JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 303 Determination of Chemical Forms of Mercury in Human Hair by Acid Leaching and Atomic Absorption Spectrometry* Karel Kratzer Petr BeneS and V6ra SpevaCkova Department of Nuclear Chemistry Faculty of Nuclear Science and Physical Engineering Czech Technical University in Prague 115 19 Prague 1 Brehova 7 Czec Republic Dana Kolihova and Jana Zilkova Department of Analytical Chemistry lnstitut of Chemical Technology Technicka 5 766 28 Prague 6 Czech Republic A simple method for the determination of sub-microgram amounts of mercury species in human hair is described. The method is based on the selective leaching of methylmercury from hair with hydrochloric acid followed by determination of the separated mercury species using cold vapour atomic absorption spec- trometry.The results obtained by the proposed method are compared with results obtained by the solvent extraction method. Both methods gave the same results and are suitable for the determination of methylmer- cury and inorganic mercury in hair with natural (10-100 ng g-') or higher contents of mercury with precision higher than 10%. Keywords Mercury species determination; hair; cold vapour atomic absorption spectrometry The levels of naturally-occurring mercury in the environment are generally low. Inorganic mercury (Hgin) can be methylated in the environment the resultant methylmercury (MeHg) is readily taken up by organisms and is released more slowly from organisms than inorganic mercury.',2 Since mercury species are excreted from the body partly by deposition in hair human hair stores information concerning exposure to mercury species chronologically over an extended p e r i ~ d .~ Various methods for the determination of MeHg in environ- mental samples have been developed. Different techniques have been used for separation of MeHg from Hg (ion exchange solvent extraction volatilization distillation).4-10 At present the most frequently used method for the separation is solvent e ~ t r a c t i 0 n . l ~ ' ~ To release mercury species bound in solid samples acid leaching is often u ~ e d ' " ~ . ' ~ ~ ~ occasionally in the presence of copper(r1) salt. However only limited knowledge still exists on the behaviour of individual mercury species in the leaching process and on possible changes in the speciation of mercury during leaching.This also applies for the isolation of mercury from hair. Therefore the leaching of MeHg and Hgin from hair was studied using hydrochloric acid at various concen- trations and the effect of the presence of copper(11) ions on the extraction was investigated. A radiotracer method was used throughout this work. The application of radioactively labelled mercury species greatly facilitates the study of the behaviour of the species in the separation. The labelled species can be easily traced during the separation by measurement of the activity of samples if isotope exchange between the labelled species and other mer- cury species is negligible or slow. The rate and extent of the isotope exchange between MeHg and Hg in aqueous solution is known" and therefore conditions could be selected to maintain a negligible isotopic exchange.The basic problem in the use of the radiotracer method for the proposed purpose was to achieve equal behaviour between the labelled mercury species added to the hair samples and the mercury species naturally present in the hair. It is not easy to prove such behaviour using model experiments practical analyses must be carried out in order to do so. In the present paper a simple and rapid procedure for the separation of MeHg from hair samples is reported and a * Presented at the XXVIII Colloquium Spectroscopicum Internationale (CSI) York UK June 29-July 4 1993. method for the determination of mercury species in hair by cold vapour atomic absorption spectrometry (CVAAS) is pro- posed.The results obtained by this method are compared with the data on MeHg concentration in the same hair determined after alkaline decomposition of hair followed by separation by solvent extraction.16 Apart from the verification of the accuracy of the determination of mercury species by the proposed methods the comparison also helps to solve the problem of the