| About this Abstract |
| Meeting |
2022 TMS Annual Meeting & Exhibition
|
| Symposium
|
Magnesium Technology
|
| Presentation Title |
Unraveling Mg <c+a> Slip Using Neural Network Potentials |
| Author(s) |
Christopher D. Barrett, Mashroor Nitol, Doyl Dickel |
| On-Site Speaker (Planned) |
Christopher D. Barrett |
| Abstract Scope |
Magnesium (Mg) activates <c+a> dislocation slip on the second order pyramidal slip plane. Historically, under c-axis compression there has been a discrepancy in the preferred pyramidal slip plane as measured by experiments versus atomistic simulations. Here we compare atomistic simulation results using several interatomic potentials including a novel artificial neural network (RANN) potential with the aim of determining whether this discrepancy arises from interatomic potential inaccuracies. The new potential shows better agreement with density functional theory and experimental calculations than previous interatomic potentials for Mg. The new results are confirmed by another independent neural network potential. We demonstrate that the basal dissociated <c+a> core is glissile at low stress, completely contrary to previously thought, and that constant stress molecular dynamics demonstrate clear preference for the 2nd order pyramidal system over the 1st order system. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Machine Learning, Magnesium, ICME |