JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 145 Plasma Mass Spectrometry Consider the Source" Invited Lecture Brenda S. Sheppard US Food and Drug Administration National Forensic Chemistry Center 7 147 Central Pkwy Cincinnati OH 45202 USA Joseph A. Caruso Department of Chemistry University of Cincinnati Mail Location 7 72 Cincinnati OH 45221 USA The use of mixed-gas helium and nitrogen plasmas as alternative sources for plasma mass spectrometry is discussed in this paper. These plasmas are used to alleviate some of the problems inherent in argon inductively coupled plasma mass spectrometry (ICP-MS). Spectroscopic and non-spectroscopic interferences as well as sensitivities for high ionization potential elements are addressed. Reduced-pressure plasmas are used for the determination of P and S.In addition applications of chromatographic techniques such as gas supercritical-fluid and high-performance liquid chromatography with alternative plasma sources are included. Some problems found in the application of argon ICP-MS can be reduced or eliminated with alternative sources and sub-ng to sub-pg levels of detection for halogens can be achieved. Keywords Inductively coupled plasma; microwave-induced plasma; mass spectrometry The need for ultra-trace level elemental analysis has been the stimulus for the development and improvement of analytical techniques. Plasma source mass spectrometry is one technique that is currently of interest for multi-element analysis at the part per billion and part per trillion levels. Methods used to increase the applicability and sensitivity of this technique are of interest. The inductively coupled plasma was developed as a source for use in atomic emission but has also been widely used as an ion source for elemental mass spectrometry.Plasma mass spectrometry was developed in the 1970s and early 1980s'-'1 and has been reviewed el~ewhere.'~-'~ In the last decade or so inductively coupled plasma mass spectrometry (ICP-MS) has become a powerful elemental analysis tool and is in some cases more valuable than atomic emission spectrometry (AES). Some advantages of ICP-MS over ICP-AES are the excellent sensi- tivity selectivity and ability for isotope dilution. The argon ICP has proven to be the most useful and widely applied source for plasma MS. It has been utilized extensively with solid liquid and gaseous sample introduction techniques. However there are ionization difficulties and problems with spectral and matrix interferences that warrant the investigation of alternative plasma source^.'^-^^ Several alternative plasma sources have been reported including helium nitrogen mixed gas reduced pressure ICPs and microwave-induced plasmas (MIP). Certain applications require investigations of glow discharge sources although these will not be discussed here.In this paper the use of mixed-gas plasma^,^^-^' helium ~ l a s m a s ~ l - ~ ~ and nitrogen plasmas5s57 as alternative sources is discussed. Studies have been carried out with both ICPs and MIPS. Reduced pressure helium plasmas formed either in a m i c r ~ w a v e ~ ~ - ~ ~ or r.f.field63*64 offer attractive possibilities with gaseous sample introduction. Also interesting are the potential applications of these plasmas for the determination of non-metals that are difficult to carry out with the argon ICP source; determination of phosphorus sulfur and halogen containing compounds are examples of these. Gas and supercritical fluid chromatographic sample introduction pro- vides an excellent opportunity to introduce gaseous samples for these compounds with sub-nanogram to sub-picogram levels of detection. * Presented in part at the XXVIII Colloquium Spectroscopicurn Internationale (CSI) York UK June 29-July 4 1993. Inductively Coupled Plasmas Argon ICPs are the most widely used ion sources for plasma MS because of their stability excellent detection limits and wide linear ranges.Trace level detection capabilities make the Ar ICP ideal for analysis of environmental and biological samples where the elements of interest are sometimes present at parts per billion to parts per trillion levels. The analysis of some of these samples are complicated by high concomitant element concentrations which could lead to polyatomic and other matrix interferences. The argon ICP is not an efficient ionization source for difficult to ionize halogens and other elements with higher ionization potentials. Additionally the formation of analyte oxides diminishes the sensitivity for some elements. Alternatives