ANALYST, NOVEMBER 1986, VOL. 11 1 1249 Simultaneous Determination of Major and Trace Elements in Urinary Calculi by Microwave-assisted Digestion and Inductively Coupled Plasma Atomic Emission Spectrometric Analysis Michael Alexander Erich Wandt* and Michel Andre Bruno Pougnett Department of Physical Chemistry, University of Cape Town, Private Bag, Rondebosch, 7700, Republic of South Africa A procedure that permits the simultaneous determination of major, minor and trace elements in urinary calculi by inductively coupled plasma atomic emission spectrometry is described. The dissolution of samples is achieved with a commercial microwave oven using a mixture of nitric and perchloric acids. Ca, Mg, P, Al, Cu, Fe, K, Li, Mn, Mo, Na, Pb, S, Sr and Zn were determined in more than 100 South African stones.Keywords: Urinary calculi; major and trace element determination; inductively coupled plasma atomic emission spectrometry; microwave-assisted digestion Since its introduction more than two decades ago, inductively coupled plasma atomic emission spectrometry (ICP-AES) has found widespread applications in the analysis of biological samples.l-3 The advantages of the ICP-AES analysis of human pathological concretions reported previously4>5 and the posi- tive results obtained in the simultaneous determination of the major elements of human calculi (Ca, Mg, P),4,5 have prompted additional investigations involving trace element determinations. In recent years, the functions of trace elements in the human body and the environment and their importance for medical practice have been increasingly recognised in the biomedical field.6.7 Although most mechanisms concerning the behaviour of trace elements in biological systems are far from being understood , many researchers are convinced that trace elements play a major role in causing diseases such as urolithiasis.8-11 In addition to the metabolic role, attention has been focused on the trace element content of human concretions.The aim of this study was to develop a method whereby both major and trace elements of urinary calculi could be quantified simultaneously. Preliminary studies showed that trace element concentra- tions in the sample solutions prepared4 were too low to allow the accurate determination of concentrations and that more concentrated samples were required.When the stone mass was increased , the time-consuming dissolution step was further lengthened, thereby limiting the speed of the analysis. A microwave-assisted digestion procedure was therefore developed. This shortened sample preparation times con- siderably owing to more rapid and efficient digestion. Subse- quently more than 100 urinary calculi obtained from Cape Town hospitals were analysed for three major (Ca, Mg and P) and 12 trace elements (Al, Cu, Fe, K, Li, Mn, Mo, Na, Pb, S, Sr and Zn). In this paper the sample preparation procedure and ICP experimental conditions developed for this purpose are described. * Present address: Council for Scientific and Industrial Research, National Accelerator Centre, Van de Graaff Group, Ion-solid Interaction Division, PO Box 72, Faure, 7131, Republic of South Africa.t Present address: Council for Scientific and Industrial Research, National Accelerator Centre, Van de Graaff Group, Nuclear Analytical Chemistry Division, PO Box 72, Faure, 7131, Republic of South Africa. Experiment a1 Instrumentation and Apparatus All experimental work was performed on a two-channel Plasma-200 ICP spectrometer (Allied Analytical Systems). This instrument is an updated model of the IL Plasma-100 used in the previous study4 and has been described by Smith et a l l 2 The model employed in this study is fitted with two scanning monochromators, the standard air path and a vacuum monochromator, which allows the observation of emission lines in vacuum below 300 nm. A commercially available microwave oven (Sharp Model R6950 E) with a stainless-steel cavity, operating at 2.45 GHz with a maximum output power of 650 W, was used.Digestions were