Cite This: Environ. Sci. Technol. 2018, 52,

Transcription

1 This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes. Cite This: pubs.acs.org/est Thermodynamics of Hg(II) Bonding to Thiol Groups in Suwannee River Natural Organic Matter Resolved by Competitive Ligand Exchange, Hg L III -Edge EXAFS and 1 H NMR Spectroscopy Yu Song, Tao Jiang, Van Liem-Nguyen,, Tobias Sparrman, Erik Bjo rn, and Ulf Skyllberg*, Department of Forest Ecology and Management, Swedish University of Agricultural Science, SE Umeå, Sweden Department of Chemistry, Umeå University, SE Umeå, Sweden School of Science and Technology, O rebro University, SE O rebro, Sweden Downloaded via on April 14, 019 at 10:5:43 (UTC). See for options on how to legitimately share published articles. *S Supporting Information ABSTRACT: A molecular level understanding of the thermodynamics and kinetics of the chemical bonding between mercury, Hg(II), and natural organic matter (NOM) associated thiol functional groups (NOM-RSH) is required if bioavailability and transformation processes of Hg in the environment are to be fully understood. This study provides the thermodynamic stability of the Hg(NOM-RS) structure using a robust method in which cysteine (Cys) served as a competing ligand to NOM (Suwannee River R101N sample) associated RSH groups. The concentration of the latter was quantified to be 7.5 ± 0.4 μmol g 1 NOM by Hg L III -edge EXAFS spectroscopy. The Hg(Cys) molecule concentration in chemical equilibrium with the Hg(II)-NOM complexes was directly determined by HPLC-ICPMS and losses of free Cys due to secondary reactions with NOM was accounted for in experiments using 1 H NMR spectroscopy and 13 C isotope labeled Cys. The log K ± SD for the formation of the Hg(NOM-RS) molecular structure, Hg + + NOM-RS = Hg(NOM-RS), and for the Hg(Cys)(NOM-RS) mixed complex, Hg + + Cys + NOM-RS = Hg(Cys)(NOM-RS), were determined to be 40.0 ± 0. and 38.5 ± 0., respectively, at ph 3.0. The magnitude of these constants was further confirmed by 1 H NMR spectroscopy and the Hg(NOM-RS) structure was verified by Hg L III -edge EXAFS spectroscopy. An important finding is that the thermodynamic stabilities of the complexes Hg(NOM-RS), Hg(Cys)(NOM-RS) and Hg(Cys) are very similar in magnitude at ph values <7, when all thiol groups are protonated. Together with data on 15 low molecular mass (LMM) thiols, as determined by the same method (Liem-Ngyuen et al. Thermodynamic stability of mercury(ii) complexes formed with environmentally relevant low-molecular-mass thiols studied by competing ligand exchange and density functional theory. Environ. Chem. 017, 14, (4), ), the constants for Hg(NOM-RS) and Hg(Cys)(NOM-RS) represent an internally consistent thermodynamic data set that we recommend is used in studies where the chemical speciation of Hg(II) is determined in the presence of NOM and LMM thiols. INTRODUCTION Passive diffusion 16,17 and active transport 18 are proposed Complexes formed between inorganic, divalent mercury mechanisms for cellular uptake of small, neutral Hg(II) 0 Hg(II) and natural organic matter (NOM) functional groups molecules, such as HgCl, Hg(SH) 0, while Hg(II) complexes are known to be important for the chemical speciation, formed with low molecular mass thiols (LMM-RSH) are bioavailability and transformation of Hg in the environment.,3 proposed to be actively transported across cell membranes. 18,19 Being a soft Lewis acid, Hg(II) has a strong affinity for thiol To add complexity, thiol (RSH) functionalities located in the functional groups in NOM (NOM-RSH), which largely very membrane (or wall) of cells will influence the biouptake control the chemical speciation of Hg(II) under oxic of Hg(II), 0 4 as a consequence of their competition with conditions. Under sulfidic conditions, inorganic sulfide species thiol groups of NOM and low molecular mass (LMM) compete with NOM-RSH functional groups for Hg(II). The compounds outside the cell. 5 7 Thus, to fully understand formation of more complex Hg(II)-sulfide molecules, such as mechanisms and rates of cellular uptake of Hg(II), information nanoparticulate (nano-hgs) and crystalline metacinnabar (β- HgS), 4,5 is further influenced by secondary effects of NOM- Received: February 16, 018 RSH on rates of nucleation and growth of HgS colloids, 4,6 11 Revised: July 4, 018 which subsequently affect biouptake and transformation of Accepted: July 7, 018 Hg. 3,1 15 Published: July 7, American Chemical Society 89 DOI: /acs.est.8b00919

2 about concentrations of all these different types of thiols, as well as data on the thermodynamics and kinetics of complexes formed between the thiols and Hg(II), is required. Although much progress has been made the last decades, there are uncertainties regarding the thermodynamic stability of complexes formed between Hg(II) and NOM-RSH functional groups, including possible mixed ligation with LMM thiols and other soft ligands such as I. 1 There is compelling evidence from Hg EXAFS spectroscopy measurements of a stable Hg(NOM-RS) structure This structure is expected to be the dominant Hg(II) form in nonsulfidic environments, even if some doubt recently was raised about its long-term stability. 31 In a critical review, the value of the formation constant for the structure Hg(NOM-RS), log K Hg(NOM RS), was constrained to 4 ±. The variability reported is expected to be mainly related to methodology, where constants selected for competing ligands (e.g., Br and thiol ligands) in experimental work, as well as pk a values for the NOM-RSH groups are particularly critical. So far, there are only two studies in which the concentration of NOM-RSH groups was quantified and a molecular reaction for the formation of the Hg(NOM-RS) structure was formulated. 3,33 Notably, the proposed Hg(NOM-RS) molecular structure has so far not been independently verified by spectroscopic measurements in any of the existing reports presenting experimental data on the stability constant log K Hg(NOM RS). Because of the exceptionally strong bonding of Hg(II) to NOM-RSH groups, thermodynamic experiments require addition of a potent competing ligand with known stability constants for Hg(II) complex formation LMM molecules containing a thiol group have been used for this purpose. 33,35,38 Liem-Nguyen and co-workers recently developed two robust mass spectrometry-based methods by which specific LMM- RSH molecules and their Hg(LMM-RS) complexes could be determined at sub-nm levels. 