The influence of sulphidizing attack on the mechanism of failure of coated superalloy under cyclic loading conditions.
A systematic study of the effect of sulphidizing atmosphere on the High Temperature Low Cycle Fatigue (HTLCF) properties of coated and uncoated unidirectionally solidified MARM002 nickel base superalloy was performed at 870°C. The coating systems investigated were, aluminide coating, three types of platinum modified aluminide coatings, and platinum coating. The creep-plasticity mode of the strain range partitioning method was used for creep-fatigue loading. A constant loading regime (Strain range 6.6 x 10-3 ) was used to test the samples in argon, air and Ar + 5%S02 and a lower strain range of3.8 x 10-3 was used to investigate the creep-fatigue properties in Ar + 5%S02 only. The results were analysed using scanning electron microscopy including spot analyses (SEM-EDS), Auger electron spectroscopy (AES) and X-ray diffraction (XRD) techniques. The synergistic effect of sulphidizing environment and the creep fatigue loading (Strain range - 0.66%) resulted in accelerated failure in all the materials systems tested, except for the TYPE I platinum aluminide coated sample. This coating displayed a "self-healing" mechanism which enhanced its fatigue life under sulphidizing conditions. In general, the coatings had an adverse effect on the fatigue properties of the material systems. This was due to the poor mechanical properties of the coating. The mechanical properties of the coating was influenced by the coating microstructure and the chemical composition. The modification of the NiAI zone with platinum in the platinum aluminide coatings improved the fatigue properties of the coating by altering the crack propagation mechanism in the NiAl zone. The higher the platinum content in this region the more brittle it became. The platinum modified aluminide coating showed an improvement in the corrosion fatigue properties in the S02 containing environment at the higher strain range when compared with the uncoated, aluminide coated and platinum coated samples. However, at the lower strain range all the coating systems performed worse than the uncoated alloy. This was mainly due to the brittle failure of the coating. The platinum modified aluminides performed the worst due to the presence of brittle platinum aluminide phases. The interdiffusion and interaction of platinum with the substrate alloying elements, resulted in this coating being ineffective for corrosion protection. The resultant coating layer produced poor corrosion-fatigue properties. Although the coating systems did show evidence of resistance to sulphidation and oxidation there were relatively ineffective under the combination of sulphidizing environment and fatigue loading due to their poor mechanical properties. The mechanism of sulphidation was consistent for all the material systems tested with oxidation proceeding first and sulphidation proceeding at the corrosion scale/substrate interface. The crack propagation in the coating and substrate was controlled by the sulphidation attack at the crack tip and failure of the oxide scales formed in the cracks.