Large Hydrogen Underground Storage

From Roads2HyCom Hydrogen and Fuel Cell Wiki - A Reliable Source of Information - Edited by Technology Experts Only

Jump to: navigation, search


The storage of large quantities of hydrogen underground is a key step for the hydrogen economy. One can store hydrogen as either a gas or a liquid. Both approaches are examined hereafter and appear to be already at an industrial level. The stakes of these sector is to keep the cost of such operation as low as possible and to avoid as much as possible leaks of hydrogen.

The most widely studied options for storing large quantities of gaseous hydrogen are underground depleted gas fields or aquifers, and caverns. This will be analyzed in the first section below. Then underground storage in buried tanks, either in compressed gas form or in liquid form will be illustrated by a few examples in the second section. Eventually the economical aspects of each technological option will also be examined.

State of the Art

Large Scale Underground Hydrogen Storage

Natural gas bas been stored underground since 1916 and much of the experience is directly applicable to hydrogen. Originally underground storage was confined to depleted oil and gas fields. They are porous rock structures not unlike those associated with aquifers. Such storage facilities tend to be extremely large; volumes of gas stored exceed 109 Nm3. Pressures can be up to 40 atm. As illustrated in the figure below, any underground storage in porous media requires the following features :

  • a stratum of porous rock, usually sand or sandstone, at a depth of 150 - 900 m below the surface ;
  • an impervious caprock of adequate thickness ;
  • a suitable geological structure such as an anticline, which usually forms a dome-shaped geologic structure.
Schematics of a depleted fossil oil or gas well, or aquifer. From Stone et al (2005)

The porosity of the sandstone or sand must be sufficiently high to provide a reasonable void place in an aggregate sense to yield an economically acceptable storage volume. The permeability must be high enough to provide an adequate rate of inlet flow (injection) and outlet transmission injection (withdrawal). On the other hand, the caprock structure must be reasonably impermeable if it is to contain the gas.

During exploitation one needs to maintain a minimum base gas or "cushion gas" which is the volume of gas intended as permanent inventory in the storage reservoir to maintain adequate pressure and deliverability rates throughout the withdrawal season. In the case of hydrogen storage it has been proposed to use a cushion gas of different nature, such as natural gas, to displace a hydrogen-rich gas (Paterson, 1983). But this is valid only if gas stratification can be maintained between cushion gas and hydrogen by avoiding interdiffusion or "fingering". Nevertheless this would also require an efficient gas separator (membrane or PSA) in the gas station at ground level.

An important point to note regarding the performance of the caprock structure is the mechanism involved in sealing the top of the underground reservoir, either a depleted field or an aquifer. This sealing occurs because of water capillary action, in which water fills all the voids of the caprock structure and must be expelled by sufficiently high pressure to overcome the capillary resistances (the threshold pressure of the caprock). Below this incremental pressure, the caprock will act as an effective barrier to the passage of any gas. This incremental or threshold pressure and the effectiveness of the sealing action is independent of the nature of the gas because it is a water-rock capillary effect. This is an important observation in that it indicates that hydrogen in underground storage will behave much like natural gas insofar as integrity against leakage is concerned.

The scientific treatment of the dynamics of gas transfer in such porous media consists in solving thermodynamic and transport equations using appropriate equation of states for the gas or gas mixtures and diffusion -permeation due to a pressure gradient according to Darcy's law to derive the gas extraction rate (Stone et al 2005). Moreover Panfilov et al (2006) have recently studied and modelled a curious physical-chemical phenomenon which can affect underground storage of H2 CO2-CH4 mixtures. It is characterized by a reduction in H2 and CO2 concentrations, concomittant with an increase of the CH4 fraction. These phenomena were observed in the Lobodice and in the Beynes town gas storages where the original stored mixture of 55% H2, 20% CO2+CO and 20% CH4 was transformed into a new gas containing 37% H2, 12% CO2+CO and 40% CH4. Simultaneously a small reduction in the total gas volume (~15%) was observed. These results are ascribed to two chemical reactions CO2+4H2 ® CH4 +H2O and CO+3H2 ® CH4+H2O induced by micro-organisms (methanogenic bacteria) present in stratal water within porous rocks.

