ICMCTF2010 Session A3-2: Thermal Barrier Coatings

Although the use of thermal barrier coatings (TBCs) has allowed for higher gas-turbine engine operating temperatures, it is engendering new materials issues. Specifically, fine sand particles ingested by aero engines deposit on the hotter TBC surfaces as molten calcium-magnesium-aluminosilicate (CMAS) glass, which penetrates the TBCs resulting in loss of strain tolerance and premature failure of the TBCs. In the case of industrial engines, impurities such as fly ash from alternative fuels, accumulate as molten deposits on the hotter TBCs, degrading them. In this context, a generic approach for mitigating molten-deposits attack on TBCs is being developed. In this approach, the compositions of metastable TBCs are tailored such that the TBCs serve as chemical-species reservoirs. These then interact with the molten deposits rendering the deposits more benign. Examples of such novel TBCs are presented, together with strategies for testing the TBCs under molten-deposits attack and thermomechanical modeling of TBCs failure.

Air plasma sprayed (APS) thermal barrier coatings (TBCs) are widely used in gas turbine engines for thermal protection. In this study, we first examine the microstructural degradation of ZrO₂-8wt.%Y₂O₃ APS TBCs with a low-pressure plasma sprayed (LPPS) CoNiCrAlY bond coat coated on IN 738LC superalloy substrate. Furnace thermal cyclic tests were carried out at 1100°C in air with 1, 10, and 50-hr dwell period with 10-minute heat-up and 10-minute forced-air-quench. Extensive and quantitative microstructural analyses were carried out to document the population and growth of micro-cracks near the YSZ/bond coat interface, growth of thermally grown oxide (TGO), and depletion of Al-rich β-NiAl phase in the bond coat. Evolution in these microstructural features was correlated with furnace thermal cyclic lifetime of TBCs. Based on the results of microstructural analyses, a lifetime approximation model was developed based on the population and propagation of micro-cracks near the YSZ/bond coat interface. The lifetime approximation method was devised by modifying the Paris’ Law that models the steady-state crack growth. Constants in the modified Paris’ Law were determined based on quantified microstructure, and related to the parameters of furnace thermal cyclic test and specific characteristics of TBC degradation.

The maximum operating temperature of conventional thermal barrier coatings is ultimately limited by the rate of evolution of the “non-transformable” t’-phase into a depleted tetragonal form susceptible to the monoclinic transformation upon cooling. Recent work has shown that the t’-phase decomposes rather rapidly into a modulated microstructure comprising a coherent array of yttria-rich and yttria-lean lamellae. These lamellae have compositions close to those expected from equilibrium considerations for the cubic and tetragonal phases, and hence the latter should be in principle transformable to monoclinic but constrained from doing so by the interleaving cubic phase. The onset of transformability requires coarsening of the microstructure and apparently loss of coherency. Hence, it is the rate of coarsening and not the initial decomposition that determines the durability of the coating in thermal cyclic environments. Understanding the formation and coarsening mechanisms of such a microstructure would provide valuable insight for increasing the durability of future thermal barrier coatings.

This contribution describes a UK collaborative project aimed at improving understanding of the thermal cycling failure mechanisms of EBPVD YSZ TBCs on CMSX4 superalloys.

The project involved the production of TBCs with different bond coats and with controlled surface morphologies, measurement of residual stress in the alumina thermally grown oxide (TGO) using luminescence, and measurement of YSZ mechanical properties and inter-layer adhesion by indentation. Generation of stress in the coating on thermal cycling, and its relief by plastic deformation and fracture, was studied by finite element modelling (FEM). The bond coats studied include two beta-structured Pt-Al types and a gamma-gamma prime structure produced by Pt diffusion without aluminising.

Luminescence piezo-spectroscopy has been shown to be capable of high spatial resolution stress mapping through the YSZ and detecting imminent coating failure. The differences in failure mechanisms between the different bond coat types have been clarified and shown to be determined by the high temperature plasticity of the bond coat which leads to rumpling in the beta BCs. but not in the gamma-gamma prime BCs. The different failure modes lead to different optimum substrate surface finishes for the different BC systems. The YSZ top coat provides a large contribution to the driving force for spallation in all cases and its mechanical properties are shown to be sensitive to its complicated microstructure and changes with thermal cycling. Finally an indentation test has been developed that reveals qualitatively how the practical adhesion of the coatings degrade with cycling.

_{The volatilization of Si-based ceramic matrix composites CMCs in moisture-laden combustion environments is alleviated by the application of an environmental barrier coating (EBC). In principle all current (e.g. BSAS) and prospective (e.g. Y2SiO5 EBC materials are susceptible to thermochemical degradation by Calcium-Magnesium Alumino-Silicate (CMAS) deposits. The mechanism involves dissolution of the parent EBC into the CMAS melt and re-precipitation as silicate reaction products. The primary reaction product for Y2SiO5 EBCs is Ca2Y8(SiO4)6O2 oxy-apatite in all cases, but the secondary products and morphological characteristics of the reaction zone depend on time, temperature and whether the environment is stagnant laboratory air or flowing H2O/O2. Mechanisms and their implications to the durability of EBCs in potential gas turbine applications are discussed in light of experiments exploring the effect of these variables. (Work sponsored by the Office of Naval Research).}

As-fabricated thermal barrier coating (TBC) systems generally consist of a superalloy substrate, a MCrAlY bond coat (M=Ni, Co, Fe), and a ceramic (usually partially stabilized zirconia) top coat. The conventional methods for fabricating the two coating layers generally derive from thermal spray and physical vapor deposition techniques. The present work demonstrates the feasibility of using an innovative method for synthesizing both the top coat and bond coat simultaneously. The selected fabrication technique, Spark Plasma Sintering (SPS), provides not only the opportunity to synthesize both coatings at once, but the process is quite rapid and can produce dense layers. This paper describes the results for the application of this method to a ZrO₂ top coat with a NiCrAlY bond coat on a Ni base Hastelloy X substrate. Select variations in densification parameters are considered. The resulting multi-layer system is characterized with optical microscopy, scanning electron microscopy (SEM), Energy-Dispersive X-ray analysis (EDAX) and X-ray Diffraction (XRD).