4-methyl-2-pentyl) phthalate isomer (BMPP), butyl cyclohexyl
phthalate, diallyl phthalate (DAP), di-n-heptyl phthalate, di-isoheptyl
phthalate, di-isononyl phthalate (DINP), hexyl decyl phthalate,
monomethyl phthalate (MMP), octyl hydrogen phthalate (MnOP)). As
many PEs have the same masses (149, 150, 105 and 104), confirmation
ions may be of relatively low intensity compared to the primary ion, and
many of the protonated phthalic anhydride masses detected were
within 10–20s of each other, a more in-depth study would be needed
to confirm identification.
Among the tentatively identified PEs, a few are particularly notable.
For example, DINP is produced and imported annually between 105
and
106
t in EEA (ECHA, 2018) and is restricted for certain uses (REACH,
2018). Additionally, DINP is an endocrine disrupting chemical at high
doses, although it has a lower toxic potential than others such as DPP
and DEHP (Hannas et al., 2011). Di-n-heptyl phthalate and diisoheptyl
phthalate have lower production volumes, but a higher toxic
potential than some of our target PEs. SAR studies found that di-nheptyl
phthalate and di-isoheptyl phthalate have higher potency for developmental
toxicity and testosterone production in fetal testis (Li et al.,
2019). These compounds may be of concern if they have high human intake
via dust ingestion, and if they were included in overall assessments
of PE mixture toxicity, the mixture toxicity would be higher. Although
we have estimated a low risk based on the set of target analytes in our
study, the presence of more than 40 suspect PEs suggests an underestimation
of both human exposure to PEs and the associated PE mixture
toxicity. This suggests for a more complete evaluation of PE exposure
and toxicity, PEs with HPV and higher toxicity potential should be prioritized
in lists of targeted analysis for indoor dust and other environmental
matrices.
4. Conclusion
This study reports the results of targeted (12 PEs and five APs) and
suspect screening of plasticizers in indoor residential dust. Dust concentrations
of APs were comparable to, or in some homes even higher than,
the concentrations of PEs. While we found no association of plasticizer
concentrations in dust with presence of PVC flooring, homes with carpeting
had higher concentrations of PEs and APs than homes without,
suggesting carpeting as an important source of plasticizers in dust. We
also found significant correlations between DMP and DEP and building
age, which might be due to past use of these LMW PEs as insecticides
in older homes.
We evaluated the risk of indoor plasticizer exposure via dust ingestion
by combining toxicity and human intake estimates. DEHP was
found to have the highest human intake but a lower toxicity compared
to other SVOCs, e.g. PCBs and OCPs. However, a cumulative antiandrogenic
risk assessment is more meaningful since mixture effects
can prevail when multiple PEs with a common mode of toxicity occur.
CRA estimated acceptable risks for children and male adults, however
the wide range of reported TRVs is a large source of uncertainty and the
biggest limitation in risk evaluation and can directly change whether
the CRA crosses the hazard threshold. Furthermore, inclusion of other exposure
routes, especially dietary intake, would lead to higher hazard quotients
associated with PE intake. Lastly, more than 40 suspect PEs were
identified through suspect screening, suggesting higher mixture toxicity
if a larger set of PEs is considered. All PEs with high production volumes
and high toxicity potential are recommended to be quantified in indoor
studies, together with APs which have comparable indoor levels to PEs.
Funding
This research was supported by Czech Science Foundation [GAČR
19-20479S], RECETOX Research Infrastructure [LM2018121], Slovak
Ministry of Health [2012/47-SZU-11], and Center of Excellence of Environmental
Health, [ITMS No. 26240120033], funded by ESIF.
Dr Codling was funded European Union's Horizon 2020 Marie
Skłodowska-Curie Actions project (PullED-MS), No 839243.
CRediT authorship contribution statement
Hale Demirtepe: Investigation, Formal analysis, Writing – original
draft, Writing – review & editing. Lisa Melymuk: Writing – review &
editing, Supervision. Garry Codling: Methodology, Writing – review &
editing. Ľubica Palkovičová Murínová: Supervision, Funding acquisition,
Writing – review & editing. Denisa Richterová: Investigation, Resources,
Writing – review & editing. Vladimíra Rašplová: Investigation,
Resources, Writing – review & editing. Tomáš Trnovec: Conceptualization,
Supervision, Funding acquisition. Jana Klánová: Conceptualization,
Supervision, Funding acquisition, Writing – review & editing.
Declaration of competing interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.scitotenv.2021.146667.
References
Ait Bamai, Y., Araki, A., Kawai, T., Tsuboi, T., Saito, I., Yoshioka, E., Cong, S., Kishi, R., 2016.
Exposure to phthalates in house dust and associated allergies in children aged 6–12
years. Environ. Int. 96, 16–23. https://doi.org/10.1016/j.envint.2016.08.025.
Amin, M.M., Ebrahimpour, K., Parastar, S., Shoshtari-Yeganeh, B., Hashemi, M.,
Mansourian, M., Poursafa, P., Fallah, Z., Rafiei, N., Kelishadi, R., 2018. Association of
urinary concentrations of phthalate metabolites with cardiometabolic risk factors
and obesity in children and adolescents. Chemosphere 211, 547–556. https://doi.
org/10.1016/j.chemosphere.2018.07.172.
ATSDR, 1995. Toxicological Profile for Diethyl Phthalate. Atlanta, Georgia.
ATSDR, 1997. Toxicological Profile for Di-n-Octylphthalate. Atlanta, Georgia.
ATSDR, 2001. Toxicological Profile for Di-n-Butyl Phthalate. Atlanta, Georgia.
Agency for Toxic Substances and Disease Registry (ATSDR), 2019. Toxicological profile for
Di(2-ethylhexyl)phthalate (DEHP). U.S. Department of Health and Human Services,
Public Health Service, Atlanta, GA.
Bekö, G., Callesen, M., Weschler, C.J., Toftum, J., Langer, S., Sigsgaard, T., Høst, A., Kold
Jensen, T., Clausen, G., 2015. Phthalate exposure through different pathways and allergic
sensitization in preschool children with asthma, allergic rhinoconjunctivitis
and atopic dermatitis. Environ. Res. 137, 432–439. https://doi.org/10.1016/j.
envres.2015.01.012.
Benson, R., 2009. Hazard to the developing male reproductive system from cumulative
exposure to phthalate esters-dibutyl phthalate, diisobutyl phthalate, butylbenzyl
phthalate, diethylhexyl phthalate, dipentyl phthalate, and diisononyl phthalate.
Regul. Toxicol. Pharmacol. 53, 90–101. https://doi.org/10.1016/j.yrtph.2008.11.005.
Bergh, C., Torgrip, R., Emenius, G., Östman, C., 2011. Organophosphate and phthalate esters
in air and settled dust - a multi-location indoor study. Indoor Air 21, 67–76.
https://doi.org/10.1111/j.1600-0668.2010.00684.x.
H. Demirtepe, L. Melymuk, G. Codling et al. Science of the Total Environment 781 (2021) 146667
7
Bertelsen, R.J., LØdrup Carlsen, K.C., Calafat, A.M., Hoppin, J.A., Håland, G., Mowinckel, P.,
Carlsen, K.H., LØvik, M., 2013. Urinary biomarkers for phthalates associated with
asthma in Norwegian children. Environ. Health Perspect. 121, 251–256. https://doi.
org/10.1289/ehp.1205256.
Bornehag, C.G., Lundgren, B., Weschler, C.J., Sigsgaard, T., Hagerhed-Engman, L., Sundell, J.,
2005. Phthalates in indoor dust and their association with building characteristics.
Environ. Health Perspect. 113, 1399–1404. https://doi.org/10.1289/ehp.7809.
Bui, T.T., Giovanoulis, G., Cousins, A.P., Magnér, J., Cousins, I.T., de Wit, C.A., 2016. Human
exposure, hazard and risk of alternative plasticizers to phthalate esters. Sci. Total Environ.
541, 451–467. https://doi.org/10.1016/j.scitotenv.2015.09.036.
Christia, C., Poma, G., Harrad, S., de Wit, C.A., Sjostrom, Y., Leonards, P., Lamoree, M.,
Covaci, A., 2019. Occurrence of legacy and alternative plasticizers in indoor dust
from various EU countries and implications for human exposure via dust ingestion
and dermal absorption. Environ. Res. 171, 204–212. https://doi.org/10.1016/j.
envres.2018.11.034.
CPSC, 2017. 16 CFR Part 1307 Prohibition of Children’s Toys and Child Care Articles Containing
Specified Phthalates.
Demirtepe, H., Melymuk, L., Diamond, M.L., Bajard, L., Vojta, Š., Prokeš, R., Sáňka, O.,
Klánová, J., Palkovičová Murínová, Ľ., Richterová, D., Rašplová, V., Trnovec, T., 2019.
Linking past uses of legacy SVOCs with today’s indoor levels and human exposure.
Environ. Int. 127, 653–663. https://doi.org/10.1016/j.envint.2019.04.001.
ECHA, 2018. European Chemicals Agency - Information on Chemicals [WWW Document].
URL https://echa.europa.eu/brief-profile/-/briefprofile/100.043.099 (accessed 9.8.18).
ECHA, 2020. Candidate List of Substances of Very High Concern for Authorisation [WWW
Document]. URL https://echa.europa.eu/candidate-list-table (accessed 6.1.20).
Fournier, K., Tebby, C., Zeman, F., Glorennec, P., Zmirou-Navier, D., Bonvallot, N., 2016.
Multiple exposures to indoor contaminants: derivation of benchmark doses and relative
potency factors based on male reprotoxic effects. Regul. Toxicol. Pharmacol. 74,
23–30. https://doi.org/10.1016/j.yrtph.2015.11.017.
German Federal Environment Agency, 2011. Substance Monograph: Phthalates – New
and Updated Reference Values for Monoesters and Oxidised Metabolites in Urine of
Adults and Children.
Giovanoulis, G., Bui, T., Xu, F., Papadopoulou, E., Padilla-Sanchez, J.A., Covaci, A., Haug, L.S.,
Cousins, A.P., Magnér, J., Cousins, I.T., de Wit, C.A., 2018. Multi-pathway human exposure
assessment of phthalate esters and DINCH. Environ. Int. 112, 115–126. https://
doi.org/10.1016/j.envint.2017.12.016.
Gray, L.E., Ostby, J., Furr, J., Price, M., Veeramachaneni, D.N.R., Parks, L., 2000. Perinatal exposure
to the phthalates DEHP, BBP, and DINP, but not DEP, DMP, or DOTP, alters sexual
differentiation of the male rat. Toxicol. Sci. 58, 350–365. https://doi.org/10.1093/
toxsci/58.2.350.
Hammel, S.C., Levasseur, J.L., Hoffman, K., Phillips, A.L., Lorenzo, A.M., Calafat, A.M.,
Webster, T.F., Stapleton, H.M., 2019. Children’s exposure to phthalates and nonphthalate
plasticizers in the home: the TESIE study. Environ. Int. 132, 105061.
https://doi.org/10.1016/j.envint.2019.105061.
Hannas, B.R., Lambright, C.S., Furr, J., Howdeshell, K.L., Wilson, V.S., Gray, L.E., 2011. Doseresponse
assessment of fetal testosterone production and gene expression levels in
rat testes following in utero exposure to diethylhexyl phthalate, diisobutyl phthalate,
diisoheptyl phthalate, and diisononyl phthalate. Toxicol. Sci. 123, 206–216. https://
doi.org/10.1093/toxsci/kfr146.
Hauser, R., Calafat, A.M., 2005. Phthalates and human health. Occup. Environ. Med. 62,
806–818. https://doi.org/10.1136/oem.2004.017590.
He, R., Li, Y., Xiang, P., Li, C., Zhou, C., Zhang, S., Cui, X., Ma, L.Q., 2016. Organophosphorus
flame retardants and phthalate esters in indoor dust from different microenvironments:
bioaccessibility and risk assessment. Chemosphere 150, 528–535. https://
doi.org/10.1016/j.chemosphere.2015.10.087.
Howdeshell, K.I., Wilson, V.S., Furr, J., Lambright, C.R., Rider, C.V., Blystone, C.R., Hotchkiss,
A.K., Gray, L.E., 2008. A mixture of five phthalate esters inhibits fetal testicular testosterone
production in the Sprague-Dawley rat in a cumulative, dose-additive manner.
Toxicol. Sci. 105, 153–165. https://doi.org/10.1093/toxsci/kfn077.
Jílková, S., Melymuk, L., Vojta, Š., Vykoukalová, M., Bohlin-Nizzetto, P., Klánová, J.,
2018. Small scale spatial variability of flame retardants in indoor dust and implications
for dust sampling. Chemosph. 206, 132–141. https://doi.org/10.1016/j.
chemosphere.2018.04.146.
Johnson, K.J., Heger, N.E., Boekelheide, K., 2012. Of mice and men (and rats): phthalateinduced
fetal testis endocrine disruption is species-dependent. Toxicol. Sci. 129,
235–248. https://doi.org/10.1093/toxsci/kfs206.
Kademoglou, K., Giovanoulis, G., Palm-Cousins, A., Padilla-Sanchez, J.A., Magnér, J., De Wit,
C.A., Collins, C.D., 2018. In vitro inhalation bioaccessibility of phthalate esters and alternative
plasticizers present in indoor dust using artificial lung fluids. Environ. Sci.
Technol. Lett. 5, 329–334. https://doi.org/10.1021/acs.estlett.8b00113.
Kay, V.R., Bloom, M.S., Foster, W.G., 2014. Reproductive and developmental effects of
phthalate diesters in males. Crit. Rev. Toxicol. 44, 467–498. https://doi.org/10.3109/
10408444.2013.875983.
Kishi, R., Ketema, R.M., Ait Bamai, Y., Araki, A., Kawai, T., Tsuboi, T., Saito, I., Yoshioka, E.,
Saito, T., 2018. Indoor environmental pollutants and their association with sick
house syndrome among adults and children in elementary school. Build. Environ.
136, 293–301. https://doi.org/10.1016/j.buildenv.2018.03.056.
Kortenkamp, A., Faust, M., 2010. Combined exposures to anti-androgenic chemicals: steps
towards cumulative risk assessment. Int. J. Androl. 33, 463–472. https://doi.org/
10.1111/j.1365-2605.2009.01047.x.
Kutz, M., 2017. Applied Plastics Engineering Handbook. Elsevier, Second Edi. ed.
Li, X., Mo, J., Zhu, Q., Ni, C., Wang, Y., Li, H., Lin, Z. kun, Ge, R.S., 2019. The structure–
activity relationship (SAR) for phthalate-mediated developmental and reproductive
toxicity in males. Chemosphere 223, 504–513. doi:https://doi.org/10.1016/j.
chemosphere.2019.02.090.
Lowell Center for Sustainable Production, 2011. Phthalates and their Alternatives: Health
and Environmental Concerns (Lowell).
Luongo, G., Östman, C., 2016. Organophosphate and phthalate esters in settled dust from
apartment buildings in Stockholm. Indoor Air 26, 414–425. https://doi.org/10.1111/
ina.12217.
Lyche, J.L., 2017. Phthalates. In: Gupta, R. (Ed.), Reproductive and Developmental Toxicology.
Academic Press.
Meeker, J.D., Calafat, A.M., Hauser, R., 2009. Urinary metabolites of di(2-ethylhexyl)
phthalate are associated with decreased steroid hormone levels in adult men.
J. Androl. 30, 287–297. https://doi.org/10.2164/jandrol.108.006403.
Mitro, S.D., Dodson, R.E., Singla, V., Adamkiewicz, G., Elmi, A.F., Tilly, M.K., Zota, A.R., 2016.
Consumer product chemicals in indoor dust: a quantitative meta-analysis of U.S.
Studies. Environ. Sci. Technol. 50, 10661–10672. https://doi.org/10.1021/acs.
est.6b02023.
Nagorka, R., Conrad, A., Scheller, C., Süßenbach, B., Moriske, H.J., 2011. Diisononyl 1,2cyclohexanedicarboxylic
acid (DINCH) and Di(2-ethylhexyl) terephthalate (DEHT)
in indoor dust samples: concentration and analytical problems. Int. J. Hyg. Environ.
Health 214, 26–35. https://doi.org/10.1016/j.ijheh.2010.08.005.
Narvaez Rincon, P.C., Suarez Palacios, O.Y., 2015. Plasticizers. In: Palsule, S. (Ed.), Encyclopedia
of Polymers and Composites. Springer, Berlin Heidelberg, pp. 1–13 https://doi.
org/10.1007/978-3-642-37179-0_73-1.
Nayef, N., Qasmi, A., Al-thaiban, H., Helaleh, M.I.H., 2019. Indoor phthalates from household
dust in Qatar : implications for non-dietary human exposure. Environ. Sci. Pollut.
Res. 26, 421–430. https://doi.org/10.1007/s11356-018-3604-8.
Net, S., Sempéré, R., Delmont, A., Paluselli, A., Ouddane, B., 2015. Occurrence, fate, behavior
and ecotoxicological state of phthalates in different environmental matrices. Environ.
Sci. Technol. 49, 4019–4035. https://doi.org/10.1021/es505233b.
NRC, 2008. Phthalates and Cumulative Risk Assessment: The Tasks Ahead. National Academies
Press, Washington, DC.
Pelletier, M., Glorennec, P., Mandin, C., Le Bot, B., Ramalho, O., Mercier, F., Bonvallot, N.,
2018. Chemical-by-chemical and cumulative risk assessment of residential indoor exposure
to semivolatile organic compounds in France. Environ. Int. 117, 22–32.
https://doi.org/10.1016/j.envint.2018.04.024.
Plastics Europe, 2020. Plastics - the facts 2020.
Promtes, K., Kaewboonchoo, O., Kawai, T., Miyashita, K., Panyapinyopol, B., Kwonpongsagoon,
S., Takemura, S., 2019. Human exposure to phthalates from house dust in Bangkok.
Thailand. J. Environ. Sci. Heal. - Part A Toxic/Hazardous Subst. Environ. Eng. 54,
1269–1276. https://doi.org/10.1080/10934529.2019.1637207.
PubChem, 2020. U.S. National Library of Medicine National Center for Biotechnology Information
[WWW Document]. URL https://pubchem.ncbi.nlm.nih.gov/ (accessed
1.31.20).
Radke, E.G., Glenn, B.S., Braun, J.M., Cooper, G.S., 2019. Phthalate exposure and female
reproductive and developmental outcomes: a systematic review of the
human epidemiological evidence. Environ. Int. 130, 104580. https://doi.org/
10.1016/j.envint.2019.02.003.
REACH, 2018. Commission Regulation (EU) 2018/2005 of 17 December 2018 Amending
Annex XVII to Regulation (EC) no 1907/2006 of the European Parliament and of
the Council Concerning the Registration, Evaluation, Authorisation and Restriction
of Chemicals (REACH) as Regar.
Sedtasiriphokin, N., Supornsilchai, V., Jantarat, C., Nosoongnoen, W., 2017. Phthalate exposure
in Thai children and adolescents. Asian Biomed. 11, 343–352. https://doi.
org/10.1515/abm-2018-0006.
Shin, H.M., Moschet, C., Young, T.M., Bennett, D.H., 2020. Measured concentrations of consumer
product chemicals in California house dust: implications for sources, exposure,
and toxicity potential. Indoor Air 30, 60–75. https://doi.org/10.1111/ina.12607.
Stanley, M.K., Robillard, K.A., Staples, C.A., 2003. Introduction. In: Staples, C.A. (Ed.),
Phthalate Esters - Handbook of Environmental Chemistry. Springer, Berlin Heidelberg,
pp. 1–7.
Staples, C.A., 2003. Phthalate esters. Handbook of Environmental Chemistry. Springer.
https://doi.org/10.1007/b11460.
Subedi, B., Sullivan, K.D., Dhungana, B., 2017. Phthalate and non-phthalate plasticizers in
indoor dust from childcare facilities, salons, and homes across the USA. Environ.
Pollut. 230, 701–708. https://doi.org/10.1016/j.envpol.2017.07.028.
Swan, S.H., Main, K.M., Liu, F., Stewart, S.L., Kruse, R.L., Calafat, A.M., Mao, C.S., Redmon,
J.B., Ternand, C.L., Sullivan, S., Teague, J.L., The Study for Future Families Research
Team, 2005. Decrease in anogenital distance among male infants with prenatal
phthalate exposure. Environ. Health Perspect. 113, 1056–1061. https://doi.org/
10.1289/ehp.8100.
Tang, B., Christia, C., Malarvannan, G., Liu, Y.E., Luo, X.J., Covaci, A., Mai, B.X., Poma, G.,
2020. Legacy and emerging organophosphorus flame retardants and plasticizers in
indoor microenvironments from Guangzhou, South China. Environ. Int. 143,
105972. https://doi.org/10.1016/j.envint.2020.105972.
Tsugawa, H., Cajka, T., Kind, T., Ma, Y., Higgins, B., Ikeda, K., Kanazawa, M., VanderGheynst,
J., Fiehn, O., Arita, M., 2015. MS-DIAL: data independent MS/MS deconvolution for
comprehensive metabolome analysis. Nat. Methods 12, 523–526. https://doi.org/
10.1038/nmeth.3393.
U.S. EPA, 2011. Exposure Factors Handbook 2011 Edition (Final Report). Washington, DC.
US EPA, 2019. Proposed Designation of Di-Ethylhexyl Phthalate (DEHP) (1,2-Benzene- Dicarboxylic
Acid, 1,2-Bis (2-Ethylhexyl) Ester) (CASRN 117-81-7) as a High-Priority
Substance for Risk Evaluation.
Varshavsky, J.R., Zota, A.R., Woodruff, T.J., 2016. A novel method for calculating potencyweighted
cumulative phthalates exposure with implications for identifying racial/
ethnic disparities among U.S. reproductive-aged women in NHANES 2001-2012. Environ.
Sci. Technol. 50, 10616–10624. https://doi.org/10.1021/acs.est.6b00522.
Venier, M., Audy, O., Vojta, Š., Bečanová, J., Romanak, K., Melymuk, L., Krátká, M., Kukučka,
P., Okeme, J., Saini, A., Diamond, M.L., Klánová, J., 2016. Brominated flame retardants
H. Demirtepe, L. Melymuk, G. Codling et al. Science of the Total Environment 781 (2021) 146667
8
in the indoor environment - comparative study of indoor contamination from three
countries. Environ. Int. 94, 150–160. https://doi.org/10.1016/j.envint.2016.04.029.
Vojta, Š., Melymuk, L., Klánová, J., 2017. Changes in flame retardant and legacy contaminant
concentrations in indoor air during building construction, furnishing, and use.
Environ. Sci. Technol. 51, 11891–11899. https://doi.org/10.1021/acs.est.7b03245.
Vykoukalová, M., Venier, M., Vojta, Š., Melymuk, L., Bečanová, J., Romanak, K., Prokeš, R.,
Okeme, J.O., Saini, A., Diamond, M.L., Klánová, J., 2017. Organophosphate esters
flame retardants in the indoor environment. Environ. Int. 106, 97–104. https://doi.
org/10.1016/j.envint.2017.05.020.
Wang, Y., Ni, C., Li, X., Lin, Z., Zhu, Q., Li, L., Ge, R.S., 2019. Phthalate-induced fetal leydig
cell dysfunction mediates male reproductive tract anomalies. Front. Pharmacol. 10,
1–11. https://doi.org/10.3389/fphar.2019.01309.
Wormuth, M., Scheringer, M., Vollenweider, M., Hungerbühler, K., 2006. What are the
sources of exposure to eight frequently used phthalic acid esters in Europeans?
Risk Anal. 26, 803–824. https://doi.org/10.1111/j.1539-6924.2006.00770.x.
Yang, C., Harris, S.A., Jantunen, L.M., Siddique, S., Kubwabo, C., Tsirlin, D., Latifovic, L.,
Fraser, B., St-Jean, M., De La Campa, R., You, H., Kulka, R., Diamond, M.L., 2019.
Are cell phones an indicator of personal exposure to organophosphate flame retardants
and plasticizers? Environ. Int. 122, 104–116. https://doi.org/10.1016/j.
envint.2018.10.021.
Yang, C., Harris, S.A., Jantunen, L.M., Kvasnicka, J., Nguyen, L.V., Diamond, M.L., 2020.
Phthalates: relationships between air, dust, electronic devices, and hands with implications
for exposure. Environ. Sci. Technol. https://doi.org/10.1021/acs.est.0c00229.
Zhang, Qinghao, Sun, Y., Zhang, Qingnan, Hou, J., Wang, P., Kong, X., Sundell, J., 2020.
Phthalate exposure in Chinese homes and its association with household consumer
products. Sci. Total Environ. 719, 136965. https://doi.org/10.1016/j.
scitotenv.2020.136965.
H. Demirtepe, L. Melymuk, G. Codling et al. Science of the Total Environment 781 (2021) 146667
9
White, Kevin B., Ondřej Sáňka, Lisa Melymuk, Petra Přibylová, and Jana Klánová. 2021.
"Application of Land Use Regression Modelling to Describe Atmospheric Levels of
Semivolatile Organic Compounds on a National Scale." Science of The Total Environment
793. https://doi.org/10.1016/j.scitotenv.2021.148520
Application of land use regression modelling to describe atmospheric
levels of semivolatile organic compounds on a national scale
Kevin B. White, Ondřej Sáňka, Lisa Melymuk⁎, Petra Přibylová, Jana Klánová
RECETOX, Faculty of Science, Masaryk University, Kamenice 753/5, 625 00 Brno, Czechia
H I G H L I G H T S
• Spatial models developed from atmospheric
levels and GIS data at 29 sampling
sites
• PAHs related to fuel consumption, industrial
sources, and topography (R2
=
0.68)
• PCBs related to elevation and soil concentrations
(R2
= 0.62)
• National maps created from PAH and
PCB models identified pollutant
hotspots
G R A P H I C A L A B S T R A C T
a b s t r a c ta r t i c l e i n f o
Article history:
Received 6 April 2021
Received in revised form 9 June 2021
Accepted 14 June 2021
Available online 19 June 2021
Editor: Jianmin Chen
Keywords:
Air pollution
Passive air sampling
Polycyclic aromatic hydrocarbons
Polychlorinated biphenyls
Spatial analysis
Despite the success of passive sampler-based monitoring networks in capturing global atmospheric distributions
of semivolatile organic compounds (SVOCs), their limited spatial resolution remains a challenge. Adequate spatial
coverage is necessary to better characterize concentration gradients, identify point sources, estimate human
exposure, and evaluate the effectiveness of chemical regulations such as the Stockholm Convention on Persistent
Organic Pollutants. Land use regression (LUR) modelling can be used to integrate land use characteristics and
other predictor variables (industrial emissions, traffic intensity, demographics, etc.) to describe or predict the distribution
of air concentrations at unmeasured locations across a region or country. While LUR models are frequently
applied to data-rich conventional air pollutants such as particulate matter, ozone, and nitrogen oxides,
they are rarely applied to SVOCs.
The MONET passive air sampling network (RECETOX, Masaryk University) continuously measures atmospheric
SVOC levels across Czechia in monthly intervals. Using monitoring data from 29 MONET sites over a two-year period
(2015–2017) and a variety of predictor variables, we developed LUR models to describe atmospheric levels
and identify sources of polycyclic aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs) and DDT
across the country. Strong and statistically significant (R2
> 0.6; p < 0.05) models were derived for PAH and
PCB levels on a national scale. The PAH model retained three predictor variables – heating emissions represented
by domestic fuel consumption, industrial PAH point sources, and the hill:valley index, a measure of site topography.
The PCB model retained two predictor variables – site elevation, and secondary sources of PCBs represented
by soil concentrations. These models were then applied to Czechia as a whole, highlighting the spatial variability
of atmospheric SVOC levels, and providing a tool that can be used for further optimization of sampling network
design, as well as evaluating potential human and environmental chemical exposures.
© 2021 Elsevier B.V. All rights reserved.
1. Introduction
Semivolatile organic compounds (SVOCs) are a broad class of atmospheric
pollutants that include polycyclic aromatic hydrocarbons
Science of the Total Environment 793 (2021) 148520
⁎ Corresponding author at: Masaryk University, Faculty of Science, RECETOX, Kamenice
753/5, Pavilion D29, 625 00 Brno, Czechia.
E-mail address: lisa.melymuk@recetox.muni.cz (L. Melymuk).
https://doi.org/10.1016/j.scitotenv.2021.148520
0048-9697/© 2021 Elsevier B.V. All rights reserved.
Contents lists available at ScienceDirect
Science of the Total Environment
journal homepage: www.elsevier.com/locate/scitotenv
(PAHs), polychlorinated biphenyls (PCBs), and organochlorine pesticides
(OCPs). An ongoing challenge in monitoring SVOCs in many regions is the
lack of sufficient spatial coverage to characterize atmospheric concentration
gradients, identify sources, evaluate human exposure, and assess the
effectiveness of national and international regulations such as the
Stockholm Convention on Persistent Organic Pollutants (POPs). While
concentrations of conventional air quality pollutants such as nitrogen oxides
(NOX), ozone (O3), and particulate matter (PM) can be rapidly measured
in “on-line” mode resulting in continuous monitoring, SVOCs
require complex equipment and daily/weekly time resolution for active
air sampling. Widespread use of passive air sampling for SVOCs has
helped to improve spatial coverage, providing a reliable technique to
quantify SVOCs across different site types and conditions without the
use of complex equipment (Bohlin-Nizzetto et al., 2020; Herkert et al.,
2018; Wania and Shunthirasingham, 2020). Yet except for national air
monitoring networks in Czechia (Kalina et al., 2018), Norway (Tørseth
et al., 2012), Spain (Muñoz-Arnanz et al., 2016), and the United
Kingdom (Graf et al., 2016), most countries in Europe have only 1 or 2
long-term SVOC monitoring sites, with monitoring of POPs being particularly
scarce (Aas and Bohlin-Nizzetto, 2018; Hulek et al., 2014).
Spatial modelling can be a powerful tool to fill gaps in the geographic
coverage of monitoring networks. Land use regression (LUR) is a spatial
modelling technique that can be used to describe the environmental
distribution of chemical concentrations at unmeasured locations across
a region based on land use and related predictor variables
(e.g., industrial emission records, traffic intensities, demographics).
LUR methods have been used extensively in spatial studies of conventional
air quality pollutants such as NOX, O3, and PM (Beelen et al.,
2013; Eeftens et al., 2012; Gulliver et al., 2018; Hoek et al., 2008;
Jedynska et al., 2017, 2014; Lu et al., 2020; Van Nunen et al., 2017), as
well as some volatile organic compounds (Amini et al., 2017). While
these studies typically investigate small-scale spatial variation within
individual cities to evaluate human exposure, some studies have used
LUR models to predict atmospheric concentrations on larger scales,
such as across Europe (de Hoogh et al., 2018, 2016; Vienneau et al.,
2013; Vizcaino and Lavalle, 2018) and even globally (Larkin et al.,
2017). However, due to the greater complexity of measuring SVOCs
and the relative scarcity of permanent monitoring sites compared to
conventional air pollutants, few attempts have been made to apply
LUR models to SVOCs. Some urban atmospheric LUR studies have included
PAHs, but typically in the context of characterizing PM composition
where the models reflect particle-phase concentrations in air
(Jedynska et al., 2014; Noth et al., 2011; Polidori et al., 2010); only
two LUR studies have considered gas-phase PAHs (Masri et al., 2018;
Noth et al., 2016). Furthermore, to our knowledge there has only been
one previous study to apply LUR methods to atmospheric PCB concentrations
(Melymuk et al., 2013), and none for OCPs.
The MONET passive air sampling network (RECETOX, Masaryk University)
is the largest national SVOC monitoring network in Europe, and
includes 29 permanent sampling sites spread across Czechia that continuously
measure atmospheric SVOC levels in 28-day intervals
(Kalina et al., 2018). Additional MONET sampling extends across the
rest of Europe as well as Africa (Přibylová et al., 2012; White et al.,
2020). However, the relatively dense spatial coverage of the MONET
network in Czechia (29 sites compared with 1–2 sites in most other
European countries), provides an opportunity to apply LUR methods
to SVOC monitoring data at a national scale. In this study we use spatial
predictor variables extracted using geographic information systems
(GIS) and two years of passive air sampling data (2015–2017) from
the 29 Czech sites in the MONET network to develop LUR models for
PAHs, PCBs, and the OCP dichlorodiphenyltrichloroethane (DDT). We
demonstrate the strength of LUR for developing generalized nationalscale
models of SVOCs that can help to improve sampling network design
for optimal coverage of sources, identify regional differences for
human and environmental exposure assessment, and assist in evaluating
the effectiveness of chemical regulations.
2. Methods
2.1. Air sampling
Continuous passive air sampling of SVOCs in Czechia is coordinated
by the MONET network at RECETOX (Masaryk University, Brno) and
has occurred at more than 60 sites across the country since 2006, with
29 sites operating as of 2017 (Fig. 1). These sites cover a wide range of
land types and regions including remote mountainous sites, rural agricultural
sites, urban city sites, and industrial sites (Table S1). Air monitoring
data for all 29 sites were used for model development, covering
a two-year period (July 2015 to June 2017) to reduce the potential influence
of seasonality and variance in atmospheric levels within a single
year, while avoiding the influence of long-term temporal trends previously
identified for SVOCs at some of these sites (Kalina et al., 2018).
Hourly meteorological conditions over the two-year monitoring period
were assessed using the MERRA-2 model (Gelaro et al., 2017), as previously
described for passive air sampling studies (Bohlin-Nizzetto et al.,
2020; Herkert et al., 2018; White et al., 2020), and were relatively uniform
between all 29 sites. Median site temperatures ranged from
7.0–9.4 °C and median site wind speeds ranged from 1.0–2.7 m/s.
Based on continuous monitoring in 28-d passive sampling intervals,
this two-year data selection resulted in 26 samples per site – except
for Churáňov, Rudolice, and Rýchory with only 25, 25, and 24 samples,
respectively – for a total of 750 samples. The sampling periods and deployment
dates for all samples at each site included in this study are
depicted in Fig. S1. Ninety percent of samples were deployed for exactly
28 d, and 99% were deployed for 26 to 30 d (Fig. S2).
MONET passive samplers consist of two stainless steel bowls containing
a 15 cm × 1.5 cm polyurethane foam (PUF) disk and are typically
deployed at 1–2 m above ground level. After the sampling period, PUF
disks are collected and analyzed for a full suite of SVOCs at the
RECETOX Trace Analytical Laboratory. Analytical methods are described
in detail in the Supplemental Materials (SM Section 2 – Analytical
Methods). In this study we examined seven indicator PCBs: 28, 52,
101, 118, 138, 153, and 180; fourteen PAHs: acenaphthylene (ACY),
acenaphthene (ACE), fluorene (FLU), phenanthrene (PHE), anthracene
(ANT), pyrene (PYR), fluoranthene (FLA), chrysene (CHR), benz[a]anthracene
(BAA), benzo[b]fluoranthene (BBF), benzo[k]fluoranthene
(BKF), benzo[a]pyrene (BAP), benzo[ghi]perylene (BGP), and indeno
[1,2,3-cd]pyrene (IP); and DDT and its degradation products (o,p’DDD,
o,p’-DDE, o,p’-DDT, p,p’-DDD, p,p’-DDE, p,p’-DDT). Several of
these compounds, particularly the high molecular weight PAHs and
PCBs, were often present within the PUF disks at concentrations below
their instrumental limit of quantification (LOQ, 3.3× the limit of detection;
Table S2). For the purposes of data analysis and model development,
these left-censored values were substituted by (√2/2) x LOQ
(Antweiler, 2015). Levels of SVOCs measured by passive sampling are
reported in ng per PUF disk per sampling period. Although SVOC levels
from passive sampling are often converted to concentrations (i.e., ng/
m3
), this conversion may be biased by a number of sources of uncertainty
(Holt et al., 2017). We chose to work with the primary data in
units of ng/PUF/d as these values are still appropriate for spatial assessment
and analysis of SVOCs (Kalina et al., 2018). To address atmospheric
levels at a national scale, and in consideration of regulations and policies
that typically address groups of chemicals of concern, we chose to develop
simplified models using the sums of SVOC compound classes
(∑7PCB, ∑14PAH, and ∑6DDT).
2.2. Predictor variables
Tool functions in ArcGIS Desktop software (v.10.5, Esri) were used to
determine values of predictor variables for all sampling sites, either at
the site itself (e.g., elevation) or in buffer zones of defined sizes (1 km
and 5 km) around each site (e.g., population, traffic density). Both buffer
zone sizes were examined for buffered variables, and the buffer with the
K.B. White, O. Sáňka, L. Melymuk et al. Science of the Total Environment 793 (2021) 148520
2
strongest single variable correlation was used in model development.
Atmospheric SVOC levels and most predictor variables were lognormally
distributed and thus log-transformed prior to modelling, except
for emission point sources and the hill-valley index (HVI). Determination
of the predictor variables is summarized in Table 1 and detailed
in the Supplementary Materials (SM Section 3 – Predictor Variables).
Annual consumption of several heating fuel types (propane-butane,
natural gas, bituminous coal, oil, coke, lignite, and wood) was estimated
for all 6354 administrative units of Czechia based on the relationship between
population and fuel consumption records from a subset of 145 administrative
units located in the north of the country. This region was
previously used as an input for PAH air concentration modelling (Sáňka
et al., 2014) and is considered representative of the country as a whole,
as it covers a diversity of settlement types (large city of >100,000 people,
as well as smaller villages and towns). Population (counts) and annual
fuel consumption (masses) for each administrative unit were logtransformed
to normalize distributions, and a linear regression was applied
(Fig. S10). The linear relationships observed between the logtransformed
fuel consumption and population (Table S3) were then
used to extrapolate annual fuel consumption in the remaining administrative
units according to settlement size (Fig. S11).
PAH and PCB emission factors for the various fuel types were estimated
according to published studies (Tables S4 and S5). While a significant
amount of uncertainty exists in the emission factors due to
differences in heater types, efficiencies, operating procedures and fuel
qualities, they convey the relative differences according to fuel efficiencies
and emissions (e.g., PAH emissions from lignite burning are 200×
greater than from home heating oil and 2000× greater than from natural
gas). As a result of these findings, coke, lignite, wood, and bituminous
coal were deemed to be the four major domestic fuel types for PAH and
PCB emissions across Czechia. In addition to the consumption of these
four individual fuel types being used as predictor variables in model development,
the sum of all four (total fuel consumption) was also
included.
2.3. Model development
Model development followed a similar approach to the standard
LUR method that has previously been applied in atmospheric studies
of PM, NOX, and PAHs (Beelen et al., 2013; de Hoogh et al., 2018, 2016,
2014, 2013; Eeftens et al., 2012; Jedynska et al., 2014). Multiple linear
regression models were derived for atmospheric levels of ∑14PAH,
∑7PCB, and ∑6DDT via a multidirectional stepwise method based on
Akike information criteria (AIC) selection using the stepAIC() function
in R software (v.3.4.2). This function adds and removes predictor variables
iteratively until it identifies a statistically significant final model
that maximizes the explained variance (adjusted R2
) in the dependent
variable (atmospheric SVOC levels). However, while the final model
may be statistically significant, individual included predictor variables
may not be, or may only negligibly affect the overall explained
variance. In the previously mentioned LUR studies, individual site- and
compound-specific models were developed to explain as much of the
variance in atmospheric levels as possible. As a result, optimization of
these final models was relatively lenient with the inclusion of all predictor
variables with p-values ≤ 0.10 that each increased the adjusted R2
by
≥0.01 (Beelen et al., 2013; de Hoogh et al., 2014, 2013; Eeftens et al.,
2012; Jedynska et al., 2014). However, our aim was to develop simplified
models for SVOC groups on a national scale to identify the most important
predictor variables. As a result, we used a more stringent
optimization process in which only predictor variables with p-values ≤
0.05 that increased the adjusted R2
by ≥0.05 were retained in the final
models selected by the stepAIC() function. Validation of each optimized
model was performed using a standard leave-one-out cross validation
(LOOCV) method (Beelen et al., 2013; de Hoogh et al., 2014, 2013;
Eeftens et al., 2012; Jedynska et al., 2014). Spatial autocorrelation of
the residuals from each optimized model were subsequently assessed
using the Moran's I statistic (Beelen et al., 2013; de Hoogh et al., 2013;
Eeftens et al., 2012; Jedynska et al., 2014). Finally, normality of residuals
for each optimized model was assessed using the Shapiro-Wilk test.
Fig. 1. MONET passive air sampling sites in Czechia included in this study. Shading indicates topography, and cities with population > 100,000 people are identified.
K.B. White, O. Sáňka, L. Melymuk et al. Science of the Total Environment 793 (2021) 148520
3
3. Results & discussion
3.1. Atmospheric SVOC levels
Average atmospheric levels of each SVOC compound class were similar
across most sites. The lowest average levels for PAHs, PCBs, and DDT
were detected at sites in the remote mountainous regions around the
border of the country, many of which were previously identified as
being the least polluted sites in an assessment of Czech background
air monitoring (Kalina et al., 2018), while the sites with the highest average
levels differed by compound class (Fig. 2).
Average atmospheric ∑14PAH levels varied over approximately one
order of magnitude, ranging from 14–124 ng/PUF/d at all sites except
Prague Radotin (255 ng/PUF/d) and Věřňovice (394 ng/PUF/d)
(Fig. 2A). Both of these sites are heavily influenced by local sources,
with Prague Radotin located at a cement factory and Věřňovice located
on the Czech-Polish border in the centre of a coal mining and industrial
region with significant coal use for domestic heating. Phenanthrene was
the most abundant individual PAH across all 29 sites, comprising 46 ±
3% of the average ∑14PAH levels, followed by fluoranthene, fluorene,
and pyrene. Together these four PAHs alone comprised 90 ± 3% of the
average ∑14PAH levels at all sites (Table S7).
Average atmospheric ∑7PCB levels ranged from 0.04–0.25 ng/PUF/
d, with only Brno Kotlarska significantly elevated above all others (0.35
ng/PUF/d; Fig. 2B). Similarly, the relative proportions of individual PCB
congeners contributing to ∑7PCB levels were comparable among all
sites, except for a significantly higher average level of PCB 28 at Brno
Kotlarska (Table S8).
Average atmospheric ∑6DDT levels ranged from 0.03–0.70 ng/PUF/
d across all sites (Fig. 2C) and were dominated by p,p’-DDE (74 ± 7%).
Four of the five sites with the highest DDT levels are located in the
South Moravia region and surrounded by large areas of agricultural
land, reflecting historical agricultural application of DDT. The other
site with high levels (Libiš) is located less than 1 km away from a former
pesticide manufacturing plant, suggesting environmental contamination
from past industrial production. This difference in dominant
sources was apparent from the differences in the composition of
∑6DDT, with the four agricultural sites having the highest p,p’-DDE
levels, while Libiš had the highest levels of the other five DDT compounds
(Table S9).
3.2. Model development
Statistically significant multiple linear regression models were derived
for atmospheric ∑14PAH and ∑7PCB levels across Czechia,
while only a simple linear regression model could be derived for
∑6DDT levels. Of all individual predictor variables, sampling site elevation
was able to explain the greatest amount of variance in the atmospheric
levels of all SVOCs, with a moderately strong negative
correlation (adjusted R2
ranging from 0.45–0.49), indicating higher
levels at lower elevations (Table S10). Most background air sampling
sites within Czechia are located in mountainous regions around the
border of the country while large cities are located at lower elevations
towards the centre. As a result, elevation of the sampling sites was
also significantly negatively correlated with most of the other selected
predictor variables: building, crop, and industrial land coverage,
Table 1
Predictor variables and data sources.
