Extensive Databases and Group Contribution QSPRs of Ionic Liquid Properties. 3: Surface Tension

Quantitative structure–property relationships (QSPR) for calculating temperature dependence of surface tension (σ) ionic liquids (ILs) in terms group contributions (GCs) is proposed and broadly presented. A statistical learning method including stepwise multiple linear regression two machine methods feed-forward artificial neural network least-squares support vector was applied to express σ as a function GCs. The models were developed using the largest experimental data compilation reported thus far (570 ILs, 1008 datasets, 6114 points). GC assignments, “reference + correction” modeling scheme, well model validation protocol adopted from previous series [Paduszyński, K. Ind. Eng. Chem. Res. 2019, 58, 5322−5338; Paduszyński, 17049–17066]. influence chemical family both cation anion on quality predictions discussed. potential applications estimating critical ILs are Finally, obtained confronted with other literature. In particular, an extensive comparative analysis presented case selected QSPRs accounting atomic topological descriptors.