Article

pubs.acs.org/EF

Wettability Evaluation of a CO2/Water/Bentheimer Sandstone

System: Contact Angle, Dissolution, and Bubble Size
N. Shojai Kaveh,*, E. S. J. Rudolph, P. van Hemert, W. R. Rossen, and K.-H. Wolf

Department of Geoscience and Engineering, Delft University of Technology, Delft 2628CN, The Netherlands

Fakultaet III, Institut fuer Lebensmitteltechnologie und Lebensmittelchemie, Technische Universitat Berlin, 14195 Berlin, Germany
*
S Supporting Information

ABSTRACT: The success of CO2 storage in deep saline aquifers and depleted oil and gas reservoirs is largely controlled by
interfacial phenomena among uid phases and rock pore spaces. Particularly, the wettability of the rock matrix has a strong eect
on capillary pressure, relative permeability, and the distribution of phases within the pore space and thus on the entire
displacement mechanism and storage capacity. Precise understanding of wettability behavior is therefore fundamental when
injecting CO2 into geological formations to sequestrate CO2 and/or to enhance gas/oil production. In this study, the contact
angles of Bentheimer sandstone/water/CO2 or ue gas have been evaluated experimentally using the captive-bubble technique in
the pressure range from 0.2 to 15 MPa. The experiments were conducted using dierent compositions of aqueous phase with
respect to CO2, i.e., unsaturated and fully saturated. It has been shown that a reliable contact-angle determination needs to be
conducted using a pre-equilibrated aqueous phase to eliminate dissolution eects. In the fully saturated aqueous phase, the
Bentheimer sandstone/water system is (and remains) water-wet even at high pressures against CO2 and/or ue gas. In these
systems, the data of the stable contact angle demonstrate a strong dependence on the bubble size, which can be mainly explained
by the gravity (buoyancy) eect on bubble shape. However, the surface nonideality and roughness have signicant inuence on
the reliability of the contact-angle determination. The results of this study prove that in order to avoid the dependency of the
contact angle on the bubble size in these systems, the eect of gravity (buoyancy) on bubble shape has to be considered by
calculation of the Bond number; for systems characterized by Bond numbers less than 0.9, the inuence of the bubble radius on
the contact angle becomes insignicant. The experimental results show that, in contrast to quartz, the phase transition of CO2
from subcritical to supercritical has no eect on the wettability of the Bentheimer sandstone/water system, which originates from
dierences in the surface charges of quartz and Bentheimer sandstone. In an unsaturated system, two dissolution regimes are
observed, which may be explained by density-driven natural convection and molecular diusion.

1. INTRODUCTION
To reduce CO2 emissions into the atmosphere, dierent
scenarios are proposed to capture and store carbon dioxide
(CO2) in geological formations (CCS). Storage strategies
include CO2 injection into deep saline aquifers,1,2 depleted gas
and oil reservoirs,35 and unmineable coal seams.6 Even though
the volumetric CO2 storage capacity is the highest in aquifers,
CO2 storage by means of CO2-enhanced gas and oil recovery
(CO2-EGR and CO2-EOR) is more economically viable.
In reservoir engineering, it has been well-identied that
interfacial interactions, i.e. ,wettability, capillarity, interfacial
tension, and mass transfer, control the ow behavior and the
displacement in porous media. Particularly, wettability has
been known as one of the most important factors determining
the residual saturation, the capillary-pressure, and the relative-
permeability functions.7,8 This means that the process of CO2
storage in hydrocarbon reservoirs and in deep saline aquifers is
inuenced by gas/liquid/rock interfacial interactions.911 The
relation between the interfacial interactions (interfacial tension,
capillarity, and wettability) is represented by the Young
Laplace equation:
where Pc is the capillary pressure, Pnw and Pw are the pressures
in the nonwetting and wetting phases, respectively, aq,CO2 is the
interfacial tension between the aqueous phase and the CO2-rich
phase, R is an eective pore radius corresponding to the narrowest
pore throat along the entire CO2 ow path,12 and is the contact
angle related to the reservoir wettability (determined through the
densest phase, here the aqueous phase; Figure 1). The capillary
pressure can be positive or negative, depending on the wetting
phase; i.e., the contact angle is smaller or greater than 90.
To identify the secure strategy for CO2 injection, not only
the wettability and its relation to uid distribution but also the
physics of the trapping mechanisms need to be understood.
Capillary trapping occurs when CO2 is immobilized in the
rock pores by capillary forces. This process depends on the
interfacial tension (IFT) between the CO2-rich and the
aqueous phase (brine), wettability of the rock, and the pore-
size distribution (eq 1).13,14 In the literature, a large amount of
data related to research with respect to CO2 storage in depleted
gas reservoirs and aquifers can be found. Current focus is on
geochemical modeling, aquifer and reservoir simulation,
long-term reservoir integrity, and risk assessment. Only a very