Predictor variable Description and data source
Fuel consumption
(ConsFuel)
See Section 2.2
PAH emissions from local heating
(HeatEmissPAH)
PCB emissions from local heating
(HeatEmissPCB)
PAH industrial source density
(PointSourcePAH)
PM10 industrial source density
(PointSourcePM10)
The European Environmental Agency's European Pollutant Release and Transfer Register (E-PRTR, http://prtr.ec.europa.eu) was used to
estimate industrial air pollution. All pollution sources within Czechia and neighbouring regions of adjacent countries with either PAH, PCB,
or PM10 records were extracted and imported into GIS. Only one official record of PCB sources was identified, thus PCBs were not
considered further. A kernel density function was applied to PAH and PM10 sources, with the emission intensity of pollution sources
selected as the population attribute. The search radius of the kernel density tool was set to 40 km. Rasters were created for both PAH and
PM10 emissions, and the corresponding values were extracted for each sampling site for use as predictor variables (Fig. S3).
PCB concentrations
in soil
(SoilPCB)
Kubošová et al. (2009) used regression trees to predict concentrations of 5 PCB congeners (101, 118, 138, 153, and 180) in soil across
Czechia at a 1 km spatial resolution. Predicted soil concentrations were imported to GIS and average values were extracted for each
sampling site using a 5 km radius buffer.
Traffic density
(DensTraf)
Road traffic density (vehicles/day) was taken from the Country Traffic Census performed by the Czech Roads and Motorways Directorate
(http://scitani2010.rsd.cz/pages/informations/default.aspx) and spatially attributed in GIS. Exact values were available for a subset of roads,
including all major roads. Roads of lower classes without exact values were assigned average values based on roads of the same class. Small streets
inside cities and villages were excluded. A GIS line density tool was used to create a grid of traffic density across the country, and the sum of all
traffic density grid values was calculated in a 1500 m radius buffer around each sampling site (Fig. S4).
Road density
(LengRoad)
Road density was determined using an OpenStreetMap dataset from Geofabrik's server (http://download.geofabrik.de/). Only roads with
car traffic were included in the analysis. The total length of roads within a 1 km and 5 km radius buffer of each sampling site was calculated (Fig. S5).
Building density
(AreadBuild)
Building density was determined using a dataset from Geofabrik's server (http://download.geofabrik.de/). The total percentage of built-up
area within a 1 km and 5 km radius of each sampling site was calculated. It is important to note that this dataset did not include building
heights, only land coverage, which could limit its explanatory potential (Fig. S6).
Population
(Pop)
The population count of each administrative unit in Czechia was taken from ArcData Prague (https://www.arcdata.
cz/produkty/geograficka-data/arccr-500). Administrative units were reshaped in GIS by excluding water bodies and forests to achieve a
more realistic population distribution. The total population was calculated within a 1 km and 5 km radius of each sampling site (Fig. S7).
Land use
(AreaCrop)
(AreaIndus)
The Corine Land Cover 2006 database (http://www.eea.europa.eu/data-and-maps/data/clc-2006-vector-4) was used to determine spatial
coverage of classes 121 (industry) and 211, 221 and 222 (agriculture) within a 1 km and 5 km radius of each sampling site (Fig. S8).
Elevation
(Elevation)
The elevation of each sampling site was acquired using GPS measurement during sampler deployment.
Hill-valley index
(HVI)
To incorporate the effect of topography into the LUR without the use of more complex contaminant air transport models, a ‘hill-valley index’ (HVI)
was developed to quantify the position of each sampling site relative to the elevation of the surrounding terrain. The HVI at each site was calculated
by dividing the elevation of the site by the median elevation of a 1.5 km radius buffer around each site (acquired from Shuttle Radar Topographic
Mission, http://www.diva-gis.org/gdata). Thus, HVI <1 indicates the site is located in a valley, while HVI >1 indicates the site is located on a hill
(see Fig. S9).
K.B. White, O. Sáňka, L. Melymuk et al. Science of the Total Environment 793 (2021) 148520
4
population, total fuel consumption, road length, and traffic density
(Table S11). Elevation is therefore a moderately strong predictor of anthropogenic
activity within Czechia. Population within 5 km of each
sampling site was also significantly correlated with atmospheric levels
of all SVOCs, however the strength of the correlation was much lower
than that of elevation (adjusted R2
ranging from 0.10–0.25). Similar to
elevation, population was also significantly correlated with most of
the other selected predictor variables. Since the strong correlation between
atmospheric SVOC levels and site elevation is a result of the specific
topography and population distribution across Czechia, we
AtmosphericSumPAHlevels
(ng/PUF/d)
Klet
Churanov
Jested
Primda
Rychory
Kosetice
Sedlonov
Rudolice
BilyKriz
Prebuz
Jesenik
Sedlec
Hodonin
Kucharovice
BrnoLisen
MokraHorakov
PragueLibus
Planavy
Svratouch
Kosor
Libis
BrnoKotlarska
Blansko
Sivice
Kyjov
Sneznik
BrnoTurany
PragueRadotin
Vernovice
1
10
100
1000
AAtmosphericSumPCBlevels
(ng/PUF/d)
Klet
BilyKriz
Jested
Kosetice
Churanov
Jesenik
Primda
Sedlonov
Rychory
Kucharovice
MokraHorakov
Sedlec
Planavy
Kosor
Svratouch
BrnoLisen
Prebuz
PragueRadotin
Sivice
Rudolice
Blansko
Kyjov
Hodonin
PragueLibus
Vernovice
Sneznik
Libis
BrnoTurany
BrnoKotlarska
0.001
0.01
0.1
1
B
AtmosphericSumDDTlevels
(ng/PUF/d)
Klet
Churanov
BilyKriz
Primda
PragueRadotin
Jesenik
Jested
Sedlonov
Rychory
Kosor
PragueLibus
Kosetice
Planavy
Prebuz
BrnoLisen
Svratouch
Blansko
MokraHorakov
Rudolice
BrnoKotlarska
Vernovice
Sivice
Sneznik
Hodonin
Kyjov
Libis
Sedlec
Kucharovice
BrnoTurany
0.001
0.01
0.1
1
C
Fig. 2. Log-boxplots of atmospheric levels (ng per PUF disk per day) of A) ∑14PAH; B) ∑7PCB; and C) ∑6DDT. Levels measured monthly (28-d) by passive sampling over a two-year
period (July 2015–June 2017) at 29 sites across Czechia (n = 24–26 at each site). Boxplots depict the interquartile range (IQR): box upper limit is the 75th percentile, horizontal line is
the 50th percentile (median), box lower limit is the 25th percentile; whiskers above and below depict 1.5 x IQR; data points outside this range (outliers) are depicted as hollow
circles. Sites are listed in order of increasing average level (depicted by triangles).
K.B. White, O. Sáňka, L. Melymuk et al. Science of the Total Environment 793 (2021) 148520
5
anticipate that population density would likely be a more general predictor
of anthropogenic activity if similar LUR methods were applied
to atmospheric SVOC levels in other countries.
While ∑6DDT was only significantly correlated with elevation and
population, ∑14PAH and ∑7PCB were both significantly correlated
with many of the other predictor variables to varying degrees. However,
most were not retained in the final models selected by stepAIC, and the
subsequent optimization simplified the models even further. Stepwise
removal of predictor variables that did not meet selection criteria during
optimization of each model is summarized in Table S12. Overall, the
models for ∑14PAH and ∑7PCB were relatively strong (adjusted R2
>
0.60) with no spatial autocorrelation (Moran's I of 0.003 and 0.048, respectively).
Conversely, the model explained variance for ∑6DDT was
lower (adjusted R2
= 0.45) and weakly spatially autocorrelated
(Moran's I of 0.142, p < 0.05). Despite this, the LOOCV R2
was <10%
lower than the adjusted R2
for all three SVOC models, indicating stable
models with minimal influence of individual sites (Beelen et al., 2013;
de Hoogh et al., 2014, 2013; Eeftens et al., 2012; Jedynska et al., 2014).
Furthermore, all three final models passed the Shapiro-Wilk test for
normality on model residuals (all >0.95) indicating appropriate log
transformations.
3.2.1. PAH model
The final optimized ∑14PAH model included three predictor variables
– total domestic fuel consumption (sum of lignite, bituminous
coal, coke, and wood), PAH point sources, and the HVI – and explained
nearly 70% of the variance in atmospheric levels (Eq. (1)).
log ∑14PAHð Þ ¼ 0:37 log ConsFuelð Þ þ 0:28 log PointSourcePAHð Þ
−1:83 HVIð Þ þ 2:35
Adjusted R2
¼ 0:68 p < 0:05ð Þ; LOOCV R2
¼ 0:62 p < 0:05ð Þ
h i ð1Þ
Surprisingly, neither road length nor traffic density were included in
our final optimized model. Previous LUR models developed for PAHs
have found traffic-related predictor variables to explain a significant
portion of the variance in atmospheric levels (Noth et al., 2016), though
the strengths of the correlations were still generally weaker than those
for other air pollutants (Jedynska et al., 2014; Melymuk et al., 2013).
Melymuk et al. (2013) attributed the weaker explanatory power of
their PAH model to more spatially complex emission sources and dynamics,
as well as greater uncertainty in passive sampling of heavier
particle-associated compounds. Jedynska et al. (2014) attributed the
poorer performance of their PAH model to a smaller influence of traffic
emissions on atmospheric PAH levels relative to other less characterized
sources, such as wood burning and industrial emissions, for which no
data were available. For our study, enough data were available in
Czechia to provide estimates for both of these primary emission sources,
with total domestic fuel consumption and industrial PAH point sources
both individually explaining modest amounts of the variance in atmospheric
levels (30% and 21%, respectively).
Jedynska et al. (2014) also proposed the incorporation of chimney
height into PAH models to account for the influence of distance and vertical
transport of PAHs from point sources to sampling sites. The atmospheric
lifetime of PAHs is generally much shorter than that of other
SVOCs such as PCBs, due to dry deposition of particle-associated compounds
as well as photochemical degradation (Lammel et al., 2009).
As a result, atmospheric levels of PAHs are strongly affected by the altitude
at which emissions occur (such as chimney height; Wu et al., 2006)
and sharply decline with increasing altitude if sources are predominantly
at ground level (Farrar et al., 2005; Tao et al., 2007). Surrounding
topography can therefore significantly compound these effects, with
higher levels measured in the lowest atmospheric layers in valleys compared
to those at mountaintops, even when affected by the same local
source (Kalberer et al., 2004). Some influence of this vertical distribution
on atmospheric levels in Czechia is reflected in the ∑14PAH
model through the inclusion of the HVI predictor variable. In agreement
with these previous studies, atmospheric PAH levels within Czechia
were significantly negatively correlated with HVI, indicating higher
levels at sampling sites located in valleys compared to those located
on hills or mountaintops.
Overall, the model is slightly concentration-dependent and overestimates
∑14PAH at the two sites with the lowest atmospheric levels
(Kleť and Churáňov), although it accurately describes the atmospheric
levels at the two most heavily polluted sites (Prague Radotin and
Věřňovice) (Fig. S12). Overestimation of LUR models for SVOCs has previously
been attributed to the possible removal of emission sources and
the resulting time-dependency of emission estimates (Melymuk et al.,
2013), which may explain the discrepancy at the sites with low atmospheric
PAH levels. However, it is also possible that Kleť and Churáňov
are more significantly affected by the negative correlation between elevation
and ∑14PAH levels than is captured by the HVI, as they are the
highest altitude sites included in this study (1060 m and 1121 m, respectively).
For the remaining sites, there appears to be a tendency for
the model to underestimate concentrations to a greater degree than
those it overestimates, suggesting additional sources of PAHs that are
not reflected in the model. In addition to primary emission sources,
Melymuk et al. (2013) also observed a correlation between atmospheric
PAH levels and proximity to contaminated sites due to high levels of
PAHs in soil which may act as a secondary source to air via volatilization.
National data on PAHs in soil were not available for model development,
thus the influence of soil and other secondary emission sources on atmospheric
∑14PAH levels in Czechia may be partially responsible for
the remaining variance (~30%) not explained by primary emission
sources and HVI in our model.
3.2.2. PCB and DDT models
Sampling site elevation was strongly negatively correlated with atmospheric
levels of PCBs and DDT in Czechia and was the only predictor
variable retained in the final ∑6DDT model (Eq. (2)), and one of two
variables retained in the final ∑7PCB model (Eq. (3)).
log ∑6DDTð Þ ¼ −0:84 log Elevationð Þ þ 1:53
Adjusted R2
¼ 0:45 p < 0:05ð Þ; LOOCV R2
¼ 0:37 p < 0:05ð Þ
h i
ð2Þ
log ∑7PCBð Þ ¼ −0:99 log Elevationð Þ þ 1:26 log SoilPCBð Þ þ 0:67
Adjusted R2
¼ 0:62 p < 0:05ð Þ; LOOCV R2
¼ 0:55 p < 0:05ð Þ
h i
ð3Þ
As previously discussed, elevation is strongly negatively correlated
with anthropogenic activity within the country, with high elevation
sites generally characterized by low population density and limited influence
of industry, agriculture, or traffic. Atmospheric levels of PCBs
and DDT at these sites are likely determined by long-range transport
and more influenced by levels in the atmospheric mixing layer of Central
Europe than local sources (Lammel et al., 2009). This is apparent
from the relative homogeneity in the atmospheric PCB and DDT levels
at the least polluted sites despite large differences in their geographic
locations, suggesting they are capturing the same regional background
level. However, elevation still only explained 47% and 45% of the variability
in atmospheric ∑7PCB and ∑6DDT levels, respectively, suggesting
the influence of local emission sources as well.
Although the production and use of PCBs in Czechia has been banned
since 1984, and despite national efforts to eliminate PCB-containing
devices and waste, PCBs remain in use in thousands of old electrical devices
and building materials, as well as in surface layers of contaminated
industrial sites (Bláha et al., 2017; Holoubek et al., 2006). Releases from
older products and materials remaining in use can still play a significant
role in atmospheric levels; for example, Melymuk et al. (2013) found
that 75% of the variability in urban/suburban atmospheric PCB levels
could be explained by just the mass of PCBs in use, storage, and building
sealants within a 1 km radius around each sampling site. Thus, while
K.B. White, O. Sáňka, L. Melymuk et al. Science of the Total Environment 793 (2021) 148520
6
rural and remote Czech sites are likely more influenced by long-range
transport and background levels, urban and surburban sites with higher
levels such as Brno Kotlarska may still be heavily influenced by nearby
primary emissions or secondary emissions from contaminated soils.
We were able to obtain ∑5PCB levels in soil across the country from
a previous study (Kubošová et al., 2009). When these soil concentrations
were included in the LUR analysis for ∑7PCB as a predictor variable,
they were retained in the final ∑7PCB model in addition to
elevation, and resulted in a substantial increase in the model explained
variance (Eq. (3)). Many SVOCs are highly persistent in soils, which can
act as a significant secondary emission source to the atmosphere if conditions
favour volatilization (Holoubek et al., 2009). Within Central
Europe, soil generally acts as a sink for high molecular weight (highly
chlorinated) PCBs, but as a source for more volatile (less chlorinated)
PCBs, particularly in summer months when temperatures are higher
(Růžičková et al., 2008). However, soil-air transfer is highly sitespecific
and depends on the physicochemical properties of both the
compound and soil, as well the contaminant burden within the soil
and local meteorological conditions (Růžičková et al., 2008). Based on
previous studies of soil contamination across Czechia (Holoubek et al.,
2009; Kubošová et al., 2009; Růžičková et al., 2008) we expect that if
we had spatially distributed records of ∑6DDT in soil, the inclusion of
these soil concentrations into the ∑6DDT model would result in a similar
or greater increase in the explained variance as that observed for the
∑7PCB model, due to the historical application of DDT directly to agricultural
land.
3.2.3. Model limitations
There are several limitations to these models related to the LUR
method itself, as well as its implementation for the compounds selected
in this study. Our study was limited to 29 sampling sites, and previous
studies have shown that LUR models developed for air pollutants
based on a small number of sampling sites have poor predictive
power when validated against external datasets, even if both the adjusted
R2
and LOOCV R2
values are high (Basagaña et al., 2012; Wang
et al., 2012), and suggest LUR models should be developed using at
least 40 (Wang et al., 2012) or even 80 sites (Basagaña et al., 2012). It
should be noted that these authors offered significantly more predictor
variables to their models (76 and 106, respectively, vs. 12 in this study),
and restricting the number of predictor variables offered to LUR models
can improve the predictive power of models developed using a small
number of sampling sites, though some overestimation of R2
is still observed
(Basagaña et al., 2012).
In addition to the small number of sampling sites, the predictive
power of our models is likely also reduced by our decision to develop
models for sum classes of SVOCs rather than individual compounds/
congeners at specific sites as is common in other LUR studies. While
broad LUR models may be more valuable for national policy and regulation,
the influence of specific sources on individual compounds may be
lost or obfuscated. For example, Melymuk et al. (2013) found that the
distribution of metal refining and processing industries explained 60%
of the variance in atmospheric PCB concentrations, but only for triand
tetra-chlorinated congeners, while paint and pigment manufacturing
was associated with other congeners. Similar differences can be observed
in PAHs, with phenanthrene being the dominant compound in
bituminous coal-, lignite-, and wood-burning emissions, while fluoranthene
dominates coke-burning emissions (Table S4). Moreover, the influence
of degradation in ambient air and during sampling can be
important for PAHs under certain conditions (Keyte et al., 2013;
Melymuk et al., 2016), however this was not accounted for in our generalized
model for Σ14PAHs.
Site selection is also an important consideration when applying LUR
models on a large spatial scale, rather than for a single city or urban area
as is more common. To optimize their predictive power, models should
be developed using sites that are representative of the population to
which the model will be applied (Wang et al., 2012). For our nationalscale
models, it was therefore important to include a broad range of
site classifications including remote, rural, suburban, urban, and industrial
sites. Although this was achieved to some extent, approximately
two thirds of the sites used in the development of the models were
background rural sites simply due to these sites being more prevalent
in the MONET network. As a result, our LUR models are likely biased towards
rural sites and may be less accurate for more industrial or impacted
sites. This is further compounded by the lack of primary and
secondary emission source data within the country, as previously
discussed.
3.3. Model application
To demonstrate the ability of LUR modelling to improve understanding
of atmospheric SVOC distributions and optimize design of air monitoring
networks, the two relatively strong models (∑14PAH and
∑7PCB) were applied at a national scale using ArcGIS software to estimate
levels across Czechia (Fig. 3). Due to the model limitations we
chose to depict atmospheric SVOC levels semi-quantitatively (high to
low) to prevent misinterpretation, however, this still provides sufficient
information to identify suspected emission hotspots and source regions.
The higher granularity of the PAH map reflects the dependence of the
model on local topography (represented by the HVI predictor variable),
suggesting that atmospheric distributions of PAHs may be more heterogeneous
than PCBs. This is supported by the fact that ∑14PAH levels include
several highly particle-bound PAHs that have more limited air
transport and thus deposit closer to sources, while PCBs are more volatile
and have greater potential for long-range transport. Furthermore,
the ∑7PCB model suggests that secondary emissions (e.g., soil) may
be the major source of atmospheric PCB levels, while PAHs are still primarily
affected by on-going primary emissions, with greater diversity of
combustion sources.
In both maps there is a tendency towards higher levels in urban regions
compared to rural/background regions, as frequently reported for
these compounds (Cetin et al., 2017; Jamshidi et al., 2007; Melymuk
et al., 2012; Motelay-Massei et al., 2005; Venier et al., 2019). In particular,
the PAH map highlights elevated atmospheric levels in the Ostrava
region, which has been identified as a hotspot of PAH contamination
due to coal mining and heavy industry (Jiřík et al., 2016; Sram et al.,
2013). Industrial emissions in this region are also linked to elevated atmospheric
PCB levels, as observed in Fig. 3. Higher PAH levels are also
observed in the North Bohemia basin, northwest of Prague, potentially
associated with extensive coal (lignite) surface mining in this area.
In addition to the industrial emissions in the Ostrava region, Fig. 3
also identifies large areas of South Moravia (Morava River basin, southeast
of Brno) and North Bohemia (Prague) as having the highest atmospheric
PCB levels across Czechia. In the 1990s, this region of South
Moravia had the highest PCB levels in all environmental matrices across
the country due to the Colorlak paint factory; Colorlak was the primary
consumer of PCBs produced in the former Czechoslovakia and resulted
in PCB-containing paints being heavily used in this region for decades
(Holoubek et al., 2007a). This was also the region most affected by a
major flood event in 1997 that damaged industrial and agricultural
chemical storage facilities and remobilized PCB burdens from sediments
– the largest environmental sink for SVOCs in Czechia (Holoubek et al.,
2007b) – to surface soils and then to air (Holoubek et al., 2007a). Similarly,
a major flood event in 2002 occurred in central Bohemia and led to
significantly increased atmospheric levels of pesticides in the region, including
DDT (Holoubek et al., 2007a). The success of the LUR-based
maps in identifying known/plausible concentration gradients and hotspots
reflects well on their utility for sampling design and source
tracking.
However, the PCB map also shows an area of high predicted air levels
northeast of Prague, which corresponds to a protected sandstone landscape.
The plausibility of this area having high PCB levels is not clear; air
sampling has not been directly performed in the region, but the high
K.B. White, O. Sáňka, L. Melymuk et al. Science of the Total Environment 793 (2021) 148520
7
levels may be an artifact of the soil organic carbon influence on PCB soil
concentrations in the Kubošová et al. (2009) soil model. The cluster
analysis by Kubošová et al. did not distinguish between mountain and
sandstone podzol soils, and it is unclear whether the predicted high
PCB burden of the mountain podzols is also applicable to the sandstone
podzols. This would require further investigation in the form of air and/
or soil sampling. Thus, while there are some clear limitations to the application
of LUR for predicting atmospheric levels, largely related to the
assumptions of the input data, the application of the PAH and PCB
models across the entire country presents strong evidence of their utility
as a tool for better understanding sources and distributions of SVOCs
in air.
4. Conclusions and policy implications
Broad spatial coverage in air monitoring of SVOCs remains challenging,
even with the widespread use of passive air sampling. Significant spatial
heterogeneity can exist in concentrations on a small scale (a few km) causing
gradients in human exposure, but can also provide crucial information
about point sources. LUR modelling is an ideal tool to better understand regional
SVOC levels by relying on high density spatial data (population, industrial
land use, topography) to highlight atmospheric gradients of
SVOCs since such spatial data are more widely available than SVOC concentrations.
We have demonstrated the success of this technique for describing
atmospheric levels of PAHs and PCBs at a national scale for the
Fig. 3. Application of the LUR models across all of Czechia, indicating the estimated gradient of low to high atmospheric levels of A) Σ14PAH, (R2
= 0.68), and B) Σ7PCB, (R2
= 0.62). Cities
with population > 100,000 are indicated on the map.
K.B. White, O. Sáňka, L. Melymuk et al. Science of the Total Environment 793 (2021) 148520
8
first time. However, the regression equations have not been tested with an
independent data set, so we consider these models to be primarily descriptive
rather than predictive. As such, they are useful to identify specific predictor
variables that correlate with atmospheric levels, and thereby
suggest potential important sources of PAHs and PCBs in Czechia.
The PAH model suggests active primary sources, particularly domestic
heating, are the most important driver of PAHs in Czech air; thus, reductions
to emissions from domestic heating (low-emission boilers,
transitioning away from wood and coal for fuel) should reduce their atmospheric
burden. In contrast, the PCB model suggests secondary
sources, particularly emission from soils and sediments, may be a key
driver of levels in Czech air. These sources are much more challenging
to actively reduce and will largely depend on environmental cycling
and the relatively slow process of environmental degradation. This is
consistent with the recently observed increasing environmental halflives
of PCBs and slowly declining/stabilizing atmospheric concentrations
(Graf et al., 2016; Muñoz-Arnanz et al., 2016; Shunthirasingham
et al., 2016), and in contrast to the previously established understanding
(up to ~2010) that primary sources were the main contributor to atmospheric
PCB levels (Melymuk et al., 2012; Robson and Harrad, 2004;
Schuster et al., 2010). While the influence of primary vs. secondary
sources likely differs by region/country, recent active reduction of PCB
stocks in many countries in response to the Stockholm Convention
deadline of 2025 for elimination of in-use PCBs (UNEP, 2017) may
have shifted the dominant sources of PCBs to the atmosphere from primary
to secondary. We also clearly show that high quality surrogate
variables covering the range of possible sources (both primary and secondary)
is a necessary part of successful LUR model development for
SVOCs, and a lack of such data, as in the case of our DDT model, hinders
the development and use of LUR as a technique.
Knowledge of atmospheric SVOC levels is necessary to assess the effectiveness
of actions for reducing SVOCs in the environment, in particular
the global Stockholm Convention on Persistent Organic Pollutants,
and the Convention on Long-Range Transboundary Air Pollution
(CLRTAP) within Europe. To these ends, it is crucial to monitor background
sites for long-term trends without the influence of local point
sources. At the same time, global monitoring needs to be cost-effective
and sustainable for decades which requires minimizing the number of
monitoring sites needed to capture both temporal trends and regional
gradients in concentrations. LUR modelling can help improve selection
of appropriate background sites, and identify areas influenced by
local/point sources to be avoided in monitoring networks. At the national
and local scales, monitoring is a crucial tool for regulators and
policymakers targeting local air quality and requires a complete overview
of influences (hot spots, legacy contamination, urban-rural gradients,
agricultural influences, etc.). Thus, LUR modelling can also help
to improve local, targeted air quality monitoring by pinpointing areas
where such impacts would be expected. In both cases, incorporation
of LUR enhances continuous atmospheric SVOC monitoring, which
should include on-going optimization of network design to improve
our understanding of sources and distributions in the longer term.
CRediT authorship contribution statement
Kevin B. White: Methodology, Validation, Investigation, Formal
analysis, Writing – original draft. Ondřej Sáňka: Methodology, Visualization,
Writing – review & editing. Lisa Melymuk: Conceptualization,
Methodology, Supervision, Writing – review & editing. Petra Přibylová:
Resources, Project administration. Jana Klánová: Funding acquisition,
Writing – review & editing, Supervision.
Declaration of competing interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
Acknowledgements
This research was supported by the RECETOX Research Infrastructure
(Czech Ministry of Education, Youth and Sports: LM2018121), the
ERA PLANET (iGOSP) project (No. 689443) of the EU Horizon 2020
programme, and a grant from Iceland, Liechtenstein and Norway
(EHP-CZ02-OV-1-029-2015). Special thanks to Jiří Kalina for guidance
on statistical analysis and modelling.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.scitotenv.2021.148520.
References
Aas, W., Bohlin-Nizzetto, P., 2018. EMEP/CCC-Report 3/2020-Heavy Metals and POP Measurements,
2018. Kjeller, Norway.
Amini, H., Yunesian, M., Hosseini, V., Schindler, C., Henderson, S.B., Künzli, N., 2017. A systematic
review of land use regression models for volatile organic compounds. Atmos.
Environ. 171, 1–16. https://doi.org/10.1016/j.atmosenv.2017.10.010.
Antweiler, R.C., 2015. Evaluation of statistical treatments of left-censored environmental
data using coincident uncensored data sets. II. Group comparisons. Environ. Sci.
Technol. 49, 13439–13446.
Basagaña, X., Rivera, M., Aguilera, I., Agis, D., Bouso, L., Elosua, R., Foraster, M., de Nazelle,
A., Nieuwenhuijsen, M., Vila, J., Künzli, N., 2012. Effect of the number of measurement
sites on land use regression models in estimating local air pollution. Atmos. Environ.
54, 634–642. https://doi.org/10.1016/j.atmosenv.2012.01.064.
Beelen, R., Hoek, G., Vienneau, D., Eeftens, M., Dimakopoulou, K., Pedeli, X., Tsai, M.Y.,
Künzli, N., Schikowski, T., Marcon, A., Eriksen, K.T., Raaschou-Nielsen, O., Stephanou,
E., Patelarou, E., Lanki, T., Yli-Tuomi, T., Declercq, C., Falq, G., Stempfelet, M., Birk,
M., Cyrys, J., von Klot, S., Nádor, G., Varró, M.J., Dedele, A., Gražulevičiene, R., Mölter,
A., Lindley, S., Madsen, C., Cesaroni, G., Ranzi, A., Badaloni, C., Hoffmann, B.,
Nonnemacher, M., Krämer, U., Kuhlbusch, T., Cirach, M., de Nazelle, A.,
Nieuwenhuijsen, M., Bellander, T., Korek, M., Olsson, D., Strömgren, M., Dons, E.,
Jerrett, M., Fischer, P., Wang, M., Brunekreef, B., de Hoogh, K., 2013. Development of
NO 2 and NO x land use regression models for estimating air pollution exposure in
36 study areas in Europe - the ESCAPE project. Atmos. Environ. 72, 10–23. https://
doi.org/10.1016/j.atmosenv.2013.02.037.
Bláha, K., Borůvková, J., Budňáková, M., Čupr, P., Eyblová, D., Gadas, P., Gruntorád, J.,
Hlavatý, E., Holoubek, I., Hůlek, R., Jarolím, O., Klánová, J., Kolář, J., Krech, L.,
Krejčová, V., Prokeš, R., Přibylová, P., Ptáčková, Ž., Puklová, V., Straková, H., Šebková,
K., Špringar, Z., Štěpánek, P., Udatný, M., Vávra, T., Veselý, I., Vytopilová, M., 2017. Updated
National Implementation Plan for Implementation of Stockholm Convention on
Persistent Organic Pollutants in the Czech Republic for the Period 2018-2023. Prague.
Bohlin-Nizzetto, P., Melymuk, L., White, K.B., Kalina, J., Madadi, V.O., Adu-Kumi, S., Prokeš,
R., Přibylová, P., Klánová, J., 2020. Field- and model-based calibration of polyurethane
foam passive air samplers in different climate regions highlights differences in sampler
uptake performance. Atmos. Environ. 238. https://doi.org/10.1016/j.
atmosenv.2020.117742.
Cetin, B., Ozturk, F., Keles, M., Yurdakul, S., 2017. PAHs and PCBs in an Eastern Mediterranean
megacity, Istanbul: their spatial and temporal distributions, air-soil exchange
and toxicological effects. Environ. Pollut 220, 1322–1332. https://doi.org/10.1016/J.
ENVPOL.2016.11.002.
Eeftens, M., Beelen, R., de Hoogh, K., Bellander, T., Cesaroni, G., Cirach, M., Declercq, C.,
Dėdelė, A., Dons, E., de Nazelle, A., Dimakopoulou, K., Eriksen, K., Falq, G., Fischer, P.,
Galassi, C., Gražulevičienė, R., Heinrich, J., Hoffmann, B., Jerrett, M., Keidel, D., Korek,
M., Lanki, T., Lindley, S., Madsen, C., Mölter, A., Nádor, G., Nieuwenhuijsen, M.,
Nonnemacher, M., Pedeli, X., Raaschou-Nielsen, O., Patelarou, E., Quass, U., Ranzi, A.,
Schindler, C., Stempfelet, M., Stephanou, E., Sugiri, D., Tsai, M.-Y., Yli-Tuomi, T.,
Varró, M.J., Vienneau, D., Klot, S. Von, Wolf, K., Brunekreef, B., Hoek, G., 2012. Development
of land use regression models for PM 2.5 , PM 2.5 absorbance, PM 10 and
PM coarse in 20 European study areas; results of the ESCAPE project. Environ. Sci.
Technol. 46, 11195–11205. https://doi.org/10.1021/es301948k.
Farrar, N.J., Harner, T., Shoeib, M., Sweetman, A.J., Jones, K.C., 2005. Field deployment of
thin film passive air samplers for persistent organic pollutants: a study in the urban
atmospheric boundary layer. Environ. Sci. Technol. 39, 42–48.
Gelaro, R., McCarty, W., Suárez, M.J., Todling, R., Molod, A., Takacs, L., Randles, C.A.,
Darmenov, A., Bosilovich, M.G., Reichle, R., Wargan, K., Coy, L., Cullather, R., Draper,
C., Akella, S., Buchard, V., Conaty, A., da Silva, A.M., Gu, W., Kim, G.K., Koster, R.,
Lucchesi, R., Merkova, D., Nielsen, J.E., Partyka, G., Pawson, S., Putman, W.,
Rienecker, M., Schubert, S.D., Sienkiewicz, M., Zhao, B., 2017. The modern-era retrospective
analysis for research and applications, version 2 (MERRA-2). J. Clim. 30,
5419–5454. https://doi.org/10.1175/JCLI-D-16-0758.1.
Graf, C., Katsoyiannis, A., Jones, K.C., Sweetman, A.J., 2016. The TOMPs ambient air monitoring
network – continuous data on UK air quality for over 20 years. Environ. Pollut.
217, 42–51. https://doi.org/10.1016/j.envpol.2016.01.033.
Gulliver, J., Morley, D., Dunster, C., McCrea, A., van Nunen, E., Tsai, M.Y., Probst-Hensch, N.,
Eeftens, M., Imboden, M., Ducret-Stich, R., Naccarati, A., Galassi, C., Ranzi, A.,
Nieuwenhuijsen, M., Curto, A., Donaire-Gonzalez, D., Cirach, M., Vermeulen, R.,
Vineis, P., Hoek, G., Kelly, F.J., 2018. Land use regression models for the oxidative
K.B. White, O. Sáňka, L. Melymuk et al. Science of the Total Environment 793 (2021) 148520
9
potential of fine particles (PM2.5) in five European areas. Environ. Res. 160, 247–255.
https://doi.org/10.1016/j.envres.2017.10.002.
Herkert, N.J., Spak, S.N., Smith, A., Schuster, J.K., Harner, T., Martinez, A., Hornbuckle, K.C.,
2018. Calibration and evaluation of PUF-PAS sampling rates across the Global Atmospheric
Passive Sampling (GAPS) network. Environ Sci Process Impacts 20, 210–219.
https://doi.org/10.1039/c7em00360a.
Hoek, G., Beelen, R., de Hoogh, K., Vienneau, D., Gulliver, J., Fischer, P., Briggs, D.J., 2008. A
review of land-use regression models to assess spatial variation of outdoor air pollution.
Atmos. Environ. 42, 7561–7578.
Holoubek, I., Adamec, V., Bartoš, M., Bláha, K., Bláha, L., Budňáková, M., Černá, M., Čupr, P.,
Demnerová, K., Drápal, J., Hajšlová, J., Hanzálková, M., Holoubková, I., Hrabětová, S.,
Jech, L., Klánová, J., Kohoutek, J., Kužílek, V., Machálek, P., Matějů, V., Matoušek, J.,
Matoušek, M., Mestřík, V., Novák, J., Ocelka, T., Pekárek, V., Petrlík, J., Petira, O.,
Provazník, K., Punčochář, M., Rieder, M., Ruprich, J., Sáňka, M., Tomaniová, M.,
Vácha, R., Volka, K., Zbíral, J., 2006. The National Implementation Plan for Implementation
of the Stockholm Convention in the Czech Republic. Brno.
Holoubek, I., Klánová, J., Jarkovský, J., Kohoutek, J., Jarkovský, J., Kohoutek, J., Jarkovský, J.,
Kohoutek, J., 2007a. Trends in background levels of persistent organic pollutants at
Kosetice observatory, Czech Republic. Part I. Ambient air and wet deposition 1996-
2005. J. Environ. Monit. 9, 557–563. https://doi.org/10.1039/b700750g.
Holoubek, I., Klánová, J., Jarkovský, J., Kubík, V., Helešic, J., 2007b. Trends in background
levels of persistent organic pollutants at Kosetice observatory, Czech Republic. Part
II. Aquatic and terrestrial environments 1996-2005. J. Environ. Monit. 9, 564–571.
https://doi.org/10.1039/b701096f.
Holoubek, I., Dusek, L., Sáňka, M., Hofman, J., Cupr, P., Jarkovský, J., Zbíral, J., Klánová, J.,
2009. Soil burdens of persistent organic pollutants–their levels, fate and risk. Part I.
Variation of concentration ranges according to different soil uses and locations. Environ.
Pollut. 157, 3207–3217. https://doi.org/10.1016/j.envpol.2009.05.031.
Holt, E., Bohlin-Nizzetto, P., Borůvková, J., Harner, T., Kalina, J., Melymuk, L., Klánová, J.,
2017. Using long-term air monitoring of semi-volatile organic compounds to evaluate
the uncertainty in polyurethane-disk passive sampler-derived air concentrations. Environ.
Pollut. 220, 1100–1111. https://doi.org/10.1016/j.envpol.2016.11.030.
de Hoogh, K., Wang, M., Adam, M., Badaloni, C., Beelen, R., Birk, M., Cesaroni, G., Cirach, M.,
Declercq, C., Dědelě, A., Dons, E., De Nazelle, A., Eeftens, M., Eriksen, K.T., Eriksson, C.,
Fischer, P., Gražulevičieně, R., Gryparis, A., Hoffmann, B., Jerrett, M., Katsouyanni, K.,
Iakovides, M., Lanki, T., Lindley, S., Madsen, C., Mölter, A., Mosler, G., Nádor, G.,
Nieuwenhuijsen, M., Pershagen, G., Peters, A., Phuleria, H., Probst-Hensch, N.,
Raaschou-Nielsen, O., Quass, U., Ranzi, A., Stephanou, E., Sugiri, D., Schwarze, P.,
Tsai, M.Y., Yli-Tuomi, T., Varró, M.J., Vienneau, D., Weinmayr, G., Brunekreef, B.,
Hoek, G., 2013. Development of land use regression models for particle composition
in twenty study areas in Europe. Environ. Sci. Technol. 47, 5778–5786. https://doi.
org/10.1021/es400156t.
de Hoogh, K., Korek, M., Vienneau, D., Keuken, M., Kukkonen, J., Nieuwenhuijsen, M.J.,
Badaloni, C., Beelen, R., Bolignano, A., Cesaroni, G., Pradas, M.C., Cyrys, J., Douros, J.,
Eeftens, M., Forastiere, F., Forsberg, B., Fuks, K., Gehring, U., Gryparis, A., Gulliver, J.,
Hansell, A.L., Hoffmann, B., Johansson, C., Jonkers, S., Kangas, L., Katsouyanni, K.,
Künzli, N., Lanki, T., Memmesheimer, M., Moussiopoulos, N., Modig, L., Pershagen,
G., Probst-Hensch, N., Schindler, C., Schikowski, T., Sugiri, D., Teixidó, O., Tsai, M.Y.,
Yli-Tuomi, T., Brunekreef, B., Hoek, G., Bellander, T., 2014. Comparing land use regression
and dispersion modelling to assess residential exposure to ambient air pollution
for epidemiological studies. Environ. Int. 73, 382–392. https://doi.org/10.1016/j.
envint.2014.08.011.
de Hoogh, K., Gulliver, J., Donkelaar, A. van, Martin, R.V., Marshall, J.D., Bechle, M.J.,
Cesaroni, G., Pradas, M.C., Dedele, A., Eeftens, M., Forsberg, B., Galassi, C., Heinrich,
J., Hoffmann, B., Jacquemin, B., Katsouyanni, K., Korek, M., Künzli, N., Lindley, S.J.,
Lepeule, J., Meleux, F., de Nazelle, A., Nieuwenhuijsen, M., Nystad, W., RaaschouNielsen,
O., Peters, A., Peuch, V.H., Rouil, L., Udvardy, O., Slama, R., Stempfelet, M.,
Stephanou, E.G., Tsai, M.Y., Yli-Tuomi, T., Weinmayr, G., Brunekreef, B., Vienneau, D.,
Hoek, G., 2016. Development of West-European PM 2.5 and NO 2 land use regression
models incorporating satellite-derived and chemical transport modelling data. Environ.
Res. 151, 1–10. https://doi.org/10.1016/j.envres.2016.07.005.
de Hoogh, K., Chen, J., Gulliver, J., Hoffmann, B., Hertel, O., Ketzel, M., Bauwelinck, M., van
Donkelaar, A., Hvidtfeldt, U.A., Katsouyanni, K., Klompmaker, J., Martin, R.V., Samoli,
E., Schwartz, P.E., Stafoggia, M., Bellander, T., Strak, M., Wolf, K., Vienneau, D.,
Brunekreef, B., Hoek, G., 2018. Spatial PM 2.5 , NO2 , O3 and BC models for Western
Europe – evaluation of spatiotemporal stability. Environ. Int 120, 81–92. https://doi.
org/10.1016/j.envint.2018.07.036.
Hulek, R., Boruvkova, J., Grehugor, J., Kalina, J., Bednářová, Z., Sebkova, K., Melkes, O., Salko,
M., Novak, R., Jarkovský, J., Dusek, L., Klanova, J., 2014. Global Monitoring Plan of the
Stockholm Convention on Persistent Organic Pollutants: Visualisation and on-Line
Analysis of Global Levels of Chemicals in Air, Water, Breast Milk and Blood [WWW
Document]. Masaryk Univ http://www.pops-gmp.org/visualization-2014 (accessed
6.17.20).
Jamshidi, A., Hunter, S., Hazrati, S., Harrad, S.J., Hazrati, S., Harrad, S.J., 2007. Concentrations
and chiral signatures of polychlorinated biphenyls in outdoor and indoor air
and soil in a major U.K. conurbation. Environ. Sci. Technol. 41, 2153–2158.
Jedynska, A., Hoek, G., Wang, M., Eeftens, M., Cyrys, J., Keuken, M., Ampe, C., Beelen, R.,
Cesaroni, G., Forastiere, F., Cirach, M., de Hoogh, K., De Nazelle, A., Nystad, W.,
Declercq, C., Eriksen, K.T., Dimakopoulou, K., Lanki, T., Meliefste, K.,
Nieuwenhuijsen, M.J., Yli-Tuomi, T., Raaschou-Nielsen, O., Brunekreef, B., Kooter,
I.M., 2014. Development of land use regression models for elemental, organic carbon,
PAH, and Hopanes/Steranes in 10 ESCAPE/TRANSPHORM European study areas. Environ.
Sci. Technol. 48, 14435–14444. https://doi.org/10.1021/es502568z.
Jedynska, A., Hoek, G., Wang, M., Yang, A., Eeftens, M., Cyrys, J., Keuken, M., Ampe, C.,
Beelen, R., Cesaroni, G., Forastiere, F., Cirach, M., de Hoogh, K., De Nazelle, A.,
Nystad, W., Akhlaghi, H.M., Declercq, C., Stempfelet, M., Eriksen, K.T.,
Dimakopoulou, K., Lanki, T., Meliefste, K., Nieuwenhuijsen, M., Yli-Tuomi, T.,
Raaschou-Nielsen, O., Janssen, N.A.H., Brunekreef, B., Kooter, I.M., 2017. Spatial variations
and development of land use regression models of oxidative potential in ten
European study areas. Atmos. Environ. 150, 24–32. https://doi.org/10.1016/j.
atmosenv.2016.11.029.
Jiřík, V., Machaczka, O., Miturová, H., Tomášek, I., Šlachtová, H., Janoutová, J., Velická, H.,
Janout, V., 2016. Air pollution and potential health risk in Ostrava region – a review.
Cent. Eur. J. Public Health 24, S4–S17. https://doi.org/10.21101/cejph.a4533.
Kalberer, M., Henne, S., Prevot, A.S.H., Steinbacher, M., 2004. Vertical transport and degradation
of polycyclic aromatic hydrocarbons in an Alpine Valley. Atmos. Environ. 38,
6447–6456. https://doi.org/10.1016/j.atmosenv.2004.06.021.
