CO2 temporary storage in big underground caverns

Abstract

After the CO2 has been captured at the source of emission,
the CO2 would have to be transported to the storage
site using different technologies. In some countries
(i.e. USA) real possibilities exist so that available and
new oil and water pipe lines could be used for such
operations. In practice it means that transportation
could be carried out with motor carriers, railway and
water carriers. If the present experiences are taken
into account and the real situation checked, such
transportation systems are mainly used in praxis. For
maximum throughput and to facilitate efficient loading
and unloading, the physical condition with respect to
pressure and temperature for the CO2 should be the
liquid or supercritical/dense phases. Temporary storage
of CO2 is of importance for finding a comprehensive
solution for long-term storage under various environmental
circumstances. Underground caverns are one
of the possibilities of temporary storage. Geotechnical
analysis of stress and strain changes that are present in
the rocks around underground caverns filled with CO2
under high pressure provides a realistic assessment of
conditions for temporary storage. This paper presents
the analysis described above, for different parameters
relating to underground storage of CO2.

Full text

237
Jakob Likar1, Jože Žarn1, Eivind Grøv2, Tina Marolt Čebašek1, Andrej Likar3
1University of Ljubljana, Faculty of Natural Sciences and Engineering, Aškerčeva cesta 12, 1000 Ljubljana, Slovenia
2SINTEF/University of Science and Technology – (NTNU), Trondheim, Norway
3Geoportal, d. o. o., Tehnološki park 21, 1000 Ljubljana, Slovenia
*Corresponding author. E-mail: jakob.likar@ntf.uni-lj.si
Received: November 5, 2015
Accepted: December 21, 2015
Original scientic paper
Abstract
After the CO2 has been captured at the source of emis-
sion, the CO2 would have to be transported to the stor-
age site using different technologies. In some countries
(i.e. USA) real possibilities exist so that available and
new oil and water pipe lines could be used for such
operations. In practice it means that transportation
could be carried out with motor carriers, railway and
water carriers. If the present experiences are tak-
en into account and the real situation checked, such
transportation systems are mainly used in praxis. For
maximum throughput and to facilitate efficient loading
and unloading, the physical condition with respect to
pressure and temperature for the CO2 should be the
liquid or supercritical/dense phases. Temporary stor-
age of CO2 is of importance for finding a comprehensive
solution for long-term storage under various environ-
mental circumstances. Underground caverns are one
of the possibilities of temporary storage. Geotechnical
analysis of stress and strain changes that are present in
the rocks around underground caverns filled with CO2
under high pressure provides a realistic assessment of
conditions for temporary storage. This paper presents
the analysis described above, for different parameters
relating to underground storage of CO2.
Key words: temporary storage, big underground
caverns, numerical modelling, boundary element
method – BEM, CO2 high-pressure
Izvleček
Po zajemu CO2, ki ni namenjen izpustu v atmosfero, je
več možnosti transportiranja večjih količin tega plina
v skladišče. V nakaterih državah (npr. ZDA) je realna
možnost za transport CO2 uporaba obstoječih ali novih
cevovodov za oskrbo območij z ogljikovimi derivati.
Prav tako je v delu več projektov, ki upoštevajo navede-
no možnost transporta CO2 v Severnem morju. Predlo-
gi za transport CO2 s cestnimi, železniškimi in vodnimi
transportnimi sistemi, ki jih je mogoče prilagoditi po-
sebnim zahtevam, upoštevajo specifične lastnosti CO2,
saj bi bila najbolj ekonomična cena transporta na enoto
dosežena, če bi bil plin v tekoči oz. superkritični gosti
fazi.
Začasno skladiščenje CO2 ima velik pomen pri iskanju
celovite rešitve dolgoročnega skladiščenja v različnih
okoljih. Podzemne kaverne so ena izmed možnosti za-
časnega skladiščenja. Geotehnična analiza napetostnih
in deformacijskih sprememb, ki so v hribinah okrog
podzemnih kavern, napolnjenih s CO2 pod visokimi
tlaki, omogoča realno oceno razmer začasnega skladi-
ščenja. V prispevku je prikazana navedena analiza za
različne parametre začasnega podzemnega skladišče-
nja CO2.
Ključne besede: začasno skladišče, velike podzemne
kaverne, numerično modeliranje, metoda mejnih ele-
mentov – BEM, visok tlak CO2
CO2 temporary storage
in big underground caverns
Začasno skladiščenje CO2 v velikih
podzemnih kavernah
ACCEPTED

Likar, J., Žarn, J., Grøv, E., Marolt Čebašek, T., Likar, A.
238
RMZ – M&G | 2015 | Vol. 62 | pp. 237–254
Introduction
Carbon dioxide (CO2) is a greenhouse gas that
occurs naturally in the atmosphere. Human ac-
tivities, such as the burning of fossil fuels and
other processes, are significantly increasing its
concentrations in the atmosphere, thus con-
tributing to the Earth’s global warming. One
technique that could limit CO2 emissions from
human activities into the atmosphere is CO2
capture and storage (CCS). It involves collect-
ing, at its source, the CO2 that is produced by
power plants or industrial facilities and storing
it away for a long time in underground geologi-
cal layers, in the oceans, or in other materials. It
should not be confused with carbon sequestra-
tion, which is the process of removing carbon
from the atmosphere through natural processes
such as the growth of forests. It is expected that
fossil fuels will remain a major energy source
until at least the middle of this century [1–5].
Therefore, techniques to capture and store the
CO2 produced, combined with other efforts,
could help stabilise greenhouse gas concen-
trations in the atmosphere and fight climate
change. CO2 could be captured from power
plants or industrial facilities that produce large
amounts of it [6–9]. Technology for CO2 capture
from small or mobile emission sources, such as
home heating systems or cars, is not sufficiently
developed yet. The question is whether it could
be realised in the near future? A significant pro-
portion of the CO2 produced by fossil fuel power
plants could potentially be captured. By 2050
the amount captured could represent 21 % to
45 % of all the CO2 emitted by human activities.
After the CO2 has been captured at the source of
emission, the CO2 would have to be transported
to the storage site. Such transportation would
require large scale infrastructures due to the
large volumes to be handled. Nowadays exist-
ing CO2 transportation systems has its basic
location in the USA, where several million tons
of CO2 are transported annually, over long dis-
tances on shore in high pressure pipelines for
use in the EOR industry (Gale et al. 2002) [10].
Using CO2 in EOR (Enhanced Oil Recovery)
projects has the advantage of adding a value to
the CO2, e.g. oil producers in the USA are willing
to pay between 9 US$/t and 18 US$/t of »end
of pipe« delivered CO2. Pipelines for off-shore
transportation of CO2 have not been applied
yet but are technologically feasible today, and
a CO2 pipeline infrastructure-off shore was in-
vestigated in the CO2 for EOR in the North Sea
(CENS) project [11]. In practice it means that
transportation could be carried out with mo-
tor carriers, railway and water carriers. If the
present experiences take into account and
check the real situations, such transportation
systems are mainly used in the food and brew-
ery industry, and the amounts transported are
within the range of some 100 000 t of CO2 annu-
ally, so that is much smaller than the amounts
associated with Carbon Capture and Storage
(CCS) (Figure 1). In contrast the transporta-
tion conditions for CO2 have some similarities
with LPG (Liquefied Petroleum Gas) technolo-
gies, which are transported by water carriers,
railway and motor carriers on a relatively large
scale. Hence, from these points of view, experi-
ences from the LPG industry could also be used
for establishing a large scale CO2 transportation
infrastructure. For maximum throughput and
to facilitate efficient loading and unloading, the
physical conditions with respect to pressure
and temperature for the CO2 should be the liq-
uid or supercritical/ dense phases. It should
be noted that pipelines suffer from pressure
drops along the transportation route, which
can result in two phase flows and operational
and material problems (e.g. cavitation) in com-
ponents such as booster stations and pumps.
Utilising pipelines still needs stable conditions
of operation where the transported media is
in the supercritical/dense phase [12–15]. This
condition occurs at temperatures higher than
60 °C and pressures above the critical pressure
of 7.38 MPa, giving a good margin for avoiding
two phase flows. For the other means of trans-
portation, i.e. motor carriers, railway and water
carriers, which have constant pressure, liquid
conditions are suitable. The density for CO2
approaches 1 000 kg/m3 as liquid, as well as
during the supercritical/dense phase.
If available conditions are weighted in the
goal to find optimal technical solutions the in-
termediate storage can be usable in different
ways. A pipeline has the advantage of provid-
ing steady state flow, i.e. a continuous flow
from the emission source to the final storage
site. That means the complex transportation
ACCEPTED