2aq,CO cos Received: January 7, 2014

Pc = Pnw Pw = 2 Revised: April 28, 2014
R (1) Published: May 1, 2014

2014 American Chemical Society 4002 dx.doi.org/10.1021/ef500034j | Energy Fuels 2014, 28, 40024020
Energy & Fuels
Figure 1. Schematic of a captive bubble explaining properties such as
contact angle, interfacial tensions, and dimensions of the bubble.
limited amount of literature data with focus on wetting
properties of aquifers and gas reservoirs at high pressures and
elevated temperatures can be found.1419 A reliable exper-
imental method to determine the wettability contributes a
better understanding of the physics of this phenomenon and
the displacement behavior of uids in reservoirs.
The main purpose of the study at hand is to examine the
wettability behavior of the system of Bentheimer sandstone/
water in the presence of CO2 at elevated pressure. Data on
the contact angle were considered for the evaluation of the
wettability of this system because (1) values of the contact
angle, particularly in the pressure range where CO2 is
supercritical, are necessary for the evaluation of the capillary
trapping;12 and (2) when dealing with supercritical CO2,
wettability cannot be measured by the Amott test and U.S.
Bureau of Mines (USBM) method due to the fact that these
methods cannot be used at high pressures. To this end, contact-
angle measurements were conducted for dierent compositions
of the aqueous phase, i.e., fully saturated and unsaturated with
respect to CO2, at a constant temperature of 318 K and at
pressures varying between 0.2 and 15 MPa. Under unsaturated
conditions with respect to CO2, the injectivity and the gas
distribution in the reservoir are not only inuenced by the rock
properties but also by the diusion of CO2 into the aqueous
phase. After the aqueous phase has been saturated with CO2,
the injectivity is mainly determined by the wetting properties of
the rock and pore morphology. Before the method can be
applied in the eld, it is thus important to evaluate the
dissolution eects and the wettability for short and long
periods. This can be done by conducting experiments with an
unsaturated aqueous phase, representing the short-term
behavior, and with a saturated aqueous phase representing
the long-term behavior.
In addition, the wettability behavior of the Bentheimer
sandstone/water system has been evaluated for synthetic ue
gas (80:20 N2/CO2). Direct injection of ue gas into a reservoir
eliminates the need for CO2 separation prior to its injection
into the eld. Moreover, nitrogen stripping can recover more
methane than reservoir pressure depletion alone.20,21 Never-
theless, ue gas injection can reduce the structural trapping
capacity for CO2. This reduction is undesirable for CCS volume
eciency. To understand this process in more detail, ue gas
injection has been evaluated from a wettability point of view,
which is missing in the literature.
4003
Article
2. THEORETICAL REVIEW
2.1. Wettability and Contact-Angle Determination.
Wettability of reservoir rocks is determined by a complex
relationship of interfacial interactions between the rock,
composed of a wide variety of minerals, and the reservoir
uids that occupy the irregular pore space. Experimentally, the
wettability is only determined in the laboratory because no
experimental method exists for in situ measurements. Dierent
techniques, either quantitative or qualitative, have been
developed to evaluate the wettability of a rock/uid system,
namely, the Amott test, the USBM test, and contact-angle
measurements. The Amott and USBM methods give a quantitative
value of the wettability of a core only at atmospheric conditions.
The contact-angle method allows the determination of the
wettability of a specic surface also at high pressures and elevated
temperatures.22 To avoid complications due to surface roughness
and heterogeneity during the experiments and data processing, the
experimental determination of the contact angle is commonly
conducted on idealized and polished surfaces. The wettability
determination on a polished surface leads to an apparent contact
angle determination. However, the average of these contact
angles can be used as an index for the wettability of the surface.
Therefore, experimentally determined contact angles are widely
used to characterize the wettability of complicated systems, in
particular at high pressures and elevated temperatures.12,14,19 For
the characterization of the surface wettability by means of the
contact angle, Youngs equation is applied. The Youngs contact
angle, Y, is a unique contact angle at equilibrium. However, in
practice, many metastable states of a droplet/bubble on a solid
exist, which might lead to an inequality of the experimentally
determined contact angle, , with the Youngs contact angle, Y. In
general, the dierence in the receding, r, and advancing contact
angle, a, is called the contact angle hysteresis and results from
surface roughness and/or heterogeneity.23 On ideal surfaces, there
is no contact-angle hysteresis, and the experimentally determined
contact angle is equal to Y.
The contact angle between the three phases ( in Figure 1)
determines whether a reservoir rock is water-wet ( < 75),
intermediate-wet (75<<105), or oil-wet ( > 105). The
contact angle can be experimentally determined in a pendant/
sessile drop cell that is adapted to allow the analysis of a
captured bubble/sessile drop at a rock surface.
2.1.1. Contact Angle and Droplet/Bubble Size. The
application of the Young equation requires well-dened
conditions, such as an ideal and perfectly smooth solid surface.
Y is exclusively determined by the liquid/solid, liquid/gas, and
gas/solid interfacial properties. Based on the Young equation,
contact angle is a single and unique contact angle and independent
of the mass and volume of the bubble/drop. However, extensive
experimental results found in the literature show that, for some
cases, the contact angle of a captive bubble/sessile droplet on a
horizontal surface is not only a physical property of the gas, liquid,
and solid materials but also varies with the bubble/droplet
size.2427 To address this phenomenon, several explanations have
been oered in the literature, i.e., surface roughness and
hysteresis,24,28 line tension,26,29 and gravity eects on droplet/
bubble geometry.30
Good and Koo27 presented the concept of line tension () to
explain the variation of the contact angle with drop size. The
so-called line tension is dened as the excess Gibbs energy per
unit length of the contact line between three coexisting phases.
Including the concept of line tension, which gives the
dx.doi.org/10.1021/ef500034j | Energy Fuels 2014, 28, 40024020
Energy & Fuels
dependency of the contact angle on bubble size,24,27 leads to
the modied Young equation.
In the literature, there are a number of experimental studies
on determining the line tension; however, the reported values
vary widely in both magnitude and sign that can be attributed
to (1) typically small values of these parameters, in particular of
the contact line length, which are used for the line-tension
determination, and/or (2) the fact that line tension is a newly
introduced physical property that is dicult to measure and based
on highly uncertain data.30,31 Values of the line tension found in the
literature are both negative and positive and vary in the range of
absolute values between 1011 and 105 J/m. Recently, Liu et al.32
reported that the linear relationship between cos and 1/R, which
is normally used for the line-tension determination, is incorrect at
the nanometer scale. In addition, by using the roughness model,
Lin et al.24 showed that the experimentally observed uctuations of
apparent contact angles, which are more scattered than the
prediction of the modied Young equation for a smooth surface,
are attributed to surface roughness. They concluded that the
modied Young equation is inappropriate to interpret the contact-
angle and bubble-size relation on a rough surface.24
These contradictory results about the line-tension determi-
nation led to a search for other approaches, which can be also
applicable for nonideal and complex systems. The other explanation
for addressing the contact-angle variation with droplet size might
be provided by considering the eect of gravity on the droplet
shape.30,31 Pethica and Pethica33 and Leja and Poling34
formulated a model explaining the inuence of gravity on the
contact angle. Vafaei and Podowski30,31 introduced models for
two dierent droplet geometries. They showed that the contact
angle for elongated droplets is size-independent and can be
described by the original Young equation. According to their
model, axisymmetric droplets are size-dependent and cannot
satisfy the original Young equation30,31 but can be described by
2lvR2 l g 1 sin
V= +
l g 2lv R0 R (2)
This approach is based on a liquid droplet surrounded by a
gas phase. Equation 2 represents a relationship between the
volume of a droplet (V), the radius of the contact circle (R), the
contact angle (), the radius of the curvature at the apex (R0),
the location of the apex (), and the properties of the droplet,
such as density (l), and surface tensions between the
coexisting phases (lv) (Figure 1).
The description of the eect of gravity on the contact angle
of a liquid droplet surrounded by a gas phase is based on the two
main forces acting on the droplet, i.e., the surface force and the
gravity, which are considered to be independent of each other. If
the gravity force dominates, the contact angle is inuenced by the
droplet volume. For a smaller droplet, the eect of gravity decreases
and surface forces dominate; with decreasing droplet radius, the
droplet shape asymptotically approaches a spherical shape. The
contact angle for a spherical droplet is exclusively determined by
the liquid/solid, liquid/gas, and gas/solid interfacial properties and
the properties of the phases such as density, and is independent of
the droplet volume, therefore fullling the Young equation.
To identify whether surface forces or gravity forces dominate,
the Bond number is used. The Bond number is a dimensionless
number described by
gL2
Bo =
lv (3)
4004
Article
where is the density dierence between the liquid and the
gas phases and L is the characteristic length, which is the
appropriate linear dimension. For bubble (droplet) systems, the
characteristic length is the bubble diameter (MD in Figure 1).35
A high Bond number indicates that the system is relatively
unaected by surface tensions; a low number indicates that
surface tension dominates. Intermediate numbers indicate a
nontrivial balance between the two forces.
Nevertheless, there are still discussions on whether the
contact angle of a liquid on a solid is inuenced by drop size
and gravity. Herzberg and Marian28 showed experimentally that
the contact-angle variation with the drop size was generated by
hysteresis rather than by gravity. In addition, Blokhuis et al.36
showed analytically that Youngs contact angle is independent
of the strength of the gravitational eld.
Fujii and Nakae37 quantitatively calculated the eect of
gravity on the contact angle using a precise drop-shape model.
They showed that gravity has no eect on the equilibrium
contact angle. They concluded that the eect that Pethica and
Pethica33 and Leja and Poling34 attributed to gravity, indeed,
originates from the dierence between the droplet shape
assumed in their models and the actual droplet shape. However,
Sakai and Fujii38 subsequently found that the solid/liquid
tension was enhanced by gravity when gas was adsorbed at a
rough solid interface. Accordingly, they concluded that the
apparent contact angle on rough surfaces, which was considered
not to be inuenced by gravity in a previous publication,37 can
be changed by gravity.
2.2. Interfacial Interactions of CO2/Water/Rock or
Mineral Systems. In the past decade, a few studies have
focused on the wettability of CO2waterrock/mineral
systems.12,14,19,3942 Chalbaud et al.17 show that reliable wettability
data are crucial for the accurate design and optimization of CO2
sequestration strategies.
The wetting properties and the interfacial interactions
between (reservoir) brine, carbonate reservoir rock, and CO2-
rich phase at various temperatures and pressures were studied
by Yang et al.39 They applied an axisymmetric drop-shape
analysis (ADSA) technique for image analysis to determine the
dynamic and equilibrium (static or stable) contact angles. They
found that the equilibrium contact angle increases with pressure
and decreases with temperature. It was assumed that this
might be attributed to the fact that the solubility of CO2 in the
aqueous phase is higher at higher pressures and lower
temperatures.
Espinoza and Santamarina19 collected literature data and
conducted a number of experiments at pressures up to 20 MPa
and a temperature of 298 K to obtain data on interfacial
tensions, contact angles, and diusion for systems with CO2,
saline water, and dierent substrates (amorphous silica, calcite,
silica coated with oil, and poly(tetrauoroethylene) (PTFE)).
They found that the contact angles on amorphous silica and
calcite substrates remained nearly constant with pressure while
dissolved NaCl in the aqueous phase increased the contact
angle by approximately 20 for SiO2 and approximately 4 for
CaCO3.19
The contact angle of CO2 on mineral samples consisting of
quartz, orthoclase, labradorite, calcite, and biotite was
determined by Mills et al. using the captive-bubble technique.14
Their ndings show strongly water-wet to water-wet conditions
for all experiments for which CO2 is in the gaseous or super-
critical state (5.513.8 MPa at 313 K). Their data show that
mica and calcite substrates become more water-wet as pressure
dx.doi.org/10.1021/ef500034j | Energy Fuels 2014, 28, 40024020
Energy & Fuels
is reduced, whereas quartz and biotite substrates become more
water-wet with increasing pressure. Comparison of the
determined contact angles for N2 and CO2 bubbles shows
that the system with CO2 is less water-wet than one with N2 at
low pressure; no clear trend was observed with increasing
pressure. Although they reported contact angles as a function of
contact circle radius, they did not draw any conclusion with
respect to this dependency.
Jung and Wan41 studied the wettability alterations of silica
surfaces, aqueous phase, and the CO2 system under pressures
from 0.1 to 25 MPa while the NaCl concentration in the
aqueous phase varies from 0 to 5.0 M. They found that, at
pressures higher than the critical pressure of CO2 (larger than
7.39 MPa), the contact angle increases steeply with increasing
pressure up to a pressure of 10 MPa. For pressures below the
critical pressure and pressures above 10 MPa, the contact angle
remained almost unchanged. They showed that the contact
angle increased with ionic strength.
Ameri et al. conrmed that the phase transition of CO2 from
a subcritical gas to a supercritical gas aects the wettability of a
hydrophobic Bentheimer sandstone.43 However, when CO2
was either subcritical or supercritical, no signicant inuence of
pressure on the wetting properties of the hydrophilic
Bentheimer was observed.
Saraji et al.42 showed that CO2 phase change and pressure
variation do not signicantly inuence the water receding
contact angles on the quartz surface. However, the water
advancing contact angles increased from 5 to 12 with
increasing pressure and temperature. They concluded that
supercritical CO2 changes the wettability of the quartz surface
toward less water-wet conditions compared to subcritical CO2.
In our previous works, we have investigated the eect of
pressure on the wettability behavior of two samples of coal of
dierent rank and Bentheimer sandstone.40,44,45 We observed
that the supercritical CO2 could alter the wettability of coal
from intermediate-wet to CO2-wet.44,45 However, no clear
trend of contact angle with pressure was observed for the CO2/
water/Bentheimer sandstone system.40
Although the wettability of idealized quartz surfaces has been
experimentally investigated by several researchers,14,19,42,46 the
results partly contradict each other, particularly concerning the
