SPE 169734 Polymer-Alternating-Gas Simulation—A Case Study

SPE 169734

Polymer-Alternating-Gas Simulation—A Case Study

Weirong Li, Jianlei Sun, and David S. Schechter, SPE, Texas A&M University

Copyright 2014, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE EOR Conference at Oil and Gas West Asia held in Muscat, Oman, 31 March–2 April 2014.

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 have not been
reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect any position of the Society of Petroleum Engineers, its
officers, or members. Electronic reproduction, distribution, or storage of any part of this paper 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 SPE copyright.

Abstract
Carbon dioxide has been used commercially to recover oil from reservoirs for more than 40 years. Currently, CO2 flooding is
the second most applied enhanced oil recovery (EOR) process in the world behind steam flooding. Water alternating gas
(WAG) injection has been a popular method to control mobility and improve volumetric sweep efficiency for CO2 flooding.
Average EOR is about 9.7% with a range between 6 and 20% for miscible WAG injection. Despite all the success of WAG
injection, sweep efficiency during CO2 flooding is a typical challenge to reach higher oil recovery.

We applied polymer-alternating-gas (PAG), in which polymer flooding is combined with miscible CO2 injection, to
improve the volumetric sweep efficiency of the WAG process in TR78 of the North Burbank Unit. High heterogeneity and
high permeability at the top layers are the two main challenges of the North Burbank Unit. Then, we compared PAG
performance with WAG and continuous gas injection (CGI). Polymer concentration and injection slug patterns were
optimized during the PAG process. Simulation results show that PAG would increase oil recovery about 14.3% compared
with 7.3% for WAG in TR78.

This study can not only be used to guide the development of the North Burbank Unit, but also to demonstrate that
encouraging recovery can be obtained in high-heterogeneity reservoirs by using PAG.

Introduction
Worldwide oil production in 2009 was about 31.4 billion bbl, with 3.5% (1.1 billion bbl) of crude oil from EOR (Cabrera and
Manrique 2010). Chemical and thermal methods represent 60 to 70% of worldwide EOR production, and gas flooding
represents up to 32% of total EOR production. According to the 2012 Worldwide EOR survey (Koottungal 2012), production
from US miscible CO2 flooding surpasses the production from steam flooding.

Although CO2 flooding is a well-established EOR technique, its density and viscosity nature is a challenge of CO2 projects.
Low density (0.5 to 0.8 g/cm3) causes gas to rise upward in reservoirs and bypass many lower portions of the reservoir. Low
viscosity (0.02 to 0.08 cp) would lead to poor volumetric sweep efficiency. In heterogeneous reservoirs with high-
permeability zones and natural fractures, the condition is even worse (Zhang et al. 2010). The following methods have been
studied to improve CO2 flooding performance.

WAG. Almost all commercial miscible gas injection projects use WAG to control mobility of gas and alleviate fingering
problems. Recovery of WAG is better than gas injection alone, and 80% of commercial WAG projects in the US are
economic (Christensen et al. 1998). However, recent studies show that most of the fields could not reach the excepted
recovery factor from the WAG process, especially for reservoirs with high-permeability zones or there are naturally fractured
(Christensen et al. 2001).

Gel. Gel application is considered the most aggressive type of conformance control. Gel acts as a blocking agent to reduce
channeling through fractures or high-permeability zones of reservoirs (Ali and Schechter 2013). The most applied gel system
in the oil industry for conformance control is hydrolyzed polyacrylamide (HPAM) with Cr (щ) acetate. Woods et al. (1986)
presented one of the earliest successful gel treatments for Lick Creek field in Arkansas. Hild and Wackowski (1999) reported
a successful gel treatment at the Rangely Weber Sand Unit in northwestern Colorado. In this treatment, a large-volume
(10,000 bbl) chromic-acetate acrylamide polymer gel was applied to improve CO2 flooding performance. The cost of the gel
treatment was estimated to be around USD 6 to 8/bbl, and the project return rate was 365%. Karaoguz et al. (2007) and
Topguder (2010) reported several field applications of gel in Bati Raman field. In General, gels can treat water coning
successfully in reservoir with vertical fracture. However, water coning through matrix reservoir is very difficult to be treated

2 SPE 169734

successfully with gels. On the other hand, conventional foams are considered effectively in matrix rock and are not applicable
in reservoir fracture channels with aperture widths on the order of greater than 0.5 mm (Robert 2007).

