SEPARATION METHODS B -a jan havlis : masaryk university :: central european institute of technology ::: mendel centre for plant genomics and proteomics faculty of science, national centre for biomolecular research ::: division of functional genomics and proteomics J (cc}©®(2)| Creative Commons: Attribution-Noncommercial-Share alike 3.0 Unported License analytical separation method analytical separation : gas chromatography :: GC : separation of macromolecules ::SEC, GPC, HCD a FFF : separation in force field :: CZE, MEKC, CIEF, ITP, CEC, ACE, NCE a CE-on-chip :: MS (Q, QqQ, IT, TOF, FT-ICR, OT) : membrane separation :: dialysis, ultrafiltration another aspects of analytical separations chiral separation separation method development and optimisation validation of analytical separation method f-"\ recommended reading V_) J. C. Giddings, Unified separation science, Wiley 1991 D. Hieger, An introduction to high performance capillary electrophoresis, Agilent Technologies 2000 C. F. Poole, The essence of chromatography, Elsevier 2003 R. L. Grob et a/., Modern practice of gas chromatography, Wiley 2004 I-1 '■ development of chromatographic method choice of separation system - suitable SF type knowing the sample, we choose SF choice of separation conditions - suitable MF type : MF composition may be derived from requested retention behaviour : practical : „unscientific", we approach problem Jrom the end" : for each case we need to do it again : out of sample physico-chemical properties we derive retention properties : scientifically correct, but uneasy 4 ŕ-\ algorithm of separation system choice sample organic substances trace metals metal chelates Mw > 2000 Mw < 2000 I IEC I RPLC sol. in org. sol. in aq. non-ionic ion. pairing cationic anionic i IEC SEC RPLC SEC sol. in org. sol. in aq. non-ionic ion. pairing ionic 1 cationic anionic I IEC less pol. LSC non-polar RPLC (-\ choice of separation conditions optimisation of separation conditions in dependence on demands aim: tRj = min; Rfy = max; ny = max; dcy / dt = max (A,B) ■B A (A,B) ^B ~ kA +kB +2 elution ratio separation ratio means: separation conditions D=f(T,u,corg,pH,l,cpufr,atpJ dependent variables (D): retention times ^> resolution, peak no., asymmetry independent variables: buffer concentration, ion-pairing agent concentration, pH, % of organic component, temperature, gradient profile... 6 f-\ we study the dependence of dependent variables on independent possibilities: modelling hard m.: based on exact physico-chemical models D=f(TlulcoiglpHIlIcbufferletc.) soft m.\ based on approximation of real function :: substitution to hard model hyper-flat of approximated function : relation between retention and separation conditions tools and process of optimisation single-criterial (semi-hard) evaluation of separation quality criteria characterising by single value the level of separation of all sample components chromatographic response function (CRF) CRF=nin|^ + (tR,min-AtR,i) i=1 v. R,min j chromatographic optimisation function (COF) COF =£a-lni + p(tD~~„ -t R R,max R,posl ) mm COF = £ R, + N° + (3(tRimax - tR]P0Sl)+ Y • (tR,prv - tRmin) 8 separation factor (S) s= 1=1 ^R.max ^R,min resolution product (RP) riR. RP = iRri) n-1 normalised retention difference (NRD) ( n-1 NRD =]J i=1 1 n-1 1 i=i J (■-\ multi-criterial evaluation of separation quality V_/ single variable approach (SVA) studies change of dependent variables while gradually changing one independent variable and keeping all other constant method: relaxation method !! omits possible relations between independent variables multiple variable approach (MVA) studies change of dependent variables while gradually changing more independent variable method : partial least square (PLS) : artificial neural network (ANN) in combination with experimental design (ED) 10 c-\ experimental design an experiment planning in a way, so that out of minimum of points we get maximum information and thus the best description of function of multi-variable function factorial design full factorial experimental design, FED : contains all possible combinations of chosen factors parameters: number of factors and levels for each factor : number of factors (f) responds to number of input variables (f components) : number of levels (L) is number of values per each input variable (L measured concentrations) : number of points of factorial design (total number of experiments n) n = Lf 11 three-level two-factorial design (/_ = 3); 32 experiments CM n .5 (0 > CM .5 (0 > variable 1 variable 1 CM > 7i 7 variable 1 two-level two-factorial design (/_ = 2) simplest; 22 experiments two-level three-factorial design (/_ = 3) 23 experiments fractional factorial experimental design (FrED) : reduces number of experiments of FED (sometimes to complex and laborious) : still describes influence of each parameter and checks possible interactions : proper in cases with expensive and time-consuming experiments 12 star design other variant of experimental design : it may be FrED variant of factorial design :: three-level two-factorial design => two-factor star design : contains (2xf+1) experiments, where f is number of factors (components) : positioning of star design points is given by position of central point : other points are located symmetrically around the centre 0 variable 1 variable 1 two-factorial star design 2xf+1 experiments three-factorial star design 2xf+1 experiments 13 c-\ central and non-central composite designs combination of factorial and star experimental design - complex hyper-flat central composite designs - centres of both plans are equal non-central composite designs - centres are not equal variable 1 five-level three-factorial central composite design 2f + 2xf+1 experiments 14 t-\ approximative methods and algorithms V_J optimisation - effort to „uncover" the numerical function of dependence of output on optimised parameters - approximation black box : algorithms do not describe the physico-chemical properties, but „only" numerically describe the dependencies between variables partial least squares (PLS) MVA, values from all components of analysed mixture are calculated at once canonical correlation (CC) 15 artificial neural networks (ANN) : mimics biologic system of mutually connected neurons processors - neurons the way of connection - network topology o c ■o c ST *< 0 hidden layers neurons are arranged in layers outputs of /7th layer are directed into each neuron in layer n + 1 first, input layer : inputs values for processing last, output layer : values are responses of whole ANN on changes of conditions of input parameters numbers of neurons in input and outputs layers are given by numbers of input and output variables inner, hidden layers : number depends on approximated function complexity 16 connection between neurons RMS 2 - j ♦ ♦ 2 4 6 hidden neurons number represented by rational number - connection weight (w) learning of prediction of output values with minimal deviation of these predicted values by ANN from values experimental - by repeated setting of numerical inputs of transformation function and watching the outputs on real value deviation - total sum of squares (TSS) sum of squares of differences of predicted and input values TSS=X(zi-OUTi)2 _|=1_ Zj - value of output variable z for given triad (x, y z), OUT-x (output) - its predicted value, n - number of elements of training set each neuron (except input) sums values from preceding layer and multiplies them with connection weight w: NETj =^(INPi-wi) + BIASi WPj - input value, wt - weight value and BIASt - value of bias, which is so-called bias parameter and is essential for correct setup of neuron value NET^ and for whole performance of network A/ETj - neuron j in neural network OUTt - transformation of sum value NET-t (output) OU^ =1/(1 + e"NETj) set training/learning - X parameter sets defined by experimental design testing - at least 3 parameter sets inside boundaries given by ED verification - at least 3 parameter sets inside boundaries given by ED (including boundaries) 18 gas chromatography : extraction G-L : extraction G-S : mobile phase (MF, gas) : stacionary phase (SF, liquid, solid, thin layer of liquid on carrier) 1941 GC history Synge and Martin : theoretic principles of GC: „...very refined separations of volatile substances should be possible in a column in which permanent gas is made to flow over gel impregnated with a non-volatile solvent......" 1952 James and Martin : practical introduction of GC; separation of volatile fatty a. 1963 GC-MS - first hyphenated technique 1980 capillary columns in GC - distinctive separation improvement 19 principal differences between GC and LC RaouIt's law o Pa =Pa'X a gas is compressible (liquid not) xA - molar ratio of A in mixture pA - pressure of saturated vapours of A molar ratio low concentrations of A, non-ideal solution kH - Henry's constant pA - partial pressure of A over mixture relation between Raoult's and Henry's laws 20 Langmuir isotherm Pa distribution constant is strongly dependent of vapour puressure and volatility of analyte V_) adsorption GC GSC distribution GC GLC distribution chromatography (GLC) vapour tension of analyte (A) over liquid phase adsorption chromatography (GSC) different adsorption of molecule A onto SF surface with active centres adsorption (distribution) GC S-SF, M-MF 22 c-\ linear flow rate of carrier gas (MF) V_) Pi X Px Po U; Uv U„ L - column length p - gas pressure u - linear flow rate indices: i - on inlet x - in point x of length o - on outlet 4 3 2 1 Pi = 0.5 MPa p0 = 0.1 MPa 0 0.0 0.2 0.4 0.6 0.8 1.0 x/L pressure gradient profile on column 0.0 0.2 0.4 0.6 0.8 1.0 x/L value profile of linear flow rate 23 average linear MF flow rate (J = Bo-(Pi-Po) B0 - specific permeability of column [m2] (Pi"Po) ~ pressure gradient [Pa] H - dynamic viscosity [Pa.s] £ - sorbent inner porosity L - column length [m] compressibility factor TQ - temperature on outlet Tco, - column temperature pw - partial pressure of water at T0 u = J-u0 . 