A hybrid computational approach for modeling thermal spray deposition

Thermal spray coatings are usually deposited by first heating the feedstock (powder) material into a series of molten or semi-molten droplets before being applied to the substrate surface. As a result of large deformation, complex interaction and material mismatch occurring during the thermal spray process, residual stresses are induced. Residual stress is one of the main contributing factors that determine the constitutive behavior and lifetime of coatings. In the present study, a new computational approach for effective prediction of residual stress evolution in thermal spray coatings has been proposed. The proposed approach combines point cloud (PC) and finite elements (FE) to model the spray process and associated residual stresses. Droplet deposition (or impact) modeling and associated deformation are modeled on PC using smooth particle hydrodynamics (SPH)-a meshless approach for modeling of violent fluid flows. The conversion of PC to FE mesh of coating splat is done using several recent algorithms for point cloud processing. Using the numerically-generated FE mesh, finite element analysis is conducted for effective prediction of the evolution of temperature and residual stresses during the process. The proposed approach is first demonstrated with a single 30 gm yttria-stabilized zirconia (YSZ) droplet that is deposited on steel alloy at 60 m/s and 2535 degrees C. Then, it is further shown that the approach can be applied to a realistic case involving multiple droplets deposition and their interactions. Both deposition and post-deposition stresses are integrated to get the final residual field. Deposition (quenching) stresses are predicted to be low and tensile. While, post-deposition (mismatch) stresses are predicted to be high and compressive. The compressive residual stress field predicted for the YSZ layer is validated by comparing with experimental and analytical results available in the literature.