Accelerating Fusion Materials Development Through Multi-Scale Simulations and Surrogate Modeling

The realization of commercially viable fusion energy hinges on the development of advanced materials capable of withstanding extreme fusion environments. Current materials lack the necessary combination of radiation resistance, low activation, and robust thermal and mechanical properties. This presentation demonstrates how open-source tools such as LAMMPS, OpenMC, MOOSE, and Dakota can be integrated into workflows that bridge atomic, mesoscale, and continuum simulations to predict material properties under fusion-relevant conditions. We also address the computational challenges inherent in these workflows by incorporating surrogate modeling techniques, which drastically reduce computational costs while maintaining accuracy. This scalability enables rapid testing and screening of potential candidate materials, facilitating the identification of alloys with enhanced performance characteristics. Through this approach, we provide a streamlined, predictive framework for materials discovery, advancing the timeline for deploying radiation-resistant, low-activation materials essential for sustainable fusion energy systems.