SPE 103228
CO2 Transport Mechanisms in CO2/Brine Coreflooding
R.B. Grigg, SPE, and R.K. Svec, SPE, New Mexico Inst. of Mining and Technology
Copyright 2006, Society of Petroleum Engineers
This paper was prepared for presentation at the 2006 SPE Annual Technical Conference and
Exhibition held in San Antonio, Texas, U.S.A., 24–27 September 2006.
This paper was selected for presentation by an SPE Program Committee following review of
information contained in an abstract submitted by the author(s). Contents of the paper, as
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at
SPE meetings are subject to publication review by Editorial Committees of the Society of
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
for commercial purposes without the written consent of the Society of Petroleum Engineers is
prohibited. Permission to reproduce in print is restricted to an abstract of not more than
300 words; illustrations may not be copied. The abstract must contain conspicuous
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.
Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.
Abstract
Concerns for the possible consequences of increased
greenhouse gas emissions will undoubtedly increase carbon
dioxide (CO2) injection into oil reservoirs for enhanced oil
recovery (EOR) and probably into other geological formations
for sequestration.1 The development of CO2 plumes and
subsequent dissolution into formation brine are essential
mechanisms in most sequestration scenarios and could aid in
understanding long-term injectivity reduction in EOR WAG
(water alternating with gas injection) projects. This paper
describes laboratory tests on sandstone and carbonate core
samples. Two types of displacement tests were performed; gas
injection to a residual brine saturation with respect to gas,
followed by brine injection to a residual gas with respect to
brine. The level of CO2 saturation in the injected brine at
reservoir pressure and temperature was varied from zero to
over 90% saturation. Sandstone and limestone rock types were
tested. This variation in CO2 saturation in the injected brine
was to determine the effect on the CO2 saturation or plume
size in the core. This information can be used in CO2-EOR-
WAG projects and carbon sequestration into geological
formations.
By injecting CO2 into brine-saturated Frio sandstone the
brine saturation was reduced to 68–71% in each of five tests.
In each test, over 90% of the reduction occurred with less than
0.33PV of CO2 injected, with very little additional brine
production after 0.5PV of CO2 injected. During brine
injection, CO2 production was equivalent to the rate expected
from brine saturated with CO2 at the reservoir conditions. This
indicates that for Frio core at high irreducible brine saturation
at the tested flow rate (~6 ft/day), the CO2 plume was reduced
through dissolution, not displacement. With increasing CO2
saturation in the injected brine, the brine volume required to
remove (dissolve) the CO2-plume increased proportionally.
Results will be used to aid in predicting injectivity in CO2-
EOR-WAG operations and CO2 plume migration and
dissolution in EOR and sequestration. The principle
mechanisms for CO2 saturation reduction are identified.
Introduction
Residual CO2 saturation is suspected to be a significant factor
for reducing injectivity during many water alternating with gas
(WAG) processes for CO2 EOR projects.2 Also, there is
increasing interest in minimizing CO2 (greenhouse gas)
emissions by sequestrating CO2 in waterflooded oil reservoirs
or in saline aquifers. Thus, this interest in increasing CO2
injection into geological formations requires a better
understanding of mechanisms and extent of CO2 plume
development and subsequent dissolution into formation brine.
This paper describes laboratory tests on Frio sandstone and
Indiana limestone. Several types of displacement tests were
performed; gas injection to a pseudo-residual brine saturation
with respect to gas injection, followed by brine injection to a
pseudo-residual gas saturation with respect to brine injection.
The level of CO2 saturation in the injected brine at reservoir
pressure and temperature was varied from zero to over 90%.
This variation of CO2 saturation in the injected brine was to
determine the effect on the CO2 saturation or plume size in the
core.
Determination of CO2 saturation in a core was sought
after injection of CO2 into a core that was originally saturated
with brine. This was then followed by the injection of brine
into the core while determining when brine displaced free CO2
versus producing CO2 dissolved in the brine. Currently, CO2 is
being injected into reservoirs nearing their waterflood
economic limit and into aquifers, thus injecting CO2 into
geological formations containing high brine saturation. To aid
in conformance control and reduce the amount of CO2
required for injection, CO2 and water are alternately injected
into oil reservoirs. Also, it is being proposed to inject CO2 into
innumerable aquifers for carbon sequestration. Thus tests are
required for both the understanding of how brine and CO2
streams flow through porous media and how their mutual
solubilities change their saturations with time.
Experimental
Material
Frio cores used in these tests were obtained from depths of
8179, 8191 and 8195 ft in the Felix Jackson # 62 Well, located
south of the S. Liberty DOE CO2 pilot site in Chambers
County, Texas . These cores were selected because they were
consolidated sandstone (see Frio Core Parameters listed in
Table 1). The DOE carbon sequestration DOE test site south
2 SPE 103228
of Houston is at a shallower depth and the test horizon is in
poorly consolidated rock. These tests were performed in the
consolidated core to simplify the development of test
procedures. Table 2 lists the composition of the synthetic brine
used in these tests, which is intended to represent the Frio
reservoir brine. Indiana limestone is quarried near Victor,
Indiana. The parameters for this core are also listed in Table 1.
Figures 1 and 2 provide permeability and porosity distribution
from mercury injection for sandstone and limestone samples,
respectively. In both cases most of the permeability is
contributed by large pore throats (~25 nm), though for
limestone there are a significant number of smaller pore
throats. The difference in the pore geometry can be seen in
Figs. 3 and 4 for sandstone and limestone, respectively.
