About this Abstract |
Meeting |
2024 TMS Annual Meeting & Exhibition
|
Symposium
|
Computational Discovery and Design of Materials
|
Presentation Title |
Unraveling the Mechanisms of Stability in CoMoFeNiCu High Entropy Alloys via Physically Interpretable Graph Neural Networks |
Author(s) |
Miguel Tenorio, James Chapman |
On-Site Speaker (Planned) |
James Chapman |
Abstract Scope |
High entropy alloys (HEA) have become a topic of significant interest due to their combinatorial nature. In particular, the HEA system CoxMo70-xFe10Ni10Cu10 has been studied extensively due to its reported superiority as a catalyst for ammonia decomposition. However, such catalytic reactions take place at elevated temperatures, potentially leading to phase separation of the HEA. Here, we aim to understand the structural features of these HEA systems that are responsible for stability within the catalyst operational temperature range. To this end, we combine density functional theory (DFT) calculations of mixing free energies with physics-inspired graph neural networks (GNN), to predict HEA stability. We show that by learning the mixing free energy with our GNN framework we can rank geometric HEA descriptors based on their importance towards stability. This work showcases the power of combining DFT with GNN to design complex materials with targeted properties, bridging the gap between simulations and experiments. |
Proceedings Inclusion? |
Planned: |
Keywords |
Computational Materials Science & Engineering, Machine Learning, High-Entropy Alloys |