Iron-Mediated Photochemical Decomposition of Methylmercury in an

Jul 21, 2010 - Sunlight-induced decomposition is the principal sink for methylmercury (CH3Hg+) in arctic Alaskan lakes and reduces its availability fo...
0 downloads 8 Views 183KB Size
Download PDF
Environ. Sci. Technol. 2010, 44, 6138–6143

Iron-Mediated Photochemical Decomposition of Methylmercury in an Arctic Alaskan Lake C H A D R . H A M M E R S C H M I D T * ,† A N D WILLIAM F. FITZGERALD‡ Department of Earth and Environmental Sciences, Wright State University, Dayton, Ohio 45435, and Department of Marine Sciences, University of Connecticut, Groton, Connecticut 06340

Received March 3, 2010. Revised manuscript received May 26, 2010. Accepted June 30, 2010.

Sunlight-induced decomposition is the principal sink for methylmercury (CH3Hg+) in arctic Alaskan lakes and reduces its availability for accumulation in aquatic food webs. However, the mechanistic chemistry of this process in natural waters is unknown. We examined experimentally the mechanism of photochemicalCH3Hg+ decompositioninfilter-sterilizedepilimnetic waters of Toolik Lake in arctic Alaska (68° 38′N, 149° 36′W), a region illuminated by sunlight almost continuously during the summer. Results from in situ incubation tests indicate that CH3Hg+ is not decomposed principally by either direct photolysis (i.e., no degradation in reagent-grade water) or primary photochemical reactions with dissolved organic material. The preeminent role of labile Fe and associated photochemically produced reactive oxygen species is implicated by tests that show1)additionsofFe(III)toreagent-gradewaterenhanceCH3Hg+ photodecomposition, 2) strong complexation of ambient Fe(III) with desferrioxamine B inhibits the reaction in lake water, and 3) experimental additions of organic molecules that scavenge hydroxyl radicals specifically among reactive oxygen species (dimethylsulfoxide and formic acid) inhibit CH3Hg+ degradation. Lake-water dilution and Fe(III) addition experiments indicate that Fe is not the limiting reactant for CH3Hg+ photodecomposition in Toolik Lake, which is consistent with prior results indicating that photon flux is a major control. These results demonstrate that CH3Hg+ is decomposed in natural surface water by oxidants, apparently hydroxyl radical, generated from the photo-Fenton reaction.

Introduction Monomethylmercury (CH3Hg+) is the highly toxic form of mercury that accumulates in biota to levels that may affect deleteriously the health of both wildlife (1) and humans (2). Aquatic ecosystems appear to be the most vulnerable to contamination with CH3Hg+ (3), which is bioconcentrated by primary producers (4) and biomagnified in food webs (1), resulting in piscivorous wildlife, including some fishes consumed by humans, often having the greatest levels. As a result, considerable research has been directed toward monitoring CH3Hg+ in wildlife tissues (e.g. ref 5) and understanding environmental factors and processes leading * Corresponding author phone: [email protected]. † Wright State University. ‡ University of Connecticut. 6138

9

(937)775-3457;

e-mail:

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 16, 2010

to its presence in natural waters (6). Comparatively little, however, is known about the nonbioaccumulative fate of CH3Hg+. With regard to residues in biota, environmental processes that reduce either the presence or bioavailability of CH3Hg+ in aquatic ecosystems are equally as important as those that create it. Indeed, bioaccumulation of CH3Hg+ by primary producers appears to be limited largely by the availability of the chemical itself (7, 8). CH3Hg+ can be demethylated biologically (9, 10) and decomposed photochemically (11), with the latter process known to occur among a variety of aquatic systems (12-16). In arctic Alaskan lakes, for example, photochemical processes destroy 66-88% of annual inputs of CH3Hg+ (17). While photodecomposition of CH3Hg+ appears to be ubiquitous, and a significant flux on both biogeochemical and toxicological scales, the mechanism by which sunlight changes the chemical structure of CH3Hg+ in natural waters is unknown. Laboratory investigations suggest that CH3Hg+ may be altered chemically by a variety of mechanisms. Aqueous CH3Hg+ can be decomposed directly by photolysis with ultraviolet (UV) radiation (λ ) 185-254 nm (18)) and indirectly by reaction with reactive oxygen species, namely hydroxyl radical (•OH 12, 19-22). Suda and colleagues (23) also have implicated singlet oxygen (1O2) as a potential agent of CH3Hg+ alteration, although other laboratory studies have found that neither 1O2 (20), superoxide (O2•- (21)), nor hydroperoxyl radical (HO2• (22)) alone affect the molecular structure of CH3Hg+ in dilute solutions. It has been proposed that attack of •OH dissociates the C-Hg bond of CH3Hg+ (i.e., demethylation (22)), although •OH also can abstract carbon-bound hydrogen atoms (24). Hydroxyl radical is known to be produced naturally by photolysis of NO3- (λ ) 313 nm 19, 25) and both Fenton (26) and photo-Fenton (27) reactions of Fe(II) and H2O2. The photo-Fenton reaction involves photochemical reduction of thermodynamically stable Fe(III) to Fe(II), which can react with naturally ubiquitous H2O2 to yield •OH, Fe(III), and OH-. Fe(III) represents Fe3+ species that are complexed with organic and inorganic ligands. We examined the mechanism by which CH3Hg+ is decomposed in an arctic Alaskan lake. The study included a suite of in situ incubation testssemploying natural water and ambient sunlightsthat utilize the laboratory investigations noted above and build upon our prior photochemical experiments in arctic lakes (13). These studies showed aqueous CH3Hg+ to be decomposed at environmentally significant rates in the Arctic and that the process was exclusively abiotic and mediated by sunlight. Here, we show that photodecomposition of CH3Hg+ in arctic lake water is 1) by an indirect pathway, 2) independent of nitrate, 3) dependent on labile iron, 4) inhibited by •OH scavenging molecules, and 5) mediated primarily by sunlight between 320 and 480 nm. Together, these results are consistent with •OH, produced by the photo-Fenton reaction, being the principal agent of CH3Hg+ decomposition in natural surface water.

