Determination of Copper Cadmium and Lead in Sediment Samples by Slurry Sampling Electrothermal Vaporization Inductively Coupled Plasma Mass Spectrometry MING-JYH LIAW AND SHIUH-JEN JIANG* Department of Chemistry National Sun Yat-Sen University Kaohsiung Taiwan 804 Republic of China Ultrasonic slurry sampling-ETV-ID-ICP-MS was applied to the determination of Cu Cd and Pb in several sediment samples. The influence of instrument operating conditions slurry preparation and non-spectroscopic and spectroscopic interferences on the ion signals and accuracy and precision of isotope ratio determination was investigated. The isotope ratios of each element were calculated from the peak areas of each injection peak. The precision of isotope ratio determination was better than 5%. The method was applied to the determination of Cu Cd and Pb in a harbour sediment reference material (PACS-1) and in a sediment sample collected from the Taiwan Straits.The accuracy was better than 6% and the precision was better than 12%. The concentrations of Cu Cd and Pb determined in the sediment samples by the ID method were compared with the results of external calibration and standard additions methods. Keywords Ultrasonic slurry sampling; electrothermal vaporization; isotope dilution; inductively coupled plasma mass spectrometry; copper; cadmium; lead; sediment sample Most analyses by ICP-MS are carried out on solutions using a conventional pneumatic nebulizer. However the type of analytical tasks which can be solved by ICP-MS can be extended by using a number of other sample introduction techniques which can be easily adapted to ICP-MS. ETV is one of the sample introduction techniques that is currently employed in ICP-MS/AES.'-13 This alternative technique to solution nebulization presents several advantages including improved sensitivity small sample size requirements and the capability for solids analysis.Perhaps the most notable benefit of ETV-ICP-MS is the possibility to perform direct solids a n a l y ~ i s . ~ ~ Ultrasonic slurry sampling is one of the methods for direct solid sample introduction that has been successfully used in ETAAS.14-25 More recently this approach has been extended to ETV-ICP-MS.293 Compared with traditional sample prep- aration methods such as acid digestion and dry ashing slurry sampling offers several benefits including reduced sample prep- aration time reduced possibility of sample contamination and decreased possibility of analyte loss prior to analysis.Furthermore slurry sampling combines the benefits of solid and liquid sampling and permits the use of conventional liquid sample handling apparatus such as an a ~ t o s a m p l e r . ~ * ~ y ~ ~ - ~ ~ Isotope dilution (ID) techniques have been applied in several previous ICP-MS application^.^^'^^ ID is well recognized as a definitive analytical technique for the determination of trace elements. Since another isotope of the same element represents the ideal internal standard for that element ID results are * To whom correspondence should be addressed. 1 Journal of I Analytical 1 Atomic Spectrometry expected to be highly accurate even when the sample contains high concentrations of concomitant elements and/or loss of the analyte element occurs during sample preparation or during sample introduction into the ICP instrument.In this work ultrasonic slurry sampling-ID-ICP-MS was used to determine the concentrations of Cu Cd and Pb in several sediment samples. The influence of instrument operating conditions slurry preparation and non-spec- troscopic and spectroscopic interferences due to the matrix on the ion signals and the precision and accuracy of isotope ratio determination was investigated. The concentrations determined in the slurries by the ID method were compared with the results of external calibration and standard additions methods. The method was used for the determination of Cu Cd and Pb in a harbour sediment reference sample (PACS-1) and in a sediment sample collected from the Taiwan Straits.EXPERIMENTAL Apparatus and Conditions A Perkin-Elmer Sciex (Thornhill Ontario Canada) ELAN 5000 ICP-MS instrument equipped with an HGA-600MS electrothermal vaporizer was used. Pyrolytic graphite coated graphite tubes and platforms were used throughout. The transfer line consisted of 80cmx6mm id PTFE tubing. The sample introduction system included a Model AS-60 auto- sampler equipped with a USS- 