Objective CO₂ breakthrough pressure, a core parameter for characterizing the sealing performance of saline aquifers as cap rocks, holds important scientific significance and application value for ensuring the safety of CO₂ geological storage engineering and assessing its storage potential. However, there is a lack of systematic studies on the variation patterns and mechanisms of breakthrough pressure under different occurrence phases of CO₂ (supercritical, liquid, and gaseous) currently. Accordingly, universal and reliable conclusions are yet to be reached.

Methods This study conducted experiments on a natural low-permeability unsaturated sandstone core. Using the step-by-step method, 13 groups of breakthrough pressure experiments were carried out under different CO₂ phases to deeply explore the variation patterns of CO₂ breakthrough pressure under these phases.

Results and Conclusions  The results indicate that the CO₂ phase is not directly correlated with the magnitude of the CO₂ breakthrough pressure. Under the experimental conditions of this study, the CO₂ breakthrough pressure at any phase transition point was higher than that under a single phase. Under the condition of a single CO₂ phase, the CO₂ breakthrough pressure decreased with an increase in pressure, exhibiting an exponential relationship with the latter. Pressure produced the greatest impact on the breakthrough pressure of supercritical CO₂ but posed the smallest impact on that of gaseous CO₂. The influence of pressure on CO₂ breakthrough pressure is caused by the synergistic effect of the interfacial tension of the CO₂-water system and the wettability (the contact angle) of the CO₂-water-rock system, as well as the two-phase density difference and viscosity ratio. Sensitivity analysis results show that the viscosity ratio was a significant factor affecting the CO₂ breakthrough pressure, with a total contribution rate of up to 39.9%. Analysis of the lg C_a-lg M displacement stability diagram further confirmed that viscosity acted as a dominant factor affecting the experimental results of this study. Comparison with the formation pressure data and practical application to engineering sites in basins reveal that the experimental conditions of this study are consistent with actual engineering. The results of this study serve as an important reference for accurately evaluating the storage capacity of saline aquifers and guiding target siting for CO₂ geologic geological storage.