Experimental Section Location. We investigated photochemical decomposition of CH3Hg+ in surface water of Toolik Lake in July of 2005, 2006, and 2007. The Arctic was selected for this study because it is illuminated by sunlight almost continuously during summer, and Toolik Lake was chosen because 1) it is adjacent to the Long-Term Ecological Research site at Toolik Field Station (68° 38′ N, 149° 36′ W; Figure 1 in ref 17), and 2) 10.1021/es1006934

 2010 American Chemical Society

Published on Web 07/21/2010

photochemical reactions are known to be a major loss mechanism for CH3Hg+ (13). Although Toolik is a relatively large (1.5 × 106 m2) and deep (mean, 11 m) lake, its waters are representative biogeochemically of many other arctic Alaskan lakes (28). Incubation Experiments. The mechanism of photochemical CH3Hg+ decomposition was investigated with a series of bottle incubation tests. Most experiments were conducted with surface water from Toolik Lake (ambient CH3Hg+ ) 0.26 pM), although comparisons were made with other natural waters, including tundra pore fluids from the Toolik watershed (0.15 pM CH3Hg+) and surface waters from Green Cabin (0.17 pM) and Island (0.08 pM) Lakes, both of which are about 20 km distant from Toolik. Water samples were collected, manipulated experimentally, and analyzed with trace-metal clean procedures (29). To exclude biological processes, all tests were conducted with 1) water passed through 0.2-µm capsule filters inside a HEPA-filtered laminar flow hood and 2) bottles that were presumably sterile after soaking in g0.6 M HCl for at least 12 h. Experimental solutions were amended with CH3Hg+ (as CH3HgCl) to initial nominal concentrations that varied between 5.6 and 19.5 pM among individual tests. CH3Hg+ additions increased concentrations about 100× over ambient but were useful for optimizing analytical precision and yield photodecomposition rate constants that are comparable to those of ambient CH3Hg+ (16). In addition to CH3Hg+, chemical additions were made to water samples in many experiments; chemical amendments were allowed to equilibrate for >12 h in the dark at about 20 °C prior to incubation, a period sufficient for added CH3Hg+ to equilibrate with natural ligands (30). All sample incubation experiments were conducted with FEP Teflon bottles, which are optically transparent to photosynthetically active radiation (PAR, λ ) 400-700 nm) yet absorb 18% of incident UV-A (λ ) 320-400 nm) and 34% of UV-B (λ ) 280-320 nm (31)). Samples were incubated at the surface of Toolik Lake (0-0.1 m water depth) under in situ sunlight and temperature conditions. Temperature and PAR were monitored continuously (32). Replicate independent samples of each water type/treatment were analyzed for CH3Hg+ at the beginning and end of each incubation test, which ranged between 2 and 5 d among experiments. As observed in prior studies (11, 13), CH3Hg+ was not decomposed appreciably in filter-sterilized natural waters incubated in the dark. Mercury Analysis. The CH3Hg+ content of water samples was quantified after direct ethylation with sodium tetraethylborate (NaTEB; Strem Chemical) as described previously (13). CH3Hg+ was determined by flow-injection gas chromatographic cold vapor atomic spectrometry (33), after calibration with aliquots of an aqueous CH3Hg+ solution that was standardized against Hg0. Direct ethylation resulted in quantitative recovery of CH3Hg+ from filtered lake water. Recovery of known CH3Hg+ additions from filtered lake water averaged 100 ( 6% (n ) 104). Moreover, there was no difference in CH3Hg+ concentration for unamended natural waters ([0.06-0.51 pM] paired t-test, p ) 0.26, n ) 7) and experimentally manipulated samples ([5.1-11.4 pM] paired t-test; p ) 0.28, n ) 6 Toolik Lake water) that were analyzed by both direct ethylation and after treatment with 0.16 M HNO3 for 4 h at 65 °C prior to analysis (13). Thus, no sample pretreatment was required to recover quantitatively CH3Hg+ from aqueous matrixes examined in this study. The estimated detection limit for CH3Hg+ in a 200-mL sample was about 0.02 pM.

