Analysis of Large Strain Hot Torsion Textures Associated With “Continuous” Dynamic Recrystallization

[+] Author and Article Information
S. M. Lim1

Centre SMS, CNRS UMR 5146, Ecole Nationale Supérieure des Mines, 158 Cours Fauriel, 42023 Saint-Etienne Cedex 2, Francelim@emse.fr

C. Desrayaud, F. Montheillet

Centre SMS, CNRS UMR 5146, Ecole Nationale Supérieure des Mines, 158 Cours Fauriel, 42023 Saint-Etienne Cedex 2, France

Out of simplicity, the basic version without grain interactions was used in the present study.

The final strain, ε¯=1.1, is low compared with strains that are typical of the steady-state regime of torsion tests. One possible reason for this could be the fact that the standard Taylor model often overestimates texture evolution (i.e., too rapid and too strong).


Corresponding author.

J. Eng. Mater. Technol 131(1), 011103 (Dec 18, 2008) (8 pages) doi:10.1115/1.3030939 History: Received February 01, 2008; Revised July 11, 2008; Published December 18, 2008

The development of ideal orientations within the steady-state region of hot torsion flow curves of fcc and bcc metals undergoing “continuous” dynamic recrystallization is analyzed. It is well known that in fcc metals, e.g., Al deformed at 400°C and above, the experimentally observed end texture consists of the twin-symmetric B(112¯)[11¯0]/B¯(1¯1¯2)[1¯10] component, whereby the (hkl)[uvw] indices correspond to the shear plane z and the shear direction θ, respectively. In bcc iron however, only one of the self-symmetric D1(112¯)[111] and D2(1¯1¯2)[111] components dominates (the former in the case of positive shear or clockwise rotation about the r-axis, and the latter during negative shear). The tendency toward a single end orientation imposes certain limitations on grain refinement, as this would ultimately imply the coalescence of subgrains of or close to this orientation, and therefore the disappearance of existing high angle boundaries (15deg). It is believed that the preference of D1 over D2, or vice versa, could be related to phenomena other than glide-induced rotations, e.g., grain boundary migration resulting from differences in work hardening rates. In this paper, the standard Taylor model is first used to predict the texture evolution in simple shear under the full-constraint rate-sensitive scheme. This is then coupled with an approach that takes into account grain boundary migration resulting from differences in dislocation densities within grains of varying orientations. The preliminary results are in agreement with experimental findings, i.e., grains with initial orientations close to D2 grow at the expense of neighboring grains during negative shear and vice versa.

Copyright © 2009 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.



Grahic Jump Location
Figure 1

(a) (111) pole figure of an AA6060 Al alloy that has experienced torsion (positive shear) at 400°C and 0.1 s−1 until a strain of 20 (1), alongside an illustration of the (r,θ,z) coordinate system in a torsion sample; and (b) (111) pole figure representation of ideal orientations observed during torsion tests in fcc metals (2)

Grahic Jump Location
Figure 2

(a) {110} pole figure of high-purity iron containing 60 ppm C after torsion (negative shear) at 450 °C and 0.01 s−1 until a strain of 40 (3); and (b) locations of typical torsion components observed in bcc metals on the {110} pole figure (4). The r-axis is located at the center of these pole figures and points out of the paper.

Grahic Jump Location
Figure 3

Definition of crystalline orientations that respect the ⟨110⟩∥r symmetry using the angle ω, where 0≤ω≤π. For example, ω=0 and ω=π are associated with the D1(112¯)[111] and D2(1¯1¯2)[111] orientations, respectively. The r-axis is located at the center of these pole figures and points out of the paper.

Grahic Jump Location
Figure 4

Simulation results involving 30 grains within the ⟨110⟩∥r fiber subjected to negative shear with MEul=1×10−3 μm3/s. Variation of (a) grain diameter, and (b) dislocation density with the number of increments (500 increments=equivalent strain of 1).

Grahic Jump Location
Figure 5

Taylor factor distribution (5) of grains within the ⟨110⟩∥r fiber prior to straining. Initially, only grains 1–30 were considered, but grains 31 and 32 were subsequently added. Grains 18 and 31 begin with the same Taylor factor, the difference being that the former tends toward D2, while the latter approaches D1 in negative shear. Likewise for grains 1 and 32, but in positive shear.

Grahic Jump Location
Figure 6

Variation of (a) diameter, and (b) dislocation density of grains 18 and 31 that belong to the ⟨110⟩∥r fiber (32 grains in all). The symbol n represents the number of increments (500 increments=equivalent strain of 1). Negative shear was simulated with MEul=1×10−3 μm3/s.

Grahic Jump Location
Figure 7

Variation of dislocation density for grains within the ⟨110⟩∥r fiber (32 grains in all) in the absence of grain boundary mobility (M=0). (a) Low values of n; (b) enlarged portion of the graph shown in (a) for grains 3 and 18; and (c) graph at high values of n, around the time when grain 3 disappears (n=263).




Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In