Kalina, J., Scheringer, M., Borůvková, J., Kukucka, P., Přibylová, P., Sáňka, O., Melymuk, L.,
Vana, M., Klánová, J., 2018. Characterizing spatial diversity of passive sampling sites
for measuring levels and trends of semivolatile organic chemicals. Environ. Sci.
Technol. 52, 10599–10608. https://doi.org/10.1021/acs.est.8b03414.
Keyte, I.J., Harrison, R.M., Lammel, G., 2013. Chemical reactivity and long-range transport
potential of polycyclic aromatic hydrocarbons–a review. Chem. Soc. Rev. 42,
9333–9391.
Kubošová, K., Komprda, J., Jarkovský, J., Sáňka, M., Hájek, O., Dušek, L., Holoubek, I.,
Klánová, J., 2009. Spatially resolved distribution models of POP concentrations in
soil: a stochastic approach using regression trees. Environ. Sci. Technol. 43,
9230–9236. https://doi.org/10.1021/es902076y.
Lammel, G., Klánová, J., Kohoutek, J., Prokeš, R., Ries, L., Stohl, A., 2009. Observation and
origin of organochlorine compounds and polycyclic aromatic hydrocarbons in the
free troposphere over central Europe. Environ. Pollut. 157, 3264–3271. https://doi.
org/10.1016/j.envpol.2009.05.028.
Larkin, A., Geddes, J.A., Martin, R.V., Xiao, Q., Liu, Y., Marshall, J.D., Brauer, M., Hystad, P.,
2017. Global land use regression model for nitrogen dioxide air pollution. Environ.
Sci. Technol. 51, 6957–6964. https://doi.org/10.1021/acs.est.7b01148.
Lu, M., Soenario, I., Helbich, M., Schmitz, O., Hoek, G., van der Molen, M., Karssenberg, D.,
2020. Land use regression models revealing spatiotemporal co-variation in NO2, NO,
and O3 in the Netherlands. Atmos. Environ. 223, 117238. https://doi.org/10.1016/j.
atmosenv.2019.117238.
Masri, S., Li, L., Dang, A., Chung, J.H., Chen, J.C., Fan, Z.H. (Tina), Wu, J., 2018. Source characterization
and exposure modeling of gas-phase polycyclic aromatic hydrocarbon
(PAH) concentrations in Southern California. Atmos. Environ. 177, 175–186. https://
doi.org/10.1016/j.atmosenv.2018.01.014.
Melymuk, L., Robson, M., Helm, P.A., Diamond, M.L., 2012. PCBs, PBDEs, and PAHs in Toronto
air: spatial and seasonal trends and implications for contaminant transport.
Sci. Total Environ. 429, 272–280. https://doi.org/10.1016/j.scitotenv.2012.04.022.
Melymuk, L., Robson, M., Helm, P.A., Diamond, M.L., 2013. Application of land use regression
to identify sources and assess spatial variation in urban SVOC concentrations. Environ.
Sci. Technol. 47, 1887–1895. https://doi.org/10.1021/es3043609.
Melymuk, L., Bohlin-Nizzetto, P., Prokeš, R., Kukučka, P., Klánová, J., 2016. Sampling artifacts
in active air sampling of semivolatile organic contaminants: comparing theoretical
and measured artifacts and evaluating implications for monitoring networks.
Environ. Pollut. 217, 97–106. https://doi.org/10.1016/j.envpol.2015.12.015.
Motelay-Massei, A., Harner, T., Shoeib, M., Diamond, M.L., Stern, G.A., Rosenberg, B., 2005.
Using passive air samplers to assess urban-rural trends for persistent organic pollutants
and polycyclic aromatic hydrocarbons. 2. Seasonal trends for PAHs, PCBs, and organochlorine
pesticides. Environ. Sci. Technol. 39, 5763–5773.
Muñoz-Arnanz, J., Roscales, J.L., Ros, M., Vicente, A., Jiménez, B., 2016. Towards the implementation
of the Stockholm Convention in Spain: five-year monitoring (2008–2013)
of POPs in air based on passive sampling. Environ. Pollut. 217, 107–113. https://doi.
org/10.1016/j.envpol.2016.01.052.
Noth, E.M., Hammond, S.K., Biging, G.S., Tager, I.B., 2011. A spatial-temporal regression
model to predict daily outdoor residential PAH concentrations in an epidemiologic
study in Fresno, CA. Atmos. Environ. 45, 2394–2403. https://doi.org/10.1016/j.
atmosenv.2011.02.014.
Noth, E.M., Lurmann, F., Northcross, A., Perrino, C., Vaughn, D., Hammond, S.K., 2016. Spatial
and temporal distribution of polycyclic aromatic hydrocarbons and elemental
carbon in Bakersfield, California. Air Qual. Atmos. Health 9, 899–908. https://doi.
org/10.1007/s11869-016-0399-y.
Polidori, A., Kwon, J., Turpin, B.J., Weisel, C., 2010. Source proximity and residential outdoor
concentrations of PM 2.5, OC, EC, and PAHs. J. Expo. Sci. Environ. Epidemiol.
20, 457–468. https://doi.org/10.1038/jes.2009.39.
Přibylová, P., Kares, R., Borůvková, J., Cupr, P., Prokeš, R., Kohoutek, J., Holoubek, I.,
Klánová, J., 2012. Levels of persistent organic pollutants and polycyclic aromatic hydrocarbons
in ambient air of Central and Eastern Europe. Atmos. Pollut. Res. 3,
494–505.
Robson, M., Harrad, S.J., 2004. Chiral PCB signatures in air and soil: implications for atmospheric
source apportionment. Environ. Sci. Technol. 38, 1662–1666.
Růžičková, P., Klánová, J., Cupr, P., Lammel, G., Holoubek, I., 2008. An assessment of air-soil
exchange of polychlorinated biphenyls and organochlorine pesticides across central
and southern Europe. Environ. Sci. Technol. 42, 179–185.
Sáňka, O., Melymuk, L., Čupr, P., Dvorská, A., Klánová, J., 2014. Dispersion modeling of selected
PAHs in urban air: a new approach combining dispersion model with GIS and
passive air sampling. Atmos. Environ. 96, 88–95. https://doi.org/10.1016/j.
atmosenv.2014.07.002.
Schuster, J.K., Gioia, R., Sweetman, A.J., Jones, K.C., 2010. Temporal trends and controlling
factors for polychlorinated biphenyls in the UK atmosphere (1991-2008). Environ.
Sci. Technol. 44, 8068–8074. https://doi.org/10.1021/es102134d.
Shunthirasingham, C., Gawor, A., Hung, H., Brice, K.A., Su, K., Alexandrou, N., DryfhoutClark,
H., Backus, S., Sverko, E., Shin, C., Park, R., Noronha, R., 2016. Atmospheric concentrations
and loadings of organochlorine pesticides and polychlorinated biphenyls
K.B. White, O. Sáňka, L. Melymuk et al. Science of the Total Environment 793 (2021) 148520
10
in the Canadian Great Lakes Basin (GLB): spatial and temporal analysis (1992–2012).
Environ. Pollut. 217, 124–133. https://doi.org/10.1016/j.envpol.2016.01.039.
Sram, R.J., Dostal, M., Libalova, H., Rossner, P., Rossnerova, A., Svecova, V., Topinka, J.,
Bartonova, A., 2013. The European hot spot of B[a]P and PM 2.5 exposure—the
Ostrava Region, Czech Republic: health research results. ISRN Public Health 2013,
1–12. https://doi.org/10.1155/2013/416701.
Tao, S., Wang, Y., Wu, S., Liu, S., Dou, H., Liu, Y., Lang, C., Hu, F., Xing, B., 2007. Vertical distribution
of polycyclic aromatic hydrocarbons in atmospheric boundary layer of Beijing
in winter. Atmos. Environ. 41, 9594–9602. https://doi.org/10.1016/j.
atmosenv.2007.08.026.
Tørseth, K., Aas, W., Breivik, K., Fjeraa, a.M., Fiebig, M., Hjellbrekke, A.-G., Lund Myhre, C.,
Solberg, S., Yttri, K.E., 2012. Introduction to the European Monitoring and Evaluation
Programme (EMEP) and observed atmospheric composition change during 1972-
2009. Atmos. Chem. Phys. 12, 5447–5481. https://doi.org/10.5194/acp-12-5447-
2012.
UNEP, 2017. Stockholm Convention on Persistent Organic Pollutants (POPs).
Van Nunen, E., Vermeulen, R., Tsai, M.Y., Probst-Hensch, N., Ineichen, A., Davey, M.,
Imboden, M., Ducret-Stich, R., Naccarati, A., Raffaele, D., Ranzi, A., Ivaldi, C., Galassi,
C., Nieuwenhuijsen, M., Curto, A., Donaire-Gonzalez, D., Cirach, M., Chatzi, L.,
Kampouri, M., Vlaanderen, J., Meliefste, K., Buijtenhuijs, D., Brunekreef, B., Morley,
D., Vineis, P., Gulliver, J., Hoek, G., 2017. Land use regression models for ultrafine particles
in six European areas. Environ. Sci. Technol. 51, 3336–3345. https://doi.org/
10.1021/acs.est.6b05920.
Venier, M., Salamova, A., Hites, R.A., 2019. How to distinguish urban vs. agricultural
sources of persistent organic pollutants? Curr. Opin. Environ. Sci. Heal 8, 23–28.
https://doi.org/10.1016/J.COESH.2019.01.005.
Vienneau, D., De Hoogh, K., Bechle, M.J., Beelen, R., Van Donkelaar, A., Martin, R.V., Millet,
D.B., Hoek, G., Marshall, J.D., 2013. Western european land use regression incorporating
satellite- and ground-based measurements of NO2 and PM10. Environ. Sci.
Technol. 47, 13555–13564. https://doi.org/10.1021/es403089q.
Vizcaino, P., Lavalle, C., 2018. Development of European NO2 Land Use Regression Model
for present and future exposure assessment: implications for policy analysis. Environ.
Pollut. 240, 140–154. https://doi.org/10.1016/j.envpol.2018.03.075.
Wang, M., Beelen, R., Eeftens, M., Meliefste, K., Hoek, G., Brunekreef, B., 2012. Systematic
evaluation of land use regression models for NO2. Environ. Sci. Technol. 46,
4481–4489. https://doi.org/10.1021/es204183v.
Wania, F., Shunthirasingham, C., 2020. Passive air sampling for semi-volatile organic
chemicals. Environ Sci Process Impacts 22, 1925–2002. https://doi.org/10.1039/
D0EM00194E.
White, K.B., Kalina, J., Scheringer, M., Přibylová, P., Kukučka, P., Kohoutek, J., Prokeš, R.,
Klánová, J., 2020. Temporal trends of persistent organic pollutants across Africa
after a decade of MONET passive air sampling. Environ. Sci. Technol. https://doi.org/
10.1021/acs.est.0c03575.
Wu, S.P., Tao, S., Liu, W.X., 2006. Particle size distributions of polycyclic aromatic hydrocarbons
in rural and urban atmosphere of Tianjin, China. Chemosphere 62,
357–367. https://doi.org/10.1016/j.chemosphere.2005.04.101.
K.B. White, O. Sáňka, L. Melymuk et al. Science of the Total Environment 793 (2021) 148520
11
Melymuk, Lisa, Pernilla Bohlin-Nizzetto, Ondřej Sáňka, Karla Pozo, and Jana Klánová. 2014.
"Current Challenges in Air Sampling of Semivolatile Organic Contaminants: Sampling
Artifacts and Their Influence on Data Comparability." Environmental Science & Technology
48 (24): 14077-91. https://doi.org/10.1021/es502164r
Current Challenges in Air Sampling of Semivolatile Organic
Contaminants: Sampling Artifacts and Their Influence on Data
Comparability
Lisa Melymuk,*,†
Pernilla Bohlin,*,†
Ondřej Sáňka,†
Karla Pozo,†
and Jana Klánová†
†
Research Centre for Toxic Compounds in the Environment (RECETOX), Faculty of Science, Masaryk University, Kamenice 5/753,
Pavilon A29, Brno 62500, Czech Republic
*S Supporting Information
ABSTRACT: With current science and policy needs, more attention
is being given to expanding and improving air sampling of semivolatile
organic contaminants (SVOCs). However, a wide range of techniques
and configurations are currently used (active and passive samplers,
different deployment times, different sorbents, etc.) and as the SVOC
community looks to assess air measurements on a global scale,
questions of comparability arise. We review current air sampling
techniques, with a focus on sampling artifacts that can lead to
uncertainties or biases in reported concentrations, in particular
breakthrough, degradation, meteorological influences, and assumptions
regarding passive sampling. From this assessment, we estimate
the bias introduced for SVOC concentrations from all factors. Due to
the effects of breakthrough, degradation, particle fractions and sampler
uptake periods, some current passive and active sampler configurations
may underestimate certain SVOCs by 30−95%. We then recommend future study design, appropriateness of sampler
types for different study goals, and finally, how the SVOC community should move forward in both research and monitoring to
best achieve comparability and consistency in air measurements.
■ INTRODUCTION AND BACKGROUND
Recent scientific and policy discussion has addressed air
sampling of semivolatile organic compounds (SVOCs)/
persistent organic pollutants (POPs), particularly regarding
long-term monitoring networks and the data needs of the
SVOC community.1−3
SVOCs are organic compounds with
vapor pressures typically between ∼1 and 10−10
Pa. Many of the
SVOCs are classified as POPs, and are subject to international
agreements and conventions such as the Aarhus protocol on
POPs (1998) under the Convention on Long-Range Transboundary
Air Pollution (LRTAP, 1979) (http://www.unece.
org/env/lrtap/) and the Stockholm Convention (SC) on POPs
(2001/2004) (www.pops.int). These conventions aim to
eliminate or restrict the production and emission of a range
of SVOCs, notably organochlorine pesticides (OCPs),
polychlorinated biphenyls (PCBs), polychlorinated dibenzo-pdioxins
and furans (PCDD/Fs), polybrominated diphenyl
ethers (PBDEs), polycyclic aromatic hydrocarbons (PAHs),
per- and polyfluorinated compounds (PFCs). To provide
information on long-term trends and effectiveness of regulatory
actions, air monitoring strategies have been established,1
including the European Monitoring and Evaluation Programme
(EMEP) for LRTAP compounds (1999) and the Global
Monitoring Plan (GMP) for SC compounds (2008).1
Monitoring networks require appropriate sampling strategies
to provide reliable and broadly comparable air data and current
regional monitoring networks, with differences in sampling
methodology and procedures, cannot adequately provide the
data needed for global assessments, such as the GMP.1
Furthermore, scientific concern is also growing regarding
currently unregulated SVOCs such as endocrine disrupting
compounds (EDCs).4
Beyond the policy demands of international conventions, the
SVOC scientific community also requires air sampling to
understand the fate and transport of SVOCs and to provide
necessary information for scientific models and human
exposure assessments.2
This type of data is often produced
through smaller-scale case studies or local- or regional-scale
monitoring networks. While less standardized than long-term
monitoring networks, some very critical knowledge of SVOCs
has come from case studies, as they allow specific questions to
be addressed in a focused manner. In particular, they provide a
platform for testing samplers (e.g., polyurethane foam (PUF)
passive air samplers,5
XAD-passive air samplers6
) and provide
data in vulnerable areas that are not covered by long-term
monitoring networks (e.g., e-waste sites in China7
or Africa,
intensive agricultural regions in India or South America8−10
).
Received: May 2, 2014
Revised: September 23, 2014
Accepted: October 20, 2014
Published: October 20, 2014
Critical Review
pubs.acs.org/est
© 2014 American Chemical Society 14077 dx.doi.org/10.1021/es502164r | Environ. Sci. Technol. 2014, 48, 14077−14091
DownloadedviaMASARYKUNIVonOctober8,2024at10:13:32(UTC).
Seehttps://pubs.acs.org/sharingguidelinesforoptionsonhowtolegitimatelysharepublishedarticles.
Together, monitoring networks and case studies provide our
information on the global distribution of SVOCs, their
atmospheric transport, fate, and relationship with sources.
However, there are still crucial gaps and biases in this
information which may lead to uncertainties in comparability
of data, namely spatial distribution and temporal coverage of
measurements, and lack of sampling standardization within and
between networks/case-studies. Problems related to spatial
distributions of established networks are discussed by Hung et
al.2
and Klánová and Harner.1
The uneven spatial distribution is
somewhat improved when one considers the contribution of
case studies (Supporting Information (SI) Figure S1b) and
new/planned monitoring networks in Australia, Southeast Asia,
and South America,2
but the issue of data comparability
remains. Additionally, there are discrepancies in the spatial
distribution of measurements and the spatial scale of
concentration variations; SVOCs vary on small scales11−13
and current sampling distributions infrequently address this.
There are also discrepancies in the frequency of sampling at
each site, with some regions having repeated continuous
monitoring capturing seasonal trends and short-term variability,
and others with only one or two reported measurements in the
past ten years (Figure 1). Although individual monitoring
networks are effective at accomplishing their respective goals
(e.g., providing data for modeling exercises, or broad spatial
coverage) the different network structures may lead to
problems with the intercomparability of scientific results.
With increasing international collaboration and data sharing
(e.g., the GMP) the question of inter-network comparability is
brought to the forefront.1
There are a wide range of techniques
used across different networks and case studies, including active
and passive air samplers, and differences in sampling volumes,
sorbents, sampling length and sampling frequencies. There has
never been an agreed mandate to establish standardized
procedures within and between SVOC networks and there is
a lack of quality standard operating procedures for SVOC
sampling. While some studies have assessed comparability
between samplers or sampling networks,14−21
efforts never
went beyond these to address the potential implications. Most
current discussion focuses on the problems related to spatial
distributions, as well as analytical quality assurance and quality
control (e.g., interlaboratory or inter-network compari-
sons14,16,17,22,23
), but little attention is given to sampling
artifacts or spatial intercomparability. Herein, we review and
discuss influences on the most frequently used SVOC air
sampling methods/techniques in both established networks and
case studies and analyze how they impact measurements and
thus data comparability. We then make recommendations on
what sampling techniques are appropriate for particular study
goals, how to enhance data comparability, and what is needed
to improve SVOC air measurements.
■ SAMPLING TECHNIQUES
Active Air Sampling Methods. Active air samplers (AAS)
are currently perceived as the most accurate method of
obtaining SVOC air concentrations as they accumulate both gas
and particle phase compounds under a controlled flow. Despite
identified sampling artifacts,24
this technique is almost
unchanged over the past 40 years.25−27
AAS are commonly classified as either high volume or low
volume samplers, with the main differences being the flow rate
and sample collection time. In high volume samplers, flow rates
are typically 15−80 m3
/hour16,28,29
yielding total sample
volumes of >400 m3
, although there is a large variation in
what is considered a high volume sample. For example, the
Integrated Atmospheric Deposition Network (IADN) typically
collects ∼820 m3
of air at American sites, and ∼350 m3
at
Canadian sites,16
while some Arctic sampling networks collect
up to 13000 m3
.30
Low volume air samplers typically have flow
rates of <3 m3
/hour.15,31−34
This often results in small sample
volumes (e.g., <200 m3
), but low volume samplers can also be
used for longer deployment times, for example, 7−14 days of
Figure 1. Global distribution of PBDE measurements in air (active and passive) from 2003 to 2013, as an example of the unbalanced spatial
distribution of SVOC sampling. Shading is used to aid in visualization of aggregated sampling distribution and frequency. Darker orange indicates a
high density of sampling (e.g., regular sampling over many years), while the lightest orange indicates collection of a single event sample (e.g., ship
cruises). Dots reflect global population density as of 2000 (http://sedac.ciesin.columbia.edu/gpw).
Environmental Science & Technology Critical Review
dx.doi.org/10.1021/es502164r | Environ. Sci. Technol. 2014, 48, 14077−1409114078
continuous sampling resulting in air volumes of 500−1000
m3
.31−33
Low volume sampling is rarely used in long-term
sampling networks; only the Coordinated Atmospheric
Monitoring Programme of OSPAR and EMEP use low volume
samplers at selected stations.31,35
Environmental parameters and SVOC concentrations vary
widely, for example, very low concentrations in remote/polar
regions or high temperatures in tropical regions, and thus it can
be problematic to use identical sampling configurations in all
locations. The use of different sampling configurations results
in different sampling artifacts. Sampling events can only be
compared if a sufficient measure of accuracy has been reached.
Current sampling procedures/protocols have wide variations in
sampled volumes that may not always reflect the best choice for
the temperature and concentrations in a region, and rather may
simply rely on established sampling protocols not tailored to
specific sampling conditions (Figure 2). While these choices
allow individual networks to have long-term temporal data
continuity, they may reduce the accuracy of results and
comparability between networks or studies. A greater analysis
of the influences of sample volume and sampling media is given
in the following sections.
Passive Air Sampling Methods. Passive air sampling
techniques for SVOCs were introduced in the end of 1990s to
simplify and reduce the costs for air monitoring of SVOCs.
Unlike AAS, passive air samplers (PAS) do not need electricity;
instead chemicals are trapped by diffusive uptake to a sorbent
material. PAS have enabled broader monitoring network
distribution and, as a result, increased the spatial breadth of
SVOC measurements.1,36
Several types of PAS have been
evaluated and used during the last 10−15 years, and a few of
them have received greater attention and been implemented in
monitoring networks (Table 1). Currently, the disk-shaped
PUF−PAS,5
and the XAD-resin based PAS6
are most used in
monitoring networks. Recently, an XAD sorbent-impregnated
PUF (SIP) PAS has also been implemented in the Global
Atmospheric Passive Sampling network (GAPS).37
Other PAS
samplers have been used (e.g., POGs,38
SPMDs39
), and new
techniques are under development,40
however these are largely
used in case studies. The available PAS differ in characteristics
Figure 2. Distribution of active air sampling volumes used in measurements of PBDEs from 2003 to 2013. Studies that did not report a sample
volume are excluded. There is a general trend of larger volumes collected in remote areas, particularly the Arctic, however there is also significant
variability in air volumes that is not directly related to the expected concentrations at a given site and this variability may affect what is reported as
concentrations in a given location (e.g., occurrence of either breakthrough or lack of detection).
Table 1. Common Active and Passive Sampling Techniques for SVOCs
sampler high volume sampler low volume sampler double-bowl PUF PAS XAD PAS SIP PAS
type of sampler active active passive passive passive
typical sampling
volume (m3
)
600 m3
100 m3
80−400 m3
100−200 m3
100−500 m3
typical sampling
rate/flow rate
15−80 m3
/hour <3 m3
/hour 0.3−20 m3
/day 0.4−0.8 m3
/
day
4−12 m3
/day
sampling time hours to month days to weeks weeks to months months to
year
weeks to months
gas-phase sorbent PUF, XAD or combination
of PUF, XAD
PUF, XAD or
combination of
PUF, XAD
PUF XAD XAD-impregnated PUF
particle-phase
sampling
GFF or QFF GFF or QFF debatable. inconsistent. with lower
sampling rate than gas-phase (10%).
no debatable. inconsistent. with lower
sampling rate than gas-phase (10%).
use in long-term
monitoring
networks?
yes (inter alia EMEP, IADN,
AMAP, MONAIRNET)
rarely (selected
OSPAR stations)
yes (GAPS, MONET) yes (GAPS) yes (GAPS)
Environmental Science & Technology Critical Review
dx.doi.org/10.1021/es502164r | Environ. Sci. Technol. 2014, 48, 14077−1409114079
and applicability, which affects data comparability, as described
below. For an in-depth review of PAS theory and application
we recommend dedicated articles on theoretical framework,41
calibration,42
indoor use,43
new developments and techni-
ques.40,44
■ SAMPLING INFLUENCES
Below, we describe and discuss some of the major sampling
influences for both AAS and PAS. All air sampling techniques
have biases that may affect measurements and thereby also the
comparability of results between different sampling events. We
consider: (i) is the sampling technique accurate−does it
provide the real picture of atmospheric levels of the target
contaminants, (ii) is the sampling technique robust−how
sensitive is it to environmental conditions, and (iii) are data
from different techniques comparable−do we get the same
number when using different sampling techniques and different
sampling approaches? A focus is often on analytical procedures
and intercomparisons of chemical analyses;17,23
comparisons of
sampling procedures/methodologies/techniques are less common.
Yet, given the range of sampling techniques, the
parameters of sample collection (e.g., sorbent type, sample
volumes) should be carefully considered, as they have the
potential to introduce significant biases. A summary of the
advantages and disadvantages of individual sampler configurations
is given in SI Table S1.
Sampling InfluencesActive Air Sampling. Sampling
Head. The most commonly used sampling heads in AAS are
the TSP (total suspended particulates) and PM10 (particles
with aerodynamic diameter <10 μm). TSP theoretically
includes all airborne particulate matter, although in practice,
the upper particle size cutoff of a TSP sampling head is between
50 and 100 μm, dependent on flow rate and ambient
conditions.45
As a result of new legislations within the
European Union (EU), recently PM2.5 (particles <2.5 μm)
has also been suggested as a sampling threshold for monitoring
networks. Different sampling heads and a specific particle cutpoint
can be useful, as particles <10 μm are most important for
atmospheric chemistry and physics46
while particles <2.5 μm
are most relevant for human health and risk assessment
estimates,47
but may result in lower data comparability. The
choice of sampling head should not result in large differences in
reported SVOC concentrations, as the majority of SVOCs are
typically associated with the finest particles (e.g., <1.5 μm),
which are sampled by the TSP, PM10 and PM2.5 sampling
heads.48−52
On average PM2.5 represents >75% of TSP by mass
and most of the SVOCs (>80%) are found in this fraction.51,52
However, there may be circumstances when the use of different
sampling heads leads to very different results, such as
environments with high TSP, or industrial settings where
SVOCs are found on coarser particles.
Sorbent and Filter Type. A range of different sorbent types
are used for sampling gas-phase SVOCs in AAS (e.g., PUF,
XAD, Tenax TA, polydimethylsiloxane), but published data on
SVOCs is dominated by sampling with PUF and XAD, and we
focus our discussion on these. PUF is a low density foam
consisting of a polymer bound by urethane links that has been
used in SVOC air sampling for over 40 years.25
XAD is a
styrene-divinylbenzene copolymer more recently used in AAS,
either independently or in conjunction with PUF. Two types of
XAD are commonly used in SVOC air sampling:53
(i)
Amberlite XAD-2, commonly used in AAS and XAD-
PAS,6,54,55
and (ii) Amberlite XAD-4, in active denuder
samplers and SIP-PAS.56,57
The main difference between
XAD and PUF is sorptive capacity; for equal amounts of
sorbent material, PUF has a much lower capacity than XAD,
largely due to differences in surface area. PUF has a specific
surface area of 0.007−0.035 m2
/g while the surface area of
XAD-2 is 300−600 m2
/g.53
On a practical basis, this means that
samplers using XAD as the gas-phase sorbent can collect larger
volumes of air before breakthrough occurs (for AAS) or be
deployed for longer periods of time before equilibrium is
reached (for PAS). However, PUF does have the advantage of
lower cost and ease-of-use compared to XAD, as manipulation
of the PUF is simple and blanks are often lower.
There are also differences in the surface bonding of particular
compounds to PUF and XAD.58,59
XAD is a stronger sorbent
for PFCs, current-use pesticides (CUPs), and other emerging
chemicals,59,60
yet the differences are not fully understood, and
research is ongoing. Both sorbents are influenced by temperature
and humidity, but the amount of variability introduced by
these parameters is unclear and needs further research.53,59
Generally, they are linked to changes in sorptive capacities,59
and thus effects can be significant when large seasonal shifts are
expected (e.g., monsoon-impacted areas). The choice of PUF
vs XAD should be based on the choice of compound to be
sampled and the sampling parameters, but is often due to
previous regional practices (e.g., PUF is more commonly used
in AAS in Canada and Europe, while XAD is more commonly
used in the U.S.). PUF/XAD combinations are seeing
increasing global use in recent years (SI Figure S2).
Three types of filters are used in AAS for SVOCs: quartz
fiber filters (QFF), glass fiber filters (GFF), and Teflon/Teflonimpregnated
filters. GFF and QFF are most common, and are
perceived to give comparable results,61
although specific studies
assessing differences between them are limited. The recommendations
on filter type instead suggest that choice of filter
should consider what, if any, additional analyses will be
completed in conjunction with SVOCs (e.g., PM, SO2, organic/
elemental carbon, or metals).45,62
While the choice of sampling head determines the upper end
of the particle sizes collected on the filter, the lower size cutoff
is related to filter choice, but is more ambiguous. Typical GFFs
or QFFs have a collection efficiency of >99%, but do not often
have a specified fine particle cutoff. Particles <50 nm may pass
through the filter and subsequently be trapped by the gas-phase
sorbent material.63
This might be an issue for ultrafine particles
(typically <100 nm) which are of concern for human
health.64,65
Although numbers of ultrafine particles are high,
their total mass is small and therefore they are challenging to
quantify and require adaptations of current instrumentation or
new sampler types. There is limited information on how
ultrafine particles are distributed within an AAS, and this may
contribute to uncertainty in gas-particle partitioning.
Breakthrough. An important consideration in AAS is what
volume of air can be collected before the sampling medium
experiences “breakthrough”, that is, the loss of compounds
downstream of the sampling medium. This can be a result of a
saturated sampling medium or desorption of compounds from
the sampling medium. Sampling design must achieve the right
balance between collecting sufficient sample for analysis of
lower concentration SVOCs, while not collecting so much
volume as to have breakthrough of more volatile SVOCs. In
breakthrough estimates or controlled experiments <10%
breakthrough is considered an acceptable threshold.66−68
Environmental Science & Technology Critical Review
dx.doi.org/10.1021/es502164r | Environ. Sci. Technol. 2014, 48, 14077−1409114080
The key factors affecting breakthrough are (1) type and
geometry of sampling medium, (e.g., PUF vs XAD), (2) target
compound properties and concentration, (3) competition for
the sampling medium (i.e., from other gas-phase compounds),
(4) sample volume, (5) sampler flow rate, (6) temperature, and
(7) humidity. Of these, sampling medium, sample volume and
sampler flow rate are the most easily controlled, and therefore
breakthrough is most often discussed in terms of what sampling
medium and sample volume are appropriate for a given class of
compounds.
Basic chromatographic relationships can be used to estimate
breakthrough,24,67
but given the number of variables influencing
breakthrough, it can be more reliable to conduct small
breakthrough experiments with the exact study configurations.
Results from early breakthrough studies67,68
tend to be used as
an accepted framework for present studies and little
consideration is given to breakthrough in typical sampling
set-ups. Despite this, measurements and calculations suggest
that in a typical high-volume AAS breakthrough of more
volatile compounds (e.g., PCB-28, HCB, HCHs, fluorene) can
occur already at <600 m3
.24,34,59,66,68−70
For example, at
tropical temperatures (e.g., around 30°C) a typical high volume
AAS (2 PUF plugs, 600 m3
of air, 24 h) can lose 10−15% of the
PCB-28 due to breakthrough (calculations in SI). However, it
should be noted that this is also highly temperature dependent:
with the same sampling configuration at 5°C the breakthrough
volume is >3000 m3
. Many current AAS configurations may
experience breakthrough for lower molecular weight SVOCs,
leading to systematic underestimation of these compounds;
consideration of this is warranted in interpretation of results.
Filter Artifacts. The filter within the AAS (GFF or QFF) can
be the site of two sampling artifacts which influence measured
concentrations: blowoff and filter adsorption. Blowoff is the
volatilization loss of SVOCs from the filter, thereby
disproportionately reducing SVOC filter masses, while filter
adsorption is the adsorptive partitioning of gaseous compounds
onto the filter, increasing the mass of SVOCs on the filter.
This is affected by sampler configurations (e.g., flow rate,
sampling time, filter type) and occurs largely under three
conditions: (i) when there are large variations in the SVOC
concentration in incoming air, (ii) when there are large
temperature variations, (iii) when there is very high particle
loading to filters. Under these conditions, the gas-particle
distribution of SVOCs in the AAS may no longer be in
equilibrium with the influent air, and filter blow-off can occur in
the case of higher temperatures/lower concentrations, or filter
adsorption in the case of lower temperatures/higher concentrations.
By using a higher flow rate or longer sampling time,
surface-sorbed SVOCs can more easily be stripped from
particles than those that are less available for exchange with the
atmosphere. In large sample volumes or in areas of high PM,
overloading of filters can cause further artifacts. If the filter
becomes physically blocked by PM, sampler flow rates can be
reduced and thus true sample volumes could be lower, resulting
in an underestimate of concentrations.
Filter adsorption and blowoff further vary by compound,
depending on partitioning coefficients, ionic interactions, and
hydrophilicity.71,72
For example, significant filter adsorption was
identified for perfluorooctanesulfonate (PFOS), and certain
perfluoroalkyl carboxylic acids (PFCAs) in a high-volume AAS,
resulting in overestimation of particle-associated fractions by up
to 80%, while other PFCs did not have significant filter
adsorption.24
If filter blowoff/adsorption occurs, bulk air concentrations
may be correctly characterized, but gas-particle partitioning
estimates may be incorrect. This is expected to be most
significant for compounds with short-term temporal variations
in concentrations71
(such as CUPs) which may not be in
equilibrium between the gas and particle phases, and those with
intermediate gas-particle partitioning that vary between largely
gas- and largely particle-associated at ambient air temperatures.
Degradation. Degradation of atmospheric SVOCs occurs as
a natural process, caused by reaction with atmospheric reactive
species, in particular hydroxyl radicals, ozone, and NO3, or by
photolysis (SI Figure S3). Degradation has been found to
significantly affect ambient atmospheric concentrations, notably
for PAHs through reaction with OH, ozone, and NO3,73,74
for
PCBs and pesticides through reaction with OH,75−78
and for
PBDEs and novel flame retardants (NFRs) through photol-
ysis.79−81
While a sampler should provide a snapshot of the
atmospheric SVOC concentrations which may be already
affected by degradation, there is also the possibility for
degradation to continue within the sampler, as atmospheric
reactive species are drawn into the sampling medium along with
the SVOCs themselves. Within-sampler degradation is an
unwanted sampling artifact that can result in underestimation
of ambient concentrations.
UV radiation has a minimal direct influence on withinsampler
degradation due to the protection provided by the
sampler housings, but there is the potential for degradation due
to ozone, hydroxyl radicals and other atmospheric reactive
species to continue within the sampler. Significant withinsampler
degradation has been observed for PAHs both on
filters82,83
and in gas-phase sorbents.84
Particle-phase PAH
concentrations can typically be underestimated by 20−40% due
to degradative losses on AAS filters, with underestimates >70%
for particularly reactive compounds (e.g., benzo[a]-
pyrene).82,83,85
Gas-phase PAHs may also experience degradation
within the sorbent, with losses of up to 50% due to
reaction with both OH and ozone.84,86
The amount of potential within-sampler degradation
depends on sampling conditions, site conditions, and
compound, but the range of variability due to these factors is
highly uncertain. For example, there are large spatial and
temporal ranges in levels of atmospheric reactive species. On a
global scale, ozone varies by a factor of 2−3, with higher levels
in populated areas,87
while OH varies by a factor of 6, with
higher levels in tropical regions,75,88
and there is potential for
even higher local-scale variability in both ozone and OH.89
Ozone can vary seasonally by a factor of 2 and diurnally to a
lesser extent87
while OH can vary up 10× both seasonally and
diurnally.75,88,90
Given the direct relationship between the
amount of degradative losses and levels of atmospheric reactive
species, within-sampler degradation may be significantly higher
(up to 100%) if the concentrations of atmospheric reactant
species are very high.84,85
This can then affect the temporal
(e.g., summer vs winter) and spatial comparability (urban vs
remote) of measured concentrations. For example, estimates
based on reaction with ozone74,91
suggest that measured
concentrations at an urban site in summer with 60 ppb ozone
would be underreported by 30% due to particle phase reaction
with ozone, while at a remote site in winter with 20 ppb ozone,
there would be minimal within-sampler degradation. While site
and season are very important for degradation, filter and
sorbent type, sampling duration, amount and composition of
PM, temperature and humidity also affect reactive losses.74
The
Environmental Science & Technology Critical Review
dx.doi.org/10.1021/es502164r | Environ. Sci. Technol. 2014, 48, 14077−1409114081
length of sampler deployment may have a minor influence on
degradation,74,85
but more characterization of these processes is
needed.
Experimental evidence only exists for PAHs, but withinsampler
degradation should also be of importance for other
reactive SVOCs, according to their susceptibility to degradation
in the atmosphere. Gas-phase compound degradation is largely
via reaction with the OH radical,73,86,92
and thus we expect
other gas-phase compounds with OH reaction rate constants in
the same range as PAHs to also have potential within-sampler
degradative losses (SI Table S2). In contrast, on-filter
degradation is largely through reaction with ozone,82,86
and
thus degradative losses of particle-associated compounds are
important mainly for PAHs (SI Table S3). Based on rate
constants and gas-particle partitioning, we estimate that PAHs,
CUPs and some NFRs may experience significant withinsampler
degradative losses.
Sampling InfluencesPassive Air Sampling. Despite
the benefits and widespread use of PAS1
they are still associated
with challenges and limitations affecting the comparability of
measurements. These are important to recognize and control,
but few of them can be fully quantified in a field deployment
setting.
Calculation of Air Concentrations. The most relevant
limitation of PAS is their lower accuracy compared to AAS.5
PAS have a semiquantitative nature, since air concentrations
can only be derived if an estimated sampling rate (m3
/day) is
applied. The sampling rate is specific to the type of sampler
(e.g., sorbent, sampler housing) and the environmental
conditions (e.g., indoor, outdoor, meteorology), and can vary
by orders of magnitude due to these differences.6,93−96
In
theory sampling rates should not vary by compound, but in
practice large variations are found due to compound-specific
behaviors or uncertainties in the methods of determining
sampling rates. For example, measured sampling rates for
PUF−PAS span 3 orders of magnitude (e.g., 0.02−20 m3
/
day),32,33,94,95,97−99
and some studies have observed systematic
variations in sampling rates according to compound physicalchemical
properties.32,33,93,100
The choice of using one general
sampling rate or nongeneric sampling rates is an important
current discussion point within the PAS community. The use of
nongeneric sampling rates (such as homologue-specific) can
“correct” for some of the potential errors a general sampling
rate may introduce under certain circumstances. For example,
using a general sampling rate of ∼4 m3
/day36,101
vs a
nongeneric sampling rate of 0.3 m3
/day for benzo[a]pyrene94
results in 13× concentration differences due to the choice of
sampling rate alone. However, a general sampling rate for all
SVOCs is often used in monitoring networks, as it is unclear
whether the corrective potential of nongeneric sampling rates is
significant considering the semiquantitative nature of PAS. It
may not be a problem when the goal is to study the sum of
compounds but may lead to errors if compound-specific
concentrations or the SVOC fingerprint is of interest. The PAS
community has not yet reached consensus regarding selection
of appropriate sampling rates.
The methods for obtaining sampling rates add an additional
level of uncertainty to their validity. Sampling rates are obtained
from three methods: (1) use of depuration compounds (DCs)/
performance reference compounds (PRCs),102,103
(2) calibration
studies with comparison to (i) continuous low volume
AAS,31,33,104
or (ii) intermittent high volume AAS,18,19,59,105
or
(3) modeling exercises106,107
(e.g., applying a PUF-air
partitioning coefficient). Method (i) is considered most
accurate, and is particularly advantageous when calibrating
compounds with high short-term variability, which could be
biased with intermittent high volume AAS depending on the
timing of sampling.31
DCs account for site-specific environmental
conditions but do not cover the full-range of
compounds, as they are not applicable for particle-associated
compounds, and cannot be used for XAD samplers due to their
high sorptive capacity.
PAS for SVOCs are used as time-integrated samplers, and
sampling rates are only applicable when the sampler is in the
linear uptake phase.41,95
The length of the linear uptake phase
varies for compounds within a SVOC class as well as between
classes and it is important to keep deployment times within the
reported linear time frames for each compound. While the
conventional PAS model assumes uniform distribution of a
compound within the sorbent,41,95
recent studies have
identified that kinetic resistance within the sorbent material
may be limiting and as such, model-based calculations may
overestimate the length of the linear uptake phase.59,100,108
This
contention is supported by experimental work.33,93
For
example, penta+hexachlorobenzene and low molecular weight
PAHs and PCBs enter a curvilinear uptake after 6−9 weeks,93
and thus air concentrations from a three-month PUF−PAS
deployment may be underestimates. Furthermore, the length of
the linear uptake phase may be affected by environmental
conditions such as temperature, wind speed and air
concentrations, but this is not yet fully characterized.
Sorbent Material. Two sorbent types are commonly used in
PAS: PUF and XAD. As with AAS, other sorbents have been
used but not applied on a large scale. Their specific
characteristics result in different sampling potentials. The
PUF sorbent has a lower capacity and higher sampling rate than
the XAD sorbent (Table 1).31
Typical deployment times in
established air monitoring networks are months (∼1−4
months) for PUF−PAS36,101,109
and up to one year for XAD-
PAS.6
As a consequence, data from the two samplers are only
comparable if sampling is conducted within the linear uptake
phase of each sampler, and time weighted average concentrations
from multiple PUF−PAS cover the same sampling
period as XAD-PAS.
Another difference between the two sorbents is the type of
compounds they can sample. While XAD is solely a gas-phase
sampler, PUF is able to accumulate both gas-phase and particleassociated
SVOCs, although particles are sampled with a lower
accuracy and more variable sampling rates.93,97
On the other
hand, PUF−PAS have shown poorer or inconsistent performance
for more volatile and polar SVOCs (e.g., CUPs, PFOS/
PFOA), for which XAD perform well.31,56,59,110
The low
sorptive capacity of PUF−PAS for most PFCs60,110
suggests
different partitioning mechanisms for polar and nonpolar
compounds. The SIP-PAS has been shown to be more effective
for PFCs,56
CUPs,37
methyl siloxanes,111
and phthalates.112
However, consistency with SIP-PAS may be challenging, as
XAD resin can be lost from the SIP disk during sampler
deployment, particularly in windy, outdoor conditions,113
thereby introducing added uncertainties.
PUF disks of different densities (0.021, 0.030, and 0.035 g
cm−3
) are currently used by different research
groups.36,105,114,115
Although density is included in the
theoretical calculation of PUF-air partition coefficients, the
comparability of results from different densities is not fully
understood. A higher density PUF disk has higher capacity for
Environmental Science & Technology Critical Review
dx.doi.org/10.1021/es502164r | Environ. Sci. Technol. 2014, 48, 14077−1409114082
SVOCs and thus a longer linear uptake phase. However, kinetic
resistance within the PUF also differs with density. This can
result in different uptake scenarios, with lower density PUFs
having enhanced rates of SVOC transfer into the interior of the
disk, whereas higher density PUF has a rapidly filling “surface
compartment” and slower filling of the interior of the disk (e.g.,
a two-phase uptake mechanism).114
Additionally, particle
uptake may be enhanced in a lower-density PUF, which
could be a contributing factor in the discrepancies found for
particle associated PAHs (0.1 vs 5.0 m3
/day) in different
studies.93,96
Sampler Housing. PAS sorbents are deployed inside a
sampler housing (chamber) to reduce the influence of
environmental factors that affect the performance of the PAS,
such as wind speed and UV radiation. Studies have indicated
that the PAS configuration, including the position of the
sampling media (both PUF and XAD) inside the housing as
well as the alignment of the housing (tight or open), affects
uptake/accumulation and sampling rates.108,116,117
The housings used for PUF−PAS consist of two stainless
steel bowls fixed on a common vertical axis around the PUF
disk. This type of housing may vary slightly in geometry and
deployment. Differences in the bowl geometries do not
significantly affect uptake114
but the way the housing is fixed
results in different uptake scenarios, especially under windy
conditions. A freely hanging sampler housing dampens the
variation in external wind speed and results in small variations
in sampling rates, while a fixed sampler results in up to 3×
higher sampling rates under windy conditions.115,118
The XADPAS
housing consists of two parts: an inner stainless steel mesh
tube in which the XAD is placed and an outer larger stainless
steel cylinder acting as a protective shelter.6
Differences in
housing configuration for XAD-PAS and air movement within
the sampler housing also cause significant variation in sampling
rates (up to 50%).108
Degradation. No study has investigated degradation in PAS
in detail. However, it is expected to affect the same compounds
and be related to the same environmental variables as in AAS.
