Assessing the effects of loading rate on fracture toughness of AISI 1020 and API 5L X80 steels with hydrogen charging: experimental and numeric simulation study

Abstract

This study focuses on the effect of hydrogen on fracture toughness of AISI 1020 and API 5L X80 steels using experimental measurement technique and numeric simulation. Hydrogen was introduced into Single Edge Notch Bend (SENB) specimens through in-situ electrochemical charging techniques. The study varied the loading rates (K̇) and observed a significant reduction in the fracture toughness of both types of steels with hydrogen presence, worsening as loading rates decreased. The findings illustrated that the current standards of hydrogen compatibility test, which specifies a loading rate that ranges from $0.1MPa\sqrt{m}/min$ to $1MPa\sqrt{m}/min$ in toughness tests, may produce non-conservative results by not fully capturing the degradation at lower loading rates (K̇ <＜$0.1MPa\sqrt{m}/min$). The low-strength steel (AISI 1020) showed a fracture mode transition from the Hydrogen-Enhanced Localized Plasticity (HELP) to the Hydrogen-Enhanced Decohesion (HEDE) as the K̇ decreases. Nevertheless, the crack re-initiation sites are always found at the tips of pre-cracks, which coincide with the areas of a high local hydrogen concentration and a high plastic strain. In contrast, the crack initiation sites in X80 steel tend to shift from the regions with high-strain (i.e. near the crack front) to areas with high hydrostatic stress (some distance away from the crack front) as the K̇ decreases, although it is still a predominantly quasi-cleavage (QC) fracture mode.

Finite element analysis further revealed that the diffusion and trapping of hydrogen atoms are significantly influenced by the gradient of hydrostatic stress and increment of plastic strain in the crack front region. In AISI 1020 steel, the concentration of trapped hydrogen significantly exceeds that of diffusible lattice interstitial hydrogen; In this case, the trapped hydrogen predominately dictates the distribution profile of total hydrogen at all K̇ conditions, whereas in X80 steel, the relative dominance of diffusible and trapped hydrogen depends on the loading rate: at lower K̇, diffusible hydrogen concentration at lattice sites exceeds that of the trapped hydrogen and there is a dynamic equilibrium relationship between the hydrogen induced by hydrostatic stress and the hydrogen trapped by plastic strain; at higher K̇, the hydrogen in trap sites dominate the hydrogen distribution. The differences in hydrogen concentration distribution are linked with different crack nucleation behaviours of the two steels under different loading rates, as well corroborated by fractographic observation of the actual fracture surface.