Metal-humic interactions: A random structural modelling approach

Abstract

The binding of the metal cations copper and calcium to peat humics and ground water (aquatic) humic and fulvic acids has been studied using ion selective electrochemistry under a variety of experimental conditions (pH, ionic strength, etc.). Competition between the metals for sites on humic acid molecules was also studied. The Random molecular model developed by Murray and Linder (1983, 1984) has been adapted in an attempt to simulate humate-metal binding behaviour. The Murray and Linder model uses a random structural approach to identify the likely metal binding sites present in each humic sample. The model contains no adjustable parameters, but instead relies on experimentally derived variables, namely, percentage carbon and hydrogen plus carboxylate and phenol concentrations. The model constructs molecules of humic acids which conform to these measurements and then searches the structures for a series of predefined metal binding sites. After generating a large number of similar structures, all of which conform to the experimental data, average concentrations of the binding sites are calculated. In the present work, the model has been extended to include nitrogen donor atoms and to search for nitrogen containing metal binding sites. In order to predict metal speciation, a version of the geochemical speciation code PHREEQE was devised in which the electrostatic effects produced by the large humic charge were accounted for. Tests on the model showed that the binding strengths predicted by the computer were consistently less than those observed by experiment, probably due to underestimation of the binding site complexity.