Optimal Aqueous Biphasic Systems Design for the Recovery of Ionic Liquids

Ionic liquid-based aqueous biphasic systems (IL-ABS) have attracted much attention in both academia and industries due to their superior performance in many applications. In order to better utilize these novel biphasic liquid–liquid systems for recovering hydrophilic ILs from their dilute aqueous solutions, a machine learning (ML)-based ABS design method is proposed for such a purpose in this work. In this method, an ML-based model, i.e., artificial neural network (ANN)-group contribution (GC) model, is employed to predict the phase equilibrium behaviors of IL-ABS. Based on the integration with a computer-aided design technique, the optimal IL-ABS is determined by formulating and solving an optimization-based mixed-integer non-linear programming problem, where the structure of IL-ABS is denoted as the input vector in the ANN-GC model. As a proof of the concept, results of the recovery of 1-butyl-3-methylimidazolium chloride ([C₄mIm][Cl]) and n-butylpyridinium trifluoromethanesulfonate ([C₄Py][TfO]) from aqueous solutions are presented. The ABS [C₄mIm][Cl]-H₂O-(NH₄)₂SO₃ (identified in this work) gives an IL recovery efficiency of 95.0 wt % and a salting-out agent input of 2.36 kg/kg IL recovery, and for the ABS [C₄mIm][Cl]-H₂O-K₂CO₃ (reported in the literature), they are 81.7 and 5.25, respectively. For the second case, our proposed ABS [C₄Py][TfO]-H₂O-KH₂PO₄ gives an IL recovery efficiency of 95.6 wt % and a salting-out agent input of 1.81 kg/kg IL recovery, and for the reported ABS [C₄Py][TfO]-H₂O-(NH₄)₂SO₄, they are 80.6 and 3.16, respectively