Journal of Colloid and Interface Science 563 (2020) 145–155

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science

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

Effect of salinity, Mg2+ and SO2

4 on ‘‘smart water”-induced carbonate
wettability alteration in a model oil system
Jin Song a, Qing Wang a, Imran Shaik b, Maura Puerto a, Prem Bikkina b, Clint Aichele b, Sibani L. Biswal a,⇑,
George J. Hirasaki a,⇑
a
Rice University, 6100 Main St., MS-362, Department of Chemical and Biomolecular Engineering, Houston, TX 77005, USA
b
Oklahoma State University, 420 Engineering North, School of Chemical Engineering, Stillwater, OK 74078, USA

g r a p h i c a l a b s t r a c t

a r t i c l e i n f o a b s t r a c t

Article history: Hypothesis: We present a systematic study of the ‘‘smart water” induced wettability alteration. This pro-
Received 7 November 2019 cess is believed to be greatly affected by the brine salinity and the presence of Mg2+ and SO2

4 in the brine.
Revised 10 December 2019 Experiments and modelling: To characterize the wettability alteration, we perform spontaneous imbibition
Accepted 11 December 2019
measurement using Indiana limestone cores and a model oil with added naphthenic acid. Both single-
Available online 12 December 2019
electrolyte-based and seawater-based ‘‘smart water” are tested to investigate the effect of Mg2+, SO2
4
and salinity on wettability alteration. Rock/brine and oil/brine zeta potentials are measured, and the elec-
Keywords:
trostatic component of disjoining pressure is calculated to understand the role of electrostatics in the
Wettability alteration
Smart water
wettability alteration. The surface concentration of charged species on the limestone surface is analyzed
Low salinity water based on a natural carbonate surface complexation model (SCM).

⇑ Corresponding authors.
E-mail addresses: js110@rice.edu (J. Song), qw10@rice.edu (Q. Wang), imran.shaik@okstate.edu (I. Shaik), Maura@rice.edu (M. Puerto), prem.bikkina@okstate.edu
(P. Bikkina), clint.aichele@okstate.edu (C. Aichele), biswal@rice.edu (S.L. Biswal), gjh@rice.edu (G.J. Hirasaki).

https://doi.org/10.1016/j.jcis.2019.12.040
0021-9797/Ó 2019 Elsevier Inc. All rights reserved.
146 J. Song et al. / Journal of Colloid and Interface Science 563 (2020) 145–155
Enhanced oil recovery Findings: Both the reduction of Na+ and addition of SO2

Spontaneous imbibition Mg2+ is found to be unfavorable for wettability alteration. Ca2+ is believed to facilitate SO2

bility alteration based on the comparison between the single-electrolyte-based and seawater-based
brines. The reduction of the Na+ surface complexation (>CaOH  Na+0.25) in low salinity brines is believed
to be a critical mechanism responsible for wettability alteration based on the SCM calculations.
1. Introduction
Wettability is known to have a significant impact on the oil
recovery in carbonate reservoirs. About 90% of carbonate reservoirs
are neutral to preferentially oil-wet [1], which makes the oil recov-
ery very challenging due to the unfavorable capillary pressure dur-
ing brine injection. Injection of hydrophilic surfactants as the
wettability modifier can successfully alter wettability [2–9] but
increases the capital investment significantly. Another approach
that requires much lower cost is the injection of ‘‘smart water”,
which are the brines with designed salinity and ionic composition
that can enhance oil recovery. ‘‘Smart water” has drawn increasing
attention from both the petroleum industry and the academia over
the past two decades [10–12]. This method requires relatively low
capital investment compared to other chemical enhanced oil
recovery (EOR) techniques and has been reported to successfully
increase oil recovery from lab scale to field scale [11]. However,
there are many cases where ‘‘smart water” has not been successful
[11,13,14]. Here we aim to understand the fundamental physico-
chemical interactions governing ‘‘smart water” EOR.
Significant research utilizing the core-scale experiments with
smart water have been reported in the literature. Strand et al. [1]
and Zhang et al. [15] performed spontaneous imbibition tests with
chalk mineral and crude oil at several temperatures up to 130 °C
and observed up to 40% additional oil recovery after introducing
Ca2+, Mg2+, and SO2
4 into the modified seawater. With the assis-
tance of zeta potential measurement and effluent ionic composi-
tion analysis, they proposed a mechanism based on electrostatic
interactions to explain the roles of Ca2+, Mg2+, and SO2
4 . In their
hypothesis, SO2
4 favors the wettability alteration by lowering the
positive surface charge of carbonate surface, thereby weakening
the electrostatic attraction between the negatively charged crude
oil and the carbonate surface. They also proposed that Ca2+ could
form a positively charged surface complex with the adsorbed
organic acids and result in acid desorption due to charge repulsion
with the rock surface. Mg2+ was believed to substitute Ca2+ in the
lattice at high temperature and take away the adsorbed organic
acids with the released Ca2+. Fathi et al. [16] performed similar
spontaneous imbibition tests using treated crude oil depleted in
water-soluble acids and treated crude oil with only water-soluble
acids. They concluded that the water-soluble acids in the crude
oil have dramatic impact on both the initial wetting state of the
rock as well as the effectiveness of low salinity water. In another
study, Fathi et al. [17] demonstrated that lowering salinity (deplet-
ing NaCl in seawater) can alter wettability and improve oil recov-
ery in spontaneous imbibition. Austad et al. [13,14] emphasized
the possible effect of anhydrite presence in the core in determining
the effectiveness of low salinity water by core flooding experi-
ments. They observed additional oil recovery in cores with initial
anhydrite presence when switching formation brine flooding to
diluted seawater flooding, while insignificant oil recovery was
observed in clean cores without anhydrite presence. Shariatpanahi
et al. [18] examined the ability of seawater-based ‘‘smart water” to
alter wettability in dolomite cores. They found that the seawater
could not significantly increase oil recovery while the 10-times-
diluted seawater was a successful EOR fluid in a dolomite system.
4 are found to contribute to wettability alteration.
4 with wetta-
Ó 2019 Elsevier Inc. All rights reserved.
Yousef et al. [19] performed core flooding experiments with a
carbonate reservoir core and crude oil. They observed approxi-
mately an additional 20% oil recovery of original oil in place (OOIP)
by diluted seawater injection. Al-Harrasi et al. [20] examined the
effect of low salinity brine (diluted seawater) by both spontaneous
imbibition and core flood testing using the same core and crude oil.
Significant additional oil recovery was observed in the sponta-
neous imbibition test, but much less increment was observed in
the core flooding. Mohanty and Chandrasekhar [21] studied the
effect of modified seawater by core flooding and demonstrated that
Mg2+ and SO2
4 have to be present for the modified seawater to
improve oil recovery. Gandomkar and Rahimpour [22] confirmed
that diluted seawater could result in higher oil recovery compared
to seawater in the core flooding experiment with limestone cores
and crude oils.
Besides core-level experiments, other efforts have been made to
decipher ‘‘smart water” EOR from different perspectives. Surface
complexation models have been developed to understand the sur-
face interactions between carbonate and the oil in different brines
[12,23–25]. Disjoining pressure calculations have been performed
to reveal the relationship between the intermolecular interactions
and the wetting behavior [10,26–28]. Molecular dynamics simula-
tion has also been applied to study the mechanism of Mg2+ and
SO2

