Advancing Plasma-Facing Materials: A Data-Driven Approach to Tungsten Alloy Design for Fusion Reactors (Materials Strand Project)

Tungsten (W) and its alloys are among the most promising candidates for plasma-facing materials (PFMs) in fusion reactors due to their high melting point, thermal conductivity, and resistance to sputtering. However, the extreme environments within fusion reactors—including high temperatures, intense neutron fluxes, plasma interactions, and irradiation bombardment—pose significant challenges to their structural integrity and longevity. These conditions can induce complex microstructural and chemical changes in W and its alloys, potentially degrading their performance. Understanding and predicting the properties of tungsten alloys under these severe conditions is crucial for developing materials capable of withstanding the operational demands of future fusion reactors.

Experimental insights are limited in this field due to accessibility constraints, making it essential to use first-principles calculations combined with machine learning to understand and design W-based alloys for PFMs. Here, we propose to employ a combined approach of density functional theory (DFT) calculations and machine learning (ML) models to explore and predict the behavior of tungsten and its alloys as PFMs under high temperatures. Key properties, such as defect formation energies, binding energies, elastic constants, hardness, and migration barriers, across various alloy compositions, will be computed. Thermal conductivity will also be considered. These properties will serve as input data for training ML models, enabling rapid predictions of tungsten alloy performance under different conditions and compositions.

The DFT results will form a database, with composition, defects, and atomic interactions as feature-space inputs in the ML models for the optimal design of PFMs. This approach will allow efficient screening of alloy candidates, guiding experimental efforts and reducing reliance on time-intensive DFT calculations.

This project aims to deliver a predictive model that identifies tungsten alloys with optimal properties for use as plasma-facing materials. The findings are expected to highlight specific alloying elements and compositions that enhance resilience to thermal stress, neutron flux, and irradiation effects. The resulting framework will provide valuable insights for designing advanced PFMs, supporting the broader goal of developing safe, efficient, and economically viable fusion energy systems.

There will be opportunities to present at and attend international conferences, and there may be opportunities for visits to industrial partner.

This project is offered by University of Liverpool. For further information please contact: Xue Yong (xue.yong@liverpool.ac.uk).

This project may be compatible with part time study, please contact the project supervisors if you are interested in exploring this.