PSI - Issue 59

Jesús Toribio et al. / Procedia Structural Integrity 59 (2024) 90–97 Jesús Toribio / Procedia Structural Integrity 00 ( 2024) 000 – 000

92

3

consisting of hydrogen-assisted micro-damage (HAMD) in the form of tearing topography surface (TTS), and finally the cleavage-like surface corresponding to final fracture in an unstable manner. This cleavage mode is the fracture pattern obtained in air when the same pre-cracked geometries of the steel under analysis are tested.

Fig. 1. Depth x TTS of he TTS zone.

Considering the TTS region as the area damaged by hydrogen, a damage depth a H may be defined: a H = a O + x TTS (2) where a O represents the depth of the pre-crack (end of fatigue pre-cracking and beginning of the HE test), while x TTS is the depth of the TTS zone, measured from the pre-crack border in a direction perpendicular to the crack line, as sketched in Fig. 1. 3. Macroscopic crack model The macroscopic crack model is based on considering the characteristic TTS area as a macroscopic crack that extends the original fatigue pre-crack (see Fig. 1) and involves linear elastic fracture mechanics (LEFM) principles, since it has been proved that LEFM is applicable in these pre-cracked samples of high-strength steel and the plane strain condition is achieved at the majority of points of the crack. On the basis of these considerations, the SIF for the total crack (fatigue pre-crack plus TTS extension) is defined by the following expression: where Y H is the dimensionless SIF for the geometry under consideration (fatigue pre-crack plus TTS extension), a H is defined in eq. (2), and  H is the remote stress applied on the cylinder (far from the crack). Considering the critical situation, i.e. , the fracture instant in hydrogen environment, K H represents the critical SIF in the hydrogen external environment and  H is the externally applied stress at the critical instant of failure in hydrogen atmosphere, given by  H = 4F H /  D 2 (4) where D is the sample diameter (12 mm) and F H the critical fracture load in the hydrogen environment. In this article, it is assumed that, according to previous studies (Toribio et al., 1992) when the maximum load in the hydrogen embrittlement test ( F H ) is achieved, the crack length is a H = a O + x TTS . This is a reasonable assumption, since the TTS is a slow crack growth topography that allows the load to increase, whereas when the cleavage fracture takes place (with a quite high crack growth rate), the fracture process develops in unstable manner and the external load suddenly drops to zero. This reasoning is consistent with the recording of load vs. time during the test, in which no decrease in load was observed until the end of the test. For the case of fracture in air environment (reference situation), the SIF is K H = Y H  H H a  (3)