Geotech Geol Eng (2015) 33:1009–1016

DOI 10.1007/s10706-015-9882-8

ORIGINAL PAPER

A Laboratory Investigation of Permeability of Coal

to Supercritical CO2
V. Vishal . T. N. Singh

Received: 4 January 2015 / Accepted: 8 April 2015 / Published online: 11 April 2015
Ó Springer International Publishing Switzerland 2015

Abstract The kinetics of gas–coal interaction during were varied between 11 and 15 MPa at a range of
coalbed methane (CBM) recovery and/or carbon 16–24 MPa confinements, and N2 was used as a
dioxide sequestration in coal has been subject of relatively neutral medium to estimate the loss in
investigation over past few years. Swelling of coal permeability due to supercritical CO2 flow. The results
matrix due to gaseous phase CO2 injection is now well indicate that high adsorption of supercritical CO2 in
established through laboratory experiments and field coal led to significant reduction in permeability. On
validations. Further, fluid exchange or flow in coal increasing the confining pressure, further decline in the
alters effective stresses in the underground environ- permeability was recorded. CO2 becomes liquid-like
ment. These significantly affect the permeability with increasing confining pressures and the coal–fluid
characteristics of coal and therefore influence the gas interactions change, causing high sorption and matrix
recovery/injection projects. Most research works on swelling, leading to reduced permeability in coal. This
coal seam sequestration have been carried out for gas or explains why substantial decline in the injection rate of
liquid phase CO2. Considering the pressure–tem- CO2 was observed progressively in most CO2 sinks.
perature conditions of deep seated coal, studies are
now being done on supercritical CO2 flow and Keywords Supercritical carbon dioxide 
adsorption in coal. A newly developed experimental Permeability  Carbon sequestration  Bituminous
set up was utilized to replicate the underground coal  India
conditions in laboratory to investigate the (1) initial
N2 permeability of coal, (2) supercritical CO2 perme-
ability of coal, (3) effects of CO2 sorption on N2 1 Introduction
permeability of coal. The temperature of the set up was
maintained at 33 °C while CO2 injection pressures India has one of the largest expanses of coal deposits in
the world. The lower Gondwana stratigraphic se-
quence of Permian age contains coal rich Raniganj and
V. Vishal (&) Barakar measures, and account for the majority of coal
Department of Earth Sciences, Indian Institute of
Technology Roorkee, Roorkee 247667, Uttarakhand, reserves in India. Raniganj, Barakar and Jharia coal-
India fields are large coalfields with huge reserves of good
e-mail: vikram12july@gmail.com quality coal. Production of CBM from these coalfields
has started, and a fairly high amount of gas is being
T. N. Singh
Department of Earth Sciences, Indian Institute of produced from these coalfields. Recent numerical
Technology Bombay, Mumbai 400076, India simulation studies on Indian coal have indicated

