Accurate prediction of electrical conductivity of ionic liquids + propylene carbonate binary mixtures

Abstract

Ionic liquids (ILs) have remarkable features which make them applicable in numerous processes as environmentally friendly and promising chemicals. However, high viscosity and moderate conductivity of pure ILs have limited their application. Propylene carbonate (PC) is one of the most important and widely used compounds which is utilized in combination with pure ILs to overcome these challenges. Electrical conductivity is one of the most significant physical properties which affect the transport capabilities of ILs. Therefore, looking for a simple and accurate model to predict the electrical conductivity of ILs-PC binary mixtures is crucial. In this communication, a new model based on the least square support vector machine (LSSVM) was developed for accurate estimation of electrical conductivities of different ILs in PC solution under various operating conditions. Coupled simulated annealing (CSA) algorithm was used for optimizing the LSSVM model. Moreover, the presented model was successfully compared with other well-known models found in the literature. The results demonstrated that the predicted values obtained by the proposed LSSVM were in excellent agreement with experimental electrical conductivity values with R2 of 0.998 and 0.983 for training and testing data, respectively. Additionally, it was observed that the results of the proposed model exhibit higher reliability and accuracy in comparison to the other correlations. Finally, a sensitivity analysis was performed on influencing factors and results showed that IL molecular weight, temperature, and IL concentration has respectively the highest effect on electrical conductivity.