Creep-fatigue-environment interaction in a bainitic 2.25wt.%Cr-1wt.%Mo steel forging

The bainitic form of 2.25Cr-1Mo steel (where the composition is in approximate weight per cent) is at present under consideration for application in thick-section coal conversion pressure vessels. The purpose of the research reported here was to determine the influence of cyclic waveform and environment on the elevated temperature fatigue behavior of this material. Uniaxial strain-controlled fatigue and creep-fatigue tests of a bainitic 2.25Cr-1Mo steel forging were performed at 565 °C (1050 °F) in both air and vacuum environments; for most tests the strain range was 0.50% fully reversed. Hold periods of up to 6 min in tension only, compression only, and combined tension and compression all significantly decreased life relative to pure fatigue. Tensile holds were more damaging than analogous compressive holds but considerably less harmful than the combined tensile-compressive hold waveform. These results were attributed to the amount of creep strain accumulated and differences in crack growth rates. The hold period duration had only a slight effect; this was explained by the kinetics of stress relaxation during the hold. Specimens preflawed with tiny electrodischarge-machined notches had their cyclic life reduced by an average of 22% with respect to smooth specimens, suggesting that crack propagation is the dominant fatigue mechanism under these loading conditions. This conclusion was subtantiated by the presence of small cracks in smooth specimens tested from 5% to 25% of normal cyclic life. Testing in vacuum improved fatigue life by approximately fourfold over air results. Owing to crack tip rewelding, compressive holds yielded the best results in vacuum (even better than pure fatigue), whereas any waveform with a tensile hold was much more damaging. Compared with vacuum, oxidation appears to enhance both crack initiation and crack growth. In conjunction, these results suggest that environment is a more important factor than creep for the test conditions investigated here. Attempts at fitting the data to an existing creep-fatigue damage model based on an environmental interaction proved unsuccessful.