Prediction of surface tension of ionic liquids by molecular approach

Originally, Quantitative Structure Property Relationship (QSPR) models for the surface tension of ionic liquids are developed based on molecular descriptors. A large data set of 930 experimental surface tension data points for 48 ionic liquids is applied to derive the model. Seven descriptors are selected by genetic function approximation to relate the surface tension of ionic liquids to their corresponding anions and cation structures. To capture the nonlinear nature of surface tension, a model based on Least-Squared Supported Vector Machine (LSSVM) is also developed. The derived models are authenticated with several statistical validation techniques.