PETROLEUM SOCIETY

CANADIAN INSTITUTE OF MINING, METALLURGY & PETROLEUM PAPER 2007-124

Basic Investigations on Enhanced Gas

Recovery by Gas-Gas Displacement
A.T. TURTA, S.S.K. SIM, A.K. SINGHAL, B.F. HAWKINS
Alberta Research Council

This paper is to be presented at the Petroleum Society’s 8th Canadian International Petroleum Conference (58th Annual Technical
Meeting), Calgary, Alberta, Canada, June 12 – 14, 2007. Discussion of this paper is invited and may be presented at the meeting if
filed in writing with the technical program chairman prior to the conclusion of the meeting. This paper and any discussion filed will
be considered for publication in Petroleum Society journals. Publication rights are reserved. This is a pre-print and subject to
correction.

Abstract In case a mixture of CO2 and nitrogen displaced the natural

gas, it was observed that there was a delay in CO2
The paper presents basic data on Enhanced Gas Recovery breakthrough, associated with a period when only a mixture of
(EGR) by gas-gas displacement for nearly depleted natural gas methane and nitrogen was produced. This is so because
reservoirs, by injecting waste gases. solubility of CO2 in connate water is considerably higher than
The soundness of the concept of gas-gas displacement for that of nitrogen. This directly leads to a higher gas recovery
enhancing gas recovery was investigated via laboratory due to a longer exploitation period, given the fact that up to
investigations, compositional modeling and economic analyses. 20% nitrogen can be tolerated in the produced stream, as
Paramount Resources is field testing the concept in their opposed to only 1% for CO2 case. For this period of methane
GRIPE Project in the Athabasca region of Alberta, to enhance and nitrogen production, there are no operational problems
production from a gas bearing stratum overlying the oil sand associated with the corrosive nature of CO2.
interval. EGR by gas-gas displacement is seen as a promising way of
This paper is a part of a series of papers presenting results prolonging the productive life and economic recovery of many
of EGR research conducted over a four year period (2003- depleting volumetric gas reservoirs.
2006). The main targets were volumetric (closed) reservoirs, in
advanced phases of exploitation. Results of basic research on
methane displaced from core samples by pure gases (pure CO2
or pure N2), as well as flue gases (mixtures of CO2 and N2) and Introduction
a CO2/ SO2 mixture, are presented.
A series of 13 gas/gas displacement tests in 30cm-long, 4cm Alberta currently has 26,000 gas pools, which are in
diameter, Berea cores were conducted at a temperature of 700C different stages of exploitation. For these gas reservoirs,
and a pressure of 6,200 kPa. Most of the tests were conducted the pools that should be considered first in the
in the presence of connate water, while others were conducted
without connate water (dry cores) to confirm and bench mark
the results by other investigators. The tests on consolidated
cores showed that for pure nitrogen and pure CO2, used as the
displacing medium, the recovery was comparable.

1
implementation of large-scale EGR-CO2 and CO2 storage 2. Review of the Technical Literature
have not yet been identified, and no screening criteria
have yet been developed. Currently, only the concept of
“disused gas reservoirs” has been advanced1 for CO2
2.1 Laboratory Investigations of Gas/Gas
storage. This concept implies that only those reservoirs Displacement
that are in an advanced stage of depletion (with very little The first gas/gas displacement tests were conducted in
marketable gas left), traditionally with extremely low Hungary2 in preparation of an EGR field test. Tests with
current pressure, or which are water invaded, should be long milled rock packs (of low permeability – 100-300
considered. In this context, they have been considered mD) in horizontal position, in the presence of connate
exclusively for CO2 storage and not for enhanced gas water saturation, at a pressure of 2500 kPa, showed a
recovery (EGR). This paper brings a new concept: that of methane recovery between 70% and 90%. The
simultaneous EGR and CO2 storage. temperature of the test was 630C, and at this temperature
CO2 was always gas. For the first time, comparative
There are two classes of gas reservoirs: displacement of methane using pure CO2 and nitrogen
„ Volumetric gas reservoirs (depletion gas was investigated. Moreover, displacement of methane
reservoirs), or closed pools using a mixture of CO2 with 20% methane was also
„ Gas reservoirs under active water drive investigated, and it was found that recovery of methane
(water drive gas reservoirs) increased; some of the methane from the injected stream
For both these classes, the gas reservoir contains connate was also recovered.
water saturation, which as a rule is immobile, and is not
produced with the gas, except by evaporation. More recent laboratory studies to evaluate feasibility of
displacing methane by CO2 (both liquid and supercritical)
In the first category, the pore volume containing gas have shown promising results3. While using 1” x 1 ft dry
remains constant over the duration of exploitation, hence carbonate cores (no connate water) held in horizontal
a ‘closed’ reservoir. Gas recovery is accompanied by a position, with pressures in the range of 500-3000 psi
decrease in static reservoir pressure, and the ultimate (3448-20,685 kPa) and temperatures in the range of 70-
recovery depends upon abandonment pressure. Recovery 1400F (21-600C), methane recovery at the CO2 break-
may be as high as 70-85% original gas in place (OGIP), through was in the range of 73% to 87%; recovery was
with little or no water production. higher at higher pressures.

