QSPR analysis of infinite dilution activity coefficients of chlorinated organic compounds in water

A quantitative structure–property relationship (QSPR) study was performed for the prediction of the natural logarithm of infinite dilution activity coefficients, ln(γ∞), of 45 chlorinated organic compounds in water. A five-descriptor correlation equation, with a squared correlation coefficient (R2) of 0.9655 and a standard error of estimation (s) of 0.613, was produced by using the stepwise multilinear regression analysis on a training set of 35 compounds. The reliability of the proposed model was further illustrated using various evaluation techniques: leave-one-out cross-validation procedure, randomization tests, and validation through an external test set of 10 compounds. All descriptors involved in the model were derived solely from the chemical structure of the compounds, which makes the model very useful in predicting ln(γ∞) of other chlorinated organic compounds not present in the dataset used for the development of the present model. In addition, the descriptors were discussed in detail to understand the interactions that affect infinite dilution activity coefficients.