Perspectives on modeling the release of hydrophobic organic contaminants drawn from model polymer release systems

Abstract

This work critically evaluates current practices in modeling the release of hydrophobic organic contaminants (HOCs) from complex matrices. Using well-constrained model polymer release systems (MPRS), we evaluate several empirical kinetic release models and one class of semi-empirical diffusion model; generally these models either do not describe release well, and are thus not useful, or they describe release well but are overparameterized, thus decreasing confidence in the fitting parameters. Results indicate that the n-phasic release often attributed to soil and sediment matrices is a non-mechanistic operational definition, arising from bias in the choice of empirical fitting expression. Further analysis illustrates pitfalls in common approaches to characterizing desorption from complex environmental matrices with respect to parameter interpretation, particularly effective diffusion coefficients. We also consider MPRS release in the context of other models and find that, though release occurs from these materials via a mechanism of anomalous (non-Fickian) diffusion, Fickian diffusion models nonetheless describe release well. This finding reconciles previous reports about anomalous release from environmental matrices with the hitherto success of Fickian diffusion models of desorption.