[SY-F4] Symposium F-4

[SY-F4] Showdown! Pitting computer simulations against time-resolved experimental measurements of grain growth in 3D

Invited

Mingyan Wang¹, Nan Wang², Jules Dake¹, Søren Schmidt³, ^○Carl Krill¹ (1.Institute of Micro and Nanomaterials, Ulm University, Germany, 2.Dept. of Physics, McGill University, Canada, 3.Dept. of Physics, Technical University of Denmark, Denmark)

Thanks to dramatic improvements in computational speed, it is now feasible to simulate the coarsening of 3D polycrystalline microstructures containing tens of thousands of grains. Indeed, such calculations have become so powerful that the simulation cells rival—and, in some cases, exceed—sample volumes that can be probed experimentally! But do the computational algorithms underlying these simulations properly capture the physics of microstructural evolution as it occurs in real materials? We have followed a two-pronged strategy to address this question, using (i) three-dimensional x-ray diffraction (3DXRD) microscopy to map the 3D network of grain boundaries in a polycrystalline specimen over the course of stepwise isothermal annealing treatments, and (ii) a phase field model to simulate 3D grain growth starting from the same initial configuration as in experiment. A grain-by-grain showdown between tactic (i) and tactic (ii) offers fresh insights into the phenomenon of grain growth—gleaned not only from discrepancies between measured and simulated size trajectories and grain shapes, but also from instances of agreement.