Computer-aided solvent selection for multiple scenarios operation of limited-known properties solute

Abstract

Solvents have been applied for both production and separation of the complex chemical substance such as the pyrrolidine-2-carbonyl chloride (C5H8ClNO). Since the properties of the target substance itself are largely unknown, the selection of the solvent is limited by experiment only. However, the reaction carried out in conventional solvents are either afforded low yields or obtained slow reaction rates. Moreover, the solvents are also highly toxic and environmental unfriendly. Alternative solvents are required to enhance the production and lessen the harmful effect toward both organism and environment. A costly, time-consuming, and laborious experiments are required for acquiring a better solvent suite for production and separation of these complex compounds; whereas, a limited improvement can be obtained. On the other hand, the combination of the state-of-the-art thermodynamic models can provide faster and more robust solutions to this solvent selection problem. In this work, a framework for solvents selection in complex chemical production process is presented. The framework combines a group-contribution thermodynamic model and a segment activity coefficient model for predicting chemical properties and solubilities of the target chemical in newly formulated solvents. A guideline for solvent selection is also included. The potential of the selected solvents is then analysed and verified. The improvement toward the production yield, production rate, and product separation is then discussed.