1

SPE 102138

Dependence on Temperature, Pressure, and Salinity of the IFT and Relative

Permeability Displacement Characteristics of CO2 Injected in Deep Saline Aquifers
D.B. Bennion, Hycal Energy Research Laboratories Ltd., and S. Bachu, Alberta Energy and Utilities Board

Copyright 2006, Society of Petroleum Engineers

the beginning of the industrial revolution in the mid 19th
This paper was prepared for presentation at the 2006 SPE Annual Technical Conference and century, and evidence supports the supposition that these
Exhibition held in San Antonio, Texas, U.S.A., 24–27 September 2006.
increasing concentrations are affecting the Earth’s climate1.
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
This increase is due to deforestation, agricultural practices and
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to mainly to the use of fossil fuels for power generation and
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 transportation. The high use of fossil fuels will continue this
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
century2, 3, thus, a major challenge in mitigating climate
for commercial purposes without the written consent of the Society of Petroleum Engineers is change effects is the reduction of CO2 emissions to the
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 atmosphere. Among various mitigation measures, CO2
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.
capture and geological storage (CCGS) will play an important
role in reducing CO2 atmospheric emissions3-5. The
Abstract technology for the injection of CO2 into deep underground
formations is well developed and currently practiced mainly
Sequestration of CO2 in deep saline aquifers is a means with by the energy and petrochemical industries for enhanced oil
great potential for reducing emissions of this greenhouse gas recovery (EOR)6 and acid gas (CO2 and H2S) disposal7.
produced from a wide range of industrial operations. The Carbon dioxide can also be sequestered in various geological
displacement characteristics of CO2 injected into deep saline media, of which deep saline aquifers have the largest storage
aquifers are essential in that they control both the migration of capacity and widest distribution, but oil reservoirs are the most
CO2 and the available pore space in the reservoir at irreducible likely to be used preferentially initially because of the positive
saturation conditions, yet very little information exists about economic potential for increased oil production through CO2
these characteristics. This paper presents continuing results of or acid gas based EOR4,6schemes. In both cases, the basic
a research study conducted by the authors to investigate the phenomenon is the displacement of a fluid (water and/or oil)
displacement characteristics of CO2-brine systems at reservoir by CO2 during injection and the displacement of CO2 by either
conditions for cases specific to the Alberta basin and water or oil after cessation of injection by influx of fluids from
representative for intracratonic and foreland basins in North adjacent aquifers or hydrocarbon sources. In the case of CO2
America. More specifically, the work reported here focuses being displaced, a significant amount of CO2 may be stored in
on a controlled series of tests in which the effects of the pore space8once the mobile oil/water phase has been
temperature, pressure and brine salinity were varied in a displaced to its irreducible saturation. Even in oil and gas
systematic manner to identify and quantify their effect on the producing regions, such as Alberta in Canada and Texas in the
interfacial tension between CO2 and equilibrium brine. The U.S.A., it may be more economic in some cases to inject CO2
results indicate that the CO2-brine IFT decreases with into a deep saline aquifer that underlies a major CO2 source
increasing pressure, while increasing temperature and brine than to transport it to oil and/or gas reservoirs some distance
salinity have an opposite effect. These results were then away, particularly if these are still producing and not available
extended under similar controlled conditions to consolidated or suited for CO2 storage or injection.
porous media to investigate the effect of interfacial tension, as
impacted by specific reservoir conditions of temperature, The processes of CO2 injection and migration within the
pressure and salinity, on the relative permeability character of storage unit, and the sequestration of CO2 in the pore space all
CO2 displacing brine for a number of different interfacial depend on the relative permeability of CO2 and formation
tension conditions. The results are presented, and the water systems and on the CO2-brine capillary pressure
significance with respect to the feasibility and optimization of character. Furthermore, numerical models for predicting the
CO2 sequestration in deep saline aquifers is discussed. fate of the injected CO2 also need information about these two
important parameters. However, because CO2 storage in
Introduction geological media, particularly in deep saline aquifers, is a
recently emerging field, up to 2005 no relevant data have been
Atmospheric concentrations of greenhouse gases such as published regarding the relative permeability and capillary
carbon dioxide (CO2) and methane (CH4) have increased since pressure of CO2-brine systems at in-situ conditions
2 SPE 102138

