Radiation damage in diopside and calcite crystals from uranothorianite inclusions
Anne-Magali Seydoux-Guillaume a,
, Jean-Marc Montel a
, Richard Wirth b
, Bernard Moine a
a
LMTG, CNRS, Université de Toulouse, IRD, OMP, 14 avenue Edouard Belin, 31400 Toulouse, France
b
GeoForschungsZentrum (GFZ) Potsdam-Division 4, Telegrafenberg, D-14473 Potsdam, Germany
a b s t r a c ta r t i c l e i n f o
Article history:
Accepted 12 April 2008
Keywords:
Pleochroic halo
Radiation damage
Uranothorianite
Calcite
Diopside
FIB/TEM
Alteration
Combining observation and simulation, radiohalos formed around uranothorianite (UTh) from the
Tranomaro granulitic skarns (SE-Madagascar) were studied. These structures consist of UTh grains
surrounded by both aluminous diopside (Cpx) and calcite (Cc1) crystals. Optical microscope and Scanning
Electron Microscope (SEM) images revealed (1) the presence of radiating cracks around the UTh probably due
to swelling of the metamict UTh, (2) a diffuse optical halo at the Cc1/UTh interface, and (3) a wide "reaction
zone" at the Cpx/UTh interface, composed of "secondary calcite" (Cc2) with low temperature sheet silicate
from the smectite () group. Samples prepared across various interfaces using Focused Ion Beam (FIB) were
investigated by Transmission Electron Microscope (TEM). In contrast to SEM observations, there is no direct
contact between Cc1 and UTh. From Cc1 to UTh, we found: (1) a large (~200­300 nm) amorphous zone (A),
enriched in U, Th and Ca, but without Si; (2) a chain (B) of very small (~20 nm) ThO2 crystals; (3) another
amorphous zone (C), which, in contrast to zone A is enriched in Si; and (4) another zone (D) made of small
amorphous Si-rich "bubbles". The organization is similar for the UTh­Cc2 interface. The presence of hydrous
minerals (smectite) and carbonate (calcite) in reaction zone and in cracks, the presence of Pb-rich inclusions
in secondary calcite, the abundance of fluid inclusions in the porous layer in calcite, the dissociation of U and
Th in the calcite­uranothorianite layer, and the ThO2 chains along interfaces, are strong indications that lowtemperature
crystallization was promoted by a fluid phase. SRIM simulation was used to calculate the effect
of  and recoil particles of the three decay chains, in Cpx, Cc and UTh. The thickness of the damaged area
calculated for  in Cpx and Cc are similar to the widths of the recrystallized areas observed in thin section
(~30 m). Corrected with the "wandering recoil effect", the size of the damaged area calculated for recoil
nuclei in Cc (~50­60 nm) is ~multiplied by 3 and is in rather good agreement with the thickness of the totally
amorphous layer at the Cc­UTh interface (~200 nm). Finally, it is emphasized that radiohalos are a point of
chemical and physical weakness in a rock and probably a starting point for alteration.
 2008 Elsevier B.V. All rights reserved.
1. Introduction
During geological time, U- and Th-rich minerals accumulate
radiation damage, mainly from  decay. Such damage destroys to a
variable extent the host crystal network, leading to an amorphous
structure, called metamict state (summary in Ewing, 1994). Damage is
caused by three types of particles. First,  particles, which are
energetic (4­8 MeV), penetrative, (10­20 m), but not very destructive
(about 100 displacement/particle); second, recoil nuclei (daughter
nuclei), which are less energetic (100­400 keV), less penetrative (20­
50 nm), but more destructive (800­2000 displacement/particle);
finally, fission nuclei, which are very energetic (150­200 MeV), very
destructive (fission tracks are 10­20 m long and 5­10 nm diameter,
and made of several ten thousands displacements), but very rare
(0.0005% of 238
U decays). Submitted to natural radiation damages, i.e.
accumulated over a long time, minerals react differently: many
minerals, like zircon (review in Ewing et al., 2003) or allanite (Headley
et al., 1981; Janeczek and Eby, 1993), become amorphous (metamict),
whereas some remain crystalline such as monazite (Meldrum et al.,
1998; Ewing and Wang, 2002; Seydoux-Guillaume et al., 2002; 2004;
reviews in Ewing et al., 2003) or apatite (Linberg and Ingram, 1964;
review in Ewing and Wang, 2002).
Radiation damage in radioactive minerals has been studied in the
geosciences for two main reasons. First, U­Th-rich minerals are used for
U­Th­Pb dating, and it is essential to understand the effects of radiation
damages on lead retentivity (Lumpkin et al., 1986a; Davis and Krogh,
2000; Romer, 2003). Second, the effect of long-term accumulation of
radiation damage is a key parameter for assessing the durability of
ceramics that could be used as nuclear-waste forms (Ewing, 1975;
Ewing et al., 1988; Ewing et al., 1995; Weber et al., 1998).
Four strategies are used to study radioactive damage in minerals (see
review in Ewing et al., 2000): (1) external irradiation by ion beams (e.g.
Wang and Ewing,1992; Meldrum et al.,1998); (2) atomistic simulations
(e.g. Crocombette and Ghaleb, 2001; Trachenko et al., 2001); (3)
synthesis of doped-crystal with short-lived isotopes (e.g. Begg et al.,
Chemical Geology 261 (2009) 318­332
 Corresponding author. Tel.: +33 5 61 33 25 97.
E-mail address: seydoux@lmtg.obs-mip.fr (A.-M. Seydoux-Guillaume).
0009-2541/$ ­ see front matter  2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.chemgeo.2008.04.013
Contents lists available at ScienceDirect
Chemical Geology
journal homepage: www.elsevier.com/locate/chemgeo
2000; Burakov et al., 2002); (4) study of old naturally radioactive
minerals in various geological contexts by using various analytical
methods like IR (Zhang and Salje, 2001), EXAFS (Farges and Calas,1991;
Harfouche et al., 2005), Raman (Nasdala et al.,1995), RMN (Farnan et al.,
2003), RPE (Balan et al., 2005), and TEM (Black et al., 1984; Lumpkin
et al.,1986a; Murakami et al.,1991; Weber et al.,1994). The most studied
minerals are those that are either used for U­Th­Pb geochronology or
expected to be good candidate for nuclear-waste storage, and many
papers have been published on zircon (review in Ewing et al., 2003),
monazite (review in Ewing and Wang, 2002; Seydoux-Guillaume et al.,
2004), titanite (Vance and Metson, 1985; Hawthorne et al., 1991;
Lumpkin et al., 1991; Farges, 1997, Zhang et al., 2002), pyrochlore and
zirconolite (Lumpkin et al., 1986b; Lumpkin and Ewing, 1988; Farges et
al., 1993; Farges, 1997), apatite (Ouchani et al. 1997), and thorite-group
(Lumpkin and Chakoumakos, 1988; Farges and Calas, 1991).
However, radioactive minerals not only irradiate themselves, but also
the surrounding non-radioactive minerals. This produces concentric
structures called "pleochroic halos" or "radiohalos", very familiar to
petrologists, who use them to identify radioactive minerals in
metamorphic or plutonic rocks. In contrast to the numerous studies
dealing with radiation effects in radioactive minerals themselves, the
effect of radiation damages in host minerals are rare. The first time that
such halo around mineral inclusions was interpreted as being due to the
radioactivity of these inclusions was 100 years ago (Mügge, 1907; Joly,
1907). After that, studies on radiohalos were published between the
seventies and the nineties (e.g. Gentry, 1973, 1974; Owen, 1988; Odom
and Rink,1989; Meunier et al.,1990). Recently, only two papers deal with
radiohalos: in biotite (Nasdala et al., 2001) and in chlorite and cordierite
(Nasdala et al., 2006). These latter studies demonstrated that radiohalos
are created by  particles and correspond only to modification of optical
characteristics of the host mineral, the various energies of the  particles
explaining the difference in size of the halos. Furthermore, those authors
found intensive damage (i.e. amorphous domains visible via TEM) only
in cordierite up to a few tens of nanometers away from radioactive
inclusions. They interpreted these damaged zones to recoil nuclei. Such
amorphous zones at the vicinity of a radioactive mineral has already
been observed by using Focused Ion Beam­Transmission Electron
Microcopy (FIB­TEM) across monazite­quartz boundaries by SeydouxGuillaume
et al. (2003); in this case, the zone was ~150 nm wide.
However, other consequences of radiation damage have not been
evaluated.Inrocks,poorlycrystallizedareas,suchasthemetamictgrainsbut
also the damaged zone around them, are actually zones, which can be used
bygeologicalfluidstoinitiatealteration.Anotherimportanteffectisswelling
produced by radiation damage, which can induce cracks around the
enclosing minerals, forming pathways for fluids to penetrate into the rock.
