Computational predictions of weld structural integrity in hydrogen transport pipelines
T. K. Mandal, J. Parker, M. Gagliano, E. Martínez-Pañeda
TL;DR
The paper addresses the risk of hydrogen embrittlement in seam welds of gas pipelines by coupling weld-process modelling with deformation-diffusion-fracture simulations to predict critical fracture pressures. It develops a multi-physics phase-field framework implemented in Abaqus to simulate diffusion, elastic-plastic deformation, and fracture with hydrogen degradation. Validations against crack-growth resistance curves for X52 and X80 indicate the approach can reproduce hydrogen-assisted fracture behavior, and the framework is applied to two seam-weld geometries to map safe hydrogen pressures. Results show that welding residual stresses, microstructural heterogeneity, porosity, and pre-existing defects can reduce the maximum admissible pressure to as low as 15 MPa, providing a mechanistic basis for assessing the hydrogen viability of existing pipelines.
Abstract
We combine welding process modelling with deformation-diffusion-fracture (embrittlement) simulations to predict failures in hydrogen transport pipelines. The focus is on the structural integrity of seam welds, as these are often the locations most susceptible to damage in gas transport infrastructure. Finite element analyses are conducted to showcase the ability of the model to predict cracking in pipeline steels exposed to hydrogen-containing environments. The validated model is then employed to quantify critical H$_2$ fracture pressures. The coupled, phase field-based simulations conducted provide insight into the role of existing defects, microstructural heterogeneity, and residual stresses. We find that under a combination of deleterious yet realistic conditions, the critical pressure at which fracture takes place can be as low as 15 MPa. These results bring new mechanistic insight into the viability of using the existing natural gas pipeline network to transport hydrogen, and the computational framework presented enables mapping the conditions under which this can be achieved safely.