eect of the transition of CO2 from subcritical to supercritical
and the inuence of pressure on wettability. These disagree-
ments are mainly due to dierences in experimental setups and
methods, the preparation and treatment of mineral plates, the
presence of contaminants, and mostly the equilibrium
condition of the aqueous phase with respect to the dissolution
of CO2. Moreover, to the best of our knowledge, there is only a
small amount of experimental data available for contact-angle
determination on natural rock samples39,40,4345 due to the
complexity of the experimental determination of the contact
angle on natural rock. In addition, there is limited information
on wettability in the presence of ue gas44,45 and hardly any
data are available on the eect of dissolved CO2 in the aqueous
phase on the wettability of sandstone reservoirs at high
pressures and elevated temperature. In our study, the contact-
angle determinations have been performed using natural rock
surfaces without any chemical treatment. The results of these
experiments contribute to a better understanding of wettability
and the displacement behavior of uids in sandstone reservoirs.
All saturation-related functions, such as relative permeability
and capillary-pressure functions and saturation distribution, are
directly inuenced by wettability of the matrix. Moreover, the
4005
Article
results of this work can be used in the capillary trapping
modeling as well as estimation of the CO2 storage capacity in
depleted gas reservoirs and saline aquifers.
3. EXPERIMENTAL SECTION
3.1. Materials. CO2 with a purity of 99.7 mol % and a synthetic
ue gas consisting of a mixture of 20 mol % CO2 and 80 mol % N2
were used (Linde Gas Benelux). Since the percentages of the other
components (i.e., NOx and SOx) usually found in industrial ue gas are
in the percentage to parts per million range,45 they were ignored in
order to establish a wettability behavior without complicating it by
reacting systems. The rock slabs were taken from a sawed Bentheimer
sandstone outcrop block from the Bentheim-area in Germany, which
was only exposed to air. The samples are mainly composed of quartz
(96%), which is homogeneously distributed throughout the rock
matrix (synthetic mineral reconstruction and X-ray diraction (XRD)
analysis of the samples are given in Appendix 1). The average porosity
and permeability of Bentheimer sandstone were about 20% and
around 23 darcies, respectively. The rectangular samples used in this
study were 30 6 12 mm3. In order to minimize the eect of surface
roughness on the contact angle, prior to the experiment one side of the
rock slabs was polished with a diamond abrasive of 10250 m.
Thereafter, the samples were washed by double distilled and
demineralized water. In the next step, the surface roughness was
determined using a LEICA 3-D stereo explorer.
To avoid the eect of surface contamination on the reliability of the
contact-angle measurement, all polishing and preparation steps were
performed without any chemical treatment.
To remove air bubbles from pore spaces, the samples were initially
evacuated. Afterward, to fully saturate the samples with water,
they were aged in evacuated double distilled and demineralized
water for 2 days.
3.2. Microscopic Image Analysis. Prior to the wettability
experiment, 2-D and 3-D microscopic images were taken from the
surface using a LEICA 3-D stereo explorer. From the 3-D images,
surface proles were obtained, which were used for the Pa factor
determination. The characterization of the surface roughness is based
on the Pa factor, which is calculated according to an internationally
recognized standard (EN ISO 4287), used to characterize the surface
proles.45 The Pa factor is the arithmetic mean of the absolute ordinate
values within a sampling length. Larger Pa values are associated with
rougher surfaces. For dierent Bentheimer sandstone samples, the
smoothest sample with the smallest Pa factor was selected for the
experiments to minimize the eect of roughness on the determination
of the contact angle. Three-dimensional images of two polished
samples with dierent Pa factors are shown in Figure 2.
After the experiment, the surface roughness of the sample was again
determined to investigate the degradation of the sample during the
experiment. The Pa factor for the smoothest Bentheimer sample was
0.032 mm before and 0.035 mm after the experiment. The minor
increase in surface roughness during the experiments could be within
the accuracy range of the experimental method to determine the
surface roughness or may have been the result of mineral reactions
with water and CO2. However, in comparison with samples of other
rock types, such as shale and coal, the increase in surface roughness
was negligible. Therefore, it is considered that the roughness of
Bentheimer sandstone remains constant during the experiments, even
at high pressures.
3.3. Experimental Setup. In this study, the captive-bubble
technique in a pendant-drop cell is used to capture CO2 and ue gas
bubbles under dierent natural sandstone surfaces at varying pressures
up to 15 MPa and a constant elevated temperature of 318 K.
A schematic drawing of the experimental setup is given in Figure 3.
The detailed description of the experimental setup can be found in
previous publications.40,43,45,47
3.4. Experimental Procedure. Before a series of experiments was
performed, all of the components of the setup were cleaned with
ethanol and distilled water to remove possible impurities. After
evacuation and a helium leakage test up to 18 MPa, the cell was lled
dx.doi.org/10.1021/ef500034j | Energy Fuels 2014, 28, 40024020
Energy & Fuels
Figure 2. Three-dimensional side view of the contact surface of polished Bentheimer sandstone samples with dierent roughness values: (a) ne-
polished sample with Pa factor of 0.032 mm and (b) less-polished sample with Pa factor of 0.059 mm. The pictures were taken with a LEICA 3-D
stereo explorer. The blue, red, and green lines give the orthogonal x, y, z directions used for the determination of factor Pa.
Figure 3. Schematic of the modied pendant-drop cell (PDC) experimental setup.
with double-distilled and demineralized water. The water level in the
cell needed to be higher than the lower end of the substrate. The
entire setup was placed in the constant temperature oven and
preheated to a temperature of 318 1 K, which was kept constant for
all of the experiments.
The amount of CO2 injected depended on which kind of aqueous
phase needed to be established for the experiment. For a wettability
study in an unsaturated aqueous phase, the initial aqueous phase was
kept pure (XCO2 = 0). Therefore, the cell was lled with water up to
the desired pressure, and a small amount of CO2 was then added to
the cell through a small needle at the bottom. For these experiments,
the changes in contact angle were mainly attributed to the CO2
dissolution rather than to the wetting properties of the Bentheimer
substrate.
To investigate the wetting properties of Bentheimer sandstone, the
experiments have to be performed in an aqueous phase that is fully
saturated with CO2. Under such conditions, the eects of CO2
dissolution and changes of the aqueous phase composition are
minimal during the experiment. To ensure complete saturation of the
aqueous phase, the mixture was circulated for at least 5 h. Thereafter,
the system was allowed to rest for at least 1 h. The pressure and the
temperature of the cell and the cabin, and the density of the mixture
were continuously monitored during the experiments.
The phase diagram of the CO2/water system in the region of high
mole fractions of water at 318 K is depicted in Figure 4. This diagram
was used to determine whether the system was within the one-phase
or the two-phase region at the given conditions. Once enough CO2
4006
Article
had been injected, circulation was initiated to ensure the mixing of
CO2 and water and to achieve faster equilibrium. In this experiment,
the overall composition of the mixture was determined through a
material balance using the mass of water and the number of CO2
moles determined by the SpanWagner equation of state (EOS).48
When the pressure in the cell and the density of the aqueous phase
reach a constant level, the system has reached equilibrium. In ue gas
experiments, the aqueous phase was initially presaturated with CO2 to
accelerate the equilibrium time. The detailed description of the
experimental procedure can be found in previous publications.40,43,45,47
After releasing a gas bubble from the tip and after a bubble had been
captured beneath the substrate, images of the bubble were taken.
These pictures were then processed in an image-analysis procedure
which allowed determination of the contact angle at the specic
pressure. Several pictures of the same bubble were taken in order to
verify the reproducibility of the determined contact angle. Contact
angles were determined using the KRU SS drop-shape-analysis
program (DSA4), analyzing the pendant-drop shape.45,50 Thereafter,
the pressure inside the cell was increased by adding more water (in
unsaturated case) or gas (in fully saturated case) to the cell.
The accuracy of contact-angle determination depends on several
parameters such as the accuracy of the instruments, the precision of
the image-analysis procedure, human error in measurements, and
alignment of the light source and endoscope with the sample. Because
of the complexity and qualitative nature of some of these parameters,
i.e., systematic and random errors, it is not straightforward to
formulate the error based on the mentioned parameters. To show the
dx.doi.org/10.1021/ef500034j | Energy Fuels 2014, 28, 40024020
Energy & Fuels
Figure 4. Phase diagram of the CO2/H2O system at a temperature of
318 K.49 In this gure, only the H2O-rich phase and part of the two-
phase region are given. The two-phase region is situated on the right
side of the curve, while the H2O-rich one-phase region is located on
the left side.
random error of measurements, the standard error is used, which is
calculated by
STDEV
SE =
n (4)
where SE is the standard error, STDEV is the standard deviation, and
n is the number of data points. More details about the accuracy of the
instruments and sensitivity of the contact-angle determination to the
image analysis can be found in Shojai Kaveh.51
4. RESULTS AND DISCUSSIONS
4.1. Contact-Angle Determination of CO2 and Flue
Gas in Pre-equilibrated Aqueous Phase. Two dierent
types of contact-angle data are reported in the literature,
namely, stable (steady) and dynamic (transient or non-
stable) contact angles.10,43,45 Variation of the contact angle
with time is generally regarded as indicating a nonstable or
dynamic contact angle. If the contact-angle experiment is
conducted within the CO2-saturated aqueous phase, stable or
steady contact angle will be measured. Variation of a stable
contact angle over time is mainly due to the surface heterogeneity
and/or roughness. In the unsaturated system, the change in
contact angle is inuenced by a number of mechanisms such as
dissolution and bubble size variation, rather than by the wetting
properties of the surface alone.
To determine whether the system is in the one-phase or two-
phase region (saturated or unsaturated with CO2), the phase
diagram of the CO2/water system is used from the literature
(Figure 4). If the amount of CO2 in the system, before releasing
a new bubble, is such that the composition describes a point on
the phase envelope (curve describing the border between the
one-phase and two-phase regions), then the system is referred
to as quasi-saturated. In such a system, by adding a CO2 bubble,
the pressure is initially slightly increased. Due to the fact that
the amount of CO2 which can be fully dissolved in the aqueous
phase increases with pressure, the addition of CO2 would mean
that the overall system composition moves slightly from the
phase envelope to the one-phase region (unsaturated case).
Therefore, one refers to this case as quasi-saturated.
Evolutions of the bubble radius and contact angle with time
at 0.58 MPa for dierent phase conditions are described in
Appendix 2. From the data at hand (Appendix 2), it can be
concluded that a reliable contact-angle determination should be
4007
Article
conducted in a pre-equilibrated (fully saturated or quasi-
saturated) aqueous phase to eliminate dissolution eects.
Dissolution aects the contact-angle determination due to the
constantly changing bubble size; this means that the system
continuously changes its conditions, thereby changing the
contact line and the contact angle.
Therefore, all contact-angle determinations in this work were
performed using a fully CO2-saturated aqueous phase to
evaluate the wetting properties of Bentheimer sandstone/
water/CO2 excluding dissolution eects. Figure 5a illustrates
the eect of pressure on the stable contact angle at a temperature
of 318 K. The values of the contact angle in Figure 5a are scattered
so that the derivation of a relation between pressure and contact
angle cannot be formulated in a straightforward manner. The
maximum variation of the contact angle of about 13 was observed
for experiments at 4.97 MPa. Another large variation of ca. 10
was found for the experiments at 1.04 MPa. There is extensive
experimental evidence in the literature conrming that the
contact angle of a captive bubble/sessile droplet on a horizontal
surface not only depends on the physical properties of the gas,
liquid, and solid phases but also varies with the bubble/droplet
size.2427 In Figure 5b, the stable contact angle is plotted as a
function of bubble radius. It can be seen that the contact
angle decreases with bubble size. Furthermore, since the bubble
size does not depend on the pressure (Figure 5c), it explains
the scattering of the contact angle as a function of pressure
(Figure 5a). Figure 6 shows the sequential digital images of the
CO2-rich bubble captured on the Bentheimer rock at 4.97 MPa
and 318 K. It can be seen that when the bubble size decreases,
the bubble shape asymptotically approaches a spherical shape.
To understand the observed contact-angle dependency on
the bubble size, dierent scenarios have been considered, i.e.,
surface roughness and hysteresis,24,28 gravity eects on droplet
geometry,30 and line tension26,29 (explanations in section 2.1.1).
Due to the contradictory results regarding the line-tension
determination in the literature and following on from the results
at 1.04 and 4.97 MPa, which are given in the Supporting
Information, the line-tension concept is not considered a proper
method to describe the dependency of the contact angle on the
bubble size for the system at hand, since surface nonideality and
roughness have a signicant inuence on the reliability of this
method.24
4.1.1. Surface Roughness. Surface roughness and/or
heterogeneity cause changes in contact angle by increasing
the chance of hysteresis.22 To minimize the eect of roughness
in this work, ne-polished samples were used. The surface rough-
ness was characterized using the so-called Pa factor which was
determined using the Leica 3-D stereo explorer (see section 3.2).
The results in Figure 9 were obtained by contact-angle deter-
mination on a sample with a Pa of 0.032 mm. However, to
identify the contribution of roughness on the contact-angle and
the bubble-size relation, the experiments were repeated using a
sample which was only roughly polished (Pa = 0.059 mm; see
Figure 2b).
Although the roughly polished sample is approximately two
times rougher than the ne-polished sample, it is still signicantly
smoother than the original rock surface (Pa = 0.272 mm). Parts a
and b of Figure 7 give the stable contact angles on the less-
polished surface as a function of pressure and bubble radius,
respectively. The range of the values of the contact angle is wider,
and contact-angle hysteresis increases: the contact angle scatters
when plotted as a function of pressure (Figure 7a). The larger
contact angle of the less-polished surface is attributed to the
dx.doi.org/10.1021/ef500034j | Energy Fuels 2014, 28, 40024020
Energy & Fuels Article