Foam-Assisted WAG (FAWAG). Bond and Holbrook (1958) first proposed the idea of using foam for mobility control.
Since then, CO2 foam with surfactant has been used as an effective mobility-reducing agent for CO2 flooding in the oil
recovery process. One of the largest full-scale field demonstrations of foam for gas mobility control was the FAWAG project
in the Snorre field on the Norwegian Continental Shelf from 1997 to 2000 (Blaker et al. 2002). Unfortunately, field
experiences showed that conventional foam with surfactant injected in water had some significant limitations. Enick et al.
(2012) concluded the problems of FAWAG: (1) the dilution of CO2 foam by subsequently injected water; (2) the inability of
foam to be effective in formations containing fractures or extremely high-permeability open flow paths; (3) the very short
propagation of the CO2 from the injection well, cold weather ice and hydrate formation, unacceptably large decreases in
injectivity associated with coinjection, and other unspecified “operational problems.”

CO2 Viscosifier. CO2 viscosifier (direct thickeners) is the most direct way to increase the viscosity of CO2, hence improving
the overall sweep efficiency. A high-molecular-weight polymer and cosolvent are blended and pressurized together with CO2
so that the CO2 viscosity can be greatly increased before CO2 is injected. A number of studies show that gas viscosifier
chemicals can increase CO2 viscosity by an order of one to two and can control CO2 mobility (Heller et al. 1985, Terry et al.
1987, McClain et al. 1996, Bae and Irani 1993, Ali and Schechter 2013). However, the main barriers of using viscosifer
include the following: (1) the large volume requirement of cosolvent makes pilot-testing costs prohibitive (Enick et al. 2012);
(2) copolymers do not dissolve in CO2 unless pressure far exceeds the minimum miscibility pressure (MMP) (Enick et al.
2012); (3) the environmental impact is negative (Kulkarni and Rao 2005).

Simultaneous Water and Solvent Injection (SWAG). Caudle and Dyes (1958) carried out the first SWAG study. They
found that one way to improve the miscible displacement sweep efficiency was to lower the mobility behind the flooding
front by injecting water with miscible gas. This reduced relative permeability to gas and lowered total mobility. Their
laboratory study showed that sweep efficiency for a five-spot pattern could be increased to 90% with SWAG. However, only
60% of oil was recovered with CGI. Stone (2004) showed that the water-above-gas process can reach better sweep efficiency
than simultaneous coinjection from the same location and can increase injectivity. Sohrabi et al. (2005) studied SWAG
residual oil and water/gas injection ratio. Faisal et al. (2009) suggested that a small slug size instead of coinjection water and
gas would increase SWAG injectivity. Aleidan and Mamora (2010) studied how salt concentration affects SWAG/WAG
recovery. Like CO2 viscosifier, most SWAG researches are still in the experimental stage.

PAG. To overcome the issues of gas breakthrough and gravity segregation, such as during SWAG, a new combination
method was proposed. This new method, termed PAG, combines features of CO2 flooding with polymer flooding to produce
a chemically enhanced WAG flooding. Coupling polymers with CO2 is expected to improve the efficiency of the current
WAG. The main feature of PAG is that the polymer is injected with water throughout the WAG process. Zhang et al (2010)
conducted polymer injection chased with gas alternative water (PGAW) experiment based on Saskatchewan crude. They
stated that coupled CO2 and polymer injection gave better recovery and efficiency than WAG and polymer flooding.
Majidaie et al. (2012) carried out the first coupled CO2 and polymer injection simulation study for light oil based on a
synthetic and homogeneous model. This study showed that PAG and WAG have almost the same recovery. He also
mentioned that chemical slug of polymer with surfactant and alkali would significantly increase oil recovery.

Workflow
In this study, we discuss PAG flooding in light oils based on the field model, TR78 of the North Burbank Unit. The
commercial software E100 was used, which is a simulator that can model both the solvent process and polymer flooding. The
main steps in the simulation study are as follows:

(1) Build pressure-volume-temperature (PVT) model with matching PVT data
(2) Calculate oil viscosity reduction and MMP from the PVT model
(3) Build a reservoir model
(4) Validate the reservoir model by matching primary and secondary production data
(5) Define polymer parameters
(6) Validate pseudo-miscible model
(7) Evaluate and compare the performance of PAG with polymer flooding, WAG and CGI
(8) Optimize polymer concentration and slug patterns to increase polymer utilization

Reservoir Simulation of PAG in TR78 of the North Burbank Unit
The North Burbank Unit, located in Osage County, was originally discovered in 1920. It is a reservoir with high-permeability
layers at the top, high Dykstra-Parsons permeability variation (VDP) factors, strong oil-wettness, and east-west trending
vertical fractures (Trantham et al. 1980). It has an extensive history of activity, including primary depletion, produced gas
cycling, and water and polymer flooding to the point of very high water cut at current conditions. The current oil production

SPE 169734 3

rate in the North Burbank Unit is approximately 1,400 bbl of oil/day from 360 active wells with a water cut of 99.5%. The
North Burbank Unit has a cumulative production of 332 million bbl of oil out of an estimated 824 million bbl of original oil
in place. Significant reserves are currently available for post-secondary production. CO2 flooding is a good choice because
MMP in the North Burbank Unit is lower than fracturing pressure, and CO2 is available from a purely anthropogenic source.