The aim of this study is to investigate, by various electron microscopy
techniques, the effect of irradiation in a radioactive mineral (uranothorianite)
and in the surrounding minerals (calcite and diopside). The
geological consequences of radioactive damage for the non-radioactive
host minerals and for the whole-rock itself will be discussed.
2. Sample description
The studied sample (SB540) is a diopside-bearing marble within
skarns from the Tranomaro area (Andranomitrohyopenpit at 46°33.15 E,
24°19.97 S) in South-East Madagascar, metamorphosed under granulitic
Fig. 1. Optical microscope images showing 3 uranothorianite (UTh) grains included in diopside (Cpx)+calcite (Cc) under crossed polar light. Note the associated structures due to
irradiation from  decay of U and Th: (1) cracks, visible only in diopside, radiating from uranothorianite; (2) a diffuse optical halo at calcite­uranothorianite interface; and (3) a wide
(~20­30 m) "reaction zone" along diopside­uranothorianite interface.
Table 1
Compositions (in wt.%) of uranothorianite (UTh), diopside (Cpx), calcite (Cc) and sheet silicate () from Fig. 2A obtained by EMP.
# (Fig. 2A) Uranothorianite (UTh) Sheet silicate () Diopside (Cpx) Calcite (Cc)
1 2 12 13 14 3 4 5 6 7 8 9 10 11
Na2O 0.00 0.00 0.08 0.00 0.00 0.07 0.00 0.00 0.00 0.12 0.22 0.21 0.29 0.00
MgO 0.00 0.00 0.00 0.01 0.00 10.56 14.02 15.50 13.75 13.79 13.72 14.05 13.99 0.66
Al2O3 0.00 0.00 0.00 0.00 0.00 19.05 18.42 18.75 19.06 8.22 8.19 8.38 8.35 0.00
SiO2 0.00 0.39 0.00 0.01 0.00 43.58 39.52 33.68 43.81 48.73 48.70 49.27 48.67 0.04
K2O 0.10 0.03 0.05 0.04 0.11 0.21 0.10 0.03 0.20 0.00 0.02 0.00 0.00 0.00
CaO 0.19 0.25 0.16 0.11 0.16 2.26 1.87 1.94 2.34 25.03 25.38 25.00 25.07 55.95
TiO2 0.00 0.00 0.00 0.00 0.00 0.25 0.18 0.09 0.24 0.74 0.72 0.83 0.68 0.00
Cr2O3 0.04 0.00 0.07 0.00 0.01 0.00 0.04 0.02 0.00 0.00 0.00 0.00 0.02 0.02
MnO 0.00 0.07 0.07 0.05 0.02 0.13 0.01 0.23 0.10 0.11 0.05 0.00 0.04 0.02
FeO 0.12 0.00 0.00 0.01 0.01 6.28 7.55 14.45 7.25 2.36 2.26 2.23 2.50 0.19
PbO 3.77 3.74 4.10 4.33 3.92 0.06 0.17 0.14 0.00 0.25 0.00 0.00 0.05 0.18
ThO2 66.37 64.80 66.33 66.12 66.99 0.04 0.68 0.10 0.00 0.03 0.00 0.06 0.08 0.00
UO2 29.24 28.50 30.27 29.76 28.92 0.10 0.00 0.00 0.00 0.00 0.06 0.00 0.01 0.00
Total 99.83 97.78 101.14 100.44 100.15 82.57 82.56 84.93 86.75 99.38 99.31 100.03 99.75 57.06
EMP operating at 15 kV, 20 nA. Standards are: albite (NaK), periclase (MgK), corundum (AlK), wollastonite (Si and CaK), sanidine (KK), perovskite (Ti and MnK), Cr2O3
(CrK), hematite (FeK), synthetic Pb-glass (PbM), synthetic ThO2-ceramic (ThM) and synthetic UO2-ceramic (UM).
319A.-M. Seydoux-Guillaume et al. / Chemical Geology 261 (2009) 318­332
conditions (4­5 kbar, 800­850 °C; Rakotondrazafy,1995; Rakotondrazafy
et al.,1996) during the Pan-African orogeny (565­580 Ma, Paquette et al.,
1994). In this region, at the western border of the Anosyan Belt,
uranothorianite mineralization is common (Moine et al., 1985; Boulvais
et al.,1998; Ramambazafy,1998; Ramambazafyet al.,1998; Boulvais et al.,
2000). They occur in skarns formed by metasomatic alteration of calcitic
marbles (Moine et al., 1985). Fluid inclusions study shows that fluids are
CO2-rich (XCO2 0.8) and in equilibrium with mineral assemblages
(Ramambazafy et al., 1998). The hydrothermal/metasomatic mobility
of Th can be explained by transport in F-rich fluids as shown by the
widespread occurrence of fluor-phlogopite and fluor-pargasite (Moine et
al., 1998; Ramambazafy, 1998).
Minerals are essentially aluminous diopside (Cpx) and calcite (Cc1)
sometimes containing uranothorianite inclusions; but spinel, pargasite,
plagioclase, phlogopite and zirconolite may also be present. Optical
microscope pictures (Fig. 1) show examples of 3 uranothorianite grains
included in diopside+calcite. The uranothorianite grains are systematically
associated with structures suggesting that radioactivity significantly
modified the host non-radioactive minerals. These include: (1)
cracks, visible only in diopside, radiating from uranothorianite; (2) a
Fig. 2. Scanning Electron Microscope (SEM) images inSecondaryElectron (SE) modefrom UTh+Cc1 +Cpx shown inFig.1 (leftone). A.Note theassociatedstructuresdueto irradiation from
 decay of U and Th: (1) cracks, visible only in diopside, radiating from uranothorianite; (2) a diffuse optical halo at calcite­uranothorianite interface; and (3) a wide (~20­30 m) "reaction
zone" along the diopside­uranothorianite interface. Within this reaction zone, 2 other phases were observed: a secondary calcite (Cc2) and a clay mineral (). 7 FIB holes and Electron
Microprobe analyses (see numbers) were also done. B. FIB hole across UTh­Cc1 boundary (# 750). Note the porosity within Cc1 and possible remnant of cracks (see arrow). C. Enlargement
within the"reaction zone" showing mixture of a secondarycalcite (Cc2) anda low-Tclay mineral (), and FIBholesdoneacross UTh- (#752)and-Cpx(# 753) boundaries. D.Enlargement
within the "reaction zone" showing fractures within diopside, with some extended into the reaction zone (arrow), and FIB hole cut across one crack (# 1092). E. Enlargement within the
"reaction zone" showing high porosity within Cc2.
320 A.-M. Seydoux-Guillaume et al. / Chemical Geology 261 (2009) 318­332
diffuse optical halo at the calcite­uranothorianite interface; and (3), a
wide (~20­30 m) "reaction zone" along the diopside­uranothorianite
interface. Only diopside in contact with uranothorianite show these
corona; other diopside crystals are pristine indicating these features are
due to irradiation coming from  decay of U and Th.
3. Analytical methods
3.1. Scanning Electron Microscope (SEM) and Electron Microprobe (EMP)
SEM images and EDX mapping were performed using the JEOL 6360
equipped with a Sahara detector from PGT at the LMTG-Toulouse.
Quantitative EMP analyses were obtained using the Cameca SX50 at
the LMTG-Toulouse operating at 15 kV and 20 nA (Table 1).
3.2. Transmission Electron Microscope (TEM) coupled with Focused Ion
Beam (FIB) technique
Since the aim of this study is to investigate the interfaces between
grains in thin section, a site-specific preparation method for TEM
analysis is needed. This method is called Focused Ion Beam (FIB) and
allows cutting site-specific TEM foils, ~15­20 m by 10­15 m, and
~100 nm thick (for technical details, see, Overwijk et al., 1993; Young,
1997; Roberts et al, 2001 and Wirth, 2004). TEM samples were milled
Fig. 3. SEM chemical maps of PbM, SiK, ThL, AlK, CaK, MgK and FeK of the zone observed in the Back Scattered Electron (BSE) image on the left. Acquisition time was
12.4 sec/image for a total of 140 images and a size of maps of 256×192×16 bits. Note of the high quantity of PbS inclusions within UTh.
321A.-M. Seydoux-Guillaume et al. / Chemical Geology 261 (2009) 318­332
by using gallium ions accelerated to 30 keV. The TEM foil is cut
perpendicular to the surface of the sample (Fig. 2), providing
information with respect to the depth of the specimen. The sitespecific
specimens were prepared with the FEI FIB200 instrument at
the GeoForschungsZentrum (GFZ)-Potsdam.