On the other hand, for many water drive reservoirs, at 2.2 Composition of CO2 –Containing Gas
abandonment, the reservoir pressure may still be high. Streams to be Used for EGR
For these reservoirs, as reservoir pressure declines below Composition of CO2 streams from different sources may
a certain level, water enters the reservoir. The water vary, and may contain impurities such as methane,
influx in many situations is almost equal to gas sulphur dioxide (SO2) and/or nitrogen. The CO2/N2
production under reservoir conditions, leading to pressure mixture refers mainly to flue gases, while the mixture of
stabilization; in this case the stabilized pressure is the CO2/ SO2/ N2 refers to the incinerated acid gases. Unlike
abandonment pressure. The water drive can be a lateral CO2 miscible flooding for oil reservoirs, where a
water drive (edge water), or a bottom water drive. In both relatively pure CO2 is a must, for gas reservoirs this
cases, the water encroaches into production wells usually condition is not critical; streams containing 12-70% CO2
in the lower parts of the pay interval. Also, in both cases, can be injected and stored. Gas streams from industrial
the volumetric sweep efficiency is low and the ultimate sources such as hydrogen plants, sweet gas processing
gas recovery is also usually low (as low as 50%-60% plants, petrochemical plants, pulp mills and sour gas
OGIP), due to both relatively low sweep, and the trapped processing plants (SGPP) may be considered. For the
(occluded) gas in the water-invaded zone. A relatively SGPP, both the injection of combusted acid gases
high value of the stabilized pressure leads to trapping of a (CO2/SO2 mixtures) and of CO2/ H2S mixtures can be
large amount of gas in the water-invaded zone, resulting considered.
in a lower recovery factor at the time of abandonment.
Upon significant water encroachment, these pools are The stream composition is important; streams composed
usually abandoned. of CO2 and methane are significantly more advantageous
for EGR and storage, as compared to streams of CO2 and
nitrogen (flue gases) or incinerated acid gases. This is so,
because when injecting these streams in gas reservoirs, an
in situ methane-CO2 separation takes place while the
mixture flows through the porous medium; due to
radically different gas solubility in water (4-30 times
higher for CO2 depending on temperature and pressure),
CO2 is partially retained in the water, while methane

2
continues to flow along. This way, some of the methane first case is that of UGS in the Cerville-Vellaine, (France)
from the injected stream can be produced with the aquifer reservoir with a total storage (inventory) capacity
original gas, while the gas mixture stored in the reservoir of 1450 million sm3 of gas, out of which 700 million was
will be enriched in CO2. the work gas (750 million sm3 – base gase)8. In that case,
a cushion gas of nitrogen of 150 million sm3 was
Laboratory tests show that when injecting a 33%/67% injected, representing 10% of the total capacity or 20% of
CO2/CH4 mixture in a water saturated porous media at 15 the base gas volume. In this sandstone reservoir, there
MPa and 400C (injection up to a complete CO2 saturation were a total of 38 (production/injection) wells, and the
of the water), the gas left in porous media has a CO2 conditions related to the mixing were characterized by
content of 40%, while the produced gas contained only field tests involving a tracer. The hydrogen-tracer test
23% methane.4,5. Thus, in practical terms, the gas lasted a few months and 12 wells (out of which 8 were
reservoir becomes a CO2-enriched gas reservoir. Based injectors) were involved; from this test a reliable
on this phenomenon, a patent aimed at CO2 –methane kinematic dispersion coefficient was determined, and
separation by injecting the stream into an aquifer was then used in the simulation of the nitrogen injection for
proposed6 . the building of the cushion gas.