Pore System Characteristics
The mercury injection data show the sandstone has a measured
porosity of 29.7% and a calculated Klinkenberg permeability
of 2700 md (see Fig. 1). Most of the PV consists of
intergranular pores (19.6% by point count) that appear to be
well interconnected. Point count data indicates that secondary
intraparticle and leached grain pores make up 3.6 vol % pore
space, or an estimated 12% of the total PV (see Fig. 3).
Micropores are created by partial to complete dissolution of
susceptible detrital grains (mostly feldspars and volcanic
grains) and are associated with authigenic clay cements. The
difference between the mercury injection porosity (29.7%) and
the point count porosity (23.2%) is generally a measure of
pore space tied up as micropores (6.5%) or 22% of the total
PV. The mercury injection of the sandstones show an initial
invasion pressure of about 3.5 psi and sharp shoulder that
leads to a flat slope from 5.0 to 9.0 psi where 50% of the pores
(mostly large to medium sized intergranular pores) are filled.
The injection curve begins to ramp up pressure until about 35–
40 psi, where the mercury is estimated to invade the micropore
space. The mercury injection pore throat radius distribution
chart estimates 18% of the total porosity as micropores
(assuming a 0.5 micron radius cutoff), which is close to the
point count estimation of micropores, which is 22% of the
total pore space.
Thin section observations show the Indiana Limestone
has a trimodal pore system of larger vugs and moldic pores,
intergranular pores and micropores in partially leached ooids
and peloids (see Fig. 4). The limestone sample has a measured
porosity of 16.0% and a calculated Klinkenberg permeability
of 118 md. Most of the PV consists of well connected
intergranular pores (8.0% by point count) and secondary
moldic pores and vugs (3.6% by point count). Fracture pores
(2.0% by point count) are probably artificial and not included
in these calculations. Micropores are created by partial
dissolution of susceptible grains like ooids, algae, peloids and
intraclasts. The difference between the mercury injection
porosity (16.0%) and the intergranular and moldic point count
porosity (11.6%) is generally a measure of pore space tied up
as micropores (4.4%) or 28% of the total PV.
The mercury injection of the Indiana limestone shows an
initial invasion pressure of about 3 psi and a gentle slope from
5.0 to 30.0 psi where the pressure begins to ramp up with little
invasion (see Fig. 2). At about 110 psi the mercury begins to
invade the micropore system. Assuming a 0.5 micron radius
pore throat cutoff, the estimated micropore volume is 44% of
the total PV. The point count estimation of micropores is
lower at about 28% of the total pore space. This difference is
likely due to sample heterogeneity, with the thin section
sample having less ooids and occasionally having plucked
grains artificially counted as vugs. The sample used for the
mercury injection sample may have had more ooids and a
higher percentage of equant calcite cement.
Core Flooding Apparatus
The core flooding apparatus is located in a temperature-
controlled air bath, with a syringe pump and separator system
outside the air bath (see Fig. 5). The dead volume of this
system (non-flow path volume) and non-core volume was
minimized by reducing the number of pressure control
devices, pressure transducers and valves in the system. All the
cores were prepared by wrapping them longitudinally in a a
lead (Pb) foil to function as a diffusion barrier between the
core and the overburden sleeve. In this way the diffusion of
the CO2 from the core into the overburden fluid is minimized
and the mass balance is optimized. During the analysis care
was taken to capture all the water using an ambient condition
separator (liquid trap) to catch the brine/water and a salt
breaker (vapor trap) to capture water vapor. For a volume
check the liquid and vapor traps were weighed before and
after each test and in a couple of cases at an intermediate
point. The wet test meter was used to determine gas
production at ambient conditions. Included in the gas
calculations were corrections for gas displaced by brine/water
in the separator.
Test Procedures
All tests were performed at the reservoir conditions of about
145°F and 2200 psig, except for a couple of comparison tests
at 100°F (these will be indicated). The overburden pressure
was maintained at 4000 psi. The brine was prepared with the
composition indicated in Table 2. In some cases the brine had
CO2 dissolved in it to represent brine that had been in contact
with CO2. The brine will be indicated as dead-brine (no
dissolved CO2), 50% CO2-brine (brine saturated to about 50%
CO2), and 90% CO2-brine (brine saturated to about 90% CO2).
Brine saturated to 100% CO2 was not used to ensure no new
free CO2 occurred from CO2 evolving from the brine. Pressure
drop across the core and/or dissolved solid changes in the
brine due to dissolution of core material or water vaporizing
into the CO2 phase could perturb brine 100% saturated with
CO2 and result in small but undesirable amounts of free CO2
forming from the injected fully saturated brine.
In all but one case the coreflood was initiated 100%
saturated with dead-brine. Dehydrated CO2 was then injected
into the core until no free brine was produced for several PVs.
The CO2 was stored outside the air bath at ambient
temperature and injected at rates from 10 to 200 cc/hr (20
cc/hr was used unless otherwise indicated) at ambient
temperature and about 2200 psig. The CO2 injection volume at
145°F was about 65% higher than at ambient temperature;
both at 2200 psig. The temperature of the air bath, core, and
injection pump were recorded. The head plus end volume of
the core system is 4.3 cc; thus in Figs. 6, 7 and 12–17 the
volumes are shown starting at -4.3 cc.
SPE 103228
Discussion of Results
Sandstone
Figure 6 compares two tests of CO2 displacing brine in Frio
Core A. In both about 7 cc of brine was produced before CO2
breakthrough. After CO2 breakthrough there was a small
quantity of brine produced and then brine production stopped
except for water dissolved in the CO2. Usually over 95% of
the brine production occurred before 1 PV of CO2 had been
injected. Any continued production after about 1 PV of CO2
had been injected was from vaporized water. The salt vapor
trap was weighed only at the end of the test and this value was
added evenly over the duration of the test scaled by injection