Where previously mined cavities are available which are, or can be made, gas tight, there is an opportunity to use these as artificial underground structures for hydrogen storage. A special case is the use of solution-mined salt caverns as illustrated in the figure below.

Schematics of the solution-mining of a salt cavern : a) initial bore hole, b) solution mining in process c) final solution-mined cavern. From Stone et al (2005)

In locations where the salt layer is between 60 to 100 metres thick it is debatable whether a vertically mined salt cavern could be used to store the required volume of hydrogen. In such locations horizontal drilling with solution mining techniques (refer figure) will be preferred unless a collection of smaller vertical solution mined caverns could be inter-connected (Stone et al, 2005). Alternatively caverns can be drilled in rocks. Lindblom (1985) made a detailed study a conceptual design for compressed hydrogen storage in mined rock caverns consisting of parallel tunnels surrounded by a water curtain as illustrated in the figure.

Unlike depleted field and aquifer storage systems, cavern storage involves large open, void spaces to be filled with gas. A more complex structural analysis is therefore required to establish feasibility. For example, if the pressure in the cavity is allowed to drop significantly below ambient pressure, a collapsing stress situation is created which might result in loss of structural integrity of the storage volume. One approach considered is to replace the gas drawn off with water so that the cavity pressure is maintained. There is of course no cushion gas requirement in this hydraulically compensated scheme and the delivery pressure is constant equivalent to the hydraulic head. The disadvantage for storage in salt caveras is that the working fluid must be saturated brine. Surface storage of this is complicated by many environmental constraints.

Schematics of an horizontal solution-mined salt cavern. From Stone et al (2005)
Concept of a compressed hydrogen storage cavern consisting of parallel tunnels drilled in the rock and surrounded by a water curtain. From Lindblom (1985)

Recently, Morrow et al (2006) of Praxair Technology Inc. have patented a method of hydrogen underground storage in a salt cavern and the subsequent reinjection into a pipeline after different compression, extraction and purification steps.

Single salt caverns can be 50 106 m3, whereas hard rock caverns would be one tenth of this size. However, modern mining machinery and techniques make the costs of mined caverns comparable to other underground options. In the Lindblom rock-cavern concept the water curtain system has three important fonctions: i) to maintain water saturation in the rock fracture system during excavation of the cavern, ii) to regulate the pressure difference between water and gas as to ensure that gas breakthrough is impossible, iii) to induce an artificially high water pressure around the cavern as to make the use of higher gas pressure possible.

Buried Tanks

Underground storage may also consist in burying compressed gas or liquid hydrogen tanks that are usually placed at ground level, in order to save ground space. Of course this option should only apply for smaller hydrogen quantities (typically 103 to 104 Nm3) and at a much smaller depth (typically a few meters) than direct gas storage in caverns.

Nevertheless the challenge of local storage of hydrogen in future hydrogen vehicle refuelling stations has triggered a new interest and proposals for this option. Indeed when referring to the largest existing gasoline fueling stations in Europe - which are scaled to refill several hundreds up to one thousand cars per day with a local storage capacity corresponding to two days of consumption - the equivalent hydrogen refueling stations in the future should be able to store up to 10 tons of hydrogen. This corresponds to a volume of the order of several 105 Nm3, or 250 m3 at 700 bars, or 140 m3 of LH2 !

Compressed Gas Tanks

Presently the largest manufactured compressed hydrogen tanks in the world (about 15000 m3) can be pressurized only up to 12-16 bar. Individual metallic cylindrical tanks at 200-300 bar can be scaled up to ~ 1000 liters (inner water volume) each for hydrogen transport in tube trailers where currently about 15 tubes are bundled together.

Large volume high pressure gas tanks made of carbon fiber-reinforced composite vessels are presently developed for CNG or H2 by a few companies.