Results and Discussion

Comparison of Natural Waters. Rates of CH3Hg+ photodecomposition in multiple surface waters were compared to provide context for tests conducted with Toolik Lake water only. These waters included tundra pore fluids from the Toolik

FIGURE 1. Rate constant (mean (1 standard deviation, n ) 3) of photochemical decomposition of CH3Hg+ versus fraction of water (0.2-µm filtered) as either Toolik Lake surface (380 µM DOC) or tundra pore water (970 µM DOC). Natural waters were diluted with reagent-grade water, amended with CH3Hg+ to an initial level of 11 pM, and incubated for 2 d at the surface of Toolik Lake (PAR ) 57 E m-2 d-1; T ) 13.2 ( 0.4 °C). watershed and surface water from Toolik, Green Cabin, and Island Lakes. Measured rate constants of CH3Hg+ photodecomposition were similar statistically among all four water sources, although dissolved organic carbon (DOC) and H+ activity were increased substantially in the pore fluid (Table S1). This suggests that pH and DOC, at least within the range of these samples (pH ) 5.3-7.6; DOC ) 380-830), do not have a major effect on CH3Hg+ decomposition and that mechanistic results obtained from Toolik Lake are applicable to other nearby lakes. Direct Photolysis vs Indirect Pathway. We examined whether CH3Hg+ was decomposed through either direct photolysis or an indirect pathway by diluting both Toolik Lake surface water and tundra pore fluids with reagent-grade water (nominal resistivity )18 MΩ-cm, DOC ) 35 µM). The rationale for this approach was that if direct photolysis were the principal mechanism, then the rate of CH3Hg+ decomposition should be largely independent of solution composition. In contrast, if a component of natural water were limiting an indirect pathway, then the rate should be reduced in more dilute solution. The rate of CH3Hg+ photodecomposition varied as a function of sample dilution for both Toolik Lake surface and tundra pore water (Figure 1). Determined photodecomposition rate constants of CH3Hg+ in natural waters diluted 10and 100-fold with reagent-grade water were similar to those in undiluted samples; however, further dilution resulted in substantially lower rates. Rate constants for natural waters diluted 104-fold (Toolik, k ) 0.000 ( 0.014 d-1; Tundra, k ) 0.001 ( 0.012 d-1) did not differ from zero and were comparable to that of CH3Hg+ photodecomposition in reagent-grade water only (k ) 0.000 ( 0.016 d-1). This indicates that, under incubation conditions imposed in this test, direct photolysis is not the principal mechanism of CH3Hg+ photodecomposition and implicates an indirect pathway that involves photochemical transformation of a constituent of natural water. This result might be anticipated based on laboratory results of Inoko (18), who observed photolysis of aqueous CH3Hg+ exposed to 185-254 nm radiation, solar wavelengths that do not strike Earth’s surface. Potential Roles of Nitrate and Iron. Laboratory studies indicate that reaction with •OH may be a mechanism by which CH3Hg+ is decomposed in natural waters (12, 19-22). As noted, •OH is known to be produced naturally by photolysis of NO3- (19) and both Fenton (26) and photoFenton (27) reactions of Fe(II) and H2O2. Toolik Lake surface VOL. 44, NO. 16, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6139

TABLE 1. Rate Constants of CH3Hg+ Photodecomposition (±1 Standard Deviation, n = 3) in Reagent-Grade and Toolik Lake Surface Waters Amended with Potential Reactants and Desferrioxamine B (DFB)a water

chemical treatment

reagent-grade water only Fe2(SO4)3 DFB Fe2(SO4)3 + DFB Toolik surface water only NO3Fe2(SO4)3 FeCl3 DFB Fe2(SO4)3 + DFB

CH3Hg+ photodecomposition (d-1) 0.000 ( 0.016 0.280 ( 0.009 0.004 ( 0.022 0.005 ( 0.012 0.083 ( 0.021 0.096 ( 0.032 0.075 ( 0.005 0.078 ( 0.008 0.008 ( 0.018 0.021 ( 0.013

a

Rate constants determined from a 3.5-d incubation (initial CH3Hg+ ) 19 pM) at the surface of Toolik Lake (PAR ) 20 E m-2 d-1; T ) 12.0 ( 0.5 °C). Fe2(SO4)3 and FeCl3 were added at 40 µM, NO3- at 100 µM, and DFB at 84 µM in all tests.

water examined in this study had about 0.5 µM NO3- and ∼5 µM total Fe. To examine initially the potential roles of these agents in mediating CH3Hg+ photodecomposition by producing •OH, we increased the level of NO3- by 200-fold (addition of 100 µM NaNO3; Fisher ACS), and, to different samples, increased Fe about 10-fold (addition of 40 µM Fe(III) as either Fe2(SO4)3 or FeCl3; Fisher ACS). In addition, desferrioxamine B (DFB, deferoxamine; Sigma) was added in excess molar quantities (84 µM) to samples of both natural and Fe(III)-amended waters, but not enough to either overwhelm levels of DOC in Toolik water (380 µM) or cause quenching of reactive oxygen species (34). DFB is a natural siderophore that is unreactive photochemically because Fe(III) is complexed with hydroxymate binding groups (35), although complexed Fe(III) can be reduced by O2•- at very slow rates (34). Hence, complexation of Fe(III) by DFB renders it inactive photochemically, and, by extension, quantitative titration of Fe(III) by DFB would be expected to attenuate photochemical decomposition of CH3Hg+, if the photoFenton reaction were important. Tests with reagent-grade and Toolik Lake water indicate that Fe(III) is a photochemically reactive agent leading to CH3Hg+ decomposition in surface waters (Table 1). Addition of Fe(III) to reagent-grade water enhanced greatly CH3Hg+ photodecomposition relative to water only, which had no significant loss, similar to samples of reagent-grade water amended with DFB. However, in tests with reagent-grade water, photochemical decomposition of CH3Hg+ by added Fe(III) was inhibited entirely in samples also amended with DFB. Hence, CH3Hg+ photodecomposition can be influenced by availability and reactivity of Fe(III) in reagent-grade water exposed to sunlight; this applies to natural waters as well. Addition of DFB inhibited photochemical decomposition of CH3Hg+ in samples of both Toolik Lake surface water and Toolik water with added Fe(III). Amendment of Toolik Lake water with Fe(III) did not promote photodecomposition, in contrast to samples of reagent-grade water. This suggests that Fe(III) is not a limiting reactant in Toolik Lake water (5 µM total Fe), as it appears to be in reagent-grade water. This assessment is consistent with results from dilution experiments with surface waters that show CH3Hg+ photodecomposition is comparable between 100-fold diluted and undiluted samples (Figure 1); that is, Fe(III) or another component of natural water may not be the limiting reactant, at least to a certain extent. Importantly, these results also are consistent with multiple investigations that indicate CH3Hg+ decomposition to be limited largely by photon flux in natural 6140

