Predicting anisotropic behavior of textured PBF-LB materials via microstructural modeling

Abstract It is well established that large temperature gradients cause strong textures in as-built metal parts manufactured via laser beam powder bed fusion. Columnar grains with a preferred crystallographic orientation dominate the microstructure of such materials resulting in a pronounced anisotropic mechanical behavior. Such materials are often studied with the help of tensile tests and corresponding numerical simulations in different loading directions. For the purpose of simulations, the microstructure is usually modeled with a statistically representative volume element (RVE). In the present study, two RVE modeling techniques, based on different texture sampling algorithms, have been compared for their property prediction capabilities. It was found that the model, based on an equally weighted crystallographic orientations set, sufficiently predicted macroscopic mechanical properties and also reduced the computational cost. Furthermore, an efficient method to rotate the boundary conditions for tensile test simulations under different loading directions was developed, thereby reducing the required number of RVE models to just one. The method was compared with an alternate method, where, an RVE model with rotated microstructure was subjected to unchanged boundary conditions. For this study, tensile test simulation results were compared with data from destructive material tests for predominantly single-phase austenitic stainless steel (EN 1.4404/AISI 316L).