Physics-based analytical modeling of materials properties in metal additive manufacturing

Abstract

Emerging additive manufacturing (AM) offers a sustainable alternative to the subtractive processes with significant potential for complex geometries and material efficiency. However, predicting and controlling the microstructure-dependent properties of AM parts, particularly metals, remains challenging due to complex multi-physical processes. This work develops a physics-based analytical modeling framework to predict the evolution of key microstructural features (texture and grain size) and their influence on material properties (elastic modulus, Poisson’s ratio, yield strength) in laser powder bed fusion (LPBF) of Ti-6Al-4V. The framework integrates: (1) a 3D thermal profile model with boundary heat transfer for a moving point heat source; (2) Johnson–Mehl–Avrami–Kolmogorov (JMAK) kinetics and Green’s function-based thermal stress analysis for grain size prediction during heating and cooling; (3) columnar-to-equiaxed transition (CET) criteria and Bunge calculation for multi-phase texture evolution; (4) a self-consistent model to predict texture-affected anisotropic elastic modulus and Poisson’s ratio; and (5) the Hall–Petch relation for grain size-dependent yield strength. Experimental validations confirm the fidelity of the thermal model (molten pool dimensions), texture simulation (pole figure intensities), and predicted properties. Crucially, the simulated effective elastic modulus (109–117 GPa) and yield strength (850–900 MPa) under consistent processing parameters align well with experimental ranges (100–140 GPa and 850–1050 MPa, respectively) and show stability regardless of layer or row settings. The Poisson’s ratio exhibits significant anisotropy (approx. 0.45–0.5 in X/Y vs. lower values in other directions). By bridging processing parameters, microstructure evolution, and final properties, this work provides a paradigm for computationally efficient prediction and optimization of AM material performance, paving the way for inverse design strategies.