| 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 |
Multiscale modeling for studying corrosion-induced hydrogen embrittlement in zirconium |
| Author(s) |
Shubham Pandey, Kyle Starkey, Volker Eyert, Erich Wimmer |
| On-Site Speaker (Planned) |
Shubham Pandey |
| Abstract Scope |
Increasing the nuclear fuel life cycle in commercial reactors is vital for achieving high fuel burnup. However, extending burnup degrades zirconium alloy cladding due to oxidation and hydrogenation. In light-water reactors, zirconium reacts with water coolant, forming a zirconia film and releasing hydrogen gas. Some hydrogen dissolves in the zirconium lattice, but excess hydrogen precipitates as hydrides (ZrHx), causing hydrogen embrittlement. This reduces the ductility and increases the likelihood of cladding failure. Predicting the microstructure and phase transitions of ZrHx and ZrO2 is critical. The phase-field method models morphological and microstructural evolution, using inputs from quantum-mechanical calculations and atomistic modeling. In this work, ab-initio calculations are employed to develop a machine-learned potential (MLP) in the Zr-O-H phase space, which is used to develop a phase-field model. The MLP and phase-field models predict phase transitions in zirconia and successfully model the columnar growth of oxide film, as seen in experimental micrographs. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Computational Materials Science & Engineering, Machine Learning, Phase Transformations |