Solidification in a Supercomputer: From Crystal Nuclei to Dendrite Assemblages

Abstract

Thanks to the recent progress in high-performance computational environments, the range of applications of computational metallurgy is expanding rapidly. In this paper, cutting-edge simulations of solidification from atomic to microstructural levels performed on a graphics processing unit (GPU) architecture are introduced with a brief introduction to advances in computational studies on solidification. In particular, million-atom molecular dynamics simulations captured the spontaneous evolution of anisotropy in a solid nucleus in an undercooled melt and homogeneous nucleation without any inducing factor, which is followed by grain growth. At the microstructural level, the quantitative phase-field model has been gaining importance as a powerful tool for predicting solidification microstructures. In this paper, the convergence behavior of simulation results obtained with this model is discussed, in detail. Such convergence ensures the reliability of results of phase-field simulations. Using the quantitative phase-field model, the competitive growth of dendrite assemblages during the directional solidification of a binary alloy bicrystal at the millimeter scale is examined by performing two-and threedimensional large-scale simulations by multi-GPU computation on the supercomputer, TSUBAME2.5. This cutting-edge approach using a GPU supercomputer is opening a new phase in computational metallurgy.

Variational formulation and numerical accuracy of a quantitative phase-field model for binary alloy solidification with two-sided diffusion

“…(4.10) in which q ⊥ (φ) is replaced by q(φ). behavior [25,30]. Accurate results for the directional solidification of a columnar dendrite and free dendritic growth in the steady state can be obtained with this model as long as W is smaller than the dendrite tip radius divided by a constant of order unity.…”

“…The validity of this model was investigated in detail by carrying out convergence tests of the simulation results with respect to W. This model exhibits excellent numerical performance [25][26][27][28]. In particular, it has recently been shown that an accurate result (i.e., a well-converged result) for steady-state dendritic growth can be obtained with this model as long as W is smaller than the curvature radius of the dendrite tip divided by a constant of order unity [30].…”

Phase-field numerical simulation of three-dimensional competitive growth of dendrites in a binary alloy

“…Simulation studies including 2-D [9][10][11][12][13][14][15][16][17][18] and 3-D dendritic growth [2,[19][20][21][22][23][24][25][26][27] , dendritic growth of pure substances [9][10][11][19][20] , dendritic growth in binary alloys [1,[12][13][14][15][21][22][23][24][25] , and dendritic growth in multiple alloys [2,[16][17][18][26][27] have been reported. Multiple dendrites simultaneously grow during the casting process and their growth process can be directly simulated by using PFM [1,[28][29][30][31] . Mutual interactions and competitions exist among grains in the simultaneous growth of multiple grains.…”

“…As an open model, the phase-field model can be coupled with various external fields such as temperature, dissolution, gravity, and fluid fields, and can effectively combine the microscale with the macroscale. With the development of the phase-field model and improvements in computing technology, quantitatively simulating and predicting the microstructure in the phase transition of actual materials by using PFM has become the focus of research into this aspect of material structures [1][2][3][4][5][6][7][8] .…”

“…Phase-field numerical simulation of threedimensional competitive growth of dendrites in a binary alloy *Li Feng 1, 2 , Ya-long Gao 1 , Chang-sheng Zhu 2,3 , Guo-sheng An 1 , Xin Deng 3 , and Bei-bei Jia 1 …”