Geochemical modelling of silver(I), copper(II) and antimony(V) sorption to soil as a function of soil properties: Development of soil assemblage models

The development of speciation and surface complexation models to calculate partitioning of metals between the soil solid phase and soil solution has made large progress. In this research, the evaluation and practical applicability of soil assemblage models to predict the sorption of Ag(I), Cu(II) and Sb(V) onto soils from Elora, Sudbury and Cobalt, Ontario, respectively.;Overall, it can be concluded that the assemblage models investigated in this thesis were able to produce a reasonable fit to the data for the adsorption of Ag(I), Cu(II) and Sb(V). Observed differences between the investigated elements with respect to their species distribution were reflected in their relative binding affinities for each surface. The Elora assemblage model revealed that Ag(I) is strongly adsorbed by the clay fraction of the soil and to a lesser extent to the organic matter fraction. The Sudbury assemblage model revealed that for Cu(II) the most important adsorbent was soil organic matter. Finally, the Cobalt assemblage model revealed that the Fe-oxide fraction was the most important sorbing surface for Sb(V).;These models are not without uncertainty and show a particular sensitivity for the number of binding sites and competitive adsorption with major cations and anions present in solution. Model predictions were also shown to rely strongly on model parameters derived from laboratory experiments for well-characterized materials, while conditions met in natural, heterogenous soils may be different from these natural systems. Nonetheless, the results of this modelling approach are encouraging, and provide a first step toward application of the assemblage model approach in site-specific risk assessments and provides an advancement over empirical approaches currently used.;The adsorption of Ag(I), Cu(II) and Sb(V) adsorption onto goethite, soil clay minerals and humic acids was investigated as a function of pH and ionic strength. Proton and elemental binding constants were developed for each surface with potentiometric titrations and modelled with FITEQL using the Constant Capacitance Model, the Diffuse Layer Model and the Triple Layer Model, and for the humic acids with a discrete ligand model. These binding constants were incorporated into soil assemblage models for each site assuming component additivity.