Kennedy et al.119
identified photodegradation when PUF−PAS
were not shielded by a sampler housing and that this is reduced
by the double bowl chamber; typical PAS housings should limit
UV exposure. The long deployment times of PAS and thereby
long exposures to atmospheric reactive species may exacerbate
degradation, both of the SVOCs and the sorbent material itself.
The effect on the structure and uptake capacity of the sorbent is
unknown. Compound-specific degradation may also be a source
of variability in sampler calibrations.
Environmental Factors/Location. The main causes of
variability in PAS are environmental variables (e.g., wind
speed, temperature, air concentrations) and most refinements
to PAS aim to quantify/control these, through calibration of
sampling rates, sampler housing design, use of DCs, etc.
Despite these efforts, environmental factors continue to lead to
bias/errors in estimated air concentrations.
Sampler housings are effective up to wind speeds of 4 m/s as
they maintain the air flow within the chamber at less than ∼1
m/s, which has a minor effect on sampling rates.115
At wind
speeds >5 m/s the sampler housing cannot sufficiently dampen
outdoor winds, resulting in a higher air velocity within the
chamber and a rapid increase in sampling rates.115
This is
supported by results from DCs, which have shown much higher
sampling rates at windy, coastal, and mountain sites.36,99,101
For
XAD-PAS, sampling rates are also significantly higher under
windy conditions.106,120
In contrast to the PUF−PAS, the
XAD-PAS sampling rates double between winds of 0 and 1 m/
s, but the effect on sampling rates is less with higher wind
speed.6,120
Temperature affects PUF and XAD-PAS in similar ways.
Temperature controls the gas-particle partitioning of SVOCs,
thus affecting the sampling rate (discussed below). However,
temperature also affects the sampling medium: for example,
higher temperature leads to higher diffusivity and thus higher
sampling rate95
and higher temperature leads to lower sorptive
capacity and thus shorter linear uptake phase. The effect of
these two factors on the overall sampling rate is complex and
varies by compound,106
but can lead to underestimation of
concentrations in warm/tropical conditions if deployment
times are the same as in colder/temperate regions. Further
complicating matters, variations in temperature also affect
ambient air concentrations (e.g., higher temperature leads to
higher volatilization and thereby higher air concentrations)
which impacts one of the major assumptions of the PAS theory,
that the air concentration is constant.
Particle-Associated Compounds. A big question when using
PAS for SVOCs is their performance for particle-associated
compounds. The XAD-PAS is considered a purely gas-phase
sampler while the PUF−PAS also accumulates particleassociated
compounds. Current findings are not consistent;
some suggest similar PUF−PAS sampling rates for gas- and
particle-phase compounds,33,96,121
others show poorer performance
and up to 100× lower sampling rates for the particleassociated
compounds.93,105
The reason for inconsistent results
is not known but may be related to site characteristics, PUF
types, or analytical differences. High TSP values and high wind
speeds may favor the uptake of particle-associated compounds
to a level similar to gas-phase compounds. Another possible
reason may be differences in PUF density, as mentioned
above.93
The influence of particle-associated compounds on
overall sampling rates can be exacerbated at colder temperatures
when a higher fraction of SVOCs are associated with
particles. The variable results imply large errors and low
comparability for particle-associated compounds between sites,
and it is difficult to draw a general conclusion that is applicable
to all or most of the sites in monitoring networks.
The ability to use PAS for particle-associated compounds is a
key need of the SVOC community, as regulated and emerging
SVOCs (PAHs, PBDEs, NFRs, PFCs, CUPs) tend to have high
particle fractions. The applicability of PAS for these compounds
is uncertain, and the subject of ongoing research.
■ CHALLENGES AND LIMITATIONS
Implications for Air Sampling. Given the aforementioned
sources of bias and potential error in AAS and PAS, it is clear
that each individual measurement is associated with some
degree of uncertainty which may be a bigger contribution than
the analytical uncertainty. However, more importantly, the
uncertainty, accuracy, and precision of the measured values vary
widely by sampling technique and sampler configuration, and
thus there may be poor comparability between different data
sets.
Key issues for comparability are (i) sampling times, sample
volumes and frequencies, (ii) PUF−PAS assumptions about
particles, (iii) PAS sampling rates, (iv) spatial coverage, and the
associated larger range of environmental variables as sampling is
expanded to more geographic regions.2
Environmental Science & Technology Critical Review
dx.doi.org/10.1021/es502164r | Environ. Sci. Technol. 2014, 48, 14077−1409114083
Table2.PercentofAmbientAirConcentrationCapturedbyGivenSamplerConfigurationsa
%ofairconcentrationcapturedbysampler
high-volumesampler
(24h,600m3
)
PUF−PAS
(90daydeployment)
XAD-PAS
(1yeardeployment)
example
compounds
vaporpressure(Pa)at
0°C,25°C
logoctanol-airpartitioningcoefficient
(KOA)at0°C,25°C
atm.half-life(hrs)
at25°C
estimatedparticlefraction
at0°C,25°C0°C25°C0°C25°C0°C25°C
PFOS0.0973,0.9815.23,4.849170.00,0.0070%(1)6%(1)1%(3)1%(3)50%(3,4)0%(3,4)
fluorene0.630,0.8198.03,6.74150.01,0.004%(1,2)3%(1,2)20%(2,3)4%(2,3)100%80%(3)
α-HCH0.105,0.3768.37,7.372240.01,0.0070%(1)20%(1)60%(3)20%(3)100%90%(3)
PCB-280.0314,0.1239.28,8.061080.03,0.00100%100%(1)90%(3)50%(3)100%100%
chlorpyrifos0.00399,0.005369.56,8.411.40.04,0.010%(2)0%(2)80%(2,3)60%(2,3)70%(2)70%(2)
endosulfan5.06×10−4
,
2.31×10−5
10.58,9.28160.16,0.0390%(2)90%(2)80%(2,3)80%(2,3)80%(2)80%(2)
PBDE-474.91×10−7
,
2.19×10−5
12.09,10.541280.64,0.15100%100%50%(4)90%(4)40%(4)90%(4)
benzo[a]pyrene0.00550,6.26×10−6
12.29,11.152.50.97,0.7660%60%(2)8%(2,4)20%(2,4)3%(4)20%(4)
PBDE-2091.01×10−8
,
2.23×10−4
16.51,15.2638101.00,1.00100%100%10%(4)10%(4)0%(4)0%(4)
a
Thepercentagesrepresentthecombinedeffectsofbreakthrough/degradation/exceedanceoflinearuptakephaseat0and25°Cand50ppbozone.Causesofthelossescorrespondingtoeachcompound
areindicatedbythenumbersfollowingthepercentage.(1)indicatesbreakthrough,(2)isdegradation,(3)isexceedanceofthelinearuptakephase,(4)isahighparticlefraction.Fulldetailsonthe
estimatesaregivenintheSI.References:PCB,HCH,andPAHphysical-chemicalpropertiesandtemperatureadjustmentsarefromBeyeretal.131
andPaasivirtaetal.;132
PBDEphysical-chemical
propertiesandtemperatureadjustmentsarefromHarnerandShoeib,133
Tittlemeieretal.,134
andWangetal.;135
PFOS,chlorpyrifosandendosulfanphysical-chemicalpropertiesandtemperature
adjustmentsarefromOdabasiandCetin136
andEpiSuite.137
Particle-phasefractionswerecalculatedusingthemethodsofHarnerandBidleman126
usingthetemperature-adjustedKOAvalues.
BreakthroughwascalculatedusingtheequationsofPankow.67
LengthofthelinearuptakephasewascalculatedusingtheequationsofShoeibandHarner.95
Degradationwasestimatedbasedonfieldstudy
datafromArmstrongetal.,138
Schaueretal.,85
TsapakisandStephanou,84
Petersetal.,139
Menichini,82
andGoriauxetal.140
Fulldetailsonallassumptions/estimatesareprovidedintheSI.
Environmental Science & Technology Critical Review
dx.doi.org/10.1021/es502164r | Environ. Sci. Technol. 2014, 48, 14077−1409114084
For example, an intermittent AAS can easily be biased for
compounds with short-term temporal concentration variability
(e.g., CUPs, combustion-related SVOCs)31
and a long-term
time weighted average concentration from a PAS cannot
provide information on episodic events. If intermittent sample
timing results in sampling on the same weekday or period of a
month, this can be biased by daily, weekly or monthly activity
patterns (e.g., weekday traffic). This can lead to an over- or
underestimation of concentrations if the sample either
coincides with or completely misses a high concentration
event. Furthermore, as we have moved to measurement of
emerging compounds and lower concentration ranges (pg/m3
and fg/m3
levels), sample collection volumes have been pushed
higher in order to achieve detection, particularly in remote
environments, without thoroughly assessing the associated
breakthrough potentials.
As described earlier, there is much uncertainty in how
particles behave in PUF−PAS. Comparability issues are
introduced because users of PUF−PAS make different
assumptions about the fraction of the particle phase sampled
by the PUF. The assumptions range from treating PUF−PAS as
gas-phase only to assuming 1−10% of particles are
sampled,93,105
to assuming all particles are sampled.96
If
PUF−PAS are treated as bulk samplers when they are only
sampling a fraction of particles, measured values may be biased
low, particularly for higher molecular weight SVOCs.
The systematic variations in PAS sampling rates by physicalchemical
properties identified in some studies,32,33,93,100
reflect
situations where the conventional PAS framework does not
apply, such as when volatile compounds enter the curvilinear
uptake phase, or when only a fraction of atmospheric particles
are sampled. Thus, the use of a general sampling rate may
introduce bias in either the low molecular weight or high
molecular weight compounds and different compound/
congener distributions.33
It is also not well quantified how
much sampling rates differ between different climate zones/
seasons, thus considering the global range of average
temperatures (SI Figure S4), large biases may be introduced
when comparing PAS measurements from very different
regions or seasons.
Spatial distributions of sampling networks further bias our
global knowledge of air concentrations.1,2
While coverage is
somewhat improved when one also considers case studies,
there continue to be areas of the world with limited
measurement coverage (Figure 1). For example, if one
considers available PBDE data for the period 2003−2013,
only 2% of the global land mass is within 100 km of a reported
PBDE air concentration, and 0.1% is within 20 km. By
population, only 30% of people live within 100 km of a
reported PBDE measurement, and only 6% within 20 km.
Southeast Asia has some coverage for PBDEs measured by case
study, but these largely consist of one-time measurements,
thereby limiting temporal information. Additionally, land-based
southern hemisphere sampling is largely PAS (SI Figure S1),
furthering the spatial bias.
Examples of Bias. To demonstrate the potential combined
effects of the aforementioned sampling artifacts and biases, we
estimated the potential errors in measured concentrations for
three typical sampler configurations: (1) a high-volume active
air sampler with GFF or QFF and two PUF plugs, collecting
∼600 m3
of air over 24 h, (2) a PUF−PAS, deployed outdoors
for 90 days, and (3) an XAD-PAS, deployed outdoors for one
year, and two temperature scenarios: 0 and 25 °C. Estimates are
summarized in Table 2 and details are given in SI Table S4.
These estimates have significant uncertainty, due to the very
different data sources, but they emphasize the large possibility
for sampling artifacts to influence measured concentrations.
The estimates suggest combined effects of breakthrough
and/or exceedance of the PAS linear uptake phase lead to large
underestimates in the reported values of the volatile SVOCs,
notably PCB-28, α-HCH, fluorene, PFOS, and by extension,
other SVOCs with similar physical-chemical properties. This
effect also influences XAD-PAS for more volatile SVOCs (e.g.,
α-HCH, fluorene, PFOS) at higher temperatures.59,106
Withinsampler
degradation, although not well-characterized, is also
estimated to contribute to large losses, particularly for fluorene,
benzo[a]pyrene, and chlorpyrifos. Similar degradation is
expected for gas-phase compounds with short reactive
atmospheric half-lives, which includes all PAHs, many CUPs
and some NFRs (SI Table S2). The particle-associated SVOCs
are under-sampled by PUF−PAS and not sampled by XADPAS.
Estimates in Table 2 assume PUF−PAS collect 10% of the
total particle fraction, but this may range from 1 to 100%. Thus,
reported concentrations may vary between 15 and 100% of the
actual concentration, depending on the fraction of particles that
is truly collected by the PUF−PAS.
Attention is frequently given to issues of analytical
comparability, but sampling comparability affects many
compounds which we assume have minimal analytical
uncertainties. For example, BDE-209, which has frequently
been identified to have difficulties in laboratory analysis,122
is
“correctly” sampled, for example, 100% of the air concentration
should be captured by a typical high volume air sampler,
whereas PCB-28, a compound that has been consistently
included in global monitoring networks and is one of the seven
typical indicator PCBs, may experience losses of up to 15%
based on breakthrough alone, and thus reported values may be
biased low in many situations.
Problematic Chemicals. Many “emerging” SVOCs have
different physical-chemical properties compared to the legacy
SVOCs and this may introduce challenges in sampling. Current
sampling networks and techniques were developed for legacy
SVOCs and may not be appropriate for providing representative
data on these newer SVOCs. For example, the use of PAS
in the future may be more challenging because target
compounds are either too volatile (volatile methylsiloxanes,
phthalates) or too involatile (CUPs, NFRs) and thus have high
uncertainty in conventional PAS. As a consequence of
regulatory actions, newer compounds are often designed to
be less persistent in the environment, but as a result are also
less persistent in samplers (leading to more within-sampler
degradation) and thus more difficult to correctly quantify.
Degradation is also of concern when measuring nitro- and
other substituted PAHs,123
and existing data on these
compounds is likely an underestimate. The lack of data on
degradation and the potential importance for many emerging
compounds highlights the need for more studies.
Furthermore, we must be aware that chemicals currently
measured by the SVOC community are a small fraction of the
chemicals in use that may be of environmental and/or human
health concern.124
Adding more chemicals to the lists of
analytes without adapting current sampler configurations will
introduce more uncertainty in measured concentrations.
Environmental Science & Technology Critical Review
dx.doi.org/10.1021/es502164r | Environ. Sci. Technol. 2014, 48, 14077−1409114085
■ RECOMMENDATIONS AND FUTURE
DEVELOPMENTS
Recommendations to Enhance Comparability. Given
the wide range of potential biases and uncertainties in air
sampling identified above, it is clear that careful consideration
must be given to sampling approaches. This consideration
should first consider the study goals and outcomes. For
example, what sampling designs are best suited to address
global distributions vs data for human exposure assessments vs
source identification, long-term vs short-term trends, and the
target SVOCs? The sampler type, frequency, and sampling
materials should correspond to the specific question and
compounds to be addressed. But ultimately, as the GMP and
other initiatives aggregate international data, study design
should also aim for comparability with global data sets.
Overall, PAS are best for broad spatial coverage and longterm
temporal trends, while AAS are best for short-term
temporal trends and coverage of a broader set of compounds.
When the intended result of a study is long-term integrated
average concentrations (e.g., monthly to yearly scale), and
particularly for largely gas-phase compounds, PAS have a clear
advantage. The simplicity of PAS (no electricity, cheaper) make
them best suited to resolve concentration gradients where levels
vary by orders of magnitude. AAS are recommended when
absolute concentrations are needed, smaller concentration
variations are expected or high time resolution is needed.
AAS are the only choice when gas-particle partitioning is
important or particle-phase concentrations are expected to be
dominant and when short-term trends (hourly to weekly) are
of interest. Because congener/compound ratios can be biased
in PAS, AAS are best for source identification/fingerprinting
techniques. Regulations for human exposure assessment are
often based on particle-associated SVOCs and the best sampler
choice is a low volume AAS providing continuous gas- and
particle-phase concentrations, while PAS are not as well suited.
However, current research continues to demonstrate the broad
applicability of PAS,96,97
and, although they have higher
uncertainty, particularly for particle-phase compounds, the
uncertainty in a temporally integrated PAS could be lower than
uncertainty in intermittent AAS for compounds with daily
concentration variations.
Sorbent choice should be considered with respect to the
length of sampling period and target compounds. In PAS, PUF
is recommended for capturing seasonal trends for stable, legacy
compounds, while XAD-PAS are recommended for longer term
(e.g., yearly, decadal trends)31
and also to cover emerging gasphase
SVOCs (more polar compounds). In AAS, for best
applicability to a broad range of SVOCs and conditions, we
recommend a combination of PUF and XAD, as this reduces
breakthrough concerns and allows quantification of both polar
and nonpolar SVOCs. While PUF and XAD are the current
focus of the majority of the SVOC community, new sorbents
should be investigated for their applicability to a wider range of
compounds.
Although there is a need for standardization in SVOC air
measurements, with large regional differences and equipment
constraints, implementation of standardization can be challenging.
This could be addressed in the long term through the
development of international standards for SVOC sampling
(e.g., an ISO standard), but in the short term resetting of all
sampling campaigns and networks to the same configurations is
not practical, nor necessarily useful. Many networks have longterm
established techniques and continuation of the same
techniques allows for temporal comparisons. Furthermore,
universal standardization of techniques would not be
appropriate for all regions. For example, sampling in high
temperatures tropical regions may require a different sampling
approach to low temperature polar regions. However, when
different techniques are used it is important for the SVOC
community to understand the biases from techniques/
configurations/approaches. Instead of implementing identical
sampling techniques, data from different sampling configurations
may be adjusted using standardization factors. Standardization
factors can account for effects of breakthrough,
temperature differences, wind, etc., and can be tailored to
specific sites/sampler types/seasons/sampling approaches. For
example, if degradation was better characterized, a standardization
factor could be used to account for the underestimation
caused at sites with high levels of reactive trace
gases. For cases where large sample volumes are needed,
standardization factors can be used to account for losses of the
more volatile SVOCs. In fact, this is already done to a certain
extent when compound-specific sampling rates are used for
PAS; these rates account for nonlinear uptake of volatile
SVOCs and lower sampling of particle-associated SVOCs.
One way standardization factors can be determined,
particularly for long-term monitoring networks, is at “supersites”
where samplers and sampling approaches from different
networks are used simultaneously.125
Quality assurance
techniques such as spiking samplers with labeled compounds
should also be used periodically across different networks to
further evaluate comparability.
However, standardization factors, once developed, should
not be treated as a perfect solution. Better consideration of
study design should be given in advance of sampler
deployments. There is sufficient theoretical and experimental
knowledge to allow researchers to estimate PAS linear uptake
phases,41,95
gas-particle partitioning of target compounds,126,127
and AAS breakthrough67
in advance of sampler deployment. A
necessary part of sampling campaign should be estimating these
effects and structuring the sampling campaign to reduce these
artifacts. New monitoring networks and case studies should aim
to harmonize, where possible, with existing networks to
maximize data comparability.
Recommendations for Future Work. Extensive experimental,
modeling and calibration work has and continues to be
done on PAS techniques with the goal of improving the
accuracy of measured concentrations and better understanding
uptake processes. This should be continued. However, limited
efforts have been made to improve AAS, despite knowledge of
uncertainties from degradation, breakthrough, etc. and this
should be an additional focus.
For PAS, there are some obvious improvements needed,
particularly in the determination of sampling rates. Currently,
sampling rates are determined either with calibration studies or
DCs, but sampling rates could be improved by combining these
two methods. Sampling rates (determined from calibration
studies) could be adjusted according to differential losses of
DCs between sampling sites to account for site-specific
influences. Further studies should also focus on understanding
the performance of PUF−PAS for particle-associated SVOCs
under different conditions and for different PUF densities, to
better understand why current studies obtain inconsistent
results.
Environmental Science & Technology Critical Review
dx.doi.org/10.1021/es502164r | Environ. Sci. Technol. 2014, 48, 14077−1409114086
For AAS there are also clear improvements needed,
particularly in the study of breakthrough and degradation.
Current sampling configurations are based on information from
breakthrough studies from 1980s and early 1990s24,67,128
when
different SVOCs were the primary concerns and concentrations
were often higher. Today, larger volumes are used to detect
compounds at the pg/m3
and fg/m3
level, particularly in low
concentration regions (e.g., polar regions, Figure 2) without a
good understanding of the associated impacts (breakthrough,
degradation). Since the use of denuders in conjunction with
AAS has yielded important information regarding the
degradation of PAHs, this avenue should be further investigated
to better understand within sampler degradation for PAHs and
other reactive SVOCs. As the relationship between ozone levels
and degradation appears relatively consistent,85
it may be
possible to estimate SVOC losses based on ozone levels, which
could be especially important in locations/times of year when
losses are expected to be large. Additionally, we should
reconsider the paradigm that AAS provide more accurate
results. We have summarized numerous well-known sampling
artifacts associated with PAS and AAS, but due to limited
experimental evidence, the question of the overall effect of
these artifacts on the accuracy of reported concentrations
remains a rough estimate (e.g., Table 2). The question of the
overall significance of these artifacts on PAS vs AAS could be
addressed with a well-designed PAS-AAS comparison.
In addition, as sampling expands to broader global coverage,
a more thorough understanding of the effect of environmental
conditions/meteorology on both AAS and PAS is needed.
Currently, sampler evaluation and testing is largely performed
in temperate regions; this may limit applicability in more
extreme environments.
There is also room for innovation in sampling techniques.
For example, source regions and the influence of specific air
masses can be assessed using new directional AAS129
and
PAS.130
Finally, the SVOC community must be progressive and
adaptable. How can sampling campaigns be designed to
produce more comparable data without compromising the
data needs of individual networks and case studies? What
compounds are most important to continue monitoring, and
what compounds should be added to global monitoring? We
have established that there are big challenges in comparability
between well-known SVOCs such as PAHs, PCBs, and PBDEs.
These must be addressed, but the SVOC community should
also move forward with air sampling methods that can meet the
future challenges of new chemicals. We must also be prepared
to address the future global developments, as the key questions
in SVOCs shift, e.g., responding to climate change-induced
meteorological shifts and new source balances due to melting of
glaciers and polar ice caps1
and the relationship between
SVOCs and human health impacts, including consideration of
possible synergistic effects.3
■ ASSOCIATED CONTENT
*S Supporting Information
Maps of global distributions of sampling networks and case
studies, description of atmospheric degradation processes,
atmospheric reactive half-lives for SVOCs, and details of
calculations for Table 2. This material is available free of charge
via the Internet at http://pubs.acs.org.
■ AUTHOR INFORMATION
Corresponding Authors
*(L.M.) Phone: +420 54949 5149; fax: +420 54949 2840; email:
melymuk@recetox.muni.cz.
*(P.B.) Phone: +420 54949 5149; fax: +420 54949 2840; email:
bohlin@recetox.muni.cz.
Notes
The authors declare no competing financial interest.
■ ACKNOWLEDGMENTS
This research was supported by the projects “Employment of
Best Young Scientists for International Cooperation Empowerment”
(CZ.1.07/2.3.00/30.0037) and “Employment of Newly
Graduated Doctors of Science for Scientific Excellence”
(CZ.1.07/2.3.00/30.009) cofinanced from European Social
Fund and the state budget of the Czech Republic. The
RECETOX research infrastructure was supported by the
projects of the Ministry of Education of the Czech Republic
(LM2011028) and (LO1214). We thank Jana Borůvková and
Roman Prokeš for assistance.
■ REFERENCES
(1) Klánová, J.; Harner, T. The challenge of producing reliable results
under highly variable conditions and the role of passive air samplers in
the Global Monitoring Plan. Trends Anal. Chem. 2013, 46, 139−149.
(2) Hung, H.; MacLeod, M.; Guardans, R.; Scheringer, M.; Barra, R.;
Harner, T.; Zhang, G. Toward the next generation of air quality
monitoring: Persistent organic pollutants. Atmos. Environ. 2013, 80,
591−598.
(3) Klánová, J.; Diamond, M. L.; Jones, K. C.; Lammel, G.; Lohmann,
R.; Pirrone, N.; Scheringer, M.; Balducci, C.; Bidleman, T. F.; Bláha,
K.; et al. Identifying the research and infrastructure needs for the
global assessment of hazardous chemicals ten years after establishing
the Stockholm Convention. Environ. Sci. Technol. 2011, 45, 7617−
7619.
(4) Bergman, A.; Heindel, J. J.; Jobling, S.; Kidd, K. A.; Zoeller, R. T.
State of the Science of Endocrine Disrupting Chemicals 2012; Bergman,
A., Heindel, J. J., Jobling, S., Kidd, K. A., Zoeller, R. T., Eds.; World
Health Organization: Geneva, 2012.
(5) Harner, T.; Shoeib, M.; Diamond, M. L.; Ikonomou, M. G.;
Stern, G. A. Passive sampler derived air concentrations of PBDEs
along an urban-rural transect: Spatial and temporal trends. Chemosphere
2006, 64, 262−267.
(6) Wania, F.; Shen, L.; Lei, Y. D.; Teixiera, C. F.; Muir, D. C. G.
Development and calibration of a Resin-Based Passive Sampling
System for Monitoring Persistent Organic Pollutants in the
Atmosphere. Environ. Sci. Technol. 2003, 37, 1352−1359.
(7) Zhang, T.; Huang, Y.; Chen, S.-J.; Liu, A.-M.; Xu, P.-J.; Li, N.; Qi,
L.; Ren, Y.; Zhou, Z.-G.; Mai, B.-X. PCDD/Fs, PBDD/Fs, and PBDEs
in the air of an e-waste recycling area (Taizhou) in China: Current
levels, composition profiles, and potential cancer risks. J. Environ.
Monit. 2012, 14, 3156−3163.
(8) Sharma, B. M.; Bharat, G. K.; Tayal, S.; Nizzetto, L.; Cupr, P.;
Larssen, T. Environment and human exposure to persistent organic
pollutants (POPs) in India: A systematic review of recent and
historical data. Environ. Int. 2014, 66, 48−64.
(9) Leyk, S.; Binder, C. R.; Nuckols, J. R. Spatial modeling of
personalized exposure dynamics: The case of pesticide use in smallscale
agricultural production landscapes of the developing world. Int. J.
Health Geogr. 2009, 8, 17.
(10) Pozo, K.; Harner, T.; Lee, S. C.; Sinha, R. K.; Sengupta, B.;
Loewen, M.; Geethalakshmi, V.; Kannan, K.; Volpi, V. Assessing
seasonal and spatial trends of persistent organic pollutants (POPs) in
Indian agricultural regions using PUF disk passive air samplers.
Environ. Pollut. 2011, 159, 646−653.
Environmental Science & Technology Critical Review
dx.doi.org/10.1021/es502164r | Environ. Sci. Technol. 2014, 48, 14077−1409114087
(11) Melymuk, L.; Robson, M.; Helm, P. A.; Diamond, M. L. PCBs,
PBDEs, and PAHs in Toronto air: Spatial and seasonal trends and
implications for contaminant transport. Sci. Total Environ. 2012, 429,
272−280.
(12) Hearn, L. K.; Kennedy, K. E.; Hawker, D. W.; Toms, L.-M. L.;
Alberts, V.; Müller, J. F. Spatial mapping of city-wide PBDE levels
using an exponential decay model. J. Environ. Monit. 2012, 14, 643−
650.
(13) Persoon, C.; Peters, T. M.; Kumar, N.; Hornbuckle, K. C.
Spatial distribution of airborne polychlorinated biphenyls in Cleveland,
Ohio and Chicago, Illinois. Environ. Sci. Technol. 2010, 44, 2797−2802.
(14) Simcik, M. F. Air monitoring of persistent organic pollutants in
the Great Lakes: IADN vs. AEOLOS. Environ. Monit. Assess. 2005, 100,
201−216.
(15) Blanchard, P.; Kallweit, D.; Brice, K. A.; Froude, F. A.; Chan, C.
H.; Neilson, M.; Holz, J.; Millat, H. A comparison of European and
North American atmospheric deposition networks: Polycyclic aromatic
hydrocarbons and lindane. J. Environ. Monit. 2006, 8, 465.
(16) Wu, R.; Backus, S. M.; Basu, I.; Blanchard, P.; Brice, K. A.;
Dryfhout-Clark, H.; Fowlie, P.; Hulting, M.; Hites, R. A. Findings from
quality assurance activities in the Integrated Atmospheric Deposition
Network. J. Environ. Monit. 2009, 11, 277−296.
(17) Schlabach, M.; Farag-clement, R.; Hung, H.; Kallenborn, R.
AMAP/EMEP/NCP Inter-Laboratory study for POP Analysis, EMEP/
CCC-Report 7/2011; Norwegian Institute for Air Research: Kjeller,
Norway, 2010.
(18) Gouin, T.; Harner, T.; Blanchard, P.; Mackay, D. Passive and
active air samplers as complementary methods for investigating
persistent organic pollutants in the Great Lakes Basin. Environ. Sci.
Technol. 2005, 39, 9115−9122.
(19) Klánová, J.; Kohoutek, J.; Hamplová, L.; Urbanová, P.;
Holoubek, I. Passive air sampler as a tool for long-term air pollution
monitoring: Part 1. Performance assessment for seasonal and spatial
variations. Environ. Pollut. 2006, 144, 393−405.
(20) Halse, A. K.; Schlabach, M.; Sweetman, A. J.; Jones, K. C.;
Breivik, K. Using passive air samplers to assess local sources versus
long range atmospheric transport of POPs. J. Environ. Monit. 2012, 14,
2580−2590.
(21) Dreyer, A.; Shoeib, M.; Fiedler, S.; Barber, J.; Harner, T.;
Schramm, K.-W.; Jones, K. C.; Ebinghaus, R. Field intercomparison on
the determination of volatile and semivolatile polyfluorinated
compounds in air. Environ. Chem. 2010, 7, 350.
(22) Brandsma, S. H.; de Boer, J.; Leonards, P. E. G.; Cofino, W. P.;
Covaci, A. Organophosphorus flame-retardant and plasticizer analysis,
including recommendations from the first worldwide interlaboratory
study. TrAC Trends Anal. Chem. 2013, 43, 217−228.
(23) Su, Y.; Hung, H. Inter-laboratory comparison study on
measuring semi-volatile organic chemicals in standards and air
samples. Environ. Pollut. 2010, 158, 3365−3371.
(24) Bidleman, T. F.; Simon, C. G.; Burdick, N. F.; You, F.
Theoretical plate measurements and collection efficiencies for highvolume
air samplers using polyurethane foam. J. Chromatogr. 1984,
301, 448−453.
(25) Bidleman, T. F.; Olney, C. E. High-volume collection of
atmospheric polychlorinated biphenyls. Bull. Environ. Contam. Toxicol.
1974, 11, 442−450.
(26) Orgill, M. M.; Sehmel, G. A.; Petersen, M. R. Some initial
measurements of airborne DDT over Pacific Northwest forests. Atmos.
Environ. 1976, 10, 827−834.
(27) Simon, C. G.; Bidleman, T. F. Sampling airborne PCB with
polyurethane foamChromatographic approach to determination of
collection effeciencies. Abstr. Pap. Am. Chem. Soc. 1978, 176, 187.
(28) Aas, W.; Breivik, K. Heavy Metals and POP Measurements, 2010,
EMEP Co-operative Programme for Monitoring and Evaluation of the
Long-Range Transmission of Air Pollutants in Europe, EMEP/CCCReport
3/2012; Norwegian Institute for Air Research: Kjeller, Norway,
2011.
(29) Hung, H.; Kallenborn, R.; Breivik, K.; Su, Y.; BrorströmLundén,
E.; Olafsdottir, K.; Thorlacius, J. M.; Leppänen, S.; Bossi, R.;
Skov, H.; et al. Atmospheric monitoring of organic pollutants in the
Arctic under the Arctic Monitoring and Assessment Programme
(AMAP): 1993−2006. Sci. Total Environ. 2010, 408, 2854−2873.
(30) Su, Y.; Hung, H.; Sverko, E.; Fellin, P.; Li, H. Multi-year
measurements of polybrominated diphenyl ethers (PBDEs) in the
Arctic atmosphere. Atmos. Environ. 2007, 41, 8725−8735.
(31) Hayward, S. J.; Gouin, T.; Wania, F. Comparison of four active
and passive sampling techniques for pesticides in air. Environ. Sci.
Technol. 2010, 44, 3410−3416.
(32) Hazrati, S.; Harrad, S. J. Calibration of polyurethane foam
(PUF) disk passive air samplers for quantitative measurement of
polychlorinated biphenyls (PCBs) and polybrominated diphenyl
ethers (PBDEs): Factors influencing sampling rates. Chemosphere
2007, 67, 448−455.
(33) Melymuk, L.; Robson, M.; Helm, P. A.; Diamond, M. L.
Evaluation of passive air sampler calibrations: Selection of sampling
rates and implications for the measurement of persistent organic
pollutants in air. Atmos. Environ. 2011, 45, 1867−1875.
(34) Batterman, S. A.; Chen, T.-C.; Chernyak, S. M.; Godwin, C.
Design and performance evaluation of a medium flow sampler for
airborne brominated flame retardants (BFRs). J. Environ. Monit. 2009,
11, 858−866.
(35) Aas, W.; Breivik, K. Heavy metals and POP measurements 2011;
EMEP Co-operative Programme for Monitoring and Evaluation of the
Long-Range Transmission of Air Pollutants in Europe. Norwegian
Institute for Air Research. Kjeller, Norway, 2011; EMEP/CCC-Report
4/2013.
(36) Pozo, K.; Harner, T.; Wania, F.; Muir, D. C. G.; Jones, K. C.;
Barrie, L. A. Toward a global network for persistent organic pollutants
in air: Results from the GAPS study. Environ. Sci. Technol. 2006, 40,
4867−4873.
(37) Koblizková, M.; Genualdi, S.; Lee, S. C.; Harner, T. Application
of sorbent impregnated polyurethane foam (SIP) disk passive air
samplers for investigating organochlorine pesticides and polybrominated
diphenyl ethers at the global scale. Environ. Sci. Technol. 2012,
46, 391−396.
(38) Farrar, N. J.; Prevedouros, K.; Harner, T.; Sweetman, A. J.;
Jones, K. C. Continental scale passive air sampling of persistent
organic pollutants using rapidly equilibrating thin films (POGs).
Environ. Pollut. 2006, 144, 423−433.
(39) Schuster, J. K.; Gioia, R.; Breivik, K.; Steinnes, E.; Scheringer,
M.; Jones, K. C. Trends in European background air reflect reductions
in primary emissions of PCBs and PBDEs. Environ. Sci. Technol. 2010,
44, 6760−6766.
(40) Mayer, P.; Wania, F.; Wong, C. S. Advancing passive sampling
of contaminants in environmental science. Environ. Sci. Process. Impacts
2014, 16, 366−368.
(41) Bartkow, M. E.; Booij, K.; Kennedy, K. E.; Müller, J. F.; Hawker,
D. W. Passive air sampling theory for semivolatile organic compounds.
Chemosphere 2005, 60, 170−176.
(42) Tuduri, L.; Millet, M.; Briand, O.; Montury, M. Passive air
sampling of semi-volatile organic compounds. Trends Anal. Chem.
2012, 31, 38−49.
(43) Bohlin, P.; Jones, K. C.; Strandberg, B. Occupational and indoor
air exposure to persistent organic pollutants: A review of passive
sampling techniques and needs. J. Environ. Monit. 2007, 9, 501−509.
(44) Esteve-Turrillas, F. A.; Yusà, V.; Pastor, A.; de la Guardia, M.
New perspectives in the use of semipermeable membrane devices as
passive samplers. Talanta 2008, 74, 443−457.
(45) Berg, T.; Dye, C.; Hanssen, J. E.; Krognes, T.; Munthe, J.;
Reissell, A.; Schaug, J.; Schmidbauer, N.; Semb, A.; Torseth, K.; et al.
EMEP Manual for Sampling and Analysis; Norwegian Institute for Air
Research: Kjeller, Norway, 2001.
(46) Finlayson-Pitts, B. J.; Pitts, J. N. J. Chapter 9: Particles in the
troposphere. In Chemistry of the Upper and Lower Atmosphere; Elsevier
Inc.: San Diego, 1999; pp 349−435.
(47) Englert, N. Fine particles and human health–a review of
epidemiological studies. Toxicol. Lett. 2004, 149, 235−242.
Environmental Science & Technology Critical Review
dx.doi.org/10.1021/es502164r | Environ. Sci. Technol. 2014, 48, 14077−1409114088
(48) Lammel, G.; Klánová, J.; Ilić, P.; Kohoutek, J.; Gasić, B.;
Kovacić, I.; Škrdlíková, L. Polycyclic aromatic hydrocarbons in air on
small spatial and temporal scales − II. Mass size distributions and gasparticle
partitioning. Atmos. Environ. 2010, 44, 5022−5027.
(49) Zhang, B.-Z.; Zhang, K.; Li, S.-M.; Wong, C. S.; Zeng, E. Y. Sizedependent
dry deposition of airborne polybrominated diphenyl ethers
in urban Guangzhou, China. Environ. Sci. Technol. 2012, 46, 7207−
7214.
(50) Chrysikou, L. P.; Samara, C. A. Seasonal variation of the size
distribution of urban particulate matter and associated organic
pollutants in the ambient air. Atmos. Environ. 2009, 43, 4557−4569.
(51) Degrendele, C.; Okonski, K.; Melymuk, L.; Landlová, L.;
Kukučka, P.; Cupr, P.; Klánová, J. Size specific distribution of the
atmospheric particulate PCDD/Fs, dl-PCBs and PAHs on a seasonal
scale: Implications for cancer risks from inhalation. Atmos. Environ.
2014, in press.
(52) Okonski, K.; Degrendele, C.; Melymuk, L.; Landlová, L.;
Kukučka, P.; Vojta, Š.; Kohoutek, J.; Klánová, J. The value of sizesegregated
data: Investigation of the size-fraction distribution of flame
retardants on atmospheric aerosols. Submitted.
(53) Lee, J. J.; Huang, K.-L.; Yu, Y. Y.; Chen, M. S. Laboratory
retention of vapor-phase PAHs using XAD adsorbents. Atmos. Environ.
2004, 38, 6185−6193.
(54) Barthel, P.; Thuens, S.; Shunthirasingham, C.; Westgate, J. N.;
Wania, F.; Radke, M. Application of XAD-resin based passive air
samplers to assess local (roadside) and regional patterns of persistent
organic pollutants. Environ. Pollut. 2012, 166, 218−225.
(55) Sun, P.; Basu, I.; Hites, R. A. Temporal trends of
polychlorinated biphenyls in precipitation and air at chicago. Environ.
Sci. Technol. 2006, 40, 1178−1183.
(56) Genualdi, S.; Lee, S. C.; Shoeib, M.; Gawor, A.; Ahrens, L.;
Harner, T. Global pilot study of legacy and emergin persistent organic
pollutants using sorbent-impregnated polyurethane foam disk passive
air samplers. Environ. Sci. Technol. 2010, 44, 5534−5539.
(57) Ahrens, L.; Harner, T.; Shoeib, M.; Lane, D. A.; Murphy, J. G.
Improved characterization of gas-particle partitioning for per- and
polyfluoroalkyl substances in the atmosphere using annular diffusion
denuder samplers. Environ. Sci. Technol. 2012, 46, 7199−7206.
(58) Kamprad, I.; Goss, K.-U. Systematic investigation of the
sorption properties of polyurethane foams for organic vapors. Anal.
Chem. 2007, 79, 4222−4227.
(59) Hayward, S. J.; Lei, Y. D.; Wania, F. Sorption of a diverse set of
organic chemical vapors onto XAD-2 resin: Measurement, prediction
and implications for air sampling. Atmos. Environ. 2011, 45, 296−302.
(60) Shoeib, M.; Harner, T.; Lee, S. C.; Lane, D. A.; Zhu, J. Sorbentimpregnated
polyurethane foam disk for passive air sampling of
volatile fluorinated chemicals. Anal. Chem. 2008, 80, 675−682.
(61) Clark, C. R.; Truex, T. J.; Lee, F. S. C.; Salmeen, I. T. Influence
of sampling filter type on the mutagenicity of diesel exhaust particulate
extracts. Atmos. Environ. 1981, 15, 397−402.
(62) Brown, A. S.; Yardley, R. E.; Quincey, P. G.; Butterfield, D. M.
Studies of the effect of humidity and other factors on some different
filter materials used for gravimetric measurements of ambient
particulate matter. Atmos. Environ. 2006, 40, 4670−4678.
(63) Lee, D. S.; Nicholson, K. W. The measurement of atmospheric
concentrations and deposition of semi-volatile organic compounds.
Environ. Monit. Assess. 1994, 32, 59−91.
(64) Ruuskanen, J.; Tuch, T.; Ten Brink, H.; Peters, A.; Khlystov, A.;
Mirme, A.; Kos, G. P. A.; Brunekreef, B.; Wichmann, H. E.; Buzorius,
G.; et al. Concentrations of ultrafine, fine and PM2.5 particles in three
European cities. Atmos. Environ. 2001, 35, 3729−3738.
(65) Cabada, J. C.; Rees, S.; Takahama, S.; Khlystov, A.; Pandis, S.
N.; Davidson, C. I.; Robinson, A. L. Mass size distributions and size
resolved chemical composition of fine particulate matter at the
Pittsburgh supersite. Atmos. Environ. 2004, 38, 3127−3141.
(66) Hites, R. A.; Basu, I.; Venier, M.; Salamova, A.; Nettesheim, T.;
Blanchard, P.; Backus, S. M.; Brice, K. A.; Froude, F. A.; Galarneau, E.
Technical Summary of the Progress of the Integrated Atmospheric
Deposition Network (IADN)2002−2008; Environment Canada,
2008; p 117.
(67) Pankow, J. F. Overview of the gas phase retention volume
behavior of organic compounds on polyurethane foam. Atmos. Environ.
1989, 23, 1107−1111.
(68) You, F.; Bidleman, T. F. Influence of volatility on the collection
of polycyclic aromatic hydrocarbon PAH vapors with polyurethane
foam. Environ. Sci. Technol. 1984, 18, 330−333.
(69) Harrad, S. J.; Hazrati, S.; Ibarra, C. Concentrations of
polychlorinated biphenyls in indoor air and polybrominated diphenyl
ethers in indoor air and dust in Birmingham, United Kingdom:
Implications for human exposure. Environ. Sci. Technol. 2006, 40,
4633−4638.
(70) Ter Schure, A. F. H.; Larsson, P.; Agrell, C.; Boon, J. P.
Atmospheric transport of polybrominated diphenyl ethers and
polychlorinated biphenyls to the Baltic Sea. Environ. Sci. Technol.
2004, 38, 1282−1287.
(71) Cotham, W. E.; Bidleman, T. F. Laboratory investigations of the
partitioning of organochlorine compounds between the gas phase and
atmospheric aerosols on glass fiber filters. Environ. Sci. Technol. 1992,
26, 469−478.
