HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (18) Citing Articles via Google Scholar Google Scholar Articles by Sarkar, D. Articles by Misra, K.C. Search for Related Content PubMed Articles by Sarkar, D. Articles by Misra, K.C. GeoRef GeoRef Citation Agricola Articles by Sarkar, D. Articles by Misra, K.C.

DIVISION S-2-SOIL CHEMISTRY

Adsorption of Mercury(II) by Variable Charge Surfaces of Quartz and Gibbsite

^a Soil and Water Sci. Dep., Univ. of Florida, Gainesville, FL, 32611-0290 USA
^b Dep. of Plant and Soil Sci., The Univ. of Tennessee, Knoxville, TN 37901-1071 USA
^c Dep. of Geological Sci., The Univ. of Tennessee, Knoxville, TN 37996-1410 USA

messington{at}utk.edu

	ABSTRACT

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results and discussion Conclusions REFERENCES

 
The influence of pH, ionic strength, ligands (Cl, SO₄, PO₄), and metals (Ni and Pb) on the adsorption of Hg(II) by quartz and gibbsite was investigated to better understand the Hg(II) adsorption process and the impact of metals and ligands on this process. The triple layer model (TLM) was used to simulate Hg(II) adsorption on both surfaces. Mercury(II) adsorption from a 0.6 µM Hg(II) solution varies as a function of pH, increasing to an adsorption maximum with increasing pH before tailing off to a constant level at high pH values. The pH at which maximum Hg(II) adsorption occurs (pH_max 4.5) is comparable to the pK_a (3.2) for the hydrolysis of Hg²⁺ to form Hg⁰₂. Further, the Hg(II) adsorption edge shifts to much higher pH values in the presence of 0.001 M and 0.01 M Cl, which also corresponds to the pH at which Hg⁰₂ is predicted to form. Only minor deviations in the degree of adsorption and the shape of the Hg(II) adsorption edge are influenced by ionic strength, suggesting the formation of inner-sphere surface complexes. However, Hg(II) adsorption can only be successfully modeled with consideration of the formation of both an outer-sphere surface complex [XO^-–HgOH⁺] and an inner-sphere surface complex [XOHg^-₂]. Swamping concentrations (0.01 M) of SO₄ and PO₄ reduced Hg(II) adsorption on quartz, a result of the predicted formation of Hg(OH)₂SO^2-₄, Hg(OH)₂H₂PO^-₄, and Hg(OH)₂–HPO^2-₄ aqueous species (the adsorption edge and pH_max were not influenced). The presence of SO₄ also decreased Hg(II) retention by gibbsite, which was also attributed to the formation of the Hg(OH)₂SO^2-₄ ion pair; however, the presence of PO₄ increased Hg(II) retention by gibbsite, which was attributed to the formation of a phosphate bridge [AlOPO₃Hg^2-₂]. Mercury(II) adsorption was decreased in the presence of 14 µM Pb and 48 µM Ni, and most noticeably in the quartz system. The adsorption of Hg(II), when in competition with Pb or Ni, could not be simulated by the TLM without the reoptimization of the Hg(II) outer- and inner-sphere log K_int values. Intrinsic Hg(II) adsorption constants derived from single-element systems could not be employed to simulate adsorption in multi-element, competitive systems.

Abbreviations: IS, ionic strength • TLM, triple layer model

	INTRODUCTION

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results and discussion Conclusions REFERENCES

 
THE ADSORPTION OF METALS such as Cd, Zn, Cu, and Pb by soils and their mineral components has been extensively examined. However, because of its relatively low natural concentration, its unique physico-chemical properties (redox speciation), and the analytical difficulties associated with its measurement, the adsorption behavior of Hg(II) has not been extensively examined. Nevertheless, anthropogenic Hg contamination of natural systems has occurred from its extensive use in many industries (one notable example is Hg discharged during nuclear weapon production at the Y-12 Plant in Oak Ridge, Tennessee), and its ubiquitous occurrence in mine leachates (Jonasson and Boyle, 1971) and wastewater solids (Beszedits, 1979). Like other trace metals, the fate of Hg(II) in the environment is largely controlled by adsorption reactions with fixed or mobile adsorbents (Schindler, 1967). The most common inorganic adsorbents in natural aquatic systems are metal oxides (including hydroxides and oxyhydroxides) and clay minerals (James and Leckie, 1974; Dudley et al., 1988; Zachara et al., 1988, 1991). Among these, metal oxides have the highest affinity for ions in solution because of their charged, reactive hydroxyl surface sites coupled with their high specific surface areas. A detailed investigation of the degree and mechanism of Hg(II) adsorption reactions involving variable-charge surfaces is essential to understanding the fate and behavior of Hg(II) in the environment.

The influence of pH and the Cl^- ion on the adsorption of Hg(II) by a variety of mineral surfaces has been documented by several investigators. MacNaughton and James (1974) studied the adsorption of Hg(II) by SiO₂ and observed a sharp increase in Hg(II) adsorption between pH 2 and 3, with an adsorption maximum at a pH of 4. The adsorption edge shifted to higher pH values when Cl was present, leading to the conclusion that Hg(II) retention was associated with the formation and adsorption of Hg⁰₂. Lockwood and Chen (1974) studied the adsorption of Hg(II) by Fe(OH)₃ and also observed a sharp increase in retention in the pH 3 to 4 range. Chloride was found to inhibit Hg(II) adsorption at neutral pH, but not in highly alkaline systems. They also concluded that Hg⁰₂ was the major Hg(II) species adsorbed. Forbes et al. (1974) studied Hg(II) adsorption by goethite and observed an adsorption maxima at pH 5. They also correlated Hg(II) retention to the formation of Hg⁰₂. Numerous other investigators have observed the abrupt increase in Hg(II) retention between pH 2 and 4, with adsorption maxima between pH 4 and 5, and a shift in adsorption maxima to higher pH values when Cl is present, irrespective of adsorbent type (Newton et al., 1976; Kinniburgh and Jackson, 1978; Farrah and Pickering, 1978; Thanabalasingam and Pickering, 1985; Barrow and Cox, 1992a, 1992b; Yin et al., 1996). These studies have also found that Hg(II) adsorption is strongly correlated to the formation of the hydrolysis products HgOH⁺ and Hg⁰₂, leading to the conclusion that these aqueous species are preferentially adsorbed by mineral surface.

