Optimizing the rotation conditions for grain refinement in equal-channel angular pressing

Equal-channel angular (ECA) pressing is rapidly becoming an established procedure for inducing microstructural refinement in polycrystalline materials through the process of intense plastic straining. To date, there have been numerous reports of the use of the ECA pressing procedure to produce grain refinement down to the submicrometer, or occasionally the nanometer, level in a range of metallic materials[1–10] and in some intermetallic compounds.[11,12] In ECA pressing, a sample is pressed through a die having two channels, equal in cross section, which intersect at an angle of F, with simple shear occurring as the sample moves through the angle subtended at the intersection of the two channels.[13,14] Since the cross section of the sample remains unchanged on passage through the die, the procedure is readily amenable to repetitive pressings of the same sample in order to introduce high total strains. Pressings are generally conducted using dies having F . 90 deg, and under these conditions the equivalent strain introduced on each passage through the die is approximately equal to 1.[15] In practice, experiments have demonstrated that the microstructural characteristics introduced by ECA pressing, including the structural homogeneity, the average shapes of the individual grains, and the misorientation angles of the boundaries between adjacent grains, are all dependent upon experimental conditions such as the number of passages through the die,[5,9,10,16] the shearing directions in each separate passage as manifested by any rotation of the sample between separate pressings,[5,9,16] and the temperature at which the straining is conducted.[10] When repetitive pressings are undertaken, it is a standard procedure to identify three different routes, designated A, B, and C, in which the pressings are conducted without any rotation of the sample, with rotation about the longitudinal axis by 90 deg between each pressing and with rotation by 180 deg between each pressing, respectively. Two recent reports have compared directly the efficiency of these different processing routes for establishing a homogeneous structure consisting of an array of equiaxed grains. In experiments by Ferrasse et al.[9] on Cu and Al alloys, it was concluded that processing by route C produces a more ho