1,39 In this work, we used cysteine (Cys) as a competing ligand to determine the stability constant for the Hg(NOM-RS) structure, using the well characterized and in research frequently applied Suwannee River NOM (R101N, International Humic Substances Society, IHSS). We used Hg L III -edge EXAFS spectroscopy to determine the concentration of NOM- RSH functional groups and to verify the Hg(NOM-RS) structure in our experiments. Thermodynamic constants were derived from data on the equilibrium concentration of the Hg(Cys) complex, as determined by HPLC-ICPMS. 1 Further, the presence of a mixed complex, Hg(Cys)(NOM-RS), with Hg(II) bonded to one Cys and one NOM-RS group was proposed, based on 13 C-Cys isotope labeling experiments. We optimized the values on the thermodynamic stability constants (log K) for the Hg(NOM-RS) and Hg(Cys)(NOM-RS) structures by a least-squares procedure using data obtained at varying Hg(II) to NOM mass ratios. Thermodynamic stabilities were further validated by proton nuclear magnetic resonance ( 1 H NMR) spectroscopy. The direct and simplistic approach, avoiding liquid liquid and solid-phase extraction steps, as well as the independent, spectroscopic determination of structures and thiol groups associated with NOM is expected to result in a lower uncertainty in the values of reported constants, as compared to previous estimates. MATERIALS AND METHODS Materials. All solutions and reagents were prepared in degassed Milli-Q water (18 MΩ cm) inside a N (g) filled glovebox (COY). Milli-Q water was deoxygenated by purging with nitrogen gas overnight in the glovebox. A Hg(II) stock solution (6.5 mm) was prepared from Hg(NO 3 ) in 1% nitric acid. The concentration of Hg(II) stock solution was verified by reversed isotope dilution analysis using ICPMS (Elan DRCe) and by combustion atomic absorption spectrometry (C- AAS, DMA-80). Stock solutions of 0.1 mm of low molecular mass thiols (Sigma-Aldrich), i.e., cysteine (Cys), homocysteine (HomoCys) and N-acetyl-L-cysteine (NACCys) were freshly prepared and stored in the glovebox no more than 4 h before use. We used Suwannee River NOM (R101N), obtained from IHSS in this study. The concentration of major elements in this material is displayed in Table S1, where in particular the concentration of Cl should be noted. Stock solutions of NOM ( 500 mg L 1 ) were freshly prepared by dissolving the NOM material in deoxygenated Milli-Q water in the glovebox. The stock solution was filtered through a 0.-μm filter (Millipore) before use, and its concentration was confirmed by a total organic carbon (TOC) analyzer (Shimadzu TOC-V + TNM1). Losses of TOC were less than % in the filtration step, indicating a close to complete dissolution of the NOM material. Chemical Equilibrium Experiments and Thermodynamic Calculations in the Hg-NOM-Cysteine System. A series of NOM solutions (10, 0, 40, 80 and 00 mg L 1 ) were prepared in duplicates in 15 ml polypropylene tubes (Sarstedt), by diluting the NOM stock solution in deoxygenated Milli-Q water added NaClO 4 to provide an inert, constant ionic medium of 10 mm. To these NOM solutions were added aliquots of Hg(II) stock solution to obtain a final volume of 10 ml and concentration of 0.5 μm Hg(II). To avoid introducing potential interferences, no ph-buffer was added. Because of the numerous NOM acid base functional groups, ph was maintained at 3.1 ± 0. in all Hg-NOM experiments. Reaction vessels were protected from light by aluminum foil and were maintained at 5 ± 1 C by a thermostat in the glovebox. The Hg-NOM solutions were gently mixed in an anaerobic glovebox for 4 h. Although the reaction of Hg(II) with NOM is expected to be kinetically controlled, 40 Hg L III -edge EXAFS data demonstrated that added Hg(II) was complexed by two NOM-RS groups after 4 h, as well as after 5 days of reaction, as depicted by reaction 1: + Hg + NOM RS = Hg(NOM RS) After 4 h of reaction between Hg(II) and NOM, μm Cys was added as a competing ligand to the NOM-RS groups, as described by reaction. Hg(NOM RS) + Cys = Hg(Cys) + NOM RS Here we define HCys as the cysteine molecule with the carboxyl group deprotonated ( COO ) and the thiol ( SH) and amino groups ( NH + 3 ) protonated, HSCH CH(NH + 3 )- COO, and Cys refers to the molecule when both the carboxyl and thiol groups are deprotonated. As established in a pilot study, a final concentration of μm of Cys was chosen to provide a measurable concentration of the Hg(Cys) complex in all experiments. Because Cys may be slowly oxidized by NOM, especially at high ph, 41 we chose to keep a low ph in our experiments. Parallel experiments with 0.5 μm Hg(II) and μm Cys prepared in deoxygenated Milli-Q water in absence (1) () 893 DOI: /acs.est.8b00919

3 of NOM, and with ph adjusted to 3.0 by nitric acid, were used as controls. We conducted Cys ligand-exchange experiments to determine the log K Hg(NOM RS) for reaction 1, using equation 3: Hg(NOM RS) KHg(NOM RS) = [ ] + { Hg }[ NOM RS ] (3) in which {} denotes activity and [] denotes equilibrium concentration in mol L 1. All activities were calculated by the extended Debye Hu ckel equation. 4 Because the NOM-RS and Hg(NOM-RS) molecules are part of a mixture of differently sized, unknown macromolecules, activities cannot be specified and we therefore use concentrations for these two components. The term {Hg + }inequation 3 was substituted { Hg(Cys) } for the quotient as derived from the law of mass K Hg(Cys) { Cys } action (equation 5) applied to reaction 4: + Hg + Cys = Hg(Cys) K Hg(Cys) yielding equation 6. K K = { Hg(Cys) } + { Hg }{ Cys } Hg(NOM RS) Hg(Cys) Hg(NOM RS) Cys = [ ]{ } { Hg(Cys) }[ NOM RS ] The value of K Hg(Cys) was set to , following Liem- Nguyen et al. 1 Of the four unknowns on the right side of equation 6, the concentration of the Hg(Cys) complex was directly determined by HPLC-ICPMS (see below) and {Cys } was calculated from [Cys ]bydifference using the mass balance equation 7. added [ Cys ] = [ Cys ] + [ HCys] + [ Hg(Cys) ] +[ Cys NOM]+[ Hg(Cys)(NOM RS) ] (7) The term Cys-NOM represents Cys that is reacted with NOM components, including losses due to degradation, as quantified in specific experiments (see below). The Hg(Cys)- (NOM-RS) structure is a proposed complex formed by a mixture of one Cys and one NOM-RS functional group, as described by reaction 8. + Hg + Cys + NOM RS = Hg(Cys)(NOM RS) The concentration of the mixed complex was estimated based on experiments in which 13 C labeled Cys was allowed to react with NOM in absence and presence of Hg(II), as described in Supporting Information, and by a final refinement of the chemical speciation model (Table S). Concentrations of HCys and Cys were calculated using a K a value of for reaction a CysH = Cys + H, K = 10 Finally, to solve the two remaining unknowns in equation 6, Hg(NOM-RS) and NOM-RS, the concentration of Hg- (NOM-RS) was calculated from the mass balance equation 10. The total Hg concentration [Hg tot ] was measured by C- AAS. Other