100 ultrasonic slurry sampler. Teflon autosampler cups were used. The USS-100 was set at 12 W (30% power) and a 25 s mixing time was used to mix slurries before injection of 10 pl sample aliquots for analysis. The experimental conditions for ICP-MS and ETV are described in Table 1.ICP operating conditions were selected to maximize sensi- tivity for the isotopes of interest in order to obtain the best precision and accuracy for isotope ratio determination. The ICP conditions were selected to maximize ion signals while a solution containing 10 ng ml-' of Cu Cd and Pb in 1 % HNO was continuously introduced with a conventional nebulizer. The sensitivity of the instrument could vary slightly from day- to-day. The ICP operating conditions used throughout this work are summarized in Table 1. Mass spectrometer parameters used for isotope ratio measurements are listed in Table 1. The measurements were made by peak hopping rapidly from one mass to another staying only a short time (dwell time) at each mass.For the best accuracy and precision for isotope ratio determination a 10 ms dwell time was used. Ion lens voltages were set to obtain the best ion signals for the elements studied simultaneously. For the measurement of Cu and Pb in the PACS-1 sediment sample an offset voltage was applied to one of the ion lenses Journal of Analvtical Atomic Svectrometrv. AuPust 1996. Vol. 1 1 (555-560 1 555Table 1 Equipment and operating conditions ICP mass spectrometer- Outer gas flow rate/l min-l Intermediate gas flow rate/l min-l Carrier gas flow rate/l min-' Rf power/kW Sampler/skimmer HGA-6000MS electrothermal vaporizer- Sample voIume/pl Drying stage (20 s ramp) Charring stage ( 3 s ramp) Cooling stage ( 5 s ramp) Vaporization temperaturePC Heating rate/"C s-' Time at maximum temperature/s Cooling stage ( 5 s ramp) Clean-up stage Heating ratePC s-' Time at maximum temperature/s Cooling stage ( 5 s ramp) Internal gas flow rate/l min-' Internal gas flow rate/l min-' Data acquisition- Dwell time/ms OmniRange setting for Cu and Pb Scan mode Sweeps per reading Readings per replicate Signal measurement mode 14 0.8 0.9 1.1 Nickel 10 120 "C for 30 s 0.3 200°C for 10 s 0.3 20°C for 5 s 2500 Maximum power heating 5 20°C for 5 s 2700 "C for 5 s Maximum power heating 5 20°C for 5 s 10 Variable Peak hopping 50 Integrated of the mass spectrometer via the OmniRange facility.This was done to reduce the sensitivity of the ICP-MS instrument. A Laser Coulter Series LS 100 (Coulter Electronics Hialeah FL USA) was used for the determination of the particle size distribution of the sediment samples.Reagents Trace-metal grade HNO (70% m/m) was obtained from Fisher (Fair Lawn NJ USA). Triton X-100 was obtained from Sigma (St. Louis MO USA). Enriched isotopes purchased from the Oak Ridge National Laboratory (Oak Ridge TN USA) included 65Cu0 ''%do and 204Pb(N03)2. Stock solu- tions of approximately 500mg 1-' of each isotope were pre- pared by dissolution of an accurately weighed amount of the material in HNO and dilution to volume. The concentrations of the spike solutions were verified by reversed spike ID-ICP-MS. Preparation of Slurries A harbour sediment reference material PACS-1 [(National Research Council of Canada (NRCC) Ottawa Canada] was obtained to demonstrate the applicability of the method to real samples.The slurry was prepared by the following pro- cedure. A 0.05 g portion of the reference material was trans- ferred into a 25ml flask. Suitable amounts of HNO and Triton X-lo0 were added so that the final solution contained 2% v/v HNO and 0.1% v/v Triton X-100. After a suitable amount of enriched isotope had been added the slurry was diluted to the mark with pure water. The slurry was then sonicated for 30min in an ultrasonic bath and 1 ml aliquots were removed as needed for analysis with the use of a pipette while the slurry was being mixed with a vortex mixer. These aliquots were then deposited in the Teflon autosampler cups for analysis. A blank was carried through the procedure as outlined above to correct for the presence of any analyte in the reagents used for sample preparation.The sediment sample 556 Journal of Analytical Atomic Spectrometry August 1996 collected from the Taiwan Straits was treated by a similar procedure. Standard Additions and External Calibration Methods A 0.05 g portion of the sediment sample was transferred into a 25ml flask. Suitable amounts of HN03 