TEM studies were carried out with the Philips CM200 TEM,
operating at 200 keV, equipped with an energy-dispersive X-ray
analyzer (EDX) with an ultra-thin window, and a LaB6 filament as
electron source, and the FEI TecnaiTMG2
F20 X-Twin, operating at
200 kV, equipped with a FEG electron source, a high angle annular dark
field (HAADF) detector, and an EDAX energy-dispersive X-ray analyzer
system; both instruments are installed at the GFZ-Potsdam. Additional
selected area diffraction (SAED) patterns were also performed at the
TEMSCAN-Toulouse with a Jeol 2010 (200 keV, LaB6) microscope.
4. Results
4.1. Microscopic study
In agreement with previous studies (Rakotondratsima, 1983;
Ramambazafy, 1998), EMP analyses showed that uranothorianite is a
solid-solution made of about two-thirds thorianite and third
uraninite. The lead content is about 3­4 wt.% PbO. The SEM­EDX map
(Fig. 3) shows that uranothorianite contains many Pb-rich inclusions,
demonstrated by EMP and TEM to be galena PbS. The chemical ages
calculated from U, Th, and Pb content range from 544 to 578 Ma, for
four analyses, in agreement with the accepted ages for those rocks, but
one analysis gives 931 Ma, due to high Pb content. We think that this
analysis was contaminated by a galena inclusion, giving an apparently
high Pb content, and then an apparently old age. Clinopyroxene is an
Al-rich diopside (Xmg =0.92, and 8.3 wt.% Al2O3).
The "reaction zone" at diopside­uranothorianite interface (Fig. 2C)
consists of a mixture of calcite (secondary calcite, Cc2), with an AlMg­Fe­silicate
fibrous phase () (Fig. 2C, D and E). EMP coupled with
TEM results (Table 1) demonstrated that it is a low-T phyllosilicate
from the smectite group. Secondary calcite is usually located at the
inner part of the reaction zone, along uranothorianite, and is always
porous, whereas smectite () is located in the outer part along
diopside. The diopside-reaction zone grain boundary is locally
indented (Fig. 2A, C and D), along the main cleavage direction. The
primary calcite­uranothorianite interface is sharp, but calcite in
contact with uranothorianite is more or less porous on a ~20 m
layer (Fig. 2B and E), as is secondary calcite in the reaction zone (Fig.
2E).
SEM confirms the presence of cracks radiating around the
uranothorianite within clinopyroxene (Fig. 2A and D). Possible
remnants of cracks were observed within Cc1 (arrow Fig. 2B). Some
fractures extend into the reaction zone (Fig. 2A and D).
4.2. Nanometric study (FIB/TEM)
TEM samples were prepared using the FIB milling technique across
the uranothorianite­primary calcite (#750; Fig. 2B), uranothorianitesecondary
calcite (#752) and diopside-reaction zone (#753) interfaces
(Fig. 2C); one foil was also cut across a fracture (#1092, Fig. 2D). From
selected area diffraction (SAD) patterns (Fig. 4B and C), we can conclude
that both primary calcite and uranothorianite are crystalline. The only
indication of structural modification in those phases, is mottled
diffraction contrast in bright field and dark field images, observed
mostly in uranothorianite. In this mineral, diffraction spots are split
(Fig. 4C), probably because of PbS inclusions (Fig. 3--PbM map).
Uranothorianite­primary calcite interface (Figs. 4 and 5): in contrast
to what was observed with SEM, there is no direct contact between
primary calcite and uranothorianite. Four different zones (A, B, C
and D) were identified (Fig. 5). From calcite to uranothorianite, we
found: (1) a large (~200­300 nm) amorphous zone (A), enriched in
U and Th, but without Si (see U­Th­Si maps in Fig. 5) and enriched
in Ca (see EDX spectrum Fig. 5); (2) a chain (B) of very small
(~20 nm) ThO2 crystals free of U (Figs. 4 and 5); (3) another
amorphous zone (C), which, in contrast to zone A is enriched in Si;
and (4) another zone (D) made of small amorphous Si-rich
"bubbles". Many fluid inclusions were observed in the calcite,
some of them displaying a "negative crystal" shape (arrow Fig. 4A).
Diopside-reaction zone interface (Fig. 6): the FIB cut foil shows
preferential dissolution of Cpx along cleavages, giving this
particular triangular-shaped boundary (Fig. 6A). Within the
Fig. 4. TEM foil # 750. A. Energy Filtered Transmission Electron Microscope (EFTEM) -- Bright Field (BF) image showing presence of an amorphous zone (A) between primary calcite
(Cc1) and UTh. Zone (B) correspond toThO2 grains and zone (C) to another amorphous zone (see Fig. 5). Many fluid inclusions were observed in the calcite, some of them displaying a
"negative crystal" shape (see arrow). B and C. Selected area diffraction (SAD) patterns in respectively primary calcite and UTh showing that both phases are crystalline. In UTh
diffraction spots are split probably because of the presence of PbS inclusions within UTh (arrows in Fig. 3).
322 A.-M. Seydoux-Guillaume et al. / Chemical Geology 261 (2009) 318­332
reaction zone an unoriented clay mineral (smectite) was observed,
with its typical layer-form (Fig. 6B and C). Many inclusions are
present in this zone (arrows on Fig. 6): Pb-rich inclusions as well as
and U+Th-rich inclusions.
Uranothorianite­secondary calcite interface (Fig. 7): the organization
is similar, but opposite to the uranothorianite­primary calcite
interface. An amorphous zone (A) enriched in U, Th and Ca and a
chain of ThO2 crystals (B) parallel to the interface was observed.
However the ThO2 layer is mostly along calcite, and not along
uranothorianite, and no Si-rich zone was observed. Secondary
calcite contains many fluid inclusions (see arrows in Fig. 7A and C)
and U+Th+Pb-rich inclusions (large inclusion Fig. 7A).
Within one crack (Fig. 8): the fracture (~2 m wide) is filled with a
mixture of clay mineral and calcite similar to the reaction zone. It
contains also ZrTiCaTh-amorphous phase (Zr), presumably metamict
zirconolite, already described in those rocks (Rakotondrazafy
et al., 1996). Nano-channels are visible at the interface between
diopside and the crack border; the channel is larger (~70 nm
compared to ~40 nm) on the upper side (Fig. 8A and B). Bright field
(BF) image of the upper channel (Fig. 8B) shows that it is filled with
nano-bubbles, i.e. fluid inclusions.
4.3. Summary of petrographic features
The aureole along the diopside­uranothorianite interface is not a
pleochroic halo such as described by Nasdala et al. (2001) in biotite or
Nasdala et al. (2006) in cordierite. It is a totally transformed zone, filled
with material that we could expect to form during low-temperature
retrogression of diopside. The transformed zone is rather constant in
thickness: 20 to 38 m, with an average at 27 m. The transition to pristine
diopside is very sharp, but sometime indented. The transformations in
primary calcite are less visible but there is actually a porous calcite layer,
again 20 to 30 m thick, although it is there more difficult to estimate. At
the nano-scale, the uranothorianite­calcite interfaces are quite complex,
with several layers, of various chemical compositions and structural states.
Those observations suggest that primary calcite and diopside located
close to uranothorianite has been crystallized at low temperature, as a
secondary porous calcite for calcite, and as a mixture of secondary calcite
and clay mineral for diopside. The presence of hydrous minerals (smectite)
and carbonate (calcite) in reaction zone and in cracks, the presence of Pbrich
inclusions in secondary calcite, the abundance of fluid inclusions in
the porous layer in primary calcite, the dissociation of U and Th in the
calcite­uranothorianite layer, and the ThO2 chains along interfaces, are all
strong indications that low-temperature crystallization was promoted by a
fluid phase.
Surprisingly, truly metamict material is very rare, limited to a narrow
(200­300 nm) amorphous layer, along the calcite­uranothorianite interface.
However all structures described above are strictly limited to the
interface between uranothorianite and other minerals, so must be more or
less related to the effect of irradiation. In order to evaluate how irradiation
by U and Th could affect the minerals studied above, we carried out a
detailed study of irradiation damages using the SRIM/TRIM software.
5. Modeling with SRIM/TRIM
Damage was modeled by using the SRIM/TRIM software package
(Ziegler, 2006) which allows simulating the effect a particle in a target,
knowing what kind of particle it is, its energy, and the density and
chemical compositions of the target. It is based on full-quantum
calculations of individual interactions of incident particles with the
atoms of the target. For our purpose the interesting outputs are the
particle path (in three dimensions), and the vacancies created in the
Fig. 5. TEM foil # 750. EFTEM-BF image (on the top on the left), and U, Th, Si-jump ratio maps of the same zone showing from calcite to uranothorianite: (1) a large (~200­300 nm) amorphous
zone (A), enriched in U and Th, but without Si (see U, Th and Si maps) and enriched in Ca (see EDX spectrum on the right); (2) a chain (B) of very small (~20 nm) ThO2 crystals free of U; (3)
another amorphous zone (C), which, in contrast to zone A is enriched in Si (see EDX spectrum on the right and Si-map); and (4) another zone (D) made of small amorphous Si-rich "bubbles".