to the 100% argon ICP have been investigated to alleviate some of these problems and have included mixed gas as well as helium ICPs.Spectroscopic interferences on analytes of interest can be a problem in ICP-MS work. For example the use of hydrochloric acid for sample dissolution or the presence of a high chloride matrix will interfere with the determination of 51V+ or 75As' because of interferences by 35Cl'60 + and 40Ar35C1 + respect- ively. Mixed-gas plasmas have been investigated for reduction of these and other matrix effects. The addition of hydrogen,33 air,34 xenon4' and helium2* to various portions of the plasma gas have been studied. These gases primarily have been added to the aerosol carrier and outer gas flows to reduce the formation of polyatomic interferences and the formation of analyte oxides. Improvements in detection limits as well as reduction of oxides and interfering species have resulted.Interfering polyatomic ions such as N2+ HN2+ NO' ArH' ClO' Arc' ClOH+ and ArO' were reduced significantly with the addition of xenon to the carrier gas. However the use of xenon can be cost prohibitive. The addition of nitrogen to the carrier and outer gas flows has been successfully used to reduce the ArCl+ and C10+ interferences on arsenic and vanadium as well as to reduce the formation of MO' and ArO' species. The main attribute of the mixed gas plasma is its ability to decompose refractory elements and reduce the formation of some plasma gas polyatomic species. However the use of some gases such as nitrogen increase the amount of mass spectral interferences at other masses and therefore must be used only in appropriate cases.Alternative gas plasmas are not only used to reduce oxides and interfering oxide species but they have also been used to146 JOURNAL OF ANALYTICAL ATOMIC SPE;CTROMETRY MARCH 1994 VOL. 9 improve the sensitivity of high ionization potential elements. For example a helium-argon ICP has been used to improve sensitivity for high ionization potential (IP) element^.^^-^' The partial replacement of argon with helium produces a plasma with better analyte ionization capabilities for some elements than a 100% argon ICP. Such a plasma is capable of ionizing elements with high first IPS such as the halogens in halide salts. Detections limits were improved for the higher IP elements such as arsenic and bromine and for some metals. For example detection limits with argon ICP-MS have been reported as 0.40 1.7 and 0.02 ng ml-' for arsenic bromine and iodine respectively.These detection limits are improved using the helium-argon ICP (20% helium) to 0.006 0.07 and 0.006 ng ml- for arsenic bromine and iodine re~pectively.~~ The helium-argon ICP can be used for the mass spectrometric detection of high IP elements without degrading sensitivity for other elements of interest. This source can be used with only a few modifications to existing commercial instrumentation. Helium ICPs have also been studied as possible sources for plasma MS.4143 Several modifications to existing instrumen- tation were necessary and include modifying the load coil plasma impedance network and torch. The helium ICP pro- duced a mass spectrum above m/z 40 that was free from background interferences. Increased sensitivity for gaseous samples containing bromine chlorine sulfur and fluorine was also found in comparison with argon ICP-MS.In further work this group has reported on the introduction of aqueous samples to helium ICP-MS. Significant improvements in sensi- tivity in comparison with argon ICP-MS have yet to be achieved. Helium ICPs show promise as ion sources for elements with high IPS. This plasma source warrants further investigation. Helium Microwave-induced Plasmas The determination of halogenated compounds at increasingly low levels is of great environmental importance. The halogens have higher ionization energies in comparison with most of the elements of interest The ionization and excitation capabili- ties of the argon ICP are not sufficiently great to achieve the sensitivity needed.In addition major argon background inter- ferences exist that hinder the determination of some elements. A helium plasma is better suited for the determination of halogens. The helium MIP is an attractive alternative to an ICP because it is compact and relatively inexpensive. The formation of helium plasmas can be accomplished with relative ease with MIPs in comparison with ICPs. Several modifications to the ICP mass spectrometer are required in order to use an MIP. The ICP torch box