Prediction of surface tension, viscosity, and thermal conductivity for common organic solvents using quantitative structure-property relationships

Predictive models for the surface tension, viscosity, and thermal conductiv ity of 213 common organic solvents are reported. The models are derived fro m numerical descriptors which encode information about the topology, geomet ry, and electronics of each compound in the data set. Multiple linear regre ssion and computational neural networks are used to train and evaluate mode ls based on statistical indices and overall root-mean-square error. Eight-d escriptor models were developed for both surface tension and viscosity, whi le a nine-descriptor model was developed for thermal conductivity. In addit ion, a single nine-descriptor model was developed for prediction of all thr ee properties. The results of this study compare favorably to previously re ported prediction methods for these three properties.