Hg(II) components, like the free Hg + ion and its (4) (5) (6) (8) (9) complexes with OH and Cl (reactions which for completeness are included in the chemical speciation model, Table S), all made negligible contributions to the mass balance. [ Hg ] = [ Hg(Cys) ] + [ Hg(NOM RS) ] tot +[ Hg(Cys)(NOM RS) ] (10) The total concentration of NOM-associated thiol groups [NOM-RS tot ] was determined by a Hg + titration procedure of these groups monitored by Hg L III -edge EXAFS, as described below. The [NOM-RS ] component in equation 6 was calculated from the mass balance (equation 11) and the pk a value of reaction 1 was set to 10.0, following Skyllberg. [ NOM RStot] = [ NOM RS ] + [ NOM RSH] + Hg(NOMRS) [ ] +[ Hg(Cys)(NOM RS) ] (11) a NOM RSH = NOM RS + H, K = 10 (1) All reactions and selected thermodynamic constants considered in our model are listed in Table S. It should be noted that because of the exceptionally strong covalent bond formation between Hg(II) and NOM-RS, electrostatic forces involving negatively charged NOM functional groups is expected to have insignificant impact on the chemical speciation. Calculations were conducted using the chemical speciation computer programs PHREEQC 43 and WinSGW. 44 Errors in the reported values of K Hg(NOM RS) and K Hg(Cys)(NOM RS) were propagated by use of equation 13, where SD i denotes the standard deviation of error source i. SD = i SD i (13) Spectroscopic Determinations. Sulfur K-edge XANES, Hg L III -edge EXAFS and 1 H NMR spectroscopy were used to determine the concentration of NOM-RS tot functional groups, to identify structures of Hg(II)-NOM complexes, and to provide independent thermodynamic data. The concentration of the Hg(Cys) complex was determined by HPLC-ICPMS according to Liem-Nguyen et al. 1 A description of these methods is given in the Supporting Information. The concentration of Cys was determined by LC-ESI-MS/MS, using a procedure described previously. 39 Finally, we used 13 C isotope labeled Cys ( 13 C-Cys) to monitor possible degradation of Cys and reaction with NOM functional groups in experiments, as described in the Supporting Information. RESULTS AND DISCUSSION Hg L III -Edge EXAFS Determinations of the Concentration of NOM-RS tot. We used a Hg(II) titration procedure, monitored by Hg L III -edge EXAFS, to determine the concentration of thiol functional groups in NOM, NOM- RS tot. The NOM samples were allowed to react with 800, 1600 and 3300 μg of Hg(II) per gram of dry mass of NOM for 5 days in the glovebox. After freeze-drying, pelleting and storage at 80 C and EXAFS measurements, the measured Hg tot concentrations were determined to be 780, 1400 and 650 μg g 1, respectively. Some loss of Hg in experiments with NOM is expected to occur as a consequence of the reduction of Hg(II) to Hg(0) by NOM under dark conditions, 40,45,46 which was 894 DOI: /acs.est.8b00919

4 Table 1. Least-Squares 1st Coordination Shell Model Fits to Full k-space Hg L III -Edge EXAFS Data for NOM Samples Added Different Concentrations of Hg(II) a First S shell First Cl shell Species composition b Hg (μg g 1 ) ΔE 0 (ev) d CN R (Å) σ (Å ) CN R (Å) σ (Å ) Hg(NOM- RS) HgCl NOM-RS tot concentration c (μmol g 1 ) (f) 0.003(f) (f) (f) 90% 10% (f) 0.003(f) 0.7.8(f) (f) 54% 46% (f) 0.003(f) (f) (f) 9% 71% 7.9 a Second coordination shell fits and merit-of-fit of full model are reported in Table S4. b Calculated as CNs/(CNs + CN Cl ) 100. c Calculated as CNs/(CNs + CN Cl ) [Hg], where [Hg] is concentration in μmol g 1 NOM. d ΔE 0 = edge-energy shift, CN = coordination number, R = bond distance, σ = Debye Waller factor. (f) values were fixed during fitting. Figure 1. Experimental Hg L III -edge EXAFS data (thin, black solid line) and model fits (dashed lines) collected at 77 K on freeze-dried NOM samples equilibrated with Hg(II) for 5 days at ph 3 in 6 mm Cl. The blue dashed line denotes a model with only one floating 1st S shell and the red dashed line denotes the full model, including 1st and nd S and Cl shells. (a) EXAFS spectra in k-space, (b) Fourier transformed (FT) spectra not corrected for phase shift, with full model fit (in dashed red), and (c) back-filtered FT spectra for the range 1..7 Å. In panel b, vertical dash lines indicate 1st S+Cl, and nd S and Cl shells. Model fits to 1st shell data are given in Table 1 and model fits to nd shell data are given in Table S4. also observed in the equilibrium experiments of this study (see below). In agreement with previous research, 8 EXAFS determinations revealed a Hg(NOM-RS) structure formed between Hg(II) and two thiol groups in NOM after 5 days of reaction. At the lowest Hg(II) concentration of 780 μg g 1, an average Hg S bond distance of.34 Å was obtained when the coordination number (CN) was fixed to.0. The result provides strong support for the Hg(NOM-RS) structure. 8,47 We did not add any chloride, but the sample contained Cl (Table S1) due to the method used by IHSS to extract and upconcentrate the Suwannee River NOM sample. In the Hg(II)- NOM suspensions prepared for EXAFS determinations, in which a minimum of water was added to improve signal-tonoise ratio, we measured 6 mm Cl ions. Because Cl ions are known to form bonds with Hg(II) that are stronger than associations with oxygen functionalities in NOM, we included a Hg Cl bond in our model and fixed it to.8 Å in agreement with the HgCl complex. 48,49 To constrain Cl and S bonding to Hg(II) in a similar way, we also fixed the Hg S bond distance (at.35 Å) in our final modeling, which is a typical distance of the Hg S bond in a Hg(SR) structure. The first coordination shell model fits to full EXAFS k-space data are reported in Table 1 and fits to FT back-filtered data are reported in Table S3. As expected, the coordination number of Cl increased with increasing addition of Hg(II) (Figure 1, Table 1). Thus, when the RS groups in NOM were increasingly saturated by Hg(II), the coordination with the more weakly bonded ligand Cl increased. Inclusion of a first shell Hg Cl bond at.8 Å significantly improved the model fits to FT back-filtered first shell data for samples added 780, 1400 and 650 μg g 1 Hg(II), by 43, 90 and 94%, respectively. Although the Hg Cl bond distance in two-coordinated Hg(II) complexes is only 0.07 Å shorter than the Hg S bond length, the extraordinary good signal-to-noise ratio in our data enabled us to separate the contribution from Hg S and Hg Cl bonding. Because of the increased contribution of Cl (and decrease in the contribution of S), the radial distance of the first coordination shell peak decreased slightly with Hg(II) loading (Figure 1b). In the second coordination shell, inclusion of one S atom at.95 Å and two Cl atoms at 3.3 and 3.5 Å, respectively, significantly improved model fits by >0% (Table S4). The second shell S distance is in agreement with previous studies of Hg(II)-NOM complexation, 8,47 reporting an organic sulfide contribution at Å, whereas the second shell Cl distances are in agreement with the crystalline structure of HgCl (s). 49 Possibly, the freeze-drying of samples gave rise to some formation of HgCl (s) crystals when water was removed, alternatively the structure observed in the second shell reflects 895 DOI: /acs.est.8b00919