3 i = 2 IPoJ 2 -1 LPoJ 3 -1 f Fm=j-F0 T. col r Po-P w V 24 retention quantities retention volume / time of /-th analyte VR. [ml], tR. [min] void volume / time of column Vm [ml], tm [min] V =F -t vR,i rM LR,i reduced retention volume / time ) VRi [ml], fRj [min] tRj = tRj — t m V =F M vR,i rM LRj K:i = vRJ - vm net retention volume VN [min] VR j corrected to carrier gas compressibility VN=FM-t'.j = V'.j = 273.15 VN specific volume Vh [ml/g] or VD [ml/m2] SI VN related to 1 g or 1 m2 SF and to 0 °C = 273.15 VN h wL-Th 25 f-\ temperature influence "l"col ^ "'"boil a "^inj — "'"col a "^"clet ^ "'"col Tinj - injection head temperature TCO| - column thermostat temperature Tdet - detector temperature t TCO| leads to faster analysis t TCO| demands t MF pressure on column inlet for keeping u through column isothermic analysis: Tcd = const. analysis with temperature gradient: TC0|2 - TcoM > 0 26 GC arrangement MF container output 27 MF delivery gas pressure containers compressor electrolyser 0.5 ml/min - 400 ml/min (HP-GC 1200 ml/min) pressure up to 400 kPa (HP-GC 1 MPa) pressure and flow control thermostating carrier gas advanced flow control (AFC) carrier gas advanced pressure control (APC) 1.0 E E 0.5 0.0 20 helium 40 hydrogen 60 80 u [cm/s] 28 f-\ carrier gas N2 (nitrogen) + cheap, safe - low thermal conductivity H2 (hydrogen) + high thermal conductivity, low viscosity - high diffusivity, explosive He (helium) + combines advantages of N2 and H2 - expensive Ar (argon) especially for ECD must be chemically inert - always necessary to remove humidity and 02 purity - pre-set guard column with molecular sieve 29 injection device loading of A onto column : more difficult than by LC tubu|ar co|umns: 1 _ 20 M| capillary columns: ~ 1 nl inject small volume and quickly : slowly and large volume (overload) =^> broad zones and resolution loss sample evaporation necessity to transform L and S samples into G state : without changing the nature of sample heated space on the beginning of the column volatility increment chemical derivatisation: silylation (N,0-bis(trimethylsilyl)acetamide) -} silanisation (dimethylchlorsilane) O-Si i \ 2 roh ♦ -J'XJ — 2rc4A ♦ -<° and acetylation (acetanhydride) \ J NHo splitless injection proportion valve 1 carrier gas inlet into column mm • w ■ » * » » • • • ■ pressure ) flow regulator with septum septum waste ..splitter waste proportion valve 2 to detector : with closed valve pressurise using proportion valve 1: flow sensor = 5 ml/min, pressure sensor = 70 kPa : septum flow set to 2 ml/min => slow flow of 3 ml/min onto column : sample introduced into injector and is carried onto column : after certain time without splitting (10 - 40 s /optimum 20 s/, splitless time), which happens after injection, the valve is open and rest of the sample is washed out it demands sample reconcentration : prevents zone broadening cold trapping : first few centimetres of column has negative temperature gradient (~ 250 °C /injection/» 40 °C capture region; ca < 150 °C than TboN) => mobility of components with high TboN is zero => their re-concentration solvent effect : first few centimetres of column has negative temperature gradient (~ 250 °C /injection/» capture region is ca 20 °C bellow solvent TboN) => sample components with low TboN condensate with solvent from the created thin film, the solvent is slowly evaporating => re-concentration of components with low TboN hyphenation of SFE with GC (cold-trapping) SFE-GC interface direct SFE-GC 7\ I splitter 1 • I. extraction cell flow restrictor septum-injection 1 capillary GC column separated analyte ! ; I N 1 ■ I I I 1 1 / separation of supercritical fluid from sample increases quality GC analysis separation by means of cold-trapping 1. Tco,intime (t = 0)<25 °C 2. df >2 urn SF JA 10 4 l5 10 ILL illlilll ll 33 a) w/o utilisation b) w/ utilisation /-\ split injection splitter allows: easy injection of small volume : is related to sharp zone entering onto column and column capacity S =--— S - degree of sample splitting, Fs + FM FM - column flow rate, Fs - splitter flow rate (proportion valve 2) disadvantages. : unsuitable for trace analysis : depends of splitter geometry today the most used way of injection : pressurise using proportion valve 1: flow sensor = 103 ml/min, pressure sensor = 70 kPa : septum flow set to 2 ml/min => slow flow of 3 ml/min onto column : pressure sensor sets proportion valve 2 to 100 ml/min => onto column 1 ml/min => through inlet MF flow quickly, 101 ml/min : sample introduced into injector and according to split equation, part goes onto column, part out to waste 35 on-column injection : injects precise amount : suitable for analytes with high TboN - no evaporation during injection instrumentally demanding - restrict pressure losses within injection overloads column with liquid (1 |jl for 50 cm of column) =^> peak broadening : solution as within splitless injection gas entrance to column is sealed with closed valve pressurise using proportion valve 1: flow sensor = 7 ml/min, pressure sensor = 70 kPa, septum flow set to 2 ml/min sample introduced into injector and carried onto column by flow rate 5 ml/min after certain time without splitting (splitless time), which happens after injection, the valve is open and rest of the sample is washed out step 1 equilibration step 1 pressurising gas injection step 2 HSE syphoning pressure valve step 2 HSE syphoning step 3 injection step 3 injection hyphenation HSE-GC injection in system with pressure equilibrium step 1 step 2 equilibration pressurising t do/z step 3 syphoning and injection L t T (-\ separation column V_J tubular : analytical : preparative capillary : open : filled length: 0.5- 10.0 m diameter: 1 - 6 mm length: 2 - 6 m diameter: > 6 mm length: 10 - 100 m diameter: 0.1 - 0.5 mm length: 0.5-50.0 m diameter: 0.3 - 1.0 mm /--> separation efficiency comparison of different column types V__' GC separation of calamus oil components A - 50 m capillary column B - 4 m tubular column column filling tubular columns cover: glass, steel, copper, polymers carriers modified infusorial earth active centres (silanols and siloxanes) =^> tailing of more polar components suppression - silylation adsorbents : unspecific (activated carbon) : specific (silicagel, alumina, molecular sieves etc.) carrier-fine, solid and inert material (spherical silicagel) serves directly as SF (GSC), or is covered by thin film of liquid phase (GLC) solid SF r non-polar : methylated polysiloxane, squalene, apolane C-87 r h ho c—c—o l h h -Jn h mildly polar : phenylated polysiloxane strongly polar : polysiloxane with CH2-CH2-CN, -CH=CH-CN, Carbowax 20M (based on PEG) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ capillary columns silica surface enlargement by etching polyimide cover => increase of mechanical stability SF universal non-polar silicon phases or immobilised Carbowax 41 wall-coated open tubular columns (WCOT) liquid SF anchored directly on the capillary wall : GLC fused silica open tubular i.d. 100-530 jjm □ (FSOT) thin wall with outer polyimide cover (mechanical stability) : GSC i.d. 320 - 530 [Jin _ film thickness 0.1 -8 jjm \ support-coated open tubular columns (SCOT) carrier is on capillary wall, SF is on it ■ GLC j.d. 320-530 jjm film layer thickness 6-60 jjm porous-layer open tubular columns (PLOT) layer of solid active sorbent on an inner capillary wall : GSC layer thickness 5-50 pm r-\ column thermostat V_) c-\ importance of temperature of GC : evaporation of liquid or solid sample : kinetic aspects of separation kept with precision of 0.1 °C; thermostat range (T,ab + 4 °C) - 450 °C optimal loading temperatures - TboN of component with highest value + 30-50 °C optimal column temperature ~ TboN of analyte column temperature > TboN => tR = 2 - 30 min minimal temperature => better resolution, but higher tR wide range of TboN of separated components => => temperature programme /column gradient (A temperature during experiment) temperature may be increased gradually or in steps 43 detectors detected compound is volatile, in gaseous state concentration dependent detector (ODD) : non-destructive, dilution with carrier gas decreases sensitivity mass dependent detector (MDD) : destructive, carrier gas interferes not, depends on introduction rate into detector ignition spirále flame ionisation detector fid collection electrode +300 V MDD signal: current created by pyrolysis of carbon sample hydrogene column : noise 10-13 : dyn. range 107 : sensitivity 109 M 44 amplifier thermal conductivity detector tcd catharometer : noise 105 : dyn. range 106 : sensitivity 108 M CDD signal: sample molecules change (decrease) thermal conductivity of carrier gas : carrier gas must have high thermal conductivity (He, H2...) : temperature dependent, universal waste electron capture detector ecd : noise 10 12 : dyn. range 105 : sensitivity 10 13 M 63Ni —Miy current measuring CDD signal: analyte molecules decrease current generated by (3-emitter : halides, nitrites, cyano-compounds, peroxides, anhydrides, organometals from column 45 nitrogen phosphorus detector npd - thermoionisation detector : noise 10 12 : dyn. range 106 : sensitivity 10 10 M MDD connections- of heating heating spirále hydrogen 6- U anode ^ rubidium or caesium glass flame air t carrier gas (nitrogen) signal: Rb/Ce glass thermoionisation electron emission enhanced by N or P presence chemoluminiscence detector inlet of fluorine pump outlet data column : noise 10 13 : dyn. range 104 : sensitivity 10 11 M irradiation CDD signal: reaction of F (strong oxidant) with analyte 46 flame photometric detector fpd region of chemoluminiscence waste thermal filter source flame air ......... photomultip. ......... into hydrogen interference filter amplifier heated metal block thermostat wall electrolytic conductivity detector elcd : noise 10-13 : dyn. range 106 : sensitivity 10 11 M : noise 10 12 : dyn. range 107 : sensitivity 10 10 M v_y MDD signal: chemoluminiscence : selective S (394 nm), P (526 nm) MDD signal: appearance of special products their conductivity measurement after mixing with solvent inlet of reactive gas column 47 photoionisation detector pid : noise 10 13 : dyn. range 107 : sensitivity 10 11 M CDD signal: UV-irradiation ionisation thermostat wall column heated ionisation cell lightproof cover waste or secondary detector UV-transparent window cooler mirror atomic emission diode array data microwave „ignition" microwave ^ reactor T column grid atomic emission detector aed /- : noise 10 14 : dyn. range 104 : sensitivity 10 11 M MDD signal: microwave induced plasma : selective according to chosen emission wavelength : very expensive 48 ionisation chamber auxiliary gas discharge chamber collection^ tables hid column waste : noise 10-14 : dyn. range 106 : sensitivity 10 12 M meter MDD signal: auxiliary gas is ionised first (He, Ar), its ions then secondary ionise sample molecules 150 V gas density balance gdb catharometer vapour bar i : noise 108 : dyn. range 103 : sensitivity 108 M waste