(72) Pankow, J. F.; Bidleman, T. F. Effects of temperature, TSP and
per cent nonexchangeable material in determining the gas particle
partitioning of organic compounds. Atmos. Environ. 1991, 25, 2241−
2249.
(73) Bunce, N. J.; Dryfhout, H. G. Diurnal and seasonal modelling of
the tropospheric half-lives of polycyclic aromatic hydrocarbons. Can. J.
Chem. 1992, 70, 1966−1970.
(74) Keyte, I. J.; Harrison, R. M.; Lammel, G. Chemical reactivity and
long-range transport potential of polycyclic aromatic hydrocarbons
A review. Chem. Soc. Rev. 2013, 42, 9333−9391.
(75) Mandalakis, M.; Berresheim, H.; Stephanou, E. G. Direct
evidence for destruction of polychlorobiphenyls by OH radicals in the
subtropical troposphere. Environ. Sci. Technol. 2003, 37, 542−547.
(76) Li, R.; Jin, J. Model predictions of toxaphene degradation in the
atmosphere over North America. Environ. Toxicol. Chem. 2013, 32,
2663−2671.
(77) Al Rashidi, M.; Chakir, A.; Roth, E. Heterogeneous oxidation of
folpet and dimethomorph by OH radicals: A kinetic and mechanistic
study. Atmos. Environ. 2014, 82, 164−171.
(78) Bai, J.; Yang, W.; Zhang, C.; Zhao, Y.; Gong, C.; Sun, X.; Zhang,
Q.; Wang, W. Theoretical study on the OH-initiated atmospheric
reaction of 1,1,1-trichloro-2,2-bis(4-chlorophenyl) ethane (DDT).
Atmos. Environ. 2013, 67, 177−183.
(79) Schenker, U.; Soltermann, F.; Scheringer, M.; Hungerbühler, K.
Modeling the environmental fate of polybrominated diphenyl ethers
(PBDEs): The importance of photolysis for the formation of lighter
PBDEs. Environ. Sci. Technol. 2008, 42, 9244−9249.
(80) Stapleton, H. M.; Dodder, N. G. Photodegradation of
decabromodiphenyl ether in house dust by natural sunlight. Environ.
Toxicol. Chem. 2008, 27, 306−312.
(81) Wang, J.; Chen, S.-J.; Nie, X.; Tian, M.; Luo, X.; An, T.-C.; Mai,
B. Photolytic degradation of decabromodiphenyl ethane (DBDPE).
Chemosphere 2012, 89, 844−849.
(82) Menichini, E. On-filter degradation of particle-bound benzo[a]pyrene
by ozone during air sampling: A review of the experimental
evidence of an artefact. Chemosphere 2009, 77, 1275−1284.
(83) Liu, K.; Duan, F.; He, K.; Ma, Y.; Cheng, Y. Investigation on
sampling artifacts of particle associated PAHs using ozone denuder
systems. Front. Environ. Sci. Eng. 2013, 8, 284−292.
(84) Tsapakis, M.; Stephanou, E. G. Collection of gas and particle
semi-volatile organic compounds: Use of an oxidant denuder to
minimize polycyclic aromatic hydrocarbons degradation during highvolume
air sampling. Atmos. Environ. 2003, 37, 4935−4944.
(85) Schauer, C.; Niessner, R.; Pöschl, U. Polycyclic aromatic
hydrocarbons in urban air particulate matter: Decadal and seasonal
trends, chemical degradation, and sampling artifacts. Environ. Sci.
Technol. 2003, 37, 2861−2868.
Environmental Science & Technology Critical Review
dx.doi.org/10.1021/es502164r | Environ. Sci. Technol. 2014, 48, 14077−1409114089
(86) Finlayson-Pitts, B. J.; Pitts, J. N. J. Chapter 10: Airborne
polycyclic aromatic hydrocarbons and their derivatives. In Chemistry of
the Upper and Lower Atmosphere; Elsevier Inc.: San Diego, 1999; pp
436−546.
(87) Fowler, D.; Amann, M.; Anderson, R.; Ashmore, M. R.;
Depledge, M.; Derwent, D.; Grennfelt, P.; Hewitt, C. N.; Hov, O.;
Jenkin, M.; et al. Ground-level ozone in the 21st century: Future
trends, impacts and policy implications Science Policy. Sci. Policy Rep. R.
Soc. 2008, 15/08.
(88) Bahm, K.; Khalil, M. A. K. A new model of tropospheric
hydroxyl radical concentrations. Chemosphere 2004, 54, 143−166.
(89) Finlayson-Pitts, B. J.; Pitts, J. N. J. Chemistry of the Upper and
Lower Atmosphere; Elsevier Inc.: San Diego, 1999.
(90) Bloss, W. J.; Evans, M. J.; Lee, J. D.; Sommariva, R.; Heard, D.
E.; Pilling, M. J. The oxidative capacity of the troposphere: Coupling
of field measurements of OH and a global chemistry transport model.
Faraday Discuss. 2005, 130, 425.
(91) Perraudin, E.; Budzinski, H.; Villenave, E. Kinetic study of the
reactions of ozone with polycyclic aromatic hydrocarbons adsorbed on
atmospheric model particles. J. Atmos. Chem. 2006, 56, 57−82.
(92) Atkinson, R. Atmospheric lifetimes of dibenzo-para-dioxins and
dibenzofurans. Sci. Total Environ. 1991, 104, 17−33.
(93) Bohlin, P.; Audy, O.; Škrdlíková, L.; Kukučka, P.; Vojta, Š.;
Přibylová, P.; Prokeš, R.; Čupr, P.; Klánová, J. Evaluation and
guidelines for using polyurethane foam (PUF) passive air samplers in
double-dome chambers to assess semi volatile organic compounds in
non-industrial indoor environments. Environ. Sci. Process. Impacts 2014,
Accepted, DOI: 10.1039/C4EM00305E.
(94) Bohlin, P.; Audy, O.; Škrdlíková, L.; Kukučka, P.; Přibylová, P.;
Prokeš, R.; Vojta, Š.; Klánová, J. Outdoor passive air monitoring of
semi volatile organic compounds (SVOCs): A critical evaluation of
performance and limitations of polyurethane foam (PUF) disks.
Environ. Sci. Process. Impacts 2014, 16, 433−444.
(95) Shoeib, M.; Harner, T. Characterization and comparison of
three passive air samplers for persistent organic pollutants. Environ. Sci.
Technol. 2002, 36, 4142−4151.
(96) Harner, T.; Su, K.; Genualdi, S.; Karpowicz, J.; Ahrens, L.;
Mihele, C.; Schuster, J. K.; Charland, J.-P.; Narayan, J. Calibration and
application of PUF disk passive air samplers for tracking polycyclic
aromatic compounds (PACs). Atmos. Environ. 2013, 75, 123−128.
(97) Heo, J.; Lee, G. Field-measured uptake rates of PCDDs/Fs and
dl-PCBs using PUF-disk passive air samplers in Gyeonggi-do, South
Korea. Sci. Total Environ. 2014, DOI: 10.1016/j.scitotenv.2014.03.073.
(98) Mari, M.; Schuhmacher, M.; Feliubadaló, J.; Domingo, J. L. Air
concentrations of PCDD/Fs, PCBs and PCNs using active and passive
air samplers. Chemosphere 2008, 70, 1637−1643.
(99) Pozo, K.; Harner, T.; Shoeib, M.; Urrutia, R.; Barra, R.; Parra,
O.; Focardi, S. Passive-sampler derived air concentrations of persistent
organic pollutants on a north-south transect in Chile. Environ. Sci.
Technol. 2004, 38, 6529−6537.
(100) Zhang, X.; Tsurukawa, M.; Nakano, T.; Lei, Y. D.; Wania, F.
Sampling medium side resistance to uptake of semivolatile organic
compounds in passive air samplers. Environ. Sci. Technol. 2011, 45,
10509−10515.
(101) Pozo, K.; Harner, T.; Lee, S. C.; Wania, F.; Muir, D. C. G.;
Jones, K. C. Seasonally resolved concentrations of persistent organic
pollutants in the global atmosphere from the first year of the GAPS
study. Environ. Sci. Technol. 2009, 43, 796−803.
(102) Persoon, C.; Hornbuckle, K. C. Calculation of passive sampling
rates from both native PCBs and depuration compounds in indoor and
outdoor environments. Chemosphere 2009, 74, 917−923.
(103) Santiago, E. C.; Cayetano, M. G. Polycyclic aromatic
hydrocarbons in ambient air in the Philippines derived from passive
sampler with polyurethane foam disk. Atmos. Environ. 2007, 41, 4138−
4147.
(104) Wilford, B. H.; Harner, T.; Zhu, J.; Shoeib, M.; Jones, K. C.
Passive sampling survey of polybrominated diphenyl ether flame
retardants in indoor and outdoor air in Ottawa, Canada: Implications
for sources and exposure. Environ. Sci. Technol. 2004, 38, 5312−5318.
(105) Klánová, J.; Cupr, P.; Kohoutek, J.; Harner, T. Assessing the
influence of meteorological parameters on the performance of
polyurethane foam-based passive air samplers. Environ. Sci. Technol.
2008, 42, 550−555.
(106) Armitage, J. M.; Hayward, S. J.; Wania, F. Modeling the uptake
of neutral organic chemicals on XAD passive air samplers under
variable temperatures, external wind speeds and ambient air
concentrations (PAS-SIM). Environ. Sci. Technol. 2013, 47, 13546−
13554.
(107) Petrich, N. T.; Spak, S. N.; Carmichael, G. R.; Hu, D.;
Martinez, A.; Hornbuckle, K. C. Simulating and explaining passive air
sampling rates for semivolatile compounds on polyurethane foam
passive samplers. Environ. Sci. Technol. 2013, 47, 8591−8598.
(108) Zhang, X.; Wong, C.; Lei, Y. D.; Wania, F. Influence of sampler
configuration on the uptake kinetics of a passive air sampler. Environ.
Sci. Technol. 2012, 46, 397−403.
(109) Dušek, L.; Borůvková, J.; Hůlek, R.; Jarkovský, J.; Kubásek, M.;
Gregor, J.; Šebková, K.; Hřebíček, J.; Holoubek, I. GENASISGlobal
environmental assessment and information system. In Environmental
Software Systems; Springer, 2011; online.
(110) Ahrens, L.; Shoeib, M.; Harner, T.; Koblizková, M.; Reiner, E.
J. Characterization of two passive air samplers for per- and
polyfluoroalkyl substances. Organohalogen Compd. 2012, 74, 339−342.
(111) Genualdi, S.; Harner, T.; Cheng, Y.; Macleod, M.; Hansen, K.
M.; van Egmond, R.; Shoeib, M.; Lee, S. C. Global distribution of
linear and cyclic volatile methyl siloxanes in air. Environ. Sci. Technol.
2011, 45, 3349−3354.
(112) Saini, A.; Chaudhuri, S. R.; Goosey, E.; Diamond, M. L.;
Shoeib, M. Calibration of passive air sampler for monitoring phthalates
in indoor air; Society for Environmental Toxicology and Chemistry
European Annual Meeting; SETAC: Glasgow, 2013.
(113) Robson, M. Personal communication; Toronto, Ontario,
Canada, 2014; [email communication with L. Melymuk, 21−4−2014].
(114) Chaemfa, C.; Barber, J. L.; Gocht, T.; Harner, T.; Holoubek, I.;
Klánová, J.; Jones, K. C. Field calibration of polyurethane foam (PUF)
disk passive air samplers for PCBs and OC pesticides. Environ. Pollut.
2008, 156, 1290−1297.
(115) Tuduri, L.; Harner, T.; Hung, H. Polyurethane foam (PUF)
disks passive air samplers: Wind effect on sampling rates. Environ.
Pollut. 2006, 144, 377−383.
(116) Tao, S.; Liu, Y.; Xu, W.; Lang, C.; Liu, S.; Dou, H.; Liu, W.
Calibration of a passive sampler for both gaseous and particulate phase
polycyclic aromatic hydrocarbons. Environ. Sci. Technol. 2007, 41,
568−573.
(117) Abdallah, M. A.-E.; Harrad, S. J. Modification and calibration of
a passive air sampler for monitoring vapor and particulate phase
brominated flame retardants in indoor air: Application to car interiors.
Environ. Sci. Technol. 2010, 44, 3059−3065.
(118) May, A. A.; Ashman, P.; Huang, J.; Dhaniyala, S.; Holsen, T.
M. Evaluation of the polyurethane foam (PUF) disk passive air
sampler: Computational modeling and experimental measurements.
Atmos. Environ. 2011, 45, 4354−4359.
(119) Kennedy, K. E.; Hawker, D. W.; Bartkow, M. E.; Carter, S.;
Ishikawa, Y.; Müller, J. F. The potential effect of differential ambient
and deployment chamber temperatures on PRC derived sampling
rates with polyurethane foam (PUF) passive air samplers. Environ.
Pollut. 2010, 158, 142−147.
(120) Zhang, X.; Brown, T. N.; Ansari, A.; Yeun, B.; Kitaoka, K.;
Kondo, A.; Lei, Y. D.; Wania, F. Effect of wind on the chemical uptake
kinetics of a passive air sampler. Environ. Sci. Technol. 2013, 47, 7868−
7875.
(121) Bohlin, P.; Jones, K. C.; Levin, J.-O.; Lindahl, R.; Strandberg,
B. Field evaluation of a passive personal air sampler for screening of
PAH exposure in workplaces. J. Environ. Monit. 2010, 12, 1437−1444.
(122) Covaci, A.; Voorspoels, S.; de Boer, J. Determination of
brominated flame retardants, with emphasis on polybrominated
diphenyl ethers (PBDEs) in environmental and human samplesA
review. Environ. Int. 2003, 29, 735−756.
Environmental Science & Technology Critical Review
dx.doi.org/10.1021/es502164r | Environ. Sci. Technol. 2014, 48, 14077−1409114090
(123) Albinet, A.; Leoz-Garziandia, E.; Budzinski, H.; ViIlenave, E.
Sampling precautions for the measurement of nitrated polycyclic
aromatic hydrocarbons in ambient air. Atmos. Environ. 2007, 41,
4988−4994.
(124) Muir, D. C. G.; Howard, P. H. Are there other persistent
organic pollutants? A challenge for environmental chemists. Environ.
Sci. Technol. 2006, 40, 7157−7166.
(125) Cape, J. N.; Bealey, W.; Coyle, M.; DiMarco, C.; Kentisbeer, J.;
Leaver, D.; Malcolm, H.; Nemitz, E.; Simmons, I.; Twigg, M.; et al.
Operation of EMEP “Supersites” in the United KingdomAnnual Report
for 2008; Centre for Ecology and Hydrology: Edinburgh, 2010; 112 p.
(126) Harner, T.; Bidleman, T. F. Octanol-air partition coefficient for
describing particle/gas partitioning of aromatic compounds in urban
air. Environ. Sci. Technol. 1998, 32, 1494−1502.
(127) Pankow, J. An absorption model of gas/particle partitioning of
organic compounds in the atmosphere. Atmos. Environ. 1994, 28, 185−
188.
(128) Hart, K. M.; Isabelle, L. M.; Pankow, J. F. High-volume air
sampler for particle and gas sampling. 1. Design and gas sampling
performance. Environ. Sci. Technol. 1992, 26, 1048−1052.
(129) Research Centre for Toxic Compounds in the Environment.
MonAirNet Project webpage. http://www.monairnet.eu/index-en.php.
(130) Ferranti, E. J. S.; Fryer, M.; Sweetman, A. J.; Solera Garcia, M.
A.; Timmis, R. J. Field-testing a new directional passive air sampler for
fugitive dust in a complex industrial source environment. Environ. Sci.
Process. Impacts 2014, 16, 159−168.
(131) Beyer, A.; Wania, F.; Gouin, T.; Mackay, D.; Matthies, M.
Selecting internally consistent physicochemical properties of organic
compounds. Environ. Toxicol. Chem. 2002, 21, 941−953.
(132) Paasivirta, J.; Sinkkonen, S. I.; Mikkelson, P.; Rantio, T.;
Wania, F. Estimation of vapor pressures, solubilities and Henry’s law
constants of selected persistent organic pollutants as functions of
temperature. Chemosphere 1999, 39, 811−832.
(133) Harner, T.; Shoeib, M. Measurements of octanol-air partition
coefficients (KOA) for polybrominated diphenyl ethers (PBDEs):
Predicting partitioning in the environment. J. Chem. Eng. Data 2002,
47, 228−232.
(134) Tittlemier, S. A.; Halldorson, T.; Stern, G. A.; Tomy, G. T.
Vapor pressures, aqueous solubilities, and Henry’s law constants of
some brominated flame retardants. Environ. Toxicol. Chem. 2002, 21,
1804−1810.
(135) Wang, Z.-Y.; Zeng, X.-L.; Zhai, Z.-C. Prediction of supercooled
liquid vapor pressures and n-octanol/air partition coefficients for
polybrominated diphenyl ethers by means of molecular descriptors
from DFT method. Sci. Total Environ. 2008, 389, 296−305.
(136) Odabasi, M.; Cetin, B. Determination of octanol-air partition
coefficients of organochlorine pesticides (OCPs) as a function of
temperature: Application to air-soil exchange. J. Environ. Manage.
2012, 113, 432−439.
(137) U.S. Environmental Protection Agency. Estimation Programs
Interface Suite (EPI Suite) for Microsoft Windows, v 4.11, 2012.
(138) Armstrong, J. L.; Fenske, R. A.; Yost, M. G.; Tchong-French,
M.; Yu, J. Comparison of polyurethane foam and XAD-2 sampling
matrices to measure airborne organophosphorus pesticides and their
oxygen analogs in an agricultural community. Chemosphere 2013, 92,
451−457.
(139) Peters, A. J.; Lane, D. A.; Gundel, L. A.; Northcott, G. L.;
Jones, K. C. A comparison of high volume and diffusion denuder
samplers for measuring semivolatile organic compounds in the
atmosphere. Environ. Sci. Technol. 2000, 34, 5001−5006.
(140) Goriaux, M.; Jourdain, B.; Temime, B.; Besombes, J.-L.;
Marchand, N.; Albinet, A.; Leoz-Garziandia, E.; Wortham, H. Field
comparison of particulate PAH measurements using a low-flow
denuder device and conventional sampling systems. Environ. Sci.
Technol. 2006, 40, 6398−6404.
Environmental Science & Technology Critical Review
dx.doi.org/10.1021/es502164r | Environ. Sci. Technol. 2014, 48, 14077−1409114091
Karásková, Pavlína, Garry P. Codling, Lisa Melymuk, and Jana Klánová. 2018. "A Critical
Assessment of Passive Air Samplers for Per- and Polyfluoroalkyl Substances." Atmospheric
Environment 185: 186-95. https://doi.org/10.1016/j.atmosenv.2018.05.030
Contents lists available at ScienceDirect
Atmospheric Environment
journal homepage: www.elsevier.com/locate/atmosenv
A critical assessment of passive air samplers for per- and polyfluoroalkyl
substances
Pavlína Karásková, Garry Codling∗
, Lisa Melymuk, Jana Klánová
Research Centre for Toxic Compounds in the Environment (RECETOX), Masaryk University, Kamenice 753/5, 625 00 Brno, Czech Republic
A R T I C L E I N F O
Keywords:
PFASs
Calibration study
Passive air samplers
Perfluorinated alkyl substances
PUF-PAS
A B S T R A C T
Since their inclusion in the Stockholm Convention, there has been a need for global monitoring of perfluorooctane
sulfonate (PFOS), its salts and perfluorooctanesulfonyl fluoride (PFOSF), along with other nonlisted
highly fluorinated compounds. Passive air samplers (PAS) are ideal for geographic coverage of atmospheric
monitoring. The most common type of PAS, using polyurethane foam (PUF) as a sorbent, was primarily
developed for non-polar semivolatile organic compounds (SVOCs) and are not well-validated for polar substances
such as the per- and polyfluoroalkyl substances (PFASs), however, they have been used for some PFASs,
particularly PFOS. To evaluate their applicability, PAS were deployed for measurement of PFASs in outdoor and
indoor air. Outdoors, two types of PAS, one consisting of PUF and one of XAD-2 resin, were deployed in an 18week
calibration study in parallel with a low-volume active air sampler (LV-AAS) in a suburban area. Indoors,
PUF-PAS were similarly deployed over 12 weeks to evaluate their applicability for indoor monitoring. Samples
were analysed for perfluoroalkyl carboxylic acids (PFCAs), perfluoroalkyl sulfonates (PFSAs), perfluorooctane
sulfonamides (FOSAs), and perfluorooctane sulfonamidoethanols (FOSEs). In outdoor air, 17 out of the 21 PFAS
were detected in more than 50% of samples, with a median ∑17PFASs of 18.0 pg m−3
while 20 compounds were
detected in indoor air with a median concentration ∑20PFASs of 76.6 pg m−3
using AAS samplers. PFOS was the
most common PFAS in the outdoor air while PFBA was most common indoors. Variability between PAS and AAS
was observed and comparing gas phase and particle phase separately or in combination did not account for the
variation observed. PUF-PAS may still have a valuable use in PFAS monitoring but more work is needed to
identify the applicability of passive samplers for ionic PFAS.
1. Introduction
Per- and polyfluoroalkyl substances (PFASs) are a diverse group of
industrial chemicals including perfluoroalkyl carboxylic acids (PFCAs),
perfluoroalkyl sulfonates (PFSAs), perfluoorooctane sulfonamides
(FOSAs) and perflurooctane sulfonamidoethanols (FOSEs). They have
been widely used for over 60 years as surfactants, lubricants, paper and
textile coatings, polishes, food packaging, and fire-fighting foams
(Prevedouros et al., 2006).
PFASs are known for unique physical and chemical properties, their
global distribution in the ocean and the atmosphere, their persistence,
bioaccumulation and potential toxicity (Giesy and Kannan, 2002).
PFASs have been detected globally in a variety of environmental media
(Langer et al., 2010; Shoeib et al., 2010; Gomez et al., 2011; Sun et al.,
2011; Wang et al., 2011; Yang et al., 2011; Codling et al., 2014; Hung
et al., 2016), biota (Giesy and Kannan, 2002), humans (Wu et al., 2017;
Poothong et al., 2017), and have been detected in remote regions due to
long-range atmospheric transport of the PFAS and their volatile precursors,
(Ahrens et al., 2011a,b), or transport via the aquatic system
and marine aerosols (Benskin et al., 2012).
A major concern in human exposure to PFASs is the effect of indoor
exposure and its potential to affect vulnerable groups such as pregnant
mothers and young children (Shoeib et al., 2005). Moreover, since the
listing of PFOS, its salts and their precursors under the Stockholm
Convention on Persistent Organic Pollutants (POPs) in 2009, the need
for consistent and comparable global monitoring is even greater, as
measurement of these compounds is a requirement under the Global
Monitoring Plan (GMP) (Klánová and Harner, 2013).
Previous atmospheric measurements of PFASs were often based on
the use of high-volume (HV) or low-volume (LV) active air samplers
(AAS), usually with polyurethane foam (PUF) and/or polystyrene-divinyl
benzene copolymeric resin (XAD) as the gas-phase sorbent
(Ahrens et al., 2011a,b). While AAS can provide information about gasparticle
partitioning and temporal resolution, AAS has limited ability to
https://doi.org/10.1016/j.atmosenv.2018.05.030
Received 2 November 2017; Received in revised form 11 May 2018; Accepted 13 May 2018
∗
Corresponding author. Masaryk University, RECETOX, Faculty of Science, Kamenice 753/5, 625 00 Brno, Czech Republic.
E-mail addresses: pavlina.karaskova@recetox.muni.cz (P. Karásková), codling@recetox.muni.cz (G. Codling), melymuk@recetox.muni.cz (L. Melymuk),
klanova@recetox.muni.cz (J. Klánová).
Atmospheric Environment 185 (2018) 186–195
Available online 14 May 2018
1352-2310/ © 2018 Elsevier Ltd. All rights reserved.
T
provide the spatial coverage needed to understand global distributions
of PFAS (Koblizkova et al., 2012; Shoeib et al., 2011), particularly in
remote areas, and are often not ideal for indoor locations due to the
disruptive nature of AAS. In human exposure assessment, personal AAS
systems highlight that single location assessment does not reflect actual
human exposure patterns (Padilla-Sánchez et al., 2017). For these
purposes, passive air samplers (PAS) are an alternative. They are silent,
normally far cheaper and require no power supply, so more PAS may be
employed at less cost than one AAS.
Various PAS have been developed thus far, primarily for monitoring
of non-polar semi-volatile organic compounds (SVOCs) in air (Shoeib
and Harner, 2002; Wania et al., 2003), and some PAS have also been
used to quantify more polar compounds such as the PFASs. Recently,
several studies have reported PFAS levels using conventional PUF-PAS
(Chaemfa et al., 2010; Liu et al., 2015), or PUF impregnated with XAD-
2 resin (a sorbent impregnated PUF disk, or SIP) (Schuster et al., 2012;
Ahrens et al., 2013) or with XAD-2 resin alone (Koblizkova et al., 2012).
Reported PFAS concentrations vary between these different PAS
methods (Ahrens et al., 2013), and the comparison of reported concentrations
could be questionable due to different sampling techniques,
sampling media, in addition to differences in the accuracy and precision
of analytical methods.
This study evaluates the performance of PAS for four PFAS classes
(PFCAs, PFSAs, FOSAs, FOSEs) in outdoor and indoor ambient air by
comparison with concentrations determined using AAS. A sampling
campaign conducted in 2012 and 2013 used LV-AAS and two types of
PAS outdoor (PUF disks and XAD cartridges) and PUF-PAS indoor. The
purpose of this survey was to assess PFAS concentrations and distributions
in air and evaluate PAS performance through determination
of sampler uptake over time. This provides information on how studies
using PUF to quantify PFASs can be interpreted, and whether existing
sampling infrastructure (e.g., global PAS networks using PUF-PAS) can
be easily adapted to include PFAS, given the importance of providing
global monitoring data under the GMP. AAS gas- and particle-phase
fractions were analysed separately, enabling determination of gas-particle
partitioning to complement the interpretation of the PAS performance.
We determined sampling rates of individual target PFASs for
PUF and XAD-PAS, and make recommendations for monitoring of
PFASs in air via PAS.
2. Materials and methods
2.1. Materials
Twenty-one analytical standards (11 PFCAs, 5 PFASs, FOSA, NMeFOSA,
N-EtFOSA, N-MeFOSE and N-EtFOSE) and 13 mass-labelled
standards (MPFBA, MPFHxA, MPFOA, M8PFOA, MPFNA, MPFDA,
MPFUnDA, MPFDoDA, MPFHxS, MPFOS, M8PFOS, dMeFOSA,
dMeFOSE) were purchased from Wellington Laboratories Inc. (Guelph,
Ontario, Canada). For details, including full IUPAC names see Table S1
in the Supplementary Information. LC/MS-grade methanol was obtained
from Biosolve b.v. (Valkenswaard, The Netherlands). HPLCgrade
water was from Fisher Scientific (Loughborough, UK). Ammonium
acetate, used as an addition to the extraction solvent and mobile
phase, was puriss p.a. grade (≥98.0%), and was obtained from Fluka
(Fluka Chemie GmbH, Buchs, Germany). All chemicals and solvents
were used without further purification.
2.2. Air sampling
LV-AAS and PUF-PAS were deployed simultaneously outdoors at a
suburban background site on the roof of the Research Centre for Toxic
Compounds in the Environment (RECETOX) at Masaryk University in
Brno, Czech Republic (49.1782 N, 16.5711 E, Figure S1) between late
April and early September 2013 (18 weeks – see Table S2). Thirty
polyurethane foam passive samplers (OUT-PUF-PAS) were deployed at
the start of the study, consisting of two stainless steel bowls (24 diameter
lower bowl and 30 cm diameter upper bowl) surrounding a PUF
disk (15 cm in diameter x 1.5 cm thick, with a density of 0.030 g cm−3
,
type T-3037, Molitan a.s., Czech Republic).
A set of triplicate OUT-PUF-PAS was collected at either 1- or 2week
intervals, generating a total of 10 sets of triplicate OUT-PUF-PAS,
each corresponding to a specific exposure time. An LV-AAS (OUT-LVAAS)
(Leckel MVS6, Sven Leckel, Ingenieurbüro GmbH, Germany) was
run continuously as a reference sampler to provide weekly time-integrated
concentrations of the targeted PFASs. Sampler flow rates were
2.3 m3
h−1
and each sample duration was one week, resulting in an
average 373.5 m3
(Table S2) per sample and generating 18 sets of reference
samples. The sampling train consisted of a Whatman®
47 mm ø
quartz fiber filter (QFF, GE healthcare, Chicago, USA) to collect the
particle phase and PUF/XAD/PUF sandwiches with 15 g of XAD-2 resin
as sorbent for gas-phase.
To complement the outdoor PUF-PAS calibration, two other PAS
were evaluated. To provide insight into the influence of sorbent and
sampler choice on PFAS uptake, three XAD-PAS were also deployed in
outdoor air in conjunction with the PUF-PAS calibration. XAD-PAS
have a lower uptake rate and typically require a longer deployment
time to collect sufficient target compound (e.g., 1 year; Wania et al.,
2003), so XAD-PAS were not collected on a weekly basis, but rather
only at 70, 98 and 126 days. The XAD-PAS consisted of steel cartridges
filled by 10 g of XAD-2 resin (OUT-XAD-PAS) (Wania et al., 2003).
Triplicates were not collected as this was only intended to screen the
possibility of using XAD-PAS for shorter durations. Additionally, to
compare PUF-PAS performance in indoor vs. outdoor air, a similar calibration
study in indoor air was evaluated. Indoor air samples (IN)
were collected from a university lecture room between late January and
late April 2012 (12 weeks – see Table S3). The indoor samples were
from a carpeted room in a new building (completed in 2011). The indoor
air calibration consisted of PUF-PAS (IN-PUF-PAS) and a reference
LV-AAS (IN-LV-AAS). The PUF-PAS were the same double-bowl PUFPAS
as used in outdoor air, and the LV-AAS used PUF as gas-phase
sorbent and QFF to collect the particle phase. Triplicate PUF-PAS were
collected every 7 days, and the LV-AAS was run continuously, with
sampling media changed every week. All samples were wrapped in
aluminium foil and stored at −18 C until extraction.
2.3. Extraction and instrumental analysis
Prior to extraction all samples were spiked with labelled standards
(M8PFOA, M8PFOS) to determine recoveries. All samples except XADPAS
were extracted with 5 mM ammonium acetate in methanol using a
B-811 automated extraction unit (Büchi, Switzerland). PUF/XAD/PUF
sandwiches were extracted as one unit. Extraction consisted of 60 min
warm Soxhlet followed by 30 min of solvent rinsing and a concentration
step to 1 mL under a stream of nitrogen. XAD-PAS were extracted by
sonication (3 times with 30 mL MeOH) and concentrated to 1 mL. The
concentrated extracts were filtered (nylon membrane, 13 mm diameter
and 0.45 μm pore size) and transferred into polypropylene centrifuge
tubes (Alpha Laboratories, UK). The filtrate was concentrated to 0.5 mL
under a gentle stream of nitrogen and diluted with 0.5 mL 5 mM ammonium
acetate in water. Extracts were centrifuged (1800 G, 10 min)
and 100 μL transferred to LC minivials for analysis (Labicom, Czech
Republic).
Separation, identification and quantification of target PFASs was
performed using high performance liquid chromatography (UHPLC)
with an Agilent 1290 (Agilent Technologies, Palo, Alto, California,
USA) connected to a QTRAP 5500 mass spectrometer (AB Sciex, Foster
City, California, USA). Chromatographic separation was performed at
20 °C on a SYNERGI 4 μm Fusion RP 80Ä 50 mm × 2 mm column with a
corresponding 4 × 2.00 mm precolumn (Phenomenex, USA). The flow
rate was 200 μL min−1
, and mobile phases consisted of methanol/5 mM
ammonium acetate in water at 55/45 (v/v; A), and methanol (B). The
P. Karásková et al. Atmospheric Environment 185 (2018) 186–195
187
gradient elution used for analyte separation has been reported previously
(Karásková et al., 2016), with 10 μL injected onto the column
with mass-labelled standards added automatically via the autosampler
(1 μL of a 100 ng mL−1
solution). The mass spectrometer was operated
in electrospray negative ionization mode (ESI-) using two MRM transitions
for each compound, except PFBA, N-MeFOSE and N-EtFOSE, at
450 °C and ion voltage 4500 V (see Table S1 for ions and conditions for
each compound). Analyst 1.6.1 was used for data integration and
evaluation.
The LV-AAS (indoor and outdoor) was considered a reference
sampler to monitor air concentrations of PFASs, as well as to calculate
average concentrations of each target analyte for PAS calibration.
2.4. QA/QC
Three QA/QC procedures were used: (a) spike/recovery experiment;
(b) monitoring of individual sample recoveries; (c) laboratory and field
blanks.
Prior to deployment, the PUF and XAD-2 resin were pre-extracted
by Soxhlet for 8 h in acetone and 8 h in methanol, dried in a fume
chamber and stored in two layers of aluminium foil and sealed in ziplock
bags. After exposure, all matrices were wrapped in aluminium foil,
labelled, placed into zip-lock bags and stored at −20 °C until analysis.
The efficiency of the extraction method was tested in a spiking experiment
using PUF, QFF and XAD with the addition of native PFASs.
Recovery of native analytes measured in spiked matrices varied from 63
to 113% for PUFs, from 61 to 112% for QFFs and from 75 to 108% for
XAD-2 resins (detailed in Table S4).
Recovery correction was applied to each sample based on the labelled
compounds added prior to extraction. The average percent recovery
for those labelled target compounds depended on sampling
matrices, ranging from 47 ± 4.6% for M8PFOA in PUF-PAS to
102 ± 7.3% for M8PFOS in PUF/XAD/PUF-LV-AAS (detailed in Table
S5).
Three PUF-PAS, and 2 QFFs and 2 XAD-2 field blanks were analysed
with each set of PUF-PAS and LV-AAS samples. Results from field
blanks were used to determine method detection limits (MDLs). MDLs
were calculated as the mean blank value plus 3 standard deviations.
When compounds were not detected in any blanks the instrumental
detection limits (IDLs) were used as MDLs. Samples with masses below
MDLs are reported as “ < MDL”. √2/2*MDL was substituted if appropriate
for the statistical analyses being performed (Antweiler, 2015).
For data visualization, GraphPad Prism 5 (GraphPad Software, Inc.,
USA) and Excel 2007 (Microsoft Corporation, USA) were used. The
results were exported to the software program SIMCA 14.0 (Umetrics,
Sweden) for multivariate data analysis. Principle component analysis
(PCA) was used to identify relationships between groups (gas and
particle phase, different PAS types, PAS and AAS bulk samples). PCA
was performed using 50% variable and observational tolerance with
variance scaling.
3. Results and discussion
3.1. Air concentration and composition
All of the 21 target PFASs were detected in at least one OUT-LV-AAS
sample. In the sum of gas and particle AAS (OUTtotalAAS), 17 out of 21
compounds were detected in > 50% of samples, and 13 compounds
in > 90% (Table S6). PFUnDA and PFDoDA were detected in < 10% of
outdoor samples and are excluded from further interpretation. The
composition, concentrations, and temporal trends of the 21 target
PFASs in OUTtotalAAS samples are provided in Fig. 1 and additional
data on PFAS concentrations in Table S7.
The range of the ∑21PFASs in OUTtotalAAS samples was
10.4–28.1 pg m−3
with a median of 18.0 pg m−3
, and dominated by
PFOS (28% of total PFASs). Ionic PFASs were dominated by PFOS
(median 5.11 pg m−3
, range: 1.9–7.68 pg m−3
), PFOA (median
2.11 pg m−3
, range 1.11–5.47 pg m−3
), PFBA (median 1.78 pg m−3
,
range 0.08–8.95 pg m−3
), and PFHpA (median 1.11 pg m−3
, range
0.39–1.89 pg m−3
). The total outdoor concentrations of ionic PFASs
were consistent with previous studies, with PFOS the most abundant
compound, followed by PFOA and the shorter chain PFCAs (Barber
et al., 2007; Kim and Kannan, 2007; Stock et al., 2007; Dreyer and
Ebinghaus, 2009; Weinberg et al., 2011a,b; Liu et al., 2015).
Concentrations of the neutral PFASs, ΣFOSA/Es in OUTtotalAAS
samples, ranged from 1.61 to 11.7 pg m−3
, and were dominated by NMeFOSE
(57% of ΣFOSA/Es, 15% of total PFASs) with a median concentration
of 1.98 pg m−3
and a range of 0.58–7.97 pg m−3
. Ahrens
et al., 2011a,b has reported a wide range of concentrations of neutral
PFASs in the atmosphere: the ΣFOSA/Es near waste treatment sites in
Canada ranged from 14.3 to 124 pg m−3
while in the Canadian Arctic
concentrations of ∑FOSA/E ranged from 0.4 to 21 pg m−3
. FOSA/Es
concentrations measured in the present study were more consistent
with rural and remote locations than dense urban or waste disposal sites
(Ahrens et al., 2011a,b; Vierke et al., 2011; Weinberg et al., 2011a,b).
Likewise, Barber et al. (2007) reported concentrations of volatile PFASs
from Northwest Europe which are one or two orders of magnitude
greater than in this study. Comparable concentrations of FOSA/Es were
reported by Dreyer and Ebinghaus (2009) during land and sea sampling
from Hamburg and into the North Sea. N-MeFOSE is a perfluorooctane
sulfonamide that was used primarily in carpets and clothing, however,
it should also be noted that the lower concentration observed in this
present study may reflect changes in use pattern, as rapid shifts in PFAS
use have been observed since some compounds became listed under the
Stockholm Convention and others came under greater scrutiny (Butt
et al., 2010).
In indoor air all of the 20 PFASs compounds were detected in at
least one sample. Eighteen out of 20 compounds were detected in all
total gas + particle (INtotalAAS) samples (Table S8). PFHxA and PFHpA
were detected in < 10% of samples, and only in the particle phase, and
are excluded from further discussion.
The median indoor concentrations measured by AAS for PFAS were
more than three times greater than outdoor (Fig. 2,Table S8), and were
dominated by PFBA (∼37% of Σ20PFASs in total gas + particle
(INtotalAAS) samples) and PFPA (22%), associated primarily with particles.
Of the more volatile compounds, N-MeFOSE and N-EtFOSE were
dominant, representing 25% and 12% of Σ18PFASs in gas phase, re-
spectively.
The median concentration measured by INtotalAAS was 76.6 pg m−3
with a range of 41.8–161.6 pg m−3
. This is lower than that observed in
a study of German schools for neutral PFASs, where the N-EtFOSA
median concentration was 243 pg m−3
, compared to 0.7 pg m−3
in this
study (Fromme et al., 2015). However, in that study the PUF/XAD/PUF
sandwich was used while in our study PUF only was used, which may
have resulted in limited sorptive capacity for gaseous PFASs. In
Norway, a survey of air and house dust for PFAS concluded that
building age, specific furnishings and clothing play a role in indoor
PFAS concentrations (Haug et al., 2011). Thus, the low concentrations
of PFAS in our indoor samples may also be due to the recent construction
of the building (2011), and new furnishings with traditional
PFAS-containing materials no longer in use.
3.2. Passive air sampler calibration
3.2.1. Outdoor PAS
Passive sampler performance was evaluated by calculating equivalent
air sample volumes for PUF-PAS and XAD-PAS. The equivalent air
volume (Veq, in m3
) sampled by PUF disk was determined based on the
ambient air concentration and accumulated mass using equation:
Veq = M/CA (Eq. 1)
P. Karásková et al. Atmospheric Environment 185 (2018) 186–195
188
Fig. 1. Gas + particle phase concentrations (pg m−3
) of individual PFASs compounds in outdoor air measured by active air sampler (OUTtotalAAS samples) during
outdoor PAS calibration.
Fig. 2. Gas + particle phase concentrations (pg m−3
) of individual target compounds in indoor air measured by active air sampler (INtotalAAS) during indoor PAS
calibration (inset figure excludes PFBA and PFPA).
P. Karásková et al. Atmospheric Environment 185 (2018) 186–195
189
Where M is the mass of compounds accumulated on the PUF disk (pg),
and CA is the concentration of target compounds in ambient air (pg
m−3
), measured with the reference AAS. The Veq values were then
plotted against deployment time to evaluate the uptake characteristics
of the PAS. The linearity of uptake to PAS was tested using least squares
regression. Calibrations were evaluated using only the gas-phase air
concentrations and also using total air concentrations, to provide insight
into whether the PAS are sampling gas-phase only or total air.
For air concentrations to be determined from PAS, the sampler
ideally should have consistent uptake of a target compound over time,
and not equilibrate within the sampler deployment period, i.e., a strong
linear relationship between Veq and time with a positive slope. If these
criteria are met, the slope of the least squares regression line is the
passive air sampling rate (Rs, m3
day−1
) of the PUF disk for the corresponding
PFAS (see Table S9). Plots of equivalent air volume and
sampling time are shown in Fig. 3.
For comparison, the sampling rate Rs was also determined based on
individual time points according to Equation (2), where t is the deployment
time. Results for this method are shown in Table S10.
Rs = M/CA × t (Eq. 2)
Sampling rates were not evaluated for compounds detected in
≤50% of total LV-AAS samples (PFUnDA, PFDoDA, PFTrDA and
PFHxS).
When calculated Veq values of OUT-PUF-PAS were plotted against
deployment time, a linear uptake (R2
> 0.5) was observed for three
PFAS (PFBA, PFPA and N-MeFOSE) (Table S9, Fig. 3), suggesting that
for these compounds the majority of the variability in the Veq is explained
by deployment time, and thus that the PUF-PAS can perform as
appropriate passive samplers for these compounds. PFHpS, PFOS, FOSA
and N-MeFOSA also had statistically significant linear uptake, but more
scatter in the data and poorer fit of the regression. Despite significant
correlations for seven of the eight PFASs, the magnitude of the slopes,
representing the PAS sampling rates, was different than what is typical
of non-polar compounds in outdoor double-bowl PUF-PAS, which typically
have sampling rates of 2–6 m3
day−1
(Bohlin et al. 2014a,
2014b). Slopes for regressions with r2
> 0.5 ranged from 0.692 to
30.4 m3
day−1
. For many compounds there was no significant difference
between gas-phase and total sampling rates, however for some
compounds the sampling rates determined based on total air concentrations
rather than gas-phase only were lower (e.g., PFPA, PFHpS),
suggesting that the PUF-PAS are capturing some fraction of particlephase
compounds, and thus that they are performing more like total air
samplers rather than gas-phase only.
Ahrens et al. (2013) evaluated the performance of PUF-PAS for a
range of PFASs; in that calibration, PFSAs were found to have consistent
uptake in the PUF-PAS with a sampling rate of approximately 3.9 m3
day−1
; this was within the range of what was observed for PFSAs in our
study where sampling rates for PFSAs ranged from 0.45 to 30 m3
day−1
.
However, Ahrens et al. did not detect any PFCAs by PUF-PAS; this
contrasts with our results where most consistent linear uptake was
observed for the PFCAs (PFBA and PFPA, with sampling rates of 30 and
26 m3
day−1
, respectively; Fig. 3). Moreover, Ahrens et al. found rapid
equilibration of FOSA, methyl and ethyl FOSAs and methyl and ethyl
FOSEs after a few weeks, whereas we observed linear uptake for at least
100 days for N-MeFOSE, more variable but still increasing uptake for
FOSA and N-MeFOSA, but no linear uptake for N-EtFOSA. N-MeFOSE is
the least volatile of the FOSE/FOSA compounds, which may account for
its better performance in the passive sampler (vapour pressure of
0.0004 Pa compared with > 0.002 Pa for other FOSE/FOSA compounds)
(Shoeib et al., 2004).