While the influence of pH and Cl on Hg(II) retention has been well established, the impact of metal cations and of other inorganic ligands has not been evaluated. The objectives of this study were to examine Hg(II) adsorption by quartz and gibbsite as a function of solution chemistry. Experimental variables include pH, ionic strength, and presence or absence of Ni, Pb, Cl, SO₄, and PO₄. Moreover, a surface complexation model, the triple layer model (TLM), was used to describe experimental data for the adsorption of Hg(II) by silica and gibbsite. This model was first proposed by Davies and Leckie (1978, 1980) as an extension of the site-binding model of Yates et al. (1974). Later, Hayes and Leckie (1987) revised the TLM to allow for both inner- and outer-sphere complexation of ions to better account for ionic strength effects. Mercury(II) adsorption on quartz and gibbsite in the presence of Pb, Ni, SO₄, PO₄, and Cl was also modeled using the TLM to more clearly elucidate a Hg(II) adsorption mechanism.

	Materials and methods

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results and discussion Conclusions REFERENCES

 
Preparation of Solids
Quartz (SiO₂) was prepared following the procedure of MacNaughton and James (1974). Natural, ground -quartz (MIN-U-SIL5, U.S. Silica, Berkeley Springs, WV) was heated at 500°C for 24 h to oxidize organic matter and then refluxed in boiling 4 M HNO₃ for 4 h to remove iron oxide impurities. This treatment was followed by repeatedly rinsing with deionized water until the pH of the supernatant was 6. The solid was dried in a 75°C oven and stored until needed. Superfine 4 gibbsite [Alcan Chemicals, Al(OH)₃] was treated for 30 min with 0.01 M NaOH to remove poorly crystallized coatings. Following this treatment, the solid was centrifuge-washed with 0.1 M NaNO₃ several times until the pH of the supernatant was 7. The slurry was stored wet at 4°C until needed.

Experimental Techniques
All reagents used were of analytical grade. A solution of 2.94 mM Hg(II) was prepared from Hg(NO₃)₂ · H₂O by dissolving the salt in a matrix of 10% HNO₃. The stock was diluted to prepare a working solution of 2.94 µM Hg(II). The background electrolyte solutions were 0.01 M, 0.1 M, and 0.5 M NaNO₃. The experimental procedure involved the equilibration of 0.1 g silica or gibbsite in 24 mL of background electrolyte solution with 6 mL of Hg(II) working solution in capped 50-mL Teflon centrifuge tubes (closed systems). This resulted in concentrations of 3.3 g L^-1 of solid and 0.6 µM Hg(II) (120 µg L^-1). Solution pH was adjusted with 0.02 M HNO₃ or 0.02 M NaOH, such that the equilibrium solutions had pH values ranging from 3 to 9 (pH measurements were performed before and after equilibration). Preliminary kinetic studies indicate that Hg(II) adsorption is characterized by a rapid initial uptake (within 1 h) followed by a much slower, continuous uptake. A 48-h reaction period was found to be sufficient to achieve equilibrium conditions when using a floor-shaker for continuous agitation. Triplicate equilibrations were performed at ambient temperature (20–25°C). The separation of the liquid from the solid phase was achieved by centrifugation at 1500 x g for 40 min. Filtration was not used in order to avoid the high loss of Hg(II) by adsorption on the 0.45-µm membrane filter and its support. After centrifugation, a 3-mL aliquot of the supernatant was withdrawn by pipette for Hg(II) analysis. During preliminary evaluations, the adsorption tube was emptied and rinsed with deionized water and refilled with 30 mL of 10% HNO₃. This tube was again shaken for 24 h and an aliquot of the acid rinse was withdrawn for analysis. This rinse step was used to account for Hg(II) retained by the Teflon centrifuge tubes. Results showed that the centrifuge tube walls did not retain Hg(II). Adsorbed Hg(II) was calculated from the difference between the Hg(II) initially added to the system and that remaining in the solution after equilibration. The dilutions induced by the pH controls were considered while computing the amount of Hg(II) adsorbed.

Mercury(II) adsorption in the presence of Cl, SO₄, and PO₄ was performed by equilibrating 0.1 g of solid in 21 mL of 0.1 M NaNO₃ background electrolyte, 6 mL of Hg(II) working solution, and 3 mL of a NaCl, Na₂SO₄, or a Na₂HPO₄ working solution (achieving 0.001 M or 0.01 M Cl, 0.001 M or 0.01 M SO₄, or 0.01 M PO₄) in capped, 50-mL Teflon centrifuge tubes. This resulted in concentrations of 3.3 g L^-1 of solid and 0.6 µM Hg(II). Solution pH was adjusted with 0.02 M HNO₃ or 0.02 M NaOH, such that the equilibrium solutions had pH values ranging from 3 to 9 (pH measurements were performed before and after equilibration). Mercury(II) adsorption in the presence of Pb and Ni was similarly performed in 0.1 M NaNO₃ background electrolyte by adding 3 mL of a Pb(NO₃)₂ or a Ni(NO₃)₂ stock solution (to achieve 14 µM Pb or 48 µM Ni) in place of the ligand stock solutions. Following the 48-h equilibration period, the tubes were centrifuged at 1500 x g for 40 min and the supernatant solutions analyzed. These experiments were performed in triplicate.

Analytical Techniques
For adsorption studies involving quartz, Hg(II) concentrations in solution were determined by the cold-vapor technique using 3% NaBH₄ in 1% NaOH as the reducing agent, and a Perkin Elmer model 5000 atomic absorption spectrophotometer, fitted with the MHS-10 hydride-generating unit (Perkin Elmer Corp., Norwalk, CT). For adsorption studies involving gibbsite, Hg(II) concentrations in solution were determined using a model FIMS-400 Perkin Elmer flow injection mercury system atomic absorption spectrometer. Dilution of the samples prior to analysis was required, and the addition of a few drops of 70% HNO₃ was necessary for generating steady signals. The pH of all the solutions was measured using a combination glass pH electrode and an Orion EA 940 ion analyzer (Orion Research, Boston, MA). Sodium, Pb, and Ni concentrations were determined using a Perkin Elmer model 5000 atomic absorption spectrophotometer. Nitrate, Cl, SO₄, and PO₄ concentrations were determined using ion chromatography (DX-100, Dionex Corp., Sunnyvale, CA).