I _^ MDD signal: pressure difference between upper and lower passage of gas in presence of eluent vapours column outlet catharometer 49 infrared detector ird waste column o w ■a gilt glass tube KBr window o c * CD : noise 10 12 : dyn. range 105 : sensitivity 10 10 M CDD signal: IR absorbance mass spectrometric detector : noise 10 14 : dyn. range 103 : sensitivity 10 15 M CDD signal: ion count universal ionisation: : electron impact (El) : chemical i. (CI) analysers: : quadrupole (Q, Qq) : ion trap (IT) : magnetic sector : time-of-flight (TOF) definition of chromatographic system in GC MF carrier gas type flow / pressure (ml.mirr1 / kPa) injection (Xpl) injection type (event, splitting rate) SF stationary phase type length, inner diameter, manufacturer, SF type, film thickness 25m x 0.32 ID J&W DB-5 DF - 1.0 temperature gradient profile initial temperature and its period, temperature increase; inlet temperature (e.g. 130 °C 1 min, 130 - 250 °C at 5 °C/min, 250 °C 5 min; 250 °C) detector basic characteristic according to type 51 f-^ analytical information in chromatogram V_/ /-\ qualitative information retention time : compound identification (standard method) spectroscopic detectors: UV-Vis spectra MS spectra (ESI / APCI; Qq / IT / o-TOF) NMR spectra (1H, 13C) retention time formulation specific retention volume (VD) = 273.15 vN ST col relative retention time (rA B) : comparison with internal standard Kovats retention indices (rA B) : linear dependence pf retention time logarithm of homologues on carbon number quantitative information peak area « amount, concentration of compound : because of narrow peaks frequently only height internal normalisation method all components are eluted (solvent does not count) all they have same/similar response factor C0/C — A0/„ i — 100-A A tot external standard method (absolute calibration; calibration curve) always same measurement conditions, same injection volumes indispensable matrix influence 'unknown A _ ^uknown 'known 'known internal standard method Aioi A _ ' XIS1 unknown 'unknown : need not to know injection volume : standard must be chemically similar to analyte Aioo A 'known IS2 known standard addition method vs : presumes calibration curve linearity V, A2 (V1+Vs) A1 V, A,, - analyte peak area, unknown concentration c, A2 - analyte peak area of unknown concentration c, after addition of standard of known concentration cs \A, - sample volume, Vs - standard solution volume 54 test measurements in GC column testing efficiency in dependence on time (at const, flow rate) we observe : normalised retention times of components : height of peaks : symmetry of peaks testing mixture for uncoated carriers n-decane, 1-aminoacetate, 3,5-dimethylpyrimidine, n-dodecane, 1-aminodecane, 2,6-dimethyl-aniline, N,N-dicyclohexylamine, 1-aminododecane and n-heptadecane MF - H2, Tinitial = 40 °C, Tterminal = 180 °C testing mixture for coated carriers (Grob test) methyl decanoate, methyl undecanoate, methyl dodecanoate, n-decane, n-undecane, n-dodecane, 1-octanol, nonanal, 2,3-butanediol, 2,6-dimethylaniline, 2,6-dimethyl-phenol, dicyclohexylamine, 2-ethylhexanoic acid MF - H2 or He, Tinitial = 40 °C, Tterminal= 100 °C, resp. 175 °C thermostability column bleeding n-C 22 MF - He, Tinitial = 40 °C, Tterminal = 300 °C 100 200 300 ■col 55 separation of macromolecules 1556 separation of macromolecules history Agricola : separation of gold using gravity in a flow of water 1870 Lord Rayleigh : basic theory on light scattering on small particles 1940 Debye and Zimm; theory on light scattering on large particles 1955 Lindquist and Storgards : gel filtration on starch („molecular sieving") 1959 Porath and Flodin : gel filtration on cross-linked dextrans (Sephadex) (GPC) 1961 Hjerten : use of synthetic gels as stationary phases : polyacrylamide 56 1962 Pedersen : protein separation on small glass spheres (HDC) 1964 Hjerten : use of natural gels as stationary phases : agarose 1966 Giddings : description of FFF method principles 1969 DiMarzio and Guttman : theory of steric exclusion for SEC 1970 v_J first commercial instrument using light scattering for mol. mass characterisation 1974 V_) Small: first HDC experiments on non-porous sorbent 1978 Noel: particle separation in empty capillary (capillary HDC) 57 theoretical fundaments of separation of macromolecules what is that macromolecule? molecule of Mw>10 000 synthetic polymers monomer, oligomer (10- 100), polymer homopolymers (PE, PP, PS, PTFE...) one repeated unit (monomer) nM -> [M]n linear heteropolymers : more of different units branched nX + mY -> XnYm 58 biological polymers M w 10 000-1 000 000 proteins peptidic bond, 21 natural amino acids (Se-Met) complicated complexes of different units, e.g. haem + globin glycans (polysaccharides, oligosaccharides) (starch, glycogen, chitin, cellulose, dextrans, pullulans) nucleic acids (polynucleotides, oligonucleotides) nucleotide = phosphate + nucleoside nucleoside = saccharide + base DNA- saccharide - deoxyribose RNA- saccharide - ribose surface forces (surface charge, ionic strength of surround) primary => secondary, tertiary, ternary structure - native form macroscopic forms 45T r random coil size of macromolecule flexible molecule description of macromolecule rod sphere contour length (L) A2 h h h \ * U L = nl n - number of bonds I - monomer length end-to-end vector length (r,) 60 mean square end-to-end distance (r2) I J radius of gyration (s2) important quantity for light scattering measurement centre of gravity s2................:£ /f\s6 s2\ = ^ n s - distance of unit from centre of gravity s2 = if monomer units are identical relative molecular mass SM separates mostly according to size = f (molecular mass, cross section, etc) 1 Mr = m- — m(12C) 12 v 7 SI definition for macromolecules mix of molecules of different molecular mass, differing in number of units = distribution number average Mr : measured by osmometry weight average Mr : measured by light scattering z-average Mr : measured by sedimentation analysis polydispersity - distribution 62 /-\ example 8 v.__/ what will be the number average, weight average molecular mass and polydispersity of polymer sample? basic modes of macromolecule separation V_ size exclusion chromatography (SEC) gel filtration chromatography (GFC) gel permeation chromatography (GPC) gel filtration (GF) hydrodynamic chromatography (HC) flow-field fractionation (FFF) sedimentation (SFFF) thermal (TFFF) electric (EFFF) gravity (FFFF) (-\ SEC, size exclusion chromatography gel permeation chromatography (GPC) gel filtration chromatography (GFC) c-\ principle : analyte is distributed between MF outside of particles and inside of particles :: sieving effect, steric exclusion :: diffusion :: pressure of carrier liquid - motion of liquid and its flow profile VR=V0ut+k^-Vin VR - retention volume K'D - distribution constant tot- total volume out - MF outside of particles in - MF inside of particles part - volume of particle material v, v, (= v t + v + v „ tot out in part vR = vout + k;v • (vtot - vout) where (V-v t) = v +v rf V v tot v out / v in 1 v part K'AV - elution constant K^v/K'd = const. flow O o r (0 Q. pore analyte molecule time molecular sieve effect : uniform pore diameter (determines cut-off) : distribution of pores with different diameter thermodynamic interpretation AG = AH-TAS = -RTIn(K) => K = g ah-tas rt as r <1 AH ~ 0 => process is entropically controlled 66 D rvin ^out 'A cin - analyte concentration inside of particles cout_ analyte concentration outside of particles VR =k1logMw+k kv k2- numeric constants 'max VR=Vout+ jKi(R,r).q>(r)dr R cp - total pore volume with diameter rto r+dr R - diameter of retained particle separation is given by ratio of diameter of pore and analyte sieve model is in many aspects not exact : flow of liquid out an in pores is different (Fout » Fin) : other interactions: adsorption, L-L distribution, electrostatic repulsion ( K'D>1) 67 gel LC mechanical separation of A molecules in particles/pores of gel based on their different size not classic LC, no chemical affinity K = ~~ D M qS-quazi SF, M-MF 68 use of SEC group separation : separation of low and high molecular groups (desalting, extraction agent removal, reaction termination between low molecular ligand and biopolymer) fractionation / purification separation of components with significant Mr difference determination of Mr comparison with standards (in line increasing Mw) polymer polydispersity and distribution analysis of ligand-biopolymer binding emerging complex has higher Mr than components (complex insulin-antibody by diabetics) concentrating samples of biopolymers dry molecular sieves remove solvent - „dry up" and concentrate sample column filling proceeding SEC : pre-filled columns : own filling - SF swelling (uniform, without bubbles) sample introduction >._✓ : injecting 1 - 5 % of column volume : either on column top or through injection adaptor elution MF not directly influences separation : solvent viscosity and elution MF ratio < 2 : water - uncharged compounds separation, or buffers pH and / keeps ion interactions minimal c-\ guarding SF v_) 0.02 % sodium azide 0.05 % trichlorobutanol (Chloreton) 0.005 % ethylmercurythiosalicylate (Mertiolate) 0.002 % Chlorhexidine ideal A . . ,, limited overload matrix influence sample wrong injection solubility calibration set of standards 4-5 defined native proteins with increasing Mw oxTGM A 1 s 1 (monostearin) AlogM Vc 71 absolute calibration basic parameter defining selectivity - hydrodynamic volume formula for limiting viscosity number of polymer [r|] derived from Einstein's equation [n]=lim(nZOzi = k_K p^o p M [n] = K-Ma =: Mark-Houwink's equation n •M = k-V independent on macromolecule structure lnJA-MA = lnJs-Ms=f(vR)| A - analyte, S - standard o vc log(n -M) = f(VR) KA • M^A+1 = Kg • MgS a«+1 Ma = Ke •MC,S+1 VA+1 ^ s K J til_ by viscosimetry 72 selectivity in relation to pore size distribution ft* increasing pore size distribution AlogMw AlogMw AlogM • I-\-1-i-► 1-(—f vR vR VR IA AA lAflft Iflft A vR vR VR f-\ separation column v_/ : classical tubular columns material - mostly soft gels : inert gel matrix (towards analyte and elution solutions) : long-term chemical stability (at different pH and temperature) : mechanical stability (resistance towards high pressure) : small amount of ionised groups : suitable particle size (5 - 250 urn) :: small particles - high resolution, low rate :: large particles - fast separation, low resolution fractionation range (FR) Mr range, in which the compounds are separated elimination limit (EL) upper limit of fractionation range 74 column fillings agarose large pores, acidic character elution: polar and non-polar solvents FR > 200 000 Sepharose Polyacrylamide low amount of polar groups; low resolution elution: polar and mild non-polar solvents FR = 1000-3 000 000 Sephacryl, Bio-Gel P mixed SF: agarose-acrylamide chemical very resistant FR = 1000-23 000 000 Bio-Gel A, Ultrogel c-\ d extra n strong adsorption effects elution: polar and non-polar solvents FR< 10 000 Sephadex f-\ styrene-DVB strong hydrophobic interactions elution: non-polar solvents FR = 400-14 000 Bio-Beads, Styragel f-\ methacrylate hydroxy methyl metacrylate + ethylendimethyl methacrylate elution: polar and non-polar solvents Spheron vinylacetate Merckogel OP-PVA silica glycomethacrylate elution: polar and non-polar solvents Separon strong hydrophilic interactions, mildly acidic elution: polar solvents Bio-Glass, Porasil, Spherosil detectors detection of separated compounds determining molecular mass and polydispersity absorption photometric detector : polymers mostly do not contain own chromophores =^> indirect detection refractometric detector : universal fluorimetric (fluorescence) detector own fluorophores (within proteins Trp, Tyr, Phe), or derivatisation 77 viscosimetric detector MDD signal: pressure unbalance in bridge Mve(Mn, Mw), MV*MW [n]-KMa^|im(n/rlsQlv)"1 p^o p Mark-Houwink's equation eluate - - - I IP inlet pressure waste different n of solutions inC1,C2,C4&C3 =>AP [n]= 4AP Pip - 2 • AP reservoir; pure solvent [r|] - limiting viscosity number [m3/kg] H* - solvent viscosity K, a - Mark-Houwink's constants (for globular macromolecules a = 0) osmometric detector vapour pressure osmometry (VPO) uses Raoult's law fast, low sample consumption, temperature interval 25-130 °C Mr = 40 - 35 000, no volatile compounds 78 T = const., saturated vapours of solvent 1) RT1 and RT2 - droplet of solvent, AT1i2 = 0, U = 0 2) RT1 - droplet of solvent, RT2 - droplet of sample (solvent + analyte) source adding droplet of sample 4< solvent vapour tension => condensation of solvent vapours into the droplet => release of condensation heat => t temperature of sample droplet, thus also of thermistor, also of solution tension pressure => Wheatstone bridge unweighing solvent vapour condensation stops when sample vapour pressure is in equilibrium with pure solvent vapour pressure due to higher temperature measured voltage, proportional to the difference of temperatures of both thermistors, is proportional to molar concentration of compound in sample thermal losses => calibration on standard of known Mr value 79 scattering of light beam on particles of suspension or colloid solution interaction of light beam electric vector with electron shell => periodic oscillations intensity, polarisation and angular distribution of scattered light depends on size and shape of scattering particles dynamic light scattering studies time fluctuations of scattered light on moving particles : information on diffusion coefficient solution (■-\ light scattering on small particles v_) /-\ macromolecules particle diameter (d) < A/20 (Rayleigh scattering) c - concentration N - number of particles; scattering centres n~0 - refractive index of solvent (dn/dc)u - particle refractive index changes at constant u => particles - secondary source of scattered light of the same wavelength L 8tt2 • Va2 Kl l 2n\ =--2—o--N • 1 + cos 0 l0 Ao-r2 V 7 intensity ratio of scattered (is) and original light l0 (non-polarised) V- unit volume A0- wavelength r - distance from particle 9 - angle measured from main light beam a = c(dn/dc\ - no 2tt • N 81 number of scattering centres N in case of identical macromolecules (monodisperse sample) m _ ^ ' Na Na - Avogadro's number iwi M - molecular mass s _ 0 2n2-n02(an/ac)2-VcM A40-r2-NA (1 + cos29) l0V(1 + cos29) Rayleigh radius K = 2TT2-n02-(an/ac)2 A40 • NA summing constants into one, K K-c _ 1 in polydisperse sample, M is substituted Re ~ M ,vlw V1 ~ 82 inter-molecular interactions and non-zero concentrations taken in account (Debye): K'C =- + 2A„c + 3A,c2+... R e M A2, A3... - virial coefficients; mostly A3 and higher are omitted A2 - phys.-chem. measure of thermodynamic solvent quality for given macromolecules good solvent A2 > 0 : macromolecule expands bad solvent A2< 0 : macromolecule shrinks 6-solvent A2= 0 : macromolecule preserves its volume 83 light scattering on large particles macromolecules particle diameter (d) > A/20 (Debye scattering) : large particles => phase shift of light scattering from different parts of molecules : phase difference is dependent on angle 9; for 9 = 0 is the difference 0 : beam interference => angular distribution of scattered light intensity P(9) P(6) = I. s(9=0) P(9) = 1- 16tt2 s2 3A sin' 0 Zimm's equation use of P(0) parameter to express scattering Kc R 9 1 P(8) M + 2A„ ■ c if(1-x)-1 = (1+x) Kc R, k0 2 Ljl 1 + 16ttz s 3A sin' 0 _1_ M + 2A2 c 84 experimental bases for calculation of gyration radius multiple angle laser light scattering (MALLS) laser- A0 source of l0 intensity refractometer (also as concentration detector) - n0 and (dn/dc)u (see constant K) is- scattered light intensity in different angles 9 in known distance r from cuvette 9 -» 0 (c = const.) blue lines, from blue slope we extract gyration radius (s2) c -» 0, slope ~ A2, interception 1/MW red line low angle laser light scattering (LALLS) at small angles 0 (< 7 °) sin2(8/2) - 0 => P(0) -> 1 KC 1 OA then ■ = -_: + 2A„C R e M for Mw> 107 or within associated systems this approximation fails instrumentation laser optics adjustable angle advantage: : absolute technique, no calibration needed Mw, A2 for - standards necessary : fast : connectible with separation technique (GPC, FFF) disadvantages: : dust- demanding high solution purity 86 HC, hydrodynamic chromatography principle : combination of steric exclusion with surface (colloid) interaction sample-filling, eventually solute retardation behind streamlines of laminar flow with profile (wall effect) sample moves with MF flow non-porous material gravity centre of large macromolecule cannot reach the channel wall (Rp) move in slower flow near to it (wall effect; given by laminar flow profile R0) => heavier (larger) molecules run through channel faster than smaller ones cannot other influences: : electric double-layer : van der Waals interactions © ©0 © w © © 0 0) 3 3 c 0) o CD => sample moves in channel hydrodynamically or electrically 87 t - polymer retention factor tj a tM - retention time of polymer and unretained component A - ratio between macromolecule radius and flow channel half-height B and C - constants dependent on channel symmetry, C also on retention model 0.001 1-1-1-'-1-'-1-'-1 0 6 0.7 0.8 0.9 1.0 A = f (t) and thus on Mw in case of tubular micro-capillary use and C —> 2.3 /-\ porous material pores of filling : 50 - 50 000 nm sample : larger molecules c-\ capillary fractionation v_J (CHDF, capillary hydrodynamic fractionation) other influences in account: : colloidal forces : non-linear inertial forces depending of flow-rate gradient and positi (lift forces; tubular pinch effect) pump exchangeable resistance capillary waste injection pi waste detector HDC capillary 1966-J. C. Giddings FFF, flow-field fractionation principle : physical field inflicts some property of analyte and creates concentration gradient dc/dx => concentration profile c(x) across channel is specific for given analyte inlet Jfield flow separation channel outlet field flow layer JS thickness (I) i ,p o o 0) 3 3 : no SF (one-phase chromatography) => no interactions with active surface : MF is carrier liquid, influences separation indirectly only : variables influencing separation may be changed continuously in wide range separation of macromolecules and particles 103 - 1015 Da FFF proceeding instrumentation f-\ pumps : wide range of adjustable flow-rates : no need for high pressure, but for pulseless flow !!! : with constant pressure and flow (reciprocal, peristaltic) f-\ injection device similar to LC : septum : multi-way valve : linear injectors (infusion) detectors similar to SEC : refractometer : photometer-absorption, fluorescence, optical rotation, scattering : other - viscosimeter, densitometer, osmometer... r 1 SdFFF, sedimentation flow-field fractionation v_J the oldest technique effective force = natural gravity or centrifugal force rotation 20000 r.p.m. (injection in steady state) G - gravity (g) or centrifugal acceleration Aq - density difference between particles and solvent dp - particle diameter GFFF: > 1 |jm SdFFF (G = 105* g) : 106 Da or > 10 nm DNA, proteoglycans, river water colloids, viruses and silicagel SF for HPLC zone 2 zone 1 94 ThFFF, thermal flow-field fractionation separation channel - two metallic (cupric) blocks the upper one is electrically heated, the lower one is water cooled => gradient 20 - 1000 °C/cm : distance teflon foil: 50 - 250 |jm temperature gradient causes slower flow at colder wall (non-isoviscose liquid) f a -1 A = w • • - I T dx j DT = thermal diffusion coefficient a - thermal diffusion factor = DT ■ T / D TFFF: to describe thermal diffusion 95 EFFF, electric flow-field fractionation walls - semipermeable cellulose membranes high voltage gradient; low absolute voltage - low current => low heating |je - electrophoretic mobility E - electric field intensity EFFF: proteins with different isoelectric point FFFF, flow-field flow fractionation external field - solvent flow orthogonal to flow of basic media tube of semipermeable material => solvent intrusion, not of analyte V0 - channel volume H - viscosity Vc - volumetric orthogonal flow d - effective Stokes diameter FFFF:> 1 nm 96 r->| electromig ration methods (■-\ basic principles of electromigration methods v_) driving force - electric field : charged particle motion in electric field : extraction L-S : electrolyte (liquid able to conduct current) : separation channel wall (carries charge) : stationary phase (SF, solid matter, micelles) mobility of ions is influenced by charge, molecule size and surrounding ions /-\ basic electromigration arrangement column arrangement (in tube, in capillary) slab arrangement (in gel) 97 1808-93 electromigration methods history first experiments in U-tubes - F. von Reuss (1808), G. Wiedeman (1856), H. Buff (1858), O. Lodge (1886), W. Whetham (1893) 1897 Kohlrausch - basic equation for ion migration in electrolyte solution 30. léta Tiselius - gel