Anal. Chem. 2009, 81, 3421–3428
Gold-Coated Silica as a Preconcentration Phase for the Determination of Total Dissolved Mercury in Natural Waters Using Atomic Fluorescence Spectrometry Kerstin Leopold,* Michael Foulkes, and Paul J. Worsfold Biogeochemistry and Environmental Analytical Chemistry Group, School of Earth, Ocean and Environmental Sciences, University of Plymouth, Portland Square, Plymouth PL4 8AA, United Kingdom A novel solid-phase preconcentration method is reported, using in-house gold-coated silica adsorbent packed in a microcolumn, for the determination of dissolved mercury in natural waters by atomic fluorescence spectrometry (AFS). The adsorbent was prepared by chemical reduction of a Au(III) solution with hydroxylamine in the presence of suspended silica particles. The resulting Au nanoparticles on the silica surface were highly efficient for adsorbing different mercury species from acidified waters without additional reagents. The acidified aqueous samples were passed over the microcolumn, either incorporated in a fully automated flow injection (FI) system directly coupled to the AFS or as part of a portable FI system for in situ preconcentration. After rinsing and drying of the column, Hg0 was released by heating and directed to the AFS cell for quantification. The method offers significant advantages because no reagents are needed for species conversion, preconcentration, sample storage, or desorption and therefore the risk of contamination is minimized and blank values are lowered. This results in a low detection limit of 180 pg L-1 using a sample volume of only 7 mL and good reproducibility, with relative standard deviations 90% in spiked river waters (spiked [Hg] ) 0, 1, 5, 10 ng L-1), and the experimental value for the certified reference material ORMS-4 (elevated mercury in river water) was 22.3 ( 2.6 ng Hg L-1 which was in good agreement with the certified value of 22.0 ( 1.6 ng Hg L-1 (recovery ) 101%). The method was successfully applied to seven different natural waters and wastewaters ([Hg] 0.5-4.6 ng L-1) from south west England. The monitoring of mercury in natural waters is of great importance due to its high toxicity and very high bioaccumulation factor (up to 106) in the food chain. The United States Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA) lists mercury and its compounds in third place on the “Priority List of Hazardous Substances”, and the European Water Framework Directive (2000/60/EG) * To whom correspondence should be addressed. Phone: +49 89 289 13764. Fax: +49 89 289 13186. E-mail:
[email protected]. 10.1021/ac802685s CCC: $40.75 2009 American Chemical Society Published on Web 03/31/2009
classifies mercury as 1 of 30 “precarious dangerous pollutants”. Consequently, the determination of mercury in waters is regulated by law. In natural waters there are three main forms of mercury found in the dissolved phase: inorganic mercury (Hg2+ and its complexes), organic mercury (predominantly monomethylmercury and dimethylmercury), and elemental mercury (Hg0). All of these compounds are highly mobile and are involved in (bio)transformation processes.1 The total dissolved mercury concentration in pristine natural waters varies from the low picogram per liter level for open ocean waters and 1-5 ng L-1 in freshwaters, to 10-20 ng L-1 in humic-rich lakes and particle-rich river waters.2-5 Consequently, for the determination of total mercury in natural waters, highly sensitive analytical detection techniques, such as atomic fluorescence spectrometry (AFS) or inductively coupled plasma mass spectrometry (ICPMS), combined with preconcentration and matrix separation techniques, e.g., cold vapor generation (CV) or solid-phase preconcentration, are required. Robust sampling and storage procedures are equally important, since the high volatility of elemental and organic Hg species and the high affinity of elemental and inorganic Hg species for surface adsorption can lead to analyte losses. Mercury contamination from sampling devices, storage vessels, and from stabilizing reagents used to prevent mercury evaporation or adsorption to the walls can also occur. In practice, the total contamination from the reagents and equipment often leads to relatively high blank values that negates any gain in sensitivity from the detection technique and hence degrades the detection limit. Therefore, in situ preconcentration is a sensible strategy to overcome the problems of contamination and analyte loss during sample transport and storage and, at the same time, potentially achieve lower detection limits. The main analytical advantages of adsorbing the analyte on a solid phase in the field rather than trying to stabilize liquid samples for transport and storage are obvious: (a) when mercury is bound to the solid phase its loss by evaporation is minimized, (b) a “solid sample” has less contact with vessel walls; hence, the (1) Ullrich, S. M.; Tanton, T. W.; Abdrashitova, S. A. Crit. Rev. Environ. Sci. Technol. 2001, 31, 241. (2) Cossa, D.; Martin, J.-M.; Sanjuan, J. Deep-Sea Res., Part II 1997, 44, 721. (3) Mason, R. P.; Rolfhus, K. R.; Fitzgerald, W. F. Mar. Chem. 1998, 61, 37. (4) Bloom, N. Can. J. Fish. Aquat. Sci. 1989, 46, 1131. (5) Meili, M. In Metal Ions in Biological Systems; Sigel, A., Sigel, H., Eds.; Marcel Dekker: New York, 1997; Vol. 34, pp 21-51.
Analytical Chemistry, Vol. 81, No. 9, May 1, 2009
3421
Figure 1. Flow injection manifold for (A) fully automated mercury determination coupled to AFS and (B) in situ preconcentration of mercury. (C) Photograph of the collector with gold-coated silica. Abbreviations: UPW, ultrapure water; SL, sample loop (7 mL); V, valve; NRV, nonreturn valve; CC, cooling coil (i.d. 1 mm, length 90 cm,