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 16, 2010

FIGURE 2. Photochemical decomposition of CH3Hg+ in 0.2-µm filtered Toolik Lake water as a function of added desferrioxamine B (DFB). Samples were amended with CH3Hg+ to an initial nominal concentration of 5.6 pM and incubated for 3 d under both dark (bottles wrapped with Al foil) and ambient sunlight (mean PAR ) 35 ( 5 E m-2 d-1) conditions at the surface (0-10 cm) of Toolik Lake (17.3 ( 0.2 °C). The dashed line represents mean rate constant for CH3Hg+ decomposition in Toolik Lake water exposed to sunlight without added DFB (k ) 0.10 ( 0.02 d-1) . waters (11, 13, 14). The lower rate for CH3Hg+ photodecomposition in lake water only (k ) 0.083 ( 0.021 d-1) as compared to that observed in reagent-grade water amended with Fe(III) (k ) 0.280 ( 0.009 d-1) may be attributed to quenching of generated radicals by DOC and other naturally occurring scavengers in lake water. In contrast to Fe, NO3- does not appear to be an important factor affecting CH3Hg+ decomposition in Toolik Lake. While photolysis of NO3- can produce •OH and thereby decompose CH3Hg+ by an indirect pathway (19), addition of NO3- to Toolik Lake surface water did not affect CH3Hg+ photodecomposition (Table 1). As with Fe(III) additions to Toolik Lake water, which also had no effect, the absence of an increase in CH3Hg+ photodecomposition (i.e., increased •OH production) may result from ambient NO3- being available in excess relative to photon flux. However, the potential significance of NO3- is negated by observation that addition of DFB, which is unreactive toward NO3-, inhibited photochemical decomposition of CH3Hg+. Nitrate-associated decomposition of CH3Hg+ also is not consistent with results from the following tests. To further investigate the potential role of Fe(III) in producing •OH and mediating CH3Hg+ photodecomposition, we experimentally titrated ambient Fe(III) in Toolik Lake surface water with DFB (Figure 2). Samples of Toolik Lake surface water were amended with 5.6 pM CH3Hg+ and from 10-8 to 10-4 M DFB (nominal), a range that encompasses ambient Fe in Toolik Lake (5 × 10-6 M). CH3Hg+ photodecomposition was unaffected when DFB was added at levels less than that of ambient total Fe (i.e., rate constant comparable to that in unamended water, dashed line in Figure 2). Photodecomposition was decreased in samples where all of the ambient Fe was titrated by the ligand. The inflection point of the Fe titration curve in Figure 2 (∼10-5 M) is in reasonable agreement with measured total Fe, most of which would be Fe(III) in oxic waters of Toolik Lake. CH3Hg+ was not degraded in similar samples incubated in the dark. These results suggest that photochemical reduction of Fe(III) has an important role in affecting CH3Hg+ decomposition in Toolik Lake, and this may be caused by production of •OH, although it is not the only oxidant produced by the photo-Fenton reaction in pH-neutral waters (36).

