Les J. Evans

Effects of organic acids on adsorption of lead onto montmorillonite, goethite and humic acid.

Adsorption isotherms for Pb onto six soil components (quartz, feldspar, kaolinite, montmorillonite, goethite and humic acid) were studied. The influence of pH, EDTA and citric acid on the adsorption of Pb onto montmorillonite, goethite and humic acid were considered. Results indicate that the experimental data fit the Langmuir Adsorption Isotherm. The adsorption capacity for Pb at pH 6 was found to be in the order: humic acid (22.7 mg g(-1)) > goethite (11.04 mg g(-1)) > montmorillonite (10.4 mg g(-1)) > kaolinite (0.91 mg g(-1)) > feldspar (0.503 mg g(-1)) > quartz (0.148 mg g(-1)). Generally, the amount of Pb adsorbed onto montmorillonite, goethite and humic acid decreased with increasing concentrations of EDTA and citric acid and with increases in alkality. However, there were two exceptions: (1) addition of citric acid increased the amount of Pb adsorbed onto humic acid; and (2) the amount of Pb adsorbed onto goethite decreased with increasing pH in the presence of EDTA. Some mechanisms involved in the adsorption reactions are discussed.

Effects of cement or lime on Cd, Co, Cu, Ni, Pb, Sb and Zn mobility in field-contaminated and aged soils.

Cement or lime can be used to treat trace element contaminated soils, reducing their mobility due to increased soil pH which enhances precipitation and adsorption, and also due to pozzolanic reactions and cementation. In the present work, an alkaline and an acidic soil both containing Cd, Co, Cu, Ni, Pb, Sb and Zn from either geogenic or geogenic and anthropogenic origin were treated with cement or calcium hydroxide. Soils were then extracted with dilute HNO(3) or NaOH solution of different concentrations to obtain extracts of different pH (pH 4-12). In untreated soils, Co, Cu, Ni and Pb in solutions were detected at alkaline pH. The addition of cement or Ca(OH)(2) reduced the mobility of every trace element at high pH, but enhanced the mobility of Cd, Co, Cu, Ni, Pb and Zn at low pH. Metal mobilisation at high pH was observed for Cu in the acidic soil due to the liberation of dissolved organic matter. Below pH 6, Sb mobility was lower in the cement-treated soil compared to the untreated soil, but the same in the Ca(OH)(2) treated soil as in the control soil. Comparison with theoretical trace element precipitates suggested that the mobility of trace elements is likely reduced at high pH by encapsulation and immobilisation within the cement matrix rather then precipitation.

Development of a coupled metal speciation-fate model for surface aquatic systems.

A coupled metal transport and speciation model (TRANSPEC) has been developed for surface aquatic systems that explicitly considers the influence of metal speciation on fate. The TRANSPEC, which is general to most metal and surface aquatic systems, is constructed by sequentially coupling the speciation/complexation module (in this application MINEQL+) with the fugacity/aquivalence approach for the fate calculations. This model formulation increases the mechanistic detail, predictive power, and fidelity to reality of current fugacity-aquivalence fate models for metals by estimating aqueous speciation and complexation, rather than relying on empirically derived partition coefficients. A pseudo-steady state version of TRANSPEC was used to simulate Zn dynamics in Ross Lake (Flin Flon, MB, Canada) that received elevated metal and organic matter inputs for over 50 years. Field studies revealed that ZnS forms soluble ZnL, Zn2+, and ZnSO4(0) increasing pore water concentrations when surficial sediments turn oxic during fall. The model results for three seasonal scenarios suggest that Zn remobilization is driven by resuspension of insoluble ZnS and the contribution of diffusion is negligible, even during fall when ZnS dissolves to increase the concentration of soluble species under oxic conditions in the sediments. The low diffusive flux is due to the binding of Zn to colloidal dissolved organic matter (DOM) for which sediment-water diffusion is relatively slow, a result that was obtained as a result of considering metal speciation in the fate calculations.

Cadmium adsorption by an organic soil: a comparison of some humic – metal complexation models

Abstract The retention of Cd by an organic soil was investigated as a function of pH and ionic strength. The adsorption of Cd at pH values from 2 to 11 at two ionic strengths (0.053 M and 0.017 M LiNO3) were found to be a function of both pH and ionic strength. Four Cd-humic complexation models were evaluated in order to test the applicability of these models to fit data from batch adsorption experiments. The models varied greatly in their complexity and implicit assumptions. Three were discrete functional group models – a simple diprotic acid model, a two diprotic acid model and the Windermere Humic Aqueous Acid (WHAM) model, and a continuous functional group model - the non-ideal competitive adsorption (NICA) model. The concentration of proton binding sites in the soil was found to be 4.51 mol kg-1. The NICA and WHAM models were more successful than either a simple diprotic acid model or a two diprotic acid model at modeling Cd complexation by the organic soil, although both underestimated adsorption at very high pH values.

