B13F-01 INVITED
Redox Transformations of Mercury in Wetlands
Wetlands are valued for their high biodiversity and for their ecosystem services. However, we still have a poor understanding of their role in the redox transformation of contaminants such as mercury. We first propose a brief overview of past studies conducted on wetlands from different latitudes. In most instances, photochemical processes are determinant in the upper portion of the water column. At the sediment/water interface, evidence is currently supporting a significant contribution of bacterial communities, as promoters of Hg(II) reduction, particularly in the presence of anoxia. A multi-year study was recently conducted on Hg redox cycling in a fluvial wetland of the St. Lawrence River, where wetland restoration could have unintended consequences. In addition to photochemistry and bacterial reduction, Hg redox cycling was affected by epiphytes living on macrophytes, through adsorption/absorption processes. Redox studies such as this one have been historically seen as having implication for water/air flux studies, since Hg(0) is volatile. We here also discuss the potential bioavailability of Hg(0) towards bacteria. An emerging axis of our wetland research effort deals with beaver dams, which are in expansion and shown to produce high levels of methylHg
B13F-02 INVITED
Chemical Speciation and Mobilization of Mercury and Methyl Mercury at Sub-anoxic Conditions in Wetlands and Sediments with Special Emphasis on Organic Thiols
A correct description of the chemical speciation of Hg and MeHg is a prerequisite for understanding biogeochemical processes like retention-mobilization, methylation-demethylation and bioaccumulation. Binding affinity experiments and spectroscopic studies have clearly established that both inorganic Hg and methyl mercury (MeHg) are strongly complexed by reduced organic functional groups (thiols, RSH) in natural organic matter (NOM). Stability constants for Hg and MeHg associated to NOM macromolecules (humic substances) have been shown to be of a similar magnitude as Hg-thiol and MeHg-thiol associations in well-defined organic molecules like cysteine. As a consequence of this strong bonding, complexes with NOM will highly dominate the speciation of Hg and MeHg, both in the solid phase and in pore water, under oxidized conditions. This also holds for environments with concentrations of halides (Cl) in the mM range. Under anoxic conditions in soils and sediments, inorganic mono-, bi- and polysulfides will form complexes with Hg. At concentrations of dissolved inorganic sulfides exceeding 10 mikroM, Hg-sulfide complexes will dominate the speciation and the contribution from organic Hg-thiols will be less than 1%. However, at slightly reduced (sub-anoxic) conditions with inorganic sulfides in the 0.1-2 mikroM range, complexes with thiols, bi- and polysulfides will all contribute to the solubility of Hg. Crucial is whether elemental sulphur is formed or not. With elemental S present, Hg-polysulfides will outcompete all other species, including neutral Hg-sulfides available for methylating bacteria. Thus, as a basis for selection of relevant chemical speciation models, independent spectroscopic information about solid phases like S80(s) and HgS(s) (using sulfur XANES, Hg EXAFS etc) is crucial. The identity of some proposed Hg-sulfides and their stability constants (e.g. HOHgSH0 and HgS5OH- ), are still uncertain and their influence on the chemical speciation will be discussed. The importance of relevant models for the chemical speciation of MeHg is discussed in the context that changes in MeHg solubility in many studies erroneously are interpreted as a consequence of methylation-demethylation and transport processes.
B13F-03
New Mechanisms of Mercury Binding to Peat
Mercury can be immobilized in the aquatic environment by binding to peat, a solid form of natural organic matter. Binding mechanisms can vary in strength and reversibility, and therefore will control concentrations of bioreactive mercury, may explain rates of mercury methylation, and are important for designing approaches to improve water quality using natural wetlands or engineered phytoremediation schemes. In addition, strong binding between mercury and peat is likely to result in the fixation of mercury that ultimately resides in coal. The mechanisms by which aqueous mercury at low concentrations reacts with both dissolved and solid natural organic matter remain incompletely understood, despite recent efforts. We have identified three distinct binding mechanisms of divalent cationic mercury to solid peats from the Florida Everglades using EXAFS spectroscopic data (FAME beamline, European Synchrotron Radiation Facility (ESRF)) obtained on experimental samples as compared to relevant references including mercury-bearing solids and mercury bound to various organic molecules. The proportions of the three molecular configurations vary with Hg concentration, and two new configurations that involve sulfur ligands occur at Hg concentrations up to about 4000 ppm. The binding mechanism at the lowest experimental Hg concentration (60-80 ppm) elucidates published reports on the inhibition of metacinnabar formation in the presence of Hg-bearing solutions and dissolved natural organic matter, and also, the differences in extent of mercury methylation in distinct areas of the Florida Everglades.
