Computational assessment of material structure weaknesses in polycrystalline materials

Abstract

By incorporating local grain orientation, grain geometry and macroscopic elastic properties, a numerical procedure has been developed for computational prediction of mesoscopic stress and strain distributions in simulated polycrystalline material samples. The numerical procedure is developed on the basis of the concept of grain-average fields, Kröner-Kneer model, Waldvogel-Rodin algorithm and a self-adaptive method. Repeated computer tests were performed to investigate mesoscopic stress variation in the samples, and find coherent interrelations of material structure weaknesses (MSWs) with local microstructure of the samples. It was found that the stronger the single crystal elastic anisotropy, the stronger the inhomogeneity of mesoscopic stress distribution. Not only the elastic anisotropy, but also the grain geometry, may produce significant local stress disturbances. It has been found that the defined 'orientation-geometry factor' and 'correlation parameter' are two adequate physical quantities which account for synergetic interactions due to grain-orientation geometry-induced anisotropy. By using the two quantities, MSWs can be well correlated with local microstructure. Computer tests also show that 250-400 conjoining grains are necessary to homogenize the mesoscopic stress distribution in the considered materials.