rate to the brine/water production plots. The time when the
vapor was actually produced is not known. In Fig. 6 the first
system had an injection rate increase from 20 to 100 cc/hr
after 200 cc of CO2 had been injected. Note the increase in the
rate of brine production for an interval following the injection
rate increase. During the second test injection was continued
overnight at a reduced injection rate of 10 cc/hr and then
increased to 100 cc/hr for a short time at the end of the test.
In each case the saturations reached what might be
considered a stable pseudo-end point. However this stable
saturation could be changed by increasing the flow rate,
decreasing pressure, or by evaporating brine saturation; thus a
true end point was not reached. For this paper it will be
referred to as only a pseudo-end point. After completion of
CO2 injection, brine was injected into the core to displace the
CO2. Figure 7 compares the first two brine injection tests
which used dead-brine in Frio Core A after the CO2 injection
in each test (see Fig. 6). In each about 4.3 cc of reservoir
condition CO2 was produced (the same as the end plate dead
volume), then the CO2 production rate decreased significantly.
After this break, approximately an additional 5 cc of CO2 at
reservoir condition were produced at a fairly constant rate.
Using values from Wiebe and Gaddy3 adjusted for dissolved
solids,4 these rates are what would be expected from CO2
dissolved in Frio brine fully saturated with CO2 at the test
conditions. The CO2 produced after the dead volume was
produced was not a free phase. The final value of produced
CO2 from the system, including blowdown to ambient
pressure, was equal to the brine produced during CO2
injection; thus a good material balance was obtained
throughout the experiments.
The second set of experiments was done in Frio Core B.
In this set the tests were performed using the same procedure
used for Frio Core A, except that the CO2 concentration in the
injected brine was varied from 0 to 90% of saturation. Figure
8 shows the production rate of CO2 in cc/min at ambient
conditions for three brines tested at a flow rate of 20 cc/hr.
These values multiplied by three would be the dissolved CO2
in a cc of brine at test conditions minus CO2 dissolved in brine
at ambient conditions. For the purpose of this study we are
considering CO2 dissolved in brine at ambient conditions to be
negligible. Figure 9 compares the production rate of CO2
during the injection of brine into Frio Core B during three
different tests. Excluded in Figures 9–11 was the first PV of
brine injection where the production of free CO2 was
occurring. See the example in Fig. 13 where early production
is plotted and is in the 100 to 200 cc/min range during the
3
brine injection rate of 20 cc/hr. Of the three tests shown in Fig.
9, each one followed the injection of CO2 into the core
saturated with dead brine. The three tests differ in the
concentration of CO2 in the injected brine. During the early
time period (see Figure 10) the production rates are essentially
equal for all three scenarios. The brine produced from this 2.4
in. core was saturated with CO2 and did not depend on the
CO2 concentration of the injected brine. Thus the brine was
saturated with CO2 over a relatively short flow path. Also
shown in Fig. 10 for comparison is the production rate of 50%
CO2-brine and 90% CO2-brine.
The injection test using 90% CO2-brine was not continued
until free CO2 was depleted in the core as in the other two
cases. Injection and production continued long enough to
verify the production rate of CO2 during the first part of the
injection. From Fig. 5, CO2 depletion in the core during the
dead-brine injection shows a rapid decline in the CO2
production rate after most of the CO2 had been produced. In
the 50% CO2-brine the drop is slower and as might be
expected the system stabilizes at a rate of about 3 cc/min,
which is the same as the content of the brine being injected
(see Fig. 8). When the pressure was released on the 50% CO2-
brine system the produced CO2 was equal to about that which
would be evolved from 1 PV of brine saturated to 50% CO2,
indicating that all the free CO2 had been removed.
Figure 11 compares the production of CO2 during the
injection of 50% CO2-brine at two temperatures, 100° and
145° F. The production rates are similar but it does appear that
the CO2 production rate is a little higher at 100°F. From the
literature the solubility of CO2 in brine is around 10–15%
higher at the lower temperature. Even though the data has
scatter similar or greater than the difference expected, the
average for the first 200 minutes in Fig. 11 is 7.3 cc/min at
145°F and 8.3 cc/min at 100°F, which is about a 14% increase
in production at the lower temperature.
The final set of tests reported here was in Frio Core C
with results shown in Figs. 12–14. Figure 13 has an expanded
production rate scale to demonstrate the rate comparison
during free CO2 production and production evolving from CO2
dissolved in brine at reservoir conditions. In these tests the
first step was started with a dry core. This was then saturated
with 100% dehydrated CO2; then dead-brine was injected into
the core. In this test about 9 cc of CO2 at reservoir conditions
were produced before production stabilized. This rate was
equal to that of CO2 evolved from brine saturated with CO2 at
100°F and 2200 psig. Then an additional 11 cc (reservoir
conditions) of CO2 were produced at a rate of about 8 cc/min
at ambient conditions. This totals 20 cc of produced CO2.
Subtracting the 4.3 cc dead volume yields 16 cc or almost
90% of the 18.1 cc core PV. Another 2 cc were produced
during the remaining injection period and subsequent
blowdown. In this test it required about 4 cc of brine or just
over 0.2 PV to establish a brine flow path. At this point free
CO2 production essentially stopped at brine breakthrough.
Shortly after brine breakthrough it appears that only CO2
dissolved in the brine was produced.
After the pressure of Frio Core C was reduced to ambient
pressure, dead-brine was injected to remove any remaining
CO2. To compare Frio Core C to Frio Cores A and B, the core
was returned to reservoir pressure, but left at 100°F. Again the
4 SPE 103228
dead volume and a few cc of addition of brine were produced.
When the production of free brine stopped then 50% CO2-
brine was injected. Figure 14 compares the production rate