Preliminary validation of a sequential fractionation method to study phosphorus chemistry in a calcareous soil

A sequential fractionation method proposed by Jiang and Gu (1989) distinguished three types of calcium phosphates (Ca-P) according to their different plant availabilities. Three extractants, NaHCO3, NH4Ac, and H2SO4 were used to extract Ca2-P, Ca8-P, and Ca10-P types, respectively, from soil. This sequential fractionation method was tested and modified for analyzing the P chemistry of a calcareous soil. The solubility test and the model diagrams of the stability of the major Ca-P minerals showed that NaHCO3 was able to extract brushite (Ca2-P type), and NH4Ac extracted brushite and β-tricalcium P (Ca8-P type) as well as hydroxyapatite (Ca10-P type). Therefore the P forms targeted by extraction with NH4Ac should include both Ca8-and Ca10-P types. The sum of the P extracted by all extractants in the sequential fractionation method in the calcareous soil was in agreement with the total P measured by the perchloric acid digestion method. A proportion of organic P measured by the sequential fractionation method was in agreement with the result from solution (31)P NMR spectroscopy. This study showed that the modified sequential fractionation method and its target P forms would be useful for quantifying and characterizing inorganic and organic P in a calcareous soil, even though it should be used in combination with other techniques, such as solution (31)P NMR spectroscopy.

A simple model to predict chromate partitioning in selected soils from China

Due to its mobility and toxicity, chromate [Cr(VI)] partitioning in soils, especially in the vadose zone, is an important environmental concern. The aim of this study was to develop a mechanism-based multi-surface complexation model using published parameters to predict the soil/water partitioning of Cr(VI) in 12 soils (previously depleted of organic matter) from China. The retention of Cr(VI) in soils was attributed to two reactive oxide surfaces: goethite and hydrous ferric oxide (HFO); however, modeling results showed that the best prediction was obtained with goethite alone, whereas the addition of HFO resulted in an overestimation of adsorption in some soils. Cr(VI) adsorption onto goethite could be described using our previously proposed CD-MUSIC model. In the absence of a specific value for the soil-reactive surface area of goethite, a general value of 45.8m2/g was used. The available phosphate in soils was identified as a strong competitor for Cr(VI) adsorption; thus, for soils with a low Fe/P ratio (<1) the effect of phosphate on Cr(VI) retention should not be neglected. The simple method presented herein can be applied to soils with a wide range of properties, pH values, and Cr(VI) loading concentrations.

Multisurface modeling of Ni bioavailability to wheat (Triticum aestivum L.) in various soils

Continual efforts have been made to determine a simple and universal method of estimating heavy metal phytoavailability in terrestrial systems. In the present study, a mechanism-based multi-surface model (MSM) was developed to predict the partition of Ni(II) in soil-solution phases and its bioaccumulation in wheat (Triticum aestivum L.) in 19 Chinese soils with a wide range of soil properties. MSM successfully predicted the Ni(II) dissolution in 0.01 M CaCl2 extracting solution (R2 = 0.875). The two-site model for clay fraction improved the prediction, particularly for alkaline soils, because of the additional consideration of edge sites. More crucially, the calculated dissolved Ni(II) was highly correlated with the metal accumulation in wheat (R2 = 0.820 for roots and 0.817 for shoots). The correlation coefficients for the MSM and various chemical extraction methods have the following order: soil pore water > MSM ≈ diffuse gradient technique (DGT) > soil total Ni > 0.43 M HNO3 > 0.01 M CaCl2. The results suggested that the dissolved Ni(II) calculated using MSM can serve as an effective indicator of the bioavailability of Ni(II) in various soils; hence, MSM can be used as an supplement for metal risk prediction and assessment besides chemical extraction techniques.

Multi-surface modeling of Ni(II) and Cd(II) partitioning in soils: Effects of salts and solid/liquid ratios

Metal partitioning in soils is a key process controlling metal bioavailability and mobility and is greatly influenced by the solid/liquid ratio. However, metal partitioning is difficult to describe either by a simple partition coefficient or by isotherm adsorption equations. This study investigated the solubility of Ni(II) and Cd(II) in 19 soils as a function of three extraction reagents (water, 0.01 M NaNO3, and 0.01 M CaCl2), five solid/liquid ratios (5-400 g/L) and field condition extracted by Rhizon samplers. Thermodynamically based multi-surface models (MSMs) that included generic parameters were used to describe metal partitioning under the studied conditions. The results showed that Ni/Cd solubility depended on the soil type, extraction reagent, and solid/liquid ratio. Soil major background cations (especially Ca2+, Mg2+, Fe3+ and Al3+) had a significant effect on the models prediction ability. The MSM was able to predict the extractable metal in 0.01 M CaCl2 in various soils at different solid/liquid ratios when soil background cations were included in the calculation; without the background cations, the model was able to predict metal partitioning only at solid/liquid ratios of <100 g/L. In addition, the model failed to predict water-extracted and 0.01 M-NaNO3-extracted Ni/Cd if background cations were not included, but could reasonably do so if they were included. More importantly, after the background cations were included, MSMs relatively well predicted the Ni/Cd content in soil pore water under 80% field capacity conditions with water as the solution matrix.