Extraction Separation and Electrothermal Atomic Absorption Spectrometric Determination of Thallium in Some Sulfide Minerals

ANALYTICAL SCIENCES MARCH 2001, VOL. 17 425 2001 © The Japan Society for Analytical Chemistry

Dragica ZENDELOVSKA*† and Trajcùe STAFILOV**

*Institute of Preclinical and Clinical Pharmacology with Toxicology, Faculty of Medicine, St. Cyril and Methodius University, 50 Divizija, 1000 Skopje, Macedonia **Institute of Chemistry, Faculty of Science, St. Cyril and Methodius University, POB 162, 1001 Skopje, Macedonia

A method for thallium determination in some sulfide minerals (sphalerite, galena, orpiment and realgar) by electrothermal

atomic absorption spectrometry has been proposed. Mineral samples were dissolved in a mixture of HNO3 and HCl. The effect of interfering elements (Zn, Pb and As) on the determination of Tl in the investigated minerals was studied. These investigations show that it is not possible to determine thallium directly from solutions obtained by dissolution of minerals in mineral acids, because these matrix elements tend to decrease the absorbance of thallium. Also, it was found that the investigated minerals contain iron that tends to increase the absorbance of thallium. Therefore, a method for extraction of thallium with isoamyl acetate from 10 mol/l HCl media was proposed. In these conditions, Zn, Pb and As remain in the inorganic phase, but iron and thallium are extracted into organic phase. Reextraction of iron from the

organic layer was performed with 4 mol/l H2SO4. The determination of thallium was performed in the organic phase using Pd as modifier. The investigated sulfide minerals originated from different mines from the Republic of Macedonia. The relative standard-deviation range was between 2.20 to 3.92%. The detection limit of the method (calculated as 3SD of the blank) was found to be 0.05 µg/g.

(Received July 24, 2000; Accepted November 9, 2000)

The detection of trace elements in mineral samples helps to It is also known that some thallium minerals are present in establish the conditions under which these minerals had been lead and zinc mines; also, in the metallurgical processes of lead formed, e.g., it helps to broaden knowledge of geology of and zinc production thallium becomes a very important different mines to determine the purity of minerals, to determine pollution problem concerning the environment. Therefore, it is the presence of very rare and important elements. Also, these important to follow the content of thallium in arsenic minerals results sometimes explain the background radiation of the orpiment and realgar from the Alsùar mine as well as lead localities, etc. Thallium is very important because of a (galena) and zinc (sphalerite) minerals which originate from possibility to apply its mineral lorandite (TlAsS2) in obtaining different lead and zinc mines from the Republic of Macedonia. knowledge in fundamental investigations. Thus, in 1976, Atomic absorption spectrometry (AAS) is one of the most Freedman et al.1 proposed applying of the reaction between often used techniques which can be employed for trace-element 205Tl and solar neutrinos, producing 205Pb, to study the neutrino analyses in minerals. The principal advantage of electrothermal flux from the sun. The aim of this idea is to determine the atomic absorption spectrometry (ETAAS), i.e., the detection content of 205Pb accumulated in thallium minerals, which is limits, which are extremely low compared with other methods produced in this nuclear reaction. For this purpose, a thallium- and are attainable for a great number of elements, makes this rich mineral is required as the target. One candidate is lorandite technique particularly suitable for determining elements present

