Using quantitative structural property relationships, chemical fate models, and the chemical partitioning space to investigate the potential for long range transport and bioaccumulation of complex halogenated chemical mixtures.

Anya Gawor, Frank Wania

Index: Environ. Sci. Process. Impacts 15(9) , 1671-84, (2013)

Full Text: HTML

Abstract

Some substances are mixtures of very large number of constituents which vary widely in their properties, and thus also in terms of their environmental fate and the hazard that they may pose to humans and the environment. Examples of such substances include industrial chemicals such as the chlorinated paraffins, technical pesticides such as toxaphene, and unintended combustion side products, such as mixed halogenated dibenzo-p-dioxins and dibenzofurans. Here we describe a simple graphical superposition method that could precede a more detailed hazard assessment for such substances. First, partitioning and degradation properties for each individual constituent of a mixture are estimated with high-throughput quantitative structure-property relationships. Placed in a chemical partitioning space, i.e. a coordinate system defined by two partitioning coefficients, the mixtures appear as 'clouds'. When model-derived hazard assessment metrics, such as the potential for bioaccumulation and long range transport, are superimposed on these clouds, the resulting maps identify the constituents with the highest value for a particular parameter and thus potentially the greatest hazard. The maps also indicate transparently how the potential for long range transport and bioaccumulation is dependent on structural attributes, such as chain length, and the degree and type of halogenation. In contrast to previous approaches, in which the mixture is represented by a single set of properties or those of a few selected constituents, the whole range of environmental fate behaviors displayed by the constituents of a mixture are being considered. The approach is illustrated with three sets of chemical substances.