| About this Abstract |
| Meeting |
2025 TMS Annual Meeting & Exhibition
|
| Symposium
|
Bridging Scale Gaps in Multiscale Materials Modeling in the Age of Artificial Intelligence
|
| Presentation Title |
Machine Learning - Kinetic Monte Carlo Investigation on Sluggish Interstitial Diffusion in Fe-Ni-Cr-Cu-Co High Entropy Alloys |
| Author(s) |
Wenjiang Huang, Xianming Bai |
| On-Site Speaker (Planned) |
Wenjiang Huang |
| Abstract Scope |
Sluggish diffusion in high entropy alloys (HEAs) has been a topic of considerable debate, potentially underpinning the exceptional properties of these materials. Traditional atomic simulations, such as classical and ab initio molecular dynamics (MD) methods, are constrained by short-time and small-size scales, rendering them less efficient for modeling atomic diffusion in HEAs with extreme chemical complexity. This study addresses this limitation by combining machine learning (ML) with Kinetic Monte Carlo (KMC) simulations to investigate interstitial dumbbell diffusion in HEAs, a critical defect governing the microstructural evolution under non-equilibrium irradiation conditions. Using the model Fe-Ni-Cr-Co-Cu HEA system, our ML models accurately and efficiently predict migration energy barriers of various types of interstitial dumbbells at different local atomic environment on-the-fly, thereby ML-KMC significantly overcoming the simulation time bottleneck. This integrated approach allows for an in-depth analysis of long-time diffusion behaviors, revealing both sluggish and chemically-biased interstitial diffusion phenomena. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
High-Entropy Alloys, Machine Learning, Modeling and Simulation |