energies

Article
An Experimental Investigation of Flow Regimes in
Imbibition and Drainage Using a Microfluidic Platform
Feng Guo and Saman A. Aryana *
Department of Chemical Engineering, University of Wyoming, Laramie, WY 82071, USA; fguo@uwyo.edu
* Correspondence: saryana@uwyo.edu; Tel.: +1-(307)-766-1992




Received: 13 March 2019; Accepted: 6 April 2019; Published: 11 April 2019


Abstract: Instabilities in immiscible displacement along fluid−fluid displacement fronts in porous

media are undesirable in many natural and engineered displacement processes such as geological
carbon sequestration and enhanced oil recovery. In this study, a series of immiscible displacement
experiments are conducted using a microfluidic platform across a wide range of capillary numbers
and viscosity ratios. The microfluidic device features a water-wet porous medium, which is a
two-dimensional representation of a Berea sandstone. Data is captured using a high-resolution
camera, enabling visualization of the entire domain, while being able to resolve features as small
as 10 µm. The study reports a correlation between fractal dimensions of displacement fronts and
displacement front patterns in the medium. Results are mapped on a two-dimensional parameter
space of log M and log Ca, and stability diagrams proposed in literature for drainage processes are
superimposed for comparison. Compared to recent reports in the literature, the results in this work
suggest that transition regimes may constitute a slightly larger portion of the overall flow regime
diagram. This two-phase immiscible displacement study helps elucidate macroscopic processes at
the continuum scale and provides insights relevant to enhanced oil recovery processes and the design
of engineered porous media such as exchange columns and membranes.

Keywords: drainage; imbibition; fractal dimension; phase diagram

1. Introduction
Study of multiphase-fluid distribution and flow in porous media is essential to several fields,
including petroleum engineering, hydrogeology, bioengineering and geoengineering [1]. Specific
examples include soil remediation, enhanced oil recovery (EOR), hydrogen fuel cells and liquid
transport through cellulose [2]. Thus, distribution and flow behavior of immiscible fluids in porous
media are active areas of both experimental and theoretical research [3]. Physical mechanisms
at pore scale for immiscible fluids appear to be well understood and are easier to model than
miscible fluids; however, predictions of transport and accurate descriptions of displacement processes
remain elusive due to a large number of factors impacting flow in porous media and lack of
comprehensive experimental data [4,5]. Using quasi-two-dimensional transparent porous media,
i.e., microfluidic devices, fluid flow can be directly controlled and monitored. Such an approach
allows virtual observation of flow distribution, displacement, interfaces between immiscible fluids
and quantitative evaluation of evolution of saturation, thereby providing significant insight into the
physical mechanisms and flow patterns at pore scale [6,7].
Two-phase displacement behavior in porous media, in the absence of gravitational forces, can be
characterized by two dimensionless numbers: capillary number, Ca, which is the ratio of viscous forces
to capillary forces; and viscosity ratio, M, which is the ratio of the two viscosities [8]. The capillary
number is defined as Ca = vµ1 /γ, where v is Darcy velocity of the invading fluid, µ is its viscosity,
and γ is the interfacial tension. M is defined as the ratio of the viscosity of the invading fluid, µ1 ,

Energies 2019, 12, 1390; doi:10.3390/en12071390 www.mdpi.com/journal/energies

Energies 2019, 12, 1390 2 of 13

to that of the displaced fluid, µ2 . Depending on these two parameters, either capillary or viscous
forces dominate, which leads to three basic displacement regimes: (1) viscous fingering, (2) capillary
fingering, or (3) stable displacement [9]. These flow regimes have been studied and delineated using
displacement experiments and pore-scale numerical simulations [10].
Lenormand et al. [11] proposed a phase diagram based on log Ca and log M to describe two-phase
immiscible displacement behavior in porous media. Schiwille et al. [12] conducted visualized
dispersion and entrapment in immiscible displacement experiments in two-dimensional packed
glass beads. Zhang et al. [13] demonstrated unstable liquid CO2 displacement of water in a porous
medium in capillary-dominated and viscous fingering regimes in microfluidic-based experiments.
Chomsurin et al. [14] and Werth et al. [15] demonstrated fluid saturation and analyzed interfacial areas
using quantitative image analysis techniques in microfluidic displacement experiments.
Due to the large ranges of imposed Ca and M values, a limited number of comprehensive
immiscible displacement experiments are conducted to observe flow behavior and fundamental
interfacial processes at the pore scale [13,16,17]. Immiscible displacement experiments conducted by
Lenormand et al. [4,9] are some of the few reported microfluidic studies so far where a large number of
fluids with a considerably high range of viscosity ratios are investigated for a variety of flow conditions.
However, issues related to microfluidic device fabrication and image analysis techniques, such as
inconsistency in pore networks across experiments and lack of quantitative phase distribution analysis,
may constrain the results. Liu et al. [18] simulated a gas-displacing-liquid drainage displacement
process using the Lattice Boltzmann method. A homogenous two-dimensional pore network consisting
of uniformly spaced square obstructions were used for the model. Based on their simulation data, three
flow regimes were identified and a different boundary line was proposed for the viscous fingering
regime compared to work by Lenormand et al. [4] and Zhang et al. [5].
Ferer et al. [19] used air to displace water in a glass microfluidic device to experiments across a
wide range of viscosities and demonstrated a crossover from capillary fingering to viscous fingering.
Furthermore, Wang et al. [10] conducted a series of experiments using supercritical CO2 to displace
water to study the crossover zone. However, the viscosity ratio between invading and residence fluids
were limited in those studies due to the choice of nonwetting and wetting fluids and their study only
focused on either drainage or imbibition [5,20].
In this effort, a series of drainage and imbibition displacement experiments covering a relatively
large range of capillary numbers and viscosity ratios are conducted. Fluid displacement and
migration behavior in a porous medium are investigated using a transparent glass microfluidic
device. Unlike most of previous studies that use porous media comprising of rectangular pore bodies
and throats [4,19,21], the pore network used in this work is a representation of a Berea sandstone,
which results in a porous medium with a complex network of channels with rigid walls in multi-fluid
displacement experiments [22]. A high-resolution camera is used to capture images of the entire
porous medium at prescribed time intervals during displacement experiments [23]. The direct
use of the camera without optical zoom provides for imaging of the medium in its entirety while
providing the resolution needed to discern features as small as 10 µm. The images enable quantitative
analysis of fractal dimensions of displacement fronts in displacement experiments. Observed flow
behaviors and fractal dimensions in displacement experiments are used to map three flow regimes
with boundaries using their respective values of log Ca and log M. This research not only extends
works of previous studies with a larger range of Ca and M using a glass microfluidic device, but also
presents improvements in visualization and analysis of phase distribution information.

