SSSAJ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 1 May 2008
Published in Soil Sci Soc Am J 72:595-604 (2008)
DOI: 10.2136/sssaj2007.0190
© 2008 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Essington, M. E.
Right arrow Articles by Anderson, R. M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Essington, M. E.
Right arrow Articles by Anderson, R. M.
Agricola
Right arrow Articles by Essington, M. E.
Right arrow Articles by Anderson, R. M.
Related Collections
Right arrow Soil Surface Chemistry
Right arrow Geochemical Processes
Right arrow Soil Chemistry

SOIL CHEMISTRY

Competitive Adsorption of 2-Ketogluconate and Inorganic Ligands onto Gibbsite and Kaolinite

Michael E. Essington* and Robert M. Anderson

Biosystems Engineering and Soil Science Dep., 2506 E.J. Chapman Dr., Univ. of Tennessee, Knoxville, TN 37996-4531

* Corresponding author (messington{at}utk.edu).


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The low-molecular-mass organic acid anion 2-ketogluconate (kG) is produced via microbial activity in rhizosphere soils. One of the mechanisms by which this organic ligand may influence the chemistry of soil systems is through adsorption by constant-potential minerals. This study examined the adsorption of kG onto gibbsite and kaolinite in the presence or absence of PO4, AsO4, and SO4 as a function of pH and ionic strength. The adsorption edge studies were performed in the pH 3 to 10 range and at ambient (20–22°C) temperatures. The adsorption of kG is a function of solution pH (decreasing with increasing pH) and independent of solution ionic strength, supporting the conclusion that kG is adsorbed by ligand exchange mechanisms. The adsorption of kG was decreased at all pH values in the presence of PO4 and AsO4, and was not significantly affected by the presence of SO4 at pH values >5. The decrease in kG adsorption in the presence of AsO4 and PO4 is further evidence that kG is adsorbed via specific retention mechanisms. The addition of kG to gibbsite containing preadsorbed PO4 did not result in PO4 displacement, regardless of the concentration of kG. Ligand adsorption was modeled using the adsorption edge data and the 1-pK basic Stern surface complexation model. The kG adsorption data was described by the formation of two inner-sphere surface species: mononuclear monodentate and binuclear bidentate. The chemical models and associated intrinsic equilibrium constants developed to describe ligand adsorption in single-adsorbate gibbsite systems were used to predict ligand retention in the kaolinite and binary-adsorbate systems. In this manner, the competitive adsorption of all ligands as a function of pH was adequately described. The findings of this study indicate that kG is specifically retained by common soil minerals and may impact the availability of PO4 and other specifically retained ligands in the rhizosphere.

Abbreviations: kG, 2-ketogluconate • LMMOA, low-molecular-mass organic acid • WSOS/DF, weighted sum of squares of residuals divided by the degrees of freedom


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Organic acid root and microbial exudates are increasingly recognized for their role in rhizosphere chemistry, specifically relative to their ability to increase nutrient availability and to detoxify metals. A wide variety of exudates exist in the soil solution, ranging in character from formic acid to complex, multi-ring phenolic acids. Typically, soil solutions are dominated by aliphatic acid anions (e.g., formate, acetate, lactate, oxalate, malonate, malate, succinate, and citrate), which are found in the <1 to 1100 µmol L–1 concentration range (Jones, 1998; Strobel, 2001; van Hees et al., 2000; Sulyok et al., 2005; Henry et al., 2007). Both plant roots and soil microbes are known to exude greater concentrations of organic acid anions (primarily citrate and malate) when subjected to Fe, P, and K (and possibly Ca and Zn) deficiencies, as a mechanism to mitigate Al toxicity effects, and in response to drought stress (Hocking, 2001; Inskeep and Silvertooth, 1988; Jones et al., 2003; Henry et al., 2007).

Organic acids influence the bioavailability and transport of metals via aqueous complexation reactions; however, their impact on ligand fate and behavior has also been established. Many organic acid ligands in soils (principally the di- and tricarboxylates) reside almost completely in the adsorbed phase (Jones, 1998; Strobel, 2001). They form specific (inner sphere) complexes with mineral surface functional groups, affecting surface charge characteristics (Yao and Yeh, 1996), compete with other specifically retained substances for surface sites (Geelhoed et al., 1998; Grafe et al., 2002; Kafkafi et al., 1988; Wijnja and Schulthess, 2000), and inhibit mineral crystallization (Huang and Violante, 1986; Jardine and Zelazny, 1996; Lebron and Suarez, 1999).