Decomposition by Hydroxyl Radical. •OH can oxidize a broad range of organic molecules (37), including CH3Hg+ (12, 19-22). We investigated whether •OH was the agent responsible for CH3Hg+ decomposition in Toolik Lake by adding well-known •OH-scavenging compounds: Both formic acid (38) and dimethylsulfoxide (DMSO (39)) scavenge •OH exclusively, with neither affecting 1O2 appreciably. Samples of Toolik Lake surface water were amended with CH3Hg+ to 11 pM (1.1 × 10-11 M) and with either formic acid or DMSO at initial concentrations from 10-8 to 10-3 M. Our hypothesis was that, if •OH were the reactant decomposing CH3Hg+, then these scavengers should inhibit CH3Hg+ degradation until they were titrated by •OH to levels comparable to that of CH3Hg+. DMSO and formic acid blocked photochemical decomposition of CH3Hg+ (Figure S1). No significant decomposition of CH3Hg+ was observed in samples amended with either DMSO or formic acid at initial levels >10-4 M. In samples that contained lesser amounts of these •OH scavengers, photodecomposition of CH3Hg+ proceeded at rates comparable to those in unamended water (dashed line in Figure S1), although initial levels of the scavengers were in great excess compared to CH3Hg+. In samples containing from 10-8 to 10-6 M of either DMSO or formic acid, we infer that •OH titrated these molecules to levels comparable to that of CH3Hg+ prior to chemically altering the organomercurial. These observations reaffirm that CH3Hg+ photodecomposition is by an indirect pathway and point to •OH as the causative agent, which is consistent with the observed role of Fe(III). Transition metals other than Fe2+ also can generate •OH through Fenton-like reactions with H2O2 (Cu1+, Co2+, Cr2+, Cr5+, V4+ 40, 41). We observed no CH3Hg+ decomposition in water samples incubated in the dark. This implies that neither Fe nor any of these other transition metals were in their reduced oxidation states (shown above) in significant concentrations in the dark and is consistent with thermodynamic estimates of the stability of each reduced metal species in oxic lake water, with the exception of Co2+, which is the most reactive and soluble form of Co under our test conditions (42). Accordingly, for any of these transition metals (other than Co2+) to be active generators of reactive oxygen species through a Fenton-like reaction, the oxidized forms of these ions (Fe3+, Cu2+, Cr3+, Cr6+, V5+) must be reduced, most likely by a photochemical process in oxic water. We found that metal ion complexation by DFB inhibits photodecomposition of CH3Hg+, and this is attributed to DFB rendering the oxidized metal precursor (e.g., Fe3+) unreactive to photochemical reduction. However, and in addition to complexing Fe3+ (log K ) 32), DFB has a high affinity for other trivalent metals (43), suggesting that the observed effect of DFB on CH3Hg+ photodecomposition could result from complexation of Cr3+. If such were the case, then the combined concentration and affinity of Cr3+ for DFB would have to be greater than those of Fe3+ to outcompete Fe3+ for DFB. Results in Figure 2, as noted, indicate that the active metal is about 10 µM, which is in good agreement with measured total Fe (5 µM). The affinity of most trivalent metals for DFB is less than Fe3+ (43), and while the stability of the Cr3+-DFB complex is unknown, typical levels of Cr(III) in oxic lake water are about 1 nM (44-46). Hence, the chemical behavior of DFB and relative concentrations of trivalent metals suggest that Fe3+ was the active metal in our tests. The implied role of Fe in photodecomposition of CH3Hg+ also is consistent with tests of solution pH and radiation wavelength. Effect of pH. Solution pH can influence the speciation and associated photoreactivity of Fe in producing •OH (27). The potential influence of H+ activity on CH3Hg+ photodecomposition was examined by adjusting the pH of Toolik Lake water (by addition of HCl or KOH) to 3.4, 4.0, 5.1, 7.6

TABLE 2. Photodecomposition Rate Constants (±1 Standard Deviation, n = 3) of CH3Hg+ in Toolik Lake Water inside FEP Teflon Bottles Treated with Radiation-Filtering Filmsa

treatmentb

radiation wavelength exposure of sample (nm)

CH3Hg+ photodecomposition (d-1)

no filter transparent mylar film yellow film

280-800 320-800 480-800

0.152 ( 0.023 0.135 ( 0.023 0.030 ( 0.016

a Samples (initial CH3Hg+ ) 11 pM) were incubated for 4 d at the surface of Toolik Lake (PAR ) 33 ( 10 E m-2 d-1; T ) 16.9 ( 0.3 °C). b Both mylar (American Micro Industries, Chambersburg, PA) and yellow films (AP6300, Star Light & Magic Inc.,www.starlight.com/gel.html) had a thickness of 0.05 mm.

(ambient), and 8.9. Water samples were amended with CH3Hg+ to a nominal concentration of 9.5 pM and incubated for 2 d at the surface of Toolik Lake (PAR ) 26 E m-2 d-1; T ) 12.6 ( 0.5 °C). We observed that H+ activity had no significant effect on rate constants ((1 SD, d-1) of CH3Hg+ photodecomposition at pH values of 7.6 or less: pH ) 3.4, k ) 0.074 ( 0.006; pH ) 4.0, k ) 0.089 ( 0.012; pH ) 5.1, k ) 0.091 ( 0.012; pH ) 7.6, k ) 0.087 ( 0.008. In contrast, photochemical decomposition of CH3Hg+ was less in water having a pH of 8.9 (k ) 0.051 ( 0.003 d-1). These observations are consistent with results 1) from our comparison of natural waters having a 100-fold difference in H+ activity, between pH 5.3 and 7.6 (Table S1), and 2) of Zepp and colleagues (27), who observed photo-Fenton production of •OH to be reduced at alkaline pH. Radiation Wavelength. Lastly, we attempted to identify more narrowly the spectrum of radiation responsible for CH3Hg+ photodecomposition by applying filters to FEP bottles (Table 2). As noted, prior studies have found CH3Hg+ to be decomposed in natural waters exposed to 280-800 nm wavelengths, and the rate of the reaction was pseudo-firstorder with respect to intensity of PAR (11, 13, 14), although the preeminent role of UV radiation has been suggested recently (16). To separate FEP bottles, we applied both a mylar film (320 nm cutoff; >85% transmittance of longer wavelengths) to exclude UV-B radiation and a yellow film (480 nm cutoff; 82% transmittance of longer wavelengths) to block UV and violet and blue visible light. We observed that exclusion of UV-B radiation with the mylar film had no significant effect on CH3Hg+ photodecomposition and that the reaction was inhibited, but not entirely, by the yellow film (Table 2). Combined, results from these two treatments indicate that CH3Hg+ degradation in Toolik Lake water is initiated primarily by radiation between 320 and 480 nm, although longer and shorter wavelengths may be important. The apparent insignificant influence of UV-B radiation on CH3Hg+ decomposition may be due, in part, to the Teflon bottles absorbing ∼34% of UV-B (31) and the mylar film attenuating 15% of longer wavelengths. Lehnherr and St. Louis (16) observed that UV-B decomposed 22-35% of aqueous CH3Hg+ in Lake 979 of the Experimental Lakes Area. However, they noted that the apparent role of UV-B radiation may be attributed partly to artifacts associated with the mylar film, which blocked about 25% of shortwave UV-A in their tests. Hence, the role of UV-B radiation in CH3Hg+ decomposition remains unclear, but results of both studies suggest longer wavelengths are relatively more important in natural surface waters. The range of radiation wavelengths initiating CH3Hg+ decomposition is consistent with the photo-Fenton reaction. As noted, •OH production by the photo-Fenton reaction requires Fe(II). Both inorganic and organic complexes of Fe3+ VOL. 44, NO. 16, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6141