2. Materials and Methods

2.1. Fluids Preparation

Deionized (DI) water (Nanopure II, Barnstead, Dubuque) and glycerol (ACS-reagent-grade,
Sigma Aldrich, Saint Louis, MO, USA) mixtures are used as the wetting phase, with viscosities
Energies 2019, 12, 1390 3 of 13

ranging from 1 cp to 250 cp. The nonwetting fluid, consisting of mineral oil mixtures including
Pentane (0.24 cp, HPLC grade, Fisher Chemical, Colorado Springs, CO, USA), Hydrobrite 550PO
(270 cp, Sonneborn, Petrolia, PA, USA) and Carnation (120 cp, Sonneborn, PA, USA), is used in the
displacement experiments. A food dye (black, Amazon) is added to the wetting fluid to increase its
contrast compared to the nonwetting phase. Methanol (HPLC grade, Sigma Aldrich, Saint Louis,
MO, USA), acetone (ACS-reagent-grade, Fisher Chemical, Colorado Springs, CO, USA) and toluene
(HPLC grade, Fisher Chemical, Colorado Springs, CO, USA) are used to clean the organic residue in
the porous medium and syringe.
In the drainage experiments, seven wetting–nonwetting fluid pairs are prepared with viscosity
ratios ranging from 0.1 to 300 (Table 1). In imbibition experiments, six wetting–nonwetting fluid
pairs are prepared with viscosity ratios ranging from 0.1 to 200 (Table 2). Viscosity is measured
using a spindle viscometer (DV2T, Brookfield, Middleboro, MA, USA). Interfacial tension values
are measured using the pendant drop method and a drop shape analysis system (Kruss DSA100).
The collected data includes three measurements for each data point; average values are reported in
Tables 1 and 2. All measurements are carried out at atmospheric pressure and room temperature
(approximately 20 ◦ C). Each fluid is degassed (Ultrasonic bath, Fisher Scientific, Colorado Springs, CO,
USA) for 10 min and filtered by a syringe filter (0.2 µm, Sterlitech, Kent, WA, USA) prior to injection in
microfluidic device.

Table 1. Drainage experiments fluids.

Interfacial
µ1 , cp Nonwetting Fluids µ2 , cp Wetting Fluids M
Tension γ, mN/m
10 Pentane + Hydrobrite 550PO 1 Water 0.1 49.26
2 Pentane + Carnation 1 Water 0.5 43.79
2 Pentane + Carnation 10 Water + Glycerol 5 30.10
0.24 Pentane 12 Water + Glycerol 50 22.99
0.24 Pentane 24 Water + Glycerol 100 21.56
0.24 Pentane 48 Water + Glycerol 200 20.55
0.24 Pentane 72 Water + Glycerol 300 19.22
0.015 CO2 10 Water + Glycerol 680.3 70.60
0.015 CO2 200 Water + Glycerol 13,605.4 42.90

Table 2. Imbibition experiments fluids.

Nonwetting Interfacial
µ1 , cp Wetting Fluids µ2 , cp M
Fluids Tension γ, mN/m
2.4 Pentane + Carnation 24 Water + Glycerol 0.1 22.48
12 Pentane + Carnation 24 Water + Glycerol 0.5 29.20
12 Pentane + Carnation 12 Water + Glycerol 1 28.90
5 Pentane + Carnation 1 Water 5 31.63
10 Pentane + Carnation 1 Water 10 32.84
50 Pentane + Hydrobrite 550PO 1 Water 50 33.27
100 Pentane + Hydrobrite 550PO 1 Water 100 35.25
200 Pentane + Hydrobrite 550PO 1 Water 200 37.01

2.2. Micromodels and Experimental Setup

The total chip dimension is 100 mm × 40 mm with a pore network region of 70 mm × 20 mm
and a porosity of approximately 40%. The average depth of the etched pore network is 10 µm.
The permeability of the microfluidic device is approximately 74 mD. The map of the complex channel
network is based on work reported by Alaskar [24]. The network represents the distribution of pores
and throats observed in a thin section of a Berea sandstone [24,25].
All displacement experiments are performed in the same microfluidic device. In drainage
experiments, the microfluidic device is flushed using methanol and DI water, in that sequence,
Energies 2019, 12, 1390 4 of 13
Energies 2019, 12, x FOR PEER REVIEW 4 of 13

