PubMed2002
The fate and risk assessment of hydrophobic substances in the terrestrial environment can be associated with large errors. These can be attributed to the partitioning and process coefficients derived in experimental studies and to the model set-up that is designed to calculate the exposure concentrations. In many cases, the concentration of xenobiotics are low in the environment, which gives the aqueous phase the characteristics of a true solution, which are in accordance with the thermodynamic description of dilute solutions. Under these circumstances, the conventional equilibrium coefficients, such as Kd, Henry's Law constant H and the bioconcentration factor, BCF, are independent of the activity coefficient of the partitioning compound in the respective phases. However, for hydrophobic substances, these coefficients are often measured in laboratory experiments, where the nominal concentration levels are above the substance saturation point within the bulk water phase. In the case of the phthalates, the hydrophobic effect induces the formation of microdroplets (third phase) in the bulk water phase, by which the system is characterised as a heterogeneous mixture. Consequently, the linearity between dissolved and sorbed concentration is no longer true. Furthermore, in the terrestrial and aquatic environment, the presence of natural Dissolved Organic Matter (DOM) will have an influence on the fate and effects of hydrophobic substances. Hydrophobic compounds show large affinity for sorption to DOM, and contrary to Fixed Organic Matter (FOM), DOM is mobile and can be transported through the soil pores with the advective flow. It is therefore crucial that dispersed or emulsified phases within the continuous aqueous phase, e.g. DOM and microemulsions of phthalates, are distinguished from true solutions in the experimental measurements of partitioning coefficients, e.g. in order not to underestimate the mobility of sorbed substance. These aspects are treated in this study, where the exposure concentration, vertical transport and microbial degradation of Di-(2-ethylhexyl)-phthalate (DEHP) is modelled in an organic rich topsoil compartment, using experimental partitioning coefficients and degradation rates from the literature. Two model set-ups are derived for the topsoil compartment, i.e. (1) a system with dilute solution of substance and (2) a system with the presence of a third phase of microdroplets. In both models, the presence of DOM is incorporated. The first model shows that the error in the calculated exposure concentration, by using partitioning coefficients derived under unfavourable experimental conditions, compared to realistic conditions, amounts to 1400%. A comparison between the two models shows that when emulsion formation is not incorporated in the model, the calculated flux will be overestimated by a factor of 60.