Modelling of grain structure using cellular automata-finite element method for additive manufacturing of metals

Abstract

The microstructure of the metal is formed in process when fabricated using additive manufacturing (AM). The microstructure determines the mechanical properties of the metal and being able to predict what microstructure would form would be beneficial. This research uses simulations to predict the grain structure. Powder bed fusion of metal using laser (PBF-L/M), often called selective laser melting (SLM), was simulated. Two models were developed, a finite element method (FEM) thermal model that translated the process parameter to temperature profiles of the moving melt pool and a cellular automata microstructure model that uses the temperature profile from the FEM model and simulate the grain growth resulting in a grain structure. Two materials, stainless steel 316L (SS 316L) and Ti34Nb, were simulated and validated with experiments. Simulated melt pool dimensions of both materials were close to experiment results. The grain structure of the SS 316L showed similar patterns found in experiments. The grain width of the Ti34Nb had some differences with experimental results likely due to the unmelted niobium causing nucleation sites that is not accounted for in the model. Simulation of both materials would require simulating multiple layers to properly relate to the electron backscatter diffraction (EBSD) results.