| About this Abstract |
| Meeting |
2025 TMS Annual Meeting & Exhibition
|
| Symposium
|
Microstructural Evolution and Material Properties Due to Manufacturing Processes: A Symposium in Honor of Anthony Rollett
|
| Presentation Title |
Understanding Twin Nucleation in Mg Alloys Through In Situ Synchroton Experiments and Machine Learning Models |
| Author(s) |
Duncan A. Greeley, Valentín Vassilev-Galindo, John E. Allison, Javier Llorca |
| On-Site Speaker (Planned) |
Javier Llorca |
| Abstract Scope |
Mechanical twinning is an important deformation mode in Mg alloys, which is triggered by the nucleation of a twin usually from a grain boundary. Although experimental evidence indicates that twins tend to nucleate in large grains suitably oriented to accommodate plastic deformation, twins also nucleate in small grains or in grains not suitably oriented for twinning. Hence, it is evident that other microstructural factors are crucial for twin nucleation.
In this work, far- and near-field high-energy X-ray diffraction microscopy (HEDM) was used to obtain a 3D twin characterization from an extruded Mg- 4wt.%Al during compression. The collected dataset, accounting for around 2000 grains characterized by more than 50 microstructural features, was used to train machine learning (ML) models for classifying if grains will or will not twin. The obtained ML models from HEDM data allowed us to unveil microstructural factors that control twin nucleation. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Machine Learning, Magnesium, |