Investigation of microstructures produced by metal additive manufacturing using 3D cellular automata finite element modeling in 316L steels and IN625 superalloys
Cameron McElfresh, Jaime Marian
Abstract
Laser-powder bed fusion additive manufacturing (LPBF-AM) of metals is rapidly becoming one of the most important materials processing techniques for advanced structural materials applications . LPBF-AM processing involves a large parameter space that makes it costly to determine what variables control the microstructural and mechanical property outcomes of the manufactured materials. Here we present a computational simulation methodology based on a three dimensional cellular automaton for fast-throughput generation of synthetic polycrystalline microstructures, and a crystal plasticity model for evaluation of elasto-plastic properties. Our approach connects the processing stage with the mechanical testing stage, thus capturing the effect of key LPBF-AM variables such as the laser power, laser spot size, scan speed, and hatch width on the properties of the simulated microstructures. Our model leverages a conduction-based thermal model calibrated against experimental data to predict melt pool geometry and its impact on mechanical property outcomes. We find that, of the three processing degrees of freedom, scan, hatch, and build directions, the build direction emerges as the one with the heaviest impact on material microstructure and mechanical properties.