323A.-M. Seydoux-Guillaume et al. / Chemical Geology 261 (2009) 318­332
target. SRIM/TRIM always assumes that the target is amorphous and
isotropic.
5.1. Irradiation by U and Th chains
Irradiation by U and Th is done by all the radioelements of the 235
U,
238
U, and 232
Th decay chains which combine  and  decays to reach the
final stable nuclei: 207
Pb, 206
Pb and 208
Pb respectively. We will neglect
the decay which produces no major damage, and we will neglect also 
decays of minor branched radioactivity. During  decay, two particles
are created: the  particle itself, which is a 4
He nucleus, and the recoil
nucleus, which is actually the nucleus of the daughter element,
belonging to elements from U to Tl. The total energy involved during
an  decay is few MeV, distributed unevenly between the two particles
(about 100 keV for the recoil, and the rest for the ). Then each  decay
produces a slow, heavy nucleus (recoil), and a light, fast particle ().
Another source of damage in radioactive minerals is the spontaneous
fission of 238
U. It is a rare event (0.00005% of the 238
U decay) but the
energy involved is about 200 MeV, and it produces two fissions nuclei
with atomic number around 100 and 140 respectively. The energies of
the particles of the three decay chains are summarized in Table 2. For 
particles, it ranges from 4 to 8.8 MeV and for recoil for from 70 to 170 keV.
Using SRIM we calculated the effect of  and recoil particles of the
three decay chains, in diopside, calcite and uranothorianite, as well as
the effect of two typical fission products 90
Y and 140
Ce. Usually 1000
particles are calculated to obtain good statistics. Two key outputs are
presented in Table 2: the average distance reached by the particle, and
the number of vacancies created. To illustrate the effect of  and recoil
particles in minerals, the vacancy distributions for the 214
Po210
Pb
decay in diopside (:7.69 MeV; recoil:146 keV), are presented in Fig.
9. For other decays or other minerals, the amount of vacancies and the
distances are different, but the shapes are similar. The effects of 
particles and recoil nuclei are fundamentally different. The  particles
are penetratives (12­38 m in diopside, 14 to 45 m in calcite) and
produce 220 to 270 vacancies in diopside and 170­200 vacancies in
calcite. The recoil nuclei are 1000 times less penetrative (27­45 nm in
diopside, 32­55 nm in calcite) but 10 times more destructive (1200­
2500 vacancies in diopside, 940­2030 vacancies in calcite). The fast 
particle produces vacancies only at the end of its displacement, when
it has been slowed down enough by ionization to interact with target
nuclei. The slow recoil nucleus creates vacancies all along its
trajectory. In three dimensions, the trajectory of a series of  particles
has the shape of a cone, while it is pear-shaped for the recoils. The
maximum distance reached by an  particle is very close to the
average distance, whereas it is about twice as long for recoil nuclei.
Fundamentally the length of the trajectory increases with increasing
energy, and decreases with increasing density of the target. In
uranothorianite the length of the trajectories are about 9 to 30 m
for  particles, so only the outer part of the uranothorianite irradiates
the surrounding minerals.
Fig. 6. TEM foil # 753. Scanning Transmission Electron Microscope Bright Field (STEM-BF) image (A) and EFTEM-BF images (B and C) showing preferential dissolution of diopside
(Cpx) along cleavages, giving this particular triangular-shaped boundary, an unoriented clay mineral (-smectite), with its typical layer-form (B and C), and many Pb-rich as well as
U+Th-rich inclusions (arrows in A and C).
324 A.-M. Seydoux-Guillaume et al. / Chemical Geology 261 (2009) 318­332
5.2. Geometry
SRIM/TRIM simulates only the effect of a particle thrown into a target
along a single direction. In the present study we are studying the effect of
a layer of radioactive material (the outer part of the uranothorianite
grain) into a non-radioactive material (diopside and calcite). The
vacancy distribution calculated by SRIM/TRIM must then be corrected.
First, because radioactive decay emits particles in random directions and
not along a single path, and second, because the emitting volume is a
layer and not a point. The main output of SRIM/TRIM is the linear density
Fig. 7. TEM foil # 752. STEM-BF image (A) and EFTEM-BF images (B and C) showing a similar organization but reversed from the uranothorianite­Cc1 interface: an amorphous zone (A)
enriched in U, Th and Ca and a chain of ThO2 crystals (B) parallel to the interface (see also image B). Secondary calcite (Cc2) contains many fluid inclusions (see arrows in A and C) and
U+Th+Pb-rich inclusions (large inclusion on the right).
Fig. 8. TEMfoil#1092. STEM--Dark Field(DF)image (A)donewith highangleannulardarkfield (HAADF)detectorandTEM-BFimages(Band C)showingTEMfoilcut acrossone fracturewithin
Cpx (see Fig. 2D). The fracture (~2 mwide) is filled with a mixture of clay mineral (), calcite (Cc), and a ZrTiCaTh-amorphous phase (Zr), presumably metamict zirconolite (A). Nano-channels,
filled with nano-bubbles (fluid inclusions) are visible at the interface between Cpx and the crack border (B); the channel is larger (~70 nm compared to ~40 nm) on the upper side (A and B).
325A.-M. Seydoux-Guillaume et al. / Chemical Geology 261 (2009) 318­332
of vacancies created by a particle, expressed in vacancy/ion/, hereafter
named VAC r linear with r being the distance (radius). A series of N
particles, along a small distance dr situated at a distance r of the emitting
volume creates a number of vacancies N Á VAC r linearÁdr. In three
dimensions, because of the random direction of the emission, the
same amount of vacancies is diluted in a fraction of sphere 4r2
Á dr.
Therefore the vacancy density per ion around an emitting center is:
VAC r sphere =
VAC r linear
4r2
1
This expression is fundamentally the same as in Nasdala et al. (2001),
but calculated by differentiation.
As shown in the first part of this paper the thickness of the
damaged area (20­30 m) is small compared to the length of the
uranothorianite­diopside/calcite interface (few hundreds of m).
Therefore we will consider the irradiation geometry to be a plane
separating a radioactive domain from a non-radioactive domain. Each
volume of the irradiated medium situated at a distance L from the
interface will receive VAC r sphere damages from all the particles
emitted from the radioactive medium and situated at a distance r. The
volume of radioactive medium situated at a distance r is:
Vrad = X Á r
2
Á dr 2
with  the solid angle of the cone defined by r and L, which is given by
X = 2 1 -
L
r
 
: 3
Combining Eqs. (1)­(3) and summing over the whole radioactive
volume, we obtain the density of vacancies created at a distance L of the
radioactive half-space:
VAC r plane =
Z 
L
1
2
Á VAC r linearÁ 1 -
L
r
 
Á dr: 4
Since we do not have an analytical expression for VAC r linear, this can
only by estimated numerically from the output of SRIM/TRIM. The
calculation of the exact expression would require calculating each
VAC r linear function for all L and all particles. In order to maintain the
calculation time reasonable we decided, for each particle and each
target, to use only the VAC r linear function calculated for the target
mineral. The consequences of this approximation will be discussed later.
5.3. The time scale
The petrological study showed that the altered aureole was formed
by recrystallization of the target minerals, calcite or diopside. To
estimate the amount of damage that the target had experienced when
recrystallization occurred, we should know the time at which damage
started, assumed to be the age of the rock formation (550 Ma), and the
time at which alteration and recrystallization occurred, which is
unknown. This initial damage at recrystallization is not zero because
some lead, presumably radiogenic has been found in the altered
aureole, but it can be from few millions to few hundredths of millions
years. This is a major uncertainty, which affect the amount of damage
and the relative proportion of damages due to the 235
U, 238
U and 232
Th
chains. However neither the size of the damaged area, which depends
only on the nature of the target and the particles energy, nor the shape
of the vacancy distribution, which depends mainly on the shape of
Table 2
Summary of the results of TRIM/SRIM calculations.