Figure 5. Stable contact angle (fully saturated system) as a function of (a) pressure and (b) the bubble radius at the apex at a temperature of 318 K.
(c) Bubble radius at the apex as a function of pressure at a temperature of 318 K. The error bars show the standard error of the values.

Figure 6. Sequential digital images of the captive CO2-rich bubble on the Bentheimer rock at 4.97 MPa and 318 K: (a) R0 = 1.90 mm, Vb = 32 L,
and = 3.1; (b) R0 = 1.73 mm, Vb = 23.56 L, and = 4.5; (c) R0 = 1.55 mm, Vb = 16.52 L, and = 6.5; (d) R0 = 1.34 mm, Vb = 11.97 L, and
= 11.3; (e) R0 = 1.08 mm, Vb = 9.09 L, and = 13.6; (f) R0 = 0.77 mm, Vb = 1.93 L, and = 22.8. Each picture was taken 30 s after the
release of the bubble. When the bubble size reduces, the geometry of the bubble approaches a spherical shape.

smaller size of the bubble in this system. However, the contact
angle of the less-polished surface (Figure 7b) behaves similarly to
the contact angle on the smooth, polished surface (Figure 5b)
even though the square of the correlation coecient for the
linear regression, r2, increases from 0.67 for the ne-polished
substrate (Figure 5b) to 0.86 for the less-polished substrate
(Figure 7b). The comparison of the data allows the conclusion
that although surface roughness aects the contact angle and its
variation with the bubble size, however, it is not a major
4008
controlling factor on the dependency of the contact angle on
bubble size in the system at hand.
4.1.2. Eect of Gravity on Bubble Geometry. In this section,
the contact-angle variation with bubble size is evaluated while
the eect of gravity on the bubble shape is taken into
account.30,31,52,53 As a reference, the analytical model of Vafaei
and Podowski (eq 2) was used, which allows the calculation of
liquid droplet volume from a given contact angle (and radius of
the contact area). This model has been adapted for a captive
dx.doi.org/10.1021/ef500034j | Energy Fuels 2014, 28, 40024020
Energy & Fuels Article

Figure 7. Stable contact angles for a less-polished surface as a function of (a) pressure and (b) bubble radius at the apex at various pressures and a
constant temperature of 318 K. The dashed line is the best linear t to the experimental data.

Figure 8. Bubble volume as a function of the contact angle at a constant temperature of 318 K and pressures of (a) 1.04 and (b) 4.97 MPa. Red
triangles show the results obtained with the analytical model (eq A3.8), and blue squares are determined from the images of the bubble.

gas bubble system by considering the eect of buoyancy force
on the bubble contour (Appendix 3). Parts a and b of Figure 8
compare the bubble volumes calculated from the analytical
expression (derived in Appendix 3) against the volumes computed
from the experimentally captured images, at pressures of 1.04 and
4.97 MPa. The experimentally determined volumes (blue squares
in Figure 8) were obtained from the numerical solution and
consecutively integrating the bubble prole based on the Young
Laplace description of the bubble image (KRUESS drop-shape
analysis50). The bubble radius at apex (R0), obtained from the
numerical solution of the bubble prole, along with the values
of the contact angle () and contact radius (R), were used to
calculate the bubble volume in analytical expression, eq A3.8
(red triangle in Figure 8). and R were obtained from the
intersection of a tangent line on the bubble prole and the
baseline describing the surface. As can be seen from Figure 8,
the results of experimentally and analytically determined
volumes are in good agreement with each other, conrming
that the contact-angle variation with the bubble volume can also
be described with the analytical expression. There are slight
dierences between the experimentally determined volumes
and the results of the analytical model for the large bubbles
with small contact angles at 1.04 MPa. This dierence might
be attributed to (1) the uncertainties in the contact-angle
determination of a strongly water-wet system with very small
4009
contact angles, which directly inuence the bubble volume
calculations using an analytical model, and/or (2) deviation of
the actual bubble curvature from a circle at the horizontal cross-
sections for large bubbles (Appendix 3).
Based on the discussion with respect to the eect of gravity
(buoyancy) on the bubble shape (section 2.1.1), it is expected
that the bubble shape asymptotically approaches a spherical
shape and, accordingly, the contact angle no longer depends on
the bubble volume. This can also be seen in Figure 6af, where
the variation of the contact angle with the bubble volume is
given at a pressure of 4.97 MPa. These sequential digital images
of the captive-CO2 bubble show that when the bubble size
decreases, the contact angle increases and the bubble shape
approaches a sphere. However, in order to identify whether
surface forces or gravity (buoyancy) forces dominate, the Bond
number (eq 3) is calculated and used for the presentation of the
contact angle. In Figure 9a,b, the stable contact angles are given
as a function of the Bond number for various bubble sizes at
pressures of 1.04 and 4.97 MPa. The experimentally determined
IFT data47 were used for calculation of the Bond numbers
(eq 3). The density of the aqueous phase was determined with
a vibrating-tube density meter (see schematic of the setup,
Figure 3). The SpanWagner EOS was used to calculate the
density of the CO2-rich bubble assuming that the water
concentration in the bubble is negligible. With this, the density
dx.doi.org/10.1021/ef500034j | Energy Fuels 2014, 28, 40024020
Energy & Fuels
Figure 9. (a) Stable contact angle as a function of the Bond number for various bubble sizes, and (b) Bond number as a function of the bubble
radius, at a constant temperature of 318 K and two pressures of 1.04 MPa (blue squares) and 4.97 MPa (red circles).
dierence between the aqueous phase and CO2 bubble () at
pressures of 1.04 and 4.97 MPa were calculated to be 1.046 and
0.957 g/mL, respectively. Following this, Bond numbers were
calculated to range between 0.3 and 3.2.
Theoretically, Bond numbers smaller than one indicate that
surface forces are dominant rather than gravity forces.
Combining this knowledge with the fact that a decreasing
bubble size results in a bubble shape approaching a sphere, it
would be expected that the contact angle does not change with
bubble size for small Bond numbers (smaller than one). In
Figure 9a, the contact angle as a function of the Bond number
is given at two dierent pressures. When Bond numbers are
larger than one, the contact angle signicantly changes with the
Bond number, which means that the gravity (buoyancy) force is
more dominant than the surface force. However, this variation
decreases for Bond numbers less than 0.9. Based on the data
from Figure 9a, pressure has no eect on the contact angle of
the bubbles with a Bond number larger than 1. However, the
contact angle of the bubbles with a Bond number less than 0.9
increases by about 5 when the pressure is increased from 1.04
to 4.97 MPa. Figure 9b shows Bond number values as a
function of the bubble radius at pressures of 1.04 and 4.97 MPa.
For bubbles with a similar radius, the Bond number at a
pressure of 4.97 MPa is larger than that at 1.04 MPa. Although
the density dierence of two coexisting phases decreases with
increasing pressure, the inuence of interfacial tension
reduction is more highlighted; this resulted in the larger
Bond number at 4.97 MPa. The variation of Bond number with
bubble radius conrms that the inuence of the bubble radius
becomes less signicant when the Bond number reduces.
According to the results presented in Figure 9 and section
4.1.1, it can be concluded that the dependency of the contact
angle on the bubble size in the system at hand can be mainly
explained by the eect of gravity (buoyancy) on bubble shape.
However, the surface nonideality and roughness have
signicant inuence on the reliability of the contact-angle deter-
mination. This variation becomes insignicant for Bond numbers
less than 0.9 (bubble maximum diameter of about 2.3 mm).
The results of this study conrm previous ndings by Vafaei
and Podowski30,31 that axisymmetric droplets are size-depend-
ent and cannot satisfy the original Young equation. In addition,
it is in a good agreement with results of Sakai and Fujii,38 which
shows that the apparent contact angle on rough surfaces can be
changed by gravity. According to the results of this study, it can
be concluded that in order to avoid the dependency of the
4010
Article
contact angle on the bubble size in these systems, the eect of
gravity (buoyancy) on bubble shape has to be considered by
calculation of the Bond number.
4.1.3. Eect of Pressure on the Contact Angle in the
Bentheimer/CO2/Water System. In the CCS eld application,
the amount of capillary-trapped CO2 depends on the wettability
of reservoir rocks at reservoir conditions,54 i.e., high pressures.
To identify the eect of pressure on the contact angle, only
those bubbles with a Bond number smaller than 0.9 were
considered (see section 4.1.2). In this way, the eect of gravity
could be excluded and the inuence of the bubble size on
the contact angle became insignicant. Figure 10 represents
the contact angle as a function of pressure for experiments
which were characterized by Bond numbers of approximately
0.9 0.02.
Figure 10 shows that the contact angle of the Bentheimer/
CO2/water system slightly increases with pressure. This slight
rise (maximum of 5) might not be correlated to the eect of
pressure, since it is within the experimental error range. This
assumption is supported by the results of Espinoza and
Santamarina,19 who observed that pressure does not have a
signicant eect of on the wettability of quartz surfaces.
However, several studies have reported the wettability
alteration of quartz surfaces from strongly to less water-wet
in contact with supercritical CO2.12,14,41,42,55 This inconsistency
originates from dierences in the surface charges of quartz and
Bentheimer sandstone. Although Bentheimer sandstone is mainly
composed of quartz, even a low content of clay (i.e., 2.5%
Kaolinite and so on) plays an important role in the surface charge
of particles.56
For the quartz surface, an increase in contact angle with
pressure has been attributed to the decreased charge of the
silica surface under the acidic pH of the CO2-equilibrated
aqueous phase.12,55,57 The pH of the aqueous phase inuences
the surface charge of the particles. According to Tokunaga,57
the surface charge of a quartz particle is negative and
approaches zero for a pH of 3. In water/CO2 solutions, the
pH decreases from 7 to 3 due to the dissolution of CO2 in the
aqueous phase.12 The pH reduction results in a decline in the
negative surface charge and, thus, reduces the electrostatic
repulsions. Consequently, the stability of the water lm
covering the surface decreases, which leads to a wettability
alteration and thus an increase in contact angle (electric double
layer eects, DerjaguinLandauVerweyOverbeek (DLVO)
theory).57
dx.doi.org/10.1021/ef500034j | Energy Fuels 2014, 28, 40024020
Energy & Fuels
Figure 10. Contact angle as a function of pressure for bubbles with Bond numbers of about 0.9 0.02 for the Bentheimer/CO2/water system
at 318 K.
Unlike quartz, the surface charge of Bentheimer sandstone
approaches zero at a pH of 8.56,58 At low pH, the charge of a
Bentheimer sandstone surface becomes more positive so that
the stability of the water lm increases.56 This means that the
pressure induced by the injection of a CO2 bubble has no eect
on the stability of the water lm and thus the wettability of the
Bentheimer sandstone (as seen in Figure 14). The sequential
digital images in Figure 14 show the CO2 bubble with similar
sizes on the Bentheimer rock at subcritical (2.83 MPa),
approximately critical (7.51 MPa), and supercritical (12.74 MPa)
pressures. The comparison of the data allows the conclusion that
CO2 phase transition has no eect on the wettability of the
Bentheimer sandstone/water system.
4.1.4. Eect of Pressure on the Contact Angle in
Bentheimer/Flue Gas/Water System. The stable contact angles
of the ue gas/Bentheimer/water system at a temperature of
318 K are presented in Figure 11a,b as a function of pressure
and ue gas bubble radius. As in the CO2 system (Figure 5a,b),
the contact angles in the ue gas/sandstone/water system are
inuenced more by bubble size than by pressure. However, the
contact angle values of ue gas bubbles are all smaller than
those of the CO2 bubbles (see Figures 5a and 11a). This can be
explained by the ratio of 20 mol % CO2 to 80 mol % N2 in the
ue gas bubble and the higher IFT values in ue gas system.47
To evaluate the eect of pressure on contact angle, the contact
angle data at four specic pressures were extracted from Figure
11b and shown in Figure 12. The comparison between data
reveals that the contact angle remains relatively unchanged,
within the ranges of the error bars, with pressure variations
(Figure 12). Thus, it can be concluded that pressure has a
negligible eect on the wettability of the ue gas/water/
Bentheimer sandstone system.
4011
Article
4.2. Contact-Angle Determination of CO2 in Unsatu-
rated Aqueous Phase. In the case where the aqueous phase
is not fully saturated with CO2, the change in contact angle is
mainly related to the dissolution of CO2 in the aqueous phase
or water in the CO2-rich phase.39,45 This class of data provides
information on the interfacial interactions between the present
phases, and, more specically, mass transfer between the liquid
and the gas phases. In the unsaturated system, the change in
contact angle is inuenced by a number of mechanisms such as
dissolution and bubble-size variation, rather than by the wetting
properties of the surface alone.
Figure 13 shows the CO2 bubble radius at the apex and the
dynamic contact angles as a function of time for dierent
pressures. In the unsaturated system, the injected CO2 bubble
completely dissolves in the aqueous phase and disappears. In
Figure 13a, it can be seen that the bubble dissolution time
depends on the initial radius of the bubble. However,
comparison between experimental dissolution curves at 4.9
and 6.7 MPa reveals that if the initial bubble size is the same,
pressure has no signicant eect on the bubble dissolution
behavior. This also conrms that the diusion coecient in the
liquid phase is independent of pressure.59 The change in the
contact angle and droplet size over time (Figure 13) can be
divided into two regimes: the rst regime (regime I) where a
CO2-rich bubble dissolves quickly in the aqueous phase and the
second regime (regime II) where CO2 gradually diuses into
the aqueous phase. For example, at 0.58 MPa after 180 s, 94%
of the initial bubble volume dissolves in the fresh aqueous
phase (Figure 14), while dissolution of the rest of the bubble
(6% of the bubble initial volume) takes 2300 s. This dissolution
trend over time at 0.58 MPa is shown in the sequential digital
images of the captive-CO2 bubble in Figure 14. In the rst
dx.doi.org/10.1021/ef500034j | Energy Fuels 2014, 28, 40024020
Energy & Fuels
Figure 11. Flue gas stable contact angle (in pre-CO2-saturated aqueous
phase) as a function of (a) pressure and (b) bubble radius at a temperature
of 318 K. The dashed line is the best linear t to the experimental data.
The error bars show the standard error of the values.
Figure 12. Flue gas stable contact angle vs bubble radius at four specic
pressures and a temperature of 318 K. The dashed line is the best linear
t to the experimental data.
regime, the contact angle increases continuously with time until
it reaches a transition point. After this point and in regime II,
the contact angle changes only slightly with time. The
transition between fast and slow dissolution regimes is clearly
visible, which allows the denition of a transition point, a
transition time, and a transition radius. The so-called transition
point is the crossing point of the tangents to the curves in
regimes I and II which can be determined experimentally
4012
Article
Figure 13. (a) Bubble radius at the apex and (b) dynamic contact
angle over time for dierent pressures and at a constant temperature of
318 K.
according to the evolution trend of the bubble size and contact
angle with time at each specic pressure. According to Figure 13,
the transition from a strong change to an almost constant value
(regime I to regime II) occurs at the same time for both contact
angle and bubble radius, which can conrm the hypothesis that
the contact angle depends on the bubble size.
The change in the contact angle over time cannot be related
to the wettability alteration of the surface from strongly to less
water-wet. The wettability behavior of a surface is a material
property. This means that the wettability does not change due
to bubble shrinkage. The increase of contact angle with time, as
seen in Figure 13b, can be explained by bubble volume
reduction (shrinkage) due to CO2 dissolution in the aqueous
phase. Because of the dependency of the contact angle on the
bubble size (section 4.1), the bubble shrinkage has an eect on
the change in the contact angles with time.
Table 1 summarizes the eect of pressure and initial bubble
size on the contact angle and transition time. In this table, min
is the contact angle of the bubble with the initial bubble radius
of Rio and max is the last accurate detectable contact angle. For
bubbles with a volume of less than 0.2 mm3, the contact angle
determination is not reliable due to the limited resolution and,
thus, the high uncertainty in the bubble prole description. The
data in Table 1 show that the transition time and radius at the
transition zone for all experiments are in the range between 3
and 5 min and between 0.37 mm and 0.53 mm, respectively,
regardless of the initial bubble size and pressure. Unfortunately,
from the data at hand, the dependency of the contact angle at
dx.doi.org/10.1021/ef500034j | Energy Fuels 2014, 28, 40024020
Energy & Fuels Article

