摘要:

Thermally Stabilized Iridium on an Integrated Carbide-coated Platform as a Permanent Modifier for Hydride-forming Elements in Electrothermal Atomic Absorption Spectrometry Part 3.* Effect of L-Cysteine Journal of Analytical Atomic Spectrometry DIMITER L. TSALEV YALESSANDRO D'ULI'JO AND LEONARD0 LAMPUGNANI CNR Istituto di Chimica Analitica Strumentale Via Risorgimento 35 561 00 Pisa Italy MARCO D I MARCO AND ROBERTO ZAMBO'VI Universitci di Pisa Dipartimento di Chimica e Chimica Industriale Via Risorgimento 35 561 00 Pisa Italy The effect of L-cysteine (0.2-2% m/v) sodium tetrahydroborate (0.05-0.4% m/v) and hydrochloric acid concentrations (0.01-0.15 mol I-') in an automated flow injection hydride generation system with in situ collection of hydrides of Sb As monomethylarsonate (MMA) dimethylarsinate (DMA) Bi and Sn in an electrothermal atomizer with an integrated platform pre-treated with 110 pg of Zr and 2 pg of Ir for permanent modification was studied.The hydride generation step was optimized by means of a full factorial design of the 33 type and the corresponding regression equations response surfaces and contour plots were derived and are discussed with a view to the improved chemical yields and possible unified conditions for the simultaneous hydride generation/trapping of the three As species or all six analytes. The best characteristic masses for integrated absorbance measurements are 83 35 31 32 110 and 104 pg for Sb As MMA DMA Bi and Sn respectively. The importance of the molar supply rates of the reagents (in pmol s-l) and their ratios rather than reagent concentralions as well as the critical role of sample acidity are emphasized.Keywords Hydride generation and trapping; electrothermal atomic absorption spectrometry; iridium and zirconium coliting; in situ enrichment; L-cysteine; antimony; arsenic; monomethylarsonate; d imethy lars inate ; bismuth; tin In previous parts of this series,'.2 we have proposed and evaluated a permanent Ir modifier stabilized on a Zr- or W-treated platform which has proved efficient for the stabiliz- ation of numerous volatile analytes (As Bi Cd Pb Sb Se Sn Te T1) in direct ETAAS' and more promisingly for in situ trapping/enrichment of volatile hydride-forming elements (HFES)~?~ and some organoelement species of As Se arid Sn with known biological and environmental occurrence and importance3 in automated FI-HG-ETAAS.2 The aim of this study was to evaluate the performance characteristics of this system in the presence of ~-cysteine,~ which appears to be one of the most promising reagents in HGAAS,3 being both an efficient pre-reductanl for As"' and Sb"' and a suitable reaction medium for several HFEs.* For Parts 1 and 2 of this series see refs. 1 and 2. t On leave from Faculty of Chemistry University of Sofia. Sofia 1 126 Bulgaria. In a series of papers by Brindle and co-worker~~-'~ on HG-MIP-AES4-8*'0-'5 and HGAAS,9 ~-cysteine~-'~ and a similar reagent ~-cystine,~.'~-'~ were introduced and analyti- cally evaluated. Further studies by several other research groups on the effects and applications of ~ - c y s t e i n e ~ * ' ~ - ~ ~ and ~-cystine~' in HGAAS,'6,'7,22 FI-HGAAS,19-21*24*25 FI-HG- ETAAS2,18 and HG-non-dispersive atomic fluorescence spectrometry ( HG-NDAFS)23 have been documented in recent years.Several analytes have been studied AS,2.5-9,14,18-21,24,25 monomethylarsonate ( MMA)24 and sn9. 15-19.22 and organotins.22 L-Cystine appears to be less convenient and efficient than L- cysteine in practical work because of its lower solubility in water; therefore higher concentrations of reagent and acid are required e.g. 3% m/v L-cystine in 5 moll-' HCl14 or 4 moll-' HCl,25 or alternatively 0.4 g of solid reagent is added to 5 ml of sample solution in a batch system;" excessive foaming is also entailed. L-Cysteine is compatible with modern continuous flow ( ~ ~ ) 7 - 9 1 6 .1 7 . 2 3 and FI ~ y s t e r n s ~ ~ ' ~ ~ ' ' - ~ ~ ~ ~ ~ and at lower con- centrations of reagent (0.5-2%) and acid (generally below 1 moll-'). Several advantages of this reagent over other pre- reductants and HG media3 have been found (i) faster and more efficient pre-reduction of AsV and SbV to their trivalent states even at lower acid concentrati~ns;~-'~~'~~~~ thus success- ful on-line pre-reduction of AsV has been a t t e m ~ t e d ; ~ ~ ' (ii) sta- bilization of analyte solutions on storage (e.g. for at least 1 (iii) better tolerance to interferen~es;~-'~"~-''~~~ (iv) sensitivity impr~vement;~-~~~'~~'~~'~ (v) identical responses from As"' AsV MMA and DMA;24 (vi) lower blanks; (vii) mild conditions for HG resulting in smaller amounts of H2 being produced' and aerosol being formed,23 and thus less disturb- ance with certain techniques such as HG-DCP-AES,4-8i'0-15 HG-NDAFS23 and HG-ETAAS.2 dimethylarsinate ( DMA),24 Sb,239,10719*20323 Ge,4*1 '-13 EXPERIMENTAL Apparatus The automated FI-HG-ETAAS system based on a Perkin- Elmer (Uberlingen Germany) Model 4100 ZL atomic absorp- tion spectrometer and a transverse-heated graphite atomizer (THGA) longitudinal Zeeman-effect background correction and an AS-7 1 autosampler,26 interfaced with a Perkin-Elmer Journal of Analytical Atomic Spectrometry October 1996 Vol.11 (989-995) 989FIAS-400 FI system has been described el~ewhere.'.~.~~ The standard THGA graphite tubes with integrated platforms (Part No. B 300-0643) were pre-treated with 1.2 pmol of Zr (1 10 pg) and then with 10.4 nmol of Ir (2 pg) as detailed in ref.1. The system was operated under the Software Version 7.21 (Part No. B0509524) using the optimized parameters and tempera- ture programmes for the FIAS-400 manifold and the THGA electrothermal atomizer previously given in refs. 1 and 2. Reagents The preparation of standard solutions and other reagents has been detailed elsewhere.'T2 Working solutions of reductant (0.05-0.400/ m/v NaBH,) were always made alkaline with 0.05% m/v NaOH (i.e. supply rates of 0.833 pmol s-' of OH-). A small amount of HC1 (50mmol 1-l) was added to the stock solution of L-cysteine (10% m/v) in order to facilitate dissolution of the reagent; this contributed to small supply rates of H+ due to this reagent solution varying from 0.1 to 1 pmol s-'.Experimental Design for Studying Hydride Generation and Trapping The effect of the concentrations of reagents viz HCl L-cysteine and NaBH was studied over the following ranges 0.01-0.15 moll-' HC1 0.2-2% m/v L-cysteine and 0.05-0.4% m/v NaBH by means of a full factorial design of the 33 type (see Table 1) according to the plan suggested by the software programme of the Statistical Graphics System Version 6 (Manugistics Rockville MD USA). Randomized experiments ( n = 29) with three replicate measurements were performed the highest mean value of the integrated absorbance for each analyte species being assigned a value of 100% response (Y%). The analyte level was fixed at 5 ng for all species. RESULTS AND DISCUSSION The effect of the concentrations of the three reagents in this chemical system viz. HCl (mol 1 - ') L-cysteine pre-reductant (YO m/v) and NaBH reductant (YO m/v) on the response for each analyte species (Y%) can be described by a regression Table 1 Factors and levels for a 33 full factorial design Level Factor -1 0 + 1 HCl/mol 1- * 0.01 0.08 0.15 NaBH (YO m/v) 0.05 0.225 0.40 L-Cysteine ( O h m/v) 0.2 1.05 2.0 equation of the following type Y% = k o + k [HCl] + k [NaBH,] + k [~-cysteine] + k [ HCl] [NaBH,] + k [ HCl] [~-cysteine] + k [ NaBH,] [~-cysteine] + k7[ HC1I2 + k8 [NaBH,I2 + k9 [~-cysteine]~ In Table 2 are given the coefficients (ko-k,) for each analyte in the above equation as derived from the 33 full factorial design.All experimental values are fitted satisfactorily by these four- dimensional (4D) response surfaces.The correspondence between the concentrations and supply rates (pmol s-') of the reagents as well their molar ratios can be evaluated from the diagram in Fig. 1. It seems that the most convenient and practically useful mode of graphic representation of these 4D surfaces would be as a series of contour plots i.e. the cross-sections of response surfaces at three selected levels of L-cysteine (a) 'low' 0.2% m/v or 1.76 pmol s-' (b) 'the centre of the experimental plan' 1.05% m/v or 9.24 pmol s - l and (c) 'high' 2% m/v or 17.6 pmol s-'. These diagrams are plotted for inorganic As (i-As"') MMA DMA Sb"' Bi"' and Sn" in Figs. 2-7 respectively while the contours at Y>90% are overlayed for all species in Fig. 8 thus showing the optimum conditions for simultaneous HG and determination.Replotting these diagrams in coordi- nates representing the actual molar concentrations of reagents HCI/ mol I-' - 0 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0 2 4 6 8 10 12 14 16 18 HCI/ pmoi s-' 0.5 0.4 3 E s 0.3 . I" m 0.2 2 O.? 0 1 Fig. 1 Calculated molar ratios of reagents within the field of exper- imental parameters. Solid line [H+]/[BH,-]; dashed line [L- cysteine]/[BH,-1; dotted line [~-cysteine]/[H+ 1. L-Cysteine concen- tration fixed at 1.05% m/v (9.24 pmol s-l) Table 2 m/v)] derived from the 33 full factorial design Model fitting results for the regression equations Y% uersus reagent concentrations [HCl/mol l - l NaBH (Yo m/v) and L-cysteine (Yo Analyte species As"' 89.8 - 9.43 - 6.40 0.109 0.113 0.0530 0.241 1.46 2.66 - 0.704 MMA 27.5 - 7.03 - 0.606 62.4 0.0806 1.39 0.361 0.385 - 2.64 - 19.7 DMA 71.5 -4.05 - 2.95 -22.6 0.195 3.23 0.145 0.628 - 2.17 3.71 BiI" 68.7 1.64 8.55 2.30 0.0187 - 1.13 - 0.806 - 0.047 - 0.696 -0.815 Sb"' 81.9 0.763 0.372 1.48 0.0123 0.0862 -0.124 - 0.038 0.103 -0.500 Sn" 91.0 0.760 3.64 0.141 0.700 - 19.9 -0.332 - 0.07 1 - 0.565 3.33 990 Journal of Analytical Atomic Spectrometry October 1996 Vol.11a HCI/ pmol s-l Contour plots for As"' normalized integrated absorba ace signal) cersus supply rates of acid and reductant at three levels of L-cysteine (a)j0.2; ( h ) 1.05; and (c) 2% m p - 0 2 4 6 8 10 12 14 16 (b) MMA I . 1 ' 1 . I ' 1 ' I ' 1 2 4 6 8 10 12 14 16 HCI/ pmol s-l Fig. 3 (a) 0.2; (b) 1.05; and (c) 2% m/v Contour plots for MMA (normalized integrated absorklance signal) versus supply rates of acid and reductant at three levels of 1.-cysteine 0 0 2 4 6 8 10 12 14 16 HCI/ pmol s-' Fig.4 Contour plots for DMA (normalized integrated absorbmce signal) uersus supply rates of acid and reductant at three levels of L-cysteine (a) 0.2; (h) 1.05; and (c) 2% m/v 2 0 2 4 6 8 10 12 14 16 HCI/ pmol s-' 0 93 9 1 d 86 I I ' 0 2 4 6 8 1 0 1 2 1 4 ' S Fig. 5 (a) 0.2; (b) 1.05; and (c) 2% mjv Contour plots for Sb"' (normalized integrated absorbance signal) versus supply rates of acid and reductant at threc levels of L-cysteine .I'ournul of Analytical Atomic Spectrometry October 1996 Vol. 11 991'na 0 . 1 . i ' - I ' 1 . I ' I ' - I 0 2 4 6 8 10 12 14 16 4w9* HCI/ pmol s-l Fig. 6 Contour plots for Bi"' (normalized integrated absorbance signal) cersus supply rates of acid and reductant at three levels of L-cysteine (a) 0.2; (b) 1.05; and (c) 2% m/v 6 8 10 12 14 16 HCI/ pmol s-' Fig.7 Contour plots for Sn" (normalized integrated absorbance signal) cersus supply rates of acid and reductant at three levels of L-cysteine (a) 0.2; (b) 1.05; and (c) 2% m/v 0 2 4 6 0 10 12 14 16 0 0 2 4 5 8 10 12 14 16 HCI/ pmol s-' I 0 2 4 6 8 10 12 14 16 4 Fig. 8 Overlay of the contour plots at 90% response (normalized integrated absorbance signal) for all analytes versus supply rates of acid and reductant at three levels of L-cysteine (a) 0.2; (b) 1.05; and (c) 2% m/v. The species symbol is placed on the side of >go% response before mixing uiz. H + (mol 1-') uersus NaBH (mol 1-') gives similar pictures to those in Figs.2-8 (not shown). Several important effects are observed in the presence of L- cysteine compared with the same system in the absence of this reagent:2 (i) improved sensitivity in FI-HG-ETAAS especially for i-As"' MMA DMA and Sn"/'" by 1.2- 2 . 5 1.8- and 1.5-fold respectively; thus the best characteristic masses (mo) for inte- grated absorbance measurements are decreased to 35 31 32 83 110 and 104 pg for i-As"' MMA DMA Sb'" Bi"' and Sn"/" respectively which agree well with the corresponding figures for the 'injected' analytes in direct ETAAS uiz. 31 30 26 92 176 and 156 pg;' (ii) substantial change in the HG pattern due to L-cysteine especially for i-As"' (shift towards much lower H + concen- trations) and Sn''/'" (signal decreases below certain H+ levels < 0.005-0.025 moll-' depending on L-cysteine concentration); (iii) strong effect of acid concentration on HG being most marked for i-As"' MMA and DMA (Figs.2-4); (iv) apparent importance of the molar supply rates and ratios of reagents uiz. HCl NaBH and L-cysteine; (v) similar behaviour of all three As species which makes possible their simultaneous determination under unified con- ditions [Fig. 8(b) and (c)]; (vi) high relative responses for Bi"'( 2 68% Fig. 6 ) Sn'1(2710/~ Fig. 7) and Sb"' (>84% Fig. 5 ) within the whole field of chemical parameters thus offering several sets of experimental conditions for the simultaneous determination of all four elements with only a minor sacrifice of sensitivity (Fig.8 ) . 992 Journal of Analytical Atomic Spectrometry October 1996 Vul. 11While all these effects are analytically important and practi- cally useful they are much easier to observe than to explain; in particular that the FI system is apparently under kinetic rather than thermodynamic control as observed also with other FI-HGAAS systems2 but not with batch HGAAS wherein a longer time is available for the generation and stripping of hydrides from the reaction medium to thr gase- ous phase.3 Expected Behaviour of the Chemical System The chemistry of the system L-cysteine-BH,-H +-analyte is complex because L-cysteine may act simultaneously in a diverse manner being (i) a pre-reductant (ii) an acid-base buffer (iii) a complexing agent for tetrahydroborate and (iv) 21; com- plexing agent for analytes. L-Cysteine as a pre-reductant L-Cysteine [ 2-amino-3-mercaptopropionic acid (HS- CH2- CHNH2-COOH) denoted as HSR] behaves as a reductant in aqueous solutions being oxidized to the disulfide form L-cysteine [ 3,3'-dithio-bis( 2-aminopropionic acid) HOOC- CHNH2-CH2-S-S-CH,-CHNH2-COOH denoted as RSSR] 2HSR -+ RSSR + 2H + + 2e L-Cysteine by itself may also further act as a reductant. The standard oxidation potential of the cystine-cysteine system is 0.074V (on the basis of polarographic data within pH 3.3-6.6).28*29 The formal potential is slightly increased at lower pH values.Thus L-cysteine is able to reduce As" and SbV to their trivalent states MMA and DMA to their trivalent thi~lates,~ Sn" to Sn" Se" to Se(0) and Te" to Te(0).The resulting solutions exhibit improved stability on storage for As"' Sb"' and Sn1',5,19,20 which could be due to complex formation between these ions and the HSR reagent. In contrast Se and Te are unstable and cannot be determined in solutions containing ~-cysteine;~' moreover the interference by Se on Ge and Sb may become stronger in the presence of this pre- r e d ~ c t a n t ' ~ ~ ~ ~ presumably due to coprecipitation of these analytes with the elemental form of the Se or Te interferent. The reduction of Bi"' to its elemental state is b rde~line;~' hence it would be worth studying the stability of L-cysteine solutions containing this element. L-Cysteine as an acid-base bufSer The acid-base equilibria for L-cysteine are also complex,32 as can be seen in the distribution diagram for this reagent (Fig. 9).Without acidification the neutral form of the reagent HSR 0 2 4 6 8 10 12 '14 PH Fig. 9 Distribution diagram for L-cysteine mole fraction (a) for different forms of reagent uersus pH behaves as a very weak acid with ~ K ~ ~ z t 3 . 