ICF13B

13th International Conference on Fracture June 16–21, 2013, Beijing, China -3- 25 50 (a) Type A: Rectangular FIB slit, =35m 8 a a a 35 (b) Type B: Semi-circular slit made by the electrical discharging, a = 74, 148 or 369 m, = 93.0, 186 or 463m d d l l 25 8 (c) Type C: Drilled hole having crack-like thin slits made by FIB, (d, l) = (100, 50) or (200, 25) m, =116or 196m 500 500 25 25 25 (d) Type D: Electrical discharged hole having crack-like thin slits made by FIB, =513m 8 Figure 3. Shapes and dimensions of artificial defects The effective area for the defect size area was estimated by considering a contour which envelopes the original irregular shape of the defect, as shown in Fig. 4 [8]. A newly proposed inclusion rating method [11] based on the statistics of extremes was used. 3. Results and discussion 3.1. Hydrogen content in specimen and the desorption behavior Figure 5 shows the hydrogen thermal desorption spectra for the H-precharged specimen [12]. Figure 6 displays the residual hydrogen content, CH, R, as a function of the total hold time after the charging, Δt, in laboratory air. The content decreased gradually with the hold time. After 100 hours, the content reduced to almost the same level of hydrogen content as in the non-charged specimen. From the reduction of hydrogen content shown in Fig. 6 [12], an apparent hydrogen diffusion coefficient at room temperature, D', was estimated by means of Demarez et al.’s solution [13]. The least-square fitting rendered the coefficient to be D' = 8.6 × 10−13 m2/s. In the H-precharged specimens, all the tensile tests were initiated within 2 hours after the charging, and they were finished within 2 minutes after the initiation of the test; i.e. around 4 mass ppm of hydrogen seemed to be present in the specimen during the tensile test. 3.2. Fracture data and fracture origins Figure 7 shows examples of the uniaxial tension stress-strain curves [12]. All the H-precharged specimens failed at a much lower stress as compared to the non-charged specimens irrespective of the presence or absence of the artificial defect, as shown in Fig. 7, i.e. the so-called hydrogen embrittlement was demonstrated in the H-precharged specimens. Moreover, the presence of an artificial 51 80.4 105.4 12.5 12.5 19 30 4.5 13 Figure 2. Shape and dimensions of test specimen in mm (Control volume: Vs = 477 mm3) (a) Non-metallic inclusion (b) Artificial defect Figure 4. Irregularly shaped defects and estimation method for effective area [8]

RkJQdWJsaXNoZXIy MjM0NDE=