摘要:

Plasma Spectrometry and Molecular Information* Plenary Lecture Journal of Analytical 1 Atomic OLIVIER F. X. DONARD AND RYSZARD LOBINSKI Laboratoire de Photophysique et Photochimie Molkculaire CNRS Universitk de Bordeaux I F-33405 Talence France Trends in species-selective analysis by GC and HPLC with plasma source element selective detection are discussed and illustrated with examples of applications. The potential of plasmas for the acquisition of auxiliary molecular information and conditions for its reliability are highlighted. The need for interaction of plasma techniques and classical organic MS techniques is emphasized. Keywords Plasma spectrometry; gas chromatography; high- performance liquid chromatography; speciation Metals exist in the environment or in biological systems as part of larger molecular structures.' Each and every such structure is the result of a series of biochemical processes defined by nature.They carry messages that are vital for the evaluation of the bioavailability of the metal and hence of its toxicity or essentiality and for the understanding of mechan- isms controlling biological life and regulatory pathways. The question of how to access this molecular information (speci- ation) has become one of the most pertinent and challenging issues to modern analytical chemistry during the past decade because of its impact on environmental chemistry eco- and clinical toxicology and to food and energy Classical approaches to speciation analysis have included the use of an intrinsically species-selective technique such as an electrochemical technique or use of soft-fragmentation mass or tandem mass spectrometry.Although these techniques can work well for standard solutions they usually fail for sub- ngg-' concentrations of an analyte in the presence of a complex matrix. As the metal of interest forms only a tiny part of the whole species detection focused on the polyatomic fragment usually lacks selectivity and suffers from high back- ground noise and isobaric matrix interferences. The two key issues for a successful analysis (sensitivity and selectivity) cannot be properly addressed other than by targeting the metal itself. Plasmas especially the ICP have become established in the field of inorganic analytical chemistry as a successful way of overcoming chemical interferences matrix signal suppression and the poor excitation efficiency associated with a chemical combustion flame." There is however an intrinsic contradic- tion between the plasma and the molecular information required.Plasma spectrometry needs a cloud of individual atoms (or atomic ions) in order to provide qualitative and quantitative information on trace metals present in the matrix being studied. As a consequence the structure of the compound has to be destroyed and the molecular information is lost during the determination stage. The paradox is that the sensitivity attainable makes plasma techniques the only ones which at the moment are capable of answering the demands of environmental and clinical trace metal species-selective analysis. *Presented at the 1996 Winter Conference on Plasma Spectrochemistry Fort Lauderdale FL USA January 8-1 3 1996. Spectrometry Information on trace-metal speciation can be addressed at three levels which is schematically illustrated in Fig.1. They include ( i ) time-resolution of analyte species prior to their entering the plasma; ( i i ) empirical formula determination; and (iii) controlled fragmentation of polyatomic ions. The isotopic or wavelength snapshot contributes to the confirmation of the identity of the target element. Most attention has been given to on-line chromatographic or electrophoretic separation of analyte species prior to the plasma. This has resulted in a new generation of analytical techniques (termed hyphenated tandem coupled and hybrid) which are able to discriminate between the different forms of a metal or metall~id."-'~ The increasing need for species-selective analytical information and the emerging market for instrumentation capable of delivering this information are making speciation analysis one of the fields capturing most interest in modern analytical chemistry.The purpose of the present paper is to discuss briefly the status of plasma techniques in trace element species-selective analysis of environmental and biological materials and to indicate the most important trends in development. DISCUSSION Species-selective Analysis Using Plasmas In species-selective analysis the plasma is used only at the final determination stage. Therefore the signal obtained carries information pertinent to the element alone but not for the whole compound. The key to a successful analysis is the quality of the separation of the species prior to their entering the plasma.Identification is based on a parameter that is external to the plasma; this is usually retention time in the case of Signal f intensity I Molecular m h ratio Empirical formula \ \ Elemental m/z ratio or emission wavelength Fig. 1 Pathways to molecular information in spectrochemical analysis Journal of Analytical Atomic Spectrometry September 1996 Vol. 11 (871 -876) 871chromatographic methods or electrophoretic mobility in the case of capillary electrophoresis. Simultaneous multiwavelength (multimass) monitoring enables the determination of several species containing differ- ent metals (metalloids) during one chromatographic run.In addition it renders possible the acquisition of more information on the eluted species in terms of the presence of other analyte lines (confirmation of elemental identity) and the mutual ratio of the constituent elements in the detected compounds (deter- mination of empirical formula). Gas chromatography with plasma detection The attractive features of capillary GC such as the high resolution and the availability of plasma source detectors with absolute detection limits at the low femtogram level have made GC with plasma source detection the technique of choice for many applications especially those related to anthropog- enic environmental contaminants. This has been supported by the commercial availability of a GC-MIP-AES system which can be easily tuned to the desired ~avelength.l~-~* The principal disadvantage of GC is the usual need for conversion of analytes into thermally stable and volatile species (derivatization).This limits the field of application to a restric- ted number of compounds and has largely contributed to GC-based techniques reaching a stage of maturity in relation to speciation analysis. The novel aspects of work now appear- ing usually concern newer designs of the interface between the GC and ICP-MS instruments on-line preconcentration prior to GC and developments at the sample preparation stage. The basic advantage that ICP-MS offers over MIP-AES is its higher tolerance to water vapour and carbon dioxide. This is particularly critical in the determination of gaseous trace contaminants in air following their cryogenic preconcentration. Cryotrapping in liquid nitrogen is very efficient but it is not selective for the analyte compounds; the thermodesorption process often does not discriminate between the organometallic analytes and matrix interferents (COz H20).Purge and trap thermal desorption ICP-MS for the determination of volatile organometallics answers the demand for a simple robust and automated system for the analysis of the gaseous air fraction for organometallic compounds. A chromatogram is shown in Fig. 2 of urban air sampled on a busy car park indicating the presence of tetraalkyllead and volatile mercury compounds. Liquid chromatography with plasma detection HPLC has usually been used in this context but a series of low-pressure columns for the speciation of redox forms and some organometallics has recently been marketed.21 Me,,Pb (2.62 ng mS) A 175 000 i 150 000 3 125000 I/ Etpb (i.5 ng m-3) II I \ s 2 loo 0 75 50 25 M%Et,Pb 000- - Pb 0 1 2 3 4 5 6 Time/min Fig.2 Chromatogram obtained for urban air (Bordeaux area) by cryotrapping (- 196 "C)-thermal desorption GC-ICP-MS; mass spec- trometer; SCIEX ELAN 5000 (Perkin-Elmer); and 0.1 m3 of air ~ I Serum 168 126 2 84 42 5 .- E v) c c .- - CT) I I I I 4.277 5.348 1.069 2.139 3.208 0' Retention time/min Fig.3 Chromatogram of a human serum sample obtained with reversed-phase HPLC-ICP-MS by monitoring m/z = 82 (selenium). Mass spectrometer SCIEX ELAN 5000 (Perkin-Elmer); column PRP-1 (Hamilton); sample diluted five times with water; injection volume 100~~1; mobile phase 2% methanol in water lOmmoll-' C,H,,S03Na at pH 4.5; flow rate 1 ml min-'.The retention time of the signal corresponds to selenocystine The need for desolvation of the mobile phase makes ICP the only source able to cope directly with this problem. ICP-MS is more popular than ICP-AES because of the much higher sensitivity and has been comprehensively evaluated as an element-selective LC detector by the group of Caruso.16 The major drawbacks of ICP-MS are the vulnerability to salt and organic solvent content of the LC mobile phase and polyatomic ion interferences which hamper ICP-MS analysis for transition metals (m/z<80). Axial ICP-AES can offer a cost-effective alternative to the light transition metals. In addition the tolerance of ICP-AES to the salt and organics load is higher than that of ICP-MS because of the higher plasma powers used and the absence of the sampler-skimmer interface.Further development of a high-resolution ICP-MS detector is still limited by the high cost of the instrumentation. The present challenge in HPLC-ICP-MS coupling is how to improve the sensitivity of detection for metal-containing compounds; conventional nebulization can hardly be con- sidered suitable because the losses in the nebulizer are too large which thus degrades the sensitivity. This technique is sufficient for some applications however as demonstrated in Figs 3 and 4. A high-pressure nebulizer (HPN) with a desolv- ation chamber is more s ~ i t a b l e . ~ ~ ' ~ Even more efficient .- t I I I Retention time/s Fig.4 Chromatographic profile of copper in a commercial mixture of metallothioneins (Sigma Rabbit liver No. M7641). 1 Copper bound to metallothionein fraction; and 2 Cu2+. Column Progel-TSK PWXL (4.0 cm x 6.0 mm); 30 ng ml-' of thionein (as Cu); mobile phase 30 mmol 1-' TRIS-HC1 pH 7.1 0.7 ml min-'; sample 50 pl injected in 22 mmol 1-' 2-mercaptoethanol in mobile phase. Mass spectrometer SCIEX ELAN 6000 872 Journal of Analytical Atomic Spectrometry September 1996 Vol. 11(although at the expense of the amount of sample handled) are the microconcentric (MCN) and direct injection nebuliz- ers (DIN). The MCN is a fixed concentric nebulizer that fits easily onto a conventional spray chamber. It operates at low sample flow rates (0.03-0.3 ml min-l) using normal aerosol carrier gas pressures (80 psi; 1 psi = 6894.76 Pa) and performs well as the interface between low pressure chromatography and ICP-MS.Shown in Fig. 5 is an example of the speciation of toxic arsenic forms in canned crab meat leachate. The DIN. is a microconcentric nebulizer with no spray chamber; it nebulizes the liquid sample directly into the central channel of the ICP torch. Compared with conventional nebuliz- ers it offers 100% sample introduction efficiency at flow rates of 10-100 pl min-'. The low dead volume (<2 pl) and the absence of a spray chamber minimize post-column peak broad- ening and memory effects. The DIN provides absolute detec- tion limits that are superior by 1-2 orders of magnitude to those obtained with conventional nebulizers. Concentration detection limits were however not Another advan- tage is that the sampler and skimmer orifices do not clog to any great extent with carbon.24 A wider range of applications of microbore HPLC-DIN-ICP-MS coupling in clinical chem- istry are clearly imminent.Reliability of Molecular Information The reliability of the information obtained is based on the reliability of the atomic information obtained and on the quality and reproducibility of the separation. Atomic information in plasma spectrometry That the signal measured definitely corresponds to the expected element being measured is crucial. Although the selectivity of plasma source spectrometers used as chromatographic detec- tors is superior to flame ionization conductivity or UV detec- tors a few concerns do arise.In AE detection the confirmation of elemental identity requires the ability to carry out a wavelength scan at the peak 1 2 0 200 400 600 800 1000 Retention time/s Fig. 5 Speciation of arsenic in commercial canned crab meat leachate by low-pressure anion-exchange chromatography-MCN-ICP MS. Mass spectrometer; HP 4500 (Hewlett-Packard); column ANX3206 (CETAC Omaha USA); sample in OSmmoll-' malonic acid; load time 55 s. Elution program 0-300 s 0.9 mmoll-' malonic; 300-400 s 5 mmol 1-' malonic acid pH 8; above 400 s 50 mmol 1-' malonic acid pH 8. Mobile phase flow rate 30 pl min-'. 1 As"' (ca. 5 ng ml-'); and 2 As" (ca. 0.5 ng ml-') apex. In the Hewlett-Packard AE detector a 40nm wide emission spectrum can be taken at the peak apex and checked for the presence of the analyte characteristic atomic emission lines and their ratio (fingerprint).20 In MS validation is obtained by the isotopic pattern.Problems can occur for monoisotopic elements e.g. arsenic which is known to suffer from an overlap by the ArCl' ion. High-resolution MS is the expensive alternative. The potential advantages of simpler background spectra with an He plasma remain largely unful- filled. In fact NO' and 02+ problems are even more severe because of the lower gas flow rates and easier entrainment of air in many He plasmas. Also the lower temperature can in some cases prevent dissociation of polyatomic ions. The most reliable proof of identity would be obtained by splitting the effluent to both ICP-AES and ICP-MS instru- ments which are considered orthogonal techniques.However apart from the fact that few laboratories have both techniques coupled to chromatography the detection limits of ICP-AES may not be sufficient for some analytes and also some analyte elements may not be determinable by ICP-MS. Compound identijication In the simplest case the identification of analyte compounds is based on the retention times of their corresponding standards. This can be influenced by the column matrix and separation conditions but within the same laboratory an acceptable reliability can be achieved. Standards however are not gener- ally available with the exception of the case of a small number of anthropogenic pollutants but not necessarily products of their degradation as well. Some compounds can be custom- synthesized and the use of MS (electron impact chemical ionization or electrospray) to confirm their structural identity is then mandatory.In practice there is often a need for the prediction of retention times by interpolation or extrapolation of those obtained for the standards available. In GC the use of the Kovat's retention index is a valid choice. Actual retention times of the mixed compounds can be confirmed using transal- kylation mixtures. Once identified the compound should be synthesized and its actual retention time then confirmed. Quantification of the mixed compounds is made by interp- olation of the signal intensity using the standards available. Discrimination of the response with the boiling-point of the analyte must be taken into account.In metal-containing pro- tein analyses a role similar to that of the Kovats index is played by the use of relative molecular mass markers in size- exclusion chromatography. Attribution of the relative molecu- lar masses to the proteins was performed after calibration of the column with for example myoglobin (17.8 kDa) thyroglo- bulin (669 kDa) alcohol dehydrogenase (150 kDa) albumin (67 kDa) which act as the relative molecular mass standards.25 The reliability of identification based on the retention time is highly dependent on the quality of separation. The better the separation the lower the probability that an unknown and unaccounted for compound elutes at the retention time of a particular analyte compound. An example is given in Fig. 6 that illustrates the decreasing possibility of interference when moving from packed to capillary columns.The chromatogram obtained using a capillary column shows between the peaks of tetrabutylin (SnBu,) and tripentylmethyltin (SnMePe,) a well separated small peak (dipentyldipropyltin SnPr,Pe,) due to a standard impurity. This peak was not shown by packed column GC analysis. Whereas the presence of an unknown compound is unlikely to be a serious source of mis-identification in the case of analysis for anthropogenic organometallic pollutants it does become a serious problem when the complexity of the species (relative molecular mass) increases and a variety of similar Journal of Analytical Atomic Spectrometry September 1996 VoZ. 11 873molar ratios of the molecular structure of the compound which is usually not the case as has been demonstrated by Huang et a1.27 Another problem is the need for an undisturbed compound band in the plasma at a given moment and the need for relatively sensitive detection of all the compound- forming elements at compromise plasma operating conditions.Complexity of the matrix leads to a high background especially on the carbon and hydrogen channels which affects the precision and often does not allow for element quantification. Under optimum conditions typical precision values vary between 2 and 5% which works well for the calculation of the formula of compounds with elemental ratios equal to whole numbers but fails for higher hydrocarbons e.g. C2oH44 where a 5% error can easily change the resulting calculated formula.The principal area of applications is low relative molecular 0 1 3 5 7 9 1 1 - 3 5 1 9 1 1 mass halogen- and sulfur-containing compounds. In general empirical formulae can be determined only for standard solu- tions as has been demonstrated for tin.28 The method fails for 5 1 l 1 1 1 Retention time/min Fig.6 Effect of resolution of the separation on the accuracy of AAS; and (b) capillary GC-AES (HP-1 column). 1 Me2Pe2Sn; 2 Pr,Sn; 3 Pr,PeSn; 4 Bu,Sn; 5 MePe,Sn; 6 Bu,PeSn; 7 Bu2Pe2Sn; 8 BuPe,Sn; and 9 Pr,Sn. Me methyl; Pr propyl; Pe pentyl; Bu butyl unknown compounds that are present in extracts of real determination of carbon and hydrogen owing to the very high backgrounds Present and low concentrations of the species of interest. Although precision cannot match that of classical microanalysis the measurements can be carried out on-line and considerably lower amounts of sample are required for analysis (pico- or nanograms instead of milligrams).analysis. (a) Packed (3y0 OV-l On Chromosorb)-quartz furnace samples. The reason is the virtual impossibility of an accurate structures could exist (metal-binding protein analysis). For this reason there is a need for orthogonal (independent) separation techniques for the same applications. The most suitable choice seems to be the parallel use of GC and HPLC (for low relative molecular mass compounds) and that of HPLC and capillary Alternative Ion Sources for Species-selective MS Analysis zone electrophoresis-(CZE) for high relative molecular mass compounds. An alternative approach is the confirmation of the molecular identity by MS or IR spectroscopy.However some incompati- bility in terms of the detection limits and matrix interference can occur. The advances in the sensitivity of electron impact MS for GC detection and electrospray MS for LC and CZE detection are likely to create valuable auxiliary tools for species-selective analysis. Since both the ICP and the MIP usually destroy any molecular structure and focus on the element itself the quest for other sources is on-going. The glow discharge has been reported to ionize a sample to a sufficient degree for MS detection at ultratrace levels and also to be capable of providing structural information at these levels.29 An alternative is the modification of atmospheric pressure ionization (API) techniques which have become a landmark in the field of LC detectors in organic analysis.These tech- Quantitative information Regarding the quantitative information required it is possible that a species will respond differently in the plasma depending on the organic part of the molecule. Internal standard cali- bration could thus turn out to be inappropriate. Use of a calibration curve or the method of standard additions requires the use of standards which are rarely available. A method to solve this problem is on-line conversion of the analyte species into forms which would respond in the same way. An example is the use of post-column HG for the determination of arsenic and selenium species. Harsh UV radiation thermochemical or microwave-assisted techniques are required in order to convert all of the species containing the element in question into the same form and thus to assure a species-independent response in the plasma.26 Empirical Formula Determination by Plasma Spectrometry The atomic information obtained as consequence of the plasma-induced radiation (or ion current) can be readily translated into molecular information in terms of the empirical formula of the compound.For this the mutual concentration ratios for all elements are necessary which in practice means the need for the simultaneous determination of all the elements that form the compound in question. The need for the quantification of a non-metal (carbon hydrogen sulfur and halogens) limits the choice of detectors to those operating with an He plasma and in particular to MIP-AES.The key to success is the independence of elemental niques produce gas-phase analyte ions directly from solution by using electrospray or ion-spray (a pneumatically assisted version of electrospray) interfaces. The problem is that under the operating conditions usual in commercial instruments the ionization energy is not sufficient to achieve efficient fragmen- tation down to the elemental ion. Recently a combination of modifications to the curtain gas the declustering potential and the geometry of the vacuum interface region of the mass spectrometer was reported to enhance the fragmentation effi~iency.~' In such a system ions from the spray chamber are sampled into the MS system through a dry N2 curtain gas which assists in the declustering of ion-solvent aggregates.The degree of adjustability of the potential difference between the orifice and skimmer (referred to as the declustering potential) allows control of the declustering reactions in the atmosphere through to the vacuum region of the mass spectrometer. As a function of this potential the system is able to operate either as an elemental analyser or as a molecular detector. A spectrum of selenomethionine obtained in the molecular and elemental mode taken at the apex of the chromatographic peak is illustrated in Fig. 7 . In Fig. 7(a) it is shown that in the molecular mode there is a noisy background and poorer sensitivity than in elemental mode [Fig. 7 ( b ) ] . The modification of the instrument allowing for its running in the elemental mode still needs to be optimized as shown by the residue of the molecular peak at m/z 196.The availability of information on the relative molecular mass of analyte species and further (by placing a second MS) fragmentation information are likely to make electrospray (ion spray) atmospheric pressure ioniz- ation (AP1)-MS a serious competitor especially for elements that suffer serious interferences in ICP-MS (e.g. As Se and Cr). 874 Journal of Analytical Atomic Spectrometry September 1996 Vol. 116x105 5x105 4x105 3x105 T~ 2x10~ 2 1x10~ G 5 100 120 140 160 180 200 A 'z 122.0 145.8 168.0 ':to 2.0X1O6 - i 90 120 150 180 m/z Fig. 7 Snapshot on the selenomethionine peak in an HPLC chroma- togram by ion spray MS. Sample selenomethionine (50 pg ml-' in H,O); Mass spectrometer modified API-300 (SCIEX); column LichroCART 250-4 (Merck); mobile phase 50% methanol 1 ml min-' Split 1 200; post-column acidification with HC1 to pH 3.4.(a) Molecular mode (orifice voltage 40 V) and (b) elemental mode (orifice voltage 400 V) Towards Analysis of Real-world Materials Very few studies have provided data on 'real-world' materials with reasonably well-substantiated analytical accuracy. Whereas few problems exist with water speciation analysis of sediments and biomaterials requires isolation of analytes from the matrix without changing their chemical form. As the organometallic species in sediments are not built into the crystal lattice leaching is the approach of choice. They can however be incorporated in tissues of a living organism so biological materials must be solubilized.Alkaline hydrolysis with tetramethylammonium hydroxide (TMAH) is faster and cheaper than enzymatic hydrolysis with a mixture of lipase and protease. Until recently the procedures reported for sediments have not only been extremely time consuming (they take from 1 h to 2 d ) but also are usually inefficient in terms of recovery of analytes and are unreliable. As shown by Zhang et ~ l . ~ ' only three out of ten sample preparation methods described in the literature for the analysis of sediments were able to recover more than 90% of tributyltin (TBT) whereas none were able to recover monobutyltin (MBT) in a non-erratic and reproduc- ible manner. High scattering of the results due to leaching problems prevented certification of MBT by the Community Bureau of Reference (BCR).32 However it would seem that a lot of these problems can be solved by the recently proposed application of low-power focused microwave-assisted technol- ~ g y .~ ~ In contrast to the common high-temperature and press- ure acid attack a focused low-power microwave field was demonstrated to be successful for quantitative extraction of organometallic species from the matrix without destroying the carbon-metal bonds.34 Also in the case of the species-selective analysis of biological materials the impact of recent developments in microwave technology for solvent extraction leaching and enhancement of extraction kinetics with the simultaneous preservation of the organometallic moiety and thus structural information is a milestone regarding the sample preparation step.Microwave assisted TMAH hydrolysis allows dissolution of a biotissue within a few minutes without degradation of the a n a l y t e ~ . ~ ~ Studies however have been restricted to small organotin and 8o 1 2 5 I 6 Time/min Fig. 8 Capillary GC-AES chromatogram of a fish tissue (NIES 11) extract obtained after simultaneous acetic acid hydrolysis ethylation with NaBEt and extraction into nonane in a microwave field (3 min). Column DB-210. 1 BuEt,Sn; 2 Pr,EtSn; 3 Bu,Et,Sn; 4 Bu,EtSn; and 5 Bu4Sn organomercury compounds. The real challenge lies in the analysis of high relative molecular mass compounds often with unknown structures which require not only sophisticated and up-to-date separation procedures but also their efficient isolation from the matrix.Hitherto sample preparation pro- cedures have been based on leaching with water or a neutral buffer followed by (ultra) centrifugation with recoveries of 50-80%. While the identity of the leached compounds often raises some questions virtually nothing is known of the identity of the compounds remaining in the solid phase. Another challenge is the integration of sample decompo- sition derivatization of polar analytes and their extraction into a non-polar solvent. Hydrolysis with acetic acid carried out in a low-power focused microwave field in the presence of sodium tetraethylborate and nonane was recently shown to shorten radically the sample preparation time for GC determi- nation of organotin compounds in biological materials.36 After a 3 min reaction time ethyl derivatives of mono- di- and tributyltin and triphenyltin were quantitatively found in the supernatant organic phase that was injected onto a capillary GC column (Fig.8). The use of a slightly polar column instead of the commonly recommended non-polar phases allowed a clean-up step to be avoided and thus to shorten the overall procedure. CONCLUSIONS The rapidly growing interest to plasma spectroscopists in speciation of elements has been followed by a rapid develop- ment of more sophisticated analytical techniques. Plasma (both ICP and MIP) techniques have an already established position in the field of species-selective analysis. They are considered to be the best detection option for modern separation tech- niques.The molecular information accessible is however based on the retention time of the metal-containing compounds which implies the availability of standards. Simultaneous AE detection permits the acquisition of data on carbon hydrogen sulfur and the halogens of the same time as that for a metal and metalloid and thus offers the possibility of structure identification. Plasma MS techniques are evolving towards the possibility of controlling the fragmentation of a species of interest so that at least some molecular information is retained. On the other hand recent advances in organic LC-MS make imminent the advent of a molecule smashing device to obtain the bare metal. Where will these approaches meet? Journal of Analytical Atomic Spectrometry September 1996 Vol.1 1 875REFERENCES 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Metals and their Compounds in the Environment ed. Merian E. VCH Weinheim 1991. Trace Element Speciation Analytical Methods and Problems ed. Batley G. E. CRC Press Boca Raton 1987. Metal Speciation Theory Analysis and Application eds. Kramer J. R. and Allen H. E. Lewis Chelsea 1988. Environmental Analysis Using Chromatography Interfaced with Atomic Spectroscopy eds. Harrison R. M. and Rapsomanikis S. Horwood Chichester 1989. Trace Metal Analysis and Speciation ed. Krull I. K. Elsevier Amsterdam 1991. Chau Y. K. Zhang S. and Maguire R. J. Analyst 1992 117 1161. Donard 0. F. X. and Martin F. M. Trends Anal. Chem. 1992 11 17. Caroli S. Microchem. J. 1995 51 64.tobinski R. in 1995 Colloquium on Analytical Atomic Spectrometry ed. Welz B. Perkin-Elmer Uberlingen 1996. Inductively Coupled Plasmas in Atomic Spectrometry eds. Montaser A. and Golightly D. W. VCH New York 2nd edn. 1992. Ebdon L. Hill S. and Ward R. W. Analyst 1986 111 1 113. Ebdon L. Hill S. and Ward R. W. Analyst 1987 112 1. Cappon C. J. LC-GC 1987 5,400. Element-specijic Chromatographic Detection by Atomic Emission Spectroscopy ed. Uden P. ACS Washington DC 1992. Hill S. J. Bloxham M. J. and Worsfold P. J. J. Anal. At. Spectrom. 1993 8 499. Vela N. P. Olson L. K. and Caruso J. A. Anal. Chem. 1993 65 585A. tobinski R. Analusis 1994 22 37. Szpunar J. Witte C. tobinski R. and Adams F. C. Fresenius’ J. Anal. Chem. 1995 351 351. 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 Lobinski R. and Marczenko Z.Spectrochemical Trace Analysis for Metals and Metalloids Elsevier Amsterdam 1996. tobinski R. and Adams F. C. Trends Anal. Chem. 1993,12,41. Gjerde D. T. Wiederin D. R. Smith F. G. and Mattson B. M. J. Chromatogr. 1993 640 13. Berndt H. Fresenius’ Z . Anal. Chem. 1988 331 321. Berndt H. and Schaldach G. J. Anal. At. Spectrom. 1994 9 39. Shum S. C. K. and Houk R. S. Anal. Chem. 1993,65,2972. Nagase M. Kondo H. and Hasebe K. Analyst 1995 120 1923. Rubio R. Alberti J. Padro A. and Rauret G. Trends Anal. Chem. 1995 14 274. Huang Y.-R. Ou Q.-Y. and Yu W.-L. J. Anal. At. Spectrom. 1990 5 115. tobinski R. Dirkx W. M. R. Ceulemans M. and Adams F. C. Anal. Chem. 1992 64 159. Caruso J. ICP In$ Newsl. 1996 21 60. Corr J. J. and Anacleto J. F. Anal. Chem. 1996 68 2155. Zhang S. Chau Y. K. Li W. C. and Chau A. S. Y. Appl. Organomet. Chem. 1991 5 431. Quevauviller Ph. Astruc M. Ebdon L. Desauziers V. Sarradin P. M. Astruc A. Kramer G. N. and Griepink B. Appl. Organomet. Chem. 1994 8 629. Donard 0. F. X. Lalere B. Martin F. and tobinski R. Anal. Chem. 1995,67,4250. Szpunar J. Schmitt V. Donard 0. F. X. and Lobinski R. Trends Anal. Chem. 1996 15 181. Szpunar J. Schmitt V. O. tobinski R. and Monod J.-L. J. Anal. At. Spectrom. 1996 11 193. Rodriguez Pereiro I. and Schmitt V. O. personal communication. Paper 6/00819D Received February 5 1996 Accepted May 8 1996 876 Journal of Analytical Atomic Spectrometry September 1996 Vol. 1 1