2-Ketogluconate (C6H9O7) is a low-molecular-mass organic acid (LMMOA) anion produced extracellularly as a microbial byproduct of glucose oxidation (Fuhrer et al., 2005). It has been isolated from the rhizosphere of several plants, including wheat (Triticum aestivum L.), corn (Zea mays L.), and pea (Pisum sativum L.) (Duff et al., 1963; Moghimi et al., 1978; Vance et al., 1996), and the production of kG has been shown to increase under conditions of P deficiency (Chiyonobu et al., 1973; Erlich, 1981; Halder and Chakrabartty, 1993; Klasen et al., 1992; Kucey et al., 1989; Neijssel and Tempest, 1975; Sokatch, 1969, p. 117–119; Webley et al., 1963; Webley and Duff, 1965). Like other LMMOAs (e.g., oxalate, malate, and citrate), kG enhances the solubility of soil minerals. Indeed, kG was as effective as citrate in enhancing the solubility of both gibbsite and goethite, reportedly through the effective complexation of aqueous Al3+ and Fe3+ (Essington et al., 2005).

The carboxylic acid moiety of kG is a relatively strong weak acid, with a pKa1 of 3.00 (Nelson and Essington, 2005). A second acidic functional group, probably the alcohol moiety in the third position (ortho to the carbonyl), has an acidity described by a pKa2 of 11.97. In "typical" soil solutions, the monovalent anion is predicted to predominate. The dissociable alcohol moiety, however, may be involved in aqueous metal complexation and surface ligand exchange reactions when solution pH values are substantially lower than pKa2, assuming a structural analogy to gluconic acid (Motekaitis and Martell, 1984; Essington et al., 2005).

In order for kG to enhanced P bioavailability in rhizosphere soil, it must be capable of competing with PO4 ions for adsorption sites, particularly those associated with constant-potential mineral surfaces. In short, kG must participate in inner sphere surface complexation reactions. The objectives of this study were to investigate the influence of pH and ionic strength on kG adsorption by gibbsite [Al(OH)3(s)] and kaolinite [Al2Si2O5(OH)4(s)], to describe the impact of competing inorganic ligands (PO4, AsO4, and SO4) on the kG adsorption edges, and to utilize a surface complexation model to elucidate the kG adsorption mechanisms.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Preparation of Solids
Hydrated alumina (SF-4) was obtained from Alcan Chemicals (Beachwood, OH). X-ray diffraction indicated SF-4 to be composed of monoclinic gibbsite without detectable impurities. Samples of SF-4 gibbsite were treated with CO2–free, 0.01 mol L–1 NaOH for 30 min to remove poorly crystalline Al(OH)3 (Sarkar et al., 1999). Following the base treatment, the gibbsite was centrifuge washed with either 1 or 10 mmol L–1 NaCl background electrolyte until a pH-neutral suspension was obtained. Well-crystallized Georgia kaolinite (KGa-1, from the Source Clays Repository of the Clay Minerals Society, West Lafayette, IN) was prepared following the procedure of Mattigod et al. (1985). A Type I water (18 {Omega}) plus kaolinite suspension was dispersed in a blender for 45 min. The suspension was then adjusted to pH 9.5 with CO2–free, 0.1 mol L–1 NaOH, and the <2.0-µm size fraction was obtained by Stoke's Law centrifugal sedimentation. The size separate was rinsed with 1 mol L–1 NaCl and adjusted to pH 3 with HCl to facilitate flocculation. Following this treatment, the kaolinite was centrifuge washed with either 1 or 10 mmol L–1 NaCl background electrolyte until the pH of the supernatant was 6. The gibbsite and kaolinite suspensions were brought to a solid/solution ratio of 50 g L–1 with background electrolyte and stored under N2 until needed.

Preparation of Solutions
All reagents were analytical grade or better. 2-Keto-D-gluconate was obtained as a hemicalcium dihydrate salt (Ca0.5C6H9O7·2H2O, 98% purity) and used to prepare a 0.1 mol L–1 Ca0.5kG solution. A stock solution of 0.1 mol L–1 Na-2-keto-D-gluconate (NakG) was prepared by a passing the Ca0.5kG solution through Na-saturated cation exchange resin (DOWEX HCR-W2, Dow Water Solutions, Midland, MI). Stock solutions of 0.1 mol L–1 AsO4, 0.1 mol L–1 PO4, and 0.1 mol L–1 SO4 were prepared in CO2–free Type I water from their Na salts (Na2HAsO4·7H2O, Na2HPO4·7H2O, and Na2SO4). These 0.1 mol L–1 stock solutions were then diluted to 4 mmol L–1 working solutions with Type I water. Also, combined ligand working solutions of 4 mmol L–1 kG + 4 mmol L–1 AsO4, 4 mmol L–1 kG + 4 mmol L–1 PO4, and 4 mmol L–1 kG + 4 mmol L–1 SO4 were prepared from the 0.1 mol L–1 stock solutions for competitive adsorption studies. The background electrolyte solutions were 0.01 and 0.001 mol L–1 NaCl prepared from solid NaCl. Solutions of 0.1 and 0.01 mol L–1 HCl and 0.1 and 0.01 mol L–1 NaOH were prepared from J.T. Baker DILUT-IT analytical concentrates (Mallinckrodt Baker, Phillipsburg, NJ) and CO2–free Type I water.