2 - 8 . 4 and pKa3 NN 10.0-10.8,32-35 being in effect a dipolar ion (zwitterion HS-CH2-CH-NH3+-COO-).32 At lower pH values (<4-5) the protonated form of the reagent HSRH+ with a much higher acid constant uiz. pK z 2 ( 1.88 +0.02;35 1.90;34 1.9632) is gradually formed on acidification. The ratio of HSRH' HSR must be rapidly increasing with decreasing pH below 3 being approximately lo-' 1 10 and lo2 at pH 3 2 1 and 0 respectively (Fig. 9). Thus L-cysteine also acts as an acid-base buffer on acidification(!) just within the analytically useful range of pH 2 f 1 and its buffering capacity must depend on the concentration of reagent and the molar ratios of L- cysteine HC1 as well as on the initial acidification (or not) of the stock solution of reagent.Indeed pH measurements show that the actual pH of a 1% m/v L-cysteine solution in water is 5.0 which then decreases on acidification to pH 1.69 in equimolar HCl medium (0.0825 moll-' of the added H f ) ; this is in agreement with the distribution diagram in Fig. 9. On the other hand 0.01 mol 1-1 solutions of HCl exhibit actual pH values of 2.0 2.4 2.9 and 3.1 in the presence of 0 0.2 1.05 and 2% m/v L-cysteine respectively i.e. pH values are higher than expected on the basis of only formal concentrations of the added HC1.Further complication of the acid-base equilibria also takes place after merging the reagents in that BH,- and OH- also act as bases and are able to neutralize part of the H + in the resulting solution instantaneously. Concurrent reac- tions of analyte reduction and tetrahydroborate hydrolysis/ decomposition are obviously taking place over very short periods of time. Hence even the mode by which the stock solutions of NaBH are made alkaline36 (or not37) will also affect the HG processes. L-Cysteine as a complexing agent for tetrahydroborate Brindle and Le12 explained the higher efficiency of tetrahydro- borate reductant in the presence of L-cysteine by the possible formation of more efficient intermediate species of the reductant BH4- +HSR-+H3B-SR- +H2 This scheme has been supported by 13B NMR spectra for L- cysteine and analogous reagents uiz.thioglycerol and penicilla- mine but not for non-thiol-containing compounds such as S- methyl-L-cysteine histidine and glycine.12 The scheme is not in conflict with that for the action of BH4- in acidic media given by Mal'tseva and K h a i ~ ~ ~ * wherein the formation of more reactive intermediate species of the reductant BH has been supposed. Hydride generation by the H3B-SR- form of the reductant can be expected to proceed faster and with higher chemical yields than HG by the protonated hydroborate BH but obviously takes place at much lower acid concentrations. Some possible explanations for this effect of H + are (i) either the H,B-SR- species cannot be formed at low pH when L-cysteine is already protonated and/or they are rapidly decomposed at low pH or (ii) higher H + levels are more detrimental in the presence of L-cysteine because of their more pronounced effect on that form of the reductant which is a weaker acid i.e.H3B-SR-Hf rather than H+BH4-. L-Cysteine as a complexing agent for the analyte Yet another possible effect of L-cysteine uiz. complex formation with the analyte ions cannot be excluded from c~nsideration;~~ moreover it may play a decisive role in the HG processes. If such complex formation is taking place it may either impair HG by 'masking' the analyte ions or improve reduction processes owing to higher chemical yields or better kinetics or both. A 'masking' effect on the analytes does not seem very probable because no signal depression is observed in the Journal of Analytical Atomic Spectrometry October 1996 Vol.1 1 993presence of L-cysteine; if still taking place it must be more pronounced for those elements that are complexed more strongly by the HS-reagents (Bi"' Sn") and at excessive amounts of L-cysteine and/or at lower acidity. In fact such a slight impairment of relative responses is observed at 2% m/v levels of L-cysteine and low H + concentrations for Bi [Fig. 6(c)] and Sn [Fig. 7(c)]. Although there are published data on the stability constants of L-cysteine with Cu" Hg" Mn'l Nil' Pb" and Zn" (within 109-1020),34,35 no data could be found on complex formation with AS'" Sb"' Sn" and Bi"'. Nevertheless these ions could be expected to be complexed to some extent by the thiol group of the reagent presumably in an increasing order in the series As"' < Sb"' < Sn" <Bi'" by analogy with the corresponding solubility products of their sulfides.As"' (a 'hard acid') should be the least inclined ion to form complexes with the HSR reagent. Hence As should be the most tolerant of these four analytes to low acid and high L-cysteine levels (see the upper left corner of the diagrams in Fig. 8). On the other hand the complexation of As species with L-cysteine should be more vulnerable to higher H + levels. This is in fact observed experimentally. Several ~ ~ r k e r ~ ~ ~ ~ ~ - ~ ~ have assumed that the formation of complexes (thiolates) between As species and reagents containing the thiol (-SH) group is taking place; eg.in the presence of mercaptoacetic a ~ i d ~ " ' ~ or its methyl ester,41 ~-cysteine~ and thi~glycerol,~~ while no positive effects have been observed with other sulfur-containing additives such as histidine12 or m e t h i ~ n i n e ~ ~ which do not contain an HS group. Haraguchi and Takatsu4' derivatized MMA and DMA to their corresponding volatile products CH,As- (SCH,COOCH,) and (CH,),AsSCH,COOCH respectively and proved by GC-MIP-AES that both As and S were present in the eluates with identical retention times i.e. As and S are present in the same molecule. Le et aL2 have proposed that the reduction of AsV MMA and DMA to their trivalent states and complex formation are taking place in the presence of L- cysteine As( SR) CH,As(SR) and (CH,),AsSR respectively and that the rate of HG from these species is faster than from hydrolysed As in aqueous solutions without added L-cysteine.If these logical assumptions by Le et aL2 are adopted then one might expect that HG will be favoured under conditions where complex formation between analyte ions and L-cysteine is promoted i.e. at lower acid levels. Experimentally HG in the presence of L-cysteine always takes place at lower acidities than in the absence of this reagent. The effect of the concen- trations of reagents and their molar ratios is particularly critical for As (Figs. 2-4). At low concentrations (deficiency) of L-cysteine [Figs. 2(a) 3(a) and 4(u)] marked differences between As"' MMA and DMA are observed. The response Y% generally decreases in the order As"' > DMA > MMA i.e.the stronger the acid the lower the efficiency of HG cf pKal =9.2 6.2 and 4.633,42 for H3As03 (CH,),AsOOH and CH3AsO(OH) respectively. This is the opposite order to that observed in the absence of L-cysteine wherein hydrides from more acidic As species viz. H3As04 and MMA are generated in more acidic media.,,,' In the absence of L-cysteine the distribution of the neutral and charged As species within this particular pH range3I may also account for the above effects log [HAs02]/[As0+] =0.34+pH According to this equation the neutral species of the analyte should prevail at higher pH values e.g. the [HAsO,]/[AsO+] ratio should be about 2.2 22 221 and 2188 at pH 0 1 2 and 3 respectively. Indeed the stepwise attack of the neutral molecule of hydrated arsenic (111) oxide by the reductant has been assumed t o be part of the reaction mechanism of HG in the absence of ~-cysteine.~* DMA is also protonated at low pH values to form a cation (CH3)2As(OH)2+ with pK,,= 1.28 k0.05 as shown by ion-exchange studies by Hansen et al.43 A detailed study of these complex effects is obviously required in order to utilize the potential of L-cysteine completely but is outside the scope of this series of papers.CONCLUSIONS Analytically useful effects on HG are observed in the presence of L-cysteine dilute acid and tetrahydroborate viz. sensitivity improvement due to better yields and kinetics of HG; unified conditions for species-independent determination of the sum of i-As"' i-AsV MMA and DMA; appropriate conditions for the simultaneous HG of arsine stibine bismuthine and stan- nane etc.The generation of arsine and methylarsines takes place within a fairly narrow range of acid concentration and at molar excess of L-cysteine and tetrahydroborate. Although working at lower acid levels is advantageous for lower blanks and reduced reagent consumption the pH adjustment of real sample digests becomes more critical in the presence of L- cysteine. The molar ratios of reagents and their supply rates are important because these chemical parameters are closely inter- related and the molar concentrations after mixing the reagents may critically affect the distribution of neutral charged and complexed species of the analyte. Among the factors contribu- ting to the total balance of reagents could also be 'details' such as the mode of stabilization of stock solutions acidification (:or not) of the L-cysteine and basification of the NaBH solutions.The actual pH obtained after mixing the reagents could be very different from the expected value because of the buffering properties of L-cysteine in the pH range 1-3 and the instantaneous neutralization of HC1 and NaOH. Financial support to D. L. T. by the Consiglio Nazionale delle Ricerche Istituto di Chimica Analitica Strumentale Pisa is gratefully acknowledged. This work was supported in part by the Ministry of University Scientific Research and Technology (Progetto Speciale Sistema Lagunare Veneziano) Italy. One of the authors (R. Z.) thanks the CNR National Committee for the Environment and Habitat Technologies Italy for financial support.REFERENCES 1 2 3 4 5 6 7 8 9 10 11 12 13 84 15 16 17 994 Journal of Analytical Atomic Spectrometry October 1996 Vol. Tsalev D. L. D'Ulivo A. Lampugnani L. Di Marco M. and Zamboni R. J. Anal. At. Spectrom. 1995 10 1003. Tsalev D. L. DUlivo A. Lampugnani L. Di Marco M. and Zamboni R. J. Anal. At. Spectrom. 1996 11 979. Dgdina J. and Tsalev D. L. Hydride Generation Atomic Absorption Spectrometry Wiley Chichester 1995. Brindle I. D. and Ceccarelli Ponzoni C. M. Analyst 1987 112 1547. Brindle I. D. and Le X.-c. Anal. Chem. 1989 61 1175. Brindle I. D. and Chen H.-w. Talanta 1991 38 1137. Brindle I. D. Alarabi H. Karshman S. Le X.-c. Zheng S.-g. and Chen H.-w. Analyst 1992 117 407. Chen H.-w. Brindle I. D. and Le X.-c.Anal. Chem. 1992,64,667. Le X.-c. Cullen R. Reimer K. J. and Brindle I. D. Anal. Chim. Acta 1992 258 307. Chen H.-w. Brindle I. D. and Zheng S.-g. Analyst 1992 117 1603. Brindle I. D. Brindle M. E. Le X.-c. and Chen H. J. Anal. At. Spectrom. 1991 6 129. Brindle I. D. and Le X.-c. Anal. Chim. Acta 1990 229 239. Brindle I. D. Le X.-c. and Li X.-f. J. Anal. At. Spectrom. 1989 4 227. Boampong C. Brindle I. D. Le X.-c. Pidwerbesky L. and Ceccarelli Ponzoni C. M. Anal. Chem. 1988 60 1185. Brindle I. D. and Le X.-c. Analyst 1988 113 1377. Beach C. and Shrader D. in 7th Annual Spring Optical Spectroscopy Workshop Varian Sugar Land TX 199 1 Shrader D. and Beach C. Presented at the X X V I I Colloquium Spectroscopicum Internationale June 9-1 4 1991 Bergen Norway Abstract D-4.3.pp. 11 1-1 19. 1118 19 20 21 22 23 24 25 26 27 28 29 30 31 32 An Y. Willie S. N. and Sturgeon R. E. Spectrochim. Actil Part B 1992 47 1403. Welz B. and Sucmanova M. Analyst 1993 118 1417. Welz B. and Sucmanova M. Analyst 1993 118 1425. Welz B. He Y. and Sperling M. Talanta 1993 40 1917 Martin F. M. and Donard 0. F. X. J. Anal. At. Spectrom.. 1994 9 1143. D'Ulivo A. Lampugnani L. Pellegrini G. and Zambclni R. J . Anal. At. Spectrom. 1995 10 969. Le X.-C. Cullen W. R. and Reimer K. J. Anal. Chim. Acta 1994 285 277. Tesfalidet S. and Irgum K. Anal. Chem. 1989 61 2079. The THGA Graphite Furnace Techniques and Recomrvended Conditions Publication B3210.20 Part No. B05(!-5538 Bodenseewerk Perkin-Elmer Uberlingen 1992. The FIAS-Furnace Technique Recommended Analytical Conllitions and General Information Publication B3212.20 Part No. 0993-5204 Bodenseewerk Perkin-Elmer CJberlingen 1993.Kolthoff I. M. Stricks W. and Tanaka N. J. Am. Chem Soc. 1955 77 4739. Encyclopedia of Electrochemistry of the Elements. Organic Section Vol. X I I Carbonyl Compounds Carboxylic Acids Esters and Anhydrides Organic Sulfur Compounds eds. Bard A. J and Lund H. Marcel Dekker New York 1978 p. 409. Romboli L. Thesis University of Pisa 1995. Pourbaix M. Atlas d 'Equilibres Electrochimique Grau thier- Villars Paris 1963. Butler J. N. Ionic Equilibrium. A Mathematical Approach Addison-Wesley Publishing Co. Reading MA 1964 p. 23 :1. 33 34 35 36 37 38 39 40 41 42 43 CRC Handbook of Chemistry and Physics ed. Weast R. C. CRC Press Boca Raton FL 67th edn. 1986-1987. Perrin D. D. Stability Constants of Metal-Ion Complexes Part B-Organic Ligands Pergamon Oxford 1979 pp. 152-153. Martell A. E. and Smith R. M. Critical Stability Constants Plenum New York 1974 vol. 1 p. 47. Qiu D. R. Vandecasteele C. Vermeiren K. and Dams R. Spectrochim. Acta Part B 1990 45 439. Mandjukov P. B. Djarkova V. and Tsalev D. L. in 6. Colloquium Atomspektrometrische Spurenanalytik ed. Welz B. Bodenseewerk Perkin-Elmer Uberlingen 1991 pp. 597-608. Mal'tseva N. N. and Khain B. S. Borogidrid Natriya [Sodium Tetrahydroborate in Russian] Nauka Moscow 1985. Anderson R. K. Thompson M. and Culbard E. Analyst 1986 111 1143. Anderson R. K. Thompson M. and Culbard E. Analyst 1986 111 1153. Haraguchi H. and Takatsu A. Spectrochim. Acta Part B 1987 42 235. Sarjeant E. P. and Dempsey B. Ionization Constants of Organic Acids in Aqueous Solution Pergamon Oxford 1979. Hansen S. H. Larsen E. H. Pritzl G. and Cornett C. J . Anal. At. Spectrom. 1992 7 629. Paper 6/04887K Received July 1 1996 Journal of Analytical Atomic Spectrometry October 1996 Vol. 11 995

