Technological high strain deformation of ‘wavy glide’ metals and LEDS

Abstract

The evolution of dislocation structures in technological high strain deformation of ‘wavy glide’ metals is studied by the examples of pure aluminium, nickel, copper and hot deformed stainless steel, and compared with the earlier observations by Langford and Cohen on drawn iron wire and Wert et al. on rolled aluminium alloys including precipitates. The various dislocation cell structures in all of these (in contrast to ‘planar glide’ metals which deform via planar arrays and/or Taylor lattices) develop in agreement with the LEDS theory to the effect of practically proving its basic tenet, the LEDS principle. Namely, dislocation rotation boundaries substantially obeying Frank's formula, i.e. LEDS's, are the dominant feature throughout. The stronger, longer and sharper of these delineate volume elements, dubbed ‘cell blocks’ (CBs). The number of active slip systems in the CBs always falls short of the five required by the Taylor criterion, and their selection and/or relative intensity of operation differs from one to the next. The CBs in turn are subdivided into ordinary dislocation cells. In the early stages of rolling, believed to represent workhardening stage III, the structure is of large plate-like arrangements of roughly equiaxed cell blocks subdivided into equiaxed dislocation cells. These plates are inclined about ±(40° ± 10°) to the rolling plane. With straining, the rotation angles increase and the scale of the structure shrinks but such that the CB size shrinks faster than the cells, eventually down to a bamboo structure. This evolution proves the dislocations forming the structure to be quite mobile, in agreement with the LEDS theory. Next, in what is believed to be the onset of stage IV, intersecting slip distorts the roughly parallel CB boundaries into S-shapes (S-bands) and thereby rotates them towards parallelity with the rolling plane into finely spaced lamellar walls subdivided by a bamboo structure of cell walls. Finally, roughly equiaxed subgrains form, no longer subdivided by dislocation cells, considered to mark stage V. Rotation angle/axis determinations reveal overlapping rotation angle distributions of CB and cell walls which extend to much larger than expected angles, and that the structure is formed of dislocations from the active slip systems. Throughout, the structure evolves in accord with the LEDS hypothesis, in that the energy per unit length of dislocation line decreases successively. Although the structures could not have been predicted, they can serve for future modelling.