AI-Driven Drying for Advanced Catalyst and Membrane Manufacturing

About the Project

Membranes and catalyst monoliths are critical components in clean fuel production and emission control systems, functioning as molecular filters and catalytic reactors whose performance and durability are strongly governed by the drying stage - a rate-limiting process that dictates microstructure, defect formation, and long-term stability. However, current industrial drying remains largely empirical, lacking real-time insight into internal moisture distribution, stress evolution, and phase transitions, and failing to systematically link formulation variables and process conditions to final product quality. Artificial intelligence and machine learning can transform drying by enabling data-driven modelling, real-time optimisation, and scalable design, outperforming traditional empirical and physics-only approaches in handling highly nonlinear, multivariate drying dynamics , while hybrid physics - AI models have demonstrated near-perfect predictive accuracy for internal concentration and mass transfer behaviour during drying processes ; moreover, the integration of AI with advanced sensors and process analytical technologies allows real-time monitoring, adaptive control, and significant improvements in energy efficiency and product quality.

This PhD project will develop an AI-enabled, dynamic drying control framework for catalyst monoliths and membranes by integrating physics-based models, machine learning, and process analytical technology (PAT) to deliver predictive control and optimisation based on material properties and drying unit design, enabling defect-free manufacturing, controlled microstructure, and energy-efficient, sustainable production. The research will be conducted in close collaboration with Johnson Matthey plc, a global leader in platinum group metals (PGMs) technologies and a pioneer in advanced catalysts, hydrogen solutions, and sustainable process technologies, with a long-standing track record of translating cutting-edge science into large-scale industrial impact, and the University of Birmingham (School of Computer Science, School of Metallurgy and Materials), providing a unique opportunity to translate cutting-edge AI research into industrial-scale impact.

Funding Notes

International applications are welcomed. This project is partially funded by Johnson Matthey, which covers 20% fee for international students and 40% fee for UK students. The remaining will need to be self-funded.