carried out in 50-ml Erlenmeyer flasks covered with small beakers to prevent spattering and possible sample losses. In this way cross-contamination was also minimised. An acid fume scrubber was constructed from a flask containing an approximately 2 M KOH solution and a beaker filled with water.13 These were connected to a desiccator, which could accommodate up to five digestion vessels. The whole assembly was stationary on a thick glass plate, which could be easily placed in the oven cavity. Samples and Standards Over 100 urinary calculi were selected from a collection of about 550 human stones from two Cape Town hospitals.The criteria for including a stone in the selection were based on the available mass (more than 250 mg was required) and on its composition as determined by X-ray powder diffraction (XRD). The aim was to obtain significant numbers of calculi in each of the three major stone groupings, viz., calcium oxalate, phosphate and uric acid stones. Chips from the larger specimens or whole stones were ground to fine powders. Approximately 250-mg aliquots of these powders were transferred into the Erlenmeyer flasks and 2 ml of concentrated nitric acid (65%) and 1 ml of concentrated perchloric (90%) acid (both Riedel-de-Haen, analytical-reagent grade) were added. The flasks were then covered with small beakers and placed in the desiccator, which was sealed with high-vacuum silicone grease. After coupling with the acid fume scrubber , the entire assembly was placed in the microwave oven.Three minutes of uninterrupted irradia- tion were usually sufficient to obtain clear solutions. Although some samples required only ca. 1 min for the reaction to be completed, all samples were exposed to the microwaves for1250 ANALYST, NOVEMBER 1986, VOL. 111 the same length of time. In a few flasks, clear solutions were not obtained and they were returned to the oven (after adding a few drops of nitric acid) for a further few minutes treatment. After cooling, the solutions were transferred quantitatively into 50-ml calibrated flasks containing 1 ml of concentrated nitric acid and 1 ml of 1% mlVTriton X-100 surfactant (BDH Chemicals).These were diluted to volume with de-ionised, distilled water, which was used throughout for all prepara- tions. All solutions were stored in polyethylene bottles, which had previously been leached with nitric acid. No precipitation was observed after storage for more than 3 months. Acid fumes produced by the digested samples could escape through small lips in the rims of the Erlenmeyer flasks. The resulting over-pressure in the desiccator then forced the evolving acid fumes through the KOH solution where they were neutralised. No acid fumes could be detected (by smell) in the oven cavity after removing the sample container, and no signs of corrosion were found after completing this study. The alkaline solution and the water were routinely changed after three successive digestions in the oven.This was found to be necessary to prevent boiling of the liquids by the fast microwave heating. Stock standard solutions (10 mg ml-1) of calcium and magnesium were prepared by dissolving calcium carbonate and magnesium metal flakes (both Johnson Matthey, Spec- pure), respectively, with a few drops of nitric acid. (NH4)H2P04 (Merck, analytical-reagent grade) was used as a phosphorus standard of the same concentration. The primary trace element standards (1 mg ml-l) used were Spectrosol solutions (BDH Chemicals), except for sulphur and lithium, for which standards were prepared from (NH4)2S04 (BDH Chemicals, analytical-reagent grade) and LiCl (Merck, Titri- sol), respectively. Standards of lower concentrations were obtained by serial dilution of these stock solutions with a blank, prepared by subjecting acids alone to the whole preparative procedure.Mixed element standards for ICP calibration purposes were prepared in the same way. ICP Experimental Conditions Optimisation of the instrumental parameters followed the procedures already d e ~ c r i b e d . ~ Firstly, the most sensitive prominent spectral lines were investigated for each (trace) element of interest for potential interferences, e.g., spectral overlap and background shifts. The effect of the major elements Ca, Mg and P and some minor elements, such as Al, K and Na, on the analyte emission were evaluated by observing the recorded intensity profiles on a video display. These were obtained by pre-nebulisation mixing14 of various concentrations of analyte and concomitants, e.g., 1 mg ml-1 of Ca and 10 pg ml-1 of Fe.The advantage of this method is that it does not require the very tedious procedure of preparing two-element mixtures of the analyte together with all potential interferents. It thus allows fast and easy evaluation of spectral lines and background correction positions. All the analytical wavelengths used are summarised in Table 1, together with other important line parameters. The intensity measurements were divided between the two monochromator systems (channels A and B) for time-saving reasons. The viewing height of channel B (vacuum monochro- mator) cannot be individually adjusted for different elements for any one analytical programme. This parameter is usually optimised and set manually for the least sensitive element in the programme. On the other hand, observation heights used in connection with channel A could be optimised for each element separately and the heights utilised are also listed in Table 1.Where necessary, automatic background correction facilities were used (Table 1). Detection limits in solution (30) achieved for the analytical lines using the given settings and instrumental conditions (Table 2) are also included in Table 1. An estimate of the detection limits for each element in the calculi (expressed as ng mg-1 of stone mass) was calculated by taking into account the dilution factor (50 ml) and an average mass of 265 mg (mean of all weighings). Multi-element standard solutions were used to optimise the instrumental conditions.The effects of varying r.f. power, nebuliser driving pressure (carrier gas flow-rate) and sample feed rate were studied. All these analytical variables are, however, interdependent and changing one might affect the optimum conditions for the others. Further, some system variables, such as the nebuliser driving pressure, cannot be optimised for each element separately. Therefore, compro- mise settings, which would give the “best” performance for all analytes, had to be chosen. These conditions are listed in Table 2. After setting the spectrometer to the optimised parameters, it was first calibrated for each element using the blank and the standards listed in Table 3. This calibration was carried out mainly to confirm the linearity of the calibration graph over the entire expected concentration range.Re-calibration of the instrument during analysis was performed after every 10-15 samples with the highest concentration standard and the blank only, except for Ca, Mg and P. Blank readings were taken after every fifth sample in order to monitor the drift of the instrumental parameters. Results and Discussion In order to check for the possible loss of elements during the microwave digestion procedure, a series of recovery tests using aqueous 10 pg ml-1 standards were undertaken. As part Table 1. Wavelength data Detection limits Ca . . Mg . . P . . A1 . . c u . * Fe . . K . . Li . . Mn . . Mo . . Na . . Pb . . s . . Sr . , Zn . . Element Wavelengthhm . . . . . . . . 315.89 . . . . . . . . 279.81 .. . . . . . . 213.62 . . . . . . . . 237.32 . . . . . . . . 324.75 . . . . . . . . 259.94 . . . . . . . . 769.90 . . . . . . . . 670.78 . . . . . . . . 257.61 . . . . . . . . 203.84 . . . . . . . . 589.59 . . . . . . . . 220.35 . . . . . . . . 182.04 . . . . . . . . 346.45 . . . . . . . . 