and Triton X-100 were added so that the final solution contained 2% v/v HNO and 0.1% v/v Triton X-100. After the mixture had been spiked with various amounts of element standards the slurries were diluted to the mark with pure water and analysed by ultrasonic slurry sampling-ETV-ICP-MS using the procedure described under Preparation of Slurries. The concentrations of Cu Cd and Pb were determined from the calibration graphs (standard additions method).In the second method of calibration (exter- nal calibration method) multi-element standard solutions were prepared with concentrations of 50-1000 ng ml-' of Cu and Pb and 1-7 ng ml-' of Cd in 2% HNO and 0.1% Triton X-100 solution. AIiquots (10 pl) of the standard solutions were analysed by ETV-ICP-MS and the calibration graphs were obtained for each element studied. The concentrations of Cu Cd and Pb in the sediment samples were then determined against the sensitivities of these calibration graphs. Isotope Dilution Calculation The analyte concentration in the sample was calculated from the following equation R,-Rt X,' M R,-R X M' c,=ct x ___ x - x - where C is the concentration of analyte C the concentration of the spike R the isotope of the spike R the natural isotope ratio R the experimentally determined isotope ratio X the natural abundance of isotope B X,' the abundance of isotope B in the enriched spike M the relative atomic mass of the analyte element and M the relative atomic mass of the spike.Owing to the mass discrimination effect intensities obtained during isotope ratio determinations were used to calculate the isotopic abundance of each element. RESULTS AND DISCUSSION Selection of ICP-MS Operating Conditions Various factors can influence the precision and accuracy of isotope ratios measured by ICP-MS. Both mass spectrometer and ICP operating conditions can affect the accuracy and precision of the measured isotope ratio^.^^-,^ Several mass spectrometer parameters could affect the pre- cision and accuracy of isotope ratio determination and the dwell time of the mass spectrometer could be the critical parameter.Since an ETV sampling device was used and the signal measurements for the two isotopes of each element were not made simultaneously too long a dwell time could have a deleterious effect on the precision of isotope ratio measure- ments for a transient signal. The dwell time must be sufficiently short to allow the ELAN software to acquire data points at a rate that is sufficient to avoid distortion of the peak shape for each isotope. Hence a 10 ms dwell time was used. Ion settings could affect the relative sensitivities of heavier and lighter isotopes and the measured isotope ratio.40 The count rate had a large effect on the precision obtained due to the counting statistics.In this work ion lens settings were adjusted to maximize sensitivity for the isotope of interest simultaneously. Since the concentrations of Cu and Pb in the PACS-1 sample were too high for the simultaneous determination of the three elements the OmniRange facility was used for Cu and Pb. Vol. 111 I I I I A Pb 201 B 20sPb/207Pb - - 63c"/65c" - 1 I I I I I I 0 2 4 6 8 10 OmniRange of Cu and Pb Fig. 1 Effect of OmniRange settings for Cu and Pb on A ion signal and B isotope ratio determination. Concentrations of Cu and Pb were 0.904 and 0.808 pg ml-' respectively This was done to reduce the sensitivity of the ICP-MS instru- ment. As shown in Fig. 1 although the signals for Cu and Pb decreased with an increase in the OmniRange value the ratios of Cu and Pb remained constant.In order to obtain optimum signals for Cu and Pb determination an OmniRange value of 5 and 8 was set for Cu and Pb determination respectively in the PACS-1 sample. ICP conditions were selected to obtain the best ion signals for the elements studied in order to obtain the best precision and accuracy for isotope ratio determination. The optimum operating conditions of the ICP mass spectrometer are listed in Table 1. Under these experimental conditions a slight but reproducible difference between the found and expected isotope ratios were obtained which was similar to those commonly seen in isotope ratio measurements with ICP-MS and was probably caused by mass discrimination in ion extraction focusing mass analysis and dete~tion.