ISSN:0267-9477    

DOI:10.1039/JA9961100871

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Investigations Into Sulfur Speciation by Electrospray Mass Spectrometry* IAN I. STEWART DAVID A. BARNETT AND GARY HORLICK Department of Chemistry University of Alberta Edmonton Alberta Canada T6G 2G2 Sulfur is an element that manifests itself in a wide variety of chemical forms in the environment and it is generally not sufficient just to determine the elemental level of sulfur in environmental samples. In this study it is shown that electrospray mass spectrometry (ESMS) has the potential to provide a powerful direct probe for sulfur species in solution samples. The basic electrospray mass spectra to be expected for species such as SO:- S20,2- and SzO$- are presented and the important role of controlling and varying collision induced dissociation (0) conditions in order to validate the species identity is illustrated.It is also shown that ESMS can be used to monitor sulfur species such as S,O:- and S20:- which are products in certain redox reactions and also thiosulfate complexes such as Ag(S2O3);-. Finally quantitative results are presented for the determination of sulfate in waste water. Keywords Sulfur; electrospray mass spectrometry; collision induced dissociation; waste water; sulfur species The chemistry of sulfur is fairly complex as it can exist in several different forms. In aqueous inorganic solutions sulfur commonly exists in anionic form which can be as simple as sulfide or as complex as one of the numerous 0x0- or peroxo- acid forms. Sulfur species are present in many different aqueous systems such as ground and surface waters as well as biological fluids.In ground-water samples complex sulfur species prob- ably originate as a result of sulfide mineral oxidation at low to neutral pHs.’ Therefore the reaction products can give some indication as to the mechanisms involved in this process and thus the determination of such species is desirable for environmental reasons as well as from the point of view of understanding general inorganic sulfur chemistry. In biological samples such as urine sulfate determination is useful in the study of urolithiasis.2 Sulfate derivatives can also exist in many different organic molecules such as sulfonated azo dyes as well as steroidal c~mplexes.~ Over the years sulfur species and inorganic anions in general have been determined for the most part by chemical chromato- graphic and spectrochemical techniques.For example oxo- sulfur species have been determined by ion-exchange chroma- tography 4*5 ion-pair chroma tograp h y ‘-’ capillary elec trop h- oresis (CE)” and spectr~photornetry.’~-~~ For the most part these studies have focused on such species as sulfate sulfite and thiosulfate as well as polythionate species containing up to 13 sulfur atoms. In some cases however these types of techniques are inadequate as they are unable to provide the desired degree of selectivity or sensitivity. As a result research- ers have moved towards developing chromatographically coupled MS techniques such as HPLC-ICP-MS and CE-CP- MS in attempts to improve intra-element speciati~n.~~-’~ These techniques are good for differentiating between different oxi- dation state metal ion species and different polyatomic ion species such as arsenite monomethyl arsonic acid (MMAA) * Presented at the 1996 Winter Conference on Plasma Spectrochemistry Fort Lauderdale FL USA January 8-1 3 1996.Journal of Analytical Journal of Analytical Atomic Spectrometry and dimethylarsinic acid (DMAA) but are not entirely effective at differentiating sulfur species.” More recently a relatively new technique electrospray ioniz- ation mass spectrometry (ESMS) has been used successfully to determine inorganic cations and anions in solution ~ a r n p l e s . ’ ~ ~ ~ ~ In particular Agnes et a1.18 have shown that it can be used successfully to determine sulfur anion species in solution directly. A variation of electrospray (ES) ‘ion spray’ has also been coupled to CE for the MS determination of inorganic cations and anions,” which is desirable when exam- ining complex samples.The technique has been described extensively in the litera- t ~ r e ” - ~ ~ and so only a qualitative description will be given here. Simply ES is an electrostatic spraying process where a liquid surface becomes disrupted when an intense electric field is applied resulting in a spray of charged droplets. With ES a solution of optimal conductivity (typically a methanolic elec- trolyte solution of the order of 1 x 10-5-1 x moll-’) is pumped (x 1-10 p1 min-l) through a stainless-steel capillary (x 100 pm id 200 pm od) which is held at a high potential relative to a counter electrode (the mass spectrometer sampling orifice).This spraying phenomena at the capillary tip affords the transfer of solution ions to the gas phase where they can be sampled by a mass spectrometer. A negatively biased capillary results in negatively charged droplets and ultimately negative gas-phase ions; the opposite is true for a positively biased capillary. The formation of the spray the charging of the droplets and formation of gas-phase ions from such droplets have been described in detail in the l i t e r a t ~ r e . ~ ~ ? ~ ~ The focus of the present paper is on the effectiveness of ESMS in investigating inorganic sulfur species in solutions and is presented in part as a continuation of previous work.’* In particular the paper will discuss some background to the generation and sampling of gas-phase sulfur ions. A variety of solutions will be investigated ranging from standard laboratory solutions to reaction mixtures to real samples.Although the focus is on the qualitative aspects some preliminary data on the quantification of sulfur species will be presented. EXPERIMENTAL Instrumentation The ES source and mass spectrometer have been described previ~usly.’~ The high-pressure sampling region of the modified SCIEX-Perkin-Elmer ELAN Model 250 ICP-MS instrument is based on a sampling plate-skimmer configuration. There have however been some modifications inside the mass spec- trometer the shadow or first photon stop in the ion optics as well as the second stop located in the bessel box have been removed. This has led to an increase in sensitivity of one to one and a half orders of magnitude.As anions are being studied the mass spectrometer was operated in negative-ion mode. For these studies the ES tip was typically biased at - 3000 V the front plate at - 600 V the sampling plate was varied (typically between -2 and -60 V) to minimize or maximize collision induced dissociation (CID) in the high- pressure region and the skimmer was held at -2 V. As the Atomic Spectrometry September 1996 Vol. 1 1 (877-886) 877skimmer potential is held constant the potential drop and hence energy is proportional to the sampling-plate voltage (see ref. 24 for a discussion of this). The curtain gas used for these experiments was nitrogen at a flow rate of about 1.3 1 min-'. The ES needle tip had a 100 pm id and was operated at a flow rate of 1.0-2.0 p1 min-' via a syringe pump (Sage Instruments Model 341A).The tip position although optimized for each set of experiments was usually set 5 mm from the front plate and 1 mm off axis. It was found that careful selection of flow rate and applied capillary potential led to stable signals without the use of discharge suppressing gases. For quantitative investigations two different procedures were employed. In one a normal ES set-up was used where sample solutions were run in a sequential manner. The samples were prepared to contain the analyte in varying concentrations and a known constant concentration of internal standard. The flow rate was typically 2.00 pl min-'. In the second a Valco high- pressure injection valve with a 0.50 p1 internal sample loop was used.The samples were prepared to include the internal standard and were injected into a flowing stream of internal standard in methanol. The flow rate of the carrier-internal stan- dard solution was 2.50 pl min-'. Selected ions were monitored as a function of time. Reagents All solutions were prepared by dissolving the ACS-grade salts in distilled de-ionized water to form a stock solution. Aliquots of the stock solution were then diluted with HPLC-grade methanol to the desired concentration. This procedure was modified for the quantitative experiments where the methanol was distilled and the reagent salt was oven dried. These procedures result in the solution being primarily methanol with a water content in the range of 1-3% by volume.RESULTS AND DISCUSSION General Preliminary studies of inorganic sulfur species were carried out by Agnes et a1.I8 using ESMS. It was shown that a wide range of anionic sulfur species could be examined. For the most part anionic sulfur species are fairly amenable to ES and it is generally the case that if a species exists in an ionic form in solution and is stable in an ES solvent it can be observed by ESMS. These two stipulations can obviously pose some limi- tations. In addressing the first point take sulfide for example the reactive solution form is typically S2- which is difficult to observe by ES. This is due primarily to the fact that simple sulfides such as H2S are only partially ionized in aqueous solutions. The first- and second-dissociation constants are pK 6.8 and pK 14.15 respectively,2s which means that for an initial solution of 1 x moll-' only about 3 .6 ~ mol 1-' exists as HS- and there are virtually no S2- ions present. In addition the equilibrium can be shifted even further to the left when the sample is diluted in methanol which has a much lower relative permittivity than water z 33 as opposed to x78. The other concern is the stability of the species both in aqueous solution and the ES solvent. For the most part the commonly occurring sulfur anions especially the 0x0-sulfur species are fairly stable in solvents such as methanol however some species such as sulfite (SO,2-) are not; this example will be addressed later in the discussion. Although the generation of gas-phase ions from the droplets is still a controversial subject in the ES literature it is evident that ions tend to leave the droplet with a stabilizing solvation sphere of significant size.It is these well solvated ions that are sampled by the mass spectrometer and Fig. 1 will be used to illustrate the processes involved in sampling a solvated anion and stripping it down to a bare polyatomic ion via high- pressure CID at the interface. 2000000 1800000 1600000 14M1ooo 7 u) + 5 1200000 2 1OOOOOO .- E 0 800000 Q) - E 600000 400000 200000 0 90 1 V 200 Fig. 1 the persulfate ion going from - 10 V potential difference to a - 30 V potential difference 3D CID spectra of ammonium persulfate ( 1 x moll-') in methanol. The plot illustrates the change in the solvated distribution of 878 Journal of Analytical Atomic Spectrometry September 1996 Vol.I 1A variable high-pressure CID plot of ammonium persulfate ( 1 x lo- mol I-’) in methanol is shown in Fig. 1 where different ‘CID energies’ are plotted as a function of intensity and m/z. It is important to note that these are not actual ‘collision energies’ but are just the potential differences between the sampling plate and skimmer which will be related to the collision energy. This can be simply understood by considering the fact that the greater the electric field in this region the greater the acceleration of an ion relative to the neutral nitrogen curtain gas (or other neutral gases in the expansion). This will in its simplest interpretation result in a greater collisional energy. The nature of the expansion will further complicate this as gas density and hence collisional frequency is a decreasing function of distance from the sampling orifice.The energy of collisions that occur in the later part of the expansion could conversely increase as a result of greater mean-free path and hence longer acceleration times between collisions. Whether the relationship between potential differ- ence and CID energy is a strictly linear relationship is unknown however the net result in all cases is that a greater potential difference affords a greater amount of stripping. Considering first the low CID potential (lOV) a low (rela- tively) intensity distribution at high m/z can be seen these are peroxodisulfate water cluster species with distributions ranging from four water ligands to 11 water ligands; the most intense peak in this distribution seems to be persulfate with seven water molecules at m/z 159.Since these species are 2- anions the distribution separation is 9 m/z units (18/2). When the potential difference is increased in increments of 5 V the distribution makes an obvious shift to lower m/z values and eventually at a CID value of 30 V yields primarily a desolvated bare persulfate anion at m/z 96. There is a minor peak at m/z 105 owing to the singly hydrated peroxodisulfate species. Past this point the energy supplied goes into molecular fragmen- tation and various decomposition products are formed. The actual process for the persulfate ion has been described in some detail by Agnes et ~ 1 . ’ ~ It is interesting to note the increase in signal intensity going from CID 10 V to CID 30 V.It would be expected that the total ion current should remain somewhat constant regardless of energy but this was not observed. One probable explanation for this is that ions leave the expansion region with greater velocities for higher potential differences and thus they have a much more efficient trans- mission through the ion optics and quadrupoles. Fig. 1 illustrates the different types of information that can be obtained and that it is important to be aware of sampling conditions while interpreting the mass spectrum. At potential differences greater than 30 V molecular fragmentation occurs and species such as HOSO,- HS04- SO3- and SO2- can be observed all of which are clearly not present in the original solution.These spectra obtained under harsher conditions have been discussed previously by Agnes et a1.I8 and are not shown here. Therefore great care must be taken when inter- preting spectra. Depending on the nature of the anion it might or might not be possible to ‘strip’ its solvation sphere down to a ‘bare’ molecular form. In the case of the persulfate ion above it is large enough to stabilize its bare gas-phase charge however other smaller anions might not be able to support multiple charges and therefore will seek to reduce their charge; this is usually achieved by fragmentation. Sulfate and thiosulfate are two such multiply charged ions that undergo charge reduction reactions in the gas phase. The sulfate anion which is perhaps the most commonly occurring of the sulfur-oxygen anion species can exist in solution in several different forms H2S04 HS04- and SO,2 - depending on the pH of the solution.For the most part it exists almost entirely as the ion in neutral solutions. Electrospray mass spectra of vanadyl sulfate in methanol are given in Fig. 2. The figure consists of two spectra a mild (relatively) CID condition spectrum in (a) and a harsher CID spectrum in (b). The distribution of solvated sulfate ions and also a certain amount of the protonated equilibrium species HS04- is shown in Fig. 2(a). The 2- water solvated species are separated by 9 m/z units corresponding to differences of one water ligand. It is observed that the sulfate anion requires at least four water ligands to maintain its two charges.Once this species is reached it will not simply lose another water ligand rather it will charge separate as shown in the scheme below in order to stabilize the charge S042-(H20)4-+HS04- +OH-(H20),+(3 -n)(HzO) Although it seems clear that when sampling with high-pressure CID the minimum water ligand number for sulfate is four the charge separation reactions could also occur to a lesser degree in the more solvated species (i.e. five water ligands). The charge separation itself is not straight forward as it could occur with some variation as indicated below S042-(H20)4+HS04-(HzO) + OH-(H20)2 +etc.. . The above reactions probably account for the presence of the HS04- species observed in Fig. 2(a). At extremely mild sam- pling conditions the HSO,- species contribution is greatly reduced.Sulfate with less than four water ligands is not observed using the high-pressure CID interface. The above studies have also been observed by low-pressure CID.26 A spectrum acquired under harsher CID conditions is shown in Fig. 2(b) where all the sulfur species have essentially been converted into one species (HS04-) prior to mass analysis. Note also the presence of various solvated hydroxide species characteristic of the charge separation step. The mass spectrum has distributions owing to basic ions such as OH- under gentler CID conditions and so the origin of these species should be viewed with caution. However as a consequence of CID if they did exist from other sources such as free OH- they would exist primarily as the bare OH- ion at m/z 17.Therefore the distribution observed in Fig. 2(b) would most probably be there as a result of the fragmentation of the solvated sulfate ion. Also note the peaks at m/z 129.5 and 179 that have not been definitely assigned but could be assigned with caution vanadyl [VO(S04)22-] and vanadate species [ V 0 2 ( S 0 ) - ] respectively. Thiosulfate although not as stable and somewhat more reactive in solution than sulfate also exists primarily as the 2 - ionic species in solution. Three sodium thiosulfate spectra are given in Fig. 3 each acquired under different CID region conditions. Under fairly gentle conditions [Fig. 3 (a)] where AV=14V a distribution of solvated thiosulfate ions with a solvation distribution ranging from three to nine water mol- ecules is observed.A distribution maximum is noted at m/z 101 corresponding to the S2032- (H20)5 species. Under gentler conditions the distribution maximum shifts to higher values. Similar to the sulfate anion a minimal solvation sphere must be maintained by the thiosulfate anion in order to maintain its charge. For thiosulfate it was found that at least three water ligands are required corresponding to an m/z of 83 any number less than this leads to some interesting results. Intuitively one might expect a charge separation similar to that of sulfate to occur as in the scheme below S2032-(H20)3 -+HS203- + OH-(H20)2 This is not strictly observed. The spectrum shown in Fig. 3(b) was acquired under slightly harsher conditions (AV = 19 V) here the solvated distribution can be seen to shift to a maximum Journal of Analytical Atomic Spectrometry September 1996 Vol.11 879200 - 150- 100- 7 'W 50- u) c 9 0 + 0 0 n = 7 n n = 8 0 7 5 8 0 8 5 90 95 100 105 110 115 120 125 130 135 140 145 150 155 160 165 7 s .- 10 2 0 30 4 0 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 fli2 Fig. 2 potential differences (a) - 12.5; and (b) -22.5 V Mass spectrum of vanadyl sulfate (1 x moll-l) in methanol. The spectra were acquired at two different sampling plate to skimmer at m/z 83 and some decomposition products resulting from the continued stripping of this species are also observed. Although there is evidence for the expected HS203- anion at (m/z 113) it is not the dominant ion formed. The dominant species formed is based on the S203- (m/z 112) ion and its solvated precursors S203-( H20) and S203-(H20)2 at m/z 130 and 148 respectively.It is unclear how these products form however an electron must be lost in coincidence with the loss of a water ligand. Examination of the third spectrum in the series [Fig. 3(c)] illustrates that the solvated species can be stripped down to the bare polyatomic ion S203- at m/z 112. The exact nature of the process going from the solvated 2- polyatomic ion to the 1- bare ion is currently under investi- gation. The sodium adduct (m/z 135) is most likely formed as a result of droplet preconcentration. The formation of the S203- species is a consistent product when sampling with the high-pressure CID source however it is unknown whether this species forms in the same dominant manner under the conditions of low-pressure CID.From the above discussion it is apparent that some ions cannot be stripped down to their 'true' polyatomic ions however they can be stripped down to characteristic poly- atomic ions as indicators. In the above case the characteristic ions are HSO,- (m/z 97) for sulfate and S203- (m/z 112) for thiosulfate. It is clear that with some background knowledge of the sampling behaviour of these species by high-pressure CID it should be possible to differentiate between the various anions in solution. A test sample was prepared of equimolar concen- trations of ammonium persulfate sodium sulfate and sodium thiosulfate where each was 1 x lo- moll-' in methanol. Two spectra are shown in Fig. 4 where spectrum (a) was acquired under fairly harsh CID conditions (AV = 30 V) and spectrum (b) was acquired under gentler conditions (AV= 10 V).Based 880 Journal of Analytical Atomic Spectrometry September on the knowledge of the potential difference and characteristic peak assignments spectrum (a) indicates that the persulfate anion the sulfate anion and the thiosulfate anion could be present however the assignment is ambiguous at best. It is not until the ions are sampled under gentler conditions [spec- trum (b)] that this is confirmed. Examination of Fig.4(b) illustrates the presence of the three distinct distributions of persulfate sulfate and thiosulfate each highlighted in its own pattern. Not all species are stable in ES solvents let alone in aqueous solutions. One such species is the sulfite ion which is only finitely stable in aqueous solutions and is reactive in methanol solutions. Upon dilution of freshly prepared aqueous sulfite solution in methanol the species CH30S02- is readily observed by ES.This is illustrated in Fig. 5 where in addition to the CH30S02- species (m/z 95) being observed the presence of the minor species HS04- is also detected. Although the origins of the HS04- species is unclear the ester could be formed via the scheme HS03- + CH30H-+H20 + CH30S02- When an aqueous sulfite solution is allowed to age ( z 1 month) the only observable species when diluted by methanol is sulfate. This indicates that sulfite slowly decomposes in aqueous solution to yield sulfate. Other species such as dithionite (S2042-) are also fairly unstable in aqueous solutions and can decompose rapidly depending on the pH.25 Its decomposition products if stable in solution can of course be detected by ESMS.From the above discussion it is obvious that the chemical nature of the species is critical when obtaining information about them by ESMS. It was shown that large multiply charged anions are able to stabilize the charge internally whereas smaller multiply charged ions need a supporting 1996 Vol. 11x = 5 I . . " " . " " ' ' " " ' " ' ' I " ' I " ' " ' " ' I " " ' " " I""""' 0 I ....,.... I ....,....,....,.... 