characteristic of deep saline aquifers. Seeking to cover this marine shales deposited in the Cretaceous during a lull in
gap in knowledge, the authors commenced in 2004 a series of tectonic activity between the Columbian and Laramide
tests on core plugs taken from several sandstone, carbonate orogenies in what was then the Western Interior Seaway that
and shale formations in the Wabamun Lake area southwest of extended all the way from the Arctic to the Gulf of Mexico14.
Edmonton in Alberta in western Canada. The results have Carbon dioxide from refineries and petro-chemical industries
been presented in a series of SPE papers9-11. The interfacial in Fort Saskatchewan area northeast of Edmonton, and/or from
tension (IFT) and relative permeability measurements were power plants in the Wabamun Lake area southwest of
performed at the specific in-situ conditions of pressure, Edmonton will likely be used for large-scale implementation
temperature and water salinity characteristics of each rock of CO2-EOR in the Pembina oil field and possibly for storage
sample. The pressure, temperature and water salinity vary outside the field itself in the water-saturated portions of this
between the various tests in an unstructured way, ranging from sandstone formation.
8.6 MPa to 27 MPa for pressure, 35°C to 75°C for
temperature, and ~28,000 ppm to 248,000 ppm for salinity, Results and Discussion
according to the in-situ conditions of the strata where the core
samples which were tested originated from. The measured The specific details of the equipment and procedures used to
IFT varied between 27 mN/m and 49 mN/m for these conduct the relative permeability and IFT measurements have
conditions. The results indicate a dependency of capillary been presented by the authors in previous papers9,10 and will
pressure on pore size distribution and of IFT on pressure, not be repeated here. Core samples from the Cardium
which was expected, but also on temperature and salinity10. sandstone were taken from Pembina area well 02/7-11-48-
Also, a weak dependency of endpoint relative permeability 9W5M. The relative permeability measurements for various
and of maximum CO2 saturation on IFT was also observed. IFT levels were conducted on the same core plug to eliminate
Because of the unstructured variability in pressure, possible effects of various pore distributions and rock
temperature and salinity conditions, it was difficult to identify characteristics on relative permeability and irreducible
and quantify the IFT dependency on these parameters, and of saturations. A number of different core samples with various
displacement characteristics on IFT. To verify the emerging permeability and porosity distributions were subjected to
pattern, a series of controlled tests was performed in which the mercury injection capillary pressure testing to select a sample
effects of temperature, pressure and brine salinity were varied that had a fairly wide distribution of pore throat sizes, as
in a systematic manner to identify and quantify their effect on opposed to “spike” type, narrow distribution. It was thought
the interfacial tension between CO2 and equilibrium brine, that media with this type of pore distribution character would
with the results being reported elsewhere12. These results, be a better candidate for illustrating the effect of interfacial
which confirmed the IFT dependency on pressure, temperature tension on pore system accessibility and, hence, changes in
and salinity, were then extended under similar controlled residual saturations and relative permeability with variations
conditions to consolidated porous media to investigate the in IFT. The characteristics and in-situ conditions of the core
effect of interfacial tension, as impacted by specific reservoir plug used in this work are presented in Table 1. Results
conditions, on the relative permeability character of CO2 specific to the work presented here include pore size
displacing brine, and the results are presented in this paper. distribution, capillary pressure and relative permeability
measurements for three different IFT conditions. Given the
Core plugs from the sandstone Cardium Formation in the IFT dependence on pressure, temperature and salinity12, the
Pembina area were used in the analysis of the effect of IFT on variation in IFT was obtained by maintaining temperature and
relative permeability. This oil field is located in west-central salinity constant at in-situ conditions typical of the Pembina
Alberta, southwest of Edmonton and of Wabamun Lake, the Cardium reservoir (43°C and 27,096 mg/l, respectively) and
area from which the rock samples used in the previous work varying only the pressure to obtain a range of water-CO2
originate (Figure 1). This formation was selected for several interfacial tensions. This allowed maintaining the majority of
reasons. First, it augments the data base of rock properties, the parameters used in the test, such as temperature, flow rate
IFT and relative permeability measurements generated and fluid composition, as constants between the various
through this research program, all the other previous analyses experiments. However, the viscosity ratio varied between the
and tests being performed on core samples from underlying tests due to the fact that increases in pressure increase the
formations. Second, the Cardium Formation is water, oil and viscosity of CO2, with little effect on the viscosity of the brine.
gas saturated, depending on location13, and it contains the This was unavoidable, as no other method of varying the IFT
Pembina oil field which is the largest oil field in Canada. was practical which would not have created similar changes in
Currently water injection is used in various parts of this giant the viscosity of either the brine or the CO2 phase (e.g., altering
oil field, but now there is growing interest in the oil and gas temperature or water salinity).
industry to use CO2 from industrial sources for CO2-EOR
tertiary recovery, and a pilot operation is currently being The relative permeability measurements were performed at
conducted in the area using CO2 brought in from a local gas operating pressures of 1,378 (200 psig) kPa, 6,890 kPa (1,000
plant. psig) and 20,000 kPa (2900 psig) respectively. The
corresponding IFT values for the Cardium brine-CO2 system
The Cardium Formation is one of several thin, isolated, for a complete range of pressures that span the operating
interbedded progradational packages of sandstones and conditions used in this study are contained in Table 2. The IFT
conglomerates of various extent embedded in a thick wedge of values for the relevant pressure points are given in Table 1.
SPE 102138 3