has to be removed and replaced with an MIP cavity a microwave generator is used instead of the r.f. generator and additional pumping capacity is added to the expansion stage of the spectrometer.The sampling orifice is also usually less than 1 mm. A fundamental study of the sampling process in a helium MIP mass spectrometer has been undertaken by Chambers et ~ 1 . ~ ' The ion transportation process was studied in order to determine what modifications to the ICP-MS interface were needed to operate an MIP-MS instrument effectively. These workers concluded that with this type of plasma it is important to minimize air entrainment and maintain a high ion flux through the interface. Smaller sampling cone and skimmer cone orifice diameters can be used to reduce air entrainment; however this will limit the ion flux. They also discussed the proper placement of the skimmer cone in relation to the Mach disk. Helium MIPs have been used as ion sources for detection of gas-phase species species in aqueous solutions as well as coupled to gas chromatography (GC) high-performance liquid chromatography (HPLC) supercritical-fluid chromatography (SFC) and electrothermal vaporization (ETV). Most of the work with the helium MIP has been conducted in the area of detection of gas-phase species.Many of the background inter- ferences associated with an argon plasma can be effectively reduced with helium plasmas as has been shown by Brown et and Satzger ct who were able to detect halogens in the low pg s-l range by modifying the MIP-MS interface. A nitrogen sheath gas was used in conjunction with a quartz bonnet to reduce the amount of entrained air in the plasma. Sensitivity for bromine chlorine and iodine introduced as gaseous mixtures of CH,Br CH,Cl and CH,I was improved by reducing the amount of air entrained in the plasma.47 Detection limits for 79Brf 35Cl+ and 1271+ were 1.2 21 and 1.8 pg s-l respectively.Additionally Douglas and co- w o r k e r ~ ~ ~ ~ found that the helium MIP source is free from ioni- zation interferences for sodium concentrations up to 100 ppm and found good correlation with the certified values of several Standard Reference Materials (Orchard Leaves Water Low- Alloy Steel) and reported on the determination of lead in blood by isotope dilution. Reduction in background interferences and improved sensi- tivity for halogens make the helium MIP ideal for use with GC. Microwave-ind uced plasmas can be easily interfaced with gas chromatographs. A gas chromatograph was coupled to an MIP-MS system for the detection of chlorinated brominated and iodinated compounds.48 A tangential flow torch was used in this study and the sampling orifice was reduced to 0.4mm with an aluminium sampler.Absolute detection limits of approximately 1 pg for bromine and iodine were found while the absolute detection limit for chlorine was 10-20 pg. Higher detection limits were found for chlorine because of raised backgrounds at m/z values of 35 and 37. Organotin compclunds are also of considerable environmen- tal importance. The most common separation method for the speciation of organotin compounds is GC. A gas chromato- graph was interfaced to a helium MIP-MS system for the detection of six orgatnotin species.49 The compounds of interest were tetravinyltin i.etraethyltin tetrabutyltin triethyltin bro- mide tripropyltin chloride and tributyltin chloride.Initially a standard tangential flow MIP torch was used in this study. It was found that tin forms refractory oxides in an MIP which deposit on the walls of the torch thereby decreasing sensitivity. Additionally the analyte can diffuse throughout the plasma prior to reaching the sampling orifice. A tantalum tube was used to introduce the effluent from the gas chromatograph directly into the centre of the plasma and reduce the distance between the plasma and the sampling orifice. The sensitivity for tin compounds was increased by a factor of ten with use of the tantalum injector. Detection limits at sub-picogram levels were achieved and linear dynamic ranges of three orders of magnitude were obtained.One disadvantage associated with helium plasmas has been their inability to tolerate aerosol introduction. The solvent load imparted by direct solution nebulization can de-stabilize the plasma. Most helium MIPs are operated at powers of less than 400 W whereas ICPs are operated at approximately 1 kW. A helium MIP has been successfully interfaced with a mass spectrometer for the analysis of aqueous aerosols. A tangential flow torch with a glass