Multiscale modeling of fracture in bcc metals
| PhD ceremony: | L. (Lei) Zhang |
| When: | December 09, 2024 |
| Start: | 14:30 |
| Supervisor: | prof. dr. ir. E. (Erik) van der Giessen |
| Co-supervisor: | F. (Francesco) Maresca, Prof |
| Where: | Academy building RUG / Student Information & Administration |
| Faculty: | Science and Engineering |
In his thesis, Lei Zhang explored fracture mechanisms in body-centered cubic (bcc) metals, focusing on iron, through multiscale modeling approaches. Bcc metals, while strong, often lack sufficient fracture toughness at low to moderate temperatures, limiting their utility in industrial applications. To understand the underlying causes of this behavior, Zhang investigated crack initiation, propagation, and the role of other defects through atomic and microscale analyses.
A key objective is to model how atomically sharp cracks propagate and how these processes influence fracture behavior. Traditional interatomic potentials fall short in accurately predicting fracture mechanics in bcc metals. This study develops machine learning interatomic potentials with expanded density functional theory (DFT) databases, significantly improving fracture predictions. Zhang also investigated the influence of loading conditions (loading rates and temperatures) on the fracture mechanisms.
Additionally, Zhang combined a discrete dislocation plasticity (DDP) framework with anisotropic elasticity and cohesive-zone modeling, to assess crack tip plasticity, dislocation emission, and the impact of microstructural interactions. Results show that dislocation emission at the crack tip can enhance fracture toughness by shielding the crack, reducing growth rates, and increasing resistance. Overall, the thesis highlights limitations in conventional models and demonstrates that machine learning-based multiscale modeling can more accurately capture fracture processes in bcc metals, providing pathways to design stronger, tougher materials for engineering applications.