Predicting Octanol/Water Partition Coefficients of Alcohol Ethoxylate Surfactants Using a Combination of Molecular Dynamics and the Conductor-like Screening Model for Realistic Solvents

We present a hybrid modeling strategy for the prediction of octanol/water partition coefficients for alcohol ethoxylate surfactants of varying chain lengths. The strategy makes use of molecular dynamics (MD) simulations for the generation of molecular conformations in the presence of a solvent. A clustering of the conformations from the MD trajectories was carried out based on principal component analysis of their dihedral angles. Representative conformations thus selected were then used in the conductor-like screening model for realistic solvents (COSMO-RS). Each conformation has been assigned a weight using an equation derived on the basis of its probability of occurrence in the MD trajectory. Experimental partition coefficients were reproduced within conformation-independent accuracy of COSMO-RS despite the size and flexibility of the ethoxy chain otherwise posing a challenge on their solvation modeling.