(TlAsS2). The Sb-As-Tl deposit at Alsùar in the Republic of in geochemical materials at concentrations on the order of ppb. Macedonia is a unique mineral deposit in the world and Errors in the atomic absorption spectrometric determination of lorandite is the most widespread Tl mineral in the deposit. thallium caused by major- and minor-element interferences are Therefore, it is important to determine exactly the concentration well known. With increased awareness of the magnitude of of lead in thallium minerals, and also the concentration of interelement interferences, their alleviation or elimination thallium in other minerals associated with thallium minerals, becomes essential. An appropriate choice of the method used such as realgar or orpiment.2Ð5 This is very important because for partitioning elements depending on their mode of occurrence these minerals, under the condition that they are co-genetic, can or chemical forms in the sample, by sequential fractionation or be used as monitors for determining the contribution of partial dissolution techniques, can result in conditions where the background reactions, which originate from natural ratio between the interfering elements and the elements to be radioactivity. determined is much greater than in the original sample. Some authors have investigated the possibility of determining † To whom correspondence should be addressed. thallium directly from the sample solution, by flame atomic E-mail: [email protected] absorption spectrometry (FAAS)2,6 or by ETAAS.7Ð10 426 ANALYTICAL SCIENCES MARCH 2001, VOL. 17

Table 1 Instrumental parameters for determining Tl by ETAAS

Wavelength 276.8 nm Slit 0.5 nm Lamp current 10.0 mA Calibration mode absorbance, peak area Background correction Zeeman Sample volume 10 µl Matrix modifier volume 5 µl Dry Temperature 120ûC Ramp time 20 s Fig. 1 Influence of Pb ( ), Zn ( ), As ( ) and Fe ( ) on the Hold time 25 s absorbance of thallium (50 ng/ml) in aqueous solutions. Pyrolysis Temperature 700ûC Ramp time 10 s Hold time 5 s Atomization solutions of Tl, As, Pb, Fe, and Zn were prepared by dissolving Temperature 2100ûC Tl(NO3)3, As2O3, Pb(NO3)2, Fe and Zn metal. The mass Ramp time 1 s concentration of the solutions was 1000 mg/l, and from these Hold time 2 s solutions other diluted solutions were prepared. Palladium Clean modifier (as a solution with a concentration of 300 mg/l) was Temperature 2200ûC prepared by the dilution of a palladium modifier solution for Ramp time 1 s graphite-furnace AAS (Merck, Germany) with a Pd Hold time 2 s concentration of 10 ± 0.2 g/l (Pd(NO ) in nitric acid). Gas argon 3 2 Mineral samples of orpiment and realgar were collected from Alsùar mine and sphalerite and galena, were collected in Zletovo, Sasa and Toranica lead and zinc mines from the Republic of Macedonia. To overcome matrix interferences, some authors have suggested the addition of different matrix modifiers, such as Procedures palladium,11Ð13 rhodium nitrate,14 mixed modifier of Pd/Mg Sphalerite. 