PSI - Issue 19

Masanobu Kubota et al. / Procedia Structural Integrity 19 (2019) 520–527 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

525

6

at the fatigue limit at least by observations using an optical microscope. One of the objectives of this study, which was characterization of the effect of hydrogen on the fatigue limit determined by crack initiation, was accomplished. Although the fatigue limit of the smooth specimen was not decreased by hydrogen, we could find the effect of hydrogen in the precursor phenomenon of crack initiation. Figures 6 and 7 show the slip bands observed on the surface of the unbroken specimen at the fatigue limit. As shown in Figures 6 (b) - (d), many small spots were observed on the surface. As shown in Figure 6 (a), there were no spots before the fatigue test. These spots were observed in every specimen regardless of the environment, stress amplitude, broken and unbroken. As shown in Figure 7, these spots were persistent slip bands. The morphology of the slip bands in the hydrogen gas was different from that in air. The development of extrusion was significant in the hydrogen gas. As shown in Figures 7 (a) and (b), the morphology of the slip bands in the nitrogen gas was similar to that in air. The development of slip bands was affected by the hydrogen. Regarding the effect of hydrogen on the development of slip bands, the number of slip bands in the hydrogen gas was significantly greater than that in air as well as in nitrogen gas as shown in Table 2. The results of the observation of the slip bands suggested hydrogen facilitated the formation of slip bands. However, Nagao et al. (2018) showed that a surface oxide film covering the specimen surface prevents hydrogen uptake. On the other hand, Staykov et al. (2014) clarified the mechanism of hydrogen uptake into a metal from hydrogen gas. They clarified that the catalytic action of the bare iron surface is necessary for the dissociation of hydrogen molecules into hydrogen atoms. Therefore, probably, exposure of a fresh surface created by slip deformation to the hydrogen gas is first necessary so that hydrogen affects the formation of the slip bands. This means that hydrogen had no effect on the first slip deformation in this study. Based on these results, it could be considered that there was no effect of hydrogen on the critical stress amplitude for crack initiation in this study.

Fig.6. Observation of unbroken specimen surface. (a) Before fatigue test; (b) In air; (c) In hydrogen gas; (d) In nitrogen gas.

Fig.7. SEM observation of unbroken specimen surface. (a) In air; (b) In hydrogen gas; (c) In nitrogen gas.

Table 2. Population of slip bands. Environment Air

Hydrogen gas

Nitrogen gas

Population (/mm 2 ) 1,600

4,000

1,700

3.3 Results of deep-notched specimen

Figure 8 shows the S-N curves of the deep-notched specimens. The fatigue life in the hydrogen gas was slightly longer than that in air unlike the smooth specimen. Most of the fatigue life of the deep-notch specimen is used for