In an atmospheric plasma spray (APS) process, in-flight powder particle characteristics, such as the particle velocity and temperature, have significant influence on the coating formation. The nonlinear relationship between the input process parameters and in-flight particle characteristics is thus of paramount importance for coating properties design and quality control. It is also known that the ageing of torch electrodes affects this relationship. In recent years, machine learning algorithms have proven to be able to take into account such complex nonlinear interactions. This work illustrates the application of ensemble methods based on decision tree algorithms to evaluate and to predict in-flight particle temperature and velocity during an APS process considering torch electrodes ageing. Experiments were performed to record simultaneously the input process parameters, the in-flight powder particle characteristics and the electrodes usage time. Various spray durations were considered to emulate industrial coating spray production settings. Random forest and gradient boosting algorithms were used to rank and select the features for the APS process data recorded as the electrodes aged and the corresponding predictive models were compared. The time series aspect of the data will be examined.

Evaluating and understanding the relationship between processing, microstructure and performance of a dielectric coating is essential for its practical usage and reliable application. In this study, the role of the powder feedstock on the properties of atmospheric plasma sprayed forsterite (Mg 2 SiO 4 ) dielectric coatings was investigated by using different forsterite powder cuts. The microstructural and porosity characteristics of the coatings associated with the spray conditions employed were assessed via scanning electron microscopy (SEM) and image analysis. The phase composition of the coatings was studied via X-ray diffraction and their crystallinity index determined. The electrical insulating characteristics were investigated using the dielectric breakdown test. The obtained electrical properties were correlated with the microstructural characteristics. Ultimately, a performance comparison between forsterite and other dielectric coatings tested in similar conditions is presented.

Si-based ceramics (e.g., SiC and Si 3 N 4 ) are known as promising high-temperature structural materials in various components where metals/alloys reached their ultimate performances (e.g., advanced gas turbine engines and structural components of future hypersonic vehicles). To alleviate the thickness recess that Si-based ceramics undergo in a high-temperature environmental attack (e.g., H 2 O vapour), appropriate refractory oxides are engineered as environmental barrier coatings (EBCs). Presently, the state-of-the art EBCs comprise multilayers of silicon (Si) bond coat, mullite (Al 6 Si 2 O 13 ) intermediate layer and BaO-SrO-Al 2 O 3 -SiO 2 (BSAS) top coat. Evaluating and understanding their mechanical properties, such as, the elastic modulus (E) and the strain-stress relationship is essential for their practical application and reliable employment. It was investigated via depth-sensing indentation the role of high-temperature treatment (1300°C), performed in H 2 O vapour environment (for time intervals up to 500 h), on the mechanical behaviour of air plasma sprayed Si/mullite/BSAS layers deposited on SiC substrates. Laser-ultrasonics was employed to evaluate the E values of as-sprayed coatings and to validate the indentation results. The fully crystalline, crack-free and near crack-free as-sprayed EBCs were engineered under controlled deposition conditions. The (i) absence of phase transformation and (ii) stability of the low elastic modulus values (e.g., ~60-70 GPa) retained by the BSAS top layers even after harsh environmental exposures provides a plausible explanation for the almost crack-free coatings observed. The measured mechanical properties of the EBCs and their microstructural behaviour during the high-temperature exposure are discussed and correlated.