elfo with glucose as medium 1937 Tiselius - first fully functional electrophoresis instrument, 1948 Nobel price 1955 Smithies - use of starch gels for elfo 1958 Hjerten - ZE in rotating tubes 1 - 3 mm 98 1959 Raymond and Winstraub - acrylamide gels, setting up gel porosity & stability 1965 Tiselius - ZE in 3 mm tubes 1967 Hjerten - elfo in tube, i.d. 1-3 mm, with inner coating against EOF 1969 Vesterberg and Svensson - IEF of proteins in ampholytes 1970 Laemmli - denaturing separation in gel, SDS and concentration gel use Everaerts - ITP on own instrument 1974 Pretorius - EOF as a MF driving force through sorbent 1974 -79 Virtanen, and Mikkers et al. - glass and teflon capillaries, i.d. 200 pm 99 1975 O'Farrell - 2D GE, presetting IEF in gel to SDS elfo 1981 Jorgenson and Lucas - borosilicate glass capillary, i.d. 75 pm 1983 Hjerten - CGE for biological samples 1984 Terabe - micellar electrokinetic chromatography /-\ 1985 Hjerten - CIEF for biological sample 1987 Karger and Cohen - high efficiency CGE for NA Knox and Grant - CEC in 50 |jm capillaries with ODS 1988 Beckmann Instruments - first commercial instrument 100 theoretical fundaments of electromigration methods separation in external field motion of free charged particle in electric field : charge and field orientation decided on direction and velocity v = [J • E = [J U v- ion motion velocity u - electrophoretic mobility [m2V_1 s_1] E - electric field intensity U - voltage I - length of voltage gradient influencing the motion by ionic atmosphere => => decrease of velocity with increase of electrolyte concentration u0 ionic (net) mobility - u at zero ionic strength 109 m2V_1 s"1 = 1 tiselius (Ti), sign implies ion polarity (anion has negative u) temperature influence: ŤT => fu0; with 1 °C about 2 % MT=MTo-[1 + 0.02.(T-T0)] T - working temperature T0 - standard, tabulated temperature 101 ion mobility estimation in a case, when value is not known (tabulated) Stokes mobility + + + + friction accelerating force force <-> ® a = 0 FE=FF E _ q-E FF 6tt • n • r * v 6tt • q • r • p M = 6nr| r a - acceleration of spherical charged particle motion q - charge H - solution viscosity r- ion radius v- ion motion velocity relation of ion mobility and diffusion coefficient z - relative charge F - Faraday constant R - gas constant T - temperature D - diffusion coefficient ion mobility estimation for small molecules Jokl equation a Mo — z • r- - b M - molecular mass a, b - empiric constants a~485x10-9 m-2 V"1 s1 b~9.6x109 m-2 V"1 s1 estimation error is ca 10 % /-\ actual ion mobility V_) Onsager equation M — Mo (0.23- M0z+Z- + 31.3-10 9 • Z+/- ' 1 + -v1 z+, z_- relative ion and counter-ion charge I - ionic strength effective mobility mobility of weak bases, acids or zwitterions resulting mobility of all ion forms n X; |jj - mobility of one ion form X: - its molar ratio i=1 free mobility mobility extrapolated to zero gel concentration migration time fully charged // Q = 1 pKai ' ' PKa2 a = 0.5 a = o// uncharged entry useful for mobility calculation ltot - separation channel total length leff - separation channel effective length tm- migration time t0- migration of neutral particle (EOF) M = I I 'tot 'eff u 1 1 —-—) t t 0 pH 12 Mtot — Meff M eff eff 'EOF tME I I 'eff 'tot tMU 104 wall is charged negatively - until said others electroosmotic flow (EOF) capillary = endo-osmotic pump capillary made of fused silica with exposed hydroxyl groups dissociation of hydroxylgroups leaves a negative charge on the inner wall + switching voltage on, liquid starts to move to cathode - it is mobilised by endoosmotic flow ! 105 : cations migrate towards cathode and carry solvent molecules in the same direction -electroosmotic flow : neutral molecules are moving in the same direction as electroosmotic flow with negligible mutual separation : anions are slowed on their way towards anode, electroosmotic flow is stronger than their electrophoretic mobility => they proceed towards cathode too EOF = 0 => no mass flow, only ion exchange 106 £ - dielectric constant 5 - zeta potential (electrostatic), appears as a consequence of charge on capillary wall H - viscosity 107 f-* influencing the EOF high EOF - electrolyte carries cationic analytes out before reaching separation low EOF - adsorption of cationic analytes some EMM modes demand EOF suppression (IEF, ITF, GE) what influences EOF? : surface wall charge : electrolyte viscosity : electric field intensity influence of voltage : change of EOF is directly proportional : low voltage => low efficiency of separation and resolution : high voltage => high Joule heat influence of ionic strength or background electrolyte concentration increasing value lowers ^-potential and thus EOF : high values increase current and thus Joule heat : high values may cause analyte salting-out and adsorption to wall : low values supports adsorption to wall and limits sample concentration : changes peak shape, if electrolyte conductivity differs much from analyte influence of organic solvent addition : decreases ^-potential and viscosity Meof :: may change selectivity, gathered only empirically 10 2.5 3.0 3.5 4.0 8 o 7 - influence of tensides : changes ^-potential, may change wall polarity; anionic tenside increases EOF, cationic decreases (if wall if negatively charged) 3 4 - 5 -2 2 109 influence of background electrolyte pH : directly proportional EOF change; low pH => low EOF, high pH :: may change charge or structure of analyte high EOF influence of temperature : changes viscosity, higher temperature => higher EOF :: thermolability of some samples influence of covalent wall surface modification : changes ^-potential and wall charge polarity influence of neutral hydrophilic polymers M eof 4-3 -2 -1 - pyrex glass teflon i i i i i r 3 4 5 6 7 8 pH pH influence on EOF : changes ^-potential (decrease) and viscosity (increase), decrease EOF by charge shielding EOF measuring outlet detector N1 | inlet + N1 B.A. Williams, G. Vigh, Anal. Chem., 68, (1996) 1174-1180 : first EOF marker injection : shifting the marker zone to detector by pressure : second EOF marker injection : shifting both marker zones to detector : voltage application - electrophoretic mobilisation N1 i N2 r N1 N2 UL N1 N2 o N1 N2 J_l N3 third EOF marker injection and consequent application of pressure - shifting all marker zones to detector 2 min 25 111 lEOF=(t3-2t2+t1)- 'eff to + t inj M = I I 'EOF 'tot u-(tM-^-^) 'eof ~ length, which marker travels during electrophoresis t,, t2, t3 - migration times of zone N2, N3 tinj- time period of marker injection by pressure leff - effective capillary length ltot - total capillary length U - applied voltage tm - time period of electrophoretic shifting tru and trd - time periods, for which the voltage (inc-/dec-)reases linearly to given value common EOF calculation Mtot — Meff + MfOF — eff I I _ 'eff 'tot tME tM-U graphical illustration of separation maximum function lsign = f (t) electrophoretic peak : also Gaussian shape as in chromatography electrophoretogram : electropherogram, electrophoregram, electrophoreogram migration time of #-th analyte tM [min] separation efficiency zones of A broaden during separation and become asymmetric reasons behind zone broadening : lateral diffusion : electrodispersion na 6 ' N = 'eff height equivalent of theoretical plate (H) (comparison of separation channels of different length) H = A + - + C-U u A = 0 : in absence of particles C = 0 : is there is no SF analogically as in chromatography increasing voltage causes increasing of flow rate, but it also releases heat and it increases rate of lateral diffusion 114 under ideal conditions (short injection length, no sorption, ...) the only influencing is diffusion (zone broadening) B _ 2D _ 2D _ 2DL u u [i-E |J-U principal difference from N in chromatography contributions to zone broadening in electromigration methods 2 2 2 2 2 2 2 ^ = ^dif + ^el.disp + ^"inj + ^heat + ^"sorp + ^det + ■■■ a^=2D-t D - diffusion coefficient t-time /-\ diffusion influence v_/ basic factor analytes with low D create sharp zones /-\ detection cell length influence should be smaller than length / width of analyte zone => better peak depicture 116 electromigration dispersion influence influences peak shape difference between conductivity of sample and electrolyte leads to : peak tailing : focusing (low sample conductivity), broadening (high sample conductivity) : ITF effect (peak fronting) because of certain ion surplus (e.g. Ch) + — © © © © i © © © © time Ms > Mbge => front gets broad and tail focuses time Ms < Mbge => front focuses and tail gets broad time Ms = Mbge => sharp zone 117 sorption influence sorption causes peak tailing n2 = k' * VEOF * Uff ads (j, , ,\2 ÍJ2. (1 + wj r-w 2 + V 4D K _ ^M,ret ^M,unret t M.unret k" - capacity factor tM ret - retained analyte migration time Kd - first order dissociation constant tM unret - unretained analyte migration time sorption could be prevented by capillary inner coating : serves to change also other system properties (reverts EOF...) injection length influence : injection length must be shorter than diffusion controlled zone width : low sensitivity demands often longer injections tinj - injection pulse length 118 Joule heat influence leads to temperature gradient and laminar flow 1 •In K sil o.d.sil + 1 r •In V "i.d.sil J K polyim o.d.polyim + 1 V 'o.d.sil J o.d.polyim h Q - output r - radius c O max. separation voltage k - thermal conductivity h - heat transfer rate off capillary 390 375 375 390 voltage d [(j m] : decreasing voltage : decreasing generated heat, low sensitivity and resolution : lowering capillary i. d. : current decrease with i. d. square, low sensitivity, adsorption! : decreasing BGE concentration : decreasing current, increasing adsorption : thermostating : draining heat resolution R _ 2• (tM,A -tMB) _ 2-At M (A,B) wA+wB wA+wB R n Ali (A,B) 4 M Au - difference, (u2 - Mi) u - median, (u2+ Mi) 12 D-(M + Meof) f-\ electromigration methods arrangement v_) li i , _K . K il ■ III II HH output voltage source 121 hydrostatic siphon effect II 1 A rozdil 1 hladin injection device typical volumes: 10 - 100 nl (capillary ~ 1 - 2 |j|) normal - longer part before detector reverse (short-end) - the other end hydrodynamic APd4 TT • t \j _ mj inj" 128-n-i tot injected volume V| inj AP - pressure difference d - capillary i. d. tinj - time length of injection 'tot- to*8' capillary length H - background electrolyte viscosity pressure LM pressure —>■ 122 electrokinetic for CGE the only possible : non-quantitative - more mobile ions go easier stacking effect sample conductivity < electrolyte conductivity => sample ions carry the current => stacking/concentration on inter-phase sample-electrolyte ©"e electrolyte weak field high conductivity strong field low conductivity sample 6— electrolyte m 5)weak field J) high conductivity Vinj — TT 'r -left t • u t U EOF ^sep injected volume Vj inj Uinj - injection voltage Usep - separation voltage r- capillary i. d. Ieff - capillary effective length tinj - injection time length tE0F - EOF marker migration time 123 voltage source typical range: 0-30 kV; recommended gradient 400 V/cm 0 - 300 mA too high voltage decreases analysis time, lead to discharges (ca 20 - 25 kV) ZE - constant voltage, ITF - constant current one electrode always grounded - that one closer to detector separation channel anode electrolyte cathode tube inter-phase sinks sample solution the oldest (proposed 1892, done 1930) glass U-tube electrophoresis in free solution : separation detection by moving inter-phase observation : coloured solution and clean electrolyte solution 124 capillary fused silica i. d. 10-200 Mm o. d. 350-400 Mm length 10 (CGE) - 100 cm; 50 - 75 cm most common outer coating - polyimide (mechanical properties) conditioning: establishing the properties of capillary inner surface surface cleaning: 1 M NaOH, 0.1 M HCI, BGE other: strong acids, organics (DMSO), detergents teflon reproducible EOF worse heat conductivity other materials based on Si02 - glass (Pyrex) 12-M 10-8 6 4 2 pH hysteresis decreasing pH increasing pH 2 3 4 t-1-1-r 5 6 7 8 pH 125 suppressing EOF, in range pH 4 - 5 relatively low (~ 0), pH 6 - 7 slowly increases at high pH is almost about 4/5 lower than in un-coated silica capillary Si-O-Si-R polyacrylamide-, arylpentafluoro-, 3-glycidoxypropyltrimethoxy-siloxan protein or amino acid, sulphonic acids, maltose, PEG, polyvinylpyrrolidon : relatively easy preparation : limited long-term stability Si-C v_J polyacrylamide using Grignard reaction : stabile between pH 2 - 10 : difficult to prepare r-\ SF from GC and LC v_) C2-18, PEG, phenylmethylsilicon : easy to hydrolyse : increased adsorption adsorbates cellulose, polyethylene glycol, polyvinyl alcohol, polyethylene imine : only short-term stability in acidic range pH 2 - 4 (PEG, PVA) : stabile in neutral pH (PEI) : relatively hydrophobic : reverts EOF (PEI) dynamic coating part of BGE, stems in the praxis of adsorbates use pH extremes reduction of coulombic interactions : pH range 2- 12 : EOF elimination at low pH, EOF high at high pH : unsuitable for proteins - denaturation : decreasing the charge differences decreases separation efficiency c-\ high BGE concentration (ionic strength) reduction of coulombic interactions : decrease of EOF often limited by Joule heat 127 hydrophilic polymers alkylcellulose, polyvinyl alcohol, dextrans, Polyacrylamide shield wall charge of capillary and decreases EOF : increases viscosity : in high concentration = entangled gel electrophoresis (CEGE) tensides anionic: sodium dodecylsulphate (SDS), cationic: cetyltrimethylammonium bromide (CTAB) non-ionic: Brij-35, BRIS zwitterionic: 3-[(-cholamidopropyl)dimethylammonio]-1 -propansulphate (CHAPS) deactivate capillary surface by hydrophobic or ionic interactions : wide possibility of compounds, easy use : decrease or revert EOF : may irreversibly denaturise protein : suitable in combination with RP-LC surfaces c-\ quaternary amines v_J decrease or revert EOF : work also as ion pairing agents (MEKC) paper / membrane 100 % cotton / cellulose 0.17-0.30 mm thick pore size 2.5 |jm electrolyte source two glasses zones electrophoretic paper acetate cellulose pore size 0.2 |jm nitrocellulose pore size 0.2 |jm visualisation nafion (PTFE, sulphonated tetrafluoroethylene) 1 - 2 nm and 5 - 6 nm I hhN bromophenol blue dimethylthionine (azure A) toluidine blue alcian blue sudan black naphthalene black hUC H9N CI ,CH3 Cut an -n n' nh nh CH3 R ch3 r=ch2-s^n:ch^ H3C CH3 130 agarose gel gel non-toxic, cheap, no additional components for polymerisation fragile 0.8% large molecules 1 - 2% common separation 4% small molecules % w/v agarose solution resulting gel structure D-galactose 3,6-anhydro-L-galactose 131 toxic (bis-acrylamide), inert fragile, reinforcement by RhinoHide™ or DurAcryl™ NhL acrylamide °- H H 0 o \ o bis-acrylamide bridge methylene-bis-acrylamide persulphate /ammonium/ - initiator tetramethylene ethylenediamine (TEMED) - catalyser gel density (cross-linking percentage; acrylamide and bis-acrylamide ratio) 4< % cross-linking => easier motion of very large molecules 12% - common for 15 kDa - 60 kDa 8% - molecules 30 kDa - 120 kDa 25%-< 15 kDa; special protocol according to Schagger-von Jagow 12%-gel viscosity cavity diameter (12%) -100 m2 s-1 ~ 4.4 nm : isocratic (continuous) (8-15 %) : discontinuous gel (4% concentration and 12 % separation) : gradient gel (Schager-von Jagow) 133 ethidium bromide (EtBr) Kongo red Coomassie blue R-250, SYPRO ruby SYBR II green silver zinc copper H2N 0.3 M CuCl 0.2 M ZnS04 0.2 M imidazole 0.02% Na2S203 0.1%AgNO3 37% HCOH 1% CHoCOOH 134 chip (CE-on-chip) injection channel simpler arrangement than LC-on-chip : easy application of driving force : simple separation channel : suitable detection ZE, ITF, IEF... electrochemical detection access point separation channel lab-on-chip LC + CE absorption photometric detector diode array detector flow window in polyimine coating light source capillary i.d. 75 urn absorbance : sensitivity 106 M indirect detection : sensitivity 104 M detectors problems : beam focusation : optical path length focusing optics - two spherical lenses /-\ prolongation of optical path bubble cell optical photomultiplier radioactive (scintillation) detector ß-particle photomultiplier tube Nal crystal photocathode optical window anode meter scintillation : sensitivity 10 10 M MDD signal: beam of ß-particles (e-) fluorescence detector laser induced fluorescence (LIF) fluorescence : sensitivity 10 11 M LIF : sensitivity 10 13 M photo-multipl. laser —;S— aperture i i '-—3> lenses \i mirror V 4 3 capillary Maser \ lens CCD camera capillary 138 r-\ amperometric detector c-\ conductivity detector ground 1 < -i Q) < elution into collection vials (10 - 15 |jl) peak detection => volume calculation / distance from capillary end pressure elution: (CZE, ITP; MEKC, IEF; CGE - no) : pressure application (5 kPa) during pre-calculated time period electrokinetic elution: (CZE, ITP, CGE, MEKC; IEF - no) : voltage application during pre-calculated time period : collection vial must contain BGE or other electrolyte elution in IEF mode: : it is necessary to consider that |j = 0 collection electrolytes: CZE 2% acetic acid ITP 2% acetic acid CGE BGE MEKC BGE IEF ampholyte electrolyte detector pressure or voltage electrolyte collection vial 142 BGE composition: buffer concentration, pH, additives injection: type, its characteristics (time, pressure, voltage) separation channel type capillary length, i. d., material, manufacturer 30 cm x 50 pm i. d., fused silica, J&W Scientific conditioning - coating, rinsing planar size (height x length x thickness), material 6.5 x 10 cm x 1 mm, Polyacrylamide continuous, discontinuous, gradient; leading colour applied voltage, current or output application time period c-\ detector I,_/ basic characteristic according to type 143 r-\ analytical information from electrophoretogram qualitative information migration time normalisation bad reproducibility; adsorption or EOF changes : on one marker (either EOF or very fasf) : on two markers inclosing separated components first: carries no charge, moves with EOF second: highest mobility peak area is function of migration velocity (migration time) only within EOF changes; within ionic strength or injection length changes - no correction effect peak area normalisation AN=A-(leff/tM)^A/t M correction of injection length change within pressure injection IS - internal standard; might be a peak in mixture f-\ basic modes of electromigration methods electrophoresis (ZE) isoelectric focusation (IEF) isotachophoresis (ITF) electrochromatography (EC) micellar electrokinetic chromatography (MEKC) affinity electrophoresis (ACE) non-aqueous electrophoresis (NCE) 145 f-\ CZE, capillary zone electrophoresis electrophoresis - greek rjAcKTpov (amber) and cpopeco (I carry) one background electrolyte (BGE) => constant electric field intensity in whole separation channel • 'A.-A background electrolyte (BGE) ▲ A A ■ ■ ■ ■ • • • • t = 0 t>0 Ma - Mr Q = selectivity of separation, analytes A and B M b 146 (-\ choice of background electrolyte : sufficient buffering capacity in chosen pH range : low background signal in detector : low mobility (large, low charged molecules) => low Joule heat c-^ additives r-\ tensides all types changes EOF; give charge to non-polar molecules changes CZE into MEKC (if the critical micellar concentration is exceeded) f-^ zwitterions CHAPS (3-[(-cholamidopropyl)dimethylammonio]-1 -propansulphate) : increases ionic strength without increase in conductivity (heat) : influences selectivity 147 f-\ chiral selectors cyclodextrins, crown-ethers ... similar to chiral additives in MF within LC (-\ metal ions K+, Na+, Cu2+, Li+ ... influence selectivity in MEKC and GE chaotropic agents urea ... solubilise NAand proteins; influence selectivity in MEKC 148 (-\ linear hydrophilic polymers methylcellulose, polyacrylamide, polyethylene glycol, polyvinyl alcohol... decrease EOF; decrease analyte adsorption in low concentrations, ZE => GE (-\ organic agents methanol, acetonitrile ... generally decrease EOF; influence selectivity in MEKC and chiral separations complexing buffers borate ... allow separation of saccharides and catechols 149 CGE, capillary gel electrophoresis ■ ■ - gel matrix t>0 classical : cross-linked gel in capillary : relatively fast, reproducible and quantitative compared to slab gel electrophoresis : on-line detection in UV-VIS without visualisation disadvantages: capillary filling (homogeneous polymerisation, bubbles...) commercially filled capillaries - high price chemical gels: Polyacrylamides - porous structure with strong covalent bonds physical gels: agarose - weak intermolecular bonds of different molecule parts entangled gel : linear gel as part of BGE : entangling medium (e.g. polymerous net) is present in background electrolyte similar to physical gels - characteristic intermolecular interactions : rapid increase in viscosity ( = f(Mw)) at liminal concentration values : N-substituted acrylamides N-acryloyl aminopropanol (AAP) N-acryloyl aminobutanol (AAB) N-acryloyl aminoethoxyetanol (AAEE) : linear Polyacrylamide mostly used polymers : cellulose derivatives : polyethylene glycol (PEG) : polyethylene oxide (PEO) methylcellulose (MC) hydroxyethylcellulose (HEC) hydroxypropylcellulose (HPC) hydroxypropylmethylcellulose (HPMC) : polyethylene alcohol (PEA) : polyvinyl alcohol (PVA) : galactomannan (GalMan) : glucomannan (GluMan) 151 => isotachophoretic polymerisation capillary and anodic space: acrylamide, bisacrylamide, triethanol amine (catalyser) cathodic space: ammonium persulphate (initiator) when the source is switched on, the initiator enters the system ITF interface chloride / persulphate keeps initiator zone sharp => supervised polymerisation such a voltage that initiator flow ~ rate of polymerisation (ca 2 - 4 V/m) 152 f-\ GE, slab-gel electrophoresis denaturing (SDS, Lammli) - separation according to Mw non-denaturing (native) - separation according to pi, shape and Mw one dimensional gel electrophoresis (1D-GE) : slab gel polymerises between glass plates, separated by spacers : loading jars are created by special spacer - comb n?ifvv¥ijuuiaruui (4%) concentration gel (12%) separation gel 1. sampling buffer is added to sample 2. sample is loaded into jars 3. gel is put in-between buffers and voltage is applied 4. gel is washed and stained electrode sample sampling jars slower analytes faster analytes container with electrolyte working electrolyte electrode basic procedure wm fl_JT~LJ 'max R.