As global interest accelerates in large-scale geological hydrogen storage to buffer renewable energy fluctuations and decarbonize industrial sectors, new data reveals that the mineral composition of underground reservoirs may play a more critical role than previously assumed. A comprehensive density functional theory (DFT) study has mapped hydrogen interactions across a spectrum of subsurface minerals, revealing that the chemistry of rocks could significantly impact hydrogen retention, retrieval efficiency, and even reservoir safety.
The study, which modeled hydrogen adsorption behavior on a curated list of minerals commonly found in potential storage formations, identified three distinct types of Hâ‚‚-mineral interactions. These include weak physisorption via van der Waals (vdW) forces, moderate electron redistribution during adsorption, and strong chemisorption leading to potential hydrogen dissociation. Each category has material implications for hydrogen containment over long durations.
Reservoirs rich in quartz, kaolinite, and muscovite exhibited only weak physisorption characteristics, suggesting a minimal likelihood of hydrogen loss. This weak interaction, dominated by vdW forces, indicates that hydrogen can be stored in such formations with limited risk of chemical entrapment or degradation. Notably, quartz—a primary constituent in sandstone reservoirs—retains its reputation for chemical inertness, reinforcing its value as a host rock for storage.
In contrast, minerals such as calcite, edenite, augite, and fayalite displayed moderate adsorption behavior involving measurable electron redistribution. While not as reactive as chemisorbing minerals, these interactions could complicate long-term storage through partial hydrogen trapping or altered desorption dynamics. The implications are especially relevant for formations containing carbonates like calcite, which are prevalent in many limestone-rich geological candidates. Changes in charge distribution may impact retrieval rates, leading to inefficiencies during production cycles.
More concerning are the findings on minerals such as albite, anorthite, fluorite, and pyrite, where strong chemisorption and evidence of hydrogen dissociation were observed. These interactions imply irreversible bonding or reaction pathways that could result in permanent hydrogen loss or mineral corrosion. Pyrite, for instance, has already been cited in experimental literature for undergoing reduction to pyrrhotite under hydrogen-rich, alkaline conditions. While earlier empirical data yielded mixed results regarding pyrite’s reactivity, the atomistic simulations underscore its potential vulnerability under specific thermodynamic conditions.
This deeper understanding comes amid growing inconsistencies in experimental findings. Whereas some studies observed enhanced porosity and mineral dissolution under hydrogen exposure, others reported negligible changes. These discrepancies often stem from the complexity of natural rocks, where multiple minerals coexist and obscure individual contributions. The controlled DFT-based approach bypasses this limitation, offering a high-resolution view of how pure minerals interact with hydrogen at the molecular level.
The study also highlights feldspars as a mineral group warranting closer scrutiny. While feldspars are not dominant in most reservoir sandstones, some formations, such as arkosic sandstones in California’s San Joaquin Basin, contain significant feldspar content. Simulations showed that albite and anorthite, key feldspar end-members, are prone to chemisorption. Given that feldspar makes up about 50% of Earth’s crust, its role in hydrogen geochemistry could be more influential than previously accounted for—especially in storage or natural hydrogen generation contexts.
Technical modeling relied on the Vienna Ab Initio Simulation Package (VASP), with DFT-D2 corrections to account for van der Waals forces. While computationally intensive and limited to small atomic systems, the approach remains among the most accurate tools for probing surface interactions and was critical for differentiating physisorption from chemisorption pathways.
These findings arrive as stakeholders in hydrogen storage—ranging from regulators to project developers—face mounting pressure to identify and certify viable storage sites. While porosity, permeability, and caprock integrity are standard selection metrics, the results suggest that mineralogical composition must now be evaluated with comparable scrutiny. Failure to do so could result in suboptimal storage performance, hydrogen loss, or even unintended geochemical side effects such as souring or formation weakening.
As the industry begins to scale projects from pilot to commercial, integrating mineral reactivity data into site assessments and modeling tools will be essential. While the current study focused on dry hydrogen-mineral interactions, ongoing work will assess how the presence of water—ubiquitous in subsurface formations—modifies these behaviors, potentially exacerbating or mitigating reactivity.
The emerging consensus is clear: in geological hydrogen storage, chemistry matters as much as physics. The mineral fabric of the Earth is not passive. Its interactions with hydrogen must be quantified, modeled, and incorporated into risk management strategies if subsurface storage is to become a reliable pillar of the clean energy transition.
