Anal. Chem. 2003, 75, 1726-1732
Speciation of Mercury by Hydrostatically Modified Electroosmotic Flow Capillary Electrophoresis Coupled with Volatile Species Generation Atomic Fluorescence Spectrometry Xiu-Ping Yan,*,†,‡ Xue-Bo Yin,†,‡ Dong-Qing Jiang,‡ and Xi-Wen He†,‡
State Key Laboratory of Functional Polymer Materials for Adsorption and Separation, and Central Laboratory, Research Center for Analytical Sciences, College of Chemistry, Nankai University, Tianjin 300071, China
A novel method for speciation analysis of mercury was developed by on-line hyphenating capillary electrophoresis (CE) with atomic fluorescence spectrometry (AFS). The four mercury species of inorganic mercury Hg(II), methymercury MeHg(I), ethylmercury EtHg(I), and phenylmercury PhHg(I) were separated as mercury-cysteine complexes by CE in a 50-cm × 100-µm-i.d. fused-silica capillary at 15 kV and using a mixture of 100 mmol L-1 of boric acid and 12% v/v methanol (pH 9.1) as electrolyte. A novel technique, hydrostatically modified electroosmotic flow (HSMEOF) in which the electroosmotic flow (EOF) was modified by applying hydrostatical pressure opposite to the direction of EOF was used to improve resolution. A volatile species generation technique was used to convert the mercury species into their respective volatile species. A newly developed CE-AFS interface was employed to provide an electrical connection for stable electrophoretic separations and to allow on-line volatile species formation. The generated volatile species were online detected with AFS. The precisions (RSD, n ) 5) were in the range of 1.9-2.5% for migration time, 1.8-6.3% for peak area response, and 2.3-6.1% for peak height response for the four mercury species. The detection limits ranged from 6.8 to 16.5 µg L-1 (as Hg). The recoveries of the four mercury species in the water samples were in the range of 86.6-111%. The developed technique was successfully applied to speciation analysis of mercury in a certified reference material (DORM-2, dogfish muscle). Mercury is an environmentally and toxicologically important element and is one of the most widely studied heavy metals because of the toxicological and biogeochemical behavior of its organic and inorganic compounds.1,2 It is considered by the * Corresponding author. Fax: (86)22 23503034. E-mail: xpyan@nankai. edu.cn. † State Key Laboratory of Functional Polymer Materials for Adsorption and Separation. ‡ Research Center for Analytical Sciences. (1) Clevenger, W. L.; Smith, B. W.; Winefordner, J. D. Crit. Rev. Anal. Chem. 1997, 27, 1-27. (2) Environmental Health Criteria for Methyl Mercury. International Programme on Chemical Safety; World Health Organization (WHO): Geneva, 1990.
1726 Analytical Chemistry, Vol. 75, No. 7, April 1, 2003
Environmental Protection Agency (EPA) as a highly dangerous element because of its accumulative and persistent character in the environment and biota.3 There are numerous potential sources of mercury to environment, including atmospheric deposition, natural geologic deposits, industrial/municipal discharges, previously contaminated sediment, and fugitive sources, such as discarded batteries or containers of elemental mercury.4 Once released into the environment, various transformations of mercury may take place. It is well-known that the toxicity of mercury is highly dependent on its chemical form. Methylmercury, the most toxic species of mercury normally found in environmental and biological materials, is a particular concern because of its accumulation as it passes through the food chain.1-3 Accordingly, the rapid and sensitive determination of mercury species in addition to total mercury is highly essential for the interpretation of their biochemical behavior or assessment of the potential danger to organisms. The methodologies currently used for mercury speciation mainly involve a chromatographic technique, either gas chromatography (GC)5-12 or high performance liquid chromatography (HPLC),13-20 coupled with a highly sensitive element-specific (3) Sa´nchez Urı´a, J. E.; Sanz-Medel, A. Talanta 1998, 47, 509-524. (4) Morel, F. M. M.; Kraepiel, A. M. L.; Amyot, M. Annu. Rev. Ecol. Syst. 1998, 29, 543-566. (5) Puk, R.; Waber, J. H. Anal. Chim. Acta 1994, 292, 175-183. (6) DeDiego, A.; Tseng, C. H.; Stoichev, T.; Amouroux, D.; Donard, O. F. X. J. Anal. At. Spectrom. 