A Dionex (Sunnyvale, CA) DX-100 ion chromatograph equipped with a 25-µL injection loop, an AS4A analytical column, an AG4A guard column, suppressed conductivity detection, a 1.8 mmol L–1 Na2CO3–1.7 mmol L–1 NaHCO3 eluent, and a flow rate of 1.5 mL min–1 was used to determine the aqueous concentrations of kG, PO4, AsO4, and SO4. The retention times for the analytes were 1.15 min for kG, 3.85 min for PO4, 5.88 min for SO4, and 6.4 min for AsO4. The method detection limit for all analytes was 1.0 µmol L–1 and sample replicates and standard checks were within ±10%. The pH measurements were performed using a Thermo Orion Ross series combination pH electrode (Thermo Fisher Scientific, Waltham, MA) calibrated using pH 4, 7, and 10 commercial buffer solutions (concentration scale). The pH readings were recorded when a <0.1 mV min–1 drift was observed.

Adsorption Edge Experiments
Batch equilibrium adsorption studies were performed in triplicate using 50-mL polypropylene centrifuge tubes. The 50 g L–1 solid suspensions of gibbsite or kaolinite were vigorously shaken and then a 4-mL aliquot (containing 0.2 g of solid) was transferred by volumetric pipette to each centrifuge tube. To this was added 15.5 mL of NaCl background electrolyte, followed by additions of HCl or NaOH to achieve a pH range of approximately 3 to 10. The tubes were then capped, vortexed, and 0.5 mL of the 4 mmol L–1 adsorptive working solution was added by volumetric pipette. This resulted in a 10 g L–1 solid/solution ratio and a 0.1 mmol L–1 initial concentration of adsorptive. For the control systems, no solid suspension was used, and 19.5 mL of the NaCl background electrolyte was added to each reaction vessel in addition to the 0.5 mL of adsorptive solution and pH adjustors. The tubes were again vortexed and placed on a platform shaker for 72 h at ambient temperature (20–22°C). The 72-h equilibration period was determined from preliminary studies. After equilibration, the liquid and solid phases were separated by centrifugation at 1000 x g for 20 min. A 5-mL aliquot of the supernatant was withdrawn by pipette for ligand analysis, and the remaining supernatant was used for pH determinations.

A series of adsorption edge studies was also performed to investigate the ability of kG to displace adsorbed PO4, and vice versa, from gibbsite. Single-adsorbate adsorption edge experiments for kG and PO4 were performed as described above in 0.01 mol L–1 NaCl. Following the centrifugal separation of the liquid and solid phases, however, a 0.5-mL aliquot of the equilibrium solution was withdrawn and replaced with a 0.5-mL aliquot containing the counter ligand, resulting in a counter ligand concentration of 0.1 mmol L–1 PO4 (for preadsorbed kG systems), or 0.1, 0.2, or 0.4 mmol L–1 kG (for preadsorbed PO4 systems). The tubes were then vortexed, placed on a platform shaker, and equilibrated for an additional 72-h period. After equilibration, the supernatant liquid and gibbsite were separated by centrifugation, and kG, PO4, and pH were determined as described above.

Data Analysis and Surface Complexation Modeling
The initial concentrations of each ligand in the adsorption systems were determined through analyses of the control systems (adsorbents absent). The equilibrium solution concentrations of each ligand in the presence of adsorbents were determined through the analyses of the supernatant solutions following the 72-h equilibration period. The concentration of adsorbed ligand, or surface excess (q, in mmol kg–1), was determined by the difference between the initial and equilibrium solution ligand concentrations. Adsorption edge plots (q vs. pH) were constructed for all systems using the negative logarithm of the mean proton concentration and the mean q data from the triplicate adsorption systems. Standard error values for pH and q were also computed; however, the ranges in SE about each mean data point were always smaller than the marker used to represent the q vs. pH data on an adsorption edge plot.

