Research Fellow in Machine Learning for Materials Design

The Chemistry Department at University College London is the oldest in England, and today is one of the best in the UK, being ranked 3rd in the UK for its world-leading research in REF2021. We are located in Bloomsbury, at the heart of London, and offer an exciting and vibrant environment in which to study in one of the UK's top universities. The Department of Chemistry at UCL is committed to supporting excellence in both research and teaching. The department offers undergraduate BSc and MSci programmes in Chemistry and currently teaches ~750 undergraduates registered in Chemistry as well as students who select Chemistry on the Natural Sciences programme and first year Chemistry for life scientists.

This post is funded through the EPSRC grant: Barocalorics for green cooling: from understanding to design. We seek to bring together a team of experimental and computational scientists to use the latest advances in machine learning and advanced characterisation to understand and design new materials for more sustainable cooling systems.

The appointee will be working in the Materials Design and Informatics Group based in UCL Chemistry and will be responsible for developing new machine learning models to understand the structure, dynamic and phase changing behaviour of a range of solid-state cooling materials. You will be part of the team collecting and analysing neutron scattering data. You will develop workflows for fitting new forcefield models designed to reproduce results from high-level electronic structure theory and neutron experiments. You will also develop new methods, drawing on concepts from information theory, to help understand and ultimately design for entropy changes across phase boundaries.

The project is a collaboration with Dr. Anthony Phillips at Queen Mary University of London whose group will synthesise new barocaloric materials and Dr. Helen Walker at ISIS Neutron and Muon Source, who will lead advanced characterisation of these systems. If you are passionate about the using simulation to understand cutting edge experimental data and was to push this field forward through the latest machine learning techniques, then we hope that you will apply.

The postholder will be expected to conduct research focused on developing workflows for fitting machine-learned potentials to multi-modal experimental and theoretical reference data. This will include performing atomistic simulations to support and complement experimental studies, as well as creating new approaches to better understand and design for entropy changes within these systems.

Customer advert reference: B04-06401

Advert information