13th International Conference on Fracture June 16–21, 2013, Beijing, China -3- Fig. 1. Shape and size (mm) of specimens. Fig. 2. Locations for EBSD and XRD analyses. distance of 6 ~ 7 mm from the center, and the cut surfaces were polished in the same way and then subjected to EBSD and XRD analyses. In the XRD analyses of the longitudinal cross-sections, the projection of X-ray beam on the cross-section was aligned in directions along x and y axes to examine the difference in microstructure in these directions. The same procedure was applied to the microstructural analysis of longitudinal cross-sections at the center of disc. 3. Results 3.1 Effect of humidity on fatigue strength Fig. 3 shows a comparison of SN curves obtained for A and B specimens which are subjected to RB tests with 50 Hz in humid atmospheres ranging 0% RH (N2 gas) to 100% RH (distilled water). Symbols in the figure are experimental data of the stress amplitude ( σa) and the number of cycles to failure (Nf), and the curves can be expressed by σa = σao + k{(Nf/Nfo) −m − 1} (1) where σao and k are the strength coefficients, m is a numerical constant and Nfo is a reference number of cycles to failure and set to be 107 cycles in the present analysis. The as-received A specimens in N2 gas show the largest fatigue strength among all of the testing conditions, which exhibits larger strength by 30 ~ 60 MPa than that of B specimens fatigue-tested in N2 gas. This is consistent with the fact that the yield and tensile strengths of A specimens are larger than B specimens. Humidity does not change the fracture strength of B specimens significantly until RH is increased to 50 %, and a marked reduction of fatigue strength occurs at the humidity larger than 50 %. Previous studies showed that this transition appeared distinctly at the humidity of about 60 %. In contrast to B specimens, the humidity-induced deterioration in A specimens is prominent from the RH of 25 %, and the fatigue strength of A specimens at the humidity from 25% RH to 85% RH is lower than that of B specimens for identical RH. These results indicate that A specimens with larger yield strength is more susceptible to the environmentally assisted fatigue cracking. Fig. 4 shows the change of the fatigue strength ( σao) for Nf = 10 7 as a function of RH. The fatigue strength of A specimens RB-tested at 50 Hz decreases monotonically with increasing RH. The fatigue strength of B specimens tested in the same condition, however, is insensitive to the humidity up to 50 % and shows a rapid decrease at higher humidity. Nevertheless, the fatigue strength of B specimens RB-tested at 6 Hz is kept to be constant up to 85% RH. The application of US loading with 20 kHz to A specimens results in a slight degradation from N2 gas to 25% RH, a high-stress-level plateau from 25% to 50% RH, a drastic degradation from 50% to 75% RH, and extrusion direction O x y z Electron beam for EBSD Electron beam for EBSD X-ray for XRD X-ray for XRD 6 ~ 7 mm 27.1 φ4 φ12 69.1 (a) RB-tested specimen. (b) US-tested specimen. R10 R20 φ8 φ5.5 φ5 20 60
RkJQdWJsaXNoZXIy MjM0NDE=