About this Abstract |
Meeting |
2022 TMS Annual Meeting & Exhibition
|
Symposium
|
Mechanical Behavior at the Nanoscale VI
|
Presentation Title |
Development of Neural Network Potential for MD Simulation and Evaluation of Mechanical Property |
Author(s) |
Takeru Miyagawa, Akio Yonezu, Kazuki Mori, Nobuhiko Kato |
On-Site Speaker (Planned) |
Takeru Miyagawa |
Abstract Scope |
Titanium nitride (TiN) has been used in various applications because of its excellent wear and corrosion resistance. TiN has a rock-salt type crystal structure, which has been studied extensively, but recently the existence of non-rock-salt type phases has been reported from first-principles calculations. However, the mechanical properties of these new phases have not been studied, and it is impossible to estimate those properties from first-principles calculations because of the scale problem.
In this study, we measured the mechanical properties of these new phases using the molecular dynamics (MD) simulation, which can handle larger scale models than first-principles calculations. Since the conventional many-body interatomic potentials for rock-salt TiN are not applicable to these new phases, in this study, the interatomic potentials applicable to the MD simulation were constructed using machine learning, and the mechanical properties of the new phases of TiN were measured. |
Proceedings Inclusion? |
Planned: |
Keywords |
Mechanical Properties, Computational Materials Science & Engineering, Titanium |