Development and exploration of an organic contaminant fate model using poly-parameter linear free energy relationships.

Octanol-based partitioning relationships, referred to as single-parameter linear free energy relationships (SP-LFERS), are often criticized for their limited applicability to polar organic substances. Therefore, SP-LFERS describing environmental phase partitioning in CoZMo-POP2, a dynamic multimedia chemical fate model, are replaced with poly-parameter linear free energy relationships (PP-LFERS) which describe temperature-dependent partitioning as the linear sum of various specific and nonspecific molecular interactions. A data set of chemicals with available solute descriptors, which quantify these molecular interactions, is compiled from the literature and, together with a data set of hypothetical chemicals, used to investigate the differences in the predictions of SP-LFER- and PP-LFER-based model in relative and absolute terms for three different emission scenarios. Model outputs are manipulated to allow the results to be displayed as a function of log K(AW) and log K(OA). Whereas the primary environmental fate is similar in both models, differences arise mostly in the environmental phases which contain only a small fraction of chemical. Larger differences in model results occur either because a difference in the predicted partitioning between water and organic matter affects the extent of soil-water runoff, or because differences in gas-particle partitioning affectthe relative deposition to aqueous and forested surfaces. The two models showed smaller differences for degradable chemicals than for chemicals assumed to be perfectly persistent Overall, however, the absolute differences between the model results are relatively small in comparison to the precision generally associated with model parametrization. Accordingly, we suggest that the quality of the available chemical input parameters should decide whether a PP-LFER model is preferable over a SP-LFER model. The PP-LFER model is further used to evaluate the effects of various molecular interactions on chemical fate, and the solute descriptor associated with van der Waals dispersive interactions is found to have the most pronounced effect on the environmental distribution of chemicals.