5 a contribution from dissolved Cl ions, combining with Hg(NOM-RS) and aqueous HgCl, and/or possible mixed HgCl(NOM-RS) structures. In either case, the inclusion of Cl in first and second shells improved the merit-of-fit for the full k-space data by, 70 and 51% for NOM samples added 780, 1400 and 650 μg g 1 Hg(II), respectively. Based on the Hg EXAFS data, the concentration of thiol groups (NOM-RS tot ) in the Suwannee River NOM sample could be calculated by equation 14. The assumption behind this equation is that all thiol groups take part in the bonding of Hg(II) before significant numbers of Cl ligands get involved. [ NOM RS ] = tot CNS CN + CN S Cl Hg [ ] (14) In this equation, CN S and CN Cl denote the first shell coordination numbers of S and Cl atoms, respectively, and [] denotes molar concentrations per mass of NOM. Using the model fits to full EXAFS k-space data, the concentration of NOM-RS tot was calculated to be 7.0, 7.5 and 7.9 μmol g 1 for samples with Hg(II) concentrations of 780, 1400 and 650 μg g 1, respectively. Thus, we obtained very similar estimates of the NOM-RS tot concentration in all three samples, suggesting our model assumptions were reasonable. As the final estimate of the NOM-RS tot, we used the average value of 7.5 ± 0.4 μmol g 1 (NOM). To further strengthen our model assumptions, we conducted theoretical thermodynamic calculations (using the data in Table S). Using the average NOM-RS tot concentration of 7.5 μmol g 1, the calculated relative abundance of the two major structures Hg(SR-NOM) (97, 54 and 8%) and HgCl (3, 45 and 69%, Table S5) was well in agreement with the EXAFS data (90, 54 and 9% for Hg(SR-NOM) and 10, 46 and 71% for HgCl ), within limits of uncertainty of both methods (Table 1, Table S5). It should be pointed out that also carboxylic and phenolic groups in NOM may form complexes with Hg(II) when NOM-RS groups are saturated. 40 However, because these associations are substantially weaker than Hg Cl bonds, we could not observe any O/N atoms in the first coordination shell of our EXAFS data. This is reasonable given that the sum of the concentration of NOM-RS tot and Cl were always in much excess of Hg(II) (considering its + charge) in the experimental systems. The insignificant contribution from O/ N containing functional groups of NOM in the bonding of Hg(II) was also predicted by the thermodynamic modeling (Table S5). By use of S K-edge XANES, we determined the concentration of reduced organic sulfur functionalities (Org- S RED, representing the sum of thiol RSH, monosulfide RSR and disulfide RSSR) 3 to account for 10% of total sulfur (Figure S). This is a small percentage of Org-S RED as compared to previous studies of NOM samples, 50,51 which can explained by the high concentration of inorganic sulfate in the Suwannee River NOM sample. Similar to Cl, sulfate ions are upconcentrated in the NOM sample due to the IHSS extraction procedure (Table S1). Recalculated to mass of NOM, Org- S RED makes up 56 μmol g 1 of NOM which is at the same level as in previous studies of similar types of NOM samples.,8 A comparison with the Hg L III -edge EXAFS determined concentration of 7.5 μmol g 1 of NOM-RS tot suggests these groups accounted for 15% of Org-S RED, which is a little lower than the average of previous studies of NOM from soils and waters, but still within the expected range of 10 35%.,8,50,5 Notably, no forms of inorganic sulfide, such as FeS(s) or nanoparticulate HgS, were detected by neither S K-edge XANES nor by the Hg L III -edge EXAFS analyses. Hg-NOM-Cysteine Ligand-Exchange Experiments. We conducted ligand-exchange experiments, allowing Cys to compete with NOM-RS functional groups for the complexation of Hg(II). Prior to the addition of Cys, Hg(II) was reacted with NOM for 4 h to form the Hg(NOM-RS) structure. The dominance of this structure was independently demonstrated by use of Hg L III -edge EXAFS for a NOM sample added Hg(II) corresponding to 800 μg g 1. The first coordination shell Hg S distance was.36 Å and the coordination number (CN).0 (data not shown). Because the values of the EXAFS parameters (R, CN, σ ) did not differ from the ones obtained after 5 days of Hg(II)-NOM reaction (Table 1), we suggest the system reached close to a chemical equilibrium within 4 h of reaction. Notably, because this EXAFS sample was prepared in only 1 ml of Milli-Q (see Supporting Information) the concentration of Cl was estimated to be 6 mm and the concentration of NOM-RS tot would be expected to have been close to saturated by the added Hg(II). Yet, inclusion of Cl as a first shell backscatter did not significantly improve model fits. After addition of Cys, the concentration of the Hg(Cys) molecule was directly measured at different times of reaction by HPLC-ICPMS. After a rapid reaction and increase within minutes, the concentration of Hg(Cys) remained constant (within 10% RSD) in the time frame from 5 min to 5 d (Figure S3). It should be noted that the concentration of Cl was below 0.15 mm in all Hg-NOM-Cys ligand-exchange experiments and the formation of aqueous HgCl n n complexes therefore can be ruled out as being of importance for the results. After the addition of Cys, Hg EXAFS data collected at a Hg/ NOM mass ratio of 800 μg g 1 and at a Cl concentration less than 0.15 mm demonstrated a first shell Hg(II) coordination with S atoms at a distance of Å (Figure S4, Table S6), which did not change with time of reaction (1 10 h). The EXAFS data can be understood from the HPLC- ICPMS measurements conducted in experiments with a similar Hg/NOM ratio (0.5 μm of Hg reacted with 00 mg NOM L 1, corresponding to 500 μg g 1, and then added μm Cys). As shown in Figure S3e, 50% of Hg(II) was in the form of the Hg(Cys) complex ( 0. μm) and the remaining 50% was bonded with NOM, obviously in the Hg(NOM-RS) structure, as demonstrated by EXAFS and thermodynamic modeling (see below). From Hg EXAFS determinations alone we cannot distinguish if the Hg(NOM-RS) and Hg(Cys) complexes only exist in these distinct forms or if they partly combine in a mixed Hg(Cys)(NOM-RS) complex. This is because Hg(II) is bonded with two thiol groups at Å in all three structures. Even if we cannot prove chemical equilibrium by reversibility in the Hg-NOM-Cys experimental system, the fact that the first coordination shell (as determined by EXAFS) remained the same at 1 h, 4 h and 5 days of reaction after Cys addition (Figure S4, Table S6), and that the measured Hg(Cys) concentration (as determined by HPLC-ICPMS), once quickly formed, did not change with time (Figure S3) indicate that chemical equilibrium was achieved among Hg(Cys), Hg(NOM-RS) and Hg(Cys)(NOM-RS) complexes already within minutes of reaction. This interpretation 896 DOI: /acs.est.8b00919

