Journal of Natural Gas Science and Engineering 71 (2019) 103002

Contents lists available at ScienceDirect

Journal of Natural Gas Science and Engineering

journal homepage: www.elsevier.com/locate/jngse

Solubility trapping as a potential secondary mechanism for CO2 T

sequestration during enhanced gas recovery by CO2 injection in
conventional natural gas reservoirs: An experimental approach
Muhammad Kabir Abba∗, Abubakar J. Abbas, Ghasem G. Nasr, Athari Al-Otaibi, Martin Burby,
Bello Saidu, Salihu M. Suleiman
The University of Salford, Manchester, UK

ARTICLE INFO ABSTRACT

Keywords: This study aims to experimentally investigate the potential of solubility trapping mechanism in increasing CO2
Enhanced gas recovery storage during EGR by CO2 injection and sequestration in conventional natural gas reservoirs. A laboratory core
CO2 storage flooding process was carried out to simulate EGR on a sandstone core at 0, 5, 10 wt% NaCl formation water
Solubility trapping salinity at 1300 psig, 50 °C and 0.3 ml/min injection rate. The results show that CO2 storage capacity was
Natural gas reservoirs
improved significantly when solubility trapping was considered. Lower connate water salinities (0 and 5 wt%)
CO2 injection
showed higher CO2 solubility from IFT measurements. With 10% connate water salinity, the highest accumu-
lation of the CO2 in the reservoir was realised with about 63% of the total CO2 injected stored; an indication of
improved storage capacity. Therefore, solubility trapping can potentially increase the CO2 storage capacity of the
gas reservoir by serving as a secondary trapping mechanism in addition to the primary structural and strati-
graphic trapping and improving CH4 recovery.

1. Introduction natural gas reservoirs have the upper hand in terms of potential and
practicality storage compared to the oil reservoirs due to their existing
The incessant utilisation of fossil fuels as sources of energy in- gas storage capability (Ding et al., 2018). Gas reservoirs have stored
variably increases greenhouse gases (GHG) emissions in the atmosphere natural gas for long periods of time safely without the penchant for
and eventually lead to the proliferation of global warming. The re- surrounding leaks. The extraction of gas from conventional natural gas
duction of these GHG emissions has become paramount, and it is reservoirs does not require complex processes of altering the reservoir
gaining significant attention globally due to its environmental con- matrix to enhance its production because primary recovery can be in
sequences. The main component of the GHGs responsible for nearly excess of 80% (van der Meer, 2005) depending on the drive mechanism.
64% of the accrued negative effect on the environment is CO2 (Ding Oil reservoirs, on the other hand, must be subjected to an array of
et al., 2018). Thus, reduction of this anthropogenic CO2 emission complex process through artificial stimulation techniques like hydraulic
cannot be overemphasised. Carbon Capture technology is a potential fracturing or matrix acidizing in order to enhance the production which
approach to reducing the CO2 emissions from heavy industries (Ding in turn tends to affect the reservoir integrity, and a potential CO2
et al., 2018; Li et al., 2013; Pereira et al., 2017) followed by the geo- leakage and contamination of adjacent freshwater aquifers will ensue
logical/underground storage and sequestration of the captured CO2 overtime as emphasised by (Xiao et al., 2016). CO2 is a medium for the
(Burton et al., 2009; Ganjdanesh and Hosseini, 2017; Mijic et al., 2014; mobilisation of oil in the reservoir during tertiary recovery and the risk
Mutailipu et al., 2018a). In all the underground storage sites, oil and associated with the leaks is the transportation of complex organic
gas reservoirs have the potential or appeal to provide additional compounds present in the crude oil like benzene, toluene, ethylbenzene
throughput in the form of economic incentives (Ding et al., 2018; Kalra and xylenes (BTEX) which are highly toxic (Cantrell and Brown, 2014)
and Wu, 2014). These incentives are realised through enhanced re- into adjacent aquifers through fissures and fracture propagation re-
covery techniques – miscible flooding technique in enhanced oil re- sulting from recovery techniques. This reason, among others as pointed
covery and CO2 injection in enhanced gas recovery processes. However, in the works of Kalra et al. (2014), highlights the choice of natural gas

∗
Corresponding author.
E-mail addresses: [email protected], [email protected], [email protected] (M.K. Abba).

https://doi.org/10.1016/j.jngse.2019.103002
Received 6 May 2019; Received in revised form 18 July 2019; Accepted 17 September 2019
Available online 23 September 2019
1875-5100/ © 2019 Elsevier B.V. All rights reserved.
M.K. Abba, et al. Journal of Natural Gas Science and Engineering 71 (2019) 103002

