Structural feature based computational approach of toxicity prediction of ionic liquids: Cationic and anionic effects on ionic liquids toxicity

Abstract

The density functional theory (DFT) based a unique model has been developed to predict the toxicity of ionic liquids using structural-feature based quantum chemical reactivity descriptors. Electrophilic indices (ω), the energy of highest occupied (EHOMO) and lowest unoccupied molecular orbital, (ELUMO) and energy gap (∆E) were selected as the best toxicity descriptors of ILs via Pearson correlation and multiple linear regression analyses. The principle components analysis (PCA) demonstrated the distribution and inter-relation of descriptors of the model. A multiple linear regression (MLR) analysis on selected descriptors derived the model equation for toxicity prediction of ionic liquids. The model predicted toxicity values and mechanism are very consistent with observed toxicity. Cationic and side chains length effect are pronounced to the toxicity of ILs. The model will provide an economic screening method to predict the toxicity of a wide range of ionic liquids and their toxicity mechanism.