Hydrogen interaction in bulk fluids for geological storage application using NMR

Hydrogen geostorage is a crucial component of decarbonization, enabling large-scale energy storage and supporting the transition to a low-carbon economy. By allowing long-term hydrogen storage in subsurface formations such as depleted oil and gas reservoirs, geostorage enhances energy security and stabilizes energy supply. This study serves as a preliminary step before investigating H₂ interactions in saturated porous rocks, focusing on hydrogen behavior in reservoir fluids using nuclear magnetic resonance (NMR). The tested fluids include water, dodecane oil (light oil), dead oil, and ozokerite wax (heavy oil) under pressures up to 1800 psi. Additionally, deuterated water and perfluorinated HT-230 were used as control fluids due to their negligible NMR signals. HT-230, commonly used as a confining fluid in core plug measurements, also provides a baseline for comparison. T₂ relaxation times served as a proxy for distinguishing free hydrogen from dissolved hydrogen in bulk liquid based on molecular interactions. Since it is sensitive to hydrogen protons in fluids, it was used to assess changes in bulk properties such as viscosity and density. The results indicate that hydrogen predominantly remains in the free phase when interacting with water, as evidenced by fast relaxation times (1–20 ms) and no observed changes in T₂ with pressure, confirming limited dissolution. Similarly, hydrocarbons—including dodecane, dead oil, and wax—showed no evidence of hydrogen dissolution under NMR, as only free-phase hydrogen signals (1–20 ms) were detected. However, visual observations of gas bubbles in dead oil suggest physical hydrogen trapping rather than true molecular dissolution, indicating hydrogen retention in a discrete gas phase without full integration into the liquid phase. In perfluorinated HT-230, an intermediate T₂ relaxation signal (100–300 ms) suggests possible hydrogen dissolution, with an estimated volume of 0.80–1.20 ± 0.02 cc at room temperature. This finding indicates that while HT-230 is generally inert, some level of hydrogen interaction may occur. Therefore, caution is advised when using HT-230 as a confining fluid in core plug tests under stress, or this signal should be excluded from analysis. Although this study was conducted at relatively low temperatures and over short experimental durations, hydrogen can be physically trapped in dead oil within our test conditions. These results provide a baseline understanding of H₂ interactions in bulk fluids, informing future core plug measurements of hydrogen retention, diffusion, and mobility in depleted oil and gas reservoirs.