reservoirs as potential sequestration site for anthropogenic CO2 emis- Table 1

sions albeit the current natural gas market compared to oil prices from Core sample dimensions and petrophysical properties.
an economic standpoint. Nonetheless, the drive to exploit the economic Core sample Length Diameter Porosimetry Gas Permeability
viability of EGR through understanding the physics of mixing between (mm) (mm) Porosity (%) (md)
the injected CO2 and the recovered CH4 and its minimisation should be
Grey Berea 76.27 25.22 20.3 217
upheld.
Accordingly, conventional natural gas reservoirs have a storage
capacity limitation compared to the most ideal choice i.e. deep saline
of Linde Group. The core sample was acquired from Kocurek USA.
aquifers (Sminchak et al., 2017) for CO2 sequestration. Deep saline
General purpose NaCl salt used in this study was supplied by Fisher
aquifer utilises a combination of different trapping mechanisms
Scientific UK.
(structural and stratigraphic, solubility, residual, mineral) to store the
injected CO2 and as such increases, significantly, its storage potential.
2.2. Experimental method
Conversely, the primary trapping mechanism for CO2 in conventional
natural gas reservoirs is structural trapping owing to the geological
A series of experiments were carried out to achieve the aim set out
seals of the reservoir (cap rock) which prevents the natural resource
in this study. The core sample was first characterised to evaluate the
from migrating to the upper strata of the overlying formations and in
petrophysical properties of the core sample for concise and dependable
this case the injected CO2. However, an additional trapping mechanism
measurements of the parameters under investigation. Brines of different
that can be explored to further improve the storage capacity of the
test salinities were prepared which were used for the investigation.
conventional natural gas reservoir i.e. the solubility trapping me-
After these preliminary preparatory tasks were carried out, a core
chanism of the CO2 in the formation water. This can potentially be
flooding process was conducted on the core sample to evaluate the
exploited in tandem with the primary trapping mechanism to increase
displacement efficiency of the process in the presence of the test con-
the storability of the natural gas reservoir. As a secondary CO2 geolo-
nate water prepared. 5 tests for individual brine concentrations were
gical trapping mechanism post injection time (Li et al., 2013), sig-
carried out. The effluent compositions were analysed using gas chro-
nificant volume of CO2 can be dissolved and stored in the formation
matography at different time intervals using the configured sampling
water given the high solubility of CO2 in water. In most cases, solubility
valve. Details of the procedure and set up are presented in our earlier
trapping mechanism in the purview of CO2 storage is usually asso-
work (Abba et al., 2017). The effluent rates were measured and re-
ciated, investigated, and adapted in deep saline aquifers alone (Chen
corded by the downstream flow meters. These provided the volumes
et al., 2018; Ding et al., 2018; Jia et al., 2017; Li et al., 2013; Nakajima
produced by displacement of CH4 by the injected CO2 and paved a way
and Xue, 2017; Oh et al., 2017; Raza et al., 2016; L M Valle et al., 2018;
to quantify the trapped or stored CO2 in the core sample after sub-
L.M. Valle et al., 2018). Experimental investigation of solubility trap-
stantial recovery of the desired CH4. The solubility through interfacial
ping mechanism in unconventional natural gas reservoirs during EGR
interaction between the different gases in different brine salinities for
is, to the author's knowledge, limited and the practicality of its effec-
all the experiments was studied using the rising bubble method of in-
tiveness in this regard needs to be evaluated to further present its po-
terfacial tension measurement. Details of the experimental set up and
tential as a viable option for CO2 sequestration.
procedure is shown in section 2.2.4.
Furthermore, CO2 displaces CH4 during EGR by advection or dif-
fusion mechanism where displacement is either controlled by inter-
stitial velocity or concentration gradient, unlike EGR by CO2 injection 2.2.1. Brine preparation
in unconventional natural gas reservoirs, like coal bed methane and The connate water of different salts concentration was prepared in a
shale gas, whose mechanisms of displacement are desorption of the CH4 round bottom flask with a magnetic stirrer on a magnetic plate. The
and adsorption of CO2 where selective adsorption depends on the clay stirring was kept at a medium rate as precautionary measure to obtain a
content of the rock (Duan et al., 2016). This displacement mechanism in homogeneous solution of the brine where all the solute was fully dis-
EGR for this study is quintessential to the assessment of the storability solved. Two brines of 5, 10 wt% NaCl concentrations were prepared to
of the injected CO2 and the recovery of the nascent CH4 and the in- simulate the formation water salinities for the investigation.
terplay between them in the conventional reservoir. The influence of
the connate water salinity and its presence in the conventional natural 2.2.2. Core sample saturation
gas reservoir during EGR has already been established in our previous For this study, 10% of characterised core sample was then saturated
work (Abba et al., 2018). Therefore, this study aims to highlight, ex- with the distilled water and prepared brines (5, 10 wt%) to represent
perimentally, the feasibility of solubility trapping, in addition to the connate water using the vacuum saturation method. Given the
structural trapping, as a secondary trapping mechanism during en- nature of the core sample used and the core flooding equipment, 10%
hanced gas recovery by CO2 injection and assess its potential to increase saturation was ideal as it provided sufficient surface area for the fluid-
the storage capacity of the reservoirs with respect to natural gas re- rock interaction within the pore matrix. Furthermore, it reduced the
covery efficiency in conventional sandstone reservoirs. risk of damage to the core holder, back pressure regulator and tubing of
the core flooding equipment from the acidic effluents formed by the
2. Materials and method interaction of the supercritical CO2 and the simulated connate water.
Equivalent volume to 10% of the pore volume of the core sample of the
In this study, a core flooding experiment was carried which involved tests brines was injected using the set-up in Fig. 1 under vacuum to
the injection of CO2 through a core sample saturated with CH4 and establish the connate water saturation in the core sample.
connate water at different salinities. The core sample used was Grey
Berea sandstone with petrophysical properties and dimensions as shown 2.2.2.1. Apparatus and procedure. Saturation set up shown in Fig. 1. To
in Table 1. The salinities of the connate water used were 0, 5, 10 wt% carry out the saturation process, an equivalent volume of 10% of the
NaCl. core sample's pore volume of brine is measured out and placed in a
hydrophobic syringe. The core sample (dried and weighted) was placed
2.1. Materials in the air tight sleeve in the vacuum chamber. Valve 1 was shut and
valve 2 was open and the vacuum pump was run. Valve 2 was then shut
High purity Carbon Dioxide and Methane with purities off and valve 1 was opened and the brine was injected and valve 1 was
of > 99.999% were used and sourced from BOC UK which is a member shut off. The set up was allowed to sit for 4 h. Slowly, the vacuum

2
M.K. Abba, et al. Journal of Natural Gas Science and Engineering 71 (2019) 103002

considered based on the gas zone density of 0.22 psi/ft, and the in-
jection rate was from our previous work). And as aforementioned, the
CO2 injection was done employing the same procedure, operational
conditions and equipment details used in our previous work (Abba
et al., 2017). The set-up works based on the principle of Darcy law
which defines fluid flow in porous media and its schematics is shown in
Fig. 2. An unsteady state flow was adopted to evaluate the mass balance
between the injected CO2 and the effluents realised – which comprised
of the displaced CH4 and the injected CO2. The concentration profile
was measured and recorded at 4–5 min intervals with the corre-
sponding effluent flowrates. And the run came to an end when there
were insignificant volumes of CH4 in the produced effluents.

