Analysis and modeling of thermal behavior and grain morphology in laser powder bed fusion of Ti-Cu alloys

The microstructure and service properties of laser powder bed fusion components are critically determined by complex thermal cycles. This study presents a computational thermal-modeling framework to predict grain evolution in Ti-Cu alloys during laser powder bed fusion. The framework incorporates key effects, including solute redistribution, Marangoni convection, and in-situ thermal cycling, within a defined semi-infinite domain accounting for conductive heat dissipation and the solute boundary layer. The transient temperature field and thermal history were calculated using a moving point heat source model. Nucleation undercooling of Ti-Cu alloys was evaluated via a free-growth model, and the initial partition coefficient was refined by incorporating the Marangoni effect. The initial grain size was predicted based on the interdependence theory, and the final grain size evolution during thermal cycling was governed by the Johnson-Mehl-Avrami-Kolmogorov model coupled with a refinement function. Results indicated that the initial grain size exhibits a distinct U-shaped distribution across the molten pool width, while the final grain size shows a peak trend along the scan path after thermal cycling. Sensitivity analysis showed that the solidification rate, impurity diffusion coefficient, and nucleation undercooling are the dominant factors influencing the grain size. This work provides an efficient numerical tool for understanding and predicting the thermal-process-microstructure relationship in metal additive manufacturing, facilitating thermal-related process optimization.