1998, 13, 623-629. (7) Munaf, E.; Haraguchi, H.; Ishii, D.; Tokeuchi, T.; Goto, M. Anal. Chim. Acta 1990, 235, 399-404. (8) Wu, J. C. G. Spectrosc. Lett. 1991, 24, 681-697. (9) Ritsema, R.; Donard, O. F. X. Appl. Organomet. Chem. 1994, 8, 571-575. (10) Armstrong, H. E. L.; Corns, W. T.; Stockwell, P. B.; O’Connor, G.; Ebdon, L.; Evans, E. H. Anal. Chim. Acta 1999, 390, 245-253. (11) Prange, A.; Jantzen, E. J. Anal. At. Spectrom. 1995, 10, 105-109. (12) Hintelmann, H.; Evans, R. D.; Villeneuve, J. Y. J. Anal. At. Spectrom. 1995, 10, 619-624. (13) Costa-Fernandez, J.; Lunzer, F.; Pereiro-Garcia, R.; Sanz-Medel, A.; BordelGarcia, N. J. Anal. At. Spectrom. 1995, 10, 1019-1025. (14) Gerbersmann, C.; Heisterkamp, M.; Adams, F. C.; Broekart, J. A. C. Anal. Chim. Acta 1997, 350, 273-285. (15) Harrington, C. F. Trends Anal. Chem. 2000, 19, 167-179. (16) Hintelmann, H.; Wilken, R. D. Appl. Organomet. Chem. 1993, 7, 173-180. (17) Hempel, M.; Chau, Y. K.; Dutka, B. J.; McInnis, R.; Kwan, K. K.; Liu, D. Analyst 1995, 120, 721-724. (18) Wan, C.-C.; Chen, C.-S.; Jiang, S.-J. J. Anal. At. Spectrom. 1997, 12, 683687. (19) Bushee, D. S. E. Analyst 1988, 113, 1167-1170. 10.1021/ac026272x CCC: $25.00
© 2003 American Chemical Society Published on Web 02/27/2003
detector, such as atomic absorption spectrometry (AAS),5,6,14,15 atomic fluorescence spectrometry (AFS),7,8,10,16,17 inductively coupled plasma mass spectrometry (ICPMS)9,10,18,19 microwave-induced plasma atomic emission spectrometry (MIPAES),11,13 and inductively coupled plasma atomic emission spectrometry (ICPAES).20 Recently, there has been an increasing interest of capillary elctrophoresis (CE) for speciation analysis.21-23 CE offers several potential advantages for speciation analysis, such as minor disturbance on the existing equilibrium between different species, a rapid and efficient separation with relatively simple instrumental setup, and a minimal sample/reagent requirement.21-23 CE with UV24 and amperometric detection25 has been employed for the determination of mercury species. On-column detection in CE with a commercially available UV detector for speciation analysis, however, possesses inherent drawbacks, such as poor detection limits, interference of coexisting species with the same mobility, and likely unstable baseline. Sensitive and element-specific detection techniques for CE speciation would provide improved performance for this application. CE coupled with ICPMS is becoming of growing importance in speciation analysis, and several excellent papers on this hyphenated technique for mercury speciation have been published.26-28 Although highly sensitive, element- and isotope-specific characteristic of ICPMS makes it very attractive as an on-line detector for CE; the high instrumental and running costs of the ICPMS instrument as well as the strict requirement that the analyst be well-trained set a serious limitation on the wide application of such a hyphenated technique. Volatile species generation atomic fluorescence spectrometry (VSG-AFS) has been proved to be a very sensitive and selective technique for the determination of mercury species.10 Compared with ICPMS, AFS also presents the advantages of much lower running costs, shorter warm-up times prior to analysis, and easy handling.10 