6 is further supported by the thermodynamic modeling (see below). Notably the measured Hg tot concentrations were 0 40% lower in the Hg-NOM-Cys systems than what was expected from the additions, and losses varied with the Hg/NOM ratio (Figure and Figure S1). All losses occurred during the first 4 Figure. Measured concentrations (±SD) of Hg(Cys) and total concentrations of Hg (Hg tot ) plotted as a function of NOM concentration. Experiments were conducted at ph = 3 and I = 10 mm NaClO 4 by pre-equilibrating 0.5 μm Hg with NOM (10 00 mg L 1 ) for 4 h followed by a reaction with μm Cys for 5 min to 5 days (Figure S3). Dotted lines show model fits. In panel a, only the Hg(NOM-RS) complex was considered and the log K Hg(NOM RS) was determined to be 40.4 ± 0.. The merit-of-fit, 100( [model exp] / [exp] ) for Hg(Cys) was 0.0%. In panel b, also the mixed complex Hg(Cys)(NOM-RS) was included in the model and log K Hg(NOM RS) and log K Hg(Cys)(NOM RS) were determined to be 40.0 ± 0. and 38.5 ± 0., respectively. The merit-of-fit for Hg(Cys) was 0.06%. the initial condition is dominated by the Hg(NOM-RS) structure. Reaction Between Cysteine and NOM and Formation of a Mixed Hg(Cys)(NOM-RS) Complex. LMM-RS compounds are known to react with RSH functional groups under the formation of organic disulfides, 54 and because Cys is sensitive to degradation in the presence of NOM, e.g., through oxidation, 41 it was important to monitor the concentration of free Cys over time in our experiments. As shown by LC-ESI- MS/MS determinations (Figure S6), Cys was not degraded in the absence of NOM. Similar results were obtained by 1 H NMR, with less than 5% of Cys degraded after 5 days (Figure S7a). However, in the presence of NOM (1000 mg L 1 ) the concentration of free Cys (100 μm) decreased by 30% in 5 days (Figure S7b), suggesting reactions taking place between Cys and NOM. Similarly, 13 C-Cys reacting with NOM showed a time-dependency (see Supporting Information for details on how 13 C data were interpreted) increasing from 1% after 1 h to 5% after 5 days (Figure 3). The 10 times higher Cys to NOM ratio may be the reason for slightly higher reactivity of Cys with NOM in the 1 H NMR experiment. h of reaction between Hg(II) and NOM in absence of Cys. In control systems without NOM, there was no Hg loss (Figure S1). This observation is in agreement with previous findings that Hg(II) is reduced to Hg(0) in the presence of NOM under dark conditions. 45,46 Losses are greatest during the first minutes to an hour of reaction when equilibrium is not established, when weaker NOM bonds are expected to be involved in the complexation of Hg(II). 40 Because the reaction tubes were sealed during reaction, losses likely occurred as Hg(0) evasion when they were opened for Hg tot analyses, or possibly formed Hg(0) could have been sorbed to the walls of the tubes. To further deepen our understanding of the kinetics of Hg(II) bonding to thiol groups belonging to different molecules, the reaction between N-acetyl-L-cysteine (NACCys), homocysteine (HomoCys) and Hg(II) was experimentally studied (see Supporting Information). Concentrations of Hg(HomoCys) and Hg(NACCys) were demonstrated to reach equilibrium within less than 3 min (Figure S5). This observation is in agreement with previous studies showing that although the Hg(II) thiol bond is very strong, the rate of ligand-exchange among Hg(II) thiol complexes is on the order of seconds. 53 The latter study demonstrated Hg(II) complex formation with a mixture of thiol groups pertaining to a relatively small, glutathione (GSH) and large molecules, hemoglobin (having eight thiol groups). These heterogenic Hg(II) thiol complexes were shown by 1 H NMR spectroscopy to have lifetimes of less than 30 s due to quick ligandexchange among GSH molecules. Similarly, it seems as rates of reactions between Cys and large macromolecules, including NOM-RS functional groups, also are very quick, at least when Figure 3. Concentration of 13 C-Cys associated with NOM in absence (black) and presence (red) of Hg(II), as a function of time of reaction. Black symbols and line denote experiments where μm 13 C- Cys was added to 00 mg L 1 of NOM. Red symbols and line denote experiments where 0.5 μm Hg(II) was pre-equilibrated with 00 mg L 1 of NOM solution for 4 h, and then μm 13 C-Cys was added. Blue symbols and dashed line designate the difference of the two experiments, and is interpreted as formation of the mixed ligand complex Hg( 13 C-Cys)(NOM-RS). Error bars represent ± SD. The experiments were conducted at ph = 3 and I = 10 mm NaClO 4 As compared to reaction with NOM alone, 13 C-Cys reacted with NOM increased by in average 7% if NOM was allowed to react with Hg(II) for 4 h before Cys was added. This increase in NOM-associated Cys is illustrated by the blue symbols and dashed line in Figure 3. We interpret this extra NOM-reacted Cys in the presence of Hg(II) as reflecting a formation of a mixed complex with the proposed composition Hg(Cys)- (NOM-RS). EXAFS data further confirmed that all Hg (within an uncertainty of ±5 10%) was indeed bonded to two thiol groups. But again, the three complexes Hg(Cys), Hg(NOM- RS) and Hg(Cys)(NOM-RS) cannot be separated by Hg EXAFS. Thermodynamic Calculations. From equations 6 1, the thermodynamic constant for the formation of the Hg(NOM-SR) complex was first calculated disregarding the existence of the proposed mixed Hg(Cys)(NOM-RS) 897 DOI: /acs.est.8b00919