Underground compressed gas storage in pressure vessels (1) immersed in a basin (4) filled with a liquid (5), with interconnecting piping and connections (2), an anchoring system (3), a cover with a slanted surface on the underside (6) for covering the basin and leading any gas leak towards and through a vent (7), with a gas detection equipment (9) protecte with a light cover (8). The basin can be filled an emptied with an appropriate water piping (10 and 11) . From Schelling et al (2006)

Gas tank with protection system for underground installation. From Poillucci (2003)

Although a compressed gas storage installation is usually place outdoor, on or above ground level, it can be buried to save ground space and also to provide improved protection from external influences such as radiation from adjacent fires or damage caused by explosions. However this alternative is rarely used because as it makes inspection of the vessels and interconnection of pipes less easy, and it requires preventive measure to prevent corrosion. Nevertheless to overcome the difficulty of inspection and pipe interconnection it has been proposed to place the tanks in a basin and submerge them afterward with a liquid such as water, so that tank protection from heating and and explosion is compatible with an easy inspection simply by lowering the water level in the basin. Leak detection could be an issue but some solutions exist such as the one proposed by Norsk Hydro (Schelling et al 2006) and illustrated in the figure. Alternatively the tank can be placed in a protection system consisting of a sack that contains the tank and a gas permeable material arranged in such a manner as to constitute a layer around the tank (Poillucci, 2003). The sack can be material impermeable to water and gas as shown in the figure above.

Liquid Tanks

At present the technology of liquid hydrogen (LH2) storage is state-of-the-art thanks to extensive applications in space propulsion. LH2 tanks for long time storage have double vacuum perlit vacuum insulation. Common stationary tanks have capacities ranging from 1500 liters (approx. 1100 Nm3 or 100 kg H2) up to 75000 liters (~ 60000 Nm3 or ~5 tons of H2 ) with radii of 1.4 m to 3.8 m and heights from 3 m up to 14 m. Since the hydrogen liquefaction capacity in North America is ten times larger than in Europe (300 tons/day in 2004 in North America, including 224 tons/day in the USA, versus 20 tons/day in Europe) there are more of such tanks in the USA, the largest of which belongs to NASA and is located at Cape Canaveral. This tank, at ground level, has a storage volume of about 3800 m3 (~ 270 tons LH2). With an outer spherical diameter of 20 m, its evaporation rate is under 0.03% per day, allowing for a storage period of several years. As compared with pressurized gas storage, this method offers more inexpensive storage costs when dealing with large quantities. The energy required for liquefaction may not be a barrier if the hydrogen is to be transported as a liquid anyway, or if the end-use application requires its fuel to be in liquid form.

The benefits of underground LH2 stationary storage can be summarized as follows: i) it decreased land usage and footprint, ii) reduced potential hazards (vandalism, fire, vehicle impact), iii) inherent spill containment. Actually this underground storage solution has been also considered for LNG storage.

Currently there are only a few examples of LH2 tank being either placed in a room underground or actually buried underground.

Main metrics

METRIC SUB-METRIC DATA / RATING UNITS Compressed Underground Hydrogen Liquid Hydrogen
Technology Accessibility Compatibility with existing technologies Rating 0-4 4 4
Number of retailers & depots Data number N/A N/A
Space available at retailers and depots Rating 0-4 N/A N/A
Global Environmental Impact
(to be coordinated with ECN)
GHG emissions associated with fuel retail and depot Data gCO2 eq / kg fuel 0 0
CO2 emissions associated with fuel production Data gCO2 / kg fuel 0 0
Local Environmental Impact
(to be coordinated with ECN)
Air quality impact (consider NOx, PM, CO, NMHC) Rating 0-4    
Noise or perception of noise from retail and depot (SPL,loudness...) Data/Rating dB(A), sone    
Land use / damage to nature (eg. fuel depots) Rating 0-4    
Efficiency Global energy efficiency Data % N/A N/A
Energy efficiency of auxiliary facilities Data % N/A N/A
Capacity & Availability
(to be coordinated with ECN)
Measured capacity of retailers and depots Data kg fuel / facility up to 1 000 000 000 nm3 1 00 000 nm3
Availability of retailers and depots Data %    
Lifetime of technology Data years ? ?
(to be coordinated with TNO)
Leaks at retail or depot per number of tanks Data no. / year    
Severity of failure Rating 0-4    

Market / Diffusion

Underground Hydrogen Storage

Basically there are four underground formations in which gas can be stored under pressure:

  • Depleted oil or gas fields,
  • Aquifers,
  • Excavated rock caverns or mines,
  • Solution-mined salt caverns.

