A study was done to determine the fatigue crack growth behavior of a PWA 1484 single-crystal nickel-base superalloy in a temperature range of 427°C to 871°C. Two distinctive failure modes were observed, which were a function of both temperature and frequency. At lower temperatures and higher frequencies crack growth occurred on the {111} octahedral slip planes at an oblique angle to the loading direction. Higher temperatures and decrease in frequencies favored angle to the loading direction. Higher temperatures and decrease in frequencies favored a Mode I type failure process. The failure mode transitions were explained by invoking arguments based on environmental damage mechanisms. The fatigue crack growth rate data were analyzed using three different crack driving force parameters. The parameters investigated consisted of the Mode I stress intensity parameter corrected for the inclined crack trajectory, and two different octahedral Mode II parameters, which are based on the calculation of resolved shear stresses on the {111} slip systems. The Mode I ΔK parameter did a fair job in correlating the data but did not collapse it into a single narrow band. The two octahedral crack driving force parameters, ΔKRSS and a newly proposed ΔKOCT, collapsed all the data into a single narrow band. In addition to correlating the fatigue crack growth rates, the two octahedral parameters also predicted the {111} planes on which the crack growth took place.

No abstract

A study was performed to determine the variables which influence hold time crack growth resistance of Alloy 10, a new powder metallurgy (P/M) superalloy. In a well controlled study, both the effect of compositional changes and variation in heat treatments were investigated. The results indicate that significant changes in the alloy's niobium, tantalum and cobalt content did not have an appreciable effect on hold time crack growth resistance. In contrast to the composition study, the heat treatments evaluated produced up to an order of magnitude changes in the crack growth resistance. Quantitative image analysis was performed to analyze the microstructural features produced by each heat treatment. It was found that the cooling precipitate size distribution is closely related to the measured hold time crack growth behavior. The larger the mean size of the cooling precipitates, the better is the resistance to hold time crack growth. It is proposed that the size and distribution of the precipitates might play an important role in determining the extent of crack tip relaxation which occurs through creep-type processes. The differences in the stress relaxation rate influence the crack driving forces and thus crack growth resistance.

Oxidation and the effects of high temperature exposures on notched fatigue life were considered for a powder metallurgy processed supersolvus heat-treated ME3 disk superalloy. The isothermal static oxidation response at 704 °C, 760 °C, and 815 °C was consistent with other chromia forming nickel-based superalloys: a TiO 2 -Cr 2 O 3 external oxide formed with a branched Al 2 O 3 internal subscale that extended into a recrystallized γ'-dissolution layer. These surface changes can potentially impact disk durability, making layer growth rates important. Growth of the external scales and γ'dissolution layers followed a cubic rate law, while Al 2 O 3 subscales followed a parabolic rate law. Cr-rich M 23 C 6 carbides at the grain boundaries dissolved to help sustain Cr 2 O 3 growth to depths about 12 times thicker than the scale.The effect of prior exposures was examined through notched low cycle fatigue tests performed to failure in air at 704 °C. Prior exposures led to pronounced debits of up to 99 % in fatigue life, where fatigue life decreased inversely with exposure time. Exposures that produced roughly equivalent 1 µm thick external scales at the various isotherms showed statistically equivalent fatigue lives, establishing that surface damage drives fatigue debit, not exposure temperature. Fractographic evaluation indicated the failure mode for the pre-exposed specimens involved surface crack initiations that shifted with exposure from predominately single intergranular initiations with transgranular propagation to multi-initiations from the cracked external oxide with intergranular propagation. Weakened grain boundaries at the surface resulting from the M 23 C 6 carbide dissolution are partially responsible for the intergranular cracking. Removing the scale and subscale while leaving a layer where M 23 C 6 carbides were dissolved did not lead to a significant fatigue life improvement, however, also removing the M 23 C 6 carbide dissolution layer led to nearly full recovery of life, with a transgranular initiation typical to that observed in unexposed specimens.

The effects on fatigue life of high temperature exposures simulating service conditions were considered for two disk superalloys. Powder metallurgy processed, supersolvus heat treated Udimet 720 and ME3 fatigue specimens were exposed in air at temperatures of 650 to 704 C, for times of 100 h to over 1000 h. They were then tested using conventional fatigue tests at 650 and 704 C, to determine the effects of exposure on fatigue resistance. Cyclic dwell verification tests were also performed to contrast the effects of intermixed exposures and fatigue cycles. The prior exposures reduced life by up to 70% and increased the scatter in life, compared to unexposed levels. Cyclic dwell tests reduced lives even more. Fractographic evaluations indicated the failure mode was shifted by the exposures and cyclic dwells from predominantly internal to often surface crack initiations. The increased scatter in life was related to the competition between internal crack initiations at inclusions or large grains producing longer lives, and surface crack initiations at an environmentally affected surface layer producing shorter lives.

No abstract

The effects of hot corrosion pits on low-cycle fatigue life and failure modes of the disk superalloy ME3 were investigated. Low-cycle fatigue specimens were subjected to hot corrosion exposures producing pits, then tested at low and high temperatures. Fatigue lives and failure initiation points were compared to those of specimens without corrosion pits. Several tests were interrupted to estimate the fraction of fatigue life that fatigue cracks initiated at pits. Corrosion pits significantly reduced fatigue life by 60 to 98%. Fatigue cracks initiated at a very small fraction of life for high-temperature tests, but initiated at higher fractions in tests at low temperature. Critical pit sizes required to promote fatigue cracking were estimated based on measurements of pits initiating cracks on fracture surfaces.

A study was done to determine the fatigue crack growth behavior of a PWA 1484 single crystal nickel base superalloy in a temperature range of 427°C to 871°C. Two distinctive failure modes were observed which were a function of both temperature and frequency. At lower temperatures and higher frequencies crack growth occured on the {111} octahedral slip planes at an oblique angle to the loading direction. Higher temperatures and decrease in frequencies favored a Mode I type failure process. The failure mode transitions were explained by invoking arguments based on environmental damage mechanisms. The fatigue crack growth rate data were analyzed using three different crack driving force parameters. The parameters investigated consisted of the Mode I stress intensity parameter corrected for the inclined crack trajectory, and two different octahedral Mode II parameters which are based on the calculation of resolved shear stresses on the {111} slip systems. The Mode I ΔK parameter did a fair job in correlating the data but did not collapse it into a single narrow band. The two octahedral crack driving force parameters, ΔKRSS and a newly proposed ΔKOCT, collapsed all the data into a single narrow band. In addition to correlating the fatigue crack growth rates, the two octahedral parameters also predicted the {111} planes on which the crack growth took place.