Computational Modelling of High Performance Permanent Magnets

Permanent magnets play an essential role in a range of technologies, with electric vehicle motors currently being the most rapidly growing area. The best-performing magnets are based on alloys of conventional magnetic transition metals like iron or cobalt, with rare-earth elements like neodymium or samarium. The incredible magnetic properties of the rare earths, specifically the lanthanide elements, are due to their highly-localised shells of 4f electrons, which form atomic-scale magnetic moments which polarise the other electrons in the material.

A key focus of industry and academia is reducing the rare-earth content in the magnet, whilst maintaining high performance. However, these magnets are complex objects, consisting of multiple grains and phases which interact both through short-range quantum mechanics and long-range magnetostatic forces. Furthermore, the highly-correlated nature of the 4f electrons poses a challenge to computational modelling, with widely-used density-functional theory (DFT)-based methods struggling to describe the lanthanides accurately.

The aim of this project is to build on several developments [Phys. Rev. Lett. 132, 056703 (2024)], [Phys. Rev. Materials 3, 101401(R) (2019)], [Phys. Rev. B 107, L020401 (2023)] to advance our ability to carry out accurate, predictive modelling of the type of rare earth magnets used in industry. Depending on the background and interests of the student, this project could focus on developing theoretical methods, implementing such methods into software, or applying the methods to study key problems. The student would benefit from interactions with internationally-leading experimental groups in the USA and Japan with expertise in synthesising and characterising these materials, as well as with other computational groups which specialise in modelling at different length scales. This DPhil project would suit a student with a materials, physics or chemistry background with an interest in electronic structure and magnetism.

For more information contact Christopher Patrick or Jonathan Yates.