2.2.4. Interfacial tension measurement (IFT)

Several works have been carried out to measure the interfacial
tension in CO2-brine, CH4-brine, CO2-brine-CH4 systems at different
conditions (Amin et al., 2010; Arabloo et al., 2016; Bagalkot et al.,
2018; Barati-Harooni et al., 2016; Chow et al., 2016; Dehghan et al.,
2015; Duchateau and Broseta, 2012; Kamari et al., 2017; Kashefi et al.,
2016; Khaksar Manshad et al., 2016; Mohammad Salehi et al., 2017;
Mutailipu et al., 2018a, 2018b; Pereira et al., 2017; Rashid et al., 2017;
Stukan et al., 2012; Yasuda et al., 2015) and the relationship between
the interfacial tension and solubility highlighted. These investigations
have shown that the forces that exist at the interfaces between two
phases or fluids interacting are a function of the densities, the tem-
perature and pressures of the fluids system. And there exist mass
transfer between the phases in contact which can be well attributed to
the solubility of one species of the fluids in another.

2.2.4.1. Apparatus and procedure. Here, the interfacial tension

measurement was done to assess the dissolution of the injected CO2
for the duration of CH4 displacement for EGR as a result mass transfer
between the phases. Each connate water solution with a specific
concentration was used to measure the IFT between the connate
water and the gases at the core flooding conditions of 1300 psig and
50 °C. The rising bubble technique was used to carry out the
measurement using the set up as shown in Figure below. A Corelab
high pressure high temperature surface energy experimental set up was
used (see Fig. 3). It comprised of the high pressure measurement cell
which can withstand pressures of up to 10,000 psig, a Rame-Hart
optical system with digital image processing software used for the IFT
determination using image analysis of the bubble, a high pressure HiP
Fig. 1. Vacuum saturation set up.
62-6-10 manual pump with a pressure rating of 10,000 psig for
charging the external phase (brine) and a Temco temperature
pressure was released, by opening the valve 2, and brought the chamber controller. The IFT is measured using DropImage software which uses
to atmospheric pressure. The now saturated core sample was weighed a theoretical algorithm to evaluate the parameter based on the bubble
to verify the saturated volume. The saturated core sample was then profile generated.
prepared by wrapping it in cling film and foil paper. The cling film Before the measurements, precautionary steps were taken to rid the
helps in preventing the foil paper from sticking to the core sample system of any contaminant and ensure proactive experiments in ob-
which makes it difficult to remove and clean while the foil paper help to taining reliable results. The accumulators (1&2), the IFT cell (4), the
preserve the vitton sleeve integrity by reducing the permeation of the injection needle (10), the delivery tubing was soaked in acetone for 2 h,
supercritical CO2 from the core sample through the sleeve. This CO2 and this step was repeated for all new samples being investigated. These
permeation can also cause overburden pressure fluctuations, in that the components were coupled back together and then evacuated using
CO2 passes through the vitton sleeve into the annulus of the core holder vacuum pump (8). Hot distilled water was then placed inside the ac-
set-up housing the hydraulic oil which simulates the overburden cumulators and then injected into IFT cell to flush the whole system.
pressure on the core sample. Dry compressed air was then used to dry the entire system in pre-
Prior to every experiment, the core sample was cleaned using paration for the IFT measurements.
Soxhlet extraction were a reflux of methanol cycles was used to remove The external phases (brine/distilled water) were charged into the
any traces of inorganic compounds (in this case NaCl salts) to restore cell using the manual pump till the desired pressure was attained and
the original state of the core sample for consistency. Drying in the oven also the temperature was set using Corelab temperature controller with
at 100 °C overnight followed. This ensured the removal of any moisture an accuracy of ± 0.3 °C as the system pressure was set in order to
and reagents used in the cleaning process. maintain the temperature at the desired temperature. After the pres-
sures and temperatures have stabilised, the gases (CH4 or CO2) were
2.2.3. Core flooding process then introduced into the drop phase accumulator. Then the manual
The core flooding process was carried out at 1300 psig and 0.3 ml/ pump was used to pressurise the gas in the accumulator to the desire
min injection rate using the set up shown in Fig. 2 (the pressure was test pressures slightly above the cell pressure. Creating the bubble

3
M.K. Abba, et al. Journal of Natural Gas Science and Engineering 71 (2019) 103002

Fig. 2. Schematics of core flooding set up.

inside the test cell was done by gently opening the injection needle As already mentioned in Section 2.2, a number of test runs were carried
valve and monitoring the development of the bubble. The bubble de- out to assess the repeatability of the experimental methodology and set
velopment and collapse were recorded using the DropImage software up and the error analysis for the best three runs prior to the actual runs
based on the Young-Laplace equation. Details of which are discussed in were selected and shown in the appendix.
Section 3.1.3.
This bubble measurement was repeated for 4 bubbles in each ex- 3.1.1. Methane recovery
periment for repeatability and acquired data reliability and the IFT First, the CH4 produced was evaluated based on the total volume of
measurement was made repeatedly on each bubble image obtained. effluents produced after the core flooding experiment was stopped.
These volumes were fractions of the original gas in place in the core
3. Results and discussion sample. The results obtained are presented in Table 2 below.
These results are presented in a graphical form in Fig. 4 which
3.1. Core flooding experiment shows the trends observed. As can be seen, the poorest CH4 recovery in
all the runs was realised in the run where 10 wt% of connate water was
The recovery efficiency of the experiment was investigated using a used. This can be attributed to the poor sweep efficiency of the injected
laboratory simulated displacement experiment to determine the con- because of the restrictive flow when CO2 traverses the core sample. This
centration profiles of the interacting gas species. This entailed injection restriction is as a result of the higher salinity (high density) connate
of the CO2 into the core sample saturated with CH4 and connate water. water sealing off the narrower pore spaces within the pore matrix due