70 80 90 100 110 120 130 140 150 160 170 70 80 90 100 110 120 130 140 150 160 170 7 0 80 90 100 110 120 130 140 150 160 170 m/z Fig. 3 Mass spectra of sodium thiosulfate (1 x potential differences (a) - 14 V; (b) - 19 V; and (c) -24 V in methanol.The spectra were acquired at different sampling plate to skimmer solvation sphere to be stable. If the ion gets to the point where it no longer has the necessary solvation then charge-separation reactions occur. It is important to be able to understand these processes in order to interpret the resulting mass spectra properly. Interpreting Reaction Mixtures Sulfur ions in solution especially various 0x0-anions can be fairly reactive. They can act both as reducing agents and oxidizing agents of various strengths in solution. For instance the thiosulfate and sulfite ions are known to have a moderate reducing character whereas the peroxodisulfate ion is known to be a powerful oxidizing agent. This characteristic can be fairly useful analytically for example Koh and co-worker~l'-~~ have determined many sulfur species spectrophotometrically based on their redox chemistry.Perhaps one of the most well known reactions is the use of thiosulfate in the volumetric determination of iodine which proceeds according to the reaction 2S2032- +I2 +2I- + S4062- where the products are iodide and tetrathionate ions. This simple reaction system was examined by ES. The mixture consisted of 1.02 x mol I-' thiosulfate in aqueous solution. Upon mixing and dilution the dark purple iodine solution immediately became clear which is characteristic of its reduction to iodide. The solution after standing for 2 h was then diluted with methanol (100-fold) and the resulting solution was electrosprayed.Two spectra acquired under different sampling conditions are given in Fig. 6 (a) at a potential difference of 15 V and (b) at 27 V. Examination of Fig. 6(u) reveals two distributions the tetra- thionate distribution stripped down to the bare tetrathionate ion at m/z 112 and accompanying hydrates as well as a distribution of thiosulfate species which were present in excess. The spectrum just highlights the hydrated species however there are species such as S4062-(CH30H) S40,2-(CH30H)2 and S4062-(CH30H)(H20) present at m/z 128 144 and 137 respectively. The iodide ion is also present at m/z 127. If the spectrum is acquired at a greater potential difference [Fig. 6(b)] the CID products shown below moll-' iodine and 2.23 x S4062- -+S203 - + S203 - S4062- +S303- + SO3- are formed which would not normally be found in solution. The tetrathionate ion because of its size can be stripped down Journal of Analytical Atomic Spectrometry September 1996 Vol.1 1 881400 300 200 100 T ln ln K O c 2 V 80 90 100 m 140 c .- ln 2 120 - 1 00 80 60 40 20 0 (b ) Sulfate Thiosulfate Persulfate I 110 120 130 140 70 80 90 100 110 120 130 140 150 160 170 nVz Fig. 4 Mixture of 0x0-sulfur species ammonium persulfate sodium thiosulfate and sodium sulfate each 1 x lop4 moll-' in methanol. Spectrum (a) was acquired with a sampling plate to skimmer potential difference of - 30 V and spectrum (b) with a potential difference of - 10 V CH,OSO,- / to its bare form at m/z 112. However increasing the CID energy leads to the decomposition products as described above and as observed in Fig.6(b). A second redox mixture system that was studied by ESMS is given in the scheme below MnO +2SO,,- +4H+ +Mn2+ +S2062- +2H,O where sulfite is oxidized by manganese(1v) oxide yielding the dithionate ion ( S 2 0 6 2 - ) . The reaction mixture consists of MnO (0.087 g in 100 ml total volume) 2.05 x mol 1-1 SO,2- and 4 x lo- moll-' HN03. The reaction mixture was allowed to stand for 1 d to ensure complete reaction upon which time it was diluted with methanol (100-fold) and then examined by ES. The results are given in Fig. 7 the spectrum was acquired under fairly harsh stripping conditions in order to illustrate that the dithionate ion can be stripped down to lJ+ . ,,- , .~ it" L, ,1.,t, :- TI.:- :- -L,,..-.,.I ^ C - - - / - o n - - A :- confirmed by the isotopic distribution and in addition there is some residual hydrate at m/z 89.If the CID energy is greatly increased the dithionate ion decomposes according to the m/Z Fig- Mass spectrum of sodium sulfite ( 1 mol 1-1) in anol. The spectrum was acquired with a sampling plate to skimmer potential difference of - 20 V scheme S206,- - SO - + SO - 882 Journal of Analytical Atomic Spectrometry September 1996 Vol. 1180 9 0 100 110 120 130 140 150 160 170 180 60 80 100 120 140 160 180 200 220 240 260 m/z Fig. 6 Mass spectra of the reaction products from a solution of iodine (1 x lo-’ moll-’) and sodium thiosulfate (2 x lo-’ moll-’) diluted in methanol 100-fold. The spectra were acquired at sampling plate to skimmer potential differences of (a) - 15; and (b) -27 V d Z Fig.7 Mass spectrum of the reaction products from a solution of MnO (1 x lo-’ moll-l) sodium sulfite (2 x lo-’ moll-’) and nitric acid (4 x lop2 moll-’) diluted in methanol 100-fold.The spectrum was acquired with a sampling plate to skimmer potential difference of -22v In addition to the 2- species the protonated and sodium adducts are also observed at higher m/z. The spectrum also shows two other ions of interest those at m/z 97 (HS04-) and m/z 111 (CH,OSO,-) for which there are two possible expla- nations. The first is the oxidation of sulfite by manganese(1v) both in aqueous and methanolic solutions to yield the two products the second is the reaction of the dithionate ion with methanol uia the following scheme S 2 0 6 2 - + CH30H+HS04- +(CH,O)SO,- It is unlikely that the above scheme occurs on its own in solution and could be acid or manganese(1v) catalysed.The above scheme could also be occurring as a result of some gas- phase chemistry. Sulfur species are also useful complexing agents and for example thiosulfate is used in photographic processes. Thiosulfate solutions are typically used to complex unphoto- lysed silver bromide from a photographic emulsion. In fact thiosulfate is able to complex with most silver ions in solution provided it is added in excess. When silver is the dominant ion an Ag2(S,0,) precipitate will form which then gradually decomposes to Ag2S. However in the case of thiosulfate being used in the dissolution and complexation of silver halides it is typically added far in excess to ensure the reaction AgX+2S2032- +Ag(S203),,- + X - The above reaction holds for any of the insoluble silver halides as the net stability constant for the above complex is of the order of 5 x lo1,.To examine this system a solution was prepared where AgCl (0.1433 g in a total volume of 100 ml) was dissolved in excess of S2032- (2.5 x lo-’ moll-l). The solution was allowed to sit for 1 d and then diluted 100-fold in methanol and examined by ESMS. The results are shown in Fig. 8. The spectrum in Fig. 8(a) was acquired under fairly gentle conditions and shows the distribution of solvated thios- ulfate as expected being added in excess and also a second distribution of solvated Ag( S,03)2 - species. This distribution ranges from five water ligands (m/z 140.3) to 11 water ligands (m/z 176.3) and the species are separated by 6 m/z units (18/3 = 6).The Ag(S20,),3- species is only stable down to five hydrates. The spectrum in Fig. 8(b) is a higher resolution spectrum and was acquired under slightly harsher sampling conditions. The isotopic fine structure due primarily to the silver and sulfur isotopes becomes compressed three-fold as it Journal of Analytical Atomic Spectrometry September 1996 Vol. 11 8830 i7j 4000; 3000 - 2000 - 1000 - x = 6 x = 7 l 5 I 10 135 140 145 150 155 160 160 170 180 190 200 210 220 230 ,m/z Fig. 8 Mass spectra of the reaction products from a solution of silver chloride ( 5 x mol l-’) dissolved by sodium thiosulfate (2.5 x lo-’ moll-’) in aqueous solution. The sample was then diluted 100-fold in methanol.The spectra were acquired at different sampling plate to skimmer potential differences (a) -9; (b) - 14 (high resolution); and (c) -24 V is a 3 - species. The spectrum in Fig. 8(c) was acquired under fairly harsh conditions and shows- what happens when the Ag( S203);- is stripped lower than the minimum stabilizing ligand number (ie. five water ligands). Consistent with the previous discussion on thiosulfate one of the thiosulfate ligands loses an electron in some unknown manner to form an electron deficient species as shown at m/z 165.5 [Ag(S203)22-]. The complex further decomposes to the Ag(S,O,)- species as observed at m/z 219 with increased CID energy. The isotopic abundances of the above species are consistent with the calculated values as shown by the two examples given in Fig.9. The above discussion illustrates the strength of ES as an investigative probe into solution chemistry where reaction mixtures or complex mixtures of species can be investigated and meaningful results obtained. In some cases however the 5 130 134 138 142 146 150 niques must be used in order to evaluate a sample properly. The use of ES or ion spray coupled techniques has focused for the most part on biological c o m p o ~ n d s ~ ~ ’ ~ ~ however some groups have investigated inorganic systems.20 It is clear that ESMS can be a powerful probe for inorganic solution species. Quantitative Studies Preliminary Data To date there has been a decided lack of information regarding the quantitative aspects of ESMS.This is probably because of the subtle complexity of the technique. The concentration response is inherently non-linear and has been described by Agnes and Horli~k.~’ There is also a lack of CRMs for many anionic compounds such as sulfur species. As a result there is a real challenge to find ways to quantitate samples accurately n 20 $ 0 .I 3 100 I? 80 60 40 20 0 160 164 168 172 176 180 m/Z c sample could be too complex and therefore separation tech- .- (b) by ESMS. such work is Currently being invef&ated in this laboratory and an in-depth study will be presented in a future Fig. 9 Relative isotopic abundances of silver thiosulfate complexes (a) Ag(S,0,)2(H20),3-; and (b) Ag(S,O,),- 884 Journal of Analytical Atomic Spectrometry September 1996 Vol. 112.001 y = 7.4800e-2 + 3.0630e+4x 0 RA2 = 0.995 f C a (*1 c C .- j~u~fatel/mol I-' Fig.10 Log-log calibration curve for sodium sulfate in methanol by ESMS. The log of the sulfate intensity as ratioed against a 1 x lop4 moll-' KI internal standard response and plotted versus the log of the sulfate concentration paper. Some preliminary results of that study which are encouraging but by no means definitive are presented here. To achieve linearity in the calibration curve the analyte signal is typically ratioed against an internal standard. In addition to account for matrix affects which can be fairly severe in some cases the preferred method of quantification is standard additions. A calibration curve for sulfate is presented in Fig. 10. Harsh CID conditions were utilized thus the signal measured was that for HS04- at m/z 97 [see Fig.2(b)].Iodide (I- at m/z 127) was used as the internal standard and was present in all solutions at a level of 1 x lo- moll-'. The response was linear over the two orders of magnitude shown by the slope of the log-log plot being 1.01 & 0.02. The correlation coefficient was 0.998. The DL for sulfate based on three times the standard deviation of the background at m/z 122 was 1 pg 1-'. The background position at m/z 122 was selected as there is no ion present at this point under the conditions sampled and thus this value should reflect the detection limit of a sample not contaminated by sulfate. Sulfate is a fairly common anion and could be present as an impurity in reagents or solvents.Therefore great care must be taken in preparation of samples and standards. As an example a sample was run consisting of just 1 x moll-' iodide in methanol. The background sulfate concentration was found to be 3.60k0.26 pg 1-' based on the above plot. In either case the above results indicate the types of sensitivities obtainable by ES and detection limits could be blank limited and thus higher than the 1 pg 1-' indicated above. The applicability of this technique to a real-life sample was investigated. A waste-water sample was examined using ESMS. The sample was originally collected from a landfill site to examine a series of cations and anions in the waste-water. Sulfate was determined to be present at 6.10f0.61 x loW3 moll-' in the sample by an independent laboratory using ion chromatography.A mass spectrum of the sample was obtained (see Fig. 11) by diluting the sample 100-fold in methanol and with a sampling plate to skimmer potential difference of 35 V. Note the presence of CH,O- (m/z 31) and CH30C02- (m/z 75) which are typical background ions indicative of basic solutions. The presence of sulfate in the sample is indicated by the signals for HS04- and NaS0,- at m/z 97 and 119 respectively. Other anions observed include F - C1- NO2- and NO3-. A major problem associated with quantification by ES is that the response of the analyte and internal standard do not necessarily behave similarly upon addition to or variation of the sample matrix. Variation in the intensity of an ion with concentration has been discussed in the 1iteratu1-e.~~ It is therefore desirable to match the matrix as closely as possible 20 40 60 CH,OCO,- HSO NaSO,(H,O)- /I NalSo4- I 80 100 120 140 160 m/z Fig.11 Mass spectrum of a waste-water sample. The sample was diluted 100-fold in methanol and acquired with a sampling plate to skimmer potential difference of -40 V to minimize these effects. The standard additions technique should allow for this. Sample preparation involved diluting 0.33 ml of the sample with KI internal standard solution (1 x lo- moll-'). Prior to dilution aliquots of a standard Na,SO solution were added to the sample to afford the standard additions. The samples were then injected from a 0.50 pl sample loop into a flowing stream of internal standard (1 x mol 1-' KI) solution. The standard additions calibration curve is shown in Fig.12. The signal level used for sulfate was the sum of the intensities for HSO,- (at m/z 97) and NaS0,- (at m/z 119) and five replicates were used for each point. These values were then normalized to give the total sulfur concentration. Using least- squares calculations the sulfate concentration was determined to be 4.87 & 0.33 x lop3 moll-'. If this value is compared with the reported value of 6.10k0.61 x lo- mol I-' (determined earlier by an independent laboratory) it was found to be 20.2% lower then expected. The difference is fairly significant and could be due to several factors such as the original sample analysis could have been inaccurate the sample that was run 1 month after collection could have had its speciation altered during this time or the response or activity of the sulfate ion or adduct ions in methanol need not be the same as in aqueous solutions.The fact that the isotope ratio was assumed to be constant and therefore only the major isotope needed to be monitored could also have contributed to error in no small amount. Although the results are not accurate enough to seem 0.30 0.15 0.10 8 0.05 u o .- .- -20 iy = 5.98056-2 + 41 11 .lx iR"2 = 0.997 [Added sulfate]/prnol I-' Fig. 12 Standard additions calibration curve for sulfate ion in a waste-water sample by ESMS. The total sulfate intensity is ratioed against a 1 x lop4 moll-' KI internal standard response and plotted versus added sodium sulfate concentration Journal of Analytical Atomic Spectrometry September 1996 Vol.11 885analytically useful they are promising especially as these are only preliminary results. CONCLUSIONS The ability of ES to generate intact gas-phase ions representa- tive of solution ions opens up a whole new avenue for speciation work. In many ways it is an ideal source to couple with MS as it retains key chemical information about the nature of the elements in a sample that MS is capable of determining (i.e. molecular form and charge state) information that for example is generally destroyed when using ICP-MS. The sampling of these ions is for the most part understood however there are areas open for further fundamental studies which could concentrate on energy deposition in the desolv- ation process (e.g. ion kinetic energies as a function of CID energy and state of desolvation and fragmentation on the resultant energies).However it is also clear from the present study that for every determination considerable work must be carried out to understand fully the qualitative results of ESMS and in particular more work is required to define the concen- tration response of ions by ES in order to quantitate them accurately and properly. Financial support by the Natural Sciences and Engineering Research Council of Canada (NSERC) and the University of Alberta are gratefully acknowledged. REFERENCES Goldhaber M. B. Am. J. Sci. 1983 283 193. Singh R. P. and Nancollas G. H. J. Chromatogr. 1988,433 373. Bruins A. P. Covey T. R. and Henion J. D. Anal. Chem. 1987 59 2642. Takano B. McKibben M. A. and Barnes H.L. Anal. Chem. 1984,56 1594. Miura Y. Tsubamoto M. and Koh T. Anal. Sci. 1994,10 595. Steudal R. and Holdt G. Chrornatogruphia 1986 651 379. 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 Rabin S. B. and Stanvury D. M. Anal. Chem. 1985 57 1130. De Backer B. L. Nagels L. J. Alderweireldt F. C. and Van Bogaert P. P. Anal. Chim. Acta 1993 273 449. Geissler M. and Van Eldik R. Anal. Chem. 1992 64 3004. Wildman B. J. Jackson P. E. Jones W. R. and Alden P. G. J. Chromatogr. 1991 546 459. Koh T. and Taniguchi K. Anal. Chem. 1974 46 1679. Koh T. Miura Y. and Suzuki M. Anal. Sci. 1988 4 267. Koh T. Miura Y. Ishimori M. and Yamamuro N. Anal. Sci. 1989 5 79. Koh T. Takahashi N. and Yokoyama K. Anal. Sci. 1994 10 765. LaFreniere K. E. Fassel V. A. and Eckels D. E. Anal. Chem. 1987 59 879. Gjerde D. T. Weiderin D. R. Smith F. G. and Mattson B. M. J. Chromatogr. 1993,640 73. Olesik J. W. Kinzer J. A. and Olesik S. V. Anal. Chem. 1995 67 1. Agnes G. R. Stewart I. I. and Horlick G. Appl. Spectrosc. 1994 48 1347. Agnes G. R. and Horlick G. Appl. Spectrosc 1994 48 655. Huggins T. G. and Henion J. D. Electrophoresis 1993 14 531. Yamashita M. and Fenn J. B. J. Phys. Chem. 1984 88 4451. Hayati I. Bailey A. I. and Tadros T. H. F. J. Colloid Interface Sci. 1987 117 205. Kebarle P. and Tang L. Anal. Chem. 1993 65 972A. Stewart I. I. and Horlick G. Anal. Chem. 1994 66 3983. Cotton F. A. and Wilkinson G. Advanced Inorganic Chemistry. A Comprehensive Text Interscience New York 5th edn. 1988 p. 500. Blades A. T. and Kebarle P. J. Am. Chem. SOC. 1994,116 10761. Smith R. D. Wahl J. H. Goodlett D. R. and Hoftstadler S. A. Anal. Chem. 1993 65 574A. Smith R. D. Barinaga C. J. and Udseth H. R. Anal. Chem. 1988 60 1948. Agnes G. R. and Horlick G. Appl. Spectrosc. 1994 48 649. Paper 5/075068 Received November 16 1995 Accepted March 15 1996 886 Journal of Analytical Atomic Spectrometry September 1996 Vol. 1 1

