Computational discovery of green deep eutectic solvents for sustainable acetonitrile–water separation

초록

Acetonitrile and water form an azeotrope that is difficult to separate by conventional distillation. Liquid–liquid extraction (LLE) using deep eutectic solvents (DESs) provides an energy-efficient and sustainable alternative for breaking this azeotrope. In this study, we present a large-scale in-silico screening strategy to identify optimal solvents for the extractive separation of acetonitrile–water mixtures. Our three-stage workflow assessed (i) intermolecular interaction energies via molecular dynamics (MD) simulations, (ii) partition coefficient (log P) from free-energy calculations, and (iii) environmental, health, and safety (EHS) profiles of the DES components using QSAR models. From an initial library of 1014 candidates, the workflow identified two ChCl-based DESs—paired with 2-methyl-4-hydroxypyridine and L-α-hydroxyisovaleric acid—that combined high extraction efficiency with favorable green metrics. Ternary LLE MD simulations confirmed the preferential partitioning of acetonitrile into the DES-rich phase, characterized by enhanced DES–acetonitrile coordination and reduced acetonitrile–water contacts. Subsequent molecular analysis revealed that effective DESs combine strong hydrogen-bond donation to acetonitrile with balanced hydrophobicity to repel water, establishing key design principles for high-performance solvents in aqueous azeotropic mixtures. This study highlights the power of computational solvent design to accelerate the discovery of greener azeotrope breakers and advance sustainable chemical separation processes. © 2025 Elsevier B.V.

키워드