Prediction of Surface Tension and Surface Properties of binary mixtures Containing Ionic Liquids Using Thermodynamics and Artificial Neural Network Models

Abstract

In this paper, surface tension and surface properties of binary mixtures containing alcohol (methanol and ethanol) ionic liquid (1-bptf, 1-b-3-mptf and 1-b-4-mptf)) have been determined using thermodynamics and theoretical methods at various temperatures (293.15-323.15) K. First, the surface tension and the surface tension deviation data over the whole mole fraction range are correlated by thermodynamic based models: Fu et al. (FLW) and Myers-Scott (MS), respectively, and then a new approach, the interaction energy between alcohol and ionic liquid (U12) has been calculated with the results of FLW model. The results show that at fixed temperatures in each system, the value of U21 increased by increasing the size of the ionic liquid cation. Second, the surface tension was predicted by using artificial neural network (ANN) method and the results were compared with thermodynamics method. On comparing the computed values of surface tension (by three methods of FLW, MS and ANN) with experimental data, satisfactory results have been observed (the mean relative standard deviations obtained from the comparison are less than 3.5%). However, among the applied methods, ANN shows the best agreement between the experimental and predicted data because it finds the best nonlinear relationship between the mole fraction and either surface tension or surface tension deviation. The results of this study provide useful information on the component interactions in the surface and the bulk phases.

Keywords

• surface tension
• ms
• artificial neural network
• flw