Creep behaviour leading to Type IV cracking for service-exposed 1.25Cr–0.5Mo steel welds

Abstract

To ensure reliability of elevated temperature components, the creep behaviour of weldment must be predicted since the ultimate failures mostly take place at this tiny region. In the case of low alloy ferritic steels, the most likely failure mode of equipment operated for long hours should be Type IV cracking, which is defined as preferential damage evolution at the Intercritical HAZ (ICZ). Despite the importance of this phenomenon, there have been some uncertainties remained unsolved. In order to elucidate the cause and accelerating factors of Type IV cracking, creep behaviours of cross-weld and the ICZ microstructure have been examined in the present work using service-exposed 1.25Cr–0.5Mo steel. Onset time to Type IV failure significantly reduced when tested by spirally notched cross-weld specimens as a result of concentrated damage accumulation at the root of a vee notch, revealing that multiaxaial stress state could play a key role in Type IV failure. The feature of creep damage suggests that grain boundary damage leading to Type IV cracking is caused by the sliding of grain boundaries around fine grains which are considered to be the products of partial transformation during welding. Heterogeneous damage evolution to the level of facet cracking surrounded by damage free grains raises the fundamental question on the validity of a generally accepted assumption, namely, that stress of grains associated with a grain boundary cavity will be off-loaded. As a matter of fact, a clear evidence that grain boundary cavitation accelerates the strain rate at the tertiary regime has not been observed in creep curves of simulated ICZ specimens, owning a bimodal microstructure expected at the ICZ in whole gauge length. Difference in the susceptibility to Type IV cracking has been found in materials with the same alloying elements and the vulnerability of the ICZ microstructure is not necessarily dependent upon creep strength of parent material. Considerable metallurgical factors to shorten the onset time to Type IV damage and the effectiveness of strain rate measurement as a potential technique for the life assessment shall be discussed.