~' Palladium as Chemical Modifier Chemical modifiers are commonly used in ETV-ICP-MS in order to reduce losses of analyte caused by condensation on different parts of the ETV cell or the transfer line that connects the ETV to the ICP-MS in~trument.'-~ Where enhancements occur in the presence of a modifier the primary role of the modifier is as a physical carrier of vaporized analyte.This effect increases the transport efficiency between the graphite furnace and the torch of the ICP-MS instrument. Palladium has been used as a chemical modifier to improve the signals of some volatile elements in many ETV-ICP-MS appli- c a t i o n ~ . ~ - ~ In this work the effect of the Pd concentration on the signal for Cd was investigated. The results obtained are shown in Fig.2. As can be seen the signal for Cd decreased as the concentration of Pd increased when slurry sampling was used. On the other hand the signal for Cd increased with an increase in the Pd concentration when an aqueous solution was introduced. Since the silicate matrix of the slurry itself acts as a physical carrier of volatile analytes the addition of extra Pd modifier may increase the total mass transferred into the ion optics to an amount that is sufficient to result in space charge and/or other ion optic perturbations which could result in a decrease in the signal. Hence except for the reagents used for slurry preparation no modifier was used in the ETV- ICP-MS analyses. 120 r---l 0 1 ' 1 " ' 1 1 0 100 200 300 400 500 600 Fig.2 Effect of Pd concentration on Cd ion signal in A slurry solution and B matrix-matched aqueous solution.The Cd concen- trations in these two solutions were similar. The slurry solution contained PACS-1 sediment diluted 50-fold with 0.1% Triton X-100 and 2% HNO,. Solution B contained the same concentrations of HNO and Triton X-100 and the equivalent amount of Cd Selection of Charring and Vaporization Temperatures Experiments were carried out to determine the optimum temperature and time for the drying charring and vaporization steps. These conditions were optimized by means of several measurements for a slurry sediment sample. The determination of the optimum charring temperature was carried out by studying the effect of different charring temperatures between 120 and 900°C on the ion signals for Cu Cd and Pb.The results obtained are shown in Fig. 3. Since no chemical modifier was used in this ETV-ICP-MS analysis Cd was evaporated and the ion signal decreased rapidly when the charring tem- perature was higher than 200 "C. Hence the charring tempera- ture was set at 200°C. In order to evaporate the elements studied completely and simultaneously the vaporization tem- perature was set at 2500°C. The ETV operating conditions used are listed in Table 1. Effect of Surfactant Concentration in the Prepared Slurry Sample ET-AAS has been successfully applied to the analysis of ~ l u r r i e s . ' ~ - ~ ~ Bendicho and de Loos-Vollebregt16 and Miller- IhliI7 have reviewed the available literature and quantified the importance of certain factors such as particle size analyte partitioning maximum slurry concentration and homogeneous slurries.Homogenization of the slurry can be achieved by ultrasonic agitation of the sample powder in solution and by stabilization of the particles using a thixotropic (thickening) 120 I o Cu 0 Cd v Pb I 100 - - d gl 8 0 - .H a $! 60 - .r( 4 e u 40 E - - 20 ' 0 ' 0 200 400 600 800 1000 Ash temperature/OC Fig. 3 Effect of charring temperature on ion signal. Vaporization temperature was set at 2500 "C. The composition of the slurry solution was the same as in Fig. 2. Journal of Analytical Atomic Spectrometry August 1996 Vol. 11 557agent which increases the viscosity of the suspension pre- venting particles from settling. The effect of several parameters of the slurry preparation on the ion signals was therefore investigated. Fig.4 shows the dependence of the ion signals on the concentration of Triton X-100 in the slurry sample.As can be seen the optimum concentration was found to be 0.1% v/v Triton X-100. In our experiments it was found that Triton X-100 not only helped to disperse the particles but also acted as a good physical carrier to enhance the signals. Under the optimum experimental conditions analyte signals were enhanced by as much as a factor of 3 in the presence of 0.1% Triton X-100. Effect of Acid Concentration The concentration of acid in the slurry solution could affect the rate of extraction of the metal ions and the precision of the ion signals.I8 If a