PSI - Issue 13

Tsubasa Kumamoto et al. / Procedia Structural Integrity 13 (2018) 710–715 Author name / Structural Integrity Procedia 00 (2018) 000 – 000

713

4

Fig. 2. Damage evolution curves plotted against local strain with and without hydrogen charging: (a) number of damage incidents per square micrometer; (b) damage area fraction; (c) average damage size. The average damage size was obtained by dividing the damage area fraction by the number of damage incidents per area. In our previous study (Koyama et al . 2014), hydrogen pre-charging affects all regimes: damage incubation; damage arrest; damage growth. It was reported that hydrogen not only shortens the damage incubation regime due to a decrease in the critical strain for decohesion in martensite regions, but also shortens the damage arrest regime due to a promotion of crack growth, e.g., cracking at a ferrite/martensite interface and in the ferrite grain interior. It is noteworthy that the hydrogen content was 0.66 mass ppm and the initial strain rate was 4 × 10 − 3 s − 1 in the previous study, while the hydrogen content was 0.145 mass ppm and the initial strain rates were 10 − 4 and 10 − 2 s − 1 in the present study. With this information, we note first in Fig. 2 that the damage evolution is significantly enhanced by hydrogen pre charging at a strain rate of 10 − 4 s − 1 , even though the change in the damage incubation regime is small. More specifically, three significant changes can be seen: (1) the damage nucleation rate becomes about twice as large as that of the uncharged specimens (Fig. 2a), (2) the end of the damage arrest regime is shortened from 67 to 45% local strain (Fig. 2c), and (3) the number of damage incidents just before the failure is decreased from 0.0065 to 0.0037  m − 2 , a 40% reduction (Fig. 2a). There is one more important finding in Fig. 2, namely, the enhancement of the damage evolution by hydrogen pre charging does not occur at a strain rate of 10 − 2 s − 1 (Fig. 2, red line) and its damage evolution curves are almost the same as those of the uncharged specimens. 3.3. Micro-damage characterization and fractography In order to discuss the significant effect of hydrogen pre-charging at a strain rate of 10 − 4 s − 1 , but almost no effect at 10 − 2 s − 1 , we have performed microstructural analysis. Fig. 3 shows examples of the damage obtained from SEM observations of the uncharged specimen and two hydrogen-charged specimens. The damages appear in black, whereas the martensite and ferrite appear in bright and dark contrasts, respectively. The damages were observed mostly within the martensite grains, irrespective of the hydrogen pre-charging and the strain rate, which is perhaps owing to the martensite-morphology-dependent micro-stress concentration. Specifically, the damages form at the necked portion of the martensite region. Furthermore, the damage tips were blunted by plastic deformation in the surrounding ferrite matrix and the damage stopped growing along the specimen thickness direction (Figs. 3a, c, and e). The constant average damage size indicates that the plasticity of ferrite is a factor controlling the damage arrest capability (Fig. 2c). From the viewpoint of damage growth, the damages were elongated along the tensile direction in the further deformed region at each test condition (Figs. 3b, d, and f). This fact also supports that the damage was arrested by further plastic deformation in ferrite.