| About this Abstract |
| Meeting |
2025 TMS Annual Meeting & Exhibition
|
| Symposium
|
Thermodynamics and Kinetics of Alloys III
|
| Presentation Title |
Global stability models of multi-principal element alloys |
| Author(s) |
Lin Wang, Zhengda He, Bin Ouyang |
| On-Site Speaker (Planned) |
Lin Wang |
| Abstract Scope |
Multi-principal element alloys (MPEAs) have rapidly gained attention due to their unique mechanical, magnetic, and catalytic properties across various fields. However, the enormous compositional space of high entropy alloys makes synthetic exploration challenging. In this work, we present a physical model and a deep learning model that can predict synthetic accessibility of MPEAs. Both models are trained on over 120,000 MPEA structures computed using density functional theory while the phase stability are evaluated with computational phase diagrams. These models enable the prediction of the decomposition energy of any MPEA made from 28 typical metals with reasonable accuracy. We will also provide a thorough comparison of the performance between the physical model and the deep learning model, highlighting some interesting trends in ultra-high entropy alloy systems. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
High-Entropy Alloys, Machine Learning, Computational Materials Science & Engineering |