In conclusion, for EGR and CO2 storage, while using In the UGS operations, the crucial idea is to avoid an
streams of CO2 with methane, the economics of CO2 intensive mixing between cushion gas and the work gas.
storage lead to: For this reason, cushion gas injection wells are placed far
¾ Increase of original gas recovery factor away from production wells and also, the cushion gas is
¾ Production of a part of the methane emplaced at the lower part of the reservoir. For example
from the CO2-methane stream along in the Cerville-Vellaine case the minimum distance
with enhanced recovery of the original between cushion gas injection wells and the closest
gas production wells is at least 2 km, while the injection
¾ For the injection of streams with higher points are at least 40 m lower than the production points.
than 70% CO2, the reservoir at terminal
conditions contains higher CO2 The second case is that of UGS in the Saint-Clair-Sur-
concentration than the stream initially Epte, France9, aquifer reservoir with a total storage
injected into the reservoir. (inventory) capacity of 550 million sm3 of gas. In that
case, a cushion gas consisting primarily of flue gas of 60
This may be very attractive when streams of CO2 and million sm3 was injected, representing approximately
methane are available and their separation is not 10% of the total capacity. Two injection wells located at
economic; this applies equally to natural as well as the lower part of structure were used to this effect. At the
industrial streams. Besides these incentives, for industrial end of the flue gas injection period, 6.4 million m3 (10%
streams, there may also be incentives related to CO2 of the cushion gas) of air was injected; however, no
storage/elimination of CO2 emission in atmosphere. traces of oxygen appeared in the production wells. The
reservoir temperature is 450C. In this reservoir a total of
2.3 Underground Gas Storage 11 injection/production wells and 17 observation wells
Displacement of natural gas by an inert gas such as were used in this UGS operation. For more than 10 years
nitrogen or CO2 to form the so-called “cushion gas” has there were no problems (including mixing problems)
been applied to a limited extent in underground gas during cyclic operation of this storage site.
storage (UGS)7-10 . A reservoir used for UGS contains
two gases: work gas and base gas. The work gas is
2.4 EGR Field Piloting
produced and injected seasonally, while the base gas is To the best of our knowledge, there has been just one
not; base gas is expanding or shrinking as the work gas is published field test of EGR. This test was conducted
produced or injected. The use of less expensive gases during 1986-1994 in the Hungarian field Budafa
such as nitrogen, flue gases, CO2, etc to substitute all or Szinfelleti, a weak water drive sandstone reservoir of 5-
part of the base gas requirement in UGS has been 40 mD permeability2. Here, EGR started when the natural
successfully tested in France7, and this substitute gas gas recovery was 67% OGIP and the injected gas was an
constitutes the cushion gas. The cushion gas is placed far impure CO2 stream, consisting of 80% CO2 and 20% CH4
away from the injection/production (I/P) zone, in such a from an adjacent natural CO2 pool.
way that cushion gas is never produced by
injection/production (I/P) wells; there is a significant The incremental gas recovery represented 11.6% OGIP,
distance (portion) between the group of I/P wells and the or 35% of the gas in place at the initiation of CO2
cushion gas location. injection. CO2 break-through occurred 1.5 years after the
start of injection; the distance between injection and
By 1985, Gaz de France were operating 9 UGS sites. Out production wells was 500 m.
of those cases, two typical ones will be presented. The

3
Given the fact that volumetric sweep efficiency during
gas/gas displacement in the reservoir is much less than Table 1: Physical Properties of Berea Core Sample
100% and there is significant reservoir heterogeneity, it is
Length (cm) 30.48
not surprising that the Hungarian field test yielded a gas
Diameter (cm) 3.81
recovery of only 35% GIP as compared to 70% GIP
obtained in the laboratory core floods. Obviously, gas Air permeability (mD) 500
reservoirs with lower heterogeneity should be preferred Porosity (%) 25.0
as candidates for EGR. Evaluation of heterogeneity Pore volume (ml) 72.4
coefficient may be based on core permeability variations
or environment of deposition pattern. One could also use Produced gas volume and composition were measured
gas tracer propagation between adjacent wells for using a wet test meter and a HP 6890 gas
determining it, as in case of underground gas storage. chromatographic system. The core pressure was set at
approximately 6200 kPa with a back pressure regulator.
Therefore, in all categories of gas reservoirs, all else Pressure and temperature were recorded in specified
being the same, preference would be for relatively intervals using a National Instruments data logger in
homogeneous and with some dip, where a gravity stable conjunction with a personal computer. Narrow bore
CO2 displacement could occur. The density of CO2 is 3-9 (1/16” O.D.) stainless steel lines were used where
times higher than that of CH4, which is significant11. possible to reduce the volume between the outlet end of
the core sample and the gas sampling loop of the gas
chromatographic system. This arrangement allowed more
3. Investigations of EGR Mechanisms: accurate determination of the breakthrough of injection
Laboratory Set up and Procedure gas.

