Recent development in machine learning of polymer membranes for liquid separation

Abstract

Emerged as a transformative technology, machine learning (ML) has demonstrated unprecedented success in the design and discovery of new materials. Over the past few years, we have witnessed the increasing applications of ML in the field of membranes. In this review, we present recent development in ML studies of polymer membranes for liquid separation, including water treatment at both process and membrane levels, pervaporation based water–organic and organic–organic separation, and solvent recovery. Along with these latest advances, we suggest directions for future ML exploration: (1) Membrane Separation Genome, (2) Membrane and process representations, (3) Deep learning algorithms, (4) ML enhanced high-throughput approach, and (5) Digital Membrane Separation. These directions provide new opportunities and challenges in this rapidly evolving field to accelerate membrane development for high-performance liquid separation.