The use of VSG-AFS as an on-line detector of CE is expected to be attractive for the speciation analysis of mercury compounds owing to the low cost, easy operation, high selectivity, and sensitivity. Nevertheless, no such work has been reported before. The objective of this work was to develop a new methodology for speciation of mercury compounds by on-line coupling of CE to VSG-AFS using a newly developed interface.29 Potential factors affecting the separation of various mercury species and sensitivity were investigated in detail. The developed method was applied for the speciation of mercury in a biological reference material DORM-2 (dogfish muscle), and a number of water samples. (20) Krull, I. S.; Bushee, D. S.; Schleicher, R. G.; Smith, S. B. Analyst 1986, 111, 345-349. (21) Dabek-Zlotorzynska, E.; Lai, E. P. C.; Timerbaev, A. R. Anal. Chim. Acta 1998, 359, 1-26. (22) da Rocha, M. S.; Soldado, A. B.; Blanco, E.; Sanz-Medel, A. J. Anal. At. Spectrom. 2001, 16, 951-956. (23) Timerbaev, A. R. Talanta 2000, 52, 573-606. (24) Medina, I.; Rubi, E.; Mejuto, M. C.; Cela, R. Talanta 1993, 40, 1631-1636. (25) Lai, E. P. C.; Zhang, W.-G.; Trier, X.; Georgi, A.; Kowalski, S.; Kennedy, S.; Muslim, T. M.; Dabek-Zlototzynska, E. Anal. Chim. Acta 1998, 364, 6374. (26) da Rocha, M. S.; Soldado, A. B.; Blanco-Gonza´lez, E.; Sanz-Medel, A. J. Anal. At. Spectrom. 2000, 15, 513-518. (27) Lee, T. H.; Jiang, S. J. Anal. Chim. Acta 2000, 413, 197-205. (28) Tu, Q.; Qvarnstro ¨m, J.; Frech, W. G. Analyst 2000, 125, 705-710. (29) Yin, X.-B.; Yan, X.-P.; Jiang, Y.; He, X.-W. Anal. Chem. 2002, 74, 37203725.
Table 1. Operating Parameters of VSG-AFS and CE parameter
setting
Atomic Fluorescence Spectrometry mercury hollow cathode lamp 60 mA (boost current) atomizer temperature 400 °C quartz furnace height 6 mm negative high voltage of -270 V photomultiplier Volatile Species Generation System flow rate of 1% v/v HCl 1.8 mL min-1 flow rate of 0.2% m/v KBH4 2.5 mL min-1 carrier gas (argon) flow rate 280 mL min-1 CapillaryElectrophoresis fused-silica capillary 50 cm × 100 µm i.d. electrolyte buffer 100 mmol L-1 H3BO3 + 12% v/v methanol pH 9.1 voltage applied 15 kV analyte injection method hydrodynamic pressure sample injected 150 nL height difference between the 2 cm inlet and outlet of capillary
EXPERIMENTAL SECTION Instrumentation. All atomic fluorescence measurements were carried out on a model XGY-1011A nondispersive atomic fluorescence spectrometer (Institute of Geophysical and Geochemical Exploration, Langfang, China). A high-intensity mercury hollow cathode lamp (Ningqiang Light Sources Co. Ltd., Hengshui, China) was used as the radiation source. A quartz tube (7-mm i.d. × 14-cm length) was used as the atomizer, into which the volatile species and the hydrogen evolved from the reactor were swept by an argon flow. A laboratory-made gas-liquid separator (GLS) as in a previous work29 was used to isolate the gas from liquid. The argon flow was controlled by a rotameter. A Chromatographic Workstation (Nanjing Qianpu Software Co. Ltd, Nanjing, China) was used for data acquisition and data treatment. The separation of the four mercury species was performed on a model TH-2000 capillary electrophoresis system (Baoding Tianhui Institute of Separation Science, Baoding, China). The power supply was operated in voltage-controlled mode (30 kV maximum). A 50-cm-long × 100-µm-i.d. fused-silica capillary (Yongnian Optical Fiber Co. Ltd., Hebei, China) was used for CE separation. New capillaries were conditioned by flushing with methanol for 30 min, 0.1 mol L-1 HCl for 30 min, doubly deionized water (DDW) for 5 min, and 0.1 mol L-1 NaOH for 1 h under N2 gas pressurization. Between two separations, the capillary was flushed with DDW for 3 min and the run buffer solution for 3 min. The capillary was