Figure 2 shows a plot of the pore size distribution of the have similar dependencies on pressure, temperature and water
Cardium Formation core sample used in the test program. The salinity, so, regardless of which one of these three primary
sample exhibits a relatively wide, unimodal type size parameters is used, the results will indicate the same trends
distribution as desired. Figure 3 provides the CO2-brine with no possibility of discerning which is the dominant
capillary pressure curves for the test sample, corrected to variable on displacement characteristics, viscosity contrast or
reservoir conditions, for the various IFT that correspond to the IFT. It is likely that both play a role, and, of course, in a
three pressure levels chosen in the study. As expected the reservoir situation it is impossible to decouple the natural
lower IFT levels have a proportional reducing effect on the variation of viscosity from changes in IFT, but is interesting to
capillary pressure values. Table 2 provides a summary of the compare the Krg values and trapped gas saturation values
endpoint permeability, relative permeability and saturation generated as a function of secondary imbibition versus those
data for the three different IFT level tests. Table 3 provides the obtained in primary drainage.
primary drainage and secondary imbibition relative
permeability data for the three IFT levels and these data have In the primary drainage process we see very consistent
been plotted in Figure 4 for the 55.4, 33.5 and 19.8 mN/m IFT reductions in irreducible water saturation and increases in Krg
tests that correspond to increasing pressures of 1,378 kPa, as the IFT is reduced. This process has a very unfavorable
6,890 kPa and 20,000 kPa at in-situ conditions of temperature mobility ratio, with the low viscosity CO2 displacing the much
and water salinity. Figure 5 shows the comparative plots of the more viscous brine. This being the case, at relatively constant
primary drainage and imbibition relative permeability curves displacement pressure levels as used in this work, one would
as a function of IFT for the three displacements. expect that, if viscous effects were dominating, relatively little
reduction in irreducible water saturation would be present
Examination of the data, particularly in Figure 5, indicates the (which is not the case). If we then examine the secondary
following trends: imbibition process, which has a very favorable viscosity ratio
in contrast to the primary drainage process, we see relatively
1. It can be clearly observed that the endpoint relative little change in trapped gas saturation and endpoint gas
permeability to CO2 increases as IFT decreases, in this permeability at lower IFT levels. Since this displacement has
case as a result of pressure increase. a vastly superior viscosity ratio to the primary drainage tests,
2. Similarly, the maximum endpoint CO2 saturation increases we would expect that, as the viscosity ratio continues to
as IFT decreases. improve at higher pressures, additional reductions in trapped
3. The value of the trapped gas saturation on secondary gas saturation and increases in endpoint relative permeability
imbibition decreases as IFT decreases with increasing to brine would occur. Once again, if IFT dominates, these
pressure, although there appears to be a limiting value as effects would be less pronounced, which is rather the case.
little reduction in trapped gas saturation was observed Thus, the comparison suggests that IFT likely plays a more
between the 33.5 mN/m (6,890 kPa) and 19.8 mN/m significant role in the changes in relative permeability and
(20,000 kPa) IFT levels. endpoint saturation character than the viscosity ratio, but the
4. The relative permeability curves for both CO2 and brine impact of viscosity cannot be excluded as a potentially
display classical hysteresis behavior by which, for each significant factor at this time without more controlled tests.
phase, relative permeability is higher when fluid saturation Developing such tests represents a significant challenge, since,
increases than when it decreases. However, contrary to as discussed, both IFT and viscosity depend on the same
expectations, the wetting phase (brine) has a more primary variables: pressure, temperature and water salinity, in
pronounced hysteresis effect than the non-wetting phase a similar manner, hence one of the two cannot be kept
(CO2). constant while the other varies on the basis of variations in any
5. Multiphase interference effects, as indicated by the of these primary variables.
curvature of the relative permeability curves, are reduced
(the curves in general became more linear) as IFT Conclusions
decreases.

