A multiphysics phase-field–diffusion framework for hydrogen-assisted fatigue crack initiation and growth in welded pipeline steels

The transport of hydrogen through existing steel pipeline networks is pivotal for the large-scale deployment of hydrogen energy. However, the vulnerability of pipeline steels to hydrogen embrittlement and cyclic degradation poses significant integrity challenges. This study introduces a comprehensive, fully coupled phase-field framework designed to predict hydrogen-assisted fatigue crack initiation and growth in welded pipeline steels, specifically utilizing welded X65 pipeline steel as the representative material system. Our approach takes into account the heterogeneous structure of weldments, encompassing the base metal (BM), heat-affected zone (HAZ), and weld metal (WM). The framework integrates a variational phase-field model to describe brittle fracture alongside stress-assisted hydrogen diffusion, incorporating a hydrogen-dependent degradation mechanism for fracture toughness via an adsorption-decohesion process. Additionally, it addresses fatigue-induced reductions in toughness through an energetic accumulation law. To accurately represent the complexity of weld heterogeneity, effective material properties are spatially varied across the BM, HAZ, and WM regions. To enhance computational efficiency for high-cycle simulations, we employ an envelope-load strategy that captures cyclic loading effects at the level of fatigue-driving energetics, without the need to explicitly resolve each load reversal. The formulation is executed within a finite-element multiphysics framework, utilizing adaptive mesh refinement to effectively manage the steep spatial gradients of damage and hydrogen concentration near evolving crack fronts. Our simulations reveal that the weld structure exhibits heightened sensitivity to hydrogen uptake and cyclic degradation, leading to accelerated crack growth and the formation of preferential crack paths, which align with observed experimental trends. Overall, this proposed framework offers a robust computational methodology for analyzing the intricate interactions among weld heterogeneity, hydrogen embrittlement, and fatigue damage in welded pipeline steels, thereby advancing our understanding of pipeline integrity in hydrogen transport applications.