reconditioned daily by flushing with 0.1 mol L-1 NaOH and DDW. Sample solution was introduced into the separation capillary using a hydrodynamic method under an N2 gas pressure of 10 kPa for 3 s, corresponding to 150 nL of the sample solution. The operating parameters of the VSG-AFS and CE are given in Table 1. The CE-AFS interface used in this work is as described previously.29 Briefly, the interface was constructed on the basis of a cross design for introducing a sheath flow around the CE capillary and a Pt electrode, which provided an electrical connection for stable electrophoretic separations and allowed on-line volatile species generation. Because of the low flow from the capillary column (e1 µL min-1), a makeup solution of 1% v/v of Analytical Chemistry, Vol. 75, No. 7, April 1, 2003
1727
HCl was used to facilitate the CE effluent delivery. This makeup liquid also served as the required medium for subsequent volatile species generation. The mixture of CE effluent and HCl solution merged with 0.2% m/v KBH4 solution in a Teflon tee for the generation of volatile species. The generated volatile species was transported directly into the atomizer of AFS with an argon gas flow for AFS detection. Reagents. All of the reagents employed were at least of analytical grade. DDW was used throughout. Boric acid (Beijing Chemicals Co., Beijing, China) and methanol (Tianjin Taixing Chemicals Co., Tianjin, China) were used to prepare the electrolyte buffer solution. The pH of the buffer solution was adjusted with 0.1 mol L-1 NaOH (Tianjin Taixing Chemicals Co.). The buffer was filtered through a 0.45-µm filter prior to use. Aqueous cysteine hydrochloride (0.1% m/v) (Sigma) was used as chelating agent and extractant. The stock solutions of methylmercury, ethylmercury, and phenylmercury of 1000 mg L-1 (as Hg) were prepared by dissolving suitable amounts of methylmercury chloride (Alfar Aesar), ethylmercury chloride (Alfar Aesar), and phenylmercury chloride (Alfar Aesar) in methanol. Mercury(II) chloride was dissolved in DDW directly to obtain the stock solution of inorganic mercury of 1000 mg L-1. All stock solutions were protected against light and stored at 4 °C in the dark. Working standard solutions containing one or more mercury compounds were prepared by stepwise diluting the stock solutions in 0.1% m/v aqueous cysteine solution just before use. A 0.2% m/v KBH4 solution was prepared by dissolving KBH4 (Tianjin Institute of Chemical Reagents, Tianjin, China) in 0.2% m/v KOH (Beijing Chemicals Co.) solution as the reductant. HCl (1% v/v) (Tianjin Taixing Chemicals Co.) was used as the carrier of the CE effluent and also as the medium of subsequent volatile species generation. Samples. A certified reference material DORM-2 (dogfish muscle, NRCC) was used to check the accuracy of the present method. Lake water and river water samples were collected locally. Immediately after sampling, the samples were filtered through a 0.45-µm filter for speciation analysis. Sample Extraction. The extraction procedure was a variation of the Westo¨o¨ method, as described in detail elsewhere.22 Briefly, 0.2500 g of DORM-2 was extracted with a mixture of 10 mL of DDW, 5 mL of concentrated HCl, and 10 mL of toluene in an ultrasonic bath at room temperature for 25 min, followed by centrifugation at 3500 rpm for 10 min. After sonification and centrifugation, the organic phase in the supernatant was collected, then the residue was extracted with another 5 mL of toluene as described above. The two organic extracts were combined and back-extracted with 5 mL of 0.1% m/v L-cysteine solution. The final aqueous extract was subjected to CE separation. RESULTS AND DISCUSSION Factors Affecting CE Separation of Mercury Species. CE separation of the four selected mercury species was first optimized with their standard solutions. A concentration of 0.1% m/v L-cysteine was included in the standard solutions for the complete complex of the mercury species, which is also comparable with that in the final extract obtained by the Westo¨o¨ method described in the Experimental Section. The optimized parameters affecting the CE separation of the mercury species include the buffer pH, 1728
