Determination of Mercury Complexation in Coastal and Estuarine

Jul 26, 2005 - competing ligand, thiosalicylic acid, in a competitive ligand exchange method in which water-toluene extraction was used to determine ...
0 downloads 0 Views 250KB Size
Environ. Sci. Technol. 2005, 39, 6607-6615

Determination of Mercury Complexation in Coastal and Estuarine Waters Using Competitive Ligand Exchange Method SEUNGHEE HAN* AND GARY A. GILL Laboratory for Oceanographic and Environmental Research (LOER), Department of Oceanography, Texas A&M University at Galveston, Galveston, Texas 77551

While many studies have examined Hg(II) binding ligand in natural dissolved organic matter, determined ligand concentrations far exceed natural Hg(II) concentrations. This ligand class may not influence natural Hg(II) complexation, given the reverse relation between ligand concentration and metal-ligand binding strength. This study used a new competing ligand, thiosalicylic acid, in a competitive ligand exchange method in which water-toluene extraction was used to determine extremely strong Hg(II) binding sites in estuarine and coastal waters (dissolved [Hg] ) 0.5-8 pM). Thiosalicylic acid competition lowered the detection limit of Hg(II) complexing ligand by 2 orders of magnitude from values found by previous studies; the determined Hg(II) complexing ligand ranged from 13 to 103 pM. The logarithmic conditional stability constants between Hg(II) and Hg(II) complexing ligand (Kcond′ ) [HgL]/([Hg2+][L′]), [L′] ) total [L] - [HgL]) ranged from 26.5 to 29.0. Applying the same method for chloride competition detected another class of ligand that is present from 0.5 to 9.6 nM with log conditional stability constants ranging from 23.1 to 24.4. A linear relationship was observed between the log conditional stability constant and log Hg(II) complexing ligand concentration, supporting the hypothesis that Hg(II) binding ligand should be characterized as a series or continuum of binding sites on natural dissolved organic matter. Calculating Hg(II) complexation using the conditional stability constants and ligand concentrations determined in this study indicates that >99% of the dissolved mercury is complexed by natural ligand associated with dissolved organic matter in estuarine and coastal waters of Galveston Bay, Texas.

Introduction In general, the chemical speciation of an element in natural water governs its biogeochemical behavior and bioavailability (1-3). Like many other metals, Hg(II) is readily complexed by various ligand groups within dissolved organic matter (DOM; 4-6). This interaction is important in controlling the solubility, mobility (7-12), and bioavailability (13-17) of Hg(II). In freshwater environments, strong correlations between concentrations of dissolved Hg and dissolved organic carbon (DOC) provide evidence of significant binding between Hg(II) and DOM (6-8, 13, 14). Complexation between Hg(II) and organic matter may increase the solubility of Hg, and may thus contribute to Hg transport from terrestrial to aquatic environments. Numerous reports stress the governing * Corresponding author phone: (858)822-4677; fax: (858)534-7313; e-mail: [email protected]. 10.1021/es048667z CCC: $30.25 Published on Web 07/26/2005

 2005 American Chemical Society

role of DOM in controlling the dissolved Hg in estuarine environments (4, 5, 9-12). Despite the importance of organic complexation in Hg biogeochemical cycling, direct assessment of Hg(II) complexation is limited compared to studies of other trace metals. Standard voltammetric methods using Hg electrodes cannot be used for Hg(II) speciation, and low concentrations of Hg(II) in natural waters inhibit the application of simple voltammetric methods. Only one voltammetric method using a gold disk electrode (18) and an analogue of a voltammetric method using SnCl2 reduction (19) have been reported; these studies have suggested that most Hg(II) in freshwater, estuarine water, and coastal seawater is associated with DOM. Recently, surface complexation models (20, 21), solid-phase extraction (SPE; 22), and the equilibrium dialysis ligand exchange (EDLE) methods (23) have been used to characterize Hg(II)-DOM complexes in natural waters (Table 1). Though numerous studies have sought to characterize Hg-DOM complexes in natural waters, reliable binding constant values for environmentally relevant concentrations of Hg(II) binding sites are unavailable. Competitive ligand exchange (CLE) is one approach currently used to determine metal speciation in natural waters. Competition between natural and added ligands for Hg(II) complexation, which is influenced by the stability constants and concentrations of each ligand, determines the extent to which a metal is bound to natural complexes. The CLE method has several advantages. For instance, a low detection limit can be set using the metal detection technique, and the concentration of the competing ligand can be varied. In this study, CLE was combined with water-toluene extraction, a method modified from the chloride competition technique developed by Benoit et al. (24). Filtered estuarine waters ranging from 0 to 35 ppt salinity were analyzed using two competing ligands: chloride and thiosalicylic acid (TSA). Because of its higher complexation coefficient (Log RHgTSAi ) 15-17, RΗgTSAi ) K1[TSA2-] + β2[TSA2-]2), TSA competition can detect Hg(II) complexing ligands at the pM level, and under higher conditional stability constants than chloride. Chloride competition can detect Hg(II) complexing ligands in the nM range, and with lower conditional stability constants, corresponding to the complexation capacity between Hg(II) and chloride (Log RΗgCli ) 11-14, RΗgCli ) β2[Cl-]2 + β3[Cl-]3 + β4[Cl-]4).

