Exploring Organic Carbon–Water Partition Ratio (KOC) Data for Mobility Hazard and Exposure Assessments Using Big Data Approaches

Chemical hazard and risk assessments often use physical-chemical properties to categorize and identify chemicals of concern. Most recently, legislation regarding the classification, labelling and packaging (CLP) has introduced new categories for chemical hazard assessment including persistent, mobile, and toxic (PMT) and very persistent and very mobile (vPvM). Mobility is determined based on the logarithmic organic carbon–water partition ratio (log₁₀ K_OC) or logarithmic octanol–water partition ratio (log₁₀ K_OW) for neutral compounds and the logarithmic octanol–water distribution ratio (log₁₀ D_OW) for ionizable compounds, depending on the regulatory agency and the degree of mobility (Arp and Hale 2022).

In this work we use all available KOC data in a consensus approach that takes into account the variability, uncertainty, different sources and types of data. Data sources include the OECD's eChemPortal, experimental datasets, and estimated values using poly-parameter linear free energy relationships (ppLFERs), OPERA, and EPI Suite. With Bayesian statistical inference approaches we aggregate values and their errors from these different sources to derive a probability distribution for the "true" log₁₀ K_OC value.

For risk assessments, the classification of mobility can then be done using the mode of the distribution with uncertainties using 95 % credibility interval. For a more risk adverse approach, the 5 % quantile of each log₁₀ K_OC distribution can be compared to the mobility thresholds.