CO2 temporary storage in big underground caverns
239
system should include appropriate intermedi-
ate storage facilities for handling the reloading
of CO2 at the middle points or at the final point
close to harbour’s facilities. There are two main
technologies for intermediate storage of LPG,
either underground in great rock and salt cav-
erns or in large steel tanks above ground which
have many disadvantages. But at present, only
the steel tank technology is used for CO2, in
the contra version that both technologies can
be applied in practice. Existing rock caverns
for LPG have storage capacities up to around
500 000 m3, which should approximately corre-
spond to 500 000 t of CO2. In the other hand salt
caverns have similar storage capacities of LPG
but are excluded in this work due to uncertain-
ties with respect to the dissolution behaviour
of CO2. Steel tanks have storage capacities up to
3 000 t of CO2
[12, 13]. Rock caverns within the LPG
industry are constructed in two different ways,
either as pressurised or as cooled caverns. If
the caverns are intended for storage of CO2,
these techniques must be combined to create
favourable conditions with respect to pressure
and temperature for the CO2. So, the construc-
tion cost of rock shelter depends mainly on the
rock quality, which is available at the decided
location. Low bearing capacity of »poor rock
quality« increases the need for strong support
measures which include lining and reinforce-
ment of the rock strata. In those cases nonlinear
increases in costs should be expected. In many
technical and scientific studies the safety and
public acceptance are not included enough. CO2
is not toxic but can be fatal, due to asphyxiation,
at concentrations exceeding around 10 % by
volume [16], levels that can be achieved at a dis-
charge as CO2 is heavier than air and, hence, will
tend to collect in depressions. Statistics from
the EOR industry clearly show that the risks for
pipeline leakage are lower than for natural gas
or hazardous pipelines. Anyway in the goal to
minimise risks, transportation of CO2 should be
routed away from large centres of population.
Another issue, which can indirectly affect the
transportation, is the public opinion concern-
ing storage of CO2. The concept of off shore
disposal on average is considered to be safer,
if the leakage is under question, than on shore
systems. For this reason the support of the pub-
lic is more easily implemented for the off shore
system [14, 15, 17–19]. The above facts and assessing
the need for the construction and use of tempo-
rary CO2 storage are the basis for further work
in this specialised field of underground con-
struction. Already in several sentences it has
also been suggested that the appropriate price
for the construction of underground structures,
essential important geological, hydrogeological
and geotechnical conditions should be present
at the selected location. Within the scope of the
presented work the addition of the basic fea-
tures of the behaviour of CO2, given orientation
are necessary and analysis of some possible
construction of temporary underground stor-
age facilities. In the analyses should take into
account the economic viability of the storage of
CO2 at pressures between 80 bar and 100 bar
at the ambient temperature of rock mass. It is
possible to take into account the technologi-
cal requirements of CO2 transport in terms of
maintaining the highest possible density of CO2
in the liquid state, which is an economically im-
portant item for establishing the final price of
permanent storage of CO2 [20–23].
Some information about CO2
producers
Coal power plants are a good example of a large
point source of CO2 emissions. Three systems
are available for power plants: post-combus-
tion, pre-combustion, and oxfuel combustion
systems. The captured CO2 must then be puri-
fied and compressed for transport and storage.
It is possible to reduce the CO2 emissions from
new power plants by about 80 % to 90 % but
this increases the cost of electricity produced by
35 % to 85 %. For industrial processes where a
relatively pure CO2 stream is produced, the cost
per ton of CO2 captured is lower. Except when
the emission source is located directly over the
storage site, the CO2 needs to be transported.
Pipelines have been used for this purpose in the
USA since the 1970s. CO2 could also be trans-
ported in liquid form in ships similar to those
transporting liquefied petroleum gas (LPG).
For both pipeline and marine transportation of
CO2, costs depend on the distance and the quan-
tity transported. For pipelines, costs are higher
when crossing water bodies, heavily congested
ACCEPTED

Likar, J., Žarn, J., Grøv, E., Marolt Čebašek, T., Likar, A.
240
RMZ – M&G | 2015 | Vol. 62 | pp. 237–254
areas, or mountains. Compressed CO2 can be
injected into porous rock formations below the
Earth’s surface using many of the same meth-
ods already used by the oil and gas industries
(Figure 1). The three main types of geological
storage are oil and gas reservoirs, deep saline
formations, and un-minable coal beds [24–28].
CO2 can for instance be physically trapped
under a well-sealed rock layer or in the pore
spaces within the rock. It can also be chemical-
ly trapped by dissolving in water and reacting
with the surrounding rocks. The risk of leakage
from these reservoirs is rather small. Storage in
geological formations is the cheapest and most
environmentally acceptable storage option for
CO2. Oceans can store CO2 because it is soluble
in water. Captured CO2 could potentially be in-
jected directly into deep oceans and most of it
would remain there for centuries. CO2 injection,
however, can harm marine organisms near the
injection point. It is furthermore expected that
injecting large amounts would gradually affect
the whole ocean. CO2 storage in oceans is gen-
erally no longer considered as an acceptable
option. Through chemical reactions with some
naturally occurring minerals, CO2 is converted
into a solid form through a process called min-
eral carbonation and stored virtually perma-
nently. This is a process which occurs naturally,
although very slowly. These chemical reactions
can be accelerated and used industrially to arti-
ficially store CO2 in minerals. However, the large
amounts of energy and mined minerals needed
make this option less cost-effective. It is techni-
cally feasible to use captured CO2 in industries
manufacturing products such as fertilisers. The
overall effect on CO2 emissions, however, would
be very small because most of these products
rapidly release their CO2 contents back into the
atmosphere. It is expected that carbon capture
and storage would raise the cost of producing
electricity by about 20 % to 50 % but there are
still considerable uncertainties. In a fully inte-
grated system including carbon capture, trans-
port storage and monitoring, the capture and
compression processes would be the most ex-
pensive steps. Geological storage is estimated
to be cheaper than ocean storage, the most ex-
pensive technology being mineral carbonation.
Overall costs would depend both on the techno-
logical choices and on other factors such as lo-
cation or fuel and electricity costs. Capture and
storage of the CO2 produced by some industrial
processes such as hydrogen production can be
cheaper than for power plants.
Basic physical and chemical
parameters of CO2
CO2 is a naturally-occurring substance made up
of carbon and oxygen, two of the more common
chemical elements on earth. Under normal at-
mospheric conditions, CO2 is a gas. It can be
compressed into a liquid, frozen into a solid
(dry ice) or dissolved in water (carbonated bev-
erages, beer and sparkling wines) (Figure 2).
In the atmosphere, CO2 comprises about 0.04 %
of the air we breathe. It also occurs naturally
in both fresh and sea water, and in the ground
Meanwhile, green plants absorb CO2 for pho-
tosynthesis and emit oxygen back into the at-
mosphere CO2 is also exchanged between the
atmosphere and the oceans and is emitted or
absorbed in other natural processes. Working
together in a natural system called the carbon
Figure 1: Possible CO2 capture and storage – CCS system [12].
Figure 2: Crystal structure of dry ice and sample of solid
carbon dioxide or »dry ice« pellets and arising atmospheric
carbon dioxide versus time [29].
ACCEPTED