Methods Sample Collection and Reagents. Filtered surface water samples were collected from Galveston Bay, and Lavaca Bay, Texas, and coastal Texas waters using a peristaltic pump system and polyethersulfone membrane filters (0.45 µm) following clean sampling procedures (25, 26). Tris(hydroxymethyl)aminomethane (TRIZMA, SigmaAldrich) and 2-amino-2-methyl-1,3-propanediol (SigmaAldrich) were used to prepare 1 M stock buffer solutions of pH 7 and pH 10, respectively. The competing ligand, thiosalicylic acid (2 µM), was prepared fresh for each titration in an acetonitrile solution using Milli-Q water. High-performance liquid chromatography (HPLC)-grade toluene and reagent-grade KCl were used without further purification. Bromine monochloride (BrCl) solution was prepared according to U.S. Environmental Protection Agency (EPA) Method 1631 (27). Low mercury contents in each reagent were verified with blank tests. Mercury standard solutions were prepared by dilution of a 1000 ppm stock standard obtained from GFS Chemicals (Powell, OH). A river water mercury standard (ORMS; National Research Council Canada) VOL. 39, NO. 17, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6607

TABLE 1. Reported Concentrations of Hg(II) Binding Ligand ([L]) and Conditional Stability Constants of HgL Complexes log Kcond

reaction

[L]

pH

sample type

HgXi + L′ ) HgLa

9.7-10.8

1.4-4.5 nM

7.2

natural water

Hg2+ + L′ ) HgL RXHn- + Hg2+ ) RXHg(n-1)- + H+ Hg2+ + L′ ) HgL Hg2+ + DOM ) HgDOM Hg2+ + L2- ) HgL Hg2+ + DOMs- ) Hg-DOMs+

21-23 11.8, 10.6

0.3-60 nM 20-40 µM

7.5 6.0

natural water DOM isolates

>30 23.2 31.6-32.2 22.8, 23.2

0.5 nM 5 nmol/mg DOM 48-217 mmol/Kg C 280 nM

7.0 3.0-3.4 6.0

wastewater DOM isolates soil organics DOM released from peat

a

method

ref

anodic stripping voltammetry Sn(II) reduction CLE-SSE

18

CLE-SPE EDLE Hg sorption modeling Hg sorption modeling

22 23, 37 20 21

19 24

HgXi ) sum of inorganic mercury species; L′ ) [Hg]t - [HgL].

and sediment mercury standard (PACS; National Research Council Canada) were used to verify the accuracy and recovery of mercury by cold vapor atomic fluorescence spectrometry (CVAFS). Water-Toluene Extraction. Mercury titrations were performed by adding increasing amounts of inorganic Hg(II) to a series of separatory funnels. In each separatory funnel, buffer solution, competing ligand solution (KCl or TSA), and Hg(II) standard (generally 1-15 nM for the chloride competition and 0.01-0.15 nM for the TSA competition) were added to 100 mL of natural sample, after which 10 mL of toluene was added. Hg(II) does not complex to any significant degree with TRIZMA or 2-amino-2-methyl-1,3-propanediol. This was confirmed by comparing Hg(II) concentrations in water-toluene extractions using the organic buffers with phosphate and borate buffers. The mixture was allowed to equilibrate for 20-24 h with intermittent shaking. After the last vigorous shaking, the water phase was drained from the funnel. Aliquots (30 mL) of the water phase were acidified to measure total Hg(II); the remaining water was used for pH measurements. Determining Mercury. CVAFS with sample reduction using NaBH4 was used to quantify Hg concentrations (25, 26). The precision of the CVAFS method was 97% were obtained when the solvent extraction solutions were treated with a weak mixed acid (0.07 N HNO3 and 0.06 N HCl) for 24 h. Direct measurement of Hg(II) in the toluene phase using back extraction resulted in poor recovery. Therefore, the Hg(II) concentration in the toluene phase was determined from the difference between total and aqueous Hg(II) measurements.

