PhD Position on Ultra-Fast Machine-Learning Interatomic Potentials for Nanoindentation of TiC Materials
For this interdisciplinary, fully funded PhD position, we seek a highly motivated and capable PhD candidate to develop and apply ultra-fast machine-learning interatomic potentials (UFPs, Xie et al., npj Comput. Mater., 2023, 10.1038/s41524-023-01092-7) for long, multi-million-atom molecular dynamics (MD) simulations of different materials families composed of Ti and C, such as titanium carbides. Together, we will identify and push the boundaries of UFPs to simulate the mechanical properties of this family of materials, in particular, hardness via nanoindentation, in agreement with and beyond experimental results.
The successful candidate will model the mechanical properties and plastic deformation of materials with diverse compositions of Ti and C and different structures, including MD-based nanoindentation simulations and defect analyses. To facilitate this, you will further develop the ultra-fast machine-learning potentials used, both methodologically and it's implementation. The current codebase can be found at https://github.com/uf3/uf3
We are looking for candidates with:
• A pertinent master’s degree in computational materials science or a related discipline
• Some knowledge of the theory of materials and experience with computational methods in materials science
• Some experience with machine learning for molecules or materials, in particular, machine-learning interatomic potentials
• Good programming skills in Python
• A friendly, motivated, positive, hands-on, initiative-taking, collaborative attitude and demeanour
• Fluency in English
Any of the following will be a plus:
• Experience developing machine-learning interatomic potentials
• Experience with UFPs
• Experience with high-performance computing, including code optimisation
• Experience with molecular dynamics, ideally with LAMMPS
• Experience with density functional theory calculations, ideally with VASP
• Contributions to a public code repository
The position is in the research group of ELLIS Fellow Matthias Rupp at the Luxembourg Institute of Science and Technology (https://list.lu).
