Ion exchange to fixed charge sites and surface complexation to hydroxylated sites are recognized to affect Cd aqueous concentrations in soil; however, the individual contribution of these reactions to Cd sorption in mineralogically complex material has not been documented. The objective of this study was to isolate the contributions of fixed charge and hydroxylated sites to Cd sorption in two Ultisol subsurface horizons that contained kaolinite, 2:1 layer silicates, and crystalline Fe and Al oxides. Cadmium sorption was measured on clay‐sized isolates (<2.0 µm) from the soil materials in NaClO4 electrolyte at ionic strength (I) = 0.1, 0.01, and 0.001 mol L−1 with pH ranging from 4 to 9 and Cd concentrations ranging from 2 × 10−8 to 3 × 10−5 mol L−1. Comparative experiments were also performed with smectite (SWy‐1), kaolinite (KGa‐1), and crystalline and amorphous Fe oxides to evaluate the potential contribution of ion exchange and surface complexation reactions on these phases to Cd sorption in the isolates. Ion exchange of Cd2+ on layer silicates dominated Cd sorption within the isolates when I was ≤0.01 mol L−1 and the pH was less than ≈6.5. Sorption of Cd on hydroxylated sites on the edges of the layer silicates or on Fe or Al oxides, and/or exchange of CdOH+ on the fixed‐charge sites of the layer silicates, dominated Cd sorption at I = 0.1 mol L−1, and when the pH was greater than ≈6.5 at any I. Below pH 6.5, the magnitude of the sorption was controlled by the cation‐exchange capacity of the isolates, which, in turn, was related to their relative proportions of kaolinite and 2:1 layer silicates. By comparing the Cd sorption on the isolates to that on (i) dithionite‐citrate‐bicarbonate (DCB) treated islates and (ii) DCB‐treated isolates augmented with Fe oxides, it was concluded that crystalline Fe oxides were not significant sorbents of Cd. Instead, the Fe oxides appeared to be associated with the layer silicates by a particle interaction that blocked access of Cd to fixed‐charge sites on the layer silicates.

The contribution of edge‐complexation reactions to metal ion sorption by smectites has not been well documented through experimentation or modeling. A model containing fixed‐charge (X−) and hydroxylated edge sites (AlOH and SiOH) was applied to sorption data for Cd on SWy‐1 smectite and a smectitic soil isolate in Na+, Ca2+, and mixed electrolytes to evaluate the potential contributions of ion exchange and surface coordination to metal cation binding. Sorption modeling was performed with FITEQL. Edge reactions were described using the triple‐layer model (TLM) and an inner sphere Cd complex on AlOH sites and outer sphere complexes on SiOH sites. The complexation reactions were parameterized using (i) Cd sorption data from silica and alumina and (ii) site concentration estimates based on cation‐exchange capacity (X−) and particle size measurements (AlOH and SiOH). Ion‐exchange reactions were described in half‐reaction format; exchange constants were fit using Cd sorption data in either Na+ or Ca2+ electrolyte. The sorption data was qualitatively described across ranges in pH and electrolyte concentration using a combination of ion exchange (CdX2) and edge (AlOCd+) complexes. Complexation of Cd to edge SiOH was calculated to be unimportant. The calculated contribution of AlOCd+ complexes increased with increases in pH and ionic strength. Good predictions of Cd sorption in mixed electrolytes were obtained using constants derived from single‐electrolyte systems. Edge sorption reactions were computed to be more important for the soil smectite than for SWy‐1 because of its smaller particle size. However, the slope and magnitude of sorption on the soil isolate was inadequately described by the model. Competitive sorption reactions on the edge sites and binding to organic material are additional factors that must be considered to adequately describe Cd sorption to smectitic soil clays.

The adsorption of CrO2‐4 to goethites varying in specific surface area and Al substitution was measured over a range in pH, sorbate and sorbent concentrations and ionic strength. The α‐(Fe, Al)OOH exhibited a lower pH50 (the pH at which 50% adsorption occurs) for the CrO2‐4 at all adsorbate concentrations (6 × 10−6 to 5 × 10−4 M CrO2‐4) than did the two α‐FeOOH samples; this difference could not be attributed to either variations in CrO2‐4 adsorption density or sorbent PZC. The overall proton coefficient (n), calculated from adsorption data for 5 × 10−7 M CrO2‐4, was identical for the two α‐FeOOH samples despite differences in the physical nature and sorbate‐to‐surface site ratio of the two samples. The observed n for the α‐ (Fe, Al)OOH system was larger than the n for the pure goethite systems. These data suggest that differences in the acid‐base properties and interaction with the background electrolyte for the Fe and Al sites affect the overall proton coefficient and CrO2‐4 adsorption on the substituted surface when compared to the pure surface. The surface coordination constants for the Triple Layer Model (TLM) were derived for CrO2‐4 adsorption on the α‐FeOOH solids using the FITEQL computer code. Using these and other TLM constants from the literature, adsorption on all three goethite samples was quantitatively simulated over a range in solute and sorbent concentrations, pH and ionic strength. The successful simulation of CrO2‐4 sorption onto α‐(Fe, Al)OOH required the use of a two‐site model which included the ionization and electrolyte complexation reactions of both Fe and Al sites. Site densities of Fe and Al were taken to be proportional to the mole fraction of each in the α‐(Fe, Al)OOH solid. However, the interaction of the CrO2‐4 species with the Al sites was not required to quantitatively describe the adsorption edge since the log K for the dominant FeOH+2‐HCrO‐4 species is three orders of magnitude greater than the ‐CrO2‐4 or ‐HCrO‐4 on Al sites.