Adsorption Modeling
The triple layer model was used to describe Hg(II) adsorption by quartz and gibbsite. The TLM is a surface coordination model developed by Davies and Leckie (1978, 1980) and later modified by Hayes and Leckie (1987) that models both inner- and outer-sphere surface complex formation of cationic, anionic, and neutral solute species. This model treats the solid–solution interface as composed of two layers of constant capacitance enveloped by a diffuse layer. The TLM uses measured or experimentally estimated surface parameters for the adsorbent and intrinsic equilibrium constants for surface reactions. The parameters are derived from experimental ion adsorption or titration data to calculate equilibrium surface and solution concentrations of the adsorptive and surface charge on the adsorbent. The surface parameters for SiO₂ and Al(OH)₃ used in the model simulations were obtained from the literature and are listed in Table 1 . Aqueous speciation reactions and associated equilibrium constants for Hg(II) species are listed in Table 2 . Surface hydrolysis constants and constants describing the complexation of the supporting electrolyte (NaNO₃) with the surfaces are listed in Table 3 .

View larger version (25K): [in this window] [in a new window]   Fig. 2 Predicted aqueous speciation of 0.6 µM Hg(II) in (a) 0.1 M NaNO3 and (b) 0.1 M NaNO3 + 0.01 M NaCl solutions using MINTEQ+ (Schecher and McAvoy, 1994) (associated constants listed in Table 2)

	Results and discussion

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results and discussion Conclusions REFERENCES

 
Effect of Ionic Strength and pH
The adsorption of Hg(II) by the variable-charge surfaces is pH dependent (Fig. 1) . Irrespective of ionic strength, Hg(II) adsorption increases with increasing pH, reaching an adsorption maximum at a pH_max of 4.3 for quartz and 4.8 for gibbsite. The percentages of Hg(II) adsorbed at the pH_max were 58% for silica and 55% for gibbsite. The increase in Hg(II) adsorption with increasing pH (to pH_max) is similar to the adsorption pattern of other hydrolyzable metal cations, whereby adsorption increases to its maximum value over a narrow pH range, giving rise to a sharp adsorption edge. This abrupt rise in the degree of adsorption is generally believed to be associated with the formation of positively charged metal–hydroxy species, which are assumed to have a very high affinity for the surface functional groups (Hodgson et al., 1964). Alternatively, it has been proposed that adsorbed hydroxide serves as a bridging ligand (Sposito, 1984), although this latter mechanism may not be appropriate in the pH range of the Hg(II) adsorption edge.

View larger version (17K): [in this window] [in a new window]   Fig. 1 Adsorption of Hg(II) (0.6 µM) by (a) quartz and (b) gibbsite (3.3 g L-1) as a function of pH and ionic strength (IS) (controlled by NaNO3)

(1)

This model was originally proposed by Davies and Leckie (1978) and successfully applied to describe retention of several specifically retained metal cations, such as Pb²⁺ (Hayes and Leckie, 1987), Cd²⁺ (Cowan et al., 1991), and Cu²⁺ and Zn²⁺ (Kooner et al., 1995). These results notwithstanding, Hg(II) adsorption by quartz and gibbsite is at a minimum when solution pH is consistent with the maximum expression of the Hg²⁺ (pH < 2.5) or the HgOH⁺ species (pH 3.1). Therefore, these surfaces do not appear to have an affinity for either Hg²⁺ or HgOH⁺ species, although the specific retention of HgOH⁺ may have a relatively minor impact on the overall adsorption of Hg(II) at pH values below pH_max.

The Hg(II) adsorption data (Fig. 1) indicates that the adsorption maxima correspond to the formation of Hg⁰₂. Indeed, the pH₅₀ values [pH value at which 50% of the Hg(II) adsorbed at the pH_max occurs] of 3.2 and 3.8 for silica and gibbsite are consistent with the p^cK_a value of 3.2 for the hydrolysis reaction (Baes and Mesmer, 1986). This observation, and the continued (but reduced) retention of Hg(II) as pH increases above pH_max, suggests that the Hg⁰₂ species is the principal Hg(II) adsorbate: a conclusion that is consistent with that of Forbes et al. (1974), MacNaughton and James (1974), Lockwood and Chen (1974), and Thanabalasingam and Pickering (1985). As solution pH increases above pH_max, Hg(II) adsorption decreases. This decrease in adsorption is gradual in the gibbsite system, reaching an adsorbed Hg(II) value that is a function of ionic strength, but invariant with pH. In the quartz system, adsorbed Hg(II) plateaus, then decreases sharply above a pH of 6, attaining an adsorbed Hg(II) value that is also a function of ionic strength and invariant with pH. These observations are consistent with those of MacNaughton and James (1974) and suggest the possible influence of surface properties, rather than solution conditions, on Hg(II) retention above pH_max.

According to Fig. 2a, the aqueous speciation of Hg(II) does not change above pH 5, where Hg⁰₂ is the predominant solution species. If it is assumed that Hg⁰₂ behaves as a ligand in the solution (Thanabalasingam and Pickering, 1985), then the decrease in the degree of adsorption when pH is greater than pH_max may be explained by the loss of exchangeable ligand (H₂O or OH^-) from the surface (assuming inner-sphere surface complexation). Irrespective of the system, the concentration of the silanol (SiOH⁰) group and the aluminol (AlOH⁰ and AlOH⁺₂) groups will decrease with increasing pH, with a concomitant increase in deprotonated groups . Although there is a decrease in exchangeable ligand on the surface, adsorbed Hg(II) achieves a constant value at high pH values. This process can be illustrated with a reaction of the type

(2)

The increased concentration of the OH^- ion in solution with increasing pH may also contribute to reduced Hg(II) retention, particularly if both ligands are competing for the same surface functional group. Further, aqueous OH^- concentrations exceed total Hg(II) when the solution pH increases above 7.7. Mattigod et al. (1985) investigated B adsorption by kaolinite and came to the conclusion that the decrease in B adsorption above pH 9.5 (where OH^-concentrations approached total B) was due to the competition between OH^- and B^-₄ for adsorption sites.