aerosol injector was used and is described in detail in the original m a n ~ s c r i p t . ~ ~ The aerosol injector prclduces an analyte rich region similar to that in an ICP. The sample introduction system consisted of a MAK nebulizer and a cooled double-pass chamber.Detection limits for chloride bromide and iodide were 39 0.18 and 0.04 ppb respectivdy. These elements were detected as positive ions at m/z 35 37 79 and 127. Detection of metals with this experimental set-up was also investigated. Detection limits were comparable or slightly improved when compared with an argon ICP. A reversed-phase HPLC system was directly coupled to a helium MIP-MS instrument for the element-selective detection of halogenated organic compounds. The tangential flow torchJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 147 with the glass aerosol injector used by Creed et aLS1 was used in this experiment. However a helium concentric nebulizer was used instead of a MAK nebulizer. The plasma was able to tolerate up to 70% methanol.Absolute detection limits were 50 pg of bromine for brominated compounds 1 pg of iodine for iodinated compounds and l o n g of chlorine for chlorinated compounds. Linear dynamic ranges were 3-4 orders of magnitude for bromine and iodine. This work shows that helium MIP-MS is a viable detector for some applications of HPLC. Helium MIP-MS can be successfully used as a mass selective detector for both HPLC and GC separations. However there are some compounds that would be better separated by SFC. Non-volatile thermally labile compounds of fairly high relative molecular mass can be separated. Also unlike HPLC SFC provides gaseous sample introduction to the plasma with virtually 100% transport efficiency. The coupling of SFC with helium MIP-MS was investigated for 1-chloronaphthalene and l-bromo-2-methylnaphthalene.s3 In this study a demountable tangential flow torch was used.The SFC and MIP systems were interfaced by connecting a transfer line from the SFC to the torch. The transfer line was run through a stainless-steel tube and wrapped in heating tape to maintain a constant temperature. Detection limits at the low to sub-picogram levels were achieved for chlorine and bromine thus indicating the potential of this technique. Electrothermal vaporization has also been used as a sample introduction device for helium MIP-MS.s4 The electrothermal vaporizer has several advantages over solution nebulization including greater sample transport efficiency and the possibility of separating analytes from interfering matrix species by use of temperature programming.Sample sizes are generally in the pl range not in them1 range. In this work a tantalum-tipped electrothermal vaporizer was used to ensure that no analyte condensed during transport to the plasma and also to facilitate the formation of an annular helium plasma. Detection limits for silver cadmium lead and bromine were 0.03 0.09 0.75 and 1.5 pg respectively. Detection limits were blank limited by contamination from furnace components. Microwave-induced plasmas are under-utilized sources for plasma mass spectrometric detection. The MIP is an excellent source for gas-phase sample introduction providing figures of merit comparable to or better than ICP. Liquid samples can be analysed with this source at sufficient powers however this source is more susceptible to non-spectroscopic matrix inter- ferences than is the argon ICP.Reduced Pressure Microwave-induced Plasmas Background interferences from entrained gases can be mini- mized with the use of a low-pressure plasma. The plasma is isolated from contamination owing to atmospheric entrainment and uses lower gas flow rates. Gas flow rates for reduced-pressure use are 250 ml min-l compared with 1 1 min- ’ for atmospheric pressure MIPs. These features mini- mize the ionization of plasma gas impurities and/or the forma- tion of polyatomic ions that interfere with the determination of certain low-mass elements. The low-pressure plasma could be the solution to certain low-mass interferences such as P (m/z 31) S (m/z 32) and C1 (m/z 35 and 37). Interferences associated with these m/z values could include l4NI6OH+ l6O 2 9 + 160180H+ I6O2H3+ and 36ArH+.The signal-to- background ratios associated with m/z values 35 37 31 and 56 were reduced using the low-pressure MIP in comparison with the atmospheric pressure MIP.” This background reduction should now allow for more accurate determinations of chlorine phosphorus and sulfur. The low-pressure MIP-MS system has been used as a detector for GC for the detection of phosphorus and sulfur in pesticide^.