Effect of material characteristics and test variables on thermal fatigue of cast superalloys. A review

Abstract Tapered disc thermal fatigue tests have been conducted on two nickel- and three cobalt-base alloys. Measurements of crack length as a function of number of thermal cycles were made on cracks growing from notches machined on the specimen periphery. In addition, some results are reported of crack initiation studies involving the detection of small initiating cracks on the specimen periphery. Both material and testing variables were studied. The former included the effect of alloy composition, solidification conditions and heat treatment. The latter included the effect of specimen geometry, temperature of the hot bath and hold time at the maximum temperature. It was shown that all these variables affected the thermal fatigue resistance although not always in a clearly defined manner. For example, the thermal fatigue ranking of the five alloys was found to be sensitive to the test conditions. As the maximum temperature and hold time at temperature increased, cobalt-base alloys showed increased resistance to thermal fatigue cracking relative to the nickel-base alloys. This represented a reversal of the trend at lower temperatures. Coarser grain size specimens had reduced crack propagation rates. Taken in conjunction with the results from a directionally solidified specimen, it is concluded that in the range of test conditions studied, slower solidification leads to reduced thermal fatigue crack propagation rate. In all cases it is shown that cracking is principally interdendritic although the details of the effect of interdendritic spacing are not understood. Two major difficulties with this type of study are the problems associated with calculating the time - temperature - strain history of the specimen and the interpretation of the effects of microstructural instabilities. Some preliminary work has been done to calculate a hysteretic loop which demonstrates the importance of creep and relaxation phenomena. Evidence is presented which indicates that microstructural instabilities may be involved in an important way in the reported hold time and maximum temperature effects. Environmental interactions have not been studied specifically but indirect observations indicate that a major role may be played by oxidation phenomena.