equal behaviour of the labelled and natural mercury species from hair in the separation process. Experiment a1 Apparatus A Radiometric Assembly NV 3102 TESLA with well-type NaI(T1) crystal was used. The atomic absorption spectro- meter was a single purpose instrument TMA-254 TESLA Czechoslovakia capable of determining sub-nanogram amounts of mercury. The solid or liquid sample is placed on a boat then automatically transferred into a combustion furnace where it is burned in a stream of oxygen.The combus- tion products pass through a catalytic furnace where the oxidation is completed. The combustion products are then passed in a stream of oxygen through an amalgamator where mercury is trapped. After heating of the amalgamator to a high temperature the entrapped mercury is released and driven to tandem measuring cells where absorbance is measured. The whole analytical run including all the parameters affecting the sensitivity and reproducibility of the determination are checked and controlled by a microprocessor. Chemicals Hydrochloric acid and sodium hydroxide (Suprapur Merck Germany) were used. Analytical-reagent grade benzene was purified by distillation.Methylmercury chloride (analytical standard Riedel-de Haen Germany) was dissolved in 0.01 moll-' sodium hydroxide to the appropriate concen- tration. A solution of 203Hg(N03)2 in 0.05 mol I-' nitric acid specific activity 250 GBq g-' of Hg (Du Pont de Nemours Germany) was prepared and mercury(11) standard solution (1 g 1 - l ) (Astasol Analytica Czech Republic) was used. All other solutions were prepared using distilled water and analyt- ical-reagent grade reagents. Methylmercury labelled with 203Hg was prepared by the isotope-exchange304 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 Hair Samples About 0.5 kg of human hair (a mixture obtained from different persons) was cut to less than 5 mm long pieces using stainless- steel scissors washed according to the procedure recommended by the International Atomic Energy Agency and the World Health Organization17 and homogenized by mixing.A portion of hair was ground down in an agate mill. Radioactively labelled hair was prepared by two alternative procedures Procedure A16 A 20 ml portion of aqueous phase containing 0.01 mol I-' acetate buffer (pH 4.7) 0.001 mol 1-' NaCl and 0.05- 0.1 pg ml-' of Hg as radioactively labelled Hg or MeHg is stirred with 1 g of hair for 1 h. The hair is separated by centrifugation washed twice with 40 ml of distilled water twice with 40 ml of acetone and air dried. Procedure B A 20 ml portion of 2 mol 1-' HC1 containing 0.05-0.1 pg ml- ' of ,03Hgin is shaken on a mechanical shaker with 1 g of hair for 100 h.The hair is separated by centrifugation washed twice with 40 ml of distilled water twice with 40 ml of acetone and air dried. Leaching of Mercury Species For the study of leaching of mercury species from hair using HC1 in the concentration range 0.1-5mol 1-' the following experiments were carried out. After determination of its initial activity (Ao) 10-150 mg of radioactively labelled (with either Me203Hg or ,03Hgin) hair were shaken in a centrifuge tube for appropriate time with 1-2ml of leaching solution which contained hydrochloric acid of the required concentration and in certain cases also 1 moll- ' CuC1 solution acidified by HC1. After centrifugation the activity of the separated aqueous phase (A) was measured and the distribution coefficient (K,) was calculated &-A V' K = - X - A m where V' is the volume of the aqueous phase and m is the mass of hair sample.Determination of Mercury Species by Acid Leaching On the basis of results obtained in the study of leaching the following procedure is recommended (Scheme 1) 90-125 mg of cut hair is shaken with 3.6-5 ml of 2 mol 1-' HCl (the ratio V m is kept at 40 ml g-') on a mechanical shaker for 4 h. The aqueous phase is separated by centrifugation and the mercury content is determined using the TMA-254. From this value the concentration of MeHg in the original hair sample is calculated. The separated hair is washed twice with 5 ml of distilled water and air dried. The content of Hg in hair is determined directly using the TMA-254 (see below). Determination of Mercury Species by Solvent Extraction For