Figure 14. Sequential digital images of the captive CO2-rich bubble on a Bentheimer surface at 0.58 MPa and 318 K: (a) t = 1 s, R0 = 1.30 mm, Vb =
7.71 L, and = 16.3; (b) t = 15 s, R0 = 1.18 mm, Vb = 5.99 L, and = 17; (c) t = 47 s, R0 = 0.97 mm, Vb = 3.36 L, and = 22; (d) t = 89 s,
R0 = 0.71 mm, Vb = 1.37 L, and = 24.3; (e) t = 136 s, R0 = 0.54 mm, Vb = 0.621 L, and = 28.9; (f) t = 180 s, R0 = 0.48 mm, Vb = 0.46 L, and
= 30.4; (g) t = 307 s, R0 = 0.44 mm, Vb = 0.35 L, and = 32.5; (h) t = 1007 s, R0 = 0.35 mm, Vb = 0.182 L, and = 34.2; (i) t = 1870 s, R0 =
0.24 mm, Vb 0.053 L, and = n.a. (due to the resolution limitation).

Table 1. Contact Angle and Other Parameters of CO2 Gas Bubble at Various Pressures and a Constant Temperature of 318 K
pressure Rio bubble min max transition time radius at transition point initial CO2 bubble mass CO2 solubility in water
(MPa) (mm) (deg) (deg) (s) (mm) (milligram)a (mol/kg)b
0.58 1.300 16.3 34.6 180 0.47 0.060 0.1219
2.7 1.765 14.4 47.8 200 0.50 1.018 0.5117
4.9 1.411 19.5 52.5 240 0.41 1.082 0.8193
6.7 1.424 15.4 59.8 250 0.53 1.783 1.0052
10.2 0.573 24.36 64.53 180 0.37 0.361 1.1923
12.0 0.850 20.85 66.37 320 0.47 1.472 1.2199
a
mb in milligram = CO2Vb. bDuan et al.60

the initial bubble radius, min, on the pressure cannot be
deducted because the initial bubble sizes are not the same at
dierent pressures.
To better understand the origin of the two regimes observed
at various pressures (Figure 13), diusion of the CO2 bubble
into the aqueous phase is described using Ficks rst law. The
equations and their derivations are given in Appendix 4. Figure 15
shows the curves of the bubble radius as a function of time at
0.58 MPa and a temperature of 318 K. Only the curves at a
pressure of 0.58 MPa are presented here, as the general
behavior is the same at dierent pressures. The blue triangles
display the experimental values, and the red solid line is the
description with Ficks law using a diusion coecient of
3.07 109 m2/s at 318 K.61 According to Figure 15, Ficks law
cannot represent the experimental data by only considering
molecular diusion. The experimental data of the bubble radius
in regime I decrease much faster than predicted by the model.
This is an indication of an accelerated mass transfer of CO2 into
the water in regime I, which cannot simply be explained by the
diusion process.
4013
It is assumed that the CO2 concentration in the bulk is zero
in regime I and the concentration of CO2 at the bubble
interface, i.e., the solubility of CO2 in water, is taken from the
literature60 at the specic pressure.
In order to explain the accelerated mass transfer observed in
regime I, dierent mechanisms have been considered, i.e.,
spontaneous imbibition of CO2 into the rock pores, density-
driven natural convection,9,62,63 the eect of surface contam-
ination (stagnant-cap model),64 the relative motion between
the bubble and the aqueous phase when a bubble rises through
the aqueous phase (internal circulation theory),59 and
Marangoni eect.9,62
CO2, as the least wetting phase, cannot overcome the
capillary pressure and consequently penetrate into the pore
spaces,40 at least for this bubble size. Therefore, accelerated
mass transfer, observed in regime I, cannot be originated by the
spontaneous imbibition of CO2 into the pores.
The onset of natural convection is determined by the
Rayleigh number (Ra) and density-driven natural convection
occurs in bulk solutions when Ra > 2100. The calculation of
dx.doi.org/10.1021/ef500034j | Energy Fuels 2014, 28, 40024020
Energy & Fuels
Figure 15. Bubble radius as a function of time at 0.58 MPa and a
constant temperature of 318 K. Blue triangles display experimental
data, and the red line gives the description with Ficks law (see also
Appendix 4) assuming a bulk phase of pure water (no CO2 initially
dissolved in the aqueous phase) and using a molecular diusion
coecient of CO2 in water, D, at 318 K from ref 61.
this dimensionless number for the system at hand (Appendix 5)
reveals that diusive transport is accelerated by density-driven
natural convection in regime I. Considering the dimensions and
conditions of the system at hand, it is expected that the entire
bubble dissolves quickly by natural convection since Ra > 2100.
However, after a certain time, known as the transition time,
dissolution becomes less convection-dominated and more
diusion-dominated. The density dierence in regime II is
smaller, but according to this criterion (Ra > 2100), natural
convection should occur in regime II as well. Nonetheless, the
results indicate that the rate of mass transfer is what one would
expect for diusion with no convection. The reason for this
sudden change observed in the mass-transfer rate is not clear
yet. It might be attributed to the formation of a CO2-saturated
water lm around the bubble due to the capillary condensation,
which reduces the driving force (concentration gradient) and
the mass-transfer rate at the interface. This phenomenon
cannot be studied by available experimental data from this
study and needs to be investigated further.
Based on the slow dissolution rate of regime II, this regime is
mainly considered to be a diusion-dominant process. Due to
the formation of a CO2-saturated water lm, the dissolution of
the bubble in this regime takes longer than the dissolution of
the bubble in the pure water phase. These ndings are in good
agreement with the experimental results of Farajzadeh.63
5. CONCLUSION
In this work, wettability of the Bentheimer sandstone/water/
CO2 and/or ue gas system has been experimentally evaluated
using the captive-bubble technique in a pressure range from 0.2
to 15 MPa (from 2 to 150 bar). The CO2 experiments were
conducted using dierent water/CO2 mixtures, i.e., unsaturated
and fully saturated, in order to evaluate the dissolution eects
and wetting properties for short and long periods. All contact-
angle determinations were performed using natural rock
surfaces which had not been treated chemically. The results
of these experiments do contribute to a better understanding of
wettability and the displacement behavior of uids in sandstone
reservoirs. The results are summarized as follows:
(a) A reliable contact angle determination should be conducted
using a pre-equilibrated (fully saturated or quasi-saturated)
4014
Article
aqueous phase to eliminate dissolution eects. Due to dissolution,
the bubble size continuously changes, so that a reliable and
reproducible contact-angle determination is not guaranteed.
(b) Experiments in the fully saturated system show that
Bentheimer sandstone is (and remains) water-wet against either
CO2 or ue gas, even at supercritical pressures. However, all
contact angles of the ue gas bubbles are smaller than those of
CO2. The data for the apparent contact angle of these systems
show a strong dependence on the bubble size.
(c) Analysis of the experimental data shows that although the
surface nonideality and roughness inuence on the reliability of
the contact-angle values and its variation with the bubble size,
however, it is not a major controlling factor on this variation in
the system at hand. This dependency can be mainly explained
by the eect of gravity (buoyancy) on bubble shape. The
inuence of the bubble radius on the contact angle of the CO2
system becomes insignicant for bubble diameters smaller than
2.3 mm (Bond numbers less than 0.9).
(d) The results of this study conrm previous ndings by
Vafaei and Podowski30,31 that axisymmetric droplets are size-
dependent and cannot satisfy the original Young equation. In
addition, it is in a good agreement with results of Sakai and
Fujii,38 which shows that the apparent contact angle on rough
surfaces can be changed by gravity.
(e) According to the results of this study, it can be concluded
that in order to avoid the dependency of the contact angle
on the bubble size in these systems, the eect of gravity
(buoyancy) on bubble shape has to be considered by calculation
of the Bond number.
(f) To identify the eect of pressure, only results with Bond
numbers smaller than 0.9 (negligible dependency to the bubble
size) were considered. From this, it is deduced that there is
no signicant eect of pressure on the contact angle of the
Bentheimer/CO2/water system. This inconsistency between
the results of Bentheimer and quartz originates from dierences
in surface charges of Bentheimer and quartz. Even if Bentheimer
sandstone is mainly composed of quartz, a low content of clay
(i.e., 2.5% Kaolinite and so on) changes the surface charge of
particles considerably.
(g) Results obtained from unsaturated aqueous phase
experiments provide important information on the interfacial
interactions and mass transfer between the aqueous and CO2
phases. In the unsaturated system, the change in contact angle
is inuenced by a number of mechanisms such as dissolution
and bubble-size variation, rather than by the wetting properties
of the surface alone.
(h) For experiments in the unsaturated aqueous phase, two
dissolution regimes were observed. Analysis of the data showed
that this behavior cannot be caused by wettability alterations of
the surface from strongly to less water-wet, but by bubble
volume reduction due to CO2 dissolution in the aqueous phase.
(i) In the CCS eld application, the amount of capillary-
trapped CO2 depends on the wettability of reservoir rocks.
Results of this study show that there is no signicant eect of
pressure on the wettability of the Bentheimer/water/CO2 or
ue gas system. Even at high pressure, the Bentheimer
sandstone remains water-wet to either CO2 or ue gas.
APPENDIX 1: SYNTHETIC MINERAL
RECONSTRUCTION AND XRD ANALYSIS OF
BENTHEIMER SANDSTONE
Tables 2 and 3 list synthetic mineral reconstruction and XRD
analysis of Bentheimer sandstone, respectively.
dx.doi.org/10.1021/ef500034j | Energy Fuels 2014, 28, 40024020
Energy & Fuels Article

