Modeling of Subgrain Growth Kinetics: 3D Monte-Carlo Simulation

  • PDF / 276,838 Bytes
  • 7 Pages / 612 x 792 pts (letter) Page_size
  • 47 Downloads / 277 Views

DOWNLOAD

REPORT


1225-HH02-04

Modeling of Subgrain Growth Kinetics: 3D Monte-Carlo Simulation Tomoaki Suzudo1, Hideo Kaburaki1, Mitsuhiro Itakura2 Center for Computational Science and e-Systems, Japan Atomic Energy Agency, Tokai-mura, 319-1195, Japan 2 Center for Computational Science and e-Systems, Japan Atomic Energy Agency, Taito-ku, 110-0015, Japan 1

ABSTRACT We used a three-dimensional Monte Carlo method to investigate subgrain growth with different initial values for average grain-boundary misorientation, and found that abnormal grain growth emerges for relatively large average misorientation, with remaining cases revealing an exponential kinetics of subgrain growth. We also found that the growth exponent was ~4.4, and that it was virtually independent of the average misorientation. Self-similarity of the misorientation distribution was observed during growth.

INTRODUCTION Cold-worked metal undergoes microstructural changes during the recovery stage, that can be classified into several sub-stages: dislocation tangling, cell formation, annihilation of dislocations within the cells, subgrain formation, and subgrain growth [1]. Subgrains are small grains nucleated inside the original grains during the recovery stage, and misorientation angles between two subgrains are typically less than 10°. Subgrain growth is the final stage of the recovery process and significantly affects the mechanical property of metals because it sometimes causes recrystallization that completely alters the microstructure. The prediction of its kinetics is thus important from the viewpoint of material design and processing. Many measurements of the change of subgrain size over time at annealing stages reveal a similar kinetics to the growth of ordinary grains, having the form, Dtn − D0n = Kt

(1)

where Dt is the subgrain size at time t, n a constant called a growth exponent and K a temperature dependent rate constant. In some cases the value of the exponent n in equation 1 is found to be 2, which is similar to that for ordinary grain growth following recrystallization [2, 3, 4, 5]. More recent experimental studies, however, give a larger n, such as n = 2.5-7 [6, 7]; the larger n means that the growth rate decreases more rapidly over time. There are also some numerical studies of subgrain growth or low-angle boundary kinetics. Humphreys [8] uses a two-dimensional model and shows that n > 2 can be given for low-angle boundary kinetics. His physical explanation of the high exponent is that grain boundaries with higher misorientation angle move faster and disappear earlier than those with low misorientation angles, thus causing the decrease in the average misorientation angle during growth θav that reduces the total driving force for the boundary motion. The numerical study by

Holm et al. [9] also shows, with their two-dimensional model, the decrease in θav over time before abnormal grain growth emerges. There is sufficient apparent evidence to suppose n>2 for subgrain growth. To the best of our knowledge, however, the above numerical studies are not ree