Accelerating Solvent Design Optimisation with Group-Contribution Machine Learning Surrogate Classifiers

Asserting the phase stability of multi-component mixtures is an important task in computer-aided mixture/blend design (CAMbD), but it is often hindered by the lack of reliable and tractable models. In this paper, we propose a group-contribution machine-learning (GC-ML) method to predict phase coexistence for a large set of ternary mixtures consisting of two solvents and one (fixed) solute. Each solvent is represented by a vector of functional group numbers, encoded by integer values. The solvent vectors are combined with mixture composition and temperature to form the input features to a GC-ML surrogate classifier, which distinguishes between four types of stable phase configurations as possible outputs: liquid (L), solid-liquid (SL), liquid-liquid (LL) or solid-liquid-liquid (SLL). To explore the performance of the trained GC-ML multi-classifier, it is embedded as a surrogate phase-stability constraint in the optimisation of an ibuprofen crystallisation process. A two-step solution strategy is proposed, iterating between a surrogate-based subproblem and a rigorous UNIFAC-based subproblem, to design binary solvent mixtures that improve the yield of ibuprofen. A high classification accuracy score of over 0.96 is achieved in identifying the correct phase configurations with the surrogate model, making it possible to accurately predict whole phase diagrams for numerous mixtures. Moreover, reliable solutions to the crystallisation design problem are generated with the two-step strategy, yielding objective function values close to those with the UNIFAC-based model at a lower computational cost. These results reveal the value of the proposed surrogate model in guiding the search for better solvent mixtures in CAMbD applications.