A predictive quantitative structure–property relationship for glass transition temperature of 1,3-dialkyl imidazolium ionic liquids

Abstract

Glass transition temperature is one of the main criteria for the evaluation of the potential options for electrolyte application. In this communication, the attention was focused on the prediction of glass transition temperature of 1,3-dialkyl imidazolium ionic liquids which could be considered as potential future electrolytes. For this purpose, the quantitative structure–property relationship (QSPR) method is employed to develop two models for determination of glass transition temperature. In this study, both anion and cation descriptors are taken into account for model development. In the first part of this study, genetic function approximation is applied for model’s parameter selection. In order to study the nonlinear relations between the selected molecular descriptors and the glass transition temperature, least square support vector machines (LSSVM) was successfully applied in second part of this study. Furthermore, the genetic algorithm (GA) optimization methods are respectively implemented to optimize the LSSVM model parameters. Consequently, we obtain two predictive models with satisfactory results qualified by the following statistical parameters: absolute average deviations of the predicted properties from existing experimental values by the GFA linear equation: 2.68%, squared correlation coefficient: 0.91 and root mean square : 5.717 K and 1.38% and 0.97 and 4.05 K evaluated by the LSSVM–QSPR model, respectively.