CO2 temporary storage in big underground caverns
241
cycle; these processes have in the past kept
the levels of CO2 in the atmosphere stable over
time. Nature’s carbon cycle normally keeps CO2
levels in balance but human activity, mostly
the burning of fossil fuels, produces more CO2
than nature can absorb. The arrows in this dia-
gram show the annual flows of carbon in billion
tones (metric tons). The human contribution is
relatively small but enough to throw the cycle
off balance. The extra CO2 stays in the atmo-
sphere, where it causes global warming. CO2
is a greenhouse gas. That is, its presence in the
atmosphere traps heat energy from the sun.
This keeps the climate warm enough for life to
continue.
As atmospheric CO2 levels increase from natu-
ral levels the climate becomes warmer, chang-
ing the natural balance in most parts of the
world. This has a wide range of major disrup-
tive impacts on the environment, natural re-
sources and human communities throughout
the world. Living things consist largely of water
and molecules containing carbon. When fuels
derived from living things such as wood or fos-
sil fuels (oil, coal or natural gas) are burned, the
carbon combines with oxygen to form CO2 that
is released into the atmosphere. People haven’t
thrown the natural carbon cycle out of balance
by burning fossil fuels (Figure 3). More CO2 is
now entering the atmosphere than can be natu-
rally absorbed, contributing to global warming.
Chemical and physical
characteristics of CO2
CO2 is one of the gases in our atmosphere,
being uniformly distributed over the earth's
surface at a concentration of about 0.033 %
or 330 × 10–6 . Commercially, CO2 finds uses as
a refrigerant (Figure 2, dry ice is solid CO2), in
beverage carbonation, and in fire extinguish-
ers. Because the concentration of carbon di-
oxide in the atmosphere is low, it is impracti-
cal to obtain the gas by extracting it from air.
Most commercial carbon dioxide is recovered
as a by-product of other processes, such as
the production of ethanol by fermentation and
the manufacture of ammonia. Some CO2 is ob-
tained from the combustion of coke or other
carbon-containing fuels. Carbon dioxide is re-
leased into our atmosphere when carbon-con-
taining fossil fuels such as oil, natural gas, and
coal are burned in air. As a result of the tremen-
dous world-wide consumption of such fossil
fuels, the amount of CO2 in the atmosphere has
increased over the past century, now rising at
a rate of about 1 × 10–6 per year. Major changes
Figure 3: Image courtesy of CO2 CRC, with values of carbon fluxes and sinks sourced from NASA Earth Science Enterprise and the
International Energy Agency [12].
ACCEPTED

Likar, J., Žarn, J., Grøv, E., Marolt Čebašek, T., Likar, A.
242
RMZ – M&G | 2015 | Vol. 62 | pp. 237–254
in global climate could result from a contin-
ued increase in CO2 concentration. In addi-
tion to being a component of the atmosphere,
carbon dioxide also dissolves in the waters of
the oceans. As carbon dioxide dissolves in sea
water, the equilibrium is established involving
the carbonate ion, CO3
2-. The carbonate anion
interacts with cations in seawater. According
to the solubility rules, »all carbonates are insol-
uble except those of ammonium and Group IA
elements.« Therefore, the carbonate ions cause
the precipitation of certain ions. For example,
Ca2+ and Mg2+ ions precipitate from large bodies
of water as carbonates [2]. Carbon dioxide does
not exist in liquid form at atmospheric pressure
at any temperature. The pressure-temperature
phase diagram of CO2 shows that liquid carbon
dioxide at 20 °C requires a pressure of 30 bar
(Figure 4). The lowest pressure at which liquid
CO2 exists is at the triple point, namely 5.11 bar
at –56.6 °C. At the critical point (31.1 °C, 73 bar
– located upper right in the phase diagram for
CO2), the temperature and pressure at which
the liquid and gaseous phases of a pure stable
substance become identical.
The high pressures needed for liquid CO2 re-
quire specialised washing machines. Clothing is
immersed in liquid CO2 in a highly pressurised
cylinder and agitated by high-velocity fluid jets
to remove soils, then dried in a high-velocity
spin cycle. Liquid CO2 has drawn high marks
in Consumer Reports’ tests for its cleaning re-
sults, and it is environmentally-friendly as it
produces no chlorinated pollutants.
Practical CO2 storage capacities
The theoretical CO2 storage capacity represents
the mass of CO2 that can be stored in hydrocar-
bon reservoirs assuming that the volume occu-
pied previously by the produced oil or gas will
be occupied in its entirety by the injected CO2.
The effective CO2 storage capacity represents
the mass of CO2 that can be stored in hydro-
carbon reservoirs after taking into account in-
trinsic reservoir characteristics and flow pro-
cesses, such as heterogeneity, aquifer support,
sweep efficiency, gravity override, and CO2 mo-
bility. However, there are also extrinsic criteria,
which need consideration when implementing
CO2 storage in oil and gas reservoirs on a large
scale and that further reduce the CO2 storage
capacity in oil and gas reservoirs to practical
levels. The storage capacity of oil reservoirs
undergoing water flooding is significantly re-
duced, making it very difficult to assess their
CO2 storage capacity in the absence of detailed,
specific numerical simulations of reservoir per-
formance. It is very unlikely that these oil pools,
and generally commingled pools, will be used
for CO2 storage, at least not in the near future.
The low capacities of shallow reservoirs, where
CO2 would be in the gas phase, make them un-
economic because of storage inefficiency [3]. On
the other hand, CO2 storage in very deep res-
ervoirs could also become highly uneconomic
because of the high cost of well drilling and of
CO2 compression, and the low ‘net’ CO2 storage
(CO2 sequestered minus CO2 produced during
Figure 4: Pressure-Temperature phase diagram for CO2 and three phases of CO2
[30].
ACCEPTED

CO2 temporary storage in big underground caverns
243
compression). Thus, the pressure window of
9 MPa to 34.5 MPa is considered as being eco-
nomic for CO2 storage in depleted hydrocar-
bon reservoirs [3], which roughly translates to a
depth interval of 900 m to 3 500 m. In terms of
CO2 storage capacity, most reservoirs are rela-
tively small in volume, and have a low capacity
for CO2 storage, rendering them uneconomic.
On the other hand, associated oil and gas reser-
voirs (oil reservoirs with a gas cap) have a CO2
storage capacity that is equal to the sum of the
individual capacities of each reservoir. Consid-
ering the size of the major stationary CO2 sourc-
es, it is most likely that only reservoirs with
large CO2 storage capacity would be considered
in the short and medium terms. Building the
infrastructure for CO2 capture, transportation
and injection is less costly if the size of the sink
is large enough and if its lifespan is long enough
to justify the needed investment and reduce
the cost per ton of sequestered CO2. Thus, only
reservoirs with individual CO2 storage capacity
greater than 1 Mt CO2 per year were selected at
the end of the capacity assessment process.
Storage mechanisms of super-critical CO2
It weighs like a liquid and flows like a gas. The
CO2 would generally be injected underground
as a so-called supercritical fluid. The somewhat
alarming term ‘super-critical’ simply means
that the CO2 has a liquid-like density and flows
like a gas, and with a decrease in pressure will
expand to form a gas without a phase transi-
tion. The CO2 density would still be less than
water. The viscosity an inverse measure of how
well the CO2 flows would be typically less than a
tenth of the brine resident in the rock. CO2 can-
not burn or explode; the only reaction that it
can undergo in the subsurface is the precipita-
tion of a solid. The injected CO2 would migrate
to the top of the rock layer because of buoyancy
forces. Real interest is the long term of trapping
the CO2 for hundreds to thousands of years,
it is imperative that the CO2 could not escape.
There are four principal ways in which the CO2
is prevented from reaching the surface such as
cap rock. Structural or stratigraphic trapping
refers to low-permeability layers of rock (cap
rock) that prevent the upwards movement of
CO2. Similar traps have held oil and gas under-
ground for millions of years. The traps are com-
prised of salt, shale or clays: they need not be
completely impermeable but have pore spaces
that are so small that the CO2 has insufficient
pressure to enter. In well characterised forma-
tions, this is a good way to ensure storage. For
instance, in Sleipner, the use of periodic seismic
surveys (using sound waves to image the sub-
surface) have shown that the injected CO2 rises
to the top of the aquifer and then spreads out
underneath low permeability cap rock layers at
the top. However, if CCS is to be applied on a
global scale, some storage sites may not be as
well characterised as major oil and gas produc-
ing basins such as the North Sea. In this case
another approach is required in case the cap
rock contains gaps or fractures or is absent.
Dissolution
Over hundreds to thousands of years, the CO2
would dissolve in the formation brine forming a
denser phase that would sink. CO2 at high pres-
sure has a reasonably high solubility in water,
although this solubility decreases as the brine
becomes more saline, as an example, a 6 % so-
dium chloride solution almost twice as salty
as sea water would dissolve approximately
30–40 kg/m3 of CO2 at temperatures of 80 °C
and pressures of 10 MPa, representative of a
reservoir at a depth of around 1 000 m where
heat from the earth’s core makes it hotter than
near the surface. While this is promising, the
dissolution of CO2 is a slow process, mediated
by molecular diffusion and the flow of the dens-
er CO2 laden brine. Simulation studies indicate
that it takes hundreds to thousands of years for
a significant fraction of the CO2 to dissolve in
typical reservoir settings [31, 32].
Reaction
The CO2 dissolved in brine forms a weakly acid-
ic solution that may react over thousands to
millions of years with the host rock, forming
solid carbonate. This is a complex geochem-
ical process but in essence, oxides in the rock
dissolve and then re-precipitate as carbonate.
The opposite can also occur, in that the acidic
brine dissolves part of the rock, increasing the
volume of the pore space and the permeability.
The speeds, extents and natures of these reac-
ACCEPTED

