Texture Development During Cold Rolling of a β-Ti Alloy: Experiments and Simulations

Microstructure evolution and texture development during cold rolling of a Ti15333 alloy were systematically investigated in the present work. Texture was simulated using mean-field [Visco-Plastic Self-Consistent (VPSC) and Taylor] models. Evolution of crystallographic texture was also simulated using the Visco-Plastic Fast Fourier Transform (VPFFT) model. The as-received samples (in the hot-forged and hot-rolled condition) were cold rolled unidirectionally up to 20, 40, 60 and 80 pct thickness reductions. Increase in the cold-rolling reduction resulted in changes in the crystallographic texture as well as grain morphology. The initial hot-rolled sample consisted of in-grain shear bands that were aligned approximately ± 35 to 40 ° with respect to the sample rolling direction. Shear band density gradually increased with the increase in cold-rolling reduction, and these bands usually represent narrow zones of intense strain. α (RD//〈110〉) and γ (ND//〈111〉) fibers were observed in all the cold-rolled samples. The volume fraction of both these fibers was found to be highest for the 80 pct deformed sample. For mean-field simulations, the normalized difference of the texture index (normalized TI_diff) was found to be a good criterion to represent the match between the simulated and experimental texture. The affine model (VPSC) was found to give a good match with the experimental texture compared to the Taylor models. The γ-fiber and α-fiber were always overestimated in mean-field VPSC simulations. Extensive shear band formation could be the possible reason for mismatch between the simulated and experimental texture. For VPFFT simulations, the general texture evolution involved the intensification of the γ-fiber and α-fiber texture. Simulated texture was reasonably well predicted quantitatively with VPFFT, analyzed based on the volume fraction of the different texture fibers/components.