^^ Malathion and diazinon were chosen as two widely available pesticides containing these elements. It was found that phosphorus reacted with the hot quartz plasma discharge tube to form phosphorus oxides on the walls of the torch and thus resulted in decreased sensitivity. The torch was modified by the addition of an air cooling jacket.A 100-fold improvement in the detection limit for phosphorus was realized with the air-cooled torch. However oxide formation was still a problem even with this improvement. Phosphorus in triethyl phosphite was detected in the 1-90ng range depending on the torch cooling. Detection of sulfur at m/z 32 was still not possible owing to a large background signal presumably from 02+. Nitrogen has a lower IP than helium and thus produces a less energetic plasma. A significant reduction in the formation of 0,’ was noted with the nitrogen plasma and made the detection of sulfur possible. Less reaction of phosphorus with the torch walls was also observed. Detection limits for phosphorus and sulfur in diazinon were 0.79 and 0.51 ng respectively .Further modifications to the low-pressure helium MIP torch were made in order to improve detection of phosphorus and some of the halogens. A water-cooled torch was developed to reduce the phosphorus interaction with the torch walls.62 A small percentage of hydrogen gas was added to the plasma to act as a reagent gas to scavenge the phosphorus before it reacted with the hot quartz walls and to reduce formation of other polyatomic species. A tantalum injection tube was used to transfer the GC effluent directly into the plasma. A seven component pesticide mixture with chlorine phosphorus sulfur and bromine containing compounds was analysed. Sub- nanogram detection limits were achieved for all elements. Several groups have investigated the use of low-pressure MIPs to obtain fragmentation of organic and organometallic corn pound^.^^^' Fragmentation in combination with total compound decomposition would allow both structural and quantitative elemental analysis with the same instrumental set-up.Poussel et aL6’ used a low-pressure surfatron MIP interfaced to a mass spectrometer and obtained soft ionization and fragmentation comparable to the conventional electron impact source for a variety of compounds including propanoic acid chloroform limonene and dodecane. The sample was injected into the expansion part of the plasma otherwise total decomposition occurred. Heppner” combined GC with low- pressure MIP-MS and obtained fragmentation of organic compounds to a greater degree. A preliminary study was conducted by Olson et using a low-pressure MIP for fragmentation of organic compounds where the sample was introduced into the expansion stage of the mass spectrometer and thus into the tail flame of the plasma.The sampling cone was modified to accommodate this method of sample introduc- tion by adding a 1 mm channel to the side of the sampler cone. In this configuration the sample was introduced midway between the sampler and skimmer orifices. This paper describes the interface in full detail.61 Low powers and flow rates produced parent ion peaks and major fragments similar to those arrived at using conventional electron impact sources for compounds such as hexane toluene o-xylene 1-chlorohexane 1-chloroheptane chlorobenzene and bromo benzene. Nitrogen Microwave-induced Plasmas Moderate-power nitrogen MIPs have been investigated as alternative sources for plasma MS.55-57 Shen et a1.55756 investi- gated the use of nitrogen MIPs for plasma mass spectrometric detection of potassium calcium chromium arsenic and sel- enium.The background mass spectrum of the argon ICP contains many argon-containing polyatomic ions (36ArH+ 40Ar2+ etc.) which interfere with the determination of potass- ium calcium chromium arsenic selenium and iron. The background mass associated with a nitrogen MIP is much less 3 8 ~ ~ ~ + 4 0 ~ ~ + 4 0 ~ ~ ~ + 4 0 ~ ~ 1 2 ~ + 4 0 ~ ~ 1 4 ~ + 4 0 ~ ~ 1 6 0 + Y 9 7148 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 complicated in comparison with that of an argon ICP.ss,s7 Background species interfering with potassium calcium and selenium are greatly reduced or eliminated.In addition ArCl' and Arc + polyatomic species are not present in significant amounts to hinder the detection of chromium arsenic and selenium.ss However an interfering polyatomic species was present at m/z 56 probably N4+ which complicated the