B13F-04
The Role of Dissolved Organic Matter in Environmental Mercury Methylation by Sulfate- Reducing Bacteria
Methylmercury (MeHg) production in the environment is controlled by many factors, including biogeochemical controls on mercury bioavailability. Strong focus has been placed on the role of sulfide concentration in determining mercury speciation and cellular uptake. However, in natural waters, dissolved organic matter (DOM) is both ubiquitous and important in influencing mercury speciation and bioavailability. We revisit this issue with experimental results from methylation assays of sulfate-reducing bacteria with a pure culture, and through synchrotron-based characterization of mercury in simulated natural waters. Pure cultures of Desulfobulbus propionicus, a sulfate-reducing bacterium (SRB) capable of fermentative growth, were allowed to methylate a mercury isotopic tracer present at <100 ng/L and equilibrated with ~1 μM aqueous sulfide and ~40 mg/L DOM. Fermentative growth conditions allowed control over ambient sulfide concentrations to favor the predicted dominance of dissolved HgS0. The DOM used was a hydrophobic fraction isolated from Florida Everglades surface water. Results showed that 5-10% of the mercury isotopic tracer was methylated in both DOM-amended and DOM-free cultures. In DOM-amended cultures, 10-20% greater cell growth was observed, suggesting an apparent slower rate of methylation in DOM-free cultures and a beneficial contribution of DOM to cell growth. We note that as much as ~10% of ambient mercury associated with DOM was also methylated, possibly explaining the observed difference in methylation rates in terms of dilution of the total bioavailable mercury pool for DOM-amended cultures. Our observations suggest that, in some cases, DOM- partitioned mercury is subject to microbial methylation at environmentally significant rates. The nature of mercury- sulfide-DOM interaction was investigated in separate experiments. No precipitation was observed in solutions containing DOM and equimolar Hg2+ and aqueous sulfide at concentrations supersaturated with respect to metacinnabar. The equilibrated Hg-S-DOM solution was loaded on hydrophobic chromatography resin and subjected to 100 μM glutathione extraction. Spectra from DOM-partitioned mercury-sulfide exhibited high similarity to metacinnabar despite the absence of precipitation. Lower mercury concentrations and better resins are being tested in order to evaluate the role of DOM-associated nanoparticulate or colloidal metacinnabar in mercury bioavailability for methylation by SRB.
B13F-05
Bacterial Influence on the Solubility of Cinnabar and Metacinnabar at New Idria, CA
Mercury in the forms of cinnabar (α-HgS) and metacinnabar (β-HgS) is generally considered to be unreactive and of little environmental concern. To determine if this current belief is valid, a consortium of bacteria (including a Thiomonas intermedia-like bacterium) was taken from the acid mine drainage (AMD) pond at the New Idria Hg Mine, San Benito Co., CA, and inoculated into filter-sterilized AMD pond water (pH = 4) containing either ground cinnabar or metacinnabar crystals (<45 μm in diameter), with sampling occurring every 3 days. Under aerobic conditions the samples showed a pronounced increase in aqueous Hg concentration over background water concentrations (350(±20)ng/L). Bacteria growing on α-HgS increased the Hg concentration to 597(±10)μg/L, while bacteria growing on β-HgS resulted in levels of 8.0(±0.2)mg/L; both maxima occurred after 18 days of incubation. Experiments conducted with (1) α- HgS or β-HgS in the presence of killed bacteria (anaerobic), (2) α-HgS with pond water (abiotic), and (3) β-HgS with AMD pond water (abiotic) showed drops in aqueous Hg to below the detection limit (0.1ng/L) within 12 days. Anaerobic growth of the bacterial consortium showed a pattern similar to those of the abiotic water-HgS experiments, except that Hg levels dropped below detection limit within 6 days. These combined results suggest that HgS degradation by this bacterial consortium is an aerobic process. Killed bacteria incubated aerobically showed a slight increase in Hg levels over background water levels (<10x increase) then dropped below detection limit. This observation suggests that enzymes might be involved in the dissolution of HgS and were still viable for ~6 days after sterilization. In aerobic living incubations, the activities of different mercury and sulfide species were estimated using the thermodynamic modeling program Minteq with AMD pond water chemistry determined by ICP-MS and total mercury and total sulfide analyses. These calculations give an equilibrium solubility product for the dissolution of HgS up to 25 orders of magnitude higher than HgS under standard conditions. When compared to calculations by Paquette et al., 1997 and Benoit et al., 1999, the bacterial consortium at New Idria causes an increase in the pK for all reported reactions including H+, HS-, and H2S of 11-13 orders of magnitude. These results indicate that the biofilm consortium at the New Idria AMD pond has a profound effect on the solubility of cinnabar and metacinnabar, suggesting that a reassessment of HgS stability in aerobic AMD environments is needed.