[Hg2+][L′]

[L′] ) [L]t - [HgL] 6608

9

where Vo is the volume of the organic phase and Va is the volume of aqueous phase. The distribution of HgCl3- and HgCl42- to the organic phase can be ignored because these species are hydrophilic at a 10:1 water/solvent ratio (24). The neutral species Hg(OH)20 and Hg(OH)Cl0 are important solution species only when chloride levels are low and pH is high. By adding chloride to low-salinity samples and buffering the extractions at pH < 7.5, these solution species become less than 1% of the total mercury; thus, these species were not included in eq 3. The concentration of aqueous Hg, [Hg]a, is assumed to be the sum of [HgCl3-], [HgCl42-], and aqueous [HgCl20]. The concentrations of [Hg2+]a and [HgL]a can be ignored because of their low concentrations

[Hg]a ≈ [HgCl3-]a + [HgCl42-]a + [HgCl20]a

(4)

The free aqueous Hg2+ concentration, [Hg2+]a, is obtained from experimentally determined [Hg]a using stability constants for HgCln (n ) 2, 3, and 4) and the chloride concentration, as follows:

[Hg2+]a )

[Hg]a β2[Cl-]a2 + β3[Cl-]a3 + β4[Cl-]a4

(5)

(1) (2)

[HgCl20]a ) β2[Hg2+]a[Cl-]a2

CLE-SSE (Solvent Solvent Extraction) Using Chloride. A conditional stability constant (Kcond′) can be defined by the concentration of free mercury ([Hg2+]), the concentration of natural ligand that is not bound by mercury ([L′]), and the concentration of Hg(II)-ligand complexes ([HgL]). Here, the term ligand (L) is defined as an Hg(II) binding site in molecules and clusters.

[HgL]

Vo [Hg]t ) [Hg2+]a + [HgCl20]a + [HgCl20]o + [HgCl3-]a + Va Vo 2[HgCl4 ]a + [HgL]o + [HgL]a (3) Va

The stability constants for HgCln (β2 ) 1014.0, β3 ) 1015.0, and β4 ) 1015.6 at I ) 0, 25 °C) were obtained from the National Institute of Standards and Technology (NIST) database 46 (28, Table 2). Ionic strength corrections were made to the stability constants using the Davies equation (29). The Davies equation provides accurate activity coefficients for monovalent salts up to the ionic strength of seawater. The relative error between the Davies equation and the Truesdell-Jones equation in determining β2(HgCl2) was 0.1 M, the relative abundance of Hg2+, Hg(OH)20, and Hg(OH)Cl0 species is very small, and it is assumed that only neutral species extract into the organic phase. The Hg(II) concentration in the organic phase at pH 7 and [Cl-] > 0.1 M is then given by [Hg]o ) [HgCl20]o. The concentration of HgCl20 in the aqueous phase was determined by MINEQL modeling. Five salinities of UV-irradiated seawater were used for a series of titrations. Each individual titration consisted of three or more additions of Hg(II). The average Kd(HgCl20) was 4.3 ( 2.7 (n ) 5). The large standard deviation may result from the sensitive variation of Kd and uncertainties in determining the experimental titration slope. Uncertainties associated with a rigorous value for Kd were estimated to (68% in determining [L] and (3% in determining log Kcond′ at the 95% confidence level. The toluene-water distribution coefficient (Kd) of Hg(TSA)0 is given by

Kd )

[Hg(TSA)0]oVo [Hg(TSA)0]aVa

(17)