ISSN:0267-9477    

DOI:10.1039/JA9961100877

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Development of a Gas Chromatography Inductively Coupled Plasma Isotope Dilution Mass Spectrometry System for Accurate Determination of Volatile Element Species. Part 1 I Selenium Speciation* Journal of Analytical Atomic Spectrometry STEFAN M. GALLUS AND KLAUS G. HEUMANNt Institute for Inorganic and Analytical Chemistry Johannes Gutenberg- University Mainz Becherweg 24 0-55099 Mainz Germany The determination of element species in the environment is often especially difficult due to their presence at low concentrations. The coupling of GC witb ICP-MS offers the advantage of transferring the total analyte into the ICP-MS instrument without any loss of analyte by nebulization. The application of IDMS results in relatively accurate results. The described GC-ICP-IDMS system consists of a gas chromatograph fitted with a capillary column for analytical separation and a diffusion cell that is used to exactly determine the mass discrimination factor for the isotope ratio measurement and to perform an element specific optimization of the plasma conditions.The construction of a relatively simple and low cost transfer line as well as the interface between GC and ICP-MS is described in detail. The applicability of the developed GC-ICP-IDMS system for the determination of volatile element species is demonstrated by the determination of selenite. Selenite is converted into a volatile piazselenol prior to determination. Selenate is determined after conversion into selenite. By applying 62% enriched 82Se selenite spike solution for the isotope dilution step the "Se 82Se and 78Se 82Se ratios respectively could be used for content calculation.Selenite (10 ng m1-I) was determined in a water sample with good agreement (within 1%) between the results obtained using the two isotope ratios. The accuracy of the results was demonstrated by the analysis of standard reference materials. The detection limit of the described method was found to be 0.02 ng ml-'. Keywords Inductively coupled plasma mass spectrometry; gas chromatographic coupling; isotope dilution; volatile element species; selenium speciation The behaviour and effect of an element e.g. in the environment essentially depends on the chemical form in which it exists. Different element species can occur due to different oxidation states or due to the co-ordination of an element to various compounds.The toxicity and volatility of many elements significantly varies from species to species; this is well known for mercury for example comparing the volatile and highly toxic dimethyl mercury with inorganic mercury compounds.' Element speciation is also an important key factor for a better understanding of the global geochemical cycle of an element.' Especially in the case of the global distribution of the trace element selenium the biogenic conversion of selenite in the ocean into the volatile dimethyl selenide is i m p ~ r t a n t . ~ ~~~ * Presented at the 1996 Winter Conference on Plasma Spectrochemistry Fort Lauderdale FL USA January 8-1 3 1996. To whom correspondence should be addressed. Owing to the great importance of element species in the environment even in trace and ultra-trace concentrations sensitive and reliable analytical methods are required.It is well known that the accuracy of analytical results is a great problem in trace element analysis. This becomes even more significant when element species have to be determined? However accu- rate analytical results are essential with respect to the judge- ment of the toxicity the bioavailability and the environmental behaviour of element species. On-line coupling of LC or GC with ICP-MS has become a powerful analytical method for element speciation in the past few years.'-' However coupling of chromatographic methods with ICP-MS does not necessarily guarantee accurate results. IDMS is a method of proven high accuracy which means that it is a technique for which the sources of systematic error are normally understood and controlled." This is the reason why IDMS is internationally accepted as a definitive meth- od." Recently Heumann and c o - ~ o r k e r s ~ ~ - ~ ~ have developed HPLC-ICP-IDMS methods which allow accurate determi- nation of element species in aquatic systems.Heavy metal complexes of humic acids and different inorganic and organic iodine species were determined by this combined IDMS tech- nique. Another single example of HPLC-ICP-IDMS is the analysis of trimethyl lead in rain water samples carried out by Brown et al.'' In all these cases 'real time' concentrations could be determined during the separation process without any external calibration which is normally the major problem in quantifying transient signals of chromatographic ICP-MS coupled systems.Although GC has been used in conjunction with ICP-MS ID techniques with GC-ICP-MS methods have not yet been applied. We therefore developed a GC-ICP-IDMS system for the accurate determination of volatile element species. The instrumental design of the GC-ICP-IDMS system and its fundamental principles are described in this paper. This system can be applied for the determination of all volatile element species if the corresponding spike compound is available the sample pretreatment including the ID step has been developed and the species to be determined is thermally stable under the GC conditions used. The reliability of this system is shown by the determination of selenite and selenate after specific chemical conversion of selenite into a volatile piazselenol.EXPERIMENTAL ICP-MS and GC Instrumentation The ICP-MS instrument used was a Perkin Elmer SCIEX ELAN 5000 quadrupole instrument with conventional equip- ment. The standard injector tube was replaced by a quartz Journal of Analytical Atomic Spectrometry September 1996 Vol. 11 (887-892) 887injector tube with an inner diameter of 1 mm (AHF Analysentechnik Tubingen Germany). The operating con- ditions which were generally used are summarized in Table 1. The gas chromatograph used was from Carlo Erba model 8160 equipped with an HP-1 capillary column (10 m x 0.53 mm; Hewlett Packard Waldbronn Germany) with a 2.65 pm thick film (dimethylpolysiloxane). A 1 m x 0.53 mm MXT guard column (intermediate polarity; deactivated fused-silica-coated steel capillary; Restek Bad Soden Germany) was attached between the on-column injector and the capillary column.For separation a helium (99.996% purity) gas flow of 15 to 20 ml min-' was used which did not influence the high argon gas flow rates (see Table 1) in the ICP-MS. When a toluene solution of piazselenol was injected on-column into the GC the following temperature program was applied injection at 110°C into the air pres- sure cooled injector followed by a temperature increase at 15 "C min-l to 130 "C (held for 2 min) to 200°C (held for 5 min) at 40°C min-l. Chemicals Nitric acid and hydrochloric acid (pro analysi Merck Darmstadt Germany) were purified by distillation under sub- boiling conditions in a quartz still.To obtain pure water deionized water was first purified by a two-fold distillation in a quartz still then distilled under sub-boiling conditions. 1,2-Diamino-4-trifluoromethylbenzene (Maybridge Tintagel UK) was purified as described in the literature.I6 Dimethyl selenide DMSe (99% purity; Strem Chemicals Newburyport MA USA) Toluol Lichrosolv dehydrated sodium sulfate (pro analysi; Merck) and selenium dioxide (puriss.; Fluka Neu Ulm Germany) were used without further purification. All gases used in this work were obtained from Linde Germany. Interface for Coupling of GC with ICP-MS The coupling unit is schematically represented in Fig. 1. The GC column is connected uia a six-port valve (C6WT Valco Schenkon Switzerland) to an MXT steel capillary (1.5 m x 0.53 mm).This is the same type of deactivated capillary as used for the guard column. The steel capillary is inserted through the injector tube of the plasma torch and ends about 3 cm in front of the load coil. The steel capillary column is inserted into a 1 m long transfer tube (inner diameter 1 mm) also made of steel which almost fits the outer diameter of the capillary. This transfer tube is heated directly by an electrical transformer (type 3234 D Statron Fiirstenwalde Gemany). The beginning of the transfer tube is connected via a modified Swagelok connector to a Table 1 ICP-MS operating conditions Rf generator Rf power/W Channeltron electron multiplier Dead time correction/ns Sampler and skimmer cone Torch Injector tube Argon flow rates11 min-l- Outer Intermediate Make up Oxygen flow rate11 min-' Scanning mode Dwell timelms Points per spectral peak Sweeps per reading Readings per replicate 40.68 MHz free running 1200 Galileo model 48 16B 80 Aluminium Standard Quartz 1 mm id 14.4 1.1 1.1-1.2 0.02 Peak hopping 10 1 at m/z=77 78 82 3 1 detector port of the GC which is heated to the same tempera- ture as the transfer tube.The temperature in the transfer tube is calibrated by a thermoelement against the applied electrical current. The temperature is found to be constant over the whole length of the transfer tube and could be reproducibly fixed by the electrical current. This allows adjustment of the temperature of the transfer tube during coupling experiments by the electrical current. No insulation is necessary when using this direct heating system to hold the temperature constant.However for safety reasons the transfer tube is covered by a PTFE tubing. The transfer tube ends at the connection to a Swagelok reducing union that fits into a laboratory-prepared glass fitting which is connected to the injector tube support. All connections are sealed with PTFE or graphite ferrules. For the introduction of piazselenol from the GC system into the ICP-MS the transfer tube was held at a temperature of 220 "C in the case of DMSe this temperature is 140°C. GC-ICP Coupling System for ID A schematic diagram of the complete instrumentation used for the determination of volatile element species by ID is shown in Fig. 2. An element species extracted into an organic solvent can be directly injected into the GC system by means of the on-column injector. The possible injection of extracts has the advantage that enrichment procedures by extraction and/or derivatizations of non-volatile species into volatile ones can be carried out prior to determination by GC-ICP-MS.Alternatively volatile element species can be directly intro- duced into the separation column. In this case cryo-focussing prior to sample introduction should be carried out to prevent unnecessary peak broadenings. The transfer tube leading to the ICP-MS instrument can either be switched to the GC capillary column or to a glass flow cell by the six-port valve. The flow cell with a volume of about 20 ml is laboratory prepared. It contains a diffusion cell consisting of a glass vial covered by a membrane which allows diffusion of the volatile compound into the flow cell.A 4ml screw cap vial (WISP Style Alltech Unterhaching Ger- many) with a PTFE-red butylgum septum (1 1 mm x 1.3 mm; Macherey & Nagel Diiren Germany) as membrane normally used as a container for GC samples is used as a diffusion cell. Both the capillary column and diffusion cell are heated by the GC oven. For calibration of the measured isotope ratios the diffusion cell containing pure chemicals of the element species to be determined is placed inside the flow cell. In the case of DMSe an oven temperature of 30°C is used for piazselenol the oven temperature is 85 "C. From here they are introduced in a controlled way into the helium gas flow entering the ICP-MS instrument under exactly the same conditions as the separated compounds from the capillary column.This is necessary because the actual mass discrimination factor in measuring isotope ratios with the quadrupole ICP-MS must always be determined for accurate results by ID. If the isotope ratio of the calibration compounds used in the diffusion cell are known for example by the natural isotopic composition of the elements or by determinations with other MS methods such as gas mass or TIMS which show only very low mass discrimination effects,17 the measured isotope ratio of the separated species in the sample can be corrected. Using the selenium isotope ratio determination as an example the mass discrimination factor can be calculated by the following equa- tion Mass discrimination factor ( 77Se 82Se) = ( 77Se 82Se),,ue/( 77Se 82Se),e,,u The introduction of compounds from the diffusion cell into 888 Journal of Analytical Atomic Spectrometry September 1996 Vol.11glass fitting injector tube support Swagelok heated transfer tube reducing union I I -H 3cm - 000 MXT steel capillary terminal to electrical I I transformer t u u 1 t plasma auxiliary I argon make up Fig. 1 Interface unit for coupling of GC with ICP-MS helium onalurnn injector GC oven I transformer transfer tu; Fig. 2. Schematic diagram of the GC-ICP-MS coupling system the ICP-MS is the best method for tuning the ion intensities. To reduce memory effects when different species are sub- sequently deposited in the diffusion cell the inner surface of the cell was deactivated by silylation. Silylation of the OH groups at the glass surface was carried out by reaction with dimethyl silicon dichloride as described in the literature." To complete the removal of the compounds introduced for signal optimization the GC oven was heated for 30 min at 200 "C and the transfer tube at 220°C.Isotope Ratio Determinations For isotope ratio determinations of selenium the isotopes 77Se 78Se and 82Se were selected. The 74Se isotope was not measured because of its low natural abundance. Interference was pre- sent by the dimeric molecular ions of the argon plasma gas 36Ar40Ar+ and 40Ar2+ on 76Se and 80Se respectively. Mass number 83 has always been controlled for possible corrections at mass number 82 which can be interfered by krypton impurities in the gas flows.Fig. 3 shows the results of the measurement of selenium isotope intensities at mass numbers 77 78 and 82 after separating the piazselenol compound described later by the GC system. The piazselenol compound for this measurement was of natural isotopic composition. The time resolved chromatogram demonstrates that ion intensities at mass numbers 77 and 82 can only be detected at the specific retention time of the piazselenol. In ICP-MS mass number 77 is very often interfered by 40Ar37C1+. One of the great advan- tages of chromatographic coupling methods in connection with ICP-MS is therefore the separation of chlorine containing compounds which prevents interferences of plasma gas mol- ecular ions such as ArCl' with isotopes of the analyte com- pound.In Fig. 3 a small contribution of the 38Ar40Ar+ dimer loo Ooo 000 000 000 0 1,7% I r 80 100 120 140 20 40 60 I Retention tirnek Fig. 3 Selenium isotope measurements within a GC-ICP-MS chroma- togram of piazselenol to the total ion current at mass number 78 is seen. This relatively constant background intensity can easily be corrected for the 78Se+ ion current in piazselenol because in the time resolved chromatogram all Ar2+ plasma gas ions appear at any time whereas the isotope peak of the element species only appear at the specific retention time of the compound to be determined. For calculation of the isotope ratios it is better to use the total area of an isotope peak instead of the peak height. Only in this case could reproducibilities in the range of 0.6 to 2% RSD be achieved from a GC chromatogram for both selenium isotope ratios 77Se 82Se and 78Se 82Se.The piazselenol was injected on-column into the GC system as a solution in toluene. Toluene and most other organic solvents appear in the first few seconds of the chromatogram with the temperature program described. To prevent carbon deposition on the cones the organic solvent had to be oxidized by oxygen gas. For the separation of piazselenol oxygen gas (purity 99.995%) was introduced with the make up argon flow for the first 50 s after injecting the sample. The use of oxygen of lower purity should be avoided because of its relatively high krypton content which causes interference. It was necessary to stop the O2 gas flow after this time because a distinct depression of the selenium isotope intensities by a factor of about three was found under these conditions compared with use of a pure argon plasma gas.This effect is not described in the literature. Relatively high solvent volumes of up to 6 pl could be applied when oxygen was introduced in the beginning of the separation. This also increased the detection power of the method com- pared with volumes of about 1 pl normally used for similar separation problems. As can be seen from the intensity curve of mass number 78 (Fig. 3) the oxygen gas also reduces the occurence of Ar2+ dimers. However the relatively low con- tribution of this ion to the 78Se+ peak under non-oxygen conditions can be easily corrected. Data processing was per- formed using inhouse programs as no commercial software for the evaluation of ID analysis of transient signals is available.Journal of Analytical Atomic Spectrometry September 1996 Vol. 1 1 889Selenite Spike Solution The 82Se isotope was selected as a spike for ID. A 45mg portion of metallic grey selenium enriched in 82Se by more than 91 YO (Chemotrade Diisseldorf Germany) was dissolved in 15 ml of concentrated nitric acid. After dissolution the acidic solution was diluted by 200ml of pure water and completely reduced