For example, there are already twenty-three salt caverns being used for natural gas or hydrogen storage in the UK. These are located in Cheshire, in Stafford and in Yorkshire. There are also another twenty-four salt caverns planned for natural gas storage (Stone et al, 2005). In France there are at least fifteen underground storage sites for natural gas, either in salt caverns or in aquifers, for total useful capacity of 110 TWh, i.e. about 30% of the present annual consumption.

Underground storage of hydrogen and synthetic gas (H2-CO) mixes is not a new concept. A major study of underground storage of gaseous hydrogen was conducted in 1979 by the Institute of Gas Technology (now known as Gas Technology Institute, GTI) in the USA (Foh et al 1979). Over the last decades there have been several examples of underground storage of pure hydrogen or synthetic gas H2-CO mixtures (Panfilov 2006) :

  • In England, at Teesside, Yorkshire, the British company ICI has stored 1 million Nm3 of nearly pure hydrogen (95% of H2 and 3-4% of CO2) in three salt caverns at about 400 m in depth for a number of years.
  • In France, at Beynes, Ile de France, the gas company Gaz de France has stored a synthetic "town gas" 50-60% hydrogen in an aquifer of 330 million Nm3 capacity between 1956 and 1974 . No gas losses or safety problems have been reported
  • In Russia, pure hydrogen was stored underground at 90 bars for the needs of the aerospace industry
  • In Germany, at Kiel, a 62% H2 town gas was stored in a salt cavern of 32000 m3 at 80-100 bar
  • In Czechoslovakia, at Lobodice a 50% H2 town gas was stored in an aquifer.

Praxair is presently constructing a large underground hydrogen storage facility to enable "peak shaving" of its hydrogen production. This facility, located in Texas, will utilize a salt cavern, and will be the first of its kind in the industrial gases industry. Connected to the Praxair’s hydrogen pipeline network which serves large in Texas and Louisiana it will significantly increase the on-demand availability of hydrogen during periods of peak demand. This peak-shaving system should be brought on line in 2007.

The report of the 1979 study by the Institute of Gas Technology (Foh et al 1979) confirmed the economic and technical feasibility of large-scale underground hydrogen storage. According to Taylor et al (1986) underground storage is the most inexpensive mean of storing large quantities of gaseous hydrogen. Indeed underground hydrogen gas storage is about two orders of magnitude cheaper than tank storage when applied to volumes of several million Nm3 of hydrogen. The lower capital investment is in salt caverns, and the highest for depleted gas wells.

In depleted gas wells the gas stored in a field is divided into active working gas and cushion gas. The latter is an inactive base gas not recoverable, and it can be considered as an investment or a minimum inventory. The ratio of working to cushion gas varies widely with a ratio greater than 2:1 generally being preferred. As hydrogen is relatively expensive commodity, the cost of the cushion gas is a very significant part of the capital charges for such large storage reservoirs. However, as the cavern is repeatedly cycled, the initial cushion gas cost is amortized (Stone et al, 2005).

Mining a salt-cavern may take a year or more to mine, which accounts for 25% to 35% of the total initial investment (Stone et al 2005). In his study of a conceptual hydrogen storage cavern of 300 million m3 at 650 m depth in competent rock, Lindblow (1985) estimated a three and half year time break-down for construction and a total cost of U.S .$39-64 million which should exceed $100 millions by now. From this we can estimate an investment cost ratio of about 0.3 € per H2 Nm3.

Buried Tanks

Compress Hydrogen

There are actually no industrial solution for large hydrogen compressed tanks and this type of solution is still under developpement. For example Lincoln Composites has already manufactured type IV vessels with plastic liner at 7000 psi (483 bar) for small size hydrogen refuelling stations (50kgH2/day); the company is presently working on a new tank with plastic liner at 5000 psi (350 bar), approximately 1.08 meter diameter by 11.5 meters length, with a storage volume of approximately 8700 liters (water volume).

Buried compressed gas tanks are relevant only for small quantities, from 100 to 1000 kg of H2 (~ 1000 to 10000 Nm3), and their cost, including excavation and basement installation is obviously much larger, in the range of 10 to 100 EUR per Nm3.