Similar to HgCl2, it is assumed that only neutrally charged Hg(II) species extract into the organic phase; the Hg(II) concentration in the organic phase at the experimental conditions is then given by [Hg]o ) [Hg(TSA)0]o. The Kd (Hg(TSA)0) was determined using the same calculation process as HgCl2. The average Kd (Hg(TSA)0) determined using Galveston Bay samples was 5.8 ( 2.5 (n ) 13). The uncertainties associated with the determination of Kd are estimated to be (13% for [L] and (1% for log Kcond′ at the 95% confidence level. CLE Conditions. Usable ranges of ligand concentration for chloride and TSA competitions were determined from the titration slope. For chloride competition, the chloride concentration should range between 0.1 and 0.6 M to produce an unsaturated (99% of the dissolved mercury is complexed by natural ligands in estuarine and coastal waters of the study region. Environmental Implications. Figure 4 plots the Hg(II) complexing ligand concentrations from this study and other similar studies (18, 19, 21, 23, 24, 37) as a function of the

FIGURE 4. Relations between the concentration of Hg(II) binding ligand (L) and conditional stability constant of HgL reported for natural water samples. Data are shown in Tables 4, 5, and 6 in this study. Log Kcond ) [HgL-(n-2)]/([Hg2+][Ln-]).

TABLE 6. Reported Stability Constants of HgL and Concentrations of Hg(II) Binding Ligand (L) in Natural Waters log Kcond′

log Kcond

[L]t (nM)

reaction pH

HgXi + L′ ) HgLa

9.7 10.2 10.6 10.8 9.8

Hg2+ + L2- ) HgL

25.6b,c 26.1 26.5 26.7 26.2

2.25 4.47 2.67 2.51 1.35

7.2

18

Hg2+ + L′ ) HgL

21.5 21.0 21.2 22.2 22.9 21.6 22.7 23.5 23.0

Hg2+ + L2- ) HgL

24.0c 23.5 23.7 24.7 25.4 24.1 25.2 26.0 25.5

8.0 20 60 3.8 6.4 24.8 3.0 4.0 0.3

7.5

19

RXHn- + Hg2+ ) RXHg(n-1)- + H+

10.6 11.8

Hg2+ + RS- ) HgRS+

22.4c 23.8

1410d 3600

6.0

24

Hg2+ + DOM ) HgDOM

23.2

Hg2+ + RS- ) HgRS+

28.7c

5-500e

7.0

23, 37

Hg2+ + DOMs- ) Hg-DOMs+

22.8 23.2

278 283

6.0

21

reaction

reaction

ref

a HgX ) sum of inorganic mercury species; L′ ) [L] - [HgL]. b Corrections made with: R 2+ 13.0 for salinity 0 and i t Hgr ()[reducible Hg]/[Hg ]) ) 10 19 ppt, RHgr ) 1013.5 for salinity 35 ppt (19); pKa ) 10. c Corrections made with pKa ) 10. d Corrections made with: [DOM]t ) 30 µM; reduced S mol fraction ) 0.12 for F1 (log K ) 23.8) and 0.05 for 2BS (log K ) 22.4). e Corrections made with [L]t ) 5 nmol/mg DOM and 1-100 mg DOM/L natural water (2).

conditional stability constant. To allow direct comparison of the stability constant, data reported by Wu et al. (18) were corrected to reflect an equilibrium constant with respect to free Hg2+ (Table 6). The ratio of Sn(II)-reducible Hg to free Hg2+ (Hgr ) [Sn(II)-reducible Hg]/[Hg2+]) determined for estuarine samples by Lamborg et al. (19) were used for the corrections; log RHgr ) 13.0 for salinity of 0 and 19 ppt, and log RHgr ) 13.5 for salinity of 35 ppt. The conditional stability constants with respect to L′ reported by Wu et al. (18) and Lamborg et al. (19) were corrected to the conditional stability constant with respect to free L- using pKa ) 10.0 as discussed previously. The binding site concentrations of 30 µM reported by Benoit et al. (24) were corrected to 3600 nM (site F1) and 1410 nM (site 2BS) using the reduced sulfur mole fraction of 0.12 mol mol-1 DOM for F1, and 0.05 mol mol-1 DOM for 2BS. In the same paper, the conditional stability constants

for RXHn- + Hg2+ ) RXHg(n-1)- + H+ were converted to those for RS- + Hg2+ ) RSHg+ using the reaction pH and the average value of 1010 for the protonation constant of organic thiol. The binding site concentration, 5 nmol mg-1 DOM, reported by Haitzer et al. (23, 37) was converted to 5-500 nM using the typical DOM concentrations (1-100 mg DOM L-1) in natural waters (2). The conditional stability constants were corrected using their pKa values obtained from modeling Hg(II)-DOM binding at different pH levels (23). The comparison in Figure 4 is similar to that used by Town and Fillela (40, 41) to assess the nature of metal binding ligands. The dataset in Figure 4 is consistent with previous studies, despite differences in methodologies and determination conditions (7.0 < pH < 9.8). The range of log conditional stability constants estimated by Haitzer et al. (23, 37) is somewhat higher than those of others. However, VOL. 39, NO. 17, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6613