As only three OUT-XAD-PAS were deployed, we did not apply a
least squares regression to the plot of Veq vs. deployment time. To obtain
an indication of the sampling rates, the sampling rate was determined
based on individual time points for sampling using Equation
(2). The calculated sampling rates are given in Table S10. We considered
the sampling rates reliable if they were consistent between the
three samplers, e.g., if the relative standard deviation of the sampling
rate was less than 30%.
Mean sampling rates for individual compounds ranged from 0.7 m3
day−1
(for PFOS) to 14 m3
day−1
(for PFBS) (Fig. 4a). The mean
sampling rate for all consistent PFASs in the XAD-PAS was higher than
the typical sampling rate for XAD-PAS, but still within the range of
what has previously been reported. Overall, the XAD-PAS have lower
Fig. 3. Equivalent air volume sampled by OUT-PUF-PAS outdoors for PFASs over a 126-day deployment. Black points represent calibration against total air; blue
points represent calibration against only gas phase. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this
article.)
P. Karásková et al. Atmospheric Environment 185 (2018) 186–195
190
variability in sampling rates and more consistent performance when
compared with previous calibrations for the non-polar compounds (e.g.,
PCBs, OCPs).
However, despite the perception that PUF-PAS sample total air,
while XAD-PAS sample gas-phase only (Melymuk et al., 2014), for both
the XAD-PAS and PUF-PAS, there were no significant differences between
the sampling rates obtained using gas-phase only vs. those using
total air concentrations for either type of sampler (Student t-test,
α = 0.05). However, this may be more due to the set of PFASs that were
sufficiently detected in both samplers to enable to determination of
sampling rates. In both sampler types, at least one compound was detected
by the LV-AAS in total air but not sufficiently in gas-phase air
(PFBA in PUF-PAS and PFTeDA in XAD-PAS), but was detected in the
passive sampler, suggesting some fraction of particle phase air is sampled
by both types of PAS.
Overall, there is contrasting evidence for the applicability of PUFPAS
for PFASs in outdoor air; and this suggests that the uptake of PFASs
to PUF is not fully understood. The complicating influences could be
poor understanding of PFAS gas-particle partitioning, or sampling artifacts
influencing the reference AAS as well as the PUF-PAS. While
XAD-PAS are perceived to be more appropriate for polar compounds
such as PFAS, they may have limited ability to collect particle-phase
compounds, and thus may not be ideal for certain less volatile PFASs.
However, we have also demonstrated that the XAD-PAS have sufficient
uptake rates for the PFAS to be deployed for shorted time periods (e.g.,
2–3 months) and thus the samplers may be applicable to assess seasonal
trends, at least in non-remote locations.
While the calibration results suggest that PUF-PAS are only appropriate
for a limited number of PFASs outdoors, we also considered
whether they could be used for qualitative purposes, e.g., detection and
identification of PFASs outdoors. We compared the outdoor air profile
of PFAS captured by AAS and PAS; this indicated some similarity between
samplers but also some notable differences (Fig. 5 and Table S6).
The LV-AAS-total and LV-AAS-gas phase profiles were very similar,
suggesting that the bulk of PFAS in outdoor air was found in the gasphase.
This supports the appropriateness of XAD-PAS for outdoor air
sampling, as it is generally a gas-phase air sampler. PFBA, PFDS and NEtFOSE
were not detected in the XAD-PAS but were small contributions
to the gas-phase LV-AAS, while PFTeDA, which was not in the gasphase
LV-AAS, was detected by the XAD-PAS. Beyond these small differences,
almost identical sets of PFASs were detected by OUTgas-phase
AAS and OUT-XAD-PAS, however the contributions of compounds
varied distinctly between the two samplers: PFOS and PFOA had higher
contributions in the LV-AAS, while PFHpA and PFBS had higher
contributions in the XAD-PAS. While these discrepancies could be rectified
in passive sampler calibration by the use of compound-specific
sampling rates, if generic sampling rates are used different profiles of
PFASs in air would be reported when using active vs. XAD-PAS sam-
pling.
The discrepancies are even larger for the PUF-PAS. The profile in the
PUF-PAS is dominated by PFBA and PFPA, which were two of the
compounds with the most consistent linear uptake in the sampler. In
contrast, PFOA and a number of other PFCAs and PFSAs are not detected
by the PUF-PAS, and the contribution of PFOS is much smaller
than in either the LV-AAS or XAD-PAS. Thus, PUF-PAS do not seem to
be appropriate for detection or quantification of a broad range of
PFASs, and may be used for quantification for only a limited number of
compounds. We discuss possible reasons for the discrepancy in following
sections.
3.2.2. Indoor PUF-PAS
PAS performance in indoor air differs from that in outdoor air because
of the more stable conditions (e.g., low air movement thus larger
boundary layer in PAS, more stable temperatures) and higher concentrations
of many compounds of interest. As noted in section 3.1, the
concentrations of PFASs in indoor air were 3× higher than in outdoor
air. These effects have conflicting influences on PAS performance. The
lower air movement should lead to lower sampling rates indoors,
however the higher concentrations and often higher temperatures could
decrease the length of the linear uptake phase of samplers, leading to
faster equilibration.
We examined the uptake of individual PFASs to the triplicate indoor
PUF-PAS, and found large differences in the uptake profiles (Fig. 6).
Unlike in outdoor air, very few compounds had increasing concentrations
in the PUF-PAS over time. Only PFBA, FOSA, N-MeFOSE and NEtFOSE
had positive slopes; the majority of compounds were detected
in the PUF-PAS but had no increase in PUF-PAS concentrations over
time, suggesting rapid equilibration in indoor air. We attribute this to
the higher temperatures compared to outdoor air in this study and
significantly higher indoor air concentrations. These results strongly
suggest that PUF-PAS are not appropriate for PFAS indoors due to the
high indoor concentrations of PFASs and the low sorptive capacity of
PUF for PFASs.
As in outdoor air, we compared the profile of PFASs captured by the
indoor LV-AAS and the PUF-PAS (Fig. 5). The INtotalAAS was dominated
by PFBA and PFPA, while INgas-phaseAAS also had significant
contributions form N-MeFOSE and N-EtFOSE. In contrast, while the
PUF-PAS detected almost the same number of compounds as the AAS,
Fig. 4. Comparison of sampling rates for total vs. gas-phase only calibrations for (a) XAD-PAS and (b) PUF-PAS determined according to Equation (2).
P. Karásková et al. Atmospheric Environment 185 (2018) 186–195
191
the profile again differed. PFOA was the highest percentage contributor,
and notably, PFPA was not detected. As with outdoor air, this
suggests limited use of PUF-PAS for PFASs, considering the lack of
linear uptake for the majority of PFASs indoors, and different overall
compound profiles.
3.3. Assessment of factors influencing PAS performance
3.3.1. Equilibration of samplers with air
PAS and AAS were compared by principle component analysis
(PCA) for all PFAS (gas and particle phase) in outdoor (Figure S2) and
indoor samples (Fig. 7) for the loading and variance. The PCA looks at
all inter-correlated quantitative dependent variables and represents
these as orthogonal variables known as principle components. The
distance between samples indicates the similarity of results. In a
loading plot, the distance from 0 indicates the weight to drive the
variance, with those at greater distance providing greater weight.
In indoor air, we observed a clear temporal shift with indoor PAS
from the 1st sample set after 7 days up to day 21, after which it appears
PAS are at equilibrium with the indoor environment, indicated by the
clustering in the PCA plot (Fig. 7a). Based on factor loading, the equilibrium
of PAS is driven by the uptake of PFPA, PFOA and PFNA, while
INtotalAAS variance is mostly due to concentrations of PFBA that were
∼10 times more abundant in the INtotalAAS, and primarily in the particle
phase, with median PFBA concentrations of 28.1 and 2.8 pg m−3
in
particle and gas phase IN-AAS, respectively. Therefore, a shorter sample
interval for PAS exposure may be applicable for indoor PAS use.
In contrast, outdoor samples show greater differences between PAS
Fig. 5. PFAS distribution in outdoor (OUT) and indoor (IN) air according to different air samplers – showing the average profile measured by LV-AAS, PUF-PAS and
XAD-PAS.
Fig. 6. Uptake to indoor PUF-PAS over 12-week (84-day) sampling period (* indicates significantly non-zero slopes).
P. Karásková et al. Atmospheric Environment 185 (2018) 186–195
192
and AAS, and no temporal trend towards equilibrium as was observed
for the indoor samples. This supports what is shown in Fig. 3, as the
majority of PFASs have increasing uptake over time in outdoor air,
while in indoor air the relationship between PUF-PAS mass and time
suggests rapid achievement of equilibrium (Fig. 6). Outdoor AAS have
low variation and neither gas-phase, particle-phase nor total concentrations
compare well with that observed in PUF-PAS (Figure S2).
Variance observed in outdoor samples were largely caused by concentrations
of PFOS, PFOA and N-MeFOSE for AAS and PFBA, PFPA and
PFHpA.
3.3.2. Gas-particle partitioning
As one concern in the use of PAS is variable and/or ineffective
particle capture by the PAS, the distribution of compounds between the
gas and particle phase may make PAS less effective for some PFAS.
Furthermore, unlike many POPs such as polychlorinated biphenyls,
PFAS are ionisable and it is therefore expected that their partitioning
behaviour will also be affected by aqueous aerosols (Kim and Kannan,
2007; Ahrens et al., 2013). In the gas phase, PFAS are in their neutral
form while in aqueous solutions they may dissociate to form ionic PFAS.
Studies of atmospheric PFAS sometimes focus only on particle- or gasphase,
or they treat the two phases as a single sample. For example,
while ionic PFASs have been reported previously (e.g. Barber et al.,
2007; Kim and Kannan, 2007; Stock et al., 2007; Dreyer and Ebinghaus,
2009; Weinberg et al., 2011a,b; Liu et al., 2015), their quantification is
often limited to only the atmospheric particle phase. Given the physicochemical
properties of many PFASs, this is not surprising as many of
the ionic PFASs are estimated to be predominantly in the particle phase
(Ahrens et al., 2011a,b). However, in this study the particle-bound
fraction, assumed to be captured by the QFF in the AAS, and the gasphase
fraction, captured by the PUF/XAD/PUF sandwich, were quantified
separately and thus were used for gas-particle partitioning assessment
to provide further insight into the observed PAS performance.
Because uncertainties in the sampling of PFAS by PUF sorbent used
indoors, we determine gas-particle partitioning only in outdoor samples,
where XAD was included as a gas-phase sorbent. For indoor
environments, dust has been used as a monitor for particle phase PFAS
(Goosey and Harrad, 2011; Padilla-Sánchez and Haug, 2016; Björklund
et al., 2009; Reiner et al., 2015), though this does not reflect inhalation
exposure. In outdoor air, the most abundant ionic compounds on QFFs
were PFBA (24% of Σ17PFASs in particle-phase) with a median of
1.3 pg m−3
, PFOS (20%) with a median of 0.3 pg m−3
and PFOA (9%)
with a median of 0.2 pg m−3
, N-MeFOSE was the dominant volatile
PFAS (7% of Σ17PFASs in particle-phase).
The particulate associated fraction of individual PFAS groups was
calculated using equation:
Φ = Cparticle phase/Cparticle+gas–phase (Eq. 3)
The obtained particle-associated fraction (Φ) in outdoor air was
found to be statistically lower for PFASs (p = 0.020) and FOSA/Es
(p = 0.002) compared to short-chain PFCAs. Statistical differences
were also observed between long-chain PFCAs and FOSA/Es
(p = 0.008) (Fig. 8).
Fig. 7. Principle component analysis (PCA) of indoor PFAS samples using the total air concentration measured by AAS (blue) and total mass captured by PAS (green).
The insets on each PCA indicate compounds within the dotted region. Figure (a) shows loading of individual air samples, with AAS labelled with “A–collection day”
while PAS are labelled by sample number and the day (D) of collection with 0 as the initial day of PAS deployment and figure (b) shows variance. (For interpretation
of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
Fig. 8. Particle associated fraction of PFASs groups calculated using Eq. (3) for
each individual sampling week in outdoor air samples.
P. Karásková et al. Atmospheric Environment 185 (2018) 186–195
193
For several compounds, partitioning can be clearly related to their
physicochemical properties, with the more volatile compounds such as
the FOSA/Es predominantly found in the gas phase (91%), with Φ of
0.06 ± 0.04. In the case of PFCAs, they distribute between both particles
and gas phase. Φ had a wide range, between 0.02 (PFHxA) and
0.94 (PFTeDA). PFASs were also primarily in the gas phase
(Φ = 0.07 ± 0.04); except PFHxS, which was predominantly detected
on particles, but as PFHxS was detected in less than 50% of all samples,
this result may have higher uncertainty. PFOA and PFOS, compounds
on which legislative bodies are currently focused, i.e. Stockholm
Convention, are predicted to be more volatile and this is confirmed by
the findings in this study (Φ (PFOA) = 0.08 ± 0.07; Φ
(PFOS) = 0.07 ± 0.04).
In general, the majority of target compounds were associated with
gas phase of ambient air, which is a useful for PAS, considering the
variable or restricted particle-phase uptake (Markovic et al., 2015).
This is similar to the findings of Arp and Goss (2009) who determined
that the majority of PFCAs would be primarily in the gas-phase and that
deposition to filters may be associated with relative humidity, thus
those PFCAs measured on filters may result from vapour phase uptake
rather than particle deposition. Particle-associated compounds in this
study increased with greater compound chain length, in keeping with
what has been seen in other studies (Vierke et al., 2011). However,
when compared to a high volume AAS sampler, the fraction of particlebound
longer chain PFASs is less in this study than that seen in some
other studies (Ahrens et al., 2012). As a range of related factors influence
partitioning behaviour of atmospheric compounds between gas
and particulate phases, including temperature, humidity, organic
matter content of aerosols, etc., the particle-associated fraction may
shift noticeably based on ambient conditions and this may influence
some of the differences observed in this study.
3.3.3. Uptake behaviour of PFAS on PUF-PAS
The poor performance of PUF-PAS for PFAS has been explored
previously and there are some theories suggesting that the partitioning
of PFAS is different compared to other compounds such as PCBs, OCPs
and PBDEs due to the hydrophilic polar features of PFAS. PUF sorptive
capacity is typically represented by the PUF-air partition coefficient
(log KPUF-Air), estimated based on the octanol-air partition coefficient,
(KOA), as described by Shoeib and Harner (2002) for hydrophobic
nonpolar POPs (Eq (4)).
= −−K Klog 0.6366 log 3.1774PUF Air OA (Eq. 4)
Using Eq. (4), log KPUF-Air was estimated for PFASs. Log KOA values
from Lei et al., (2004) for N-EtFOSA, N-MeFOSE, and N-EtFOSE and
Kim et al. (2015) for other PFAS were used. Log KPUF-Air for PFASs
ranged from −0.5 for PFBA to 1.7 for PFTeDA (SI Table S11).
In comparison, PCBs have log KPUF-Air from 1.7 for PCB 18 to 3.7 for
PCB 156 (Shoeib and Harner, 2002), and PBDEs from 2.7 for BDE 17 to
4.6 for BDE 126 (Harner and Shoeib, 2002). Thus, the estimated KPUF-Air
values suggest weaker partitioning of gaseous PFASs to PUF than for the
hydrophobic POPs. Moreover, other studies of the partitioning of PFAS
compounds in the atmosphere have indicated that the gas-particle
partitioning relationships that have been derived for non-polar compounds
do not fit well for PFAS compounds based upon their KOA, e.g.,
that partitioning to PUF is even weaker than predicted based on Eq. (4)
(Shoeib et al., 2008). It has been suggested that the carbon-fluorine
chain length, atmospheric moisture content, functional groups and
airborne particles may be key factors in the sorption behaviour of PFAS
to PUF-PAS (Ahrens et al., 2013).
4. Conclusions and recommendations
The use of XAD-PAS as passive air samplers appears to be a useful
tool in the measurement of PFASs but there is still variability between
those compounds detected by active air and passive samplers. PUF-PAS
may have some limited use in characterizing some of the sulfonate
PFAS compounds, but the variable accumulation of many PFASs in this
PAS make it a poor media for long-term monitoring. Moreover, the
qualitative determination of PFAS profiles is not appropriate with PUFPAS
due to different sorptive capacities for different classes of PFASs.
The use of PUF-PAS indoor environments still needs further exploration
as equilibrium was reached rapidly and more study on the limitations
are needed.
Further study is needed for PUF-PAS applicability for fluorotelomer
alcohols (FTOHs) as well as comparison to other atmospheric PAS
samplers such as ENV+ (Biotage, Uppsala, Sweden) and SIPs. The use
of PAS samplers to predict individual exposure is limited due to the
number of environments to which a person is exposed and personal
passive samplers may be key to this understanding (Padilla-Sánchez
et al., 2017).
In order to fulfil properly global monitoring needs, caution should
be given to interpretation of global distributions of PFASs when based
on PUF-PAS. Considering the weakly significant uptake rates in the
outdoor calibration for PFOS, and better uptake for PFCAs, PUF-PAS
may be acceptable to identify large differences in concentration gradients
on global scales (e.g., semi-quantitative/order of magnitude
differences), however they are not able to provide accurate concentrations
for a broad set of PFASs. XAD-PAS are a more reliable option
which give profiles that more closely resemble those from AAS, and
it may be possible to achieve seasonal resolution with XAD-PAS. Other
PAS options (e.g., SIP-PAS) are also useful for seasonal resolution and
broader compound coverage. Given the importance of establishing reliable
long-term monitoring for PFASs, passive sampling techniques for
these compounds should continue to be investigated and optimized.
Acknowledgements:
This research was supported by the Ministry of Education, Youth
and Sports and the European Structural and Investment Funds in the
Czech Republic: CETOCOEN PLUS project (CZ.02.1.01/0.0/0.0/
15_003/0000469) and the RECETOX research infrastructure
[LM2015051 and CZ.02.1.01/0.0/0.0/16_013/0001761]. We also
thank Roman Prokeš for his contribution to the sampling campaign.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://dx.
doi.org/10.1016/j.atmosenv.2018.05.030.
References
Ahrens, L., Shoeib, M., Del Vento, S., Codling, G., Halsall, C., 2011a. Polyfluoroalkyl
compounds in the canadian arctic atmosphere. Environ. Chem. 8 (4), 399–406.
Ahrens, L., Shoeib, M., Harner, T., Lee, S.C., Guo, R., Reiner, E.J., 2011b. Wastewater
treatment plant and landfills as sources of polyfluoroalkyl compounds to the atmosphere.
Environ. Sci. Technol. 45 (19), 8098–8105.
Ahrens, L., Harner, T., Shoeib, M., Lane, D.A., Murphy, J.G., 2012. Improved characterization
of gas–particle partitioning for per- and polyfluoroalkyl substances in the
atmosphere using annular diffusion denuder samplers. Environ. Sci. Technol. 46 (13),
7199–7206.
Ahrens, L., Harner, T., Shoeib, M., Koblizkova, M., Reiner, E.J., 2013. Characterization of
two passive air samplers for per- and polyfluoroalkyl substances. Environ. Sci.
Technol. 47 (24), 14024–14033.
Antweiler, R.C., 2015. Evaluation of statistical treatments of left-censored environmental
data using coincident uncensored data sets. II. Group comparisons. Environ. Sci.
Technol. 49 (22), 13439–13446.
Arp, H.P.H., Goss, K.-U., 2009. Gas/particle partitioning behavior of perfluorocarboxylic
acids with terrestrial aerosols. Environ. Sci. Technol. 43 (22), 8542–8547.
Barber, J.L., Berger, U., Chaemfa, C., Huber, S., Jahnke, A., Temme, C., Jones, K.C., 2007.
Analysis of per- and polyfluorinated alkyl substances in air samples from Northwest
Europe. J. Environ. Monit. 9 (6), 530–541.
Benskin, J.P., Ahrens, L., Muir, D.C.G., Scott, B.F., Spencer, C., Rosenberg, B., Tomy, G.,
Kylin, H., Lohmann, R., Martin, J.W., 2012. Manufacturing origin of perfluorooctanoate
(PFOA) in atlantic and canadian arctic seawater. Environ. Sci.
Technol. 46 (2), 677–685.
Björklund, J.A., Thuresson, K., De Wit, C.A., 2009. Perfluoroalkyl compounds (PFCs) in
P. Karásková et al. Atmospheric Environment 185 (2018) 186–195
194
indoor dust: concentrations, human exposure estimates, and sources. Environ. Sci.
Technol. 43 (7), 2276–2281.
Bohlin, P., Audy, O., Skrdlikova, L., Kukucka, P., Pribylova, P., Prokes, R., Vojta, S.,
Klanova, J., 2014a. Outdoor passive air monitoring of semi volatile organic compounds
(SVOCs): a critical evaluation of performance and limitations of polyurethane
foam (PUF) disks. Environ. Sci. Process. Impacts 16 (3), 433–444.
Bohlin, P., Audy, O., Skrdlikova, L., Kukucka, P., Vojta, S., Pribylova, P., Prokes, R., Cupr,
P., Klanova, J., 2014b. Evaluation and guidelines for using polyurethane foam (PUF)
passive air samplers in double-dome chambers to assess semi-volatile organic compounds
(SVOCs) in non-industrial indoor environments. Environ. Sci. Process.
Impacts 16 (11), 2617–2626.
Butt, C.M., Berger, U., Bossi, R., Tomy, G.T., 2010. Levels and trends of poly- and perfluorinated
compounds in the arctic environment. Sci. Total Environ. 408 (15),
2936–2965.
Chaemfa, C., Barber, J.L., Huber, S., Breivik, K., Jones, K.C., 2010. Screening for PFOS
and PFOA in European air using passive samplers. J. Environ. Monit. 12 (5),
1100–1109.
Codling, G., Vogt, A., Jones, P.D., Wang, T., Wang, P., Lu, Y.L., Corcoran, M., Bonina, S.,
Li, A., Sturchio, N.C., Rockne, K.J., Ji, K., Khim, J.-S., Naile, J.E., Giesy, J.P., 2014.
Historical trends of inorganic and organic fluorine in sediments of Lake Michigan.
Chemosphere 114, 203–209.
Dreyer, A., Ebinghaus, R., 2009. Polyfluorinated compounds in ambient air from shipand
land-based measurements in northern Germany. Atmos. Environ. 43 (8),
1527–1535.
Fromme, H., Dreyer, A., Dietrich, S., Fembacher, L., Lahrz, T., Völkel, W., 2015. Neutral
polyfluorinated compounds in indoor air in Germany – the LUPE 4 study.
Chemosphere 139, 572–578.
Giesy, J.P., Kannan, K., 2002. Peer reviewed: perfluorochemical surfactants in the environment.
Environ. Sci. Technol. 36 (7), 146A–152A.
Gomez, C., Vicente, J., Echavarri-Erasun, B., Porte, C., Lacorte, S., 2011. Occurrence of
perfluorinated compounds in water, sediment and mussels from the Cantabrian Sea
(North Spain). Mar. Pollut. Bull. 62 (5), 948–955.
Goosey, E., Harrad, S., 2011. Perfluoroalkyl compounds in dust from Asian, Australian,
European, and North American homes and UK cars, classrooms, and offices. Environ.
Int. 37 (1), 86–92.
Harner, T., Shoeib, M., 2002. Measurements of Octanol−Air partition coefficients (KOA)
for polybrominated diphenyl ethers (PBDEs): predicting partitioning in the environment.
J. Chem. Eng. Data 47 (2), 228–232.
Haug, L.S., Huber, S., Schlabach, M., Becher, G., Thomsen, C., 2011. Investigation on perand
polyfluorinated compounds in paired samples of house dust and indoor air from
Norwegian homes. Environ. Sci. Technol. 45 (19), 7991–7998.
Hung, H., Katsoyiannis, A.A., Brorström-Lundén, E., Olafsdottir, K., Aas, W., Breivik, K.,
Bohlin-Nizzetto, P., Sigurdsson, A., Hakola, H., Bossi, R., Skov, H., Sverko, E., Barresi,
E., Fellin, P., Wilson, S., 2016. Temporal trends of persistent organic Pollutants
(POPs) in arctic air: 20 years of monitoring under the arctic monitoring and assessment
programme (AMAP). Environ. Pollut. 217, 52–61.
Karásková, P., Venier, M., Melymuk, L., Bečanová, J., Vojta, Š., Prokeš, R., Diamond,
M.L., Klánová, J., 2016. Perfluorinated alkyl substances (PFASs) in household dust in
Central Europe and North America. Environ. Int. 94, 315–324.
Kim, S.-K., Kannan, K., 2007. Perfluorinated acids in air, rain, snow, surface runoff, and
lakes: relative importance of pathways to contamination of urban lakes. Environ. Sci.
Technol. 41 (24), 8328–8334.
Kim, M., Li, L.Y., Grace, J.R., Yue, C., 2015. Selecting reliable physicochemical properties
of perfluoroalkyl and polyfluoroalkyl substances (PFASs) based on molecular descriptors.
Environ. Pollut. (Barking, Essex : 1987) 196, 462–472.
Klánová, J., Harner, T., 2013. The challenge of producing reliable results under highly
variable conditions and the role of passive air samplers in the Global Monitoring Plan.
Trac. Trends Anal. Chem. 46, 139–149.
Koblizkova, M., Genualdi, S., Lee, S.C., Harner, T., 2012. Application of sorbent impregnated
polyurethane foam (SIP) disk passive air samplers for investigating organochlorine
pesticides and polybrominated diphenyl ethers at the global scale.
Environ. Sci. Technol. 46 (1), 391–396.
Langer, V., Dreyer, A., Ebinghaus, R., 2010. Polyfluorinated compounds in residential and
nonresidential indoor air. Environ. Sci. Technol. 44 (21), 8075–8081.
Lei, Y.D., Wania, F., Mathers, D., Mabury, S.A., 2004. Determination of vapor pressures,
Octanol−Air, and Water−Air partition coefficients for polyfluorinated sulfonamide,
sulfonamidoethanols, and telomer alcohols. J. Chem. Eng. Data 49 (4), 1013–1022.
Liu, B., Zhang, H., Yao, D., Li, J., Xie, L., Wang, X., Wang, Y., Liu, G., Yang, B., 2015.
Perfluorinated compounds (PFCs) in the atmosphere of Shenzhen, China: spatial
distribution, sources and health risk assessment. Chemosphere 138, 511–518.
Markovic, M.Z., Prokop, S., Staebler, R.M., Liggio, J., Harner, T., 2015. Evaluation of the
particle infiltration efficiency of three passive samplers and the PS-1 active air
sampler. Atmos. Environ. 112, 289–293.
Melymuk, L., Bohlin, P., Sáňka, O., Pozo, K., Klánová, J., 2014. Current challenges in air
sampling of semivolatile organic contaminants: sampling artifacts and their influence
on data comparability. Environ. Sci. Technol. 48 (24), 14077–14091.
Padilla-Sánchez, J.A., Haug, L.S., 2016. A fast and sensitive method for the simultaneous
analysis of a wide range of per-and polyfluoroalkyl substances in indoor dust using
on-line solid phase extraction-ultrahigh performance liquid chromatography-time-offlight-mass
spectrometry. J. Chromatogr. A 1445, 36–45.
Padilla-Sánchez, J.A., Papadopoulou Poothong, E.S., Haug, L.S., 2017. Investigation of
the best approach for assessing human exposure to poly- and perfluoroalkyl substances
through indoor air. Environ. Sci. Technol. 51, 12836–12843.
Poothong, S., Thomsen, C., Padilla-Sanchez, J.A., Papadopoulou, E., Haug, L.S., 2017.
Distribution of novel and well-known poly-and perfluoroalkyl substances (PFASs) in
human serum, plasma, and whole blood. Environ. Sci. Technol. 51 (22),
13388–13396.
Prevedouros, K., Cousins, I.T., Buck, R.C., Korzeniowski, S.H., 2006. Sources, fate and
transport of perfluorocarboxylates. Environ. Sci. Technol. 40 (1), 32–44.
Reiner, Jessica L., Blaine, Andrea C., Higgins, Christopher P., Huset, Carin, Jenkins,
Thomas M., Kwadijk, Christiaan J.A.F., Lange, Cleston C., et al., 2015.
Polyfluorinated substances in abiotic standard reference materials. Anal. Bioanal.
Chem. 407 (no. 11), 2975–2983.
Schuster, J.K., Gioia, R., Harner, T., Lee, S.C., Breivik, K., Jones, K.C., 2012. Assessment of
sorbent impregnated PUF disks (SIPs) for long-term sampling of legacy POPs. J.
Environ. Monit. 14 (1), 71–78.
Shoeib, M., Harner, T., 2002. Characterization and comparison of three passive air
samplers for persistent organic Pollutants. Environ. Sci. Technol. 36 (19),
4142–4151.
Shoeib, M., Harner, T., Ikonomou, M., Kannan, K., 2004. Indoor and outdoor air concentrations
and phase partitioning of perfluoroalkyl sulfonamides and polybrominated
diphenyl ethers. Environ. Sci. Technol. 38 (5), 1313–1320.
Shoeib, M., Harner, T., Wilford, B.H., Jones, K.C., Zhu, J., 2005. Perfluorinated sulfonamides
in indoor and outdoor air and indoor Dust: occurrence, partitioning, and
human exposure. Environ. Sci. Technol. 39 (17), 6599–6606.
Shoeib, M., Harner, T., Lee, S.C., Lane, D., Zhu, J., 2008. Sorbent-impregnated polyurethane
foam disk for passive air sampling of volatile fluorinated chemicals. Anal.
Chem. 80 (3), 675–682.
Shoeib, M., Vlahos, P., Harner, T., Peters, A., Graustein, M., Narayan, J., 2010. Survey of
polyfluorinated chemicals (PFCs) in the atmosphere over the northeast Atlantic
Ocean. Atmos. Environ. 44 (24), 2887–2893.
Shoeib, M., Harner, T.M., Webster, G., Lee, S.C., 2011. Indoor sources of poly-and perfluorinated
compounds (PFCS) in Vancouver, Canada: implications for human exposure.
Environ. Sci. Technol. 45 (19), 7999–8005.
Stock, N.L., Furdui, V.I., Muir, D.C.G., Mabury, S.A., 2007. Perfluoroalkyl contaminants in
the canadian arctic: evidence of atmospheric transport and local contamination.
Environ. Sci. Technol. 41 (10), 3529–3536.
Sun, H., Li, F., Zhang, T., Zhang, X., He, N., Song, Q., Zhao, L., Sun, L., Sun, T., 2011.
Perfluorinated compounds in surface waters and WWTPs in Shenyang, China: mass
flows and source analysis. Water Res. 45 (15), 4483–4490.
Vierke, L., Ahrens, L., Shoeib, M., Reiner, E.J., Guo, R., Palm, W.-U., Ebinghaus, R.,
Harner, T., 2011. Air concentrations and particle–gas partitioning of polyfluoroalkyl
compounds at a wastewater treatment plant. Environ. Chem. 8 (4), 363–371.
Wang, T., Lu, Y., Chen, C., Naile, J.E., Khim, J.S., Park, J., Luo, W., Jiao, W., Hu, W.,
Giesy, J.P., 2011. Perfluorinated compounds in estuarine and coastal areas of north
Bohai Sea, China. Mar. Pollut. Bull. 62 (8), 1905–1914.
Wania, F., Shen, L., Lei, Y.D., Teixeira, C., Muir, D.C.G., 2003. Development and calibration
of a resin-based passive sampling system for monitoring persistent organic
Pollutants in the atmosphere. Environ. Sci. Technol. 37 (7), 1352–1359.
Weinberg, I., Dreyer, A., Ebinghaus, R., 2011a. Landfills as sources of polyfluorinated
compounds, polybrominated diphenyl ethers and musk fragrances to ambient air.
Atmos. Environ. 45 (4), 935–941.
Weinberg, I., Dreyer, A., Ebinghaus, R., 2011b. Waste water treatment plants as sources of
polyfluorinated compounds, polybrominated diphenyl ethers and musk fragrances to
ambient air. Environ. Pollut. 159 (1), 125–132.
Wu, M., Sun, R., Wang, M., Liang, H., Ma, S., Han, T., Xia, X., Ma, J., Tang, L., Sun, Y., Xu,
G., 2017. Analysis of perfluorinated compounds in human serum from the general
population in Shanghai by liquid chromatography-tandem mass spectrometry (LCMS/MS).
Chemosphere 168, 100–105.
Yang, L., Zhu, L., Liu, Z., 2011. Occurrence and partition of perfluorinated compounds in
water and sediment from Liao River and Taihu Lake, China. Chemosphere 83 (6),
806–814.
P. Karásková et al. Atmospheric Environment 185 (2018) 186–195
195
Melymuk, Lisa, Pernilla Bohlin Nizzetto, Tom Harner, Kevin B. White, Xianyu Wang, Maria
Yumiko Tominaga, Jun He, Jun Li, Jianmin Ma, Wan-Li Ma, Beatriz H. Aristizábal, Annekatrin
Dreyer, Begoña Jiménez, Juan Muñoz-Arnanz, Mustafa Odabasi, Yetkin Dumanoglu, Baris
Yaman, Carola Graf, Andrew Sweetman, Jana Klánová. 2021. "Global Intercomparison of
Polyurethane Foam Passive Air Samplers Evaluating Sources of Variability in SVOC
Measurements." Environmental Science and Policy 125: 1-9.
https://doi.org/10.1016/j.envsci.2021.08.003
Environmental Science and Policy 125 (2021) 1–9
Available online 24 August 2021
1462-9011/© 2021 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Global intercomparison of polyurethane foam passive air samplers
evaluating sources of variability in SVOC measurements
Lisa Melymuk a,
*, Pernilla Bohlin Nizzetto b
, Tom Harner c
, Kevin B. White a
, Xianyu Wang d
,
Maria Yumiko Tominaga e
, Jun He f
, Jun Li g
, Jianmin Ma h
, Wan-Li Ma i
, Beatriz H. Aristiz´abal j
,
Annekatrin Dreyer k
, Bego˜na Jim´enez l
, Juan Mu˜noz-Arnanz l
, Mustafa Odabasi m
,
Yetkin Dumanoglu m
, Baris Yaman m
, Carola Graf n
, Andrew Sweetman n
, Jana Kl´anov´a a
a
RECETOX, Masaryk University, Brno, Czech Republic
b
NILU - Norwegian Institute for Air Research, Kjeller, Norway
c
Air Quality Processes Research Section, Environment and Climate Change Canada, Toronto, Canada
d
Queensland Alliance for Environmental Health Sciences (QAEHS), The University of Queensland, Australia
e
CETESB - S˜ao Paulo State Environmental Company, S˜ao Paulo, Brazil
f
Department of Chemical and Environmental Engineering, University of Nottingham Ningbo China, Ningbo, China
g
State Key Laboratory of Organic Geochemistry, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, Guangzhou, China
h
College of Urban and Environmental Sciences, Peking University, Beijing, China
i
International Joint Research Center for Persistent Toxic Substances (IJRC-PTS), Harbin Institute of Technology, Harbin, China
j
Hydraulic Engineering and Environmental Research Group (GTAIHA), Universidad Nacional de Colombia, Manizales, Colombia
k
Eurofins GfA GmbH (Now Operating Under the Name ANECO Institut für Umweltschutz GmbH & Co), Germany
l
Department of Instrumental Analysis and Environmental Chemistry, IQOG-CSIC, Madrid, Spain
m
Department of Environmental Engineering, Dokuz Eylul University, Buca-Izmir, Turkey
n
Lancaster Environment Centre, Lancaster University, UK
A R T I C L E I N F O
Keywords:
Passive air sampling
Global air monitoring
Persistent organic pollutants
Semi-volatile organic compounds
Stockholm Convention
PUF disk
A B S T R A C T
Polyurethane foam passive air samplers (PUF-PAS) are the most common type of passive air sampler used for a range of
semi-volatile organic compounds (SVOCs), including regulated persistent organic pollutants (POPs) and polycyclic
aromatic hydrocarbons (PAHs), and emerging contaminants (e.g., novel flame retardants, phthalates, current-use
pesticides). Data from PUF-PAS are key indicators of effectiveness of global regulatory actions on SVOCs, such as
the Global Monitoring Plan of the Stockholm Convention on Persistent Organic Pollutants. While most PUF-PAS use
similar double-dome metal shielding, there is no standardized dome size, shape, or deployment configuration, with
many different PUF-PAS designs used in regional and global monitoring. Yet, no information is available on the
comparability of data from studies using different PUF-PAS designs. We brought together 12 types of PUF-PAS used by
different research groups around the world and deployed them in a multi-part intercomparison to evaluate the variability
in reported concentrations introduced by different elements of PAS monitoring. PUF-PAS were deployed for 3
months in outdoor air in Kjeller, Norway in 2015–2016 in three phases to capture (1) the influence of sampler design
on data comparability, (2) the influence of analytical variability when samplers are analyzed at different laboratories,
and (3) the overall variability in global monitoring data introduced by differences in sampler configurations and
analytical methods. Results indicate that while differences in sampler design (in particular, the spacing between the
upper and lower sampler bowls) account for up to 50 % differences in masses collected by samplers, the variability
introduced by analysis in different laboratories far exceeds this amount, resulting in differences spanning orders of
magnitude for POPs and PAHs. The high level of variability due to analysis in different laboratories indicates that
current SVOC air sampling data (i.e., not just for PUF-PAS but likely also for active air sampling) are not directly
comparable between laboratories/monitoring programs. To support on-going efforts to mobilize more SVOC data to
contribute to effectiveness evaluation, intercalibration exercises to account for uncertainties in air sampling, repeated at
regular intervals, must be established to ensure analytical comparability and avoid biases in global-scale assessments of
SVOCs in air caused by differences in laboratory performance.
* Corresponding author.
E-mail address: lisa.melymuk@recetox.muni.cz (L. Melymuk).
Contents lists available at ScienceDirect
Environmental Science and Policy
journal homepage: www.elsevier.com/locate/envsci
https://doi.org/10.1016/j.envsci.2021.08.003
Received 13 January 2021; Received in revised form 14 July 2021; Accepted 3 August 2021
Environmental Science and Policy 125 (2021) 1–9
2
1. Introduction
Long-term global data on atmospheric levels of semi-volatile organic
compounds (SVOCs), including polycyclic aromatic hydrocarbons
(PAHs) and persistent organic pollutants (POPs), such as polychlorinated
biphenyls (PCBs), organochlorine pesticides (OCPs), and
polybrominated diphenyl ethers (PBDEs), are a fundamental need in
efforts to reduce emissions and minimize human and environmental
exposure. This need has been formalized in the requirements of international
actions, such as the Stockholm Convention on POPs (Articles 11
and 16) implemented through the Global Monitoring Plan (GMP), the
UNECE Convention on Long-Range Transboundary Air Pollution
(CLRTAP), and the development of a Global Earth Observation System of
Systems (GEOSS) to increase our understanding of global processes and
to underpin decision-making through sharing of accessible, high quality
interoperable environmental data.
The GMP has a clear policy mandate to collect comparable, harmonized
and reliable information on POP levels in core environmental
matrices, one of which is ambient air. The Global Observation System
for Persistent Organic Pollutants (GOS4POPs) is an initiative within the
Group on Earth Observations (GEO) to increase the availability and
quality of Earth observation data on POPs, and improve data availability
and interoperability across POP monitoring networks, providing support
for international conventions on toxic compounds (Stockholm Convention,
CLRTAP) and on-going international programs (e.g., GMP, European
Monitoring and Evaluation Programme (EMEP)).
Polyurethane foam passive air samplers (PUF-PAS) are widely used
in international air monitoring of POPs (Borůvkov´a et al., 2015;
Mu˜noz-Arnanz et al., 2016; Pozo et al., 2009; Wania and Shunthirasingham,
2020) and other semi-volatile organic compounds (SVOCs).
The spatial coverage and ease-of-use of PUF-PAS has been crucial in
enabling the development of international air monitoring programs such
as GAPS (Global Atmospheric Passive Sampling) and MONET (Lee et al.,
2007; Mu˜noz-Arnanz et al., 2018, 2016; Pozo et al., 2006; Pˇribylov´a
et al., 2012; Rauert et al., 2018; Roscales et al., 2018a; White et al.,
2021), and their use in many individual case studies has greatly
increased our knowledge of atmospheric levels of SVOCs. Following the
entry-into-force of the Stockholm Convention in 2004, the GMP was
established to secure monitoring data in core media (ambient air, breast
milk, human blood) and became a strong driver for the development of
passive air sampling programs to address global data gaps, especially
given the simplicity and relatively low cost of passive air samplers
(Kl´anov´a and Harner, 2013). The first GMP Report (UNEP, 2009) called
for improved collaboration within and among regions, and establishment
of strategic partnerships with expert laboratories and programs to
address the challenges in setting up new POP monitoring programs that
can continually adapt to include newly listed POPs. To mobilise such
data and ensure their interoperability, we must move towards more
harmonised monitoring frameworks with comprehensive datasets.
While internal consistency of data within individual programs is
necessary to assess long-term trends, the comparability of data among
different programs must also be improved so that datasets can be combined
for more effective global assessment. However, despite the
intended goal of global-scale comparability, differences in analytical
methods and sampler configurations between institutes and monitoring
programs may affect performance (Holt et al., 2017; Markovic et al.,
2015; Roscales et al., 2018b) and decrease the comparability of international
monitoring data (Su and Hung, 2010).
The simple design of the PUF-PAS has led to many individuallydesigned
versions around the globe, all following the same original
PUF-PAS concept (Shoeib and Harner, 2002) of a PUF disk protected by
a metal double-dome housing, but without standardized geometry. In a
previous comparison of three samplers, differences in sampler design
were found to have no discernable effect on PUF-PAS uptake rates
(Chaemfa et al., 2008), however, today the use of PUF-PAS has greatly
expanded due to ease of deployment and use, and sampler designs differ
to a much greater extent. At least 15 different designs are regularly used,
with differences in dome size and shape, placement of the PUF disk
relative to the gap between domes, size and density of the PUF disk itself,
and deployment practices (i.e., fixed versus freely hanging). In addition,
there are clear differences across laboratories in analytical methodology
applied to PUF processing and SVOC analysis, which have the potential
to lead to large variabilities in reported concentrations (Su et al., 2011;
Su and Hung, 2010). Current efforts to harmonize and synthesize global
SVOC monitoring combine data collected from different PUF-PAS
sampler designs and analyzed in different laboratories (e.g., GMP incorporates
PUF-PAS data from five different air monitoring networks
globally; Fig. S1), but lack information on how the variability introduced
by physical PUF-PAS design parameters compares to the analytical
variability between laboratories.
To evaluate the comparability of global SVOC data, we established
an international intercomparison in 2015 to evaluate sources of variability
in PUF-PAS-generated data. Institutes from 12 countries
(Australia, Brazil, Canada, China, Colombia, Czechia, Germany, Mexico,
Norway, Spain, Turkey, UK) participated in the intercomparison,
covering many of the major research groups using PUF-PAS and
including most of the monitoring networks/laboratories that have reported
PUF-PAS data to the Stockholm Convention GMP Data Warehouse
for the 3rd
Global Monitoring Report on Persistent Organic
Pollutants. We note that such an exercise is only possible due to the
simplicity and small size of the PUF-PAS samplers, whereas a similar
effort for active air samplers would not be feasible for logistical reasons.