ISSN:0267-9477    

DOI:10.1039/JA9961100989

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Behaviour of Various Arsenic Species in Electrothermal Atomic Absorption Spectrometry I Journal of I Analytical Atomic Spectrometry I I VERA I. SLAVEYKOVA," FARAMARZ RASTEGAR AND MAURICE J . F. LEROY Ecole Europeenne de Chimie Polymeres et Materiaux de Strasbotirg Laboratoire de Chimie Analytique et Minerale U R A 405 CNRS 1 rue Blaise Pascal 67 008 Strasbourg Cedex France The behaviour of arsenite (As"') arsenate (As') monomethylarsonate (MMA) dimethylarsinate (DMA) arsenobetaine ( AsB) and arsenocholine ( AsC) in pyrolytic graphite coated graphite (pyrocoated) tubes was investigated. The influence of a tungsten carbide coating on the thermal pre- treatment losses of the analytes and on the analytical signals in aqueous and methanolic solutions was studied. Inorganic species MMA and DMA are less volatile in pyrocoated tubes; a tungsten carbide coating produces a good thermal stabilization but a marked 'dip' in the pyrolysis curves is observed in aqueous solutions.No pronounced stabilizing effect for the highly volatile AsB and AsC was observed in tungsten- treated tubes or in the presence of palladium chloride. The determination of these species requires the addition of palladium nitrate in both pyrocoated and tungsten-treated tubes. A comparison of the stabilizing action of palladium as its chloride and nitrate was made. Palladium nitrate exhibits efficient stabilizing action for each of the species studied whereas palladium chloride is efficient only for inorganic 4s species in pyrocoated tubes. The tungsten treatment of the tube and addition of palladium nitrate leads to a further increase in the pyrolysis temperature and better sensitivity for the As species.Tungsten treatment plays an important role in improving the performance of palladium chloride particularly in the determination of organically bound species. The effective stabilization and relative 'levelling-off' of the signal for each of the As species (except for AsC) in methanolic solution was observed in the presence of a palladium modifier in tungsten- treated tubes. The in situ separation and determination of As"' and AsB in tungsten-treated tubes was attempted. Howevtx because of the presence of significant amounts of AsB under the conditions used for As"' determination complete separation of these species was not possible.Keywords Arsenic species; electrothermal atomic absorption spectrometry; tungsten carbide coating; palladium modijie 1. Arsenic exists in environmental and biological samples in different oxidation and binding states. A knowledge of the speciation and transformation of As is important because each of the species possesses unique physical and chemical proper- ties which determine their bioavailability and physiological/ toxicological effects in the environment.''2 ETAAS provides limited potential for speciation. Usually the advantages of ETAAS are used for total As determination3 or off-line detection of As species after separation by HPI 1C4-7 and various extraction as well as the trapping of hydrides on the treated surface of a graphite This is reflected in the appearance of only a few publications dealing with the behaviour of As species in ETAAS. Krivan and Arpadjan13 have investigated the behaviour of As"' and AsV in a graphite furnace by means of an 76As * On leave from Faculty of Chemistry University of Sofia 1 lames Bourcier Sofia 1126 Bulgaria.radiotracer. The influence of different matrices such as HCl NaC1 HNO and urine and of various chemical modifiers including W and Pd was studied. A mixed W+Pd+citric acid modifier has been applied to determine As"' and AsV by ETAAS after extraction ~hromatography.'~ Tsalev et al.I5 showed that Ce" thermally stabilizes arsenite arsenate monomethylarsonate (MMA) and dimethylarsinate (DMA) up to 1100-1300°C and improves their response in ETAAS. LarsenI6 evaluated the analytical sensitivity for six As species using conventional and fast furnace programmes as well as a Pd + Mg modifier.Nevertheless systematic investigations of the behaviour of the individual As forms during the thermal pre-treatment and atomization stage are lacking. The aim of this work was to investigate the behaviour of arsenite (As"') arsenate (As") MMA DMA arsenobetaine (AsB) and arsenocholine (AsC) by ETAAS. Aqueous and methanolic solutions of these species were studied because they are the main components of the various extraction systems and HPLC effluents. The efforts were aimed at using the advantages provided by the W Pd and W+Pd modifiers" and by treatment of the graphite tubes with W and to compare their influence on the atomization signal and thermal pre-treatment losses of the As species.The W and W + Pd modifiers are very effective thermal stabilizers for inorganic As.'* Palladium alone and in combi- nation with other substances is one of the most popular and efficient modifiers used in ETAAS.I7 Pre-treatment of the graphite tube with W and other carbide-forming refractory elements has been widely employed in ETAAS offering certain advantages as reviewed elsewhere." Attention was directed to a detailed investigation of the pyrolysis stage which is the most critical step for the analysis of highly volatile species. The investigation of the behaviour of the various As species by ETAAS is of interest in both fundamental and practical aspects. In analytical practice it is important to apply a correct calibration procedure and to obtain reliable results for the total As in a sample of environmental or biological origin as well as with a view to developing methods for in situ speciation for some of the species.This preliminary study could be a basis for the development of ETAAS methods free of systematic errors for As determination in various undigested biological samples when As is present in a different form than in the reference solutions. EXPERIMENTAL Apparatus Measurements were made with a Varian ( Palo Alto CA USA) SpectrAA 400 Zeeman atomic absorption spectrometer equipped with a GTA 96 graphite atomizer and a programm- able sample dispenser. The operating parameters were set as recommended by the manufacturer except that a bandpass of Journal of Anizlytical Atomic Spectrometry October 1996 Vol.11 (997-1 002) 997Table 1 Temperature programme for the GTA 96 graphite atomizer loo-. 4 3 80 0 W 9) 6o s 40 a 20 Step No. 1 2 3 4 5 6 7 8 9 0 11 - - - - T/"C 90 120 120 var" var" 120 120 120 2300 2300 2600 t'ls 10 10 20 10 20 10 1 1 1 3 2 Ar flow rate/ 1 min-' 3.0 3.0 3.0 3.0 3.0 3.0 3.0 0 0 0 3.0 Read No No No No No No No No Yes Yes No *See figures. 0.5 nm was used. Pyrolytic graphite coated graphite (pyro- coated) partition tubes and integrated absorbance measure- ments were used throughout. The optimized heating programme for the GTA 96 graphite atomizer is shown in Table 1. A 'cool down' step was incorpor- ated in the heating programme in order to normalize atomiz- ation conditions for experiments at different pre-treatment temperatures.Reagents Stock standard solutions each containing 1000 mg 1-' of AS"' As" MMA DMA AsB and AsC in doubly distilled water were prepared using the following reagents sodium arsenite NaAsO,. (Rectapur; Prolabo Paris France); sodium arsenate Na2HAs0,.7H,0. (Rectapur; Prolabo); sodium monomethyl- arsonate CH3As0,Na2.6H,0. (Carlo Erba; Milan Italy); sodium dimethylarsinate (CH,),As02Na.3H,0. (Rectapur; Prolabo); arsenobetaine C,H1,As02*5H20. (Service Central de Microanalyse CNRS; Solaize France); and arsenocholine C,H,,AsOBr(CNRS). Working solutions containing 50 or 100 pg I-' of As were prepared daily from each of the stock standard solutions by dilution in doubly distilled water. The W chemical modifier (0.1% m/v) was prepared by dissolution of ammonium paratungstate (NH,),,H,( W207)6 (Fluka Buchs Switzerland) in doubly distilled water with gentle heating.Solutions (0.2 and 0.1% m/v) of Pd modifier as Pd(N03) in 15% HNO (Merck Darmstadt Germany) and PdCl in 10% HC1 (Riedel-de-Haen Hannover Germany) were used in the thermal pre-treatment study. Procedures Treatment of the graphite tube with the W modifier was realized by injection of 50 pl of a 0.1% m/v solution of the W modifier into the tube and thermal treatment by using the following temperature programme 90 "C for 20 s 120 "C (15+20s) 300°C (10+20s) 1200°C (lO+lOs) 2300°C (lo+ 2 s) 2500 "C for 2 s. The procedure was repeated three times. Such thermal treatment is believed to provide a smooth carbide coating with a minimized risk of high temperature volatilization of W03 and WC.,' The following procedure for in situ removal of As contami- nation from the W modifier was developed and applied in some thermal pre-treatment studies.A 20p1 aliquot of the modifier solution was injected and As was removed by pre- heating to 1700 "C for 10 s. After cooling to 40 "C the sample aliquot was placed in the graphite tube and the temperature programme in Table 1 was applied. A similar approach to in situ modifier purification has been used in Cd determination in serum in the presence of a Mg+ Pd modifier.21 Thermal Pre-treatment Study The maximum permissible pyrolysis temperatures were deter- mined by systematic variation of the pyrolysis temperature at a fixed atomization temperature. The effect of any drift in sensitivity was avoided by randomizing measurements and using the mean of two or three values at each temperature.Normalized absorbance measurements were used assigning a value of 100% to the plateau absorbance for the species. When a pronounced plateau in the pyrolysis curves was absent the maximum integrated absorbance signal was used to nor- malize the signals. The influence of the pyrolysis time on analyte losses and absorbance signals has been studied in detail p r e v i ~ u s l y . ~ ~ . ~ ~ On the basis of that study a pyrolysis time of 20 s was chosen as being the optimum and experiments were carried out at fixed pre-treatment times. RESULTS AND DISCUSSION Study of the Behaviour of As Species in Aqueous Solutions in the Absence of Modifier The behaviour of As in pyrocoated tubes depends on its oxidation and binding states.Typical examples of thermal pre- treatment curves for MMA and AsB are presented in Figs. 1 and 2 respectively. The inorganic and methylated species exhibit a maximum in the pyrolysis curves at temperatures of about 700-800 "C. At lower pyrolysis temperatures losses in the integrated absorbance signal were observed. The dimin- ution of the signal at low pre-treatment temperatures could be explained by redistribution of the sample in the tube as a result As is less efficiently atomized in the cooler region towards the ends of the tube. The temperature gradient in the '20 T fi 0 I 0 200 400 600 800 1000 1200 1400 1600 1800 Temperature PC Fig. 1 Pyrolysis curves for MMA in pyrocoated tubes (A) W-treated tubes (B) and in the presence of in situ purified W (20 pg) (C).T 2300 "C; aqueous solutions I2O T 04 I 0 200 400 600 800 1000 i200 1400 1600 1800 Temperature /OC Fig. 2 Pyrolysis curves for AsB in pyrocoated tubes (A) W-treated tubes (B) and in the presence of in situ purified W (20 pg) (C) and 6 pg Pd in W-treated tubes (D). T 2300 "C; aqueous solutions 998 Journal of Analytical Atomic Spectrometry October 1996 voz. 11Table 2 Characteristic masses m in picograms for different As species in aqueous solutions W-treated tubes Pd as its nitrate of chloride and W-treated tubes plus Pd modifier as its nitrate or chloride respectively ~ ~ As species Tube/modifier As"' AsV MMA DMA AsB AsC Pyrocoated 39 35 36 36 56 72 W 38 35 36 32 36 66 PdCl 40 43 51 45 48 65 WNO,) 25 26 26 25 26 40 W-PdCI 25 25 25 28 26 37 W-Pd(NO,) 20 19 20 20 20 28 central part of the furnace at low temperatures should favour condensation of both As vapours and compounds which vaporize at low temperature without decomposition.The vapo- rization of volatile As compounds at low temperature and their transformation into a less volatile form during thermal treatment could also be taken into account to explain the observed effect. By using MS at atmospheric pressure Styris et aL2' have shown that in the absence of a modifier As"' in HNO medium begins to lose As as AsO(g) at 490K; the losses are significant at 770K when As,(g) appears in the gas phase. AsB and AsC are more volatile and it is impossible to analyse these species without losses at ashing temperatures higher than 200 "C.Variation of the atomization temperature shows that the maximum integrated absorbances are obtained between 2000 and 2300 "C; however a temperature of 2300 "C was chosen because it gave the best ratio of peak height peak area and good response for all the species studied. The values of the characteristic masses m which correspond to the mass of analyte producing an integrated absorbance signal equal to 0.0044 s at pyrolysis temperatures correspond- ing to the maximum absorbance signal in the pyrolysis curves were calculated and are summarized in Table 2. The values shown are for aqueous solutions of the sodium salts of the As species without addition of acids. Chemical Modification With Tungsten Palladium and Tungsten Plus Palladium Efficient thermal stabilization is one of the prime requirements of a modifier.The utilization of higher pyrolysis temperatures has beneficial effects as has been reviewed in ref. 17. The potential of the applied modifiers was investigated with respect to their stabilizing power and improvement in the sensitivity. The stabilizing action of two forms of Pd viz. the nitrate and chloride in pyrocoated and W-treated tubes was compared. On the basis of a previous study,26 6 pg of Pd modifier was chosen as being the optimum. The effects of W tube treatment and in some experiments an in situ pre-purified W modifier were also studied. Aqueous Solutions The maximum pyrolysis temperatures for the studied systems in aqueous solutions are compiled in Table 3. Further details and specific features of the pyrolysis curves are presented below.In contrast to pyrocoated tubes signal losses for inorganic and methylated As species were not observed at low temperatures in W-treated tubes; however a minimum in the pyrolysis curves between 800 and 1000°C was clearly estab- lished. A similar effect has been observed for Se species uiz. Se" SeIV and SeV1 in the presence of 20 pg of Ni and 2.5 pg of Cu as m0difie1-s.