Analytical Chemistry, Vol. 75, No. 7, April 1, 2003
the concentration of methanol and boric acid in the buffer solution, and the applied voltage. The pH of the buffer solution used for CE separation is usually the most important operational parameter. A pH range of 8.010.0 for the buffer solution (100 mmol L-1 boric acid and 12% v/v methanol) was investigated, at which the complexes formed by cysteine with mercury species are charged negatively. It was found that the Hg-cysteine complex could be easily baseline-separated from the MeHg-, EtHg-, and PhHg-cysteine complexes in the pH range studied, while the separation time increased with increasing of the pH of buffer solution. However, the resolution of the MeHg-, EtHg-, and PhHg-cysteine complexes was much dependent on the pH of the buffer solution. At pH 1) was achieved only in the pH range of 9.0-9.2. Accordingly, a buffer pH of 9.1 was chosen for baseline separation of all the four mercury species. The effect of the concentration of boric acid on the separation was examined in the concentration range of 50-150 mmol L-1. Increase in the concentration of boric acid improved the resolution of the four mercury species but increased their migration times and electric current. A concentration of 100 mmol L-1 boric acid was used as a compromise to ensure a baseline separation in a minimal time and a reasonable electric current (∼50 µA). Methanol, as the most commonly used organic modifier in CE, can be used to improve the resolution by decreasing EOF and improving resolvability of analytes. The effect of the content of methanol in the buffer solution on the separation of the four mercury species was tested. It was shown that the migration time increased with increasing the content of methanol. The resolution for the three organomercury-cysteine complexes increased from 0 to 12% v/v methanol but decreased with further increase in the content of methanol from 12 to 25% v/v. Moreover, the distinctly bad peak shape was observed in the presence of 25% v/v methanol. Accordingly, 12% v/v of methanol was included in the buffer solution to improve the separation of the four mercury species. Hydrostatically Modified Electroosmotic Flow (HSMEOF). Although laminar flow is often avoided in most CE separation because it produces a parabolic rather than flat flow profile, it can be used for specific purpose in CE separation. Hydrodynamically modified electroosmotic flow (HDMEOF) opposite to the bulk flow within the CE capillary had been introduced by pressurizing the sample vial during high voltage in electrokinetic injection and CE separation to increase the injected quantities of analyte.30 Bulk flow control31 and laminar flow produced by the nebulizer of ICPMS32 had been used to improve the resolution of CE. In this work, the electroosmotic flow (EOF) was modified by applying a hydrostatical pressure opposite to the direction of EOF. (30) Magnuson, M. L.; Creed, J. T.; Brockhoff, C. A. J. Anal. At. Spectrom. 1997, 12, 689-695. (31) Kar, S.; Dasgupta, P. K. Microchem. J. 1999, 62, 128-137. (32) Kinzer, J. A.; Olesik, J. W.; Olesik, S. V. Anal. Chem. 1996, 68, 3250-3257.
Figure 1. Electropherograms of the mixture of 500 µg L-1 (as Hg) of each individual mercury species standards. Height difference between inlet and outlet of capillary: (a) 1.0, (b) 2.0, and (c) 2.0 cm. Voltage applied: (a) 15, (b) 15, and (c) 20 kV. All other conditions as shown in Table 1.
