View article

In this work, new quantitative structure-property relationship (QSPR) models were developed to predict melting and freezing points of deep eutectic solvents (DESs). Modified particle swarm optimization (MPSO) based on multiple linear regression (MLR) was applied to two data sets of DESs, constituted of chloride, diverse cations and various hydrogen bond donors (HBD), and optimum subsets of descriptors were selected to develop two linear models. The experimental data sets of 91 melting points and 90 freezing points were divided such that 80% of the data as training sets were used to develop the models, and remaining 20% as test sets for external validation of the models. The melting and freezing point models resulted in squared correlation coefficient values of 0.795 and 0.764, respectively. The obtained results in this paper indicate the good capability of QSPR models for prediction of melting and freezing …