In general the results are classic of what would be expected as The results of the study indicate that:
a function of reducing IFT in a gas-oil system. The immediate
conclusion would of course be that IFT has a strong effect on 1. As expected for constant temperature and water salinity,
relative permeability for CO2-brine systems, which we believe the brine-CO2 IFT has a strong dependency on pressure.
is the case; however, the effect of variable viscosity ratio Interfacial tension is found to be significantly reduced by
between the various tests must also be considered. It is increases in pressure.
interesting to plot both endpoint relative permeability to water 2. For the variable IFT CO2-brine drainage and imbibition
and maximum CO2 saturation on primary drainage as a relative permeability experiments conducted on a single
function of both IFT and viscosity ratio (Figure 6). reservoir rock sample with a defined pore size distribution
Examination of these data indicates virtually identical trends it was found that:
with both variables. This is because the relative decrease in
a. The endpoint relative permeability to CO2 increases
the brine-to-CO2 viscosity ratio, caused by the increase in CO2
as IFT decreases.
viscosity due to pressure increase, is of a similar magnitude to
the decrease in IFT. Furthermore, both IFT and viscosity ratio
4 SPE 102138

b. The maximum endpoint CO2 saturation increases as References

IFT decreases.
1. IPCC (Intergovernmental Panel on Climate Change),
c. The value of the trapped gas saturation on
Climate Change 2001: Synthesis Report (eds. R.T. Watson
secondary imbibition decreases as IFT decreases.
and the Core Writing Team), Cambridge University Press,
d. The relative permeability curves for both CO2 and Cambridge, U.K., and New York, NY, U.S.A., 397 p.,
brine display classical hysteresis behavior by 2001.
which, for each phase, relative permeability is
higher when fluid saturation increases than when 2. Bajura R.A., The role of carbon dioxide sequestration in
fluid saturation decreases. However, contrary to the long term energy future; Fifth International
expectations, the wetting phase (brine) has a more Greenhouse Gas Technologies Conference (GHGT-5),
pronounced hysteresis effect than the non-wetting Cairns, Australia, (eds. D.J. Williams, R.A. Durie, P.
phase (CO2). McMullan, C.A.J. Paulson, and A.Y. Smith), CSIRO
e. Multiphase interference effects, as expressed by the Publishing Collingwood, VIC, AU, 52-58, 2001.
curvature of the relative permeability curves, are
being reduced as IFT decreases (the curves in 3. IEA (International Energy Agency), Prospects for CO2
general became more linear). Capture and Storage, OECD/IEA, Paris, France, 249 p.,
3. Analysis of the data suggests that in these experiments the 2004.
variation of the IFT with pressure is roughly proportional
to the variation of the viscosity ratio of the brine/CO2 4. IPCC (Intergovernmental Panel on Climate Change), IPCC
displacement over the same pressure range. This makes it Special Report on Carbon Dioxide Capture and Storage
difficult to conclude if the observed variations in (eds. B. Metz, O. Davidson, H.C. de Coninck, M. Loos and
displacement characteristics can be solely attributed to IFT L.A. Mayer), Cambridge University Press, Cambridge,
effects or to viscosity ratio effects as well. U.K., and New York, NY, U.S.A., 442 p., 2005.
4. Analysis of the drainage and imbibition data and additional
unpublished work and experience lead the authors to 5. Socolow, R.H., Can we bury global warming? Scientific
believe that the observed variation in displacement American, v. 293, no. 1, 49-55, 2005.
characteristics of CO2-brine systems in this test program
can be more strongly attributed to changes in IFT than to 6. Moritis G., Special Report: Enhanced oil recovery-2002
variations viscosity ratio, however, additional testing worldwide EOR survey; Oil and Gas Journal, v. 100, no.
would be required in order to conclusively state this. 15, 43-47, Apr. 15, 2002.