~~ The observed losses were avoided when a sufficiently high thermal pre-treatment temperature was applied and the original absorbance signal was restored. Various factors could be responsible for the observed 'dip' in the pyrolysis curves. Insufficiently high atomization tempera- tures differences in the heating rates and different starting pyrolysis temperatures are unlikely to influence the signal because the 'dip' appears when higher atomization tempera- tures and a 'cool down' step temperature programme are used.The transformation of As species into volatile forms which would be expected to evaporate at temperatures higher than 500 "C probably accounts for the losses observed. The earlier appearance of losses of the As species AS'" AsV MMA and DMA in W-treated tubes in comparison with pyrocoated tubes could be explained by hindrance of the formation of intercal- ation compounds between As species and graphite in W-treated tubes. The original signal is restored at 1000-1200°C; this could be related to the specific properties of tungsten carbides. It is known that tungsten carbides are oxidized at temperatures higher than 1000°C; moreover they can play the role of oxygen carrier.Ig It is possible that the resulting tungsten oxide prevents further reduction of oxygen-containing As species by binding them and delaying their vaporization.The more volatile AsB and AsC are not thermally stable in the W-treated tube; however the sensitivity is still about 1.5 times better (see Table2) than that for the pyrocoated tube alone. It should be noted that W-treated tubes show a marked drift in sensitivity after about 60 firings. The original sensitivity is restored when the procedure for W tube treatment described under Experimental is re-applied. In order to provide additional insight into the reasons for the 'dips' in the pyrolysis curves in W-treated tubes experi- ments were carried out with an in situ purified W modifier.In the presence of 20 pg of W the pronounced 'dip' in the pyrolysis curves for inorganic As species MMA and DMA is not observed and the maximum loss-free pre-treatment tem- peratures are of the same order as in the W-treated tubes. Even the very volatile AsB and AsC are stable up to 1000 and 900°C respectively. The stabilizing effect could be due to the embedding of As species (by means of isomorphous substi- tution or simple embedding) in the modifier oxides or oxocar- bides which would lead to a decrease in the As partial pressure and reduce analyte losses during pyrolysis. Table 3 Maximum pyrolysis temperatures for As species in the presence of various modifiers Species P yrocoated * W - treated 20 Pg w PdCl Pd(NO3 )2 W-PdCl W-Pd (NO,) As"' 800 1600t 1600 1300 1300 1500 1500 AsV 800 1500t 1500 1300 1300 15002 1500 MMA 800 13001- 1300 400§ 1300 14002 1400 DMA 800 1200t 1250 4 w 1200 14002 1400 AsB 200 200 1000 400s 1200 1400 1400 AsC 120 200 900 40@ 1200 1400 1400 *Corresponding to the maximum integrated absorbance signal in the pyrolysis curves.?A 'dip' in the pyrolysis curves was observed. $A signal increase at low temperature was observed. $The signal diminished gradually with temperature. Journal of Analytical Atomic Spectrometry October 1996 Vol. 1 1 999The observed differences in the behaviour of inorganic As species MMA and DMA in pyrocoated and W-treated tubes compared with that of AsB and AsC could become a major source of systematic errors when undigested natural samples are analysed directly in a graphite furnace.As"' and AsV are usually used to prepare the standard solutions whereas AsB is the predominant species in some biological materials par- ticularly marine tissues. Differences in the behaviour of As species were observed in pyrocoated tubes in the presence of PdC1 and Pd(N03) as modifiers. Inorganic species are stabilized efficiently up to 1300"C regardless of the form in which the Pd modifier is introduced into the tube. For organically bound species a pronounced plateau up to 1200-1300 "C was found in the presence of Pd(N03) whereas a gradual decrease in the integrated absorbance signal with increasing pyrolysis temperature was found for MMA DMA AsB and AsC in the presence of PdC1 as is illustrated in Fig.3 for MMA. The low efficiency of the stabilizing action of Pd as its chloride could be explained by the formation of volatile chloride-containing analyte species which are not stabilized by the modifier.28 The experimental results of Kamiya et using thermogravimetry differential thermal analysis and X-ray diffraction indicate a transformation interval of 675-755°C for the decomposition of PdC1 to Pd. Thus residual chloride would be expected to have an effect on analyte stabilization and atomization. Depending on the form of the Pd modifier differences in the peak profiles were found for the various As species. In the presence of PdCl a small additional peak was observed which was negligible for the inorganic species and more pronounced for the organic species (see Fig.4). An increase in the pyrolysis temperature led to a reduction of this additional peak. Its appearance could be related to the action of residual chloride in the system; moreover the magnitude of this small additional peak correlates with the volatility of the chloride- 120 T t""l 8 "t $ 40 d 20 0 \ h,- A 0 - l I 0 200 400 600 800 1000 1200 1400 1600 1800 Temperature /OC Fig.3 Comparison of the stabilizing effect of Pd as its chloride or nitrate for MMA in pyrocoated tubes (A) and (B) and also W-treated tubes (C) and (D) respectively d) c e 8 2 -0 Pd as a chloride .05; 3.0 Time/s v . e 2 e F c a Fig. 4 Overlay of the absorbance profiles for 80 pg 1-' MMA in the presence of 6 pg Pd as its nitrate or chloride respectively; 20 pl injections methanol solutions containing species of the As forms and increases in the order As"' AsV < MMA < DMA < AsB < AsC.In the presence of Pd as its nitrate no double peaks were observed. In addition the absorbance signal was delayed in comparison with that in the presence of chloride. The effects discussed above are illustrated for MMA in Fig. 4. The sensitivity for the analytes in pure aqueous solutions using PdC1 was lower by approximately a factor of 1.5 as compared with that obtained in the presence of Pd(N03),. Similarly differences in the performance of Pd prepared from its chloride or nitrate have been reported for other element^.^^'^^'^^ The combination of W tube treatment and addition of Pd(N03) leads to a further increase in the pyrolysis tempera- ture for all the species studied and to an improvement in the sensitivity by a factor of 1.5-2.A considerable improvement in the performance of the PdCl modifier was obtained with W-treated tubes (Fig. 3). A pronounced enhancement of the maximum pyrolysis tempera- ture and decrease in the influence of residual chloride were found. It is difficult to explain the increase in the absorbance signal observed at low temperatures for AsV MMA and DMA in the presence of PdCl in a W-treated tube. Redistribution of the analyte during pre-treatment the availability of chloride in the system and the conversion of PdCl into an active metal form could be some of the reasons. However a complete explanation is not yet possible. The W treatment also results in an earlier appearance of absorbance profiles for As species in the presence of the Pd modifiers in comparison with pyrocoated tubes.This could be due to the formation of smaller Pd droplets during pyrolysis whereas in pyrocoated tubes more pronounced agglomeration of the particles on the graphite would be expected as has been shown in ref. 32. It should be noted that the maximum loss-free pyrolysis temperatures for As in W-treated tubes in the presence of Pd are of the same order as those for As"' in the presence of a mixed W+Pd modifier." This finding is not unexpected because Pd is thermally stabilized in W-treated platforms33 in a manner similar to that observed for the corresponding mixed W + Pd modifier.34 The analytical performance of a Pd modifier in the presence of chloride is of practical importance since many real matrices/ digests may contain an excess of chloride. Experimental results show that treatment of the tubes with W results in an improvement in their action as a thermal stabilizer and makes them less prone to chloride interferences particularly in the determination of AsB and AsC.The influence of the modifier on the characteristic masses of the As species studied is summarized in Table 2. For AS'" AsV MMA and DMA the sensitivity is similar for both pyrocoated and W-coated tubes. Palladium as its nitrate improves the sensitivity whereas Pd as its chloride produces a similar or poorer response for the As species in comparison with pyro- coated tubes. Tungsten treatment of the tube and addition of Pd further improves the characteristic masses and Pd(NO,) in combination with W provides the best sensitivity.The behaviour of the As species is very similar in the presence of the Pd(NO,) modifier in both pyrocoated and Table4 Characteristic masses m in picograms for different As species in methanolic solutions W-treated tubes (W) and W-treated tubes plus Pd modifier ( W-Pd) As species Modifier As"' AsV MMA DMA AsB AsC W 31 53 53 55 69 93 W-Pd 25 38 32 36 37 56 1000 Journal of Analytical Atomic Spectrometry October 1996 Vol. 11W-treated tubes; therefore Pd( NO3) is more suitable than the other modifiers studied for stabilizing the As species particularly in undigested biological samples. The role of Pd(NO,) in a graphite furnace as a thermal stabilizer also approaches the requirement of an 'analyte species isoformer'. =.'00 A Ej - a 60 8 3 * 20 a 40 D Methanolic Solutions The problems accompanying the analysis of organic scllutions in ETAAS are summarized in ref. 35. Taking into account the thermal pre-treatment resiilts for aqueous solutions only the role of W treatment and Pd(N03) modifier was studied. Tungsten treatment produces a s tabiliz- ing effect for inorganic and methylated As species. Analytical pre-treatment losses ('dips') were not observed and the pyrolysis curves showed a pronounced plateau (see Fig. 5). AsB and AsC in methanol were not stabilized in W-treated tubes and the maximum thermal pre-treatment temperature decreased in the order As"' As" > MMA > DMA >> As 13 AsC. The significant differences in the stabilizing action of Pd(NO,) in aqueous and methanolic solutions of the As species were not ascertained completely but the sensitic ity was impaired (see Table 4).The decrease in the sensitivity could be explained by spreading of injected sample further along the surface of the tube and as a result the As is less efficiently volatilized and atomized from the cooler region towards the ends of the tube. A decrease in the diffusion path of the species before leaving the tube36 could also be a factor. The addition of a Pd modifier in W-treated tubes not only assists the efficient thermal stabilization for all six As specie but also leads to a delay in the absorbance signal as illustrated in Fig. 6 for AsB. The peaks appear in a relatively short time and are narrower in W-treated tubes for each of the As species studied.The addition of Pd produces a shift in the peak maximum signal broadening and an increase in the integrated absorbance. - - - - - 120 T A'\ B 0 4 ..I---I 0 200 400 600 800 1000 1200 1400 1500 1800 Temperature /OC Fig.5 Pyrolysis curves for DMA in W-treated tubes (A) anct in the presence of 6 pg Pd in W-treated tubes (B). T 2300°C; methanolic solutions 0.50 3000 9 -. 2 4- 2 4 2 9 $ e F -0.05 * 0 0 30 Time/s Fig.6 Effect of W treatment and Pd modifier on the absorbance profiles for 100 pg 1-' AsB; 20 pl injections methanolic solutions Table 5 Determination of As"' and AsB in model solutions Concentration/pg I-' Total As As"' AsB Expected 50 57 107 Determined 61 50 111 Expected 10 99 109 Determined 29 82 111 Expected 90 11 101 Determined 92 16 108 Attempt at in situ Speciation The observed differences in the behaviour of AsB and AsC and of the inorganic and methylated As species suggested that in situ speciation of As in a W-treated graphite furnace might be possible.An attempt was made to separate and determine AsB and As"' in solutions containing these species in ratios of 1 1 1 9 and 9 1. AsB is the major As species in marine samples. As"' is the most toxic of the As species. These two species were chosen because of their similar sensitivity and a pronounced difference in their volatility during the pyrolysis stage in W-treated tubes. The procedure involves total As determi- nation in W-treated tubes with pyrolysis at 200°C and measurement of the As"' concentration after pyrolysis at 1400°C.The calibration is performed with AsB and As"' standards respectively. The results obtained are presented in Table 5. As can be seen the values obtained for total As are in good agreement with the expected values. However the results for As"' are systematically higher by about 20% for an As"' AsB ratio of 1 1 by about 50% for an As"' AsB ratio of 1 9 and in good agreement for an As"' AsB ratio of 9 1. The presence of significant amounts of AsB under the conditions used for As"' determination makes the complete separation and correct determination of As"' impossible. Unfortunately the tested procedure does not allow the in situ speciation of As"' and AsB in W-treated tubes. CONCLUSIONS The behaviour of several As species in ETAAS was investigated.It was found that their properties depend on the oxidation and binding states of As. On the basis of analyte behaviour in pyrocoated and W-treated tubes the As species can be classified conditionally into two groups. The first group includes inor- ganic and methylated As species. They exhibit a maximum in their pyrolysis curves at 600-700°C in pyrocoated tubes. A pronounced 'dip' in their pyrolysis curves between 800 and 1000°C in W-treated tubes and subsequent restoration of the original absorbance signal in aqueous solution is observed. The second group includes AsB and AsC which are more volatile in pyrocoated tubes and pre-treatment losses are possible even at drying temperatures as low as 120-200°C. As a result of the treatment of the tube with W the sensitivity can be improved but no significant stabilization is observed.In the presence of Pd as its nitrate or chloride the behaviour of the As species depends on the form of the modifier. Palladium as its nitrate efficiently stabilizes each of the species up to 1200-1300 "C whereas Pd as its chloride exhibits effective stabilizing action only for the inorganic species. The absorbance signal diminishes gradually at temperatures higher than 700°C for organic As species. The W treatment of the graphite tubes in combination with the Pd( N03)2 modifier produces excellent stabilizing action for all the species studied regardless of their oxidation and binding states and also improves and 'levels-off' the sensitivity (except for AsC). Journal of Analytical Atomic Spectrometry October 1996 VoE.11 1001Thus Pd(N03)2 in both pyrocoated and W-treated tubes is more suitable than the other modifiers for stabilizing As species in various undigested biological and environmental samples and is closer to the requirements of an 'analyte species isoformer'. A procedure for the in situ separation and determination of As"' and AsB in a model solution in W-treated tubes by ETAAS was tested; however because of the presence of signifi- cant amounts of AsB under the conditions used for As"' determination complete separation and determination of these species was not possible. REFERENCES 1 I 7 3 4 5 6 7 8 9 10 11 12 Seiler A. and Sigel H. Handbook on Metals in Clinical and Analytical Chemistry Marcel Dekker New York 1989.Florence T. M. in Trace Element Speciation Analytical Methods and Problems ed. Batley D. E. CRC Press Boca Raton FL Tsalev D. L. J. Anal. At. Spectrom. 1994 9 405. Brinkman F. E. Blair W. Jevett K. and Iverson W. J. Chromatogr. Sci. 1977 15 493. Chau Y. and Wong P. Fresenius' J. Anal. Chem. 1991 339 640. Environmental Analysis Using Chromatography Interfaced with Atomic Spectroscopy eds. Harrison R. M. and Rapsomanikis S. Ellis Horwood Chichester 1989. Ebdon L. Hill S. Walton A. P. and Ward R. W. Analyst 1988 113 1159. Chung C. H. Iwamoto E. Yamamoto M. and Yamamoto Y. Spectrochim. Acta Part B 1984 39 459. George G. M. and Frahm L. J. J. Assoc. Of. Anal. Chem. 1986 69 838. Sturgeon R. Willie S. Sproule G. Robinson S. and Berman S. Spectrochim. Acta Part B 1989 44 667.Zhang L. Ni Z.-m. and Shan X.-q. Spectrochim. Acta Part B 1989 44 339. Shuttler I. L. Feuerestein M. and Schlemmer G. J. Anal. At. Spectrom. 1992 7 1299. 1989 pp. 319-341. 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 Krivan V. and Arpadjan S. Fresenius' Z . Anal. Chem. 1989 335 743. Russeva E. Havezov I. and Detcheva A. Fresenius' J. Anal. Chem. 1993 347 320. Tsalev D. L. Mandjukov P. B. and Stratis J. A. J. Anal. At. Spectrom. 1987 2 135. Larsen E. H. J. Anal. At. Spectrom. 1991 6 375. Tsalev D. L. Slaveykova V. I. and Mandjukov P. B. Spectrochim. Acta Rev. 1990 13 225. Slaveykova V. I. and Tsalev D. L. Anal. Lett. 1992 23 1921. Volynsky A. Zh. Anal. Khim. 1987 42 1541. Byme J. P. Hughes D. M. Chakrabarti C. L. and Gregoire D. C. J. Anal. At. Spectrom. 1994 9 913. Bulska E. Grobenski Z. and Schlemmer G. J. Anal. At. Spectrom. 1990 5 203. Slaveykova V. I. and Tsalev D. L. Spectrosc. Lett. 1991,24 139. Slaveykova V. I. and Tsalev D. L. J. Anal. At. Spectrom. 1992 7 365. L'vov B. V. Polzik L. and Yatsenko L. F. Talanta 1987,34 141. Styris D. Prell L. and Redfield D. Anal. Chem. 1991 63 503. Slaveykova V. I. and Tsalev D. L. Spectrosc. Lett. 1992,25,221. Welz B. Schlemmer G. and Voellkopf U. Spectrochim. Acta Part B 1984 39 501. Tesfalidet S. and Irgum K. Anal. Chem. 1988 60 2031. Kamiya N. Noshino K. and Ota K. Nippon Kagaku Kaishi 1988 12 1944. Voth-Beach L. M. Spectroscopy 1987 2 21. Voth-Beach L. M. and Shrader D. E. J . Anal. At. Spectrom. 1987 2 45. Qiao H. and Jackson K. Spectrochim. Acta Part B 1991 46 1841. Tsalev D. L. D'Ulivo A. Lampugnani L. Di Marco M. and Zamboni R. J . Anal. At. Spectrom. 1995 10 1003. Tsalev D. L. Slaveykova V. I. and Mandjukov P. B. in 5 Colloquium Atomspektrometrische Spurenanalytik ed. Welz B. Bodenseewerk Perkin-Elmer Uberlingen 1989 pp. 178-205. Komarek J. and Sommer L. Chem. Listy 1988 82 1151. Gilutdinov A. Kh. and Fishman I. S. Spectrochim. Acta Part B 1984 39 171. Paper 6/01 305 H Received February 23 1996 Accepted June 13 1996 1002 Journal of Analytical Atomic Spectrometry October 1996 Vol. 11