determination of iron at this isotope. The reduction or elimin- ation of argon-containing polyatomic species allows for the determination of calcium potassium chromium arsenic and selenium at their major isotopes. Significant improvement over the argon ICP for some detection limits was achieved using the nitrogen MIP. Detection limits for 39K and 40Ca were 0.48 and 0.22 ppb with the nitrogen MIP and 1000 ppb for 39K and 5 ppb for 44Ca with the argon ICP.s6 Arsenic and selenium were free from ArCl' interferences with the nitrogen plasma.The nitrogen MIP-MS system allowed determination of iso- tope ratios for 10 ppm of calcium (40Ca:44Ca) and 100 ppb of potassium (39K:41K) chromium (s2Cr:s3Cr) and selenium (soSe:78Se) with less than 5% error.s6 Nitrogen MIPS have been shown to be an effective source for mass spectrometric detection. High-purity nitrogen can be obtained almost any- where at reasonable cost. Low-pressure Inductively Coupled Plasma Atmospheric pressure ICPs have numerous polyatomic inter- ference species associated with them. The entrainment of air coupled with the plasma gas causes the formation of these species. Low-pressure ICPs can be generated to reduce the impact of these interferents on analytical determinations. The low-pressure ICP is interfaced to the mass spectrometer in a similar manner as the low-pressure MIP.A modification to the sampling cone has been made incorporating an ultra-Torr fitting.63 The torch was connected to the fitting. The pumping capacity of the expansion stage was also increased. No modifi- cations were made to the matching network. Two types of low-pressure torch were investigated. The first was a regular Fassel style torch with a water-cooled jacket.63 However it was found that the jacket was not necessary and the torch was re-designed. The second torch was a quartz discharge tube similar to an MIP torch connected to gas lines with ultra- Torr fittings.64 Low-pressure ICPs have been generated using argon air carbon dioxide nitrogen and helium.The first torch was coupled to a gas chromatograph for the detection of 1 -bromon~nane.~~ The second low-pressure torch was also interfaced with a gas chromatograph for the detection of halides in organic compounds.64 Bromobenzene benzylbro- mide chlorobenzene and chloroheptane were investigated. Detection limits were in the 3-8 pg range for all compounds. These detection limits are comparable to those found with helium MIP.48 The advantage was the relative ease of setting up the low-pressure ICP system in comparison with an MIP. Further studies to improve the matching network are necessary to reduce the reflected power. This should assist in improving the analytical performance. Conclusions Inductively coupled plasma mass spectrometry has become an accepted method for trace metal analysis.Alternative plasma sources can be used successfully to improve determinations of various elements such as the halides and to eliminate interfering polyatomic species. The use of an argon-nitrogen ICP is an effective means of controlling some interferences and is not cost prohibitive as are other methods. This plasma has been successfully used by a number of workers to improve the detection of elements such as arsenic. In specific cases mixed- gas plasmas can be easily used with present instrumentation and minimal cost to improve detectibility of elements of interest in particular high ionization potential elements. Modest modifications to existing instrumentation can be made in order to use low-pressure plasmas when air entrainment creates interfering po1;yatomic species.The MIP shows high potential for use as an alternative source for analysis of gas-phase samples. At present this technique is under-utilized because commercial instrumen- tation is not readily available. In addition the nitrogen MIP shows great promise as an alternative source since it provides comparable detection levels to arsenic ICP-MS but with a less expensive more readily available and highly purified plasma gas. The development of commercial instrumentation should take the MIP from use as a research tool to everyday use in service laboratories. There is still much work to be done in the area of alternative plasma sources to improve trace element determinations. 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 36 37 38 References Houk R.S. Fassel. V. A. Flesch G. D. Svec H. J. Gray A. L. and Taylor C. E. Anal. Chem. 1980 52 2283. Gray A. L. Proc. Xoc. Anal. Chem. 1974 11 182. Gray A. L. Analyst 1975 100 289. Gray A. L. Anal. Chem. 1975 47 600. Houk R. S. Svec H. J. and Fassel V. A. Appl. Spectrosc. 