New insights into hydrogen-assisted intergranular cracking in nickel

TL;DR

This work addresses how grain boundary character governs hydrogen-assisted intergranular cracking in pure nickel under 4–14 wppm hydrogen. Using controlled electrochemical charging, TDS, tensile testing, and SEM-EBSD, it shows that $\Sigma$-3 GBs are consistently most resistant to hydrogen-induced cracking, and that embrittlement progresses with hydrogen concentration mainly via decohesion rather than plasticity. The study reveals surface charging raises surface cracking without changing GB susceptibility and highlights a decohesion-dominated mechanism modulated by GB energy and hydrogen trapping across $\Sigma$ classes. These findings offer a physics-based basis for grain boundary engineering in Ni to mitigate hydrogen embrittlement, and clarify differences between Ni and Ni alloys where plasticity-assisted cracking can prevail.

Abstract

We characterize the grain boundary (GB) susceptibility to hydrogen-assisted intergranular cracking in pure nickel as a function of coincident site lattice value ($Σ$-n), over a wide range of hydrogen concentrations (4 to 14 wppm). Cracks on the surface and within the bulk material were identified across the entire gauge region of the specimens. The susceptibility of GBs to crack initiation and propagation was evaluated by separating cracks containing single GB or multiple GBs. A larger loss in fracture strain, a smaller reduction in area, and an increase in the percentage of intergranular fracture indicated a higher degree of embrittlement at elevated hydrogen concentrations. The number of cracks was significantly higher on the surface than in the bulk for the most severe hydrogen charging conditions ($\geq$ 8 wppm), while a similar number was observed for lower concentrations. The propensity for hydrogen-assisted intergranular cracking at different types of GBs on the surface and in the bulk material was consistent, indicating that while cathodic charging can promote surface cracks, it does not significantly impact the GBs relative susceptibility. The $Σ$-3 boundaries were the most resistant to cracking, as evidenced by the considerably lower fraction of these GBs exhibiting intergranular cracking at all hydrogen concentrations considered. This contrasts literature findings for Ni alloys and can be explained by the segregation energies and reductions in the cohesive strength with hydrogen, with less favorable trapping at the $Σ$-3 boundaries. No evidence of plasticity-mediated cracking initiation was observed.

Figures (12)

• Figure 1: Baseline microstructure: (a) IPF map and (b) three GB categories showing $\Sigma$-3 in red, other $\Sigma$-low in blue and general in black.
• Figure 2: Surface and bulk intergranular cracking analysis: (a) material removal to expose the bulk, (b) IPF maps of surface and bulk, (c) IPF map of bulk material with a crack, and (d) GB characterization around the crack (depicted by a zigzag line).
• Figure 3: The longitudinal surface of post-deformation specimens with highlighted cracks: (a) single-GB and (b) multi-GB surface cracks; (c) single-GB and (d) multi-GB bulk cracks after removing half of the material thickness.
• Figure 4: Engineering stress-strain curves (left) and fracture surfaces (right) of uncharged specimen and specimens with hydrogen concentration of 3.5, 3.9, 8.2 and 14.5 parts per million in weight (wppm).
• Figure 5: Quantifying HE in pure Ni: (a) Reduction in area, $RA/RA_{\text{Uncharged}}$, and intergranular area, $A_{\text{IG}}$, as functions of hydrogen concentration, indicating increased embrittlement with higher hydrogen concentration. Fracture surfaces for (b) 3.5 wppm with a mix of microvoids and intergranular fracture, whereas for (c) 14.5 wppm, a larger fraction of intergranular fracture corresponds to a higher embrittlement level.