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 |
Revealing the Impact of Hydrogen on Iron: Large-Scale Quantitative Atomistic Analysis with Highly Accurate and Transferrable Machine Learning Interatomic Potentials |
Author(s) |
Shigenobu Ogata |
On-Site Speaker (Planned) |
Shigenobu Ogata |
Abstract Scope |
Experimentally observing the behavior of hydrogen in materials directly is challenging, and this poses a significant barrier to elucidating the impact of hydrogen on materials. Atomic simulations have been employed to address this issue. The accuracy of atomic simulations depends on the choice of interatomic interactions used; particularly in iron – hydrogen system, there is a scarcity of reliable interatomic interaction models capable of adequately representing the interactions of hydrogen with grain boundaries, surfaces, and dislocations in iron, as well as hydrogen diffusion and the dynamics of hydrogen and defects. We have developed a high-efficiency, high-precision machine learning potential for the iron-hydrogen system to overcome this situation. In this presentation, I will introduce the results obtained from large-scale atomic simulations using this machine-learning potential, detailing the effects of hydrogen on defect formation and propagation during plastic deformation and elucidating the mechanisms of crack initiation within grains and at grain boundaries. |
Proceedings Inclusion? |
Planned: |
Keywords |
Modeling and Simulation, Mechanical Properties, |