7 complex. In a second step, we calculated the concentration of the mixed Hg(Cys)(NOM-RS) structure as the difference between free 13 C-Cys concentrations measured in systems with and without Hg(II) addition (Figure 3), by also including equation 15. K [ Hg(Cys)(NOM RS) ] = { Hg }{ Cys }[ NOM RS ] Hg(Cys)(NOM RS) + (15) The calculation was done by an iterative procedure where the error-sum-of-squares was minimized by adjusting the two parameters K Hg(NOM RS) and K Hg(Cys)(NOM RS) while keeping the quotient between them (=37) the same as determined by the 13 C-Cys experiment. The log K Hg(NOM RS) was calculated to be 40.4 ± 0., when the mixed Hg(Cys)(NOM-RS) was disregarded, and 40.0 ± 0. in the presence of this mixed complex. In the latter case, log K Hg(Cys)(NOM RS) was calculated to be 38.5 ± 0.. Fits to data of the two alternative models are indicated by the dotted lines in Figure a,b. The sources of uncertainties propagated by equation 13 for the log constants (±0.) are listed in Table S8. Because the relative error of the fits did not differ between the two alternative models (0.0 and 0.06%, respectively, Figure ), evidence for a mixed complex relies entirely on the observations from 13 C-Cys experiments, and not on the thermodynamic model fitting. We used 1 H NMR spectroscopy measurements to validate the thermodynamic models. We determined the equilibrium concentrations of free Cys and Hg(Cys), Figure S8. The Hg(II) concentration was required to be much higher than in experiments using HPLC-ICPMS, but the Hg(II) ratio to NOM (3000 μg g 1 ) was kept similar to experiments in which 0.5 μm Hg(II) was added to 0 or 40 mg L 1 of NOM. The measured concentrations of Hg tot, Cys and Hg(Cys) are tabulated together with the theoretical concentrations calculated from our final model, with log K Hg(NOM RS) and log K Hg(Cys)(NOM RS) set to 40.0 and 38.5, respectively, in Table S7. As further illustrated in Figure S9, we obtained a good correspondence between measured and model predicted concentrations of Hg(Cys) and free Cys by use of 1 H NMR. We consider our reported thermodynamic stability for the Hg(NOM-SR) structure to be the most robust reported so far. The direct determination of the Hg(Cys) complex formed with the competing ligand, the continuous monitoring and correction for losses of Cys and the independent spectroscopic determination and verification of the concentration of NOM- RS tot groups and the Hg(NOM-SR) structure, respectively, are all major improvements as compared to previous laboratory studies on Hg(II)-NOM complexation.,3 35,38 In none of the previous studies have the modeled structure of the Hg(II)- NOM complex been spectroscopically verified. As described by Black and co-workers, 34 the CLE-SPE methodology employed in several of these studies to separate Hg(II)-NOM complexes from Hg(II) complexes formed with competing ligands, is subjected to operational difficulties related to the dynamic character of hydrophilic and hydrophobic fractions of NOM and how these fractions may be affected by and interact with the competing LMM thiol ligands and the C-18 column employed for separation. With the simplistic and direct approach used here, operational constraints are kept at a minimum. Environmental Implications. Although NOM is known to significantly influence speciation and bioavailability of Hg(II), the details on the thermodynamics and kinetics of Hg(II) structures formed with NOM functionalities need to be sharpened. In this study, the structure and thermodynamic stability of the Hg(NOM-RS) complex was verified by Hg L III -edge EXAFS. It should be noted that NOM-RSH is a representation of all thiol groups in NOM and that the pk a and log K Hg(NOM RS) therefore can be considered average constants. The size of the constant for Hg(NOM-RS) in essence confirms the magnitude of critically reviewed constants for this chemical structure in NOM. It should, however, be noted that our reported values on the thermodynamic constants (similar to other ligand-exchange studies) heavily depends on the pk a value of the competing ligand (HCys) and on the log K value of the formation of Hg(Cys). For every log-unit increase in log K Hg(Cys) the log K for the formation of Hg(NOM-RS) and Hg(Cys)(NOM-RS) will increase by one log-unit and for every log-unit increase in the pk a for HCys the K Hg(NOM RS) and K Hg(Cys)(NOM RS) will increase by two and one log-units, respectively. At acidic and neutral ph conditions, the thiol groups of HCys and NOM-RSH will be protonated and the relative stabilities of Hg(Cys), Hg(NOM-RS) and Hg(Cys)(NOM-RS) can be described by a general reaction 16, where RSH denotes a thiol group of either Cys or NOM: + + Hg(RS) Hg + RSH = Hg(RS) + H, log K (16) Formulated in agreement with equation 16, the log K Hg(Cys), K Hg(NOM RS) and K Hg(Cys)(NOM RS) have the values 0.3, 0.0 and 19.9, respectively (calculated as , and , respectively). This means that all three complexes in essence (within experimental errors of ±0.) have equal thermodynamic stabilities at ph-values when the thiol groups are protonated. This is a very important finding, since it tells us that regardless of values chosen for the pk a of HCys and log K for Hg(Cys), which vary substantially in the literature, we may adjust constants to be consistent with our finding that Hg(Cys), Hg(NOM-RS) and Hg(Cys)- (NOM-RS) have equal stabilities at and below neutral ph. This would be particularly important in systems where thiol compounds compete and control the chemical speciation and bioavailability of Hg(II), such as at surfaces of bacteria. To extend this reasoning further, we argue for the usage of internally consistent thermodynamic models. Using the same HPLC-ICPMS methodology as in this work, Liem-Nguyen et al. 1 reported log K Hg(RS) values for 15 different LMM thiols. Because Cys was part of that study, it means that this entire set of constants for LMM thiols is consistent with the values on log K Hg(NOM RS) and K Hg(Cys)(NOM RS) reported in this work. We therefore recommend using the stability constants for Hg(II) complex formation with LMM thiols from the study of Liem-Nguyen et al. 1 in combination with the thermodynamic constants reported for Hg(NOM-RS) and Hg(Cys)(NOM- RS) as reported here. In our experiments, concentrations of Cys ( μm) were about times higher than normally encountered in soils and waters. Yet, the indicated large stability of the Hg(Cys)(NOM-RS) complex and the assumption that other LMM thiols may also form similar mixed complexes with NOM-RS suggest these complexes may be important for transportation and biouptake of Hg(II), e.g., by methylating bacteria. A simple calculation, in which concentrations of Hg(II) (1 10 ng L 1 ), NOM (5 100 mg L 1 ), Cys (1 100 nm), ph 3 7 were varied to cover typical ranges in soils and 898 DOI: /acs.est.8b00919