can be reduced photochemically to yield Fe(II) by wavelengths between 250 and 450 nm (27, 47, 48). This spectrum is within the “active” radiation range determined in the current study (principally, 320-480 nm) and consistent with results from multiple field experiments that show decomposition is either related to the intensity of PAR or initiated principally by wavelengths in the PAR or UV-A spectra (11, 13, 14, 16). Results of this investigation, however, do not support •OH generation by NO3- photolysis as being a major sink for CH3Hg+ in Toolik Lake. NO3- is photolyzed by radiation in the UV-B spectrum (19, 25), wavelengths that had no significant effect on CH3Hg+ decomposition in our tests, although incident UV-B flux was attenuated ∼34% by the FEP bottle. The relatively insignificant role of nitrate photolysis also is supported by our other tests with solution pH variation and added DFB (Table 1). Environmental Implications. As noted, photochemical degradation of CH3Hg+ is pronounced in oligotrophic arctic lakes of the Alaskan tundra (13, 17). Indeed, we estimated previously that 66-88% of CH3Hg+ inputs could be decomposed photochemically over the approximate 100-d, ice-free period. We established that the photolytic decomposition of CH3Hg+ was mediated primarily by PAR except at the lake surface, where decomposition was enhanced due to the additional influence of UV radiation. Additionally, we suggested that greater light attenuation associated with naturally productive water bodiessthose with higher amounts of organic matter and especially those affected by eutrophicationsmay experience less photodecomposition and greater availability of CH3Hg+ for bioaccumulation. This current work reinforces these hypotheses while adding mechanistic and biogeochemically intriguing information related to the potential role of Fe in the production of reactive oxygen species (i.e., •OH) that can decompose CH3Hg+ in aqueous systems. Given that photochemically mediated degradation of CH3Hg+ is intensified in iron replete, low-organic waters such as those found in many arctic lakes, the efficacy of such photolysis reactions would be expected to be reduced in iron-limited lacustrine systems and euphotic zone of the open ocean, especially oligotrophic regions where much, if not all, of the Fe is complexed by siderophores or other strong ligands (49). Also, we posit that incident radiation is the principal constraint on CH3Hg+ photodecomposition in systems with sufficient labile Fe for the photo-Fenton reaction. This could explain why photon flux-normalized rates of CH3Hg+ photodecomposition in surface water are consistent among multiple freshwater systems that are disparate physicochemically (13, 16), but none likely limited by Fe availability. We suggest that significant variability in horizontal and vertical distributions of CH3Hg+ should be evident in fresh and marine waters that are affected, for example, by seasonal changes in incident radiation and variations of DOC and primary production that attenuate the flux of solar radiation. Such differences in Fe and sunlight availability should influence CH3Hg+ bioaccumulation.

Acknowledgments We thank Andrew Rose, Carl Lamborg, and three anonymous reviewers for providing helpful comments on earlier versions of the manuscript. Prentiss Balcom and Kathryn Bluske assisted with sample preparation. This study was supported by the U.S. National Science Foundation-Office of Polar Programs (0425562).

Supporting Information Available One table and one figure. This material is available free of charge via the Internet at http://pubs.acs.org. 6142

