PSI - Issue 54

Margo Cauwels et al. / Procedia Structural Integrity 54 (2024) 233–240 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

237

5

3.3. Fracture surfaces Generally, the fracture appearance of a SENT specimen is characterized by several zones. First, the machined notch and fatigue pre-crack, then the blunting step (also referred to as the stretch zone), followed by the stable crack growth region. The final region is the post-mortem brittle fracture zone. Fig. 5 gives the overview of the fracture surfaces for all three conditions, with the end of the stable crack growth region of interest delineated by the dashed white line.

Fig. 5. Fracture surface appearance of an (a) uncharged reference specimen, (b) H charged, ex-situ tested specimen and (c) H charged, in-situ tested specimen. The end of the stable crack growth region is delineated by the dashed white line, the fatigue pre-crack is indicated by a blue double-headed arrow. Quasi-cleavage in (b) is delineated by the dashed yellow line. A white arrow indicates a split in the crack growth region in (c). Dash-dot white line marks the branching crack front in (c). A typical example of the stretch zone on the fracture surface is shown in Fig. 6 (a) for an uncharged, air-tested specimen. For the ex-situ sample, instead of this stretch zone, the end of the fatigue pre-crack is characterized by a crack perpendicular to the crack growth direction and parallel to the tensile direction. As can be seen in Fig. 6 (b), the split clearly marks the boundary between the fatigue pre-crack (below the split) and the dimpled ductile tear region above it. This coincides with the direction of the banded microstructure and the hard bands, indicating that the crack branching shown in Fig. 4 (b) is due to cracking or delamination along the banded microstructure. Given the orientation of the test specimen, this type or split is called a ‘crack arrester’ type. Along the edge of the split quasi cleavage facets can also be found, as shown in Fig. 6 (c) in the encircled area. The limited effect of hydrogen on the fracture toughness value for this test seen in the mechanical data may then be partially attributed to this split causing the initiating crack to blunt before growing, even though the fracture micromechanism is no longer fully ductile. Furthermore, the appearance of splits will affect the measurement of the crack tip opening displacement (CTOD), since when the crack grows beyond the crack arrester split, the crack tip will still appear more blunted than is the case. This is also illustrated by comparing Fig. 4 (a) and (b). For the in-situ sample, there is also no clearly identifiable stretch zone on the fracture surface. In Fig. 6 (d), the transition from the fatigue pre-crack, characterized by striations, to the ductile tear zone, in this case characterized by quasi-cleavage, can be observed, while there is no clear evidence of a crack blunting stage on the fracture surface. The lack of crack blunting stage is supported by camera images of the growing crack taken during the test. For the ex-situ test, since the specimen is exposed to air for a relatively long time after pre-charging, it is likely that diffusible hydrogen near the specimen edge, as well as near the fatigue pre-crack tip will have diminished significantly, since hydrogen diffusion in this type of ferritic-pearlitic steel is relatively high. Still, the hydrogen concentration was sufficient to promote splitting or crack growth along the L direction. For in-situ specimens, the hydrogen content at the crack tip will not only be higher at the start of the test, since there is pre-charging immediately prior to the application of load, but hydrogen can continually enter during the SENT test itself. In this case, the higher hydrogen