Current Trends in QSAR and Machine Learning Models of Ionic Liquids: Efficient Tools for Designing Environmentally Safe Solvents for the Future

To support the initiative “Green Chemistry,” ionic liquids (ILs) are one of the promised alternatives to traditional organic solvents. A reasonable degree of chemical and thermal stability, high polarity, promising solvating characteristics, negligible volatility, low melting point, high ionic conductivity, and minimal environmental release make the ILs “green solvents” to consider for multiple usages. However, recent studies have raised doubts about their nontoxic claim and reported related acute to chronic toxicity and the risk of bioaccumulation varying on structural elements. A series of potential applications of ILs in the chemical industry and different industrial sectors compel us to check all possible toxicity associated with ILs. The experimental toxicity evaluation of ILs is a time-consuming task involving money and ample resources. Therefore, in silico methodologies like quantitative structure–activity relationship (QSAR) models and machine learning (ML) tools offer the opportunity to fill the toxicity data gaps by predicting the toxicity of new and untested ILs. Predictive QSAR models are also interpretative for the environmentally safe-by-design of ILs. The present chapter discusses QSAR and ML models and the mechanism of toxicity of ILs to introspect their health and environmental risk assessment issues.