The objectives of the laboratory investigations were (1) to

determine methane recovery during its displacement with 3.2 Experimental Procedures
CO2, N2, and their mixtures, and (2) to determine effect
Nitrogen, methane and carbon dioxide gases (99.5%+
of flow rates on recovery efficiency. All the
purity) were supplied by Praxiar Inc. Injection gases of
investigations were performed on core with connate
specified composition were prepared by mixing
water saturation, as this is a normal state in all gas
appropriate amount of gases in a pistoned cylinder.
reservoirs. However, in order to compare our results with
Composition of the resulting mixtures was analyzed by
other similar results from the literature, some tests were
gas chromatography. The volume of methane gas
conducted with dry core (without connate water
contained in the Berea core sample was determined with
saturation), although this does not represent conditions in
a combination of the following two methods.
a gas reservoir.
(1) Gas expansion method:
3.1 Equipment Set-up After the core sample was saturated with methane gas at a
specified pressure and temperature, the core sample was
The experimental apparatus consists of four main depressurized to atmospheric pressure, and the volume of
components: the injection system, core holder, production gas released was determined with a gasometer.
system and data recording system (Figure 1).
(2) Cumulative Volume of Methane Produced:
The volume of gas produced during the entire
The injection system consisted of several floating piston displacement was logged with the wet test meter. The
cylinders connected to an HPLC water pump. The methane content of gases collected in each time interval
cylinder containing the injection gas was placed inside a (approximately 7 minutes) were also determined from gas
constant temperature oven. During the core flood test, chromatographic analyses. Using this information, the
water was injected into the cylinder containing the cumulative volume of methane gas produced during the
injection gas at a specified flow rate, displacing the entire test was calculated. This value should be a good
injection gas into the core sample. approximation of the volume of original gas in place.

The displacement test proceeded as follows: the core

A Berea core measuring 30.4 cm long with 3.8 cm
sample was evacuated and then fully saturated with
diameter was used for all tests. Physical properties of the
methane gas at a pressure of 6200 kPa and 70 oC. A
core sample are presented in Table 1. The core was
stable pressure drop was then established by injecting
confined within a lead sleeve and mounted in the core
methane gas into the core sample at a specified flow rate.
holder.
After the injection pressure had stabilized, the pressure of
the injection gas was adjusted to the same value as the

4
injection pressure. In this manner, when the injecting gas
was changed from methane to the displacing gas (such as
N2 or CO2), disturbance of the displacement front due to
abrupt change in pressure drops was minimized.
Table 3: Summary of Test Results (dry core)
The core sample was fully saturated with 3% sodium
chloride solution at a pressure of 6200 kPa and 70 oC.
Methane was then injected into the core at a slow rate Test # Injection gas Methane Methane
displacing the brine into a window cell where the composition recovery at Recovery with
cumulative volume of brine displaced was determined. B/T (1% 10%
After the core sample had reached irreducible brine contamination) contamination
saturation, the flow rate of methane gas was set to the (%OGIP) (%OGIP)
target value to be used during the test. Once again, the 3 N2 69 81
stabilized injection pressure was then used to set the 4 CO2 67 77
pressure of the injection gas and to minimize disturbance 7* N2 47 76
of the displacement front at the start of the test. During 8* CO2 48 64
the displacement, the composition of the effluent gas was
9 CO2 64 76
analyzed every 5-7 minutes. The test was terminated
when the effluent contains less then 0.2% of methane. *Velocity reduced

Effect of displacing gas

4. Investigations of EGR Mechanisms: Tests 3 and 4 were conducted to compare the
Laboratory Results displacement efficiency of nitrogen and carbon dioxide
under similar pressure, temperature and flow rate
4.1 Displacement Tests without Irreducible conditions (Figures 2 and 3). During the displacement of
Brine Saturation methane with nitrogen, the methane recovery at
The first tests were conducted in dry Berea core. A breakthrough (arbitrarily defined as methane recovery
summary of the test conditions is presented in Table 2. when the produced gas contains 1% impurity) was
Key results of the tests are included in Table 3. approximately 69% original gas in place (OGIP). When
carbon dioxide was used to displace methane, the
The main objectives of this series of tests were to breakthrough recovery was approximately 67% of OGIP.
investigate the effect of injection gas composition The viscosity of carbon dioxide and nitrogen gas (0.021
(nitrogen versus CO2). In addition, the test results can be and 0.020 centipoises respectively) were both higher than
used for comparison with published data reported by that of methane (0.0134 centipoises) under the test
Mamora and Seo3, since their experiments were also conditions. Thus, the mobility ratio should be favourable
conducted in dry core. for both displacements.

Table 2: Summary of Test Conditions (dry core) Test results show that the displacement efficiency of pure
CO2 and nitrogen were quite similar under the test
Test # Inj. Temp Injection Viscosity conditions (displacement conducted in a horizontal
Gas gas orientation). Generally, our tests confirmed the results
Density obtained in almost similar conditions as those reported by
o Mamora and Seo3.
C kg/m3 mPa.s
CH4 70 37.54 0.0134
3 N2 70 60.2 0.0208
4 CO2 70 121.9 0.0197 4.2 Displacement Tests with Irreducible
7* N2 70 58.5 0.0207 Brine Saturation
8* CO2 70 120.8 0.0195 After completing five displacement tests with dry core
9 CO2 100 102.8 0.0206 samples, four tests were conducted with irreducible brine
*Velocity reduced saturation in the ranges of 17-25%. The volume of
irreducible brine saturation in the core sample was
obtained from an average of two values. The first value
was deduced from the volume of brine displaced out of
the core sample during methane displacement of the fully
brine saturated core. A second value was determined by
comparing the volumes of original gas in place in the
core sample with and without irreducible brine saturation.

5
In general, the differences between the two values are brine in the water-wet Berea core tends to occupy the
less than 5 percent. narrower flow paths and smaller pore space, only the
larger channels are opened for gas flow. Second, the
A summary of the test conditions is presented in Table 4. breakthrough of CO2 is delayed due to dissolution of the
Key results of the tests are included in Table 5. leading edge of the CO2 displacement front along the
small-scale higher permeability pathways. Thus, the
presence of irreducible brine saturation had a net effect of
Table 4: Summary of Test Conditions (with irreducible reducing the small-scale heterogeneity of the core and
brine) consequently resulted in a more efficient displacement.
Test Injection Temp Injection Viscosity
# Gas gas Table 6: Effect of Irreducible Brine Saturation on
density Displacement Efficiency
0
C kg/m3 mPa.s
Test Inject Swi Methane Methane
# ion recovery at 1% Recovery
11 CO2 70 120.9 0.0197 Gas contamination at 10%
12* 14% CO2 70 66.38 00205 contamination
in N2 % % OGIP % OGIP
13 14% CO2 70 65.40 0.0204 8 CO2 0 48 64
in N2 11 CO2 18 61 71
18* CO2 70 123.77 0.0197 18 CO2 19 62 70
*Velocity reduced Legend: Swi = Irreducible water saturation

Displacement with flue gas:

Table 5: Summary of Test Results (with irreducible brine)
In order to evaluate the feasibility of using flue gas to
Test Injectio Swi Methane Methane enhance gas recovery, two displacement runs, Tests 12
# n Gas recovery @ Recovery and 13, were conducted using a synthetic gas mixture
1% @ 10% containing 14% CO2 and 86% N2. It is important to
contamina- contamina- mention that we could not locate any similar tests in the
tion tion open literature, to which to compare our tests results. Our
% % OGIP % OGIP tests seem to be the first reported. Therefore, their
11 CO2 18 61 71 efficiency was compared with that of a basic pure gas
12* 14% 18 66 76 displacing a pure gas.
CO2 in
N2
13 14% 20 75 82 The results were very encouraging. As shown in Table 7,
CO2in methane recovery at N2 breakthrough was 66% when flue
N2 gas was used to displace methane. These results
18* CO2 19 62 70 compared favourably with the breakthrough recovery of
61-62%, obtained by using pure carbon dioxide. In order
*Velocity reduced
to examine and compare the results more closely, the
Legend: Swi = Irreducible water saturation
experimental data from Tests 11 and 12 were plotted
together in Figure 4. It can be seen that, in Test 11, the
displacing gas (carbon dioxide) had its breakthrough at
0.6 pore volume (PV) injected, while the nitrogen from
Effect of irreducible brine saturation
the displacing gas (Test 12) broke through at around 0.66
The results of these three tests are presented in Table 6. It PV injected. The carbon dioxide component of the flue
can be seen that, for the two tests conducted in the gas broke through at a much later time due to its higher
presence of irreducible brine saturation, namely Tests 11 solubility in water (at 0.81 PV).
and 18, the methane recovery is quite similar (61%
versus 62%). On the other hand, the methane recovery
from Test # 8, which was conducted in the absence of Also, from the graphs of Figure 8, it can be seen that in
irreducible brine, was substantially lower. fact the exploitation of production wells can be easily
extended up to 10-20% N2 in the produced stream, as
significant amounts of CO2 exist only when the N2
There are reasons for improved displacement efficiency content exceeds 30%. If up to 20% N2 in the produced
in the presence of irreducible brine. First, the irreducible stream is allowed, than methane recovery increases up to