to selenite by adding concentrated hydro- chloric acid. The isotope ratio 77Se 82Se of this solution was found to be about 1 150; this composition is too extreme for precise isotope ratio determinations by ICP-MS. By adding selenite of natural isotopic composition and by further addition of concentrated hydrochloric acid a stock solution with a more optimized isotopic distribution of about 1:20 for 77Se 82Se (isotopic abundances 77Se = 3.13% 78Se = 9.73% "Se = 61.9%) was prepared. The concentration of this stock solution was determined by inverse IDMS to be (5.94f0.14) x 1014 selenium atoms per gram using a standard selenite solution of natural isotopic composition.It was found that selenite is stable in this stock solution for many months at 5°C in the dark in a PFA container. In ICP-MS the precision of the isotope ratio measurement is better the closer the isotope ratio is to unity. The isotope ratios of the isotope diluted samples should therefore not distinctly exceed the range between 0.1 and 10. This means that the optimum spike addition depends on the concentration of the species compound to be determined in the sample.I About 2 ml sample I Sample Treatment for Selenite and Selenate Speciation with If non-volatile element species such as selenite are to be determined by GC-ICP-IDMS they must be converted into a volatile compound which is thermally stable under the GC and transfer tube conditions. The specific formation of different piazselenols from selenite was used in the past for the determi- nation of selenite by GC separation and ECD or electron impact ionization MS.19-22 For high detection sensitivities by ECD nitro- or halogeno-groups must be substituted at the piazselenol. A more element-specific detection compared with ECD is possible by ICP-MS. Chlorine and bromine substituted piazselenols as described by Dilli and Sutiknolg and applied by Tanzer,20 are not acceptable in ICP-IDMS because 77Se will be interfered by 40Ar37C1 and high ion intensities of 79Br and 81Br can contribute to the neighbouring 78Se and 82Se isotopes.Nitro-compounds as used by Reamer and Veillon,22 have relatively long retention times so the 1,2-diamino-4- trifluoromethylbenzene was selected as the compound for forming piazselenol with Se'" in acidic solutions by equation (2). This 5-trifluoromethylpiazselenol has relatively short reten- tion times of about 100 s under the GC conditions used in this work (see Fig. 3). GC-ICP-IDMS + 2 H 2 0 + H30+ The different sample treatment steps for selenite speciation in aquatic systems by converting selenite into the fluoropiazse- lenol subsequent extraction of this compound into toluene and determination by GC-ICP-IDMS is schematically rep- resented in Fig.4. The derivatization is carried out in polypro- pylene centrifuge tubes with screw caps (Sigma-Aldrich Deisenhofen Germany). About 2 ml of the sample are diluted by adding about 10ml of pure water. The dilution step is necessary to obtain optimum acidic conditions for reaction (2) Dilution with about 10 ml pure water I 1 Addition of "Se0s2- spike I I Measurement of nSe?2Se and "Se?*Se (correction by mass discrimination factor) I Fig. 4. Sample treatment for selenite speciation in aquatic systems by GC-ICP-IDMS and to avoid reduction of selenate after addition of the spike solution acidified with concentrated hydrochloric acid. Afterwards about 1 ml of the 82Se-enriched selenite spike is added to make total use of one of the main advantages of the IDMS technique which is that after ID has taken place loss of substance has normally no effect on the analytical res~1t.l~ The amount of spike added is optimized with respect to the isotope ratio of the isotope diluted sample.Then 20Opl of a 10 mmol 1- solution of 1,2-diamino-4-trifluoromethylbenzene in 0.25 mol 1-l hydrochloric acid is added and the solution is heated at 70 "C for 30 min. Under these conditions the selenite is totally converted into the corresponding pia~selenol.~'.~~ A single step extraction of the piazselenol is subsequently carried out with 1 ml of toluene. The toluene phase is dried with about 50mg of dehydrated sodium sulfate. To increase the injected selenium amount about 90% of the toluene is evaporated at 80°C after removal of the sodium sulfate with 0.45 pm PTFE syringe filters (Sartorius Gottingen Germany).Portions (1-6 p1) of this concentrated extract are introduced into the on-column injector of the GC system. Selenate is determined by reduction to selenite at 80°C for 1-4 h with about 5 moll-' hydrochloric acid after addition of the 82Se0,2- pike.^^,^' Afterwards the solution is diluted with water to obtain about 0.5 mol 1-1 acidic solution and steps 4 to 8 of Fig. 4 are then carried out as described before. With reference to the larger volume of the aqueous phase 2ml of toluene are used for extraction.The selenate content is deter- mined by subtraction of selenite from the total selenium content after reduction of selenate.However for the determi- nation of selenate in natural waters by this procedure it must be shown that no organoselenium compounds are present which can be converted into inorganic selenium under the sample treatment condition^.^^,^^ The piazselenol is separated from other compounds by the capillary column and then analysed by ICP-MS for its 77Se 82Se and 78Se 82Se isotope ratios respectively. Prior to each set of measurements the actual mass discrimination factor is determined by introducing piazselenol of natural isotopic composition from the diffusion cell into the ICP-MS for correction of the measured isotope ratio of the isotope diluted sample. The evaluation of the selenite concentration for IDMS 890 Journal of Analytical Atomic Spectrometry September 1996 Vol.11analyses in general is carried out as described in more detail el~ewhere.~.'~ RESULTS AND DISCUSSION GC-ICP-MS Coupling System One of the great advantages of coupling GC with ICP-MS is that the total amount of the separated analyte is used for detection whereas in the case LC coupled with ICP-MS more than 95% of the analyte is lost when a conventional nebulizer is used. For example a selenite solution introduced by a conventional nebulizer into the ICP-MS used in this work results in a detection limit of about 1 ng ml-'. Application of a highly efficient direct injection nebulizer in connection with anion-exchange separation resulted in detection limits for selenite and selenate of 7-8 ng m1-'.28 On the other hand after conversion of selenite into fluoropiazselenol and analysis by the described GC-ICP-IDMS system the detection limit was 0.02 ng ml-' determined by the measurement of the 77Se 82Se ratio.A comparison of technical details of GC-ICP-MS systems described in the literature with the corresponding coupling system used in this work is summarized in Table 2 (see also Fig. 1 and Fig. 2). Using steel transfer lines instead of those made of PTFE or glass has one advantage in that they give good and fast heat transfer which is especially important if temperature programs are to be applied for the transfer line. With a temperature limit of 400°C deactivated steel as a capillary material is applicable at a broader temperature range than PTFE.Another advantage of the steel capillary is its mechanical stability. Vibrations at the end of the capillary column in the torch (Fig. l) which produce unstable ion currents can be almost totally prevented. The direct heating of the transfer tube needs less instrumentation and is faster than indirect heating by an aluminium rod. The transfer tube is also heated much more homogeneously than with a resist- ance wire. It could be shown that the temperature in the whole transfer tube was constant by direct heating without any insulation of the tube. Optimization of the plasma (gas flows and torch position) can be carried out with volatile species of the element to be determined. By introducing these compounds from the diffusion cell into the mass spectrometer (Fig.2) with the same gas flows and under the same temperature conditions as used for the GC separation process the instrumental conditions for the special analytical problem can be tested best. This is never possible if the instrumental conditions for a direct introduction of gaseous compounds into the plasma torch are checked by a liquid introduction (nebulizer) system. However determination of the actual mass discrimination factor for an isotope ratio measurement which is necessary for the IDMS technique can only be obtained by a compound of the same element. Table 2 Comparison of different GC-ICP-MS coupling systems Technical detail Transfer capillary Heating of transfer tube Optimization of plasma and other instrumental conditions This work Deactivated steel Direct heating of steel tube by electrical current By volatile element- specific compounds using a diffusion cell under GC conditions Literature PTFE,'v2' quartzg Aluminium rod,30 indirect heating by resistance Gas flow with Hg,30 injection of standard solution (nebulizer system)' Isotope Dilution Technique As mentioned accuracy in element speciation is a general problem. In the case of coupling systems calibration can especially be influenced by matrix effects and the treatment of the sample.For example it was shown that in HPLC-ICP-MS different matrix compositions in the various chromatographic fractions of separated element species can place limitations on the ac~uracy.'~,'~ When applying ID only an isotope ratio and not an absolute amount or concentration must be deter- mined.As the isotopes of an element and of element species respectively show identical behaviour during (chemical) treat- ment of the sample and also with respect to matrix effects the isotope ratio is not influenced by these parameters. Possible chemical isotope effects which are normally in the range of less than 1 per ml can be neg1e~ted.I~ However the mass discrimination effect especially in ICP-MS cannot be neglected for accurate analyses by ICP- IDMS. The determination of natural 78Se "Se isotope ratios by nitrogen MIP-MS also shows mass bias.32 Catterick et ~ 1 . ~ ~ stated that the determination of the mass discrimination factor is an important topic for accurate ICP-IDMS analyses. The mass discrimination factor with respect to equation (1) must therefore be determined by a corresponding calibration com- pound. This can be best carried out by continuous introduc- tion of a volatile compound from a diffusion cell into the mass spectrometer under the same conditions as for the GC separation (Fig.2). The mass discrimination factors for the selenium isotope ratio measurements of this work are listed in Table 3. From these results one can see that this factor for both isotope ratios determined 77Se 82Se and 78Se "Se is independent of the selenium compound when applying piazselenol and dimethyl selenide respectively. The mass discrimination factors were calculated on the basis of the natural isotopic composition of selenium with 'true' values recommended by IUPAC.34 The factor for selenium isotopes was found to be relatively high by about 3% per mass unit but was constant at least for a whole set of measurements and the day-to-day variation was not significant. This enables the necessary corrections to be made for accurate IDMS results. The reason for this relatively high mass discrimination effect could not be explained up to now.A possible discrimination by the detector dead time could be neglected because count rates were usually less than 100000 s-' at mass number 78. Also mass calibration and ion lens potentials were corrected before each set of measurements. However the finding of relatively high mass discrimination factors was confirmed by V O ~ I ~ ' who found the mass discrimi- nation factor for zinc to be 3% (measurement of 64Zn:66Zn) and for zirconium to be 2.3% ("Zr 94Zr) per mass unit using the same ICP-MS with a conventional nebulizer. Selenium Speciation by GC-ICP-IDMS Important selenium species which can be found in the environ- ment are listed in Table 4 together with possible chromato- graphic ICP-MS coupling systems for determination.When applying ID in connection with the GC-ICP-MS an exactly known quantity of the corresponding spike enriched with 82Se should be added to the sample at the beginning of the analytical procedure. The sample treatment process must guarantee that no conversion of one selenium species into another one takes place and that derivatization processes are specific for only one species. Two different water samples the NIST standard reference material SRM 1643b Trace Elements in Water and the selenium species reference material CRM 602 from BCR Brussels were analysed by GC-ICP-IDMS for selenite and selenate using the sample treatment procedure represented in Fig.4 (Table 5 ) . Journal of Analytical Atomic Spectrometry September 1996 Vol. 11 891Table 3 Mass discrimination factors for selenium isotope ratio measurements by GC-ICP-MS Isotope ratio Selenium compound Piazselenol Dimethyl selenide ~~ Measured Na t Mass discrimination factor 77Se ”Se 0.762 0.874 1.15 78Se 82Se 2.44 2.72 1.12 77Se 82Se 0.766 0.874 1.14 78Se ”Se 2.46 2.72 1.11 Table 4 Selenium species in the environment References Species Trimethylselenonium ion Selenoamino acids Selenium-HUS compounds Selenite selenate Dimethyl selenide DMSe Dimethyl diselenide DMDSe Dimethylselenone Possible ICP-MS method IC-ICP-M S * RPC-ICP-MSt RPC-ICP-M S IC-ICP-MS or GC-ICP-MS after derivatization GC-ICP-MS GC-ICP-MS GC-ICP-MS * IC =ion chromatography t RPC = reversed-phase chromatography Table 5 Selenite and selenate concentrations (in ng m1-l) in certified reference water samples by GC-ICP-IDMS NIST SRM 1643b BCR CRM 602 77Se 82Se 78Se 82Se 77Se 82Se 7sSe 82Se Selenite 10.6f0.3 10.7k0.5 5.0k0.7 5.1k0.8 Selena te < 0.02 < 0.02 7.4+ 1.3 7.42 1.1 G C- IC P- ID MS- Certijied values- Selenite - Selenate - Total selenium 9.7 + 0.5 5.8 k 0.4 7.7 If 0.7 - For both samples the reliability of the results was proved by measuring both isotope ratios 77Se 82Se and 78Se 82Se.The results listed in Table 5 show that the calculated concentrations on the basis of both measured isotope ratios are identical within the limits of the given standard deviations (1s) of 0.3 to 0.5% (three independent determinations).The accuracy of the results is shown by the agreement of the GC-ICP-IDMS data with the certified values. CONCLUSION The GC-ICP-IDMS system described here allows in principle accurate determinations of all volatile element species if a corresponding spike compound is available. Most spike com- pounds for element speciation can easily be synthesized as has been demonstrated in this work for selenite. The costs for an isotopically enriched spike are usually negligible compared with the other costs of the analysis because less than 1 pg of a spike compound is necessary per analysis.We thank the ‘Deutsche Forschungsgemeinschaft’ for financial support. 1 2 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 Organometallic Compounds in the Environment ed. Craig P. J. Longman UK 1986. Frimmel F. H. Fresenius’ J. Anal. Chem. 1994 350 7. Tanzer D. and Heumann K. G. Atmos. Environ. 1990,24A 3099. Heumann K. G. in Metal Speciation in the Environment ed. Broekaert J. A. C. Giicer S. and Adams F. Springer Heidelberg 1990 NATO AS1 Series vol. G23 pp. 153-168. Crews H. M. Dean J. R. Ebdon L. and Massey R. C. Analyst 1989 114 895. Gercken B. and Barnes R. M. Anal. Chem. 1991 63 283. Hansen S. H. Larsen E. H. Pritzl G. and Cornett C. J. Anal. At. Spectrom. 1992 7 629. Feldmann J. Grumping R. and Hirner A. V. Fresenius’ J.Anal. Chem. 1994,350 228. Prange A. and Jantzen E. J . Anal. At. Spectrom. 1995 10 105. De Bikvre P. Fresenius’ J. Anal. Chem. 1990 337 766. Heumann K. G. Mass Spectrom. Rev. 1992 11 41. Heumann K. G. Rottmann L. and Vogl J. J. Anal. At. Spectrom. 1994 9 1351. Rottmann L. and Heumann K. G. Anal. Chem. 1994,66 3709. Rottmann L. and Heumann K. G. Fresenius’ J. Anal. Chem. 1994,350 221. Brown A. A. Ebdon L. and Hill S . J. Anal. Chim. Acta 1994 286 391. Al-Attar A. F. and Nickless G. J. Chromatogr. 1988 440 333. Heumann K. G. in Inorganic Mass Spectrometry ed. Adams F. van Grieken R. and Gijbels R. Wiley New York 1988 Knapp D. R. Handbook of Analytical Derivatization Reactions Wiley New York 1979 p. 20. Dilli S. and Sutikno I. J . Chromatogr. B 1984 300 265. Tanzer D. PhD Thesis University of Regensburg 1990. Elaseer A. and Nickless G. J. Chromatogr. A 1994 664 77. Reamer D. C. and Veillon C. Anal. Chem. 1981 53 2166. Johannson K. and O h A. J. Chromatogr. 1992,598 105. Bye R. and Lund W. Fresenius’ J. Anal. Chem. 1988 332 242. Pettersson J. and Olin A. Talanta 1991 38 413. Tanzer D. and Heumann K. G. Anal. Chem. 1991,63 1984. Ornemark U. and Olin A. Talanta 1994 41 1675. Shum S. C. K. and Houk R. S. Anal. Chem. 1993 65 2972. Hintelmann H. Evans R. D. and Villeneuve J. Y. J. Anal. At. Spectrom. 1995 10 619. Kim A. W. Foulkes M. E. Ebdon L. Hill S. J. Patience R. L. Barwise A. G. and Rowland S . J. J. Anal. At. Spectrom. 1992 7 1147. Peters G. R. and Beauchemin D. Anal. Chem. 1993 65 97. Yoshinaga J. Shirasaki T. Oishi K. and Morita M. Anal. Chem. 1995 67 1568. Catterick T. Handley H. and Merson S. At. Spectrom. 1995 16 229. IUPAC Commission on Atomic Weights and Isotopic Abundances Pure Appl. Chem. 1991 63 991. Vogl J. personal communication University of Mainz 1996. pp. 301-376. Paper 6/01 71 5K Received March 11 1996 Accepted June 24 1996 892 Journal of Analytical Atomic Spectrometry September 1896 Vol. 11

