microemulsion-enhanced displacement of oil in porous media ... › view › uq:676409 ›...
TRANSCRIPT
Accepted Manuscript
Title: Microemulsion-enhanced Displacement of Oil in PorousMedia Containing Carbonate Cements
Authors: Tianzhu Qin, Gina Javanbakht, Lamia Goual,Mohammad Piri, Brian Towler
PII: S0927-7757(17)30665-9DOI: http://dx.doi.org/doi:10.1016/j.colsurfa.2017.07.017Reference: COLSUA 21791
To appear in: Colloids and Surfaces A: Physicochem. Eng. Aspects
Received date: 28-3-2017Revised date: 6-7-2017Accepted date: 7-7-2017
Please cite this article as: Tianzhu Qin, Gina Javanbakht, Lamia Goual, MohammadPiri, Brian Towler, Microemulsion-enhanced Displacement of Oil in Porous MediaContaining Carbonate Cements, Colloids and Surfaces A: Physicochemical andEngineering Aspectshttp://dx.doi.org/10.1016/j.colsurfa.2017.07.017
This is a PDF file of an unedited manuscript that has been accepted for publication.As a service to our customers we are providing this early version of the manuscript.The manuscript will undergo copyediting, typesetting, and review of the resulting proofbefore it is published in its final form. Please note that during the production processerrors may be discovered which could affect the content, and all legal disclaimers thatapply to the journal pertain.
1
Microemulsion-enhanced Displacement of Oil in Porous Media Containing Carbonate
Cements
Tianzhu Qin,1 Gina. Javanbakht,2 Lamia Goual,3* Mohammad Piri,4 and Brian Towler5
1-4Department of Chemical and Petroleum Engineering, University of Wyoming, Laramie, WY 82071,
USA
5School of Chemical Engineering, University of Queensland, Brisbane St Lucia, QLD 4072, Australia
* Corresponding author: Lamia Goual ([email protected])
Graphical abstract
Brine
Surfactant
Microemulsion
Highlights
The impact of rock characteristics on the complex fluid-rock interactions is investigated
MEs enhance the immiscible displacement (or mobilization) of NAPL by decreasing IFT
and oil droplet size
MEs promote the desorption of NAPL from rock surfaces (or solubilization) thereby
altering their wettability
At low concentration, MEs outperform surfactants in rocks containing carbonate cements
2
ABSTRACT
Subsurface porous formations containing nonaqueous phase liquids (NAPLs) such as crude oils are often
targets for surfactant flooding during enhanced oil recovery (EOR) or aquifer remediation processes
designed to mobilize and solubilize oil. Recent studies suggested that putting surfactants into micro-
emulsified state prior to injection might improve their performance. Most of these studies proved that
microemulsion (ME) efficiency depends on test conditions and the proper selection of their chemical
formulations and brine chemistry. However, the impact of rock characteristics on the complex fluid-rock
interactions is still unclear, especially in heterogeneous rocks. The goal of this fundamental study was to
examine the effect of MEs on oil displacement in three different aged rocks (Berea, Edwards, and
Tensleep) and identify the test conditions in which MEs outperform surfactants. The effectiveness of
surfactants and MEs was evaluated at two concentrations from different sets of spontaneous imbibition
tests and petrographic analyses. Several mechanisms such as reduction of interfacial tension (IFT), oil
emulsification, and wettability alteration were responsible for the improved recovery. Wettability
alteration of aged cores by surfactants and MEs led to an early oil removal by spontaneous imbibition
while emulsification increased the ultimate amount produced. Both additives lowered the IFT from 12 to
less than 1 mN/m. However, high ME concentration was able to decrease the size of oil droplets by one
order of magnitude, significantly enhancing oil mobilization in all three rocks. The solubilization
capability of MEs was superior in Tensleep due to their unique ability to penetrate dolomite cements and
alter their wettability.
Keywords: Microemulsion, surfactant, oil, wettability, imbibition, carbonate cement.
1. INTRODUCTION
Surfactant flooding or flushing processes hold particular promise for cleaning up or recovering the oil left
in subsurface formations after secondary waterflooding or traditional pump and treat methods. The oil
phase is usually entrapped in the form of ganglia within the fine pores of reservoir rocks due to large
capillary forces.1 With their amphiphilic structure, surfactants to accumulate at oil/water interfaces and
lower the IFT, enhancing the mobilization of small and deformable oil droplets through pores and
throats.2 They may also adsorb on rock surfaces and promote the micellar solubilization of surface-active
oil species (such as asphaltenes), thereby restoring the wettability of the rock. The extent by which
wettability is altered depends on the type of surfactant and the rock properties.3 Surfactants with an
intermediate hydrophile-lipophile balance (HLB) number favor the mobilization and, to a lower extent,
the solubilization of oil in porous media.4 Mobilization in this case manifests by the in-situ formation of
bi-continuous Winsor Type III microemulsions, consisting of bilayer vesicles in equilibrium with oil and
water phases.5
A major limitation of surfactant flooding is the chemical losses encountered within the first few inches of
3
the formation due to adsorption on minerals. Recent studies suggest that putting surfactants into micro-
emulsified state prior to injection may improve their performance. Because MEs exhibit lower adsorption
propensity on mineral rocks,6 better demulsification capacity,7 and higher stability at elevated temperature
and brine pH/salinity compared to surfactant solutions,8 they are considered as good carrier systems that
can deliver surfactants deeper into the formation,9 and maintain the surfactant concentration in the
displacement front.10 Microemulsions are dynamic systems in which the interface is continuously and
spontaneously fluctuating.11 They can form a wide variety of structures and phases depending upon the
proportions of their components and other characteristics.12
Microemulsions typically consist of thermodynamically stable, isotropic, and macroscopically
homogeneous dispersions of two immiscible fluids (hydrocarbon and brine) stabilized by a surfactant and
co-surfactant.13-15 The presence of both surfactant and co-surfactant has a dual effect on lowering the
IFT.16,17 The hydrocarbon phase (also called carrier fluid) could be C6-C18 alkanes,18 biodiesel,19 or natural
oils such as pine oil,20 palm oil,21 and terpenes like d-limonene,22,23 which are usually preferred for their
solvent qualities and biodegradability.24 The surfactant should be able to create stable microemulsions
when used in specific proportions. The most commonly used ones are biodegradable nonionic surfactants
with an HLB number between 8 and 18. Examples include linear alcohol ethoxylates with various
ethylene oxide numbers (EON),23,25 alkyl polyglucosides,21 as well as those from the tergitol family,18 and
polysorbate (or tween) family.19 The co-surfactant consists of short or medium chain primary, secondary
and tertiary alcohols with 1-20 carbon atoms such as 2-propanol,18,23 isoamyl alcohol,26 and glycerol
monooleate.27 The alcohol serves as a coupling agent between the solvent and the surfactant, thereby
stabilizing the microemulsion by increasing the total interfacial area,28 enhancing the surfactant mass
transfer to the interface and decreasing the repulsive forces between its hydrophilic groups.9 Furthermore,
the alcohol lowers the freezing point and the viscosity of MEs to prevent the formation of rigid structures
like liquid crystals and gels at the interface.18
Microemulsions have had several applications in the environmental and petroleum industries. Recent
investigations have demonstrated that in-situ flushing with surfactant/alcohol mixtures enhances the
remediation of NAPL-contaminated aquifers. The best surfactants had HLB numbers between 12 and
14.29 Soil core data indicated that approximately 90-95% of the most prevalent NAPL constituents were
removed.30 Dilute concentrations of microemulsions in brine were also found to be effective in
remediating damaged wells during drilling and stimulation treatments. The enhanced fluid recovery and
relative permeability were the result of IFT and CA alterations.18,31 Hydraulic fracturing operations,
particularly in tight gas reservoirs, have seen a continuous surge of interest in utilizing MEs to reduce
flowback pressure and allow piston-like flow during the recovery of injected fluids, resulting in minimal
losses.10 Unlike wellbore remediation methods, microemulsion flooding involves more than one well and
can sweep a much larger area of the reservoir. Several microemulsion formulations in tertiary oil recovery
experiments were able to recover a significant portion of the residual oil by drastically reducing the IFT to
ultra-low values, enhancing the spontaneous emulsification and displacement of the oil in porous media.
