Royal Society University Research Fellow and Assistant Professor
Durham University
Research Project
Machine learning classical density functional theory
Molecular dynamics and Monte Carlo simulations are powerful techniques to understand the behaviour of liquids. However, they can be computationally expensive, and much of the time, the exact details of what individual atoms or molecules do are unnecessary to understand the problem at hand. Approaches rooted in classical density functional theory (cDFT), on the other hand, provide the average equilibrium density and free energy simply by minimising a free energy functional. However, the free energy functional is, in general, not known.
In this project, we will use machine learning techniques to establish the free energy functional for various liquids, including water, molten salts, and mixtures. This will pave the way to understanding liquids and liquid-solid interfaces accurately and at much reduced computational cost.
Project results
With machine learning representations of classical density functional theory, we have investigated the adsorption of a binary mixture under confinement. Our results show that features associated with the azeotropic point (a point on the phase diagram where liquid and vapor have the same composition) persist under thermodynamic conditions far from the azeotropic point. This result may have implications for chemical separation processes.
How has your research benefitted from using Bede?
We have used Bede to train neural representations of classical DFT. Going forward, I anticipate that we will start using it generate training data from molecular simulations.
Has using Bede enabled you to apply for further research funding, and how does it relate to your initial project?
I have my Royal Society University Research Fellowship renewal pending, for which I plan to use resources from Bede.