Fig. 3. IFT set up - (1&2) accumulators (3) Rame-Hart digital camera (4) IFT cell (5) Monitor (6) Vent Valve (7) Vacuum Valve (8) Heating element (9) Injection
Needle (10) data logger and temperature controller (11) stability controller (12) manual pump (13) gas bottles (CH4 and CO2) (14) Light source.

4
M.K. Abba, et al. Journal of Natural Gas Science and Engineering 71 (2019) 103002

Table 2
CH4 production in pore volumes for all the runs.
Time (min) PV Time (min) PV Time (min) PV Time (min) PV
Produced Produced Produced Produced
10 wt% CH4 5 wt% CH4 Distilled CH4 Dry CH4

0.17 2.03 0.17 5.13 0.15 7.07 0.16 7.02

5.32 8.31 5.33 21.13 5.49 23.62 5.33 27.82
10.66 9.16 10.67 24.00 10.83 26.27 10.67 29.64
15.99 9.57 15.82 17.12 15.99 10.82 15.83 9.94
21.16 9.90 21.16 7.78 21.32 7.27 21.16 6.86
26.49 3.15 26.49 3.86 26.66 5.10 26.49 5.63
31.66 1.63 31.83 2.70 31.98 3.35 31.82 4.97
37.01 1.16 37.16 1.86 37.16 2.46 37.16 4.53
42.32 0.89 42.33 1.15 42.48 1.91 42.32 4.17
47.66 0.57 47.67 0.58 47.82 1.17 47.82 3.75
53.82 0.31 53.33 0.41 55.98 0.35 54.66 3.31
59.16 0.18 59.49 0.41 61.33 0.11 60 2.99
64.32 0.19 65.16 0.42 66.66 0.08 65.16 2.85

to its density compared to the other runs with lower connate water analysis will be accentuated in the runs with 10% of their pore volumes
concentration with lower densities. Because of the forced-homogeneity saturated with connate water of different salinities (0, 5, 10 wt%). The
actualised by the presence of the connate water in the pore matrix, less concentration profile also presented, notably, the variation of the
time was spent by the CO2 as it was injected through the core sample breakthrough times with the salinities. This variation was explained in
and also early CO2 breakthrough as seen in the concentration profile in our previous works where significant pressure drop was seen when CO2
Fig. 5. was displacing CH4 at a 10 wt% connate condition (Fig. 6). The same
The original gas in place (OGIP) in the core sample before the restrictive flow comes into play when explaining the variation in
commencement of the flooding process was obtained using the same breakthrough times. The higher the salinity of the connate water the
relation in Eq. (1) as employed in our previous work (Abba et al., 2018). more pore throat sealing effect was noticed. Distilled water saturated
run did not fully plug the pore throats instead it made them narrower
vb (1 Sw )
G= and the flow channels became more tortuous. Similarly, 5 wt% connate
Bg (1) water run had lower pressure drop compared to the 10 wt% connate
3
Where G is the original gas in place in scm , vb is the bulk volume in water runs. This means that the pore channels were not significantly
cm3, Sw is initial water saturation fraction, Bg is gas formation volume reduced thereby allowing more unrestricted flow through the pore
factor in cm3/scm3 for the purpose of this research. This was then used matrix.
to evaluate the CH4 recovery factor shown in Table 3.
Furthermore, the CH4 recovery was highest when there was no 3.1.2. Carbon dioxide injection and recovery
connate water saturation. This is obvious because there was no reduc- Using a simple form of gas material balance and mass conservation,
tion in the original pore volume for the gas to occupy and hence more the volume of CO2 injected, and CO2 produced can be evaluated to
volume for nascent CH4. Higher volume of CH4 was realised in the core assess the production efficiency of each injection strategy.
sample during the dry run and thus higher recovery was observed. This
VCO2 in = (VCO2 Accumulated + VCO2 Produced) (2)
will serve as the benchmark to which other tests are pitted against. So,

Fig. 4. Graphical representation of CH4 volumes produced from all the experiments.

5
M.K. Abba, et al. Journal of Natural Gas Science and Engineering 71 (2019) 103002

Fig. 5. Concentration profiles of CO2 produced.

Table 3 Table 4
CH4 recovery factor evaluation for each test. CO2 produced during EGR for all the experimental runs.
Tests Swi (%) CH4 Produced PV OGIP Recovery Tests Swi PV injected PV PV Accumulated % CO2 Stored
PV Factor (%) % Produced

10 wt% 10 13.62 82.87 16.44 10 wt% 10 36.10 13.31 22.79 63.13

5 wt% 10 53.11 82.87 64.09 5 wt% 10 36.10 29.44 6.66 18.45
Distilled 10 56.17 82.87 67.78 Distilled 10 36.10 27.72 8.38 23.21
Dry 0 81.55 83.23 97.98 Dry 0 36.10 33.65 2.45 6.79

Here, CO2 was injected at a constant flowrate rate and the effluents % connate water yielded the most significant results in terms of CO2
produced were recorded and analysed. Produced CO2 results obtained storage with 63.13% of the total pore volumes injected stored in the
and analysed are shown in Table 4. core sample. This is further established and reaffirmed in Fig. 7 where
From Table 4, it suffices to say that the experimental run with 10 wt the same run yielded the least CO2 recovered compared to the other

Fig. 6. dP changes during all experimental runs at different connate water salinities at 1300 psig and 50 °C.