Table 2. Synthetic Mineral Reconstruction of Bentheimer bubble with time (Figure 16c). After introducing the CO2
Sandstone bubble, mass transfer occurs between the bubble and the
aqueous phase. The driving force for this process is the
mineral concn wt (%)
dierence in chemical potentials of each specic component in
quartz 91.70 the two coexisting phases. The time required to reach a stable
kaolinite 2.50 bubble size depends on the chemical potential gradients in the
montmorillonite 0.18 two phases. Since adding a CO2 bubble to the equilibrated
orthoclase 4.86 system only slightly disturbs the equilibrium, the mass transfer
dolomite 0.26 between the bubble and the aqueous phase is negligible and
calcite 0.15 does not aect the contact angle so that the wetting properties
hematite 0.16 of the substrate are determined. For such a system, any
rutile 0.03 variation of the contact angle with time can be attributed to
pyrite 0.01 surface heterogeneity and roughness.
calcium phosphate 0.07 According to Figure 16, the contact angle variation is
halite (NaCl) 0.03 proportional to the change in bubble volume which shows the
dependency of the contact angle on the bubble size in the
Table 3. XRD Analysis of Bentheimer Sandstone system at hand. These observations are in agreement with the
compd name concn wt (%) absolute error (%) results discussed in section 4.2.
Al2O3 1.931 0.04
In Table 4, the basic parameters allowing the direct
CaO 0.208 0.01
comparison of experiments in the unsaturated, quasi-saturated,
Cl 0.020 0.004
and fully saturated system at 0.58 MPa are given. min is the
Co3O4 0.001 0.0008
contact angle of the initial bubble with the radius of Rio. In the
Cr2O3 0.005 0.002
unsaturated system, the CO2 bubble disappeared after 2700 s.
Fe2O3 0.172 0.01
In the quasi-saturated system the bubble rst shrank and then
K2O 0.827 0.03
became stable after 4800 s. In the fully saturated system, the
MgO 0.064 0.008 stable bubble was obtained after 180 s. In the two latter
Na2O 0.022 0.004 systems, Rfinal and final are the bubble radius and the contact
NiO 0.002 0.001 angle after establishing the equilibrated bubble.
P2O5 0.030 0.005 From the data at hand, it can be concluded that a reliable
PbO 0.003 0.002 contact-angle determination should be conducted in a pre-
SO3 0.020 0.004 equilibrated (fully saturated or quasi-saturated) aqueous phase
SiO2 96.616 to eliminate dissolution eects. Dissolution aects the contact-
SrO 0.002 0.001 angle determination due to the constantly changing bubble size;
TiO2 0.072 0.008 this means that the system continuously changes its conditions,
thereby changing the contact line and the contact angle.

ZnO 0.002 0.001
ZrO2 0.004 0.002
APPENDIX 3: THEORETICAL ANALYSIS OF GAS

APPENDIX 2: EVOLUTION OF THE BUBBLE RADIUS

AND CONTACT ANGLE WITH TIME FOR
DIFFERENT PHASE CONDITIONS
To determine whether the system is in the one-phase or two-
phase region, the phase diagram of the CO2/water system is
used from the literature (Figure 4). Figure 16 shows the
evolution of the bubble radius and contact angle with time at
0.58 MPa for dierent phase conditions and overall
compositions of the CO2/water system. The data show that,
for pre-equilibrated systems, the bubble dissolution rate is
smaller than the dissolution rate in the unsaturated system. For
an unsaturated system (one-phase region), the amount of CO2
in the system is smaller than the maximum amount which can
be dissolved in the aqueous phase at a given temperature and
pressure. Hence, when adding a CO2 bubble, the bubble
dissolves completely in the aqueous phase (Figure 16a).
In the quasi-saturated system, the bubble initially dissolves
slowly and the bubble size decreases. When equilibrium is
established, the bubble size remains constant (Figure 16b). The
time to reach equilibrium depends on the initial composition of
CO2 in the water-rich phase.
In the fully saturated system, equilibrium is already
established when injecting CO2 into the system. Still, releasing
CO2 in the equilibrated system causes a slight disturbance,
which can be recognized by an initial volume reduction of the
CAPTIVE BUBBLE ON A HORIZONTAL SUBSTRATE
The following theoretical analysis and derivations are based on
the analytical model of Vafaei and Podowski,31 which allows the
calculation of liquid droplet volume from a given contact angle
(and radius of the contact area). In their approach, the droplet
is circular in all its horizontal cross-sections. This assumption
does not apply to vertical sections. This model has been
adapted for a captive gas bubble system surrounded by the
aqueous phase by considering the eect of buoyancy force on
the bubble contour. Forces in the z-direction acting on an
axially symmetric captive bubble, as shown in Figure 17, should
be considered.
The force balance for a slice between z and z + dz is
dFg dFb + Fp(z) Fp(z + dz) F sin (z)
+ F sin( + d )(z + dz) 2rPL dr = 0 (A3.1)
where the individual forces are
gravity force:
dFg = g gr 2(z) dz (A3.2)
buoyancy force:
dFb = l gr 2(z) dz (A3.3)

4015 dx.doi.org/10.1021/ef500034j | Energy Fuels 2014, 28, 40024020

Energy & Fuels Article

Figure 16. Evolution of bubble radius (shown as blue triangles at right y-axis) and contact angle (shown as red squares at left y-axis) over time at
0.58 MPa and 318 K in (a) unsaturated, (b) quasi-saturated, and (c) fully saturated aqueous phase. The error bars in these panels are given based on
the standard error of the values.

Table 4. Parameters Characterizing the Experiments at surface tension force:

0.58 MPa in the Unsaturated, Quasi-saturated, and Fully F (z) = 2r(z)lv (A3.5)
Saturated System
R0 0 Rfinal final where the vertical pressure distribution is given by
aqueous-phase condition (mm) (deg) (mm) (deg) time (s)
2lv
unsaturated (fresh water) 1.18 17 0 35.1 2700 p(z) = g gz + PL +
quasi-saturated with CO2 1.02 17.3 0.62 26.06 4800 R0 (A3.6)
fully saturated with CO2 1.23 16.7 1.21 20.4 180
where PL is the pressure determined in the aqueous phase; lv is
the liquidvapor interfacial tension at pressure PL; R0 is the
bubble radius at apex, and g is the density of the bubble.
By accounting for the individual forces, eq A3.1 gives

r 2(z)(l g )g
d dr
= [r 2(z) p(z) 2r(z)lv sin (z)] 2rPL
dz dz
(A3.7)
Integrating both sides of eq A3.7 yields
2lvR2 1 sin
Vb = 0 r 2 dz = +
g R 0

R

(A3.8)

Figure 17. Force balance in the z-direction for a captive bubble.

where is the location of the apex, is the density dierences
between the bubble and aqueous phase, and R is the radius of
the contact circle. Data of the contact angle (), the radius of
pressure force: the curvature at the apex (R0), and the radius of the contact
circle (R) are obtained using the numerical method from the
Fp(z) = r 2(z) p(z) (A3.4) YoungLaplace description of the bubble prole. and lv
4016 dx.doi.org/10.1021/ef500034j | Energy Fuels 2014, 28, 40024020
Energy & Fuels Article

are determined experimentally at the specic pressure and b

CCO2 = CCo 2
at r (A4.6)
temperature.