213.86 Background correction Monochromator - Air path Rhs Air path - Air path - Vacuum Rhs Air path - Vacuum - Air path - Air path - Air path Rhs Vacuum - Air path Lhs Vacuum Lhs Vacuum - Air path - Vacuum Viewing height/mm 25 14 14 14 5 14 14 2 - - - - - 14 - Solution/ 30 40 90 54 5 4 940 7 5 8 30 70 110 30 5 vgl-’ Calculus/ ng mg- * 6 8 17 10 1 1 180 1.5 1 1.5 6 14 21 6 1ANALYST, NOVEMBER 1986, VOL. 111 125 1 of a general investigation of microwave-assisted dissolution, many elements in addition to those determined in calculi were included in this part of the study (Al, Cd, Co, Cr, Cu, Fe, K, Mn, Mo, Na, Ni, Pb, S, Se, Sr, Ti, V and Zn).Multi-element solutions were subjected to the same preparative steps as the samples, in duplicate. No loss of any of the elements was observed. These favourable results do not, however, exclude the possibility of elemental loss if the particular element is incorporated in a more complex matrix or in samples where volatile species can be formed. A major shortcoming in the verification of the method was the lack of a suitable reference material. As no certified (urinary) stone standards are available, the accuracy of the entire analytical procedure was evaluated using US National Bureau of Standards Standard Reference Materials (NBS SRMs).l3 In general, good agreement with certified or reported values was obtained.Three comparatively large calculi were chosen to test the precision of the method described. Five separate aliquots were prepared from each of these and mean concentrations found are listed in Table 4 together with the standard deviations (SD) and relative standard deviations (RSD). For the three major elements (Ca, Mg and P) the reproducibility was found to be comparable to that in the previous study4 and ranged from 1.5 to 2.9% RSD. The precision for the trace elements varied from 1.0 to 74.4% RSD. However, when concentration values below five times the detection limit are excluded, a mean RSD of 5.1% is obtained. Table 2. ICP operating conditions Power .. . . . . . . . . . . . . Plasma coolant gas flow . . . . . . . . Sample feed rate . . . . . . . . . . Nebuliserdrivingpressure . . . . . . Aerosolcarrierflow-rate . . . . , . Pump delay . . . . . . . . . . . . Peak window . . . . . . . . . . Integration time . . . . . . . . . . Observation height . . . . . . . . Number of readings . . . . , . . . 1.2kW 13 1 min-1 1 ml min-’ 206.8 kPa (30 lb in - 0.4 1 min-l 30 s 0.067 nm 3s 2-25 mm, varied 3 Table 3. Calibration standards Element Standarddmg 1-1 Ca . . . . . . . . , . . . . . 2000,1000,500,100 Mg, P . . . . . . . . . . . . . . 1000,500,200,100 Na, K . . . . . . . . . . . . . . 100,50,10 S . . . . . . . . . . , . . . . . 20,10,5,2,1,0.5 Al,Cu,Fe,Li,Mn,Mo,Pb,Sr,Zn . . 10,5,2,1,0.5 Aliquots of 16 large stones, which had been analysed for Ca, Mg and P in the previous study,4 were prepared and re-analysed.Mean deviations of 6.9, 5.6 and 4.2% for Ca, Mg and P, respectively, were obtained when comparing the two analyses. These figures show excellent agreement between the conventional “h~t-plate”~ and microwave-assis- ted digestion procedures. Potassium could not be determined at the same time as the other elements because of the low sensitivity of the 769.90 nm line. In order to measure low concentrations of this element, the standard photomultiplier tube in channel A (Hamamatsu R 106 UH) was replaced by a more red-sensitive photomulti- plier (Hamamatsu T 955). This model is sensitive up to 930 nm compared with 650 nm for the standard tube, thus giving better results at longer wavelengths.Lithium and Mn could not be determined in any of the calculi because of their very low concentrations. It can therefore be concluded that if these elements occur in stones they do so at concentrations below ca. 1 ng mg-1. Copper, Mo and Pb concentrations were found to be below the detection limits (1, 1.5 and 14 ng mg-1) in 68, 51 and 64% of all cases, respectively. The determination of Si was at first attempted. However, as a result of the poor precision achieved during trial analyses the attempt was abandoned. One reason for the difficulties generally experienced when determining this element lies in the necessity of stabilising the silicon ion in solution. In