ORIGINAL PAPER Thiourea catalysis of MeHg ligand exchange between natural dissolved organic matter and a thiol-functionalized resin: a novel method of matrix removal and MeHg preconcentration for ultratrace Hg speciation analysis in freshwaters Brian R. Vermillion & Robert J. M. Hudson Received: 8 December 2006 / Revised: 9 February 2007 / Accepted: 13 February 2007 / Published online: 7 March 2007 # Springer-Verlag 2007 Abstract Ultratrace analysis of dissolved MeHg in fresh- waters requires both dissociation of MeHg from strong ligands in the sample matrix and preconcentration for detection. Existing solid phase extraction methods generally do not efficiently adsorb MeHg from samples containing high concentrations of natural dissolved organic matter. We demonstrate here that the addition of 10–60 mM thiourea (TU) quantitatively releases MeHg from the dissolved matrix of freshwater samples by forming a more labile complex (MeHgTU + ) that quantitatively exchanges MeHg with thiol-functionalized resins at pH∼3.5 during column loading. The contents of these columns were efficiently eluted with acidified TU and MeHg was analyzed by Hg– TU complex ion chromatography with cold-vapor atomic fluorescence spectrometry detection. Routinely more than 90% of MeHg was recovered with good precision (average relative standard deviation of 6%) from natural waters— obtained from pools and saturated sediments of wetlands and from rivers—containing up to 68.7 mg C L -1 dissolved organic matter. With the preconcentration step, the method detection limit of 0.29 pg absolute or 0.007 ng L -1 in 40- mL samples is equivalent to that of the current state-of-the- art as practiced by skilled analysts. MeHg in 20–50-mL samples was completely trapped. On the basis of our knowledge of the chemistry of the process, breakthrough volume should depend on the concentrations of TU and H + . At a TU concentration of 12 mM breakthrough occurred between 50 and 100 mL, but overall adsorption efficiency was still 85% at 100 mL. Formation of artifactual MeHg is minimal; only about 0.7% of ambient MeHg is artifactual as estimated from samples spiked with 4 μgL -1 Hg II . Keywords Methylmercury . Thiourea . Solid-phase extraction . Porewater . Speciation analysis Introduction The myriad of factors governing the production, mobiliza- tion, and bioaccumulation of monomethylmercury (MeHg) in aquatic systems make predicting its ambient concen- trations from total Hg inputs a profound challenge for environmental scientists and natural resource managers alike. Both those seeking to develop such predictive capabilities and those mapping the distributions of MeHg are critically dependent on having efficient and cost- effective methods of analyzing this neurotoxin in environ- mental media. Each medium poses unique challenges for the analyst, as suggested by the fact that MeHg is generally present at below parts per trillion levels in freshwater systems yet can bioaccumulate via aquatic food webs to levels of parts per million in fish [1]. Analysis of MeHg in fish tissue requires extracting relatively high levels of the analyte from a matrix with an extremely high affinity for it, while analysis of MeHg in water requires detecting MeHg at ultratrace levels in the presence of an about tenfold excess of inorganic Hg [2]. In freshwaters, the difficulty of this ultratrace analysis is compounded by the presence of an interfering component of the sample matrix, dissolved organic matter (DOM). Anal Bioanal Chem (2007) 388:341–352 DOI 10.1007/s00216-007-1207-1 B. R. Vermillion (*) : R. J. M. Hudson Department of Natural Resources and Environmental Sciences, University of Illinois Urbana-Champaign, W512A Turner Hall, 1102 South Goodwin Ave, Urbana, IL 61801, USA e-mail: bvermill@uiuc.edu