The PUF-PAS intercomparison consisted of three phases to address the
following questions:
o what is the variability introduced by differences in PUF-PAS sampler
designs and deployment practices? (Phase 1)
o what is the variability introduced by differences in analytical
methods/performance between laboratories? (Phase 2)
o what is the overall variability/comparability between PUF-PASderived
air concentrations for POPs from different programs/laboratories?
(Phase 3)
This study evaluates the variability in SVOC measurements across
these three phases due to differences in sampler design and laboratory
performance to assess the comparability of reported SVOC monitoring
data from PUF-PAS across the globe.
2. Methods
Laboratory groups known to routinely use PUF-PAS to quantify
SVOCs in air were contacted and invited to join the study. In all, the
study included 15 participating research institutes (Table S1) using 12
different PUF-PAS sampler designs (Fig. 1). The institutes supplied their
own PUF disks and PAS housings. PUF-PAS designs differed by housing
dimensions (dome shape, internal volume, overhang, gap diameter)
and/or type of PUF disk (details of the individual designs are given in
Table S2). All equipment was kept separated under strict regimes. The
PUF disks and housings were pre-cleaned at the Norwegian Institute for
Air Research (NILU) before deployment in each phase of the study. All
samples in the intercomparison study were deployed at the same site,
located in Kjeller, outside Oslo, Norway. The site is semi-rural near grass
fields, with a mix of residential, and office buildings at a short distance.
Meteorological parameters corresponding with the deployment periods
of each study phase are given in Tables S3 and S4. The study was divided
into three phases, each addressing a key aspect of monitoring data
comparability, as described below.
2.1. Phase 1 – different samplers, same PUFs, same laboratory
The objective of Phase 1 was to isolate and identify the specific influence
of different PAS housings on sampler uptake. For Phase 1, 16
L. Melymuk et al.
Environmental Science and Policy 125 (2021) 1–9
3
passive air samplers (consisting of 12 different designs) were collected
from the 15 participating laboratories. Each laboratory’s PAS design
(different housing and installation parameters, but with identical PUF
disks) were deployed simultaneously at the field site in Kjeller, Norway
for 80 days from April 1, 2016 to June 13, 2016. Samplers were
deployed along a 50 m section of a wire fence at a height of 2 m (Fig S2).
This therefore addressed differences in both sampler design and installation
parameters, including fixed rigid installations for some samplers
and free-swinging installations for others, following the method of the
participating laboratory. The PUF disks had a density of 2.70 × 104
g/
m3
, mass of 5.9 g, diameter of 14.1 cm, and thickness of 1.4 cm. Average
daily ambient temperature during Phase 1 deployment was 9.2 ◦
C (range
− 4.6 to +28.5 ◦
C) and average wind speed was 2.8 m/s (range 1.3–5.6
m/s) (Table S3). After 80 days, the PUF-PAS were collected, and PUF
disks and three field blanks were packed individually in pre-cleaned
aluminum foil and shipped to the Trace Analytical Laboratories of
RECETOX, Czechia for SVOC analysis.
All PUF disks were analyzed according to accredited analytical
methods (ˇCSN EN ISO 17025: 2018) for 8 PCB congeners (7 indicator
PCBs + PCB 11), 12 OCPs (chlorobenzenes, hexachlorocyclohexanes HCHs,
dichlorodiphenyltrichloroethane and associated metabolites DDX
compounds), 29 PAHs and 10 PBDE congeners; compounds are
Fig. 1. Sampler designs used in the study and their basic dimensions. V indicates volume, G indicates the area of the horizontal gap between upper and lower dome,
V/G is the ratio of the dome volume to gap area, and overhang is the distance the upper dome extends over the lower dome.
L. Melymuk et al.
Environmental Science and Policy 125 (2021) 1–9
4
listed in Table S5. Full details on the analytical methods used by
RECETOX can be found in Kalina et al. (2017). Recoveries were tracked
using deuterated PAHs (d8-naphthalene, d10-phenanthrene, d12-perylene)
and non-environmental PCBs (PCB 30, PCB 185) (Table S6). PAH,
PCB and OCP masses were adjusted for recoveries based on the closest
corresponding recovery standard. PBDEs were quantified by isotope
dilution. Method detection limits (MDLs) were determined based on the
field blanks; MDL=[avg. mass in field blanks]+3*[standard dev. of field
blanks] (Table S7). If a compound was below detection in all field
blanks, the instrumental detection limit was taken as the MDL. All results
are reported as mass per PUF disk without conversion to air
concentration.
2.2. Phase 2 – same samplers, same PUFs, different laboratories
The objective of Phase 2 was to identify purely analytical variability
between laboratories. Fourteen identical PUF-PAS samplers (Sampler 15
from Fig. 1) were deployed at the Norwegian field site (deployment
height 2 m) for 81 days from September 11, 2015 to December 1, 2015.
Average daily ambient temperature during deployment was 5.9 ◦
C
(range -12.0 to +19.6 ◦
C) and the average wind speed was 1.9 m/s
(range 0–4.9 m/s) (Table S4). After 81 days each PUF disk and a corresponding
field blank were collected, wrapped in pre-cleaned
aluminum foil and sealed in plastic zip-top bags, packed in a padded
envelope and sent to the 15 participating laboratories. Participating
laboratories were asked to analyze the PUFs according to their in-house
methods and report masses for seven PCB congeners, 10 OCPs, 10 PBDE
congeners, and 16 PAHs to an Excel template. All results were reported
as mass per PUF disk without conversion to air concentration. Details of
the individual methods for each laboratory are given in Table S8. Most
laboratories used Soxhlet extraction while three laboratories used
accelerated solvent extraction (ASE) and one used a Büchi system. Seven
different solvent combinations were used, while only three different
clean-up methods were used. Not all laboratories reported all sets of
compounds, resulting in data for PCBs from 11 laboratories, OCPs from
11 laboratories, PBDEs from 10 laboratories, and PAHs from 9 laboratories.
Two laboratories that received PUF samples did not report any
results.
2.3. Phase 3 – different samplers, different PUFs, different laboratories
The objective of Phase 3 was to identify the full variability in SVOC
measurements due to the combined effect of different sampler designs
and laboratory analyses. This reflects the “realistic” variability that
would occur between different studies/monitoring networks. In this
Phase, 14 laboratories sent a PAS housing and PUF disk to NILU, and
each laboratory’s own PUF-PAS configuration (considering housing,
PUF disk and installation parameters) was deployed at the Norwegian
field site (deployment height 2 m), concurrent with Phase 2 from
September 17, 2015 to December 3, 2015. After 77 days the PUF disk
and a corresponding field blank were collected and shipped with the
Phase 2 samples. As with Phase 2, participating laboratories were asked
to analyze the PUFs for seven PCB congeners, 10 OCPs, 10 PBDE congeners
and 16 PAHs (Table S5) and report results to an Excel template.
Laboratories used the same analytical methods as for Phase 2 (Table S8)
and reported identical sets of compounds, resulting in records for PCBs
from 11 laboratories, OCPs from 11 laboratories, PBDEs from 10 laboratories,
and PAHs from 9 laboratories. Two laboratories that received
PUF samples did not report any results.
2.4. Quality assurance/quality control
Each participating laboratory reported their internal standards,
instrumental detection limits, and method detection limits for Phases 2
and 3.
All PUFs were sequentially pre-cleaned by Soxhlet at NILU
laboratories with 24 h toluene, 8 h acetone, 8 h hexane, and then dried
under vacuum. Field blanks were included in all three phases. Each PUF
disk sample sent to participating laboratories was paired with a field
blank of the same PUF disk type. PUF disks were only numbered and
were not separately identified as field blank or sample. All field blanks
were pre-cleaned at the same laboratory (NILU), using the same method.
Thus, any variability in levels in the field blanks should be due to
contamination during transport or laboratory procedures.
Data received from the Excel template spreadsheets were compiled
separately for each compound group. Each sampler was assigned a
number code for Phase 1 data (1–16) and each laboratory was assigned a
letter code for Phases 2 and 3 data (A–M) to anonymize all results.
Inconsistencies or missing values in reported data were addressed
individually with participating laboratories. Data handling and statistical
evaluation was done through MS Excel and R software.
3. Results and discussion
Twelve different sampler designs were received, differing in dome
shape, internal volume, overhang, and gap diameter. Not all domes were
hemispherical (some had a straight-sided conical shape), thus individual
dome volumes were measured based on the mass of water that could fill
each dome. Dome gap dimensions and overhangs were measured for the
assembled sampler design, and the surface areas of the gap between
upper and lower dome, i.e., the main space for air diffusion into the PAS
housing, were calculated assuming circular geometry. All sampler
housings also allowed additional diffusion through holes in the bottom
of the lower dome, although the number of holes varied from 4 to 8
depending on the sampler. Samplers and associated indicators of
sampler geometry are shown in Fig. 1. Internal volumes (V) ranged from
3340 to 6750 cm3
, and surface areas of the gap between upper and lower
dome (G) from 75 to 171 cm2
(Fig. 1). Differences in the PUF disks were
smaller, with diameters of 13.2–14.1 cm, thicknesses of 1.1–1.5 cm, and
densities of 0.020 to 0.031 g/cm3
. Full dimensions of samplers and PUF
disks are given in Table S2.
3.1. Phase 1 – different samplers, same PUFs, same laboratory
The objective of Phase 1 was to isolate and identify the influence of
different sampler designs on sampler uptake by comparing different
housings fitted with identical PUF disks, deployed simultaneously at the
same site, and analyzed in a single laboratory. The differences in the
masses of SVOCs sampled by each PAS should therefore give insight into
the variability in uptake introduced only by sampler geometry (dome
sizes, gap and overhang) and deployment (e.g., fixed vs. free swinging).
Compounds that were below detection limits in 50 % or more of the
PAS were excluded from further interpretation. This resulted in the
exclusion of eight of 29 PAHs (naphthalene, biphenyl, acenaphthylene,
cyclopenta(cd)pyrene, perylene, dibenzo(ah)anthracene, dibenzo(ac)
anthracene, and anthanthrene), five of 12 OCPs (β- and δ-HCH, and o,p’DDD,
p,p’-DDD and o,p’-DDE), two of seven PCBs (PCB-138, PCB-180),
and six of ten PBDEs (BDE-66, BDE-85, BDE-99, BDE-100, BDE-153,
BDE-154). Any results <MDL for the remaining compounds were
substituted with 0.5*MDL for statistical analysis. For PAHs and PCBs,
this resulted in substitution of 3% of records, and no substitutions for
OCPs; for PBDEs the substitution was required for 37 % of records, thus
there is higher uncertainty in the statistical analysis of the PBDEs than
for the other compound groups.
We evaluated the variability in the individual samplers by normalizing
masses measured per compound to the median masses for all
samplers (Fig. 2a). Compounds were mostly within a relatively narrow
range, spanning 519–939 ng/PUF for Σ21PAHs (Table S9), 4.86–12.3
ng/PUF for Σ6PCBs (Table S10), 0.80–2.21 for Σ6DDXs (Table S11),
2.54–5.50 ng/PUF for Σ4HCHs (Table S11), and 59.5–109 pg/PUF for
Σ9PBDEs (excluding BDE-209; range for BDE-209 was < MDL–1300 pg/
PUF) (Table S12). Individual compound masses were within one order of
L. Melymuk et al.
Environmental Science and Policy 125 (2021) 1–9
5
magnitude of the median for all of the sampler designs, with the
exception of one record of BDE-209 (Fig. 2a). In general, the variability
increased with increasing molecular weight of compounds, e.g., high
molecular weight PAHs and PBDEs had the highest variability.
Five of the samplers (Samplers 1–5) were identical configurations of
the commercially available Tisch sampler design (TE-200, Tisch Environmental,
Cleves, OH). We considered these samplers as replicates and
used them to assess the typical range of variability between identical codeployed
samplers (i.e., due to environmental conditions and laboratory
uncertainty rather than differences in sampler configuration). These
Tisch replicates were normally distributed (Shapiro-Wilks test, α = 0.05,
except for p,p’-DDE, and BDE-47 and 209), so means and standard deviations
were used to evaluate their distribution. The relative standard
deviation (RSD) of the five replicates ranged from 1.98 % for acenaphthene
to 55.6 % for BDE-183 (Table S13). The highest RSDs were
observed for the higher molecular weight compounds, i.e. 5- and 7-ring
PAHs and BDEs 183 and 209 (average 22.7 % vs. 8.9 % for all other
compounds), which likely reflects two possible effects on higher
molecular weight compounds: (1) higher analytical uncertainty and (2)
variable sampler uptake of particulates. Lower ambient levels of higher
molecular weight compounds may lead to greater measurement uncertainties
as MDLs are approached. The larger variability in the uptake
of particle-bound compounds to PUF-PAS has been extensively discussed
in other publications (e.g., Holt et al., 2017; Markovic et al., 2015).
However, we note that the Tisch sampler is reported as having high
particle infiltration in Markovic et al. (2015), suggesting that even when
particle infiltration is high, there remains higher variability/uncertainty
for particle uptake than gaseous compounds.
The RSD determined from the Tisch samplers was assumed to
represent the typical uncertainty in a sampler due to environmental
variability and laboratory uncertainty and was therefore used to flag
cases when the variability between PAS was beyond the range of typical
differences between identical samplers. Upper and lower boundaries
were calculated per compound as the median of all sampler masses ±3
times the percent uncertainty (Tables S9-S12). We identify the specific
cases and compounds where variability exceeded these thresholds.
Fig. 2. Variability in analytes detected by PUFPAS
due to differences in (a) sampler geometry
and installation (Phase 1, n = 16), (b) analytical
methods between laboratories (Phase 2, n =
13), and (c) combined sampler and analytical
differences (Phase 3, n = 13). Boxes represent
the 25th
to 75th
percentiles, with the median
(50th
percentile) as a horizontal black line. Individual
SVOC levels were normalized to the
median. Whiskers represent ±1.5 times the
interquartile range (IQR) with individual points
indicating outliers.
L. Melymuk et al.
Environmental Science and Policy 125 (2021) 1–9
6
No single sampler design had all results within the acceptable range
(median±3xRSD of Tisch samplers), however most samplers had only a
few compounds outside of this range (Tables S9-S12), with the Tisch
samplers (Samplers 1–5) demonstrating the least variability across all
compound groups (Fig. 3a). A few samplers had more substantial deviations
from the median values: Sampler 6 measured 24 % lower total
SVOC masses than the median, with OCPs particularly low; Sampler 13
recorded 38 % lower total SVOC masses, with OCPs and PCBs particularly
low (Fig. 3a); Sampler 8 recorded 11 % higher total SVOC masses
(Fig. 3a). BDE-209 was substantially more variable than other compounds,
with most cases < MDL and five samplers with notably high
values.
We also examined the correlation between the measured masses and
the sampler geometry parameters. The strongest correlation was with
the overhang distance (i.e. the overlap between the top and bottom
domes), with a significant negative relationship between overhang distance
and mass of SVOCs collected by the PUF (Spearman ρ of -0.796, p
< 0.01 for correlation between overhang distance and total SVOCs), e.g.,
larger overhang is correlated with lower mass. This suggests reduced
airflow to the inside of the sampler housing due to overhang, and
consequent lower uptake of compounds, especially particle-bound
compounds (Markovic et al., 2015). Despite large ranges in the other
sampler geometries (e.g., sampler volumes ranging from 3340 to 6750
cm3
), no other sampler geometry parameters had significant correlations
with the mass of SVOCs collected.
While individual differences were generally small, small systematic
errors across many individual compounds can lead to significant differences
in compound group totals, e.g., indicators such as ΣEPA-16
PAHs and Σ7PCBs, which are frequently applied in policy and effectiveness
evaluation. For example, ΣEPA-16 PAHs is 482 ng/PUF in the
lowest reporting sampler (Sampler 13) vs. 873 ng/PUF in the highest
(Sampler 10), and for Σ7PCB the mass collected was 2440 pg/PUF vs.
5120 pg/PUF in Samplers 13 vs. 8. While these are the extremes of the
set, it does suggest that uncertainties introduced by differences in
sampler configuration can account for up to 50 % variation in reported
sample masses. Moreover, it is known that at sites with meteorological
Fig. 3. Reported SVOC masses normalized to
median for (a) Phase 1, (b) Phase 2, and (c)
Phase 3. Phase 1 shows variability between
different PUF-PAS sampler designs (1–16)
analyzed at a single laboratory; Phase 2 shows
variability between co-deployed, identical PUFPAS
samplers analyzed in 13 different laboratories
(A–M); and Phase 3 shows differences in
co-deployed different samplers analyzed in
different laboratories (Labs A-M). Samplers 1–5
in Phase 1 are identical Tisch TE-200 samplers.
Note the smaller y-axis scale in Phase 1
compared to Phases 2 and 3.
L. Melymuk et al.
Environmental Science and Policy 125 (2021) 1–9
7
extremes (e.g., coastal sites with very high wind speeds, polar sites) the
protective effect of the sampler housing on PAS uptake rates is variable,
and the effects of differences between sampler configurations could be
exacerbated, leading to variation beyond what was seen in this test case.
3.2. Phase 2 – same samplers, same PUFs, different laboratories
Results for Phase 2 indicate the extent of variability between SVOC
measurements introduced by differences in transport, laboratory
handling and analysis of PUF disks. Major differences were identified in
two aspects: field blank contamination and sample masses.
Field blank masses varied over 2 orders of magnitude, spanning
187–8260 pg/PUF for Σ7PCBs, <3–640 pg/PUF for HCB, 50–35600 pg/
PUF for Σ3HCHs, <50–98400 pg/PUF for Σ6DDXs, 17.9–1670 ng/PUF
for Σ16PAHs, and <25–3250 pg/PUF for Σ10PBDEs (Table S14). The 2–3
order of magnitude range in field blanks suggests either errors in analysis
and/or large variations in contamination during transport and
processing, with the implication being that choices regarding blank
treatment can have a large impact on reported values. We also note that
in three cases, masses of PBDEs and DDXs reported in field blanks
exceeded those reported in the samples (Fig. S2). PBDEs and other flame
retardants are often identified as a particular challenge for analysis in
PUF samples due to their prevalence in equipment and electronics, and
low ambient levels at many locations, including the NILU site in this
study.
Despite the significant contribution of blanks to total samples for a
few laboratories (laboratories A, J, K; Fig. S1), we did not further adjust
samples for blanks as we had only one field blank per laboratory, and in
practice many laboratories use blanks only as a quality control in longterm
monitoring, rather than for data adjustment.
Phase 2 sample results clearly indicate that the variability due to
laboratory analysis is much higher than that introduced by sampler
geometry identified in Phase 1. The Σ16PAH reported spanned
1830–5870 ng/PUF, a substantially larger span than that observed in
Phase 1 (Table S15). Variations were even higher for POPs, with reported
values spanning 3 or more orders of magnitude: Σ3HCHs spanning
283–263000 pg/PUF, Σ6DDX from 171 to 70100 pg/PUF (Table
S16), Σ7PCBs from 378 to 29300 pg/PUF (Table S17) and Σ10PBDEs
from 0.35 to 1950 pg/PUF (Table S18).
We explored whether the choice not to adjust sample masses for field
blank contamination led to such large ranges, but the effect was limited.
For example, when blanks were subtracted from measured values
Σ7PCBs ranged 378–23000 pg/PUF, Σ3HCHs ranged 243–227000 pg/
PUF, Σ6DDX ranged 140–14200 pg/PUF, and the range for PBDEs did
not change, suggesting that the large variability in their reported masses
is independent of differences in blank contamination.
As with Phase 1, we assessed the variability by normalizing sampled
SVOC masses to the median of the whole set of reported masses (Fig. 2b).
It is clear from the range of the normalized data, with some compounds
covering a range of ~0.1–1000 around the median, that much greater
differences in reported masses are introduced by laboratory analysis
than by differences in sampler design (where the range of normalized
masses was much less than 0.1–10 times the median, Fig. 2a). We note
that some of the compounds with the largest ranges in Phase 2 (e.g., PCB
180, dibenzo[a,h]anthracene, β-HCH) were below detection in Phase 1
in all samples, which inherently suggests that the variability for these
compounds is smaller than the LOD. Of the compounds detected in both
phases, all compounds except chrysene had at least a 2x higher span in
reported values in Phase 2 compared with Phase 1, and the span of
values was more than 50x higher for BDE-28, BDE-47 and p,p’-DDT.
This substantial increase in variability in Phase 2 compared with
Phase 1 indicates that, due to the uncertainty introduced by analysis in
different laboratories, close to 50 % of laboratories report SVOC masses
with order of magnitude differences from a consensus values, i.e., median
of all laboratories (Fig. 3b). We compared the deviations from
consensus values with the analytical methods used across the
laboratories (Table S8) but did not find any relationship with the basic
parameters of extraction method, solvent, clean-up or instrumental
analysis and the laboratory performance. However, it is clear from the
difference between Phases 1 and 2 that the sample processing and
analysis are an important contributor to the differences in the reported
values between laboratories. We are unable to identify specific method
influences on performance due to the large differences in methods used
across the 12 laboratories (e.g., eight different extraction methods),
which prevents us from making generalizations.
The differences introduced by analytical methods are higher than
what has been typically identified by global interlaboratory evaluations
on POPs, e.g., the UNEP-supported Bi-ennial Global Interlaboratory
Assessments on POPs (Nilsson et al., 2014; van Bavel et al., 2012; van
der Veen et al., 2017; Van Leeuwen et al., 2013). However, the Bi-ennial
Global Interlaboratory Assessments on POPs included injection-ready
test mixtures and environmental matrices high in organic matter or
lipid content, or else spiked air samples (Fiedler et al., 2020, 2017;
Nilsson et al., 2014), with relatively high concentrations. These concentrations
were higher than what are found in typical rural/remote air
samples, and many interlaboratory evaluations have reported lower
precision and accuracy at lower concentrations (Melymuk et al., 2018,
2015; Su and Hung, 2010). The low concentrations in our samples
collected from semi-rural Norway likely contributed to the poorer performance
in our study. Further, our study captured the full scope of
variability in sample processing, as laboratories received the PUF material,
and were required to extract and purify the samples. This contrasts
with the Bi-ennial Global Interlaboratory Assessments on POPs,
where laboratories analyzed air extracts not requiring further clean-up
(van der Veen et al., 2017); the variability introduced from extraction
and clean-up of complex matrices is a large contributor to the differences
between laboratories (Abalos et al., 2013; Melymuk et al., 2018; Su and
Hung, 2010; Van Leeuwen et al., 2013).
If these differences reflect a recurring deviation by certain laboratories,
this suggests that international comparisons of SVOC data could
be highly biased when analyses are performed at different laboratories.
This does not impact internal reporting within individual laboratories,
therefore comparability within individual laboratory research studies
and monitoring programs, (e.g., to assess regional spatial differences or
temporal changes), should still be valid. Yet when comparisons involve
merging data from multiple studies/monitoring networks, (e.g., for
model comparisons), there may be significant biases introduced in the
interpretation on a global level. This may also apply to data reported
from active samplers, which typically follow similar methods of laboratory
analysis.
3.3. Phase 3 – different samplers, different PUFs, different laboratories
Phase 3 gives an indication of overall variability, encompassing
differences between sampler configurations identified in Phase 1 and
between laboratory analysis identified in Phase 2. As with Phase 2, large
differences were observed in both field blanks and reported sample
masses. Field blanks had a similar distribution and range to those of
Phase 2 (Table S19) and in two cases the blank mass exceeded that of the
sample (Fig. S3). As in Phase 2, no blank adjustment was applied to the
reported sample masses.
Phase 3 showed variability in sample masses comparable to that of
Phase 2. The Σ16PAH ranged 1640–5010 ng/PUF (Table S20), a similar
range to that observed in Phase 2. As in Phase 2, variations were higher
for POPs, with ranges spanning >3 orders of magnitude: from 247 to
182000 pg/PUF for ΣHCH, 123–58200 pg/PUF for ΣDDX (Table S21),
283–25900 pg/PUF for Σ7PCBs (Table S22), and 0.44–3110 pg/PUF for
ΣPBDEs (Table S23). As with Phases 1 and 2, we assessed the variability
by normalizing masses to the median of the whole set of reported masses
(Fig. 2c). The range of the normalized data, covering ~0.001–1000
around the median, was similar to Phase 2 (Fig. 2b), and much greater
than Phase 1 (Fig. 2a).
L. Melymuk et al.
Environmental Science and Policy 125 (2021) 1–9
8
Phase 3 combined three factors affecting variability (1) sampler
design, (2) sample transport and laboratory analysis, and (3) differences
in PUF disk parameters (size, density; Table S2). While we did not have a
separate study phase to isolate the influence of differences in PUF disks,
the similarity of results from Phases 2 and 3 suggest that this is minimal,
and likely comparable to or less than the variability introduced by
sampler design. This is supported by previous work identifying 40–60 %
differences in particle uptake and distribution in high density vs. low
density PUF disks (Chaemfa et al., 2009), which was a larger range in
density than that seen within our study.
As in Phase 2, a large fraction of laboratories reported masses that
differed by more than one order of magnitude from the median (Fig. 3c),
suggesting that, when PUF-PAS data are reported by different laboratories
using different samplers, ~50 % of laboratories are reporting
values outside the boundaries of acceptable uncertainty, which can
create a major challenge in the global comparability of SVOC data from
PUF-PAS monitoring.
The similarity in the ranges and collected masses between Phases 2
and 3 suggests the major influence on differences in comparability between
laboratories is sample transport, processing and analytical variation,
and this source of variability overwhelms any smaller differences
due to sampler design identified in Phase 1.
4. Conclusions and implications
Phase 1 of this international intercomparison revealed that variations
in the double-dome PAS housings used by different research groups
contributes relatively little to uncertainties in sampled masses of PCBs,
PAHs, PBDEs, and OCPs for a 3-month deployment, with differences in
reported masses due to PUF-PAS sampler configurations not exceeding
50 %. Any difference in uptake between samplers appears related to
differences in airflow into the sampler housing due to the amount of
overhang of the upper dome over the lower.
Phase 2 of the study, which assessed uncertainty associated with
laboratory performance (i.e., each laboratory analyzing the same type of
sample), showed a substantial increase in uncertainty. Reported masses
varied by an order of magnitude or greater, with ~10× differences between
individual masses and medians for many compounds, as well as a
few more extreme outliers (Fig. 2). Phase 3 of the study confirmed the
results of Phase 2; similar ranges of reported values and large deviations
from medians for some individual laboratories/compounds show that
the major influence on comparability between laboratories is sample
processing and analytical variation, and this source of variability overwhelms
any smaller differences due to sampler design and differences in
PUF disk size and density. Considering only the three laboratories that
currently report PUF-PAS data to the Stockholm Convention GMP, the
variability is lower, but there still exist order-of-magnitude differences
in reported masses of Σ7PCBs, HCB and Σ6DDX.
The high level of uncertainty observed in Phases 2 and 3 of this study
indicates that current global air monitoring data are not directly comparable
between different laboratories/monitoring programs for most of
the SVOCs included in this study. However, this does not mean that the
data are not internally consistent (i.e., within a program using a single
laboratory) for deriving valid spatial and temporal trends. Yet on a
global scale, it is clear that merging data from multiple laboratories must
be done with caution. With current levels of uncertainty, it is not feasible
to compare results between laboratories/monitoring programs without
prior assurance of comparability in reported data. It is also clear that the
uncertainty is not limited to passive sampling but also pertains to active
air sampling results if similar extraction, clean-up and analytical procedures
are followed. This uncertainty may be due to several factors
including instrumental methods, sample processing procedures, potential
laboratory contamination due to solvents and other sources, and
differences in analytical standards (not assessed in the current study).
Additional uncertainties will also arise if concentrations are adjusted to
volumetric units (e.g., pg/m3
), since slightly different conventions may
be used among laboratories for estimating effective air sample volumes.
These uncertainties can be resolved through ongoing participation in
intercalibration exercises and adoption of best practices. The high
variability between laboratories means that a crucial part of any efforts
to integrate and evaluate global spatial patterns of POPs in air must
require and implement intercalibration to assess and account for uncertainties,
repeated at regular intervals for both active and passive air
sampling. Some examples of such international repeated intercalibration
exercises already exist, most notably the AMAP/EMEP intercomparison
exercises (Schlabach et al., 2012; Tkatcheva et al., 2013).
Establishing such actions as a part of research infrastructure is even
more necessary given the efforts to mobilize data from additional
monitoring networks to contribute to global data repositories.
Effectiveness evaluation of international SVOC actions (Stockholm
Convention, CLRTAP) relies on the provision of high-quality air monitoring
data. PUF-PAS are a valuable tool to provide this information,
particularly given that the slight differences in PUF-PAS sampler designs
do not greatly affect data comparability. Yet without establishment of
frameworks to ensure analytical comparability, our understanding of
global spatial POP distributions will be biased by the regional differences
in analytical performance.
CRediT authorship contribution statement
Lisa Melymuk – Conceptualization, Methodology, Writing, Formal
Analysis; Pernilla Bohlin Nizzetto– Conceptualization, Methodology,
Resources, Review & Editing; Tom Harner – Conceptualization, Resources,
Review & Editing, Supervision; Kevin B. White – Formal
Analysis, Visualization, Review & Editing; Xianyu Wang – Investigation,
Resources, Review & Editing; Maria Yumiko Tominaga – Investigation,
Resources, Review & Editing; Jun He – Investigation,
Resources, Review & Editing; Jun Li – Investigation, Resources, Review
& Editing; Jianmin Ma – Investigation, Resources, Review & Editing;
Wan-Li Ma – Investigation, Resources, Review & Editing; Beatriz H.
Aristiz´abal – Investigation, Resources, Review & Editing; Annekatrin
Dreyer – Investigation, Resources, Review & Editing; Bego˜na Jim´enez
– Investigation, Resources, Review & Editing; Juan Mu˜noz-Arnanz –
Investigation, Resources, Review & Editing; Mustafa Odabasi – Investigation,
Resources, Review & Editing; Yetkin Dumanoglu – Investigation,
Resources, Review & Editing; Baris Yaman – Investigation,
Resources, Review & Editing; Carola Graf – Investigation, Resources,
Review & Editing; Andrew Sweetman – Investigation, Resources, Review
& Editing; Jana Kl´anov´a – Resources, Supervision, Review &
Editing.
Declaration of Competing Interest
The authors report no declarations of interest.
Acknowledgements
This work was supported by the EU Horizon 2020 program (ERA
PLANET: 689443, e-SHAPE: 820852 and CETOCOEN EXCELLENCE
TEAMING 2-857560), the Czech Ministry of Education, Youth and
Sports (RECETOX RI: LM2018121), the European Structural and Investment
Funds (CETOCOEN EXCELLENCE TEAMING 2: CZ.02.1.01/
0.0/0.0/17_043/0009632), and a grant from Iceland, Liechtenstein and
Norway (EHP-CZ02-OV-1-029-2015). This publication reflects only the
authors’ view and the European Commission is not responsible for any
use that may be made of the information it contains. We thank the
RECETOX Trace Analytical Laboratories for Phase 1 sample processing
and analysis, and Ondˇrej S´aˇnka for the GMP site map. Partial funding for
the participation of the GAPS Network was provided through the
Chemicals Management Plan (Government of Canada) and the Northern
Contaminants Program (NCP) and analytical support was provided by
Cassandra Rauert, Ky Su and Anita Eng.
L. Melymuk et al.
Environmental Science and Policy 125 (2021) 1–9
9
Appendix A. Supplementary data
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.envsci.2021.08.003.
References
Abalos, M., Abad, E., Van Leeuwen, S.P.J., De Boer, J., Lindstr¨om, G., Van Bavel, B.,
Fiedler, H., 2013. Results for PCDD/PCDF and dl-PCBs in the first round of UNEPs
biennial global interlaboratory assessment on persistent organic pollutants. TrAC Trends
Anal. Chem. 46, 98–109. https://doi.org/10.1016/j.trac.2012.11.003.
Borůvkov´a, J., Gregor, J., ˇSebkov´a, K., Bedn´aˇrov´a, Z., Kalina, J., Hůlek, R., Duˇsek, L.,
Holoubek, I., Kl´anov´a, J., 2015. GENASIS – Global Environmental Assessment and
Information System [Online] [WWW Document]. Masaryk Univ., 2015 [cit. 2017-
06-19]. Available http//www.genasis.cz. Version 2.0. ISSN 1805-3181.
Chaemfa, C., Barber, J.L., Gocht, T., Harner, T., Holoubek, I., Kl´anov´a, J., Jones, K.C.,
2008. Field calibration of polyurethane foam (PUF) disk passive air samplers for
PCBs and OC pesticides. Environ. Pollut. 156, 1290–1297.
Chaemfa, C., Wild, E., Davison, B., Barber, J.L., Jones, K.C., 2009. A study of aerosol
entrapment and the influence of wind speed, chamber design and foam density on
polyurethane foam passive air samplers used for persistent organic pollutants.
J. Environ. Monit. 11, 1135–1139.
Fiedler, H., van der Veen, I., De Boer, J., 2017. Bi-ennial Global Interlaboratory
Assessment on Persistent Organic Pollutants – Third Round 2016/2017. Geneva.
Fiedler, H., van der Veen, I., de Boer, J., 2020. Global interlaboratory assessments of
perfluoroalkyl substances under the Stockholm Convention on persistent organic
pollutants. TrAC - Trends Anal. Chem. 124, 115459 https://doi.org/10.1016/j.
trac.2019.03.023.
Holt, E., Bohlin-Nizzetto, P., Borůvkov´a, J., Harner, T., Kalina, J., Melymuk, L.,
Kl´anov´a, J., 2017. Using long-term air monitoring of semi-volatile organic
compounds to evaluate the uncertainty in polyurethane-disk passive samplerderived
air concentrations. Environ. Pollut. 220, 1100–1111. https://doi.org/
10.1016/j.envpol.2016.11.030.
Kalina, J., Scheringer, M., Borůvkov´a, J., Kukuˇcka, P., Pˇribylov´a, P., Bohlin-Nizzetto, P.,
Kl´anov´a, J., 2017. Passive air samplers as a tool for assessing long-term trends in
atmospheric concentrations of semivolatile organic compounds. Environ. Sci.
Technol. 51, 7047–7054. https://doi.org/10.1021/acs.est.7b02319.
Kl´anov´a, J., Harner, T., 2013. The challenge of producing reliable results under highly
variable conditions and the role of passive air samplers in the Global Monitoring
Plan. Trends Analyt. Chem. 46, 139–149.
Lee, S.C., Harner, T., Pozo, K., Shoeib, M., Wania, F., Muir, D.C.G., Barrie, L.A., Jones, K.
C., 2007. Polychlorinated naphthalenes in the Global Atmospheric Passive sampling
(GAPS) study. Environ. Sci. Technol. 41, 2680–2687.
Markovic, M.Z., Prokop, S., Staebler, R.M., Liggio, J., Harner, T., 2015. Evaluation of the
particle infiltration efficiency of three passive samplers and the PS-1 active air
sampler. Atmos. Environ. 112, 289–293.
Melymuk, L., Goosey, E., Riddell, N., Diamond, M.L., 2015. Interlaboratory study of
novel halogenated flame retardants: INTERFLAB. Anal. Bioanal. Chem. 407,
6759–6769. https://doi.org/10.1007/s00216-015-8843-7.
Melymuk, L., Diamond, M.L., Riddell, N., Wan, Y., Vojta, ˇS., Chittim, B., 2018.
Challenges in the analysis of novel flame retardants in indoor dust: results of the
INTERFLAB 2 interlaboratory evaluation. Environ. Sci. Technol. 52, 9295–9303.
https://doi.org/10.1021/acs.est.8b02715.
Mu˜noz-Arnanz, J., Roscales, J.L., Ros, M., Vicente, A., Jim´enez, B., 2016. Towards the
implementation of the Stockholm Convention in Spain: five-year monitoring
(2008–2013) of POPs in air based on passive sampling. Environ. Pollut. 217,
107–113. https://doi.org/10.1016/j.envpol.2016.01.052.
Mu˜noz-Arnanz, J., Roscales, J.L., Vicente, A., Ros, M., Barrios, L., Morales, L., Abad, E.,
Jim´enez, B., 2018. Assessment of POPs in air from Spain using passive sampling from
2008 to 2015. Part II: Spatial and temporal observations of PCDD/Fs and dl-PCBs.
Sci. Total Environ. 634, 1669–1679. https://doi.org/10.1016/j.
scitotenv.2018.04.164.
Nilsson, H., van Bavel, B., van der Veen, I., 2014. United Nations Environmental
Programme, Bi-ennial Global Interlaboratory Assessment on Persistent Organic
Pollutants – Second Round 2012/2013. Geneva, Switzerland.
Pozo, K., Harner, T., Wania, F., Muir, D.C.G., Jones, K.C., Barrie, L.A., 2006. Toward a
global network for persistent organic pollutants in air: results from the GAPS study.
Environ. Sci. Technol. 40, 4867–4873.
Pozo, K., Harner, T., Lee, S.C., Wania, F., Muir, D.C.G., Jones, K.C., 2009. Seasonally
resolved concentrations of persistent organic pollutants in the global atmosphere
from the first year of the GAPS study. Environ. Sci. Technol. 43, 796–803. https://
doi.org/10.1021/es802106a.
Pˇribylov´a, P., Kares, R., Borůvkov´a, J., Cupr, P., Prokeˇs, R., Kohoutek, J., Holoubek, I.,
Kl´anov´a, J., 2012. Levels of persistent organic pollutants and polycyclic aromatic
hydrocarbons in ambient air of Central and Eastern Europe. Atmos. Pollut. Res. 3,
494–505.
Rauert, C., Harner, T., Schuster, J.K., Eng, A., Fillmann, G., Castillo, L.E., Fentanes, O.,
Villa Ibarra, M., Miglioranza, K.S.B., Moreno Rivadeneira, I., Pozo, K., Aristiz´abal
Zuluaga, B.H., 2018. Atmospheric concentrations of new persistent organic
pollutants and emerging chemicals of concern in the group of Latin America and
Caribbean (GRULAC) region. Environ. Sci. Technol. 52, 7240–7249. https://doi.org/
10.1021/acs.est.8b00995.
Roscales, J.L., Mu˜noz-Arnanz, J., Ros, M., Vicente, A., Barrios, L., Jim´enez, B., 2018a.
Assessment of POPs in air from Spain using passive sampling from 2008 to 2015.
Part I: Spatial and temporal observations of PBDEs. Sci. Total Environ. 634,
1657–1668. https://doi.org/10.1016/j.scitotenv.2018.03.043.
Roscales, J.L., Mu˜noz-Arnanz, J., Ros, M., Vicente, A., Jim´enez, B., 2018b. Does the
number of field blanks influence reported air POP concentrations in monitoring
programs based on PUF-PAS? Organohalogen Compd. 80, 509–512.
Schlabach, M., Farag-clement, R., Hung, H., Kallenborn, R., 2012. AMAP/EMEP/NCP
Inter-laboratory Study for POP Analysis 2010. EMEP Co-operative Programme for
Monitoring and Evaluation of the Long-range Transmission of Air Pollutants in
Europe.
Shoeib, M., Harner, T., 2002. Characterization and comparison of three passive air
samplers for persistent organic pollutants. Environ. Sci. Technol. 36, 4142–4151.
Su, Y., Hung, H., 2010. Inter-laboratory comparison study on measuring semi-volatile
organic chemicals in standards and air samples. Environ. Pollut. 158, 3365–3371.
https://doi.org/10.1016/j.envpol.2010.07.041.
Su, Y., Hung, H., Stern, G., Sverko, E., Lao, R., Barresi, E., Rosenberg, B., Fellin, P., Li, H.,
Xiao, H., 2011. Bias from two analytical laboratories involved in a long-term air
monitoring program measuring organic pollutants in the Arctic: a quality assurance/
quality control assessment. J. Environ. Monit. 13, 3111–3118. https://doi.org/
10.1039/c1em10513b.
Tkatcheva, V., Ali, B., Reiner, E.J., 2013. Northern Contaminants Interlaboratory Quality
Assurance Program (NCP III – Phase 7) [WWW Document]. URL. www.aadnc-aandc.
gc.ca/eng/1100100035611/1100100035612.
UNEP, 2009. Stockholm Convention Global Monitoring Plan for Persistent Organic
Pollutants: First Global Monitoring Report. Geneva.
van Bavel, B., van Leeuwen, S., de Boer, J., 2012. Bi-ennial Global Interlaboratory
Assessment on Persistent Organic Pollutants – First Round 2010/2011. Geneva,
Switzerland.
van der Veen, I., Fiedler, H., de Boer, J., 2017. Bi-ennial Global Interlaboratory
Assessment on Persistent Organic Pollutants – Third Round 2016/2017.
Organohalogen Compd. 79, 575–578.
Van Leeuwen, S.P.J., De Boer, J., Van Leeuwen, S.P.J., Van Bavel, B., 2013. First
worldwide UNEP interlaboratory study on persistent organic pollutants (POPs), with
data on polychlorinated biphenyls and organochlorine pesticides. TrAC - Trends
Anal. Chem. 46, 110–117. https://doi.org/10.1016/j.trac.2012.12.020.
Wania, F., Shunthirasingham, C., 2020. Passive air sampling for semi-volatile organic
chemicals. Environ. Sci. Process. Impacts 22, 1925–2002. https://doi.org/10.1039/
D0EM00194E.
White, K.B., Kalina, J., Scheringer, M., Pˇribylov´a, P., Kukuˇcka, P., Kohoutek, J.,
Prokeˇs, R., Kl´anov´a, J., 2021. Temporal trends of persistent organic pollutants across
Africa after a decade of MONET passive air sampling. Environ. Sci. Technol. 55 (14),
9413–9424. https://doi.org/10.1021/acs.est.0c03575.
L. Melymuk et al.
Jílková, Simona, Lisa Melymuk, Šimon Vojta, Martina Vykoukalová, Pernilla Bohlin-Nizzetto,
and Jana Klánová. 2018. "Small-Scale Spatial Variability of Flame Retardants in Indoor Dust
and Implications for Dust Sampling." Chemosphere 206: 132-41.
https://doi.org/10.1016/J.CHEMOSPHERE.2018.04.146
Small-scale spatial variability of flame retardants in indoor dust and
implications for dust sampling
Simona Jílkova a
, Lisa Melymuk a, *
, Simon Vojta a, 1
, Martina Vykoukalova a
,
Pernilla Bohlin-Nizzetto b
, Jana Klanova a
a
Research Centre for Toxic Compounds in the Environment (RECETOX), Masaryk University, Brno, Czech Republic
b
Norwegian Institute for Air Research (NILU), Kjeller, Norway
h i g h l i g h t s
 Flame retardant levels in indoor dust varied significantly between and within rooms.
 Up to 1000-fold differences exist in flame retardant levels within the same room.
 Levels of hexabromobenzene were elevated in computer room dust.
 Composite dust samples are recommended due to within-room spatial variability.
a r t i c l e i n f o
Article history:
Received 24 January 2018
Received in revised form
18 April 2018
Accepted 22 April 2018
Available online 23 April 2018
Handling Editor: Myrto Petreas
Keywords:
Indoor dust
Flame retardants
Spatial variability
Sampling methods
a b s t r a c t
Indoor dust is often used to evaluate levels of organic compounds indoors, particularly for compounds
with indoor sources, such as flame retardants (FRs). Yet there are uncertainties about the type of information
that can be obtained from indoor dust. This study reports detailed dust sampling to assess
spatial variability in indoor dust concentrations, the relationship between FR sources and dust, and the
implications when interpreting dust concentrations. Multiple dust samples were collected from a range
of surface types in three large rooms: a residential flat, a university seminar room, and a university
computer room. Samples were analysed for polybrominated diphenyl ethers (PBDEs), novel halogenated
flame retardants (NFRs) and organophosphate esters (OPEs).