6
M.K. Abba, et al. Journal of Natural Gas Science and Engineering 71 (2019) 103002

Fig. 7. CO2 volumes recovered in pore volumes as functions of time.

runs. Also, the restrictive flow during the run as a result of the sealing decreased as salinity decreased. That cannot be substantiated because
effect by the connate water aided the storage of the injected CO2 which the storage efficiency in the distilled water run is fairly higher than that
was characterised by the large pressure drop observed during the in- of the 5 wt% run. This could be due to CO2 solubility in brine and
jection. Next, experimental run with the core sample saturated with distilled water. Further investigation is thus required to substantiate the
distilled water provided stored 23.21% of the total pore volumes in- claim and expatiate the trend observed.
jected. This was followed closely by the run with 5 wt% connate water It is a well-known fact that the CO2 is highly soluble in water. The
and the least efficient storage scenario was the core sample with no mutual solubilities of CO2 and CH4 and connate water at different
connate water with the storage of 6.79% of the total pore volume in- salinities was investigated next using IFT measurement to further drive
jected. Given the similar flow behaviour of the injected CO2 in terms of and explain the narrative already postulated.
pressure drops between the distilled water and 5 wt% runs, it was ex-
pected that the storage efficiency will be very close.
3.1.3. IFT measurements
Consequently, to assess the displacement efficiency in terms of CH4
The experimental fluid-fluid IFT measurement was carried out using
recovery and CO2 sequestration, Tables 2 and 3 were combined to
the rising bubble technique described in Section 2.2.4.1. This technique
produce Fig. 8.
capitalises on the buoyancy of the gas bubble with respect to the brine
From the Figure, there seem to be an increase in recovery of CH4 as
used, in that its ability to rise through the denser fluid is exploited. The
the salinity decreases, this can be attributed to the pore matrix of the
IFT measurement is evaluated based on the profile of the gas bubble in
core sample and the flow physics of the fluid as discussed in section
the brine created in the IFT cell which is deduced using the Young-
3.1.1. For CO2 storage, however, the storage efficiency arguably
Laplace equations:

Fig. 8. Efficiency of flooding process terms of CO2 storage and CH4 recovery.

7
M.K. Abba, et al. Journal of Natural Gas Science and Engineering 71 (2019) 103002

Table 5 3 min where the bubble sizes became constant and hence the IFTs. The
IFT measurement of CO2 at different brine salinities (1300 psig 50OC). CH4 IFT results shown in Fig. 13 are similar to those obtained by
Time (s) Distilled 5 wt% (mN/m) 10 wt% (mN/m) (Yahaya et al., 2018) at the equilibrium conditions relevant to this
Water (mN/m) work.
From the results, it follows the same trend as that observed when
0.0 55.23 62.30 65.51
the measuring the CO2 IFT in brine, in that the rate of IFT decrease is
1.0 54.89 61.10 64.53
1.9 52.12 59.89 63.63
consequential to the brine salinity, with the lowest IFT measured be-
2.9 48.11 57.19 63.51 tween the CH4 and the brine. This reaffirms that the higher the salinity
4.0 44.22 55.22 63.48 of the brine the lower the gas solubility. The graphical representation of
5.0 38.16 53.45 63.41 the IFT variation with time is shown in Fig. 13.
6.0 33.67 52.32 63.40
Once the equilibrium between CH4 and the brine was attained, CO2
6.9 28.32 51.75 63.38
8.0 24.33 50.11 63.07 was now injected at the same pressure into the CH4 saturated brine to
9.0 22.12 48.29 62.36 evaluate the IFT. This was to simulate the rate of CO2 dissolution in the
reservoir during the displacement. It is noted that the connate water in
the reservoir was saturated with the CH4 prior to injection, so this step
gde2 in IFT determination of the CO2 in a CH4 saturated brine gives a re-
=
H (3) presentation of the trends observed in Fig. 14.
It is clear that the gases had the highest interfacial tension in the
Where
brine with the highest salinity and lowest interfacial tension value in
1 d distilled water. This explains why more CO2 seemed to accumulate
=f s during the run with distilled water (Table 4 and Fig. 8) compared to the
H de (4)
run with 5 wt%.
Δρ is the density difference between the two fluids, γ is the inter- The primary trapping mechanisms in the conventional natural gas
facial tension, g is the acceleration due to gravity, de is equatorial reservoirs during EGR by CO2 injection and sequestration are the
diameter of the drop, ds is the diameter of the bubble at de from the structural and stratigraphic trapping as seen in Table 4 in the dry run
apex, H is the bond number which is a function of the ratio of ds/de. which only about 7% of the injected CO2 was stored in the core sample.
The densities of the phases were evaluated using PVTsim at the test This storage efficiency was considerably increased to about 63% of the
conditions of 1300 psig and 50 °C. The IFT was first measured when the injected CO2 when connate water was introduced into the core sample.
external phase (connate water) was not saturated with the drop phase This highlights the feasibility that in addition to structural trapping,
(CO2) to observe the development and collapse of the bubble generated. solubility trapping is realisable during EGR and tends to increase the
The results for all the test fluids are shown in Table 5 where mea- storage capacity of the conventional natural gas in terms of storage
surements were taken continuously as the bubble shrunk and collapsed. while maintaining substantial recovery of the hydrocarbon resource.
The shrinking of the bubble signified the rate of mass transfer over
the interface between the gas bubbles generated and brine phase in the 4. Conclusion
cell. As seen in Fig. 9, Fig. 10, and Fig. 11, the rate of shrinkage of the
gas bubble is more pronounced in the distilled water experiment and In this research, Berea sandstone core sample was used as the
the rate decreased as the salinity of the connate water sample increased. conventional porous media to carry out a core flooding process to
The IFT decreased rapidly in the distilled water which explained the evaluate the production of CH4 and CO2 during EGR scenario in the
shrinkage observed. However, IFT rate decreased at a slower rate when presence of connate water to realise the effects of its presence. CO2
the salinity increased to 5 w% and even slowest at 10 wt% connate storage was highest in the run where the connate water salinity was
water. This is represented graphically in Fig. 12. 10 wt% which is attributed to the restrictive flow of the injected CO2 to
After the results of the rate of shrinkage and IFT variation with time displace the CH4 and was characterised by low CO2 and high CH4 re-
in the unsaturated brine were obtained, the next step was to evaluate coveries. Here, structural trapping mechanism was dominant and also
CO2 IFT when the brine was saturated with the injected CH4. The ex- solubility trapping to an extent. However, solubility trapping me-
ternal phase of the experiment (brine) was saturated with the CH4 by chanism is most pronounced during the distilled (0 wt%) and 5 wt%
injecting the gas through the injection needle which pressurised the salinity runs where both runs had similar CO2 and CH4 production, but
system to the test pressures. IFT measurements were taken at time in- the distilled water run had the higher CO2 accumulation. This is due to
tervals at the test conditions. Full equilibrium was achieved after about the higher solubility of the CO2 in distilled water as seen in their