Likar, J., Žarn, J., Grøv, E., Marolt Čebašek, T., Likar, A.
244
RMZ – M&G | 2015 | Vol. 62 | pp. 237–254
tions depend principally on the mineralogy of
the rock. Dissolution and precipitation both
render the CO2 less mobile over time. The stor-
age security increases over hundreds to thou-
sands of years. The problem is that these are
slow processes: in the worst case, by the time
a significant fraction of the CO2 has dissolved,
much of the CO2 may already have escaped to
the surface.
Capillary trapping
The final process, which is more rapid, is capil-
lary trapping. This occurs when water displaces
CO2 in the pore space. Figure 5 shows this pro-
cess coupled with dissolution at the field scale
and illustrates CO2 trapped at the pore scale.
Figure 5 shows the increasing storage effec-
tiveness for CO2 with depth and in the critical
depth CO2 is in gaseous state (balloons), below
critical depth it is in liquid-like state (drop-
lets). Volumetric relationship shown by blue
numbers (e.g. 100 m3 of CO2 at surface would
occupy 0.32 m3 at a depth of 1 km). Simulation
studies of CO2 storage have emphasised the
importance of this mechanism. This process
is well established in the oil industry. Water is
used to displace oil from reservoirs but typical-
ly only around half the oil is recovered as much
remains trapped in the pore space. Further
water injection simply leads to excessive recy-
cling of water from injection to the production
wells with little or no further oil recovery this
is why three barrels of water are recovered for
every barrel of oil on average. The CO2 would be
trapped when it is displaced by water due to a
regional movement of groundwater or when a
buoyant CO2 plume migrates upwards. Recent
work has suggested that pumping out saline
water (brine) from the aquifer and then re-in-
jecting would enhance this natural process,
leading to the proposal of an injection scheme
where CO2 and brine are injected together fol-
lowed by chase brine. The idea is to design
injection so that all the CO2 is trapped during
the injection phase, making significant leakage
very unlikely [31, 32].
Pressure responses
In the oil industry there is a net removal of fluid
from the subsurface. The pressure in the reser-
voir drops and the rock, water and hydrocarbon
expand to fill the space vacated by hydrocar-
bon. In most reservoirs, the natural expansion
of rock and water surrounding the reservoir is
insufficiently fast to prevent a very rapid drop
in pressure. To compensate for this, to maintain
pressure and push the oil out, water is usually
injected hence the comments on water produc-
tion in the preceding paragraph. In gas fields
this is unnecessary, simply allowing the pres-
sure to decrease allows the gas to expand and
be produced. The obvious storage solution is to
inject CO2 to replace the oil and gas produced
in old hydrocarbon fields. This has some ad-
vantages i.e. the field must have a good cap
rock in order to contained the hydro-carbon
for millions of years and so safe storage would
be possible, the injection of CO2 can enhance
Figure 5: Density of CO2 versus ground depth and Different types of CO2 trapping [12].
ACCEPTED

CO2 temporary storage in big underground caverns
245
oil and gas production, giving some economic
pay-back, next there is a pipeline infrastructure
in place for injection. The injected CO2 would
cause the reservoir pressure to rise again, re-
placing the volume of produced hydrocarbons.
The main disadvantage is that the extra pro-
duction causes more CO2 to be burnt when ex-
tra oil and gas is produced typically at least as
much CO2 as is stored. The CO2 displaces brine
and the increased fluid pressure tends to ex-
pand the pore space, pushing the rock apart. If
the fluid pressure is too high, this can fracture
the rock, creating cracks through which the CO2
could escape. This squeezing of the subsur-
face also leads to regional pressure increases,
which again could cause extensive fracturing,
or the seepage of salty water to displace fresh
water, contaminating drinking water supplies.
The experience of Sleipner and other sites
where large volumes of CO2 have been injected
without significant increases in pressure pro-
vides evidence that large aquifers do have sub-
stantial storage capacities. Some indications
have shown there are huge volumes in which
the pressure can be dissipated. In reality the
amount of CO2 that can be stored is a function
of how the injection is engineered, how many
wells are drilled, what sort of wells and wheth-
er or not brine is produced. The storage design
depends on economics and the field properties,
and so it is usually unrealistic to talk of a single
capacity estimate.
Dynamic capacity
The first consideration is injectivity, or dynam-
ic storage capacity. This means, can the CO2 be
injected at the rate required in the single well.
The use of additional wells or horizontal wells
through layers of high permeability come with
an additional cost but would allow CO2 to be
injected more rapidly [33]. Current field experi-
ence indicates that a single well can readily in-
ject up to 1 Mt of CO2 per year but more than
one injection well would therefore be required
for large storage projects, especially if CO2 is
collected from several sources before injection.
Static capacity
Large, regionally-extensive aquifers almost
certainly have sufficient storage capacity even
under rather modest constraints on pressure
increase. The second concern is the extent of
the CO2 itself, since this indicates the potential
footprint for any escape. Simulation studies
suggest than in highly heterogeneous systems,
the lowest storage capacity is around 2 % of the
pore space. As the pore space itself is around
25 % of the rock volume, this represents around
0.5 % of the total rock volume, which is similar
to the capacity estimated using pressure con-
straints. The storage capacity and storage se-
curity could be improved, through improved
injection design. If it is known that there is a
good cap rock (such as in hydrocarbon reser-
voirs), CO2 could be allowed to accumulate
under the cap rock, where it could occupy the
majority of the pore space. In the oil industry,
it is standard practice to inject gas and water
together or in alternating slugs, as the mobility
of the combination of the two phases has a low-
er mobility than CO2 alone, leading to a more
stable displacement and a more efficient sweep
of the reservoir. In contrast, CO2 alone has a
very high mobility (low viscosity) and tends
to rise to the top of the reservoir and channel
along high permeability channels. The results
of a simulation study on a North Sea aquifer in-
dicated that, with only a short period of brine
injection, the vast majority of the CO2 could be
capillary trapped, ensuring permanent storage
[11, 34]. It is never possible to guarantee that a cap
rock will be impermeable to CO2, or that the
permeability would be sufficiently high to al-
low rapid injection. Compressibility is defined
as the fractional change in volume for a unit
increase in pressure. When CO2 is injected at
high pressure, it compresses the resident brine
and pushes the rock apart, increasing the pore
volume. The combination of rock and brine has
a compressibility of around 10–9 Pa–1. An aqui-
fer at a depth of 1 000 m would typically have a
pressure of around 10 MPa to avoid fracturing
it would be wise to limit the pressure increase
to between 10 % and 50 %. Hence the pres-
sure should increase by no more than 1 MPa
to 5 MPa. This leads to a fractional change in
volume of the order 1–5 × 10–3 or 0.1–0.5 % a
regionally extensive aquifer some 100 km long
and 100 km wide with permeable layers of a
total thickness of 1 km has a total rock volume
of the order of 104 km3 or 1013 m3. Using typi-
cal density of 600 kg/m3, this would allow the
ACCEPTED

