Chromatographic estimation of the soil-sorption coefficients of organic compounds

Soil-sorption coefficient (KOC) assessment is essential before placing a new chemical substance on the market.Since experimental KOC measurements are tedious, time-consuming and costly, alternative approaches to estimating KOC from other descriptors have been proposed. We review the use of chromatographic retention (log k) values, obtained in different types of chromatographic systems, as descriptors to establish predictive log KOC-log k models.We highlighted critical aspects of such models and performed a comparative study using data in the literature. We compare two strategies to deal with the large variability of experimental log KOC data. We contrast the results of this study with the main conclusions reported in the literature.