The influence of inelastic damage on creep, fatigue and fracture toughness

Abstract

A significant number of advanced gas cooled reactor (AGR) plant components with operating temperatures in the range of 500-650 °C undergo creep-fatigue loading conditions. Some of these components have previously operated in the creep regime and subsequently had their operating temperatures reduced. The aims of this project are to determine the effects of service exposure, thus prior creep damage, on fatigue crack growth and fracture behaviour of the material and to include these effects in fatigue and fracture assessments. In laboratory testing, creep damage can be introduced into the material globally, where uniform creep damage will form through a section, or locally, where creep damage is confined to a small region local to the crack tip. The local creep damage (LCD) approach has been employed in this research and the global creep damage (GCD) tests will be performed in future work. The material examined is an ex-service Type 316H stainless steel, which is widely used in AGR plant components. Creep damage local to the crack tip has been introduced into the material by interrupting creep crack growth (CCG) tests performed at 550 °C, which is the typical operating temperature of the plant components. Fatigue crack growth (FCG) and fracture toughness tests have been performed on these LCD specimens and the results are compared to those of conducted on the material with no creep damage. The yield stress of Type 316H stainless steel is relatively low at 550 °C. In order to limit the extent of plasticity in CCG tests during loading and unloading, the material has been uniformly pre-compressed (PC) to 8% plastic strain at room temperature prior to specimen manufacturing. It has been shown that the pre-compression process increases the yield stress and reduces the tensile failure strain of the material compared to the as-received (AR) material’s tensile response. A substantial drop in the rupture time and creep ductility has also been exhibited in the uniaxial creep behaviour of the PC material compared to AR, whereas the creep strain rate remains almost unchanged. To interrupt CCG tests on LCD specimens at a desired normalised crack length of around a/W = 0.5, tests have been performed on nominally identical specimens to calibrate the CCG behaviour of the material. The results from these tests have been analysed in terms of the creep fracture mechanics parameter C* and compared to the available test data on AR material at 550 °C. It has been shown that the creep crack growth rate, da/dt, in PC material is around an order of magnitude larger than those of seen in short term tests on AR material and the data follow the trend of long term tests on AR material. FCG and fracture toughness tests have been performed on LCD specimens at room temperature to investigate the potential effects of prior creep damage on subsequent fatigue and fracture behaviour of the material, outside the creep regime. The results from these tests have been compared to those of performed on PC material containing no creep damage. A considerable drop in the fracture energy of the material has been observed in LCD specimens in regions close to and away from the creep damage zone, compared to those of obtained from PC material. Assuming a sharp crack tip in LCD specimens, though discontinuous, the FCG rate per cycle, da/dN, data have been correlated with the stress intensity factor range, ΔK. A severe delay in the crack initiation and also a significant drop in the crack growth rate per cycle in the regions close to the creep damage zone have been observed in the FCG data on the LCD specimen compared to the PC material. However, further away from the creep damage zone the FCG data from LCD specimen converge with the trends attained from the PC material. Further FCG and fracture toughness tests on LCD specimens need to be performed in future work to confirm the observed trends. Also to examine the influence of uniform creep damage in the absence of micro cracks (which form ahead of the crack tip when the local creep damage approach is employed) on the subsequent FCG and fracture behaviour of the material, GCD tests need to be conducted in future work and the results to be compared to those of obtained from LCD specimens.