0.05 to 0.5 g of a powdered sample of sphalerite 15Ð17 18 19 (NO3)2, ascorbic acid or EDTA. was dissolved in 15 ml of 2 mol/l HNO3. The solution was Separation by extraction has been extensively used to isolate evaporated to dryness and the residue was dissolved in 20 ml of thallium from possible interferences in a sample matrix and its 10 mol/l HCl. A few drops of H2O2 were added and the solution determination by FAAS20Ð31 or ETAAS.3,32Ð38 Some authors was boiled for 5 min. It was then transferred to a separatory suggest the extraction of thallium as different chalogenide funnel. A 10-ml volume of isoamyl acetate was added and the complexes, such as bromide complex into methyl isobutyl mixture was shaken for 5 min. After the phases were separated, ketone (MIBK)22,27,37 or into isopropyl ether,21,23,26 iodide reextraction of the present iron in the organic layer was 24,25,29,32,35 complexes into MIBK or chloride complexes of performed with 10 ml of 4 mol/l H2SO4. Thallium was thallium.31,34 Other authors suggest different dithiocarbamates determined in the organic phase by applying 10 µl of the for thallium extraction into MIBK,20,31,33,38 butyl acetate31 or solution and 5 µl of a Pd modifier into a graphite furnace. toluene.36 Galena. 0.05 to 0.5 g of a powdered sample of galena was 3 In our previous investigation, we applied isoamyl acetate for dissolved in 10 ml conc. HNO3 and 2 ml of H2O2. After the thallium extraction from different minerals. However, the solution was evaporated to dryness, the residue was dissolved in presence of iron interferes with thallium determination because 20 ml of 10 mol/l HCl and the solution was transferred into a of its coextration with Tl. In this paper, we present a modified separatory funnel. The same procedure as that in the case of method for the determination of thallium in some arsenic sulfide sphalerite was then followed. minerals (orpiment-As2S3 and realgar-As4S4), zinc (sphalerite- Realgar and orpiment. 0.05 to 0.5 g of a powdered sample of ZnS) and lead (galena-PbS) minerals, with the extraction of realgar and orpiment were dissolved in 10 ml conc. HNO3 and 2 thallium by isoamyl acetate and the reextraction of iron using ml H2O2. The same procedure as that in the case of sphalerite sulfuric acid. Thallium determination was performed by was then followed. Zeeman ETAAS. Results and Discussion Experimental Since thallium occurs at low concentrations in mineral samples, Instrumentation it is necessary to carry out a preliminary separation to A Varian SpectrAA 640Z Zeeman electrothermal atomic concentrate it and, at the same time, to separate it from the absorption spectrophotometer with a PSD-100 autosampler was matrix elements. For sample dissolution, acid digestion is used. A thallium hollow-cathode lamp was used as a source. usually used. However, acid digestion produces solutions that The operating conditions for the thallium determination are contain different ions of elements dissolved from the sample given in Table 1. matrix, which may interfere with the determination of thallium. The interference of Pb, Zn and As as matrix elements was Reagent and samples studied. Because these sulfide minerals also contained a high All reagents and standards were of analytical grade. Stock concentration of iron,39,40 the influence of iron upon the ANALYTICAL SCIENCES MARCH 2001, VOL. 17 427

