About this Abstract |
Meeting |
2023 TMS Annual Meeting & Exhibition
|
Symposium
|
Local Ordering in Materials and Its Impacts on Mechanical Behaviors, Radiation Damage, and Corrosion
|
Presentation Title |
Neural-network Based Atomistic Simulation on Chemical Order Formation Kinetics in Medium Entropy Alloys |
Author(s) |
Shigenobu Ogata, Jun-ping Du |
On-Site Speaker (Planned) |
Shigenobu Ogata |
Abstract Scope |
We studied the chemical ordering in CrCoNi medium-entropy-alloys (MEA) using a large-scale atomic model with a neural network potential (NNP) trained by density functional theory and molecular dynamics/Monte-Carlo (MD/MC) hybrid simulated annealing method. The MD/MC simulations from high to low temperatures show a chemical ordering transition at 800 K from the chemical short-range order to the superlattice ordering consisting of Cr(110)/CoNi and Cr(100)/CoNi nano-sized superlattice domains. In addition, the kinetics of chemical ordering evolution through vacancy diffusion is studied using the NNP with kinetic Monte-Carlo (kMC) simulation method combined with a novel neural network technique of fast activation energy evaluation. The kMC simulations at various temperatures lead to a diagram of Time-Temperature-Chemical ordering (TTC) relation, which can be used to tune the degrees of chemical ordering by controlling annealing temperature and time. |
Proceedings Inclusion? |
Planned: |
Keywords |
High-Entropy Alloys, Computational Materials Science & Engineering, Modeling and Simulation |