B13F-06
The Influences of Dissolved Organic Matter on Mercury Biogeochemistry in Mesocosm Experiments in the Florida Everglades
Interactions of mercury (Hg) with dissolved organic matter (DOM) play important roles in controlling reactivity, bioavailability and transport of Hg in aquatic systems. Laboratory experiments using a variety of organic matter isolates from surface waters in the Florida Everglades indicate that DOM binds Hg very strongly and is the dominant ligand for Hg in the absence of sulfide. These experiments have also shown that the presence of DOM influences the geochemical behavior of cinnabar (HgS) through the stabilization of nanocolloidal HgS resulting in relatively high Hg concentrations under supersaturated conditions with respect to HgS, a common condition in waters containing measurable sulfide concentrations. In this paper, the results of in-situ mesocosm experiments designed to directly measure the effects of DOM -Hg interactions on Hg biogeochemistry will be described. In these experiments, mesocosms (wetland enclosures), located in the central Everglades region of Water Conservation Area 3A (WCA 3A15), were amended with isotopically enriched Hg (200Hg, 202Hg), sulfate (SO4=) and the hydrophobic organic acid (HPOA) fraction of DOM from a site (F1) in the eutrophic northern Everglades. The use of stable isotope spikes in these studies allowed us to examine the delivery of Hg to surface soils (which are the predominant zones of methylation); partitioning of Hg and MeHg among phases (which impacts bioavailability); net MeHg production; loss of Hg and MeHg through photodemethylation, reduction and volatization; and bioaccumulation. The F1 HPOA isolate, obtained using XAD resins, was more aromatic, had a greater specific ultra-violet absorbance and had previously been shown to be more reactive with Hg than the DOM present at the 3A15 site. The F1 HPOA isolate formed strong DOM-Hg complexes (KDOM') = 1023.2 L kg-1 at pH = 7.0 and I = 0.1) and effectively inhibited the precipitation of HgS in laboratory experiments. Select mesocosms were amended with either F1-HPOA or SO4= resulting in a range of concentrations for each constituent. For the DOM amended mesocosms, DOC concentrations increased from 50-100% and the overall SUVA increased from 2.9 to 3.7 L mg C-1 m-1 relative to control mesocosms, indicating that both the concentration and overall reactivity of the DOM in the amended mesocosms had been altered substantially. In these mesocosms, the concentrations of both ambient and isotopically enriched dissolved Hg increased significantly compared to controls. Greater concentrations of both dissolved ambient and labeled methylmercury were also observed in the DOM amended mesocosms indicating that the added DOM increased Hg bioavailabilty of both Hg pools for methylation. In addition, DOM shielded Hg and MeHg from photodemethylation and volatilization, however, it inhibited subsequent MeHg bioaccumulation. Overall, the addition of DOM resulted in increased concentrations of labeled methylmercury comparable to those measured in mesocosms amended with SO4= suggesting that DOM is an important constituent influencing the methylation of Hg. This effect is likely due to increased concentrations of dissolved Hg in the DOM amended mesocosms.
B13F-07
Mercury Emission to the Atmosphere from Experimental Manipulation of DOC and UVR in Mesoscale Field Chambers in a Freshwater Lake
Mesocosm experiments in an optically transparent lake allow the manipulation of both dissolved organic carbon (DOC) and incident ultraviolet radiation (UVR) in order to study mercury reduction and emission processes. In the absence of UVR and presence of visible light, mercury emission is very low (~0.3 ng/m2/h). When UVR is permitted in the mesocosm chambers, mercury emission increases, with emission rates ranging from 0.3 ng/m2/h to 2.5 ng/m2/h. At concentrations between 1.5 and 2.5 mg/L DOC, mercury emission does not appear to depend on either the concentration or the optical properties of the DOC. In particular the addition of 1.0 mg/L DOC from a nearby wetland to a photobleached mesocosm did not increase the emission of mercury. The similarities between mercury emission from highly photobleached 1.5 mg/L DOC and from terrestrially enriched 2.5 mg/L DOC suggest that the moieties responsible for mercury reduction are far in excess of that needed for mercury reduction. Using the measured flux rate of mercury from the water surface, we calculated a dissolved gaseous mercury (DGM) concentration that would need to be present to drive the emissive flux. The buildup of DGM was used to approximate a kinetic rate constant for the net mercury reduction in this system of approximately 0.17 h-1, which is consistent with existing published values. http://www.lehigh.edu/~scp2