ISSN:0267-9477    

DOI:10.1039/JA9961100997

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Rapid Determination of Selenium in Soils and Sediments Using Slurry Sampling- Electrothermal Atomic Absorption Spectrometry Journal of Analytical Atomic Spectrometry IGN ACIO LO PEZ-G A R C ~ A M ATEO s ANCHEZ-M ER LO s AND MANUEL H ERN ANDEZ-c ORDOB A* Department of Analytical Chemistry Faculty of Chemistry Unimmity of Murcin E-30071 -Murcia Spain An electrothermal atomic absorption spectrometric procedure for the rapid determination of selenium in soil and sediment samples is presented. The samples were suspended in concentrated (40% m/v) hydrofluoric acid containing 1% m/v nickel nitrate and then injected into the electrothermal atomizer. A fast heating programme with no conventional ashing step was used. Platform atomization at 2300°C and Zeeman-effect background correction were used.Calibration was performed directly by using aqueous standards prepared in a 10% v/v hydrofluoric acid solution. The determination limit was 0.1 pg g-' of selenium when using the maximum recommended slurry concentration of 10% m/v. The results obtained for five certified reference materials show the reliability of the procedures. Keywords Electrothermal atomic absorption spectrometry; fast heating programme; slurry sampling; selenium; chemical modijication; soil; sediment The dual nature of selenium as both an essential and potcntially toxic element has led to growing interest in the development of rapid and reliable methods for its determination. The high sensitivity of ETAAS makes it a suitable technique for such a purpose. However the determination involves a number of problems.Most importantly volatile selenium compounds may be lost during the dissolution stage or during tlie pre- atomization heating steps inside the atomizer' and in addition severe spectral and chemical interferences may hinder the determination.'-I5 It is therefore not surprising that in spite of the great interest focused on slurry sampling-ETAAS methods,I6 there are few references to their use in selenium determination. To the best of our knowledge following the work of Ebdon and Parry on coal analy~is,'~ the slurry sampling ETAAS approach has only been used for determining selenium in fly ash,'* rnilk,lg flour2' and seafood.21 The amount of selenium in soils is highly variable the levels mainly reflecting the weathering of parent materials all hough anthropogenic contributions may also be of importance.The mean level for a variety of soils has recently been reported22 to be about 0.5 pg g-'. This is a level within the range of ETAAS a technique which is widely used nowadays in labora- tories dealing with environmental analysis. Based on the fact that slurry sampling-ETAAS methodology is particularly suitable for trace element analysis in soils and a study was made to develop rapid procedures for determining selenium in this type of sample. The results reported here show that good analytical results can be otrtained with a considerable saving of time provided that Zeeman- effect background correction is used and that the samples are previously slurried in a concentrated hydrofluoric acid schhon.* To whom correspondence should be addressed. EXPERIMENTAL Instrumentation Measurements were made with an ATI-Unicam (Cambridge UK) 939 QZ atomic absorption spectrometer equipped with a GF90 electrothermal atomizer. The spectrometer was pro- vided with both a Zeeman-based and a deuterium-arc back- ground corrector. The source of radiation was a selenium hollow cathode lamp (Photron Victoria Australia) operated at 10 mA. Measurements were performed at 196.0 nm using a spectral bandwidth of 1.0 nm. For comparative purposes measurements were obtained with a Perkin-Elmer (Norwalk CT USA) Model 1 lOOB atomic absorption spectrometer equipped with a deuterium-arc background corrector and an HGA-400 electrothermal atomizer the source of radiation in this instance being an electrodeless discharge lamp operated at 200 mA from an external power supply (Perkin-Elmer System 2).Pyrolytic graphite coated graphite tubes with platforms obtained from ATI-Unicam (reference 9423 393 95 19 1 ) and Perkin-Elmer (reference B050 5057) were used. Argon was used as the purge gas the flow rate being 300 ml min-' except during atomization when the flow was stopped. For preparation of the sample a Fristch Pulverisette (Idar-Oberstein Germany) agate ball-mill of 80 ml capacity was used. Reagents A selenium (IV) standard solution (1000 pg ml-') was obtained from Panreac (Barcelona Spain) and diluted as necessary to obtain working standards. Concentrated hydrofluoric acid (40% m/v) and all other chemicals used were obtained from FIuka (Buchs Switzerland).High-quality water obtained using a Milli-Q water purification system (Millipore; Milford MA USA) was used exclusively. Plasticware [poly(propylene)] of the type commonly used for clinical purposes was used for storing and handling the solutions or suspensions containing hydrofluoric acid. Procedure Five soil and sediment samples with certified selenium contents were used throughout this work. The samples were first ground for 15 min using the ball-mill and the resulting powders were kept in tightly closed plastic containers until analysis. No sieving was carried out. The suspensions were prepared by weighing the samples directly into Eppendorf tubes and then adding 1000 pl of the concentrated hydrofluoric acid solution containing 1 YO Ni(N03)2.6H20.The suspensions were manu- ally shaken (caution concentrated hydrofluoric acid solutions cause severe burns. Eyes and skin must be protected) and then left for 15 min after which the suspensions were again shaken and immediately 10 p1 aliquots were taken and manually injected into the electrothermal atomizer. The experimental Journal of Analytical Atomic Spectrometry October 1996 Vol. 11 (1003-1006) 1003Table 1 Instrumental and furnace parameters Wavelengt h/nm Bandpass/nm Background correction Atomizer type Injection volume/pl Calibration range/pg 1-' Characteristic mass/pg Slurry volume/ml Slurry concentration (YO m/v) Chemical modifier Hydrofluoric acid concentration (Yo v/v) 196.0 1 .o Zeeman-effect Pyrolytic graphite coated 10 5-200 25 1 0-10 1 % m/v Ni( N03)2*6H20 100 graphite tube + L'vov platform Step TemperaturePC RampPC s-l Hold/s 1 400 20 5 2* 2300 0 4 3 2650 0 4 * The flow of argon was stopped during the atomization step.conditions used for the preparation of the suspensions as well as the instrumental settings of the spectrometer are summarized in Table 1. The fast heating programme recommended is also given in Table 1. Calibration was performed using aqueous standards prepared in a solution containing 10% v/v concen- trated hydrofluoric acid and 1% m/v nickel nitrate. RESULTS AND DISCUSSION Halls26 demonstrated several years ago that the drying and ashing stages normally included in conventional furnace pro- grammes can be replaced by a modified drying stage thus shortening and simplifying the heating cycle.This fast heating m e t h ~ d o l o g y ~ ~ . ~ ~ was used throughout the present work. The temperature and the holding time for the modified drying stage are dependent on the characteristics of the electrothermal atomizer used. A 400°C drying temperature with a heating rate of 20°C s-l and a hold time of 5 s was used in all the experiment^.^^ In spite of the relatively low temperature inside the atomizer at the end of the modified drying stage prelimi- nary experiments showed that a chemical modifier was neces- sary because of pre-atomization loss of analyte. Consequently nickel nitrate was incorporated in the suspension medium. In addition it is known that when suspensions prepared from samples with a high silica content are atomized the graphite atomizer severely deteriorate^.^'*^' Bendicho and de Loos- V ~ l l e b r e g t ~ ' .~ ~ were the first to overcome this difficulty by adding hydrofluoric acid to the suspension medium. This simple approach which has subsequently been ~ s e d ~ ~ > ~ ~ * ~ ~ for the analysis of other samples with a high silica content was followed here. For preliminary experiments the samples were suspended in a medium containing 5% v/v concentrated hydrofluoric acid in addition to the nickel modifier and deuterium-arc background correction was used. Matrix Effects Preliminary experiments showed severe matrix effects. The atomization profile of aqueous selenium standards was unaffected by the presence of hydrofluoric acid but the analyt- ical signals obtained from suspensions prepared from sediment samples containing similar amounts of selenium were very low.The matrix effect was so severe that the signal obtained from a 120 pg 1-l selenium solution virtually disappeared when solid sediment was added to the solution to make a 0.5% m/v suspension (Fig. 1). In an attempt to clarify the exact nature of this depressing effect a number of experiments were per- formed by preparing suspensions from pure A1203 Fe203 O.* It J . . . 1 0.0 0.2 .................. . . . . . :: 0.0 L L :::It A I (c;l . . . . . . . . ................ g 0.2 5 0.0 0 1.5 1 .o 0.5 - . . . . 0.OIl 4 .... + I ..... (el 2.0 1.5 . . . . . . 1.0 1 _ . . !.A I - . . 0.5 ..... 0.0 0 1 2 3 4 Time/s Fig. 1 Absorbance uersus time curves for selenium (solid line) and background (dashed line) for (a) a 120 pg 1-' selenium solution in a 5% v/v hydrofluoric acid medium containing the nickel modifier; (b) (c) ( d ) and (e) as above in the presence of 0.5 0.4 0.6 and 3% m/v of sample BCSS- 1 Fe203 A1203 and SiO respectively.Deuterium-arc background correction. Si02 and soil samples always using an aqueous selenium solution containing 5% hydrofluoric acid and the nickel modi- fier as the suspension medium. The concentration of the suspensions prepared from oxides was chosen in such a way that a typical 5% m/v soil suspension was simulated. The selenium atomization profiles some of which are shown in Fig. 1 suggested that the severe depressing effect was mainly due to the presence of solid SO2. In addition the presence of iron was shown to lead to a significant baseline drift as a consequence of its spectral interference which cannot be corrected by the deuterium device.In order to avoid this baseline drift Zeeman-effect background correction was used for further experiments. Since the above results suggested that silica was mainly responsible for the matrix effect the possible benefit of increas- ing the hydrofluoric acid concentration in the suspension medium was considered. As expected the matrix depressing effect considerably decreased when the amount of hydrofluoric acid was increased. Fig. 2 shows the analytical signal obtained from 3% m/v suspensions prepared from NIST SRM 2711 Montana Soil sample when solutions containing different concentrations of hydrofluoric acid were used as the suspension medium. It is clear that a very high percentage of hydrofluoric acid was needed to obtain the maximum and constant signal.The use of concentrated hydrofluoric acid solutions which contain about 40% by mass of the pure acid can be hazardous for the operator and several attempts were made to avoid its use. However because the level of selenium in soils and sediments is very low suspensions containing a high percentage of solid matter must be used which necessitates the direct use of concentrated acid solutions. This can be seen in Fig. 3(a) where results for some of these experiments are shown. A number of suspensions were prepared from the Montana Soil sample with different amounts of solid matter using 500 p1 of concentrated hydrofluoric acid. After 15 min of contact between the sample and the acid 500 pl of pure water were 1004 Journal of Analytical Atomic Spectrometry October 1996 Vol.110.30 . 