Neuro-evolutionary neural network for the estimation of melting point of ionic liquids

Abstract

Ionic Liquids (ILs) are salts known for their low melting point, wide liquid phase, and their low toxicity. Also, ILs have an extensive range of applications. Choosing the “best” IL for an application requires the prior knowledge of the physicochemical properties of all the existing ILs which is currently inadequate, furthermore, the synthesis of ILs is generally expensive and time-consuming; thus, a large-scale study is infeasible. Therefore, an estimation system of the melting points could solve partially this problem, the estimation is complex since the ILs exhibit unconventional behavior and the information available may be inaccurate. This paper presents a neuro-evolution neural network for the estimation of the melting point of ILs.