and total production of CO2 during the injection of the 50%
CO2-brine with the injection of dead-brine into the dry CO2
core. The results are little different from Frio Cores A and B,
in that free CO2 production terminated before the dead volume
was produced. This had occurred once before and in both
cases it appeared that some brine had remained in the end
plates of the core and/or in the lines. In one case, after CO2
injection about 1 cc of brine was blown out of the lines after
they were isolated from the core. In the other tests the system
lines had been brine/water free. The slope of the cumulative
production for the second tests is less than that of the first test
because the CO2 dissolved in the 50% CO2-brine subtracted
from the produced CO2 when calculating the produced CO2.
Limestone
Several tests on Indiana limestone were conducted using the
same procedure used for the Frio sandstone. All tests were
performed at about 100°F and pore pressure of about 2200
psig. In the first four tests the core was initially saturated with
dead-brine and displaced with CO2 until no free water was
being produced, usually requiring 2–3 PV after the last
production of free water was detected. In all but the third tests,
the CO2 injection rate was 20 cc/hr at room temperature or
about 21.7 cc/hr at core conditions. In series three the injection
rate was initially 21.7 cc/hr and then increased incrementally
to 43.4, 86.8, and 130.2 cc/hr while the incremental produced
water was 7.8, 0.3, 0.1, and 0.1 cc, respectively. Again, the
first 4.3 cc produced was from the line volume resulting in
brine production from the core of about 4.0 cc and vapor
production caught in the salt trap of 1.4 for a total of about 5.4
cc of brine from the core. The brine/water production in the
third test is compared to the other three tests in Fig. 15. There
is a contrast of the production at similar injection rates
between the first two and last two tests of almost 1 cc (see Fig.
15).
Figure 16 compares flow tests at three different times
during these tests. As the number of tests and PV of fluid
injected into the core increased, the pressure drop versus flow
rate decreased. This is an indication that the core permeability
was increasing with time or PVs of fluid injected. Tests
performed after the last test indicated almost no pressure drop
at all tested flow rates (20–200 cc/hr), indicating a very high
permeability. In earlier tests with limestone, total core
permeability increased over time until a solution channel
through the core had been formed, and then the permeability
drastically increased.5,6 In each case there had been plugging
or deposits advancing ahead of the solution channel.
Figure 17 compares the reservoir volumes of CO2
produced for each test during the injection of brine. For Tests
1 and 2 dead-brine was injected and for both there was a good
material balance. For Tests 3 and 4, 50% CO2-brine was
injected. Again in both cases there was a good material
balance. Shown in Fig. 18 is the comparison of CO2
production rate during brine injection. As shown in Fig. 17,
rapid production occured until the line volume of 4.3 cc was
produced and thereafter production was at or below the rate
from saturated brine.
In Tests 1 and 2 the core had very little CO2 remaining at
blow down. It is of interest to compare Tests 3 and 4 in Fig.
18. Though the two had similar CO2 saturations the production
dropped much more quickly to the baseline for the 50% CO2-
brine. At blowdown both had CO2 remaining, but Test 4 had
almost 3 cc compared to about 1 cc remaining in Test 3. It is
believed that the difference is due to the formation of the
solution channel, where most of the flow bypassed the bulk of
the core.
In Test 5 the core was dried and saturated with CO2.
Dead-brine was then injected into the core. About 7 cc
reservoir volume of CO2 was produced before production was
from CO2 dissolved in the produced brine. Figure 19 compares
the CO2 production volume of Test 5 with Test 1. Again the
first 4.3 cc was from the line, leaving less than 3 cc of brine
injected into the core before a flow channel was established.
This low saturation was due to the solution channel that had
been established. CO2 production continued for a time but the
produced brine was undersaturated in CO2. Figure 20
compares the production rate in Test 5 with Test 1. In fact,
CO2 production dropped to less than 1 cc/min at ambient
conditions with only a fraction of the total having been
produced, less than 5 cc at core conditions or less than one
third the total core volume of 16 cc. In Test 1 where there had
only been a few cc of CO2 saturation, brine saturated with CO2
was produced much longer. It was expected that there was at
least another 10 cc of CO2 remaining in the core. At that rate
of production it would take at least another 5000 minutes to
produce the remaining CO2. At blow down another 8 cc of
core condition CO2 was produced. Comparison of Test 5
production with the expected volume of CO2, assuming the
core was saturated with CO2 did not yield a good material
balance. The total CO2 produced was 12.5 cc compared to the
previously determined 16.3 cc core total PV. It is expected
that due to the solution channel brine had not been entirely
removed from the core and thus some of the low permeability
zones of the core had remaining brine saturation. The bimodal
pore throat size range determined in the Mercury tests had
been artificially increased to include at least one channel of
relatively large macroscopic pore throats.
Summary and Conclusions
These were relatively short cores (2.25 to 3.13 in.), about 1.5
in.in diameter and therefore care must be taken when
extrapolating results to reservoir scale. There are a number of
interesting results:
1. In the range of 0.2 to 0.3 PV fraction of free CO2
saturation was required to establish a CO2 flow path, after
which there was little brine production except through
evaporation, which is a slow process. The CO2 saturation
can be increased by increasing flow rate, reducing
pressure, and water evaporation.
2. At the end of CO2 injection there is a relatively low CO2
saturation and high brine saturation in the core, thus no
reduction in CO2 saturation was required to return to brine
flow. In the case of water-wet systems this and the
previous conclusion are not a surprise.
3. Brine is equilibrated with CO2 in a short time frame over
a relatively short distance.
SPE 103228 5

