A support vector machine analysis to predict density of mixtures of methanol and six ionic liquids.

Abstract: Ionic liquids (ILs) are typically mixed together and/or with conventional solvents, and other organic and inorganic compounds to inhibit unfavorable characteristics. Methanol is a widely used solvent and additive in many industrial applications and can be beneficially combined with ILs. Ionic liquids in isolation have some intrinsic disadvantages such as high viscosity. Pumping viscous liquids is a challenge for most industrial applications. This undesirable feature is typically tackled by combining ILs with specific solvents. Here, the binary attributes of IL-solvent combinations are assessed and correlated utilizing 731 data records from published sources. A support vector machine (SVM) algorithm is applied to establish reliable correlations between binary density of the IL systems and the methanol component they contain. Error analysis of the results suggests that the proposed SVM model is highly reliable for the purpose of determining the density of IL-methanol systems with a high degree of accuracy such as coefficient of determination (R¯) of greater than 0.99.Graphical abstract: [ABSTRACT FROM AUTHOR]