Although specimen smectities (e.g., SWy‐1) are often used as analogues of the exchanger phase in smectitic soils, few comparisons of metal ion sorption on specimen and soil smectites have been made. In this study, the sorption of Cd was measured on SWy‐1 and on clay‐sized separates from two smectitic subsoils to evaluate the selectivity of specimen and soil‐derived smectites for Cd. Sorption was measured in clay suspensions (≈1 mmolc L−1 equivalent charge concentration at pH 6.0) in Na+, Ca2+, and Na+‐Ca2+ perchlorate solutions across pH 4.5 to 8.5 and at ionic strengths (I) ranging from 0.005 to 0.1. Ionic strength and electrolyte cation valence strongly influenced Cd sorption by SWy‐1 and the soil smectites. Ion exchange dominated Cd sorption at low ionic strength in Na+ electrolyte (I = 0.005–0.014). Increasing Na+ concentrations to I = 0.1 or changing the electrolyte cation to Ca2+ at I = 0.003 to 0.006 suppressed ion exchange. When ion exchange was suppressed, Cd sorption to both specimen and soil smectites showed little dependence on ionic strength and increased with pH. Except at the lowest Na+ concentration (I = 0.005), conditional equilibrium constants (Kv) for Cd2+ exchange increased with increases in both ionic strength and pH. These increases were ascribed to Cd complexation reactions to edge sites on the layer silicates whose effects became evident only under conditions that suppressed ion exchange. At pH 6 and I = 0.05–0.01, SWy‐1 did not exhibit any preference for Na+, Ca2+, or Cd2+. The smectitic soil separates, in contrast, showed (i) sorption behavior that increased sharply with pH, (ii) preference for Cd in Na+ and Ca2+ electrolytes, and (iii) variation in Kv with ionic strength, pH, and surface coverage. The contrasting sorption behavior of the soil smectites was hypothesized to result from (i) a greater edge surface area, which increased the contribution of oxidelike complexation reactions to Cd sorption, and (ii) the presence of minor associated organic material and Fe oxides that functioned as co‐complexants for Cd.

A multisite model was applied to Cd‐sorption data across a range in pH and at ionic strength (I) = 0.1 and 0.01 on specimen layer silicates and on untreated and dithionite‐citrate‐bicarbonate (DCB) treated isolates from two Utisols. The two isolates contained different mass concentrations of kaolinite, 2:1 layer silicates, Al‐substituted crystalline Fe oxides, and gibbsite. The multisite model included the exchange of Cd2+ on fixed‐charge sites on the layer silicates, complexation to hydroxyl sites on the layer silicates and Fe oxides, and electrostatic interaction between positively charged sites on Fe oxides and fixed‐charge sites on the layer silicates. Site concentrations for the specimen layer silicates and DCB‐treated and untreated isolates were estimated from cation‐exchange capacity (CEC) data and N2‐specific surface area. Equilibrium constants for specific surface reactions were estimated from Cd‐sorption data on the specimen layer silicates, KGa‐1 and SWy‐1, and the DCB‐treated soil isolates. The combined modeling results indicated that (i) Cd2+ ion exchange to fixed‐charge sites was the dominant sorption reaction for pH < 6.5 and I = 0.01, (ii) mass action with Na suppressed ion exchange at I = 0.01, (iii) Cd complexation to hydroxyl sites produced strong pH dependency in Cd sorption and was responsible for all the Cd sorption at I = 0.01 and above pH 6.5 at I = 0.01, and (iv) Fe oxides decreased Cd sorption by blocking access of Cd to exchange sites. The modeling results were unable to determine whether Fe oxides were important sorbents for Cd. Either with or without Cd sorption on Fe oxides, this model was able to describe the Cd sorption on the soil isolates reasonably well.