Modelling of recrystallisation textures in aluminium alloys: I. Model set-up and integration

A modelling scheme is presented to predict recrystallisation textures of aluminium alloys. It is assumed that the formation of textures during recrystallisation is governed by both oriented nucleation and oriented growth. An essential step is the prediction of the nucleation textures. The approach presented is based on orientation-dependent statistical information on the deformation substructure, which is obtained from an interactive combination of the deformation texture model GIA (grain interaction) and the dislocation density-based work hardening model 3IVM (three internal variables model). Substructural quantities characteristic for each single grain in the simulation are the dislocation density, the probability of shear band formation, the strength of in-grain orientation gradients, and the number of active slip systems. The latter is connected to subgrain morphology to predict orientation-dependent recovery. The nucleation texture model ReNuc distinguishes three nucleation mechanisms: random nucleation at shear bands, grain boundary nucleation, and nucleation in transition bands. The relative contribution of the three nucleation mechanisms to the nucleation texture is predicted based on the frequency of their substructural features in the simulated deformed state. The simulated dislocation density of each deformed matrix grain is used as a driving force during the growth simulation with the statistical recrystallisation texture model StaRT. The model allows a quantitative prediction of recrystallisation texture for hot forming or annealing after cold forming.