large percentage of the analyte is extracted into the liquid phase the precision will approach that obtain- able with a conventional liquid digest.In addition the analysis will be more representative of the analyte concentration in the original solid sample. Also as described by Gregoire et d.,' the presence of mineral acid in ETV-ICP-MS analysis could affect the ion signals. These workers found that analyte signals were enhanced by as much as a factor of 2 in the presence of 1% v/v HNO,. The effect of the amount of HN03 in the slurry sample on the ion signals was therefore studied. The results obtained are shown in Fig. 5. As can be seen the ion signal reached a maximum when the HNO concentration was l * O t 0.0 0.2 0.4 0.6 0.8 1.0 1.2 (Triton X-1 001 (% v/v) Fig. 4. Effect of Triton X-100 concentration in the slurry sample on ion signal.The composition of the slurry sample was similar to that given in Fig.2 except that the concentration of Triton X-100 was varied. For the ETV-ICP-MS operating conditions see Table 1. All data are relative to the first point I 0 Cu 0 Cd v Pb Fig.5 Effect of HNO concentration on ion signal. All data are relative to the first point 2% v/v. Moreover the ion signal gradually decreased when the acid concentration was higher than 2%; this could be due to the loss of analyte during the charring step. Hence 2% HNOJ was used in all slurry preparations. Effect of Dilution Factor An important factor in the slurry technique is the slurry concentration. However dilution of the slurry can only be carried out within a limited range; precision is degraded when working with highly diluted slurries because only a small number of particles remain in the slurry.On the other hand if the slurries are more concentrated matrix effects and the accumulation of injected samples will occur and pipetting efficiency will deteriorate. The effect of the dilution factor on the ion signal and precision of the ion signal was therefore investigated. As shown in Fig. 6 although the ion signals decreased with an increase in the dilution factor the dilution factor (m/v) did not affect the precision of the ion signals significantly if it was greater than 100. However the precision became worse when the dilution factor was greater than 2000; this could be due to the smaller number of particles in the slurry sample. However accumulation of the injected sample was observed when a small dilution factor was used; hence in order to ensure sample homogeneity good analyte signals and complete vaporization of the introduced sample a dilution factor of 500 was used.Effect of Particle Size The optimum grinding time using tungsten carbide beads was determined by measuring the analyte signals in several slurries of marine sediment samples that had been ground for 0-90 min. The effect of grinding time of the Taiwan Straits' sediment on the ion signal and precision of the ion signal is shown in Fig. 7. As can be seen the grinding time did not affect the ion signal and precision of the ion signal significantly. This could be because the particle size of the original sediment samples was sufficiently small.The large uncertainty in the Cd signal could be due to the extremely low concentration of Cd in this 250000 4 d 0 g 200000 3 100000 ' 150000 4 (d .r( U El 50000 0 l-l I 15 10 5 Cu b "'Cd v 'Fpb I ' I I 100 1000 10000 Dilution factor Fig.6 Effect of dilution factor of the slurry on A ion signal and B the uncertainty of the signal. All slurry samples contained 0.1% Triton X-100 and 2% HNO and various amounts of PACS-1 sediment 558 Journal of Analytical Atomic Spectrometry August 1996 Vol. 1 Ilo' 9 10' ~ 0 20 40 60 80 100 Grinding time/min Fig. 7 Effect of grinding time of the sediment on the ion signal and the uncertainty of the ion signal. All slurry samples contained 0.1% Triton X-100 2% HNO and 0.2% m/v Taiwan Straits sediment sediment sample. In other experiments the particle size distri- bution of these sediment samples was determined by laser diffraction. The results obtained are shown in Fig.8. As can be seen the particle size of the original sediment samples was less than 100 pm. After 60 min grinding time the mean particle size of the sediments did not change significantly. Hence the sediment samples were analysed directly without a pre- grinding step. Non-spectroscopic and Spectroscopic Interferences For high-salt content samples 63Cu is interfered with by