123
1010 Geotech Geol Eng (2015) 33:1009–1016

possibilities of CO2 injection for enhanced recovery of Carboniferous coals (approx. 99 %) and Tertiary
methane (Vishal et al. 2013a, 2015a). Few studies have coals. The Permo-Carboniferous coals have a high
also been carried out on shale from this region, also gas content and have mostly been explored for their
using numerical modelings (Vishal et al. 2015b). Deep CBM potential. Most of these coals are located closer
seated coal deposits are also being viewed as potential to the large point sources of emission of anthropogenic
locales for storage of CO2 for greenhouse gas mitiga- CO2 and may be potential targets for storage of the
tion. Inferior grade/non-mineable coals, with a suitable greenhouse gases (GHG), with or without enhanced
trapping mechanism, are prime targets for CO2 storage. CBM recovery. There are coalfields where coals occur
It is important to understand the coal–fluid interaction at relatively shallow depths, which allow for open cast
at such depths for long term, safe and secure disposal of mining of the commodity, while at other places, the
greenhouse gases. coal seams are quite deep. The depth range may vary
At depths greater than 800 m, injected CO2 is likely from 100 m to over 1000 m. It is believed that the
to behave like supercritical fluid inside the coal seam. coal-seam sequestration of CO2 may be targeted in
The transition of CO2 from a subcritical to supercritical coals ranging from 300 m and deeper. If the potential
state occurs at 7.38 MPa and 31.8 °C (Suehiro et al. or sink capacity of coal seams deeper than 1200 m
1996). The fluid behaviour as well as the response of (below the current limit of mining) is established, the
coal to supercritical CO2 is quite different, and hence, deeper seams could offer huge sink space to the
research based on real term down-hole conditions for anthropogenic greenhouse gas.
CO2 sequestration in coal mass must be encouraged. It is The coal samples in this study were collected from
expected that coal matrix swelling over longer satura- Barakar Formation of Jharia basin. It is a sickle shaped
tion period under supercritical CO2 will influence its coalfield truncated with a major boundary fault on the
permeability. Several field results have revealed the southern flank. It is a part of the East–West aligned
decline in CO2 permeability with time and this Damodar–Koel group of Gondwana basins of India.
highlights the need of understanding coal response About a 3000 m thick succession of Lower Gondwana
under various scenarios of CO2 injection. Field test rocks are preserved that outcrop over an area of nearly
results from Ishikari basin in Japan showed a reduction 465 km2. The sedimentary succession unconformably
of nearly 70 % of injected CO2, from the first year to the overlies the Archaean basement, beginning with the
second year of injection (Fujioka et al. 2010). A similar Talchir Formation, followed upward by fluvial and
scenario was observed in Williston basin coal seam in fluvio-lacustrine successive sediments of Barakar,
North Dakota, where in coal matrix swelling had led to Barren Measures and Raniganj Formations deposited
approximately a 10 time-decline in CO2 permeability within an intracratonic extensional setting. The
(Botnen et al. 2009). The CO2 injection rate reduced by geological map of Jharia basin is shown in Fig. 1. A
nearly 50 % during the initial years of CO2 injection in brief geological outline of Jharia is mentioned in
the San Juan Basin (USA) (Reeves 2001). If the Table 1 (after Chandra 1992). The litho-units are
disposition of these target coal seams were worked almost horizontal with a dip amount varying up to 8°.
out, it would indicate that the variation in coal perfor- Coal samples from seam no. XIII were collected as
mance depends not only on coal characteristics but also drill cores and used for permeability tests. The core
on the fact that CO2 transits from one phase to another sample with a length to diameter ratio of 2 was
during sequestration. Studies related to supercritical polished and prepared for tests as shown in Fig. 2.
CO2 permeability in coal are few and there are not many
published literature on such behaviour of Indian bitu-
minous coal, which otherwise offers considerable CO2 3 Experimental Methodology and Permeability
storage potential at varying depths. Estimation

Experimental set ups with triaxial confinements are

2 Field Works and General Geology of the Area used in testing of geomaterials to incorporate the
down-hole conditions, in terms of overburden pres-
On the basis of mode of origin and period of sure, confining stresses, temperature, etc. (Shukla et al.
deposition, Indian coals are classified as Permo- 2010; Vishal et al. 2013b). In this study, coal core

123
Geotech Geol Eng (2015) 33:1009–1016 1011

Fig. 1 Geological map of Jharia Basin highlighting the study area (Chandra 1992)

Table 1 General geology Age Formation Dominant lithology

of Jharia coalfield (Chandra
1992) Recent Weathered Sandy soil, clay, alluvium, gravel
Unconformity
Jurassic Igneous flow and Dolerite dykes, lamprophyres, etc.
intrusions
Upper Permian Raniganj Arkosic sandstones, shales and coal seams
Middle Permian Barren Measure Shales, carbonaceous shales and buff
coloured sandstone
Lower Permian Barakar Thick coal seams along with shales,
carbonaceous shales, grits and
buff coloured sandstones
Upper Carboniferous Talchir Greenish shales and fine grained sandstones
Archean Basement Metamorphics

sample was loaded in a holder and confining pressure pressure (or lowering the downstream pressure). This
was applied. The system was then charged with test pulse decays over time as the differential pressure
gas (N2 and CO2) and adequate time was given to causes fluid to flow through the sample. Data was
allow upstream and downstream pressures to equili- recorded throughout the process and used in conjunc-
brate. A pulse was then created by raising the upstream tion with known system volumes to calculate gas flow

123
1012 Geotech Geol Eng (2015) 33:1009–1016

were maintained during the tests to replicate the

deeper down-hole conditions. Further, the set up was
stabilized at 33 °C, well above the critical point
temperature for CO2, ensuring a complete super-
critical CO2 flow across the sample. N2 was initially
passed through coal at different injection pressures to
set the base value of N2 permeability of the sample. In
the next stage, supercritical CO2 was injected into the
coal at different inlet pressures. The downstream
conditions were thoroughly monitored to ensure a
complete supercritical CO2 flow. A series of N2
injection was followed in the same range on inlet
pressure, and permeability changes were determined.
The experimental conditions are shown in Table 2.

