| About this Abstract |
| Meeting |
2025 TMS Annual Meeting & Exhibition
|
| Symposium
|
Elucidating Microstructural Evolution Under Extreme Environments
|
| Presentation Title |
Accelerating Nuclear Material Discovery: Integrating Machine Learning With In-Situ Ion Irradiation Experiments |
| Author(s) |
Kevin Field, Hangyu Li, Ian Steigerwald, Ethan Poselli, Robert B. Renfrow, T.M. Kelsy Green, Boopathy Kombaiah, Charles A. Hirst |
| On-Site Speaker (Planned) |
Kevin Field |
| Abstract Scope |
This talk explores the innovative integration of machine learning (ML) with in-situ ion irradiation experiments to accelerate the discovery of nuclear materials and irradiation phenomena. First, we introduce a real-time ML pipeline for defect detection, tracking, and quantification, highlighting methods for stabilizing in-situ videos with mobile key points and enabling 3D reconstructions. In the second part, we demonstrate the framework’s effectiveness with a use case involving MX-type precipitate-bearing steels for fusion energy systems. Specifically, we conducted in-situ TEM ion irradiations under fusion-relevant conditions to observe dynamic MX-precipitate dissolution and growth. The ML-based quantification results, with full temporal and spatial fidelity, are combined with post-irradiation characterization and benchmarked against ex-situ irradiations and theoretical models. This comprehensive approach provides a holistic understanding of radiation-induced microstructural evolution in complex alloys under extreme environments. This talk sets the stage for additional talks on advancing nuclear materials research through ML and experimental synergies. |
| Proceedings Inclusion? |
Planned: |