r = R io at t = 0 (A4.7)
APPENDIX 4: STEADY-STATE DIFFUSION By assuming a spherical shape for the bubble (truncated by the
THROUGH THE BUBBLE VARIABLE AREA solid surface),
In this section we follow the derivation for steady-state
diusion through the spherical bubble variable area into an Vb = rb3[2 + 3 cos s cos3 s]
3 (A4.8)
innite body of liquid65 and then scale the mass ux by the
fraction of the full spherical surface represented by the CO2 Ab = 2rb 2(1 + cos s) (A4.9)
bubble against the solid surface. Diusion of CO2 from a CO2-
rich bubble into the aqueous phase is described using Ficks Integrating eq A4.4 with boundary conditions (eq A4.5 and
rst law.65 A mass balance is written at a thin spherical shell A4.6) yields
with the thickness of r around the bubble (Figure 18). Since s b
W = 2 (1 + cos s)D(CCo 2
CCo )r
2 b (A4.10)
W is the constant molar rate of mass transfer (molar ux area,
eq A4.4) and is equal to the rate of dissolution of the bubble at
any instant:
d mb d
W= = (CO Vb)
dt dt 2 (A4.11)
Combining eqs A4.8, A4.10, and A4.11 gives
drb
[2 + 3 cos s cos3 s]CO rb 2
2 dt
s b
= 2 (1 + cos s)D(CCO2
CCO )r
2 b (A4.12)
Figure 18. Mass balance in the r-direction for a captive bubble through Integrating eq A4.12 with initial condition (eq A4.7) yields
the variable area.
s
(CCO CCOb
)D 2(1 + cos s)
2
rb = R io 2
2 2 2
3
t
the bubble dissolves in the liquid phase, the diusion area CO
2
(2 + 3 cos s cos s)
cannot be considered to be constant and changes along the (A4.13)
diusion direction (r) and also with time.65 In this case, at any
instant in time, t, the CO2 bubble radius is rb. It is assumed that the CO2 concentration in the bulk (CbCO2)
A steady-state mass balance (zero accumulation rate) on the far from the bubble is zero. In this model, the bubble geometry
spherical shell (r) gives65 is assumed to be spherical. This may have caused an error in
calculations, in particular for larger bubbles. The average
d deviation for the volume of bubbles compared to spheres, at
(AbNCO2) = 0
dr (A4.1) 0.58 MPa, was 5.31%.
which yields
AbNCO2 = constant = W (A4.2)
APPENDIX 5: RAYLEIGH NUMBER CALCULATION
FOR A CAPTIVE-BUBBLE SYSTEM
where NCO2 is the molar ux of CO2, Ab is the gasliquid The onset of natural convection in the porous medium is
interfacial area, and W is the constant molar rate of mass determined by the Rayleigh number (Ra), which is dependent
transfer. In this case, it is assumed that water diusion into the on the properties of the uids and geometry of the system:
CO2-rich phase (bubble) is negligible because of the small k gL
diusion coecient of water into CO2. Therefore, the system is Ra =
D (A5.1)
considered as diusion of CO2 through nondiusing water.
According to Ficks law where k is the permeability of the porous medium, is the
density dierence between the boundary layer uid and that far
dCCO2
NCO2 = D away, g is the local gravitational acceleration, L is the
dr (A4.3) characteristic length scale of convection, is the porosity,
where D is the CO2 diusion coecient into water. It should be is the dynamic viscosity, and D is the diusivity of the
noted that D depends on temperature but not greatly on characteristic that is causing the convection. In porous media
pressure (D = 3.07 109 m2/s at 318 K).59 the interfaces will be unstable for Rayleigh numbers above
Combining eq A4.2 and A4.3 yields 42 40.66
For bulk solutions eq A5.1 converts to
dCCO2
A bD =W gL3
dr (A4.4) Ra =
D (A5.2)
In this case, initial and boundary conditions are
s In these systems, density-driven natural convection occurs
CCO2 = CCo at r = rb (A4.5)
2 when Ra > 2100.
4017 dx.doi.org/10.1021/ef500034j | Energy Fuels 2014, 28, 40024020
Energy & Fuels Article

Accordingly, for the captive-bubble system at 0.58 MPa and
318 K, eq A5.2 gives
(1060 1000 kg/m 3) (9.8 m/s2) [(2 102)3 m 3]
Ra =
[0.65 103 kg/(ms)] [3 109 (m 2/s)]
= 24 108 > 2100
Therefore, the calculation of this dimensionless number for
the system at hand reveals that diusion is accelerated by
density-driven natural convection in regime I. The density
dierence in regime II is smaller, but according to this criterion,
natural convection should occur in regime II as well.
Nonetheless, the results indicate that the rate of mass transfer
is what one would expect for diusion with no convection
(section 4.2, Figure 15).
Nomenclature
Ab = gasliquid (bubble) interfacial area
Bo = Bond number
CCO2 = CO2 concentration
CsCO2 = CO2 concentration at the bubble interface
CbCO2 = CO2 concentration in the aqueous phase (bulk)
CbCO
1
= CO2 concentration in the bulk phase in regime I
b2 2
min = contact angle of the bubble with the initial bubble
radius of Rio
max = last accurate detectable contact angle
final = contact angle after establishing the equilibrated
bubble
= porosity of the porous medium
= dynamic viscosity
CO2 = density of CO2 in the bubble pressure
g = density of the gas phase (bubble)
l = density of the liquid phase
= density dierence between the bubble and aqueous
phase
aq,CO2 = interfacial tension between the aqueous phase and
the CO2-rich phase
w,cCO = interfacial tension between water and CO2
lv = interfacial tension between the aqueous phase and the
gas phase
wg = interfacial tension between CO2 and water
sv = surface energy between the solid and the gas phases
sl = surface energy between the solid and the aqueous
phases
= line tension
= bubble height; location of the apex

CCO2 = CO2 concentration in the bulk phase in regime II
D = CO2 molecular diusion coecient into water
D1 = eective diusion coecient in regime I ASSOCIATED CONTENT
D2 = eective diusion coecient in regime II *
S Supporting Information

Fb = buoyancy force Text describing information concerning line-tension determi-

Fg = gravity force nation for a CO2/water/Bentheimer system and a gure
Fp = pressure force showing contact angle versus bubble radius or inverse length of
F = surface tension force the contact line at two dierent pressures. This material is
available free of charge via the Internet at http://pubs.acs.org.

g = local gravitational acceleration
H = contact-angle hysteresis
L = characteristic length scale of convection AUTHOR INFORMATION
NCO2 = molar ux of CO2 Corresponding Author
mb = bubble mass *E-mail: n.shojaikaveh@tudelft.nl. Tel.:+31152789671.
MD = bubble maximum diameter Notes
p(z) = vertical pressure distribution The authors declare no competing nancial interest.
Pa = characterization factor of the surface roughness
Pb = total pressure in the bubble
Pc = capillary pressure
ACKNOWLEDGMENTS
The research reported in this work is carried out as a part of the
PL = pressure of the aqueous phase CATO2 project (CO2 capture, transport, and storage in The
Pnw = pressure of the nonwetting phase Netherlands). This research is conducted in the Laboratory of
Pw = pressure of the wetting phase Geoscience and Engineering at Delft University of Technology.
R = contact radius; the radius of the contact circle We gratefully thank the technical sta of the Laboratory,
rb = bubble radius particulary J. Etienne, M. Friebel, K. Heller, and J. van Meel.
R0 = radius of curvature at apex We also thank Dr. Cas Berentsen and Amin Ameri for their
Ri0 = initial radius of bubble at apex constructive input during this study.
R02 = radius of bubble at apex at transition time
Rfinal = bubble radius after establishing the equilibrated
bubble
REFERENCES
(1) Ofori, A. E.; Engler, T. W., Eects of CO2 Sequestration on the
Ra = Rayleigh number Petrophysical Properties of an Aquifer Rock. Canadian Unconventional
t = time Resources Conference, Calgary, Alberta, Canada; Society of Petroleum
V = volume of a droplet Engineers: Richardson, TX, USA, 2011.
Vb = volume of a bubble (2) Chadwick, A.; Arts, R.; Bernstone, C.; May, F.; Thibeau, S.;
W = constant molar rate of CO2 Zweigel, P. Best practice for the storage of CO2 in saline aquifers:
= apparent contact angle (experimentally determined Observations and guidelines from the SACS and CO2STORE Projects;
contact angle) British Geological Survey: Nottingham, U.K., 2007.
(3) Arts, R. J.; Vandeweijer, V. P.; Hofstee, C.; Pluymaekers, M. P.
Y = Youngs contact angle D.; Loeve, D.; Kopp, A.; Plug, W. J. The feasibility of CO2 storage in
a = advancing contact angle the depleted P18-4 gasfield offshore the Netherlands (the ROAD
r = receding contact angle project). Int. J. Greenhouse Gas Control 2012, 11 (Suppl. (CATO: CCS
s = contact angle for a spherical droplet/bubble Research in the Netherlands)), S10S20.
= contact angle for the bubble with innite radius of the (4) Velasquez, D.; Rey, O.; Manrique, E., An overview of carbon
solidliquid contact circle dioxide sequestration in depleted oil and gas reservoirs in Florida,

