PhD Position in Enhancing ageing predictions through Machine Learning Force Fields (MLFFs)

Research theme: Computational Materials Chemistry 

This 3.5 year PhD position is fully funded by an external sponsor (AWE). The stipend was costed as follows: £19,815 (Y1), £20,410 (Y2), £21,023 (Y3), and £10,827 (Y4, 6m). Home students are eligible for this project. If you need guidance on this, please email the supervisor. 

The goal of this project is to explore the behaviour of hydrogen in lithium oxides, hydroxides and hydrides using a combination of solid-state density-functional theory (DFT) and machine-learning force fields (MLFFs).

DFT methods will be used to study materials of interest including pristine and defective Li2O, LiOH, LiH and their major surfaces. The simulations will be used to train machine-learned forced fields (MLFFs) to explore hydrogen diffusion using molecular-dynamic (MD) simulations. The use of MLFFs will allow for larger simulation cells and longer timescales to be accessed, whilst retaining the accuracy of DFT calculations, which we hope will allow for more accurate prediction of e.g. diffusion coefficients than is possible with existing techniques.

In the first year, you will carry out an in-depth literature review on the methodology and materials to be studied, and you will be trained in the main techniques and software to be used during the project and perform initial simulations and MLFF training/validation on the bulk materials.

In the second year, you will identify and study defects and hydrogen diffusion pathways in the bulk materials, extend/validate the MLFF for these simulations, and use it to perform simulations at larger length scales and longer timescales than are accessible with DFT.

In the third year, you will then investigate the interactions of hydrogen with material surfaces and surface diffusion, again using DFT and an updated MLFF. As a stretch goal, if time allows you will also explore the use of grand-canonical Monte-Carlo (GCMC) simulations to study hydrogen solubility.

In the final six months of the project, you will consolidate the simulations from across the project, write up a thesis, and engage in knowledge transfer with AWE (e.g. through workshops).

You will gain a well-rounded experience of computational materials modelling, including cutting-edge machine-learning techniques, as well as experience of using high-performance computing (HPC) and programming/scripting in languages such as the UNIX shell (bash) and Python.

The project is fully funded and sponsored by AWE, and you will interact regularly with supervisors from AWE’s modelling team as well as the supervisory team and research group at Manchester.

Applicants should have, or expect to achieve, at least a 2.1 honours degree or a master’s (or international equivalent) in a relevant science or engineering related discipline.

To apply, please contact the supervisors for this project; Dr Skelton - jonathan.skelton@manchester.ac.uk and Prof Kaltsoyannis - nikolas.kaltsoyannis@manchester.ac.uk. Please include details of your current level of study, academic background and any relevant experience and include a paragraph about your motivation to study this PhD project.

Advert information