4 Results and Discussion

Fig. 2 Coal core specimen from seam no. XIII in Jharia
coalfield used for permeability tests using N2 and supercritical
It is important to investigate the influence of fluid
CO2
phase of CO2 on permeability of coal. Figure 4 is the
unicomponent phase diagram for carbon dioxide.
rates and permeability. A simple schematic of the set Siriwardane et al. (2009) investigated the influence
up conceptualized, fabricated and installed in Rock of liquid and gas states of CO2 on permeability of
Science and Rock Engineering laboratory is given in Pittsburg coal as a function of confining pressure and
Fig. 3. time while Vishal et al. (2015c) compared the effects
The changes in pressure was obtained and used for of liquid and supercritical CO2 flow in Indian coal.
calculation of the flow rate using the following Massarotto et al. (2010) conducted laboratory ex-
equation as also applied in previous studies (Vishal periments to assess changes in reservoir properties
et al. 2013c): following exposure to supercritical CO2 and water.
dP Depending upon the depth of the burial of target coal
Q¼ bV: ð1Þ seam and the regional geothermal gradient, it is
dt
expected that the injected phase of CO2 may slip into
where Q is the downstream flow rate, dP/dt is the rate liquid or supercritical phases and the response of coal
of change in the downstream pressure with time, V is to such phases may not be the same as earlier studies
the downstream volume and b is the adiabatic on gaseous CO2. Thus, in this study, the flow
compressibility. The corresponding permeability (K) characteristics of Indian coal in the supercritical phase
for each coal specimen under each condition for the of CO2 were considered and discussed in detail.
particular phase of CO2 was calculated using Darcy’s
equation: 4.1 N2 Permeability of Coal
2Qpo lL
K¼
: ð2Þ In the first set of experiments, N2 was used as a
A p2i  p2o
relatively non-sorptive medium and passed through
where Q is the downstream flow rate, l is the viscosity the specimen to determine the original permeability of
of injected fluid, po is the downstream pressure, pi is the coal specimen. This was the base value of original
the upstream pressure, L and A are the mean length and N2 permeability of coal for comparison with the
cross-sectional area of the coal specimen, second stage of N2 flow during the experiments. It was
respectively. expected that the passage of supercritical CO2 through
In this study, a series of tests were run on a coal core the sample would alter the physical structure of the
specimen under triaxial conditions. High pressures coal and hence, the changes in N2 permeability would

123
Geotech Geol Eng (2015) 33:1009–1016 1013

Fig. 3 A simple schematic of the experimental set up

Table 2 Test conditions for nitrogen and supercritical carbon

dioxide injection through the coal specimen
Confining Supercritical CO2 N2 injection
pressure (MPa) injection pressure (MPa) pressure (MPa)

16 11 5
12 7
13 9
14 11
15 13
20 11 –
12 Fig. 4 P–T Phase diagram for carbon dioxide
13
Downstream pressure development began almost
14
instantly and the permeability of coal was determined
15
for all cases of injection. It was observed that N2
24 11 –
permeability of coal increased with increasing injec-
12
tion pressure. N2 has extremely low adsorption in coal
13
and does not induce swelling in coal matrix (Harpalani
14
and Chen 1995). Therefore, injection of N2 at higher
15
pressures caused a reduction in effective stresses
which eventually led to the widening of fractures and
the resultant high permeability. N2 permeability tested
be established when tested after the flow of CO2 at 16 MPa confining pressure, increased from
through the coal. Figure 5 shows the initial N2 0.0071 mD at 5 MPa injection pressure to
permeability of coal at different injection pressures. 0.0088 mD at 12 MPa injection pressure.

123
1014 Geotech Geol Eng (2015) 33:1009–1016

Fig. 5 Variation of permeability of N2 is linear with increase in

injection pressure Fig. 6 Supercritical CO2 permeability in different confining
pressures, an overall reduction of permeability takes place
4.2 Supercritical CO2 Permeability of Coal which may be attributed to coal matrix swelling (legend
indicates different confining pressures)

