A Quantitative Property-Property Relationship for Estimating Packaging-Food Partition Coefficients of Organic Compounds

Organic chemicals encapsulated in beverage and food packaging can migrate to the food and lead to human exposures via ingestion. The packaging-food (Kpf) partition coefficient is a key parameter to estimate the chemical migration from packaging materials. Previous studies have simply set Kpf to 1 or 1000, or provided separate linear correlations for several discrete values of ethanol equivalencies of food simulants (EtOH-eq). The aim of the present study is to develop a single quantitative property-property relationship (QPPR) valid for different chemical-packaging combinations and for water or different EtOHeq values. We compiled datasets of measured Kpf from 3 studies, which contained 302 data points of 152 chemicals in LDPE and HDPE (low and high density polyethylene) at 25 °C for EtOH-eq values ranging from 0% (water) to 95%. A multiple linear regression (MLR) model was developed to predict Kpf as a function of the chemical’s Kow, the EtOH-eq, the packaging type and an interaction term between Kow and EtOH-eq. The model shows good fitting performance of the experimental datasets with adjusted R-square of 0.92. All predictors are highly significant except the packaging type, probably because only two packaging types are included. This preliminary QPPR demonstrates that the Kpf for various chemicalpackaging-food combinations can be estimated by a single linear correlation. Based on more than 1000 collected Kpf in 15 materials, we will present extensive results for other packaging types and different temperatures. This QPPR provides a comprehensive correlation method to estimate the Kpf for a wide range of chemical-packaging-food combinations, and thus facilitate high-throughput estimates of human exposures to chemicals encapsulated in food contact materials.