PSI - Issue 19

Masanori Nakatani et al. / Procedia Structural Integrity 19 (2019) 312–319 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

317

6

Fig. 4. S - N diagram of non-charged and H-charged FG specimens with small defects.

The surface crack-growth in the run-out specimens is exhibited in Fig. 5 (b). The final lengths of the cracks were revealed to be approximately the same in all conditions. However, in contrast to the crack-growth in the non-charged specimens, cracks in the H-charged specimens developed gradually at relatively high cycles, as indicated by the arrows. Considering the crack-growth curve in Fig. 5 (b) and the fact that some smooth specimens fractured at an N higher than 10 7 cycles ( cf. Fig. 4), it is very likely that these cracks would re-accelerate above 10 7 cycles and ultimately lead to failure. Despite the possibility of crack-growth re-acceleration, the progression of these cracks was extremely slow when the d a /d N was below 10 − 11 m/cycle, as evidenced by Fig. 5 (c). In other words, the crack-growth per cycle of these cracks was a few orders of magnitude smaller than the radius of an atom. The crack-growth was slow enough to prevent cracks from leading to failure below N = 10 7 cycles, although there was the possibility of their failure above 10 7 cycles of N . Nevertheless, it can be confirmed that the threshold Δ Κ th for small crack-growth diminished as a result of H-charging, although the existence of the Δ Κ th in the H-charged specimens remains debatable. In the fractured specimens (square marks in Fig. 5 (c)), the impact of hydrogen on crack-growth behavior varied according to the Δ K value. Near the threshold, the crack in the non-charged specimen displayed the behavior typical of a small crack, i.e. , an initial decrease, followed by a sudden acceleration to accompany the increase in crack-growth.

Fig. 5. (a) Crack-growth in EDM-notched FG specimens and (b) at run-out of EDM-notched FG specimens. (c) Crack-growth rates in EDM notched FG specimens.