acidic media this is usually achieved by including HF (fluoride ion) in the digest, which is precluded in this instance as a result of incompatibility with the torchs’ quartz tubes.Further, erosion of the silica ICP torch and subsequent introduction of spurious Si into the plasma are additional unwanted possibilities.15 In digests of cystine calculi, sulphur concentrations exceeded the dynamic range of the calibration. These samples were therefore diluted five-fold and then analysed again. On the assumption that these stones were 100% cystine, good agreement with the expected stoicheiometric sulphur value (26.7% mlm) was obtained. Of the 102 calculi analysed in this study, 14 belong to the calcium oxalate (CaOx) group, 18 to the CaOx - apatite (APA) group, 45 to the struvite (STR) - APA group and 19 to the uric acid (UA - UAD) - CaOx group.Two calculi containing urates, one STR - calcium oxalate monohydrate (COM) and three cystine stones were also found.16 In the CaOx group, COM was the major component (ie., >50%) in 12 cases and calcium oxalate dihydrate (COD) in 2, whereas in the CaOx - APA group COM was predominant in 3 calculi, COD in 11 and APA in 3. In the STR - APA group, struvite occurred in 25 calculi at concentrations greater than 50% and apatite in 14. In the UA - CaOx group, UAD (uric acid Table 4. Replicate analysis of three stone samples (mean of five preparations) Concentration? % mlm ng mg-1 Stone* Ca Mg P A1 Cu Fe K Mo Na Pb S Sr Zn S1, APA - STR Mean 4.21 7.80 12.11 36.8 N.d. N.d. 3280 4.2 3431 N.d. 134 68.0 206 RSD, % 2.9 1.8 1.9 7.5 - - 5.3 24.5 2.4 - 3.6 13.2 2.8 SD 0.12 0.14 0.23 2.7 - - 172 1.0 83 - 4.8 9.0 5.9 S2, APA - COM Mean 29.19 0.25 9.46 71.9 3.9 23.0 1003 2.2 5913 114 1851 141 1381 SD 0.45 0.004 0.17 3.6 0.7 1.6 104 0.5 120 12.7 39.0 4.9 13.5 RSD, Yo 1.5 1.8 1.8 5.0 19.1 7.0 10.4 22.4 2.0 11.2 2.1 3.5 1.0 S3, UA - UAD Mean 0.40 N.d.0.04 N.d. 2.4 3.7 649 N.d. 459 N.d. 463 N.d. 4.5 SD 0.01 - 0.001 - 1.8 1.9 80 - 21.7 - 7.7 - 0.9 RSD, % 2.9 - 2.5 - 74.4 51.4 12.3 - 4.7 - 1.7 - 20.1 * APA, apatite; COM, calcium oxalate monohydrate (whewellite); STR, magnesium ammonium phosphate hexahydrate (struvite); UA, uric t N.d., not detectable. acid; UAD, uric acid dihydrate.1252 ANALYST, NOVEMBER 1986, VOL. 111 Table 5. Elemental concentrations in four major stone groups. Ca, Mg and P expressed as 70 rnirn, other elements as ng mg-1.Value(s) below the detection limit are set to half of the detection limit concentration and are identified by (d). Ca Mg P A1 c u Fe K Mo Na Pb S Sr Zn Element . . * . . . . . * . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Range Mean Median . . Range Mean Median . . Range Mean Median . . Range Mean Median . . Range Mean Median . . Range Mean Median . . Range Mean Median . . Range Mean Median . . Range Mean Median . . Range Mean Median . . Range Mean Median . . Range Mean Median . . Range Mean Median All instances (n = 102) 15.82 17.43 2.38 0.25 6.67 6.67 32 31 <0.50-10 1.4 0.50(d) <0.50-156 27 20 <90-6756 1737 1325 <0.75-8.3 2.2 0.75(d) 129-50042 <0.0 1-31.43 <0.01-9.50 <0.01-16.79 4.0-89 4670 2695 <7.0-139 22 7.0(d) 134-271 000 8564 542 <3.0-622 121 90 <0.50-1381 270 166 CaOx (n =14) 23.18 24.94 0.09 0.03 0.07-1.44 0.99 0.33 <5.0-89 25 17 0.59 0.50( d) 6.0-51 30 25 1031 53 1 1.3 0.75( d) 92% 10636 2763 1651 <7.0-131 27 7.0(d) 10.22-28.1 1 <0.0 1-0.87 <0.5&1.7 367-4059 <O. 75-4.1 372-1633 1022 1127 12-622 113 60 21-1316 210 65 CaOx - APA (n = 18) 23.34-29.56 26.63 26.53 0.30 0.18 1.22-1 1.11 4.51 3.48 30 29 <0.50-10 2.1 0.50( d) 9.0-129 48 33 350-3270 1063 810 <0.75-3.1 1.2 0.75( d) 1428-15212 3884 2616 <7 A-119 47 43 345-2081 906 792 44-462 126 94 483 418 0.03-1.85 12-72 113-1381 APA - STR ( n = 45) 1.30-31.43 15.25 15.35 0.56-9.50 5.02 5.41 5.80-16.79 12.65 13.04 27-73 47 46 <0.50-71 1.1 0.50(d) <0.50-156 24 18 2914 2702 <0.75-8.3 3.3 3.4 1002-12392 6671 6437 <7.0-139 19 7.0(d) 134-1428 478 424 12-459 180 148 26-994 345 303 1042-6756 UA - CaOx (n = 19) <0.01-23.68 5.81 1.22 <0.01-0.15 0.03 0.01 <0.01-1.58 0.18 0.04 4.0-21 8.7 5.0(d) <O.50-5.9 1.6 0.50( d) <0.50-79 18 11 <90- 1265 393 289 <0.75-3.0 1.3 0.75( d) 129-1738 662 459 <7.0-14.1 7.4 7.0(d) 341-1097 605 572 C3.0-126 18 3.0(d) 2.1-131 16 8.2 dihydrate) was detected in 4 stones, whereas COM was the prevailing phase in 3 of them, compared with UA in 14. Range, mean and median elemental concentrations measured are summarised in Table 5 for the four major stone groups. Comparison of these results with those obtained by other researcherslOJ7-19 was found to be difficult, because of the different approaches in publishing concentration values.Often these are reported as “as received,” “dry mass” or “YO ash.” In instances where no mean ash portions or component information are given to supplement the data, calculation of meaningful concentration values becomes an impossible task. One possible reason for the inconsistent concentration values observed could be the degree to which different stone groups (e.g., calcium oxalates, APA - STR, UA, etc.) are present in the total number of calculi sampled in each instance. This parameter influences reported concentration values to a great extent. For example, from Table 5 , it can be clearly seen that trace elements are far less concentrated in uric acid stones than in apatite stones. Strontium and zinc levels especially are exceptionally low in the former group.An additional reason for the disagreement in results might be the way in which mean concentrations are evaluated. Levinson et al., 10 for example, set all concentration values that were too low for determination at zero. In this study, these values were arbitrarily set at half the detection limit. In other studies, these figures are not considered at all when calculating the mean. To further illustrate this, Table 5 shows the mean (median) Pb concentration in CaOx calculi as 27 (7.0) ng mg-l; when discarding all values below the lower limit of detection, the mean (median) lead concentration becomes 62 (44) ng mg-l! In most studies, however, detection limits or figures for accuracy and precision are not stated.Differences observed between similar investigations might, nevertheless, be real, i.e., stem from regional differences of the collection sites, giving rise to factors such as dissimilar drinking waters or diets. Investigating a causative connection between trace element content and stone formation was, however, beyond the scope of this study. Further, the limited number of such studies does not yet permit any conclusions to be drawn in this direction from the available data. A more systematic approach is therefore warranted for trace element studies in human concretion analysis. Conclusion Inductively coupled plasma atomic emission spectrometry has been successfully applied to the simultaneous determination of major, minor and trace elements in urinary stones.This technique was found to be an extremely useful tool in the analysis of calculi and could be easily adapted to accommodate additional elements. Rapidity, cost effectiveness and ease of operation have been achieved (i) by using a microwave-assisted wet digestion procedure, (ii) by employing a single set of operating conditions and calibration graphs and (iii) by determining all elements directly without pre-concentration. The microwave decomposition procedure employed in this study undoubtedly had a positive influence on the analytical accuracy andANALYST, NOVEMBER 1986, VOL. 111 1253 precision of the results. It is acknowledged, however, that the large sample mass required for precise concentration measurements precludes some of the smaller calculi for analysis by this method.Also, accuracy is limited by inhomo- geneities that are invariably found in the sampled aliquots. The lack of stone standards constitutes another limitation in the attempt to improve analytical reliability. Most recently an “animal bone” standard (IAEA H-5) was introduced ,20,21 which currently provides the closest match between a standard and stone matrix. Unfortunately, this