PSI - Issue 77

Behzad Vasheghani Farahani et al. / Procedia Structural Integrity 77 (2026) 424–431 Behzad V. Farahani et al./ Structural Integrity Procedia 00 (2026) 000–000

426

3

in ABAQUS© under both air and hydrogen exposure. It considers both a fully homogeneous BM configuration and a heterogeneous configuration reflecting the actual WM and HAZ properties. 2. Material and Computational Framework Fig. 1-a) shows a GW section from the studied X70 pipeline, from which a notched round bar (NRB) sample with a 6 mm notch radius (R6) was extracted longitudinally with the notch centred in the WM. This geometry serves as the benchmark in this study.

a) b) Fig. 1: Macrograph of a girth weld section from the studied pipeline: a) NRB R6 sample extracted with the notch centered in the weld; fusion lines are highlighted in blue, and b) Vickers hardness distribution , HV 10 with a proof load of 10 kgf . The Gurson-based damage model employed here extends the classical framework to represent ductile fracture mechanisms in hydrogen-exposed steels. Ductile fracture in polycrystalline metals proceeds through microvoid nucleation, growth, and coalescence. It should be noted that void nucleation and growth reduce load capacity, while coalescence triggers localized failure (Noell et al., 2023). The original Gurson model (Gurson, 1977), refined by Tvergaard and Needleman (Tvergaard & Needleman, 1984), describes void evolution but cannot predict the onset of localized failure. To overcome this, Thomason’s plastic limit load model (Thomason, 1985) was integrated, improving coalescence prediction. This enhancement, termed the Complete Gurson Model (CGM) proposed by Zhang (Z. L. Zhang et al., 2000), offers comprehensive fracture representation across stress triaxialities and hardening conditions. Coupling the Gurson framework with hydrogen diffusion and trapping effects further captures plasticity-dominated failure. Hydrogen effects were incorporated by accelerating void-driven ductile damage evolution. A refined method for void nucleation parameters improves predictions of crack initiation and growth in hydrogen-exposed steels. Nevertheless, the numerical model of hydrogen-induced degradation follows the approach developed and validated at the Laboratory Soete, Ghent University (Depraetere et al., 2021); readers are referred there for additional details. 3. Element-Specific Material Characterization for Numerical Modeling In the studied NRB-R6 sample, the corresponding microstructural heterogeneity across the WM and HAZ necessitates accounting for a spatially varying material response in the numerical analysis. To address this, an element specific property assignment strategy is employed: constitutive properties for each mesh element are derived and converted from local hardness measurements obtained from HV maps of the pipeline girth weld, c.f. Fig. 1-b). 3.1. Conversion of hardness values Numerous studies have focused on deriving tensile mechanical properties from HV measurements. Amongst others, Hertelé et al. (Hertelé et al., 2015) introduced a methodology to extract nonlinear mechanical properties, such as yield stress , , ultimate tensile strength, , hardening exponent, , and the flow stress-strain curve, directly from a hardness value, . The following equations are then introduced:

Information classification: Internal