The specific adsorption of Hg⁰₂ is also supported by the small change in the pH-dependent adsorption with variation in ionic strength. For both quartz and gibbsite, the change in ionic strength had only a minor impact on the Hg(II) adsorption edge and pH_max. This observation is similar to one of MacNaughton and James (1974). Mercury(II) does not significantly complex with nitrate (Fig. 2a), and Na and nitrate undergo nonspecific adsorption by oxide surfaces (Sposito, 1984); thus, they are not likely to influence the specific retention of Hg(II). Clearly, the behavior of Hg(II) at the solid–solution interface appears to be quite similar to specifically adsorbed species, although a lack of response to ionic strength alone may not support such a mechanism (Manceau and Charlet, 1994).

Mercury(II) adsorption from 0.1 M NaNO₃ by quartz and gibbsite was simulated using the TLM and Eq. [3] and [4] (which are consistent with Eq. [1] and [2]), assuming inner-sphere surface complexation.

(3)

(4)

With this approach, the Hg(II) adsorption edge could be adequately simulated for both quartz and gibbsite using Eq. [3] only (Fig. 3a and 4a) . However, the model did not predict the sharp decrease in Hg(II) adsorption at higher pH values. Indeed, adsorption was predicted to remain constant in the quartz system or gradually decrease in the gibbsite system as pH increased above pH_max. Since aqueous speciation could not account for the decrease in Hg(II) adsorption, an alternative adsorption mechanism, involving outer-sphere complexation to form [XO^-–HgOH⁺], is proposed

(5)

along with inner-sphere complexation to form [XOHg^-₂] (described in Eq. [4]). However, because the adsorbed Hg(II) species in Eq. [4] is neutral, the TLM cannot distinguish between the formation of an outer- vs. an inner-sphere complex. An inner-sphere surface complex [XOHg^-₂] is assumed because of the high polarizability of Hg(II).

View larger version (57K): [in this window] [in a new window]   Fig. 3 Triple layer modeling of Hg(II) (0.6 µM) adsorption by quartz (3.3 g L-1) with the outer-sphere surface species [SiO-–HgOH+] and the inner-sphere surface species as a function of solution chemistry: (a) 0.1 M NaNO3; (b) 0.1 M NaNO3 + 0.01 M NaCl with the additional outer-sphere surface species [SiOHgOHCl-]; (c) 0.1 M NaNO3 + 0.01 M Na2SO4; (d) 0.1 M NaNO3 + 0.01 M Na3PO4 with the additional bridging surface species [SiOPO3Hg2-2]; (e) 0.1 M NaNO3 + 48 µM Ni(NO3)2; (f) 0.1 M NaNO3 + 14 µM Pb(NO3)2

View larger version (58K): [in this window] [in a new window]   Fig. 4 Triple layer modeling of Hg(II) adsorption by gibbsite (3.3 g L-1) with the outer-sphere surface species [AlO-–HgOH+] and the inner-sphere surface species [AlOHg-2] as a function of solution chemistry: (a) 0.1 M NaNO3; (b) 0.1 M NaNO3 + 0.01 M NaCl with the additional outer-sphere surface species [AlOHgOHCl-]; (c) 0.1 M NaNO3 + 0.01 M Na2SO4; (d) 0.1 M NaNO3 + 0.01 M Na3PO4 with the additional bridging surface species [AlOPO3Hg2-2]; (e) 0.1 M NaNO3 + 48 µM Ni(NO3)2; (f) 0.1 M NaNO3 + 14 µM Pb(NO3)2

Effect of Inorganic Ligands
The presence of ligands can increase, decrease, or have no effect on the adsorption of metals, depending on the properties of both the adsorptive and the adsorbent. A large number of Hg(II) adsorption studies that involve a variety of adsorbents have been performed in Cl systems. In the presence of Cl, the Hg(II) adsorption edge shifts to higher pH values, irrespective of the adsorbent (MacNaughton and James, 1974; Farrah and Pickering, 1978; Thanabalasingam and Pickering, 1985; Yin et al. 1996). In addition to Cl^-, ligands such as SO^2-₄, HPO^2-₄, and H₂PO^-₄ may also potentially impact Hg(II) retention.

In the presence of Cl, the Hg(II) adsorption edge (Fig. 5) is consistent with the formation of the Hg⁰₂ species (Fig. 2b). The pH_max was also independent of surface type: pH 8 in the 0.01 M Cl systems. Reducing the Cl concentration from 0.01 M to 0.001 M (quartz system only) shifted the Hg(II) adsorption edge and pH_max toward lower pH values. This adsorption shift is also consistent with the formation of Hg⁰₂ (aqueous speciation not shown). The percent Hg(II) adsorbed in the 0.01 M Cl systems at pH values above the pH_max are 35% and 24% for quartz and gibbsite, compared to 31% and 17% in the same pH range but in the absence of Cl. These data further support the conclusions of others (e.g., Yin et al., 1996) that Cl reduces Hg(II) retention through the formation of soluble Hg(II)–Cl species that have little to no affinity for the variable-charge surfaces.

View larger version (19K): [in this window] [in a new window]   Fig. 5 Adsorption of Hg(II) (0.6 µM) by (a) quartz and (b) gibbsite (3.3 g L-1) as a function of pH and in the presence of Cl, SO4, and PO4 [ionic strength (IS) is 0.1 (NaNO3)]

(6)

It has been suggested that the retention of [HgOHCl]⁰ may contribute to adsorbed Hg(II) (Yin et al., 1996). Because the sorbed species is neutral, the TLM cannot distinguish between outer- vs. inner-sphere complexation. We have elected to identify the surface complex as an inner-sphere species, using the same line of reasoning for the formation of the [XOHg^-₂] species. Using the log K_int values previously generated, and by optimizing the log K_int for [XOHgOHCl^-] formation, Hg(II) adsorption in both the quartz and gibbsite systems was accurately simulated (Fig. 3b and 4b). The optimized log K_int value for Eq. [6] in the quartz and gibbsite system is shown in Table 3. The TLM results illustrate that [XOHgOHCl^-] and [XOHg^-₂] dominate in the quartz system, while [XOHgOHCl^-] is of lesser importance in the gibbsite system.

