Cyclic Flow Characteristics of Sandstones during Geological Hydrogen Storage in Saline Aquifers

Abstract

Geological hydrogen storage in saline aquifers has emerged as a promising solution to mitigate climate change and aid the transition to a low-carbon society. While this technology offers significant benefits, it also uncovers research gaps that need to be addressed. Economically, it is essential to utilize storage sites over extended periods, allowing for the injection of hydrogen into these porous rocks during low-demand periods and its extraction during high demand. However, this process is not uniform: the amount of hydrogen produced in the high-demand season differs from the quantity initially injected in the low-demand season, primarily due to residual gas trapping. Moreover, with repeated cycles of injection and withdrawal, the risk of enhanced residual trapping increases, due to the hysteresis effects.   Understanding the impact of multiple cycles on residual trapping and the influence of rock type on this process is vital for site selection in hydrogen storage in saline aquifers. More importantly, to model this cyclic process accurately across different rock types, experimental data for each specific rock type are required. To this end, various sandstone rock types were selected for cyclic flow core tests at the School of Geosciences, University of Edinburgh, using our state-of-the-art in-house core flooding system. These tests, conducted under high-pressure, high-temperature conditions with a simulated brine composition to better reflect real-case scenarios, aimed to evaluate the impact of rock type on fluid flow characteristics and residual trapping during hydrogen injection and withdrawal. Core samples, sourced from reservoir rocks and cleaned with solvents, were analysed for porosity and permeability, aligning with specific reservoir log data. The selection process was tailored to encompass a broad range of rock types. Influential parameters were identified and isolated using an integrated, in-house methodology, ensuring the only variable was rock type. In addition to core tests, Mercury Injection Capillary Pressure (MICP) tests, tracer tests on each core sample, and Inductively Coupled Plasma (ICP) analysis of effluent samples were conducted a comprehensive interpretation of the results.   Our observations reveal that rock type significantly influences residual trapping and fluid flow through porous media. Different rock types demonstrated distinct behaviours in terms of residual trapping and relative permeability endpoints. Concerning the impact of multiple cycles on residual trapping, the heterogeneity of the core sample proved to be a critical factor; in homogeneous samples, residual trapping remained constant after the initial cycle. In contrast, for heterogeneous samples, the degree of heterogeneity led to an increase in residual trapping over successive cycles. This phenomenon was primarily attributed to hysteresis effects, as no significant geochemical reactions were detected in the tests.