The basic Stern formulation of the 1-pK surface complexation model (SCM) coupled with the q vs. pH adsorption edge data was used to develop chemical models that describe ligand (kG, PO4, AsO4, and SO4) adsorption (Westall and Hohl, 1980; Lützenkirchen, 1998; Rietra et al., 1999; Olsson et al., 2003). The SCM computations were conducted using FITEQL 4.0 software (Herbelin and Westall, 1999). The SCM requires values for surface parameters of the adsorbent (e.g., specific surface, capacitance), formation constants for all aqueous species, and intrinsic equilibrium constants for all surface complexes considered in the chemical model, including surface hydrolysis and electrolyte adsorption constants. The solid and suspension characteristics used by the SCM are given in Table 1 . Formation constants for aqueous species were obtained from the literature and adjusted for ionic strength using the Davies equation (Table 2 ). Parameters describing the chemical processes at the solid–solution interface, including surface hydrolysis and counterion complexation reactions for the 0.001 and 0.01 mol L–1 NaCl systems were also obtained from literature sources (Tables 3 and 4 ), with the exception of the gibbsite aluminal ({equiv}AlOH) protolysis constant (1-pK model), which was determined by potentiometric titration. For gibbsite and kaolinite, it was assumed that the singly coordinated aluminal surface functional groups were the only reactive functional groups relative to ligand exchange on the solid surfaces. The doubly coordinated surface functional groups on gibbsite and kaolinite and the silanol groups on kaolinite were considered nonreactive relative to ligand exchange.


View this table:
[in this window]
[in a new window]

  		Table 1. Solid and suspension properties used in the 1-pK basic Stern surface complexation modeling of 2-ketogluconate, PO4, AsO4, and SO4 adsorption by gibbsite and kaolinite.

 

View this table:
[in this window]
[in a new window]

  		Table 2. Aqueous speciation reactions and associated equilibrium constants (K, at 25°C) used in modeling the adsorption of 2-ketogluconate (kG), PO4, AsO4, and SO4 by gibbsite and kaolinite.{dagger}

 

View this table:
[in this window]
[in a new window]

  		Table 3. The 1-pK basic Stern model surface complexation reactions and associated intrinsic equilibrium constants (Kint) for ligand adsorption by gibbsite.

 

View this table:
[in this window]
[in a new window]

  		Table 4. The 1-pK basic Stern model surface complexation reactions and associated intrinsic equilibrium constants (Kint) for ligand adsorption by kaolinite.

 
The surface complexation of kG by the aluminal functional group may occur via any combination of inner and outer sphere mechanisms. Some of the potential kG adsorption mechanisms are illustrated in Fig. 1 . Surface complexation constants for ligand adsorption were fit simultaneously to the adsorption data at the two ionic strengths. The surface species that were ultimately used to describe ligand adsorption were selected according to a number of criteria. FITEQL optimizes the intrinsic equilibrium constants for user-defined surface complexation reactions by combining a nonlinear least squares routine with the chemical model that describes solid–solution interface speciation, mineral surface parameters, and aqueous speciation under mass- and charge-balance constraints. A goodness-of-fit parameter calculated by FITEQL, defined as the weighted sum of squares of residuals divided by the degrees of freedom (WSOS/DF), is a measure of the overall variance associated with the model predictions. Values of WSOS/DF between 0.1 and 20 generally indicate that the user-defined chemical model adequately describes the ligand adsorption data when using the FITEQL default error parameters. These error values are a function of the goodness-of-fit of the chemical model as well as the standard deviation in the experimental data (defined as the analytical error). The goal of the modeling was to find a chemical model with the least number of surface species (simplest model) that generated the lowest values of WSOS/DF, that was applicable to both ionic strength conditions, and that was capable of predicting adsorption by multiple surfaces and in binary-adsorbate systems. For example, surface species used to model kG adsorption to gibbsite in 0.01 mol L–1 NaCl must also be capable of modeling kG adsorption to gibbsite in 0.001 mol L–1 NaCl, as well as to model kG adsorption to gibbsite in the presence of other ligands, and to provide a reasonably good description of adsorption by kaolinite.


Figure 1
View larger version (31K):
[in this window]
[in a new window]

  		Fig. 1. Schematic representation of possible 2-ketogluconate (kG) surface species according to the 1-pK basic Stern surface complexation model.

 
The chemical model and the FITEQL-optimized intrinsic constants that describe the adsorption of kG, PO4, AsO4, and SO4 to gibbsite (single-ligand adsorption) were used to predict adsorption of kG in the presence of PO4, AsO4, or SO4 to gibbsite (binary-ligand adsorption), as well as to predict single- and binary-ligand adsorption to kaolinite. In these systems, the goodness-of-fit parameter generated by FITEQL (WSOS/DF) is an indication of the accuracy of the chemical models to predict adsorption in binary-ligand systems (or the kaolinite systems).


