Experimental evaluation of hydrogen injection and withdrawal performance in residual oil-saturated carbonate core plugs

Storage of hydrogen in depleted reservoirs is increasingly considered a viable option for large-scale subsurface energy storage. However, most H₂ hydrodynamic studies remain limited to simplified H₂-brine systems and do not capture the residual-oil conditions expected in depleted oil reservoirs. In addition, hydrogen withdrawal is often evaluated by brine displacement, whereas field production is mainly driven by pressure drawdown. In this work, high-pressure core flooding experiments were conducted on carbonate core samples conditioned with residual oil to evaluate hydrogen storage and withdrawal efficiency. Three gas systems were investigated: H₂, 60%H2–40%CH4, and 60%H2–40%N2 mixtures at 1200 psi and 70 °C. The withdrawal stage was performed via controlled pressure drawdown using a drum-type gas meter, followed by porosity measurements and effluent ion analysis to assess potential fluid-rock interactions. Results show that 60%H₂–40%CH₄ achieved the highest initial gas saturation (30.4%), followed by 60%H₂–40%N₂ (29.6%) and pure H₂ (26.6%), indicating improved drainage efficiency with cushion gas injection. In contrast, recovery was highest for pure H₂ (85%), compared with 60%H₂–40%N₂ (75%) and 60%H₂–40%CH₄ (67%). Residual gas saturation ranged from 3% to 10%, depending on the gas system. Post-flood porosity measurements showed no measurable change in pore volume, while effluent ion analysis indicated a slight increase in Ca concentration without evidence of bulk matrix alteration. These findings provide practical guidance for selecting cushion gas and designing withdrawal strategies in depleted carbonate oil reservoirs, where maximizing storage capacity must be balanced against hydrogen recoverability.