typical DOM concentrations measured in Galveston Bay were 1-5 mg L-1, suggesting that the lowest limit of the ligand concentration range determined by Haitzer et al. should be compared to the data of this study. The continuous binding characteristics of dissolved organic ligands shown in Figure 4 agree with those of other metals (40, 41). This result suggests that stronger binding sites are used at lower metal concentrations, and progressively weaker binding sites complex with higher metal concentrations. The binding coefficients (RHgL ) [HgL]/[Hg2+]) calculated from Kcond′ × [L] in Figure 4 suggest that lower concentrations of binding sites with higher stability constants are more important for Hg speciation than are higher concentrations of weaker binding sites. Thus, much of the Hg-DOM binding determined at nM binding site concentrations may not be relevant for Hg(II) speciation in pristine aquatic environments with low Hg concentrations (up to 50 pM). This agrees with recent reports that noted that as the ratio of [Hg(II)]/[DOM] changes, so does the binding strength between Hg(II) and DOM (20, 37). These reports demonstrated that higher stability constants are determined at lower [Hg(II)]/[DOM] ratios, indicating that the binding between Hg(II) and DOM under natural conditions is controlled by a small fraction, including the reactive thiol functional group, of DOM molecules. Using the DOC concentrations in Tables 4 and 5, and assuming that approximately 50% of organic matter is carbon, KDOM as defined in Haitzer et al. (37; KDOM ) [HgDOM]/[Hg2+]/(DOM); (DOM) ) concentration of DOM in kg L-) was estimated for the Galveston Bay samples as follows: 23.2-23.6 L kg-1 for chloride competition, and 23.1-24.7 L kg-1 for TSA competition, with the Hg/DOM concentration ratios ranging from 1 × 10-4 to 4 × 10-4 µg Hg mg-1 DOM. The KDOM range determined in this study agrees with those observed by Haitzer et al. (log KDOM ) 23.2 ( 1.0 L kg-1) at Hg DOM-1 concentration ratios below 1 µg Hg mg-1 DOM (37). The importance of each class of DOM binding sites can be estimated using known stability constants and natural concentrations of model binding sites. Assessing formation constants for each class of model binding sites, including carboxylic, amino, sulfide, and thiolate (42), demonstrates that only sulfide and thiolate ligands fall within the range of log K in Figure 4. Assuming 0.5-2% of DOM is total sulfur concentration and 20% of total sulfur is reduced sulfur (42), total reduced sulfur concentration in Galveston Bay water is estimated to 50-800 nM. The lowest limit of the estimated range agrees with the experimental determination of reduced sulfur in filter-passing Galveston Bay waters, i.e., 20-60 nM (43). Reduced sulfur sites in DOM far exceed the amount of Hg available in natural waters, suggesting that only small fractions of reduced sulfur can actually complex with natural Hg(II). Little is known about the forms of reduced sulfur sites in DOM. While the pH dependence of conditional distribution coefficient of Hg(II)-DOM suggests importance of thiolate (-SR) (23), metal-sulfide (-SM) associated with DOM can be a important binding site for Hg(II). Smith et al. (44) demonstrated that Cr(II)-reducible sulfide, which is a valid surrogate for -SM binding site on DOM, correlates to total ligands related to strong binding constant for Ag(I), B-type metal like Hg(II). The importance of Cu(II) as a complexing metal for dissolved sulfide was suggested by thermodynamic calculation using conditional stability constants between sulfide and several metals (45, 46). Low nM levels of dissolved sulfide reported for surface waters of Galveston Bay (47) support the possible complexation of sulfide by dissolved Cu. Even if most dissolved sulfide is complexed by Cu, greater binding strength between Hg(II) and sulfide (48, 49) suggests that Cu(II)-SH associated with DOM can be a binding site for Hg(II). Recently, a metastable sulfide cluster has been measured in oxic and anoxic freshwater environments (50). 6614