4. In the systems tested, once CO2 injection stopped, the Table 1. Core Parameters
CO2 phase did not travel but was transported only to the
extent that unsaturated brine was available to dissolve Frio Sandstone Indiana
CO2. A B C Limestone
5. The injection of brine into a 100% CO2 phase required 0.2 Depth [ft] 8179 8191 8195 Quarried
to 0.3 PV fraction saturation to establish a brine flow
Diam [in] 1.47 1.44 1.47 1.51
path.
Length [in] 2.393 2.4 2.25 3.13
6. In this case the sandstone and limestone initially
performed similarly. This was changed when through Bulk vol [cc] 66.55 64.05 62.58 91.85
dissolution of the rock matrix a solution channel was Pore vol [cc] 18.51 18.01 18.29 16.30
formed in the limestone creating a dominant flow path Por [%] 27.8 28.1 29.2 17.7
that significantly altered the flow behavior of the core.

Acknowledgements Table 2. Synthetic Brine Composition (Frio Receipt)

Appreciation is expressed to the Department of Energy and the
Component mg/L Species mg/L
State of New Mexico for funding and other support. We also
NaCl 82,753 Cl 57,460
thank Shinichi Sakurai an Associate at the Bureau of CaCl2 8,584 Na 32,603
Economic Geology, John A. and Katherine G. Jackson School MgCl2 2,152 Ca 3,100
of Geosciences at the University of Texas at Austin for the KCl 362 Mg 549
core samples. We thank Core Laboratories Advanced NaHCO3 186 K 190
Technology Center in Houston for the Mercury injection and HCO3 135
Total Dissolved Solids 94,037 94,037
thin section work. We appreciate the aid of Liz Bustamante for
her assistance in editing this manuscript.