Likar, J., Žarn, J., Grøv, E., Marolt Čebašek, T., Likar, A.
246
RMZ – M&G | 2015 | Vol. 62 | pp. 237–254
storage of 6 Gt to 30 Gt of CO2 but application
of CCS at a global scale for several decades;
it would still need to store CO2 in many large
aquifer units around the world.
Temporary CO2 storage in existing
or new underground caverns
Temporary storage of CO2 in underground facil-
ities requires detailed analysis of all influencing
factors, which include in addition to the geo-
logical structure of the area and engineering
geological, hydrogeological and geotechnical
evaluation includes depth below the ground
surface and not least information about the
inhabited environment. Temporary storage of
CO2 has a practical goal in the case finding out
a location of proper sound rock mass between
the place of CO2 capture and the compressed
and final storage phases. In Figure 6 the pro-
posed location of temporary storage is possible
between the previous mentioned primary and
final technological procedure of long term stor-
age of CO2.
For such storage the specific conditions laid
down by the goal of optimal CO2 pressure and
temperature should be taken into account to
achieve the appropriate density during its
storage. Based on the phase diagram of CO2
(Figure 4) the gas pressure has to be calculated
from 80 bar to 90 bar at temperatures between
10 °C and 15 °C [35]. In practice there may be
other combinations of temperature and pres-
sure which depend of the natural conditions of
a potential storage area. In these decisions, it is
necessary to have sufficient reliable data of the
rock environment including projected depth of
storage, natural rocks temperature, and finally
information about possible seismic activity.
Existing types of underground caverns
Man-made cavities
Great Britain has a very long history of mining
and there are very few minerals that have not
been worked underground at some stage in the
past. Coal mining was by far the most extensive
but metal mining formerly also covered large
areas. However, many other minerals, including
oil shale, fireclay, ball clay, fuller’s earth, lime-
stone, building stone, silica sand, fluorspar, bar-
ytes, slate and, notably the evaporate minerals
salt, gypsum/anhydrite and potash have all
been mined to a greater or lesser extent. All
these activities have created underground cav-
erns of varying sizes and shapes over a wide
range of geological settings. For most minerals
this has produced voids, which are unstable,
particularly where early mining methods were
employed [36]. The type of void created and its
suitability for storage use depends on the rock
worked and the type of mining method used.
Modern room and pillar mining is used for gen-
erally flat-lying, sedimentary strata. Typically
25–50 % of the rock is left in the form of square
pillars to provide a permanent support for the
roof. Rock salt mining is a good example. Mod-
ern salt solution mining techniques also have
the capacity to produce stable cavities ideally
suited for certain types of storage.
Salt caverns and abandoned coal mines
Salt occurs in nature either in solid form as
rock salt (halite) in beds ranging from a few
centimetres to hundreds of meters thick, or in
solution as brine. Salt caverns are constructed
in naturally occurring thick salt domes, deep
underground. Salt can be found in almost every
part of the world with some exceptions around
the Pacific Rim. Salt caverns are a proven me-
dium for hydro-carbon storage as salt acts as a
natural sealant, trapping the natural gas inside
the cavern. Salt caverns for gas storage use are
formed with a leaching process by pumping
hot water to dissolve the salt and removing the
resulting brine via a single well, which then
serves for gas injection and withdrawal. The
storage capacity for a given cavity volume (sev-
Figure 6: A Schematic Illustrating Carbon Dioxide Capture
and Storage (© British Geological Survey) [12].
ACCEPTED

CO2 temporary storage in big underground caverns
247
eral hundreds of thousands to several million
cubic metres) is proportional to the maximum
operating pressure, which depends on the
depth. Salt caverns are typically much smaller
than depleted gas reservoirs and aquifers, usu-
ally covering only one-hundredth of the acreage
taken up by a depleted gas reservoir. As such,
they are particularly suited for short-term stor-
age of natural gas because of their high deliv-
erability as well as the ability to quickly switch
from injection to withdrawal. As with deplet-
ed gas reservoirs and salt caverns, CO2 stored
in coal mines is inspired by storage projects
for natural gas in abandoned coal mines, the
oldest of which dates back to 1961. One of the
typical cases is the Leyden coal mines, located
near Denver Colorado, which were in operation
from 1903 until 1950, producing 5.4 Mt sub-bi-
tuminous coals from two horizontal seams at
210 m and 225 m depth in the upper Creta-
ceous Laramin formation. The second case is
two abandoned mines converted into natural
gas storage reservoirs, both located in the gas-
sy Hainaut coalfield in southern Belgium [23, 37].
Experts carried out a detailed feasibility study
on using abandoned coal mines for long-term
CO2 storage, with special reference to a Belgian
colliery. CO2 stored in an abandoned coal mine
may exist in the gas phase, in solution in wa-
ter and adsorbed on remaining coal. The stor-
age capacity has been estimated at between
7.5 Mt to 12.5 Mt CO2, which maybe small but
accounts for approximately 3 % to 6 % of the
emission reduction for Belgium required under
the Kyoto agreement. The technical set-up of
an abandoned coal mine storage project is rela-
tively simple. Unlike unminable coal seams, CO2
induced swelling is not an issue here. In fact,
the seams surrounding former mine workings
are naturally stimulated and thus high injec-
tion rates can be achieved. On the other hand,
fractured rock which exists around an aban-
doned coal mine may provide leakage paths
for CO2 which would be unacceptable for a
storage site [23, 37]. The same authors have sug-
gested some special requirements which need
to be met in order to obtain a safe and stable
reservoir with sufficient capacity. Firstly, the
highest level of the mine should be at least 500
m deep, with well-sealed shafts and a tight,
mostly dry cap rock. Secondly, in order to pre-
vent mine flooding, the storage pressure should
be higher than the hydrostatic pressure of the
surrounding strata. This overpressure, typical-
ly around 130 % of the hydrostatic pressure, in
turn places a stringent leak-proof requirement
on the top seal of the reservoir and the existing
shafts.
Existing underground caverns used for different
storages
In some places around the world you can find
many underground caverns over the last six
decades. Some of them were done connected
to military activities, defence regulations and
many other requests in the goal to improve
conditions for storage energetic, water and air
masses. Big advantages were done in Norway,
where all the mentioned time has clearer strat-
egy in which way can underground available
space been used. In the scientific and techni-
cal literature can found many usable technical
solutions including geological and geotechni-
cal assessments of hoisted rock masses in the
goal to find proper economic and environmen-
tal acceptation. It is not unknown, that natu-
ral physical, chemical and geotechnical char-
acteristics of rocks mass in these cases have
played enormous important game because the
hoisted media has more than one influence on
the potential used of available underground
space. In the next chapter high attention will
be paid to Norwegian and other Scandinavian
experiences [9, 36–38].
Self-standing capacity
Most rock mass have a certain self-supporting
capacity, although this capacity may vary with-
in a wide range. An appropriate engineering
approach is to take this capacity into account
when designing permanent support. As for
any type of underground structures the selec-
tion of the site location, orientation and shape
of the caverns are important steps preceding
the dimensioning and the laying out of the un-
derground site. Rock strengthening may, how-
ever be needed to secure certain properties/
specified capacities, the same way as is the case
for any other construction material. The fact
that, the rock mass is not a homogenous ma-
ACCEPTED

