PhD in Machine Learning for X-ray Spectroscopy
About the Project
Supervisor: Dr Keith Butler (UCL)
Application deadline: Tuesday 21st July 2026
Start date: 1st October 2026
Location: UCL and Harwell
We invite applications for an exciting PhD studentship at the interface of chemistry, machine learning, and advanced experimental characterisation. The project will develop new machine-learning approaches to interpret high-resolution X-ray absorption spectroscopy data, with the goal of uncovering the atomic-scale structures that control the performance of catalytic materials.
Functional oxide and zeolitic catalysts are central to challenges in the circular economy, clean energy, and environmental protection. However, their activity often depends on subtle defects, vacancies, oxidation-state changes, or unusual local coordination environments that are difficult to determine experimentally. This project will combine state-of-the-art HERFD-XANES measurements, computational chemistry, and machine learning to build workflows that link experimental spectra directly to physically meaningful atomistic structures. The work will draw on emerging approaches for ML-based XANES prediction and extend them by coupling spectrum prediction with structure optimisation and DFT/TDDFT validation.
The student will be based in UCL Chemistry and Diamond Light Source, gaining experience in both a leading university chemistry department and a world-class synchrotron facility. The project will suit a chemistry student with strong programming skills, experience in machine learning, and an enthusiasm for applying computational methods to real experimental data. Prior experience using ML for experimental data analysis, spectroscopy, catalysis, materials chemistry, or atomistic simulation would be especially valuable.
The successful applicant must have or expect to achieve at least a 2.1 honours or equivalent at Masters level degree in chemistry, engineering, analytical chemistry, materials science or related subject.
Find Your Best Opportunity
Tell them AcademicJobs.com sent you!
