Interfacial interaction and roughness parameters effects on the residual stresses in DCL-TBC system with different thickness distributions

Compared with the traditional Single-Ceramic-Layer Thermal Barrier Coating system (SCL-TBCs), the application of the Double-Ceramic-Layers Thermal Barrier Coating system (DCL-TBCs) can strengthen the thermal insulation performance and ameliorate the interfacial oxidation behavior. The residual stresses due to the mismatch of thermal expansion, are so important that they can lead to the spallation and delamination failure of the entire system. In present work, considering the evolution of failure positions in the DCL-TBCs, the effects of interfacial interaction and roughness parameters on the residual stresses were investigated under two different thickness distributions of the ceramic layer. Therein, the numerical models with the general periodic boundary have been developed. Results show that the morphology of the yttria partially stabilized zirconia (YSZ)/thermally grown oxide (TGO) interface can affect the stress distribution along the lanthanum zirconate (LZ)/YSZ interface in the TBCs with low thickness of YSZ, but this influence can be negligible if the YSZ is thick enough. The qualitative distribution of the residual stress along the LZ/YSZ interface is similar for various interfacial roughness parameters, and the maximum tensile stress is always presented at the peak of the LZ/YSZ interface. But for the YSZ/TGO interface, the increase in the roughness parameters can lead to the change of the maximum tensile stress position and even the transition of the stress state in the local region. Also, as the reference for judging the maximum tensile stress location, the critical wavelength was obtained based on the parametric calculation procedure, and the relationship between the critical wavelength and the YSZ/TGO interface amplitude has also been revealed. Moreover, the analysis results can also help to understand the failure mechanisms of the DCL-TBCs and provide a reference for further process optimization.