FR levels in dust varied significantly between and within rooms. Levels typically ranged over one order
of magnitude within a room, and up to four orders of magnitude for a few OPEs. The spatial distribution
of FRs related (in some cases) to proximity to sources, surface properties, and dust surface loadings.
Differences also existed between surface and floor dusts, e.g., the contribution of TBOEP to
P
OPEs was
higher in floor than surface dust, which has implications for human exposure assessment; adults typically
have more contact with elevated surfaces, while young children have greater contact with floor
surfaces. Overall, significant spatial heterogeneity exists in indoor dust, even in seemingly homogeneous
indoor spaces, thus hampering comparability between studies and locations when single samples are
collected. Composite samples are strongly recommended to limit the influence of spatial heterogeneity.
© 2018 Elsevier Ltd. All rights reserved.
1. Introduction
Flame retardants (FRs) are chemicals used in electronics, furniture,
and building materials to reduce flammability in order to meet
fire safety regulations (de Wit, 2002). Two types of organic FRs are
commonly used: (1) halogenated flame retardants (HFRs) and (2)
organophosphate esters (OPEs). HFRs consist of brominated and
chlorinated flame retardants (Bergman et al., 2012), including
polybrominated diphenyl ethers (PBDEs) and so-called “novel”
halogenated flame retardants (NFRs) used as replacements for
banned PBDEs (Betts, 2008). OPEs are used as both FRs and plasticizers
and, as with the NFRs, OPE production also increased after
restrictions on PBDEs (van der Veen and de Boer, 2012).
Due to the nature of FR use, high concentrations are found in
* Corresponding author. RECETOX, Masaryk University, Kamenice 753/5, pavilion
A29, 62500 Brno, Czech Republic.
E-mail address: melymuk@recetox.muni.cz (L. Melymuk).
1
Present address: Graduate School of Oceanography, University of Rhode Island,
Narragansett, Rhode Island 02882-1197, USA.
Contents lists available at ScienceDirect
Chemosphere
journal homepage: www.elsevier.com/locate/chemosphere
https://doi.org/10.1016/j.chemosphere.2018.04.146
0045-6535/© 2018 Elsevier Ltd. All rights reserved.
Chemosphere 206 (2018) 132e141
indoor environments, and differences in indoor levels have been
noted between countries (Harrad et al., 2008b; Sj€odin et al., 2008;
Venier et al., 2016; Vykoukalova et al., 2017) or within countries
(Karlsson et al., 2007; Stapleton et al., 2008); between types of
buildings (Ali et al., 2011; Cequier et al., 2014; Harrad et al., 2008a);
as well as between rooms within a building (Al-Omran and Harrad,
2018; Kuang et al., 2016; Muenhor and Harrad, 2012; Stapleton
et al., 2008). Comparisons of different rooms in the same buildings
have identified that measured concentrations are influenced
by the presence of furniture and electronics. For example, Muenhor
and Harrad (2012) suggested that where and when an indoor dust
sample is taken exerts a substantial influence on the level of
contamination detected, and relates to the proximity to potential
sources such as electronics and carpet. Harrad et al. (2009) found
decreasing concentrations of hexabromocyclododecane with
increasing distance from a TV. Al-Omran and Harrad (2018) identified
consistently higher FR concentrations in dust from elevated
surfaces than from floors, and differences in dusts from two areas of
the same rooms. Kuang et al. (2016) identified higher levels of
PBDEs and NFRs in living rooms and bedrooms than in kitchens,
however, this could not be directly linked to any source. Rather, it
was suggested that environmental factors (e.g., higher moisture in
kitchen) lead to lower dust concentrations in kitchens. Thus, a key
question that arises considering all of the indoor spatial differences
between and within buildings is how much of the variation is due
to real differences in the indoor environments, and how much is an
artifact of the choice of sampling location within a given room.
Moreover, in broad indoor studies covering many locations, single
dust samples are often considered representative of whole-room or
even whole-building conditions, however, it is not clear whether
this assumption is valid.
To address this question, the main objectives of this study were
(1) to identify the differences in FR profiles and levels obtained by
two sampling methods (wet wipes and vacuuming), (2) to identify
the range of concentrations within individual rooms and thus
identify what effect the choice of sampling location may have on
reported concentrations, and (3) to determine the extent to which
concentration levels are influenced by room type and proximity to
room elements (electronics, furnishings, different usage of the
space, etc.), and identify whether greater heterogeneity in room
furnishings and use leads to greater heterogeneity in indoor dust.
These objectives were addressed by detailed dust sampling (>9
samples per room) in three indoor environments.
2. Methods
2.1. Sampling strategy
Three indoor environments were chosen for sampling. The first
was a residential flat, where the entrance, living room and kitchen
were sampled in March 2014. The second and third were university
classroomseone classroom without computers (seminar room - SR)
and one with computers (computer room - CR), sampled in July
2014. The seminar and computer rooms were the same size and
shape and located in the same university building, differing only by
building storey and room equipment. The classrooms are normally
vacuumed 3 times/week and desks are wiped 4e5 times/week, but
sampling took place after the conclusion of the academic year and
there was no regular activity in the lecture room in the two weeks
prior to sampling. Both locations were not vacuumed/dusted for
two weeks prior to sampling.
In the flat, eight vacuum samples were collected from floors,
which included the entrance area, living, dining room, kitchen, and
from the sofa. The flat was not cleaned in the week prior to sample
collection. Ten wipe samples were collected from horizontal
surfaces, including electronics, a chair, and kitchen furnishings, and
one vertical surface (i.e. windows). Samples were also collected
from areas with high dust accumulation (F4, F6, and F14 in Fig. 1),
i.e., places with infrequent cleaning, such as under a sofa and on an
inaccessible window sill. In the classrooms, four vacuum samples
were collected from the carpet and five wipe samples were
collected from desks in both SR and CR. Additionally, six wipe
samples were collected from monitors (all-in-one PCs) and keyboards
in CR. The sampling locations are shown in Fig. 1 and more
details are given in Table S1 in the Supplementary Data.
2.1.1. Sampling
Two dust sampling methods were used e wet wipe sampling for
settled dust on smooth elevated surfaces and a vacuum cleaner
with sock insert for floor dust and dust from the sofa.
The wet wipe samples were collected using laboratory kimwipes.
The kimwipes were pre-cleaned via soxhlet extraction in
dichloromethane (DCM) for 8 h. At the site, the kimwipes were
moistened by approximately 2e3 ml of propan-2-ol and were used
to wipe each target surface. Depending on surface area and amount
of dust, 1e5 wipes were used for each surface, and combined into
one sample per surface. The area of the sampled surface was
measured. Wipes were packed in aluminium foil and sealed in a
plastic bag. Wipe samples were transported to the laboratory in a
cooler box and stored at À18 C until processing.
Dust samples from floors and fabric surfaces (e.g., sofa) were
collected using a household vacuum cleaner with polyester sock
inserts. Socks were precleaned via soxhlet extraction in DCM for
8 h, and before sampling and between samples, the vacuum nozzle
and tube were cleaned with propan-2-ol. To collect each sample, a
polyester sock was inserted into the front of the vacuum tube and
held in place by the vacuum nozzle. Each target surface was vacuumed
and its area measured. The sock was removed from the
vacuum cleaner, sealed by a plastic cable tie, packed in aluminium
foil and sealed in a plastic bag. Socks were transported to the laboratory
in a cooler box and stored at À18 C until processing.
2.2. Sample extraction and clean-up
The wipes were extracted using automated warm Soxhlet
extraction in a Büchi B-811 automatic extractor, with DCM as the
extraction solvent. Before extraction, samples were spiked with
isotopically labelled (13
C) internal standards, including triphenyl
phosphate, BDE 28, 47, 99, 100, 153, 154, 183 and 209, HBB, PBBZ,
syn- and anti-DDC-CO, BTBPE, and DBDPE (all from Wellington
Laboratories, Inc., Guelph, Ontario, Canada, except DDC-CO from
Cambridge Isotope Laboratories, Inc., Tewksbury, MA, USA). The
soxhlet extraction was cycled for 40 min, followed by 20 min where
solvent was diverted to concentrate the extract. The extract was
then concentrated to less than 10 ml and quantitatively transferred
to a vial. Then the extract was split 3:7 by weight (30% and 70%
aliquots) and each fraction was concentrated to a volume of 1e2 ml
by nitrogen flow.
Vacuum-collected dust samples were extracted via sonication.
Before extraction, samples were weighed and sieved with a 500 mm
sieve (Newark Wire Cloth Company, USA) to remove coarse particles
(e.g., hair, large fibres). Approximately 100 mg of the sieved
dust was used for extraction. The sock was rinsed with 20 ml of 1:1
hexane:acetone (v/v) and this volume was added to the accurately
weighed dust sample and the mass difference between unrinsed
sock and rinsed sock was included to the mass of extracted dust.
Before sonication, the sample was spiked with the same set of internal
standards as the wipe samples. The sample and 20 ml of
hexane:acetone were sonicated for 10 min, then the sample was
allowed to settle for 20 min and the solvent supernatant was
S. Jílkova et al. / Chemosphere 206 (2018) 132e141 133
transferred to a vial. An additional 10 ml of solvent was added to the
sample and it was sonicated again, allowed to settle and transferred
to the same vial. This step was repeated once more. The extracts
were split 3:7 (30% and 70% aliquots) by weight and each part was
concentrated to a volume of 1e2 ml by nitrogen flow.
The 70% aliquot of the extract was cleaned on a sulfuric acid
modified silica column (1 cm activated silica, 5 g silica modified by
sulfuric acid and 1 cm cleaned silica), eluted with 30 ml of 1:1
hexane:DCM (v/v). The 30% fraction was cleaned on a column of 5 g
activated silica and 1 cm sodium sulphate. This aliquot was further
fractionated by elution with 20 ml DCM (Fraction 1) followed by
20 ml of 7:3 acetone:DCM (v/v) (Fraction 2). Six of the Fraction 1
samples and seven of the 70% aliquots required additional clean up
(see Table S2).
After the column chromatography, 40 ml nonane was added to
the samples, and they were further concentrated and quantitatively
transferred to gas chromatography vials, and recovery standards
(BDEs 77 and 138) were added. The 70% aliquot was used for PBDE
analysis, the 30% F1 fraction for NFR analysis, and the 30% F2
fraction for OPE analysis.
2.3. Instrumental analysis
Samples were analysed for 10 PBDEs: BDE 28, BDE 47, BDE 66,
BDE 85, BDE 99, BDE 100, BDE 153, BDE 154, BDE 183 and BDE 209;
19 NFRs: TBP-AE, TBP-BAE, TBX, DDCeCOeMA, PBEB, PBT, TBPDBPE,
HBB, DBHCTD, EH-TBB, BTBPE, syn- and anti-DDC-CO, BEHTEBP,
DBDPE, a-, and b-, g-, and d-DBE-DBCH, a- and b-TBCO, PBBZ,
TBCT, PBB-Acr; and 17 OPEs: TIBP, TNBP, TCEP, TCIPP, DBPP, BDPP,
TDCIPP, TPHP, EHDPP, TBOEP, CDP, TEHP, o-, m- and p-TMPP, TIPPP,
TDMPP, TDBPP, TBPP. The full compound names and CAS numbers
are given in Table S3.
Details of the instrumental analysis can be found in the
Supplementary Material.
Fig. 1. Sampling locations in flat (F1eF18), seminar room (SR1eSR9) and computer room (CR1eCR14).
S. Jílkova et al. / Chemosphere 206 (2018) 132e141134
2.4. Quality assurance/quality control (QA/QC)
Average recoveries for spike-recovery tests are reported in
Table S4 and sample recoveries in Table S5. Field blanks (kimwipes
and vacuum bags) were transported to the field site, manipulated
as per the samples (e.g., wipes were exposed to ambient air but not
directly to sampled materials, polyester sock inserts were inserted
to the vacuum nozzle, but without the vacuum cleaner being on),
and then treated as per the samples for the analytical process. Six
vacuum sock field blanks and six kimwipe field blanks were
collected, three of each from the flat and the classrooms. Solvent
blanks were also analysed: six from Büchi extraction and six from
ultrasonic extraction. Analysis of the blanks demonstrated that the
majority of blank contamination was coming from the field blanks,
so field blanks were used to determine method detection limits
(MDLs). Solvent and field blanks are reported in Table S6. MDLs
were calculated based on the average of the field blank plus three
times the standard deviation of the blanks (Table S7). For compounds
that were not detected in the blanks, the instrument
detection limit was used as the MDL. Values < MDL were recorded
as such, and values > MDL were subtracted based on the average of
the blank of the corresponding matrix. Mass fraction was calculated
using sample mass (nanograms of analyte divided by grams of
sample mass) and area density was calculated using sample area
(nanograms of analyte divided by square meters of area). For the
purposes of statistical analyses, values below detection were
replaced by √2/2*MDL (Antweiler, 2015). Compounds with low
detection (generally <60% detection frequency in a given matrix/
room) were not used for statistical analysis.
Statistical analyses were performed in Statistica 12, MS Office
Excel 2010 and GraphPad Prism (Version 5).
3. Results and discussion
3.1. Detection and general composition of FRs
Eleven compounds: TBX, DDCeCOeMA, TBCO, TBCT, PBB-Acr,
DBPP, BDPP, o-TMPP, TDMPP, TDBPP, and TBPP, were not detected,
or infrequently detected at low levels with no spatial patterns and
will not be discussed further. In the following text we focus on 10
PBDEs, 14 NFRs and 12 OPEs which were broadly detected
(Table S3). Hereafter,
P
DDC-CO refers to the sum of the syn- and
anti-isomers,
P
DBE-DBCH refers to the sum of the a-, b-, g-, and disomers
and
P
TMPP to the sum of m- and p-TMPP.
The FRs in all samples were dominated by OPEs, which were on
average 96% of all FRs, but a number of individual NFRs and BDE 209
were also significant contributors (Fig. 2; for higher resolution of
NFRs and PBDEs see Fig. S1). TCIPP was detected in every sample; it
was the highest contributor in the flat samples (74% of
P
12OPEs)
and, along with TBOEP, was one of the two highest contributors in
classrooms. TCIPP is frequently used in polyurethane foams, as well
as textiles and plastics, while TBOEP is used in floor finish products
(Brandsma et al., 2014). BDE-209 was the highest PBDE, contributing
78%, 91% and 97% to
P
10PBDEs in the flat, seminar room and
computer room, respectively. For the NFRs, BEH-TEBP and DDC-CO
were predominant in most samples, however, more differences
between rooms/samples were observed, e.g., contributions from
HBB to
P
14NFRs ranged from <1% in the flat and seminar room to
8% in the computer room, while
P
DBE-DBCH contributed 20% to
P
14NFRs in the flat, and less than 1% in SR and CR. BEH-TEBP and
DDC-CO are widely used FRs in polymer materials such as polyurethane
foam, wire and cable insulation, PVC, polypropylene, ABS,
and epoxy resins (Covaci et al., 2011; Xian et al., 2011). DBE-DBCH is
used in home insulation, wire coatings, fabrics and appliances; and
HBB is used largely on the Asia market in textiles, electronics, and
textiles (Covaci et al., 2011). These differences and possible explanations
will be discussed further in the following sections.
Measured mass fractions and area densities of all FRs in individual
samples are given in Tables S8-S9. Summary statistics are
given in Table 1.
3.2. Differences according to sampling method
The two sampling methods (wet wipes and vacuumed dust)
gave a wide range of mass fractions and area densities. As dust mass
fractions could not be obtained from the wet wipe samples, the
differences according to sampling method can only be compared in
terms of area density and percentage composition profiles.
3.2.1. Comparison of area density
The area densities of FRs were highly dependent on the dust
surface loading, i.e., the mass of dust per unit area. The highest area
density (ng/m2
) of FRs was observed in wipe samples from areas of
long-term dust accumulation in the flat. This category represents
areas of less frequent cleaning (e.g., top of kitchen cabinets, inaccessible
windowsill) where dust is more likely to accumulate for a
longer time period, and the dust surface loading is higher. The area
densities in these areas were an order of magnitude higher than the
area densities from surfaces which were used/disturbed/cleaned on
a daily basis, and they are likely not representative of typical indoor
human exposure. Moreover, compounds that were below detection
limits in other samples can often be detected in these types of
samples; for example, PBEB was detected only in this type of
sample from the flat, so sampling these types of surfaces could be
good strategy to obtain detectable levels of the broadest set of FRs,
if an objective is merely to identify their presence indoors, e.g.,
when screening for what FRs are in use.
Window wipe samples had lower area densities (~pg/m2
) than
other samples (~ng/m2
) from the flat. The windows are a vertical
surface with low dust surface loading, as opposed to the other
sampled surfaces, which were largely horizontal, suggesting that
differences in gravitational settling of dust on surfaces strongly
influences the measured levels. Vertical surfaces will have little or
no contribution from settled dust, and will mostly reflect levels due
to partitioning or impaction. Also, analytes on windows may be
more susceptible to photodegradation.
The influence of the dust surface loading can be most easily
identified from the floor dust samples, as for these samples we
could determine the mass of dust per area of floor (Fig. S2). The
floor below the sofa (location F14) was the dustiest place in the flat,
with 3.98 g dust/m2
while the area below the TV table (location
F11) had the lowest dust loading (0.103 g dust/m2
), suggesting more
frequent use and cleaning. The area density of FRs was much higher
at F14 than F11:
P
10PBDEs was 405 ng/m2
at F14 vs 33.2 ng/m2
at
F11,
P
14NFRs was 138 ng/m2
at F14 vs. 33.4 ng/m2
at F11, and
P
12OPEs was 319 ng/m2
at F14 vs 7770 ng/m2
at F11. Moreover, the
area density was strongly correlated with the dust surface loading
(Spearman R's of 0.81e0.98, Table S10), with the exception of
compounds which we hypothesize to have direct nearby sources,
such as BEH-TEBP in sofa upholstery. No such correlation was
observed for mass fraction (ng/g). The lack of relationship with
mass fraction contrasts with the results of Al-Omran and Harrad
(2018), who identified dilution of PBDEs and BEH-TEBP in dust in
areas of high dust loading. The absence of a dilution effect in the flat
suggests that, except in cases of directly adjacent sources (e.g, BEHTEBP
from sofa), there are sufficient FR sources and sufficient
equilibration of the system to load all dust with similar levels of FRs
and therefore dust loading is the main control on area density of
FRs in dust.
In the classrooms, the locations at the fronts of the rooms (SR6
S. Jílkova et al. / Chemosphere 206 (2018) 132e141 135
and CR14) had the lowest dust loadings, likely reflecting less use of
these areas of the room. However, overall the dust surface loading
in the classrooms was much more homogeneous than in the flat
(0.062e0.209 g/m2
in classrooms vs. 0.103e3.98 g/m2
in the flat)
due to the greater diversity of activities in the flat, and less systematic
cleaning in homes than in institutional buildings.
Differences in dust loading are not frequently considered when
reporting and interpreting indoor dust, as dust is often reported in
mass fraction (e.g., ng/g), however, there is a question of how
appropriate mass fraction units are to human exposure estimates.
Dust ingestion and dermal transfer from dust are highly uncertain
(Abdallah et al., 2015; Lioy et al., 2002; Wilson et al., 2013), and
some exposure pathways, e.g., particularly dermal exposure, may
be more appropriately evaluated using area density (Lioy et al.,
2002). Moreover, infant and toddler exposure to compounds
through mouthing of objects and ingestion of dust may be significantly
higher if the dust surface loading is higher. Thus, dust
loading on surfaces should be an important consideration in sampling,
both for detection of compounds and evaluations of
exposure.
3.2.2. Comparison of profiles
In the flat, vacuumed dust and wipe samples yielded similar
profiles of FRs (Fig. 2). TCIPP was the primary contributor in most
dust and wipe samples, comprising 11.0e96.0% of the total OPEs,
and CDP was also a significant contributor in both dust and wipes
(0.1e51.9% of total OPEs). However
P
TMPP had a higher content in
wipes than in floor dust samples (on average 2.7% in wipes vs. 0.2%
in dust), suggesting either differences in the surfaces sampled by
wipes vs. by vacuum or differences in the ability of vacuum vs.
wipes to collect dust (e.g., collecting slightly different fractions of
dust).
In the classrooms, the FR profiles had more distinct differences
between vacuumed dusts and wipes. Most notably, TBOEP was the
major contributor to floor dust samples, while in the wipe samples
TCIPP was the highest contributor. The area densities of TBOEP
(Fig. 3b) also reflect these differences. In the classrooms, the
distinct differences in the profiles are more likely due to differences
in the sampled material rather than the sampling technique, and
suggest a source of TBOEP in the carpeted floor, or the trapping of
specific materials in the carpet that are not trapped by the elevated
surfaces, e.g., the carpet captures differently sized particles than
other surfaces.
3.3. Differences between rooms
Average sample compositions by room were compared to
identify the influence of different furnishings and equipment
Fig. 2. Percentage composition of FRs in (a) flat and (b) classroom floor dust and (c) flat and (d) classroom wipes. Samples are labelled according to the sampling locations in Fig. 1.
For dust samples, compounds with mass fraction >500 ng/g in at least one sample are shown individually, and
P
7OPEs,
P
12NFRs, and
P
9PBDEs include all remaining compounds.
For wipe samples, compounds with area density >100 ng/m2
(or >100 ng/sample for electronics) in at least one sample are shown individually, and
P
4OPEs,
P
12NFRs, and
P
9PBDEs include all remaining compounds. For better resolution of HFRs see Fig. S1.
S. Jílkova et al. / Chemosphere 206 (2018) 132e141136
within the same building, and differences in buildings and room
types within the same region. The floor dust samples were
compared by mass fraction (ng/g) and wipe samples were
compared by area density (ng/m2
).
3.3.1. Seminar vs computer room
The seminar room (SR) and computer room (CR) are identical in
size, shape, location, and building materials, differing only by one
floor, and by the equipment in the rooms. Thus any significant
differences between the two rooms can be attributed to room
equipment, and most likely, to the presence of computers in CR
compared with SR.
Six FRs had differences in detection frequencies between CR and
SR, and of these HBB, EH-TBB, TBP-BAE, TBP-AE, and DBDPE had
higher detection frequencies in CR than SR. Only TNBP was detected
in 4 of 5 samples in SR and only in 1 of 4 in CR. The most notable
difference between the two rooms was for HBB, which was
detected in all floor dust and wipe samples in CR, but only in two
wipe samples in SR and in none of the floor dust samples. Previous
air sampling suggested computer operation as a source of HBB
emissions in this room (Vojta et al., 2017).
Considering compounds with consistent detection in both
rooms, we compared dust mass fractions for floor dust samples and
area densities for wipe samples from SR and CR with a MannWhitney
U test (Tables S11 and S12). BDE 209, BTBPE and PBBZ
had significantly higher mean area densities in CR than in SR, while
TCIPP had significantly higher mean floor dust mass fraction in CR
than in SR and TEHP higher mean in SR. The combination of
detection frequencies and U-tests strongly suggest a source of HBB
in CR that is not present in SR, supporting previous air sampling
(Vojta et al., 2017), and possible differences in sources of TBP-AE,
EH-TBB, BTBPE, DBDPE, TNBP, and TCIPP.
3.3.2. Classrooms vs flat
Greater differences between the flat and classrooms were expected,
given the differences in type and age of building materials,
furniture (desks and computers in the university building and
kitchen equipment, sofa, and TV in the flat), and flooring (carpet in
classrooms, parquet and small rug in flat). Firstly, the suite of
detected FRs differed between the two sites. Five compounds were
detected in the classrooms but not in the flat: PBT, EHDPP, p-TMPP,
and TIPPP.
P
DDC-CO was the only compound that was detected in
the flat but not in the classrooms.
For compounds with consistent detection in both flat and
classrooms, we again compared dust mass fractions for floor dust
and area densities for wipe samples with a Mann-Whitney U test
(Table S13 and S14). The mean mass fractions of BDE 209, TCIPP,
DBDPE, and TBOEP were higher in the classrooms. Conversely,
Table 1
Area density and mass fraction in flat, seminar and computer room. The shaded rows (floors) refer to the same samples, but with different metrics (mass fraction and area
density). Red numbers indicate samples with the relative standard deviation > 100%, indicating a large range in concentrations within the same room.
S. Jílkova et al. / Chemosphere 206 (2018) 132e141 137
BTBPE,
P
DBE-DBCH, TIBP and TNBP were higher in the flat. In
wipes, 12 compounds were significantly higher in the flat
(Table S14).
Fig. 3 shows room comparisons for HBB, TBOEP, TCIPP, and
DBDPE based on area densities (ng/m2
) of all samples. TBOEP and
DBDPE had on average higher levels in the classrooms than the flat,
but it is apparent from Fig. 3 that this was only due to the elevated
levels in floor dust, not on surfaces. This suggests the carpet as a
likely source of TBOEP and possibly DBDPE. Interestingly, TBOEP is
mostly used in floor polish (Brandsma et al., 2014; Cristale et al.,
2016; van der Veen and de Boer, 2012), and these rooms were
fully carpeted, however it is also used in rigid and flexible polyurethane
foam (Cristale et al., 2016; van der Veen and de Boer,
2012), and in plastics and rubbers (van der Veen and de Boer,
2012). DBDPE has similar usage to BDE 209: in textiles, polypropylene
and other polymeric materials (Covaci et al., 2011). Thus,
the high levels of these compounds on classroom floors may be due
to carpet materials or carpet backing, and with releases exacerbated
abrasion of the carpet material during use and vacuuming.
However, DBDPE also had higher area density in the flat in surface
wipe samples than in floor dust from the flat or surface wipes from
the classrooms. While TCIPP had a higher mass fraction (ng/g) in
the classrooms, the higher area density (ng/m2
) of TCIPP was in the
flat (Fig. 3c), and the elevated levels were particularly notable in the
wipe samples. This difference suggests differences in the fractions
of TCIPP on individual particles, e.g., small particles with high levels
of TCIPP in the classrooms, such as from building-specific abrasive
sources (floor or wall materials, insulation, thermoplastics and
polyurethane foam) (Kemmlein et al., 2003; Liesewitz, 2000).
3.4. Spatial variability within rooms
To further address the objective of the study, we examined the
small-scale spatial differences in the rooms, and based on these,
identified how dust sampling can be adjusted to either capture or
minimize these differences, as befits the study objectives.
Area densities and mass fractions in the rooms typically ranged
over one order of magnitude in different samples from the same
room. The magnitude of small-scale spatial variation is shown for
selected FRs through room maps of the flat and CR (Fig. 5) according
to area density, mass fraction, and percentage composition. More
details are given in Table 1 and in Tables S8 and S9. Much of the
heterogeneity in the area densities and sample composition does
not appear to be directly linked to obvious sources, such as 5Â
higher area densities of
P
DDC-CO in one side of the computer
room compared with the other (CR12 vs. CR13), or 3e5-fold
Fig. 3. Area density of (a) HBB, (b) TBOEP, (c) TCIPP and (d) DBDPE in flat and classroom floor dust and wipe samples. Horizontal lines show mean and whiskers show standard
deviation.
S. Jílkova et al. / Chemosphere 206 (2018) 132e141138
variation in individual PBDE congeners, even amongst the hard
floor samples from the flat.
However, in some cases, we can relate the spatial differences to
sources like electronics or foam in the sofa according to the spatial
distribution in the room. In the flat, the contribution of BDE 209 to
P
10PBDEs decreased with increasing distance from the TV by 13%
(Fig. 4a), which supported the hypothesis of electronics as a source
of BDE 209. BEH-TEBP had a similar profile to BDE 209, with an even
larger decrease; from 74% to 5% of
P
14NFRs (Fig. 4b), except in the
kitchen, where the contribution was 70%, which could indicate
another source in this area. BEH-TEBP is often used in combination
with EH-TBB in Firemaster 550 as an FR in foams with a ratio of 1:4
BEH-TEBP:EH-TBB, but also independently in applications such as
cable insulation (Covaci et al., 2011). In the flat, the area density of
BEH-TEBP was one order of magnitude higher than EH-TBB, suggesting
independent usage in electronics rather than as Firemaster
550 in foams, which agrees with the dependence on distance from
the electronics. However, the trends in BDE 209 and BEH-TEBP
were only observable in the floor dust; wipe samples did not
show any pattern. This could be due to additional confounding
factors that challenge the interpretation of wipe samples, such as
debromination of selected FRs via photodegradation on the windows,
or higher temperatures on electronics surfaces selectively
degrading certain compounds. Moreover, all floor areas had lower
levels of BEH-TEBP (whether in ng/m2
and ng/g) than the sofa
surface (Fig. 5aec), which could lead to underestimation of exposure
via dust if only floor dust is measured. The decreasing percentage
contribution of BEH-TEBP to
P
14NFRs with increasing
distance from the TV shown in Fig. 4 is also visible in Fig. 5c.
However, the sofa (F13), armchairs (F5), carpet (F12) and floor
below sofa (F14) dust have high contributions of BEH-TEBP, which
suggests foam in upholstered furniture as another source.
TCIPP and also HBB displayed strong spatial gradients that may
be related to source proximity (Fig. 5). For TCIPP, areas of long term
dust accumulation (F4, F6) had the highest area densities (Fig. 5d),
as seen for BEH-TEBP. Also similarly to BEH-TEBP, the sofa (F13) and
floor below sofa (F14) had high area densities, but in contrast to
BEH-TEBP, the TV and DVD (F1 and F2) had low area densities of
TCIPP. This suggests spatial distributions influenced by the use of
TCIPP in foams (Stapleton et al., 2009), but not in electronics. TCIPP
on the kitchen floor (F18) had a high mass fraction but low area
density compared with other locations. The dust loading is low on
the kitchen floor, but the small amount of particles that are there
have a high mass of TCIPP. This contrasts with location F14 (below
sofa), where the comparison of area density and mass fraction
suggests there is a high particle loading but with lower amounts of
TCIPP on the particles. This demonstrates how the combined effects
of room use/cleaning patterns and proximity to sources complicate
interpretation of indoor dust levels.
HBB in the computer room had the highest values on the floor in
the corner of the room (CR14) in all units (area densityeng/m2
;
mass fractioneng/g and percentage content - %) (Fig. 5gei). This
could be due to more limited tracking of particles from outside to
the front of the room (like soil from shoes), and thus higher
contribution of within-room particle sources with elevated levels of
HBB. The comparison between SR and CR (section 3.3.1) identified
consistently higher levels of HBB in CR than in SR, suggesting
computers as a source, however from the room maps the source is
not as clear. The keyboards and monitors (CR1eCR6) had a wide
range of percentage content of HBB (Fig. 5i), and thus this does not
clearly indicate either the PCs or keyboards as HBB sources. We also
expected greater homogeneity in floor dust in terms of both ng/m2
and ng/g, as the carpet is uniform throughout the room; however,
nearby samples, e.g., CR11eCR13, had more than 2Â differences in
levels. This might be caused by room ventilation patterns (windows
are opened only on left side of room), but no reason is clear.
4. Conclusions and recommendations
To evaluate small scale spatial differences in indoor dust, two
types of indoor environments were evaluated: a residential flat and
two classrooms (one with and one without computers); and two
types of sampling methods were chosenevacuum and wipe samples.
Distinct differences between the flat and classrooms were
observed for 12 compounds: BDE 209,
P
DDC-CO,
P
DBE-DBCH,
DBDPE, TIBP, TNBP, TCIPP, PBT, EHDPP, TBOEP, p-TMPP and TIPPP.
Comparison of seminar room and computer room levels suggested
a computer-related source of HBB. Individual rooms demonstrated
significant within room variability in FR levels in dust, both in mass
fractions (ng/g) and area densities (ng/m2
). Levels typically ranged
over one order of magnitude within a room, with larger ranges for
the OPEs than PBDEs. Spatial variability in the flat was expected due
to the heterogeneity in furniture and room usage, and cleaning
frequencies. Spatial patterns suggested the TV, sofa and upholstered
chairs, and possibly kitchen appliances as possible sources of
FRs. Smaller spatial differences were expected in classroom samples,
as furnishing and room use are more homogenous, but we
observed differences between nearby locations (same carpet, same
sampled method) with no clear explanation.
When there are very strong differences in sources between
rooms (e.g., HBB between computer and seminar room) then the
difference can be observed regardless of where the samples are
taken, but more subtle differences are overwhelmed by selection of
sampling location. For example, within the flat, if single samples
were collected to represent the whole room (as is typical practice
when comparing levels across many buildings), significant bias
could be introduced in the reported values. The mean level of BEHTEBP
in the flat is 195 ng/g, but if we collected a single sample from
Fig. 4. Spatial variability of (a) BDE 209 as a fraction of
P
10PBDEs and (b) BEH-TEBP as a fraction of
P
NFRs in floor dust relative to distance from a potential indoor source.
S. Jílkova et al. / Chemosphere 206 (2018) 132e141 139
an area in the dining room (e.g., F17) we would report only 7 ng/g;
while based on the area around the TV table (e.g., F11) we would
measure 418 ng/g, e.g., 60-fold differences in reported levels based
only on location of floor sampling within the room. If extrapolated
to estimate human exposure to dust in the home, this would thus
result in 60-fold differences in exposure to BEH-TEBP simply due to
the location of sample collection within the home.
Moreover, this only considers the difference between floor
samples; clear differences also exist between surfaces and floors
(e.g., TBOEP in classroom desks vs. floor), which could lead to overor
underestimation of exposure due to usage of the location (i.e.
children play more on carpet, but adults have more contact with
raised surfaces, e.g., desks). Additionally, we observed that the
range of surface dust loading influences levels; as a result, area
density and mass fraction do not correlate in many cases, and thus
careful consideration should be given to selection of the dust metric
most appropriate to the study goals.
The strategy of the sampling depends on the goal of the
research. If simply detecting a wide range FRs is a goal, it is ideal to
sample in areas with high dust loadings, such as areas of long term
dust accumulation. However, this strategy would be inappropriate
for exposure assessment; rather, dust exposure is best assessed
with a focus on most frequently used areas of an indoor environment,
and this should be given careful consideration because there
is significant spatial heterogeneity, even in seemingly homogeneous
indoor spaces. When single samples are collected, composite
samples are strongly recommended to limit the influence of spatial
heterogeneity. Dust is an accessible and useful matrix to evaluate
indoor concentrations of compounds of concern, and a necessary
component of exposure assessment, however, to obtain coherent
and comparable data it is necessary to make appropriate choices in
sampling method, sampled area and metric for sample
interpretation.
Fig. 5. Maps of area densities, mass fraction and percentage content of a-c) BEH-TEBP and d-f) TCIPP in flat and g-i) HBB in CR.
S. Jílkova et al. / Chemosphere 206 (2018) 132e141140
Acknowledgements
This research was supported by the RECETOX Research Infrastructure
(LM2015051 and CZ.02.1.01/0.0/0.0/16_013/0001761). We
thank Roman Prokes for assistance in sample collection.
Appendix A. Supplementary data
Supplementary data related to this article can be found at
https://doi.org/10.1016/j.chemosphere.2018.04.146.
References
Abdallah, M.A.-E., Pawar, G., Harrad, S.J., 2015. Evaluation of 3D-human skin
equivalents for assessment of human dermal absorption of some brominated
flame retardants. Environ. Int. 84, 64e70. https://doi.org/10.1016/
j.envint.2015.07.015.
Al-Omran, L.S., Harrad, S.J., 2018. Within-room and within-home spatial and temporal
variability in concentrations of legacy and “novel” brominated flame retardants
in indoor dust. Chemosphere 193, 1105e1112. https://doi.org/10.1016/j.
chemosphere.2017.02.096.
Ali, N., Harrad, S.J., Goosey, E.R., Neels, H., Covaci, A., 2011. “Novel” brominated flame
retardants in Belgian and UK indoor dust: implications for human exposure.
Chemosphere 83, 1360e1365. https://doi.org/10.1016/j.chemosphere.2011.02.
078.
Antweiler, R.C., 2015. Evaluation of statistical treatments of left-censored environmental
data using coincident uncensored data sets. II. Group comparisons.
Environ. Sci. Technol. 49, 13439e13446.
Bergman, A., Ryden, A., Law, R.J., de Boer, J., Covaci, A., Alaee, M., Birnbaum, L.S.,
Petreas, M.X., Rose, M., Sakai, S., Van den Eede, N., van der Veen, I., 2012. A novel
abbreviation standard for organobromine, organochlorine and organophosphorus
flame retardants and some characteristics of the chemicals. Environ. Bar
Int. 49, 57e82.
Betts, K.S., 2008. New flame retardants detected in indoor and outdoor environments.
Environ. News. https://doi.org/10.1021/es801070p.
Brandsma, S.H., de Boer, J., van Velzen, M.J.M., Leonards, P.E.G., 2014. Organophosphorus
flame retardants (PFRs) and plasticizers in house and car dust and
the influence of electronic equipment. Chemosphere 116, 3e9.
Cequier, E., Ionas, A.C., Covaci, A., Marce, R.M., Becher, G., Thomsen, C., 2014.
Occurrence of a broad range of legacy and emerging flame retardants in indoor
environments in Norway. Environ. Sci. Technol. 48, 6827e6835. https://doi.org/
10.1021/es500516u.
Covaci, A., Harrad, S.J., Abdallah, M.A.-E., Ali, N., Law, R.J., Herzke, D., de Wit, C.A.,
2011. Novel brominated flame retardants: a review of their analysis, environmental
fate and behaviour. Environ. Int. 37, 532e556.
Cristale, J., Hurtado, A., Gomez-Canela, C., Lacorte, S., 2016. Occurrence and sources
of brominated and organophosphorus flame retardants in dust from different
indoor environments in Barcelona. Spain. Environ. Res. 149, 66e76. https://doi.
org/10.1016/j.envres.2016.05.001.
de Wit, C.A., 2002. An overview of brominated flame retardants in the environment.
Chemosphere 46, 583e624.
Harrad, S.J., Abdallah, M.A.-E., Covaci, A., 2009. Causes of variability in concentrations
and diastereomer patterns of hexabromocyclododecanes in indoor dust.
Environ. Int. 35, 573e579. https://doi.org/10.1016/j.envint.2008.10.005.
Harrad, S.J., Ibarra, C., Abdallah, M.A.-E., Boon, R., Neels, H., Covaci, A., 2008a.
Concentrations of brominated flame retardants in dust from United Kingdom
cars, homes, and offices: causes of variability and implications for human
exposure. Environ. Int. 34, 1170e1175. https://doi.org/10.1016/j.envint.2008.05.
001.
Harrad, S.J., Ibarra, C., Diamond, M.L., Melymuk, L., Robson, M., Douwes, J.,
Roosens, L., Dirtu, A.C., Covaci, A., 2008b. Polybrominated diphenyl ethers in
domestic indoor dust from Canada, New Zealand, United Kingdom and United
States. Environ. Int. 34. https://doi.org/10.1016/j.envint.2007.08.008.
Karlsson, M., Julander, A., van Bavel, B., Hardell, L., 2007. Levels of brominated flame
retardants in blood in relation to levels in household air and dust. Environ. Int.
33, 62e69. https://doi.org/10.1016/j.envint.2006.06.025.
Kemmlein, S., Hahn, O., Jann, O., 2003. Emissions of organophosphate and brominated
flame retardants from selected consumer products and building materials.
Atmos. Environ. 37, 5485e5493. https://doi.org/10.1016/j.atmosenv.2003.
09.025.
Kuang, J., Ma, Y., Harrad, S.J., 2016. Concentrations of “legacy” and novel brominated
flame retardants in matched samples of UK kitchen and living room/bedroom
dust. Chemosphere 149, 224e230. https://doi.org/10.1016/j.chemosphere.2016.
01.092.
Liesewitz, A., 2000. Substituting Environmentally Relevant Flame Retardants:
Assessment Fundamentals.
Lioy, P.J., Freeman, N.C.G., Millette, J.R., 2002. Dust: a metric for use in residential
and building exposure assessment and source characterization. Environ. Health
Perspect. 110, 969e983.
Muenhor, D., Harrad, S.J., 2012. Within-room and within-building temporal and
spatial variations in concentrations of polybrominated diphenyl ethers (PBDEs)
in indoor dust. Environ. Int. 47, 23e27. https://doi.org/10.1016/j.envint.2012.06.
001.
Sj€odin, A., P€apke, O., McGahee, E., Focant, J.-F., Jones, R.S., Pless-Mulloli, T., Toms, L.M.L.,
Herrmann, T., Müller, J.F., Needham, L.L., Patterson, D.G., 2008. Concentration
of polybrominated diphenyl ethers (PBDEs) in household dust from
various countries. Chemosphere 73, S131eS136.
Stapleton, H.M., Allen, J.G., Kelly, S.M., Konstantinov, A., Klosterhaus, S.L.,
Watkins, D.J., McClean, M.D., Webster, T.F., 2008. Alternate and new brominated
flame retardants detected in U.S. house dust. Environ. Sci. Technol. 42,
6910e6916.
Stapleton, H.M., Klosterhaus, S.L., Eagle, S., Fuh, J., Meeker, J.D., Blum, A.,
Webster, T.F., 2009. Detection of organophosphate flame retardants in furniture
foam and U.S. house dust. Environ. Sci. Technol. 43, 7490e7495.
van der Veen, I., de Boer, J., 2012. Phosphorus flame retardants: properties, production,
environmental occurrence, toxicity and analysis. Chemosphere 88,
1119e1153. https://doi.org/10.1016/j.chemosphere.2012.03.067.
Venier, M., Audy, O., Vojta, S., Becanova, J., Romanak, K., Melymuk, L., Kratka, M.,
Kukucka, P., Okeme, J.O., Saini, A., Diamond, M.L., Klanova, J., 2016. Brominated
flame retardants in indoor environment - comparative study of indoor
contamination from three countries. Environ. Int. 94, 150e160. https://doi.org/
10.1016/j.envint.2016.04.029.
Vojta, S., Melymuk, L., Klanova, J., 2017. Changes in flame retardant and legacy
contaminant concentrations in indoor air during building construction,
furnishing, and use. Environ. Sci. Technol. 51, 11891e11899. https://doi.org/10.
1021/acs.est.7b03245.
Vykoukalova, M., Venier, M., Vojta, S., Melymuk, L., Becanova, J., Romanak, K.,
Prokes, R., Okeme, J.O.J.O., Saini, A., Diamond, M.L., Klanova, J., 2017. Organophosphate
esters flame retardants in the indoor environment. Environ. Int. 106,
97e104. https://doi.org/10.1016/j.envint.2017.05.020.
Wilson, R., Jones-Otazo, H.A., Petrovic, S., Mitchell, I., Bonvalot, Y., Williams, D.T.,
Richardson, G.M., 2013. Revisiting dust and soil ingestion rates based on handto-mouth
transfer. Hum. Ecol. Risk Assess. An Int. J. 19, 158e188. https://doi.org/
10.1080/10807039.2012.685807.
Xian, Q., Siddique, S., Li, T., Feng, Y.-L., Takser, L., Zhu, J., 2011. Sources and environmental
behavior of dechlorane plusea review. Environ. Int. 37, 1273e1284.
S. Jílkova et al. / Chemosphere 206 (2018) 132e141 141