About this Abstract |
Meeting |
MS&T22: Materials Science & Technology
|
Symposium
|
Grain Boundaries, Interfaces, and Surfaces: Fundamental Structure-Property-Performance Relationships
|
Presentation Title |
Learning the Grain Boundary Solute Drag Hypersurface |
Author(s) |
Fadi Abdeljawad, Malek Alkayyali |
On-Site Speaker (Planned) |
Fadi Abdeljawad |
Abstract Scope |
Even minute amounts of dopants or impurities at grain boundaries (GB) result in profound changes to GB dynamics. GB segregation has been the subject of active research efforts; however, dynamic solute drag has not been systematically explored. The challenge here is that GB solute drag greatly depends on several thermodynamic properties of the alloy and kinetic processes, including GB solute diffusion, mobility, and migration velocity−solute drag is a hypersurface. Based on recently developed theoretical and machine learning models, we explore GB solute drag in regular solution alloys. The governing equation describing solute drag is solved numerically to generate data to train, test, and validate a neural network model, which is then used to establish the solute drag hypersurface. Our results reveal a plethora of solute drag trends that are used to explain experimental observations of sluggish grain coarsening in a wide range of alloys. |