CORRELATION OF SHORT-TERM TO LONG-TERM OXIDATION TESTING FOR ALUMINA FORMING ALLOYS AND COATINGS

Engineering long cyclic oxidation life of high temperature materials requires success on two fronts. First a slow growing protective oxide scale must form during the elevated temperature exposure. To satisfy this aspect, alumina-forming alloys and coatings are widely accepted as leading materials for use in this environment and are the focus of this discussion. The second aspect is the formation of an adherent oxide that resists spallation during thermal cycling. The driving force for spallation is the stored elastic strain energy that develops from stresses in the oxide scale. Once this stored elastic strain energy exceeds the oxide-substrate interfacial toughness, cracking and subsequent spallation occurs followed by rapid oxidation of the substrate. With advances in alloy and coating development resulting in higher operating temperatures and increased service lives, researchers are faced with excessive laboratory time and cost required to perform a long-term cyclic oxidation test.The challenge is to predict long-term oxidation behavior from short-term experiments. Since the rate limiting step to high temperature oxidation is a thermally activated process, previous investigations were performed at increased exposure temperatures for rapid degradation of the alloys and coatings to rank material performance. Others have mechanically induced oxide spallation to give insight on the adherence of oxide scales prior to spontaneous failure. In this investigation, short-term testing is employed to gain insight on long-term performance and to determine inputs into a cyclic oxidation model for life-time prediction. This model operates in an iterative process where one iteration is a thermal cycle. The amount of oxide formed during the high temperature segment is calculated followed by the amount that is lost due to scale spallation during cooling. Retained oxide at the end of this cycle is used as the starting point for the following iteration. The two inputs into this model are the oxide scale growth and spallation behavior. Scale growth behavior corresponds to the isothermal growth kinetics that are experimentally determined by thermogravimetric analysis. Oxide scale spallation behavior is quantified by two short-term experiments of a novel acoustic emission experiment during a 24 hour exposure and the stress measurement of the scale after an exposure to the temperature of interest. Results from these short-term tests and modeled cyclic oxidation are compared to life-times from long-term cyclic oxidation tests