L-Leucine Propyl Ester–Fatty Acid-Based Pseudo-Protic Ionic Liquids: Synthesis, Extraction Ability, and Ecotoxicity Prediction by Machine Learning

We synthesized low-toxicity L-leucine propyl ester linoleate and L-leucine propyl ester oleate pseudo-protic ionic liquids (ILs) for benign extraction of Ni(II), Co(II), and Mn(II). The extraction ability order for both ILs was Ni(II) > Co(II) > Mn(II). In addition, we developed a machine learning model with an eXtreme Gradient Boosting regressor algorithm to evaluate and predict the ecotoxicity level of the ILs. An evaluation of the proposed regression model by cross-validation indicates that the model is reliable, with an R² value of 0.71. The prediction results indicate that the newly synthesized ILs are much less toxic than a commercially available IL (methyltrioctylammonium chloride) that is often used for metal extraction.