1981 37 380. Houk R. S. Montaser A. and Fassel V. A. Appl. Spectrosc. 1983 37 425. Douglas D. F. Quan E. S. K. and Smith R. G. Spectrochim. Acta Part B 1983 38 39. Date A. R. and Gray A. L. Analyst 1981 106 1255. Date A. R. and Gray A. L. Analyst 1983 108 159. Date A. R. and Gray A. L. Spectrochim. Acta Part B 1983 38 29. Gray A. L. and Date A. R. Analyst 1983 108 1033. Douglas D. J. and Houk R.S. Prog. Anal. At. Spectrosc. 1985 8 1. Houk R. S. and Thompson J. J. Mass Spectrom. Reu. 1988 7 425. Hieftje G. M. and Vickers G. H. Anal. Chim. Acta 1989 216 1. Beauchemin D. McLaren J. W. and Berman S. S. Spectrochim. Acta Part B 1987 42 467. Gregoire D. C. Spectrochim. Acta Part B 1987 42 895. Tan S. H. and Horlick G. J. Anal. At. Spectrom. 1987 2 745. Gillson G. R. Douglas D. J. Fulford J. E. Halligan K. W. and Tanner S. D. Anal. Chem. 1988,60 1472. Vandecasteele C. Nagels M. Vanhoe H. and Dams R. Anal. Chim. Acta 1988 211 91. Hutton R. C. and Eaton A. N. J. Anal. At. Spectrom. 1988,3,547. Brotherton T. J. Shen W. L. and Caruso J. A. J. Anal. At. Spectrom. 1989 4 39. Thompson J. J. and Houk R. S. Appl. Spectrosc. 1987 41 801. Crain J. S. Hod; R. S. and Smith F. G.Spectrochim. Acta Part B 1988 43 1355. Olivares J. A. and Houk R. S. Anal. Chem. 1986 58 20. Gregoire D. C. Appl. Spectrosc. 1987 41 897. Satzger R. D. Anal. Chem. 1988 60 2500. Sheppard B. S. Shen W. L. Davidson T. M. and Caruso J. A. J. Anal. At. Spectrom. 1990 5 697. Sheppard B. S. Shen W. L. and Caruso J. A. J. Am. SOC. Mass Spectrom. 1991 2 355. Sheppard B. S. Caruso J. A. Heitkernper D. T. and Wolnik K. A. Analyst 19'32 117 971. Wang J. Evans 13. H. and Caruso J. A. J. Anal. At. Spectrom. 1992 7 929. Craig J. M. and Beauchemin D. J. Anal. At. Spectrom. 1992 7 937. Hill S. J. Ford M. J. and Ebdon L. J. Anal. At. Spectrom. 1992 7 719. Louie H. and Soo S. Y.-P. J. Anal. At. Spectrom. 1992 7 557. Lam J. W. H. and Horlick G. Spectrochim. Acta Part B 1990 45 1313.Lam J. W. H. arid Horlick G. Spectrochim. Acta Part B 1990 45 1327. Lam J. W. and McLaren J. W. J. Anal. At. Spectrom. 1990 5 419. Evans E. H. and Ebdon L. J. Anal. At. Spectrom. 1989 4 299. Evans E. H. and Ebdon L. J. Anal. At. Spectrom. 1990 5 425.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 149 39 40 41 42 43 44 45 46 47 48 49 50 51 52 Beauchemin D. and Craig J. M. Spectrochim. Acta Part B 1991 46 603. Smith F. G. Wiederin D. R. and Houk R. S. Anal. Chem. 1991 63 1458. Montaser A. Chan S. K.. and Koppenaal D. W. Anal. Chem. 1987 59 1240. Koppenaal D. W. and Quinton L. F. J . Anal. At. Spectrom. 1988 3 667. Nam S. Masamba W. R. L. and Montaser A. Anal. Chem. 1993 65 2784. Douglas D. J. and French J. B. Anal. Chem. 1981 53 37.Chambers D. M. Carnahan J. W. Jin Q. and Hieftje G. M. Spectrochim. Acta Part B 1991 46 1745. Brown P. G. Davidson T. M. and Caruso J. A. J. Anal. At. Spectrom. 1988 3 763. Satzger R. D. Fricke F. L. Brown P. G. and Caruso J. A Spectrochim. Acta Part B 1987 42 705. Mohamad A. H. Creed J. T. Davidson T. M. and Caruso J. A. Appl. Spectrosc. 1989 43 1127. Suyani H. Creed J. Caruso J. and Satzger R. D. J. Anal. At. Spectrorn. 1989 4 777. Creed J. T. Mohamad A. H. Davidson T. M. Ataman G. and Caruso J. A. J. Anal. At. Spectrom. 1988 3 923. Creed J. T. Davidson T. M. Shen W. L. Brown P. G. and Caruso J. A. Spectrochim. Acta Part B 1989 44 909. Heitkernper D. Creed J. and Caruso J. A. J . Chrom. Sci. 1990 28 175. 53 54 55 56 57 58 59 60 61 62 63 64 Olson L. K. and Caruso J. A. J . Anal. At. Spectrom. 1992,7,993. Evans E. H. Caruso J. A. and Satzger R. D. Appl. Spectrosc. 1991 45 1478. Shen W. L. Davidson T. M. Creed J. T. and Caruso J. A. Appl. Spectrosc. 1990 44 1011. Shen W. L. Davidson T. M. Creed J. T. and Caruso J. A. Appl. Spectrosc. 1990 44 1003. Wilson D. A. Vickers G. H. and Hieftje G. M. Anal. Chem. 1987 59 1664. Story W. C. Olson L. K. Shen W. L. Creed J. T. and Caruso J. A. J. Anal. At. Spectrom. 1990 5 467. Heppner R. A. Anal. Chem. 1983 55 2170. Poussel E. Mermet J. M. Deruaz D. Beaugrand C. Anal. Chem. 1988 60 923. Olson L. K. Story W. C. Shen W. L. and Caruso J. A. J. Anal. At. Spectrom. 1990 5 471. Story W. C. and Caruso J. A. J. Anal. At. Spectrom. 1993,8,571. Evans E. H. and Caruso J. A. J . Anal. At. Spectrom. 1993,8,427. Castillano T. M. Giglio J. J. Evans E. H. and Caruso J. A. presented at FACSS XIX Philadelphia PA USA September 20-25 1992 paper no. 532. Paper 3/04991 D Received August 17 1993 Accepted November 4 1993