8 waters, demonstrates that the Hg(Cys)(NOM-RS) complex can be expected to generally be more abundant than Hg(Cys), and in cases even more abundant than Hg(NOM-RS) (Figure S10). Future work is needed to directly confirm the structure of Hg(Cys)(NOM-RS) and its bioavailability. The previously proposed HgNOM-RS + (Hg + + RS = HgRS + ) species 55,56 was not detected by our EXAFS measurements. Because one-coordinated Hg(II) complexes are not thermodynamically stable, 57 a more correct structure of the HgSR + species is a two-coordinated complex with one thiol and one oxygen (or nitrogen) functionality, Hg(NOM-RORS), where the mathematic formulation suggests the two functional groups are belonging to the same molecule (forming a bidentate complex). According to the linear free energy relationships reported by Dyrssen and Wedborg, 56 the log K for such a HgRS + molecule should be. Our thermodynamic modeling constrained the log K for such a complex to be smaller than 6. Neither did our EXAFS determinations on Hg-NOM-Cys systems revealed any sign of O/N functionalities, indicative of a formation of the HgCys + or Hg(NOM-RS) + complex, where Hg(II) forms a bidentate complex with a combination of thiol and carboxyl functionalities of a single Cys or NOM molecule. Given ±5% detection limit of possible O/N functionalities by EXAFS, we estimate the log K for the HgCys + species to be smaller than 6. Our HPLC-ICPMS measurements further constrained the log K to a maximum of 4. Within method uncertainties, these values are in fair agreement with data from Liem-Nguyen et al. 1 Notably, their reported log K value of for HgRS + complexes formed with a variety of LMM thiols was incorrectly expressed as monodentate complexes in which Hg(II) was complexed by separate carboxyl and thiol ligands, Hg( I RO)( II RS). A correct modeling of the same experimental data with Hg(II) complexed by carboxyl and thiol groups of the same molecule, Hg(RORS), results in a log K value for HgCys + of 5 ± 0.5 (for details, see Supporting Information). ASSOCIATED CONTENT *S Supporting Information The Supporting Information is available free of charge on the ACS Publications website at DOI: /acs.est.8b Materials and methods (PDF) AUTHOR INFORMATION Corresponding Author *Ulf Skyllberg. Phone: +46 (0) ; ulf. skyllberg@slu.se. ORCID Yu Song: Erik Bjo rn: Ulf Skyllberg: Notes The authors declare no competing financial interest. ACKNOWLEDGMENTS This work was financially supported by the Swedish Research Council (VR) project Sino-Swedish Mercury Management Research Framework SMaReF ( ) and the VR project ( ) to U.S., and by the Kempe Foundations (JCK-1501, SMK-745, SMK-143). We acknowledge Dr. Roberto Boada and Ann-Kathrin Geiger at the Diamond Light Source (Beamline I0-scanning), for assistance with the Hg L III -edge EXAFS spectroscopy measurements (project SP9157). We also acknowledge Dr. Chenyan Ma at the Beijing Synchrotron Facility (Beamline 4B7A), Chinese Academy of Sciences, for assistance with the S K-edge XANES spectroscopy measurements. REFERENCES (1) Liem-Nguyen, V.; Skyllberg, U.; Nam, K.; Bjo rn, E. Thermodynamic stability of mercury(ii) complexes formed with environmentally relevant low-molecular-mass thiols studied by competing ligand exchange and density functional theory. Environmental Chemistry 017, 14 (4), () Skyllberg, U.Competition among thiols and inorganic sulfides and polysulfides for Hg and MeHg in wetland soils and sediments under suboxic conditions: Illumination of controversies and implications for MeHg net production. J. Geophys. Res.-Biogeosci. 008, 113 (G), DOI: /008JG (3) Hsu-Kim, H.; Kucharzyk, K. H.; Zhang, T.; Deshusses, M. A. Mechanisms regulating mercury bioavailability for methylating microorganisms in the aquatic environment: a critical review. Environ. Sci. Technol. 013, 47 (6), (4) Gerbig, C. A.; Kim, C. S.; Stegemeier, J. P.; Ryan, J. N.; Aiken, G. R. Formation of nanocolloidal metacinnabar in mercury-dom-sulfide systems. Environ. Sci. Technol. 011, 45 (1), (5) Drott, A.; Bjorn, E.; Bouchet, S.; Skyllberg, U. Refining thermodynamic constants for mercury (II)-sulfides in equilibrium with metacinnabar at sub-micromolar aqueous sulfide concentrations. Environ. Sci. Technol. 013, 47 (9), (6) Ravichandran, M.; Aiken, G. R.; Ryan, J. N.; Reddy, M. M. Inhibition of precipitation and aggregation of metacinnabar (mercuric sulfide) by dissolved organic matter isolated from the Florida Everglades. Environ. Sci. Technol. 1999, 33 (9), (7) Waples, J. S.; Nagy, K. L.; Aiken, G. R.; Ryan, J. N. Dissolution of cinnabar (HgS) in the presence of natural organic matter. Geochim. Cosmochim. Acta 005, 69 (6), (8) Deonarine, A.; Hsu-Kim, H. Precipitation of mercuric sulfide nanoparticles in NOM-containing water: implications for the natural environment. Environ. Sci. Technol. 009, 43 (7), (9) Slowey, A. J. Rate of formation and dissolution of mercury sulfide nanoparticles: The dual role of natural organic matter. Geochim. Cosmochim. Acta 010, 74 (16), (10) Aiken, G. R.; Hsu-Kim, H.; Ryan, J. N. Influence of dissolved organic matter on the environmental fate of metals, nanoparticles, and colloids. Environ. Sci. Technol. 011, 45 (8), (11) Poulin, B. A.; Gerbig, C. A.; Kim, C. S.; Stegemeier, J. P.; Ryan, J. N.; Aiken, G. R. Effects of Sulfide Concentration and Dissolved Organic Matter Characteristics on the Structure of Nanocolloidal Metacinnabar. Environ. Sci. Technol. 017, 51 (), (1) Zhang, T.; Kim, B.; Levard, C.; Reinsch, B. C.; Lowry, G. V.; Deshusses, M. A.; Hsu-Kim, H. Methylation of mercury by bacteria exposed to dissolved, nanoparticulate, and microparticulate mercuric sulfides. Environ. Sci. Technol. 01, 46 (13), (13) Graham, A. M.; Aiken, G. R.; Gilmour, C. C. Effect of dissolved organic matter source and character on microbial Hg methylation in Hg-S-DOM solutions. Environ. Sci. Technol. 013, 47 (11), (14) Zhang, T.; Kucharzyk, K. H.; Kim, B.; Deshusses, M. A.; Hsu- Kim, H. Net methylation of mercury in estuarine sediment microcosms amended with dissolved, nanoparticulate, and microparticulate mercuric sulfides. Environ. Sci. Technol. 014, 48 (16), (15) Pham, A. L.-T.; Morris, A.; Zhang, T.; Ticknor, J.; Levard, C.; Hsu-Kim, H. Precipitation of nanoscale mercuric sulfides in the presence of natural organic matter: Structural properties, aggregation, and biotransformation. Geochim. Cosmochim. Acta 014, 133, DOI: /acs.est.8b00919