E (MeV) Alpha particles E (keV) Recoil nuclei
Diopside Calcite Uranothorianite Diopside Calcite Uranothorianite
dist dam dist dam dist dam dist dam dist dam dist dam
238
U
238
U234
U 4.20 12.3 228 14.6 172 9.7 184 71 27 1231 33 983 16 971
234
U230
Th 4.78 14.8 225 17.5 186 11.5 190 83 29 1414 36 1129 16 1124
230
Th226
Ra 4.69 14.4 221 17.1 178 11.2 198 83 29 1381 36 1117 17 1113
226
Ra222
Rn 4.79 14.7 224 17.5 179 11.5 190 86 29 1447 36 1145 18 1141
222
Rn218
Po 5.49 18.0 238 21.4 184 14.0 203 90 31 1485 41 1315 19 1312
218
Po214
Pb 6.00 20.5 236 24.4 193 15.9 206 112 35 1826 43 1456 22 1467
214
Po210
Pb 7.69 30.0 251 35.7 197 22.9 224 146 42 2300 53 1830 24 1836
210
Po206
Pb 5.31 17.1 236 20.4 185 17.1 209 103 34 1694 42 1357 20 1341
232
Th
232
Th228
Ra 4.01 11.8 219 14.0 173 9.0 136 69 27 1191 32 937 15 927
228
Th224
Ra 5.42 18.1 232 21.4 186 13.7 194 95 31 1577 38 1247 19 1224
224
Ra220
Rn 5.69 19.4 231 23.0 177 14.6 210 102 32 1647 40 1329 19 1310
220
Rn216
Po 6.29 22.5 233 26.7 189 16.9 203 114 35 1848 43 1459 21 1443
216
Po212
Pb 6.78 25.2 252 29.9 193 18.8 213 126 38 2032 46 1607 23 1595
212
Bi208
Tl (36%) 6.09 21.5 231 25.5 194 16.1 203 115 34 1863 43 1470 21 1450
212
Po208
Pb (64%) 8.79 37.8 268 44.7 201 27.7 225 166 45 2592 55 2029 44 2038
235
U
235
U231
Th 4.69 14.3 221 17.1 178 10.9 192 81 29 1386 35 1095 16 1057
231
Pa227
Ac 5.06 16.0 231 19.0 181 12.1 192 89 30 1497 36 1194 18 1179
227
Th223
Ra 6.04 20.8 239 24.7 190 15.6 206 108 34 1754 42 1406 21 1382
223
Ra219
Rn 5.87 19.9 235 23.7 180 15.0 199 107 35 1742 42 1400 21 1376
219
Rn215
Po 6.82 24.9 243 29.6 191 18.6 214 127 37 2035 47 1623 24 1597
215
Po211
Pb 7.39 28.2 245 33.5 197 21.0 204 140 41 2244 50 1785 26 1750
211
Bi207
Tl 6.62 23.8 241 28.3 185 17.8 207 128 38 2055 47 1634 26 1604
Fission
90
Y (100 MeV) 15.8 m 39000 vacancies/ion 9.10 m 43500 vacancies/ion
140
Ce (100 MeV) 17.5 m 81000 vacancies/ion 8.50 m 95700 vacancies/ion
E: energy of the particle, in MeV (alpha) or in keV (recoil); dist: average distance done by the particle in m (alpha) or in nm (recoil); dam: damage in vacancies/ion.
326 A.-M. Seydoux-Guillaume et al. / Chemical Geology 261 (2009) 318­332
individual particle damaging and of the geometrical corrections, are
affected. All the calculations have been carried out assuming 550 Ma
as the irradiation time. As a result of this major uncertainty, we made
no attempt to estimate dpa (displacements per atom) from the
number of vacancies in target minerals.
5.4. Results
Uranothorianite has been auto-irradiated since its formation. It is
possible to calculate the number of vacancies directly from the average
damage in Table 2, integrated over 550 Ma, and converted it in dpa
(displacement per atom). The calculated average of 240 dpa over
550 Ma, when compared to the dose necessary to completely
amorphized thorite (ThSiO4), ~0.20 dpa at room temperature (Meldrum
et al., 1999), suggests that the uranothorianite should be completely
amorphous. However, because electron diffraction patterns (SAD--
Fig. 4) demonstrate that it is crystalline, uranothorianite must selfanneal
by some processes. As a comparison, amorphization of UO2
seems impossible, because of the highly rapid recombination of the
bonding; for an irradiation dose equivalent to 25 dpa at ~170 °C, UO2 is
still crystalline (Matzke and Turos, 1992; Matzke and Wang, 1996).
Results for diopside and calcite are presented in Figs. 10 and 11.
Spontaneous fission of 238
U is such rare event that it can be neglected.
The shapes of the curves, representing the damage due to  particles
as a function of distance from uranothorianite are similar. Most
damage is accumulated at short distances (about 15 m in diopside
and 17 m in calcite). The more distant part corresponds to the area
damaged only by the more energetic particles emitted close to the
uranothorianite surface. The total damaged area is then about 35 m
wide in calcite and 30 m in diopside. Recoils nuclei create 1000 times
more damage than  particles, with a very steep decrease for a
damaged zone of 60 nm in calcite and 50 nm in diopside.
Fig. 9. Projection of the vacancy distribution created by a 7.69 MeV  particle (A) and the corresponding recoil (B) nuclei (210
Pb at 146 keV) in diopside. The fast  particle produces
vacancies only at the end of its displacement, when it has been slowed down enough by ionization to interact with target nuclei. The slow recoil nucleus creates vacancies all along its
trajectory. Note that vertical and horizontal scales are very different in the two figures.
327A.-M. Seydoux-Guillaume et al. / Chemical Geology 261 (2009) 318­332
The thickness of the damaged area calculated for  in diopside is
similar to the width of the recrystallized area observed in thin section.
The thickness of the porous calcite layer around along the primary
calcite­thorianite interface is more difficult to estimate but is also
similar to the size of the area damaged by  as calculated above.
However the characteristic distances calculated for the damage of recoil
nuclei (50 to 60 nm) are far away from the size of the totally amorphous
layer seen at the calcite­uranothorianite interface (150­250 nm).
5.5. Effect of wandering recoils
Modeling the effect of recoil nuclei by a succession of individual
ions projected from uranothorianite into the target is, actually, a poor
representation of the real process occurring during the radioactive
decay of a chain. In decay chains the recoil nucleus is the next
radioactive element. During radioactive decay, the recoil moves of few
tens of nm, in a random direction. Therefore, there is a possibility, for
decay occurring close to the uranothorianite surface, that after decay
the recoil is indeed located out of the uranothorianite. Then next
decay will occur inside the diopside (or calcite), and the new recoil
will move again, in a random direction. For damages created by  this
effect can be neglected, because the path of  is much longer than the
possible displacement of the recoil, but for the recoil itself, the damaged
area can be significantly enlarged by the "wandering recoil effect": a
series of recoils that would move always in the same direction, although
very unlikely, can ends few hundredths of nm inside diopside (or
calcite). In order to quantitatively estimate this effect we simulated this
effect by the following procedure: (1) for each decay, a random direction
is chosen; (2) the average damage is distributed along the path, as
calculated by SRIM/TRIM; (3) the recoil (daughter) nucleus is displaced
in that direction, at the average distance shown in Table 2; (4) the next
decay in the chain is processed from this position, as described in (1)(2)­(3).
For each chain and for both diopside and calcite we simulated
1000 decay chains. At the end of the calculation, the radial distribution of
the damages is calculated. Strictly, this procedure only simulates a decay
chain initiated exactly at the interface. Again, because we do not know
the time scale, but only the shape of the curves, only the characteristics
distances have real meanings in this simulation. The results are
presented in Fig. 12. The main result is that the thickness of the area
damaged by the recoils is approximately multiplied by 3 because of the
Fig. 10. Spatial distribution of the vacancies created in calcite by  particles (A) and recoil nuclei (B) emitted by uranothorianite. The calculation includes all particles from the three
decay chains, integrated for 550 Ma. It assumes that uranothorianite is a semi-infinite emitting medium irradiating a semi-infinite calcite half-space, separated by a plane. Note that
vertical and horizontal scales are very different in the two figures.
328 A.-M. Seydoux-Guillaume et al. / Chemical Geology 261 (2009) 318­332
"wandering recoil effect". With this correction, the size of the area
damaged by recoil nuclei in calcite is in rather good agreement with the
thickness of the totally amorphous layer at the calcite­uranothorianite
interface (Figs. 4 and 7).
6. Interpretation and discussion
6.1. A reasonable history of radiohalos in Tranomaro skarns
From petrographical study and damage modeling, the evolution of
the rock can be reconstructed as follows. The initial rock is a hightemperature
marble, mainly made of aluminous diopside and primary
calcite, with some large uranothorianite grains. As soon as the latter
is formed, radioactive decays start, with three consequences: (1)
radiogenic Pb accumulates in uranothorianite; (2) radiation damage
accumulates in uranothorianite and in the contiguous minerals; and
(3) the uranothorianite grains swells because of radiation damage. For
uranothorianite similar to the minerals studied here, Evron et al.