ISSN:0267-9477    

DOI:10.1039/JA9961100887

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Speciation of Arsenic Compounds by Ion Chromatography With Inductively Coupled Plasma Mass Spectrometry Detection Utilizing Hydride Generation With a Membrane Separator* Journal of Analytical Atomic Spectrometry MATTHEW L. MAGNUSON JOHN T. CREED AND CAROL A. BROCKHOFF United States Environmental Protection Agency National Exposure Research Laboratory Human Exposure Research Division 26 W. Martin Luther King Drive Cincinnati OH 45268 USA Ion chromatography (IC) was used to speciate four of the environmentally significant toxic forms of arsenic arsenite arsenate monomethylarsonic acid and dimethylarsinic acid. Hydride generation (HG) was used to convert the species to their respective hydrides. These hydride species were detected with ICP-MS. The gas-liquid separator for the HG unit was based on microporous PTFE tubing.Two novel features which reduce noise are incorporated into the membrane-based HG unit firstly gas is added to the liquid stream prior to entering the membrane and secondly the flow of carrier gas through the gas-liquid separator forms a 'feedback' loop. Absolute detection limits based on 3.140 from 7 replicates for the four arsenic species listed above were 0.6 3.1 1.1 and 0.7 pg respectively. The overall IC-HG-ICP-MS system produced RSD values of 1-6% over 30 min and 2-6% over a week for the four compounds. Two saline reference materials NASS-4 and SLEW-2 (National Research Council of Canada) were analysed to determine concentrations of the four arsenic species and the sum of the arsenic concentration was compared with the certified total arsenic value for each of the reference materials.Keywords Speciation; arsenic; ion chromatography; hydride generation; microporous membrane gas-liquid separator; inductively coupled plasma mass spectrometry Arsenic species vary in their toxicity by several orders of magnitude. To accurately evaluate the risk of environmental exposure to arsenic it is essential to quantify all arsenic species. Risk assessment made on the basis of total arsenic determi- nation can greatly over-estimate the actual risk associated with an exposure because total arsenic concentration can be high while the more toxic forms of arsenic may be present at much lower concentrations. Therefore the detection of the separated arsenic species requires excellent analytical sensitivity. Adequate sensitivity can be obtained by ICP-MS through the use of HG.'-' The greatest exposure risk to arsenic may result from environmental exposure to arsenite (As"') arsenate (As') monomethylarsonic acid (MMA) and dimethylarsinic acid (DMMA).The inorganic forms are likely to be found in water as products of industrial release and thus speciation provides a better risk assessment. MMA and DMMA are less toxic than inorganic arsenic and readily form hydride~.~*~*'' Highly derivatized organo-arsenicals such as arseonbetaine and arsen- ocholine are found as detoxified metabolic products in animal tissue and produce virtually no response to the HG reac- t i ~ n . ~ * ~ . " Thus they are unlikely to interfere with the detection * Presented at the 1996 Winter Conference on Plasma Spectrochemistry Fort Lauderdale FL USA January 8-1 3 1996.Journal of Analytical of As"' As' MMA and DMMA because of the specificity provided by HG. High-performance liquid chromatography (HPLC) often in the form of ion chromatography (IC) has proven useful for the speciation of arsenic corn pound^.^-^^^-^^ Ion chromatogra- phy eluents typically contain high concentrations of salts which quickly build up on the nebulizer ICP torch and/or sampling cones of the ICP-MS interface leading to instrumental drift. Hydride generation in conjunction with IC eliminates this problem because the arsenic is converted to its gaseous hydride which can be separated from the liquid eluent and introduced in the ICP-MS without the associated dissolved solids.However HG typically utilizes relatively high concentrations of hydrochloric acid from which conventional gas-liquid separators produce a 'residual' aerosol.26 This 'residual' aerosol is transported to the ICP where it forms 40Ar35C1 (m/z=75) which interferes with the detection of monoisotopic 75As (m/z = 75). The most direct method for dealing with this interference is to use a mass spectrometer which has sufficient resolution to separate 40Ar35C1 from "As. High resolution mass spec- trometers due to higher cost are not economical for environ- mental monitoring. On the other hand the chloride interference present in the 'residual' aerosol can be eliminated by using a microporous membrane as the gas-liquid separator. The mem- brane transports the gaseous hydride leaving the chloride interference in the liquid phase.'*4*6i7*26-34 The findings presented here reflect the development of a robust IC-HG-ICP-MS system which has relatively short analysis time picogram detection limits and good long-term reproducibility. The principle optimizations shown are in (a) gas-liquid separator noise; (b) IC condition; and (c) HG conditions.The system's accuracy is evaluated by analysing two National Research Council of Canada (NRCC) certified reference materials (CRMs) NASS-4 and SLEW-2 to deter- mine concentrations of the four arsenic species and the sum of the arsenic concentration was compared with the certified total arsenic value for each of the reference materials. EXPERIMENTAL Instrumentation The ICP-MS instrument was a Hewlett-Packard 4500 (HP 4500) benchtop ICP-MS.Optimized system parameters for the HP 4500 with HG were similar to solution nebulization parameters. The standard HP 4500 utilizes a nickel sampler cone (1.0 mm orifice) and a nickel skimmer cone (0.4 mm orifice). The instrument was tuned by adding 1 ppb arsenic to the HC1 used for HG allowing optimization on a continuous arsenic signal. The rf power was set at 1200 W the plasma gas flow rate at 15 1 min-' and the intermediate gas flow rate at Atomic Spectrometry September 1996 Vol. 11 (893-898) 8931.0 1 min-'. The three-dimensional positioning of the torch relative to the sampling cone dramatically affected the overall noise. A sampling depth of 5.2mm was used with the torch position slightly (- 0.1 mm horizontally 0.8 mm vertically set in the instrument software) off the axis of the sampling orifice thereby reducing noise while maintaining a signal.Chromatographic data were collected in the time resolved analysis mode with an integration time of 2 s per point and peak areas were measured with the chromatographic inte- gration software provided with the instrument. For consistency the integration of the peak area used the average level of the background to define the beginning and end of each peak. The ion chromatograph was a Dionex GPM-2 system with a Hamilton PRP-X100 250 mm anion-exchange column and a 20 mm guard column. Fig. 1 is a schematic of the IC-HG-ICP-MS system. The sample is injected into the ion chromatograph using a 50 pl loop. The IC effluent is mixed in a three-way PTFE manifold with HC1.The HCl is delivered using the HP 4500's built-in peristaltic pump. This mixture flows into a second four-way manifold where NaBH is delivered using a peristaltic pump and argon 'sweep' gas is delivered using the mass-flow control- ler on the HP 4500 which conventionally controls the blend gas flow. After passing through a 27 cm glass bead mixing coil (not shown in Fig. l) the gas-liquid mixture proceeds through the membrane gas-liquid separator (MGLS). The sweep gas the arsines and the H2 migrate across the microporous mem- brane into the argon carrier gas which is introduced directly into the central channel of the ICP torch via a short length of Tygon tubing. Important parameters for the IC-HG-ICP-MS system are summarized in Table 1.The lower half of Fig. 1 is an expanded view of the MGLS. The components for the MGLS depicted in Fig. 1 are commer- cially available inert materials and the construction is similar to one described elsewhere.' This arrangement provides for a gas flow path around the tubular microporous membrane. In the course of the MGLS design (Fig. l) volumes were mini- mized to avoid dead volumes. In practice because of the high gas flow rates compared with the total volume of the separator the exact volume of the separator was not critical although for some microporous membrane tubing separator volume is i m p ~ r t a n t . ~ The membrane utilized was expanded polytetra- fluoroethylene (ePTFE) microporous tubing which has excel- lent gas transport rate and efficiency and is available from International Polymer Engineering Tempe AZ USA.Reagents All reagents and solutions were handled and prepared in a Class-100 clean air hood to avoid arsenic contamination. The Sweep gas y q + T H q chromatograph HCI NaBH4 ,,*' Membrane gasfiquid separator I Carrier gas FIOW constrictor OUT /sweep gas + carrier gas \ - tarsines+H2 sweep gas + arsines +H2 Fig. 1 Schematic of the IC-HG-ICP-MS system Table 1 IC-HG-ICP-MS parameters Ion chromatography- Anion exchange column Guard column Elution mode IC eluent (10 mmol I-' P0,-3-10 mmol 1-' NO3-) Hydride generation system- HCl (10% m/m) NaBH (1% m/m) Sweep Gas (argon) Carrier Gas (argon) ICP-MS- Single ion monitoring Detection mode Integration time per point in chromatogram Hamilton PRP-X100 250 mm x 4.1 mm 20 mm x 4.1 mm Isocratic 1.5 ml min-' 1.5 ml min-' 0.75 ml min-' 40.0 ml min-' 1.28 1 min-' m/z = 75.0 Pulse counting 2.0 s distilled water used was deionized to 18 MQ by a Millipore system.Solutions were prepared in fresh Nalgene polyethylene or Teflon bottles. The injection loops of the IC solution were loaded with polypropylene syringes (Becton Dickinson Franklin Park NJ USA). The syringes were rinsed once with the solution prior to injection. The HC1 (Fisher Fair Lawn NJ USA; ACS +certified) was used after determining that the arsenic concentration within this acid was lower than a 'high-purity' acid. Dilutions were made (m/m) with distilled water. The NaBH (Alfa AESAR Johnson Matthey Ward Hill MA USA; 97 + % pure) solution was made up on a m/m basis.The NaBH was stabilized by adding 7.5 ml of 50% m/m NaOH (Fisher) per litre of solution and fresh solutions were prepared daily. The arsenic solutions were prepared on an arsenic mass basis. Arsenite was derived from solid arsenic( 111) trioxide (SPEX Industries Edison NJ USA) diluted to 1000 ppm in 1% nitric acid and arsenate was prepared from a 1000 ppm standard of orthoarsenic acid in 2% nitric acid (SPEX). Monomethylarsonic acid and dimethylarsinic acid (both 98 % pure) were obtained from Chem Service Chemicals (West Chester PA USA). Working solutions of the arsenic species were prepared by dilution from a 1 ppm (in arsenic) standard in distilled water. The concentration of these 1 ppm standards in distilled water was confirmed by ICP-AES. Two CRMs were used to verify accuracy.NASS-4 is an open ocean water reference material and SLEW-2 is an estuarine water reference material. Both are stabilized with nitric acid and certified for total arsenic by the NRCC. The two CRMs were analysed as received. The arsenic background varied with the source of the phosphate ion and sodium phosphate dibasic heptahydrate (Fisher) was used because it contained a lower arsenic back- ground than other sources of phosphate ion available in our laboratory. The source of nitrate for the IC buffer was concen- trated HNO (Ultrex 11 J. T. Baker Phillipsburg NJ USA). RESULTS Membrane Gas-Liquid Separator The gas-liquid separator in the HG system is based on microporous ePTFE. ePTFE is porous because during the manufacturing process 'nodal' spaces are created as the tubing expands.The distance between the nodes can be controlled and affects the 'density' of the membrane. A lower density (0.45 g ~ m - ~ ) material provided excellent gas transmission. However liquid was observed to permeate through the mem- 894 Journal of Analytical Atomic Spectrometry September 1996 Vol. 1 1brane. A high-density material (0.9 g cm-,) maintained good gas transmission while inhibiting the liquid transport across the membrane. A 15cm length of 0.94mm i d x 1.35mm od tubing removed>95% of the gas from the gas-liquid mixture under the HG conditions of 10% HCl and 1% NaBH used in most of these experiments. More gas was observed to exit the MGLS with increasing HCl and NaBH concentration. This was empirically determined by observing the amount of gas exiting the MGLS.This high gas transport efficiency is important because it allows optimal analyte transmission at a rate conducive to maintaining good chromatographic peak shape. Based on the amount of 40Ar37C1 observed it was concluded that not much liquid permeated through the mem- brane. The ratio of counts per s at m/z=75 to m/z=77 was about 100 1 indicating that most of the background signal at m/z= 75 is due to 75As. There are two novel features for the gas-liquid separator which serve to reduce noise in the ICP-MS signal. These features were evaluated using 1% NaBH mixed with 30% HCl containing 1 ppb AS'" so that the ion chromatograph was not used to introduce the arsenic standard. The use of the novel features reduced the observed RSD from -30% RSD to -5% RSD.These values were determined for a continuous arsenic signal using two hundred 0.1 s integrations. These conditions produce more hydrogen gas than the conditions used in the results section but serve to illustrate the reduction in noise. The first novel feature is the addition of sweep gas in the four-way mixing manifold. When the HCl and the NaBH mix in addition to generating arsines they produce excess hydrogen. Without the sweep gas the gas-liquid mixture entering the MGLS is not very homogeneous i.e. small liquid plugs are separated by gas plugs of varying size. The resulting gas pressure pulsations cause visual fluctuations in the plasma and are thought to increase the noise in the ICP-MS signal.8 By contrast with the sweep gas added the flow entering the gas-liquid separator is much more homogeneous thus reducing the gas pressure pulses entering the plasma and in turn the noise.The RSD is decreased to approximately 50% of its original value (from -28 to - 12%). A sweep gas flow of 40 ml min-' was used which is the lowest flow available with the mass flow controller. To investigate lower flow rates an external mass flow controller was used to supply -15mlmin-' but the noise level was comparable to the 40mlmin-' flow rate. Lower noise was not obtained for higher flow rates (40-70 ml min-') and at >70 ml min-' some increase in the noise of the ICP-MS signal was observed. The second novel feature involves the flow path of gas through the gas-liquid separator.The path parallel to the microporous ePTFE tubing (Fig. l) functions as a 'feedback' loop which tends to dampen pulsations in the plasma caused by excess hydrogen evolved during HG. This reduces the noise to about 50% of its original value (from -12 to -5%). Without the upper feedback loop the plasma visually fluctuates by several millimeters. In the upper gas flow path made from 4.8 mm id Tygon tubing is a small flow constrictor (Fig. l) a Teflon tube. The constrictor serves to divert a small amount of gas into the path containing the ePTFE tubing. Without the constrictor water vapour will condense on the inside of the tubing of the gas-liquid separator. The diameter of the constrictor was chosen as a balance between this concern and the need to control plasma pulsations.The diameter of the constrictor was not optimized but decreased until the con- densed vapour was not observed. For the length of constrictor (1 1 mm) the diameter used was 3.5 mm. Ion Chromatography Speciation of arsenic compounds by HPLC often in the form of IC has attracted much attenti~n.~-~*'-~' In the separation the retention times of As"' AsV MMA and DMMA are dependent on the sample matrix the eluent composition their pK,'s and the hydrophobicity of the column. Gradient elution5 provides excellent separation. Isocratic elutions are desirable because they typically shorten analysis time and will not change the HG conditions during the separation. Isocratic reverse-phase ion-pairing and anion-exchange separations have been proposed based on a variety of chromatographic In an anion-exchange separation the resolution of As"' and DMMA is dependent on the hydr~phobicity.'