standard was not available for this investigation. An independent check of the accuracy of ICP-AES (and XRD) analyses was obtained by participations in three interlaboratory tests (urinary calculus analyses) organised by the German Society for Clinical Chemistry. These tests confirmed that good accuracy and precision were obtained with (quantitative) ICP-AES and (qualitative) XRD tech- niques.16 Some of the trace elements explored were found to be at concentrations too low to be measured with ICP-AES. However , the detection limits achieved under the compromise settings selected for analytical parameters in this study could possibly be improved by optimising the conditions specifically for the low concentration elements occurring in calculi. Further improvement may be achieved by employing pre- concentration or more sensitive techniques. Future technical developments of the sample introduction system of ICPs and new sample preparation techniques will without doubt over- come these present shortcomings. The quantitative determination of trace elements in calculi is essential for understanding their aetiology. It is now accepted that the crystallisation processes occurring during the formation of stones are influenced by these elements, even if these are present in minute concentrations only.8.9 Although their functional role in urinary calculi is still unknown, many of the elements studied in this investigation were shown to promote or inhibit the precipitation of calculi.s.9 It is hoped that a comprehensive (multivariate) statistical analysis of the data obtained22 will contribute to a better understanding of the stone-forming process. The financial support of the Council for Scientific and Industrial Research (CSIR, South Africa) and the Medical Research Council (MRC, South Africa) in the person of Dr.Rodgers is gratefully acknowledged. The authors thank Miss B. Collocot for the loan of the microwave oven.References 1. 2. 3. 4. 5. Mermet, J. M., and Hubert, J., Prog. Anal. At. Spectrosc., 1982, 5 , 1. Schramel, P., Spectrochim. Acta, Part B, 1983, 38, 199. Olsen, S. D., Rama, D. B. K., and Bohmer, R. G., ChemSA, 1985, 11, 144. Wandt, M. A. E., Pougnet, M. A. B., and Rodgers, A. L., Analyst, 1984, 109, 1071. Wandt, M. A. E., Pougnet, M. A. B., and Rodgers, A. L., in Schwille, P. O., Smith, L. H., Robertson, W. G., and Vahlensieck, W. , Editors, “Proceedings of the 5th Interna- tional Symposium on Urolithiasis and Related Clinical Research,” Plenum, New York, 1985, p. 699. Feinendegen, L. E., and Kasperek, K., Trace Elem. Anal. Chem. Med. Biol., Proc. Int. Workshop, lst, 1980, p. 1. Schramel, P., and Xu-Li-Giang, ICP Inf Newsl., 1982,7,429. Meyer, J. L., and Angino, E. F., Invest. Urol., 1977, 14, 347. Hesse, A., Schneider, H.-J., and Berg, W., Zentralbl. Pharm., 1978, 117, 753. Levinson, A. A., Nosal, M., Davidman, M., Prien, E. L., Sr., Prien, E. L., Jr., and Stevenson, R. G., Invest. Urol., 1978,15, 270. 11. Thomas, W. C., Jr., Proc. SOC. Exp. Biol. Med., 1982, 170, 321. 12. Smith, S. B., Jr., Schliecher, R. G., Dennison, A. G., and McLean, G. A . , Spectrochim. Acta, Part B, 1983, 38, 157. 13. Pougnet, M. A. B., and Wandt, M. A. E., ChemSA, 1986,12, 16. 14. Pougnet, M. A. B., Orren, M. J., and Haraldsen, L., Int. J. Environ. Anal. Chem., 1985, 21, 213. 15. Lichte, F. E., Hopper, S . , and Osborn, T. W., Anal. Chem., 1980, 52, 120. 16. Wandt, M. A. E. , Dissertation, Department of Physical Chemistry, University of Cape Town, 1986. 17. Donev, I . , Mashkarov, S . , Maritchkova, L., and Gotsev, G., J. Radioanal. Chem., 1977, 37, 441. 18. Nagy, Z . , Szabo, E., and Kelenhegyi, M., Z. Urol., 1963,56, 185. 19. Hesse, A., Dietze, H.-J., Berg, W., and Hienzsch, E., Eur. Urol., 1977, 3, 359. 20. Lee, J., ICP Inf. Newsl., 1983, 8, 553. 21. Mahanti, H. S., and Barnes, R. M., Anal. Chim. Acta, 1983, 151, 409. 22. Wandt, M. A. E . , and Underhill, L. G., J. Urol., submitted for publication. 6. 7. 8. 9. 10. Paper A61163 Received May 23rd 1986 Accepted June 6th, 1986