(1994) estimated a macroscopic volume expansion of ~1.5%. With
increasing time, swelling creates radial cracks in diopside and calcite,
and radiation damage accumulates and weakens calcite and diopside
around uranothorianite. A heavily damaged (amorphous) thin layer
is created along the interface by the recoils, and a partially or totally
damaged layer, 20­40 m is created by . Since no Helium bubbles
was observed within uranothorianite by TEM, it is supposed that
radiogenic Helium diffused out of the crystal through the cracks. After
some time, a low-temperature fluid infiltrates through the rock, using
the cracks and the weak interfaces as preferential pathways. The
damaged part of diopside is retrogressed as a mixture of clay and
secondary calcite. The damaged part of primary calcite recrystallized
as a secondary porous calcite layer. At this stage, cracks in calcite are
healed, whereas, cracks in diopside are partially filled by various
secondary material (Fig. 8; calcite, zirconolite, clay mineral). Uranothorianite
is also affected by fluids, as shown by the presence of Pb,
U, and Th in the reaction zones (Fig. 6) and in the amorphous layers A
and C (Figs. 4 and 5), and the presence of ThO2 chains (Figs. 4, 5 and 7).
However the fact that the chemical ages are in good agreement with
the age of the rock indicates that this mineral is not deeply penetrated
Fig.11. Spatial distribution of the vacancies created in diopside by  particles (A) and recoil nuclei (B) emitted by uranothorianite. The calculation includes all particles from the three
decay chains, integrated for 550 Ma. It assumes that uranothorianite is a semi-infinite emitting medium irradiating a semi-infinite diopside half-space, separated by a plane. Note that
vertical and horizontal scales are very different in the two figures.
329A.-M. Seydoux-Guillaume et al. / Chemical Geology 261 (2009) 318­332
by fluids. During fluid infiltration, some elements are partially
redistributed, including elements which are not obviously present in
the thin section such as Zr. After this hydrothermal episode, the newly
formed phases are again affected by radioactive damage.
The exact significance of the features observed along the
uranothorianite and calcite is at this stage not totally explained.
However the B zone, a chain of ThO2 grains, suggest an in-situ
reprecipitation of ThO2 during partial dissolution of uranothorianite,
as it is commonly observed in experiments with ThO2 (SeydouxGuillaume
et al., 2002; Heisbourg et al., 2003; Heisbourg et al., 2004).
6.2. Some properties of uraninite, calcite, and diopside relative to radiation
damages
The observations made on this sample support several hypotheses
on the behavior of calcite, diopside and uranothorianite under
irradiation. Uranothorianite is remarkably resistant to irradiation
damage. This is not unexpected, since uraninite, which is isostructural
with uranothorianite remains crystalline (Evron et al. 1994; Janeczek
et al. 1996). By contrast, a thorium silicate (ThSiO4) that accumulated
approximately the same dose as the uranothorianite studied here
(~5×1020
/g) is completely amorphous (Seydoux-Guillaume et al.,
2007). This observation is consistent with higher resistance to
amorphization by radiation damage (or greater ability to self-anneal)
in minerals for which the long-range ionic forces dominates over the
short-range covalent forces (Trachenko, 2004), as in the case for
uranothorianite when compared to thorite.
Contrary to uranothorianite, diopside seems to be highly sensible
to radiation damage. The average size of the retrogressed zone is
27 m. At this distance the damage is only 2% of the maximum damage
created by the  at the contact with uranothorianite. For 550 Ma of
irradiation, this would correspond to 0.0014 dpa, so it means that the
resistance of diopside relative to low-temperature fluids is modified
even if about 1/1000 atoms are displaced. We should recall that
Fig. 12. Spatial distribution of the vacancies created in diopside (A) and calcite (B) by wandering nuclei. The model simulates the random trajectories of recoil nuclei during the
successive decays in the three decay chains, and estimates, via TRIM/SRIM simulation the vacancies created along the path. The integration time is 550 Ma. Those diagrams should be
compared to Figs. 10 and 11(B).
330 A.-M. Seydoux-Guillaume et al. / Chemical Geology 261 (2009) 318­332
550 Ma is the maximum for the irradiation time since it assumes that
retrogression occurred at the present time.
The limit between pristine diopside and the retrogressed area is
very sharp; there is no intermediate zone in which diopside would be
partly retrogressed. This suggests that there is a threshold in the
sensitivity of damaged diopside to low temperature fluid. Below a
certain value of damage the diopside behaves as an undamaged
diopside, and is not affected by alteration. We should recall that
diopside is a common mineral in heavy minerals concentrates in river,
so it is not normally destroyed by low-T fluids.
In calcite the porous layer, which can be assumed to be the
equivalent of the retrogressed zone in diopside, is ~10 to 15 m wide,
as estimated from Fig. 2. At this distance damage is 45 to 25% of the
maximum (550 Ma)  damage, for 0.013 to 0.0055 dpa. Calcite seems
then more resistant to irradiation damage than diopside, probably due
to the type of interatomic forces in that structure (Trachenko, 2004). It
is important to remember, that diopside crystals that are not in
contact with uranothorianite in the rest of the rock are pristine. This
demonstrates that irradiation is responsible for the destabilization of
diopside by fluids. However, calcite directly in contact with
uranothorianite (see presence of Ca within amorphous zone A (Figs.
4A and 5) is amorphous, showing the strong effect of "continuous
irradiation by wandering recoil". It should be also noted that calcite is
always crystalline, even at the contact with uranothorianite where the
damage rate can be up to 0.055 dpa.
Outside the modified area around uranothorianite, the rock is very
fresh; the only indication of the low-T fluid infiltration is secondary
material along the diopside­uranothorianite interfaces and, at nanoscale,
in swelling structures, i.e. cracks (Fig. 8). This definitely shows that
uranothorianite grains create weak zones in the rock, by destroying the
surrounding minerals, and by creating swelling cracks.
The thickness of the retrogressed area at the diopside­uranothorianite
interface is approximately constant all around the thin section. It
means that the effect of irradiation damage in diopside does not depend
on the orientation of the diopside crystals although it is an anisotropic
chain-silicate. The only visible indication of the anisotropy of diopside is
the local presence of indentation at the pristine-retrogressed limit, due
to limited fluid infiltration along cleavage surfaces.
6.3. Reliability of the conclusions
The above conclusions are based on a detailed study of structures
surrounding a single uranothorianite grain. The representativity of the
studied zone can then be legitimately questioned. Many features
observed around this particular grain are visible in the whole thin
section, such as the radiating cracks, the presence of retrogressed zone
along the diopside­uranothorianite interface, and the presence of a
porous layer along the calcite­uranothorianite. Those features are also
present in all studied samples of the same rock. Observations at the
nano-scale by coupling FIB/TEM analyses, like the presence of a wide
amorphous zone along the calcite­uranothorianite boundary, or the
presence of chains of ThO2 crystals, which are time-consuming,
cannot easily be multiplied all around a thin section, but are regular in
shape and size, and are in agreement with damage model. Therefore
we are quite confident that these are general features in this sample.
Moreover, similar observations have been reported in other studies
(Seydoux-Guillaume et al., 2003, 2007; Nasdala et al., 2006; SeydouxGuillaume,
unpublished), suggesting that it is common in rocks
containing radioactive minerals. Some observations are clearly
accidental, and may not be representative, such as the presence of
zirconolite grains in the fractures.
6.4. Reliability of the model
In this study the SRIM/TRIM model provided several important
constraints. This software has the advantage of being the most
sophisticated program available for non-specialist; on the other hand
it has some limitations; for example it considers all materials as
isotropic, and neglects the effect of crystalline structures. However,
the fact that the size of damaged area in diopside is constant whatever
its orientation provides an a posteriori justification of this assumption.
Diopside is a chain-based, highly anisotropic structure, and if radiation
damage is sensitive to crystal anisotropy, it should be visible in this
mineral.
7. Conclusions
Radiohalos are complex structures, which deserve detailed study.
Since radiation damage has a visible effect at the thin section scale, but
also affects the structure of crystals at the atomic scale, studies must
be conducted with a variety of techniques, including FIB/TEM for the
nano-scale. Radiohalos are unique sources of information for understanding
the behavior of minerals submitted to irradiation damages,
even for non-radioactive minerals. This approach allows to study the
effects of long-term irradiation and is thus very complementary to
other methods that involve doping experiments with short-lived
isotopes, which are far more expensive.
In the present study we have demonstrated the high resistance of
uranothorianite to self-irradiation damage, in agreement with previous
results (Evron et al.,1994), and, by contrast, the high sensitivity of
diopside to radiation damage from neighbouring uranothorianite.
TRIM/SRIM simulations have been shown to be able to correctly
simulate radiation damage, even in complex minerals such as diopside.
Finally, we would like to emphasize that radiohalos are actually a
point of chemical and mechanical weakness in a rock and probably a
starting point for alteration.
