Store hydrogen in natural gas deposits
Can hydrogen be stored for longer in the underground of natural gas fields? Researchers evaluated as a part of the project H2STORE geohydraulic, mineralogical, geochemical and biogenic interactions in underground storage of hydrogen in depleted natural gas deposits.
|Project status||Project completed|
|Project duration||August 2012 until December 2015|
The storage of electrical power transformed into hydrogen within the pore space of sandstones of depleted gas reservoirs might offer an option for a long-term storage of energy in large volumes - one basic requirement in implementing the demanded energy change of the German government. The underground storage of hydrogen is (well) established in salt caverns. However such caverns have a limited storage capacity over only short time spans (~ weeks) and the availability of appropriate salt deposits is restricted. In contrast (depleted) gas reservoirs are widespread and have the potential to take large volumes of hydrogen for several months. Because the impact of (injected) hydrogen on rock composition of sandstone reservoirs was almost unknown, H2STORE investigated such potential mineralogical, (geo- and hydro-) chemical, petrophysical and microbiological reactions and their relevance to storage capacity and reservoir sealing.
Use sandstone pores
Salt caverns have only a limited storage capacity and the availability of appropriate salt deposits is restricted. The storage of hydrogen in the pore space of sandstones of depleted gas reservoirs might be another option. These reservoirs are commonly overlain by thick clay layers, which have prevented any gas escape for millions of years. Moreover such (depleted) gas reservoirs are widespread and have the potential to take large volumes of hydrogen.
H2STORE investigated the feasibility of a long-term storage (~ several months) of electrical power transformed into hydrogen in the geological underground. Therefore mainly (reservoir) sandstone units of 4 German depleted natural gas deposits and storage sites and 2 locations in Austria and Argentina, which are recently used as test sites for an injection of H2-CH4 gas mixtures at field scale, were studied. Thereby interactions induced by hydrogen-bearing laboratory experiments with hydrogen and reservoir components were evaluated by chemical, mineralogical, petrophysical and microbiological means on sandstones and their associated formation fluids before and after these tests. The runs were conducted at T = ~ 40 – 120 °C and p = ~ 4 - 20 MPa (= ~ 700 - 3.500 m depths), which corresponds to present-day conditions in the investigated reservoirs. The obtained data sets were used for numerical simulations, modelling the alteration of the rocks by special software programs.
Methane is the option
The basic concept for underground hydrogen storage in depleted gas reservoirs is the fact that such reservoirs have proven their capability to retain their natural gas content over millions of years and that they are used over decades as storage sites for gas supply. Moreover most recent investigations show that such structures are also appropriate for CO2 storage. The relevance of the H2STORE project in studying mainly geochemical/mineralogical, physico-chemical and bio-chemical features and alteration processes of reservoir sandstones (and their sealing caprocks) is unpredictable. Besides considering the safe, long-term storage of large volumes of energy, also new insights in the generation of "green", synthetic methane ("power-to-gas technology") were intended, because some bacteria can produce methane by metabolism processes from hydrogen and carbon dioxide. However, probably caused by the site-specific high salinity of formation fluids which were used in the laboratory experiments, such a reaction was not realized in the H2STORE study.
Basic research for future pore stores
The H2STORE H2STORE project conducted basic research to increase the knowledge on the behaviour of hydrogen in natural porous media at enhanced temperature and pressure conditions. A direct and immediate economic use was not anticipated. However some potential limiting parameters, such as fluid salinity, temperature and pressure/depths for industrial, large-scale geological hydrogen underground storage in the pore space of reservoir sandstone were identified. Thus this study provides an essential part of any economic appraisal of future wind energy-hydrogen underground storage projects. Moreover these results may influence even some other fields of energy supply, like the storage and transport of hydrogen in pipelines, the construction of wind farms, and the production of "green" methane.
The topics and milestones of the six subprojects involved in H2STORE were:
- Well core sampling at distinct German locations and delivery of samples from Austria and Argentina of depleted natural gas deposits and gas storage sites.
- Analyzing the mineralogical/geochemical composition and the petrophysical features (e.g. porosity, permeability) of the rocks, with a special emphasis on samples used in laboratory experiments.
- Performing laboratory runs at reservoir conditions on rock samples. Thereby the samples were exposed to site-specific formation fluids, enriched in hydrogen. By this approach the impact of hydrogen on reservoir components (mineral content, formation fluid, residual hydrocarbon species, biocenosis) established in mineral dissolution/precipitation processes, variations of petrophysical properties and in microbial biocenosis was shown.