verification of the proposed leaching method the extraction procedure16 was modified (Scheme 2) 1.4 ml of 10 mol I-' NaOH is added to 1 g of hair in a centrifuge tube.The tube is kept in a thermostat at 90-95°C for 30min. Then 5ml of distilled water are added to the dissolved sample and the pH is adjusted to 0.5-1.0 using concentrated H2S0,. The sample is cooled to room temperature 1 g of solid KI is added and MeHg is extracted by 30min shaking with 4ml of benzene. After separation of the phases by centrifugation the content of MeHg is determined directly in the organic phase and/or I 1 Hair (90-125 mg)+2 rnol I-' HCI (3.6-5 ml) v/m=40 ml g - ' L 1 I 1 v 1 Shaken for 4 h 1 t I r 1 I I MeHg in aqueous phase I 1 Hair washed with H,O (2 x 5 rnl) I determined by AAS I Air dried 1 1 I Hgi determined by AAS I Scheme 1 Hair (1 g) + 10 rnol I-' NaOH (1.4 ml) heated for 30 min at 90-95 "C H,O added (5 rnl) +concentrated H,SO to pH 0.5-1 Cooling Solid KI ( 1 g) + benzene (4 ml) added I I Shaken for 30 rnin 1 Centrifuged I r 7 1 .MeHg in organic phase (1.5 rnl) determined by AAS I I Organic phase (2 ml) $2 rnol I-' NaOH (2 ml) re-extracted (15 min) IL 1 Centrifuged i___r_l I MeHg in aqueous phase (1.5 rnl) I determined by AAS Scheme 2 after re-extraction into 2 moll-' NaOH using the TMA-254 instrument. Determination of Mercury Using the TMA-254'*-'' A 50-200 pl amount of the solution to be analysed (cf. Schemes 1 and 2) or 10-25 mg of cut hair is introduced into the analyser and thermally treated in a programmed process.The sample is initially dried and then combusted in the stream of oxygen. The mercury vapour is trapped quantitatively on the surface of a gold amalgamator. Mercury preconcentrated in this way is then evaporated into the two measuring cells of the system. Results and Discussion The distribution of labelled mercury species between hair and HCl solution of various concentrations after 4 h of leaching isJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 305 shown in Fig. 1 (cut hair) and Fig. 2 (ground hair). No differ- ences have been found between leaching of mercury species from ground hair labelled by Procedures A and B. However in the case of cut hair the distribution coefficient of Hg depends on the method of spiking used (Fig.1 curves C and E). The presence of Cu2+ in the leaching solution causes a decrease in K of both MeHg and Hgi and makes the separation of these forms less efficient. The dependence of K on time of leaching using 2 moll-' HC1 (Fig. 3) indicates that the equilibrium is reached within approximately 4 h. Only for cut hair labelled with Hg by Procedure A does the quick transfer of the label into the aqueous phase followed by the slow re-uptake by hair occur (Fig. 3 curve C). Similar uptake is observed when non-spiked hair is treated with 2 moll-' HCl containing labelled Hg (curve D). The differences in the behaviour of Hg spiked on cut hair by Procedures A and B suggest that with Procedure A Hg is adsorbed on the surface of hair whereas in Procedure B it is absorbed into hair.On the basis of the above results while respecting the natural level of mercury species in hair and the sensitivity of the analytical method used the leaching with 2 mol I-' HCl for 4 h at the ratio Vrn=40 ml g-' was chosen for separation of mercury species in hair. The applicability of the proposed method was verified by comparison of the results obtained by analysis of the same hair sample using this method (Scheme 1) 1 I x 1 O 4 x103 LY \ I 1x102 - E 10 1 1 x lo-' I 1 I I I 0 1 2 3 4 5 6 IHCll/mol 1-l Fig. 1 Distribution of labelled mercury species between hair and HCl solution after leaching from cut hair for 4 h A MeHg; B MeHg in the presence of 1 mol I-' Cu2+; C Hg prepared by solvent extraction; D Hgin prepared by solvent extraction in the presence of 1 moll-' Cuz+; and E Hg prepared by acid leaching Fig.2 Distribution of labelled mercury species between hair and HCl solution after leaching from ground hair for 4 h A MeHg; By MeHg in the presence of 1 mol I-' Cu2+; C Hgi,; and D Hgin in the presence of 1 moll-' CuZ+ 1 x lo3 r 1 x 102 - E 9 1x10 -- _ - - - - - - - - _ _ - 1 x lo-' lLIZl 0 20 40 60 80 100 Time/h Fig.3 Dependence of K on time of leaching using 2moll-' HCl A MeHg cut hair; B Hg