Here a flow opposite to EOF within the CE capillary was generated by placing the inlet of the separation capillary lower than the outlet. The flow induced in this way is referred to as hydrostatically modified electroosmotic flow (HSMEOF). Like HDMEOF, this HSMEOF can be utilized to improve separation efficiency and resolution of the mercury species as a result of an increase in the residence time of analytes in a separation capillary by decreasing EOF. Compared with HDMEOF,30 the generation of HSMEOF is much simpler and more cost-effective. Although the HSMEOF inversely directed to EOF improves the resolution, it also increases the separation time. To ensure good resolution and meanwhile to keep separation time as short as possible for the CE separation, the effects of the height difference between the inlet and outlet of separation capillary and the applied voltage on the CE separation of the mercury species in a 50-cm × 100-µm-i.d. fused-silica capillary were investigated. Shown in Figure 1 are three typical electrophorograms of a mixture of the four mercury species standards obtained from such investigation. It was found that a height difference of 2 cm between the inlet and outlet of the CE capillary in conjunction with an applied voltage of 15 kV gave a sufficient resolution for baseline
separation of the mercury species with good electrophoretic peaks while the separation time was kept as short as possible (15 min, see Figure 1b). Effects of Concentration and Flow Rate of the Make-Up Liquid. In the current CE-VSG-AFS, a makeup liquid should be employed not only to complete the electrophoresis circuit and to facilitate the transportation of the CE effluent, but also to provide a favorable medium for the ensuing volatile species generation. To this end, diluted HCl solution was employed as the makeup liquid. Figure 2 shows the effect of HCl concentration in the makeup solution at a flow rate of 1.8 mL min-1 on the signals of the four mercury species. For all the species studied, as HCl concentration increased, the signal first increased to a maximum and then decreased. The optimal concentrations of HCl were found to be 0.8, 1, 1, and 1% v/v for EtHg(I), MeHg(I), PhHg(I), and Hg(II), respectively. As a compromise, a 1% v/v of HCl solution was employed as the makeup liquid. Studies on the influence of the flow rate of the makeup solution show that the signal intensities of the four species increased gradually as the flow rate increased from 1.0 to 1.8 mL min-1 and Analytical Chemistry, Vol. 75, No. 7, April 1, 2003
1729
Figure 2. Effect of HCl concentration in the makeup solution on the fluorescent intensities of organomerurials of 500 µg L-1 each (as Hg) and inorganic mercury of 200 µg L-1 (as Hg). All other conditions as shown in Table 1.
then decreased with further increase of the flow rate from 1.8 to 4.0 mL min-1. These results could be understood in terms of VSG efficiency and dilution of the CE effluent. Below a flow rate of 1.8 mL min-1, VSG efficiency might dominate the signal intensity, whereas over the flow rate of 1.8 mL min-1, the dilution of the CE effluent probably controlled the signal. Accordingly, a flow rate of 1.8 mL min-1 was selected for the makeup solution. Effects of Concentration and Flow Rate of KBH4 Solution. Figure 3 shows the influence of KBH4 concentration at a flow rate of 2.5 mL min-1 on the signals of the four mercury species. As KBH4 concentration increased from 0.001 to 0.01% m/v, the signal intensities of the four mercury species slightly increased. Further increase in KBH4 concentration from 0.01 to 0.2% m/v resulted in a strong increase in the signal intensities of all three organic mercury species but a decline in the signal intensity of inorganic mercury. As a compromise, a 0.2% m/v of KBH4 solution was used as the reductant for volatile species generation, where the sensitivity of organomercurial species was improved by sacrificing that of inorganic mercury. The influence of the flow rate of 0.2% m/v KBH4 solution was investigated with the makeup solution of 1% v/v HCl at a flow rate of 1.8 mL min-1. The optimal flow rate of KBH4 solution
Figure 3. Influence of KBH4 concentration on fluorescent intensities of organomercurials of 500 µg L-1 each (as Hg) and inorganic mercury of 200 µg L-1 (as Hg). All other conditions as shown in Table 1.