The presence of 0.01 M SO₄ resulted in a decrease in Hg(II) adsorption at pH_max (Fig. 5). Adsorbed Hg(II) was reduced by 16% (58 to 44%) by the inclusion of 0.01 M SO₄ in the quartz system. A ten-fold dilution of the SO₄ (0.001 M) in the quartz system did not greatly impact Hg(II) adsorption relative to the 0.01 M SO₄ system (data not shown); although a slight shift in the Hg(II) adsorption edge to a lower pH value (0.3 pH units) and a slight increase in the percentage of Hg(II) adsorbed at the pH_max (2%) was noted. The adsorption of Hg(II) by gibbsite was reduced from 57 to 42% (15% decrease) in the presence of 0.01 M SO₄. Although SO₄ influenced the adsorption maxima, the pH_max was not impacted by the presence of SO₄: 4.7 in 0.01 M SO₄ and 4.5 in the absence of SO₄.

Thanabalasingam and Pickering (1985) observed a reduction in Hg(II) retention by manganese oxide in the presence of SO₄. They attributed decreased Hg(II) adsorption to the formation and subsequent precipitation of Hg(II)–SO₄ complexes. The predicted speciation of Hg(II) in the presence of 0.01 M SO₄ (using MINTEQ⁺, data not shown) indicates that only a minor amount [2% of total Hg(II)] of HgSO⁰₄ forms at pH values <3. Above a pH of 3.5, HgSO⁰₄ is not predicted to form in solution, and the predicted speciation of aqueous Hg(II) is identical to that illustrated in Fig. 2a. While these observations do not support the hypothesis of Thanabalasingam and Pickering (1985), it must be recognized that thermodynamic data describing the formation of Hg(II)–SO₄ complexes may not be complete. Indeed, only HgSO⁰₄ is considered by MINTEQ⁺. On the other hand, the reduced retention of Hg(II) in the presence of SO^2-₄ may be explained through the formation of basic Hg(II)–SO^2-₄ complexes {e.g., (HgOH–SO₄)^- and [Hg(OH)₂–SO₄]^2-}.

The adsorption results of He et al. (1997) tend to support an outer-sphere retention mechanism for the adsorption of SO^2-₄ by kaolinite. Hence, the reduction in Hg(II) adsorption might be due to sulfate ion competition with the various Hg(II) species for sorption sites. The inclusion of reactions that describe the formation of [XOH⁺–SO^2-₄] and [XOH⁺–SO₄H⁺] [with adjustable log K_int values] as components of the TLM could not account for the reduced adsorption of Hg(II) in the quartz or gibbsite Hg(II)–SO₄ systems. Indeed, the chemical models that included either one or both of the SO₄ surface species would not converge. Furthermore, the use of published values (He et al., 1997) as fixed parameters would still not account for the reduced Hg adsorption (actually simulated Hg adsorption was unaffected by modifications in sulfate adsorption parameters).

It was then assumed that the formation of an aqueous Hg(II)–SO^2-₄ species would be necessary to account for the reduced retention of Hg(II). As a result of this postulation and the TLM application, the [Hg(OH)₂ SO₄]^2- species was predicted to form. The inclusion of this species in the chemical model allowed for an acceptable description of the Hg(II) adsorption data (Fig. 3c) for quartz; however, the modeled adsorption of Hg(II) by gibbsite overestimated experimental retention when it employed the optimized aqueous association constant for [Hg(OH)₂SO₄]^2- presented in Table 2 (Fig. 4c). Figure 6a illustrates that reoptimization of the log K_int values for [XOHg^-₂] and [XO^-–HgOH⁺] (Table 3) is required to model the experimental Hg(II) adsorption data; reoptimization of the [Hg(OH)₂SO₄]^2- association constant alone resulted in an overestimation of Hg(II) adsorption by gibbsite.

View larger version (27K): [in this window] [in a new window]   Fig. 6 Triple layer modeling of Hg(II) adsorption by gibbsite (3.3 g L-1) with the outer-sphere surface species [AlO-–HgOH+] and the inner-sphere surface species [AlOHg-2] as a function of solution chemistry [reoptimized log Kint values]: (a) 0.1 M NaNO3 + 0.01 M Na2SO4; (b) 0.1 M NaNO3 + 48 µM Ni(NO3)2; and (c) 0.1 M NaNO3 + 14 µM Pb(NO3)2

Phosphate forms numerous inner-sphere oxide surface complexes [XOPO⁰₂], [XOPO₂OH^-], and [XOPO^2-₃] (Goldberg and Sposito, 1984). In the Hg(II)–PO₄–quartz system, phosphate was hypothesized to form an inner-sphere complex [SiOPO₂OH^-] via ligand exchange according to the reaction listed in Table 3. Similarly to the SO₄ system, reduced Hg(II) retention by quartz could be associated with the formation of aqueous Hg(II)–PO₄ species. Further, and although not evident in the quartz system, the enhanced retention of Hg(II) by gibbsite may also suggest a phosphate bridging mechanism; the observed Hg(II) adsorption behavior in the quartz system is influenced by both aqueous speciation and surface phosphate complexation of Hg(II). Consequently, log K_int values for the formation of [SiOPO₂OH^-] and [SiOPO₃Hg^2-₂], and the formation constants for the [Hg(OH)₂–H₂PO₄]^- and the [Hg(OH)₂–HPO₄]^2- aqueous species were the adjustable parameters used to model Hg(II) adsorption in the Hg(II)–PO₄–quartz system. The convergence of the TLM and the goodness-of-fit of the model curve to the experimental data (Fig. 3d) support the hypothesis that both surface and solution phosphate chemistry impact Hg(II) adsorption, with the formation of an aqueous Hg(II)–PO₄ species dominating.

For the Hg(II)–PO₄ gibbsite system, it was necessary to account for an increased retention of Hg(II), relative to adsorption in the absence of PO₄. Like the quartz system, it was postulated that competing surface and aqueous speciation reactions were responsible for the observed retention of Hg(II). When the log K_int values for the formation of [AlOPO^2-₃] and [AlOPO₃Hg^2-₂] were defined as adjustable parameters (and the formation constants for [Hg(OH)₂–H₂PO₄]^- and [Hg(OH)₂–HPO₄]^2- that were generated from the quartz system were used), the increased retention of Hg(II) by gibbsite was predicted (Fig. 4d). It is also evident that the [AlOPO₃Hg^2-₂] species is predicted to comprise a dominate fraction of the total Hg(II) adsorbed.