Fig. 9. Bubble shrinkage of CO2 bubble in Distilled water L: Onset R: End (10 s interval).

8
M.K. Abba, et al. Journal of Natural Gas Science and Engineering 71 (2019) 103002

Fig. 10. Bubble shrinkage of CO2 in 5 wt% brine L: Onset R: End (10 s interval).

Fig. 11. Bubble shrinkage of CO2 in 10 wt% brine L: Onset R: End (10 s interval).

interfacial tensions – with distilled water having a value of 36.12 mN/m trapping mechanism in the reservoir connate water to increase the
and 48.20 mN/m for 5 wt% brine at the same conditions. CO2 seques- storage capacity. Considering only structural and stratigraphic trapping
tration during EGR is not just focused on the primary trapping me- mechanisms which were simulated using the dry run (no water sa-
chanisms of geological sequestration but can also exploit the solubility turation), only 7% of the injected CO2 was stored and a substantial

Fig. 12. Graphical representation of CO2 IFT decrease as a function of time at under-saturated aqueous conditions (1300 psig 50OC).

9
M.K. Abba, et al. Journal of Natural Gas Science and Engineering 71 (2019) 103002

Fig. 13. CH4 IFT as function of time at equilibrium (1300 psig 50OC).

Fig. 14. CO2 IFT as a function of time at saturated conditions.

volume of 63% of the injected CO2 was sequestered when solubility for CO2 storage by showcasing the potential of solubility trapping as a
trapping was considered by introducing connate water saturation. This secondary trapping mechanism which increases the storage capacity
goes on to show there is a potential for additional storage capacity (the limitation) of natural gas reservoir. Future work will entail in-
through secondary trapping mechanisms. The salinity of the connate vestigation of effect of other types of salts on this process and at even
water plays a vital role in promoting the trapping – in this case struc- higher concentrations and saturations.
tural trapping which resulted from the density of the connate water
sealing off the narrow pore spaces within the pore matrix as evident in
the 10 wt% connate water run. A substantial volume of CH4 was re- Acknowledgements
covered in all the cases which is a win-win scenario for the technique.
The recovered CH4 from conventional natural gas reservoir can offset The authors wish to acknowledge Petroleum Technology
part of the cost of the sequestration process whilst providing good se- Development Fund (PTDF) for the studentship, and also the Spray
questration site CO2 storage. This study shows that structural and Research and Petroleum & Gas research groups of University of Salford,
stratigraphic trapping mechanisms are not the only exploitable avenue Manchester UK for their consultation.

10
M.K. Abba, et al. Journal of Natural Gas Science and Engineering 71 (2019) 103002

Appendix

Error analysis of the core flooding runs

Dry Runs Distilled water 5 wt% NaCl 10 wt% NaCl

Time Average Standard Time Average Standard Time Average Standard Time Average Standard
(min) Mol. Frac CO2 Deviation CO2 (min) Mol. Frac CO2 Deviation (min) Mol. Frac CO2 Deviation (min) Mol. Frac CO2 Deviation
(%) (%) (%) (%) (%) (%) (%) (%)

0.17 0.0 0.0 0.17 0.0 0.0 0.16 0.0 0.0 0.15 0.0 0.0
5.32 0.0 0.0 5.33 0.0 0.0 5.33 0.0 0.0 5.49 0.0 0.0
10.66 0.0 0.0 10.67 2.3 2.5 10.67 1.7 2.9 10.83 6.3 4.7
15.99 1.0 1.0 15.82 6.5 3.7 15.83 61.8 2.8 15.99 70.7 4.0
21.16 4.9 4.5 21.16 70.8 11.0 21.16 74.6 3.1 21.32 78.6 3.2
26.49 70.2 4.0 26.49 89.2 3.3 26.49 84.2 4.1 26.66 82.8 2.6
31.66 81.1 3.8 31.83 92.0 1.7 31.82 90.4 2.5 31.98 85.1 1.0
37.01 91.0 1.0 37.16 94.7 1.4 37.16 94.1 2.4 37.16 86.6 1.5
42.32 93.5 1.3 42.33 96.6 0.7 42.32 95.8 2.1 42.48 88.0 1.0
47.66 96.1 0.8 47.67 97.6 0.6 47.82 97.0 1.0 47.82 89.5 0.4
53.82 97.5 0.5 53.33 97.9 0.8 54.66 97.0 1.0 55.98 90.3 0.9
59.16 98.5 0.4 59.49 98.2 0.3 60.01 97.3 1.2 61.33 90.6 1.2
64.32 98.5 0.5 65.16 98.2 0.3 65.16 97.5 1.2 66.66 90.8 1.4

11
M.K. Abba, et al. Journal of Natural Gas Science and Engineering 71 (2019) 103002

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.jngse.2019.103002.

