Investigation of Primary Process Parameters for Laser Powder Bed Fusion of High Temperature Alloys through Physical and Virtual Experimentation

Recent innovations in the aerospace and energy industries require materials capable of maintaining sufficient mechanical strength at high operating temperatures. A category of materials that possesses these capabilities are refractory metals. However, due to several factors, such as their very high strength at room temperature, they are difficult to shape through conventional metal removal manufacturing practices. Because of this, additive manufacturing methods, such as laser powder bed fusion are being explored for cost-effective fabrication of complex net-shape geometries from high temperature materials. In response to this demand, the impact of primary processing parameters for laser powder bed fusion for these materials were investigated using laboratory experiments, as well as physics-based modeling. These two approaches were believed to provide complimentary information and data that would enable the establishment and exploration of complex relationships not readily apparent through only physical experimentation. The two alloys examined in this study were a nickel-based alloy, IN718, and a niobium-based alloy, C103.
Physical experimentation was confined to the IN718 alloy due to unavailability of the refractory metal. The results of the laboratory experiments were examined utilizing statistical methods, such as analysis of variance and linear regression, to determine the influence of primary process parameters, such as laser power, scan speed, and powder size distribution, on critical melt pool geometries of melt pool width and depth. Virtual simulations were conducted with both materials using ANSYS Fluent to capture thermal and flow behaviors of the melt pool. The results from the physical experiments showed that the melt pool depth was more sensitive to the primary parameters comprising the energy density function, laser power and velocity, as well as mean powder size, compared to bead width. Virtual simulations also reflected this sensitivity with melt pool measurements. However, underlying phenomena like vaporization of material, recoil pressure, and creation of a free surface must be accounted for to accurately simulate bead width.