PVT Model. The Peng-Robinson equation of state (EOS) was used in this study for fluid modeling of the North Burbank
Unit. After lumping 40 components into 11 pseudo-components, molecular weight, critical pressure, critical temperature,
binary interaction coefficients, and Pedersen viscosity coefficients were regressed to match experimental data with simulated
data. Figs. 1 to 5 show the matching results after regression, which increase our confidence in the PVT model that we set up
and used in predictions by the reservoir simulator. It is worth mentioning that as CO2 fraction increases from 0.0 to 0.75 with
a constant temperature of 122 degF, the swelling factor for light oil increases to 1.58 (Fig. 5). Table 1 shows the 11-pseudo-
component fluid system and the parameters used in both the compositional model and black oil model.

Based on the PVT model we obtained, we calculated fluid viscosity and MMP. As CO2 mole fraction increases from 0.0 to
0.75, fluid viscosity declines from 3 to 1 cp (Fig. 6). Calculated MMP for the North Burbank Unit oil is about 1,680 psi (Fig.
7). Considering initial reservoir pressure is about 1,600 psi and the highest injection pressure is about 2,100 psi, CO2 miscible
flooding is possible for this light oil formation.

100%

Final ROV Exp. ROV

Liquid Relative Volume 99%

98%

97%

0 1000 2000 3000 4000 5000

Pressure, psia

Fig. 1—Constant composition expansion (CCE) liquid relative volume matching result

10

Final Oil Visc. Exp. Oil Visc.

8

Oil Viscosity, cp 6

4

2

0

0 1000 2000 3000 4000 5000

Pressure, psia

Fig. 2—CCE oil viscosity matching result

4 SPE 169734

56

Final Oil density Exp. Oil density

Oil Density, lb/ft354

52

50 1000 2000 3000 4000 5000
0 0.80
Pressure, psia
3000
2500 Fig. 3—CCE oil density matching result
2000
1500 Final Psat Exp. Psat
1000
Saturation Pressure, psia 500

0 0.20 0.40 0.60
0.00 CO2 Mole Fraction

Fig. 4—Saturation pressure matching result

SPE 169734 5

2.0 Final S.F. Exp. S.F.
1.5
Swelling Volume Factor
1.0

0.5

0.0 0.20 0.40 0.60 0.80
0.00 CO2 Mole Fraction

Fig. 5—Swelling factor matching result

Table 1—Component Fluid System and Parameters

Component Specific Mole Weight, Pc, Tc, Acentric Factor Composition,
Gravity g/mol atm K % (in 2010)
N2
CO2 0.81 28.01 33.50 126.20 0.04 0.17
CH4 0.82 0.18
C2H6 0.30 44.01 72.80 304.20 0.23 1.61
C3H8 0.36 1.18
IC4 to NC4 0.51 16.04 45.40 190.60 0.01 1.77
IC5 to NC5 0.58 2.89
FC6 0.63 30.07 48.20 305.40 0.10 3.64
C7 to C10 0.69 3.56
C11 to C17 0.75 44.10 41.90 369.80 0.15 30.53
C18 to C30 0.82 29.04
0.91 58.12 37.19 421.70 0.19 25.43

72.15 33.34 466.14 0.24

84.00 32.46 507.50 0.28

113.99 30.10 583.94 0.37

183.39 23.72 691.94 0.58

431.20 14.31 826.85 0.94

6 SPE 169734

4

3

Fluid Viscosity, cp 2

1

0 0.8
0 0.2 0.4 0.6
CO2 Mole Fraction
Fig. 6— Correlation of fluid viscosity versus CO2 mole fraction

100

96

Oil Recovery Factor, % 92

88

84

80 1400 1600 1800 2000 2200 2400 2600
1200 Pressure, psia

Fig. 7—Calculated MMP from software

Reservoir Model. TR78 in the North Burbank Unit, a typical section with high heterogeneity of the reservoir and high
permeability at the top layers, was selected for the purpose of modeling. TR78 is characterized by a gridded network with
permeability and porosity parameters specified for each block. For this model, a 0.5 × 0.5-mi reservoir section was divided
into 60 grid blocks in the x-direction, 60 grid blocks in the y-direction, and 10 grid blocks in the z-direction. In the x- and y-
directions, the grid blocks are 44 ft in length. The grid blocks in the z-direction vary from 4 to 24 ft, which results in a pay
zone of 90.7 ft. Fig. 8 shows x-horizontal permeability (kh) in the model. The vertical permeability (kv) is 0.01 times the x-
horizontal permeability, while y-horizontal permeability is three times the x-horizontal permeability. Table 2 presents the
input reservoir rock and fluid properties used for the simulation.