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 17, 2005

This cluster may occur discretely, or in association with a metal co-bonded to another ligand in DOM. Thus Hg(II) may coordinate on the sulfur site of the CuS cluster, which coordinates with a -NH2 or -OOC in the DOM (44). Limitations of CLE-SSE. The CLE-SSE speciation method has several limitations in its ability to comprehensively describe ligand characteristics, as well as Hg speciation. First, small amounts (∼10%) of hydrophilic ligands that exist at the nM level in natural waters cannot be determined from chloride competition, and small amounts (∼10%) of hydrophobic ligands which exist at the pM level in natural waters cannot be determined from TSA competition. Second, the slopes of natural water titration curves were generally lower than those of UV-treated samples, suggesting that Hg(II) complexation results from a spectrum of organic ligands, rather than a discrete organic ligand. Third, the concentrations and binding strengths of organic ligands were determined for pH 7.0 and pH 10 instead of natural pH. To make corrections for a pH-independent stability constant requires making assumptions about the acid-base character of the Hg(II) binding ligand; careful assessment of this assumption is warranted. Fourth, lower concentrations of natural organic ligands that complex Hg(II) with higher binding strengths may exist. Higher competition strengths than those found using TSA competition are required to determine Hg(II) complexing organic ligands at natural Hg(II) levels.

Acknowledgments We give special thanks to Dr. Peter H. Santschi and Dr. Alexandru T. Balaban for helpful discussions. We also thank Ron Lehman and Dr. Key-Young Choe for technical advice. This research was supported by the Calfed Bay-Delta Mercury Project (USBR 99FC200241) and by a graduate fellowship from the Texas Institute of Oceanography, Texas A&M University, Galveston, Texas.

Supporting Information Available Tables of titration results and non-linear regression equations. This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Santschi, P. H. Factors controlling the biogeochemical cycles of trace elements in fresh and coastal marine waters as revealed by artificial radioisotopes. Limnol. Oceanogr. 1988, 33 (4), 848866. (2) Buffle, J. Complexation Reactions in Aquatic Systems: An Analytical Approach; Ellis Horwood: Chichester, U.K., 1990. (3) Santschi, P. H.; Lenhart, J. J.; Honeyman, B. D. Heterogeneous processes affecting trace contaminant distribution in estuaries: The role of natural organic matter. Mar. Chem. 1997, 58, 99-125. (4) Guentzel, J. L.; Powell, R. T.; Landing, W. M.; Mason, R. P. Mercury associated with colloidal material in an estuarine and an openocean environment. Mar. Chem. 1996, 55, 177-188. (5) Stordal, M. C.; Gill, G. A.; Wen, L.-S.; Santschi, P. H. Mercury phase speciation in the surface waters of three Texas estuaries: Importance of colloidal forms. Limnol. Oceanogr. 1996, 41 (1), 52-61. (6) Cai, Y.; Jaffe, R.; Jones, R. D. Interactions between dissolved organic carbon and mercury species in surface waters of the Florida Everglades. Appl. Geochem. 1999, 14 (3), 395-407. (7) Mierle, G.; Ingram, R. The role of humic substances in the mobilization of mercury from watersheds. Water, Air, Soil Pollut. 1991, 56, 349-357. (8) Watras, C. J.; Morrison, K. A.; Host, J. S.; Bloom, N. S. Concentration of mercury species in relationship to other sitespecific factors in the surface waters of northern Wisconsin lakes. Limnol. Oceanogr. 1995, 40 (3), 556-565. (9) Coquery, M.; Cossa, D.; Sanjuan, J. Speciation and sorption of mercury in two macro-tidal estuaries. Mar. Chem. 1997, 58, 213-227. (10) Bilinski, H.; Kwokal, Zˇ .; Plavsˇic, M.; Wrischer, M.; Branica, M. Mercury distribution in the water column of the stratified Krka

(11) (12) (13) (14)

(15)

(16) (17)

(18)

(19)

(20) (21)

(22) (23) (24)