~ The Hamilton PRP-X100 25 cm anion-exchange column provides a suitable separation of As"' from DMMA.In previous studies with this co1umn,3.9,17,18,23,25 pH=6.0 and 8.0 provided optimal reso- lution. Because the reported levels of phosphate b ~ f f e r ~ ' ~ ~ ~ ~ were found to produce unwanted arsenic background an eluent with lower phosphate concentration was used in this work. To speed up elution at these low concentrations nitrate was added to increase the ion strength. Nitrate was chosen because we could add it in the form of high-purity HN03 to dibasic sodium phosphate. Thus the HNO simultaneously increased the ionic strength and adjusted the pH while minimiz- ing contributions to the arsenic background. Adequate reso- lution and shortened analysis time were obtained for 10 mmol I-' Po4- adjusted to pH = 6.0 with HNO (resulting in about 10 mmol 1-lN03-). Hydride Generation The respective hydrides of the arsenic species3' were generated through the NaBH reaction in HCl.Compromise in optimiz- ing these reagents is necessary in order to achieve good sensitivity for all four species. Fig. 2 represents the changes in peak area for all four arsenic species as a function of percent HCl (m/m) at a 1% NaBH concentration. Each species was injected with an equivalent of 50 pg of arsenic. Each datum in Fig. 2 represents a single measurement. The data were all collected in an isocratic mode so that eluent composition did not influence the HG efficiency.The respective arsenic hydrides from the four species are detected with different efficiencies due to their formation mechanisms and gas transport efficiencies. The AS'" MMA AsV species all show the greatest increase in signal up to 10% HCl. This is contrary to DMMA" which increases up to 5% HCl and then decreases dramatically with increasing HCl concentration. Fig. 3 is a plot of peak area of the four species as a function of percent NaBH (m/m) at a 10% HC1 concentration. Each species was injected with an equivalent of 50pg of arsenic. Each datum in Fig. 3 represents a single measurement. All species show the greatest signal increase up to 1% NaBH,. Between 1% NaBH and 2% NaBH the As"' response decreases significantly those of MMA and DMMA remain virtually unchanged and AsV response increases somewhat.At o I ( - I 0 10 20 30 40 [HCl] (% d m ) Fig. 2 YO HC1 (m/m) for 1% NaBH (m/m) Peak area changes of the four arsenic species as a function of Journal of Analytical Atomic Spectrometry September 1996 Vol. 11 895800 I r+.. I- .- 0 1 2 3 4 [NaBH,] (% d m ) Fig. 3 Peak area changes of the four arsenic species as a function of YO NaBH4 (m/m) for 10% HCl (m/m) 4% NaBH AS'" DMMA and MMA have approximately equal intensities. The intensity for AsV at 4% NaBH is less than the other three species but the absolute difference between the four species is smaller than at lower NaBH concentrations. For example at 1% NaBH the ratio of As"' to AsV is approximately 3.7 while at 4% NaBH the ratio is approxi- mately 1.6.The decrease in peak area between 2% NaBH and 4% NaBH may be attributed to a decrease in the ionization efficiency of the plasma with the increase in H2 gas present or H2 competing with the arsines for transport across the membrane. In selecting compromise HC1 and NaBH concentrations we sought to maximize sensitivity for the most toxic species AS"' while retaining good sensitivity for the other less toxic species particularly AsV and DMMA. As a compromise we selected 10% HCl and 1% NaBH,. Fig. 4 is a chromatogram obtained of the arsenic species is distilled 18 MIR deionized water with IC-HG-ICP-MS with the compromise conditions of 10% HCl and 1% NaBH,. The injection contained 24 28 43 120 pg arsenic for AS'" DMMA MMA and AsV respect- ively and the eluent was 10 mmol 1-' P0,-3-10 mmol 1-' NO3- at pH=6.0.The peak at 0.4min is that of arsenic injected simultaneously with the sample injection downstream from the column and serves as a zero point to time the elutions because the chromatograph controls were not interfaced to the ICP-MS software. The amount of arsenic injected for the timing peak was empirically adjusted (for graphical purposes) so that the intensity of the timing peak would be similar to the intensity of the largest peak in the chromatograph. Table 2 contains the method detection limits (MDL) deter- mined as 3.14 times the standard deviation (c,,-') of 7 blank- subtracted replicate injections of a low level standard36 con- taining the four compounds in distilled 18 MIR deionized water.The AsV detection limit is higher because it is a later eluting peak and also because the compromise HCl and NaBH concentrations do not produce the optimal AsV response (Figs. 2 and 3). Different HCl and NaBH coiicentrations would o 1.00 2.00 3.00 4.00 5.00 6.00 7.00 Time/min Fig. 4 Chromatogram (m/z = 75) obtained in distilled deionized water with IC-HG-ICP-MS. Mass of arsenic injected was 24,28,43 120 pg for AS'" DMMA MMA and As" respectively Table 2 Method detection limits (MDL) for IC-HG-ICP-MS in distilled deionized water As"' AsV MMA DMMA Absolute MDL*/pg 0.6 3.1 1.1 0.7 Calculated MDL*t/ppt 12 62 22 14 Amount injected/pg 2.3 7.9 3.0 3.3 * Calculated from 3 . 1 4 ~ - ~ of 7 replicate injections. t Based on a 50 p1 injection. decrease the AsV MDL but could possibly increase others namely DMMA (Figs.2 and 3). Among gas-liquid separators based on microporous mem- branes,l ,4,6,7,28-34 the membrane chosen plays a role in the stability of the overall system. For example if the membrane allows passage of liquid across it liquid droplets may be introduced sporadically into the plasma resulting in increased ICP-MS noise. For a silicon rubber membrane the trans- mission efficiency seemed to change with time affecting the long- and short-term stability of the m/z=75 signal of an FI-HG-ICP-MS system.' For simplicity and precision how- ever good long- and short-term stability are desirable in an IC-HG-ICP-MS system. The IC-HG-ICP-MS system described here has good short-term stability. For example the RSD values for the four analyte peak areas in Fig.4 have RSD values of 1-6% for three sequential runs over -25 min. The day-to-day system stability was judged by repeating these replicate runs on 4 separate days of a week (days 1 2 3 and 7 ) . In Fig. 5 the average of the 3 replicates for each day is indicated by a vertical bar so for each analyte on the x-axis there are four bars. The day-to-day RSD value calculated from the average of the four bars is indicated above each analyte (Fig. 5). Certified Reference Materials (CRM) Fig. 6 is a chromatogram of undiluted NASS-4 obtained by IC-HG-ICP-MS using 10% HCl and 1% NaBH,. The eluent was 10 mmol I-' P04-3-10 mmol I-' NO3- at pH =6.0. NASS-4 open ocean water is a CRM with a salinity of 31 300 ppm. The peak at 0.3 min is that of arsenic injected simultaneously with the sample injection downstream from the column and serves as a zero point to time the elution.The amount of arsenic injected for the timing peak was empirically adjusted (for graphical purposes) so that the intensity of the timing peak would be similar to the intensity of the largest peak in the chromatograph. The retention times of As"' and AsV in NASS-4 differ from those obtained in distilled water (Fig. 4) as a result of loading of the column by the chloride (-20000 ppm) present in undiluted NASS-4. Nevertheless both peaks are resolved although the AsV peak is somewhat broadened as a result. The signal intensity decreases between 3.0 and 4.4 min. This decrease in intensity may be due to chloride which has Peak area vs time A%(JJJ) JIMMA klMA As(\) Fig.5 Stability of the IC-HG-ICP-MS system over 4 d 896 Journal of Analytical Atomic Spectrometry September 1996 Vol. 1 11.00 2.08 3.00 4.00 5.00 6.00 Time/min Fig. 6 Chromatogram (m/z = 75) of undiluted NASS-4 Open Ocean Water by IC-HG-ICP-MS approximately this retention time on this column in a different buffer system.25 To verify this an NaCl solution (-9000 ppm) was fortified with As"' and the resulting chromatogram con- tained a drop in signal intensity in the 3.0-4.4 min range. The width and magnitude of the signal suppression is correlated to NaCl concentration. It is not clear how the presence of chloride reduces signal intensity. One possibility is that the chloride 'plug' may cause a change in the efficiency of the HG reaction or the chloride 'plug' may exclude the phosphate buffer which contains residual arsenic.High chloride concentration also causes peak splitting for analytes with retention times similar to chloride on this analytical column.25 If NASS-4 is fortified with MMA and DMMA both the MMA and DMMA peaks split and accurate integration of MMA and DMMA intensities are difficult. If the fortified NASS-4 is diluted 1 :3 with distilled water the splitting of the MMA and DMMA peaks in the fortified sample is eliminated and accurate integration of peak areas is possible. When unfortified NASS-4 is diluted 1 3 with distilled deionized water no peaks are observed at retention times corresponding to MMA and DMMA. Given this dilution the detection limits are a factor of 4 greater than those reported in Table2 assuming that the detection limits for these com- pounds in NASS-4 are similar to those in distilled water.SLEW-2 is an estuarine reference material with a salinity of 11 600 ppm. Fig. 7 is the chromatogram of undiluted SLEW-2 obtained by IC-HG-ICP-MS using 10% HCl and 1 YO NaBH,. The eluent was 10 mmol I-' P0,-3-10 mmol I-' NO3- at pH = 6.0. The lower level of chloride (- 6000 ppm) in SLEW-2 permits direct investigation of the levels of DMMA and MMA for which peaks appeared at appropriate retention times (Fig. 7). The identification of the peak labelled as MMA (Fig. 7) is based on the retention time of MMA in a fortified sample of SLEW-2. The chloride eluting in SLEW-2 (Fig. 7) between 4.2 and 5.0 min is narrower than in NASS-4 (Fig.6) W corresponding to the lower concentration of chloride in SLEW-2. The peak at 1.1 min (Fig. 7) is that of arsenic injected simultaneously with the sample injection downstream from the column and serves as a zero point to time the elution. The amount of arsenic injected for the timing peak was empirically adjusted (for graphical purposes) so that the intensity of the timing peak would be similar to the intensity of the largest peak in the chromatograph. Thus the retention window of the chloride relative to the timing peak is between 3.1 and 3.9 min which is less than NASS-4 (2.7 to 4.1 min). NASS-4 and SLEW-2 both certified for total arsenic were analysed using the IC-HG-ICP-MS system. Concentrations were calculated from the peak areas for the individual species (Figs.6 and 7) and were summed to arrive at the total arsenic concentration (Table 3). An external calibration was used to calculate the concentrations from the peak areas. The cali- bration was composed of five blank-subtracted replicates of the solution used in Fig. 4 and a linear pulse-counting detector response is assumed. The total arsenic concentration in NASS-4 is within the 20,-~ limits provided on the CRM certificate and the SLEW-2 is slightly outside the 20 - limits. However in neither CRM does the accuracy of the total certified value differ from the measured value by more than 4% (- 0.030 ppb for SLEW-2). It may be possible to lower this 3% value by the use of standard addition techniques which are not as susceptible to matrix affects as external calibration.Practically however the additional time and inaccuracies induced by the use of standard additions may offset any improvements in the accuracy. In terms of precision the 2 0 ~ ~ value determined here is somewhat larger than the certified value for NASS-4 and much larger for SLEW-2 (Table 3). One reason for this larger 20,-' value is that two and four peak areas were summed for NASS-4 and SLEW-2 respectively. The greater number of peak area measurements contributes to the imprecision compared with the imprecision of the certified value for total arsenic which was made with one measurement (with several replicates). Probably the dominating cause of the imprecision is that the AsV peaks were broadened by the presence of chloride as discussed above.This causes the integration of the peak area to be less precise as does 'tailing' of the chloride peak into the AsV peak. Dilution of the CRM's would reduce the chloride tailing but also would reduce peak area. However imprecisions caused by dilution errors and in the integration of the smaller peak may offset any improvement in the precision of the AsV measurement. CONCLUSIONS The IC-HG-ICP-MS system described has been used to analyse two saline certified reference materials. Despite con- Table 3 Arsenic determination for certified reference materials NASS-4 and SLEW-2 NASS-4 f 20 - 1 (ppb)* As"' 0.29 f 0.01 AsV 0.94k0.14 MMA < 0.080 DMMA < 0.050 Total 1.23+0.14? Certified8 1.26 + 0.09 SLEW-2 X+2an- (ppb)* 0.093 f 0.009 0.66 & 0.1 1 < 0.022 0.034 & 0.01 5 0.79 f 0.1 1$ 0.821 f 0.008 * Five replicate determinations were used.(total) was deter- mined by ~=(CO?)'/~ where ci and cri are the concentration and on-l J of each component i. 0 1.00 2.00 3 M 400 &00 600 7.00 -f MMA and DMMA values not included in X or o - ~ for total for total arsenic Ti me/min arsenic concentration. $ The MMA value was not included in X or 8 Certified for total arsenic. Fig. 7 Chromatogram (m/z = 75) of undiluted SLEW-2 Estuarine Water by IC-HG-ICP-MS concentration. Journal of Analytical Atomic Spectrometry September 1996 Vol. 1 I 897cerns that substances in these materials may interfere with the HG reaction and also cause matrix effects the total arsenic determined with IC-HG-ICP-MS compared favorably with the certified values.Arsenic in these species was composed predominately of AsV but contained a significant amount of As"'. The lower level determined for the more toxic As"' demonstrates the need to accurately quantify this species at low levels in order to evaluate the risk of environmental exposure to arsenic. This work was performed while Matthew L. Magnuson held a National Research Council-US EPA Associateship at the National Exposure Research Laboratory in Cincinnati OH USA. REFERENCES 1 2 3 4 5 6 7 8 9 10 11 12 13 Creed J. T. Magnuson M. L. Brockhoff C. A. Chamberlain I. and Sivaganesan M. J. Anal. At. Spectrom. 1996 11 504. Hwang C.-J. and Jiang S.-J. Anal. Chim. Acta 1994 289 205. Le X.-C. Cullen W. R. and Reimer K. J. Talanta 1994 41,495. Story W. C. Caruso J.A. Heitkemper D. T. and Perkins L. J . Chromatogr. 1992 30 427 and related personal communications. Roehl R. Alforque M. M. and Riviello J. presented at the Winter Conference on Plasma Spectrochemistry Jan. 6-1 1 1992. Branch S. Corns W. T. Ebdon L. Hill S. and O'Neill P. J. Anal. At. Spectrom. 1991 6 155. Wang X. Viczian M. Lasztity A. and Barnes R. M. J. Anal. At. Spectrom. 1988 3 821. Thompson M. and Walsh J. N. A Handbook of Inductively Coupled Plasma Spectrometry Blackie & Sons Glasgow 1983 ch. 6. Alberti J. Rubio R. and Ruaret G. Fresenius' J. Anal. Chem. 1995 351 415. Lopez M. A. Gomez M. M. Palacios M. A. and Camara C. Fresenius' J. Anal. Chem. 1993 346 643. Woller A. Mester Z. and Fodor P. J. Anal. At. Spectrom. 1995 10 609. Thomas P. and Sniatecki K.J. Anal. At. Spectrom. 1995,10,615. Kawabata K. Inoue Y. Takahashi H. and Endo G. Appl. Org. Chem. 1994 8 245. 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 Demesmay C. Olle M. and Porthault M. Fresenius' J. Anal. Chem. 1994,348 205. Inoue Y. Kawabata K. and Endo G. J. Chromatogr. A. 1994 675 149. Hwang C. and Jiang S. Anal. Chim. Acta 1994 289 205. Branch S. Ebdon L. and ONeill P. J. Anal. At. Spectrom. 1994 9 33. Gailer J. and Irgolic K. J. Appl. Org. Chem. 1994 8 129. Schlegel D. Mattusch J. and Dittrich K. J. Chromatogr. A. 1994,683 261. Larsen E. H. Pritzl G. and Hansen S . H. J. Anal. At. Spectrom. 1993 8 1075. Larsen E. H. Pritzl G. and Hansen S . H. J. Anal. At. Spectrom. 1993 8 557. Gjerde D. T. Wiederin D. R. Smith F. G. and Mattson B. M. J. Chromatogr. A. 1993 640 73. Beauchemin D. Siu K. W. M. McLaren J. W. and Berman S . S. J. Anal. At. Spectrom. 1989 4 285. LaFreniere K. E. Fassel V. A. and Eckels D. E. Anal. Chem. 1987 59 879. Low G. K. C. Batley G. E. and Buchanan S . J. Chromatographia 1986 22 292. Buckley W. T. Budac J. J. and Godfrey D. V. Anal. Chem. 1992 64 724. Brockmann A. Nonn C. and Golloch A. J. Anal. At. Spectrom. 1993 8 397. Tao H. Miyazaki A. and Bansho K. Anal. Sci. 1990 6 195. Cave M. R. and Green K. A. J. Anal. At. Spectrom. 1989,4,223. Barnes R. M. and Wang X. J. Anal. At. Spectrom. 1988,3 1083. Wang X. and Barnes R. M. J. Anal. At. Spectrom. 1988,3 1091. Nakata F. Sunahara H. Fujimoto H. Yamamoto M. and Kumamaru T. J. Anal. At. Spectrom. 1988 3 579. Motomizu S. Toei J. Kuwaki T. and Oshima M. Anal. Chem. 1987 59 2930. Pacey G. E. Straka M. R. and Gord J. R. Anal. Chem. 1986 58 502. Kaise T. Yamauchi H. Hirayama T. and Fukui S. Appl. Org. Chem. 1988,2 339. Glaser J. A. Foerst D. L. McKee G. D. Quave S. A. and Budde W. L. Enuiron. Sci. Technol. 1981 15 1426. Paper 6/00893C Received February 7 1996 Accepted May 15 1996 898 Journal of Analytical Atomic Spectrometry September 1996 Vol. 1 1

