Since modern gas turbine suppliers have come under immense cost pressure in order to remain economically competitive, the advancement of MCrAlY coatings (M= Ni and/or Co) has become of great industrial interest. MCrAlYs are oxidation and corrosion resistant coatings applied onto hot gas path parts of gas turbines. Adapting them onto the more and more versatile operation conditions of modern gas turbines has proved to be a necessity in order to ensure maximum service life of each part and the turbine. Still today the general processing route is to adapt the chemical composition and properties on an experimental “trial and error” basis. This, however, is a time and cost consuming process, because every potential system has to be validated and field tested later.
Experience has shown that modern thermodynamic software is a powerful tool in order to predict properties and structure of materials through calculation of system related state variables and simulation of diffusion processes. At the present point in time these programs can only simulate diffusion between solid and/or liquid phases, though. A straightforward consideration of the atmospheric influences is not possible due to unavailability of mobility databases for oxygen and relevant oxides (e.g. Al2O3, Cr2O3). However, since the oxide scale acts as an oxidation barrier and also leads to element depletion in the coating, its diffusion controlled formation and growth is crucial for the lifetime of the coating and the whole part.
Therefore, in this research the oxidation of MCrAlY was considered by combining the diffusion simulation software DICTRA with predictive oxidation models using mathematical and FDM approaches. Investigations were conducted on an industrial relevant γ/β CoNiCrAlY - IN738LC alloy system at 900°C and 1000°C. Accuracy of the developed models was assessed based on experimental results from Microscopy, REM, EDX and HT-XRD of long term heat treated MCrAlY overlay systems.
Latest results refer to a good correlation of simulated and experimental data. CoNiCrAlY and IN738LC specific phases as well as surface and substrate near element depletion areas were successfully predicted by the models and could sufficiently be correlated to experimental results. That implicates the potential of thermodynamic software for such use cases. The presented data also forms the foundation of a lifetime assessment model for MCrAlY overlay systems, which frames the primary target of this ongoing project.