Carbon dioxide was passed through the coal specimen

at higher injection pressures to obtain supercritical other fractures. Closure of cleats due to matrix
flow. Four different injection pressures were used in swelling may take place even during CO2 injection
three sets of confining pressure conditions and the as also observed by previous works (Fokker and van
permeability in each case was determined (Table 2). der Meer 2004; Vishal et al. 2015d). Bae and Bhatia
The rate of change in pressure with time was recorded (2006) argued that the volume of CO2 adsorbed onto
with each case of supercritical CO2 injection. Tem- the matrix surface of coal increases with increasing
perature of the set-up was kept constant at 33 °C. This pressure, which causes a larger degree of matrix
was to ensure that CO2 passing through the coal mass swelling and hence, reduced permeability. On increas-
remained well above its phase transition temperature ing the confining pressure, CO2 permeability of coal
and pressure. The downstream pressure started devel- decreased further. This may be attributed to the
oping soon with the injection of CO2. However, it development of higher effective stresses on coal,
took nearly an hour for CO2 pressures in coal to which hinders the flow of CO2. Large volumes of CO2
equilibrate. The estimation of permeability was done get preferentially adsorbed on to the fractured and
for the part of experiment where the (dP/dt) curves micro porous surfaces of coal. While the permeability
remain linear. According to review of White et al. reduced with increasing injection pressures initially,
(2005), supercritical CO2 has a higher potential to there was no major change in coal permeability in the
displace the existing gases from coal while Krooss last phase of injection. The reason for this pattern may
et al. (2002) found that coal had a higher adsorption be that the permeability reduction due to minimal
affinity to supercritical state of CO2 than its subcritical swelling is compensated for, by the dilation due to
state. Therefore, changes in coal permeability were increase in inlet pressure.
expected due to the flow of supercritical CO2.
Figure 6 shows different permeability values for 4.3 N2 Permeability of CO2-Treated Coal
the tested coal specimen at three confining pressures.
It can be observed that the permeability of coal using In the final stage of this study, once the supercritical
CO2 is nearly 60 % less than N2 permeability for the CO2 permeability for coal was determined, N2 was
same injection pressure of 11 MPa. According to again passed through the sample at injection pressures,
Figs. 5 and 6, an increase in N2 injection pressure the same as before, to estimate the loss in permeability
leads to an increase in permeability by nearly 5 % of coal. Nitrogen was used to compare the reduction in
while CO2 injection leads to permeability reduction by flow due to flow of supercritical CO2.
around 10 %. The initial decrease in CO2 permeability Figure 7 shows N2 permeability of coal measured
of coal may be related to the coal matrix swelling in the final stage. It is evident that original N2
which occurs during CO2 flow through the cleats and permeability undergoes significant decrease due to

123
Geotech Geol Eng (2015) 33:1009–1016 1015

insitu pressure and temperature conditions. This study

was carried out to investigate the flow of supercritical
CO2 Indian bituminous coal. It was found that the
supercritical CO2 permeability of coal was nearly
60 % less as compared to its original N2 permeability
for the same value of injection pressure. An increase in
confining pressure caused further reduction in perme-
ability, this implies that higher injection pressures
would be needed for sequestration of CO2 in deeper
coal seams. Initial increase in CO2 injection pressure
led to reduction in coal permeability and may be
Fig. 7 N2 permeability of coal tested after the flow of attributed to the swelling of coal matrix in the presence
supercritical CO2 through the sample, where marked reduction of supercritical CO2, even during the period of
is observed from the original permeability value (confining
pressure = 16 MPa) injection. On contrary, N2 permeability increased
linearly with injection pressure implying that it was
not significantly adsorbed during flow. The original N2
permeability undergoes a reduction of nearly 25 % at
5 MPa injection pressure from pre-CO2 flow to post-
CO2 flow phase. The change in reduction, although,
declines, as N2 offers partial recovery of the loss in
permeability. A part of the change in permeability
would also be on account of the changes in confining
pressure from high to less and reverse during the
experiments.

Acknowledgments This research was conducted as a part of

Fig. 8 Percentage reduction in N2 permeability of coal before the DST INSPIRE Faculty Award Grant (IFA-13-EAS-07). VV
and after treatment with supercritical CO2 at different injection is thankful to the Department of Science and Technology,
pressures (confining pressure = 16 MPa) Government of India, New Delhi for the research grant.

