Characterization and assessment of hydrogen leakage mechanisms in salt caverns

Salt caverns are widely regarded as a suitable option for the underground storage of hydrogen. However, an accurate assessment of the hydrogen leakage through the walls of salt caverns into the surrounding formations remains crucial. In this work, the flow of hydrogen into the surrounding formation is evaluated by assuming that salt rock consists of bundles of tortuous nano-capillary tubes. A formulation was then proposed to model the flow in linear and radial domains. The formulations are based on a newly proposed unified gas flow model that is valid for the entire range of Knudsen numbers and accounts for gas slippage, bulk diffusion, and Knudsen diffusion. A finite-difference approximation with an iterative procedure was then used to treat the nonlinearity and solve the presented formulations. The formulations were validated against the experimental data reported in the literature. The results obtained indicated that for hydrogen flow over a wide range of pore radii and operating pressures and temperatures, the slippage flow regime must be considered. In a salt cavern with relevant dimensions and operating conditions, the cumulative hydrogen leakage after 30 years of cyclic storage was only 0.36% of the maximum storage capacity. It was also noticed that most of the leaked hydrogen would flow back into the salt cavern at times when the pressure in the salt cavern is lower than the surrounding pressure, e.g. during production and subsequent idle times. At low storage pressure and very tight salt rock, diffusion was the most important mechanism for hydrogen transport. At a high pressure though, viscous flow became the predominant leakage mechanism. The presence of a thin interlayer such as mudstone, carbonate, and anhydrite in the body of the salt rock can have a significant impact on the amount of leakage. It appeared that although increasing the maximum operating pressure from 120 to 135 bar only led to an 11.9% increase in the maximum storage capacity, the hydrogen loss increased significantly from 0.007% at 120 bar to 0.36% at 135 bar. Furthermore, although the absolute leakage rate for natural gas storage was higher than that for hydrogen storage, the relative leakage rate in relation to the maximum salt cavern capacity was much lower. The leakage range was also lower for natural gas storage compared to hydrogen storage. The formulations presented and the results obtained in this study can help to have a better understanding of the salt caverns when it comes to large-scale hydrogen storage.