Research Case Study - Designing Solvent Molecules to Optimise Process Performance
The performance of a manufacturing process for a given product or product range depends not only on the design of the units and the choice of operating conditions, but also on the choice of materials used for processing, such as solvents and catalysts. This can be illustrated by considering some of the possible impacts of making a different choice of solvent: a judicious choice can lead to an increase in the rate of a reaction by several orders of magnitude, the elimination of side products, an improved effectiveness of a separation, or even a change in the shape of particles produced in a crystalliser, and hence in the quality of the product. Furthermore, solvents account for a significant portion of the environmental impacts and energy requirements of processes; there is much scope for improving manufacturing performance in this area.
Under this research theme, the Molecular Systems Engineering (MSE) Group develop predictive models and systematic methods that make the identification of the best solvent possible. The aim is to enable engineers to identify the most appropriate design by manipulating decision variables at the process and molecular scales simultaneously, recognising explicitly the close and complex interactions between the solvent choice and process performance. Such approaches serve as a guide to the designer, rapidly narrowing in on a few possible options within a large space of possibilities, which can then be tested experimentally.
The MSE group have successfully developed an approach to identify solvents that enhance the rates of chemical reactions among a design space of thousands of possible molecules. This is very useful in the development of pharmaceutical or agrochemical manufacturing processes, where most reactions take place in a solvent and where the choice of reaction medium has many implications on downstream processing. Standard techniques for solvent choice rely on extensive experimental investigations. The MSE Group's approach combines quantum mechanical calculations for a small number of solvents with the use of advanced optimisation techniques to explore the full design space. It has been shown to lead to an improvement in the reaction rate of 40%. Using similar principles, the group have developed a systematic approach to design natural gas cleaning processes for the removal carbon dioxide, which is present in increasing quantities in natural gas fields due to injection. Here, the MSE Group have built on formal statistical mechanics approaches and process modelling (Figure 1) to predict the impact of the solvent on process performance, demonstrating the strong link between process economics and molecular structure. Based on these advances, the group are currently extending the range of design problems which can be tackled within this integrated molecular and process design framework.