|M.Sc Student||Lev Efrat|
|Subject||Void Growth in Heavily Defomed Metals - The Effects of|
|Department||Department of Mechanical Engineering||Supervisor||Dr. Shmuel Osovski|
|Full Thesis text|
A large number of experiments in the micron-scale exhibit size dependency leading to drastic changes in the mechanical behavior of the material with decreasing size. In order to explain this phenomenon, strain gradient theories were developed. Out of the many non-local theories which were developed in the last two decades, the mechanism-based strain gradient plasticity theory (MSG) by Gao et al. is appealing in its simplicity and ease of relating it to easily measured material properties .The MSG theory adopts a Taylor model formulation and connects between the strain gradients and dislocation interactions. In this work, a versatile user subroutine that allows for any dislocation- based material model to be incorporated in the strain gradient framework was implemented into the commercial FE software ABAQUS. A well-established dislocation density based model was used in conjunction with the MSG theory to account for the mechanical behavior of heavily deformed Cu.
To examine the effects of strain gradients, a study case with two stages was constructed. The first stage is a simulation with pure shear boundary conditions, allowing the strain gradients to develop with the dislocation cell. The second stage is a void cell simulation, analyzed in the aspects of void growth, localization and coalescence, as well as the roles of the severe plastic deformation and the strain gradients.
It was found that the strain gradients influence is greater when the evolution rate of the dislocation cell is decreasing. The void growth was affected by the strain gradients, and the behavior of the model material was found to be considerably different after being subjected to various level of deformation prior to the introduction of the void. The void growth and shape were found to be depended mostly on the stress state, and different growth mechanisms were detected.