0.00 1 1 0 20 40 60 80 100 Hydrofluori acid concentration (?h v/v) Fig. 2 Effect of the percentage of concentrated hydrofluoric ;acid in the suspension medium on the selenium peak area. A and B mr:asure- ments obtained 15 min and 24 h after the suspension prephxation respectively added to reduce the acid concentration and measurements were taken from the whole slurry (curve A) and aftx the suspensions had been filtered through chromatographic mem- brane filters from the supernatant solution (curve B). The results showed that interference could only be overcoine for slurry concentrations below 5 %. The signal obtained from the supernatant solutions was as expected unaffected by the matrix effect which appeared to be solely due to the slurried solid matter.On the other hand when the samples were directly slurried in concentrated hydrofluoric acid there was a linear relationship [Fig. 3(b)] between the percentage of solid matter in the slurry and the analytical signal obtained from the whole suspension (15 min after its preparation). Other experiments performed using dilute hydrofluoric acid solutions as the suspension medium and mild heating treatments with a heating block or a microwave oven produced the same result. In order to avoid the matrix effect completely and relialdy for high slurry concentrations (up to 10% m/v) it is necessary for the samples to be slurried directly in concentrated hydrofuoric acid. This does not involve the total dissolution of the sample and in fact selenium was only partially transferred to the liquid phase by the action of hydrofluoric acid.By filtering the solid residues through chromatographic membrane filter and measuring selenium in the supernatant solutions it was shown that only about 70-75% of the total selenium present was extracted. This value increased to 80-85% when the suspen- sions were filtered 24 h after preparation. On the other hand one important practical point must be 0.3 (a) I I I I I I 1 o.2; 0.1 00 0 2 4 6 8 1 0 Slurry concentration (./o) Fig.3 Effect of the slurry concentration on selenium peak area. (a) Using a 50% v/v hydrofluoric acid solution as the suspension medium. A and B signals obtained from the whole suspensions and from the supernatant solutions respectively; (b) using concentrated hydrofluoric acid.A B and C signals from suspensions prepared from samples SRM 2711 PACS-1 and BCSS-1 respectively noted since the use of concentrated hydrofluoric acid solutions may at first sight appear to be harmful to the electrothermal atomizer. In order to obtain the results summarized here nearly 1000 heating cycles were performed and the quartz windows of the atomization device were periodically checked. No signs of deterioration were noted nor did the pyrolytic graphite platforms show any indication of damage or prema- ture ageing. The presence of 1% nickel nitrate in the suspension medium proved suitable for avoiding premature loss of the analyte.On the other hand the atomization temperature was varied in the range 1900-2600 "C maximum analytical signals being obtained at 2300°C. Calibration and Results In order to confirm that concentrated hydrofluoric acid com- pletely overcame the matrix effect suspensions were prepared from five soil and sediment samples and standard additions calibration graphs were obtained. The slopes of these graphs (Table 2) did not show significant differences (95% confidence level) from those obtained with aqueous standards which validates the direct and more simple calibration. It is important to point out that working selenium standards prepared in concentrated hydrofluoric acid proved unstable. For this reason aqueous standards were prepared in a dilute (10% v/v) hydrofluoric acid solution.Using the conditions recommended a characteristic mass of 25 pg Se was obtained. The detection limit was calculated using the criterion based on three times the standard deviation. For this purpose the standard deviation obtained from a plot of the analytical signal against the slurry concentration for a representative sample was used instead of the standard devi- ation of the blank.35 Using this approach the detection limit for a 10% m/v suspension was calculated to be 0.03 pg g-' Se which corresponds to a quantification limit of 0.1 pg 8-l. Table 3 summarizes the final results obtained for selenium determination in five certified soil and sediment samples using both the standard additions method and the direct calibration against aqueous standards prepared in 10% v/v hydrofluoric acid solution. There were no significant differences (95% confidence level) between the results obtained and the certi- fied values.Once the final experimental conditions had been optimized and taking into account that the deuterium-arc corrector is still the most commonly used correction device in most routine analysis laboratories the need to use Zeeman-effect back- ground correction was reconsidered. This was facilitated by the fact that the spectrometer used was provided with both correction systems which allowed a more realistic comparison than that based on two different spectrometers. Fig.4 shows selenium atomization profiles obtained from a 5% m/v suspen- sion prepared from the PACS-1 sample. Although the slopes of the calibration graphs obtained from aqueous standards were similar when either correction system was used (1.78 x s pg-' 1 for Zeeman and deu- terium-arc devices respectively) it is evident that Zeeman- and 1.89 x Table 2 Slopes of the standard additions calibration graphs Sample Slope* + S / I O - ~ s pg-' I Aqueous solution 1.78 & 0.02 NIST SRM 2709 (San Joaquin Soil) 1.75 f 0.04 NIST SRM 2711 (Montana Soil) 1.81 k0.03 NIST SRM 2704 (River Sediment) 1.78 0.03 1.74 & 0.04 NRCC PACS-1 (Marine Sediment) NRCC BCSS-1 (Marine Sediment) 1.82 f 0.03 * Each graph was constructed with four points. Three measurerncnts were obtained from each point.Journal of Analytical Atomic Spectrometry October 1996 Vol. 11 1005Table 3 Results for the determination of Se in certified reference materials Content* +s/pg g-' Sample Description Certified Direct calibration Standard additions 0.43 f 0.06 0.44 0.02 0.43 -t 0.02 NRCC BCSS-1 Marine Sediment NRCC PACS-1 Marine Sediment 1.09f0.11 1.03 f 0.07 1.05 f 0.06 NIST SRM 2704 River Sediment 1.12 f 0.05 1.1 3 f 0.05 1.10 f 0.05 NIST SRM 2709 San Joaquin Soil 1.57k0.06 1.60 f 0.04 1.57 k 0.05 NIST SRM 2711 Montana Soil 1.52 k0.14 1.49 +_ 0.03 1.51 k0.04 * Mean k standard deviation for five suspensions... . . . . . . _ U - 0.20 1 0.8 5 0.15 0.6 0.10 0.4 I fbi I k- AT-/ . . . ... '. . . . . . . .. . . . 0 1 2 3 Time/s Fig. 4 Absorbance versus time curves for selenium (solid line) and background (dashed line). A 5% m/v PACS-1 Marine Sediment slurry prepared in concentrated hydrofluoric acid containing 1 YO nickel nitrate was used.(a) Deuterium-arc background correction system; (b) Zeeman-effect background correction system effect background correction is mandatory. This was confirmed by using another spectrometer (Perkin-Elmer Model 1100B) equipped with a deuterium-arc background corrector and a selenium electrodeless discharge lamp. The slope of the cali- bration graph was similar (1.80 x lop3 s pg-' 1) but the atomiz- ation profiles of selenium obtained from suspensions proved unsuitable for quantification. The foreseeable improvement in performance due to the electrodeless discharge lamp could not be checked on the ATI-Unicam instrument since at present this radiation source is not available. CONCLUSION The direct suspension of soil and sediment samples in concen- trated (40% m/v) hydrofluoric acid permits the rapid and reliable determination of selenium by ETAAS without the need for a dissolution stage.The acid acts as a true chemical modifier by simplifying the matrix during the heating cycle. Suspensions as concentrated as 10% m/v can be introduced into the electrothermal atomizer with no damage to the graphite material or to the quartz windows of the atomizer. However personal safety precautions when handling concen- trated hydrofluoric acid solutions must be taken into account. Zeeman-effect background correction is mandatory. The results confirm that slurry methodology is particularly suitable when dealing with trace analysis in soil and sediment samples since background levels are not high and the time-consuming and hazardous dissolution stage is avoided altogether.This work was financially supported by the Spanish DGICYT (Project PB93-1138). M. Sanchez-Merlos acknowledges a fel- lowship from DGICYT. REFERENCES 1 Radziuk B. and Thomassen Y. J. Anal. At. Spectrom. 1992 7 397. 2 Saeed K. and Thomassen Y. Anal. Chim. Acta 1981 130 281. 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 Saeed K. and Thomassen Y. Anal. Chim. Acta 1982 143 223. Carnrick G. R. Manning D. C. and Slavin W. Analyst 1983 108 1297. Bauslaugh J. Radziuk B. Saeed IS. and Thomassen Y. Anal. Chim. Acta 1984 165 149. Welz B. Schlemmer G. and Mudakavi J. R. J . Anal. At. Spectrom. 1988 3 93. Nham T. T. and Brodie K. G. J. Anal. At. Spectrom.1989,4,697. DoEekalova H. DoEekal B. Komarek J. and Novotny I. J. Anal. At. Spectrom. 1991 6 661. Aller A. J. and Garcia-Olalla C. J. Anal. At. Spectrom.. 1992 7 753. Garcia-Olalla C. and Aller A. J. Anal. Chim. Acta 1992,259,295. Kumpulainen J. and Saarela K.-E. J . Anal. At. Spectrom. 1992 7 165. Welz B. Bozsai G. Sperling M. and Radziuk B. J. Anal. At. Spectrom. 1992 7 505. Johannessen J. K. Gammelgaard B. Jerns O. and Hansen S. H. J. Anal. At. Spectrom. 1993 8 999. Laborda F. Viiiuales J. Mir J. M. and Castillo J. R. J. Anal. .4t. Spectrom. 1993 8 737. Volynsky A. B. and Krivan V. J . Anal. At. Spectrom. 1996 11 159. Bendicho C. and de Loos-Vollebregt M. T. C. J. Anal. At. Spectrom. 1991 6 353. Ebdon L. and Parry H. G. M. J. Anal. At. Spectrom. 1988,3 131.Bradshaw D. and Slavin W. Spectrochim. Acta Part B 1989 44 1245. Wagley D. Schmiedel G. Mainka E. and Ache H. J. At. Spectrosc. 1989 10 106. Bendicho C. and Sancho A. At. Spectrosc. 1993 14 187. Lopez-Garcia I. Viiias P. Campillo N. and Hernandez- Cordoba M. J. Agric. Food Chem. 1996 44 836. Neal R. H. in Heavy Metals in Soils ed. Alloway B. J. Chapman and Hall London 2nd edn. 1995 ch. 12. pp. 260-283. Hinds M. W. Latimer K. E. and Jackson K. W. J. Anal. At. Spectrom. 1991 6 473. Bermejo-Barrera P. Barciela-Alonso C. Aboal-Somoza M. and Bermejo-Barrera A. J. Anal. At. Spectrom. 1994 9 469. Lopez-Garcia I. Sanchez-Merlos M. and Hernandez- Cordoba M. Anal. Chim. Acta 1996 328 19. Halls D. J. Analyst 1984 109 1081. Hoenig M. and Cilissen A. Spectrochim. Acta Part B 1993 48 1003. Halls D. J. J. Anal. At. Spectrorn. 1995 10 169. Eames J. C. and Matousek J. P. Anal. Chem. 1980 52 1248. Miiller-Vogt G. and Wendl W. Anal. Chem. 1981 53 651. Bendicho C. and de Loos-Vollebregt M. T. C. Spectrochim. Acta Part B 1990 45 679. Bendicho C. and de Loos-Vollebregt M. T. C. Spectrochim. Acta Part B 1990 45 695. Lopez Garcia I. Arroyo Cortez. J. and Hernandez-Cordoba M. J. Anal. At. Spectrom. 1993 8 103. Lopez-Garcia I. Arroyo Cortez J. and Hernandez-Cordoba M.. Anal. Chim. Acta 1993 283 167. Miller J. C. and Miller J. N. Statistics for Analytical Chemistry Ellis Horwood Chichester 3rd edn. 1993 p. 11 5. Puper 6/03660K Received Muy 28 1996 Accepted July 10 1996 1006 Journal of Analytical Atomic Spectrometry October 1996 Vol. 1 1