References.
1. Rogers, J.D. and Reid B. Grigg, R.B.: “A Literature Analysis of
the WAG Injectivity Abnormalities in the CO2 Process,” SPE
Reservoir Eval. & Eng., Vol. 5, No. 5, October 2001
2. Benson, S.M.: “Overview of Geologic Storage of CO2,” Carbon
Dioxide Capture for Storage in Deep Geologic Formations,
Volume 2, Introduction, C.C. Thomas and S.M. Benson (Eds.),
2005 Elsevier Ltd.
3. Wiebe R. and Gaddy, V.L.: “The Solubility for Carbon Dioxide
in Water at Various Temperatures from 12 to 40°C and at
Pressure to 500 Atmospheres. Critical Phenomena,” J Amer.
Chem. Soc., Vol. 62, 815-817, (April 1940).
4. Enick, R.M.: “CO2 Solubility in Water and Brine under
Reservoir Conditions,” Chem. Eng. Comm. (1990), Vol. 90, pp
23-33.
5. Grigg, R.B. and Svec, R.K.: “Co-Injected CO2-Brine
Interactions with Indiana Limestone,” SCA2003-19, Society of
Core Analyst prepared for the SCA 2003 Pau, France, 21-24
September 2003.
6. Wellman, T.P, Grigg, R.B., McPherson, B.J., Svec, R.K., and
Lichtner, P.C.: “Evaluation of CO2-Brine-Reservoir Rock
Interaction with Laboratory Flow Tests and Reactive Transport
Modeling,” paper 80228 prepared for presentation at the 2003
SPE International Symposium on Oilfield Chemistry, Houston,
Feb. 5–8.
6 SPE 103228

