Analytical modeling of microstructure features in metal additive manufacturing

Appearing additive manufacturing (AM) offers a sustainable approach contrary to traditional methods, aligning with global decarbonization efforts. Understanding grain evolution in metal AM is crucial for controlling final structural material properties and predicting continuum-level properties at different length scales. Crystallographic orientations (texture) and grain size are essential in determining material properties. They influence anisotropy, void growth, and coalescence behaviors. However, although various approaches have investigated microstructure evolution, the relationship between grain size and texture during microstructure evolution has yet to be fully explored, primarily based on analytical methodology. Specifically, the primary aim of this research is to study the relationship between microstructure features that affect material properties. This can deepen the understanding of microstructure evolution during the AM process and improve aspects such as the selection of processing parameters, validation of microstructure accuracy, and component mechanical design. This work utilized a physics-based analytical model to predict the 3D temperature distribution with integrated considerations of heat transfer boundary conditions, a point-moving heat source solution, and heat conduction, convection, and radiation based on processing parameters. With the temperature profile, the grain size is predicted by Johnson-Mehl-Avrami-Kolmogorov (JMAK) and grain refinement models, and texture is constructed by Hunt's model, Bunge's calculation, and basic thermodynamics rules. Grain size's impact on texture is investigated. Then, the association between texture and grain size is examined using the simulated microstructural results. Fitting curves drawn exhibit negative logarithmic relationships between grain size and texture intensity. In the process, Ti-6Al-4V is used for analytical modeling. This work newly provides a direct, fast analytical approach to link microstructure features with a quantitative relationship that can be used for applications such as real-time prediction, processing optimizations in industries.