ISSN:0267-9477    

DOI:10.1039/JA9961101003

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Journal of Analytical Atomic Spectrometry ERRATUM I I Waste Reduction in Inductively Coupled Plasma Mass Spectrometry Using Flow Injection and a Direct Injection Nebulizer JEFFREY S . CRAIN AND JAMES T. KIELY Analytical Chemistry Laboratory Chemical Technology Division Argonne National Laboratory 9700 South Cuss Avenue Argonne IL 60439-4831 USA Journal of Analytical Atomic Spectrometry 1996 11 525 On page 526 Table 2 should appear as follows Table 2 ICP-MS analyses obtained with FI-DIN and CPN Analyte concentration (mean f s; n = 18) Sample Nebulizer Blank$ FI-DIN CPN CPN CPN CPN CPN SRM 1643C$ FI-DIN ICPMS-2Al FI-DIN Waste #377 FI-DIN Waste #40Y FI-DIN ( Ni* 3.2 f 0.4 0.8 f 0.2 57f9 80 f 20 I l k 1 13f2 0.8 f 0.4 0.4f0.1 1.44 f 0.03 0.6 f 0.2 Cd 0.03 f 0.01 0.0 12 & 0.008 13.0 & 0.3 12.9 & 0.4 10.0 f 0.2 10.1 f 0.2 1.3 1 f 0.03 1.34f0.03 0.065 f 0.002 0.067 & 0.002 Pbt 0.07 f 0.04 0.1 1 f0.05 37f 1 38f3 10.0 f 0.2 11f1 1.6f0.1 1.62 k 0.05 0.7 1 & 0.02 0.71 f 0.03 U 0.003 f 0.004 0,008 +_ 0.007 nd§ nd§ 9.9 f 0.2 9.9 f 0.2 3.20 +_ 0.09 3.19 & 0.09 0.62 f 0.02 0.60 f 0.01 *Based on 58Ni. ?Based on 208Pb. $Concentrations expressed in pg 1- '. §Element not detected. YConcentrations expressed in mg 1-'. Journal of Analytical Atomic Spectrometry October 1996 Vol. I I 1007