While it is clear that aqueous speciation influences Hg(II) adsorption in the Cl system, and potentially in the SO₄ and PO₄ systems, the mechanisms by which ligands impact Hg(II) retention depends on surface type. Preliminary evaluations (data not shown) indicated that 3.5% of the total SO₄ (0.01M) and 3% of the total PO₄ (0.01 M) is retained by quartz, irrespective of pH. Similarly, 7% of the total SO₄ and 9% of the total PO₄ is retained by gibbsite. For both solid systems, there was no appreciable change in ligand adsorption as a function of pH. Although these are small percentages relative to the total SO₄ and PO₄ present in the systems, adsorbed SO₄ and PO₄ overwhelm total Hg(II) and the total concentration of adsorption sites (Table 1). The observed reduction in Hg(II) adsorption in the presence of SO₄ is predominantly influenced by aqueous speciation. The fact that Hg(II) was still adsorbed in the presence of swamping PO₄ and SO₄ may be attributed to a high specificity of the surfaces for Hg(II) relative to the ligands. However, this evaluation does not hold for Hg(II) retention on gibbsite in the presence of PO₄. Apparently, the retention of PO₄ species enhances Hg(II) retention, perhaps by the surface complexation of Hg⁰₂ by specifically adsorbed PO₄ that results in a bridging surface complex.

Effects of Metal Cations
Studies that examine the competitive adsorption of one metal cation on another by Fe and Al oxides have been performed by several investigators (Benjamin and Leckie, 1981a, b; Farley et al., 1985). It has been demonstrated that the competitive effects are often very significant and have the capacity to alter the entire adsorption character of the metal. Therefore, an accurate prediction of Hg(II) retention by variable charge surfaces in natural and contaminated soils and water systems requires a comprehensive understanding of the competition among Hg(II) and other metal cations.

Both Ni and Pb have been reported to undergo strong specific retention by a variety of metal oxides and soils (Ellis and Knezek, 1972; Forbes et al., 1976; Benjamin and Leckie, 1981a, b; Hendrickson and Corey, 1981; Bowman et al., 1981; Elrashidi and O'Connor, 1982; Hayes and Leckie, 1987; Kooner, 1993). Correspondingly, the presence of small concentrations of Pb and Ni (14 µM and 48 µM, respectively) exerted a pronounced influence on the retention of Hg(II) by quartz (Fig. 7) . In the quartz system, the adsorption of Hg(II) in the presence of Pb and Ni was reduced at the pH_max by 38 and 31%. In the gibbsite system (Fig. 7), Hg(II) adsorption at the pH_max was reduced by 14 and 11% by Pb and Ni. The observation that Pb has a greater impact on Hg(II) adsorption than does Ni is not consistent with the concentrations of these metals. On the other hand, the adsorption edge (pH₅₀) for Pb is 6.1 and 6.6 in the quartz and gibbsite systems, while pH₅₀ for Ni is 7.1 and 7.0 (data not shown); thus, the greater adsorption of Pb, relative to Ni, below pH 7 may account for the relatively lower adsorption of Hg(II), irrespective of solid type. The surface functional group (SiOH⁰ vs. AlOH⁰) also influenced the impact of Pb and Ni on Hg(II) adsorption. As reported above, the presence of competing metals has a greater impact on Hg(II) retention through the SiOH⁰ group.

View larger version (19K): [in this window] [in a new window]   Fig. 7 Adsorption of Hg(II) (0.6 µM) by (a) quartz and (b) gibbsite (3.3 g L-1) as a function of pH and in the presence of Ni and Pb [ionic strength (IS) is 0.1 (NaNO3)]

In order to model Hg(II) adsorption on quartz and gibbsite in the presence of Pb and Ni, Hg(II) was assumed to form the [XO^-–HgOH⁺] and [XOHg^-₂] species, as described in Table 3. Lead and Ni were predicted to form the inner-sphere surface species: [XOPbOH⁰] and [XONiOH⁰]. The species formation reactions and associated log K_int values are listed in Table 3. The simulation approach was unable to describe the competition induced by Ni and Pb on Hg(II) adsorption (Fig. 3e, f and 4e, f). Indeed, the modeled adsorption of Hg(II) in the presence of these metals was almost indistinguishable from Hg(II) adsorption that was predicted in their absence. This is understandable, as below pH 6 only minor quantities of Ni and Pb are retained by the quartz and gibbsite surfaces (data not shown).

Using the Hg(II) adsorption data (in the presence of Pb or Ni), intrinsic constants for the Hg(II) adsorption reactions were reoptimized to produce a simulated adsorption envelope that more closely modeled the experimental data. For Hg(II) adsorption by quartz in the presence of both Ni and Pb (Fig. 8a, b) , the best fit was obtained by reoptimizing only the Hg(II) surface complexation constants (Table 3). As is evidenced by Fig. 8, this approach resulted in an acceptable fit of the experimental data.

View larger version (29K): [in this window] [in a new window]   Fig. 8 Triple layer modeling of Hg(II) adsorption by quartz (3.3 g L-1) with the outer-sphere surface species [SiO-–HgOH+] and the inner-sphere surface species [SiOHg-2] as a function of solution chemistry [reoptimized log Kint values]: (a) 0.1 M NaNO3 + 48 µM Ni(NO3)2; and (b) 0.1 M NaNO3 + 14 µM Pb(NO3)2

	Conclusions

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results and discussion Conclusions REFERENCES

 
The experimental and simulated results presented above question the validity of applying current representations of surface complexation models to binary systems. It was shown that the modeled adsorption of Hg(II) by quartz in the binary metal systems, which used adsorption constants derived from noncompetitive (single element) adsorption systems, grossly underestimated the impact of Ni and Pb. Benjamin and Leckie (1981a, b) studied competitive adsorption of Cd, Cu, Zn, and Pb on amorphous iron oxyhydroxide and obtained similar results. They concluded that if the metals compete for the same sites, then adsorption of a more strongly bound metal forces the less strongly bound metal to attach to weaker binding sites (multi-site hypothesis). In such cases, adsorption of the less strongly bound metal decreases when the more strongly bound metal is added.

