Fundamental Investigation of Directed Energy Deposition: Understanding the Process Through In-situ Monitoring and Numerical Simulations
Directed Energy Deposition (DED), one of the additive manufacturing techniques, has seen rapid growth over the last decade in multiple industries, including aerospace, medical devices, and energy systems. Although showing great potential and having proven itself a promising manufacturing alternative, the complex physics and highly non-equilibrium processes involved in DED exhibit unique challenges in optimizing the design parameters and controlling the overall quality of the final parts. Often times the entire process is treated as a “black box”, in which the effort has been focused on correlating the final built properties with input parameters, without fully understand the governing physics. This dissertation aims to provide a deeper understanding of the fundamental physics at each stage during the DED process, with the aid of in situ process monitoring tools and numerical simulations. In the first study, particle flow behavior of a gas atomized metallic powder (Inconel 718) and a ceramic powder (TiC) during the coaxial nozzle injection was studied through both experimental and numerical analysis. A 3-dimensional COMSOL numerical model was formulated to simulate the particle trajectories under the influence of gas flow, particle-gas interactions, and particle-wall collisions. High-speed photography was used to examine and compare the powder flow behavior of the Inconel 718 and TiC particles. The results reveal distinct differences in particle velocity and spatial distribution between the Inconel 718 and TiC particles, due to the dissimilar particle sizes, morphologies, material densities, and particle collision behavior. Next, thermal behavior of coated particles during the interaction with the molten pool was studied by constructing heat and phase transfer models using COMSOL software. Transient temperature and phase fields were calculated for coated and uncoated stainless steel (SS316L) and Zn-Al particles under different particle size, coating thickness, molten pool temperature and coating material conditions. Particle residence time values, defined as how long the particle can exist before undergoing a certain phase change, were determined from the calculations. The results show large variations in residence time due to the external parameters, especially the thermal diffusivity of the coated materials. Recommendations regarding coating material selection for retaining the Zn-Al particle were made based on the calculation results. In the third study, macroscopic thermal histories for a cuboidal build were analyzed in a 3-dimensional heat transfer model. Hatch rotation angle of 0 and 90 degrees were compared. A more randomized thermal history was observed for the 90-degree hatch rotation angle, which may inhibit epitaxial grain growth due to the dissimilar heat flux from pre-deposited layers. Finally, the tensile deformation behavior and phase formation in two nickel-based superalloys, Inconel 718 and Haynes 282, plus TiC-reinforced metal matrix composites of these alloys, were characterized experimentally for DED as-deposited samples. Scheil solidification calculations were also performed to understand the role of Nb (in the Inconel 718) and excess Ti and C (from the partially dissolved TiC reinforcement particles) on the phase formation sequence during solidification. The results reveal that the excess Ti+C leads to Nb-rich carbide formation in the Inconel MMCs, which depletes the Nb in the matrix and thus limits the formation of Laves and ” phases. This leads to the Inconel MMC matrix having a similar composition as in the Nb-free Haynes MMC matrix, and yields similar tensile behavior in the two sets of bulk, as-deposited MMC samples. These studies, combined, provide insight into the importance of materials properties and phase transformations in comprehensively understanding the DED process to facilitate expanded application in the industry.