SPE 169734 7

0 X-horizontal Permeability, md 300
50 100 150 200 250
1
Layer 2
3
4
5
6
7
8
9
10
11
12

a) X-horizontal permeability distribution b) X-horizontal permeability value by layers

Fig. 8—X-horizontal permeability model

Table 2—Reservoir Rock and Fluid Properties

Reservoir Rock Reservoir Fluid

Parameters Values Parameters Values
Size of Model, ft 2,640×2,640×90.7 Water Density, lb/ft3 62.97

Number of Grid 60×60×10 Water Viscosity, cp 0.5
VDP 0.85 Oil Density, lb/ft3 50 to 52

kv/kh 0.01 Oil Viscosity, cp 2 to 4

Porosity 0.15 to 0.27 Initial Oil Saturation 0.61 to 0.80

Initial Pressure, psi 1350 Initial Water Saturation 0.20 to 0.39

Permeability, mD 6 to 230

Reservoir Model Validation Before CO2 Flooding. A simulation was run to match oil and water production during primary
depletion and secondary development with the reservoir model described above. Liquid production rate was used as the
primary constraint in the simulation. A good match of oil rate and water cut was reached (Fig. 9), which validates the
reservoir model.

8 SPE 169734

10000 Oil Rate, bbl/day Oil Rate--simulation result 1.00 Water Cut
1000 Oil Rate--production data 0.75
100 Water cut--simulation result 0.50
Water cut--production data

10 0.25

1 1927 1941 1954 1968 1982 1995 2009 0.00
1913 Year 2023

Fig. 9—History matching result of oil rate and water cut

Parameters for Polymer Flooding. Rock adsorption and polymer viscosity are two important parameters for polymer
flooding. The correlations of rock adsorption and polymer viscosity with polymer concentration shown in Figs. 10 and 11
were used for the polymer flooding simulation in this study. We also assumed a residual resistance factor (RRF) value of 1.5
at 0.50 lb/stb in this study.

40

Polymer Viscosity, cp 30

20

10

0 0.100 0.200 0.300 0.400 0.500 0.600
0.000

Polymer Concentration, lb/stb
Fig. 10—Correlation between polymer concentration and polymer viscosity

SPE 169734 9

60

Rock Adsorption, ug/g 50
40

30

20

10

0 1.20
0.00 0.20 0.40 0.60 0.80 1.00

Polymer Concentration, lb/stb
Fig. 11—Correlation between polymer concentration and rock adsorption

Pseudo-Miscible Model Validation. A pseudo-miscible model was used in E100 to model the miscible and solvent. E100
introduces a mixing parameter, ω, which determines the amount of mixing between the miscible fluids within a grid block. A
value of zero corresponds to the case of a negligible dispersion rate, whereas a value of one corresponds to complete mixing.
Because the fluid model was changed from a compositional model to a black oil model, it was necessary to test the fluid
model consistency with existing data. In this study, the correlation of mixing parameter ω and pressure was obtained (Fig.
12) by matching production data predicted by the pseudo-miscible model with the compositional model (Fig. 13). The
highest ω value is 0.6, and the minimum miscible pressure is 1,700 psi (Fig. 12), which is similar to the MMP value from the
experiment and PVT calculation.

0.8

0.6

Mixing Parameter 0.4

0.2

0 3500
0 500 1000 1500 2000 2500 3000

Pressure, psi
Fig. 12— Correlation between mixing parameter and pressure

10 SPE 169734

400 Compositional Model
Pseudo-miscible Model
300

Oil Rate, bbl/day 200

100

0
0 1000 2000 3000 4000 5000 6000 7000 8000

Days

Fig. 13— Production matching between pseudo-miscible model and composition model

PAG Versus Polymer Flooding, WAG and CGI. In the production history of NBU, surfactant/polymer flooding and
polymer flooding have been carried in 1980s. The commercializing surfactant/polymer pilot in Tract 97 is considered
economic unfavorable particularly in areas of high heterogeneity (Joseph 1983, and Moffitt et al. 1993). While the freshwater
polymer flood project at NBU block A (included 84 producers and 36 injectors) is technically and economically successful,
even with the large drop in oil prices in 1986. In this project, 4.2 million lbm of polyacrylamide and 4.0 million lbm of 2.9
aluminum citrate crosslinking were injection and incremental oil recovery exceeded 2.5 MMSTB of oil (Tracy and Dauben
1982, Joseph and Paul 1982, Zornes et al. 1986, Moffitt et al. 1993). It is worth to compare the performance between polymer
flooding and gas flooding in this field.

