Predicting octanol/water partition coefficient using solvation free energy and solvent-accessible surface area

摘要

The regression model for octanol/water partition coefficients ( K ow ), is founded with only two molecular descriptors available through quantum chemical calculations: solvation free energy (Δ G S ), and solvent accessible surface area (SASA). For the properties of 47 organic compounds from 17 types, the model gives a correction coefficient (adjusted for degrees of freedom) of 0 959 and a standard error of 0 277 log unit. It is a suitable way to predict the partition properties that are related to solute solvent interactions in the water phase.