Mathematical Model of the Phase Diagrams of Ionic Liquids-Based Aqueous Two-Phase Systems Using the Group Method of Data Handling and Artificial Neural Networks

Abstract

Development of ionic liquid-based aqueous two-phase systems as a new viewpoint in the expansion of research in the field of biological materials separation depends on accurate determination of phase diagram. In this work, the efficiency of artificial neural network was studied aiming to forecast the formation possibility of phase diagrams of aqueous two-phases systems for the ability of range of ionic liquids composed of different anions with a selected salt. In order to investigate effects of the anion of ionic liquids on phase diagram, this study was performed on 472 of experimental data. On the basis of the accurate set of statistical measurements obtained, a good agreement between the experimental data points and the predicted values was gained. Furthermore, the group method of data handling was applied to model the molality of ionic liquids and a reasonable agreement was obtained between experimental data and the predicted values of this model.