Work

Classifier surrogates to ensure phase stability in optimisation-based design of solvent mixtures

Year: 2024

Type: article

Abstract: The ability to guarantee a single homogeneous liquid phase is a key consideration in computer-aided mixture/blend design (CAM b D). In this article, we investigate the use of a classifier surrogate of the phase stability condition within a CAM b D optimisation model for designing solvent mixtures with guaranteed phase stability properties. We show how to develop such classifiers for describing multiple candidate mixtures over a range of compositions and temperatures based on the generation of phase stability data using thermodynamic models such as UNIFAC. We test the approach on two solvent design case studies and illustrate its effectiveness in enabling the in silico design of stable mixtures, simultaneously providing a probability of phase stability as an interpretable metric. • A surrogate phase stability model is developed and embedded within solvent design. • The ANN-based classifier is trained on UNIFAC-generated phase stability data. • It achieves high reliability over many solvent mixtures and temperatures. • Improved solutions are found for molecular and process design problems.

Source: Digital Chemical Engineering

DOI: 10.1016/j.dche.2024.100200

Authors: Tanuj Karia, Gustavo Chaparro, Benoît Chachuat, Claire S. Adjiman

Institution: Imperial College London

Language: English

Topic: Phase Equilibria and Thermodynamics

Subfield: Biomedical Engineering

Field: Engineering

Domain: Physical Sciences

Open Access status: gold

Estimated APC paid: $1,800

Funders: Imperial College London, Engineering and Physical Sciences Research Council, Department of Chemical Engineering, Universiti Teknologi Petronas