4 induced wettability alteration by Sakuma et al. [29] and
Andersson et al. [30] Multiple surface characterizations including
surface forces measurement have been performed to understand
the change of calcite surface after being exposed to low salinity bri-
nes by Chen et al. [31] Moreover, contact angles of carbonate/oil/
brine system are utilized to characterize the rock wettability
before and after exposing to the ‘‘smart water” [21,32–34]. How-
ever, the consistency of contact angle data among different studies
is typically unsatisfactory.
Even though a large amount of work has been done in the crude
oil system using modified seawater as the ‘‘smart water”, the
underlying mechanisms governing the oil-brine-rock interfaces
are still not conclusive. Here we utilize a model oil composed of
naphthenic acids, one of the surface-active components in crude
oil, to better understand these mechanisms. A model oil study
can provide valuable insights because most of the proposed mech-
anisms for ‘‘smart water” are based on the interaction between
ions and naphthenic acids in the oil.
Moreover, diluted seawater is investigated as the ‘‘smart water”
in most of the literature work. Dilution of seawater lowers the
salinity and the concentration of divalent ions (Mg2+ and SO2
4 )
simultaneously, making it difficult to distinguish the individual
effects of lowering ionic strength and modifying divalent ions con-
centration. Therefore, a systematic study of the roles of the brine
composition (potential determining ions and salinity) in well-
characterized model systems is required and currently unavailable
in the literature. In this work, Indiana limestone cores and model
oil with added naphthenic acid are tested by different series of
low salinity brines in spontaneous imbibition. Two groups of
‘‘smart water” are tested: one group is prepared by modifying 4x
diluted seawater; the other group is prepared with single elec-
trolyte (NaCl/Na2SO4/MgSO4). Zeta potential, disjoining pressure
and surface complexation modeling are also used to assist in
 J. Song et al. / Journal of Colloid and Interface Science 563 (2020) 145–155 147

analyzing the effects of salinity and divalent ions, Mg2+ and SO2

4 ,
on wettability alteration.
2. Experimental section
2.1. Core material
Indiana limestone was chosen as the representative of carbon-
ate minerals. Seven Indiana limestone cores (Kocurek Industries)
of 1-inch length, 1-inch diameter, were selected for the sponta-
neous imbibition tests. The permeability was approximately
50mD. The mineralogy of the limestone was characterized by a
JEOL JXA 8530F Hyperprobe electron probe microanalyzer (EPMA)
with energy dispersive X-ray spectroscopy (EDX). The composition
is shown in the supplementary material Table S1. The Indiana
limestone is primary CaCO3 (93% by molar fraction). Anhydrite
presence in the rock sample is expected to be negligible. This
was confirmed by equilibrating 20 mL 1 wt% NaCl solution with
0.5 g crushed Indiana limestone powder for a week and then test-
ing the supernatant by inductively coupled plasma atomic emis-
sion spectroscopy (ICP-OES). Sulfate was found to be below the
detection limit in the brine.
2.2. Model oils
A model oil composed of 87% (vol%) dodecane (Sigma-Aldrich,
99%) and 13% (vol%) toluene (Sigma-Aldrich, 99%), with added
naphthenic acid (Sigma-Aldrich, AN = 230) was used to model
the crude oil. The dodecane and toluene represents the long chain
saturates and aromatic compounds, respectively. The volume ratio
of dodecane/toluene was chosen based on lab experience to mimic
the phase behavior of a crude oil. Naphthenic acid was added to
obtain a model oil with total acid number (TAN) = 1.5 mg KOH/g.
The model oil is abbreviated as ‘‘AN1.5MO”. TAN is measured by
a modified ASTM titration method [35] proposed by Fan and Buck-
ley and confirmed to be 1.5 mg KOH/g oil. Moreover, model oil
(‘‘MO”) without any added acid was also tested as the control oil.
2.3. Brines
Single-electrolyte brines are also tested. This series of brines were
chosen to examine the effect of the change in the electrostatic
double-layer and correlate the electrostatic interactions to individ-
ual potential determining ions.
All the brines were pre-equilibrated with calcite rock powder
and AN1.5 MO (vol/vol = 1:1) for a week at ambient conditions
until the solution pH remained constant. During the pre-
equilibration process, the rock was not in direct contact with the
oil phase; therefore, the naphthenic acid in the oil would not be
depleted by adsorption onto the rock minerals. The purpose of
pre-equilibration was to avoid rock dissolution and acid desorption
from the rock surface to the brine. Moreover, the pH was constant
during the imbibition test if the brine was pre-equilibrated. All
brine samples were prepared without air contact and equilibration
was done in a closed vessel.
Due to the dissolution of calcite, Ca2+ was present in all the
equilibrated brines even though no calcium salt is added in most
of the brines. Ca2+ is one of the most important potential determin-
ing ions in a carbonate mineral system. Therefore, the Ca concen-
tration was measured by ICP-OES to monitor possible variation of
Ca level after switching to a different brine. The equilibrated brines
are filtered by 0.22 mm PTFE filters and diluted in 1% nitric acid for
the ICP-OES measurement.
2.4. Core preparation
The Indiana limestone cores were first dried to constant weight
in an oven. Then the cores were vacuumed and saturated with the
initial brine (5 M NaCl) at room temperature. The porosity and pore
volume were calculated based on the change of core weight and
the density of the initial brine (1.1805 g/cm3 for 5 M NaCl). After-
wards, the cores were placed in glass centrifuge tubes and
immersed in the oil for water drainage using a high-speed ultra-
centrifuge at 11,000 rpm for at least 24 h. The equivalent capillary
pressure was calculated to be 80psi. The displaced brine volume
was measured, and the initial oil saturation was calculated. After
the oil saturation, the cores were aged for 2 weeks at 120 °C to
obtain oil-wet surfaces.

A 5 M NaCl solution was chosen as the representative formation
brine. This ionic strength (I) is characteristic of carbonate reser-
voirs and is used as the initial brine to saturate the cores before
oil saturation. Neither Ca2+ nor Mg2+ is added to the formation
brine so that the individual role of the divalent ions and salinity
can be investigated separately in the diluted brines. A series of
low salinity brines are prepared, as shown in Table 1. Diluted
seawater-based brines include the following: 4x diluted seawater
(‘‘4dSW”), 4x diluted seawater with 3x concentrated SO2
4 noticeable oil loss due to evaporation was observed during the
(‘‘4dSW3S”), 4x diluted seawater depleted in SO2
4 (‘‘4dSW0S”),
and 4x diluted seawater depleted in Mg2+ (‘‘4dSW0Mg”).
2.5. Procedure for spontaneous imbibition tests
The oil-wet cores were first placed into dry Amott imbibition
cells. Then the cells were filled with the initial brine (5 M NaCl)
and transferred into a water bath at 90 °C. The top narrow neck
of each Amott cell was not immersed in the water bath and was
elongated by connecting it to a glass pipette of the same OD as
the top of the cell. This extension functioned as a condenser. No
experiment. The recovered oil was read daily through the gradu-
ated section of the Amott cell.

Table 1
Recipe of the tested brines (before equilibration) and their equilibrium pH.