ArNa' and 65Cu is interfered with by ArMg'. Hence an experiment was performed to check the interferences caused by these two molecular ions. A stock slurry sample of PACS-1 was prepared by the method described under Preparation of Slurries. The sample was then spiked with 30 pg ml-' of Na and 15 pg ml-' of Mg.These concentrations are about half the concentrations of Na and Mg in the prepared slurry sample. As shown in Table 2 although the ion signals decreased slightly when additional Na and Mg were added the isotope ratio of Cu was not affected by Na and Mg at these concen- trations. Also listed in Table 2 are the interferences of 47TiO+ and 49TiO+ ions on 63Cu and 65Cu. The spiked Ti concen- trations were equivalent to about half the concentration of Ti in the prepared slurry sample. As can be seen the isotope ratio of Cu was not affected by the Ti at these concentrations. This experiment demonstrated that the concentration of Cu in the 1 10 100 Particle diamster/pm Fig. 8 Typical particle size distribution curves for the Taiwan Straits sediment sample with grinding times of A 0; B 30; and C 60min.Differential volume (YO) was calculated from 0.40 to 900 pm. Mean particle sizes were 19 9 and 8 pm for 0 30 and 60 min grinding times respectively Table 2 Effect of various matrices on Cu isotope ratio determination* Solution composition PACS-1 solution PACS-1 solution PACS-1 solution PACS- 1 solution + 30 pg ml-' Na + 15 pg ml-' Mg + 30 pg ml-' Na + 15 pg ml-' Mg PACS- 1 solution + 4 pg ml-' Ti PACS-1 solution + 8 pg ml-' Ti Peak area/counts s-' 63cu 65cu 65cu 63cu 33 400 f 1700 30 900 f 970 16 600 f 460 15 500 & 460 0.497 f 0.015 0.502 f 0.01 1 28 900 f 1900 14 300 f 760 0.497 f 0.030 29 300 k 1000 14 200 f 200 0.496 f 0.020 3 1 000 k 960 27 400 f 2020 15 700 f 590 13 900 k 900 0.507 f 0.009 0.506 k 0.007 * PACS-1 solution is a PACS-1-diluent = 1 +499 solution; the dilu- ent is a mixture of 2% HNO and 0.1% Triton X-100 solution.Values are mean of seven measurements standard deviation. sediment samples can be determined directly by ETV-ID- ICP-MS without significant interferences. Determination of Cu Cd and Pb in Sediment Samples by In order to validate the ultrasonic slurry sampling-ETV-ID- ICP-MS method the concentrations of Cu Cd and Pb in the PACS-1 harbour sediment reference sample were determined. The results obtained are shown in Table 3. The determined concentrations are in good agreement with the certified values. This experiment indicated that Cu Cd and Pb could be readily quantified by the proposed method.The results of the external calibration and standard additions methods are also listed in Table 3. Although the results obtained by the use of standard additions agreed with the certified values and with the ID method the concentrations determined by external calibration were slightly lower than the certified values; this could be due to non-spectroscopic interferences and/or the poorer analyte transport efficiency when a slurry sample is analysed. These results are in contrast to those of previous ~ t u d i e s . ~ . ~ The concentrations of Cu Cd and Pb in the sediment collected from the Taiwan Straits were determined with the proposed method. As shown in Table 4 the concentrations of these elements are much lower than those in the PACS-1 ETV-ID-ICP-MS Table3 ment sample by ultrasonic slurry sampling-ETV-ID-ICP-MS Determination of Cu Cd and Pb in PACS-1 harbour sedi- Concentration*/pg g- Analysis method c u Cd Pb External calibration 444 f 63 1.94 f 0.16 339 & 29 Standard additions 464 f 38 2.22 f 0.20 435 f 38 ID 426 f 26 2.31 f0.28 419f 18 Certified valuet 452 f 16 2.38 f 0.20 404 & 20 * Values are mean of three measurements f standard deviation.NRCC certified value. Values are given in 95% confidence limits. Table 4 Determination of Cu Cd and Pb in Taiwan Straits sediment sample by ultrasonic slurry sampling-ETV-ID-ICP-MS ~ ~ Concentration*/pg g- ' Analysis method c u Cd Pb External calibration 13.1 f 1.2 0.40f0.19 Standard additions ID 22.8 k 2.1 19.0k 1.3 0.27f0.12 23.9 f 0.4 16.7f0.4 0.29f0.14 21.9 f 1.2 * Values are mean of three measurements f standard deviation.Journal of Analytical Atomic Spectrometry August 1996 Vol. 1 1 559sample. No OmniRange setting was used for Cu and Pb determination in this sample. Detection limits based on the usual definition as the concentration of the analyte yielding a signal equivalent to three