PSI - Issue 33

S. Schoenborn et al. / Procedia Structural Integrity 33 (2021) 757–764 S. Schoenborn, T. Melz, J. Baumgartner, C. Bleicher / Structural Integrity Procedia 00 (2019) 000–000

762

6

Fig. 5. S/N curve for 1.4521 in air (without results at 0.1 Hz) and in pressurized hydrogen for the sharply notched specimens K t ≈ 2.7.

The influence of a pressurized hydrogen atmosphere results in an earlier failure compared to the tests in air, especially for the tests in the LCF and HCF regime and lifetimes N f < 1∙10 6 cycles. In the VHCF regime, no difference in the number of cycles between air and pressurized hydrogen could be observed. The results of the mildly and the sharply notched specimens show, that the influence of compressed hydrogen causes a rotation of the S/N curve in the HCF regime. The knee point of the Woehler-line remains unchanged. However, with increasing notch factor the slope gets steeper. For the mildly notched specimen, this means that the slope of the Woehler-line changes from k = 10.7 in air to k = 12.6 in pressurized hydrogen atmosphere. This can be observed even more distinctly for the sharply notched specimens. Here, the slope of the SN curves determined in air changes from k = 8.6 to k = 12.9 under pressurized hydrogen. This leads to the fact that the influence of hydrogen increases with an increasing notch factor, which is associated with a reduction in lifetime and fatigue strength. The fatigue investigations on the unnotched and notched specimens indicated that an absorption of hydrogen into the material only occurs at higher (local) stress amplitudes or that only at these stress amplitudes the absorbed hydrogen changes the fatigue behavior. This is even more pronounced if the exposure time is increased by reducing the test frequency. This could be explained by the HELP theory. The results suggest that the plastic strain must be sufficiently high enough for the material to be susceptible to the hydrogen environment. This influence of hydrogen on the cyclic fatigue behavior could be explained by the decohesions respectively HEDE theory, after which the binding forces are lowered by the interaction of atomic hydrogen and superimposed high stresses. In summary, it can be said that several mechanisms usually act together for the damage under hydrogen. Therefore, to qualify the influence of the plastic strain component in fatigue tests in a pressurized hydrogen atmosphere more in detail, strain-controlled fatigue strength tests were carried out with unnotched specimens. Results in Fig. 6 show a significant influence on the fatigue behavior for tests carried out in pressurized hydrogen compared to air. With increasing total strain amplitude, the effect of hydrogen increases, resulting in lower fatigue life at the same strain amplitudes and a decrease of the cyclic ductility coefficient ε f '. One assumption is that with increasing proportions of plastic strain, dislocation movement is also increasing, which may lead to persistent slip bands formation that cause oxide layer breakdown on the surface and thus facilitate the absorption of hydrogen. An amplitude of ε a,t ≈ 0.8 % results in a factor of 26 in the reduction of lifetime under hydrogen compared to test results in air. However, this life-reducing influence of the hydrogen only becomes clear when the results are plotted in the form of the total strain ε a,t over the number of cycles N i . On the other hand, the stress-strain curves do not show any significant differences in the curves or the cyclic strain limits R’ p0.2 , Fig. 7. This shows a slight cyclical hardening both in air and in pressurized hydrogen in comparable orders of magnitude.