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Adsorption Edge Studies
The retention of kG by gibbsite and kaolinite is dependent on solution pH and independent of ionic strength (Fig. 2 ). Adsorption is at a maximum in acidic systems (pH <4), averaging 6.82 mmol kg–1 for gibbsite and 3.82 mmol kg–1 for kaolinite, and decreases with increasing pH. Assuming a site density of singly coordinated aluminal groups at the edge of the gibbsite crystal of 8 nm–2, and that the relative contribution of edge area to the total surface area is approximately 20% (Hiemstra et al., 1999), then the concentration of reactive aluminal groups on gibbsite is 9.3 mmol kg–1. Thus, kG is computed to occupy approximately 73% of the available binding sites on gibbsite at the pH of maximum adsorption (assuming monodentate surface complexation). For kaolinite, the concentration of singly coordinated aluminal groups is estimated at 4.305 mmol kg–1 (Sposito, 1984). Again, at the pH of maximum adsorption and assuming monodentate complexation, kG occupies approximately 89% of the available binding sites on kaolinite.


Figure 2
View larger version (24K):
[in this window]
[in a new window]

  		Fig. 2. The adsorption of 2-ketogluconate (kG) by gibbsite and kaolinite as a function of pH and ionic strength. In (a), the symbols represent experimentally determined kG adsorption and the solid lines represent the 1-pK basic Stern optimized fit to the experimental data using the chemical models described in Tables 2, 3, and 4. The dashed lines represent the model-predicted adsorption of kG by kaolinite using the surface complexation model developed for kG adsorption by gibbsite. The dotted lines represent the model-predicted adsorption of kG by gibbsite using the {equiv}AlOH–1kG–1.5(s) species rather than the {equiv}Al2O2H–1kG–(s) species. Graph (b) illustrates model-predicted adsorbed kG species composition as a fraction of the total predicted adsorbed kG for gibbsite (filled symbols) and kaolinite (open symbols).

 
The observation that kG adsorption is independent of ionic strength strongly suggests that the retention of this LMMOA occurs via an inner sphere surface complexation mechanism. In this regard, the adsorption envelopes of kG are similar to those of PO4 (Fig. 3 ), a ligand that also forms inner sphere surface complexes on gibbsite and kaolinite (Laiti et al., 1996; Van Emmerik et al., 2007). The maximum adsorption of PO4 is greater than that of kG, however, ranging from 9.14 mmol kg–1 for gibbsite to 5.45 mmol kg–1 for kaolinite (pH <4). Similarly, maximum AsO4 adsorption exceeds that of kG, ranging from 8.58 mmol kg–1 for gibbsite to 5.51 mmol kg–1 for kaolinite (pH <4). Like kG and PO4, AsO4 adsorption decreases with increasing pH (Fig. 3). Furthermore, both PO4 and AsO4 are computed to occupy essentially all singly coordinated aluminal groups on gibbsite and kaolinite at the pH of maximum adsorption. This implies that PO4 and AsO4 have larger intrinsic affinities than kG for the gibbsite and kaolinite surfaces. Unlike PO4, however, AsO4 adsorption by gibbsite is apparently a function of ionic strength when solution pH is greater than approximately 6. This observation would tend to suggest that outer sphere retention mechanisms dominate in neutral to alkaline systems for AsO4. This ionic strength dependency is not observed for AsO4 adsorption to kaolinite, however, and is not observed for AsO4 adsorption to gibbsite in the mixed ligand systems (discussed below). Furthermore, inner sphere complexation of AsO4 by aluminal surface functional groups across a wide range of pH values has been established in the scientific literature (Arai et al., 2001; Ladeira et al., 2001; Weerasooriya et al., 2004), although Catalano et al. (2006) reported that AsO4 adsorption by alumina may occur via both inner and outer sphere mechanisms. Of the ligands examined, SO4 retention is the lowest, with an adsorption maximum averaging 5.39 mmol kg–1 in 0.001 mol L–1 NaCl and 3.20 mmol kg–1 in 0.01 mol L–1 NaCl for gibbsite (pH <5). For kaolinite, the average SO4 adsorption maximum is 1.14 mmol kg–1 (pH <5), irrespective of ionic strength. The dependence of SO4 retention on ionic strength suggests that the dominant adsorption mechanism is outer sphere (He et al., 1997), although inner sphere surface complexation of SO4 has been proposed in acidic systems involving the aluminal group (He et al., 1996; Wijnja and Schulthess, 2000).