ranged from 2.0 to 3.0 mL min-1 for all four mercury species studied. Therefore, a flow rate of 2.5 mL min-1 for 0.2% m/v of KBH4 solution was used for further work. Argon Flow Rate. Studies on the effect of the argon flow rate on the signal intensities of the four mercury species show that the optimal argon flow rate ranged from 250 to 320 mL min-1. Lower signal intensity below the flow rate of 250 mL min-1 likely resulted from incomplete release of the volatile species from the reaction mixture. At higher flow rates (>320 mL min-1), the dilution of the evolved volatile species and short residence time of the analyte species in the atomizer would be dominant, leading to the decrease of the signal intensities. Thus, an argon flow rate of 280 mL min-1 was used to maintain the maximum signal with good precision. Effect of Atomizer Temperature. As shown in Figure 4, the signal intensities of the four mercury species exhibited different dependence on the temperature of the atomizer. As the temperature of the atomizer increased, the signal intensity of inorganic mercury declined, whereas those of MeHg(I) and PhHg(I) rose in the temperature range from room temperature to 600 °C. For EtHg(I), the signal intensity increased with the temperature of the atomizer from room temperature to 400 °C and then decreased as the temperature of the atomizer further increased from 400 to 600 °C. Moreover, higher temperatures of
Table 2. Characteristic Performance Data of the CE-VSG-AFS for Mercury Speciation
precision (RSD, n ) 5), % migration time peak area peak height detection limits, µg L-1 calibration functiona corr coeff recoveryb, % river water 1 river water 2 lake water 1 lake water 2
EtHg(I)
MeHg(I)
PhHg(I)
Hg(II)
1.9 5.0 4.4 15.9 A ) 94.2C + 5.6 0.9959
2.1 6.3 6.1 16.5 A ) 103.3C + 13.2 0.9971
2.5 1.8 2.3 13.3 A ) 117.8C + 11.2 0.9918
1.9 4.2 5.2 6.8 A ) 579.1C + 26.6 0.9919
87.9 88.2 96.3 91.8
92.7 96.9 94.7 96.6
98.7 86.6 104 96.7
106 107 111 99.9
a A, peak area (mV‚s); C, concn (mg L-1). b Recovery for spiking with 200 µg L-1 (as Hg) of each individual organomercurial and with 100 µg L-1 (as Hg) of Hg(II).
1730 Analytical Chemistry, Vol. 75, No. 7, April 1, 2003
Table 3. Method Detection Limits for Hg Species Using Different CE-Hyphenated Techniques Hg(II) note CE-VSG-AFS (this work) CE-UV26 CE-ICPMS26 CE-DF-ICPMS26
conventional concentric Meinhard nebulizer conventional concentric Meinhard nebulizer
CE-VSG-ICPMS26 CE-ICPMS27 CE-ICPMS28 CE-ICPMS28
microconcentric nebulizer MCN-100 microconcentric nebulizer MCN-100 cross-flow nebulizer
MeHg(I)
EtHg(I)
µg L-1
pg
µg L-1
pg
µg L-1
pg
injected volume, nL
6.8 500 81
1.0 81 28
16.5 680 128
2.5 111 45
15.9 750 275
2.4 122 96
150 163 350
54
24
84
38
450
30 80
7
13.6
2.3
170
25.3
170
25 1 170 6.0 112
11 0.2 1.0 19
149
230 100
Figure 4. Influence of atomizer temperature on fluorescent intensities of organomercurials of 500 µg L-1 each (as Hg) and inorganic mercury of 200 µg L-1 (as Hg). All other conditions as shown in Table 1.