Acknowledgements
FIB/TEM analyses have been done thanks to the financial support
for travels to Potsdam from PROCOPE (2005­2006) No. 09638ZB. The
authors want to thank Ph. De Parseval and T. Aigouy for their technical
assistance with the Electron Microprobe and with the SEM, and L.
Datas and L. Weingarten for their technical assistance with the TEM at
the TEMSCAN service from UPS. Constructive comments from R.C.
Ewing and an anonymous reviewer were appreciated; we also want to
thank J. Hanchar for his excellent editorial work. Thanks to M. Jessel
for correcting the English language.
References
Balan, E., Allard, T., Fritsch, E., Sélo, M., Falgures, C., Chabaux, F., Pierret, M.P., Calas, G.,
2005. Formation and evolution of lateritic profiles in the middle Amazon basin:
insights from radiation-induced defects in kaolinite. Geochimica et Cosmochimica
Acta 69, 2193­2204.
Begg, B., Hess, N.J., Weber, W.J., Conradson, S.D., Scheiger, M.J., Ewing, R.C., 2000. XAS
and XRD study of annealed 238
Pu and 239
Pu substituted zircons (Zr0.92Pu0.08SiO4).
Journal of Nuclear Materials 278, 121­224.
Black, L.P., Fitzgerald, J.D., Harley, S.L., 1984. Pb isotopic composition, colour, and
microstructure of monazites from a polymetamorphic rock in Antarctica. Contrib.
Mineral. Petrol. 85, 141­148.
Boulvais, P., Fourcade, S., Gruau, G., Moine, B., Cuney, M., 1998. Persistence of premetamorphic
C and O isotopic signatures in marbles subject to Pan-African
granulite facies metamorphism and U­Th mineralization (Tranomaro, South East
Madagascar). Chemical Geology 150, 247­262.
Boulvais, P., Fourcade, S., Moine, B., Gruau, G., Cuney, M., 2000. Rare-earth elements
distribution in granulite-facies marbles: a witness of fluid­rock interaction. Lithos 53,
117­126.
Burakov, B.E., Hanchar, J.M., Garbusov, V.M., Zirlin, V.A., 2002. Synthesis and investigation
of Pu doped single crystal zircon ZrPuSiO4. Radiochimica acta 90, 95­97.
Crocombette, J.P., Ghaleb, D., 2001. Molecular dynamics modeling of irradiation damage
in pure and uranium doped zircon. Journal of Nuclear Materials 295, 167­178.
Davis, D.W., Krogh, T.E., 2000. Preferential dissolution of 234
U and radiogenic Pb from recoil-damaged
lattice sites in zircon: implications for thermal histories and Pb
isotopic fractionation in the near surface environment. Chemical Geology 172,
41­58.
Evron, R., Kimmel, G., Eyal, Y., 1994. Thermal recovery of self-radiation damage in
uraninite and thorianite. Journal of Nuclear Materials 217, 54­66.
331A.-M. Seydoux-Guillaume et al. / Chemical Geology 261 (2009) 318­332
Ewing, R.C., 1975. The crystal chemistry of complex niobium and tantalum oxides IV.
The metamict state: discussion. American Mineralogist 60, 728­733.
Ewing, R.C., 1994. The metamict state: 1993--the centennial. Nuclear Instruments and
Methods in Physics Research B91, 22­29.
Ewing, R.C., Wang, L.M., 2002. Phosphates as nuclear waste forms. In: Kohn, M.J.,
Rakovan, J., Hughes, J.M., Ribbe, P.H. (Eds.), Phosphates: Geochemical, Geobiological
and Materials Importance. . Reviews in Mineralogy and Geochemistry, vol. 48.
Mineralogical Society of America, pp. 673­699.
Ewing, R.C., Weber, W.J., Clinard Jr., F.W., 1995. Radiation effects in nuclear waste forms.
Progress in Nuclear Energy 29, 63­127.
Ewing, R.C., Chakoumakos, B.C., Lumpkin, G.R., Murakami, T., Greegor, R.B., Lytle, F.W.,
1988. Metamict minerals: natural analogues for radiation damage effects in ceramic
nuclear waste forms. Nuclear Instruments and Methods in Physics Research B32,
487­497.
Ewing, R.C., Meldrum, A., Wang, L.M., Wang, S.X., 2000. Radiation-induced amorphisatio.
In: Redfern, S.A.T., Carpenter, M.A., Ribbe, P.H. (Eds.), Transformation
Processes in Minerals. . Reviews in Mineralogy and Geochemistry, vol. 39.
Mineralogical Society of America, pp. 319­361.
Ewing, R.C., Meldrum, A., Wang, L.M., Weber, W.J., Corrales, L.R., 2003. Radiation damage
in zircon. In: Hanchar, J.M., Hoskin, P.W.O. (Eds.), Zircon. . Reviews in Mineralogy and
Geochemistry, vol. 53. Mineralogical Society of America, pp. 387­425.
Farges, F., 1997. Coordination of Ti4+ in silicate glasses: A high-resolution XANES
spectroscopic study at the Ti K edge. Am. Mineral. 82, 36­43.
Farges, F., Calas, G., 1991. Structural analysis of irradiation damage in zircon and thorite.
An X-ray absorption spectroscopic study. American Mineralogist 76, 60­73.
Farges, F., Ewing, R.C., Brown, G.E., 1993. The structure of aperiodic, metamict (Ca,Th)
ZrTi2O7 (zirconolite): an EXAFS study of the Zr, Th and U sites. Journal of Material
Research 8, 1983­1995.
Farnan, I., Balan, E., Pickard, C.J., Mauri, F., 2003. The effect of radiation damage on local
structure in the crystalline fraction of ZrSiO4: investigating the Si-29 NMR response
to pressure in zircon and reidite. American Mineralogist 88, 1663­1667.
Gentry, R.V., 1973. Radioactive halos. Annual review in Nuclear Science 23, 347­362.
Gentry, R.V., 1974. Radiohalos in a radiochronological and cosmological perspective.
Science 184, 62­66.
Harfouche, M., Farges, F., Crocombette, J.-P., Flank, A.-M., 2005. XAFS and molecular
dynamics study of the structural environment around actinides and network
formers in natural minerals analogues of ceramics for nuclear waste storage.
Physica Scripta T115, 928­930.
Hawthorne, F.C., Groat, L.A., Raudsepp, M., Ball, N.A., Kimata, M., Spike, F.D., Gaba, R.,
Halden, N.M., Lumpkin, G.R., Ewing, R.C., Greegor, R.B., Lytle, F.W., Ercit, T.S.,
Rossman, G.R., Wicks, F.J., Ramik, R.A., Sherriff, B.L., Fleet, M.E., McCammon, C.,1991.
Alpha-decay damage in titanite. American Mineralogist 76, 370­396.
Headley, T.J., Ewing, R.C., Haaker, R.F., 1981. Amorphous structure of metamict minerals
observed by TEM. Nature 293, 449­450.
Heisbourg, G., Hubert, S., Dacheux, N., Ritt, J., 2003. The kinetics of dissolution of Th1 -xUxO2
solid solutions in nitric media. Journal of Nuclear Materials 321, 141­151.
Heisbourg, G., Hubert, S., Dacheux, N., Purans, J., 2004. Kinetic and thermodynamic studies
of the dissolution of thoria-urania solid solutions. Journal of Nuclear Materials 335,
5­13.
Janeczek, J., Eby, R.K., 1993. Annealing of radiation damage in allanite and gadolinite.
Physics and Chemistry of Minerals 19, 343­356.
Janeczek, J., Ewing, R.C., Oversby, V.M., Werme, L.O., 1996. Uraninite and UO2 in spent
nuclear fuel: a comparison. Journal of Nuclear Materials 238, 121­130.
Joly, J., 1907. Pleochroic halos. Philosophical Magazine 13, 381­383.
Lumpkin, G.R., Chakoumakos, B.C., 1988. Chemistry and radiation effects of thoritegroup
minerals from the Harding pegmatite, Taos County, New Mexico. American
Mineralogist 73, 1405­1419.
Lumpkin, G.R., Ewing, R.C., 1988. Alpha decay damage in minerals of the pyrochlore
group. Physics and Chemistry of Minerals 16, 2­20.
Lumpkin, G.R., Chakoumakos, B.C., Ewing, R.C., 1986a. Mineralogy and radiation effects
of microlite from the Harding Pegmatite, Taos County, New Mexico. American
Mineralogist 71, 569­588.
Lumpkin, G.R., Ewing, R.C., Chakoumakos, B.C., Greegor, R.B., Lytle, F.W., Forltyn, E.M.,
Clinard Jr., F.W., Boatner, L.A., Abraham, M.M., 1986b. Alpha-recoil damage in
zirconolite (CaZrTi2O7). Journal of Material Research 1, 564­576.