Figure 3
View larger version (25K):
[in this window]
[in a new window]

  		Fig. 3. The adsorption of (a) PO4, (b) AsO4, and (c) SO4 by gibbsite (G) and kaolinite (K) as a function of pH and ionic strength. The solid lines represent the 1-pK basic Stern optimized fit to the experimental data using the chemical models described in Tables 2, 3, and 4. The dashed lines represent the model-predicted ligand adsorption by kaolinite using the surface complexation model developed for gibbsite.

 
In multiple-ligand systems, ligand adsorption is influenced by direct competition for surface sites and by the change in the electrostatics of the solid–solution interface that occurs on ligand adsorption. The former effect predominates when adsorbates are introduced simultaneously and when the affinities of the competing ligands for the surface are comparable, whereas when a competing ligand is preadsorbed or when one ligand has a stronger affinity for the surface relative to a competing ligand, the latter effect predominates. The competitive adsorption studies indicate that kG retention is influenced by the presence of inorganic ligands (Fig. 4 and 5 ). Both PO4 and AsO4 had the greatest impact on kG adsorption, reducing kG retention by as much as 42% on gibbsite and 87% on kaolinite (at pH 5) relative to kG adsorption in the absence of these ligands. Furthermore, the presence of PO4 and AsO4 reduced kG retention in all but the highly alkaline systems (pH >9, where the retention of all ligands is at a minimum). By comparison, the influence of SO4 on kG adsorption is minimal and restricted to the pH <5 to 6 range. At pH 4, kG adsorption is reduced by approximately 14 and 30% by SO4 on gibbsite and kaolinite, relative to kG adsorption in the absence of SO4. These findings suggest that the adsorption of SO4 in acidic systems may have a significant inner sphere component, again an interpretation that is supported in the literature.


Figure 4
View larger version (15K):
[in this window]
[in a new window]

  		Fig. 4. The single-ligand adsorption of 2-ketogluconate (kG) and the competitive binary-ligand (L) adsorption of kG and (a) PO4, (b) AsO4, (c) and SO4 at equal total concentrations of 0.1 mmol L–1 and 10 g gibbsite L–1 as a function of pH and ionic strength. The solid lines represent the 1-pK basic Stern optimized fit to the experimental data. The dashed lines represent the model-predicted adsorption of kG in the binary systems using the surface complexation model developed for adsorption in the single-ligand gibbsite system.

 

Figure 5
View larger version (15K):
[in this window]
[in a new window]

  		Fig. 5. The single-ligand adsorption of 2-ketogluconate (kG) and the competitive binary-ligand (L) adsorption of kG and (a) PO4, (b) AsO4, (c) and SO4 at equal total concentrations of 0.1 mmol L–1 and 10 g kaolinite L–1 as a function of pH and ionic strength. The solid lines represent the 1-pK basic Stern optimized fit to the experimental data. The dashed lines represent the model-predicted adsorption of kG in the binary systems using the surface complexation model developed for adsorption in the single-ligand gibbsite system.

 
The simultaneous addition of kG and inorganic ligands also decreased the retention of the inorganic ligands (Fig. 6 and 7 ). This effect is relatively small for PO4 and limited to acidic systems. Adsorption of PO4 on kaolinite is unaffected by kG. On gibbsite, however, PO4 adsorption is reduced by approximately 19% (at pH 4) relative to PO4 adsorption in the absence of kG. These findings further illustrate the greater affinity of PO4 for gibbsite and kaolinite surfaces relative to kG. Arsenate retention is reduced throughout the entire pH range on both gibbsite and kaolinite. Relative to adsorption in single-adsorbate systems, AsO4 retention is reduced by 45% on gibbsite and 58% on kaolinite in the binary-adsorbate systems. Although AsO4 adsorption is greater than that of kG in the single-adsorbate systems, it would appear that these two ligands have similar affinities for the adsorbents. The reduction in SO4 retention was particularly noticeable in the binary-adsorbate gibbsite, where SO4 adsorption in the single-adsorbate systems is significant. Due to the low affinity of SO4 for kaolinite, however, the simultaneous addition of kG has no detectable impact on SO4 retention.


Figure 6
View larger version (16K):
[in this window]
[in a new window]

  		Fig. 6. The single-ligand adsorption of (a) PO4, (b) AsO4, (c) and SO4 and the competitive binary adsorption of ligands (L) and 2-ketogluconate (kG) at equal total concentrations of 0.1 mmol L–1 and 10 g gibbsite L–1 as a function of pH and ionic strength. The solid lines represent the 1-pK basic Stern optimized fit to the experimental data. The dashed lines represent the model-predicted adsorption of kG in the binary systems using the surface complexation model developed for adsorption in the single-ligand gibbsite system.

