TY - JOUR
T1 - Three-dimensional phase field simulation of the effect of anisotropy in grain-boundary mobility on growth kinetics and morphology of grain structure
AU - Suwa, Yoshihiro
AU - Saito, Yoshiyuki
AU - Onodera, Hidehiro
PY - 2007/7
Y1 - 2007/7
N2 - The kinetics and topology of grain growth in three dimensions have been simulated using a phase-field model with anisotropic grain-boundary mobilities. In order to perform large scale calculations we have applied both modifications of algorithms and parallel coding techniques to Fan and Chen's phase-field algorithm. The kinetics of abnormal grain growth together with isotropic grain growth is presented. It is observed that the grains of a minor component which are at the beginning surrounded preferentially by boundaries with high mobility grow faster than the grains of a major component until the texture reverses completely. In our simulation program, data arrays for storing orientation field variables are divided into each message passing interface (MPI) domain, and only the information about the cluster enumeration is shared with whole system. This enables us to perform large scale calculations such as 3203 grid points. In order to follow up the time-dependent growth exponent and distribution functions in a material with a strong texture, a large scale calculation is indispensable.
AB - The kinetics and topology of grain growth in three dimensions have been simulated using a phase-field model with anisotropic grain-boundary mobilities. In order to perform large scale calculations we have applied both modifications of algorithms and parallel coding techniques to Fan and Chen's phase-field algorithm. The kinetics of abnormal grain growth together with isotropic grain growth is presented. It is observed that the grains of a minor component which are at the beginning surrounded preferentially by boundaries with high mobility grow faster than the grains of a major component until the texture reverses completely. In our simulation program, data arrays for storing orientation field variables are divided into each message passing interface (MPI) domain, and only the information about the cluster enumeration is shared with whole system. This enables us to perform large scale calculations such as 3203 grid points. In order to follow up the time-dependent growth exponent and distribution functions in a material with a strong texture, a large scale calculation is indispensable.
KW - Anisotropic grain boundary mobilities
KW - Grain growth
KW - Microstructure
KW - Parallel computation
KW - Phase field
UR - http://www.scopus.com/inward/record.url?scp=34248182548&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=34248182548&partnerID=8YFLogxK
U2 - 10.1016/j.commatsci.2006.10.025
DO - 10.1016/j.commatsci.2006.10.025
M3 - Article
AN - SCOPUS:34248182548
SN - 0927-0256
VL - 40
SP - 40
EP - 50
JO - Computational Materials Science
JF - Computational Materials Science
IS - 1
ER -