Objective The large-scale deployment of Carbon Capture, Utilization, and Storage (CCUS) technology necessitates the advancement and engineering application of supercritical CO₂ pipeline transport. However, conventional crack arrest models fail to accurately predict the crack arrest toughness of these pipelines. Furthermore, existing stress correction formulas yield highly inconsistent results that deviate significantly from current CO₂ pipeline design standards. Consequently, the lack of an established crack arrest control system presents a major barrier to widespread CCUS adoption, highlighting an urgent need for a more cost-effective, engineering-applicable crack arrest evaluation model.

Methods To address the above issue, full-scale burst test data of CO₂ pipelines publicly available worldwide were adopted in this study, covering different pipe diameters, wall thicknesses, steel grades, and test conditions. Existing resistance-driving force curves were modified via the crack arrest toughness parameter correction method, based on the crack arrest criteria of dynamic fracture mechanics and relevant CO₂ pipeline design standards. The crack arrest evaluation performances of different correction coefficients were compared, and the optimal correction coefficient k was obtained by fitting the measured data. Meanwhile, a novel crack arrest evaluation model for supercritical CO₂ pipelines was established by replacing the crack tip pressure p_c with the saturation pressure p_s.

Results The research results indicated that existing crack arrest evaluation models were not fully applicable to CO₂ pipelines, as traditional crack arrest stress formulas led to large calculation errors and a remarkable overestimation of pipeline crack arrest capacity. When the fitted correction coefficient of the crack arrest toughness parameter (k) was set to 3.09, the proposed model demonstrated high stability and optimal crack arrest evaluation performance within the sample range. This modified model eliminated the non-conservative defects of traditional prediction models and broadened the application scope of crack arrest evaluation diagrams. Additionally, a crack arrest evaluation scheme was formulated for the design of Pipeline L22 in the Phase II CCUS Project of Yanchang Oilfield. Corresponding crack arrest control recommendations were put forward, including increasing wall thickness to enhance material crack arrest toughness and installing external crack arresters.

Conclusion The crack arrest evaluation model developed in this study demonstrates high accuracy and engineering applicability. To accommodate variations in pipeline material and diameter resulting from large-scale CCUS deployment, future research should utilize finite element simulation technology with two-way fluid-structure interaction in conjunction with existing full-scale burst test data.