The reservoir performance during PAG was compared with polymer flooding, WAG and CGI. Simulation results show
that oil recovery from PAG with a polymer concentration of 0.06 lb/stb is more than polymer flooding (with a polymer
concentration of 0.10 lb/stb for twenty years), CGI and WAG (Fig. 14), which indicates that oil recovery increases with
polymer injection.

We forecasted oil production rate for the three different EOR process (Fig. 15). The oil rate by CGI is lower than WAG
and PAG mainly due to gas breakthrough and gas override. However, the highest oil rate of CGI is reached after 24 months
injection, while it takes 34 and 30 months for WAG and PAG to reach oil rate peak, respectively. Peak oil rate from polymer
flooding is lower than gas injection, and the oil rate declines slower than gas flooding. PAG has two oil rate peaks. One is
due to gas injection (Phase I), and the other is a result of polymer injection (Phase II).

Fig. 16 shows the gas-oil ratio of these three processes. Gas production occurs after 6-month CO2 injection for CGI, while
it takes 10 months for WAG and PAG. Gas breakthrough occurs after 2 year CO2 injection for CGI, and it takes 4 years for
WAG. We do not find significantly gas breakthrough in PAG precess. PAG does improve the breakthrough. The gas ratio of
PAG is much lower than WAG, which means that PAG would reduce gas production and more CO2 would be left in
reservoir.

SPE 169734 11

16

12

Recovery Factor,% 8

4

0 CGI WAG Polymer Flooding PAG-0.06
lb/stb
400 0.10 lb/stb
350
300 EOR methods
250
200 Fig. 14—Recovery factor of different EOR processes
150
100 CGI
WAG
50 PAG
0 Polymer flooding

Oil Rate, bbl/day 0

1000 2000 3000 4000 5000 6000 7000 8000
Days

Fig. 15—Production rate of different processes

12 SPE 169734

160 CGI WAG PAG

120
Gas/Oil Ratio, Mscf/bbl
80

40

0
0 1000 2000 3000 4000 5000 6000 7000 8000

Days
Fig. 16—Gas-oil ratio of different processes

Fig. 17 shows the water saturation distribution of the first layer for WAG and PAG, respectively. Water sweep efficiency
of the PAG process is significantly increased compared with WAG. We compared the percentage of water injected into each
zone for WAG and PAG (Fig. 18). More water is injected into middle- and lower-permeability zones in PAG (40%) than in
WAG (27%). The improvement of vertical and areal sweep efficiency is a consequence of improving water/oil mobility ratio.
In addition, the polymer reduces the contrasts in permeability by preferentially plugging the high-permeability zones. This
forces the water to flood the lower-permeability zones and increases the sweep efficiency.

PAG WAG
Water Saturation Injector

Producer

Fig. 17—Layer-1 water saturation distribution after WAG/PAG flooding

SPE 169734 13

0% 10% 20% 30% 40% 50% 60% 70% 80%

High
Permeability

Zone

Middle WAG PAG
Permeability

Zone

Low
Permeability

Zone

Fig. 18—Percentage of water injected into different permeability zones for WAG and PAG

Sensitivity Analysis—Polymer Concentration. A sensitivity analysis was performed on polymer concentration which
affects the process significantly. We performed four simulation cases with different concentrations of polymer injected with
water all the time, which are presented in Table 3. Different polymer concentration yields different injection fluid viscosity.
In this study, bottom-hole pressure at injectors is set to 2,100 psi which is fracturing pressure in this field.

Table 3—Chemical Concentrations for Different Cases

Case Name Polymer Concentration,
lb/stb

PAG-1 0.06

PAG-2 0.10

PAG-3 0.14

PAG-4 0.18

Fig. 19 indicates that increasing polymer concentration would significantly reduce water injectivity, especially when
polymer concentration is larger than 0.10 lb/stb. Polymer injection would not change gas injectivity, however. Increasing
polymer concentration would reduce the oil rate in Phase I, which is dominated by gas injection. After gas breakthrough
(Phase II), increasing the concentration from 0.06 to 0.10 lb/stb would increase oil rate. However, increasing the
concentration from 0.10 to 0.14 lb/stb would reduce oil rate due to injectivity problems (Fig. 20). Similarly, as polymer
concentration increases from 0.06 to 0.10 lb/stb, oil recovery increases from 12 to 15%. But concentration higher than 0.10
lb/stb would not recover significantly more oil (Fig. 21). Considering the recovery and polymer consumption, a polymer
concentration of 0.10 lb/stb was used for the sensitivity study of slug patterns.

14 SPE 169734

16

Water Injection Volume, MM stb 12

8

4

0 PAG—2 PAG—3 PAG—4
PAG—1 0.10 lb/stb 0.14 lb/stb 0.18 lb/stb
0.06 lb/stb

Fig. 19—Water injectivity decreasing with polymer concentration

350
PAG-0.06 lb/stb

300 PAG-0.10 lb/stb
PAG-0.14 lb/stb

250 PAG-0.18lb/stb

Oil Rate, stb/day 200

150

100

50

0
0 1000 2000 3000 4000 5000 6000 7000 8000

Days
Fig. 20—Oil rate of different polymer concentrations

15

12

Recovery Factor, % 9

6

3

0

PWAG—1 PWAG—2 PWAG—3 PWAG—4

0.06 lb/stb 0.10 lb/stb 0.14 lb/stb 0.18 lb/stb

Fig. 21—Recovery factor of different polymer concentrations

SPE 169734 15

Sensitivity Analysis—Slug Pattern. To identify which slug pattern yields a better recovery, four different schemes were
conducted (Fig. 22). Pattern-1 always injects a polymer of 0.10 lb/stb with water. Pattern-2 adds a polymer of 0.10 lb/stb to
water after gas breakthrough. Pattern-3 injects a polymer of 0.10 lb/stb with water only at the beginning. Pattern-4 injects a
polymer of 0.10 lb/stb with water at the beginning and decreases polymer concentration to 0.05 lb/stb at the end of the WAG
process.

Pattern-1 High concentration of polymer
Pattern-2 Low concentration of polymer
Pattern-3 Water
Gas

Pattern-4

Fig. 22—Slug patterns of four different schemes

Slug patterns affect the recovery factor of PAG (Fig. 23). Pattern-2, which adds a polymer of 0.10 lb/stb to water after gas
breakthrough, has the lowest recovery. It indicates that earlier polymer injection is preferred. Recovery from Patterns-1, -3,
and -4 are similar, which suggests that both polymer injection in the beginning and lower polymer concentrations (Patterns-3
and -4) are good choices to reduce polymer consumption. Polymer utilization is calculated as

ܶ‫݊݋݅ݐ݌݉ݑݏ݊݋ܿ ݎ݁݉ݕ݈݋݌ ݈ܽݐ݋‬
ܲ‫ ݊݋݅ݐܽݖ݈݅݅ݐݑ ݎ݁݉ݕ݈݋‬ൌ ܶ‫ ܩܣܹܲ ݉݋ݎ݂ ݈݅݋ ݈ܽݐ݋‬െ ܶ‫ ܩܣܹ ݉݋ݎ݂ ݈݅݋ ݈ܽݐ݋‬

Polymer utilization ranges from 1.80 to 2.30 lb/stb for these four patterns in the PAG process (Fig. 24). Thus, the PAG
process has high polymer utilization as successful polymer flooding. Pattern-3 has the best polymer utilization. In this
pattern, inject polymer with a concentration of 0.10 lb/stb for 15 years in the PAG process and then chased with the WAG
process for 5 years. Oil recovery increased by this pattern is forecasted to be 14.3%, which is 7.0% higher than conventional
WAG.

15

14

Recovery Factor,% 13

12

11

10 Pattern—2 Pattern—3 Pattern—4
Pattern—1

Fig. 23—Recovery factor of different slug patterns

16 SPE 169734

2.80

Polymer Utilization, lb/stb 2.40

2.00

1.60

1.20 Pattern—2 Pattern—3 Pattern—4
Pattern—1

Fig. 24—Polymer utilization of different slug patterns

Discussion
Using the K-value in STARS (CMG) is the other method that could model the PAG process. Future work may include
comparison of PAG results between STARS and E100. Next, a coreflooding experiment would be carried out to verify the
PAG performance that is simulated by this study. Synthetic models of WAG in different reservoirs and fluid conditions
would be further discussed. In this study, fractures around injectors are not considered, which would lead to significant
injectivity reduction when polymer injection, further study should consider fractures around wellbore by adjusting skin
factor.

Conclusions
A new EOR method named PAG is proposed to improve the efficiency of the conventional WAG process in TR78 of the
North Burbank Unit. The following conclusions were made for this study:

1. Simulation results show that PAG postpone gas breakthrough and would significantly reduce gas/oil ratio and
improve water and gas sweep efficiency.

2. After optimizing polymer injection concentration and slug patterns, the following development strategy was
suggested for TR78: inject polymer with a concentration of 0.10 lb/stb for 15 years in the PAG process and then
chased with the WAG process for 5 years.

3. Oil recovery increased by PAG in TR78 is forecasted to be 14.3%, which is 7.0% higher than WAG.
4. Polymer utilization is about 1.80 lb/stb, which is economically feasible.

Nomenclature

CCE Constant composition expansion

CGI Continuous gas injection

EOR Enhanced oil recovery

EOS Equation of state

FAWAG Foam-assisted water alternating gas

HCPV Hydrocarbon pore volume

HPAM Hydrolyzed polyacrylamide

Kv Vertical permeability

Kh Horizontal permeability

MMP Minimum miscibility pressure

PVT Pressure-volume-temperature

PAG Polymer-alternating-gas

RRF Residual resistance factor

SWAG Simultaneous water and solvent injection

VDP Dykstra-Parsons permeability variation coefficient

WAG Water alternating gas

Acknowledgments
Financial support for this study was provided by Chaparral LLC. The authors wish to thank DeLon Flinchum and Matt Stover

SPE 169734 17

for their contribution of ideas and data. Thanks are also due to Randy Seright for his suggestions on preparing the manuscript.

References
Aleidan, A.A and Mamora, D.D. 2010. SWACO2 and WACO2 Efficiency Improvement in Carbonate Cores by Lowering

Water Salinity. Paper SPE 137548-MS presented at Canadian Unconventional Resources and International Petroleum
Conference, Calgary, Alberta, Canada, 19-21 October.
Ali H.A. Al-Ali and David S. Schechter. 2013. Application of Polymer Gels as Conformance Control Agents for Carbon
Dioxide EOR WAG Floods. Paper SPE 164096 presented at 2013 SPE International Symposium on Oilfield
Chemistry. The Woodlands, TX, USA, 8-10 April.
Bae, J.H., and Irani, C.A. 1993. A Laboratory Investigation of Viscosified CO2 Process. SPE Advanced Technology Series
1(1): 166-171. SPE-20467-PA.
Blaker, T., Aarra, M., Skauge, A., et al. 2002. Foam for Gas Mobility Control in the Snorre Field: The FAWAG Project. SPE
Res Eval & Eng 5(4): 317-323. SPE-78824-PA.
Bond, D.C. and Holbrook, O.C. 1958. Gas Drive Oil Recovery Process. US Patent No. 2,866,507.
Cabrera, F. and Manrique, E. 2010. Projected EOR Oil Production. Workshop EOR, 2010.
Caudle, B.H. and Dyes, A.B. 1958. Improving Miscible Displacement by Gas-Water Injection. Paper SPE 911-G presented at
32nd Annual Fall Meeting of Society of Petroleum Engineers, Dallas, TX, October 6-9.
Christensen, J.R., Stenby, E.H. and Skauge, A. 1998. Review of WAG Field Experience. Paper SPE 39883-MS presented at
International Petroleum Conference and Exhibition of Mexico, Villahermosa, Mexico, 3-5 March.
Christensen, J.R., Stenby, E.H. and Skauge, A. 2001. Review of WAG Field Experience. SPE Res Eval & Eng 4(2): 97-106.
SPE-71203-PA.
Enick, R.M., Olsen, D.K., Ammer, J. R., et al. 2012. Mobility and Conformance Control for CO2 EOR via Thickeners, Foams
and Gels—A Literature Review of 40 years of Research and Pilot Tests. Paper SPE 154122-MS presented at SPE
Improved Oil Recovery Symposium. Tulsa, Oklahoma. 14-18 April 2012.
Faisal, A., Bisdom, B., Zhumabek, B., et cl. 2009. Injectivity and Gravity Segregation in WAG and SWAG Enhanced Oil
Recovery. Paper SPE 124197-MS presented at SPE Annual Technical Conference and Exhibition, New Orleans,
Louisiana, 4-7 October.
Heller, J.P., Dandge, D.K., Card, R.J., et al. 1985. Direct Thickeners for Mobility Control of CO2 Floods. SPE Journal 25(5):
679-686. SPE-11789-PA.
Hild, G.P. and Wackowski R.K. 1999. Reservoir Polymer Gel Treatments To Improve Miscible CO2 Flood. SPE Res Eval &
Eng 2(2):196-204. SPE-56008-PA.
Karaoguz, O.K., Topguder, N.N.S., Lane, R.H., et al. 2007. Improved Sweep in Bati Raman Heavy-Oil CO2 Flood: Bullhead
Flowing Gel Treatments Plug Natural Fractures. SPE Res Eval & Eng 10(2):164-175. SPE-89400-PA.
Kulkarni, M.M. and Rao, D.N. 2005. Experimental investigation of miscible and immiscible Water-Alternating-Gas (WAG)
process performance. Journal of Petroleum Science and Engineering 48(1–2): 1–20.
Koottungal, L. 2012. 2012 Worldwide EOR Survey. Oil & Gas Journal Online, http://www.ogj.com/articles/print/vol-
110/issue-4/general-interest/special-report-eor-heavy-oil-survey/2012-worldwide-eor-survey.html
Joseph C. Trantham and Paul D. Moffitt. 1982. North Burbank Unit 1,440-Acre Polymer Flood Project Design. Paper SPE
10717-MS presented at the SPE Enhanced Oil Recovery Symposium, Tulsa, Oklahoma, 4-7 April.
Joseph C. Trantham. 1983. Prospects of Commercialization, Surfactant/Polymer Flooding, North Burbank Unit, Osage
County, OK. Journal of Petroleum Technology 35(5): 872-880. SPE-9816-PA.
Majidaie, S., Khanifar, A., Onur, M., et al. 2012. A Simulation Study of Chemically Enhanced Water Alternating Gas
(CWAG) Injection. Paper SPE 154152-MS presented at the SPE EOR Conference at Oil and Gas West Asia, Muscat,
16-18 April.
McClain, J.B., Betts, D.E., Canelas, D.A., et al. 1996. Characterization of Polymers and Amphiphiles in Supercritical CO2
using Small Angle Neutron Scattering and Viscometry. Proceedings of the 1996 Spring Meeting of the ACS,
Division of Polymeric Materials, Vol. 74, 234-235. New Orleans, LA.
Moffitt, P.D., Zornes, D.R., Moradi-Araghi, Ahmad, McGovern, J.M. 1993. Application of Fresh water and Brine Polymer
Flooding in the North Burbank Unit, Osage County, Oklahoma. SPE Journal 8(02): 128-134. SPE-20466-PA.
Robert D. Sydansk. 2007. Polymers, Gels, Foams, and Resins. In Petroleum Engineering Handbook Vol. 5, ed. Edward D.
Holstein, Chap. 13, 1219-1224, Richardson: Society of Petroleum Engineers.
Sohrabi, M., Danesh, A., and Tehrani, D.H. 2005. Oil Recovery by Near-Miscible SWAG Injection. Paper SPE 94073-MS
presented at the SPE EUROPEC/EAGE Annual Conference, Madrid, Spain, 13-16 June.
Stone H.L., 2004. A Simultaneous Water And Gas Flood Design With Extraordinary Vertical Gas Sweep. Paper SPE 91724-
MS presented at SPE International Petroleum Conference in Mexico, Puebla Pue., Mexico, 7-9 November.
Topguder, N.N.S. 2010. A Review on Utilization of Crosslinked Polymer Gels for Improving Heavy Oil Recovery in Turkey.
Paper SPE 131267 presented at SPE EUROPEC/EAGE Annual Conference and Exhibition, Barcelona, Spain, 14-17
June.
Tracy G.W. and Dauben D.L., 1982. An Evaluation of the North Burbank Unit Tertiary Recovery Pilot Test. Report No.
DOE/BC/10033-2. Bartlesville, Oklahoma.

18 SPE 169734

Terry, R.E., Zaid, A., Angelos, C., et al. 1987. Polymerization in Supercritical CO2 to Improve CO2/Oil Mobility Ratios.
Paper SPE 16270-MS presented at the SPE International Symposium on Oilfield Chemistry, San Antonio, TX, 4-6
February.

Trantham, J.C., Threlkeld, C.B., and Patterson Jr., H.L. 1980. Reservoir Description for a Surfactant/ Polymer Pilot in a
Fractured, Oil-wet Reservoir-North Burbank Unit Tract97. Journal of Petroleum Technology 32(9): 1647-1656. SPE-
8432-PA.

Woods, P., Schramko, K., Turner, D. et al. 1986. In-Situ Polymerization Controls CO2/Water Channeling at Lick Creek.
Paper SPE 14958-MS presented at the SPE Enhanced Oil Recovery Symposium, Tulsa, Oklahoma, 20-23 April.

Zhang, Y., Huang, S.S., and Luo, P. 2010. Coupling Immiscible CO2 Technology and Polymer Injection to Maximize EOR
Performance for Heavy Oils. Journal of Petroleum Technology 49(5):252-33. SPE-137048-PA.

Zornes, D.R., Cornelius, A.J., and Long, H.Q. 1986. An Overview and Evaluation of the North Burbank Unit Block A
Polymer Flood Project, Osage County, Oklahoma. Paper SPE 14113-MS presented at International Meeting on
Petroleum Engineering, Beijing, China, 17-20 March.