Goldberg and Traina (1987) observed that single element adsorption constants over-predicted the competition between orthophosphate and selenite and between orthophosphate and silicate on the goethite surface. They argued that it was not possible to closely predict binary element sorption from single element adsorption constants because of site heterogeneity, which would cause some of the sites to selectively adsorb one element over another. Furthermore, as a result of competition, an element would become adsorbed to less energetic sites than are encountered during single element adsorption. To avoid this problem, they recommended that the constants be reevaluated or optimized using the competition data. In the present investigation, this approach worked well for describing Hg(II) adsorption onto quartz in the presence of Ni or Pb, but was less satisfying when applied to the gibbsite system. Nevertheless, there are several limitations associated with this approach (Cowan et al., 1991). This method is reasonable if the objectives of the modeling are to predict competition for similar experimental conditions. The new constants would then have limited value for predicting competition for other solution compositions. Furthermore, reoptimization would require that constants be derived for every solution composition of interest. Instead, if the objective of the modeling is to develop an approach for predicting multi-element adsorption from basic data, such as that describing the adsorption of single element, then another approach must be developed.

In the present study, a very different modeling result was obtained when gibbsite was the adsorbent, relative to when quartz was the adsorbent. Both competing metals (Pb and Ni) had a relatively minor impact on Hg(II) adsorption, and the competition data was modeled (when pH > pH_max) by slightly modifying the adsorption constants from single-adsorptive systems. This indicates that the nature of the surface has a significant impact on the adsorption of an element in the presence of another. Irrespective of the fact that Pb and Ni were present in higher concentrations, they were not preferred over Hg(II) for adsorption sites on gibbsite. Moreover, it should be noted that the first approach works satisfactorily when the competition is low (e.g., competitive studies on gibbsite), but it fails when the competition becomes intense (e.g., competitive studies on quartz). These limitations diminish the universal applicability of adsorption modeling, but at the same time they point to the importance of actual experimental or field data to adequately describe adsorption behavior of elements in systems with variable solution and solid phase chemistry.Ballistrieri Murry 1982

	ACKNOWLEDGMENTS

 
The authors express their appreciation to David Bass of the Perkin-Elmer Corporation for the use of a flow injection mercury spectrophotometer. Financial support from the Geological Society of America, and The University of Tennessee, Department of Geological Sciences, Institute of Agriculture, and Office of Academic Research is gratefully acknowledged.

	NOTES

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results and discussion Conclusions REFERENCES

 
Contribution from the Dep. of Plant and Soil Sciences, The Univ. of Tennessee.

Received for publication June 15, 1998.

	REFERENCES

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results and discussion Conclusions REFERENCES

 