9 (16) Mason, R. P.; Reinfelder, J. R.; Morel, F. M. Uptake, toxicity, and trophic transfer of mercury in a coastal diatom. Environ. Sci. Technol. 1996, 30 (6), (17) Barkay, T.; Gillman, M.; Turner, R. R. Effects of dissolved organic carbon and salinity on bioavailability of mercury. Appl. Environ. Microbiol. 1997, 63 (11), (18) Schaefer, J. K.; Rocks, S. S.; Zheng, W.; Liang, L.; Gu, B.; Morel, F. M. Active transport, substrate specificity, and methylation of Hg(II) in anaerobic bacteria. Proc. Natl. Acad. Sci. U. S. A. 011, 108 (1), (19) Schaefer, J. K.; Morel, F. M. High methylation rates of mercury bound to cysteine by Geobacter sulfurreducens. Nat. Geosci. 009, (), (0) Mishra, B.; O Loughlin, E. J.; Boyanov, M. I.; Kemner, K. M. Binding of HgII to high-affinity sites on bacteria inhibits reduction to Hg0 by mixed FeII/III phases. Environ. Sci. Technol. 011, 45 (), (1) Graham, A. M.; Bullock, A. L.; Maizel, A. C.; Elias, D. A.; Gilmour, C. C. Detailed assessment of the kinetics of Hg-cell association, Hg methylation, and methylmercury degradation in several Desulfovibrio species. Appl. Environ. Microbiol. 01, 78 (0), () Mishra, B.; Shoenfelt, E.; Yu, Q.; Yee, N.; Fein, J. B.; Myneni, S. C. B. Stoichiometry of mercury-thiol complexes on bacterial cell envelopes. Chem. Geol. 017, 464, (3) Wang, Y.; Schaefer, J. K.; Mishra, B.; Yee, N. Intracellular Hg(0) Oxidation in Desulfovibrio desulfuricans ND13. Environ. Sci. Technol. 016, 50 (0), (4) Liu, Y. R.; Lu, X.; Zhao, L.; An, J.; He, J. Z.; Pierce, E. M.; Johs, A.; Gu, B. Effects of Cellular Sorption on Mercury Bioavailability and Methylmercury Production by Desulfovibrio desulfuricans ND13. Environ. Sci. Technol. 016, 50 (4), (5) Zhao, L.; Chen, H.; Lu, X.; Lin, H.; Christensen, G. A.; Pierce, E. M.; Gu, B. Contrasting Effects of Dissolved Organic Matter on Mercury Methylation by Geobacter sulfurreducens PCA and Desulfovibrio desulfuricans ND13. Environ. Sci. Technol. 017, 51 (18), (6) Moreau, J. W.; Gionfriddo, C. M.; Krabbenhoft, D. P.; Ogorek, J. M.; DeWild, J. F.; Aiken, G. R.; Roden, E. E. The Effect of Natural Organic Matter on Mercury Methylation by Desulfobulbus propionicus 1pr3. Front. Microbiol. 015, 6, (7) Dunham-Cheatham, S.; Mishra, B.; Myneni, S.; Fein, J. B. The effect of natural organic matter on the adsorption of mercury to bacterial cells. Geochim. Cosmochim. Acta 015, 150, (8) Skyllberg, U.; Bloom, P. R.; Qian, J.; Lin, C. M.; Bleam, W. F. Complexation of mercury(ii) in soil organic matter: EXAFS evidence for linear two-coordination with reduced sulfur groups. Environ. Sci. Technol. 006, 40 (13), (9) Skyllberg, U.; Drott, A. Competition between disordered iron sulfide and natural organic matter associated thiols for mercury (II) An EXAFS study. Environ. Sci. Technol. 010, 44 (4), (30) Nagy, K. L.; Manceau, A.; Gasper, J. D.; Ryan, J. N.; Aiken, G. R. Metallothionein-like multinuclear clusters of mercury(ii) and sulfur in peat. Environ. Sci. Technol. 011, 45 (17), (31) Manceau, A.; Lemouchi, C.; Enescu, M.; Gaillot, A. C.; Lanson, M.; Magnin, V.; Glatzel, P.; Poulin, B. A.; Ryan, J. N.; Aiken, G. R.; Gautier-Luneau, I.; Nagy, K. L. Formation of Mercury Sulfide from Hg(II)-Thiolate Complexes in Natural Organic Matter. Environ. Sci. Technol. 015, 49 (16), (3) Skyllberg, U.; Xia, K.; Bloom, P. R.; Nater, E. A.; Bleam, W. F. Binding of mercury (II) to reduced sulfur in soil organic matter along upland-peat soil transects. Journal of Environmental Quality 000, 9 (3), (33) Khwaja, A. R.; Bloom, P. R.; Brezonik, P. L. Binding constants of divalent mercury (Hg+) in soil humic acids and soil organic matter. Environ. Sci. Technol. 006, 40 (3), (34) Black, F. J.; Bruland, K. W.; Flegal, A. R. Competing ligand exchange-solid phase extraction method for the determination of the complexation of dissolved inorganic mercury (II) in natural waters. Anal. Chim. Acta 007, 598 (), (35) Gasper, J. D.; Aiken, G. R.; Ryan, J. N. A critical review of three methods used for the measurement of mercury (Hg+)-dissolved organic matter stability constants. Appl. Geochem. 007, (8), (36) Ravichandran, M. Interactions between mercury and dissolved organic matter a review. Chemosphere 004, 55 (3), (37) Miller, C. L.; Liang, L.; Gu, B. Competitive ligand exchange reveals time dependant changes in the reactivity of Hg-dissolved organic matter complexes. Environmental Chemistry 01, 9 (6), (38) Hsu, H.; Sedlak, D. L. Strong Hg (II) complexation in municipal wastewater effluent and surface waters. Environ. Sci. Technol. 003, 37 (1), (39) Liem-Nguyen, V.; Bouchet, S.; Bjorn, E. Determination of subnanomolar levels of low molecular mass thiols in natural waters by liquid chromatography tandem mass spectrometry after derivatization with p-(hydroxymercuri) benzoate and online preconcentration. Anal. Chem. 015, 87 (), (40) Jiang, T.; Skyllberg, U.; Wei, S.; Wang, D.; Lu, S.; Jiang, Z.; Flanagan, D. C. Modeling of the structure-specific kinetics of abiotic, dark reduction of Hg(II) complexed by O/N and S functional groups in humic acids while accounting for time-dependent structural rearrangement. Geochim. Cosmochim. Acta 015, 154, (41) Chu, C.; Erickson, P. R.; Lundeen, R. A.; Stamatelatos, D.; Alaimo, P. J.; Latch, D. E.; McNeill, K. Photochemical and Nonphotochemical Transformations of Cysteine with Dissolved Organic Matter. Environ. Sci. Technol. 016, 50 (1), (4) Stumm, W.; Morgan, J. J. Aquatic chemistry: chemical equilibria and rates in natural waters; John Wiley & Sons, 01; Vol. 16. (43) Parkhurst, D. L. User s guide to PHREEQC: A computer program for speciation, reaction-path, advective-transport, and inverse geochemical calculations, (44) Karlsson, M.; Lindgren, J. WinSGW, a user interface for SolGasWater, 006. (45) Gu, B.; Bian, Y.; Miller, C. L.; Dong, W.; Jiang, X.; Liang, L. Mercury reduction and complexation by natural organic matter in anoxic environments. Proc. Natl. Acad. Sci. U. S. A. 011, 108 (4), (46) Zheng, W.; Liang, L.; Gu, B. Mercury reduction and oxidation by reduced natural organic matter in anoxic environments. Environ. Sci. Technol. 01, 46 (1), 9 9. (47) Manceau, A.; Lemouchi, C.; Rovezzi, M.; Lanson, M.; Glatzel, P.; Nagy, K. L.; Gautier-Luneau, I.; Joly, Y.; Enescu, M. Structure, Bonding, and Stability of Mercury Complexes with Thiolate and Thioether Ligands from High-Resolution XANES Spectroscopy and First-Principles Calculations. Inorg. Chem. 015, 54 (4), (48) Bower, J.; Savage, K. S.; Weinman, B.; Barnett, M. O.; Hamilton, W. P.; Harper, W. F. Immobilization of mercury by pyrite (FeS ). Environ. Pollut. 008, 156 (), (49) Boada, R.; Cibin, G.; Coleman, F.; Diaz-Moreno, S.; Gianolio, D.; Hardacre, C.; Hayama, S.; Holbrey, J. D.; Ramli, R.; Seddon, K. R.; Srinivasan, G.; Swadzba-Kwasny, M. Mercury capture on a supported chlorocuprate(ii) ionic liquid adsorbent studied using operando synchrotron X-ray absorption spectroscopy. Dalton Trans 016, 45 (47), (50) Qian, J.; Skyllberg, U.; Frech, W.; Bleam, W. F.; Bloom, P. R.; Petit, P. E. Bonding of methyl mercury to reduced sulfur groups in soil and stream organic matter as determined by X-ray absorption spectroscopy and binding affinity studies. Geochim. Cosmochim. Acta 00, 66 (), (51) Skyllberg, U.; Qian, J.; Frech, W.; Xia, K.; Bleam, W. F. Distribution of mercury, methyl mercury and organic sulphur species in soil, soil solution and stream of a boreal forest catchment. Biogeochemistry 003, 64 (1), (5) Liem-Nguyen, V.; Skyllberg, U.; Bjorn, E. Thermodynamic Modeling of the Solubility and Chemical Speciation of Mercury and Methylmercury Driven by Organic Thiols and Micromolar Sulfide 8300 DOI: /acs.est.8b00919