the flow of CO2. Close to 25 % reduction is observed

for the first injection pressure value. This reduction in References
N2 permeability is mainly due to shrinkage of flow
paths in the coal. The magnitude of reduction as shown Bae J, Bhatia SK (2006) High-pressure adsorption of methane
in Fig. 8, decreased as the loss in permeability was and carbon dioxide on coal. Energy Fuel 20:2599–2607
Botnen LS, Fisher DW, Dobroskok AA, Bratton TR, Greaves
partly recovered during the flow of N2 as explained by KH, McLendon TR et al (2009) Field test of CO2 injection
Kiyama et al. (2011). Yet, under all circumstances, N2 and storage in lignite coal seam in North Dakota. J Energy
permeability of coal increased with increase in Procedia 1(1):2013–2019
injection pressure, which shows that minimal or no Chandra D (1992) Jharia coalfield. Mineral Resources of India,
Geol Soc India, Bangalore, pp 1–149
swelling was induced in coal due to N2 flow. Fokker PA, van der Meer LGH (2004) The injectivity of coalbed
Therefore, this study highlights the influence of high CO2 injection wells. Energy 29(9, 10):1423–1429
magnitude reduction in permeability of coal under Fujioka M, Yamaguchi S, Nako M (2010) CO2-ECBM field
supercritical CO2 conditions, which must be addressed tests in the Ishikari coal basin of Japan. Int J Coal Geol
82:287–298
during sequestration of CO2 in coal. Harpalani S, Chen G (1995) Estimation of changes in fracture
porosity of coal with gas emission. Fuel 74(10):1491–1498
Kiyama T, Nishimoto S, Fujioka M, Xue Z, Ishijima Y, Pan Z
5 Conclusions et al (2011) Coal swelling strain and permeability change
with injecting liquid/supercritical CO2 and N2 at stress-
constrained conditions. Int J Coal Geol 85(1):56–64
CO2 when injected underground changes its phase Krooss BM, van Bergen F, Gensterblum Y, Siemons N, Pagnier
from subcritical to supercritical depending on the HJM, David P (2002) High-pressure methane and carbon

123
1016 Geotech Geol Eng (2015) 33:1009–1016

dioxide adsorption on dry and moisture equilibrated Vishal V, Ranjith PG, Singh TN (2013b) CO2 permeability of
Pennsylvanian coals. Int J Coal Geol 51:69–92 Indian bituminous coals: implications for carbon seques-
Massarotto P, Golding SD, Bae JS, Iyer R, Rudolph V (2010) tration. Int J Coal Geol 105:36–47
Changes in reservoir properties from injection of super- Vishal V, Ranjith PG, Pradhan SP, Singh TN (2013c) Perme-
critical CO2 into coal seams—A laboratory study. Int J ability of sub-critical carbon dioxide in naturally fractured
Coal Geol 82(3):269–279 Indian bituminous coal at a range of down-hole stress
Reeves S (2001) Geological sequestration of CO2 in deep, un- conditions. Eng Geol 167:148–156
mineable coal beds: an integrated research and commer- Vishal V, Singh TN, Ranjith PG (2015a) Influence of sorption
cial-scale field demonstration project. In New Orleans, LA, time in CO2-ECBM process in Indian coals using coupled
USA: Society of Petroleum Engineers Annual Technical numerical simulation. Fuel 139:51–58
Conference and Exhibition Vishal V, Jain N, Singh TN (2015b) 3-D modeling of
Shukla R, Ranjith PG, Haque A, Choi X (2010) A review of propagation of hydraulic fractures in shale at different in-
studies on CO2 sequestration and caprock integrity. Fuel jection pressures. Sustain Environ Res 25(4):1–8
89(10):2651–2664 Vishal V, Sudhakaran A, Singh TN, Pradhan SP (2015c) Liquid
Siriwardane H, Haljasmaa I, McLendon R, Irdi G, Soong Y, and supercritical CO2 permeability of bituminous coal at
Bromhal G (2009) Influence of carbon dioxide on coal multiple downhole stress conditions. Int J Coal Geol
permeability determined by pressure transient methods. Int (submitted for publication)
J Coal Geol 77(1–2):109–118 Vishal V, Ranjith PG, Singh TN (2015d) An experimental in-
Suehiro Y, Nakajima M, Yamada K, Uematsu M (1996) Critical vestigation on behaviour of coal under fluid saturation,
parameters of xCO2 ? (1 - x)CHF3 for x = (1.0000, 0.7496, using acoustic emission. J Nat Gas Sci Eng 22:428–436
0.5013, and 0.2522). J Chem Thermodyn 28:1153–1164 White CM, Smith DH, Jones KL, Goodman AL, Jikich SA,
Vishal V, Singh L, Pradhan SP, Singh TN, Ranjith PG (2013a) LaCount RB et al (2005) Sequestration of carbon dioxide in
Numerical modeling of Gondwana coal seams in India as coal with enhanced coalbed methane recovery: a review.
coalbed methane reservoirs substituted for carbon dioxide Energy Fuel 19(3):659–724
sequestration. Energy 49:384–394

123