Lumpkin, G.R., Eby, R.K., Ewing, R.C., 1991. Alpha-recoil damage in titanite (CaTiSiO5):
direct observation and annealing study using high resolution transmission electron
microscopy. Journal of Material Research 6, 560­564.
Linberg, M.L., Ingram, B., 1964. Rare-earth silicatian apatite from the Adirondack
Mountains, New-York. US Geological Survey Professional Paper 501-B, B64­B65.
Matzke, H.j., Turos, A., 1992. Ion implantation studies of UO2 and UN. Journal of Nuclear
Materials 188, 285­292.
Matzke, H.j., Wang, L.M., 1996. High-resolution transmission electron microscopy of ion
irradiated uranium oxide. Journal of Nuclear Materials 231, 155­158.
Meldrum, A., Boatner, L.A., Weber, W.J., Ewing, R.C., 1998. Radiation damage in zircon
and monazite. Geochimica et Cosmochimica Acta 62, 2509­2520.
Meldrum, A., Zinckle, S.J., Boatner, L.A., Ewing, R.C.,1999. Heavy-ion irradiation effects in
the ABO4 orthosilicates: decomposition, amorphization, and recrystallization.
Physical Review B59, 3981­3992.
Meunier, J.D., Sellier, E., Pagel, M., 1990. Radiation damage rims in quartz from uranium
bearing sandstones. Journal of Sedimentary Petrology 60, 53­58.
Moine, B., Rakotondratsima, C., Cuney, M., 1985. Les pyroxénites  urano-thorianite du
sud-est de Madagascar, conditions physico-chimiques de la métasomatose. Bulletin
de Minéralogie 108, 325­340.
Moine, B., Ramambazafy, A., Rakotondrazafy, M., Ravololomiandrinarivo, B., Cuney, M.,
de Parseval, P., 1998. The role of fluorine-rich fluids in the formation of the
thorianite and sapphire deposits of S.E. Madagascar. 8th Goldschmidt Conference.
Mineralogical Magazine, vol. 62A, pp. 999­1000.
Mügge, O., 1907. Radioaktivität als Ursache der pleochroitischen Höfe. Zentralblatt
Mineralogie Geologie 71, 529­532.
Murakami, T., Chakoumakos, B.C., Ewing, R.C., Lumpkin, G.R., Weber, W.J., 1991. Alphadecay
event damage in zircon. Am. Mineral. 76, 1510­1532.
Nasdala, L., Irmer, G., Wolf, D., 1995. The degree of metamictization in zircons a Raman
spectroscopic study. European Journal of Mineralogy 7, 471­478.
Nasdala, L., Wenzel, M., Andrut, M., Wirth, R., Blaum, P., 2001. The nature of radiohaloes
in biotite: experimental studies and modeling. American Mineralogist 86, 498­512.
Nasdala, L., Wildner, M., Wirth, R., Groshopf, N., Pal, D.C., Möller, A., 2006. Alpha particle
haloes in chlorite and cordierite. Mineralogy and Petrology 86, 1­27.
Odom, A.L., Rink, W.L., 1989. Giant radiation-induced colored halos in quartz solution to
a riddle. Science 246, 107­109.
Ouchani, S., Dran, J.C., Chaumont, J., 1997. Evidence of ionization annealing upon
helium-ion irradiation of pre-damaged fluorapatite. Nuclear Instruments and
Methods. Physical Research, vol. B132, pp. 447­451.
Overwijk, M.H.F., van den Heuvel, F.C., Bulle-Lieuwma, C.W.T., 1993. Novel scheme for
the preparation of transmission electron microscopy specimens with a focused ion
beam. Journal of Vacuum Science and Technology 11, 202.
Owen, M.R., 1988. Radiation damage halos in quartz. Geology 16, 529­532.
Paquette, J.L., Nédélec, A., Moine, B., Rakotondrazafy, M., 1994. U­Pb, single zircon
evaporation and Sm­Nd isotopic study of a granulite domain in SE Madagascar.
Journal of Geology 102, 523­538.
Rakotondratsima, C., 1983. Les pyroxénites  Uranothorianite du Sud-Est de Madagascar:
étude pétrographique, minéralogique et géochimique. Thesis, Claude
Bernard University-Lyon, 226 pp.
Rakotondrazafy, M.,1995. La hibonite (CaAl12O19) du Sud-Est de Madagascar. Caractres
et modalités de formation dans les skarns  thorianite du facis granulite
(Unpublished). Thesis, Antananarivo University, Madagascar.
Rakotondrazafy, M., Moine, B., Cuney, M., 1996. Mode of formation of hibonite
(CaAl12O19) within the U­Th skarns from the granulites of S­E Madagascar.
Contribution to Mineralogy and Petrology 123, 190­201.
Ramambazafy, A., 1998. Granites et fluides en relation avec les skarns  thorianite dans
les granulites du S.E. de Madagascar. Thesis, Paul Sabatier Toulouse III University,
302 pp.
Ramambazafy, A., Moine, B., Rakotondrazafy, M., Cuney, M., 1998. Signification des
fluides carboniques dans les granulites et les skarns du Sud-Est de Madagascar.
Compte Rendu de l'académie des Sciences 327, 743­748.
Roberts, S., McCaffrey, J., Giannuzzi, L., Stevie, F., Zaluzec, N., 2001. Advanced techniques
in TEM specimen preparation. In: Xiao-Feng, Zhang., Ze, Zhang (Eds.), Progress in
transmission electron microscopy, vol. 1. Springer Series in Surface Sciences, 38, pp.
336­342.
Romer, R., 2003. Alpha-recoil in U­Pb geochronology: effective sample size matters.
Contribution to Mineralogy and Petrology 145, 481­491.
Seydoux-Guillaume, A.M., Paquette, J.L., Wiedenbeck, M., Montel, J.M., Heinrich, W.,
2002. Experimental resetting of the U­Th­Pb systems in monazite. Chemical
Geology 191, 165­181.
Seydoux-Guillaume, A.M., Goncalves, P., Wirth, R., Deutsch, A., 2003. TEM study of
polyphasic and discordant monazites: site specific specimen preparation using the
Focused Ion Beam technique. Geology 31, 973­976.
Seydoux-Guillaume, A.M., Wirth, R., Deutsch, A., Schärer, U., 2004. Microstructure of 24 ­
1928 Ma concordant monazites: implications for geochronology and nuclear waste
deposits. Geochimica et Cosmochimica Acta 68, 2517­2527.
Seydoux-Guillaume, A.M., Wirth, R., Ingrin, J., 2007. Contrasting response of ThSiO4 and
monazite to natural irradiation. European Journal of Mineralogy 19, 7­14.
Trachenko, K.O., 2004. Understanding resistance to amorphization by radiation
damage. Journal of Physics: Condensed Matter 16, 1491­1515.
Trachenko, K.O., Dove, M.T., Salje, E.K.H., 2001. Atomistics modeling of radiation
dammage in zircon. Journal of Physics: Condensed Matter 13, 947­952.
Vance, E.R., Metson, J.B., 1985. Radiation damage in natural titanites. Physics and
Chemistry of Minerals 12, 255­260.
Wang, L.M., Ewing, R.C., 1992. Ion beam induced amorphization of complex ceramic
materials -- minerals. Materials Research Society Bulletin 13, 38­44.
Weber, W.J., Ewing, R.C., Catlow, C.R.A., Diaz de la Rubia, T., Hobbs, L.W., Kinishita, C.,
Matzke, H.J., Motta, A.T., Nastasi, M., Salje, E.H.K., Vance, E.R., Zinkle, S.J., 1998.
Radiation effects in crystalline ceramics for the immobilization of high-level
nuclear waste and plutonium. Journal of Material Research 13, 1434­1484.
Weber, W.J., Ewing, R.C., Wang, L.M., 1994. The radiation-induced crystalline-toamorphous
transition in zircon. Journal of Materials Research 9, 688­698.
Wirth, R., 2004. Focused Ion Beam (FIB): a novel technology for advanced application of
micro- and nanoanalysis in geosciences and applied mineralogy. European Journal
of Mineralogy 16, 863­876.
Young, R.J., 1997. Application of the focused ion beam in materials characterization and
failure analysis. Microstructural Science 25, 491­496.
Zhang, M., Salje, E.K.H., 2001. Infrared spectroscopic analysis of zircon: radiation damage
and the metamict state. Journal of Physics: Condensed Matter 13, 3057­3071.
Zhang, M., Salje, E.K.H., Bismayer, U., Groat, L.A., Malcherek, T., 2002. Metamictization and
recrystallization of titanite: an infrared spectroscopic study. American Mineralogist
87, 882­890.
Ziegler, J.F., 2006. SRIM: the stopping and range of ions in matter. Instruction Manual.
332 A.-M. Seydoux-Guillaume et al. / Chemical Geology 261 (2009) 318­332