times the standard deviation of the blank signal were 0.055 0.024 and 0.038 pg g-’ for Cu Cd and Pb respectively. Better detection limits are to be expected with reagents of higher purity. CONCLUSION The use of ultrasonic slurry sampling-ETV-ID-ICP-MS pro- vides a simple rapid and accurate technique to determine routinely Cu Cd and Pb in sediment samples. The concen- trations determined in the slurries by the ID method were compared with the results of the external calibration and standard additions methods. The results obtained by the use of ID agreed with the certified values and with those obtained by standard additions.However the concentrations deter- mined by external calibration were slightly lower than the certified values; this could be due to non-spectroscopic inter- ferences and/or the poorer analyte transport efficiency when a slurry sample is analysed. The proposed method should be useful for the direct analysis of other solid samples. Other applications of the ultrasonic slurry sampling-ETV-ICP-MS system are under investigation in this laboratory. This research was supported by a grant from the National Science Council of the Republic of China under Contract NSC 8 5-262 1 -M- 1 10-009. REFERENCES 1 2 3 4 5 6 7 8 9 10 Gregoire D.C. Goltz D. M. Lamoureux M. M. and Chakrabarti C. L. J. Anal. At. Spectrom. 1994 9 919. Gregoire D. C. Miller-Ihli N. J. and Sturgeon R. E. J. Anal. At. Spectrom. 1994 9 605. Fonseca R. W. and Miller-Ihli N. J. Appl. Spectrosc. 1995 10 1403. Gregoire D. C. and Sturgeon R. E. Spectrochim. Acta Part B 1993 48 907. Ediger R. D. and Beres S. A. Spectrochim. Acta Part B 1992 47 907. Wei W. C. Chen C. J. and Yang M. H. J. Anal. At. Spectrom. 1995 10 955. Gregoire D. C. and Naka H. J. Anal. At. Spectrom. 1995,10,823. Sparks C. M. and Holcombe J. Spectrochim. Acta Part B 1993 48 1607. Hasting D. W. Emerson S. R. and Nelson B. K. Anal. Chem. 1996 68 371. Lamoureux M. M. Gregoire D. C. Chakrabarti C. L. and Golt D. M.Anal. Chem. 1994,66 3208. 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 39 40 41 Byrne J. P. Hughes D. M. Chakrabarti C. L. and Gregoire D. C. J. Anal. At. Spectrom. 1994 9 913. Becker S. and Hirner A. V. Fresenius’ J Anal. Chem. 1994 350 260. Zaray Gy. and Kantor T. Spectrochim. Acta Part B 1995 50 489. Soares M. E. Bastos M. L. and Ferreira M. A. J. Anal. At. Spectrom. 1994 9 1269. Vinas P. Campillo N. Garcia I. L. and Cordoba M. H. Analyst 1994 119 1119. Bendicho C. and de Loos-Vollebregt M. T. C. J. Anal. At. Spectrom. 1991 6 353. Miller-Ihli N. J. Anal. Chem. 1992 64 964. Miller-Ihli N. J. J. Anal. At. Spectrom. 1994 9 1129. Bermejo-Barrera P. Barciela-Alonso C. Aboal-Somoza M. and Bermejo-Barrera A. J. Anal. At. Spectrom.1994 9 469. Miller-Ihli N. J. Fresenius’ J. Anal. Chem. 1990 337 271. Anzano J. M. Martinez-Garbayo M. P. Belarra M. A. and Castillo J. R. J. Anal. At. Spectrom. 1994 9 125. Miller-Ihli N. J. Spectrochim. Actu Part B 1995 50 477. Hoenig M. Regnier P. and Wollast W. J. Anal. At. Spectrom. 1989 4 631. Robles L. C. and Aller A. J. Talanta 1995 42 1731. Vinas P. Campillo N. Garcia I. L. and Cordoba M. H. Spectrochim. Actu Part B 1995 50 527. Dean J. R. Ebdon L. and Massey R. J. Anal. At. Spectrom. 1987 2 369. Gregoire D. C. and Lee J. J. Anal. At. Spectrom. 1994 9 393. van Heuzen A. A. Hoekstra T. and van Wingerden B. J. Anal. At. Spectrom. 1989 4 483. Beary D. C. Brletic K. A. Paulsen P. J. and Moody J. R. Analyst 1987 112 441. McLaren J. W. Beauchemin D. and Berman S. S. Anal. Chem. 1987 59 610. Bowins R. J. and McNutt R. H. J. Anal. At. Spectrom. 1994 9 1233. McLaren J. W. Mykytiuk A. P. Willie S. N. and Berman S. S. Anal. Chem. 1985 57 2907. Beauchemin D. McLaren J. W. Mykytiuk A. P. and Berman S. S. J. Anal. At. Spectrom. 1988 3 305. McLaren J. W. Beauchemin D. and Berman S. S. J. Anal. At. Spectrom. 1987 2 277. Lu P.-L. Huang K.-S. and Jiang S.-J. Anal. Chim. Acta 1993 284 181. Hwang T.-J. and Jiang S.-J. J. Anal. At. Spectrom. 1996 11 353. Longerich H. P. Fryer B. J. and Strong D. F. Spectrochim. Acta Part B 1987 42B 39. Garbarino J. R. and Taylor H. E. Anal. Chem. 1987 59 1568. Longerich H. P. At. Spectrosc. 1989 10 112. Gregoire D. C. Anal. Chem. 1987 59 2479. Jiang S.-J. Palmieri M. D. Fritz J. S. and Houk R. S. Anal. Chim. Acta 1987 200 559. Paper 6/0250 7B Received April 10 1996 Accepted May 21 1996 560 Journal of Analytical Atomic Spectrometry August 1996 Vol. 11