Computer-Aided Entrainer Design for Extractive Distillation Using Infinitely Sharp Splits

Abstract We present an integrated product-process design framework for ranking heavy entrainer candidates for minimum-boiling azeotrope separation, corresponding to the Serafimov class (1.0-1a), the most common such case. The framework combines computer-aided molecular design (CAMD) with the infinitely sharp split (ISS) shortcut method. This enables rapid computation of the minimum entrainer-to-distillate ratio ((E/D)min) and reflux ratio (Rmin) from group-contribution property predictions. Both entrainer screening and novel candidate design were validated through rigorous process simulations and economic optimization. For the well-studied acetone/methanol separation, the top-ranked novel candidate achieved a competitive total annual cost relative to that of established benchmarks, confirming the (E/D)min-based ranking. The generalizability of this ranking was further supported by an independent literature case for isopropanol dehydration, where the predicted entrainer ordering matched the reported economics. These results show that CAMD-ISS, while not capturing the regeneration difficulty directly, is a more reliable indicator of process economics than the infinite-dilution selectivity alone.