• Baes C.F., Jr., Mesmer R.E. The hydrolysis of cations. Malabar, FL: Krieger, 1986.
• Ballistrieri L.S., Murry J.W. The adsorption of Cu, Pb, Zn, and Cd on goethite from major ion seawater. Geochim. Cosmochim. Acta 1982;46:1253-1265.
• Barrow N.J., Cox V.C. The effects of pH and chloride concentration on mercury sorption: I. By goethite. J. Soil Sci. 1992;43:295-304 a.
• Barrow N.J., Cox V.C. The effects of pH and chloride concentration on mercury sorption: II. By a soil. J. Soil Sci. 1992;43:305-312 b.
• Benjamin M.M., Leckie J.O. Multiple-site adsorption of Cd, Cu, Zn, and Pb on amorphous iron oxyhydroxide. J. Colloid Interface Sci. 1981;79:209-221 a.
• Benjamin M.M., Leckie J.O. Competitive adsorption of Cd, Cu, Zn, and Pb on amorphous iron oxyhydroxide. J. Colloid Interface Sci. 1981;83:410-419 b.
• Beszedits, S. 1979. Mercury removal from effluents and wastewater. In J.O. Nriagu (ed.) The biogeochemistry of mercury in the environment. Elsevier Press, New York.
• Bowman R.S., Essington M.E., O'Connor G.A. Soil sorption of nickel: Influence of solution composition. Soil Sci. Soc. Am. J. 1981;45:860-865.[Abstract/Free Full Text]
• Cowan C.E., Zachara J.M., Resch C.T. Cadmium adsorption on iron oxides in the presence of alkaline-earth elements. Environ. Sci. Technol. 1991;25:437-446.
• Davies J.A., Leckie J.O. Surface ionization and complexation at the oxide/water interface: II. Surface properties of amorphous iron oxyhydroxide and adsorption of metal ions. J. Colloid Interface Sci. 1978;67:90-107.
• Davies J.A., Leckie J.O. Surface ionization and complexation at the oxide-water interface: III. Adsorption of anions. J. Colloid Interface Sci. 1980;74:32-43.
• Dudley L.M., McLean J.E., Sims R.C., Jurinak J.J. Sorption of copper and cadmium from the water soluble fraction of an acid mine waste by two calcareous soils. Soil Sci. 1988;145:207-214.
• Ellis, B.G., and B.D. Knezek. 1972. Adsorption reactions of micronutrients. In J.J. Mortvedt et al. (ed.) Micronutrients in agriculture. SSSA Book Ser. 4, SSSA, Madison, WI.
• Elrashidi M.A., O'Connor G.A. Influence of solution composition on sorption of Zn by soil. Soil Sci. Soc. Am. J. 1982;46:1153-1158.[Abstract/Free Full Text]
• Farley K.J., Dzombak D.A., Morel F.M.M. A surface precipitation model for the sorption of cations on metal oxides. J. Colloid Interface Sci. 1985;106:226-242.
• Farrah H., Pickering W.F. The sorption of mercury species by clay minerals. Water Air Soil Pollut. 1978;9:23-31.
• Forbes E.A., Posner A.M., Quirk J.P. The specific adsorption of inorganic Hg(II) species and Co(III) complexes on goethite. J. Colloid Interface Sci. 1974;49:403-409.
• Forbes E.A., Posner A.M., Quirk J.P. The specific adsorption of divalent Cd, Co, Cu, Pb, and Zn on goethite. J. Soil Sci. 1976;27:154-166.
• Goldberg S., Sposito G. A chemical model of phosphate adsorption by soils: Reference oxide minerals. Soil Sci. Soc. Am. J. 1984;48:772-778.[Abstract/Free Full Text]
• Goldberg S., Traina S.J. Chemical modeling of anion competition on oxides using the constant capacitance model–mixed-ligand approach. Soil Sci. Soc. Am. J. 1987;51:929-932.[Abstract/Free Full Text]
• Hayes K.F., Leckie J.O. Modeling ionic strength effects on cation adsorption at hydrous oxide/solution interfaces. J. Colloid Interface Sci. 1987;115:564-572.
• He L.M., Zelazny L.W., Baligar V.C., Ritchey K.D., Martens D.C. Ionic strength effects on sulfate and phosphate adsorption on -alumina and kaolinite: Triple layer model. Soil Sci. Soc. Am. J. 1997;61:784-793.[Abstract/Free Full Text]
• Hendrickson L.L., Corey R.B. Effect of equilibrium metal concentrations on apparent selectivity coefficients of soil complexes. Soil Sci. 1981;131:163-171.
• Herbelin, A., and J. Westall. 1996. FITEQL. A computer program for determination of chemical equilibrium constants from experimental data. Version 3.2. Rep. 96-01. Dep. of Chemistry, Oregon State Univ., Corvallis.
• Hodgson J.F., Tiller K.G., Fellows M. The role of hydrolysis in the reaction of heavy metals with soil-forming materials. Soil Sci. Soc. Am. Proc. 1964;28:42-46.
• Huang, C.P. 1981. The surface acidity of hydrous solids. In M.A. Anderson and A.J. Rubin (ed.) Adsorption of inorganics at solid–liquid interfaces. Ann Arbor Science, Ann Arbor, MI.
• James, R.O., and J.O. Leckie. 1974. Control mechanisms for trace elements in natural systems. In E. Rubin (ed.) Aqueous-environmental chemistry of metals. Ann Arbor Science, Ann Arbor, MI.
• Johnson L.M. Development of stability constants for the sorption of uranium to gibbsite: A surface complexation approach. Clemson, SC: Clemson Univ, 1995 M.S. Thesis..
• Jonasson, I.R., and R.W. Boyle. 1971. Geochemistry of mercury. In J.E. Watkin (ed.) Mercury in man's environment. Proc. R. Soc. Can. Symp., Ottawa, Canada.
• Katz L.E., Hayes K.F. Surface complexation modeling: I. Strategy for modeling monomer complex formation at moderate surface coverage. J. Colloid Interface Sci. 1995;170:477-490.
• Kinniburgh D.G., Jackson M.L. Adsorption of mercury(II) by iron hydrous oxide gel. Soil Sci. Soc. Am. J. 1978;42:45-47.[Abstract/Free Full Text]
• Kooner Z.S. Comparative study of the adsorption behavior of copper, lead, and zinc onto goethite in aqueous systems. Environ. Geol. 1993;21:242-250.
• Kooner Z.S., Cox C.D., Smoot J.L. Prediction of adsorption of divalent heavy metals at the goethite/water interface by surface complexation modeling. Environ. Toxicol. Chem. 1995;14:2077-2083.
• Lockwood R.A., Chen K.Y. Adsorption of Hg(II) by ferric hydroxide. Environ. Lett. 1974;6:151-166.[ISI][Medline]
• MacNaughton M.G., James R.O. Adsorption of aqueous mercury(II) complexes at the oxide/water interface. J. Colloid Interface Sci. 1974;47:431-440.
• Manceau A., Charlet L. The mechanism of selenite adsorption on goethite and hydrous ferric oxide. J. Colloid Interface Sci. 1994;168:87-93.
• Mattigod S.V., Frampton J.A., Lim C.H. Effect of ion pair formation on boron adsorption by kaolinite. Clays Clay Miner. 1985;33:433-457.[Abstract]
• Meng X., Letterman R.D. Modeling ion adsorption on aluminum hydroxide modified silica. Environ. Sci. Technol. 1993;27:1924-1929.
• Newton D.W., Ellis R., Paulsen G.M. Effect of pH and complex formation on mercury(II) adsorption to bentonite. J. Environ. Qual. 1976;5:251-254.[Abstract/Free Full Text]
• Riese A.C. Adsorption of radium and thorium onto quartz and kaolinite: A comparison of solution/surface equilibria models. Boulder, CO: Colorado School of Mines, 1982 Ph.D. Thesis..
• Schecher, W.D., and D.C. McAvoy. 1994. MINEQL⁺: A chemical equilibrium program for personal computers. User's manual. Version 3.0. Environ. Res. Software, Hallowell, ME.
• Schindler, P.W. 1967. Heterogeneous equilibria involving oxides, hydroxides, carbonates, and hydroxide carbonates. In W. Stumm (ed.) Equilibrium concepts in natural water systems. Advances in Chemistry Ser. 67, Am. Chem. Soc., Washington, DC.
• Smith R.M., Martell A.E. Critical stability constants. Volume 4: Inorganic complexes. New York: Plenum Press, 1976.
• Sposito G. The surface chemistry of soils. New York: Oxford Univ. Press, 1984.
• Thanabalasingam P., Pickering W.F. Sorption of mercury(II) by manganese (IV) oxide. Environ. Pollut. Ser. B 1985;10:115-128.
• Yates D.E., Levine S., Healy T.W. Site bonding model of the electrical double layer at the oxide/water interface. J. Chem. Soc. Faraday Trans. 1974;170:1807-1818.
• Yin Y., Allen H.E., Li Y., Huang C.P., Sanders P.F. Adsorption of mercury by soil: Effects of pH, chloride, organic matter. J. Environ. Qual. 1996;25:837-844.[Abstract/Free Full Text]
• Zachara J.M., Cowan C.E., Schmidt R.L., Ainsworth C.C. Chromate adsorption by kaolinite. Clays Clay Miner. 1988;36:317-326.[Abstract]
• Zachara J.M., Cowan C.E., Resch C.T. Sorption of divalent metals on calcite. Geochim. Cosmochin. Acta 1991;55:1549-1562.

This article has been cited by other articles:

				M. E. Essington and R. M. Anderson Competitive Adsorption of 2-Ketogluconate and Inorganic Ligands onto Gibbsite and Kaolinite Soil Sci. Soc. Am. J., May 1, 2008; 72(3): 595 - 604. [Abstract] [Full Text] [PDF]

				M. E. Essington, J. B. Nelson, and W. L. Holden Gibbsite and Goethite Solubility: The Influence of 2-Ketogluconate and Citrate Soil Sci. Soc. Am. J., June 2, 2005; 69(4): 996 - 1008. [Abstract] [Full Text] [PDF]

M. Arias, M. ARIAS, M. T. BARRAL, J. DA SILVA-CARVALHAL, J. C. MEJUTO, and D. RUBINOS Interaction of Hg(II) with kaolin-humic acid complexes Clay Minerals, March 1, 2004; 39(1): 35 - 45. [Abstract] [Full Text] [PDF]