Likar, J., Žarn, J., Grøv, E., Marolt Čebašek, T., Likar, A.
248
RMZ – M&G | 2015 | Vol. 62 | pp. 237–254
terial should not disqualify the utilisation of its
self-standing and load bearing capacity. Typ-
ically, rock support application in Norwegian
oil and gas storage facilities consists mainly of
rock bolting and sprayed concrete. The appli-
cation of cast-in-place concrete lining in such
facilities has been limited to concrete plugs and
similar structures and is normally not applied
for rock support purposes. The rock support
measures are typically not considered as con-
tributing to the containment, other than indi-
rectly by securing the rock contour and thus
preventing it from loosening. Furthermore, the
Norwegian tunnelling concept applies widely
as a drained concept, meaning that the rock
support structure is drained and the water is
collected and lead to the drainage system. Thus
the rock support is not designed to withstand
the full hydrostatic pressure in the rock mass
because the self-load-bearing capacity was
applied in the design process. The experience
with large underground caverns was obtained
in Norway during the development of hydro-
electric power schemes for which purpose a to-
tal of 200 underground plants were construct-
ed. Commonly the caverns for power-houses
and hydrocarbon storage were all typically
sized to some 15–20 m width, 20–30 m height
and tens-hundreds meter length. That geomet-
rical data, based on past experiences can be
usable for CO2 storage systems. Various types
of monitoring to follow-up the behaviour of the
rock mass and the support structures are avail-
able and used to document the stability and be-
haviour of the rock mass [38–40].
Identication of design parameters
The locations of the rock caverns are normal-
ly fixed within the design concept and being
based on information gathered during a com-
prehensive pre-investigation phase, however,
depending on the actual rock mass conditions
as encountered during tunnelling in the ap-
proach to the designed and planned location,
relocation of the underground structure may
of course take place. Several underground proj-
ects in Norway have experienced changed lo-
cations and local optimisation to better adapt
to the actual rock mass conditions. It is com-
mon to take into account the next information
relating to:
―rock types and mechanical properties
―characteristics and frequency, spacing of
rock mass discontinuities
―in-situ rock stresses
―groundwater conditions.
During the approach to the planned location
of the cavern(s) the rock mass is thoroughly
mapped, joint systems are observed and char-
acterised, weakness zones are interpreted,
in-situ rock stresses are measured, ground
water is monitored (Figure 7). If these condi-
tions are not in accordance with the expected
and required quality of the rock mass, it may
be conclusively decided to shift the location of
the storage caverns to other adjacent caverns
and tunnels, or make some layout adjustments.
Typically, the final layouts of the caverns, their
locations, geometries, alignments, lay-outs of
the tunnel system and rock support design may
not be finally decided upon until the above in-
formation is obtained from the excavation of
the approaches of access tunnels. Numerical
analyses as well as analytical calculations are
useful tools for the designing and planning of
the caverns. These must of course be verified
during the construction phase by adequate
monitoring and follow-up of the stability of the
underground caverns.
Figure 7: Underground cavern under construction [36].
ACCEPTED

CO2 temporary storage in big underground caverns
249
Assessment of the CO2 pressure
impact of on the walls of caverns
The statically base of the CO2 pressure acting on
the caverns’ walls we can apply for calculating
stress and strain relationships in the lining and
surrounding rock structure. According to the
strain nonlinear softening constitutive model
of practical rock, the pressure tunnel with liner
is analysed. The model is considers the influ-
ence of intermediate principal stress σ2. Stress
distribution laws of surrounding rock plastic
zone of tunnel, the mechanism of load bearing
and acting relationship between surrounding
rock and support are studied. Some import-
ant concepts of the working status of practical
tunnel surrounding rock are obtained: such a
superior certain limit [Smax] of self-support geo-
stress and inferior certain limit [Smin] of support
less tunnel surrounding rock. The relations be-
tween [Smax] and geo-stress, between [Smin] and
geo-stress are given. Calculation shows that
the assumed model agrees well with practical
conditions of the rocks. Analysis shows that the
ideal plastic model and the brittle model are
special cases of the proposed solution.
It is well known that the stability of tunnel sur-
rounding rock is decided by the interaction re-
sults of stresses in the surrounding rock and its
strength, i.e. surrounding rock states. If its sur-
rounding rock is in elastic or plastic state after a
tunnel is driven, the surrounding rock is stable.
However, if its surrounding rock is in a broken
state after the tunnel is driven, the surrounding
rock is unstable. In addition, lots of in-situ ob-
servation data have shown that a broken zone
exists widely in surrounding rocks of tunnels.
Therefore, it can be seen that the thickness of
the broken zone, a geometrical parameter in-
dicating the broken range in the surrounding
rocks of tunnels, can be taken as a compre-
hensive index of stresses in the surrounding
rock and its strength to evaluate the stability
of surrounding rock of a deep tunnel. Kastner’s
solution is often used in elastic-plastic analysis
for surrounding rock of a circular tunnel. It is
well known that Kastner’s formula is based on
an ideal elastic-plastic model. This leads to the
Kastner’s solution and is far away from corre-
sponding actual values in surrounding rock.
Following along the path pioneered by Kastner,
researchers have published different solutions
for surrounding rocks of circular tunnels [41].
However, these solutions are restricted to very
simple material models, such as the simple lin-
ear relationship between stress-strain.
Numerical analysis used
BEM– EXAMINE 2D
For preliminary analysis the stability assess-
ment of unlined rock caverns in more stages
without and with CO2 pressure, the 2D model-
ling was used [42]. The main advantages of the
presented modelling were in the parametri-
cal analysis which shows what influences of
each of them on stability are present. Using
BEM to determine the strength factor of the
host rock mass the calculations were done in
the cases when the caverns with dimensions
W/H 21 m/31 m were empty and in the case
where CO2 was filled with 8 MPa pressure
(Figure 8). Technology of CO2 pumping in to
the caverns in the analysis are not included in
detail because in the presented type of prelim-
inary analysis is unnecessary. The important
question is the factor of safety of the cavern’s
stability explains the real situation enough as
to which can be present in the underground
environment. Results of calculations, which
are shown below, clearly explain that the in-
fluence of CO2 pressure 8 MPa on the unlined
rock cavern wall is very high. At the same case
it is clear that the depth of the cavern location
below ground surface has an important influ-
ence on stability, too. The results of calculation
shows that in the proposed geometry of an un-
derground unlined cavern with no loaded and
loaded with CO2 inner pressure Pa = 8 MPa, the
host rock would not have enough strength in
practically all calculated cases. The geotechni-
cal characteristics of the host rock which were
used in the present calculations are the same as
used in the paper Thermal Behaviour of Rock in
Relation to Underground Gas Storage, prepared
by Ming Lu (2007) [43].
―Horizontal stress ratio = 1.5
―Out of plane stress ratio = 1.7
―E-modulus = 30 GPa
― Poisson’s ratio= 0.28
― Friction angle = 38°
ACCEPTED

Likar, J., Žarn, J., Grøv, E., Marolt Čebašek, T., Likar, A.
250
RMZ – M&G | 2015 | Vol. 62 | pp. 237–254
―Cohesion = 0.5 MPa
―Tensile strength = 0.7 MPa
―Unit weight = 27 kN/m3.
For detailed analysis of the geotechnical sta-
bility of the virtual temporary CO2 storage an-
other geotechnical parameters can be used.
It’s no doubt, that in previous experience in
underground big caverns construction in bet-
ter rock mass conditions, the stability analysis
should not be a part of the problem. Short look-
ing through geological environments which are
possible locations of the future CO2 storages
gave optimistic plans for underground space
used for such types of projects.
The caverns stability analysis was given in
the comparison of calculated strength factors
(FS) for different load and geometric cases. In
the next figures the results of calculations are
shown in the name of the Strength factor, as re-
sults of the mentioned analysis. In the first case
only one cavern was analysed. First analysis of
the single unlined cavern stability was done
for empty available volume space (Figure 8)
and the cavern filled with CO2 (Figure 9). In the
figures the geometry of the proposed cavern is
shown, too. The dimensions which are includ-
ed in the analysis are close to that used in con-
struction practice in Norway.
Based on previous research of primary stress
states at different locations, the coefficient
of primary stress ratio 1.5 is accepted. In Fig-
ures 8, 9, 10, 11 and 12 where the results of 2D
numerical modelling are shown, it can found
that the cavern in the proposed rock mass envi-
ronment is unstable without installation of the
support system. The greater differences exist
between the depths 100 m, 200 m and 300 m
but deeper location i.e. 400 m has no import-
ant influence on the calculated strength factors.
They are similar to the factors, which were
found for the case where the virtual cavern
was at the 300 m depth. In a similar way the 2D
analysis was done for a single cavern filled with
CO2 under 8 MPa pressure. Geometrical and
loading position and the results of analyses for
four different location depths it can be found
in Figure 11. The results, which are shown in
Figure 11, were not looked at as surprising re-
garding those given in the input data. They are
understandable, as the pressure of CO2 in this
case even improved the stability of the cavern.
The case where two caverns are located 58 m
between axis shows that that the unlined cav-
erns are unstable without some support mea-
sures. The stress influence between caverns is
higher in the greater depths. That is easily un-
derstandable because the coefficient of prima-
ry stress state is 1.5 which means that in such
primary conditions the axes distance should be
longer. Calculated SF for the unlined caverns
where only left loaded with CO2 pressure at dif-
ferent depths. Interesting results are shown in
Figure 11 where the effect of CO2 pressure on
the temporary stability of the left cavern is evi-
dent. The main positive influence on stability is
present at the depths below the 100 m for both
cases presented in Figure 11 and Figure 12.
The presented results of 2D calculations are in-
formative. The purpose of this numerical anal-
ysis was to show some limitations which are
necessary when planning or updating existing
underground caverns, which were probably
used for temporary storage of CO2. Indeed, the
geotechnical input parameters that were con-
sidered in this analysis took the pessimistic val-
ues, so that it is possible with a greater degree
of optimism to look at more favourable rock
mass circumstances. For further activities in
this field of underground space used, detailed
plan for the necessary in-depth research and
analysis of real sites, which are potential sites
in the future, would be needed.
Figure 8: Geometry of single cavern with no loading with
CO2 pressure and calculated FS for different depths of cavern
positions.
ACCEPTED

