The influence of test method on failure mechanisms and durability of a thermal barrier system

Failure mechanisms that occur in commercial thermal barrier systems with a Pt-aluminide bond coat and an electron beam vapor deposited thermal barrier coating have been compared for two different test protocols. The two methods consist, respectively, of a furnace cycle test and a burner rig test that differ in their configuration (disk relative to cylinder) and in their cycle time/rate. It has been shown that the failure mechanism is the same for both tests. It involves a displacement instability of the thermally grown oxide (TGO). Detailed measurements reveal substantial differences in microstructure evolution en route to failure. For the burner rig configuration, the TGO is much thinner, the cracks in the TBC are relatively longer and the amplitudes of the TGO instabilities are smaller. Simulations have been used to rationalize the differences in terms of the different hot times and by virtue of the occurrence of vertical separations in the TBC in the burner rig configuration.