4018 dx.doi.org/10.1021/ef500034j | Energy Fuels 2014, 28, 40024020

Energy & Fuels
USGS Petroleum Province 50. Fourth LACCEI International Latin
American and Caribbean Conference for Engineering and Technology
(LACCEI2006), Mayaguez, Puerto Rico; 2006.
(5) Damen, K.; Faaij, A.; van Bergen, F.; Gale, J.; Lysen, E.
Identification of early opportunities for CO2 sequestrationWorld-
wide screening for CO2-EOR and CO2-ECBM projects. Energy 2005,
30 (10), 19311952.
(6) Bergen, F. V.; Krzystolik, P.; Wageningen, N. v.; Pagnier, H.; Jura,
B.; Skiba, J.; Winthaegen, P.; Kobiela, Z. Production of gas from coal
seams in the Upper Silesian Coal Basin in Poland in the post-injection
period of an ECBM pilot site. Int. J. Coal Geol. 2009, 77 (12), 175
187.
(7) Anderson, W. G. Wettability Literature SurveyPart 1: Rock/
Oil/Brine Interactions and the Effects of Core Handling on
Wettability. J. Pet. Technol. (1969) 1986, 38 (10), 11251144.
(8) Morrow, N. R. Wettability and Its Eect on Oil Recovery. J. Pet.
Technol. (1969) 1990, (12).
(9) Arendt, B.; Dittmar, D.; Eggers, R. Interaction of interfacial
convection and mass transfer effects in the system CO2water. Int. J.
Heat Mass Transfer 2004, 47 (1718), 36493657.
(10) Yang, D.; Gu, Y.; Tontiwachwuthikul, P. Wettability
determination of the crude oilreservoir brinereservoir rock system
with dissolution of CO2 at high pressures and elevated temperatures.
Energy Fuels 2008, 22 (4), 23622371.
(11) Chalbaud, C.; Robin, M.; Lombard, J. M.; Martin, F.; Egermann,
P.; Bertin, H. Interfacial tension measurements and wettability
evaluation for geological CO2 storage. Adv. Water Resour. 2009, 32
(1), 98109.
(12) Chiquet, P.; Broseta, D.; Thibeau, S. Wettability alteration of
caprock minerals by carbon dioxide. Geofluids 2007, 7, 112122.
(13) Pentland, C. H.; Al-Mansoori, S.; Iglauer, S.; Bijeljic, B.; Blunt,
M. J. Measurement of Non-Wetting Phase Trapping in Sand Packs.
SPE Annual Technical Conference and Exhibition, Denver, CO, USA;
Society of Petroleum Engineers: Richardson, TX, USA, 2008.
(14) Mills, J.; Riazi, M.; Sohrabi, M. Wettability of common rock-
forming minerals in a CO2-brine system at reservoir conditions.
International Symposium of the Society of Core Analysts, Austin, Texas,
USA, 2011.
(15) Akbarabadi, M.; Piri, M. Geologic storage of carbon dioxide: an
experimental study of permanent capillary trapping and relative
permeability. International Symposium of the Society of Core Analysts,
Austin, Texas, USA, 2011.
(16) Alotaibi, M. B.; Azmy, R. M.; Nasr-El-Din, H. A. Wettability
Studies Using Low-Salinity Water in Sandstone Reservoirs. SPE
Reservoir Eval. Eng. 2011, 14, 713725.
(17) Chalbaud, C.; Robin, M.; Bekri, S.; Egermann, P. Wettability
impact on CO2 storage in aquifers: Visualisation and quantication
using micromodel tests, pore network model and reservoir simulations.
International Symposium of the Society of Core Analysts, Calgary,
Canada, 2007.
(18) Jiamin, W.; Yongman, K.; Jongwon, J. Wettability alteration
upon reaction with scCO2: Pore scale visualization and contact-angle
measurements. Goldschmidt2011, Prauge, Czech Republic, Aug. 14
19, 2011.
(19) Espinoza, D. N.; Santamarina, J. C. Water-CO2-mineral systems:
Interfacial tension, contact angle, and diusionImplications to CO2
geological storage. Water Resour. Res. 2010, 46W07537.
(20) Shi, J.-Q.; Durucan, S. Modelling of Mixed-Gas Adsorption and
Diusion in Coalbed Reservoirs. SPE Unconventional Reservoirs
Conference, Keystone, CO, USA; Society of Petroleum Engineers:
Richardson, TX, USA, 2008.
(21) Puri, R.; Yee, D., Enhanced Coalbed Methane Recovery. SPE
Annual Technical Conference and Exhibition, New Orleans, LA, USA;
Society of Petroleum Engineers: Richardson, TX, USA, 1990.
(22) Anderson, W. G. Wettability Literature SurveyPart 2:
Wettability Measurement. J. Pet. Technol. (1969) 1986, 38 (11),
12461262.
4019
Article
(23) Kwok, D. Y.; Neumann, A. W. Contact angle measurement and
contact angle interpretation. Adv. Colloid Interface Sci. 1999, 81 (3),
167249.
(24) Lin, F. Y. H.; Li, D.; Neumann, A. W. Effect of Surface
Roughness on the Dependence of Contact Angles on Drop Size. J.
Colloid Interface Sci. 1993, 159 (1), 8695.
(25) Drelich, J.; Miller, J. D.; Hupka, J. The Effect of Drop Size on
Contact Angle over a Wide Range of Drop Volumes. J. Colloid Interface
Sci. 1993, 155 (2), 379385.
(26) Gaydos, J.; Neumann, A. W. The dependence of contact angles
on drop size and line tension. J. Colloid Interface Sci. 1987, 120 (1),
7686.
(27) Good, R. J.; Koo, M. N. The effect of drop size on contact angle.
J. Colloid Interface Sci. 1979, 71 (2), 283292.
(28) Herzberg, W. J.; Marian, J. E. Relationship between contact
angle and drop size. J. Colloid Interface Sci. 1970, 33 (1), 161163.
(29) Good, R. J.; Elbing, E. Generalization of Theory for Estimation
of Interfacial Energies. Ind. Eng. Chem. 1970, 62 (3), 5478.
(30) Vafaei, S.; Podowski, M. Z. Analysis of the relationship between
liquid droplet size and contact angle. Adv. Colloid Interface Sci. 2005,
113 (23), 133146.
(31) Vafaei, S.; Podowski, M. Z. Theoretical analysis on the effect of
liquid droplet geometry on contact angle. Nucl. Eng. Des. 2005, 235
(1012), 12931301.
(32) Liu, Y.; Wang, J.; Zhang, X. Accurate determination of the
vapor-liquid-solid contact line tension and the viability of Young
equation. Sci. Rep. 2013, 3, No. PMC3684806.
(33) Pethica, B. A.; Pethica, P. J. C. 2nd International Congress of
Surface Activity; Butterworths, London, 1957; p 131.
(34) Leja, J.; Poling, W. International Mineral Process Congress; 1960;
1960; p 325.
(35) Lowe, H.; Hardt, S. Chemical Micro Process Engineering; John
Wiley & Sons: New York, 2004.
(36) Blokhuis, E. M.; Shilkrot, Y.; Widom, B. Youngs law with
gravity. Mol. Phys. 1995, 86 (4), 891899.
(37) Fujii, H.; Nakae, H. Effect of gravity on contact angle. Philos.
Mag. A 1995, 72 (6), 15051512.
(38) Sakai, H.; Fujii, T. The Dependence of the Apparent Contact
Angles on Gravity. J. Colloid Interface Sci. 1999, 210 (1), 152156.
(39) Yang, D.; Gu, Y.; Tontiwachwuthikul, P. Wettability
Determination of the Reservoir BrineReservoir Rock System with
Dissolution of CO2 at High Pressures and Elevated Temperatures.
Energy Fuels 2008, 22 (1), 504509.
(40) Shojai Kaveh, N.; Berentsen, C.; Rudolph-Floter, S. E. J.; Wolf,
K. H.; Rossen, W. R., Wettability Determination by Equilibrium
Contact Angle Measurements: Reservoir RockConnate Water
System with Injection of CO2. SPE Europec/EAGE Annual Conference;
Society of Petroleum Engineers: Copenhagen, Denmark, 2012.
(41) Jung, J.-W.; Wan, J. Supercritical CO2 and Ionic Strength Effects
on Wettability of Silica Surfaces: Equilibrium Contact Angle
Measurements. Energy Fuels 2012, 26 (9), 60536059.
(42) Saraji, S.; Goual, L.; Piri, M.; Plancher, H. Wettability of
Supercritical Carbon Dioxide/Water/Quartz Systems: Simultaneous
Measurement of Contact Angle and Interfacial Tension at Reservoir
Conditions. Langmuir 2013, 29 (23), 68566866.
(43) Ameri, A.; ShojaiKaveh, N.; Rudolph, E. S. J.; Wolf, K. H.;
Farajzadeh, R.; Bruining, J. Investigation on Interfacial Interactions
among Crude OilBrineSandstone RockCO2 by Contact Angle
Measurements. Energy Fuels 2013, 27 (2), 10151025.
(44) Shojai Kaveh, N.; Wolf, K. H.; Ashrafizadeh, S. N.; Rudolph, E.
S. J. Effect of coal petrology and pressure on wetting properties of wet
coal for CO2 and flue gas storage. Int. J. Greenhouse Gas Control 2012,
11 (Suppl. (CATO: CCS Research in the Netherlands)), S91S101.
(45) Shojai Kaveh, N.; Rudolph, E. S. J.; Wolf, K.-H. A. A.;
Ashrafizadeh, S. N. Wettability determination by contact angle
measurements: hvBb coalwater system with injection of synthetic
flue gas and CO2. J. Colloid Interface Sci. 2011, 364 (1), 237247.
dx.doi.org/10.1021/ef500034j | Energy Fuels 2014, 28, 40024020
Energy & Fuels Article

(46) Chiquet, P.; Daridon, J.-L.; Broseta, D.; Thibeau, S. CO2/water

interfacial tensions under pressure and temperature conditions of CO2
geological storage. Energy Convers. Manage. 2007, 48 (3), 736744.
(47) Shojai Kaveh, N.; Rudolph, E. S. J.; Rossen, W. R.; Hemert, P.
v.; Wolf, K. H. Interfacial Tension and Contact Angle Determination
in Water-Sandstone Systems with Injection of Flue Gas and CO2. IOR
2013From Fundamental Science to Deployment; EAGE: Saint
Petersburg, Russia, 2013.
(48) Span, R.; Wagner, W. A New Equation of State for Carbon
Dioxide Covering the Fluid Region from the Triple-Point Temper-
ature to 1100 K at Pressures up to 800 MPa. J. Phys. Chem. Ref. Data
1996, 25 (6), 15091596.
(49) Shyu, G.-S.; Hanif, N. S. M.; Hall, K. R.; Eubank, P. T. Carbon
dioxide-water phase equilibria results from the Wong-Sandler
combining rules. Fluid Phase Equilib. 1997, 130 (12), 7385.
(50) Song, B.; Springer, J. Determination of Interfacial Tension from
the Profile of a Pendant Drop Using Computer-Aided Image
Processing: 1. Theoretical. J. Colloid Interface Sci. 1996, 184 (1),
6476.
(51) Shojai Kaveh, N. Interfacial interactions and wettability evaluation
of rock surfaces for CO2 storage; Delft University of Technology: Delft,
The Netherlands, 2014.
(52) Das, A. K.; Das, P. K. Equilibrium shape and contact angle of
sessile drops of different volumesComputation by SPH and its
further improvement by DI. Chem. Eng. Sci. 2010, 65 (13), 4027
4037.
(53) Ren, H.; Xu, S.; Wu, S. T. Effects of gravity on the shape of
liquid droplets. Opt. Commun. 2010, 283 (17), 32553258.
(54) Iglauer, S.; Fern, M. A.; Shearing, P.; Blunt, M. J. Comparison
of residual oil cluster size distribution, morphology and saturation in
oil-wet and water-wet sandstone. J. Colloid Interface Sci. 2012, 375 (1),
187192.
(55) Dickson, J. L.; Gupta, G.; Horozov, T. S.; Binks, B. P.; Johnston,
K. P. Wetting phenomena at the CO2/water/glass interface. Langmuir
2006, 22 (5), 21612170.
(56) Peksa, A. E.; Zitha, P. L. J.; Wolf, K. H. A. A., Role of Rock
Surface Charge in the Carbonated Water Flooding Process. 75th
EAGE Conference & Exhibition incorporating SPE EUROPEC 2013;
EAGE: London, U.K., 2013.
(57) Tokunaga, T. K. DLVO-Based Estimates of Adsorbed Water
Film Thicknesses in Geologic CO2 Reservoirs. Langmuir 2012, 28
(21), 80018009.
(58) Sharma, M. M.; Yen, T. F. Interfacial electrochemistry of oxide
surfaces in oil-bearing sands and sandstones. J. Colloid Interface Sci.
1984, 98 (1), 3954.
(59) Cussler, E. L. Diusion: Mass transfer in uid systems; Cambridge
University Press: Cambridge, U.K., 2009.
(60) Duan, Z.; Sun, R.; Zhu, C.; Chou, I. M. An improved model for
the calculation of CO2 solubility in aqueous solutions containing Na+,
K+, Ca2+, Mg2+, Cl, and SO42. Mar. Chem. 2006, 98 (24), 131
139.
(61) Frank, M. J. W.; Kuipers, J. A. M.; van Swaaij, W. P. M.
Diffusion Coefficients and Viscosities of CO2 + H2O, CO2 + CH3OH,
NH3 + H2O, and NH3 + CH3OH Liquid Mixtures. J. Chem. Eng. Data
1996, 41 (2), 297302.
(62) Okhotsimskii, A.; Hozawa, M. Schlieren visualization of natural
convection in binary gasliquid systems. Chem. Eng. Sci. 1998, 53
(14), 25472573.
(63) Farajzadeh, R. Enhanced transport phenomena in CO 2
sequestration and CO2 EOR. Delft University of Technology: Delft,
The Netherlands, 2009.
(64) Vasconcelos, J. M. T.; Orvalho, S. P.; Alves, S. S. Gasliquid
mass transfer to single bubbles: Effect of surface contamination. AIChE
J. 2002, 48 (6), 11451154.
(65) Dutta, B. K. Principles of mass transfer and separation processes;
PHI Learning: New Delhi, India, 2009.
(66) Lapwood, E. R. Convection of a fluid in a porous medium.
Math. Proc. Cambridge Philos. Soc. 1948, 44 (04), 508521.

4020 dx.doi.org/10.1021/ef500034j | Energy Fuels 2014, 28, 40024020