(25) (26) (27) (28) (29) (30) (31)

river estuary (Croatia): importance of natural organic matter and of strong winds. Water Res. 2000, 34 (7), 2001-2010. Choe, K.-Y.; Gill, G. A.; Lehman, R. Distribution of particulate, colloidal, and dissolved mercury in San Francisco Bay estuary. 1. Total mercury. Limnol. Oceanogr. 2003, 48 (4), 1535-1546. Conaway, C. H.; Squire, S.; Mason, R. P.; Flegal, A. R. Mercury speciation in the San Francisco Bay estuary. Mar. Chem. 2003, 80, 199-225. Driscoll, C. T.; Yan, C.; Schofield, C. L.; Munson, R.; Holsapple, J. The mercury cycle and fish in the Adirondack Lakes. Environ. Sci. Technol. 1994, 28, 136A-143A. Driscoll, C. T.; Blette, V.; Yan, C.; Schofield, C. L.; Munson, R.; Holsapple, J. The role of dissolved organic-carbon in the chemistry and bioavailability of mercury in remote Adirondack lakes. Water, Air, Soil Pollut. 1995, 80, 499-508. Sjo¨blom, A.; Meili, M.; Sundbom, M. The influence of humic substances on the speciation and bioavailability of dissolved mercury and methylmercury, measured as uptake by Chaoborus Larvae and loss by volatilization. Sci. Total Environ. 2000, 261, 115-124. Lawson, L. M.; Mason, R. P. Factors comtrolling the bioaccumulation of mercury and methylmercury by the estuarine amphipod Leptocheirus. Environ. Pollut. 2001, 111, 217-231. Gorski, P. R.; Cleckner, L. B.; Hurley, J. P.; Sierszen, M. E.; Armstrong, D. E. Factors affecting enhanced mercury bioaccumulation in inland lakes of Isle Royale National Park, USA. Sci. Total Environ. 2003, 304, 327-348. Wu, Q.; Apte, S. C.; Batley, G. E.; Bowles, K. C. Determination of the mercury complexation capacity of natural waters by anodic stripping voltammetry. Anal. Chim. Acta 1997, 350, 129134. Lamborg, C. H.; Tseng, C.; 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 (15), 3316-3322. Skyllberg, U.; Xia, K.; Bloom, P. R. Binding of mercury(II) to reduced sulfur in soil organic matter along upland-peat soil transects. J. Environ. Qual. 2000, 29 (3), 855-865. Drexel, R. T.; Haitzer, M.; Ryan, J. N.; Aiken, G. R.; Nagy, K. L. Mercury (II) sorption to two Florida Evegrades peats: evidence for strong and weak binding and competition by dissolved organic matter released from the peat. Environ. Sci. Technol. 2002, 36, 4058-4064. Hsu, H.; Sedlak, D. L. Strong Hg(II) complexation in municipal wastewater effluent and surface waters. Environ. Sci. Technol. 2003, 37, 2743-2749. Haitzer, M.; Aiken, G. R.; Ryan, J. Binding of mercury(II) to aquatic humic substances: Influence of pH and source of humic substances. Environ. Sci. Technol. 2003, 37, 2436-2441. Benoit, J. M.; Mason, R. P.; Gilmour, C. C.; Aiken, G. R. Constants for mercury binding by dissolved organic matter isolates from the Florida Evergrades. Geochim. Cosmochim. Acta 2001, 65 (24), 4445-4451. Gill, G. A.; Bruland, K. W. Mercury speciation in surface freshwater systems in California and other areas. Environ. Sci. Technol. 1990, 24, 1392-1400. Choe, K.-Y.; Gill, G. A. Isolation of colloidal monomethyl mercury in natural waters using cross-flow ultrafiltration techniques. Mar. Chem. 2001, 76, 305-318. U.S. Environmental Protection Agency. Method 1631: Mercury in Water by Oxidation, Purge and Trap, and Cold Vapor Atomic Fluorescence Spectrometry; USEPA: Washington, DC, 2002. NIST. Database 46: NIST Critically Selected Stability Constants of Metal Complexes Database, Version 8.0; National Institute of Standards and Technology: Gaithersburg, MD, 2004. Stumm, W.; Morgan, J. J. Aquatic Chemistry; John Wiley & Sons: New York, 1996. Ruzic, L. Theoretical aspects of the direct titration of natural waters and its formation yield for trace metal speciation. Anal. Chim. Acta 1982, 140, 99-113. van den Berg, C. M. G. Determination of the complexing capacity and conditional stability constants of complexes of copper (II)

(32)

(33) (34) (35) (36)

(37) (38) (39) (40) (41) (42) (43) (44)

(45) (46) (47)

(48)

(49)