The results suggest that reservoir conditions strongly impact 7. Bachu, S., and Haug, K., In-situ characteristics of acid-gas
the value of CO2-brine displacement characteristics in a injection operations in the Alberta basin, western Canada:
typical lower permeability porous media with a broad pore Demonstration of CO2 geological storage; Geologic
size distribution which includes significant fractions of Storage of Carbon Dioxide with Monitoring and
microporosity. The data suggest that these effects may be less Verification, v. 2 (ed. S.M. Benson), Elsevier, London,
pronounced in higher permeability porous media where a 867-876, 2005.
greater level of accessibility to larger pores could be obtained
at higher IFT conditions, but once again, additional studies 8. Kumar, A., Noh, M., Pope, G.A., Sepehrnoori, K., Bryant,
would be required in order to verify this supposition. S., and Lake, L.W., Reservoir simulation of CO2 storage in
deep saline aquifers; Paper SPE 89343, presented at the
Acknowledgements SPE/DOE Fourteenth Symposium on Improved Oil
Recovery, Tulsa, OK, USA, April 17-21, 2004.
The authors express appreciation to the Alberta Energy and
Utilities Board for permission to publish the data and to
9. Bennion, B., and Bachu, S., Relative permeability
Donna Leach and Dan Magee for assistance in the preparation
characteristics for CO2 displacing water in a variety of
and formatting of the manuscript and the figures.
potential sequestration zones in the Western Canada
Sedimentary Basin; Paper SPE 95547, 15 p., presented at
the 2005 SPE Annual Technical Conference and
Exhibition, Dallas, TX, October 9-12, 2005.67

10. Bennion, B., and Bachu, S., The impact of interfacial

tension and pore size distribution/capillary pressure
character on CO2 relative permeability at reservoir
conditions in CO2-brine systems; Paper SPE 99325, 10 p.,
presented at the SPE/DOE Fifteenth Symposium on
Improved Oil Recovery, Tulsa, OK, USA, April 22-26,
2006.
SPE 102138 5

11. Bennion, B., and Bachu, S., Supercritical CO2 and H2S- 14. Leckie, D.A., Upper Zuni sequence, Middle Cretaceous to
brine drainage and imbibition relative permeability Lower Tertiary; Western Canada Sedimentary Basin – a
relationships for intergranular sandstone and carbonate Case History (ed. B.D. Ricketts), Canadian Society of
formations; Paper SPE 99326, 13 p., presented at the SPE Petroleum Geologists, Calgary, AB, 269-284, 1989.
Europec/EAGE Annual Conference and Exhibition,
Vienna, Austria, June 12-15, 2006. 15. Janos, T, Tibor, B., and Szucs, P., Direct determination of
relative permeability from nonsteady constant pressure and
12. Bachu, S., and Bennion, B., Effects of in-situ conditions on rate displacements; Paper SPE 67318, presented at the SPE
relative permeability characteristics of CO2-brine systems; Production and Operations Symposium, Oklahoma City,
Environmental Geology, in press. OK, USA, March 24-27, 2001.

13. Michael, K., and Bachu, S., Fluids and pressure 16. Wellman, T.P., Grigg, R.B., McPherson, B.J. and Svec,
distributions in the foreland-basin succession in the west- R.K., Evaluation of CO2 – brine reservoir rock interaction
central part of the Alberta basin, Canada: Evidence for with laboratory flow tests and reactive transport modeling;
permeability barriers and hydrocarbon generation and Paper SPE 80228, presented at the International
migration; AAPG Bulletin, v. 85, no. 7, 1231-1252, 2001. Symposium on Oilfield Chemistry, Houston, TX, USA,
February 5-7, 2003.

Table 1 – Summary of test parameters.

Parameters Value
Sample Depth (m) 1626.33
Net Overburden Pressure (kPa) 11,000
In-situ Temperature (°C) 43
In-situ salinity (ppm) 27,096
Sample Length (cm) 3.73
Sample Diameter (cm) 3.77
Flow Area (cm2) 11.16
Displacement Rate (cm3/hr) 10.0
Average Porosity (%) 15.3
Pore Volume (cm3) 6.37

Table 2 – Fluid properties and displacement characteristics for the three different IFT values that correspond to
different pressures at constant temperature and water salinity (see Table 1 for the last two parameters).

Pressure IFT Viscosity (mPa·s) Krg Sgmax Krw Trapped

(psi) (kPag) mN/m Brine CO2 Gas
Saturation
200 1,378 56.2 0.648 0.014 0.2978 0.621 0.4044 0.224
1000 6,890 33.2 0.649 0.035 0.4557 0.729 0.8600 0.107
2900 20,000 19.8 0.651 0.075 0.5260 0.803 0.9053 0.102
6 SPE 102138

Table 3 – Drainage and imbibition data for CO2-brine relative permeability displacement tests performed for
variable IFT: a) 56.2 mN/m; b) 33.2 mN/m, and c) 19.8 mN/m, that correspond to various pressures and constant
temperature and water salinity. For test conditions see Tables 1 and 2.

a) IFT = 56.2 mN/m (P = 1,378 kPa)

Drainage Imbibition

b) IFT = 33.2 mN/m (P = 6,890 kPa)

Drainage Imbibition

c) IFT = 19.8 mN/m (P = 20,000 kPa)

Drainage Imbibition
SPE 102138 7

Figure 1 – Location of Pembina Cardium oil field in Alberta, Canada, and of the well sampled for testing of relative permeability and
displacement characteristics of CO2-brine systems in the Cardium Formation.

Figure 2 –Pore size distribution of the core sample from the Figure 3 –Capillary pressure curves for various IFT for the
Cardium Formation in well 02/7-11-48-09W5M. same core sample from the Cardium Formation in well 02/7-
11-48-09W5M.
8 SPE 102138

Figure 4 – Relative permeability curves (drainage and imbibition) for a core sample from the Cardium Formation in Wabamun Lake
area, Alberta, Canada, for CO2-brine systems at in-situ temperature of 43°C and water salinity of 27,096 ppm, and for various IFT: a)
56. 2 mN/m, b) 33.5 mN/m, and c) 19.8 mN/m, that corresponds to increasing pressures of 1,378 kPa, 6,890 kPa and 20,000 kPa,
respectively.
SPE 102138 9

Figure 5 – Dependence on IFT of the relative permeability of a rock sample from the sandstone Cardium Formation in Alberta,
Canada: a) drainage, and b) imbibition.

Figure 6 – Comparison of endpoint relative permeability to gas (primary drainage), maximum gas saturation, endpoint relative
permeability to water (secondary imbibition) and trapped gas saturation for: a) different IFT, and b) different viscosity ratios, both
corresponding to the same in-situ temperature and water salinity and pressures of 1,378 kPa, 6,890 kPa and 20,000 kPa, respectively.