CO2 temporary storage in big underground caverns
251
Specic requirements for lining caverns for
providing storage of CO2
If the CO2 pressure has over 100 bar, when the
cavern is closed and the surrounding rocks
have temperature around 10 °C, the big advan-
tage would not achieved. The main reason is
when the pressure of CO2 is lower than 45 bar
it will act as a gas (gas law for real gases). When
the temperature of the CO2 is the same as the
temperature of surroundings rocks i.e. approx.
10 °C, the equilibrium pressure between liquid
and gas is 45 bar. If CO2 continues to be pumped
into the cavern, the gas will follow the charac-
teristics which are adequate to fluid and will
still continue volume of fluid increases. The
volume of gas above the liquid CO2 is less until
the cavern full of liquid CO2, but the pressure is
still 45 bar. If CO2 pressure is slowly increased
with additional pumping, say from 45 bar to
150 bar, liquid CO2 is indeed a bit compressed,
but the compressibility of fluid is very small, so
the pressure increase from 45 bar to 150 bar
gain just fluid volume [26, 35]. In the case that
the higher pressure CO2 by the offline, faster
pumping gas into the cavern, the balance is
not changed. At the beginning CO2 pressure is
of course somewhat higher but when the gas
is cooled to 10 °C, the equilibrium pressure
would be 45 bar. Of course, it may take sever-
al days / weeks / months for the surrounding
rock environment and CO2 in the caverns to
reach equilibrium. At the beginning, when the
Figure 11: Layout of two unlined caverns, left filled with CO2
with 8 MPa working pressure, right with no inside pressure
and calculated SFs for different depths.
Figure 9: Geometry of the single cavern for CO2 storage with
pressure 8 MPa and calculated FS for a single unlined cavern
located at different depths.
Figure 10: Two unlined caverns not loading with CO2 pressure
and calculated SF for different depths below the surface.
Figure 12: Layouts of two caverns filled with CO2 under
working pressure 8 MPa. Calculated Strength factors for
the unlined two caverns, loaded with 8 MPa CO2 pressure at
different depths.
ACCEPTED

Likar, J., Žarn, J., Grøv, E., Marolt Čebašek, T., Likar, A.
252
RMZ – M&G | 2015 | Vol. 62 | pp. 237–254
pumping gas is warm (25 °C), the pressure
above 45 bar may even have 100 bar or high
value. When the surroundings cool; more and
more the pressure would decrease and in the
end, if time allows, would stabilise at 45 bar
(if there is already some liquid).The density of
liquid at 10 °C should be somewhere around
0.45 kg/l, at the critical temperature it was
0.47 kg/l. The problem could arise if CO2 were
to be warmed at the critical temperature i.e.
31 °C. In that case the CO2 would be changed to
gas which causes increasing pressure because
it would behave as a gas and no longer a liquid.
Conclusions
The underground storage of industrial quan-
tities of carbon dioxide is technically possible,
and CO2 storage both in saline water-filled
reservoir rocks and in oil and gas fields has
reached the demonstration stage. Nonethe-
less, the indications are that underground CO2
storage could have a significant impact on our
greenhouse gas emissions.
Underground storage may involve substantial
construction work, including major surface in-
frastructure provision (access roads, rail links,
pipelines, head works buildings) and not just
restricted to the immediate locality. All of these
may create more or new impacts in terms of
amenities and traffic. Offshore storage may re-
quire new or expanded onshore installations
and infrastructure. Concerns in relation to sta-
bility, pollution and safety would also need to
be addressed.
In some situations underground storage can
be considered as a three stage activity; a short-
term development stage (the ‘temporary’ work
involved in construction of the void and the
associated surface works including infrastruc-
ture); a long-term operational stage (the per-
manent use of the resulting void); and finally a
decommissioning and post abandonment stage
(when the planning impacts arising from the
presence of the facility and the infrastructure
may need relevant considerations for a consid-
erable period of time.
Storage of CO2 underground could reduce
construction costs and offer protection from
storms, accidents, arson, acts of terrorism and
also prevent ‘shrinkage’ (loss by theft). It may
also provide an ideal ambient environment in
terms of stable humidity and temperature and
be dry, reducing energy costs for heating or air
conditioning. However, some storage facilities
may be wet, hot or very dry with issues of air
quality inhibiting access and might raise con-
cerns about managing fire or pollution events.
Ventilation, access and fire escape structures
may be required at the surface. Surface stability
would also be a major planning consideration.
Experience with the use of water curtains at the
three Norwegian air storages discussed herein,
at pressures from 4 MPa to 8 MPa, is encourag-
ing. It has been found that a properly designed
water curtain totally eliminates any gas leakage
from the storage, even for a storage pressure
head that is only twice the thickness of the rock
overburden. A water curtain may provide not
only a cost-effective method for restricting gas
leakage from unlined hard rock caverns; cur-
rently it also appears to be the only practical
way of totally preventing gas leakage from high
pressure storage.
The main advantages of CO2 underground stor-
age in rock mass formation are very wide in-
cluding utilising rock mass properties, environ-
mentally-friendly, protection during wartime,
operation and maintenance, and not least pro-
tection from natural catastrophes.
The rock mass has a number of important pa-
rameters that are utilised in the underground
storage of hydrocarbon products. These capac-
ities allow a variety of storage conditions and
enable a number of diverse types of products to
be stored in unlined rock caverns.
With the current knowledge of the mechani-
cal and thermodynamic behaviour of the rock
mass and the current use of such storage facil-
ities the proven technologies could take place
during the construction process.
As far as the environmental aspects are con-
cerned the experience from Norwegian un-
derground storage projects are unreservedly
positive. So far product leakages have not been
reported at any of these projects indicating
clearly that the applied concept and techniques
for obtaining the required confinement are ap-
propriately proven.
For a subsurface solution, dedicated systems
for collection and handling of various types of
ACCEPTED