Table 2 Content of thallium in the investigated minerals

Mineral/ Tl added/µg Tl calc./µg gÐ1 Tl found/µg gÐ1 R, % mine

Sphalerite-Zletovo (0.1 g) 1 0.0 — 1.56 — 2 0.5 6.56 6.4 97.6 3 1.0 11.56 11.3 97.7 4 2.0 21.56 21.43 99.4 Sphalerite-Toranica (0.08 g) 1 0.0 — <0.05 — 2 0.5 6.25 6.25 100.0 3 1.0 12.5 12.1 96.8 Fig. 2 Absorption signal of the same concentration of thallium in 4 2.0 25.0 25.0 100.0 the inorganic phase (a) and in the organic phase after extraction (b). Galena-Zletovo (0.1 g) (c) Background signal. 1 0.0 — <0.05 — 2 0.5 5.0 5.0 100.0 3 1.0 10.0 9.8 98.0 4 2.0 20.0 19.3 96.5 absorbance of thallium was investigated. A series of solutions Galena-Sasa (0.1 g) with the same concentration of thallium and various 1 0.0 — <0.05 — concentrations of interfering elements were prepared, so that the 2 0.5 5.0 5.0 100.0 concentration of Tl was similar to the concentration in the 3 1.0 10.0 10.0 100.0 4 2.0 20.0 19.3 96.5 sample solutions. The results from these investigations are Orpiment-Alsˇar (0.1 g) given in Fig. 1. These data show that Tl can not be determined 1 0.0 — 152.8 — directly from a solution obtained by dissolving minerals in 2 3.0 182.8 178.9 97.9 mineral acids, because Pb, Zn and As tend to decrease and iron 3 5.0 202.8 195.8 96.5 tends to increase the absorbance of thallium. Therefore, we Realgar- Alsˇar (0.1 g) proposed to modify the method for thallium extraction as a 1 0.0 — 64.0 — chloride complex into isoamyl acetate3 and its determination by 2 3.0 94.0 93.5 99.5 Zeeman ETAAS 3 4.0 104.0 103.5 99.5 For this purpose, the first step in our work was to check the optimal concentration of HCl for this extraction. It was found that the highest efficiency of extraction (99.2%) could be achieved at a concentration of 10 mol/l of HCl by one-step anions (chlorides, sulfates, nitrate). extraction (instead of 3 times extraction as in previous work) Using the proposed method, some samples of the investigated and a 1:2 volume ratio of the organic and inorganic phases. minerals with and without standard additions (with a mass of This extraction was applied for arsenic minerals and for the first 0.1 g, except in the case of sphalerite-Toranica sample = 0.08 g, time for lead and zinc minerals. To check whether the As, Pb, was taken) were dissolved, and thallium was extracted and Zn and Fe coextract with Tl, a series of solutions with the same determined in the organic layer. The results given in Table 2 concentration of thallium and different concentrations of show satisfactory recovery values. interfering elements were prepared and thallium was extracted The method was applied to one reference standard sample (G- by the proposed procedure. It was found that in the organic 2). The value of Tl obtained by this method (1.03 µg/g) had the extracts only 0.2 Ð 0.35% of all arsenic, lead and zinc, coextract same general order of magnitude as the certified value (1.02 with thallium. However, it was found that iron is completely µg/g). Calibration curves were made using the proposed extracted together with thallium. extraction procedure for standard solutions of the investigated Iron is also extracted by isoamyl acetate3,41 and in the case elements. The relative standard deviation for this method when the content of iron is high it can influence on thallium ranges between 2.20 to 3.92%. The detection limit of the absorbance. To avoid these interferences, we found that iron method (calculated as 3SD of the blank) was found to be 50 can be successfully back-extracted with a solution of sulfuric ng/g. acid. For this purpose, it was found that 4 mol/l is the optimal concentration of H2SO4. At this concentration of sulfuric acid, thallium completely remained in the organic phase. The Conclusion optimal volume ratio of organic and inorganic phase was 1:1. It is known that thallium chlorides and nitrates interfere with A method is described for the determination of thallium in some thallium determination by ETAAS caused by their volatilization sulfide minerals (sphalerite, galena, orpiment and realgar) by in the pyrolysis stage.42,43 Therefore, the addition of a matrix Zeeman electrothermal atomic absorption spectrometry. To modifier is necessary. Our investigation has shown that avoid potential matrix interferences, thallium was extracted as a thallium can be efficiency atomized at temperatures above chloride complex into isoamyl acetate. 2100ûC by the addition of a palladium modifier. In this case, almost the same absorption signal was obtained for the same concentration of thallium in the inorganic phase before and in References the organic phase after the extraction of standard solutions (Fig. 2). As can be seen, the background absorption signal (Fig. 2) is 1. M. S. Freedman, C. M. Stevens, E. P. Honwitz, L. H. lower than the absorption signals of thallium. This signal Fuchs, J. Sherner, L. S. Goodman, and W. J. Childs, probably appeared because of the presence of some inorganic Science, 1976, 193, 1117. 428 ANALYTICAL SCIENCES MARCH 2001, VOL. 17