Nomenclature

CH4 Methane
CO2 Carbon dioxide
γ Interfacial Tension IFT (mN/m)
ρ Density (g/cm3)
g Acceleration due to gravity (m/s2)
de Bubble equatorial diameter (mm)
ds Bubble diameter from bottom tip of bubble to height de (mm)
H Bond number
f Function
PV Pore volume

References interfacial tension measurement. Fluid Phase Equilib. 295, 230–236. https://doi.org/
10.1016/j.fluid.2010.05.020.
Arabloo, M., Ghazanfari, M.H., Rashtchian, D., 2016. Wettability modification, interfacial
Abba, M.K., Abbas, A.J., Nasr, G.G., 2017. Enhanced gas recovery by CO2 injection and tension and adsorption characteristics of a new surfactant: implications for enhanced
sequestration: effect of connate water salinity on displacement efficiency. In: SPE Abu oil recovery. Fuel 185, 199–210. https://doi.org/10.1016/j.fuel.2016.06.088.
Dhabi International Petroleum Exhibition & Conference, . https://doi.org/10.2118/ Bagalkot, N., Hamouda, A.A., Isdahl, O.M., 2018. Dynamic interfacial tension measure-
188930-MS. ment method using axisymmetric drop shape analysis. MethodsX 5, 676–683.
Abba, M.K., Al-Othaibi, A., Abbas, A.J., Nasr, G.G., Mukhtar, A., 2018. Experimental https://doi.org/10.1016/j.mex.2018.06.012.
investigation on the impact of connate water salinity on dispersion coefficient in Barati-Harooni, A., Soleymanzadeh, A., Tatar, A., Najafi-Marghmaleki, A., Samadi, S.J.,
consolidated rocks cores during Enhanced Gas Recovery by CO2 injection. J. Nat. Gas Yari, A., Roushani, B., Mohammadi, A.H., 2016. Experimental and modeling studies
Sci. Eng. 60, 190–201. https://doi.org/10.1016/j.jngse.2018.10.007. on the effects of temperature, pressure and brine salinity on interfacial tension in live
Amin, R., Sidiq, H., Kennaird, T., Van der Steen, E., 2010. Gas-gas experimental oil-brine systems. J. Mol. Liq. 219, 985–993. https://doi.org/10.1016/j.molliq.2016.