Name Ionic strength(M) Na+(mM) Ca2+(mM) Mg2+(mM) Cl
(mM) HCO

3 (mM) SO2

4 (mM) pHa Group
5 M NaCl 5.000 5000 0 0 5000 0 0 8.0 Initial brine

4dSW 0.164 115.0 3.25 11.25 131.5 0.5 6 7.5 Seawater based
4dSW3S 102.5 3.25 11.25 95.0 0.5 18 7.4
4dSW0S 121.0 3.25 11.25 149.5 0.5 0 7.4
4dSW0Mg 148.8 3.25 0 142.8 0.5 6 7.6

LowSal NaCl 164 0 0 164 0 0 7.8 Single electrolyte

LowSal Na2SO4 109 0 0 0 0 55 7.7
LowSal MgSO4 0 0 41 0 0 41 7.7
a
The brine pH was measured after the brine was equilibrated with calcite and AN1.5MO (1:1 vol/vol). The same equilibrated brines were used in the imbibition test.
148 J. Song et al. / Journal of Colloid and Interface Science 563 (2020) 145–155

The initial brine was switched to the designed low salinity bri-
nes after the oil reading was stable for at least 5 days. Nine cores
(A ~ I) were tested with different brine series. The porosity (/), ini-
tial oil saturation (Soi) and the testing brine sequence were sum-
marized in Table 2.
2.6. Amott-Harvey wettability index (IA-H) measurement
To evaluate the initial wettability of Indiana limestone core and
its wettability after oil aging in AN1.5MO, the Amott-Harvey wet-
tability index (IA-H) was measured for two Indiana limestone cores
(J and K) using 5 M NaCl brine. Core J was saturated with MO with-
out any oil aging step. Core K was saturated with AN1.5MO and
then aged in AN1.5MO for 2 weeks at 120 °C, the same as core A ~ I.
There were four sequential steps for IA-H measurement: (1)
brine spontaneous imbibition; (2) brine forced imbibition; (3) oil
spontaneous drainage; and (4) oil forced drainage. IA-H was calcu-
lated based on the definition:
V osp V wsp
IA
H ¼
tion with the two major surface sites: >CO3H+0.75 and >CaOH
0.75.
V osp þ V of V wsp þ V wf
Vosp and Vof were the volume of displaced oil in the (1) and (2)
steps, respectively. Vwsp and Vwf were the volume of displaced brine
in the (3) and (4) steps, respectively. All steps were done at room
temperature and the forced imbibition/drainage are achieved by
centrifuging the core at 10,000 rpm for 48 h in total (after first
24 h, the core was flipped to eliminate capillary end effect). The
tested oil was MO for core J and was AN1.5MO for core K.
2.7. Zeta potential measurement
To analyze the role of electrostatic interactions in determining
the rock wettability, both oil/brine and rock/brine zeta potential
were measured in various brines. The brines used to prepare the
rock/brine and oil/brine dispersion were first allowed to equili-
brate with the AN1.5MO and the Indiana limestone rock (3 phases
equilibration) for 1 week until the pH of the brine was stable. The
container for the sample can be treated as a closed system because
the brine is filled up to the brim, leaving negligible air space in the
bottle. Then the brine was extracted and filtered through a 0.5-mm
stainless steel filter. The same equilibrated brine was used for
preparing both rock/brine and oil/brine dispersion. A cartoon
demonstrating the equilibration method is provided in the supple-
mentary material Fig. S1. The zeta potential was measured by a
light scattering analyzer (DelsaMax PRO) based on electrophoretic
phase analysis light scattering (PALS) method.
Ground Indiana limestone powder (0.2 g) and equilibrated brine
(25 mL) were mixed to obtain 0.8%wt. sample for rock/brine zeta
potential measurements. A sonifier (Branson) was used to sonicate
the sample for 1 min to facilitate powder dispersion. After
sonication, the large particles were allowed to settle for 1 h. Then
the supernatant was extracted for the measurement. Five samples
from each bottle was measured to obtain the mean value and stan-
dard deviation.
AN1.5 model oil (25 mL) and equilibrated brine (25 mL) were
mixed together to prepare the dispersion for oil/brine zeta poten-
tial measurement. A sonifier (Branson) was used to sonicate the
sample for 30 s to facilitate oil dispersion. After sonication, a period
of 1 h was given to allow the oil droplet density to decrease to a
proper level for the zeta potential measurements. The supernatant
was extracted, and each sample was measured 5 times.
2.8. Surface complexation modeling
To probe the concentration variation of surface charged species
in different brines, the carbonate surface complexation model
(SCM) previously proposed by Song, et al.[36] was adopted to
model the Indiana limestone rock. Based on the model assump-
tions, the surface charge of limestone is a result of ion complexa-
Surface coverage of organic impurities (>CO3HA
5.25), as well as
the presence of silica impurity (>SiOH) are both considered to
account for the surface charge difference between Indiana lime-
stone and ideal pure calcite. Model parameters from this previous
work were used directly without re-fitting (shown in Table 3). A
geochemical database, PHREEQC Version 3.3.9 (database phreeqc.-
dat), was used for the model calculation. The modeling procedure
can be found in the reference by Song, et al. [36]. For the experi-
ments, the zeta potentials were measured in brines pre-
equilibrated with the acidic AN1.5MO in order to represent the
actual brine chemistry in the spontaneous imbibition experiments.
Therefore, the solubility of naphthenic acid in the brines must be
considered in the SCM. The dissociation constant (pKa) for naph-
thenic acids can vary depending on their origins. Previous litera-
ture report pKa between 3.9 and 5.0 for naphthenic acids based
on titration curves [37,38]. Therefore, a model naphthenic acid
whose pKa = 4.8 was assumed in the calculation to account for
the impact of acid solubility on the pH of the equilibrated brines.
Small quantities of naphthenic acid were added to the brine in
the model to ensure that the brine pH after equilibration was equal
to the measured pH (7.7–8.0). Based on the calculations,
0.8–3.3 mM of the model naphthenic acid was needed to account
for the naphthenic acid partitioning for a given brine. Note that
the calculation is not sensitive to pKa of naphthenic acid as long
as pKa < 7.7 because the added acids will almost completely disso-
ciate in the tested brines. The calculated zeta potential and calcium
concentration from SCM were compared to the actual measure-
ments to check the validity of model. Finally, the concentrations
of surface charged species were analyzed to provide insights on
wettability alteration.

Table 2
Summary of experimental design of the spontaneous imbibition tests.

# /(%) Soi (%) Initial brine Oil Testing brine sequence Purpose
A 16 81 5 M NaCl AN1.5 MO 5 M NaCl ? 4dSW ? 4dSW3S Effect of Mg2+and SO42
(seawater based)
B 17 83 5 M NaCl AN1.5 MO 5 M NaCl ? 4dSW0S
C 18 85 5 M NaCl AN1.5 MO 5 M NaCl ? 4dSW0Mg

D 16 78 5 M NaCl AN1.5 MO 5 M NaCl ? LowSal NaCl ? LowSal Na2SO4 Effect of Na+, Mg2+and SO2

4 (single electrolyte)
E 17 81 5 M NaCl AN1.5 MO 5 M NaCl ? LowSal Na2SO4
F 17 86 5 M NaCl AN1.5 MO 5 M NaCl ? LowSal MgSO4

G 17 79 5 M NaCl AN1.5 MO 5 M NaCl ? LowSal NaCl Reproducibility (repeat)

H 17 78 5 M NaCl AN1.5 MO 5 M NaCl ? I = 0.4 M NaCl ? LowSal NaCl
I 17 85 5 M NaCl AN1.5 MO 5 M NaCl ? LowSal MgSO4

J 17 83 5 M NaCl MO 5 M NaCl Amott-Harvey wettability index

K 16 81 5 M NaCl AN1.5 MO 5 M NaCl
 J. Song et al. / Journal of Colloid and Interface Science 563 (2020) 145–155 149

Table 3 additional oil is recovered in 4dSW3S in the tertiary recovery

Model parameters for the surface complexation model of the Indiana limestone from mode. From the comparison, the presence of SO2
is confirmed
4
Song et al. [36].
to be beneficial for oil recovery while Mg2+ is not.
Surface reaction Log10K Surface composition Spontaneous imbibition using single-electrolyte solutions was
>CaOH
0.75 + H+
> CaOH+0.25
2 0.30 CaCO3 (96%) also tested to further compare the individual effect of Mg2+, SO2
4
>CO3H+0.75 + OH

> CO
0.25
3 + H2O 0.50 >CaOH
0.75 4.95 nm
2 and reduction in ionic strength (Fig. 1(b)). Low salinity NaCl brine

0.75
>CaOH + Ca
> CaOH  Ca+1.25
2+
1.98 >CO3H+0.75 4.80 nm
2 without any added divalent ions is found to also trigger wettability

0.75
>CaOH + Mg
> CaOH  Mg+1.25
2+
1.66 >CO3HA
5.25 0.15 nm
2
>CO3H+0.75 + CO2

1.25 alteration and results in 17% additional oil recovery. Higher oil
3
> CO3H  CO3 1.99 SiO2 (4%)
>CO3H+0.75 + SO2

4
> CO3H  SO4

1.25
1.00 >SiOH 2.3 nm
2 recovery is observed in the two sulfate-containing brines: 25% in
>CO3H+0.75 + HCO
3
> CO3H  HCO3

0.25
0.25 LowSal Na2SO4 and 21% in LowSal MgSO4. Additionally, LowSal
>CaOH
0.75 + Na+
> CaOH  Na+0.25 0.09 Na2SO4 is also tested in a tertiary mode following the LowSal NaCl
>CO3H+0.75 + Cl

> CO3H  Cl
0.25
0.50
brine for Core D. The final oil recovery in LowSal Na2SO4 is almost
>SiOH
> SiO
+ H+
4.0 [39]
>SiOH + Ca2+
> SiOCa+ + H+
9.7 [40] identical as that for Core E, which is tested in LowSal Na2SO4 in sec-
ondary mode. This comparison shows the good reproducibility of
the imbibition test.
Reproducibility of the oil recovery in LowSal NaCl and LowSal
3. Results and discussion
MgSO4 is examined by duplicating the experiments. 4dSW is also
re-tested, but in tertiary mode, following LowSal MgSO4 brine.
3.1. Wettability after oil aging
Good reproducibility is observed for all the repeated cases. The fig-
ures of the oil recovery history of the repeated cases are provided
The wettability index (IA-H) of Indiana limestone cores were
in the supplementary material. A summary of all the final oil recov-
measured to investigate the initial wettability and the wettability
ery results is provided in Fig. 2. The effectiveness of low salinity
after AN1.5MO oil aging. The wettability index IA-H is 0.4 for the
‘‘smart water” with the same ionic strength ranks as: 4dSW3S 
core J, which was saturated with the non-surface-active MO and
4dSW0Mg > LowSal Na2SO4  4dSW > LowSal MgSO4 > 4dSW0S
tested without oil aging. IA-H is positive when the brine can more
 LowSal NaCl. For both seawater-based brines and single-
favorably displace the oil comparing to the oil displacing the brine
V V
electrolyte solutions, SO2
4 is confirmed to be beneficial and Mg
2+

(V osposp
þV
> V wspwsp
þV
). Therefore, a positive IA-H indicates a water-wet is unfavorable for wettability alteration at 90 °C.
of wf

state of the Indiana limestone core before oil aging. However, IA-
H is
0.1 for the core K, which was saturated with the acidic
AN1.5MO and tested after oil aging at 120 °C for 2 weeks. The neg-
ative IA-H shows the limestone core becomes preferentially oil-wet
after oil aging. The reversal of the IA-H comparing core K and core J
demonstrates the effect of naphthenic acid presence and oil aging.
The adsorbed naphthenic acid on the limestone surface after the oil
aging process is responsible for the oil-wetness.
The oil recovery of Core A and Core J during spontaneous imbi-
bition experiments are also compared in the supplementary mate-
rial Figs. S1, S2 to show the effect of wettability. The water-wet
Core J has 27% oil recovery (of OOIP) in 5 M NaCl at room temper-
ature. However, slightly oil-wet Core A has only 6% oil recovery in
5 M NaCl at 90 °C. The oil recovery for Core A is very limited, even
though the higher temperature is a favorable condition for oil
recovery. This difference demonstrates that the capillary imbibi-
tion of the brine will be significantly inhibited if the core becomes
preferentially oil-wet.
3.2. Effect of Mg2+, SO2

4 and salinity
Wettability alteration in the ‘‘smart water” is quantified by the
additional oil recovery in the spontaneous imbibition experiment
after replacing the initial brine with the ‘‘smart water”. Capillary
imbibition is inhibited when the core is preferentially oil-wet.
Therefore, additional oil recovery in spontaneous imbibition test
is assumed to indicate wettability alteration.
Core A–C were all first tested in 5 M NaCl and then immersed in
4dSW, 4dSW0S and 4dSW0Mg respectively to investigate the role
of Mg2+ and SO2
4 in seawater-based brines (Fig. 1(a)). The oil
recovery in the high salinity 5 M NaCl brine is very limited in all
the cases (3–6%). A significant amount of oil is recovered after
switching to low salinity brines, clearly indicating wettability
alteration. An additional 28% (of OOIP) oil was observed for Core
C in 4dSW0Mg, which is 4x diluted seawater depleted in Mg2+.
Depletion of SO2
4 in the diluted seawater results in the lowest
additional oil recovery (20% of OOIP) among the three experiments.
For Core A, the additional oil recovery in 4dSW is 24%. A further 5%
A similar benefit of SO2
4 has also been observed in the sponta-
neous imbibition test with crude oils [1,15]. Strand et al. [1] pro-
posed a two-step mechanism explaining the effect of SO2
4 on
wettability alteration: (1) SO2
4 adsorption reduces the positive
charge of the carbonate surface so that the co-adsorption of Ca2+
is increased; (2) Ca2+ and adsorbed carboxylic acids (RCOO
) form
complexes (RCOO-Ca+), and the positively charged complexes
(RCOO-Ca+) are released from the positively charged rock surface
due to electrostatic repulsion. This proposed synergetic effect
between Ca2+ and SO2
4 also explains the observation that a higher
SO2
4 concentration does not always result in a higher oil recovery.
For instance, 4dSW0Mg (6 mM SO2
4 ) is more effective than LowSal
Na2SO4 (55 mM SO2
4 ) on wettability alteration. The two brines
have similar Na+ concentration (109 mM and 149 mM) and neither
of them contains Mg2+. However, 4dSW0Mg brine contains added
Ca2+ while LowSal Na2SO4 brine only has Ca2+ as a result of calcite
dissolution. It is possible that the limiting factor for the wettability
alteration in LowSal Na2SO4 is the low Ca2+ concentration.
To analyze the effect of Ca2+, the calcium concentrations were
measured for some of the brines after equilibrating them with cal-
cite rock and AN1.5MO (Fig. 3). Note that even though there are no
added calcium salts in the single-electrolyte brines, Ca2+ still exist
as a result of calcite dissolution. Naphthenic acid partitioning from
the AN1.5MO further promotes calcite dissolution, and results in
an even higher Ca2+ concentration. The calcium level (including
Ca2+ and all other Ca2+-complexes) was found to be roughly
8 mM in seawater-based brines and 5 mM in single-electrolyte
brines. The seawater-based brines have 3 mM added Ca2+ (shown
in Table 1), which is exactly the increment compared to the
single-electrolyte brines. Based on the hypothesis by Strand et al.
[1], SO2

4 is only effective when Ca
2+
is present because the release
of carboxylic acids from mineral is due to Ca-complexes formation.
The additional 3 mM Ca2+ in seawater-based brines can indeed
contribute significantly to the wettability alteration if the Ca2+ is
the limiting ion while SO2
4 is abundant. In that case, 55 mM
SO2
2

4 may be the same as 6 mM SO4 because the wettability alter-
ation is limited by relatively low Ca2+ concentration.
The presence of Mg2+ is unfavorable because the depletion of
Mg2+ (4dSW0Mg) results in higher additional oil recovery when
150 J. Song et al. / Journal of Colloid and Interface Science 563 (2020) 145–155

Fig. 1. Oil recovery history for core A–F in the spontaneous imbibition of (a) seawater-based brines and (b) single-electrolyte brines at 90 °C. Dashed lines mark the date when
brines were replaced.

Fig. 2. Summary of additional oil recovery compared to 5 M NaCl for all the seawater-based and single-electrolyte brines. All tests were performed with AN1.5MO and
Indiana limestone cores at 90 °C.
 J. Song et al. / Journal of Colloid and Interface Science 563 (2020) 145–155 151

Fig. 3. ICP-OES results of calcium concentration in most of the tested brines after equilibrating with calcite and AN1.5MO.

compared with the 4dSW case, and the addition of Mg2+ (MgSO4)
results in lower additional oil recovery when compared to the Na2-
SO4 test. These observations are contradictory to the results by
Zhang et al. [15] with crude oil spontaneous imbibition results
and the density function theory (DFT) calculations by Sakuma
et al. [29], who showed that substitution of MgSO4 ion pairs onto
calcite can increase the water affinity to calcite surface, while the
substitution of SO2

4 alone does not. We hypothesize that the effect
of Mg2+ is a function of the rock/oil/brine interaction in a specific
system. Zhang et al. utilized a system with crude oils from the
North Sea and chalk while a limestone/model oil system is used
in our study. The working mechanism of Mg2+ may associate with
the specific system of Zhang et al.’s work. On the other hand, our
observation agrees with the contact angle investigation by Shaik
et al. [41] They observed higher water advancing contact angles
(ACA) in MgSO4 brine (I = 0.164 M) compared to Na2SO4 brine
(I = 0.164 M) for oil-wet calcite plates. The ACA is initially 128°
in 5 M NaCl for the oil-aged Iceland spar plates. After the ‘‘smart
water” treatment at 120 °C, the ACA reduces to 26° in Lowsal
Na2SO4 and it only reduces to 49° in Lowsal MgSO4. The tested
brine and oil in the work by Shaik et al. have the same composition Fig. 4. Summary of zeta potential measurements and additional oil recoveries from
spontaneous imbibition experiments for all tested brines.
as that studied in this work.

3.3. Role of electrostatics: Surface complexation modeling and

disjoining pressure
Electrostatic interactions have been generally considered a key
factor in wettability alteration process triggered by low salinity
water. If there is sufficient electrostatic repulsion between rock/
brine and oil/brine interfaces, the disjoining pressure may stabilize
the brine film between the rock and oil, resulting in a water-wet
condition [27,42]. Fig. 4 summarizes the rock/brine and oil/brine
zeta potential in all tested brines along with the additional oil
recovery results. The zeta potentials are measured at 25 °C, which
may deviate from the actual value at 90 °C due to the temperature
effect on the ion binding to the limestone surface. However, the
values at ambient temperature are an important indicator of the
charge interaction between rock/oil/brine during the imbibition
test. Also, the trend of zeta potential variation is not expected to
change dramatically at different temperatures [43]. The oil/brine
zeta potential with AN1.5MO remains strongly negative in all
tested brines, while the Indiana limestone/brine zeta potential is
small in magnitude with either positive or negative sign. The inten-
sity of electrostatic interaction cannot be directly compared quan-
titatively among different cases because both the rock/brine and
oil/brine zeta potentials vary in different brines. Therefore, the
electrostatic component of disjoining pressure (Pelectrostatics) is then
calculated based on the zeta potential measurements to quantify
the electrostatic interaction. A model of two flat charged plates is
assumed. In the cases where the two charged plates have varying
surface potentials of the same sign or one has a zero potential,
the assumption of constant surface charge or constant surface
potential can be unreasonable at small separation (less than one
Debye length) [44]. Therefore, we choose to use the linear superpo-
sition approximation (LSA) to calculate the electrostatic compo-
nent of the disjoining pressure (Eq. (1)), which has been shown
to provide reasonable approximation lying between the results of
constant surface charge and constant surface potential cases [44].
P(d) is the electrostatic component of disjoining pressure as a func-
tion of separation, d, n is the number of cations (or anions) per unit
volume (/m3), k is Boltzmann’s constant, and T is the system tem-
perature in degrees Kelvin. The reduced potential is defined as
ci ¼ zew
kT
, wherew is the surface potential. The w can be calculated
from the zeta potential by assuming a slip plane distance
xs ¼ 0:0013
I0:5
[45], where I is the ionic strength of the brine. j is the
reciprocal Debye length, which is also a function of the ionic
strength I. Equation (1) is derived for calculating the interaction
in z:z symmetrical electrolytes such as NaCl and MgSO4. For the
152 J. Song et al. / Journal of Colloid and Interface Science 563 (2020) 145–155
case of Na2SO4, z = 1 is assumed because using the expression for a
1:1 electrolyte to calculate the interaction in a 1:2 electrolyte solu-
tion has been shown to be acceptable in the literature [46]. More-
over, 4dSW brine is also considered as a 1:1 electrolyte solution
because it primarily contains NaCl. The calculated Pelectrostatics
curves are summarized in Fig. 5.
PðdÞ ¼ 64nkTc1 c2 expð
jdÞ
The extent of wettability alteration aligns with the intensity of
rock/oil electrostatic repulsion if comparing within a group
(Na2SO4 > MgSO4 > NaCl and 4dSW0Mg > 4dSW > 4dSW0S). How-
ever, the trend is inconsistent when comparing the single-
electrolyte group and the seawater-based group. For example,
there is unfavorable rock/oil electrostatic attraction in 4dSW
because the Indiana limestone is positively charged. However,
the additional oil recovery in 4dSW is even higher than that in
LowSal MgSO4, where the rock is negatively charged and has elec-
trostatic repulsion to the oil. Moreover, the rock/oil electrostatic
repulsion should be stronger in LowSal Na2SO4 than 4dSW0Mg
according to the zeta potentials. But the additional oil recovery
in 4dSW0Mg is higher than that in LowSal Na2SO4.
The inconsistency between electrostatic repulsion and wettabil-
ity alteration indicates the positive effect of Ca2+. The Indiana lime-
stone rock has more positive zeta potentials in the seawater-based
brines because of the 3 mM added Ca2+. The added Ca2+ facilitates
removing the adsorbed carboxylic acids on the rock due to RCOO-
Ca+ complexation [1], even though it is unfavorable in terms of
electrostatics. The effect of Ca2+ has also been discussed in previous
section when comparing the oil recovery results in 4dSW0Mg and
LowSal Na2SO4. The higher Ca2+ concentration is responsible for
the better outcome in 4dSW0Mg, even though SO2
4 concentration
is lower in 4dSW0Mg.
Disjoining pressure calculations and zeta potential measure-
ments only consider the average of surface charge and fail to pro-
vide information about particular charged species on the limestone
surface. Therefore, a surface complexation model (SCM) is also
employed to provide insights of low-salinity-triggered wettability
alteration. The surface concentration of all the charged surface spe-
cies can be calculated from a SCM which captures the electrostatic
properties of the mineral. A previously proposed SCM for Indiana
limestone [36] is used without re-fitting any of the model param-
eters. The impact of naphthenic acid partitioning from AN1.5MO to
Fig. 5. Electrostatic component of disjoining pressure for Indiana limestone/AN1.5MO in various tested brines. Calculation is based on the zeta potential results in this work.
the brine is accounted for by assuming a certain concentration of
model naphthenic acid (pKa = 4.8) [37,38] in the brine so that the
calculated brine pH after phase equilibrium matches with the mea-
sured pH.
The experimental and SCM-predicted values of the calcium con-
centration and zeta potential for four tested brines are summarized
in Fig. 6. The general trends for both calcium concentration and
ð1Þ
zeta potential are successfully captured by the model when com-
paring these tested brines. However, calcium concentration is
always under-predicted by roughly 2–3 mM and the rock/brine
zeta potentials are over-predicted by 4 and 7 mV for 4dSW and
LowSal MgSO4, respectively. The more positive zeta potential cal-
culated from the SCM compared with experimental results is likely
because naphthenic acid adsorption on the rock surface is
neglected in the SCM. The impact of naphthenic acid is only consid-
ered in the bulk brine phase by including a model naphthenic acid
in the SCM, but it is not considered at the interface. Adsorption of
the naphthenic acid on limestone surface will result in a more neg-
ative rock/brine zeta potential.
Since the SCM successfully predicts the qualitative trend of cal-
cium concentration and zeta potential, it is further used to calcu-
late the surface concentration of charged species on Indiana
limestone surface (Fig. 7). Surface charge density contribution (C/
m2) from various species is calculated by multiplying the surface
concentration (mol/m2) of a specie and its charge in coulomb
(C/mol). The summation of all the positive and negative contribu-
tions yields the net surface charge density of the rock. Although the
zeta potential in 5 M NaCl cannot be measured due to the electrical
conductivity limitation, the SCM prediction in 5 M NaCl is provided
to help understand the effect of salinity. >CO3H  A
5.25 and >SiO

represent the organic and inorganic impurities on Indiana lime-
stone surface which accounts for the charge difference of natural
limestone and synthetic calcite (Table 3). The same assumption is
used in the previous work by Song et al. [36] and the readers are
referred to their paper for detailed justification. The charge contri-
bution from these two species remain constant in all the tested
brines.
A significantly different mix of species is found for the rock min-
eral in 5 M NaCl and other brines in Fig. 7. In the high salinity initial
brine, 91% of major surface site >CaOH
0.75 is converted to the Na-
complex >CaOH  Na+0.25, and 67% of the other major site
>CO3H+0.75 is converted to the Cl-complex >CO3H  Cl
0.25. Even
 J. Song et al. / Journal of Colloid and Interface Science 563 (2020) 145–155 153

Fig. 6. Comparison between measurements and prediction from surface complexation model (SCM) for Ca concentration and rock/brine zeta potential in various equilibrated
brines.

Fig. 7. Comparison of surface charge density contribution from various species in different brines based on surface complexation modeling results.

though Na+ and Cl
are generally considered indifferent to the cal-
cite surface, their surface interaction with calcite is still significant
if present in a high concentration such as 5 M [12,36]. Based on this
observation, the calcite surface interaction with Na+ in high salinity
brine is hypothesized to be a critical mechanism responsible for
the oil-wetness. A similar finding has also been reported by Liu,
et al. [47], who performed a comprehensive study of calcite wetta-
bility alteration by quantum molecular dynamics (MD) simulation.
The MD simulation reveals that Na+ resides closer to the calcite
surface compared to the divalent ions Ca2+and Mg2+, and it modi-
fies the interfacial water structure. As a result, a high Na+ concen-
tration reduces the adhesion energy of calcite/water and makes
calcite less water-wet. In another MD simulation study by Koleini
et al. [48], the Na+ ion is also found to be in the Stern layer, and clo-
ser to the calcite surface than the divalent ions. A direct visualiza-
tion study confirming that Na+ indeed exists in the Stern layer is
reported by Ricci et al. [49] using atomic force microscopy
(AFM). Moreover, another AFM study by Guo and Kovscek [50] also
demonstrates that Na+ reduces the decay length of short-range
non-DLVO repulsion and disturbs the innermost water structure.
These previous MD simulations and AFM studies are consistent
with our SCM calculation of charged species concentration, indicat-
ing that the reduction of NaCl concentration triggers wettability
alteration due to removal of Na+ from calcite surface. This also
agrees well with the spontaneous imbibition results. Interestingly,
there is no clear link between additional oil recovery and the other
charged surface species. Moreover, the contribution of CO2
3 to sur-
face charge is minimal in all the tested brines, even though CO2
3 is
considered to be a potential determining ion for the calcite surface.
One reason why CO2
3 ions show almost no impact is that its bulk
concentration is too low (~0.01 mM) in all the tested brines. CO2

3 is
only present as a result of calcite dissolution and it is mostly con-
verted to HCO
3 at the pH around 8.0.
4. Conclusions
The effect of salinity, Mg2+ and SO2

4 on ‘‘smart water”-induced
wettability alteration is examined using Indiana limestone cores
and a model oil with added naphthenic acid. Low salinity brines
with or without added divalent ions are found to trigger wettabil-
154 J. Song et al. / Journal of Colloid and Interface Science 563 (2020) 145–155
ity alteration and improve oil recovery in the spontaneous imbibi-
tion test. The SO2

4 in the low salinity brines further promotes wet-
tability alteration, while the presence of Mg2+ is unfavorable for
wettability alteration. Such observation is consistent in both
single-electrolyte brines and seawater-based brines. Moreover,
other promotive mechanisms other than ionic strength and SO2
4 Appendix A. Supplementary material
is found when comparing single-electrolyte brines and seawater-
based brines. The additional oil recovery in LowSal Na2SO4, a
single-electrolyte solution with 55 mM SO2
4 is lower than that in
4dSW0Mg, a seawater-based brine with only 6 mM SO2
4 . Both bri-
nes have identical ionic strength, very similar Na+ concentration
and no Mg2+. A higher Ca2+ concentration in 4dSW0Mg is believed
to facilitate wettability alteration, as Ca2+ has also been identified
as a wettability modifier in the literature [1,15].
The role of electrostatic interaction is then investigated by zeta
potential measurements, disjoining pressure calculation, and sur-
face complexation modeling. The intensity of rock/oil electrostatic
repulsion aligns with the additional oil recovery in spontaneous
imbibition for the three single-electrolyte brines (Na2SO4 > MgSO4 >-
NaCl). However, electrostatic repulsion does not explain the better
oil recovery in seawater-based brines. Wettability alteration is
observed in 4dSW in which the rock/oil have electrostatic attrac-
tion. The better performance of seawater-based brines is explained
by the higher Ca2+ concentration, which promotes the wettability
alteration via removal of adsorbed hydrophobic carboxylic acids.
According to the SCM calculation, over 90% of the major surface
site >CaOH
0.75 converts to its Na+-binding complex
>CaOH  Na+0.25 in the high salinity 5 M NaCl brine. The reduction
of >CaOH  Na+0.25 concentration is significant in all the low salin-
ity brines. The interaction with Na+ is believed to be responsible for
the oil-wetness for the limestone surface. Such hypothesis is sup-
ported by MD simulation studies [47,48] and AFM studies
[49,50], which show that Na+ is in the closest proximity to calcite
surface and cause the calcite/water adhesion energy to decrease.
Therefore, one of the key mechanisms of low-salinity-induced wet-
tability alteration is the reduction of Na+ near the rock surface.
Further investigation is required to illuminate how the divalent
ions affect limestone wettability in low salinity brines. Moreover, a
different optimal low salinity or ionic strength may exist for differ-
ent systems. The ionic strength of low salinity brines is kept con-
stant as 0.164 M in this work but the impact of varying ionic
strength should also be studied in the future.
CRediT authorship contribution statement
Jin Song: Methodology, Software, Formal analysis, Investigation,
Writing - original draft. Qing Wang: Investigation. Imran Shaik:
Methodology, Validation. Maura Puerto: Methodology, Resources,
Writing - review & editing. Prem Bikkina: Writing - review & edit-
ing. Clint Aichele: Methodology, Validation. Sibani L. Biswal: Con-
ceptualization, Methodology, Resources, Writing - review & editing,
Supervision, Project administration, Funding acquisition. George J.
Hirasaki: Conceptualization, Methodology, Resources, Writing -
review & editing, Supervision, Funding acquisition.
Declaration of Competing Interest
The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared
to influence the work reported in this paper.
Acknowledgement
We acknowledge the financial support from Abu Dhabi National
Oil Company (ADNOC) and the Rice University Consortium for Pro-
cesses in Porous Media (Houston, TX, USA). We thank Mr. Leilei
Zhang for his help in the EDX measurement for the core sample.
We also thank Dr. Yongchao Zeng, Dr. Pengfei Dong and Mr. Zhuq-
ing Zhang for their insightful discussions.
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.jcis.2019.12.040.
References
[1] S. Strand, E.J. Høgnesen, T. Austad, Wettability alteration of carbonates—effects
of potential determining ions (Ca2+ and SO42
) and temperature, Colloids
Surf. Physicochem. Eng. Asp. 275 (1–3) (2006) 1–10, https://doi.org/10.1016/
j.colsurfa.2005.10.061.
[2] G. Hirasaki, D.L. Zhang, Surface chemistry of oil recovery from fractured, oil-
wet, carbonate formations, SPE J. 9 (02) (2004) 151–162, https://doi.org/
10.2118/88365-PA.
[3] D. Leslie Zhang, S. Liu, M. Puerto, C.A. Miller, G.J. Hirasaki, Wettability
alteration and spontaneous imbibition in oil-wet carbonate formations, J. Pet.
Sci. Eng. 52 (1–4) (2006) 213–226, https://doi.org/10.1016/j.
petrol.2006.03.009.
[4] G. Sharma, K.K. Mohanty, Wettability Alteration in High Temperature and High
Salinity Carbonate Reservoirs; Society of Petroleum Engineers, 2011,
https://doi.org/10.2118/147306-MS.
[5] M. Mohammed, T. Babadagli, Wettability alteration: a comprehensive review
of materials/methods and testing the selected ones on heavy-oil containing
oil-wet systems, Adv. Colloid Interface Sci. 220 (2015) 54–77, https://doi.org/
10.1016/j.cis.2015.02.006.
[6] D.C. Standnes, T. Austad, Wettability alteration in carbonates, Colloids Surf.
Physicochem. Eng. Asp. 216 (1–3) (2003) 243–259, https://doi.org/10.1016/
S0927-7757(02)00580-0.
[7] A. Kumar, A. Mandal, Critical investigation of zwitterionic surfactant for
enhanced oil recovery from both sandstone and carbonate reservoirs:
adsorption, wettability alteration and imbibition studies, Chem. Eng. Sci. 209
(2019) 115222, https://doi.org/10.1016/j.ces.2019.115222.
[8] N. Pal, N. Saxena, K.V. Divya Laxmi, A. Mandal, Interfacial behaviour,
wettability alteration and emulsification characteristics of a novel
surfactant: implications for enhanced oil recovery, Chem. Eng. Sci. 187
(2018) 200–212, https://doi.org/10.1016/j.ces.2018.04.062.
[9] P. Pillai, A. Mandal, Wettability modification and adsorption characteristics of
imidazole-based ionic liquid on carbonate rock: implications for enhanced oil
recovery, Energy Fuels 33 (2) (2019) 727–738, https://doi.org/10.1021/acs.
energyfuels.8b03376.
[10] H. Ding, S. Rahman, Experimental and theoretical study of wettability
alteration during low salinity water flooding-an state of the art review,
Colloids Surf. Physicochem. Eng. Asp. 520 (2017) 622–639, https://doi.org/
10.1016/j.colsurfa.2017.02.006.
[11] M.A. Sohal, G. Thyne, E.G. Søgaard, Review of recovery mechanisms of ionically
modified waterflood in carbonate reservoirs, Energy Fuels 30 (3) (2016) 1904–
1914, https://doi.org/10.1021/acs.energyfuels.5b02749.
[12] J. Song, Y. Zeng, L. Wang, X. Duan, M. Puerto, W.G. Chapman, S.L. Biswal, G.J.
Hirasaki, Surface complexation modeling of calcite zeta potential
measurements in brines with mixed potential determining ions (Ca2+, CO2- 3 ,
Mg2+, SO2- 4 ) for characterizing carbonate wettability, J. Colloid Interface Sci.
506 (2017) 169–179, https://doi.org/10.1016/j.jcis.2017.06.096.
[13] T. Austad, S.F. Shariatpanahi, S. Strand, C.J.J. Black, K.J. Webb, Conditions for a
low-salinity enhanced oil recovery (EOR) effect in carbonate oil reservoirs,
Energy Fuels 26 (1) (2012) 569–575, https://doi.org/10.1021/ef201435g.
[14] T. Austad, S.F. Shariatpanahi, S. Strand, H. Aksulu, T. Puntervold, Low salinity
EOR effects in limestone reservoir cores containing anhydrite: a discussion of
the chemical mechanism, Energy Fuels 29 (11) (2015) 6903–6911, https://doi.
org/10.1021/acs.energyfuels.5b01099.
[15] P. Zhang, M.T. Tweheyo, T. Austad, Wettability alteration and improved oil
recovery by spontaneous imbibition of seawater into chalk: impact of the
potential determining ions Ca2+, Mg2+, and SO42
, Colloids Surf.
Physicochem. Eng. Asp. 301 (1–3) (2007) 199–208, https://doi.org/10.1016/
j.colsurfa.2006.12.058.
[16] S.J. Fathi, T. Austad, S. Strand, Effect of water-extractable carboxylic acids in
crude oil on wettability in carbonates, Energy Fuels 25 (6) (2011) 2587–2592,
https://doi.org/10.1021/ef200302d.
[17] S.J. Fathi, T. Austad, S. Strand, Water-based Enhanced Oil Recovery (EOR) by
‘‘smart water”: optimal ionic composition for EOR in carbonates, Energy Fuels
25 (11) (2011) 5173–5179, https://doi.org/10.1021/ef201019k.
[18] S.F. Shariatpanahi, P. Hopkins, H. Aksulu, S. Strand, T. Puntervold, T. Austad,
Water based EOR by wettability alteration in dolomite, Energy Fuels 30 (1)
(2016) 180–187, https://doi.org/10.1021/acs.energyfuels.5b02239.
[19] A.A. Yousef, S.H. Al-Saleh, A. Al-Kaabi, M.S. Al-Jawfi, Laboratory investigation
of the impact of injection-water salinity and ionic content on oil recovery from
carbonate reservoirs, SPE Reserv. Eval. Eng. 14 (05) (2011) 578–593, https://
doi.org/10.2118/137634-PA.
 J. Song et al. / Journal of Colloid and Interface Science 563 (2020) 145–155
[20] A. Al Harrasi, R.S. Al-maamari, S.K. Masalmeh, Laboratory Investigation of Low
Salinity Waterflooding for Carbonate Reservoirs; Society of Petroleum
Engineers, 2012. https://doi.org/10.2118/161468-MS.
[21] K.K. Mohanty, S. Chandrasekhar, Wettability Alteration with Brine
Composition in High Temperature Carbonate Reservoirs; Society of
Petroleum Engineers, 2013, https://doi.org/10.2118/166280-MS.
[22] A. Gandomkar, M.R. Rahimpour, Investigation of low-salinity waterflooding in
secondary and tertiary enhanced oil recovery in limestone reservoirs, Energy
Fuels 29 (12) (2015) 7781–7792, https://doi.org/10.1021/acs.
energyfuels.5b01236.
[23] H. Mahani, A.L. Keya, S. Berg, R. Nasralla, Electrokinetics of carbonate/brine
interface in low-salinity waterflooding: effect of brine salinity, composition,
rock type, and PH on f-potential and a surface-complexation model, SPE J. 22
(01) (2017) 053–068, https://doi.org/10.2118/181745-PA.
[24] H. Sharma, K.K. Mohanty, An experimental and modeling study to investigate
brine-rock interactions during low salinity water flooding in carbonates, J. Pet.
Sci. Eng. (2017), https://doi.org/10.1016/j.petrol.2017.11.052.
[25] A.A. Eftekhari, K. Thomsen, E.H. Stenby, H.M. Nick, Thermodynamic analysis of
chalk–brine–oil interactions, Energy Fuels 31 (11) (2017) 11773–11782,
https://doi.org/10.1021/acs.energyfuels.7b02019.
[26] G.J. Hirasaki, Wettability: fundamentals and surface forces, SPE Form. Eval. 6
(02) (1991) 217–226, https://doi.org/10.2118/17367-PA.
[27] Q. Xie, A. Saeedi, E. Pooryousefy, Y. Liu, Extended DLVO-based estimates of
surface force in low salinity water flooding, J. Mol. Liq. 221 (2016) 658–665,
https://doi.org/10.1016/j.molliq.2016.06.004.
[28] M.J. Alshakhs, A.R. Kovscek, Understanding the role of brine ionic composition
on oil recovery by assessment of wettability from colloidal forces, Adv. Colloid
Interface Sci. 233 (2016) 126–138, https://doi.org/10.1016/j.cis.2015.08.004.
[29] H. Sakuma, M.P. Andersson, K. Bechgaard, S.L.S. Stipp, Surface tension
alteration on calcite, induced by ion substitution, J. Phys. Chem. C 118 (6)
(2014) 3078–3087, https://doi.org/10.1021/jp411151u.
[30] Sci. Rep. 6 (1) (2016), https://doi.org/10.1038/srep28854.
[31] L. Chen, G. Zhang, L. Wang, W. Wu, J. Ge, Zeta potential of limestone in a large
range of salinity, Colloids Surf. Physicochem. Eng. Asp. 450 (2014) 1–8, https://
doi.org/10.1016/j.colsurfa.2014.03.006.
[32] H. Mahani, A.L. Keya, S. Berg, W.-B. Bartels, R. Nasralla, W.R. Rossen, Insights
into the mechanism of wettability alteration by low-salinity flooding (LSF) in
carbonates, Energy Fuels 29 (3) (2015) 1352–1367, https://doi.org/10.1021/
ef5023847.
[33] S. Rashid, M.S. Mousapour, S. Ayatollahi, M. Vossoughi, A.H. Beigy, Wettability
alteration in carbonates during ‘‘smart waterflood”: underlying mechanisms
and the effect of individual ions, Colloids Surf. Physicochem. Eng. Asp. 487
(2015) 142–153, https://doi.org/10.1016/j.colsurfa.2015.09.067.
[34] M. Lashkarbolooki, S. Ayatollahi, Investigating injection of low salinity brine in
carbonate rock with the assist of works of cohesion and adhesion and
spreading coefficient calculations, J. Pet. Sci. Eng. 161 (2018) 381–389, https://
doi.org/10.1016/j.petrol.2017.12.010.
[35] T. Fan, J.S. Buckley, Acid number measurements revisited, SPE J. 12 (04) (2007)
496–500, https://doi.org/10.2118/99884-PA.
[36] J. Song, S. Rezaee, L. Zhang, Z. Zhang, M. Puerto, O.B. Wani, F. Vargas, S.
Alhassan, S.L. Biswal, G.J. Hirasaki, Characterizing the influence of organic
155
carboxylic acids and inorganic silica impurities on the surface charge of
natural carbonates using an extended surface complexation model, Energy
Fuels (2019), https://doi.org/10.1021/acs.energyfuels.8b03896.
[37] T.E. Havre, J. Sjöblom, J.E. Vindstad, Oil/water-partitioning and interfacial
behavior of naphthenic acids, J. Dispers. Sci. Technol. 24 (6) (2003) 789–801,
https://doi.org/10.1081/DIS-120025547.
[38] R. Huang, P. Chelme-Ayala, Y. Zhang, M. Changalov, M. Gamal El-Din,
Investigation of dissociation constants for individual and total naphthenic
acids species using ultra performance liquid chromatography ion mobility
time-of-flight mass spectrometry analysis, Chemosphere 184 (2017) 738–746,
https://doi.org/10.1016/j.chemosphere.2017.06.067.
[39] J.S. Buckley, K. Takamura, N.R. Morrow, Influence of electrical surface charges
on the wetting properties of crude oils, SPE Reserv. Eng. 4 (03) (1989)
332–340, https://doi.org/10.2118/16964-PA.
[40] P.V. Brady, N.R. Morrow, A. Fogden, V. Deniz, N. Loahardjo, Winoto,
Electrostatics and the low salinity effect in sandstone reservoirs, Energy
Fuels 29 (2) (2015) 666–677, https://doi.org/10.1021/ef502474a.
[41] I.K. Shaik, J. Song, S.L. Biswal, G.J. Hirasaki, P.K. Bikkina, C.P. Aichele, Effect of
brine type and ionic strength on the wettability alteration of naphthenic-acid-
adsorbed calcite surfaces, https://doi.org/10.1016/j.petrol.2019.106567.
[42] Y.-L. Chen, L. Zhang, J. Song, G. Jian, G. Hirasaki, K. Johnston, S.L. Biswal, Two-
step adsorption of a switchable tertiary amine surfactant measured using a
quartz crystal microbalance with dissipation, Langmuir 35 (3) (2019) 695–
701, https://doi.org/10.1021/acs.langmuir.8b03150.
[43] H. Mahani, R. Menezes, S. Berg, A. Fadili, R. Nasralla, D. Voskov, V. Joekar-
Niasar, Insights into the impact of temperature on the wettability alteration by
low salinity in carbonate rocks, Energy Fuels 31 (8) (2017) 7839–7853, https://
doi.org/10.1021/acs.energyfuels.7b00776.
[44] J. Gregory, Interaction of unequal double layers at constant charge, J. Colloid
Interface Sci. 51 (1) (1975) 44–51, https://doi.org/10.1016/0021-9797(75)
90081-8.
[45] F. Heberling, T.P. Trainor, J. Lützenkirchen, P. Eng, M.A. Denecke, D. Bosbach,
Structure and reactivity of the calcite-water interface, J. Colloid Interface Sci.
354 (2) (2011) 843–857, https://doi.org/10.1016/j.jcis.2010.10.047.
[46] J. Israelachvili, Intermolecular and Surface Forces, third ed., Elsevier (2011).
[47] J. Liu, O.B. Wani, S.M. Alhassan, S.T. Pantelides, Wettability alteration and
enhanced oil recovery induced by proximal adsorption of Na +, Cl
, Ca 2 +, Mg
2 +, and SO 4 2
ions on calcite, Phys. Rev. Appl. 10 (3) (2018), https://doi.org/
10.1103/PhysRevApplied.10.034064.
[48] M.M. Koleini, M.F. Mehraban, S. Ayatollahi, Effects of low salinity water on
calcite/brine interface: a molecular dynamics simulation study, Colloids Surf.
Physicochem. Eng. Asp. 537 (2018) 61–68, https://doi.org/10.1016/
j.colsurfa.2017.10.024.
[49] M. Ricci, P. Spijker, F. Stellacci, J.-F. Molinari, K. Voïtchovsky, Direct
visualization of single ions in the stern layer of calcite, Langmuir 29 (7)
(2013) 2207–2216, https://doi.org/10.1021/la3044736.
[50] H. Guo, A.R. Kovscek, Investigation of the effects of ions on short-range non-
DLVO forces at the calcite/brine interface and implications for low salinity oil-
recovery processes, J. Colloid Interface Sci. 552 (2019) 295–311, https://doi.
org/10.1016/j.jcis.2019.05.049.