6
84-87%; the difference between 61-62%, for pure CO2 • The tests with dry cores showed that for
use, and 84-87% recovery (at 20% N2 in the produced pure gases used as displacing medium, the
stream) for flue gas injection, is very significant. recovery is very similar.
• When using CO2 as a displacing agent it
was found that the recovery was higher in
the presence of irreducible water saturation
The relatively high displacement efficiency of the flue than in its absence.
gas can significantly improve the economics of the EGR • Gas recovery is always better for flue gases,
process due to its relative abundance and lower cost as compared to the recovery when using pure
compared to pure CO2. In addition, the delay in the CO2 or nitrogen. This is due to the
significant contrast in solubility of gases
production of CO2 when flue gas was used (as the (from mixtures) in connate water; solubility
displacing agent) means that operators may worry less of CO2 is considerably higher than that of
about corrosion in the production wells. In field practice, nitrogen and methane, and causes a delay in
allowable concentration of nitrogen in the sales gas CO2 break-through, which is directly
stream is in fact much higher than that for CO2 (about associated with higher gas recovery and
2%), thus use of flue gas for EGR could mean that the fewer operational problems associated with
produced gas could contain much higher amount of the corrosion nature of CO2.
• The advantages related to the use of gas
impurity. mixtures, in particular flue gases -
containing high contrast-solubility gases in
Table 7: Comparison of Displacement the mixture – can also be utilized in
Efficiency Between Flue Gas and CO2 optimization of EGR operations, as well as
Test Injection Swi Methane Methane Methane optimization of CO2 storage.
# Gas recovery at Recovery Recovery at
1% with 10% 20% N2
contamina- contamina- contamina-
tion tion tion
% % OGIP % OGIP % OGIP
11 CO2 18 61 71 -
12* 14% CO2 18 66 76 84
in N2
13 14% CO2 20 75 82 87
in N2
18* CO2 19 62 70 -
*Velocity reduced
Note: At 20% N2 in the effluent gas, CO2 content is still
less than 1% vol.

Conclusion
1. In the technical literature there is little information on
Enhanced Gas Recovery (EGR) by gas/gas displacement.
Previous investigations in this area have been reported only in
USA and Hungary. In USA some preliminary tests investigated
the efficiency of methane recovery by CO2 displacement in
cores, in the absence of connate water saturation. In Hungary
some laboratory tests investigated the efficiency of methane
recovery by CO2 and nitrogen displacement in long sand-pack
systems, in the presence of connate water saturation, in
preparation of a field EGR test. Actually this field test – which
constitutes the only EGR field test completed and reported so
far - was performed during 1986-1994 in a weak water drive gas
reservoir and showed a gas recovery of 11.6 % original gas in-
place (OGIP); the CO2 break-through occurred after 1.5 years.