- Investigations of samples (before and) after the experiments concerning the given components (see item 3) were conducted and the achieved data sets were allocated to subprojects for their numerical simulation attempts (see item 5).
- Numerical simulations (modelling) on: (a) the distribution of injected hydrogen in the underground, (b) the behaviour of gas mixtures and formation fluids, (c) the mineralogical and petrophysical alteration of the reservoir and cap rocks, (d) the interaction of microorganisms and the hydrogen-bearing formation fluids and reservoir rocks were realized.
Aims of the project partners
Subproject TP1 (Clausthal University of Technology - TU Clausthal) investigated experimentally the geo-chemical alteration of the reservoir and caprocks, induced by hydrogen and how these alterations affected the petrophysical properties of the reservoir rocks and the barriere properties of the caprocks. According to this objective, batch experiments on hydrogen-reservoir/caprock-brine systems were carried out with a high pressure - high temperature (HPHT-) autoclave as well as petrophysical routine and special core analyses (SCAL) to determine the alteration of fluid transport and barriere properties. A major outcome of this work was that the alteration/modification of petrophysical features is site-specific, ranging from a lack of such variations to moderate and high changes in rock porosities.
In subproject TP2 (TU Clausthal, Energieforschungszentrum Niedersachsen, Université de Lorraine) the spatial extension and dynamic alteration in the composition of the injected hydrogen with special reference of bacterial population dynamics and hydrodynamic instabilities were calculated by means of numerical modeling and the simulation of gas mixing and reactive transport processes in a porous underground hydrogen storage site. Effects of biochemical reactions on the rocks, their organic contents, reservoir fluids and hydrogen as well as the biological response to varying fluid compositions and mineralogical alterations were studied. It is suggested that e.g. the replacement of formation fluids by high rates of hydrogen injection is unstable and lateral fingering will occur, that during cyclic storage operations of hydrogen recovery rates of 80 – 95 mol% H2 can be expected and that an accumulation of methanogenic archaea is likely, which will metabolize H2 into methane.
In subproject 3 of the University Jena mineralogical, geo- and hydrochemical analyses of reservoir- and caprocks and their associated formation fluids as well as numerical simulations of the (connectivity of the) pore space and fluid transport features, based on micro tomographic data sets were conducted. These data were determined before and after the laboratory batch experiments of TP 1. They confirm that at specific reservoir conditions hydrogen can promote mineral reactions like mineral dissolution, but that such processes are absent in storage sites characterized by less extreme environments. The strong impact of H2 on the (in some reservoirs) observed mineral dissolution processes was substantiated by PHREEQC geochemical numerical simulations. In these sites the mineralogical reactions also provoked modifications of petrophysical parameters (as obtained in TP 1) and thus in fluid transport behaviour.
The working group "microbial geo-engineering" of the German Research Centre for Geosciences (GFZ) Potsdam was executing subproject 4, which investigated potential interactions between hydrogen and the metabolisms of microbes. In their laboratory experiments variations in composition, abundance and activity of the microorganisms induced by hydrogen exposure were realized. Thereby the activity of sulfate reducing bacteria (SRB) was most impressive, whereas no indications of methane formation by archaea were found.
Subproject 5 was located at the GFZ Potsdam and mainly performed laboratory experiments to investigate the interaction of hydrogen and potential reservoir rocks at the site-specific conditions of the location Ketzin/Brandenburg. Here, only minor variations in fluid composition and rock porosity before and after the tests were detected. These changes are interpreted as some kind of artefacts induced during (fluid) sampling, (rock) sample preparation and/or analytical methods. Additionally, tests to determine the rate of hydrogen solubility in water were conducted. These experiments indicate that hydrogen is much more soluble in saline waters than suggested before.
In subproject 6 (GFZ Potsdam) numerical modelings of fluid-fluid and fluid-rock interactions were conducted to improve the knowledge on geochemical reactions and to evaluate the relevance und integration of thermodynamic and kinetic data in such simulations. These calculations were combined with a plausibility test on the relevance of the observed (very minor) modifications induced by autoclave experiments in TP 5 at the Ketzin/Brandenburg location. An outcome of this study is that only a reduction of pyrite (FeS2) to pyrrhotite (FeS) is likely at the specific site conditions of Ketzin, whereas any other mineral reactions are improbable. Even the reduction of pyrite to phyrrhotite is kinetically strongly impeded in a way that during the operating time of a potential hydrogen underground storage operation at Ketzin this reaction is without any relevance. However, this is only true if abiotic reactions are regarded and at the site-specific conditions of Ketzin, when also taking into account microbiological processes and/or different p-/T-conditions and formation water compositions (e.g. salinities) this statement has to be evaluated.