Early core flooding experiments with Berea sandstone and microemulsion slugs of finite size showed that
4
the residual oil left behind the flood decreased with increased surfactant concentration and flooding rate.32
Moreover, higher ME viscosity may enhance the flooding efficiency in reservoir sandstones.26 Other
studies with sandpacks indicated that MEs could produce 4.3% more oil than typical polymer flooding
due to the ultra-low IFT values.27 More recent work with aged carbonate rocks examined the influence of
wettability alteration on oil recovery using microemulsion solutions (7 wt% in brine) formulated with
different types of surfactants.25 Cationic surfactants had the highest potential for wettability alteration
from neutral-wet to water-wet, leading to incremental recoveries over 26% after conventional
waterflooding. Exceptionally high oil recovery factors (over 90%) were also obtained with Tensleep rock
by spontaneous imbibition of brine containing complex nanofluids.33
To better understand the wettability alteration of aged calcite surfaces, molecular dynamics (MD)
simulations were performed with model oil molecules and a microemulsion based on a linear alcohol
ethoxylate.34 The simulations revealed that polar molecules of oil (such as naphthenic acids) irreversibly
adsorbed on calcite and served as a platform upon which further non-polar molecules could accumulate
and form a weakly bound adsorbed layer. The hydrocarbon phase of MEs was effective at swelling the
adsorbed layer and reducing its interaction with polar components, leading to easier removal when
subjected to a water flow. MD flow simulations with MEs were also performed in aged graphite
capillaries to represent kerogen pores in shale.35 The hydrocarbon and surfactant molecules played
separate roles that were both essential in solubilizing the adsorbed oil film. The surfactant acted as a
linker that diminished slip effects at the oil/water interface, thereby enabling the hydrocarbon solvent to
penetrate the oil film and displace its molecules.
Almost all existing laboratory studies indicate that MEs have a superior ability to mobilize and solubilize
oil in porous media, compared to surfactants alone. Yet, microemulsion flooding is considered
controversial in EOR and has been debated for many years. The use of ME flooding is mainly dependent
on fluid-rock interactions, which are complex especially in heterogeneous rocks. Furthermore, the impact
of rock petrophysical attributes on these interactions is still unclear. The objective of this study was to
examine the effect of MEs on oil recovery or cleanup from different aged rocks and identify the test
conditions in which MEs outperform surfactants alone. We used two different outcrops and a
heterogeneous reservoir rock to demonstrate the dependence of ME performance on surface mineralogy,
roughness, and wettability state of oil-bearing rocks. Novel insights are gained on the unique ability of
MEs to alter the wettability of carbonate cements often ubiquitous in heterogeneous rocks.
2. MATERIALS AND METHODS
2.1 Rocks
The rocks consist of two outcrops: Berea sandstone and Edwards limestone, and a reservoir rock from
Tensleep formation in Wyoming. The rocks were drilled and cut into small cores 1 inch in diameter and 5
cm in length. Their physical properties are provided in Table 1S of the Supporting Information. Berea
and Edwards cores were baked at 110 C for 24 hours to remove any water. Tensleep cores were first
cleaned by flooding them with a 50/50 volume mixture of toluene/methanol at 1000 psi and 80 C until
the produced solution was colorless, then baked at 120 C for 24 hours in a constant temperature oven
5
(DKN402, Yamato) to remove the solvents. The porosity and Klinkenberg-corrected permeability of these
core samples were measured simultaneously by an automated porosimeter and permeameter (AP-608,
Coretest system).
2.1.1 Mineralogy and pore size distribution
The mineralogy of core samples was evaluated using a QEMSCAN 650F from FEI. Imaging by this
technique was captured at 25 kV and 6.2 nA and the beam was optimized within 0.05 nA of 10.00 nA. X-
ray detectors were used to generate 3.0 3.0 mm mineralogy maps with an optical resolution of 0.73
microns per pixel. The pore size distributions of these cores with 5 mm diameter and 20 mm length were
measured using an FEI HeliScan micro-CT scanner with a resolution of 1.5 µm per pixel at 100 kV and
52 µA. The digital images were analyzed by using Avizo software.
2.1.2 Petrographic thin section analysis
In order to analyze rock wettability after the spontaneous imbibition tests with different solutions, three
sets of petrographic thin sections of Edwards, Berea and Tensleep rocks were prepared by Wagner
petrographic company. The thin sections of the clean rocks were also provided as a reference. The size of
the thin sections was 46 24 mm. Blue epoxy impregnation, K-feldspar stain, plagioclase stain and
calcite stain were applied on the thin sections. A petrographic microscope (Zeiss AXIO, Scope. A1) was
used for visualization of thin sections. AXIO vision software was also selected to analyze the thin section
images.
2.2 Fluids
We used a medium gravity crude oil from Tensleep formation in Wyoming as a NAPL phase (Table 2S).
The oil was first centrifuged at 6000 rpm for one hour and then filtered with 0.5 μm filter. The brine
phase consisted of 0.1 M NaCl in distilled water. A linear alcohol ethoxylate, Biosoft N25-9 (Stepan Co.)
was used as a nonionic surfactant (Table 2S). This surfactant was also used as an emulsifying agent in
microemulsions where d-limonene (MP Biomedicals) was the hydrocarbon solvent and 2-propanol
(Fisher Scientific) was the co-surfactant. The surfactant was added to brine to prepare solutions at high
(3.3 wt%) and low (0.3 wt%) concentrations. Both concentrations were selected above the critical micelle
concentration (CMC) of the surfactant to enhance its efficiency. Although not viable, the high
concentration was merely used to better understand differences in the fundamental mechanisms of
surfactant and ME floodings.
In order to visualize the structures of these MEs, a Tecnai TF20 S-Twin High Resolution Transmission
Electron Microscope (HRTEM) from FEI was used at 220 kV. MEs were carefully transferred on silicon
dioxide coated carbon TEM grids (SPI supplies) and dried overnight before the imaging process. ImageJ
software was used for image analysis.
2.3 Dynamic interfacial tension and contact angle
6
The interfacial tension between Tensleep oil and different brine solutions (0.1 M NaCl, 0.3 wt%
surfactant, and 0.8 wt% ME containing 0.3 wt% surfactant) were measured by the pendant drop and
rising/captive bubble tensiometer, which was described in a previous study.36 The needle diameter was
selected between 0.364 mm to 1.762 mm in order to approach a bond number close to unity. Images were
captured every 1 min and analyzed using Axisymmetric Drop Shape Analysis (ADSA). The dynamic IFT
of brine solutions with low additive concentration (0.3 wt% surfactant and 0.8 wt% ME containing 0.3 wt%
surfactant) was measured at least 3 times and the average values were reported in figures with error bars.
The IFT of brine solutions with high additive concentration (3.3 wt% surfactant and 8 wt% ME
containing 3.3 wt% surfactant) were also measured by a spinning drop tensiometer (SITE100, Kruss) due
to their low values. The rotation speed was set to 3260 rpm and the measurement interval was 30 s.
Measurements of the static contact angle of oil/water/rock systems were performed using the IFT/CA
apparatus.36 The rocks were cut into small substrates that were vacuumed at 10-7 psi for 12 hours then
immersed in oil. After aging in oil for 7 and 14 days at 60 C, they were gently placed in the IFT/CA cell.
Brine (with and without chemical additives) was then transferred to the cell until the substrates were fully
immersed. The oil inside the substrates formed several small oil droplets on the surface of the rock, as it
was released by spontaneous imbibition of brine. Images of these droplets were taken after 24 hours and
analyzed by ImageJ software to estimate the contact angles. Figure 1S illustrates an example of these
images with Edwards rock substrates after being aged in oil for 0, 7, and 14 days.
2.4 Droplet size distribution
In order to estimate the oil drop size distributions in brine solutions, oil and brine (50/50 volume ratio)
with different additive concentrations were mixed for 5 hours at a speed of 500 rpm. The rag layers (or
Winsor Type III microemulsions) formed between these phases were diluted 20 times in the same brine
solutions to enhance their transparency to light, then immediately transferred into cuvettes. A particle
sizer and zeta potential analyzer (Brookhaven Instruments Co.) was used to measure the droplet size of
emulsions via dynamic light scattering technique (DLS). Each measurement was conducted for 1 minute
and repeated at least 5 times to reduce experimental error. ZetaPlus Particle Sizing Software was used to
analyze the droplet size distribution of these emulsions.
2.5 Spontaneous imbibition
For the spontaneous imbibition tests, the cores were first vacuumed at 10-7 psi for 12 hours. Oil was then
injected into the vacuum cell to saturate the cores for 24 hours. The cores were aged in Tensleep oil at 60
C for 7 and 14 days. After the aging process, the cores were weighed, placed in Amott cells, and then
immersed in different brine solutions (0.1 M NaCl, 0.3 wt% surfactant, 3.3 wt% surfactant, 0.3 wt% ME,
and 3.3 wt% ME). The produced oil was recorded in time until recovery was complete.
3. RESULTS
3.1 Rock characterization
7
The mineral map and composition of the three core samples are provided in Figure 1 and Table 3S.They
show that Berea consists of 85% quartz and a small fraction of other minerals, such as kaolinite, K-
feldspar, illite, albite, etc. Edwards is homogeneous with respect to mineralogy and contains over 99% of
calcite. Tensleep consists of 84% quartz and small fractions of other minerals, which are similar to Berea.
However, in contrast to Berea, some of the pores in Tensleep are filled with gypsum/anhydrate and most
of the pores contain dolomite microcrystals. These microcrystals may have formed during the
precipitation of evaporite minerals within Sabkha sediments. Although the percentage of these
microcrystals is only about 3.7 %, the dolomite coatings can change the mineralogy of the pore surface
and increase its roughness, thereby affecting wettability and fluid flow in the porous rock.
Figure 1 also displays the micro-CT images of the three rocks under study. Berea is a fine-grained, well
sorted, and well-rounded sandstone. Its pore size distribution has two peaks, which represent their average
pore and throat sizes. The average pore and throat sizes of Berea are about 55 m and 3 m, respectively.
In contrast, the average pore size of Edwards is about 65 m, which is higher than Berea. The pore size
distribution of Edwards is much wider than Berea, which means that Edwards has a larger number of
small and large pores than Berea. Thus, the size distribution of Edwards makes it easier to trap oil inside
the pores compared to Berea. Tensleep has fine to medium, subrounded detrital quartz grains that are
embedded in a pervasive microcrystalline dolomite cement, seen here as white particles inside the pores.37
These cements enhance the roughness of grain surfaces, in agreement with the mineralogy map. Abundant
gypsum/anhydrite patches or stripes are also distributed in the rock with almost no porosity. This matrix
heterogeneity is a result of variation in lithologies, depositional structures, and diagenetic modifications.38
Tensleep has the widest pore size distribution, which indicates that there is a large amount of big and
small pores available in this rock. The average pore size of Tensleep is about 60 m.
3.2 Microemulsion phase behavior
The pseudo-ternary phase diagram of MEs was prepared by first mixing surfactant and alcohol at a fixed
weight ratio of 2.5/1 to form Smix. D-limonene was then added at different weight ratios (1/9, 2/8, 3/7, 4/6,
5/5, 6/4, 7/3, 8/2, 9/1) and the phase diagram was obtained by titrating these mixtures with brine. Figure
2a presents the phase diagram of this system where the solid arrows indicate the path of microemulsion
dilution in brine until the desired surfactant concentrations were reached). We found that mixtures of
surfactant, d-limonene, brine, and 2-propanol generated at a specific weight ratio of 2/1/1/0.8 provided
stable and transparent microemulsions over a period of 3 months at ambient conditions. When Smix to d-
limonene ratio was lower than 4/6 (area below dashed line), oil-in-water (o/w) emulsions formed and
extended over a wide area with increasing water content. The size of emulsions was large enough to
scatter light and as such they appeared as cloudy milky white colloidal solutions. Conversely, when Smix
to d-limonene ratio was higher than 4/6 (area above dashed line), transparent MEs were obtained. The
high concentration of Smix contributed to a lower IFT and higher emulsification ability, which could
promote the formation of MEs. More details of the preparation procedure are provided elsewhere.39
8
HRTEM micrographs of MEs with low surfactant concentration are presented in Figure 2b. The average
size of d-limonene droplets was about 90 nm. The dark layers around these droplets represent surfactant
molecules adsorbed at d-limonene/brine interfaces. At these interfaces, the hydrophobic tails of surfactant
remain in the d-limonene while the hydrophilic heads of the surfactant stretch toward the aqueous phase.
In addition to the adsorbed surfactant molecules on the d-limonene/water interfaces, there was a large
fraction of small surfactant micelles in the solution with an average diameter of 3.3 nm. This indicates
that free micelles and adsorbed surfactant molecules coexist in microemulsions.
3.3 Dynamic interfacial tension and drop size distribution
The effect of MEs on the dynamic interfacial tension between oil and brine was studied at ambient
conditions. Two ME concentrations were used: 8 wt% ME (containing 3.3 wt% of surfactant), and 0.8 wt%
ME (containing 0.3 wt% of surfactant). Surfactant solutions in brine (0.3 and 3.3 wt%) were also tested
for comparison. Without any additives, the IFT between oil and brine was about 11.8 mN/m after 600
mins. It sharply decreased to 0.8 mN/m and 0.3 mN/m upon introduction of small concentration of
surfactant and ME, respectively (Figure 3a, left). The 2-propanol molecules in ME could partition
between the brine and oil phases and behave in many respects as nonionic co-surfactants, further reducing
the IFT.40 At high concentration, IFT reached equilibrium within the first 100 minutes to stabilize at 0.39
and 0.26 mN/m with surfactant and ME, respectively (Figure 3a, right). Although the difference between
these low IFT values was rather small, it had a large impact on the size of oil droplets. Figure 3b presents
the size distribution of oil droplets measured by DLS at different additive concentrations in brine. The oil
droplets were taken from the diluted rag layer formed between oil and brine phases according to the
procedure described in Section 2.3. We found that the average drop sizes generated by low additive
concentration were similar and in the 100-200 m range. However, drops at high ME concentration (D =
9.2 m) were less polydisperse and more than one order of magnitude smaller than those formed by high
surfactant concentration (D = 156 m). While concentration did not affect drop sizes in surfactant
solutions since it was above their CMC, it had a significant impact in MEs. At high concentration, the
amount of d-limonene and 2-propanol in ME was large enough to significantly reduce the size of these
droplets. This in turn had significant implications for oil mobilization in porous media, as shown in the
next sections.
3.4 Wettability alteration
3.4.1 Effect of aging time
The wettability alteration of Berea sandstone, Edwards limestone, and Tensleep sandstone over different
aging times with oil was first examined through contact angle measurements with brine alone. The
substrates were immersed in oil inside a sealed container for 7 and 14 days at 60 C. During the aging
process, asphaltene molecules adsorbed on the rock surfaces, altering their wettability.4 Since adsorption
is a kinetic process, wettability alteration is a function of aging time. After 7 days of aging, the contact
angles became almost constant, indicating that a minimum of one week was needed to alter the wettability
9
of these substrates, as shown in the top curves of Figure 4. Aging the rocks for 14 days slightly increased
the static contact angles, especially for Edwards limestone. Both Tensleep and Edwards exhibited the
highest wettability alteration with an increase in CA from 78 to 150 and from 72 to 166, respectively.
This could be explained by the strong adsorption of oil polars onto calcite in Edwards and dolomite in
Tensleep.41 Similar trends were observed with Berea sandstone where wettability was altered from water-
wet to weakly oil-wet after 14 days of aging.
The effect of additives on the wettability alteration of Berea, Edwards, and Tensleep substrates was also
investigated at high concentration. Figure 4a shows that both surfactant and ME restored the wettability
of aged Berea substrates from weakly oil-wet to water-wet, with a contact angle of 66 and 49,
respectively. Surfactant molecules adsorbed with their hydrophobic tails on the mineral surface and
created a hydrophilic layer with their polar heads,3 enhancing the momentum transfer efficiency across
the oil/water interface and promoting the solubilization of the adsorbed oil layers.35,42 The high
performance of MEs was likely due to the ability of d-limonene to penetrate these layers and reduce the
overall percentage of polar components, leading to easier removal.34,35 The same trends were observed
with Edwards, with more than 50% decrease in contact angle in the presence of surfactant and ME
(Figure 4b). Nevertheless, these angles were slightly higher than those on Berea due to the strong
adsorption of polars on calcite. The contact angle on 14 day-aged Tensleep substrates decreased
considerably from 150 to 57 upon the addition of ME. Unlike Berea and Edwards, the surfactant ability
to alter the wettability of Tensleep rock was relatively limited. Indeed, the rock surface was still neutral-
wet after the addition of high surfactant concentration, with a contact angle decrease of only 40 (Figure
4c). Figure 5a reveals that the contact angle distribution of oil droplets on the substrates immersed in
surfactant solutions spans over a wide range of values (i.e., from 70 to 140). In other words, a small
fraction of the substrate surface became water-wet while a larger fraction remained neutral-wet or oil-wet.
By comparison, the contact angle distribution with ME was narrower and ranged from 50 to 80,
indicating that the wettability of a large fraction of the rock has been restored to its water-wet condition.
The main difference between Tensleep and the other rocks is the presence of dolomite microcrystals
throughout the pores (Figure 1). These rough microcrystalline cements may make it harder for surfactant
alone to solubilize oil from their surfaces.
3.4.2 Effect of additive concentration
To further understand the effect of ME on wettability alteration, the contact angles of oil droplets released
from 14 day-aged rocks during imbibition in brine, surfactant and ME with different surfactant
concentrations (i.e., 0.3 and 3.3 wt%) were measured, as shown in Figure 5b. Concentration had little
effect on the wettability of Berea, indeed both low and high surfactant and ME concentrations altered the
wettability of Berea from oil-wet to water-wet. The impact of concentration was more prominent on
Edwards rock, with a 40 decrease in CA at high surfactant and ME concentrations. This implies that
surfactant molecules interacted similarly with calcite whether they were alone or within microemulsions.
In contrast, increasing concentration in surfactant solutions did not affect the neutral wettability of
10
Tensleep, since the contact angles on dolomite cement could only be altered by microemulsions. By
increasing surfactant concentration in MEs, we are also enhancing the amount of d-limonene solvent and
therefore the ability of this phase to penetrate and swell the adsorbed oil layers, further altering rock
wettability from 96 to 57.
3.5 Spontaneous imbibition
3.5.1 At high additive concentration
Spontaneous imbibition tests with brine, microemulsions, and surfactant solutions with a surfactant
concentration of 3.3 wt% were first conducted on Berea, Edwards and Tensleep rocks at different aging
times to investigate the performance of MEs. This high concentration was chosen to significantly enhance
the differences between surfactant and ME. The volume of oil produced from the cores during
spontaneous imbibition was recorded for at least 30 days until no more oil was produced.
3.5.1.1 Berea sandstone
Figure 6 shows the effect of high surfactant and ME concentration on oil recovery from Berea at different
aging times prior to imbibition. Earlier production was recorded with surfactant compared to brine due to
wettability alteration towards more water-wet conditions, although the final amount produced was very
similar (i.e., about 55 %). This is because Berea is a well-sorted sandstone rock with relatively
homogeneous pore sizes (Figure 1a). Brine and surfactant solution could invade into most of the pores,
leaving some oil behind due to unfavorable threshold capillary pressures of altered surfaces and bypassing
of oil caused by pore-scale displacement mechanisms, which is more prone in water-wet surfaces. The
addition of ME to brine could significantly enhance oil removal from Berea and produce 86% of oil,
which was 35% more than surfactant and brine. This increase in the amount of oil recovery was
independent of aging time, suggesting that emulsification was the driving force and not wettability
alteration. The sharp decline of IFT from 11.8 to 0.26 mN/m promoted the formation of very small oil
droplets (Figure 3, right) that were easier to mobilize through the porous rock because their sizes were
smaller than the pores and throats of Berea.
Figure 2S in the Supporting Information provides thin sections of clean Berea as well as 14 day-aged
Berea after spontaneous imbibition with brine, surfactant, and ME. Wettability alteration can be seen as
thin oil layers adsorbed on grain surfaces that brine alone could not remove. Imbibition with surfactant
and ME restored wettability to a great extent by solubilizing most of the adsorbed oil, in agreement with
Figures 4a and 5b.
3.5.1.2 Edwards limestone
Figure 7 shows oil recovery from spontaneous imbibition tests conducted on aged and non-aged Edwards
cores using brine and high concentration of surfactant and ME. The strong wettability alteration of
carbonates from 72 to 166 upon contact with oil (Figure 4b) is an important parameter that controls
fluid displacement in this type of rocks. As the rock becomes strongly oil-wet after 14 days, brine alone is
11
not able to invade the pores resulting in a one order of magnitude drop in oil removal with aging time,
from 40% to 4%. Surfactant on the other hand could produce more oil than brine. In non-aged rocks for
example, the slight increase in recovery with surfactant compared to brine was mainly attributed to a
decrease in IFT. As a result, surfactant could invade some of the larger pores of Edwards that brine alone
could not. The oil removal with surfactant was almost independent of aging time, in agreement with
Figure 4b. MEs exhibited the same wettability behavior as surfactants, according to Figures 4b and 5b.
Therefore the constant 25-30% additional oil removal with MEs compared to surfactant in Figure 7 was
mainly due to the increased emulsification of oil droplets (Figure 3b, right), thereby enhancing their
mobilization in Edwards limestone, similar to Berea sandstone. Note that during the imbibition of non-
aged Edwards with high ME concentration, the volume of produced oil was 3.68 ml, which indicates that
3.68 ml of the brine phase (containing 0.07 ml of d-limonene) has invaded into the pores to displace this
oil. If limonene was produced with the oil, it will constitute only 2% of the oil recovery.
The thin section analysis of Edwards cores used in the spontaneous imbibition tests indicates that the
thickness of adsorbed oil layers on grain surfaces has been reduced by high concentration surfactant and
ME (Supporting Information, Figure 3S). Due to the strong interactions between oil polar molecules and
carbonate, surfactant and ME could solubilize some adsorbed layers but not to the same extent as Berea.
3.5.1.3 Tensleep sandstone
Figure 8 shows the effect of high concentration additives on oil recovery from aged and non-aged
Tensleep cores, and their comparison with brine. Aging this rock in oil delayed and curbed oil production
by spontaneous imbibition in brine due to wettability alteration from water-wet to oil-wet (Figure 4c). A
thin section of this core is shown in the Supporting Information (Figure 4S) where wettability alteration
was intermediate between those of Berea and Edwards, in agreement with Figure 4. This explained why
brine could displace less than 20% of oil from Tensleep. The behavior of high concentration surfactant
solutions was very peculiar in this rock. While initially larger than 20%, oil removal with surfactant
decreased with aging time until it became zero with 14 day-aged cores. This gradual decrease was
consistent with the systematic increase in contact angle with aging time, as shown in Figure 4c with
surfactant. Examination of the thin section in Figure 4Sc revealed that the surfactant formed large bi-
continuous phases at the surface of dolomite cements that could not detach within the time scale of the
experiment. At such a high concentration, surfactant micelles were able to penetrate the microporous
structure of dolomite cement, due to the low IFT, and trap oil in the form Winsor type I microemulsions.
The small oil droplets further clustered and percolated into larger bi-continuous phases that continuously
grew and diffused through the porous cement.43 These phases extended deeper into the oil adsorbed on
dolomite surface and did not detach from it because the surfactant could not alter its wettability. The
accumulation of these phases on dolomite eventually blocked the water channels within the first day,
significantly impeding oil recovery or cleanup. This behavior was not observed with microemulsions. In
fact, MEs enhanced oil removal from 51% to 61% with increasing aging time (Figure 8). The high
performance of MEs could be explained by two distinct mechanisms: (i) enhanced mobilization of very
small emulsified oil droplets due to the low IFT between oil and ME solution, similar to Berea and
12
Edwards, and (ii) solubilization of oil adsorbed on dolomite cement, in accord with CA data in Section
3.2.2. The second mechanism was confirmed by the thin section of Figure 4Sd where a large contrast
between surfactant and ME was observed. Indeed, MEs were very effective at cleaning up the surface of
sand grains including dolomite cements. Due to its lower viscosity, the d-limonene solvent present in
MEs could penetrate and swell the oil trapped in dolomite cements, leading to more effective
desorption.34,35,44 A schematic explaining this mechanism is illustrated in Figure 5S of the Supporting
Information.
3.5.2 At low additive concentration
Another set of spontaneous imbibition tests were undertaken to investigate the impact of more
economically viable additive concentration (i.e., 0.3 wt%) on oil recovery from 14 day-aged rocks. The
results displayed in Figure 9 indicate that the performance of surfactant and ME was very similar in
Berea and Edwards but differed in Tensleep. At low concentration, the oil/brine IFT dropped from 12
mN/m to 0.8 mN/m and 0.3 mN/m with surfactant and ME, respectively. However, the amount of
limonene and 2-propanol in ME was not large enough to reduce the size of oil droplets compared to the
surfactant alone (Figure 3b, left). As a result, the amount of oil mobilized by surfactant and ME was
comparable and equal to 62% in Berea and 50% in Edwards (Figure 9a,b). The variations seen in
Tensleep were mainly attributed to the unique ability of MEs to solubilize oil from carbonate cements, as
detailed in the previous section, although their cleaning power was reduced at low concentration (Figure
9c). Overall, ME could recover 28% of oil from Tensleep, which was 7 % higher than surfactant alone
and 16 % higher than brine. These recoveries were lower than those in Berea and Edwards due to the
presence of anhydrite beddings with almost no porosity in Tensleep. In all the rocks examined here, oil
recoveries at low additive concentration were lower than those at high concentration, except for surfactant
imbibition in Tensleep. In this case, low surfactant concentration produced 20% of oil, compared to none
at high concentration. This is because surfactant micelles could not penetrate dolomite cement and formed
large bi-continuous phases that restricted oil displacement in this rock.
3.5.3 Effect of rock type on oil removal
Berea is a well-sorted, relatively homogeneous rock that mainly consists of quartz and a small fraction of
other minerals. Because of its relatively uniform pore size distribution and water-wetness, brine,
surfactant and ME floodings exhibited the highest ultimate oil recovery with Berea among all three rocks.
Edwards is homogenous with respect to mineralogy (over 99% of calcite) but heterogeneous with respect
to pore size distribution. Therefore, Edwards tend to trap more oil inside the pores and can easily be
altered to oil-wet during the aging process. MEs and surfactant recovered less oil from non-aged Edwards
compared to Berea. After 14 days of aging in oil, the ultimate oil recovery with brine from Edwards was
significantly lower than that from 14 days aged Berea due to the oil-wet state of Edwards rock. Tensleep
consists of 84% of quartz and small fractions of other minerals, which are similar to Berea. However, in
contrast to Berea, some of the pores in Tensleep are filled with gypsum/anhydrate. This resulted in a poor
connectivity between the capillary elements of this rock and consequently led to more trapped oil.
13
Therefore, brine, surfactant and MEs exhibited the lowest ultimate oil recovery from Tensleep among all
three rocks. The existence of pervasive dolomite microcrystals inside Tensleep affected the performance
of surfactant significantly. In contrast to Edwards and Berea, high concentration of surfactant could not
recover any oil from this rock within the time scale of the experiment due to the existence of dolomite
cement while MEs exhibited exceptional performance in oil removal. The results of this study indicate
that ME flooding is an attractive remediation or EOR process in oil-wet heterogeneous formations
containing microcrystalline carbonate cements.
4. CONCLUSIONS
The impact of an environmentally friendly ME formulation on the cleanup or recovery of oil was
examined using spontaneous imbibition tests and petrographic analyses. Two different outcrops, Berea
sandstone and Edwards limestone, and a heterogeneous reservoir rock, Tensleep sandstone, were used to
demonstrate the dependence of ME performance on mineral texture and wettability state. The fluids
consisted of 1 M NaCl brine, two ME solutions in brine containing 0.3 wt% and 3.3 wt% of a linear
alcohol ethoxylate surfactant, and two surfactant solutions at the same concentrations. Rock substrates
and cores were aged with crude oil for 7 and 14 days to establish a wide range of wettability states. The
cores were then subjected to imbibition tests to investigate the effectiveness of surfactants and MEs in
enhancing oil recovery or removal. A systematic analysis of IFT values, drop size distributions, thin
sections, and CA data at different aging times was used to explain the fluid displacement mechanism
observed during the imbibition process. Below are the main conclusions summarized for each rock:
1- In Berea and Edwards, the ultimate recovery with high ME concentration constantly
outperformed that of surfactant and brine in all wettability states due to the emulsification of oil
into very small droplets. The amount of d-limonene solvent and 2-propanol co-surfactant in ME
was large enough to reduce the size of oil droplets by more than one order of magnitude, resulting
in 25-35% additional oil removal over surfactant solutions. On the other hand, the performance of
low surfactant and ME concentration was alike due to comparable effects on IFT, emulsification,
and wettability alteration. Therefore, the use of ME over surfactant alone was not justified at low
concentration.
2- Solubilization of adsorbed oil by surfactant and ME was comparable in Berea and Edwards but
very different in Tensleep due to the presence of dolomite cement. The solubilization of oil by
high surfactant concentration led to the formation of bi-continuous phases at cement surfaces that
blocked water channels, significantly impeding oil removal. Microemulsions outperformed
surfactants in Tensleep rock even at low concentrations due to the unique ability of d-limonene
solvent to penetrate microcrystalline dolomite cements of Tensleep and alter their wettability.
5. ACKNOWLEDGMENTS
14
The authors would like to thank Newfield Exploration and the National Science Foundation (Career
Award #1351296) for financial support. The authors are also grateful to Elizabeth Barsotti for
QEMSCAN analysis, Wendi Kuang for microCT imaging, and Dr. Joshua Stecher for HRTEM imaging.
6. REFERENCES
[1] Bennion, D. B.; Thomas, F. B.; Bietz, R. F.; Bennion, D. W. Water And Hydrocarbon Phase
Trapping In Porous Media-Diagnosis, Prevention And Treatment. Journal of Canadian Petroleum
Technology 1996, 35 (10).
[2] Sharma, M. K.; Shah, D. O.; Brigham, W. E. The Influence of Temperature on Surface and
Microscopic Properties of Surfactant Solutions in Relation to Fluid Displacement Efficiency in
Porous Media. AIChE J. 1985, 31 (2), 222–228.
[3] Standnes, D. C.; Austad, T. Wettability Alteration in Chalk: 2. Mechanism for Wettability
Alteration from Oil-Wet to Water-Wet Using Surfactants. Journal of Petroleum Science and
Engineering 2000, 28 (3), 123–143.
[4] Javanbakht, G.; Goual, L. Mobilization and Micellar Solubilization of NAPL Contaminants in
Aquifer Rocks. Journal of Contaminant Hydrology 2016, 185–186, 61–73.
[5] Winsor, P. A. Hydrotropy, Solubilisation and Related Emulsification Processes. Trans. Faraday
Soc. 1948, 44, 376.
[6] Pursley, J. T.; Penny, G.; Holcomb, D. Microemulsion Additives Enable Optimized Formation
Damage Repair and Prevention. Society of Petroleum Engineers, SPE 86556 presented at the
International Symposium and Exhibition on Formation Damage, Lafayette, Louisiana, 2004.
[7] Mahmoudkhani, A.; O’Neil, B.; Wylde, J. J.; Kakadjian, S.; Bauer, M. Microemulsions as
Flowback Aids for Enhanced Oil and Gas Recovery after Fracturing, Myth or Reality: A Turnkey
Study to Determine the Features and Benefits, Society of Petroleum Engineers, SPE 173729
presented at the Symposium on Oilfield Chemistry, The Woodlands, Texas, USA, 2015.
[8] Pietrangeli, G. A.; Quintero, L.; Jones, T. A.; Darugar, Q. Treatment of Water in Heavy Crude Oil
Emulsions with Innovative Microemulsion Fluids. Society of Petroleum Engineers, SPE 171140
presented at the Heavy and Extra Heavy Oil Confrrence, Medellin, Colombia, 2014.
[9] Rostami, A.; Nguyen, D. T.; Nasr-El-Din, H. A. Improving Relative Permeability to Gas in Tight
Sandstone Formations by Using Microemulsions. International Petroleum Technology Conference,
IPTC 17675 presented at the International Petroleum Technology Conference, Doha, Qatar, 2014.
[10] Penny G., A. Zelenev and N. Lett, Nano-fluid System Improves Postfrac Oil and Gas Recovery in
Hydrocarbon Rich Gas Reservoirs, SPE 154308 Presented at the 18th SPE Improved Oil Recovery
Symposium, Tulsa OK, 14-18 April 2012.
[11] Gogarty, W. B.; Tosch, W. C. Miscible-Type Waterflooding: Oil Recovery with Micellar Solutions.
Journal of Petroleum Technology 1968, 20 (12), 1407–1414.
[12] Walker, D.; Britton, C.; Kim, D. H.; Dufour, S.; Weerasooriya, U.; Pope, G. A. The Impact of
Microemulsion Viscosity on Oil Recovery. Society of Petroleum Engineers, SPE 154275 presented
at the Improved Oil Recovery Symposium, Tulsa, Oklahoma, USA, 2012.
15
[13] Ezrahi, S.; Aserin, A.; Garti, N. Aggregation Behavior in One-Phase (Winsor IV) Microemulsion
Systems. Handbook of Microemulsion Science and Technology. New York: Informa. Inc. 1999,
185–246.
[14] Salager, J.-L.; Antón, R. E.; Sabatini, D. A.; Harwell, J. H.; Acosta, E. J.; Tolosa, L. I. Enhancing
Solubilization in microemulsions—State of the Art and Current Trends. J. Surfact. Deterg. 2005, 8
(1), 3–21.
[15] Bohlen, D. S.; Vinson, P. K.; Davis, H. T.; Scriven, L. E.; Xie, M.; Zhu, X.; Miller, G. W. Generic
Patterns in the Microstructure of Midrange Microemulsions. Organized Solutions, Surfactant
Science Series 1992, 145–158.
[16] Fanun, M. Microemulsions: Properties and Applications; CRC Press, 2008.
[17] McClements, D. J. Nanoemulsions versus Microemulsions: Terminology, Differences, and
Similarities. Soft Matter 2012, 8 (6), 1719–1729.
[18] Bera, A.; Mandal, A. Microemulsions: A Novel Approach to Enhanced Oil Recovery: A Review. J.
Petrol Explor. Prod. Technol. 2015, 5 (3), 255–268.
[19] Karambeigi, M. S.; Nasiri, M.; Haghighi Asl, A.; Emadi, M. A. Enhanced Oil Recovery in High
Temperature Carbonates Using Microemulsions Formulated with a New Hydrophobic Component.
Journal of Industrial and Engineering Chemistry 2016, 39, 136–148.
[20] Poprawski, J.; Catté, M.; Marquez, L.; Marti, M.-J.; Salager, J.-L.; Aubry, J.-M. Application of
Hydrophilic–lipophilic Deviation Formulation Concept to Microemulsions Containing Pine Oil and
Nonionic Surfactant. Polym. Int. 2003, 52 (4), 629–632.
[21] Jeirani, Z.; Jan, B. M.; Ali, B. S.; See, C. H.; Saphanuchart, W. Pre-Prepared Microemulsion
Flooding in Enhanced Oil Recovery: A Review. Petroleum Science and Technology 2014, 32 (2),
180–193.
[22] Lu, W.-C.; Chiang, B.-H.; Huang, D.-W.; Li, P.-H. Skin Permeation of D-Limonene-Based
Nanoemulsions as a Transdermal Carrier Prepared by Ultrasonic Emulsification. Ultrasonics
Sonochemistry 2014, 21 (2), 826–832.
[23] Pursley, J. T.; Holcomb, D. L.; Penny, G. S. Composition and Process for Well Cleaning. Patent
US7380606 B2, June 3, 2008.
[24] Pérez-Mosqueda, L. M.; Trujillo-Cayado, L. A.; Carrillo, F.; Ramírez, P.; Muñoz, J. Formulation
and Optimization by Experimental Design of Eco-Friendly Emulsions Based on D-Limonene.
Colloids and Surfaces B: Biointerfaces 2015, 128, 127–131.
[25] Castro Dantas, T. N.; Soares A, P. J.; Wanderley Neto, A. O.; Dantas Neto, A. A.; Barros Neto, E.
L. Implementing New Microemulsion Systems in Wettability Inversion and Oil Recovery from
Carbonate Reservoirs. Energy & Fuels 2014, 28 (11), 6749–6759.
[26] Santanna, V. C.; Curbelo, F. D. S.; Castro Dantas, T. N.; Dantas Neto, A. A.; Albuquerque, H. S.;
Garnica, A. I. C. Microemulsion Flooding for Enhanced Oil Recovery. Journal of Petroleum
Science and Engineering 2009, 66 (3–4), 117–120.
[27] Jeirani, Z.; Mohamed Jan, B.; Si Ali, B.; Noor, I. M.; See, C. H.; Saphanuchart, W. Formulation,
Optimization and Application of Triglyceride Microemulsion in Enhanced Oil Recovery. Industrial
Crops and Products 2013, 43, 6–14.
16
[28] Bellocq, A. M. Effects of Alcohol Chain Length and Salt on Phase Behavior and Critical
Phenomena in SDS Microemulsions. Handbook of Microemulsion Science and Technology, Marcel
Dekker, New York 1999, 139–185.
[29] Fiorenza, S.; Miller, C. A.; Oubre, C. L.; Ward, C. H. NAPL Removal: Surfactants, Foams, and
Microemulsions, Lewis Publishers, Boca Raton, Florida, 2000.
[30] Jawitz, J. W.; Annable, M. D.; Rao, P. S. C.; Rhue, R. D. Field Implementation of a Winsor Type I
Surfactant/Alcohol Mixture for in Situ Solubilization of a Complex LNAPL as a Single-Phase
Microemulsion. Environ. Sci. Technol. 1998, 32 (4), 523–530.
[31] Yang, J.; Jiang, W.; Guan, B.; Lu, Y.; Qiu, X. Optimizing Nanoemulsions as Fluid Flowback
Additives in Enhancing Tight Gas Production. Journal of Petroleum Science and Engineering 2014,
121, 122–125.
[32] Healy, R. N.; Reed, R. L.; Carpenter, C. W. A Laboratory Study of Microemulsion Flooding
(Includes Associated Papers 6395 and 6396). Society of Petroleum Engineers Journal 1975, 15 (1),
87–103.
[33] Towler, B. F.; Lehr, H. L.; Austin, S. W.; Bowthorpe, B.; Feldman, J. H.; Forbis, S. K.; Germack,
D. Spontaneous Imbibition Experiments of Enhanced Oil Recovery with Surfactants and Complex
Nano-Fluids, Journal of Surfactants and Detergents 2017, 20 (2), 367–377.
[34] Lowry, E.; Sedghi, M.; Goual, L. Molecular Simulations of NAPL Removal from Mineral Surfaces
Using Microemulsions and Surfactants. Colloids and Surfaces A: Physicochemical and Engineering
Aspects 2016, 506, 485–494.
[35] Bui, K.; Akkutlu, I. Y.; Zelenev, A.; Saboowala, H.; Gillis, J. R.; Silas, J. A. Insights Into
Mobilization of Shale Oil by Use of Microemulsion. SPE Journal 2016, 21 (2), 613–620.
[36] 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), 6856–6866.
[37] Sheldon, R. P. Physical Stratigraphy of the Phosphoria Formation in Northwestern Wyoming;
School of Mineral Sciences, 1957.
[38] James, W. C. Sandstone Diagenesis in Mixed Siliciclastic-Carbonate Sequences; Quadrant and
Tensleep Formations (Pennsylvanian), Northern Rocky Mountains. Journal of Sedimentary
Research 1992, 62 (5), 810–824.
[39] Goual, L. Microemulsions and Uses Thereof to Displace Oil in Heterogeneous Porous Media,
Provisional Patent 15/367,478, University of Wyoming (2016).
[40] Zana, R. Aqueous Surfactant-Alcohol Systems: A Review. Advances in Colloid and Interface
Science 1995, 57, 1–64.
[41] Thomas, M. M.; Clouse, J. A.; Longo, J. M. Adsorption of Organic Compounds on Carbonate
Minerals. Chemical Geology 1993, 109 (1), 201–213.
[42] Kumar, K.; Dao, E. K.; Mohanty, K. K. Atomic Force Microscopy Study of Wettability Alteration
by Surfactants. SPE Journal 2008, 13 (2), 137–145.
[43] Vollmer, D.; Vollmer, J.; Eicke, H.-F. Droplet Cluster vs. Bicontinuous Structures in Percolating
Water-in-Oil Microemulsions. Europhysics Letters 1994, 26 (5), 389–394.
17
[44] Javanbakht, G.; Arshadi, M.; Qin, T.; Goual, L. Micro-Scale Displacement of NAPL by Surfactant
and Microemulsion in Heterogeneous Porous Media. Advances in Water Resources 2017, 105,
173–187.
a) Berea
b) Edwards
c) Tensleep
0
0.05
0.1
0.15
0.2
1 10 100
f (r
)
r (µm)
Berea
0
0.05
0.1
0.15
0.2
1 10 100
f (r
)
r (µm)
Edwards
0
0.05
0.1
0.15
0.2
1 10 100
f (r
)
r (µm)
Tensleep
18
Figure 1. Size distribution (left), micro-CT image of core cross-section (center), and mineralogy map
(right) of a) Berea, b) Edwards, and c) Tensleep. The two peaks in pore size distributions represent
average pore and throat sizes. A large fraction of Tensleep pores contain dolomite microcrystals (or
cement) that are not found in Berea or Edwards.
a)
b)
Figure 2. a) Phase diagram of surfactant/limonene/water/2-propanol system. The surfactant/2-propanol
ratio is fixed at 2.5/1. Smix is the fraction of (surfactant + alcohol) in the (surfactant + alcohol + d-
limonene) mixture. When Smix/d-limonene ratio < 6/4 (see area below dashed line where Smix < 0.6), milky
white oil-in-water (o/w) emulsions form with increasing water content and extend over the large grey area
in the phase diagram. On the other hand, when Smix/d-limonene ratio > 6/4, colorless and stable
microemulsions form. b) HRTEM micrographs of low ME concentration showing small 3 nm large
ME (3.3 wt%)
Emulsion
ME (0.3 wt%)
19
surfactant micelles in the bulk phase and microemulsions with an average diameter of 90 nm. Surfactants
form a thin layer at the limonene/brine interface to stabilize MEs.
a) Interfacial tension
b) Size distribution
20
Figure 3. Effect of additive concentration in brine on the a) oil/brine dynamic interfacial tension and b)
size distribution of oil droplets in diluted rag layer between oil and brine. Low and high additive
concentrations are shown on the left and right sides, respectively.
Figure 4. Average static contact angles of oil droplets released from aged a) Berea, b) Edwards, and c)
0
20
40
60
80
100
120
140
160
180
0 2 4 6 8 10 12 14 16
Co
nta
ct a
ngl
e (
de
gre
e)
ME 3.3wt% Surfactant 3.3 wt% Brine
a) Berea
0
20
40
60
80
100
120
140
160
180
0 2 4 6 8 10 12 14 16
Co
nta
ct a
ngl
e (
de
gre
e)
b) Edwards
0
20
40
60
80
100
120
140
160
180
0 2 4 6 8 10 12 14 16
Co
nta
ct a
ngl
e (
de
gre
e)
Aging time prior to imbibition (days)
c) Tensleep
21
Tensleep rocks during imbibition in brine, surfactant and ME. Both surfactant and ME alter the
wettability of Berea and Edwards from oil-wet to water-wet, regardless of aging time. With Tensleep
however, the surfactant is not as effective as ME in altering wettability, especially at longer aging times.
Figure 5. Static contact angle of oil droplets released from 14 day-aged rocks during imbibition in brine,
surfactant and ME. a) Angle distribution on Tensleep at high surfactant and ME concentration. Relatively
narrow distribution is achieved at low CA values with ME. b) Average angle on Berea, Tensleep and
Edwards at different surfactant and ME concentrations. The difference between CA at low and high ME
concentration is more prominent on Tensleep and Edwards.
0
0.1
0.2
0.3
0.4
0.5
0.6
30 40 50 60 70 80 100 110 120 130 140 150 160 170
Fre
qu
en
cy
Contact angle (degree)
Brine Surfactant 3.3 wt% ME 3.3 wt%a)
0
20
40
60
80
100
120
140
160
180
Tensleep Berea Edwards
Co
nta
ct a
ngl
e (
de
gre
e)
0.1 M NaCl Surfactant 0.3 wt% Surfactant 3.3 wt%
ME 0.3 wt% ME 3.3 wt%
b)
22
Figure 6. Effect of high concentration surfactant and ME on oil recovery by spontaneous imbibition in
Berea at different aging times: a) non-aged, b) 7 day-aged, and c) 14 day-aged.
0
20
40
60
80
100
0.1 1 10 100 1000
Oil
reco
very
(%
)Brine Surfactant 3.3 wt% ME 3.3 wt%
a) 0 days
0
20
40
60
80
100
0.1 1 10 100 1000
Oil
reco
very
(%
)
b) 7 days
0
20
40
60
80
100
0.1 1 10 100 1000
Oil
reco
very
(%
)
Time (hours)
c) 14 days
23
Figure 7. Effect of high concentration surfactant and ME on oil recovery by spontaneous imbibition in
Edwards at different aging times: a) non-aged, b) 7 day-aged, and c) 14 day-aged.
0
20
40
60
80
100
0.1 1 10 100 1000
Oil
reco
very
(%
)Brine Surfactant 3.3 wt% ME 3.3 wt%
a) 0 days
0
20
40
60
80
100
0.1 1 10 100 1000
Oil
reco
very
(%
)
b) 7 days
0
20
40
60
80
100
0.1 1 10 100 1000
Oil
reco
very
(%
)
Time (hours)
c) 14 days
24
Figure 8. Effect of high concentration surfactant and ME on oil recovery by spontaneous imbibition in
Tensleep at different aging times: a) non-aged, b) 7 day-aged, and c) 14 day-aged.
0
20
40
60
80
100
0.1 1 10 100 1000
Oil
reco
very
(%
)Brine Surfactant 3.3 wt% ME 3.3 wt%
a) 0 days
0
20
40
60
80
100
0.1 1 10 100 1000
Oil
reco
very
(%
)
b) 7 days
0
20
40
60
80
100
0.1 1 10 100 1000
Oil
reco
very
(%
)
Time (hours)
c) 14 days
25
Figure 9. Spontaneous imbibition of low surfactant and ME concentration with 14 day-aged rocks: a)
Berea, b) Edwards, and c) Tensleep. The performance of surfactant and ME is very similar in Berea and
Edwards but differs in Tensleep.
0
20
40
60
80
100
0.1 1 10 100 1000
Oil
reco
very
(%
)Brine Surfactant 0.3 wt% ME 0.3 wt%
a) Berea
0
20
40
60
80
100
0.1 1 10 100 1000
Oil
reco
very
(%
)
b) Edwards
0
20
40
60
80
100
0.1 1 10 100 1000
Oil
reco
very
(%
)
Time (hours)
c) Tensleep