2. A comprehensive study of EGR by gas/gas displacement

was undertaken. A series of 9 gas/gas displacement tests in 1ft-
long, 1.5in diameter cores were performed at a temperature of
700C and a pressure of 6200 kPa. Most of the tests were
conducted in the presence of connate water, while a few were
done on dry cores. From the results of these tests the following
conclusions were made:

7
References 11) Oldenburg, C.M, et al.: “Mixing of CO2 and CH4 in
1) Fisher, L., Sloan, T., and Mortensen, P. : “Costs for Gas Reservoirs: Code Comparison Studies” Energy and
Capture and Sequestration of CO2 in Western Fuels, 15, 293-298, 2002.
Canadian Geologic Media” Presented at 53rd Annual
Technical Meeting of the Petroleum Society, June
11-13, 2002, Calgary, Canada

2) Papay J: “Improved Recovery of Conventional

Natural Gas". Published in the German Magazine
Erdoel, Erdgas, Kohle; Part I: Theoretical
Discussion of Recovery Methods. NO 6, PP 302-
308, January 1999 and Part II: Results of a Pilot
Test; NOS 7-8, PP 354-355, July -August 1999

3) Mamora, D.D, and Seo, J.G.: “Enhanced Gas

Recovery by Carbon Dioxide Sequestration in
Depleted Gas Reservoirs” Presented at SPE Annual
Technical Conference and Exhibition, September 29-
october 2, 2002, Houston, USA

4) Cucuiat, M.I., Turta A T. et al. : “Non-Conventional

CO2 Sources for Enhanced Oil Recovery”.
International Symposium on CO2 Enhanced oil
Recovery, March 8-11, 1983, Budapest, Hungary

5) Turta A T. and Cucuiat, M.I.: “Recovery of Oil from

an Oil Reservoir by CO2 Miscible Displacement”
USA Patent 4,343,362, August 10, 1982

6) Cornelius, A.J. and Needham, R.B.: “CO2 Removal

from Hydrocarbon Gas in Water Bearing
Underground Reservoir” USA Patent 4,187,910,
February 12, 1980

7) Misra, B.R., Foh, S.E., Shikari, Y.A., Berry, R.M. and

Labaune, F.: “The Use of Inert Gas in Underground
Natural Gas Storage” SPE 17741, SPE Gas
Technology Symposium, Dallas June 13-15, 1988

8) Laille, J.P., Coulomb, C. and Tek, M.R.:

“Underground Storage in Cerville-Vellaine, France: A
Case History in Conversion and Inert Gas Injection as
Cushion Substitute” SPE 15588, presented at 61st Annual
Technical Conference of SPE, New Orleans, LA October
5-8, 1986

9) Laille, J.P., Molnard, J.E. and Wents, A.:”Inert Gas

Injection as Part of the Cushion of the Underground
Storage of Saint-Clair-Sur-Epte, France “SPE 17740,
presented at SPE Gas Technology Symposium, Dallas,
TX, June 13-15, 1988

10) Carriere J.F., Fasanino G. and Tek M.R.: ”Mixing in

Underground Storage Reservoirs” presented at 60st
Annual Technical Conference and Exhibition of SPE, Las
Vegas, NV, September 22-25, 1985

8
 Core holder
BPR
1 footBerea
BereaCore
core

1/16” lines to GC
Oven Sampling loop

Gas

Temp= 700 C
Pressure = 6000 kPa Wet Test
Meter

Figure 1: Schematic Diagram of Enhanced Gas Recovery Coreflood Test

9
 100 10

90 9

C u m u la tiv e C H 4 P r o d u c e d (S L )
80 8
% CH4

% C H 4 a n d N 2 in g a s ,
70 7
% N2

C H 4 re c o v e ry
60 % CH4 recovery 6

50 Cum CH4 (L) 5

40 4

30 3

20 2

10 1

0 0
0.0 0.5 1.0 1.5 2.0
Pore Volume
Figure 2: Decrease of methane concentration and increase of displacing gas (N2) concentration in
the produced gas for the test 3

10
 100 5.0

90 4.5

% C H 4 a n d C O 2 in p ro d u c e d g a s , 80 4.0

C u m u la tiv e p ro d u c e d C H 4 (S L )
C H 4 R e c o v e ry (% O G IP )
70 3.5

% CH4
60 3.0
% CO2
50 2.5
CH4 recovery
40 2.0
Cum CH4

30 1.5

20 1.0

10 0.5

0 0.0
0.00 0.50 1.00 1.50 2.00

PV injected

Figure 3: Decrease of methane concentration and increase of displacing gas (CO2) concentration in the
produced gas for the test 4

11
 100

90 % CH4 (with CO2)

80 % CO2 (with CO2)
70
% CH4 Recovery
60 (with CO2)
% N2 (flue gas)
50
% C1 (flue gas)
40

30 % CO2 (flue gas)

20 CH4 recovery (flue

gas) (%OGIP)
10

0
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00
PV

Figure 4: Comparison of displacement efficiency for CO2 and flue gases (test
11 versus test 12)

12