Ceramic thermal barrier coatings for gas turbine blades
The maximum service temperature of state-of-the-art Nickel base superalloys is approximately 1000°C. A further significant increase of this value by future alloy development is not expected. To improve efficiency of stationary gas turbines and jet engines by increasing turbine inlet temperature in spite of this limitation in material development, internally cooled turbine blades combined with thermal barrier coatings (TBCs) made of ZrO2 stabilized by Y2O3 are used, allowing surface temperatures of more than 1200°C. Together with film cooling of the blade surface, turbine inlet temperatures up to 1400°C can currently be reached. Newly developed thermal barrier coatings based on pyrochlores may lead to a further increase of surface temperature up to 1400°C and therewith either result in decreasing efforts for cooling or in a further increase of turbine inlet temperature.
To exploit this great potential of ceramic TBCs, the integrity of the coating must be assured under all possible service loadings. The adhesion and integrity of the thermal barrier coating is affected by the thermomechanical loading of the blade during start-stop processes and load changes as well as by time dependent degradation of the layer by sintering, oxidation and interdiffusion with the substrate material.
The deformation and damage mechanisms of thermal barrier coatings produced by plasma spraying or electron beam physical vapor deposition during isothermal annealing, thermocyclic exposure and thermomechanical fatigue with and without temperature gradients across the coating are investigated by the department "Materials Mechanics" of IEK-2 in the framework of several industry- and public funded research projects. Local mechanical properties of the single layers of TBC systems are investigated by microindentation at temperatures up to 600°C. Mechanical testing at free-standing TBCs is performed at temperatures up to 1300°C. The results of these experiments are used to establish phenomenological and FEM-based models for life prediction of thermal barrier coatings.
As an example, Fig. 3 shows the simulated stress distribution within a thermal barrier coating after thermocyclic loading considering interface roughness, elastic-viscoplastic deformation of all components, oxidation at the interface between bondcoat and thermal barrier coating as well as sintering of the ceramic topcoat. A comparison with a scanning electron micrograph of the crack path gives evidence of the correlation between predicted peak stresses and the crack propagation observed in thermal cycling tests. Based on these results obtained within a collaborative research center funded by the German research foundation a global life prediction model is developed which considers oxide growth as well as crack kinetics, the latter predicted using a fracture mechanics approach. Experimental validation of this model is one of the tasks of a further research project partly funded by industry.