cut hair prepared by solvent extraction; C Hgin cut hair prepared by acid leaching; D uptake of Hg by cut hair; E MeHg ground hair; and F Hgin ground hair and the solvent extraction method (Scheme 2). The results are given in Tables 1 and 2. The results show that consistent values were obtained for the content of methylmercury using both methods.The differences are not statistically significant. Results obtained by direct determination of total mercury in hair are presented in Table 3. Good agreement was reached between these results and the sum of contents of mercury species found by the acid leaching method (Table 1). This confirms that the results obtained by analysis of hair after acid leaching represents the content of Hg,. The good agreement of the data obtained on the speciation of natural mercury in hair using two very different separation methods also suggests that natural mercury species from hair behave in the separation process similarly to radioactively labelled species used in the development of both separation methods. This confirms the validity of basic assumptions of Table 1 Determination of mercury species in hair by AAS using the acid leaching method Mass of hair/ mg 125 107 117 115 111 103 110 112 Mean _+ CI* MeHg found/ ng g-' of Hg 135 124 136 128 128 132 125 120 129 f 5 Hg found/ ng g-' of Hg 283 29 1 287 298 28 5 283 309 293 291 f 8 Total Hg/ ng g-' of Hg 418 415 423 426 413 415 434 413 420 f 12 ~~ * CI = 95% confidence interval.Table2 Determination of MeHg in hair by AAS using the solvent extraction method Mass of hair/ g 0.888 1.032 1.088 1.076 1.024 1.000 1 .om Mean f CI* MeHg found in ng g- of Hg 144 114 113 120 127 123 122 123 10 C6H61phase/ MeHg found in NaOH phase/ ng g-' of Hg 103 121 99 99 118 130 135 115 f 12 * CI = 95 % confidence interval.306 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL.9 Table 3 Direct determination of total mercury in hair by TMA-254 Mass of hair/ mg 11.52 15.90 16.12 16.31 15.42 23.46 11.38 17.24 18.03 20.03 Hg found/ ng 5.09 6.96 6.82 6.93 6.67 9.75 4.98 7.42 7.82 8.47 Hg total/ ng 8-l of Hg 441 437 423 424 432 41 5 437 430 43 3 422 Mean k CI* 429 & 6 * CI = 95% confidence interval. radiotracer methods applied in the development of the procedure. The proposed leaching method allows a simple and rapid determination of both MeHg and inorganic Hg mercury in a 100mg sample of human hair. The method is suitable for serial analysis. We thank the International Atomic Energy Agency Vienna Austria for funding this work. References Stary J. and Kratzer K. Int. J. Environ. Anal. Chem. 1980,8 189.Stary J. and Kratzer K. Radiochem. Radioanal. Lett. 1980,437. Bencze K. Fresenius’ J. Anal. Chem. 1990 337 867. May K. Stoeppler K. and Reisinger M. Toxicol. Environ. Chem. 1987,13 153. 5 6 7 8 9 10 11 12 13 May K. and Stoeppler K. Fresenius’ 2. Anal. Chem. 1984 317 248. Gage J. C. Analyst 1961 86 457. Westoo G. Acta Chem. Scand. 1968 22 2277. Zelenko V. and Kosta L. Talanta 1973 20 115. Dermelj M. Horvat M. Byrne A. R. and Stegnar P. Chemosphere 1987 16 877. Horvat M. May K. Stoeppler M. and Byrne A. R. Appl. Organometall. Chem. 1988 2 850. Horvat M. Byrne A. R. and May K. Talanta 1990 37 207. Horvat M. Water Air and Soil Pollution 1991 56 95. IAEA-UNEP-FAO-IOC Reference Method for Marine Pollution Studies No 13 Rev. 1. United Nations Environment Programme (UNEP) Regional Seas Programme Activity Centre Geneva. 1992. 14 Stary J. Havlik B. Prasilova J. Kratzer K. and Hanusova J. Radiochem. Radioanal. Letters 1978 35 47. 15 Stary J. and Prasilova J. Radiochem. Radioanal. Lett. 1976 26 193. 16 Kratzer K. Benes P. and Spevackova V. Report on the Second Research Co-ordination Meeting. NAHRES-13 pp. 25-3 1. IAEA Vienna 1992. 17 UNEP-WHO-IAEA Reference Method for Marine Pollution Studies No. 46. United Nations Environment Programme (UNEP) Regional Seas Programme Activity Centre Geneva 1987. 18 Miholova J. Mader P. Szakova J. Slamova A. and SvatoS Z. Fresenius’ J. Anal. Chem. 1993 345 256. 19 HlavaE R. Doleial J. Kolihova D. Sychra V. PietroS L. Valenta S. Puschel P. and Formanek Z. Proceedings of the XIV Colloquium Spectroscopicum Internationale (CSI) Garmisch-Partenkirchen Germany 1985. Paper 3/05092K Received August 23 1993 Accepted November 2 1993