PSI - Issue 13

Taro Suemasu et al. / Procedia Structural Integrity 13 (2018) 1088–1092 Author name / Structural Integrity Procedia 00 (2018) 000 – 000

1091

4

3. Experimental results and discussion

Fig. 4 shows the SEM images of the fatigue cracks propagating from the pre-crack tip after N = 2000 cycles. The propagation lengths of the fatigue cracks from the pre-crack tip are 220 μm and 178 μm for the fatigue crack shown in Figs. 4 (a) and (b), respectively. In addition, it can be observed that there is a gap between the surfaces of the fatigue cracks. The width of the gap is less than 1 μm but the gap is locally wider than 1 μm. Fig. 4(c) shows a magnified view of the fatigue crack. The presence of the gap between the fatigue crack surfaces indicates that they are not in contact with each other under no loading. Therefore, the contact between the fatigue crack surfaces occurs when the amount of displacement of the fatigue crack at the time of shear stress loading is larger than the width of the gap. We investigated the contact between the fatigue crack surfaces and calculated the rate of decrease of K II . Consequently, for the fatigue crack shown in Fig. 4 (a), the rate of decrease of K II was 5.6% and 3.4% when the loaded shear stress was +150 MPa and − 150 MPa, respectively. For the fatigue crack shown in Fig. 4 (b), the rate of decrease of K II was 4.7% and 1.2% when the loaded shear stress was +150 MPa and − 150 MPa, respectively. Therefore, in the fatigue cracks shown Fig. 4 (a) and (b), the decrease in the Mode II stress intensity factor range owing to the contact between the fatigue crack surfaces was 4.5% and 3.0%, respectively, and it can be considered that there was no RISS in the fatigue crack propagation under Mode II loading on the textured material. As the gap is formed between the fatigue crack surfaces, it is considered that the RISS effects were very small because no contact between the fatigue crack surfaces occurred. In the case of fatigue cracks that propagate in the same direction as the pre-crack from its tip under Mode II loading such as the fatigue crack shape obtained in this study, fatigue crack propagation occurs owing to the dislocation emission from the Frank – Read source in a region near the crack tip rather than the crack tip [Hamada et al. (2018c)]. In the case where the dislocation is emitted from the Frank – Read source in a region near the crack tip, a persistent slip band (PSB) is formed near the fatigue crack tip in the same direction as the crack caused by the cyclic shear loading and voids are formed in the dipole dislocation wall of the PSB; these voids coalesce, resulting in a fatigue crack [Zhai et al. 1990 and 1995]. It is considered that gaps occurred between the fatigue crack surfaces in the fatigue crack shape obtained in this study owing to the coalescence of voids. In addition, as the width of the gap between the failure surfaces was almost uniform, it is believed that this gap was caused not by wear but by fatigue crack propagation. Fig. 5 shows the SEM image of the fatigue crack surfaces. We could not obtain a streak-pattern-like trace of the contact between the fatigue crack surfaces. From the analysis of the fatigue crack shape on the side surface of the specimen, it can be observed that RISS did not exist, but it is considered that the influence of contact between fatigue crack surfaces was small, even inside the specimen. Moreover, surface asperities between the failure surfaces were connected at a few locations, and striation was not confirmed. This is also believed to be due to the fatigue crack propagation owing to the coalescence of the voids.

(a) Fatigue crack on the right side of the pre-crack

(b) Fatigue crack on the left side of the pre-crack

(c) Magnified image of the fatigue crack

Fig. 4 SEM images of a fatigue crack after N = 2000 cycles