Liquid Hydrogen

Underground storage for the LH2 tank in the BP hydrogen refuelling station located in London. The picture of the tank is from Cryolor (tank manufacturer, subsidiary of Air Liquide), and the schematic drawing of the installation is taken from Roach (2004)

The first example of LH2 tank was operated for the bus refuelling station of BP in London within the HyFleet-CUTE project. The cryogenic tank of liters capacity ( tons of LH2), manufactured by Air Liquide for BOC, is placed in a basement, with a technical room for operators separated from the main body of the tank for safety reasons (Roach 2004 ), as illustrated in figure.

The second example is the concept proposed by Linde (Trill and Wolf, 2006) for future hydrogen filling stations offering both LH2 and CGH2, as illustrated in figure below. This concept of underground LH2 storage will soon be applied in the construction of a hydrogen refuelling station for BMW in Münich in 2007.

Linde concept for future hydrogen filling stations for GH2 and LH2 with underground storage tanks.

Eventually the third example is the concept proposed and developed by Air Products for the Shell Hydrogen refuelling station in Washington DC, as illustrated in the figure. The originality of this solution is all connections are accessible at ground level after the tank has been vertically slided in a cylindrical cavity. This concept applies for storage volumes from 1500 up to 9000 gallons (~ 400 up ~ 2400 liters) of LH2, equivalent to 27 - 167 kg, or 320 - 1920 Nm3 of H2, and the boil off rate is less than 0.5% per day.

Air Products' underground LH2 tank being installed at the Shell Hydrogen refuelling station in Washington DC in 2004.

Main industrial players

The main industrial players for large hydrogen storage systems are actually the major international companies that are selling industrial gases namaly Praxair, Air Products, BOC, Linde, or l'Air Liquide.

The research and development projects are mainly focused on reducing cost and hydrogen leaks. Safety can also be an issue.


  • EIA-NGD: United States Government Energy Information Administration / Natural Gas Division
    The basics of underground natural gas storage
    (August 2004), [1]
  • Foh S., Novil M., Rockar E. and Randolph P.
    Underground hydrogen storage
    Final report, Institute of Gas Technology, Chicago, issued as BNL 51275 (December 1979)
  • Lindblom U.E.
    A conceptual design for compressed hydrogen storage in mined caverns
    Int J. Hydrogen Energy, 10 (1985) pp. 667-675
  • Morrow J.M., Corrao M., and Hylkema S.
    Method of storing and supplying hydrogen to a pipeline
    US Patent US7,078,011 B2, July 18, 2006
  • Panfilov M., Gravier G., and Fillacier S.
    Underground storage of H2 and H2-CO2-CH4 mixtures
    Proceedings of the 10th European Conference on the Mathematics of Oil Recovery (ECMOR), Amsterdam, The Netherlands, Sept 4-6, 2006
  • Paterson L.
    The implications of fingering in underground hydrogen storage
    Int. J. Hydrogen Energy, 8 (1983) pp. 53-59
  • Poillucci G.
    Gas tank with protection system for underground installation
    European patent EP1435481 B1, September 26, 2003
  • Roach J.
    Liquefied gas storage installation
    European patent application EP1631768 A1, June 8, 2004
  • Schelling, J., Kittilsen, P Nilsen, S.H., and Grøønstadt, L. K
    Plant for storing gas under pressure
    World patent application, WO2006088378, August 24, 2006
  • Stone H.B.J., Veldhuis I., and Richardson R.N
    An Investigation into Large-Scale Hydrogen Storage in the UK
    Proceedings International Hydrogen Energy Congress and Exhibition IHEC 2005, Istanbul, Turkey, 13-15 July 2005
  • Taylor J.B., Alderson J.E.A., Kalyanam K.M., Lyle A.B., and Phillips L.A.
    Technical and economic assessment of methods for the storage of large quantities of hydrogen
    Int J. Hydrogen Energy, 11 (1986) pp. 5-22
  • Trill, R. , and Wolf J.
    Filling station for hydrogen
    European Patent EP 1360084 B1, August 30, 2006
Personal tools
Site administrator:

Presented by Institut fr Kraftfahrzeuge

Contributing projects:


Funded by the European Commission

Funded by the European Commission