ISSN:0267-9477    

DOI:10.1039/JA9961101007

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

CUMULATIVE AUTHOR INDEX JANUARY-OCTOBER 1996 Abou-Shakra Fadi R. 61 Acheson Barbara M. 765 Adams Freddy C. 201 Akatsuka Kunihiko 69 Akiyama Masayuki 69 Alberini G. 731 Alcantara Elena 917 Allen Lori B. 526 Alvarado Jorge S. 923 Angeles Quijano M. 407 Apostoli Pietro 519 Arbore Philippe 929 Arruda Marco A. Z. 169 Ascanelli M. 731 Ashino Tetsuya 577 Augagneur Sylvie 71 3 Barinaga Charles J. 317 Barnes Ramon M. 343 Barnett David A. 877 Barren James 279 Barrero Moreno Josefa M. 929 Bavazzano Paolo 5 19 Beato Emilio Romero 37 Begerow Jutta 303 913 Belkin M. 491 Bengtson Arne 829 Benoy D. A. 623 Benzo Zully 447 Bergdahl Ingvar A. 735 Berndt Harald 703 Besteman Arthur D. 479 Betti Maria 855 Bin He 165 Birolleau Jean-Claude 759 Blades M. W. 43 Bloxham Martin J. 145 509 Bogaerts A.841 Bohlen Alex Von 537 Botto Robert I. 675 Brebion Sophie 497 Bredendiek-Kamper Susanne Brenner I. B. 91 Brockhoff Carol A. 504 893 Broekaert Jose A. C. 661 739 Brown Francine Byrdy 633 Byrne John P. 549 Caimi Stefano 773 Caldwell Kathleen L. 339 Camara Carmen 407 Camblor Juan Pablo 591 Canals Antonio 949 Can0 Pavon J. M. 107 Caroli Sergio 773 Caruso J. A. 491 633 Cavalchi B. 731 Ceulemans Michiel 201 Chakrapani G. 8 15 Chamberlain Isa 504 Chapple Graeme 549 Chen Zhongxing 805 Chenery Simon 53 177 Chiappini Remo 497 Chirinos Jose 253 Chizhik Andrei S. 649 Ciani I. 731 Clevenger Wendy L. 393 Coan P. 731 Concepcion Perez-Conde M. Conrad Gregoire D. 765 Cordero Bernard0 Moreno 37 537 797 407 Crain Jeffrey S. 523 Creed John T. 504 893 Dams Richard 543 Dannecker Walter 723 Daskalova Nonka 567 Davoli V.731 Dawson John B. 967 Debrah Ebenezer 127 De Gendt Stefan 937 De Grootte F. 623 Denoyer Eric R. 127 Depalma Jr. Patrick A. 483 De Regt J. M. 623 Di Marco Marco 979 939 Dietze Hans-Joachim 643 661 Ding W-W. 225,421 Dolan Scott P. 307 Donard Olivier F. X. 871 Dorfman Ethel 811 D’Ulivo Alessandro 979. 989 Dunemann Lothar 303 913 Ebdon Les 427 Efstathiou Constantinos E. 31 Eiden Gregory C. 317 Einhauser Thorsten J. 747 El-Hagrasy Maha A. 379 El-Kourashy Abdel-Ghany 379 Ellis Lyndon A. 259 Elmahadi Hayat 99 Erickson Mitchell D. 92 3 Fang Zhaolun 1 Fell Gordon S. 297 Feng Xinbang 287 Fernandez Sanchez Maria L. Fey F. H. A. G. 623 Flint Colin D. 53 Foulkes Michael 427 Frame Eileen Skelly 279 Fryer Brian J. 805 Gachanja Anthony 145 Galanski Markus 747 Gallego Mercedes 169 Gallus Stefan M.887 Garcia De Torres A. 107 Garcia Alonso Ignacio J 929 Garcia Sanchez Soledad 37 Garcia Albert0 Menendez 561 Gauthier Gilles 787 Gavrieli Ittai 8 11 Gentscheva Galja 567 Gercken Berthold 371 Gijbels R. 841 Goodall Phillip S. 57 469 Greenway Gillian M. 907 Grtgoire D. C. 359 Greibrokk Tyge 117 Grubb Anders 735 Giinther Detlef 899 Gutierrez Ana Maria 407 Halicz Ludwik 811 Hall Gwendy E. M. 779 787 Hang Wei 835 Haraguchi Kensaku 69 Harrison W. W. 835 849 Hartmann C. 237 Hasanen Erkki K. 365 Hasegawa Noriyuki 513. 601 Hayashi Yasuhisa 513 601 Helliwell T. R. 133 Hernandez-Cordoba Manuel 723 571 1003 Hernandis Vicente 949 Heumann Klaus G. 887 Hieftje G. M. 401 613 Hill Steve J. 145 509 Holclajtner-Antunovic Ivanka Holderbeke Mirja Van 543 Horlick Gary 877 Houk R.S. 247 Hutton Robert 187 Hwang Tarn-Jiun 139 353 Ilkov Atanas 313 Imai Shoji 513 601 Ince Ahmet T. 967 Infante Heidi Goenaga 571 Ingeneri Kristofor 849 Ivanova Elisaveta 567 Iversen Bent Schack 591 Jackson Simon E. 805 899 Jager Ralf 661 Jakubowski Norbert 797 Jalkanen Liisa M. 365 Jarvis Kym E. 917 Jiang Shiuh-Jen 139 353 555 Jin Qinhan 331 Jin Qun 331 Johnson Stephen G. 57 469 Jonkers J. 623 Kabil Mohamed A. 379 Karayannis Miltiades I. 595 Karpati Peter 773 Katoh Takunori 69 Kelly S. A. 133 Kenneth Marcus R. 821 Keppler Bernhard K. 747 Kerl Wolfgang 723 Kiely James T. 523 Kim S. 91 Kingston H. M. 187 Klenerman L. 133 Klockenkamper Reinhold 537 Klockow Dieter 537 Knight Kevyn 53 KO Fu-Hsiang 413 Koch Lotar 929 Koh Lip Lin 585 Kojima Isao 607 Koller Dagmar 187 907 Koppenaal David W.317 Kotrebai Mihaly 343 Krivan Viliam 159 371 Krushevska Antoaneta 343 Kumamaru Takahiro 11 1 Lafontan Silvyane 759 Lampugnani Leonard0 979 Lasztity Alexandra 343 Lau Nancy 479 Lerat Yannick 213 Leroy Maurice J. F. 997 Li Gangqiang 401 Li Jason 683 Liang Feng 331 Liaw Ming-Jyh 555 Littlejohn D. 207 463 Liu Don-Yuan 479 Liu Huiying 307 Lobinski Ryszard 193 713 871 Lonardo Robert F. 279 Longerich Henry P. 805 899 Lopez-Garcia Ignacio 1003 Lorthioir Stephane 759 Luan Shen 247 325 989 Lund Walter 943 Luterotti Svjetlana 973 Lutman A. 731 Lyon Thomas D. B. 297 Magnuson Matthew L. 504 Mahoney Patrick P. 401 MaloviC Gordana 325 Maquieira Angel 99 Marcus R. Kenneth 483 Masera Eric 213 Massart D.L. 149 237 Matveev Oleg I. 393 Mauchien Patrick 21 3 Mazzoli A. 731 McCandless Tom E. 667 McCrindle Robert I. 437 McGaw Brian 297 Medina Bernard 713 Michel Robert G. 279 Moens Luc 543 Mohamad Ghazi A. 667 Moissette A. 177 Monod Jean-Louis 193 Montaser Akbar 307 Montero Thais 447 Montoro Rosa 271 Mordoh Leah S. 393 Moser-Veillon Phylis B. 727 Mostafa M. A. 455 Murillo Miguel 253 Murty D. S. R. 815 Naka Hirohito 359 Nakamura Seiji 69 Nelms Simon M. 907 Nicolaou Georgos 929 Nishiyama Yasuko 601 Nogay Donald J. 187 OHanlon Karen 427 Ohtsuka Hideyuki 69 Olson L. K. 491 633 Panayi Antonia 591 Pang Ho-Ming 247 Parsons Patrick J. 25 Paschal Daniel C. 339 Patriarca Marina 297 Patterson Kristine Y. 727 Pavel Jiri 371 Pedersen-Bjergaard Stig 117 Pelchat Jean-Claude 779 787 Pelchat Pierre 787 Penninckx W.237 Perez P a v h Jose Luis 37 Perico Andrea 519 Perkins C. V. 207 463 Pilidis George A. 595 Pilon Fabien 759 Pinto Carmelo Garcia 37 Piperaki Efrosini A. 31 Pollmann Dagmar 797 849 Poluzzi V. 731 Polydorou Christoforos K. 3 1 Puchades Rosa 99 Quentmeier Alfred 537 Quintal Manuelita 447 Rademeyer Cornelius J. 437 Raspopovic Zoran 325 Rastegar Faramarz 997 Rayman Margaret P. 61 Remy Bernard 213 Rhoades Jr. Charles B. 751 Risnes Anna 943 Riter Ken L. 393 893 Journal of Analytical Atomic Spectrometry October 1996 Vol. 11 1009Roberts David J. 231 259 Roberts N. B. 133 Rosendahl Kerstin 519 Rubio Marcelo 123 Ruette Fernando 447 Ruiz Joaquin 667 Sabbioni Enrico 591 Sabine Becker J. 643 661 723 Sadler D. A. 207 463 Saito Kengo 513 601 Sanchez Hector J.123 Sanchez-Merlos Mateo 1003 Sanz-Medel Alfredo 561 571 Saprykin Anatoli I. 643 Schelles Wim 937 Schickling Claudia 739 Schmitt Vincent O. 193 Schram D. C. 623 Schutz Andrejs 735 Schwartz Robert S. 307 Seifert Gotthard 643 Senofonte Oreste 773 Shepherd T. J. 177 Shkolnik J. 91 Sierraalta Anibal 447 Siitonen Paul H. 526 Siles Cordero M. T. 107 Sivaganesan Manohari 504 Slaveykova Vera I. 997 Slavin Walter 25 Slavova Petranka 567 Smeyers-Verbeke J. 149 237 Smith Benjamin W. 393 479 Snook Richard D. 967 Stalikas Constantine D. 595 Stewart Ian I. 877 Stuewer Dietmar 797 Sturgeon R. E. 225 421 Sun Han-Wen 265 Szpunar Joanna 193 713 Taillade Jean-Michel 497 Takada Kunio 577 Takayanagi Asako 607 Tao Guanhong 1 Tao Shiquan 11 1 Taylor Daniel B. 187 Taylor Richard P.765 Thomaidis Nikolaos S. 31 Thompson Harold C. Jr. 526 Thompson Michael 53 Ting Bill G. 339 Tittes Wolfgang 797 Todoli Jose L. 949 Trentini P. 731 Treshchalov Aleksei B. 649 Tripkovic Mirjana 325 Tsalev Dimiter L. 979 989 689 Turfeld Martha 9 13 Turner Andrew D. 231 Tyson Julian F. 127 Uria Enrique Sanchez 561 Vaive Judy E. 779 787 Valcarcel Miguel 169 Van Der Mullen J. A. M. 623 Van Grieken Rene E. 937 Vance Donald E. 861 Vanhaecke Frank 543 Vankeerberghen P. 149 Vanko D. A. 667 Veillon Claude 727 Vklez Dinoraz 271 Velichkov Serafim 567 Vereda Alonso E. I. 107 Vill Arnold A. 649 Volynsky Anatoly B. 159 Wagatsuma Kazuaki 957 Wagner 11 Eugene P. 689 Walsh H. P. J. 133 Ward Neil I. 61 Wayne David M. 861 Wee Yeow Chin 585 Weir D. G. 43 Westheide Jochen Th.661 Wildhagen Dieter 371 Williams John G. 917 Wills Julian D. 917 Winefordner James D. 393 479 Wittmeier Adolph 287 Wong Ming Keong 585 Worsfold Paul J. 145 509 Wu Shaole 287 Xu Shukun 1 Yaiiez Jorge 703 Yang Jinfu 739 Yang Karl X. 279 Yang Kuei-Lin 139 Yang Li-Li 265 Yang Mo-Hsiung 413 Yang Wenjun 331 Ybaiiez Nieves 271 Yoshida Thomas M. 861 You Jianzhang 483 Yuzefovsky Alexander I. 279 Zamboni Roberto 979 989 Zander A. 91 Zhang De-Ciang 265 Zhang Hanqi 331 Zhao Yu-Hui 287 Zhe-Ming Ni 165 Zhou Chao Yan 585 Zhu Jim J. 675 Zochowski Stan W. 53 Zong Yan Y. 25 689 101 0 Journal of Analytical Atomic Spectrometry October 1996 Vol. 11

ISSN:0267-9477    

DOI:10.1039/JA9961101009

出版商:RSC    

年代:1996

数据来源: RSC