12
M.K. Abba, et al. Journal of Natural Gas Science and Engineering 71 (2019) 103002

04.013. impact of non Darcy flow, phase misciibility and gas compressibility. Water Resour.
Burton, M., Kumar, N., Bryant, S.L., 2009. CO2 injectivity into brine aquifers: why re- Res. WR014893, 4163–4185. https://doi.org/10.1002/2013WR014893 (Received).
lative permeability matters as much as absolute permeability. Energy Procedia 1, Mohammad Salehi, M., Omidvar, P., Naeimi, F., 2017. Salinity of injection water and its
3091–3098. https://doi.org/10.1016/j.egypro.2009.02.089. impact on oil recovery absolute permeability, residual oil saturation, interfacial
Cantrell, K.J., Brown, C.F., 2014. Source term modeling for evaluating the potential tension and capillary pressure. Egypt J. Pet. 26, 301–312. https://doi.org/10.1016/j.
impacts to groundwater of fluids escaping from a depleted oil reservoir used for ejpe.2016.05.003.
carbon sequestration. Int. J. Greenh. Gas Control 27, 139–145. https://doi.org/10. Mutailipu, M., Liu, Y., Jiang, L., Zhang, Y., 2018a. Measurement and estimation of
1016/j.ijggc.2014.05.009. CO2–brine interfacial tension and rock wettability under CO2 sub- and super-critical
Chen, L., Wang, M., Kang, Q., Tao, W., 2018. Pore scale study of multiphase multi- conditions. J. Colloid Interface Sci. https://doi.org/10.1016/j.jcis.2018.09.031.
component reactive transport during CO2dissolution trapping. Adv. Water Resour. Mutailipu, M., Liu, Y., Jiang, L., Zhang, Y., 2018b. Measurement and estimation of
116, 208–218. https://doi.org/10.1016/j.advwatres.2018.02.018. CO2–brine interfacial tension and rock wettability under CO2 sub- and super-critical
Chow, Y.T.F., Maitland, G.C., Trusler, J.P.M., 2016. Interfacial tensions of the (CO2 + N2 conditions. J. Colloid Interface Sci. 534, 605–617. https://doi.org/10.1016/j.jcis.
+ H2O) system at temperatures of (298 to 448) K and pressures up to 40 MPa. J. 2018.09.031.
Chem. Thermodyn. 93, 392–403. https://doi.org/10.1016/j.jct.2015.08.006. Nakajima, T., Xue, Z., 2017. Trapping mechanisms in field scale: results from Nagaoka
Dehghan, A.A., Masihi, M., Ayatollahi, S., 2015. Phase behavior and interfacial tension geologic CO2Storage site. Energy Procedia 114, 5015–5022. https://doi.org/10.
evaluation of a newly designed surfactant on heavy oil displacement efficiency; ef- 1016/j.egypro.2017.03.1650.
fects of salinity, wettability, and capillary pressure. Fluid Phase Equilib. 396, 20–27. Oh, J., Kim, K.Y., Han, W.S., Park, E., 2017. Transport of CO2in heterogeneous porous
https://doi.org/10.1016/j.fluid.2015.03.028. media: spatio-temporal variation of trapping mechanisms. Int. J. Greenh. Gas Control
Ding, S., Xi, Y., Jiang, H., Liu, G., 2018. CO2storage capacity estimation in oil reservoirs 57, 52–62. https://doi.org/10.1016/j.ijggc.2016.12.006.
by solubility and mineral trapping. Appl. Geochem. 89, 121–128. https://doi.org/10. Pereira, L.M.C., Chapoy, A., Burgass, R., Tohidi, B., 2017. Interfacial tension of CO2+
1016/j.apgeochem.2017.12.002. brine systems: experiments and predictive modelling. Adv. Water Resour. 103,
Duan, S., Gu, M., Du, X., Xian, X., 2016. Adsorption equilibrium of CO2 and CH4 and their 64–75. https://doi.org/10.1016/j.advwatres.2017.02.015.
mixture on Sichuan basin shale. Energy Fuels 30, 2248–2256. https://doi.org/10. Rashid, S., Harimi, B., Hamidpour, E., 2017. Prediction of CO2-Brine interfacial tension
1021/acs.energyfuels.5b02088. using a rigorous approach. J. Nat. Gas Sci. Eng. 45, 108–117. https://doi.org/10.
Duchateau, C., Broseta, D., 2012. A simple method for determining brine-gas interfacial 1016/j.jngse.2017.05.002.
tensions. Adv. Water Resour. 42, 30–36. https://doi.org/10.1016/j.advwatres.2012. Raza, A., Rezaee, R., Bing, C.H., Gholami, R., Hamid, M.A., Nagarajan, R., 2016. Carbon
03.008. dioxide storage in subsurface geologic medium: a review on capillary trapping me-
Ganjdanesh, R., Hosseini, S.A., 2017. Geologic carbon storage capacity estimation using chanism. Egypt J. Pet. 25, 367–373. https://doi.org/10.1016/j.ejpe.2015.08.002.
enhanced analytical simulation tool (EASiTool). Energy Procedia 114, 4690–4696. Sminchak, J.R., Babarinde, O., Gupta, N., 2017. Integrated analysis of geomechanical
https://doi.org/10.1016/j.egypro.2017.03.1601. factors for geologic CO2Storage in the midwestern United States. Energy Procedia
Jia, W., Pan, F., Dai, Z., Xiao, T., McPherson, B., 2017. Probabilistic risk assessment of 114, 3267–3272. https://doi.org/10.1016/j.egypro.2017.03.1458.
CO2Trapping mechanisms in a sandstone CO2-EOR field in Northern Texas, USA. Stukan, M., Abdallah, W., East, S.M., 2012. In: Interfacial Tension (IFT) and Surface al-
Energy Procedia 114, 4321–4329. https://doi.org/10.1016/j.egypro.2017.03.1581. teration Interplay. Spe, pp. 11–14. https://doi.org/10.2118/161279-ms.
Kalra, S., Wu, X., 2014. CO2 Injection for Enhanced Gas Recovery. SPE West. North Am. Valle, L.M., Rodríguez, R., Grima, C., Martínez, C., 2018. Effects of supercritical CO2
Rocky Mt., pp. 16–18. injection on sandstone wettability and capillary trapping. Int. J. Greenh. Gas Control
Kamari, A., Pournik, M., Rostami, A., Amirlatifi, A., Mohammadi, A.H., 2017. 78, 341–348. https://doi.org/10.1016/j.ijggc.2018.09.005.
Characterizing the CO2-brine interfacial tension (IFT) using robust modeling ap- van der Meer, B., 2005. Carbon dioxide storage in natural gas reservoirs. Oil Gas Sci.
proaches: a comparative study. J. Mol. Liq. 246, 32–38. https://doi.org/10.1016/j. Technol. 60, 527–536. https://doi.org/10.2516/ogst:2005035.
molliq.2017.09.010. Xiao, T., McPherson, B., Pan, F., Esser, R., Jia, W., Bordelon, A., Bacon, D., 2016. Potential
Kashefi, K., Pereira, L.M.C., Chapoy, A., Burgass, R., Tohidi, B., 2016. Measurement and chemical impacts of CO 2 leakage on underground source of drinking water assessed
modelling of interfacial tension in methane/water and methane/brine systems at by quantitative risk analysis. Int. J. Greenh. Gas Control 50, 305–316. https://doi.
reservoir conditions. Fluid Phase Equilib. 409, 301–311. https://doi.org/10.1016/j. org/10.1016/j.ijggc.2016.04.009.
fluid.2015.09.050. Yahaya, A.A., Akpan, E.U., Enyi, G.C., Nasr, G.G., Abbas, J., 2018. Experimental in-
Khaksar Manshad, A., Olad, M., Taghipour, S.A., Nowrouzi, I., Mohammadi, A.H., 2016. vestigation of methane-water and methane-brine IFT measurements using pendant
Effects of water soluble ions on interfacial tension (IFT) between oil and brine in drop (rising bubble) method Journal of engineering technology. J. Eng. Technol. 6,
smart and carbonated smart water injection process in oil reservoirs. J. Mol. Liq. 223, 394–407.
987–993. https://doi.org/10.1016/j.molliq.2016.08.089. Yasuda, K., Mori, Y.H., Ohmura, R., 2015. Interfacial tension measurements in water-
Li, B., Tchelepi, H.A., Benson, S.M., 2013. Influence of capillary-pressure models on CO2 methane system at temperatures from 278.15 K to 298.15 K and pressures up to 10
solubility trapping. Adv. Water Resour. 62, 488–498. https://doi.org/10.1016/j. MPa. Fluid Phase Equilib. 413, 170–175. https://doi.org/10.1016/j.fluid.2015.10.
advwatres.2013.08.005. 006.
Mijic, A., LaForce, T.C., Muggeridge, A., 2014. Co2 injectivity in saline aquifers: the

13