2. T. Stafilov, T. Todorovski, B. Grozdanova, and Lj. 23. G. P. Sighinolfi, At. Absorption Newsl., 1973, 12, 136. Spandzùeva, Nucl. Instr. Meth. Phys. Res., 1988, A271, 321. 24. N. K. Roy and A. K. Das, J. Indian Chem. Soc., 1990, 67, 3. T. Stafilov, A. Lazaru, and E. Pernicka, Acta Chim. Slo., 976. 1993, 40, 37. 25. B. Ebarvia, E. Makalalad, N. Rogue, and I. Rubeska, J. 4. T. Balic« Zùunic«, T. Stafilov, and D. Tibljasù, Geologica Anal. At. Spectrom., 1988, 3, 199. Macedonica, 1993, 7, 45. 26. M. Fratta, Can. J. Spectrosc., 1974, 19, 33. 5. T. Stafilov, S. Aleksovska, and V. Jordanovska, N. Jb. 27. H. G. Schrader, Metal, 1974, 28, 705. Miner. Abh., 1994, 167, 401. 28. N. T. Voskresenskaya, N. F. Pchelinsteva, and T. I. 6. E. Norval and W. H. Gries, Anal. Chim. Acta, 1976, 83, Tsekhonya, Zh. Anal. Khim., 1981, 36, 667. 393. 29. I. Janousek, Hutn. Listy, 1982, 37, 737. 7. H. Heinrichs, Fresenius Z. Anal. Chem., 1979, 294, 345. 30. C. H. Wachter and W. Weisweiler, Mikrochim. Acta, 1982, 8. H. Heinrichs and H. Keltsch, Anal. Chem., 1982, 54, 1211. 307. 9. H. Gorbauch, H. H. Rump, G. Alter, and C. Schmitt-Henco, 31. P. Hannaker and T. C. Hughes, Anal. Chem., 1977, 49, Fresenius Z. Anal. Chem., 1984, 317, 236. 1485. 10. I. Leim, G. Kaiser, M. Sager, and G. Toelg, Anal. Chim. 32. M. A. De Kastro, R. Bugagao, B. Ebarivia, N. Roque, and Acta, 1984, 158, 179. I. Rubeska, Geostand. Newsl., 1988, 12, 47. 11. B. Welz, G. Schlemmer, and R. J. Mudakavi, Anal. Chem., 33. R. Keil, Fresenius Z. Anal. Chem., 1981, 309, 181. 1988, 60, 2567. 34. M. Ikramuddin, At. Spectrosc., 1983, 4, 101. 12. M. T. Rettberg and M. L. Beach, J. Anal. At. Spectrom., 35. X. Guo, M. Heashi, R. R. Brooks, and R. D. Reeves, Anal. 1989, 4, 427. Chim. Acta, 1992, 266, 127. 13. Y. Zheng and Y. Wang, Spectrosc. Lett., 1994, 27, 817. 36. B. Magyar, B. Wampler, and A. Wuersch, Fortschr. 14. Y. Funato and M. Takasaka, Anal. Sci., 1991, 7, 455. Atomspektrom. Spurenanal., 1984, 1, 659. 15. M. S. Reichart, At. Spectrosc., 1992, 13, 178. 37. C. M. Elson and C. A. R. Albuqqerkue, Anal. Chim. Acta, 16. J. Creed, T. Martin, L. Lobring, and J. O’Dell, Environ. Sci. 1982, 134, 393. Technol., 1992, 26, 102. 38. W. Schmidt and F. Dietl, Fresenius Z. Anal. Chem., 1983, 17. G. Schlemmer and B. Welz, Spectrochim. Acta, 1986, 41B, 315, 687. 1157. 39. A. Lazaru and T. Stafilov, Geologica Macedonica, 1993, 7, 18. X. Shan, S. Yuan, and Z. Ni, Can. J. Spectrosc., 1986, 31, 73. 35. 40. D. Zendelovska and T. Stafilov, Geologica Macedonica, 19. K. Matsusaki and T. Yoshino, Bunseki Kagaku, 1986, 35, 1999, 13, 91. 931. 41. R. M. Dangall, T. S. West, and P. Young, Anal. Chem., 20. E. M. Donaldson and M. Wang, Talanta, 1986, 33, 35. 1966, 38, 358. 21. S. S. Murti, I. V. S. Rao, S. C. S. Rajan, and J. 42. D. Tsalev, V. Slaveykova, and P. Mandjukov, Spectrochim. Subrahmanyan, Talanta, 1989, 36, 601. Acta Rev., 1990, 13, 225. 22. A. E. Hubert and T. T. Chao, Talanta, 1985, 32, 568. 43. T. Stafilov, Spectrochim. Acta, Part B, 2000, 55, 893.