(50) (51)

with natural organic ligands in seawater by cathodic stripping voltammetry of copper-catechol complex ions. Mar. Chem. 1984, 15, 1-18. Bodini, M. E.; Valle, M. A. D. Redox chemistry and spectroscopy of 2-mercaptobenzoic acid and its manganese (II) and (III) complexes in dimethylsulphoxide. Polyhedron 1990, 9 (9), 11811186. Alniaimi, N. S.; Alsaadi, B. M. Thiosalicylic acid complexes of divalent zinc, cadmium, mercury and lead. J. Inorg. Nucl. Chem. 1974, 36 (7), 1617-1622. Koul, M. K.; Dubey, K. P. Potentiometric determination of the stability constants of bisthiosalicylato mercurate. J. Inorg. Nucl. Chem. 1973, 35, 2571-2573. van den Berg, C. M. G.; Nimmo, M.; Daly, P.; Turner, D. R. Effects of the detection window on the organic copper speciation in estuarine waters. Anal. Chim. Acta 1990, 232, 149-159. Miller, L. A.; Bruland, K. W. Determination of copper speciation in marine waters by competitive ligand equilibration/liquidliquid extraction: an evaluation of the technique. Anal. Chim. Acta 1994, 284, 573-586. Haitzer, M.; Aiken, G. R.; Ryan, J. N. Binding of mercury(II) to dissolved organic matter: The role of the mercury-to-DOM concentration ratio. Environ. Sci. Technol. 2002, 36, 3564-3570. Laglera, L. M.; van den Berg, C. M. G. Copper complexation by thiol compounds in estuarine waters. Mar. Chem. 2003, 82, 71-89. Averyt, K. B.; Kim, J. P.; Hunter, K. A. Effect of pH on measurement of strong copper binding ligands in lakes. Limnol. Oceanogr. 2004, 49 (1), 20-27. Town, R. M.; Fillela, M. Dispelling the myths: is the existence of L1 and L2 ligands necessary to explain metal ion speciation in natural water? Limnol. Oceanogr. 2000, 45 (6), 1341-1357. Town, R. M.; Fillela, M. Implications of natural organic matter binding heterogeneity on understanding lead(II) complexation in aquatic systems. Sci. Total Environ. 2002, 300, 143-154. Ravichandran, M. Interactions between mercury and dissolved organic mattersa review. Chemosphere 2004, 55 (3), 319-331. Tang, D.; Warnken, K. W.; Santschi, P. H. Organic complexaion of copper in surface waters of Galveston Bay. Limnol. Oceanogr. 2001, 46 (2), 321-330. Smith, D, S.; Bell, R. A.; Kramer, J. R. Metal speciation in natural waters with emphasis on reduced sulfur groups as strong metal binding sites. Comp. Biochem. Physiol., Part C: Pharmacol., Toxicol. Endocrinol. 2002, 133, 65-74. Dyrssen, D. Sulfide complexation in surface seawater. Mar. Chem. 1988, 24, 143-153. Al-Farawati, R.; van den Berg, C. M. G. Metal-sulfide complexation in seawater. Mar. Chem. 1999, 63, 331-352. Tang, D.; Santschi, P. H. Sensitive determination of dissolved sulfide in estuarine water by solid-phase extraction and highperformance liquid chromatography of methylene blue. J. Chromatogr. A 2000, 883, 305-309. Benoit, J. M.; Gilmour, C. C.; Mason, R. P.; Heyes, A. Sulfide controls on mercury speciation and bioavalibility to methylating bacteria in sediment porewater. Environ. Sci. Technol. 1999, 33, 951-957. Benoit, J. M.; Mason, R. P.; Gilmour, C. C. Estimation of mercurysulfide speciation in sediment pore waters using octanol-water partitioning and implications for availability to methylating bacteria. Environ. Toxicol. Chem. 1999, 18 (10), 2138-2141. Rozan, T. F.; Lassman, M. E.; Ridge, D. P.; Luther, G. W., III. Evidence for Fe, Cu and Zn complexation as multinuclear sulfide clusters in oxic river waters. Nature 2000, 406, 879-882. Yeh, S. J.; Lo, J. M.; Shen, L. H. Determination of extraction constants for metal diethyldithiocarbamates and chloride mixed metal diethyldithiocarbamates by substoichiometric extraction. Anal. Chem. 1980, 52, 528-531.

Received for review August 26, 2004. Revised manuscript received June 3, 2005. Accepted June 14, 2005. ES048667Z

VOL. 39, NO. 17, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6615