The porous architecture of coatings has a significant influence on the coating performances and thus should be properly designed for the intended applications. For simulating the coating properties, it is necessary to determine the numerical representation of the coating microstructure. In this study, YSZ coatings were manufactured by suspension plasma spray (SPS). Afterwards, the porous architecture of as-prepared coatings was investigated by the combination of three techniques, imaging analysis, Ultra Small Angle X-ray Scattering (USAXS), and X-ray transmission. A microstructural model for reconstructing the porous architecture of the SPS coating was subsequently computed according to the collected experimental results. Finally, the coating thermal properties were simulated based on the model and were compared with the experimental results.

YSZ coatings were deposited by suspension plasma spraying and a parametric study was performed with different process parameters. Afterward, the porosity of the as-prepared coatings was investigated by SEM imaging and X-ray transmission and a multivariate analysis of the collected data was carried out. The results show that total porosity correlates negatively with suspension mass load, but positively with original powder size, spray step, substrate roughness, and spray distance, which was found to have the greatest impact. A porosity prediction model was also developed and its practical use is discussed.

Suspension plasma spray (SPS) is far more complicated than conventional plasma spray and requires a deep knowledge about the influence of process parameters and their correlations. In this study, YSZ coatings were manufactured by SPS with six different process parameters such as plasma power, suspension mass load, original powder size, substrate surface topology, spray distance, and spray step. Afterwards, the porosity of as-prepared coatings was investigated by image method and X-ray transmission technique. A multivariate analysis on the collected experimental data was carried out by employing mathematical statistics methods. The results showed that: 1) Coating porosity has a negative correlation with plasma power and suspension mass load and a positive correlation with the original powder size, spray distance, spray step, and substrate roughness; 2) Spraying distance is the main factor affecting to coating porosity, followed by suspension mass load and substrate surface roughness, respectively. A linear model for porosity prediction was developed and was verified by experiments. The mechanism by which process parameters influence coating porosity is also discussed.