Company: New Mexico Tech PPRC Sample: FSSm Host Plug Company: New Mexico Tech PPRC Sample: LSm Host Plug
Well: n/a Depth, feet: n/a ambient na na Well: n/a Depth, feet: n/a ambient na na
Field: n/a Klinkenberg Permeability, md: 2700. - - Field: n/a Klinkenberg Permeability, md: 118. - -
Formation: n/a Permeability to Air, md: 2770. - - Formation: n/a Permeability to Air, md: 127. - -
Location: n/a Swanson Permeability, md: 907. - - Location: n/a Swanson Permeability, md: 64.7 - -
File: HOU-060360 Porosity, fraction: 0.297 - - File: HOU-060360 Porosity, fraction: 0.160 - -
maximum Sb/Pc, fraction: 1.63 maximum Sb/Pc, fraction: 0.341
R35, microns: 14.1 R35, microns: 5.19
R50 (median pore throat radius): 11.8 R50 (median pore throat radius): 0.639

PORE THROAT RADIUS DISTRIBUTION PORE THROAT RADIUS DISTRIBUTION

Micro Meso Macro Micro Meso Macro

Incremental Intruded Pore Space, fraction

Incremental Intruded Pore Space, fraction

Cumulative Intruded Pore Space, fraction

Cumulative Intruded Pore Space, fraction

1.0 0.20 1.0 0.20
0.9 0.18 0.9 0.18
Cumulative
0.8 0.16 0.8 Incremental 0.16
0.7 0.14 0.7 0.14
0.6 0.12 0.6 0.12
0.5 R35 parameter 0.10 0.5 R35 parameter 0.10
0.4 Radius, microns at SHg = 0.35 0.08 Radius, microns at SHg = 0.35
0.4 0.08
0.3 Cumulative 0.06 0.3 0.06
0.2 Incremental 0.04 0.2 0.04
0.1 0.02 0.1 0.02
0.0 0.00 0.0 0.00
0.001 0.01 0.1 1 10 100 0.001 0.01 0.1 1 10 100
Pore Throat Radius, microns Pore Throat Radius, microns

1.0
PERMEABILITY DISTRIBUTION PERMEABILITY DISTRIBUTION
1.0
0.9 0.9
Permeability (fraction of total)

Permeability (fraction of total)

0.8 0.8
0.7 0.7
0.6 0.6
0.5 0.5
0.4 0.4
0.3 0.3
0.2 As outlined by Purcell (1949) and As outlined by Purcell (1949) and
Burdine, Gournay, & Reichertz (1950)
0.2
Burdine, Gournay, & Reichertz (1950)
0.1 0.1
0.0 0.0
0.001 0.01 0.1 1 10 100 0.001 0.01 0.1 1 10 100
Pore Throat Radius, microns Pore Throat Radius, microns

Fig. 1. Mercury injection pore throat and permeability Fig. 2. Mercury injection pore throat and permeability
distribution plots for Frio sandstone. distribution plots for Indiana limestone.
SPE 103228 7

Cr
ip
M

Vol

ig
SS Vol For
Cr V

Frac
LGr
Bry

Ooid Cr
1 mm

Fring Cr
Vol
ip mp

Plag Equant
Ooid ip
Qtz mp

Kaol mp

Ooid
Vol
ss

Cr
ip ign 0.2 Intraclast

Fig. 3. Frio sandstone thin sections. Key: intergranular pores (ip),
leached grain pores (LGr), micropores (mp), volcanic (Vol).
kaolinite (Kaol), quart (Qtz), plagioclase (Plag), igneous (Ign), and
sandstone (SS).
Fig. 4. Indiana limestone thin sections. Key: intergranular pores
(ig) pores, micropores (mp)I, ooids (Ooid), secondary moldic
pores (M), vugs (V), crinoids (Cr), intraclasts (Intraclast),
bryozoans (Bry), forams (For), fractures (Frac), and fringing
(Fring) and equant (Equant) calcite cement..
8 SPE 103228

BRINE / CO2 Air Bath

MIXER

P P

CORE BPR

CO2 EQUIL Overburden

COIL

VAPOR TRAP GAS

PUMP METER

LIQUID TRAP
Fig. 5. Schematic of the coreflood apparatus.

6 8
100 cc/hr
20 cc/hr 7
4
CO2 production [cc/min]

6
Brine produced [cc]

100 cc/hr
2 10 cc/hr 5
20 cc/hr
4
0

3
-2
2
CO2 inj. #1 CO2 inj. # 2 50% CO2 calibration [100 F]
-4 50% CO2 calibration [145 F]
1
90% CO2 calibration [145 F]
50% CO2 2nd calibration [145 F]
0
-6
0 10 20 30 40 50
0 100 200 300 400 500
CO2 [cc] Time [min]

Fig. 6. Comparison of brine production during CO2 injection of Fig. 8. CO2 production from brines saturated to ~ 50% CO2 or ~
two tests in Frio Core A. 90% CO2; each at a flow rate of 20 cc/hr .
10
90% CO2 sat. inj. brine
8 9 50% CO2 sat. inj. brine
blow down 8 0% CO2 sat. inj brine
6
CO2 production [cc/min]

7
CO2 prod. res.con. [cc]

4
6
change to 100cc/hr from 20cc/hr
2 5

4
0
3

-2 2
Brine inj. #2 Brine inj. #1 1
-4
0
-6 0 100 200 300 400 500 600 700 800
0 100 200 300 400 500 Time [min]
Brine Injeciton [cc] . Fig. 9. Production rate of CO2 during the injection of brine into
Fig. 7. Comparison of CO2 production during brine injection for Frio Core B for three different tests, each at different
two tests in Frio Core A. concentrations of CO2 in the injected brine.
SPE 103228 9

10 20 200

9 180

Produced CO2 res. cond. [cc]

Producted CO2 STD [cc/min]

8 15 160
CO2 production [cc/min]

7 140

6 10 120

5 100

4 5 80
50% sat. brine calibration
3 90% sat. brine calibration CO2 produced Production rate 60
90% CO2 sat. brine inj
2 50% CO2 sat. brine inj. 0 40
0% CO2 sta. brine inj.
1 20
0 -5 0
0 50 100 150 200 250 0 200 400 600 800 1000 1200 1400
Time [min] Time [min]
Fig. 10. Comparison of CO2 production rates during brine Fig. 13. Comparison similar to Fig. 12 for Frio Core C except the
injection into Frio Core B. scale for CO2 production rate is expanded to include the
maximum production rate during production of the CO2 phase.
12
50 % sat. brine inj @ 145 F
50% sat. brine inj. @ 100 F
20 10
10
9
CO2 production [cc/min]

Produced CO2 res. cond. [cc]

Producted CO2 STD [cc/min]

8 15 8
CO2 prod., CO2 initial
sat. 7
CO2 prod., brine initial
6 sat.
`
10 6
CO2 prod. rate, CO2
initial sat
5
4 CO2 prod. rate, brine
initial sat.
5 4

2 3

0 2
0
0 100 200 300 400 500 600 700 800 900 1
Time [min]
-5 0
Fig. 11. Comparison of the CO2 production rate during the 0 200 400 600 800 1000 1200 1400
injection of 50% CO2-brine at two different temperatures, 100˚ and Time [min]
145°F for Frio Core B. Fig. 14. Comparison of CO2 production during dead-brine
injection into the Frio Core C 100% saturated with CO2 and
20 10 injection of 50% CO2-brine into the core after being saturated with
dead-brine and then flooded with CO2.
18 9
6
Produced CO2 res. cond. [cc] .

Producted CO2 STD [cc/min] .

16 8
5
14 7 4
CO2 production
12 6 3
CO2 prod. rate
Vol brine [cc]

10 5 2

1
8 4
0
6 3
-1
4 Emptied Separator 2
-2
2 1
-3
0 0 Produced brine #1 Produced brine #2
-4
0 200 400 600 800 1000 1200 1400 Production brine #3 multi rate Production brine #4
Time [min] -5
0 50 100 150 200 250
Fig. 12. Comparison of total CO2 production at reservoir
conditions and production rate from the separator at ambient Time [min]
conditions for CO2 during the test at 100°F for Frio Core C. Fig. 15. Comparison of brine production during CO2 injection of
four tests in Indiana limestone.
10 SPE 103228

70 14
13
12
60
11
10
50 9

Res. vol. CO2 [cc]

8
7
Delta P [psi]

40 6
5
4
30
3
2
20 1
19-Feb-06 0
22-Feb-06 -1 CO2 pro. 5 100% CO2 CO2 production 1
10 2-Mar-06 -2
-3
-4
0 -5
0 100 200 300 400 500 600 700 0 50 100 150 200 250 300 350 400 450 500
Flow Rate [cc/hr] Time [min]
Fig. 16. Pressure drop versus flow rate for Indiana limestone on Fig. 19. Cumulative gas production in reservoir volume for core
three different days. The indication was a decrease in starting with a saturation of 100% CO2 after formation of the
permeability with time. solution channel compared to one partially saturated with CO2
before the solution channel formed in the limestone.

6 1000.0
5 CO2 production 1
CO2 production 5
Produced CO2 res. cond. [cc]

CO2 production [cc/min]

3 100.0

0 10.0

-1

-2
1.0
-3
CO2 production 3 CO2 production 4
-4
CO2 production 1 CO2 production 2
-5
0.1
0 100 200 300 400 500 600 700 800 900
0 100 200 300 400 500
Time [min]
Time [min]
Fig. 17. Comparison of CO2 production during brine injection for Fig. 20. Gas production rates at ambient conditions for limestone
four tests in Indiana limestone. core saturated starting with a saturation of 100% CO2 with a
solution channel versus one partially saturated with CO2 before
20
Gas production #1 Gas production #2 the solution channel formed.
Gas production #3 Gas production #4
18

16
CO2 production [cc/min]

14

12

10

0
0 100 200 300 400 500 600 700 800 900
Time [min]

Fig. 18. Production rates for CO2 at ambient conditions at a brine

injection rate of 20 cc/hr.