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 16, 2010

Literature Cited (1) Wiener, J. G.; Krabbenhoft, D. P.; Heinz, G. H.; Scheuhammer, A. M. Ecotoxicology of mercury. In Handbook of Ecotoxicology; Hoffman, D. J., Rattner, B. A., Burton, G. A., Jr.; Cairns, J., Jr., Eds.; Lewis Publishers: Boca Raton, FL, 2003; pp 409-463. (2) Mergler, D.; Anderson, H. A.; Chan, L. H. M.; Mahaffey, K. R.; Murray, M.; Sakamoto, M.; Stern, A. H. Methylmercury exposure and health effects in humans: A worldwide concern. Ambio 2007, 36, 3–11. (3) Scheuhammer, A. M.; Meyer, M. W.; Sandheinrich, M. B.; Murray, M. W. Effects of environmental methylmercury on the health of wild birds, mammals, and fish. Ambio 2007, 36, 12–18. (4) Mason, R. P.; Reinfelder, J. R.; Morel, F. M. M. Uptake, toxicity, and trophic transfer of mercury in a coastal diatom. Environ. Sci. Technol. 1996, 30, 1835–1845. (5) U.S. Environmental Protection Agency. The National Survey of Mercury Concentration in Fish Data Base Summary 1990-1995; EPA publication 823-R-99-014; U.S. EPA: Washington, DC, 1999. www.epa.gov/waterscience/fish/mercurydata.html (accessed December 9, 2009). (6) Benoit, J. M.; Gilmour, C. C.; Heyes, A.; Mason, R. P.; Miller, C. L. Geochemical and biological controls over methylmercury production and degradation in aquatic systems. In Biogeochemistry of Environmental Important Trace Metals; Cai, Y., Braids, O. C., Eds.; American Chemical Society: Washington, DC, 2003; pp 262-297. (7) Pickhardt, P. C.; Folt, C. L.; Chen, C. Y.; Klaue, B.; Blum, J. D. Algal blooms reduce the uptake of toxic methylmercury in freshwater food webs. Proc. Natl. Acad. Sci. 2002, 99, 4419– 4423. (8) Orihel, D. M.; Paterson, M. J.; Blanchfield, P. J.; Bodaly, R. A.; Hintelmann, H. Experimental evidence of a linear relationship between inorganic mercury loading and methylmercury accumulation by aquatic biota. Environ. Sci. Technol. 2007, 41, 4952–4958. (9) Matilainen, T.; Verta, M. Mercury methylation and demethylation in aerobic surface waters. Can. J. Fish. Aquat. Sci. 1995, 52, 1597–1608. (10) Schaefer, J. K.; Yagi, J.; Reinfelder, J. R.; Cardona, T.; Ellickson, K. M.; Tel-Or, S.; Barkay, T. Role of the bacterial organomercury lyase (MerB) in controlling methylmercury accumulation in mercury-contaminated natural waters. Environ. Sci. Technol. 2004, 38, 4304–4311. (11) Sellers, P.; Kelly, C. A.; Rudd, J. W. M.; MacHutchon, A. R. Photodegradation of methylmercury in lakes. Nature 1996, 380, 694–697. (12) Gårdfeldt, K.; Sommar, J.; Stro¨mberg, D.; Feng, X. Oxidation of atomic mercury by hydroxyl radicals and photoinduced decomposition of methylmercury in the aqueous phase. Atmos. Environ. 2001, 35, 3039–3047. (13) Hammerschmidt, C. R.; Fitzgerald, W. F. Photodecomposition of methylmercury in an arctic Alaskan lake. Environ. Sci. Technol. 2006, 40, 1212–1216. (14) Hines, N. A.; Brezonik, P. L. Mercury inputs and outputs at a small lake in northern Minnesota. Biogeochemistry 2007, 84, 265–284. (15) Whalin, L.; Kim, E.-H.; Mason, R. Factors influencing the oxidation, reduction, methylation and demethylation of mercury species in coastal waters. Mar. Chem. 2007, 107, 278–294. (16) Lehnherr, I.; St. Louis, V. L. Importance of ultraviolet radiation in the photodemethylation of methylmercury in freshwater ecosystems. Environ. Sci. Technol. 2009, 43, 5692–5698. (17) Hammerschmidt, C. R.; Fitzgerald, W. F.; Lamborg, C. H.; Balcom, P. H.; Tseng, C.-M. Biogeochemistry of methylmercury in lakes and tundra watersheds of arctic Alaska. Environ. Sci. Technol. 2006, 40, 1204–1211. (18) Inoko, M. Studies on the photochemical decomposition of organomercurials-methylmercury(II) chloride. Environ. Pollut. Ser. B 1981, 2, 3–10. (19) Zepp, R. G.; Holgne´, J.; Bader, H. Nitrate-induced photooxidation of trace organic chemicals in water. Environ. Sci. Technol. 1987, 21, 443–450. (20) Suda, I.; Takahashi, H. Degradation of methyl and ethyl mercury into inorganic mercury by other reactive oxygen species besides hydroxyl radical. Arch. Toxicol. 1992, 66, 34–39. (21) Suda, I.; Totoki, S.; Takahashi, H. Degradation of methyl and ethyl mercury into inorganic mercury by oxygen free radicalproducing systems: involvement of hydroxyl radical. Arch. Toxicol. 1991, 65, 129–134.