to facilitate
facilitate subsequent
subsequent saturation
saturation by thebywetting
the wetting
fluids,fluids,
followed followed
by the by the injection
injection of a of
of a of dyed dyed
wetting
fluids. Nonwetting fluids are injected through one side of microfluidic device using a syringe pumpa
wetting fluids. Nonwetting fluids are injected through one side of microfluidic device using
syringe
(PHD pump
Ultra, (PHD Apparatus)
Harvard Ultra, Harvard Apparatus)
at various at various
volumetric volumetric
flowrates (q); the flowrates
flowrates (q);
rangedthe from
flowrates
0.03
ranged from 0.03 to 6 µL/min. The other side of the microfluidic
to 6 µL/min. The other side of the microfluidic device is open to the atmosphere. At each flowrate, device is open to the
atmosphere.
the nonwetting Atfluid
eachisflowrate,
injected tothe nonwetting
displace wetting fluid
fluidis until
injected to displace fluid
the nonwetting wetting fluidthe
crossed until the
entire
nonwetting
porous medium fluidandcrossed the entire
saturation reachedporous medium and
a quasi-steady state.saturation reacheddevice
The microfluidic a quasi-steady
is cleaned state.
after
The
each microfluidic
displacement device is cleaned
experiment after each displacement
according to experiment
the following according
cleaningthe following
to protocol:
cleaning protocol: isopropanol/ethanol/water
isopropanol/ethanol/water (1:1:1), 2 M HCl (1:1:1), 2 M HCl
solution, DIsolution,
water, DI a water,
basic a basic
solutionsolution
(DI
(DI water/NH 4 OH/H 2 O 2 at 5:1:1) and DI water. Likewise, in
water/NH4OH/H2O2 at 5:1:1) and DI water. Likewise, in imbibition experiments, the same imbibition experiments, the same
experimental setup
experimental setup isis used.
used. The The microfluidic
microfluidic device
device is is flushed
flushed by by methanol
methanol and and toluene
toluene followed
followed by by
injection of
injection of nonwetting
nonwetting fluids.fluids. Dyed
Dyed wetting
wetting fluids
fluids are
are injected
injected through
through thethe microfluidic
microfluidic device
device using
using
the syringe pump at various volumetric flowrates ranging
the syringe pump at various volumetric flowrates ranging from 0.06 to 6 µL/min. from 0.06 to 6 µL/min.
The microfluidic
The microfluidic device
device is is placed
placed horizontally
horizontally on on aa flat
flat stage
stage above
above aa homogeneous
homogeneous light light source
source
which allows
which allows the the light
light to
to penetrate
penetrate the themicrofluidic
microfluidicdevice.
device.Displacement
Displacementimages imagesare
areacquired
acquiredwith witha
Phase One IQ3 digital camera with a monochromatic 60 MP sensor. The
a Phase One IQ3 digital camera with a monochromatic 60 MP sensor. The entire experimental setup entire experimental setup is
shown
is shown ininFigure
Figure 1. 1.

Figure
Figure 1.
1. Microfluidic
Microfluidic displacement
displacement experimental
experimental setup.
setup.

3. Results
3. Results and
and Discussion
Discussion
High-resolution images
High-resolution images of of the
the medium
medium underunder continuous
continuous flowflow conditions
conditions areare captured
captured during
during
dynamic displacement experiments at prescribed time intervals until the saturation
dynamic displacement experiments at prescribed time intervals until the saturation of the invading of the invading
fluid reaches
fluid reaches aa quasi-steady
quasi-steady state
state for
for each
each ofof the
the flow
flow rates.
rates. Fractal
Fractal dimensions,
dimensions, D Df,, of
of displacement
displacement
fronts are
fronts are computed
computedusing usingthetheboxboxcounting
counting method
method following
following Aryana
Aryana et [26],
et al. al. [26],
which which
serves serves
as a
as a measure of the interface roughness for drainage and imbibition experiments.
measure of the interface roughness for drainage and imbibition experiments. In the case of drainage In the case of
drainage experiments,
experiments, the relation thebetween
relation the
between
inversetheofinverse
flow rateof flow ratewidth
and the and theof width of stabilized
stabilized zone,
zone, a quasi-
a quasi-steady region around the water-oil displacement front exhibiting
steady region around the water-oil displacement front exhibiting sharp variations of the displacing sharp variations of the
displacing phase saturation, is investigated and the results are compared
phase saturation, is investigated and the results are compared with the findings by Leas and with the findings by Leas
and Rapoport
Rapoport [27] [27]
and and Barenblatt
Barenblatt et [28].
et al. al. [28].
FlowFlow regimes
regimes areare importantininthe
important thedesign
designand andoperation
operation of of
improved oil recovery and EOR processes; flow regimes affect displacement
improved oil recovery and EOR processes; flow regimes affect displacement efficiency, overall efficiency, overall recovery
and surface
recovery andinjection recyclingrecycling
surface injection capacity capacity
needs. Aneeds.stableAviscous displacement
stable viscous often results
displacement in less
often results
bypassed
in resident
less bypassed fluid compared
resident to unstable
fluid compared regimes.
to unstable If a displacement
regimes. falls within
If a displacement falls the stable
within theregime,
stable
one can expect a reasonably high recovery and displacement efficiency. On
regime, one can expect a reasonably high recovery and displacement efficiency. On the other hand, the other hand, unstable
flow regimes
unstable flow may lead may
regimes to significant losses of economically
lead to significant recoverablerecoverable
losses of economically resources from petroleum
resources from
assets [29],assets
petroleum or losses
[29],inorstorage
losses incapacity
storageincapacity
geologic instorage
geologicofstorage
CO2 , due
of COto 2bypassing of resident
, due to bypassing of
resident fluids [30]. Flow instabilities are caused by heterogeneities in the media and nonlinearities
in the flow equations [26,31–34]. Bypassing of resident fluids promoted in unstable flow regimes
Energies 2019, 12, 1390 5 of 13

Energies 2019, 12, x FOR PEER REVIEW

fluids [30]. Flow instabilities are caused by heterogeneities in the media and nonlinearities 5inofthe
13

flow equations [26,31–34]. Bypassing of resident fluids promoted in unstable flow regimes often leads
often leads to early breakthrough of the injectant in producers, which may put unanticipated stress
to early breakthrough of the injectant in producers, which may put unanticipated stress on surface
on surface facilities needed to handle, recycle and reinject the fluids [35–37].
facilities needed to handle, recycle and reinject the fluids [35–37].
3.1.
3.1. Drainage
Drainage
Based
Based on on drainage
drainage displacement
displacementimages, images,shown shownin inFigure
Figure2, 2, itit is
is observed
observed thatthat at
at small
small values
values
of Ca (log Ca < −6.3) and small values of M (log M < 2), patterns
of Ca (log Ca < −6.3) and small values of M (log M < 2), patterns in displacement interfaces in displacement interfaces form an
form
interconnected flow path with randomly distributed forward
an interconnected flow path with randomly distributed forward and lateral fingers. This trendand lateral fingers. This trend in
displacement
in displacement interface
interfacepatterns
patterns indicates
indicates a capillary
a capillaryfingering
fingeringpatternpatternininwhich which capillary
capillary forces
forces
dominate
dominate the displacement [21]. In capillary fingering, the nonwetting fluid in a pore progresses
the displacement [21]. In capillary fingering, the nonwetting fluid in a pore progresses
preferentially
preferentially through
throughthe thelargest
largestpore porethroat
throat duedueto toa relatively
a relatively lower entry
lower pressure;
entry hence,
pressure; the
hence,
nonwetting
the nonwetting fluidfluid
may progress
may progressin anyindirection, even perpendicular
any direction, even perpendicular to the mainto thedirection of flow or
main direction of
in the backward direction [38]. At large values of M (log M > 2.5), flow
flow or in the backward direction [38]. At large values of M (log M > 2.5), flow paths with multiple paths with multiple loosely
connected or narrow-disconnected
loosely connected or narrow-disconnected fingeringfingering
are also areobserved. In such cases,
also observed. In such viscous forces appear
cases, viscous forces
to dominate the flow, and viscous fingers are important features of
appear to dominate the flow, and viscous fingers are important features of the interface affectingthe interface affecting the overallthe
flow behavior. In cases where log Ca > −4.9, log M < −0.3, the flow
overall flow behavior. In cases where log Ca > −4.9, log M < −0.3, the flow path shows a uniform path shows a uniform interface
front, distributed
interface over the over
front, distributed entire thewidth
entireofwidth
the microfluidic device.device.
of the microfluidic This displacing front front
This displacing results in a
results
high final saturation and would be considered a near-stable displacement,
in a high final saturation and would be considered a near-stable displacement, where low viscosity where low viscosity
wetting
wetting fluids
fluids are
are displaced
displaced by by aa relatively
relatively viscous
viscous nonwetting
nonwetting fluid fluid at at aa high
high injection
injection rate
rate [4].
[4]. For
For
capillary
capillarynumbers
numbersranging rangingbetween
between loglogCaCa= −5.0
= −and log Ca
5.0 and log= Ca
−6.3,= both
−6.3,capillary and viscous
both capillary fingers
and viscous
are observed.
fingers Nonwetting
are observed. fluid distribution
Nonwetting is related
fluid distribution to physical
is related to physicalmechanisms
mechanisms responsible for the
responsible for
displacement and the stability of the interface [39]. Based on work by
the displacement and the stability of the interface [39]. Based on work by Lenormand et al. [4,9] and Lenormand et al. [4,9] and our
observations
our observations (Figure 3), in3),
(Figure theindrainage
the drainage process, the nonwetting
process, the nonwetting fluidfluid is stopped at theatentrance
is stopped of a
the entrance
channel
of a channel until the pressure exceeds the capillary entry pressure. After passing this throat, the
until the pressure exceeds the capillary entry pressure. After passing this throat, the
nonwetting
nonwetting fluidfluid spontaneously
spontaneously invades
invades the the adjacent
adjacent larger
larger intersections.
intersections.

Figure
Figure2.
2.Image
Imageof
ofdisplacement
displacementfronts
fronts in
indrainage
drainage experiments.
experiments.
Energies 2019,
Energies 2019,12,
12,1390
x FOR PEER REVIEW 66 of 13
13
Energies 2019, 12, x FOR PEER REVIEW 6 of 13

Figure 3. An image of nonwetting fluids (white) displacing wetting fluids (black) in the microfluidic
device.
Figure 3. An image
Figure 3. imageofofnonwetting
nonwetting fluids
fluids (white)
(white) displacing
displacing wetting
wetting fluidsfluids (black)
(black) in the in the microfluidic
microfluidic device.
device.
3.1.1.
3.1.1. Phase Diagram
Diagram
3.1.1.The
Phase Diagram
The fractal dimensions (D (Dff))of
ofthe
thefronts
frontsprior
priortotoreaching
reachingthe theoutlet
outletare
areplotted
plotted asas a function
a function of
of log
log M M
The and
and log
log Ca
fractal Ca in Figure
in Figure(D
dimensions 4. The
4.f)The characteristic
characteristic
of the fronts prior of of
to DD that is
f fthat is the
reaching of significance
of significance to flow
to flow
outlet are plotted regimes is its
as a function itsof
gradient;
gradient;
log M and aasharp
sharp
log Cachange
change
in Figure ininthe4.gradient
the gradient
The of of
DfDcorresponds
f corresponds
characteristic of Df that to ato
istransition between
aoftransition toflow
between
significance flow regimes.
flow Inisthis
regimes.
regimes In
its
work,
this values
work, of
valuesD , along
of D , with
along images
with of phase
images of saturation,
phase are
saturation, used to
are estimate
used
gradient; a sharp fchange in the gradient of Df corresponds to a transition between flow regimes. In
f to the boundary
estimate the locations
boundary
for
thisvarious
locations
work,for flow
values regimes.
various
of D flow regimes.
f, along with images of phase saturation, are used to estimate the boundary
locations for various flow regimes.

Figure 4. Drainage
Figure 4. Drainage fractal
fractal dimension
dimension versus
versus log
log M
M and
and log
log Ca.
Ca.
Figure 4. Drainage fractal dimension versus log M and log Ca.
A phase diagram including boundaries for drainage displacement are shown on a log Ca − log M
A phase diagram including boundaries for drainage displacement are shown on a log Ca − log
plot in Figure 5. These boundaries may be affected by pore size distribution and domain size [5,9,18].
M plot in Figure
A phase 5. These
diagram boundaries
including may for
boundaries be affected
drainageby pore size distribution
displacement are shown on anda domain
log Ca − size
log
In Figure 4, the low region corresponding to log M > 2.3, log Ca < −5.3 on the Df surface reflects the
[5,9,18].
M plot inInFigure
Figure5.4,These
the low region corresponding
boundaries may be affectedto logby M > 2.3,
pore sizelog Ca < −5.3 on
distribution andthe Df surface
domain size
transition from capillary fingering regime to viscous fingering regime. The slope at approximately log
reflects the
[5,9,18]. transition
In Figure 4, thefrom
low capillary fingering regime
region corresponding to log to M
viscous fingering
> 2.3, log regime.
Ca < −5.3 The
on the Df slope
surfaceat
M < −0.5 and log Ca > −4.3 appears to indicate the transition from stable displacement to transition
approximately
reflects log M from
the transition < −0.5 and logfingering
capillary Ca > −4.3 appears
regime to indicate
to viscous the transition
fingering regime. The from stable
slope at
regime. The low region at 0.7 < log M < 2 and log Ca > −6.6 on the Df surface corresponds to the
displacement tolog
approximately transition
M < −0.5 regime.
andThe
loglowCa region
> −4.3atappears
0.7 < log to
M <indicate
2 and log Catransition
the > −6.6 on the Df surface
from stable
change from capillary fingering to transition regime. The boundaries (red solid boxes) of various flow
corresponds to
displacement to the changeregime.
transition from capillary fingering
The low region to <transition
at 0.7 log M < 2regime.
and log The
Ca >boundaries
−6.6 on the D(red solid
f surface
regimes for drainage displacement based on our microfluidic experiment study are shown in Figure 5.
boxes) of various
corresponds to the flow
change regimes for drainage
from capillary displacement
fingering based
to transition on ourThe
regime. microfluidic
boundariesexperiment
(red solid
study are
boxes) of shown
variousinflow
Figure 5.
regimes for drainage displacement based on our microfluidic experiment
study are shown in Figure 5.
Energies 2019, 12, 1390 7 of 13
Energies 2019, 12, x FOR PEER REVIEW 7 of 13
Energies 2019, 12, x FOR PEER REVIEW 7 of 13

Figure 5.
Figure 5. Drainage
Drainage flow
flow phase
phase diagram.
diagram. The solid (red)
The solid
solid (red) boxes
boxes areare based
based on
on our
our results;
results; the
the shaded
shaded
The (red) boxes
areas are
areas are based
based on
on work
work by
by Zhang
Zhang et
et al.
al. [5];
[5]; the
the dot-dash
dot-dash line
line is
is based
based on
on the
the study
study byby Liu
Liu et al.
et al. [38];
al. [38];
[38];
dot-dash line is based on the study by Liu et
and regimes delineated
delineated with
with dashed
dashed lines
lines are
arebased
basedononwork
workby byLenormand
Lenormandet etal.
al.[9].
[9].
and regimes on work by Lenormand et al. [9].

3.1.2. Width
Width of
3.1.2. Width of Stabilized
Stabilized Zone
Zone
3.1.2. of Stabilized Zone
Rapoport
Rapoport et et al. [28]
[28] proposed
proposed the the concept
concept of of the
the stabilized
stabilized zone,
zone, which
which isis aaa quasi-steady
quasi-steady region
region
Rapoport al. [28] proposed the concept of the stabilized zone, which is quasi-steady region
around
around the
thenonwetting/wetting
nonwetting/wetting displacement
displacement front front
that determines
that the transition
determines the between
transitionnonwetting
between
around the nonwetting/wetting displacement front that determines the transition between
and wettingand
nonwetting fluids.
wetting Results
fluids.from thisfrom
Results studythisimply a linear
study imply
imply relationship between the the
width of
nonwetting and wetting fluids. Results from this study aa linear
linear relationship
relationship between
between width
the width
the
of stabilized
the stabilized zone
zoneandandthethe
injection
injectionrate
rateininwhich
which the
the width
width appears
appears to
to be
be decreasing
decreasing inversely
inversely
of the stabilized zone and the injection rate in which the width appears to be decreasing inversely
proportionally
proportionally with with the
the displacement velocity [27]. In our study, the width of
displacement velocity of the
the stabilized
stabilized zonezone
proportionally the displacement [27]. In our study, the width of the stabilized zone
around
around the
the displacement
displacement front
front decreases
decreases when
when the
the injection
injection rate
rate is
is less
less than
than approximately
approximately 33 µL/min.
µL/min.
around the displacement front decreases when the injection rate is less than approximately 3 µL/min.
After
After increasing
increasingthe theinjection raterate
injection beyond
beyond3 µL/min, the width of the of stabilized zone beginszonetobegins
increase,
After increasing the injection rate beyond 33 µL/min,
µL/min, the width
the width the stabilized
of the stabilized zone begins to
to
see Figure
increase, see 6. This
see Figure observation
Figure 6. 6. This is in
This observationagreement
observation is is in with
in agreement results
agreement with reported
with results by
results reported Barenblatt
reported by et al.
by Barenblatt [27].
Barenblatt et et al.
al. [27].
[27].
increase,

Figure 6. Relation
Figure 6. Relation between
between the
the width
width of
of the
the stabilized
stabilized zone
zone of
of the
the interfaces
interfaces and
and the
the inverse
inverse flow
flow rates
rates
Figure 6. Relation between the width of the stabilized zone of the interfaces and the inverse flow rates
in drainage displacement.
in drainage displacement.
in drainage displacement.

3.2. Imbibition
3.2. Imbibition
Recorded images
Recorded images of
of imbibition
imbibition displacement
displacement fronts,
fronts, Figure
Figure 7,
7, show
show that
that when
when log
log M
M >> 0,
0,
displacement fronts exhibit both viscous and capillary fingering patterns; viscous fingers appear
displacement fronts exhibit both viscous and capillary fingering patterns; viscous fingers appear to to
 and log M < 1.5, capillary forces are significant, and the interconnected large capillary fingers appear
to dominate the observed flow pattern. After breakthrough, these capillary fingers grow laterally and
coalesce resulting in relatively large amounts of nonwetting fluid being entrapped in the pore
network. At log Ca > −5.5 and log M > 1.5, viscous forces become comparable to capillary forces and
several
Energies 2019,narrow-disconnected
12, 1390 viscous fingers appear in the flow pattern. When log M < 0.0 log Ca > 13
8 of
Energies 2019, 12, x FOR PEER REVIEW
−4.0, the displacing front is relatively flat and little to no trapping of the wetting fluid is observed 8 of 13

inside the porous medium; this behavior indicates a stable flow regime.
3.2.dominate
Imbibition at high Ca and capillary fingers seem to dominate low values of Ca [40]. At log Ca < −5.5
Displacement is affected by the geometry of the medium such as its aspect ratio. In imbibition,
and log M < 1.5, capillary forces are significant, and the interconnected large capillary fingers appear
this may leadimages
Recorded to the wetting phase flowing
of imbibition along corners
displacement fronts,when the 7,
Figure poreshowcenters
thatare occupied
when log M with > 0,
to dominate the observed flow pattern. After breakthrough, these capillary fingers grow laterally and
the nonwetting
displacement phase,
fronts alongboth
exhibit the wall roughness
viscous for strongly
and capillary water-wet
fingering systems,
patterns; or in the
viscous bulk of
fingers the
appear
coalesce resulting in relatively large amounts of nonwetting fluid being entrapped in the pore
channels, at
to dominate making
highCaCatheandflow of the fingers
capillary wettingseemphase potentiallylow
toforces
dominate more complicated
values of Ca than thatCaof< the
−5.5
network. At log > −5.5 and log M > 1.5, viscous become comparable to [40]. At log
capillary forces and
and nonwetting
log M < 1.5,phase [4,41].
capillary This
forces effect
are is observed
significant, and by
the comparing
interconnectedimbibition
large and
capillarydrainage
fingers phase
appear
several narrow-disconnected viscous fingers appear in the flow pattern. When log M < 0.0 log Ca >
distribution images, see Figures 3 and 8.
to dominate the observed
−4.0, the displacing frontflow pattern. flat
is relatively After breakthrough,
and these capillary
little to no trapping fingers fluid
of the wetting growislaterally
observed and
Based on Lenormand’s study [42], there are three synergistic mechanisms that govern the flow
coalesce
inside resulting
the porous inmedium;
relativelythislarge amounts
behavior of nonwetting
indicates fluidregime.
a stable flow being entrapped in the pore network.
behavior in imbibition: meniscus (interface) displacement, movement of the nonwetting fluid due to
At log Ca > − 5.5 and log M > 1.5, viscous forces become
Displacement is affected by the geometry of the medium such as its comparable to capillary forces
aspect ratio. and several
In imbibition,
the displacement of the meniscus, and flow of the wetting fluid (injectant) from the inlet boundary to
narrow-disconnected viscous fingers appear in the flow pattern. When
this may lead to the wetting phase flowing along corners when the pore centers are occupied log M < 0.0 log Ca >with−4.0,
the meniscus. In this study, since the pore network is a representation of a thin section of a Berea
thethe
displacing
nonwetting front is relatively
phase, along the flat androughness
wall little to nofortrapping
stronglyof the wetting
water-wet fluid is
systems, orobserved
in the bulk inside
of thethe
sandstone, the aspect ratio between pores and channels vary, and thus, both pore invasion and
channels,
porous making
medium; this the flow of
behavior the wetting
indicates a phase
stable flow potentially
regime. more complicated than that of the
collapse in a channel mechanism occur during the displacement.
nonwetting phase [4,41]. This effect is observed by comparing imbibition and drainage phase
distribution images, see Figures 3 and 8.
Based on Lenormand’s study [42], there are three synergistic mechanisms that govern the flow
behavior in imbibition: meniscus (interface) displacement, movement of the nonwetting fluid due to
the displacement of the meniscus, and flow of the wetting fluid (injectant) from the inlet boundary to
the meniscus. In this study, since the pore network is a representation of a thin section of a Berea
sandstone, the aspect ratio between pores and channels vary, and thus, both pore invasion and
collapse in a channel mechanism occur during the displacement.

Figure
Figure 7.7.Image
Imageofofdisplacement
displacement fronts
fronts in
in imbibition
imbibitionexperiments.
experiments.

Displacement is affected by the geometry of the medium such as its aspect ratio. In imbibition,
this may lead to the wetting phase flowing along corners when the pore centers are occupied with
the nonwetting phase, along the wall roughness for strongly water-wet systems, or in the bulk
of the channels, making the flow of the wetting phase potentially more complicated than that of
the nonwetting phase [4,41]. This effect is observed by comparing imbibition and drainage phase
distribution images, see Figures 3 and 8.
Figure 7. Image of displacement fronts in imbibition experiments.

Figure 8. An image of wetting fluids (black) displacing nonwetting fluids (white) in the microfluidic
device.

Figure
Figure 8. An
8. An image
image of wetting
of wetting fluids
fluids (black)
(black) displacing
displacing nonwetting
nonwetting fluids
fluids (white)
(white) inmicrofluidic
in the the microfluidic
device.
device.
Energies 2019, 12, 1390 9 of 13

Based on Lenormand’s study [42], there are three synergistic mechanisms that govern the flow
behavior in imbibition: meniscus (interface) displacement, movement of the nonwetting fluid due to
the displacement of the meniscus, and flow of the wetting fluid (injectant) from the inlet boundary
to the meniscus. In this study, since the pore network is a representation of a thin section of a Berea
sandstone, the aspect ratio between pores and channels vary, and thus, both pore invasion and collapse
in a channel mechanism occur during the displacement.
Energies 2019, 12, x FOR PEER REVIEW 9 of 13

PhaseEnergies
Diagram
2019, 12, x FOR PEER REVIEW 9 of 13
Phase Diagram
The
Phasefractal dimension values for imbibition experiments are shown in Figure 9. Large gradient
TheDiagram
fractal dimension values for imbibition experiments are shown in Figure 9. Large gradient
change of Df values correspond to transition of flow between different regimes. For instance, the slope
changeTheof Dfractal
f values correspond
dimension to transition
values of flow
for imbibition between different
experiments are shown regimes.
in FigureFor9.instance, the slope
Large gradient
on the surface between log M = −1.0 and log M = −0.3, and between log Ca = −3.3 and log Ca = −3.7,
onchange
the surface between
of Df values log M = to
correspond −1.0 and logofM
transition = −0.3,
flow and different
between betweenregimes.
log Ca =For −3.3 and logthe
instance, Caslope
= −3.7,
suggests that flow is transitioning away from the stable regime. A low region appears between
suggests that flow
on the surface is transitioning
between log M = −1.0 away
andfrom
log Mthe stable
= −0.3, andregime.
betweenA log
lowCa region
= −3.3appears
and log between
Ca = −3.7,log
logMMsuggests
==0.0
0.0and
and
that log
logflowMMis==transitioning
1.0 when
1.0 when log log Ca is
away
Ca is
frombetween
the stable
between −6.8
−6.8 regime.
and −
and−5.3,
A5.3,
low which
region
which maymay
appearscorrespond
betweentolog
correspond to flow
flow
transitioning away
M = 0.0 andaway
transitioning from
log M fromcapillary
= 1.0 fingering.
when log
capillary There
Ca is between
fingering. is a
There is −6.8deep
a deep region
andregion on the
−5.3, which surface
on the may when
correspond
surface 1.7 <
when 1.7to< loglog
flowM << 2,
M
which may indicate
2, transitioning
which away
may indicate flow
fromtransitioning
flowcapillary to viscous
fingering.
transitioning toThere fingering.
viscous FlowFlow
is a fingering.
deep region regime
on the boundaries
surface
regime when 1.7(red
boundaries solid
< log
(red <boxes,
Msolid
shown 2, in Figure
which may 10) for
indicate imbibition
flow displacement
transitioning to are
viscous estimated
fingering. based
Flow on
regime
boxes, shown in Figure 10) for imbibition displacement are estimated based on valuesf of Df (Figure values of
boundariesD (Figure
(red solid9) and
imagesboxes, shown
9) andofimages
fronts of in
(FigureFigure
fronts8). 10) for
(Figure 8). imbibition displacement are estimated based on values of D f (Figure
9) and images of fronts (Figure 8).

Figure 9. Imbibitionfractal
fractal dimension with
withrespect
respecttotologlog
MMandand
log log
Ca. Ca.
Figure9.9.Imbibition
Figure Imbibition fractal dimension
dimension with respect to log M and log Ca.

Figure 10. Imbibition

Figure flow
10. Imbibition phase
flow phasediagram
diagram with boundariesofof
with boundaries three
three flow
flow regimes
regimes (red(red boxes).
boxes).
Figure 10. Imbibition flow phase diagram with boundaries of three flow regimes (red boxes).
4. Conclusions
4. Conclusions
This study reports observation from a series of two-phase immiscible displacement experiments
across a wide range of injection rates using a microfluidic device featuring a quasi-2D complex
This study reports observation from a series of two-phase immiscible displacement experiments
network of channels. The experiments were conducted across a relatively large range of a parameter
Energies 2019, 12, 1390 10 of 13

4. Conclusions
This study reports observation from a series of two-phase immiscible displacement experiments
across a wide range of injection rates using a microfluidic device featuring a quasi-2D complex network
of channels. The experiments were conducted across a relatively large range of a parameter space
defined by viscosity ratio, M, and capillary number, Ca: log M = −1.0 to 4.2 and log Ca = −7.6 to −4.0
for drainage, and M = −1.0 to 2.3 and log Ca = −6.9 to −3.3 for imbibition.
Quantitative analysis, such as calculation of fractal dimension values of the displacement fronts,
are made possible by the high-resolution images. In the absence of gravity effects, the interplay
between capillary and viscous forces determine the character of flow patterns. The insights provided
by the quantitative analysis provides a measure to approximate and delineate various flow regimes.
Boundaries of various flow regimes are estimated for both drainage and imbibition based on phase
distribution images and the fractal dimension values of the displacement fronts. The Df -log M-log
Ca plots reveal that relatively large gradient changes of Df may correspond to a transition between
different flow regimes. The observations from drainage displacement experiments are in relatively
close agreement with flow regime boundaries suggested by Lenormand et al. [4]. In the drainage flow
regime diagram, the regions that correspond with transitions between distinct flow regimes appear to
constitute a slightly larger area of the overall diagram than inferred in literature from observations
in homogenous media strictly consisting of uniformly spaced cylindrical pillars [5]. As is evident
by the fractal dimension values, which is a measurement of the space filling capacity of a shape [43],
stable flow regime fills the two-dimensional space of the medium more fully than regimes exhibiting
fingering behavior; stable regimes displace the resident fluids uniformly and result in more efficient
recovery of the resident fluids. On the other hand, both capillary and viscous fingering regimes
develop fronts that bypass some of the resident fluid. The fingers in capillary dominated regime
are dispersed and fill the two-dimensional space more fully; hence their fractal dimension values
are higher compared to viscous dominated instabilities where few prominent fingers emerge and
overtake all the smaller ones. As a result, displacement fronts in the viscous fingering regime have
lower factual dimension values. Transition out of and into a flow regime is signified by a change in the
character of the displacement front, i.e., a relatively uniform front in the stable viscous, a multitude
of interconnected small fingers in capillary dominated, and few prominent ones in viscous fingering
regime. The change is reflected in the gradient of the fractal dimension values in the two-dimensional
parameter space of log Ca and log M: as a displacement transitions from one flow regime to another,
there is a sharp change in the aforementioned gradient of fractal dimension values.
This study extends the available experimental data on two-phase immiscible displacement
behavior using a water-wet microfluidic device, and the analysis and quantification of the space
filling capacity of the displacement fronts help estimate the boundaries of various flow regimes in
both imbibition and drainage. Flow regimes have significant implications in various multiphase flow
in porous media applications, such as geologic storage of CO2 and recovery operations of petroleum
reservoirs. In CO2 storage applications, flow regime affects storage capacity of the target formation.
In subsurface petroleum systems, flow regimes affect recovery efficiency and recycling operational
needs related to EOR processes.

Author Contributions: Conceptualization, S.A.A.; methodology, S.A.A.; formal analysis, S.A.A. and F.G.;
investigation, F.G.; data curation, F.G.; writing—original draft preparation, F.G.; writing—review and editing,
S.A.A.; visualization, F.G. and S.A.A.; supervision, S.A.A.; funding acquisition, S.A.A.
Funding: The corresponding author gratefully acknowledges the Donors of the American Chemical Society
Petroleum Research Fund (55795-DNI9) for partial support of this research. The analysis and writing stages of this
research were supported as part of the Center for Mechanistic Control of Water-Hydrocarbon-Rock Interactions
in Unconventional and Tight oil Formations (CMC-UF), an Energy Frontier Research Center funded by the U.S.
Department of Energy, Office of Science under DOE (BES) Award DE-SC0019165.
Energies 2019, 12, 1390 11 of 13

Acknowledgments: The corresponding author is grateful to Roland Horne and Cynthia Ross; the base image of the
sandstone was created by Cynthia Ross for Roland Horne, who graciously shared it with the corresponding author.
Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the design of the
study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to
publish the results.

Nomenclature
Non-Dimensional Numbers:
Ca capillary number; vµ1 /γ
M viscosity ratio; µ2 /µ1
Df fractal dimension
Acronyms and abbreviations:
EOR Enhanced oil recovery
Deionized DI
Greek Letters:
µ dynamic viscosity, cp
γ the interfacial tension, mN/m

References
1. Kenneth, C.C.; McDonald, K.; Marble, J.; Russo, A.E.; Brusseau, M.L. The Impact of Transitions between
Two-Fluid and Three-Fluid Phases on Fluid Configuration and Fluid-Fluid Interfacial Area in Porous Media.
Water Resour. Res. 2015, 51, 7189–7201.
2. Isha, S.; Bedeaux, D.; Kjelstrup, S.; Vassvik, M.; Sinha, S.; Hansen, A. Ensemble Distribution for Immiscible
Two-Phase Flow in Porous Media. Phys. Rev. 2017, 95, 023116.
3. Lincoln, P. Diffusion-Limited Aggregation and Two-Fluid Displacements in Porous Media. Phys. Rev. Lett.
1984, 52, 1621–1624.
4. Roland, L. Liquids in Porous Media. J. Phys. Condens. Matter 1990, 2, 79–88.
5. Changyong, Z.; Oostrom, M.; Wietsma, T.W.; Grate, J.W.; Warner, M.G. Influence of Viscous and Capillary
Forces on Immiscible Fluid Displacement: Pore-Scale Experimental Study in a Water-Wet Micromodel
Demonstrating Viscous and Capillary Fingering. Energy Fuels 2011, 25, 3493–3505.
6. Gvirtzman, H.; Roberts, P.V. Pore Scale Spatial Analysis of Two Immiscible Fluids in Porous Media. Water
Resour. Res. 1991, 27, 1165–1176. [CrossRef]
7. Grate, J.W.; Changyong, Z.; Wietsma, T.W.; Warner, M.G.; Anheier, N.C.; Bernacki, B.E.; Orr, G.; Oostrom, M.
A Note on the Visualization of Wetting Film Structures and a Nonwetting Immiscible Fluid in a Pore Network
Micromodel Using a Solvatochromic Dye. Water Resour. Res. 2010, 46, 11602–11608. [CrossRef]
8. Lenormand, R.; Zarcone, C.; Sarr, A. Mechanisms of the Displacement of One Fluid by Another in a Network
of Capillary Ducts. J. Fluid Mech. 1983, 135, 337–353. [CrossRef]
9. Lenormand, R.; Touboul, E. Numerical Models and Experiments on Immiscible Displacements in Porous
Media. J. Fluid Mech. 1988, 189, 165–187. [CrossRef]
10. Wang, Y.; Changyong, Z.; Ning, W.; Oostrom, M.; