SPE 89477
Effect of CO2 Impurities on Gas-Injection EOR Processes
P.Y. Zhang, SPE; S. Huang, SPE; S. Sayegh, SPE, Saskatchewan Research Council
X.L. Zhou, Protroleum Technologies Ltd.
Copyright 2004, Society of Petroleum Engineers Inc.
This paper was prepared for presentation at the 2004 SPE/DOE Fourteenth Symposium on
Improved Oil Recovery held in Tulsa, Oklahoma, U.S.A., 17–21 April 2004.
This paper was selected for presentation by an SPE Program Committee following review of
information contained in a proposal 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 a proposal of not more than 300
words; illustrations may not be copied. The proposal 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
Carbon dioxide flooding is a promising enhanced oil recovery
method both on technical and, if operating costs are properly
controlled, economic grounds. Injecting this greenhouse gas
also has environmental merits. Flue gas from power plants is a
ready source of CO2; however extracting CO2 for enhanced oil
recovery from such a source will increase project costs.
Furthermore, to reduce both the net CO2 utilization and the
cost of purchasing gas, it is usually necessary to recycle the
produced CO2 with as little purification as possible. Therefore,
understanding the roles of impurities in fluid phase behaviour
and miscibility characteristics is necessary for designing a
cost-effective CO2 enhanced oil recovery process.
Laboratory studies of the effect of CO2 impurities on
phase equilibrium and minimum miscibility pressure (MMP)
were conducted on two Saskatchewan light oils covering a
range of densities from 29.5°API to 38°API. The results
indicate that the MMP for these light oils could increase
unfavourably as the N2 and/or CH4 concentration increased in
the CO2 stream. The MMP changes as the type and
concentration of impurities in the injected CO2 stream change.
However, coreflood tests showed that the near-miscible CO2
displacement might employ the same mechanisms as miscible
CO2 flooding to mobilize and displace oil; thus, good oil
recovery can be achieved in the vicinity of the MMP. While
laboratory measurements are essential in the evaluation of a
gas injection process, an equation of state (EOS) simulation
was demonstrated to be a useful tool in analyzing the phase
behaviour of various injection gases, reservoir fluids, and the
gas-oil interactions.
Introduction
Carbon dioxide gas flooding, when properly designed, has
proven to be one of the most promising tertiary oil recovery
technologies, not only on technical and economic grounds, but
also from an environmental point of view. CO2 flooding can
significantly increase oil recovery from depleted reservoirs
and be eligible for greenhouse gas credits due to CO2
sequestration. A good example of this is the CO2 flood in the
Weyburn field in southeastern Saskatchewan, Canada. The
construction of a CO2 pipeline to deliver large amounts of CO2
from North Dakota to the Weyburn area greatly improved the
feasibility and the economics of this CO2 flooding venture.
It was reported that CO2 flooding is not sensitive to
reservoir lithology, but it is sensitive to reservoir oil
characteristics.[1] In general, reservoirs that have performed
well under waterflooding typically will perform well with CO2
flooding. For gas injection projects, two different
displacement mechanisms—miscible and immiscible—can
occur when CO2 is injected into an oil-containing porous
medium. In miscible flooding, which is an ideal displacement
mechanism for enhanced oil recovery (EOR), multiple-contact
miscibility between the injected gas and oil can be achieved at
pressures above the minimum miscibility pressure and below
the first contact miscible pressure. If the injected fluid is
miscible with the reservoir oil, the interfacial tension at the
displacing front is zero, and the capillary forces that trap the
oil in the reservoir rocks disappear. Thus, the relative
permeability becomes irrelevant resulting in an efficient
displacement. The MMP is a very important parameter that is
required in a well-designed CO2 flood project. If it is operated
well below the MMP, the miscibility between the oil and the
injected gas will most likely not be achieved, and the above-
mentioned advantages will not be available. In this case, oil
recovery will be less than that at pressures near or above the
MMP. Therefore, to properly design a CO2 EOR project, it is
necessary to have a good knowledge of the thermodynamic
properties of the oil–gas mixtures and completely understand
the physical and chemical mechanisms involved in oil and
gas miscibility.
Unfortunately, CO2 sources are rarely pure; the most
likely CO2 source is usually a stream extracted from the flue
gas of a power plant, which contains a certain concentration of
N2 and other gases. Purifying the flue gas will increase
operating costs significantly. In addition, to reduce the net
CO2 utilization and the cost of purchasing CO2, field
operations often employ staged floods and recycle the
produced gas, which usually contains CH4, without processing
it. Key questions are: How can a flooding process maintain
miscibility if flue gas or produced gas is added to the CO2
injection stream? Under what kind of operating conditions can
the miscibility between the injected solvent and the reservoir
2 SPE 89477
oil be maintained without risking the entire miscible flooding
project? In general, contamination of the CO2 injection stream
by N2 or CH4 could substantially increase the CO2 MMP,
while the presence of H2S, C2H6, or intermediate hydrocarbons
(such as C3 or C4) can reduce the CO2 MMP.[2]
Studies aimed at addressing the effect of impure CO2
streams on the MMP were carried out at Saskatchewan
Research Council (SRC) with light oils mixed with pure or
impure CO2. The primary effort was focused on measuring the
minimum miscibility pressure and saturation pressures
(pressure-composition phase diagrams) for these mixtures. It
is beneficial to measure the effects of impure CO2 streams on
the miscibility because CO2 sources diluted by impurities are
common in practice and can substantially improve the
economic prospects of an EOR project.
In addition, gravity-stable CO2 displacement tests with
composite reservoir cores were conducted to evaluate the
flood efficiency at pressures above, near and below the MMP
for tertiary oil recovery of the Weyburn light oil.
Experimental Details
Phase Behaviour, MMP and Saturation Pressure Studies.
The pressure–volume–temperature (PVT) properties, the
MMP, and the bubblepoint pressure of an oil–CO2 mixture
depend on the oil characteristics and the reservoir temperature.
Therefore, laboratory procedures are very important for
evaluating various oil–CO2 mixtures under simulated
reservoir conditions.
Two light oils were collected from Weyburn and
Steelman fields in southeast Saskatchewan and used to
investigate the effects of CO2 impurity on solubility, PVT
properties, and MMP behaviour. The samples were
characterized physically and chemically. Basic sediment and
water in the oil was removed by high-speed centrifugation.
The data, along with the molecular weight and chemical
composition, are listed in Table 1. The stock tank oils from
Weyburn and Steelman fields had API oil gravity of 29.5° and
38°API, and asphaltene contents of 6.7% and 1.0% by
weight, respectively.
The PVT properties and MMP of the oils mixed with pure
CO2 and contaminated CO2 as injection gases were
investigated. The MMP experiments were conducted in the
rising bubble apparatus (RBA) at the reservoir temperature of
59°C for Weyburn reservoir fluid, and 61°C for Steelman field
stock tank oil. Details of the RBA and experimental
procedures can be found in previous SRC papers and
reports.[3-7] Table 2 shows the measured minimum miscibility
pressures for the two light oils using nine different gases, of
which the compositions are also listed in the table.
CO2 streams containing C2 or C3 were also used in the gas
flooding process to study the MMP reduction from the pure
CO2 and the oil. Such a reduction of MMP is beneficial when
an adjustment is required to maintain miscibility in the field
injection process. The experiments found that the CO2 MMP
of Steelman reservoir fluid was 16.5 MPa at 67°C. This
pressure was considered too high for gas injection in the field.
It was decided to bring down the CO2 MMP to a desirable
operating pressure through enrichment of the injection gas.
Two types of hydrocarbon solvents (C2H6 and C3H8) in
combination with CO2 mixed with Steelman reservoir fluid
were studied to compare the effect on MMP of using the
hydrocarbon gas in the CO2 stream.
Coreflood Procedures. Three coreflood tests were conducted
using Weyburn reservoir fluid to investigate the oil recovery
behaviour resulting from CO2 injection. The ultimate oil
recoveries of the tests at operating pressures above, near, and
below the measured MMP at 59°C were compared. These data
were later used for numerical simulation to predict the field
recovery at Weyburn. All coreflood tests were conducted in a
linear core displacement mode. The coreflood system consists
of a core holder, an injection system, a production system and
a data acquisition system. More details of the coreflood
apparatus can be found in previous SRC papers and reports.[8-
12]
After preserved Marly composite cores were saturated with
the dead oil and aged to re-establish wettability, the
reconstituted Weyburn reservoir fluid was injected to displace
the dead oil. The live oil injection was continued until the
produced gas−oil ratio (GOR) was constant and equaled that
of the originally injected reservoir fluid. Following the live oil
injection, the initial waterfloods were carried out in a
horizontal model, whereas the tertiary CO2 injections were
conducted in a gravity-stabilized (vertical downward
direction) mode. After the coreflood tests, the plugs were
cleaned by Dean-Stark Analysis to determine the residual oil
and water saturations. The final oil recovery and connate
water saturation of the composite cores were determined by
material balance.
Results and Discussion
Effect of Gas Solvent on Saturation Pressure of Mixture.
Measurement of the solubility of injection gases in oils is the
basis of the so-called “swelling tests”. These tests also provide
the opportunity of measuring some of the physical properties
of these mixtures, such as their densities, swelling factors, and
viscosities. These data are essential for characterizing fluids in
equation of state simulations, and help understand the various
oil recovery mechanisms, such as swelling and
viscosity reduction.
The saturation pressures of several mixtures of the
Weyburn reservoir fluid as a function of gas solvent (pure and
impure CO2) concentration at 59°C were measured and are
plotted in Figure 1. The figure depicts an almost linear
increase in saturation pressure of the fluid mixture with
addition of CO2 to the reservoir fluid. The saturation pressure
increased from 5.5 MPa at 23.9 mol% CO2 loading, to 8.49
MPa at 44.1 mol% CO2.
The most likely source of CO2 for field flooding projects
is that extracted from stack gas, followed by oilfield-produced
gas. These gas streams have nitrogen and methane as the
predominant components. To reduce operating costs, it is
expected that the produced CO2 from the field can be recycled
and reinjected without further purification. The curves in
Figure 1 indicate that the saturation pressures increased
greatly when CO2 was contaminated by N2 and CH4. For
example, whereas the saturation pressure of Weyburn
reservoir fluid containing about 58 mol% CO2 was 10.8 MPa,
it rose to 13 MPa when the CO2 was contaminated by 9.9
SPE 89477
mol% CH4. When the impurity in the CO2 stream consisted of
5.1 mol% N2 + 5.1 mol% CH4, the saturation pressure was as
high as 16.3 MPa. All three mixtures had essentially the same
reservoir fluid/gas molar ratio: ~42/58 (reservoir fluid to gas).
The experimental results indicate that CH4 in the CO2 stream
is less effective than N2 in raising the saturation pressure of
Weyburn reservoir fluid.
Figure 2 shows a comparison of saturation pressure
changes with pure CO2 and with a CO2–C2H6 mixture as
solvents in Steelman reservoir fluid. Because the reservoir
fluids were reconstituted from two separate oil samples
collected from two wells, which resulted in having two
baseline live oil saturation pressures, the saturation-pressure
increase ratio versus the solvent concentration is reported here.
The saturation-pressure increase ratio is defined as:
P mixture − P liveoil
Ratio = sat sat × 100
P liveoil
sat
The two curves in Figure 2 show clearly that the
saturation pressure increased much faster using CO2 as
injection gas than using a mixture of 60 mol% CO2 + 40 mol%
C2H6. For example, at a solvent concentration of 50 mol%, the
saturation-pressure increasing ratio for solvent CO2 was
116%, which is almost twice the value for the CO2 + C2H6
mixture. The results suggest that CO2 in combination with
intermediate hydrocarbons, such as ethane, is a favourable
solvent for controlling the rise of the saturation pressure of the
fluid systems.
Effect of Gas Solvent on MMP. Minimum miscibility
pressure has been used as an essential criterion for screening
and selecting reservoirs for the miscible displacement process.
For CO2 injection projects, it has been recognized that the CO2
MMP is not sensitive to reservoir lithology, but depends upon
the purity of CO2, reservoir oil characteristics, and
reservoir temperature.
The MMP of pure CO2 with the Weyburn reservoir fluid
was determined by using a rising bubble apparatus. A distinct
pattern was depicted by the CO2 bubbles (see Figure 3). The
left photo in Figure 3 shows the shape of a rising bubble at a
pressure below the MMP. The tail formation on the bubble
shown in the right photograph indicates a dramatic reduction
in the interfacial tension and the diminishing interface
between the CO2 phase and the oil phase as the pressure rises
above the MMP.
Because of the depositional environment of Weyburn
reservoir, the oil-bearing strata dip from north (shallow depth)
to south (substantially deeper). The depth of the reservoir on
the west side is deeper than on the east side. The oil
characteristics can be significantly different within the
reservoir area. The measurements from previous SRC tests[7,
12]
on samples collected from three different locations are
mapped in Figure 4. The stock tank oil density was as high as
885.3 kg/m3 (28.3°API) at the south well (8-12-6-14 W2M)
and as low as 861.9 kg/m3 (32.7°API) at the north well (12-25-
6-14 W2M). At the east well (14-17-6-13 W2M) the density
was measured to be 884.0 kg/m3 (29.5°API).
3
The MMPs at these three locations in the Weyburn reservoir
are also presented in Figure 4. Under the same temperature
(59°C) condition, the CO2 MMPs at the north well (12-25-6-
14 W2M) and at the east well (14-17-6-13 W2M) were 11.5
MPa and 12 MPa, respectively. However, the CO2 MMP was
as high as 15.4 MPa at the south well (8-12-6-14 W2M). The
different oil properties resulted in noticeably different CO2
MMPs within the Weyburn reservoir, wherein the lighter oils
(high API gravity) have lower MMPs than the heavier oils
(lower API gravity).
For a CO2 EOR project, determining the CO2 MMP and
the level of contaminants the CO2 stream can tolerate are key
factors in the technical success of the EOR project and its
economics feasibility. Table 2, and Figure 5 summarize the
MMP behaviour of Weyburn reservoir fluid with pure CO2
and three impure CO2 solvent compositions (referred to as
Gas-2 to Gas-4).[7] The MMP for the Weyburn reservoir fluid
with these impure CO2 solvents at 59°C varied from 14.5 MPa
to 20.5 MPa. These results indicate that pure CO2 and impure
CO2 containing up to 9.9 mol% CH4 or 3.1 mol% N2 + 2.8
mol% CH4 are promising EOR solvents owing to their
relatively low miscibility pressure compared to the reservoir
pressure. However, when N2 contamination in the CO2 stream
was increased to 5.1 mol% in combination with 5.1 mol%
CH4, the MMP rose above the reservoir fracture pressure.
The MMP measurements for the Steelman stock tank oil[3]
with six gases (CO2, Gas-5 to Gas-9) are also listed in Table 2.
The results in Figure 6 show that the presence of SO2 and H2S
in the CO2 gas stream can reduce the CO2 MMP, whereas N2
and O2 tend to have the opposite effect. It indicates that an
impurity may decrease the CO2 MMP if it enhances the
solubility of the driving gas.
If the CO2 minimum miscibility pressure is too high, i.e.
close to the reservoir fracture pressure, it will be impossible to
apply a miscible EOR process. In this case, some intermediate
hydrocarbon gas solvents can be added to the CO2 stream to
bring the CO2 MMP down to a reasonable level. Figure 7
illustrates the MMP measurements for two hydrocarbon gas
solvents mixed with Steelman reservoir fluid.[10] The MMP of
16.5 MPa for pure CO2 decreased by nearly 25% for the
addition of about 16 mol% propane (C3H8) in CO2, and by
approximately 45% for the addition of 37 mol% propane in
CO2. In comparison, even when 40 mol% ethane (C2H6) was
added to the CO2 stream, the MMP was determined to be 13.6
MPa, a reduction of only 18% from the MMP with pure CO2.
The larger reduction in CO2 MMP for propane than for ethane
at the same hydrocarbon concentration in CO2 indicates that
propane is a more effective enriching agent for CO2
than ethane.
EOS Simulation. An EOS simulation package[13] with the
required functionalities and an extensive standard component
library was utilized for tuning and matching the experimental
results and predicting the MMP with respect to various
injection gases with impurities. The Peng-Robinson (PR)
equation of state model was used in this work. The MMP
algorithm is based on the tie-line intersection approach[14,15]
that calculates the minimum miscibility pressure analytically
when an equation of state model is defined.
4 SPE 89477
For Weyburn oil, the available experimental data are the
saturation pressure of the reservoir fluid and the swelling test
results with the injection gas, which provide the measured
saturation pressure points at selected solvent concentrations in
the mixture. Since all data involved are related to the fluid
phase behaviour, only saturation pressure points were selected
for the tuning processes. The plus fraction C6+ of the original
reservoir fluid was first characterized into five pseudo-
components using its molecular weight and specific gravity.
Typically, in the compositional analysis, the heavy end
compositions are corrected by using the stabilized stock tank
oil molecular weight measured using the freezing point
depression method. The molecular weight measurement
usually has an error in the range of 10%, which implies that it
is adequate to adjust the plus fraction molecular weight by
10%. In this simulation, 4.2% of the plus fraction molecular
weight was added to improve the tuning results of the EOS
model. Using the automatic regression function, the model
was tuned to match the pressure-composition (P-x) data
resulting from the swelling test given in Figure 8. The
selected regression parameters include the critical pressure,
critical temperature and accentric factor of the heaviest
pseudo-component, and all binary interaction coefficients. All
P-x data matched the experimental results very well as shown
in Figure 8. The tuned model was then used for calculating the
MMPs of the oil−gas system. The calculated MMP results are
compared in Table 2. All MMP predictions were consistent
with the experimental measurements, except the system with
5.1% N2 for which the MMP is slightly off. All calculations
show the miscibility mechanism as combined vaporizing
and condensing.
For Steelman stock tank oil, there were no phase
behaviour data available for tuning the EOS model.
Nevertheless, the EOS model can predict the MMP for the oil
system with the various gas mixtures listed in Table 2 using
default parameter values. The C6+ fraction of the dead oil,
which amounts to over 96% of the oil composition, was again
characterized into five pseudo-components using its molecular
weight and specific gravity. Applying similar reasoning as
above, 2.3% of the plus fraction molecular weight was added
to improve the prediction results of the model. It was found
that setting all binary interaction coefficients relating to CO2
to zero can improve the MMP prediction with pure CO2. The
same EOS model was used to predict the MMPs with different
solvent mixtures. The results are compared in Table 2. The
Steelman oil MMPs with pure CO2, CO2 + H2S, and CO2 +
SO2 are adequate, while the MMPs with the remaining three
gas streams show significant discrepancies, which appear to
have some correlation with the existence of N2 in the gas
stream. The gas streams containing H2S and SO2 appear to be
able to reduce the MMP value, which is consistently
demonstrated in the EOS results. All calculations show the
miscibility mechanism as combined vaporizing
and condensing.
Effect of Miscibility on Oil Recovery. Three gravity-
stabilized tertiary CO2 coreflood experiments (Runs 2, 4, and
5) using the stacked composite Weyburn cores (preserved/
restored) were conducted to determine the effect of injection
pressure, (above, near and below the MMP) on oil recovery.
As discussed in the previous section, the CO2 MMP for the
Weyburn reservoir fluid was determined to be 11.5 MPa.
Therefore, the CO2 injection pressures selected for the
coreflood tests were 10 MPa (below the MMP, Run 5), 12
MPa (near and at the MMP, Run 4), and 16 MPa (above the
MMP, Run 2).
Figure 9 presents the cumulative oil recovery as a
function of pore volumes of fluid injected for the three tests. It
should be noted that all these total (or ultimate) oil recovery
values include the initial waterflood, CO2 and extended
waterflood (EWF) stages. Figure 10 is a comparison of
enhanced oil recovery (CO2 plus EWF stages only) expressed
as residual oil in place for the tests. Overall, the 12 MPa flood
(Run 4) exhibited the highest tertiary (57.0% ROIP or 29.9%
IOIP) and total oil recovery (77.3% IOIP). The coreflood
conducted at 10 MPa (Run 5) displayed the lowest tertiary
(39.1% ROIP or 21.5% IOIP) and total oil recovery (66.5%
IOIP). Run 2 carried out at 16 MPa showed tertiary (46.4%
ROIP or 25.9% IOIP) and total oil recovery (70.1% IOIP)
comparable to but slightly lower than that obtained for the 12
MPa run, possibly because of the considerably smaller CO2
slug size. These test results indicate that the tertiary CO2 flood
conducted at pressures near or above the MMP would result in
a higher oil recovery than those conducted below the MMP.
Conclusions
The following conclusions can be drawn on the basis of
experimental and theoretical studies presented above.
1. The minimum miscibility pressure for Weyburn reservoir
fluid with CO2 containing about 10% impurity can
increase by 70% from 11.8MPa to 20.5MPa. The MMP is
a strong function of the impurity in the CO2 stream and
reservoir oil characteristics.
2. Pure CO2 and impure CO2 containing up to 9.9 mol% CH4
or 3.1 mol% N2 + 2.8 mol% CH4 are promising solvents
for enhanced oil recovery in the Weyburn field because of
their relatively low MMP. However, any further
contamination would make it necessary to further purify
the CO2 or enrich it with intermediate hydrocarbon gases.
3. For Steelman oil, the addition of hydrocarbon gas in the
CO2 stream was shown to reduce the MMP. Carbon
dioxide containing 37 mol% propane reduced the pure
CO2 MMP by 45%. Propane was a more effective
enriching agent in the CO2 stream than ethane for
MMP reduction.
4. EOS simulation was demonstrated to be an effective tool
in analyzing the miscibility of oil−gas systems. A tuned
EOS model adequately predicted the MMP with various
injection gases when they were within the tuning scope.
The by-product of the MMP calculation was the ability to
determine the miscibility mechanism of the fluid system.
5. The gravity-stabilized tertiary CO2 floods using the
stacked composite Weyburn core showed that floods
conducted at near or above the MMP are likely to result in
a higher oil recovery than that conducted below the MMP.
Near-miscible CO2 displacement can result in good
oil recovery.
SPE 89477 5

Acknowledgment
The authors wish to express their thanks to Saskatchewan
Research Council for permitting publication of this paper.
Acknowledgment is gratefully extended to B. Tacik for her
assistance in the preparation of the manuscript.

References
1. “CO2 Flooding Increase Recovery,” www.pttc.org/solution/
12.htm, July 9, 2003.
2. Alston, R.B., Kokolis, G.P. and James, C.F.: “CO2 Minimum
Miscibility Pressure: A Correlation for Impure CO2 Streams and
Live Oil Systems,” Society of Petroleum Engineers Journal,
(April 1985).
3. Huang, S. and Dong, M.: “Potential of Greenhouse Gas Storage
and Utilization through Enhanced Oil Recovery,” Saskatchewan
Research Council Publication No. P-110-468-C-99,
(September 1999).
4. Huang, S. and Dyer, S.B.: “Miscible Displacement in the
Weyburn Reservoir − A Laboratory Study,” Journal of
Canadian Petroleum Technology, (September 1993), 42−52.
5. Srivastava, R.K. and Huang, S.: “New Interpretation for
Determining Minimum Miscible Pressure by Rising Bubble
Apparatus for Enriched-Gas Drives,” SPE 39566 presented at
the 1998 SPE India Oil and Gas Conference and Exhibition held
in New Delhi, India, April 7-9, 1998.
6. Dong, M., Huang, S., Schnell, B. and Meyer, G.: “Analysis of
Miscibility Development by Rising Bubble Technique for CO2
Miscible Injection,” paper CIM 200-37 presented at the
Petroleum Society’s Canadian International Petroleum
Conference 2000, Calgary, June 4-8, 2000.
7. Dong, M., Huang, S., Dyer, S.B., and Mourits, F.M.: “A
Comparison of CO2 Minimum Miscibility Pressure
Determinations for Weyburn Crude Oil,” Journal of Petroleum
Science and Engineering (2001) 31, 13-22.
8. Srivastava, R.K. and Huang, S.: “Technical Feasibility of CO2
Flooding in Weyburn Reservoir − A Laboratory Investigation,”
Journal of Canadian Petroleum Technology,
(November 1997), 48−55.
9. Srivastava, R.K., Huang, S.S. and Dong, M.: “Laboratory
Investigation of Weyburn CO2 Miscible Flooding,” Journal of
Canadian Petroleum Technology, (February 2000), 41−51.
10. Dong, M., Huang, S. and Srivastava, R.: “A Laboratory Study
on Near-Miscible CO2 Injection in Steelman Reservoir,” Journal
of Canadian Petroleum Technology, (February 2001), 53−61.
11. Dong, M., Huang, S. and Srivastava, R.: “Coreflood Studies of
Tertiary CO2 Flood in Naturally Fractured Midale Formation in
Southeast Saskatchewan,” Journal of Canadian Petroleum
Technology, (February 2002), 41−46.
12. Zhang, P.Y., Freitag, N., and Huang, S.: “Baseline Study of Oil
Recovery Parameters at Pre-CO2-Injection Conditions−Weyburn
CO2 Monitoring and Storage Study,” Saskatchewan Research
Council Publication No. P-110-557-C02, (February 2002).
13. PVTprop – An EOS Simulation Package, Version 4.0;
Protroleum Technologies Ltd., http://www.pvtprop.com
14. Wang, Y. and Orr Jr., F.M.: “Calculation of Minimum
Miscibility Pressure,” SPE 39683 presented at the 1998
SPE/DOE Improved Oil Recovery Symposium held in Tulsa,
Oklahoma, April 19-22, 1998.
15. Jessen, K., Michelsen, M.L. and Stenby, E.H.; “Effective
Algorithm for Calculation of Minimum Miscibility Pressure”,
SPE 50632 presented at the 1998 SPE European Petroleum
Conference held in The Hague, The Netherlands,
October 20-22, 1998.
6 SPE 89477

Table 1−Chemical and Physical Properties of Cleaned Dead Oils

Samples Weyburn (14-17-6-13 W2M) Steelman (8-16-4-6 W2M)
Temperature (°C) 15 20 59 15 20 67
3 878.2 834.4
Density (kg/m ) @ 1 atm 874.0 846.1 832.0 804.3
(29.5°API) (38°API)
Viscosity (mPa·s) @ 1 atm - 11.3 4.2 4.02 3.44 1.64
BS&W (vol%) Trace Trace
Oil MW (g/g-mol) 232 191
Saturates (wt%) 39.5 57.1
Aromatics (wt%) 45.4 32.2
Resins (wt%) 8.4 9.7
Asphaltenes (wt%) 6.7 1.0

Table 2−Comparison of Experimental and EOS Calculated Minimum Miscibility Pressures

1 2
Oil Sample Weyburn Live Oil at 59°C Steelman STO at 61°C
Exp. Calc. Exp. Calc.
Error Error
Type of Gas in CO2 Stream MMP MMP MMP MMP
(%) (%)
(MPa) (MPa) (MPa) (MPa)
Pure CO2 11.7 12.0 2.6 11.8 12.0 1.7
Gas-2 (94.1% CO2 + 3.1% N2 + 2.8% CH4) 14.5 14.0 3.4 - -
Gas-3 (90.1% CO2 + 9.9% CH4) 16.0 16.3 1.9 - -
Gas-4 (89.8% CO2 + 5.1% N2 + 5.1% CH4) 20.5 18.3 10.7 - -
Gas-5 (70% CO2 + 30% H2S) - - 10.4 10.8 3.8
Gas-6 (70% CO2 + 30% SO2) - - 8.7 7.7 11.5
Gas-7 (85% CO2 + 15% N2) - - 33.0 22.5 31.8
Gas-8 (65% CO2 + 15% N2 + 20% SO2) - - 34.0 17.3 49.1
Gas-9 (80% CO2 + 5% N2 + 5% O2 + 10% SO2) - - 22.5 16.8 25.3
1− Weyburn Live Oil was reconstituted by the separator oil and separator gas collected from Well 14-17-6-13 W2M.
2− Steelman Stock Tank Oil (STO) was collected from Well 3B-15-4-6 W2M.

20
Pure CO2 (Gas-1)
18
90.1% CO2 + 9.9% CH4 (Gas-3)

16 89.8% CO2 + 5.1% N2 + 5.1% CH4 (Gas-4)

Saturation Pressure (MPa)

14

12

10

2
0 10 20 30 40 50 60 70
Gas Solvent Concentration in Reservoir Fluid (mol%)

Figure 1− Saturation Pressure as a Function of Gas Solvent Concentration in

Weyburn Reservoir Fluid at 59°C
SPE 89477 7

120

CO2 (Oil Sample from Well 3-21-4-5 W2M)

60% CO2 + 40% C2H6 (Oil Sample from Well 8-16-4-6 W2M)
Saturation Pressure Increase Ratio (%)

100

80

60

40

20

0
0 5 10 15 20 25 30 35 40 45 50 55
Gas Concentration in Reservoir Fluid (mol%)

Figure 2−Saturation Pressure Increasing Ratio as a Function of Gas Concentration in

Steelman Reservoir Fluid at 67°C

MMP = 12 MPa

At 7.3 MPa and 59°C At 12.9 MPa and 59°C

(Below MMP) (Above MMP)

Figure 3−CO2 MMP from Rising Bubble for Weyburn Oil Sample at 59°C (Well 14-17-6-13 W2M)
8 SPE 89477

R 14W2 R 13W2

25 30 29
12-25-6-14W2 (32.7°API)
MMP = 11.5 MPa

24 19 20
14-17-6-13W2 (29.5°API)
MMP = 12 MPa
T6 T6

13 18 17

8-12-6-14W2 (28.3°API)
MMP = 15.4 MPa

12 7 8

Figure 4−CO2 MMP Distribution within Weyburn Reservoir

22
89.8% CO 2 + 5.1% N 2 + 5.1% CH 4

20

18
Minimum Miscibility Pressure (MPa)

90.1% CO2 + 9.9% CH4

16
94.1% CO2 + 3.1% N2 + 2.8% CH4
MMP of Pure CO2
14

Pure CO2
12

10

0
Gas-1 Gas-2 Gas-3 Gas-4
Type of Gas Solvent in System

Figure 5−Effect of Contaminated CO2 on MMP for Weyburn Reservoir Fluid at 59°C
SPE 89477 9

35

Minimum Miscibility Pressure (MPa) 30

25

20

MMP of Pure CO2

15

10

0
Pure CO2 Gas-5 Gas-6 Gas-7 Gas-8 Gas-9
Type of Gas Solvent in System

Figure 6−Effect of Contaminated CO2 on MMP for Steelman Stock Tank Oil at 61°C

17

Propane in CO2
16
Ethane in CO2
Minimum Miscibility Pressure (MPa)

15

14

13

12

11

10

8
0 10 20 30 40 50 60 70 80 90 100
Gas Solvent Concentration in CO2 Stream (mol%)

Figure 7−Measured MMP for Steelman Reservoir Fluid with CO2/C2 and CO2/C3
Gas Mixtures at 67°C
10 SPE 89477

20
Calculated for Pure CO2 (Gas-1)
Exp. Data for Pure CO2 (Gas-1)
18
Calculated for Gas-4
Exp. Data for Gas-4
16 Calculated for Gas-3
Exp. Data for Gas-3
Saturation Pressure (MPa)

14

12

10

2
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Solvent Concentration (mole fraction)

Figure 8−Measured and Matched Saturation Pressure for Mixtures of Weyburn Reservoir Fluid
with Pure and Impure CO2 at 59°C

80

70
Cumulative Oil Recovery (% IOIP)

60

50

Started EWF
40

30
Started CO 2 Injection
20
12 MPa (Run 4)

10 16 MPa (Run 2)

10 MPa (Run 5)
0
0 1 2 3 4 5 6 7 8 9 10 11 12
Fluid Injected (PV)

Figure 9−Effect of CO2 Pressure on Cumulative Oil Recovery for Weyburn Reservoir Fluid at 59°C
SPE 89477 11

60

50
Total Enhanced Oil Recovery (%ROIP)

40

30

20

10

0
10 12 16
CO2 Injection Pressure (MPa)

Figure 10−Effect of CO2 Injection Pressure on Total Enhanced Oil Recovery for Weyburn
Reservoir Fluid at 59°C