Speaker
Description
The analysis of QENS data has historically used analytical models that describe the physics of some diffusive phenomena. However, this approach is often limited by the assumptions of these models. Classical molecular dynamics simulations offer a complementary approach to probe diffusion in materials, covering similar time and length scales to QENS. However, typically, it isn't easy to obtain quantitative agreement between simulation and experiment.
In this presentation, we will introduce recent work to improve the use of simulation in the analysis of QENS data by improving the simulation analysis we can perform and the simulations themselves. Specifically, we will discuss the impact that accurate estimation of diffusion coefficients from simulations can have on QENS analysis [doi:10.1021/acs.jctc.4c01249] and the role the machine learning interatomic potentials can have on the analysis of QENS data. We will focus on simple model systems but highlight the importance this may have for polarised QENS measurements of complex systems, including energy materials.