-A. d retention factor 154 two dimensional gel electrophoresis (2D-GE) two dimensions: 1. IEF 2. SDS-GE 1. isoelectric focusation (IEF) immobilised pH-gradient in gel strip sample *1 decreasing pi 1st dimension IEF 2. denaturing gel elfo (SDS-GE) SDS is not in gel since polymerisation (as with 1D) : micelles would be created IEF strip on SDS gel ■ a [-> ij 2nd dimension SDS-GE decreasing pi necessary to cool more than 1D (5- 12 °C) H,C=C-H -N V -C=CH~ H 2 as cross-linking agent piperazine diacrylyl (PDA), diallyltartarate diamide (DATD), bisacrylyl cystamine (BAC) H,C=C-C-NI-K -( VHN-C-C=CH, H H„ H, H 2 OH OH O H.C=C—L. 2 H -SH-SH* O J—C=CH, H 2 sodium thiosulphate in gel - low background with Ag-staining 155 in 2D density gradient (9-16 %) is used in connected containers are mixed A) solution without cross-linker B) solution with max cross-linker concentration : at outflow, increasing cross-linker gradient is formed gradient profile is given by the shape of containers new- non-linear pH gradients in IEF pump after staining densitometry :: UV-Vis :: fluorimetry optical 1> density (%) 3=> CCD camera focusing optics passed light stained gel light filter T \-1-1 i H1-9-~ ill B mixing record evaluation 156 : prior to analysis, sample is denatured (+ EtSH, 95 °C, 5 min) :: breaking of di-sulphidic bonds :: turn into random coil conformation : leading colour :: bromphenole blue denaturing GE HO merkaptoethanol 'dismisses S-S bridges OH non-denaturing (native) GE : separation of acidic and basic proteins - separately: %7 AAA—>| SDS | unit charge : leading colour :: bromphenole blue for acidic :: methylene blue for basic N CH3 N CI" CH3 : separation of acidic and basic proteins - together :: giving them a unit charge without denaturation blue native PAGE clean native PAGE (BN-PAGE) - CBB R-250 (~ 1 g to 1 g of protein) (CN-PAGE) - n-dodecyl-p-maltoside and digitonin 157 polyacrylamidove gel electrophoresis - PAGE : for separation of proteins in native and denaturing mode; 1D and 2D agarose gel electrophoresis - AGE : for nucleic acids separation 0.8% 50 - x1000 kbp only one mode (1D) 1 - 2% 20 - 50 kbp NAs already have unit charge 4% < 20 kbp leading colours : xylene and bromophenol blue, cresol red, orange G separation conditions TRIS-acetate EDTA (TAE) : low voltage, large molecules (50 - xOOO kbp) TRIS-borate EDTA (TBE) : 20 - 50 kbp sodium borate (SB) : high voltage (35 V/cm), small molecules < 5 kbp new technique similar to slab GE - primarily preparative :: mostly SDS-PAGE :: native isoelectrofocusing QPNC-PAGE (quantitative preparative native continuous) suitable for on-line connection with detection techniques (MS) 491 prep cell separation buffer lution buffer cooling core collection channel gel column separation zones elution chamber fraction collectior separation buffer uffer recyclation 159 /-\ CIEF, capillary isoelectrofocusing V_J isoelectrofocusation - greek looq (same), rjAsKTpov (amber) and latin focus solution contains ampholytes during separation, the pH gradient is established pH = pi, analyte is not moving, movement towards detector only due to EOF (or pressure) mixture of ampholytes and sample C • D ▲ F E a G • a B • ea C . F H E a • G F a D B • a G F ^ F B H A • a D • B h a • H E A • D G A A low pH < pH gradient —> high pH A A B B a a c c DD EE ■ ■ F F • • G G H H A A B B a A c c DD EE ■ ■ F F • • G G H H t>0 160 zones are sharp, self-focusation effect v_) (»/( !<5pHj * - I 5x )) wA - zone width x - length coordinate resolution in IEF rapH> /e- I 5x ; / I aPHj E - electric field intensity [V/cm] dpH / dx - pH gradient du / dpH - mobility slope at given CITF, capillary isotachoforesis V_) isotachophoresis - greek looq (same), tqxus (speed) and cpopeco (i carry) two electrolytes : leading - leading ion has absolutely highest mobility in system : terminal (trailing) - terminal ion has absolutely lowest mobility in system => electric field intensity increases from leading to terminal ion T A* ■ ^ leading electrolyte (L) t = 0 terminal electrolyte (T) A A A A ■ ■ • • • • L t>0 component concentration in zone is according to Kohlrausch co-function analytical concentration of compound A, cA: Ml-M ci Ma-Mci M l for strong univalent electrolytes CI - analyte counter-ion 162 f-\ self-focusing effect zones are sharp and do not broaden => concentrating minor components in few orders if ion L because of diffusion goes to zone X, because of t E also increases its migration velocity and it goes back to zone L if ion X because of diffusion goes to zone L, because of 4< E also decreases its migration velocity and it goes back to zone X 333 isotachophoretogram typical detection - resistance; others methods - conductivity, thermometry, UV-Vis r-\ MEKC, micellar electrokinetic chromatography one electrolyte containing ionogenic tenside over critical micellar concentration => micelles are created analyte is separated between micelles and electrolyte acc. distribution coefficient (K) MEKC may be seen as ZE of two entities - analyte and micelles with it analyte does not enters micelles => K = 0, analyte enters completely => K = oc • tenside / micelles ▲ t = 0 t>0 k'= ^mO(^Ivl/^lv1,mic)) = K-(VS/VM) k" - capacity factor tm- void retention time tM- retention time Vmic- retention time of micelles 164 commonly used tensides anionogenic : sodium dodecylsulphate ... cationogenic : cetyltrimethylammonium bromide, septonex to decrease migration velocity of micelles non-ionogenic tenside (Triton X-100) is added micelles may be substituted with microemulsion or poly ions addition of organic phase: solvatation changes, micellar structures, smoother setting - mixture of tensides resolution in MEKC R = ra-1> ( k'2 ] { i-(tm/tM) ) I4 j Ui-(tm/tM))-k;J efficiency selectivity a - selectivity retardation n - number of theor. plates disadvantage: difficult reproducibility 165 TLE, thin layer electrochromatography paper electrophoresis, slab electrochromatography charged (mostly negative) SF; often silicagel, cellulose and its derivatives analyte is separated between SF and electrolyte acc. distribution coefficient (K) charged surface groups I— -r T- O 0 0 0 0 0" compact C£^) C+) (+) © layer ---------------------- diffusion I__solvent j_I-1 EOF fast: applied voltage is driving force; comparing to TLC where it is capillary elevation : fast also comparing to capillary variant (up to three orders of magnitude) : voltage 160 V/cm => migration velocity 100 |jm.s-1 166 (-\ CEC, capillary electrochromatography charged (mostly negative) SF; porous particles of o.d. 1.5 - 5.0 |jm column: either broader (320 |jm) or narrower capillary (50, 75 or 100 |jm) analyte is separated between SF and electrolyte acc. distribution coefficient (K) : applied voltage is separation driving force => flow of the liquid is not laminar : EOF is created on the surface of SF rather than on a wall of separation channel low currents: max 10 |jA Joule heat 0.1 W.cnrv2 (1500x more heat than within pressure heating by HPLC) charged (-) stationary phase Oo° OoO 00( o o o oQo0 o oo0 c O O ( 0 O < ■f Ob 3^.c o o o °9°< o o o o0°c . D.o. :)• o #- o o 167 C18 bound on silicagel (reverse CEC) (3-CD bound on silicagel (chiral CEC) SCX cation exchanger (-CH2CH2CH2S03H) 90 % SF for separation equilibrium 10 % SF (pure silica) for EOF stabilisation HIT 5HV Imrn PCI CIE33CN testing mixture thiourea GR 57888X, GR 57994X fluticason propionate, des-6-a-fluoro-fluticason propionate thiourea indicates EOF components 2 and 3 determine hydrophobicity components 4 and 5 determine resolution HO*. COSCH2F CM,? ^OCOEl O + electric field advantages : higher efficiency than HPLC :: up to 300 000 plates / m (i.e. 3 - 4x) frit column frit window : may use very small particles :: no high back pressure : separation of neutral, lipophilic and water-insoluble analytes : low sample and MF consumption : isocratic and gradient elution : may use MS detection : same instrumentation as for CZE, CEC or CLC 320 urn i.d. 75 urn i.d. 280 urn o.d. disadvantages inlet frit column outlet frit window : column :: filled capillaries with frits; fragility : bubbles (EOF differences, Joule heat) : electrokinetic injection (internal standard) : lower sensitivity 169 ŕ-\ AE, affinity electrophoresis V_) uses combination of separation in filed and affinity separation affinity separation - specific interaction of analyte and ligand enzyme nucleic acid antigen receptor coenzyme, substrate, inhibitor complementary chain, histone antibody signal molecule afinant carrier analyte A analyte B - EOF in capillary and in gel : separation highly selective : purification shot-gun : interaction study compatibility association constants separation c-\ blotting Southern blot - DNA solution Northern blot - RNA Western blot - proteins r-\ immunoelectrophoresis 1D gel immunophoresis 2D gelova immunophoresis f-\ NAE, non-aqueous electrophoresis V_) separation in non-aqueous solvents 1978 - non-aqueous TLE 1984 - non-aqueous CE (NACE) advantages : elimination of levelling effect of solvent => higher selectivity of separation : low current : separation of hydrophobic (water-insoluble) analytes solvent choice : volatility : ability to solve BGE and analyte : viscosity : dielectric constant : transparency in UV 172 solvents water content max 1 % amphiprotic neutral (+ ;+): protogenic (+ ;-): protophilic (- +): dipol. protophilic (- +): MeOH, glycerol, phenol, te/f-butylalcohol sulphonic a., formic a., acetic a. liquid ammonium, formamide, N-methylformamide DMSO, dimethylformamide, THF, 1,4-dioxan, pyridine aprotic : dipol. protophilic : inert (-;-) : AcN, acetone, nitrobenzene, sulpholane, PC (-;-) : alif. hydrocarb., benzene, 1,2-dichloret., tetrachlorom. relatively basic or acidic (*;*) background electrolytes : ammonium acetate, sometimes with addition of acetic a. or sodium acetate : quaternary ammonium salts : Tris, magnesium acetate, citric a., formic a., trifluoroacetic a. ... additives: polyalcohols and surfactants => decreasing EOF 173 validation of analytical separation method procedure - demonstration and documentation of quality of analytical separation method by means of establishment of defined criteria and by estimation of values of these criteria statistical proof of reliability of separation method : including whole manipulation/analytical background validated property - subject of validation : identity and concentration of principal substance : impurity concentration : physico-chemical parameter when to validate? when introducing new method when transferring validated methods (e.g. out of development into target laboratory; published validated methods) when checking competence of system when revalidating method; revalidation conditions should be strictly given 174 /-\ kinds of validation internal validation _^ in a frame of one laboratory f->. pilot validation : limited number of samples : piloting the suitability of chosen analytical method for full scale validation : validation parameters: selectivity, robustness, reproducibility full validation : demonstration of method suitability for intended use : all required validation parameters 175 /-\ validation by method transfer V_/ : introduction of published validated analytical method : validation parameters: laboratory accuracy and reproducibility f-\ retrospective validation : checking the validity of previously fully validated method : checking the calibration line (linearity and sensitivity) : validation parameters: reproducibility C-"\ external validation inter-laboratory comparison tests : internal validation + comparative method validation from more laboratories : validation parameters: repeatability 176 validation programme summarises basic rules: : for planning and organisation of analytical data validation : for introduction and use of such defined parameters in praxis f-\ items of validation programme a) operating sequence b) validation parameters c) system revalidation conditions d) validation protocol e) literature (critical research and consultations) 177 c-\ operating sequence v_) : complete analytical formula serving to reproduce whole analytical method : contains all needful instructions: precise, detailed and complete : must be optimised and as such used with statistical check of measurements characteristics of operating sequence scope of method use, sequence principle, chemical reactions and interactions of determined component, analyte and matrix, range of content of determined component, measurement principle and units chemicals chemical purity of chemicals used, their processing and purification, preparation of solvents, agents and support chemicals, stability and concentration standard operation procedure mechanical sample procedure, chemical sample procedure, calibration, measurement, calculations and evaluation 178 (■-N method identity v_J C-\ validation parameters measure of agreement tightness between independent results under defined conditions independent result - result obtained uninfluenced by any previous result on the same or similar sample expression: standard deviation of results (sx) absolute value ofsx- if not dependent on content (X) relative value ofsx (%) - if dependent on content relative standard deviation - if standard deviation is constant in whole range of measured values; related to the highest value of set x 179 (-\ standard deviation v_) characterises deviation of individual values x, around average x coincidental quantity - it is not valid characteristics of given analytical method must be specified : must content all sources of variability (also those of operation sequence -sample decomposition, dilutions, extractions, dissolutions, final instrumental measurement) : changing the operation sequence - revalidate the standard deviation value for obtaining must be sufficiently high number of samples of the same material not from one series, but from long-term measurements 180 /-\ reproducibility consistency of method under conditions of reproducibility : depends only on coincidental error distribution; has no relation to accuracy tightness of identity between mutually independent results of tests obtained under conditions of reproducibility conditions mutually independent results of tests by repeated use of the same test method on identical material, in the same laboratory, by the same operator using the same instruments, during short time range max H * ^x sx is standard deviation, q is tabulated value of studentised distribution 181 conditions of sxand reproducibility determination at least 5 levels (H), sample number m > 20, parallel measurement number nA=2 validation protocol: all measured quantities, calculated sx if we presume Rmax = f (H), we need to test linear dependence Rmax =a + b-H exponential dependence log Rmax = c + d • log H for a = 0 and d = 1 are these equation equivalent in case b = 0 (for majority of cases b < 0,1) is Rmax (resp. sx) constant in a whole range of X values if sxis not dependent on content, the relative standard deviation is calculated in regard to the highest set value xmax 182 method precision/accuracy v_) tightness of identity of obtained value with real one source of real value : standard : reference material : validated independent methods : reference laboratory (same method) yield of method R-nex/nref must be 0.95 - 1.05 for each concentration level yield test I. m - number of parallel determination of reference sx - determined out of min. 7 values on one concentration level; RSDmax = 3 % if t > ta; method is subjected to systematic error 183 /-\ yield test II. test to if systematic error is : constant ±f(c) : proportional = f (c) t.=- 1-a tb = : b 3b if tb > ta; method is subjected to constant systematic error eliminable - new blank experiment if ta > ta; method is subjected to proportional systematic error C-\ calibration choice linear or non-linear? linearise? and if non-linear, so which and why such? aspect of linearity: rxy > 0.98 min: 5 points in concentration scale, 3 points per each point of scale evaluation of importance of segment b recalibration new adjustment of parameters a and b : difference test of new and old values by F-test :: if not similar, it is necessary to calibrate again t = b'-b t = a'-a sensitivity dy _ df(x) dx dx first derivation of calibration function 185 limit of detection analyte concentration at which the signal is statistically different from noise uses blank experiment max deviation of baseline (hmax) in range of 20-fold of w1/2 of signal peak Ylod ~ 3 ' ^max =^ ^Lop ~ Ylod ' ah ; ah is calibration on peak height y = ah ■ x limit of quantification analyte concentration at which the relative standard deviation predicted from calibration is small (~ 0.1) uses blank experiment max deviation of baseline (hmax) in range of 20-fold of w1/2 of signal peak yloq =10- hmax => xloq = Yloq I 3h; ah is calibration on peak height y = ah ■ x 186 selectivity ability of precise and accurate determination of analyte in matrix presence determination comparison of analyte signal in standard sample and in sample with matrix all minimally 3x and at concentration close to LOQ : determine quantity and deviation of background signal : determine the difference importance of background signal to substance concentration at LOQ interferent < 1% of response close to LOQ 187 t-\ robustness extent of influence of individual parameter deviation on resulting determination robustness optimisation : choose purpose quantity/function; has an extreme in optimum; Z : consider and choose factors, which may influence result; Qt : for each Qj choose extreme of purpose quantity/function - min or max reduces multifactorial analysis by Plackett and Burman : use of 2-level reduced experimental design : minimal number of runs m (= 4), minimal number of factors n (= m - 1 =3) : to each factor assign two extreme values higher (+) and lower (-) : in the first line, m/2 factors is + and (m/2)-1 is - : each next line has same representation, but different composition : last line has all - if factor number < than possible (m-2) =^> use of dummy factors (+1 or-1) ((m/2)-1) dummy factors tests errors by prediction of main effect 188 run factors purp. funct. Q1 Q2 Q3 Q4 Q7 1 + + + — + — — 2 + + — + — — + z2 3 — — + — — + + z3 4 — + — — + + + z4 5 + — — + + + — z5 6 — — + + + — + z6 7 — + + + — + — z7 8 z8 weight w1 W2 w3 w4 w5 w6 W7 if Wj > W = sw ■ ta; influence of factor Qj is statistically important statistical testing deviation agreement agreement of mean values outlying values F-test Student t-test Grubbs T-test Q-test according to Dean-Dixon Cochran test C-test instrumental validation validation by manufacturer - norms ISO 9000 - 9004 other, individual validation program of instrumentation 190 C-N revalidation s_/ conditions cannot be generally defined each change in the analytical system must lead to its revalidation influence on final outcome should be considered individually revalidation should not be complex, only as a partial step of validation program (e.g. calibration, sensitivity); standard deviation must be retrospectively determined (resp. Rmax), i.e. influence of revalidation on value of Rmax resp. sx validation protocol in regard to particular validation program : records all measurements, calculations : results and conclusions are clearly defined mention the date of individual tests, name of responsible operator and names of all collaborators, which worked on validation program 191 (-\ scheme of validation procedure v_/ 192