Background

Femtoseconds and faster are the natural timescale of electrons and ions in materials. Over the last 40 years (Nobel Prizes in 1993 and 2024), scientists have developed exquisite laser-based tools for tracking material dynamics on these timescales. The problem is such methods (i) typically lack simultaneous nanoscopic spatial resolution, which is necessary to track individual atomic motions, and (ii) are unable to capture stochastic events like ion-hops or lattice fluctuations which often dictate material properties.

In recent years ultrafast X-ray free electron laser (XFELs) have been developed that have the potential to alleviate the two above challenges and follow random processes in materials with nanometre spatial resolution and up-to attosecond time resolution. These XFELs are especially well suited to studying process like ion hops which govern the charging rates battery materials, and lattice fluctuations which drive phase changes in these systems that dictate their durability.

One bottleneck however is the complexity of the design and interpretation of XFEL experiments. This primarily stems from the challenging materials (large/disordered unit-cells, long sampling timescales) and light-matter interactions involved. Consequently, computational tools, rooted in physics, that can predict and rationalise XFEL observables are desperately needed such that XFEL results can reach their full potential.

Aim

This research aims to utilise the latest advances of computational methods (machine learning potentials) and analytical calculation methods (quantum dynamics) to develop robust methods for predicting and interpreting observables from XFEL experiments. It will particularly focus around THz pump, X-ray diffraction probe experiments used to capture ultrafast ion ‘hops’ in battery solid-electrolytes and X-ray photon correlation spectroscopies being implemented to elucidate lattice fluctuations in these systems. But the methodologies developed be applicable to a wide range of materials, making impactful tools for the scientific community. The models and predictions in the project will be tested against real experimental data and used to drive the design of new XFEL experiments.

Outcomes

• Understanding of the ion hopping mechanism in battery solid-electrolyte materials by modelling XFEL experiments that measure changes in lattice structure of battery solid-electrolytes after driving of an ion hop.
• Tools for the predication and interpretation of signals in X-ray photon correlation experiments designed used to measure fluctuating lattice dynamics around phase transitions in battery electrodes.
• Design of new XFEL experimental geometries, accounting for material properties and X-ray light-matter interactions, to capture fluctuating chemical processes in material systems.
• Robust software that can be used by other XFEL users that leverages the latest advances in scientific machine learning and pen-and-paper methods.