Unveiling keyhole-induced porosity and grain orientation in laser powder bed fusion (LPBF) of maraging 300 steel through coupled computational fluid dynamics and cellular automaton

Abstract

Laser Powder Bed Fusion (LPBF) is a prominent additive manufacturing technique for fabricating complex metal components, but its applications in high performance high pressure die casting (HPDC) molds made by maraging 300 steel are challenged by defect formation and microstructural inconsistency. In this study, a multiscale simulation framework is developed by integrating Computational Fluid Dynamics (CFD) and Cellular Automaton (CA) to investigate both molten pool dynamics and grain structure evolution under LPBF conditions alongside microstructure analysis. Unlike prior approaches, the proposed CFD–CA model explicitly couples pore formation mechanisms arising from keyhole oscillations with solidification-driven grain orientation in maraging 300 steel used for HPDC. The CFD model captures transient temperature fields and melt flow driven by recoil pressure and surface tension, which reveals the formation of keyhole induced pores due to oscillations at the gas liquid interface. CFD outputs are interpolated into a CA model, which predicts grain nucleation, growth and microstructure evolution based on local cooling rates and solidification dynamics. The CA simulations demonstrate that increased scan velocity refines grain size while lower scan velocity generates shallower gradients and slower cooling to promote deeper thermal penetration, yielding elongated columnar grains with laser grain angles decreased to 68°. The grain structure shows that the orientation of columnar grains aligns with the maximum thermal gradient and higher undercooling promotes nucleation. The higher scan velocity and laser power alter the local thermal gradient in a way that favors a more vertically oriented grain structure which increases laser grain angles. By establishing a predictive process-structure–property linkage, this work pioneers the design strategy for LPBF.