Microstructural damage of thermal barrier coatings due to CMAS

Over recent years, due to a constant desire for higher efficiency engines and hence increased turbine entry temperatures and a proportional reduction in [Carbon dioxide] emissions, there is a need to understand how molten slags (CMAS: Calcia magnesia alumina-silicate), including volcanic ash, affect engine life. Thermal barrier coatings (TBC) are employed together with cooling technology to protect engine hardware from the high temperature seen within the turbine and combustion zones. At current operating temperatures, CMAS can adhere to the TBC surface resulting in premature degradation of the coating. The columnar, high porosity microstructure of electron beam physical vapour deposited (EB-PVD) TBCs make them particularly susceptible to CMAS/molten deposit attack. CMAS attack of PYSZ is reported in literature to be characterised by penetration of the melt along the columnar structure, chemically attacking the TBC whereupon yttria is leached from PYSZ and into the melt, creating an yttria depleted interaction zone. A new approach for classifying and reporting CMAS attack on TBCs is introduced in this thesis and a degradation map is created to acknowledge that the mechanism and severity of CMAS damage is related to variation in the CMAS compositions. CMAS degradation of EB- PVD has been extensively studied by previous authors, all reporting similar degradation mechanism with varying degree of severity. In this study, this category of CMAS degradation mechanism is termed “classic” CMAS attack. The primary aim of this study has been to investigate the damage caused by volcanic ash and CMAS to materials used within an aerospace gas turbine engine. The thesis investigates two aspects. It is recognised that, debris ingested by the engine will cause erosion damage to components in the cooler section of the engine (compressor), thus the first part examines this issue. A series of erosion tests with Eyjafjallajokull volcanic ash and similar sized MIL spec silica sand have been undertaken with two compressor-typical materials (Ti-6Al-4V and Inconel 718). The results were consistent with volcanic ash behaving like fine silica sand both at room and at compressor operating temperatures. The measured erosion rates are consistent with a ductile erosion mechanism with peak rates of material loss at lower impact angle. The results would appear to fit classical ductile erosion models where the material loss depends on particle velocity and follows a power with an exponent close to 2.4.