 

Figure 7
View larger version (15K):
[in this window]
[in a new window]

  		Fig. 7. The single-ligand adsorption of (a) PO4, (b) AsO4, (c) and SO4 and the competitive binary adsorption of ligands (L) and 2-ketogluconate (kG) at equal total concentrations of 0.1 mmol L–1 and 10 g kaolinite L–1 as a function of pH and ionic strength. The solid lines represent the 1-pK basic Stern optimized fit to the experimental data. The dashed lines represent the model-predicted adsorption of kG in the binary systems using the surface complexation model developed for adsorption in the single-ligand gibbsite system.

 
The adsorption envelopes of ligands in the single-adsorbate systems indicate that kG, PO4, and AsO4 have high affinities for the gibbsite and kaolinite surfaces. Furthermore, the binary-adsorbate studies indicated that kG competes with the inorganic ligands for gibbsite and kaolinite surfaces when the ligands are introduced simultaneously. The data also suggest that kG impacts SO4 retention by shifting the electrostatics of the solid–solution interface to more negative values rather than through direct competition for adsorption sites. The preadsorbed-ligand studies indicate that PO4 has a much stronger affinity for the aluminal group than does kG (Fig. 8 ). The adsorption edge of kG in the binary-adsorbate preadsorbed gibbsite systems with PO4 is similar to that of the simultaneous system; however, kG has little impact on preadsorbed PO4, even when aqueous kG concentrations are increased from 0.1 to 0.4 mmol L–1 (Fig. 9 ). Essentially, kG is not an effective ligand for displacing adsorbed PO4. Thus, in rhizosphere soils, the ability of kG to enhance PO4 availability may be derived through the enhanced dissolution of Al and Fe minerals (as illustrated by Essington et al., 2005) rather than through the direct competition and displacement of PO4 from adsorption sites.


Figure 8
View larger version (14K):
[in this window]
[in a new window]

  		Fig. 8. The influence of solution composition and preadsorbed PO4 or 2-ketogluconate (kG) on kG adsorption by gibbsite as a function of pH in 0.001 mol L–1 NaCl.

 

Figure 9
View larger version (18K):
[in this window]
[in a new window]

  		Fig. 9. The influence of solution composition and preadsorbed PO4 or 2-ketogluconate (kG) on PO4 adsorption by gibbsite as a function of pH in 0.01 mol L–1 NaCl.

 
Surface Complexation Modeling
The adsorption edge studies indicated that kG forms inner sphere surface complexes on gibbsite and kaolinite surfaces. Several kG retention mechanisms can be postulated, but only three inner sphere surface species can be imposed to adequately describe the adsorption edges: {equiv}AlOkG0.5–(s), {equiv}AlOH–1kG1.5–(s), and {equiv}Al2O2H–1kG–(s) (Fig. 1). The formation of each surface complex is described in the following reactions:

Formula 1[1]

Formula 2[2]

Formula 3[3]

Chemical models that use the monodentate {equiv}AlOkG0.5–(s) species and either the monodentate {equiv}AlOkGH–11.5– (s) or the bidentate {equiv}Al2O2H–1kG–(s) species are capable of describing the kG adsorption edge data (Fig. 2); however, the model including Eq. [1] and [3] routinely generated lower WSOS/DF values (particularly in the binary-adsorbate systems), although in the single-adsorbate gibbsite systems the differences between the two models were indistinguishable. For example, a WSOS/DF value of 9.909 was obtained when using Eq. [1] and [2], and 6.076 when using Eq. [1] and [3] (on gibbsite) (Fig. 2). Therefore, the bidentate {equiv}Al2O2H–1kG–(s) species (Eq. [3]), along with the monodentate {equiv}AlOkG0.5–(s) species (Eq. [1]), were used to account for kG surface complexation at the aluminal site. As illustrated in Fig. 2, the {equiv}AlOkG0.5–(s) species predominates in acidic systems while the {equiv}Al2O2H–1kG–(s) species predominates in neutral to alkaline systems. In order for the bidentate complex to form, an alcohol moiety must dissociate at pH values that are substantially lower than pKa2 = 11.97 (Nelson and Essington, 2005). Thus, the surface must induce the dissociation of the hydroxyl moiety, not unlike the surface-induced deprotonation of the citrate hydroxyl (postulated by Filius et al., 1997), or the metal (i.e., Al3+ and Fe3+) induced dissociation of organic ligands in aqueous complexation reactions (Essington, 2008).

