Hydrogen embrittlement testing of alloy 718 for oil and gas applications

Slow strain rate (SSR) tensile testing of smooth specimens is useful for comparing the susceptibilities of different alloys and microstructures to hydrogen embrittlement in the testing environment and is the most commonly performed test method, mostly due to the short duration of the test. There are several accelerated fracture mechanics test methods that have been developed to produce useful design data in the form of threshold stress intensity factors (Kth) for crack growth in a hydrogen-rich environment and have a much shorter test duration compared to constant displacement/constant load fracture mechanics testing. One type of accelerated fracture mechanics test method is the rising step load (RSL) test method, which attempts to simulate the loading conditions experienced by components in application by holding the specimen at constant displacements for a majority of the test.;In this investigation, alloy 718 was annealed and aged to produce microstructures with variations in grain size, grain boundary precipitation, and strength level. The peak-aged and over-aged conditions exhibited a high strength and a low strength, respectively, and were produced through heat treatments that conformed to the API standard. An under-aged condition had no grain boundary delta phase and a similar yield strength to the over-aged condition. A high delta condition was produced through a double aging heat treatment that precipitated large delta phase particles on a significantly greater fraction of grain boundaries compared to the other conditions and produced a comparable strength level as the peak-aged condition. A high temperature and time annealing heat treatment was performed on these four conditions which resulted in a relatively large grain size. A small grain size condition was also produced by performing a lower temperature and time annealing heat treatment and was peak-aged.;Grain boundary precipitation was evaluated through scanning electron microscopy (SEM), and the mechanical properties of the different microstructures were determined. The effect of microstructure on hydrogen embrittlement susceptibility of alloy 718 was primarily assessed through SSR experiments performed with cathodic hydrogen charging and in air on all microstructural conditions. RSL tests were performed on circular notch tensile (CNT) specimens of the peak-aged, over-aged, and under-aged microstructures to evaluate how the hydrogen embrittlement susceptibilities and fracture modes produced in the RSL test compare to the SSR test.;The extent of grain boundary delta phase precipitation had the strongest effect on hydrogen embrittlement susceptibility. The under-aged condition exhibited significantly greater hydrogen embrittlement resistance than the other microstructural conditions and a hydrogen-affected fracture mode consisting of primarily transgranular cleavage cracking. The high delta condition exhibited the lowest hydrogen embrittlement resistance and a fracture surface with a brittle region composed of intergranular fracture with a serrated appearance on the facets. The peak-aged, over-aged, and small grain size conditions exhibited hydrogen cracking regions composed of mostly smooth intergranular cracking with some transgranular cracking. The volume fraction, coherency, and/or size of the primary strengthening gamma' and gamma '' was primarily responsible for the strength levels of the different microstructures and was also determined to affect hydrogen embrittlement resistance. The effect of gamma' and gamma'' precipitation on hydrogen embrittlement resistance was best evidenced by the higher strength peak-aged condition which had lower hydrogen embrittlement resistance compared to the lower strength over-aged condition despite having less grain boundary delta phase precipitation. Reducing the grain size for the peak-aged condition had no statistically significant effect on hydrogen embrittlement resistance.;RSL tests of the CNT specimens for the peak-aged, over-aged, and under-aged conditions were successful in causing crack growth to begin at the notch due to hydrogen embrittlement mechanisms while the specimen was held at a constant displacement. RSL testing produced threshold stress intensity factors for crack growth (Kth) with cathodic hydrogen charging that exhibited the same trend as the total elongation ratios produced by SSR testing. The brittle hydrogen-affected regions produced in the RSL tests showed the same fracture modes as were observed in the SSR fracture surfaces. The similarity in fracture modes from RSL and SSR testing indicates that the mechanisms for crack propagation in the two test methods may be similar. The mechanism for crack propagation may be independent of plasticity around the crack, or dynamic straining in the SSR test may have had a similar effect as the plastic zone around the crack in the RSL test. Smaller step sizes for the RSL tests resulted in lower Kth values with less uncertainty for all conditions at all step times studied. For future hydrogen embrittlement testing, the RSL test has the ability to provide Kth values through constant displacement conditions that are similar to application and with a much shorter test duration than constant displacement fracture mechanics testing. (Abstract shortened by UMI.).