the atomizer would lower the baseline noise. Yin et al.33 also reported that the sensitivity of organo-mercury species determination could be improved by a thermolysis decomposition of volatile organomercurials. Accordingly, an atomizer temperature of 400 °C was chosen as a compromise for the four mercury species. Analytical Performance. The analytical characteristic data of the present CE-VSG-AFS for speciation of the four mercury species are summarized in Table 2. The precisions (RSD) of the migration time, the peak area and peak height for five replicate injections of a mixture of 500 µg L-1 (as Hg) of three individual organomercurial species and 200 µg L-1 (as Hg) of inorganic mercury were in the range of 1.9-2.5, 1.8-6.3, and 2.3-6.1%, respectively. The detection limits (3σ) of the four species based on peak area measurement ranged from 6.8 to 16.5 µg L-1. The recoveries of the spikes (200 µg L-1 (as Hg) of each organomercurial species, and 100 µg L-1 (as Hg) of inorganic mercury) of the natural water samples ranged from 86.6 to 111%. A comparison of the detection limits obtained by several CEhyphenated techniques for mercury speciation is made in Table (33) Yin, X.-F.; Frech, W.; Hoffmann, E.; Lu ¨ dke, C.; Skole, J. Fresenius’ J. Anal. Chem. 1998, 361, 761-766.
Figure 5. Electropherograms of (a) the extract of a certified reference material DORM-2 (dogfish muscle) obtained by Westo¨o¨ procedure; (b) the extract of DORM-2 spiked with organomercurials of 500 µg L-1 each (as Hg) and inorganic mercury of 200 µg L-1 (as Hg); (c) a mixture of standard aqueous solution with organomercurials of 500 µg L-1 each (as Hg) and inorganic mercury of 200 µg L-1 (as Hg). All other conditions as shown in Table 1.
3. It can be seen that the detection limits obtained with the present method are lower than those obtained using CE-ICPMS (quadrupole and double focusing) with a conventional nebulizer,26 similar to those obtained CE-ICPMS (quadrupole) with a microconcentric nebulizer28 and by CE-VSG-ICPMS (quadrupole).22 Additional advantages of the developed methodology for mercury speciation are its low instrumental and running costs and easy operation, as compared with CE-ICPMS. Validation of the Developed CE-VSG-AFS Technique for Mercury Speciation. The accuracy of the present method was demonstrated by analyzing a certified reference material (DORM2, dogfish muscle) with a certified methylmercury content. Figure 5 shows the electropherograms for the extract of DORM-2 Analytical Chemistry, Vol. 75, No. 7, April 1, 2003
1731
Table 4. Analytical Results for the Speciation of Methylmercury in a Certified Reference Material methylmercury content, µg g-1 (as Hg) sample
certified
determined (mean ( σ, n ) 5)
DORM-2 (dogfish muscle)
4.47 ( 0.32
4.53 ( 0.21
obtained as described in the Experimental Section (Figure 5a), the spiked extract with a mixture of four mercury species standards (Figure 5b), and a mixture of four mercury species standards (Figure 5c) under the conditions in Table 1. Only one peak in the extract of DORM-2 was observed, which was identified as methylmercury by comparing with Figure 5b,c. From Figure 5b,c, we can see the identical position of each individual mercury species, both in the spiked extract and in the standard mixture. The concentration of methylmercury in DORM-2 was quantified using a simple external calibration method based on peak area measurements. The result is given in Table 4. Good agreement was obtained between the certified value and the determined value by the developed method, indicating that the present CE-VSG-
1732 Analytical Chemistry, Vol. 75, No. 7, April 1, 2003
AFS technique can be used with confidence for the determination of methylmercury in biological samples. CONCLUSIONS The results in the present work have demonstrated the feasibility of the developed CE-VSG-AFS technique for mercury speciation in biological materials. Compared with CE-ICPMS techniques, this CE-VSG-AFS technique provides similar or lower detection limits along with the additional advantages of low instrumental and running costs and easy operation for mercury speciation. ACKNOWLEDGMENT This work was supported by the National Natural Science Foundation of China for Distinguished Young Scholars (no. 20025516) and the Research Foundation for the Excellent Young Teachers, State Education Ministry.
Received for review October 30, 2002. Accepted January 28, 2003. AC026272X