CO2 temporary storage in big underground caverns
253
spill could be planned thus limiting the spread
of any spill. Bringing these storage tanks be-
low the surface allows valuable surface areas
to be utilised for other purposes; recreational,
cultural and residential. In addition unsightly
structures can be hidden away underground.
Protection from natural disasters and catastro-
phes such as earthquakes is a beneficial ad-
vantage of underground storage. It has been
acknowledged that subsurface structures have
several intrinsic advantages in resisting earth-
quake motions. Experience and calculations
show this clearly.
The total construction cost would be within the
range of 150–310 USD per m3 storage which
would be many times competitive in the open
construction market.
For any decision about working pressure and
temperature of CO2 a phase diagram should
be used regarding the names and technologi-
cal possibilities. Two potential solutions exist;
first include pressure 80 bar to 100 bar at the
normal rock temperature 8 °C to 12 °C. In this
case additional isolation and a freezing sys-
tem aren’t needed. The second case is close to
CO2 transport parameters at pressure of about
7 bar and temperature –50 °C. The final deci-
sion depends of the financial and technological
closed cycle of the CO2 long-term storage.
References
[1] Bachu, S. (2003): Screening and ranking of sedimen-
tary basins for sequestration of CO2 in geological
media. Environmental Geology; 44 (3), pp. 277–289.
[2] Bachu, S, Lytviak, A. T. (2005): Carbon Dioxide
Storage Capacity in Uneconomic Coal Beds in Alberta:
Potential and Site Identification. Report to Plains CO2
Reduction (PCOR) Partnership, Alberta Geological
Survey, Edmonton, AB.
[3] Bergman, P. D., Winter, E. M. (1995): Disposal of car-
bon dioxide in aquifers in the U. S. Energy Conversion
and Management, pp. 523–526.
[4] Doughty, C., Pruess, K. (2004): Modelling supercriti-
cal carbon dioxide injection in heterogeneous porous
media. Vadose Zone Journal; pp. 837–847.
[5] Hadlow, R. E. (1992): Update of industry experience
with CO2 injection. SPE Paper 24928, presented at
the Sixty Seventh SPE Annual Technical Conference,
Washington D.C.
[6] Hitchon, B., Gunter, W. D., Gentzis, T., Bailey, R. T.
(1999): Sedimentary basins and greenhouse gases:
a serendipitous association. Energy Conversion and
Management; pp. 825–843.
[7] Holt, T., Jensen, J. I., Lindeberg, E. (1995): Under-
ground storage of CO2 in aquifers and oil reservoirs.
Energy Conversion and Management; pp. 535–538.
[8] International Energy Agency Greenhouse Gas R&D
Programme (IEA GHG) (2005): Building the Cost
Curves for CO2 Storage: North America. Cheltenham,
U.K.
[9] Barbo, T. F., Danielsen, S. W. (1980): Bacterial
impair-ment of water curtain between oil storage
caverns in rock. Proc. Intl. Symp. Subsurface Space,
Rockstore 80, Stockholm, Sweden, Pergamon Press,
Oxford, UK. pp. 245–250.
[10] Gale, J., Davidson, J. (2002): Transmission of CO2
– safety and economic considerations. Sixth Inter-
national Conference on Greenhouse Gas Control
Technologies, Kyoto.
[11] Markussen, P. (2002): A CO2-infrastructure for EOR in
the North Sea (CENS). Sixth International Conference
on Greenhouse Gas Control Technologies, Kyoto.
[12] Storage Capacity Estimation, Site Selection and Char-
acterization for CO2 Storage Projects(2008): CO2CRC,
Report No: RPT08-1001, pp. 60.
[13] Hendriks, C. A. (2000): Cost of carbon dioxide
removal by underground storage. Fifth International
Conference on Greenhouse Gas Control Technolo-
gies, Cairns, pp. 967–972.
[14] Concawe (2002): Western European cross-country
oil pipelines 30 year performance statistics, CON-
CAWE report.
[15] Guijt, W. (2004): Analyses of incident data show
US, European pipelines becoming safer. Oil and Gas
Journal; pp.68–73.
[16] Golomb, D. (1997): Transport systems for ocean dis-
posal of CO2 and their environmental effects. Energy
Convers Manage; 38, pp. 279–286.
[17] Odenberger, M., Svensson, R. (2003): Transportation
systems for CO2 - application to carbon sequestration.
MSc thesis, Chalmers University of Technology, Dept.
of Energy Conversion, http://www.entek.chalmers.
se/klon/msc/.
[18] Chandler, H. M. (2000): Heavy Construction Cost
Data - 14th Annual Editions. R. S. Means Company,
Inc. Kingston, MA, USA.
ACCEPTED

Likar, J., Žarn, J., Grøv, E., Marolt Čebašek, T., Likar, A.
254
RMZ – M&G | 2015 | Vol. 62 | pp. 237–254
[19] European Gas Pipeline incident Data Group. (2002):
5th EGIG report 1970–2001 Gas Pipeline Incidents,
document EGIG 02.R.0058.
[20] Ffooks, R. (1993): Natural gas by sea, the develop-
ment of a new technology. Royal Institution of Naval
Architects, London.
[21] Gale, J., Davison, J. (2002): Transmission of CO2 - safe-
ty and economic considerations. GHGT-6.
[22] Hendriks, C. A., Wildenborg, T., Feron, P., Graus,
W. (2005): Capture and Storage, prepared for EC,
DG-ENV, Ecofys Energy and Environment, report Nr.
M70066.
[23] Piessens, K., Dusar, M. (2003b): CO2 sequestration in
abandoned coal mines. Proceedings of the Interna-
tional Coalbed Methane Symposium, Tuscaloosa,
paper 0346, 11 p.
[24] United Nations. (2002): Carbon sequestration––
avoiding CO2 emission from fossil fuels. Ad hoc Group
of Experts on Coal and Thermal Power, pp. 18–19.
[25] Doctor, R. D. (2000): Transporting carbon dioxide
recovered from fossil-energy cycles. Fifth Internation-
al Conference on Greenhouse Gas Control Technolo-
gies, Cairns, pp. 567–571.
[26] Vendrig, M., Spouge, J., Bird, A., Daycock, J., Johnsen,
O. (2003): Risk analysis of the geological sequestra-
tion of carbon dioxide. Report no. 8, Department of
Trade and Industry, London, UK.
[27] Veritec (1988): On-bottom stability design of subma-
rine pipelines. Recommended Practice E305.
[28] West, J. M. (1974): Design and operation of a super-
critical CO2 pipeline-compression system, SACROC unit,
Scurry County, Texas. Society of Petroleum Engineers
Permian Basin Oil and Gas Recovery Conference,
paper SPE 4804.
[29] VipulNaik (2014): Carbon dioxide, climate sensitivity,
feedbacks, and the historical record: a cursory exam-
ination of the Anthropogenic Global Warming (AGW)
hypothesis. http://lesswrong.com/lw/kfx/carbon_di-
oxide_climate_sensitivity_feedbacks_and/.
[30] ASCO Carbon Dioxide Presentation. (2012): ASCO
Product Catalogue. http://www.ascoco2.com/
fileadmin/user_upload/ascojet.com/Dokumente/
ASCO_2015_DE/index.html#.
[31] Sarv, H., John, J. (2000): Technical and economic
aspects of large-scale CO2 storage in deep oceans. 25th
International Technical Conference on Coal Utiliza-
tion and Fuel Systems, Clearwater, Florida.
[32] Sarv, H. (2001): Further technological evaluation of
CO2 storage in deep oceans. 26th International Techni-
cal Conference on Coal Utilization and Fuel Systems,
Clearwater, Florida.
[33] Bock, B. R. (2003): Economic Evaluation of CO2 Stor-
age and Sink Enhancement Options. TVA Public Pow.
Institute.
[34] Skovholt, O. (1993): CO2 transportation system.
Energy Convers Manage; 34, pp. 1095–1103.
[35] UK Department of Trade and industry (2002): Car-
bon Capture and Storage. Report of DTI International
Technology Service Mission to the USA and Canada,
Advanced Power Generation Technology Forum.
[36] Kjørholt, H. (1991): Gas tightness of unlined hard
rock caverns. Doctoral thesis, Norwegian Inst. of
Technology, Trondheim, Norway.
[37] Piessens, K., Dusar, M. (2003a): CO2 geothermic in
abandoned coal mines. Proceedings of the Interna-
tional Coalbed Methane Symposium, Tuscaloosa,
paper 0345, 10 p.
[38] Kjørholt, H., Broch, E. (1992): Hard rock CAES cav-
erns – leakage prediction and prevention. Proc. 2nd
Int. Conf. Compressed and Air Energy Storage, San
Francisco, USA.
[39] Rathe, L. (1975): An innovation in surge-chamber
design. Water Power and Dam Construction, pp.
244–248.
[40] Frøise, S. (1987): Hydrocarbon storage in unlined
rock caverns: Norway’s use and experience. TUST.
[41] Pan, Y., Wang, Z. Q. (2004): Research on relationship
of load-displacement for cavern surrounding rock
with strain nonlinear softening. Journal of Rock
Mechanics and Engineering; pp.1515–1521.
[42] Rocscience; EXAMINE 2D. (2012): https://www.roc-
science.com/help/examine2d/webhelp/pdf_files/
tutorials/Tutorial_01_Quick_Start.pdf
[43] Ming. L. (2010): Rock engineering problems related
to underground hydrocarbon storage. Journal of
Rock Mechanics and Geotechnical Engineering;
pp. 289–297.
ACCEPTED