(22) Chen, J.; Pehkonen, S. O.; Lin, C.-J. Degradation of monomethylmercury chloride by hydroxyl radicals in simulated natural waters. Water Res. 2003, 37, 2496–2504. (23) Suda, I.; Suda, M.; Hirayama, K. Degradation of methyl and ethyl mercury by singlet oxygen generated from sea water exposed to sunlight or ultraviolet light. Arch. Toxicol. 1993, 67, 365–368. (24) Schuchmann, M. N.; von Sonntag, C. Hydroxyl radical-induced oxidation of 2-methyl-2-proponal in oxygenated aqueous solutionsA product and pulse radiolysis study. J. Phys. Chem. 1979, 83, 780–784. (25) Zafiriou, O. C.; True, M. B. Nitrate photolysis in seawater by sunlight. Mar. Chem. 1979, 8, 33–42. (26) Fenton, J. J. H. Oxidation of tartaric acid in presence of iron. J. Chem. Soc. 1894, 65, 899–910. (27) Zepp, R. G.; Faust, B. C.; Holgne´, J. Hydroxyl radical formation in aqueous reactions (pH 3-8) of iron(II) with hydrogen peroxide: the photo-Fenton reaction. Environ. Sci. Technol. 1992, 26, 313–319. (28) Kling, G. W.; O’Brien, W. J.; Miller, M. C.; Hershey, A. E. The biogeochemistry and zoogeography of lakes and rivers in arctic Alaska. Hydrobiologia 1992, 240, 1–14. (29) Gill, G. A.; Fitzgerald, W. F. Mercury sampling of open ocean waters at the picomolar level. Deep Sea Res. 1985, 32, 287–297. (30) Lamborg, C. H.; Tseng, C.-M.; Fitzgerald, W. F.; Balcom, P. H.; Hammerschmidt, C. R. Determination of the mercury complexation characteristics of dissolved organic matter in natural waters with “reducible Hg” titrations. Environ. Sci. Technol. 2003, 37, 3316–3322. (31) Amyot, M.; Gill, G. A.; Morel, F. M. M. Production and loss of dissolved gaseous mercury in coastal seawater. Environ. Sci. Technol. 1997, 31, 3606–3611. (32) Arctic Long Term Ecological Research. http://ecosystems.mbl.edu/ARC/Datatable.html (accessed Nov 14, 2009). (33) Tseng, C.-M.; Hammerschmidt, C. R.; Fitzgerald, W. F. Determination of methylmercury in environmental matrixes by online flow injection and atomic fluorescence spectrometry. Anal. Chem. 2004, 76, 7131–7136. (34) Rose, A. L.; Waite, T. D. Reduction of organically complexed ferric iron by superoxide in a simulated natural water. Environ. Sci. Technol. 2005, 39, 2645–2650. (35) Barbeau, K.; Rue, E. L.; Trick, C. G.; Bruland, K. W.; Butler, A. Photochemical reactivity of siderophores produced by marine heterotrophic bacteria and cyanobacteria based on characteristic Fe(III) binding groups. Limnol. Oceanogr. 2003, 48, 1069– 1078. (36) Vermilyea, A. W.; Voelker, B. M. Photo-Fenton reaction at near neutral pH. Environ. Sci. Technol. 2009, 43, 6927–6933.

(37) Pignatello, J. J.; Oliveros, E.; MacKay, A. Advanced oxidation processes for organic contaminant destruction based on the Fenton reaction and related chemistry. Crit. Rev. Environ. Sci. Technol. 2006, 36, 1–84. (38) Devasagayum, T. P. A.; Steenken, S.; Obendorf, M. S. W.; Schulz, W. A.; Sies, H. Formation of 8-hydroxy(deoxy)guanosine and generation of strand breaks at guanine residues in DNA by singlet oxygen. Biochemistry 1991, 30, 6283–6289. (39) Basu-Modak, S.; Tyrell, R. M. Singlet oxygen: A primary effector in the ultraviolet A/near-visible light induction of the human heme oxygenase gene. Cancer Res. 1993, 53, 4505–4510. (40) Moorhouse, C.; Halliwell, B.; Grootveld, M.; Gutteridge, J. Cobalt(II) ion as a promoter of hydroxyl radical and possible crypto-hydroxyl radical formation under physiological conditionssdifferential-effects of hydroxyl radical scavengers. Biochim. Biophys. Acta 1985, 843, 261–268. (41) Stohs, S. J.; Bagchi, D. Oxidative mechanisms in the toxicity of metal ions. Free Radical Biol. Med. 1995, 18, 321–336. (42) Saito, M. A.; Rocap, G.; Moffett, J. W. Production of cobalt binding ligands in a Synechococcus feature at the Costa Rica upwelling dome. Limnol. Oceanogr. 2005, 50, 279–290. (43) Duckworth, O. W.; Bargar, J. R.; Jarzecki, A. A.; Oyerinde, O.; Spiro, T. G.; Sposito, G. The exceptionally stable cobalt(III)desferrioxamine B complex. Mar. Chem. 2009, 113, 114–122. (44) Balistrieri, L. S.; Murray, J. W.; Paul, B. The biogeochemical cycling of trace metals in the water column of Lake Sammamish, Washington: Response to seasonally anoxic conditions. Limnol. Oceanogr. 1992, 37, 529–548. (45) Beaubien, S.; Nriagu, J.; Blowes, D.; Lawson, G. Chromium speciation and distribution in the Great Lakes. Environ. Sci. Technol. 1994, 28, 730–736. (46) Achterberg, E. P.; van den Berg, C. M. G.; Boussemart, M.; Davison, W. Speciation and cycling of trace metals in Esthwaite water: A productive English lake with seasonal deep-water anoxia. Geochim. Cosmochim. Acta 1997, 61, 5233–5253. (47) Van der Zee, J.; Krootjes, B. B. H.; Chignell, C. F.; Dubbelman, T. M. A. R.; Van Steveninck, J. Hydroxyl radical generation by a light-dependent Fenton reaction. Free Radical Biol. Med. 1993, 14, 105–113. (48) Laglera, L. M.; Van den Berg, C. M. G. Wavelength dependence of the photochemical reduction of iron in arctic seawater. Environ. Sci. Technol. 2007, 41, 2296–2302. (49) Rue, E. L.; Bruland, K. W. Complexation of iron(III) by natural organic ligands in the Central North Pacific as determined by a new competitive ligand equilibration/adsorptive cathodic stripping voltammetric method. Mar. Chem. 1995, 50, 117–138.

ES1006934

VOL. 44, NO. 16, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6143