ISSN:0267-9477    

DOI:10.1039/JA9961100893

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Laser Ablation Inductively Coupled Plasma Mass Spectrometric Transient Signal Data Acquisition and Analyte Concentration Calculation* Analytical Atomic Spectrometry HENRY P . LONGERICH SIMON E. JACKSON AND DETLEF GUNTHER Department of Earth Sciences and Centre for Earth Resources Research Memorial University of Newfoundland St. John's Newfoundland Canada A1 B 3XB Laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) produces complex time-dependent signals. These require significantly different treatment both during data acquisition and reduction from the more steady-state signals produced by solution sample introduction. This paper discusses in detail data acquisition and reduction considerations in LA-ICP-MS analysis. Optimum data acquisition parameters are suggested.Equations are derived for the calculation of sample concentrations and LOD when time-resolved data acquisition is employed sensitivity calibration is obtained from reference materials with known analyte concentrations and naturally occurring internal standards are used to correct for the multiplicative correction factors of drift matrix effects and the amount of material ablated and transported to the ICP. Keywords Laser ablation inductively coupled plasma mass spectrometry; transient signal data acquisition; transient signal analyte concentration calculations; limit of detection Laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) is an extremely powerful analytical technique for spatially resolved in situ analysis of solids for a wide variety of trace elements.The often complex rapidly changing transi- ent signals produced can provide not only quantitative concen- tration data but other information about the sample (i.e. the presence of inclusions and chemical zoning within an ablation volume) and can permit meaningful analyses of contained inclusions both solid' and liquid. A wide variety of different data acquisition conditions have been reported for LA analy- sis.2-6 The conditions cited in a number of papers together with the default settings which are used in the software of some ICP-MS instrumentation suggest that data acquisition parameters for transient signal data acquisition have not been considered in detail by some users and manufacturers. The purpose of the present paper is to discuss data acqui- sition and reduction considerations in LA-ICP-MS analysis and to provide guidance on optimum data acquisition param- eters and data reduction algorithms.Many of the consider- ations are applicable to other transient sample data acquisition protocols. DATA ACQUISITION To achieve the maximum benefit from LA-ICP-MS especially when heterogeneous or multi-phase samples are being analysed * Presented at the 1996 Winter Conference on Plasma Spectrochemistry Fort Lauderdale FL USA January 8-1 3 1996. it is very important to acquire and visually examine the signals as a function of time. Without real-time data acquisition and display important information is lost and is unavailable to the analyst (Fig. 1). Further the ability to observe the signals while the ablation process is taking place is important as it allows the operator to make changes to the laser operating parameters (e.g.laser power and focus) during an analysis. To obtain background intensity a 'gas background' signal (i.e. the signal when the sample is not being ablated) can be obtained during the initial part of the acquisition before 30 60 Time/s Fig. 1 Intensity uersus time for LA of (a) a synthetic garnet and glass sample and (b) a fluid inclusion in quartz. The intensity is on a linear scale and has been normalized to the range. In (a) the A1 and Y are contained in the garnet. During the ablation the underlying glass was penetrated as signified by the quickly rising signal of Ba. In processing the signal from the garnet integration of the analytes would end when the Ba signal increased sharply.In (b) opening of a fluid inclusion is indicated by rising intensities of Bi and T1 (dissolved in the fluid) following ablation through the host quartz (shown by the Si signal) Journal of Analytical Atomic Spectrometry September 1996 Vol. 11 (899-904) 899ablation is started. Selected signal intervals are then integrated and net count rates are determined for each element from background count rates and gross analyte ablation count rates measured subsequently in time. While a very few specialized ICP-MS instruments are now available which use simultaneous detection most instruments use quadrupole mass selection and are operated in a fast peak hopping mode in which intensities for various m/z values are detected sequentially.Laser ablation signals being noisy even when compared with other transient signals require a data acquisition sequence that very rapidly jumps between m/z values in order to approach simultaneous detection. Instrument manufacturers use different terminology for the various data acquisition parameters that define the sequence which has no doubt added confusion among operators attempting to keep the syntactic definitions in mind. In this laboratory there are both SCIEX and Fisons/VG quadrupole-based ICP-MS instrumentations but the terminology of the SCIEX software will be used in the present paper where applicable. ‘Number of elements’ ‘quadrupole settling time’ ‘dwell time’ ‘sweeps per reading’ ‘readings per replicate’ ‘number of replicates’ and ‘points per spectral peak’ are all parameters (not independent) that define a data acquisition procedure. A quadrupole mass spectrometer is a sequential device in which a sweep is defined as one cycle of the mass spectrometer in which data are acquired on each of the selected m/z values.Readings and replicates are available in the SCIEX but not the VG software. ‘Readings’ are the mean of a number of sweeps. The mean of a set of ‘readings’ is a ‘replicate’. Quadrupole Settling Time Quadrupoles are very fast peak hopping mass selectors. However they require a finite time to settle which imposes a limitation on scanning speed. Quadrupole settling time is the time for the detector signal to stabilize (to some arbitrary level) following a discontinuous change (peak hop) in the selected m/z value.The quadrupole settling time can be software controlled (and might or might not be adjustable by the operator) or can be hardware controlled. Further the time could or could not be different depending upon the magnitude of the m/z jump. The largest required settling time is expected when the change of m/z is the largest. If intensity data are acquired sequentially in order of increasing m/z the ‘fly back’ a term borrowed from television technology when the large jump is made from the highest to the lowest selected m/z (e.g. U to Li) requires the greatest settling time. The effect of choosing a settling time which is too short will be exacerbated when the jump is from or over a very intense peak to a small spectral peak and will result in a detected signal which is biased high.Conversely when the two peaks are of the same magnitude and close together the error will be much smaller. When short dwell times are used the settling error will be larger than when using longer dwell times. The settling time for quadrupole mass spectrometers is between 0.1 to 10 ms. If the time is set at too low a value it could be necessary to insert an additional ‘dummy’ mass in the sequence from which data are acquired but are not used. It was found necessary to do this in several data acquisition protocols. However where selectable a quadrupole settling time of 1.5 ms was found to be adequate for most but not all applications. It is the short settling time of quadrupole mass analysers that makes them useful for transient signal data acquisition.Instruments that use magnetic mass analysers are by compari- son slower to settle following an m/z jump. The time for magnetic sector instruments to settle is variable and depends upon the magnitude of the m/z jump and the m/z values involved since magnetic sectors do not have linear relationship of mass to magnet current. Although recent developments have produced significant improvements in switching speed in mag- netic instruments they are at the present time less suitable for transient sampling applications than quadrupole-based instruments. Dwell Time Knowledge of the quadrupole settling time is required in order to make an appropriate selection of the dwell time the integrating time on each selected m/z. Often the software allows dwell time to be independently selectable for different m/z values but for simplicity only analyses in which a constant dwell time is used for all m/z values will be described.In some data acquisition software the dwell time might not be treated as an independent variable but instead could be a dependent variable which is set indirectly by adjusting the time per slice (sweep) until the desired dwell time results. Selecting this parameter requires a compromise between a long dwell time which maximizes counting efficiency (the time spent acquiring data relative to system overhead) and an infinitely short dwell time that allows the system to approach in a mathematical sense the ideal of simultaneous data acquisition. A reasonable compromise is to select a dwell time that is approximately six times the quadrupole settling time (e.g.84 ms dwell time for a 1.5 ms quadrupole settling time). Then approximately 15% of the total time is lost to overhead while approximately 85% of the total time is used for data acqui- sition. Dwell times longer than 10ms result in small but insignificant increases in counting efficiency and fewer mass sweeps in the total acquisition period resulting in reduced ability to characterize fast changing transient signals precisely. Owing to the presence of noise from the mains frequencies (60 Hz in North America and 50 Hz in many other countries) dwell times which are an integer multiple of the mains fre- quency are recommended since noise-power spectral measure- ments made on ICPs often show noise that is related to power line frequencies.With 60 Hz power 60 Hz harmonics will be present along with 120Hz the frequency of rectified 60Hz power. When three-phase power sources are used (common with older vacuum tube ICP power supplies) 180 Hz will also be present. Since half a cycle is 83ms one cycle (1 60s) is 16$ms dwell times of 84 16 or 50ms are recommended. In countries where the mains are 50 Hz (one cycle is 20 ms) slightly longer dwell times of 10 20 or 40 are recommended. In most of the present work 50ms were used for solution nebulization data acquisition and ca. 8.3 ms for laser sampling data acquisition. A typical silicate mineral element analysis element menu consists of 25 elements two major element internal standards (e.g. Ca and Ti) fifth-row elements (Sr Y Zr and Nb) sixth- row elements including all the lanthanides (Ba La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf and Ta) and the actinide elements (Th and U).More specialized analyses can determine fewer elements but a set of 25 elements is very typical. Using the suggested compromise dwell time of 8; ms ( tdwell) with a quadrupole settling time of 1.5 ms gives 85% counting efficiency i.e. 85% of the time is used in acquiring data and 15% of the time consumed in overhead. Measuring one point per spectral peak (see below) the time per sweeps (isweep) is given by tsweep = lZelernents( tdwell -k tquadsettling) where tqudsettling is the quadrupole settling time and nelernents is the number of elements. With a sequence of 25 elements a dwell time of 84 ms and 1.5 ms quadrupole settling time tsweep = 246 ms.A typical analysis consisting of approximately 1 min of background measurement followed by approximately a 1 min measurement of sample ablation requires approxi- 900 Journal of Analytical Atomic Spectrometry September 1996 Vol. 11mately 480 sweeps and gives an integrated acquisition time for each element of 2 s. Sweeps per Reading Some data acquisition software allow the averaging of a number of sweeps into a ‘reading’. If the instrument data acquisition software does not allow this it can be accomplished off-line by averaging (typically 3-10) sweeps. Averaging of the signals is helpful to reduce the noise and allow the user to distinguish small analyte signals from the background (note that one detected ion in 8.3 ms equates to a count rate of 120 counts s-l which is large compared with instrument back- grounds that are often less than 10 counts s-’ for quadrupole instruments).Combining sweeps also reduces the size of the resulting data files which can be very large (e.g. 25 elements x 480 sweeps per analysis). However excessive aver- aging (the limit being when all of the data are integrated into a single replicate) can result in the loss of important information contained in fleeting signals or spikes related to real and interesting small inclusions of different phases. The choice of the number of sweeps per reading is thus a compromise between noise reduction (large number of sweeps per replicate) and loss of information from real fine structure in the signals.Experience has shown that integration should not reduce the replicate time to less than approximately 1 s (three sweeps per reading). The number of readings per repli- cate is always set to one and the number of replicates is selected by the operator for the appropriate total acquisition time. Points per Spectral Peak Most ICP-MS data acquisition software allow the user to select the minimum step size along the m/z axis of a spectrum. This step size could be for example 0.1 u (10 points per m/z value) or 0.05 u (20 points per m/z value). Users often report the use of three or more points per peak where a point at the peak centre is acquired along with one or more on the high and one or more on the low m/z side. Some users select a large number of points per peak (e.g.seven or nine) the equivalent of a total integration. Consideration of the consequences of the selection of the number of points per peak clearly indicates that one is the optimum number of points per peak because one point per peak results in (1) the maximum signal (2) the minimum change of intensity with any change in calibration of m/z (3) the maximum counting efficiency (fraction of total time used for data acquisition) when the total sweep time is fixed and (4) the best abundance sensitivity. (1) Signal. That one point per peak results in the maximum signal is intuitively obvious since using more than one point per peak results in the averaging of a measurement at the peak centre where intensity is the largest with measurements made off the peak centre where intensities are smaller.(2) Mass calibration. That measurements made using one point per peak are less sensitive to a change or drift in the calibration of m/z is not intuitively obvious and the expectation that using three or more points per peak will minimize the effects of drift in m/z calibration is presumably the reason that users often select more than one point per peak. The reader can confirm that it is better to use only one point per peak by acquiring signal intensities over a spectral peak and calculating a moving average of a fixed number of points across the peak. As the set of data points selected is moved from the peak centre the mean changes by a larger amount as the number of points per peak is increased. (3) Counting eficiency.When more than one point per peak is acquired several quadrupole settling times per peak are used. Assuming a constant quadrupole settling time (1.5 ms) is used for all m/z changes and a dwell time of 85ms dwell one point per peak and a 25 element analysis a counting effici- ency of 85% is obtained. If three points per peak are chosen and the same time per sweep is maintained [246ms=25 x (8$ + lS)] the dwell time would have to be reduced to 1.78 ms and the efficiency would be lowered to 54%; this results in a reduced total integration time per element of from 2.0 to 1.3 s. (4) Abundance sensitivity. In inorganic quadrupole MS abundance sensitivity is the most important measure of mass spectrometer resolution. When adjusting m/z calibration and resolution the operator will probably use as a measure of resolution the peak width at some fraction of the maximum intensity (e.g.1 m/z unit at 10% of the maximum peak intensity). However abundance sensitivity which is a measure of the interference from adjacent peaks is a more important measure of resolution that measures the interference owing to tailing of the signal of an adjacent ion on the m/z peak being determined. The abundance sensitivity is the ratio of the intensity at rn/z m t 1 or m- 1 relative to the intensity at m/z m when there is a large ion signal at m/z m and there are no ions of rn/z rn + 1 or m - 1 present. It is intuitively obvious that abundance sensitivity will be degraded if off-peak points which lie closer to adjacent peaks are included in the integration.Data Acquisition and Sampling Protocol Once the data acquisition parameters have been selected a sampling sequence (i.e. samples calibration materials and reference standards) must be established. While most data acquisition parameters will be similar regardless of application the sample sequence will vary significantly depending upon application. Summarized below is the data acquisition and sampling protocol used in this laboratory which specializes in high spatial resolution analysis of geological minerals in petrographic sections. Typical data acquisition parameters These are as follows dwell time = 85 ms; quadrupole settling time = 1.5 ms; number of m/z sweeps for total analysis = 480 where 240 are used for background and 240 are used for ablation; sweeps per reading = 3; number of elements = 25 (typical but can vary from 2 to 50); points per peak = 1; and total analysis time = ca.120 s. Data acquisition (a) Acquire approximately 1 min of data with the laser beam blocked to obtain the background. Note that this is an instrumental background as opposed to a reagent blank. More time can be used but the uncertainty in the background does not improve greatly with increased time since the uncertainty is inversely proportional to the square root of the integration time. When very low instrument backgrounds are encountered it is possible that zero ions can be detected giving a count rate of zero with accompanying analytically unrealistic esti- mated detection limit of zero. In this situation one can be added to the total number of background ions to produce a higher non-zero estimate of the background.It is rec- ommended that the laser actually be on and firing with the beam blocked during background acquisition time. This is because in the past electrical pick up by the ion detector of electrical noise from the laser firing circuits has been observed. This also helps to stabilize the laser output. (b) Ablation is commenced (the beam path unblocked) and data are acquired for up to about 1 min using a pulse energy and repetition rate predetermined to produce the required Journal of Analytical Atomic Spectrometry September 1996 Vol. 11 901ablation intensity (i.e. pit size signal). The time can be reduced when boring very shallow pits or when the ablation penetrates through the sample into the support as indicated either visually or by the signal for one or more diagnostic elements in the sample the epoxy or the underlying glass support.Petrographic selections are now prepared in this laboratory using epoxy spiked with Bi an element which occurs at very low concentrations in most samples so that the concentration of Bi is 200 pgg-' in the solid epoxy. This is achieved by premixing bismuth 2-ethylhexanoate (Alfa Aesar Johnson Matthey Ward Hill MA USA) with epoxy resin (Buehler epoxide resin 20-8 130-032 Tech-Met Ontario Canada). When mounting the rock section to a glass slide the resin is mixed as usual four parts resin with one part hardener (Buehler epoxide hardener 20-8 132-032). Acquiring data Data are acquired in runs of up to 20 analyses.Each run starts and ends with two data acquisitions on a calibration material. The 20 analysis limit ensures that calibration is performed on no more than about an hourly basis in order to monitor and correct for the drift of the inter-element sensitivities (ie. analyte-internal standard) with time. Reference materials are also analysed at least twice during a run to monitor accuracy. Selection and Integration of Background and Analyte Data In the analysis of the data it is first necessary to select the data intervals which are to be integrated. Data Selection Choose the time intervals (replicates) over which to integrate (the mean is more convenient to calculate noting that an integration is the mean count rate multiplied by time) the background and the ablation count rates (Fig.1). The first background interval is not used and 2-3 replicates immediately before the initiation of the LA are not used. The first few (typically 1-10 depending upon the application) ablation repli- cates are generally rejected in order to avoid the effects of laser removal of surface contamination and to allow the signal to stabilize. Data integration is terminated at the point when the laser starts to sample the support an underlying mineral phase or an inclusion or other anomalous chemical heterogeneity within the sample or the signal falls significantly (< FZ 50%) either owing to termination of ablation or laser defocusing. While rejection of this tail results in loss of some signal it gives rise to higher signal-to-background noise for the inte- grated interval and consequently improved detection limits.7 Calculate mean background intensity Defining variables nb = the number of sweeps in the back- ground; and tdwell = the dwell time.For each element calculate the mean background intensity (counts s-'). Calculate mean gross analyte ablation intensity Defining variables n = the number of sweeps of gross analyte signal. For each element calculate the mean gross intensity (counts s-l). Apply background correction Subtract from the gross mean analyte intensity (counts s-') the mean background intensity (counts s-l). Since the inte- gration time of the background is not usually the same as the ablation integration time the mean count rates are used. Concentration Calculation In LA analysis calibration can be carried out using external calibration when samples are carefully matrix matched and the time and power of the ablation are reproduced precisely.However in the same manner that internal standards are used routinely to improve precision and accuracy in solution nebul- ization ICP-MS the use of 'naturally occurring' internal stan- dards in LA produces a more robust calibration. The use of internal standards corrects for the sources of bias which are multiplicative i.e. sensitivity drift matrix effects and differences in the volume (mass) of the sample ablated relative to the calibration material (ablation sensitivity correction). Biases that are additive (e.g. polyatomic ion interferences) are not corrected by the use of internal standards.Since the addition of internal standards is not easily and routinely carried out with solid samples 'naturally occurring' elements are used as internal standards. These are elements that are found in both the samples and calibration material and for which the con- centrations are known in both materials. The concentration of the internal standard can be obtained from an analysis using an alternative method or from the known elemental stoichiometry when crystalline materials are analysed. The concentration of the analyte element in the sample (CANSAM) is given by the count rate for the analyte (RAN,,,) in the sample divided by the normalized sensitivity (S) as follows The normalized sensitivity (S) is the sensitivity determined on a calibration standard (CAL) corrected for the volume (mass) of sample ablated.When using naturally occurring internal standards the sensitivity (counts s-' per unit of concentration) normalized to the mass of the sample (SAM) ablated in the determinations is where RANCAL is the count rate of the analyte in the calibration material; CANCAL is the concentration of the analyte in the calibration material; RISsAM is the count rate of the internal standard in the sample; RISCAL is the count rate of the internal standard in the calibration material; CIS, is the concentration of the internal standard in the sample; and CIscALI is the concentration of the internal standard in the calibration material. The sensitivity (S) is prone to amass dependent drift which can change with time. This is a particular problem in the analysis of geological samples where a light major element is generally used as an internal standard which could suffer from significant drift in sensitivity relative to much heavier trace element analytes.In the absence of more than one internal standard it is assumed that this relative drift occurs linearly with time. Consequently a correction to the sensitivity ratios of analyte and internal standard is applied using a linear interpolation (with time) between the calibration samples ana- lysed at the beginning and the end of the run. Limit of Detection The amount of material ablated in LA sampling is often significantly different for each analysis. Consequently LOD are different for each analysis and must be calculated for each individual acquisition.The first step in the calculation is to find the standard deviation of an individual sample count rates in the selected background interval using the well known equation for the calculation of the standard deviation of an individual determi- nation in a population (in commercial spread-sheet software 902 Journal of Analytical Atomic Spectrometry September 1996 Vol. 1 1this function is often referred to as the sample standard deviation as opposed to the population SD) Uindividual = Jz (x - n b - 1 where Xi are the individual measurements and nb is the number of determinations of the background accumulated. This standard deviation is a measure of the distribution of the individual determinations in the set of nb samples of the count rate.However what is required is not a measure of the distribution of the individual determinations of the back- ground but the distribution of the mean of nb background measurements. That is if the data acquisition was repeated a large number of times how would the mean background count rate be distributed or what would be the standard deviation of the mean. The standard deviation of the mean is given by the well known but less well understood equation Oindividual 6mean background = - In the calculation of the LOD it is necessary to estimate the variation of an ablation signal that would be found if the sample contained no analyte. It is assumed that the standard deviation of an individual determination from the signal derived from ablation of a sample with zero concentration is the same as it is for a gas background interval.Thus the standard deviation of an individual determination in the ablation region is also assumed to be dindividual. The standard deviation of the mean ablation signal is then Oindividual O'mean ablation = - A3 where na is the number of slices in the ablation time interval that are integrated. The net count rate for an analyte in any analysis is the difference between the gross count rate for the selected ablation interval and the background count rate The rules of propagation of error through the subtraction give gRnet2 = ORablatian' + ORbaskgroun where the o values are of the mean since it is the means that are subtracted and it is the standard deviations of the means which are of interest. Thus g.. . 2 individual Oindividua12 +- - Oindividua? (k + ;) na or 1 OR, = Oindividual 4; + If na=nb=n as is often the situation then the following simplification results Oindividual Users often omit the square root of two in LOD calculations which results in a lower estimate of the LOD. Consideration of the propagation of error through the subtraction described above gives a more realistic estimate of the LOD. Defining the LOD as three times the standard deviation of a sample that contains zero analyte then This converts the variation in the net intensity from count rates into appropriate concentration units. The sensitivity (S) is the sensitivity normalized to the amount of sample ablated and includes all multiplicative factors that are corrected by the use of the naturally occurring internal standard.Note that the standard deviation that is calculated is an estimate of the variation in a population. There is an uncer- tainty in the calculated standard deviation which follows a chi-squared distribution having the notable characteristics that negative values are not possible and that the distribution tails asymmetrically in the positive direction. When the standard deviations are compared the uncertainty in the standard deviations must be considered; consequently when one or even more experiments demonstrate a lower standard devi- ation than does another experiment it might not be possible to conclude that the difference is statistically significant. Only when a large number of estimates of the standard deviation are obtained do the estimates of the standard deviation approach the true standard deviation.CONCLUSIONS While most of the discussions in the present paper are amenable to experimental demonstration many of the conclusions can be more easily demonstrated using mathematical modelling and statistical considerations. For example experiments to demonstrate the effect of calibration drift in m/z are difficult to carry out since it is not easy to operate an instrument with a known drift. Furthermore experiments can be carried out using conditions that provide accurate and precise analysis over a short period of time but do not demonstrate what happens under conditions of severe instrumental drift or when unusual samples are analysed. From practical and theoretical considerations together with simple mathematical modelling the following acquisition par- ameters are recommended for data acquisition using LA sampling with quadrupole-based ICP-MS detection ( 1) ana- lyses should be performed using time-resolved data acquisition software; (2) a pre-ablation background of similar duration to the ablation should be acquired; (3) for instruments with a quadrupole settling time c a .1.5 ms a dwell time is rec- ommended that is about six times the quadrupole settling time and is a multiple of the mains frequency i.e. 8$ ms for 60 Hz mains or 10 ms for 50 Hz mains; (4) one point per peak should always be used; ( 5 ) pre-integration of small groups of sweep^^-^ provides some smoothing and reduces data file size while allowing fine detail to be recognized; and (6) for all real samples that are heterogeneous to some extent selection of signal intervals for integration should be performed only after careful inspection of time-resolved signals.To calculate sample concentrations and LOD the necessary equations are summarized below. The sensitivity ( S ) is calcu- lated when using naturally occurring internal standards from the signals obtained from a calibration material of accurately known concentrations. Sensitivity is normalized to the mass or volume of sample ablated in a determination. Correction for mass dependent drift should be applied using measurements on the calibration material made before and after the sample data are acquired. The concentration of an analyte in a sample is then given by The LODs in LA-ICP-MS which must be calculated for each Journal of Analytical Atomic S p e c t r o m e t r y September 1996 VoZ. 11 903analyte in each analysis are given by Ingo Horn acquired the garnet sample data shown as an example of time-resolved data. 1 S n 3cindiviciual Ji +- Detection limit = This work was carried out as part of the development of the ICP-MS facility at Memorial University of Newfoundland. Two ICP-MS instruments and the LA system were purchased with Natural Sciences and Engineering Research Council of Canada (NSERC) equipment grants and the system is operated with the assistance of an NSERC infrastructure grant and numerous NSERC research grants including one to H.P.L. A grant from Fisons Instruments with matching funding from NSERC under the Industrial Oriented Research programme is also acknowledged. The generous support of Memorial University to the operation of the facility is also acknowledged. REFERENCES Taylor R. P. Jackson S. E. Webster J. L. and Jones P. Geochim. Cosmochim. Acta submitted. Pearce N. J. G. Perkins W. T. and Fuge R. J. Anal. At. Spectrom. 1992 7 595. Garbe-Schoenberg C.-D. and McMurtry G. M. Fresenius’ J. Anal. Chem. 1994 350 264. Jarvis K. E. and Williams J. G. Chem. Geol. 1993 106 251. Durrant S. F. Fresenius’ J. Anal. Chem. 1994 349 768. Jackson S. E. Longerich H. P. Dunning G. R. and Fryer B. J. Can. Mineral. 1992 30 1049. Denoyer E. At. Spectrosc. 1992 13 93. Paper 51075071; Received November 16 199.5 Accepted February 27 1996 904 Journal of Analytical Atomic Spectrometry September 1996 Vol. 11

ISSN:0267-9477    

DOI:10.1039/JA9961100899

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

CUMULATIVE AUTHOR INDEX JANUARY-SEPTEMBER 1996 Abou-Shakra Fadi R. 61 Acheson Barbara M. 765 Adams Freddy C. 201 Akatsuka Kunihiko 69 Akiyama Masayuki 69 Alberini G. 731 Allen Lori B. 526 Angeles Quijano M. 407 Apostoli Pietro 519 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 Bavazzano Paolo 519 Beato Emilio Romero 37 Becker J. Sabine 643 661 Begerow Jutta 303 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 Can0 Pavon J. M. 107 Caroli Sergio 773 Caruso J. A. 491 633 Cavalchi B. 731 Ceulemans Michiel 201 Chakrapani G. 815 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. Cordero Bernard0 Moreno 37 Crain Jeffrey S. 523 Creed John T. 504 893 Dams Richard 543 Dannecker Walter 723 723 537 797 407 Daskalova Nonka 567 Davoli V. 731 Debrah Ebenezer 127 De Grootte F. 623 Denoyer Eric R. 127 Depalma Jr. Patrick A.483 De Regt J. M. 623 Dietze Hans-Joachim 643 661 Ding W-W. 225,421 Dolan Scott P. 307 Donard Olivier F. X. 871 Dorfman Ethel 811 Dunemann Lothar 303 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 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 Sanchez Soledad 37 Garcia Albert0 Menhdez 561 Gauthier Gilles 787 Gavrieli Ittai 811 Gentscheva Galja 567 Gercken Berthold 371 Ghazi A.Mohamad 667 Gijbels R. 841 Goodall Phillip 469 Goodall Phillip S. 57 Gregoire D. Conrad 359 765 Greibrokk Tyge 117 Grubb Anders 735 Gunther Detlef 899 Gutitrrez Ana Maria 407 Halicz Ludwik 81 1 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 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 313 723 571 325 Imai Shoji 513 601 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 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 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 Koh Lip Lin 585 Kojima Isao 607 Koller Dagmar 187 Koppenaal David W. 3 17 Kotrebai Mihaly 343 Krivan Viliam 159 371 Krushevska Antoaneta 343 Kumamaru Takahiro 11 1 Lafontan Silvyane 759 Lasztity Alexandra 343 Lau Nancy 479 Lerat Yannick 213 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 Lorthioir Stephane 759 Luan Shen 247 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 483 821 Masera Eric 213 Massart D. L. 149 237 Matveev Oleg I. 393 Mauchien Patrick 2 13 Mazzoli A. 731 McCandless Tom E. 667 McCrindle Robert I. 437 McGaw Brian 297 Medina Bernard 71 3 Michel Robert G. 279 Moens Luc 543 Moissette A. 177 893 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 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 11 7 Pelchat Jean-Claude 779 787 Pelchat Pierre 787 Penninckx W. 237 Perez Pavon 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. 31 Puchades Rosa 99 Quentmeier Alfred 537 Quintal Manuelita 447 Rademeyer Cornelius J. 437 RaspopoviC Zoran 325 Rayman Margaret P. 61 Remy Bernard 213 Rhoades Jr. Charles B. 751 Riter Ken L. 393 Roberts David J. 231 259 Roberts N. B. 133 Rosendahl Kerstin 519 Rubio Marcelo 123 Ruette Fernando 447 Ruiz Joaquin 667 Sabbioni Enrico 591 Sadler D.A. 207 463 Saito Kengo 513 601 Sanchez Hector J. 123 Sanz-Medel Alfredo 561 571 Saprykin Anatoli I. 643 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 Slavin Walter 25 Slavova Petranka 567 Journal of Analytical Atomic Spectrometry September 1996 Vol. 11 905Smeyers-Verbeke J. 149 237 Smith Benjamin W. 393 479 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 53 689 Ting Bill G. 339 Tittes Wolfgang 797 Trentini P. 731 Treshchalov Aleksei B. 649 TripkoviC Mirjana 325 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 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 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 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 739 689 Yang Karl X. 279 Yang Kuei-Lin 139 Yang Li-Li 265 Yang Mo-Hsiung 413 Yang Wenjun 33 1 Ybaiiez Nieves 271 Yoshida Thomas M. 861 You Jianzhang 483 Yuzefovsky Alexander I. 279 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 COPIES OF CITED ARTICLES The Royal Society of Chemistry Library can usually supply copies of cited articles. For further details contact The Library Royal Society of Chemistry Burlington House Piccadilly London W1V OBN UK. Tel +44 (0) 171-437 8565; fax +44 (0) 171-287 9798; Telecom Gold 84; BUR210; Electronic Mailbox (Internet) LIBRARY@RSC.ORG.If the material is not available from the Society’s Library the staff will be pleased to advise on its availability from other sources. Please note that copies are not available from the RSC at Thomas Graham House Cambridge. Atomic Spectroscopy Group Study Bursaries The Atomic Spectroscopy Group Analytical Division RSC invites applications from UK scientists working in the field of analytical spectrometry for study bursaries. These typically will have a value not exceeding f500 and are intended to afford applicants the opportunity for professional development. Specific activities for which the study bursary might be considered include attendance at meetings workshops and seminars or support for study visits to other laboratories.Applications should include a statement of the purpose for which the bursary is sought (1 page A4) and a summary of recent work (1 page A4). The submission should be sent to Dr S. J. Hill Chairman A.S.G. Department of Environmental Sciences University of Plymouth Drake Circus Plymouth PL4 8AA at least 2 months before funds are required. 906 Journal of Analytical Atomic Spectrometry September 1996 Vol. 11

ISSN:0267-9477    

DOI:10.1039/JA9961100905

出版商:RSC    

年代:1996

数据来源: RSC