Developing a new method for calculating the nuclei of orientations during aluminum alloys recrystallization

A new approach to modeling texture formation in aluminum alloys du ring their recrystallization is proposed. It is based on a previously developed method for calculating texture formation during deformation, which involves solving the problem of deformation of a solid at the macro level by the finite element method. In each layer with a homogeneous stress-strain state, a domain is defined, which is divided into finite elements. Each element corresponds to a crystallite having a specific crystallographic orientation. Next the domain deformation problem is solved based on the boundary conditions obtained from the solution at the macro level. A new approach to the nuclei calculation during recrystallization involves the use of finite elements with a high approximation order or the division of one crystallite into several finite elements. This approach makes it possible to find areas with deformation heterogeneity and thereby more accurately determine the number and type of nuclei of the recrystallized structure. The model is able to predict the number of nuclei formed at the boundaries of new grains, shear and cubic bands, as well as in areas with crystal lattice distortions due to the presence of intermetallides in the structure. To verify the performance of the model, we solved the problem of texture formation during self-annealing in a roll of AA8011 alloy after its hot rolling in a continuous train. Alloy AA8011 refers to deformable alloys of the Al – Fe system, which are likely to forming a sharp cube texture, therefore, the correct calculation of the nuclei of this orientation for alloy AA8011 is relevant. The results showed that, without taking into account the inhomogeneity of the deformation, the model predicts a smaller number of cubic textures and a larger number of β-fiber textures compared to experimental data. Taking into account the inhomogeneity of the deformation makes it possible to more accurately predict the proportion of cube and β-fiber textures. At the same time, to calculate the amount of Goss texture, the proposed method requires further development.The research was carried out at the expense of a grant from the Russian Science Foundation, project 24-19-00590, https://rscf.ru/project/24-19-00590/.