PSI - Issue 77

A. Hell et al. / Procedia Structural Integrity 77 (2026) 41–48 Author name / Structural Integrity Procedia 00 (2026) 000–000

45 5

A steep decline of c H is visible with increasing penetration depth in the charging profiles of compact tension samples. The red lines in Fig. 2a and 2b indicate the displayed cross sections in the 1D numerical analysis. Higher concentrations are situated in the path of the fatigue crack. At crack lengths > 4.5 mm, which are reached in J - ∆ a curve testing, lower c H values imply a need for sufficient diffusion time and enrichment at the crack tip during mechanical testing. 3.3. J- ∆ a-curve measurement The crack growth rates obtained during 4.5 mm fatigue precrack generation are shown in Fig. 3a. Fatigue crack growth (FCG) is unaffected from the presence of internal hydrogen after 96 hours charging in the specimens tested with 10 Hz and 19 kN maximum force, probably attributed to the high precracking frequency. In contrast, slightly shifted da/dN -curves are obtained for H-charged samples at testing conditions of 5 Hz and 27 kN. The higher stress intensities and the lower fatigue frequency leads to an acceleration of crack growth in hydrogenated P355NH. With the 5 Hz – 27 kN approach, fast precrack growth occurs even below 10000 cycles in all tested samples, which might influence J - ∆ a -results as it possibly indicates large strains at the crack tips. For this instance the 10 Hz – 19 kN methodology was applied for all J - ∆ a -measurements in this work to minimize the risk of influencing the quasi-static stable crack growth resistance evaluation while still being able to perform precracking in a comparably short period of time, minimizing H-desorption. A sufficient precrack size was now reached at 20000 – 30000 cycles.

Fig. 3. (a) FCG-rates measured with respect to precracking parameters; (b) J - ∆ a -curves obtained for the CT50-specimen geometry.

Fig. 3b shows the J - ∆ a -curves obtained for uncharged and 96 hours charged P355NH. A progressive reduction of slope in the curves of charged steel occurs, indicating a decrease in ductile, stable crack growth resistance. The cracks in hydrogenated material extend further than in uncharged pressure vessel steel. A discussion regarding the H concentration profile follows in section 3.4. However, no unstable crack growth was observed in the tests and the material remains predominately ductile. For this instance, a hydrogen-related facilitation of ductile crack extension is more likely than strongly embrittling mechanisms like intergranular cleavage. An influence of hydrogen on void initation and coalescence is expected, leading to increased stable crack growth. The results of a J 0.2 -evaluation for the CT50-specimen geometry using blunting line and 0.2 mm offset line criterions based on ASTM E1820 are displayed in Table 1 and indicate promoted ductile crack initiation.