A NEW MODEL TO PREDICT GRAIN NUCLEATION DURING DYNAMIC RECRYSTALLIZATION

Deformation in metals and alloys is accompanied by high hardening rates and high dislocation content. Static annealing of deformed metals or forming operations at elevated temperatures can lead to static and dynamic recrystallization (SRX and DRX). The resultant texture and hence the properties of such a material are determined by the nucleation and growth of recrystallized grains conditioned by the deformation. Accurate prediction of nucleation in the deformed state is essential to predict final microstructure and texture. This study presents a new approach to determining nucleation of recrystallized grains based on local gradient in the dislocation density tensor in the deformed material. The Nye tensor, which measures local lattice curvature and is an indicator of the stored energy, is calculated using a crystal plasticity finite element code that accounts for deformation mechanisms as well as texture and its evolution. Nucleation is postulated to occur in regions of the microstructure that have high (locally) stored energy adjacent to a low Nye tensor (relatively undeformed) zone. Texture resulting from different starting microstructure are simulated and compared to identify the importance of initial texture on the final DRX microstructure in an aluminum alloy and the modifications to the approach to study recrystallizing of Mg alloys are identified.