About this Abstract |
| Meeting |
2027 TMS Annual Meeting & Exhibition
|
| Symposium
|
Modeling, AI Applications, and Method Development in Nuclear Materials
|
| Presentation Title |
Point Defect Evolution in BCC-Complex Composition Alloys Investigated via Machine-Learning Potentials |
| Author(s) |
Shu-Ming Chen, Nick Crnkovich, Gaurav Arora, Muhammad Waqas Qureshi, Adrien Couet, Kavin Ammigan, Dane Morgan, Izabela Szlufarska |
| On-Site Speaker (Planned) |
Shu-Ming Chen |
| Abstract Scope |
Body-centered cubic (BCC) alloys are promising candidates for radiation-resistant materials. However, atomic-level understanding of the defect kinetics in BCC Complex Composition Alloys (CCAs) remains limited due to complex chemical environments. In this work, we employ machine learning potentials (MLIPs) coupled with Monte Carlo/molecular dynamics (MC/MD) simulations to investigate the defect behavior near B2/BCC interfaces and in radiation-induced segregation (RIS) in BCC-CCAs.
Specifically, building on the atomic cluster expansion framework, we developed an MLIP for Al0.5MoTaTi alloys. Using this potential, we simulated the evolution of B2/BCC interfaces across a wide temperature range and elucidated the ability of B2/BCC interfaces to act as defect sinks. In parallel, a moment tensor potential for CrMnV was constructed to determine point-defect-mediated diffusion coefficients via MD simulations, enabling RIS prediction.
Taken together, these results provide mechanistic insight into defect behavior in BCC-CCAs and establish a computational foundation for the design of radiation-tolerant alloys. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Computational Materials Science & Engineering, High-Entropy Alloys, Machine Learning |