PSI - Issue 7

Ning Wang et al. / Procedia Structural Integrity 7 (2017) 376–382

379

4

N. Wang et al. / Structural Integrity Procedia 00 (2017) 000–000

3.2. Fatigue crack initiation In Fig. 3, significant difference is observed between these two types of specimens even at similar fatigue lifetime, i.e., specimen surface crack initiation for as-received specimens, while interior micro-defect for hydrogen charged specimens. Fig. 3a and b shows cracks initiated from a surface inclusion and an interior inclusion, respectively. In Fig. 3b, a fish-eye pattern is observed and an inclusion is found at the center. No apparent GBF can be found around the inclusion. Fig. 4a is a follow up observation with the help of OM, and results indicate the optically dark area (ODA), according to Murakami et al. (2002), is filled with the whole fish-eye pattern. This area is reported to be crack growth areas affected with hydrogen [Furuya et al. (2002)]. As for the surface crack initiation, as shown in Fig. 4b, the optically dark area is not as obvious as that for interior cracking. It seems that the formation of ODA is essentially related to micro-defects, a more probable crack initiation site in case of hydrogen due to the hydrogen trapping process. This indicates the contribution of hydrogen to fatigue crack initiation behavior.

(a) (b) Fig. 4 OM observation of fracture surface for hydrogen charged specimen (a) ( σ =460 MPa, N f =3.11 × 10

7 cycles) and as-received specimen (b)

( σ =505 MPa, N f =5.5 × 10

7 cycles)

3.3. Effect of hydrogen on interior cracking process Fig. 5 shows the morphologies of interior cracking of hydrogen charged specimens. It is observed in Fig. 5a and b that there are several non-metallic inclusions on the fracture surface though with one as fatal. The internal inclusions are spherical and mainly consist of Al and O elements based on Energy Dispersive Spectrometry analysis [Zhu et al. (2015a)]. They are most probably in the form of oxide Al 2 O 3 , which are formed during the forging process of the steel. The average equivalent diameters of inclusions at fracture origins are 9-82 µ m. It seems that several inclusions have been activated as potential crack initiation sites for hydrogen charged specimens, indicating the hydrogen element tended to be trapped by micro-defects. A similar case can be observed in Fig. 5d where a cluster of inclusions nucleate a fatigue crack but is dominated by the main crack originated from the near surface inclusion. According to Murakami et al. (1994), fatigue strength is inversely proportional to micro-defect size, the higher the inclusion size, the lower the fatigue strength. The trapping of hydrogen is more likely to occur at larger micro defects, and thus is more probable to dominate crack initiation. This implies the fatigue strength of hydrogen charged specimens is lower than that of as-received specimens, as shown in Fig. 2. It is worth noting in Fig. 5f and h that faceted crack growth is observed around critical non-metallic inclusions. The existence of facets, cleavage or intergranular, implies the contribution of hydrogen to early stage of interior cracking process, and indicates the easiness for interior crack growth. It is known that fatigue crack growth is faster and threshold of fatigue crack propagation will be lower at hydrogen environment. This explains why the fatigue lifetime at certain stress levels are shortened significantly. Interestingly, such kind of facets was not reported in high strength steels in VHCF regime [Yang, et al. (2010)]. Whether the strength level plays a role in the hydrogen effect