PSI - Issue 13

Akira Maenosono et al. / Procedia Structural Integrity 13 (2018) 694–699 Akira Maenosono et al. / Structural Integrity Procedia 00 (2018) 000 – 000

696

3

Table 1. Chemical composition of the alloy used [mass%]

Material

Al

V

O

Fe

Ti

Ti-6Al-4V

6.29

4.35

0.155

0.225

Bal

2.2. Fatigue tests The specimen surface was etched by immersion in a chloride solution (1 vol% HF + 3 vol% HNO 3 + 96 vol% H 2 O) for approximately 3 s to observe the microstructure. Seven notches were introduced at the center of the prior β grain by a focused ion beam (FIB), with an acceleration voltage of 30 kV, and a beam current of 15 nA. The length, width, and depth of the FIB notch were 100, 7, and 33 μm, respectively. After the FIB machining, 5 μm was removed from the surface by mechanical polishing with colloidal silica. Fatigue tests were carried out at a stress amplitude σ a = 400 MPa, stress ratio R = 0, and frequency 1.3 × 10 -2 Hz using SEM. As a setup for observations of crack growth within each colony, a specimen with six notches was tested for up to 2.5 × 10 3 cycles. SEM images capturing fatigue cracks were taken at cycles 0, 1, 5, 10, 50, 100, 500, and later every 500 cycles. As another setup for observation of crack growth across a prior β grain boundary, a specimen with a single notch was tested for up to 2.8 × 10 3 cycles. In addition, at 2779 and 2788 cycles, in situ observations under one loading cycle were carried out to characterize fatigue crack growth behaviors when the crack passed through a grain boundary. In these observations, images were taken every 100 MPa during loading and unloading. After the fatigue test, crystallographic characterization around cracks was performed by means of EBSD. The specimen surface for the EBSD measurement was mechanically polished again until mirror finishing. The EBSD measurements were carried out at an acceleration voltage of 15 kV, a beam current of 76 nA, and a beam step of 2 μm . For the sake of simplicity, we here note the crack growth behavior until the cracks meet colony boundaries. We assume that the specimen is all α phase because of its volume ratio. We divided the crack propagation path into two parts, each FIB notch tips to crack tips, because each tip of FIB notch located other colonies. As reported in previous studies of Ti – 6Al – 4V alloys, the crack plane of the small fatigue cracks tends to be along basal plane (Bantounas et al., 2009; Bridier et al., 2008). Therefore, we calculated the fraction of the fatigue crack plane along the basal plane for each grain having a FIB notch. Then, the fraction of fatigue cracks along the basal plane was plotted against the angle between notch alignment and basal plane, as in Fig. 2. The fraction of the crack plane along the basal plane is defined as follows. The crack lengths parallel to the basal plane trace, L P , and vertical to the basal plane trace, L V , are 1 cos n i P i L L  = =  (1) 1 sin n i V i L L  = =  (2) where L i is the true crack length of number i, and φ is the angle between notch alignment and the crack growth direction. Then, the fraction of fatigue cracks along basal plane F is defined as 3. Results and discussion 3.1. Crystallographic feature of crack plane

F L =

P

(3)

L L +

P V

As seen in Fig. 2, the fatigue cracks tend to be along the basal plane. The crack growth behavior of the lowest F is zigzag crack propagation alternately along the basal and prismatic planes. As a minor portion, fatigue crack growth perpendicular to the loading direction was also observed with a low F . However, the F becomes high when the angle between notch alignment and basal plane trace is higher than 20°. In other words, when sufficient resolved shear stress acts on the basal plane, the crack tends to develop along the basal plane. This indicates that the fatigue crack propagates preferentially when mode II loading is applied at the crack tip. Therefore, in this research, we focus on mode II fatigue crack growth, as presented in the next section.