Channel die tests on Al and Cu polycrystals: Study of the prestrain history effects on further large strain texture

Abstract

Al and Cu polycrystals have been prestained in a channel die to a compressive strain of 0.04-0.08 (small prestrain) and 0.40-0.55 (large prestrain) and then secondarily strained in different orientations—still in the channel die—up to a second strain of 0.75 in compression. The polycrystal hardening anisotropy is measured at small and large strain and reported under a form similar to the latent hardening ratio curves for single crystals, say here the ratio of the initial (back extrapolated) yield stress for the secondary straining to the maximum stress at the end of the prestrain. Texture measurements have been performed at the end of the secondary straining for the copper samples and are reported under the form of {111}, {200} and {110} pole figures. The observed differences on the measured textures are discussed with regard to the prestrain orientation and amplitude with special attention to the associated polycrystal hardening anisotropy and its more probable causes, geometrical or microstructural. The main observations are: (1) for copper, there is a clear hardening anisotropy after small strains, which does not seem due to texture because of too mild associated rotations, but more reasonably to an intracrystalline “latent hardening” effect in polycrystals. This effect appears greater in Cu than in Al in agreement with theoretical expectations (from the stacking fault energy of the two materials) and experiments on single crystals confirming the above interpretation. This hardening anisotropy depends on the way the load path is changed and differs for the various secondary orientations; (2) a weak hardening anisotropy occurs after large strains where one can expect more effect of the associated texture, which here appears to act in the opposite direction of that due to the latent hardening; (3) the correlation between changes on deformation textures at large strain and path changes during the sample loading appears as follows: while a path change after a large prestrain has a strong effect on the texture pattern after a secondary strain of 0.75 both on intensity and location of the maxima, it has only a mild effect after a weak prestrain but still showing noticeable differences for some orientations on the intensity maxima, which are the orientations associated with the highest measured hardening values at the end of the prestrain.