Ketogluconate adsorption on kaolinite could be predicted using the surface complexation model developed for gibbsite (Fig. 2, Table 3). Similarly, the adsorption of PO4, AsO4, and SO4 on kaolinite was well predicted using the models developed for gibbsite (Fig. 3). These findings are similar to those of Sarkar et al. (2000). They used the chemical models and associated equilibrium constants used to describe Hg(II) adsorption by quartz (silanol) and gibbsite (aluminal) sites to successfully predict the Hg(II) adsorption edge on kaolinite. Although kG adsorption was successfully predicted (WSOS/DF = 28.951) (Table 5 ), albeit underpredicted in acidic media (pH <5), the FITEQL optimization of the surface complexation constants resulted in an improved fit (WSOS/DF = 8.112) (Table 4). The reoptimization of the gibbsite models for inorganic ligand adsorption to kaolinite resulted in modest reductions in WSOS/DF (Table 5); however, the new intrinsic constants did not alter the predicted ligand adsorption edges (Fig. 3).


View this table:
[in this window]
[in a new window]

  		Table 5. FITEQL-generated goodness-of-fit parameters (WSOS/DF) for the 1-pK basic Stern models used to describe ligand adsorption by gibbsite and kaolinite.

 
The surface complexation modeling of competitive adsorption, using intrinsic constants obtained from single-adsorbate systems, has been met with mixed results. The successful application of single-adsorbate models to binary-adsorbate systems has been demonstrated by Balistrieri and Murray (1982), Zachara et al. (1988), and Nowack and Stone (2006). Other studies have observed limited success with the procedure, however, requiring the reoptimization of the intrinsic constants (Goldberg and Traina, 1987; Geelhoed et al., 1998; Sarkar et al., 1999, 2000). For the gibbsite kG–PO4 system, the single-adsorbate models (Table 3, Fig. 2 and 3) underpredicted kG adsorption and overpredicted PO4 adsorption in acidic media (Fig. 4 and 6). For the kG–AsO4 and kG–SO4 gibbsite systems, however, competitive ligand adsorption was well predicted, although the reoptimized surface complexation constants resulted in improved fits (Table 5). In the competitive kG–PO4 and kG–AsO4 kaolinite systems, kG adsorption was greater than that predicted by the single-adsorbate models in acidic media (Fig. 5); however, kG retention in the competitive kG–SO4 kaolinite system, and inorganic ligand adsorption, was adequately predicted by the single-adsorbate models (Fig. 5 and 7). Attempts to optimize the intrinsic constants for competitive kG–PO4 and kG–AsO4 adsorption by kaolinite were not able to improve on the single-adsorbate models, as the resulting models could not account for increasing kG retention with decreasing pH.


   CONCLUSIONS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
Low-molecular-mass organic acids play a key role in the soil environment through their ability to chelate metal nutrients and contaminants and to compete for adsorption sites with strongly retained ligands. Previous evaluations indicate that kG is as effective as citrate in enhancing the solubility of gibbsite and goethite, presumably via the strong aqueous complexation of Al3+ and Fe3+ (Essington et al., 2005). Adsorption edge studies indicate that kG retention is dependent on pH and independent of ionic strength. Ketogluconate competes with PO4 and AsO4 for aluminal sites on gibbsite and kaolinite surfaces, but is only affected by SO4 in acidic systems. Ketogluconate is incapable of displacing adsorbed PO4, however, even when kG concentrations are increased. The experimental evidence strongly suggests that kG is retained at the gibbsite and kaolinite surfaces by inner sphere surface complexation mechanisms. The kG adsorption envelopes are predicted through the application of the 1-pK basic Stern surface complexation model by assuming the formation of the monodentate {equiv}AlOkG0.5–(s) and bidentate {equiv}Al2O2H–1kG–(s) surface species. Ketogluconate adsorption, as well as inorganic ligand adsorption, by kaolinite can be adequately described using the chemical models developed for adsorption by gibbsite. Thus, the reactivity of the aluminal group on gibbsite may potentially be used as an analog to model ligand adsorption by other minerals that bear this functional group. In general, ligand adsorption in binary-adsorbate systems can also be described using the chemical models developed for single-adsorbate systems, although reoptimization of the intrinsic surface complexation constants often improved the model descriptions of the adsorption edge data. The results of this evaluation, when coupled with the existing body of literature, indicate that kG is an exudate that affects the chemistry of the microenvironment that surrounds soil microbes, and potentially of the rhizosphere.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
This project was supported by the National Research Initiative of the USDA Cooperative State Research, Education and Extension Service, Grant no. 2001-35107-10165.

All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.

Received for publication May 24, 2007.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Essington, M. E.
Right arrow Articles by Anderson, R. M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Essington, M. E.
Right arrow Articles by Anderson, R. M.
Agricola
Right arrow Articles by Essington, M. E.
Right arrow Articles by Anderson, R. M.
Related Collections
Right arrow Soil Surface Chemistry
Right arrow Geochemical Processes
Right arrow Soil Chemistry

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome