applied

sciences
Article
Compositional Modeling of Impure CO2 Injection for
Enhanced Oil Recovery and CO2 Storage
Hye-Seung Lee 1 , Jinhyung Cho 2 , Young-Woo Lee 1 and Kun-Sang Lee 1, *

1 Department of Earth Resources and Environmental Engineering, Hanyang University, 222 Wangsimni-ro,
Seongdong-gu, Seoul 04763, Korea; seung7185@hanyang.ac.kr (H.-S.L.);
youngwoolee@hanyang.ac.kr (Y.-W.L.)
2 Center for Climate, Environment Change Prediction Research, Ewha Womans University, 52 Ewhayeodae-gil,
Seodaemun-gu, Seoul 03760, Korea; jh.cho@ewha.ac.kr
* Correspondence: kunslee@hanyang.ac.kr; Tel.: +82-222-202-240

Abstract: Injecting CO2 , a greenhouse gas, into the reservoir could be beneficial economically,
by extracting remaining oil, and environmentally, by storing CO2 in the reservoir. CO2 captured
from various sources always contains various impurities that affect the gas–oil system in the reservoir,
changing oil productivity and CO2 geological storage performance. Therefore, it is necessary to
examine the effect of impurities on both enhanced oil recovery (EOR) and carbon capture and
storage (CCS) performance. For Canada Weyburn W3 fluid, a 2D compositional simulation of water-
alternating-gas (WAG) injection was conducted to analyze the effect of impure CO2 on EOR and CCS
performance. Most components in the CO2 stream such as CH4 , H2 , N2 , O2 , and Ar can unfavorably
increase the MMP between the oil and gas mixture, while H2 S decreased the MMP. MMP changed
according to the type and concentration of impurity in the CO2 stream. Impurities in the CO2 stream





 also decreased both sweep efficiency and displacement efficiency, increased the IFT between gas and
Citation: Lee, H.-S.; Cho, J.; Lee,
reservoir fluid, and hindered oil density reduction. The viscous gravity number increased by 59.6%,
Y.-W.; Lee, K.-S. Compositional resulting in a decrease in vertical sweep efficiency. In the case of carbon storage, impurities decreased
Modeling of Impure CO2 Injection for the performance of residual trapping by 4.1% and solubility trapping by 5.6% compared with pure
Enhanced Oil Recovery and CO2 CO2 WAG. As a result, impurities in CO2 reduced oil recovery by 9.2% and total CCS performance
Storage. Appl. Sci. 2021, 11, 7907. by 4.3%.
https://doi.org/10.3390/
app11177907 Keywords: CCUS; CO2 -EOR; carbon capture and storage; CCS-EOR; impure CO2 ; water alternating
gas (WAG)
Academic Editor: Fulvia Chiampo

Received: 3 August 2021

Accepted: 25 August 2021
1. Introduction
Published: 27 August 2021
The CO2 -enhanced oil recovery (EOR) method has been widely applied to numerous
Publisher’s Note: MDPI stays neutral
field projects around the world [1,2]. CO2 -EOR was mostly aimed at recovering residual
with regard to jurisdictional claims in
oil initially, but it is converted to not only increase recovery but also achieve reduction
published maps and institutional affil- greenhouse gases by injecting CO2 or flue gas. Recently, the application has been extended
iations. to unconventional hydrocarbons such as gas hydrates [3]. Existing CO2 -EOR aimed to pro-
duce maximum oil and inject minimum CO2 , but carbon capture, utilization, and storage
(CCUS) consider the increase in oil recovery and greenhouse gas storage [4–6]. With atmo-
spheric CO2 concentrations continuing to rise worldwide, CCUS has the potential to reach
Copyright: © 2021 by the authors.
net-zero [7]. As a key technology used to reduce carbon dioxide emissions from fossil-
Licensee MDPI, Basel, Switzerland.
fueled power plants and other industrial activities, CCUS helps provide energy security
This article is an open access article
by lowering emissions while achieving economics. CCUS is considered as an important
distributed under the terms and technology for moving toward a global low-carbon economy. There are many sources of
conditions of the Creative Commons CO2 such as natural gas reservoirs containing CO2 , industrial or anthropogenic sources,
Attribution (CC BY) license (https:// and natural CO2 reservoirs. Depending on the capture approaches (post-combustion,
creativecommons.org/licenses/by/ oxy-fuel, or pre-combustion), various impurities such as N2 , O2 , H2 S, CH4 , Ar, and water
4.0/). are contained in the CO2 stream [8–12]. The CO2 concentration used for EOR commonly

Appl. Sci. 2021, 11, 7907. https://doi.org/10.3390/app11177907 https://www.mdpi.com/journal/applsci

Appl. Sci. 2021, 11, 7907 2 of 14

requires extremely high purity (around 90–98%), especially for miscible displacement [13].
A purification process is generally required to obtain the appropriate CO2 concentration
from the source gas [14]. The purity of captured CO2 largely depends on the type of capture
technology. The type and amount of impurities in the CO2 stream affect the fluid proper-
ties and thermodynamics of CO2 differently, exhibiting beneficial or negative effects on
CO2 -EOR: fluid solubility, phase behavior of the oil and gas system, and miscibility [15,16].
A small number of impurities significantly changes the physical properties and phase
behavior of the CO2 mixtures. The solubility in water depends on the type and amount of
impurities in the CO2 stream [17–19]. The solubility of CO2 in water increases with increas-
ing pressure and decreasing temperature. Most impurities have low solubility in water
compared with CO2 . Impurities reduce the partial pressure of CO2 and therefore reduce
solubility in water. CO2 impurities can affect all stages of CO2 -EOR, from CO2 capture
to flooding. Since the level of contaminant removal is determined by process technology
and cost, the actual level of CO2 purity required depends not only on the requirements of
transportation and storage but also on the economics of the process. Separating impurities
greatly increases the capture cost. Higher CO2 purity is associated with higher costs of
capture. The cost of a low purity CO2 stream (82.9 wt.% CO2 ) is 42% lower than high purity
CO2 (99.98 wt.% CO2 ) in a capture plant [14]. It can be economical to inject the impurities
contained in the CO2 together. Several studies have investigated the effects of contaminants
in the CO2 stream on minimum miscibility pressure (MMP) to determine how impurities
affect MMP compared with pure CO2 . In general, impurities including CH4 , SO2 , O2 , and
N2 have been found to increase MMP, while H2 S, C2 H6 , or intermediate hydrocarbons
such as C3 or C4 reduce MMP, positively affecting miscible displacement [8,16,20–22].
Based on the effect of individual impurities on the phase behavior and MMP of the CO2 –
hydrocarbon system, the effect of impurity mixtures on MMP and EOR performance has
also been examined [16]. In addition, several studies have investigated the change of
EOR performance and storage capacity when impurities and CO2 are co-injected in the
aquifer [10,11]. The presence of contaminants will affect the trapping mechanism since it
changes the wettability, interfacial tension, and buoyancy of CO2 [23,24]. However, studies
on CO2 -EOR performance coupled with CO2 storage considering the composition of the
actual impure CO2 stream from carbon sources are insufficient. Previous studies have
focused on laboratory investigation, and discussions on reservoir-scale studies considering
the composition of actual CO2 are insufficient. Therefore, it is important to investigate
the impact of impurities on CO2 properties combined with the performance of EOR and
carbon storage. In this study, compositional simulation was used to investigate the impact
of impurities on the phase behavior of CO2 , CO2 displacement mechanism, and carbon
storage when injected without removing impurities from the oxy-fuel combustion stream.

2. Materials and Methods

2.1. Fluid Modeling
The Weyburn reservoir, located in southeast Saskatchewan, became a target for CO2 -
EOR and carbon sequestration after water flooding production [25]. Since the reservoir has
been conducting CO2 monitoring and operated as a storage project, Weyburn W3 fluid
data [25] is used to examine the interactions of reservoir fluid and CO2 for the purpose
of CCS-EOR. The composition of the reservoir oil and fluid properties used for the EOS
calculations are summarized in Table 1 [26,27].
Regression was used to match the W3 fluid experimental measurements and fluid
model data by tuning the EOS parameters. Table 2 shows PVT properties according to the
dissolved gas mole fraction [25]. The properties are saturation pressure, gas oil ratio (GOR),
gas solubility, formation volume factor (FVF), and swelling factor (SF). Fluid characteriza-
tion, lumping of components, and matching with laboratory data are performed through
regression using a modification of the adaptive least-squares algorithm of Dennis et al. [28].
The Peng–Robinson equation of state (PR-EOS) was used to construct the fluid model.
Table 3 presents a comparison of experimental and simulation PVT properties of the fluid.
Appl. Sci. 2021, 11, 7907 3 of 14

Through the regression method, results from the fluid model were properly matched to the
experimental measurements and verified reliability.

Table 1. Oil composition and properties for equation of state (EOS) calculations.

Critical Critical Molecular

Acentric
Component Composition Pressure Temperature Weight Parachor
Factor
(kPa) (K) (g/gmol)
N2 0.0207 3394.4 126.2 28.0 0.04 41
CO2 0.0074 7376.5 304.2 44.0 0.23 78
H2 S 0.0012 8936.9 373.2 34.1 0.1 80
Methane, C1 0.0749 4600.2 190.6 16.0 0.01 77
Ethane, C2 0.0422 4883.9 305.4 30.1 0.1 108
Propane, C3 0.0785 4245.5 369.8 44.1 0.15 150
Butane, C4 0.0655 3722.7 416.5 58.1 0.19 186
Pentane, C5 0.0459 3379.4 464.9 72.1 0.24 228
C6–9 0.2156 3019.6 556.3 102.5 0.33 297
C10–17 0.2202 2017.5 692.2 184.0 0.58 508
C18–27 0.1027 1327.0 808.4 306.2 0.89 771
C28+ 0.1252 1155.1 915.5 566 1.1 1112

Table 2. PVT properties of the reservoir fluid and CO2 mixtures at 140 ◦ F.

Saturation
Dissolved Gas GOR Gas Solubility
Pressure FVF (m3 /m3 ) SF (m3 /m3 )
Mole Fraction (sm3 /m3 ) (sm3 /m3 )
(MPa)
0.0058 2.9 19 0 1.087 1.074
0.158 4.5 42 23 1.143 1.13
0.412 8 113 94 1.308 1.292
0.439 8.4 125 106 1.336 1.32
0.521 9.9 158 139 1.409 1.392
0.595 11.4 221 202 1.546 1.527
0.641 12.6 263 244 1.634 1.614
0.826 19.7 875 856 2.694 2.668

Table 3. Comparison of the data for Weyburn fluid and the fluid model.

Parameters W3 Fluid Model Difference (%)

Oil density at saturation pressure(kg/m3 ) 806.4 805.7 0.09
Saturation pressure (kPa) 4920 4916 0.08
Viscosity at saturation pressure (mPa·s) 1.76 1.76 0.0
Formation volume factor (m3 /m3 ) 1.12 1.108 1.07
API (◦ ) 31 34.48 −11.23
MMP with CO2 (kPa) 14,196 13,900 2.09

2.2. Minimum Miscibility Pressure (MMP)

CO2 is widely used in EOR through injection into an oil reservoir to increase oil
recovery. CO2 is used as a solvent, reducing oil viscosity and expanding the volume of oil
in the reservoir. These phenomena can be expected to be more prominent if the oil and gas
are more miscible. If the injected gas and fluid are not completely miscible, some of the gas
does not dissolve into the oil and moves faster than the fluid, forming a separate phase.
This causes an early breakthrough at the production well and lowers the productivity of
the oil reservoir. To avoid this circumstance, determining whether the gas and fluid are
under miscible conditions at the reservoir pressure is of significant interest. The lowest
pressure to achieve miscibility by a multiple contact process under the given pressure
and temperature is called the minimum miscibility pressure (MMP). Below the MMP,
immiscible displacement occurs and consequently the oil recovery is reduced.
Appl. Sci. 2021, 11, 7907 4 of 14

Methods for measuring MMP include the slim-tube test, the rising-bubble method,
and the vanishing interfacial tension (VIT) test [29–33]. The slim-tube test has been widely
used to measure MMP and is the most used method [34]. Since these methods are time-
consuming, expensive, and sometimes less accurate, it is better to use the simulation
method to determine the MMP. A slim-tube simulation model is conducted to measure the
MMP. The slim tube is saturated with oil, and gas is subsequently injected to displace the
oil in the slim tube. The amount of gas and oil is determined, and the recovery is defined.
When the recovery is at least 90%, it is said that oil and gas are miscible. This process is
repeated at different pressures, and the oil recovery is recorded at certain pressure. MMP
from the slim-tube simulation is estimated as the lowest pressure with a recovery of 90% of
the oil–gas system [35].

2.3. Evaluation of CO2 Streams

CO2 can be captured from various sources through various technologies, and impuri-
ties such as N2 , O2 , CH4 , H2 S, Ar, and water may exist in the CO2 stream. The amounts of
impurities contained in CO2 depends on the degree of fuel oxidation and characteristics of
the fuel. Among the three general CO2 capture methods, pre-combustion, post-combustion,
and oxy-fuel combustion, the oxy-fuel combustion process generates the highest impurities
content in the CO2 stream. Due to the use of lower purity oxygen, which is relatively
cost-effective, significant amounts of air-derived impurities such as N2 , O2 , and Ar are pro-
duced. Table 4 presents the main impurities from different CO2 capture technologies [13].
In this study, the composition of the gas obtained from the oxy-fuel scenario with the lowest
CO2 purity level was used to predict how many contaminants will affect performance and
to compare the CO2 -EOR and CO2 storage efficiency of impure and pure CO2 .

Table 4. Summary of CO2 impurities from different CO2 capture technologies.

Oxy-Fuel Combustion
Pre- Post-
Raw/ Double Combustion Combustion
Distillation
Dehumidified Flashing
CO2 (% v/v) 74.8–85.0 95.84–96.7 99.3–99.4 95–99 99.6–99.8
O2 (% v/v) 3.21–6.0 1.05–1.2 0.01–0.4 0 0.015–0.0035
N2 (% v/v) 5.80–16.6 1.6–2.03 0.01–0.2 0.0195–1 0.045–0.29
Ar (% v/v) 2.3–4.47 0.4–0.61 0.01–0.1 0.0001–0.15 0.0011–0.021
NOx (ppmv) 100–709 0–150 33–100 400 20–38.8
SO2 (ppmv) 50–800 0–4500 37–50 25 0–67.1
SO3 (ppmv) 20 20
H2 O (ppmv) 100–1000 0 0–100 0.1–600 100–640
CO (ppmv) 50 50 0–2000 1.2–10
H2 S/COS (ppmv) 0.2–34,000
H2 (ppmv) 20–30,000
CH4 (ppmv) 0–112

2.4. Solubility
A Solubility of gas in connate water is considered by including Henry’s law as fol-
lows [36].
f iw = yiw Hi (1)
where f iw is the fugacity of the component i, yiw is the composition of the component i in
the aqueous phase and Hi is the Henry’s law constant. The Henry’s law constant can be
expressed by [37]:
Vi ( p − p∗ )
lnHi = lnHi ∗ + (2)
RT
where Hi ∗ is the Henry’s law constant for component i at reference pressure, p∗ is the
reference pressure, Vi is the partial molar volume of component i at infinite dilution, R is
Appl. Sci. 2021, 11, 7907 5 of 14

the universal gas constant, and T is the temperature. The values are obtained from Li and
Nghiem’s method [36].

2.5. Reservoir Modeling

A compositional reservoir simulator, CMG-GEM, was used to simulate a
two-dimensional (2D) vertical reservoir discretized into 52 × 1 × 20 grid blocks of volume
3.048 × 3.048 × 1.524 m3 . The water-alternation-gas (WAG) injection scenario was used to
investigate the miscibility and sweep efficiency with gravity override. In order to compare
differences in the performance of EOR and CCS depending on the impurity content, the gas
composition in Table 5 was used. The solubility of each component in water is considered
by including Henry’s law to represent gas solubility. Although impure gas was composed
of 8 impurities, minor impurities at ppm levels such as H2 O, NOx , SOx , etc. were neglected.
The injection process was conducted for 12 years, with water flooding in the first 2 years
and WAG cycles of 10 years (Figure 1). During the 5 WAG cycles, water and gas were in-
jected sequentially. Two injection cases were analyzed: 100% CO2 + 0% impurities (Case 1)
and 85% CO2 + 15% impurities (Case 2) (Table 6). Each gas injection rate was constant at
566 ft3 /day at surface condition. The initial reservoir condition is presented in Table 7.
The producing bottom hole pressure was set to 2000 psi close to the MMP of 100% CO2 ,
leading to miscible flooding in the case of pure CO2 otherwise immiscible flooding in the
case of impure CO2 .

Table 5. Composition of CO2 impurities from the low purity Oxy-fuel stream.

Components Low Purity Level Oxy-Fuel Stream

CO2 (vol%) 85.0
O2 (vol%) 4.70
N2 (vol%) 5.80
Ar (vol%) 4.47
H2 O (ppmv) 100
NOx (ppmv) 100
SO2 (ppmv) 50
SO3 (ppmv) 20
CO (ppmv) 50

Figure 1. Water alternating gas (WAG) injection period.

Table 6. Classification of cases according to impurities content.

Case Injection Gas Contents

1 100% CO2 + 0% Impurities
2 75% CO2 + 15% Impurities
Appl. Sci. 2021, 11, 7907 6 of 14

Table 7. Reservoir initial and operating conditions.

Parameters Values
Depth (ft) 4050
Initial reservoir pressure (psi) 4000
Reservoir temperature (◦ F) 145
Permeability in I, J, K-direction (md) 50, 50, 5
Porosity (%) 0.3
Initial oil saturation (%) 0.7
Initial water saturation (%) 0.3
Producing bottom hole pressure (psi) 2000

3. Results
3.1. Physical Properties of Impure Gas
MMP, Density, and Viscosity
A change in the physical properties of the gas will impact the displacement and sweep
efficiencies. The slim-tube simulation method was conducted to determine CO2 MMP
for the reservoir fluid, which was matched against the experimental data. It has been
recognized that MMP between the oil and injected solvents depends on the purity of CO2 .
In general, the component of H2 S or intermediate hydrocarbons such as C3 , C4 reduce
the MMP, whereas the presence of N2 and CH4 in the CO2 stream increase the MMP [38].
The MMP between CO2 mixtures containing 10% of each representative impurity in the
CO2 such as CH4 , H2 , N2 , and O2 , and W3 fluid was calculated. The results from simulated
MMPs using the slim-tube method with different mole fractions of CO2 and impurities
are shown in Table 8. Impurities of CH4 , H2 , N2 , and O2 effected an increase in the MMP,
while H2 S decreased the MMP. Therefore, CO2 containing either components that increase
(CH4 , H2 , N2 , O2 ) or decrease (H2 S) MMP changes the miscibility.

Table 8. Minimum miscibility pressures (MMPs) calculation using the slim-tube method with various
CO2 mixtures.

Components MMP (kPa)

CO2 100% 13,900
CO2 90% + CH4 10% 18,616
CO2 90% + H2 10% 16,079
CO2 90% + N2 10% 18,933
CO2 90% + O2 10% 14,555
CO2 90% + Ar 10% 18,650
CO2 90% + H2 S 10% 12,714

Table 9 presents the impact of impure gas in the CO2 on the gas properties. The MMP
of pure CO2 for this reservoir fluid was determined to be 13,900 kPa. Presence of the impure
gas content increased the MMP to 19,927 kPa, while decreasing the density and viscosity,
and had a negative impact on the miscibility of the gas and oil. The density of the injection
gas impacted both injection rate and pressure. Density changes due to the addition of
impurities reduced the gas compressibility and caused differences in bottomhole pressure
at given injection wellhead pressure.

Table 9. Minimum miscibility pressures (MMPs) calculated using the slim-tube method with pure
and impure CO2 mixtures.

Case 1 Case 2
Minimum Miscibility Pressure (kPa) 13,900 19,927
Density (kg/m3 ) 689 552
Viscosity (mPa·s) 0.0702 0.0493
Appl. Sci. 2021, 11, 7907 7 of 14

3.2. Effects of Impurities in CO2 Streams on EOR Efficiency

3.2.1. Vertical Sweep Efficiency
The decrease in vertical sweep efficiency due to impurities in CO2 is explained by the
viscosity of the oil. Figure 2 shows the distribution of oil viscosity when 0.22 PV of gas is
injected, indicating the decrease in vertical sweep efficiency due to impurities. In Case 2,
impurities in the CO2 stream cause the injected gas to move forward faster since it hinders
the role of CO2 in lowering the viscosity of oil. The blue swept area on the front represents
the area where the injected fluid is in maximum contact with the oil, indicating the gas
front. The minimum viscosity in Case 1 is 0.48 mPa·s and in Case 2 is 0.59 mPa·s. Therefore,
the addition of 15% impurities generated a 22.9% decrease in the oil viscosity, indicating
less gas acts as a solvent to decrease the oil viscosity after multiple contacts due to the
impurities in CO2 .

Figure 2. Oil viscosity distribution when 0.22 PV of gas is injected: (a) Case 1, (b) Case 2.

In horizontal gas flow through the cross-sectional 2D model, the vertical sweep
efficiency was significantly reduced due to gravity segregation. The steeper the slope of
the gas front of the sweep area, the more the injected gas contacts with the reservoir fluid.
Since the difference in density and gravity between the injected fluid and reservoir fluid
causes an override, the injected gas moves to the upper layer, reducing the vertical sweep
efficiency. The change in vertical sweep efficiency due to the addition of impurities can be
quantified by the viscous gravitational number (Ngv ), which includes the terms of flooding
velocity, density, and viscosity of fluid. The viscous gravitational can be expressed by [39]:

time for horizontal flow k v krs ∆ρgcosαAL

Ngv = = (3)
time for vertical flow qs µs h

where k v is the vertical permeability, krs is the relative permeability, ∆ρ is the density
difference between injected and reservoir fluids, g is the gravitational acceleration, A is
the cross-sectional area, L is the length of the reservoir, qs is the flow rate of the solvent,
µs is the solvent viscosity, and h is the height of the reservoir. The range that determines
whether gravity or viscous forces dominate is as follows [39]:

Ngv < 0.1 : Viscous forces dominate;

0.1 < Ngv < 10 : Intermediate;
Ngv > 10 : Gravity dominates.

The decrease in vertical sweep efficiency was quantified from the viscous gravity
number, as shown in Table 10. In Case 2, the viscous gravity number increased by as much
as 59.6% by adding impurities into the CO2 . Since a larger viscous gravity number prevents
the lower layer of the reservoir from being sufficiently swept, impurities in the injecting
CO2 stream lower the vertical sweep efficiency.
Appl. Sci. 2021, 11, 7907 8 of 14

Table 10. Viscous gravity numbers of Cases 1 and 2.

Model Viscous Gravity Number, Ngv

Case 1 2.1
Case 2 5.2

3.2.2. Displacement Efficiency

Above the MMP, the injected gas is completely mixed with the oil and the interfacial
tension (IFT) between the injected gas and fluid would be negligible, affecting the relative
permeability curves. The density and viscosity also decrease, leading to miscible displace-
ment. The effect of adding impurities on the IFT between gas and oil in the grid block
coordinates (12, 1 10) was investigated. The lowest IFT of Case 1 was 0.10 mN/m between
the CO2 and reservoir fluid, whereas for Case 2 it was 0.65 mN/m, a 547% higher value.
The addition of impurities increased the IFT of Case 2 as shown in Figure 3, indicating
that higher pressure would be required to achieve the miscible condition. In the same grid
block of (12, 1, 10), the effect of reducing the density of oil was hindered as impurities were
added to the CO2 stream at the beginning of the injection period. After multiple contacts,
Case 1 gave the lowest oil density value of 785 kg/m3 (2.5% reduction in initial oil density),
while Case 2 gave a value of 799 kg/m3 (0.83% reduction in initial oil density). However, at
some point, the oil density of Case 1 was greater than that of Case 2 since the intermediate
and heavy components of the oil were displaced (Figure 4).

Figure 3. Interfacial tension between CO2 -impure gas and the oil during CO2 and impure CO2 WAG
injection.
The cumulative gas injection under reservoir conditions obtained from CO2 and
impure CO2 WAG is indicated in Figure 5. Impurities in the CO2 stream have low critical
temperatures and have relatively lower compressibility under the reservoir conditions [13].
CO2 containing impurities would cause an increase in molar volume, i.e., impure CO2
would occupy more volume than pure CO2 under the same conditions. Although the
injection rate was the same at the surface condition, the total amount of injected gas of
Case 2 was greater than Case 1 under reservoir conditions. The gas moves faster in the
upper layer in Case 2 than in Case 1, which causes an early gas breakthrough, as shown in
Figure 5. This phenomenon occurs because the gas does not lower the viscosity of the oil
compared with pure CO2 , and some of the injected gas forms a gas channel, causing early
gas breakthrough in the production well, as shown in Figure 6.
Appl. Sci. 2021, 11, 7907 9 of 14

Figure 4. Oil density during CO2 and impure CO2 WAG injection.

Figure 5. Cumulative gas injection during the CO2 and impure CO2 WAG injection under reservoir
conditions.
The average reservoir pressure is presented in Figure 7. As the impurities are less
compressible than CO2 , they occupy a larger volume in the pore, increasing the average
reservoir pressure. However, due to significant reduction in the density of the injected
stream caused by impurities, a higher flow pressure is required to achieve the same bottom
hole pressure in the pure CO2 case. Therefore, co-injection of impurities and CO2 causes a
relatively higher oil production compared with CO2 injection at the early stage of WAG
cycles. This trend is reversed due to the decrease in vertical sweep and displacing efficiency
from the impurities, and the oil recovery is higher for Case 1 as shown in Figure 8. As a
result, impurities in CO2 reduce the oil recovery by 9.2%.
Appl. Sci. 2021, 11, 7907 10 of 14

Figure 6. Cumulative gas production during the CO2 and impure CO2 WAG injection under standard
conditions.

Figure 7. Average reservoir pressure during the CO2 and impure CO2 WAG injection.

3.3. Effects of Impurities in CO2 Streams on Carbon Storage Efficiency

Figures 9 and 10 indicate the carbon sequestration by residual trapping and solubility
trapping. Considering that impurities have less viscosity than CO2 , the presence of impuri-
ties increases the mobility and thus increases the horizontal movement of the injected gas.
Since the density of impurities is lower than that of CO2 , the injected gas with impurities
migrates upward easily. A significant volume of the injected gas moves and accumulates
upward, which negatively affects residual trapping. Residually trapped CO2 was reduced
by 18.6% in Case 2 compared with Case 1. The solubility trapped CO2 also decreased by
19.8% with the addition of impurities into CO2 . Since the absolute amount of injected CO2
into the reservoir was less in Case 2 than in Case 1 due to the addition of impurities, the
ratio of sequestered CO2 to injected CO2 was quantified. Figure 11 compares the ratio of
Appl. Sci. 2021, 11, 7907 11 of 14

CO2 stored by residual trapping and solubility trapping. In Case 2, the performance of
residual trapping was reduced by 4.1% and solubility trapping by 5.6% compared with
Case 1. As a result, impurities in CO2 reduced the total CCS performance by 4.3%.

Figure 8. Oil recovery factor during the CO2 and impure CO2 WAG injection.

Figure 9. Residual CO2 trapped by hysteresis during CO2 and impure CO2 WAG injection.
Appl. Sci. 2021, 11, 7907 12 of 14

Figure 10. Solubility trapped CO2 during CO2 and impure CO2 WAG injection.

Figure 11. The comparison of trapping index: (a) Case 1, (b) Case 2.

4. Discussion
Impurities captured together with CO2 are subjected to a purification process to
generate high purity CO2 , which significantly increases the capture cost. Therefore, it is
cost-effective to inject the CO2 and impurities together. However, impure CO2 adversely
affects CO2 -EOR performance, resulting in reduced oil recovery. The change in oil recovery
depending on the purity level would be controlled and considered. Though not considered
in this study, it is important to develop the economic model to estimate the appropriate
purity level of the CO2 stream by using economic factors such as capital expenditures
(CAPEX), operating expenditure (OPEX), tax credits, oil price, and CO2 price depending on
the purity level, sources and technology. For field applications, since the optimal operating
conditions to increase the performance of CCS-EOR depend on the characteristics of the
reservoir, it is necessary to develop a design for each site, such as different injection
scenarios, based on economic analysis for field application.
Appl. Sci. 2021, 11, 7907 13 of 14

5. Conclusions
This study aimed to investigate the effects of impurities in the CO2 on CO2 -EOR and
CCS performances. The degree of MMP change depended on the type of impurity since
most non-condensable impurities increased the MMP, negatively affecting the miscible state
except for H2 S. A compositional simulation was used to analyze the effect of impurities on
both CO2 -EOR efficiency and carbon storage. A 2D vertical model was used to investigate
vertical sweep efficiency considering gravity override during the WAG injection period.
Two cases were analyzed: pure CO2 (Case 1) and impure CO2 , with a composition of 15%
impurities captured from oxy-fuel combustion (Case 2). The vertical sweep efficiency and
displacement efficiency were affected by the impurities and eventually affected the oil
recovery and CO2 -EOR performance. Impurities in the CO2 stream increased the IFT and
hindered the reduction in oil density. The viscous gravity number increased by 59.6%,
confirming the decrease in vertical sweep efficiency due to the addition of impurities.
The IFT on a specific grid block increased by 547% in Case 2. The effect of reducing oil
density as a solvent was also inhibited from 2.5% to 0.83%. As a result, impurities in CO2
reduced oil recovery by 9.2%. Due to CO2 impurities, residual trapping performance was
reduced by 4.1%, solubility trapping decreased by 5.6%, and the total CCS performance
was reduced by 4.3%.

Author Contributions: Conceptualization, H.-S.L. and K.-S.L.; methodology, H.-S.L., J.C., Y.-W.L.
and K.-S.L.; software, H.-S.L. and J.C.; validation, H.-S.L. and K.-S.L.; formal analysis, H.-S.L.;
investigation, H.-S.L., J.C. and Y.-W.L.; resources, H.-S.L. and K.-S.L.; data curation, H.-S.L.; writing—
original draft preparation, H.-S.L.; writing—review and editing, H.-S.L. and K.-S.L.; visualization,
H.-S.L., J.C. and Y.-W.L.; supervision, K.-S.L.; project administration, K.-S.L.; funding acquisition,
K.-S.L. All authors have read and agreed to the published version of the manuscript.
Funding: This work was supported by the National Research Foundation of Korea (NRF) grant
funded by the Korea government (MSIT) (No. 2020R1F1A1070406). Jinhyung Cho was supported by
a National Research Foundation of Korea (NRF) grant (No. 2020R1I1A1A01067015).
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: The data presented in this study are available on request from the
corresponding author.
Acknowledgments: The authors are grateful to the Computer Modelling Group Ltd. (CMG), Calgary,
Canada for technical support.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Gozalpour, F.; Ren, S.R.; Tohidi, B. CO2 EOR and storage in oil reservoir. Oil Gas Sci. Technol. 2005, 60, 537–546. [CrossRef]
2. Holm, L. Carbon dioxide solvent flooding for increased oil recovery. Trans. AIME 1959, 216, 225–231. [CrossRef]
3. Terzariol, M.; Goldsztein, G.; Santamarina, J.C. Maximum recoverable gas from hydrate bearing sediments by depressurization.
Energy 2017, 141, 1622–1628. [CrossRef]
4. Kuuskraa, V.A.; Godec, M.L.; Dipietro, P. CO2 utilization from “next generation” CO2 enhanced oil recovery technology.
Energy Procedia 2013, 37, 6854–6866. [CrossRef]
5. IEAGHG. CO2 Storage in Depleted Oilfields: Global Application Criteria for Carbon Dioxide Enhanced Oil Recovery; IEAGHG Publica-
tions: Cheltenham, UK, 2009.
6. Khelifa, T.; Maini, B.B. Evaluation of CO2 based vapex process for the recovery of bitumen from tar sand reservoirs. In Proceedings
of the SPE International Improved Oil Recovery Conference in Asia Pacific, Kuala Lumpur, Malaysia, 20–21 October 2003.
7. Hassanpouryouzband, A.; Joonaki, E.; Edlmann, K.; Haszeldine, R.S. Offshore geological storage of hydrogen: Is this our best
option to achieve net-zero? ACS Energy Lett. 2021, 6, 2181–2186. [CrossRef]
8. Choubineh, A.; Helalizadeh, A.; Wood, D.A. The impacts of gas impurities on the minimum miscibility pressure of injected
CO2 -rich gas–crude oil systems and enhanced oil recovery potential. Pet. Sci. 2019, 16, 117–126. [CrossRef]
9. Chapoy, A.; Burgass, R.; Tohidi, B.; Austell, J.M.; Eickhoff, C. Effect of common impurities on the phase behavior of carbon-
dioxide-rich systems: Minimizing the risk of hydrate formation and two-phase flow. SPE J. 2011, 16, 921–930. [CrossRef]
Appl. Sci. 2021, 11, 7907 14 of 14

10. Chapoy, A.; Nazeri, M.; Kapateh, M.; Burgass, R.; Coquelet, C.; Tohidi, B. Effect of impurities on thermophysical properties and
phase behaviour of a CO2 -rich system in CCS. Int. J. Greenh. Gas Control 2013, 19, 92–100. [CrossRef]
11. Jin, L.; Pekot, L.J.; Hawthorne, S.B.; Salako, O.; Peterson, K.J.; Bosshart, N.W.; Jiang, T.; Hamling, J.A.; Gorecki, C.D. Evaluation
of recycle gas injection on CO2 enhanced oil recovery and associated storage performance. Int. J. Greenh. Gas Control 2018, 75,
151–161. [CrossRef]
12. Ziabakhsh-Ganji, Z.; Kooi, H. An equation of state for thermodynamic equilibrium of gas mixtures and brines to allow simulation
of the effects of impurities in subsurface CO2 storage. Int. J. Greenh. Gas Control 2012, 11, S21–S34. [CrossRef]
13. IEAGHG. Effects of Impurities on Geological Storage of CO2 ; IEAGHG Publications: Cheltenham, UK, 2011.
14. Kolster, C.; Mechleri, E.; Krevor, S.; Mac Dowell, N. The role of CO2 purification and transport networks in carbon capture and
storage cost reduction. Int. J. Greenh. Gas Control 2017, 58, 127–141. [CrossRef]
15. Bachu, S.; Bennion, D.B. Chromatographic partitioning of impurities contained in a CO2 stream injected into a deep saline aquifer:
Part 1. Effects of gas composition and in situ conditions. Int. J. Greenh. Gas Control 2009, 3, 458–467. [CrossRef]
16. Wilkinson, J.R.; Leahy-Dios, A.; Teletzke, G.F.; Dickson, J.L. Use of CO2 containing impurities for miscible enhanced oil recovery.
In Proceedings of the International Oil and Gas Conference and Exhibition in China, Beijing, China, 8–10 June 2010.
17. Savary, V.; Berger, G.; Dubois, M.; Lacharpagne, J.; Pages, A.; Thibeau, S.; Lescanne, M. The solubility of CO2 + H2 S mixtures in
water and 2 M NaCl at 120 ◦ C and pressures up to 35 MPa. Int. J. Greenh. Gas Control 2012, 10, 123–133. [CrossRef]
18. Shabani, B.; Vilcáez, J. Prediction of CO2 -CH4 -H2 S-N2 gas mixtures solubility in brine using a non-iterative fugacity-activity
model relevant to CO2 -MEOR. J. Pet. Sci. Eng. 2017, 150, 162–179. [CrossRef]
19. Dhima, A.; de Hemptinne, J.; Jose, J. Solubility of hydrocarbons and CO2 mixtures in water under high pressure. Ind. Eng.
Chem. Res. 1999, 38, 3144–3161. [CrossRef]
20. Zhang, P.Y.; Huang, S.; Sayegh, S.; Zhou, X.L. Effect of CO2 impurities on gas-injection EOR processes. In Proceedings of the
SPE/DOE Symposium on Improved Oil Recovery, Tulsa, OK, USA, 17–21 April 2004.
21. Luo, E.; Hu, Y.; Wang, J.; Fan, Z.; Hou, Q.; Ma, L.; Dai, S. The effect of impurity on miscible CO2 displacement mechanism.
Oil Gas Sci. Technol.–Revue d’IFP Energ. Nouv. 2019, 74, 1–20. [CrossRef]
22. Sayegh, S.; Huang, S.; Zhang, Y.; Lavoie, R. Effect of H2 S and pressure depletion on the CO2 MMP of Zama Oils. J. Can.
Pet. Technol. 2007, 46, 34–41. [CrossRef]
23. Blanco, S.T.; Rivas, C.; Fernández, J.; Artal, M.; Velasco, I. Influence of methane in CO2 transport and storage for CCS technology.
Environ. Sci. Technol. 2012, 46, 13016–13023. [CrossRef]
24. Ren, Q.; Chen, G.; Yan, W.; Guo, T. Interfacial tension of (CO2 + CH4 )+ water from 298 K to 373 K and pressures up to 30 MPa.
J. Chem. Eng. Data 2000, 45, 610–612. [CrossRef]
25. Srivastava, R.K.; Huang, S.S.; Dong, M. Laboratory investigation of Weyburn CO2 miscible flooding. J. Can. Pet. Technol. 2000, 39,
41–51. [CrossRef]
26. Peng, D.Y.; Robinson, D.B. A new two-constant equation of state. Ind. Eng. Chem. Fundam. 1976, 15, 59–64. [CrossRef]
27. Robinson, D.B.; Peng, D.-Y. The Characterization of the Heptanes and Heavier Fractions for the GPA Peng-Robinson Programs; Research
Report 28; Gas Processors Association: Tulsa, OK, USA, 1978.
28. Dennis Jr, J.E.; Gay, D.M.; Walsh, R.E. An adaptive nonlinear least-squares algorithm. ACM Trans. Math. Softw. (TOMS) 1981, 7,
348–368. [CrossRef]
29. Yellig, W.F.; Metcalfe, R.S. Determination and prediction of CO2 minimum miscibility pressures (includes associated paper 8876).
J. Pet. Technol. 1980, 32, 160–168. [CrossRef]
30. Elsharkawy, A.M.; Poettmann, F.H.; Christiansen, R.L. Measuring CO2 minimum miscibility pressures: Slim-tube or rising-bubble
method? Energy Fuels 1996, 10, 443–449. [CrossRef]
31. Dong, M.; Huang, S.; Dyer, S.B.; Mourits, F.M. A comparison of CO2 minimum miscibility pressure determinations for Weyburn
crude oil. J. Pet. Sci. Eng. 2001, 31, 13–22. [CrossRef]
32. Ayirala, S.C.; Rao, D.N. Comparative evaluation of a new gas/oil miscibility-determination technique. J. Can. Pet. Technol. 2011,
50, 71–81. [CrossRef]
33. Zhang, K.; Gu, Y. Two different technical criteria for determining the minimum miscibility pressures (MMPs) from the slim-tube
and coreflood tests. Fuel 2015, 161, 146–156. [CrossRef]
34. Ahmad, W.; Vakili-Nezhaad, G.; Al-Bemani, A.S.; Al-Wahaibi, Y. Uniqueness, repeatability analysis and comparative evaluation
of experimentally determined MMPs. J. Pet. Sci. Eng. 2016, 147, 218–227. [CrossRef]
35. Vulin, D.; Gaćina, M.; Biličić, V. Slim-tube simulation model for CO2 injection EOR. Rudarsko-Geološko-Naftni Zbornik 2018, 33,
37–49. [CrossRef]
36. Li, Y.K.; Nghiem, L.X. Phase equilibria of oil, gas and water/brine mixtures from a cubic equation of state and Henry’s law. Can.
J. Chem. Eng. 1986, 64, 486–496. [CrossRef]
37. Stumm, W.; Morgan, J.J. Aquatic Chemistry: Chemical Equilibria and Rates in Natural Waters; John Wiley & Sons: Hoboken, NJ, USA,
2012; Volume 126.
38. Lake, L.W. Enhanced Oil Recovery; Prentice-Hall: Hoboken, NJ, USA, 1989.
39. Chugh, S.; Baker, R.; Cooper, L.; Spence, S. Simulation of horizontal wells to mitigate miscible solvent gravity override in the
Virginia Hills Margin. J. Can. Pet. Technol. 2000, 39, 28–34. [CrossRef]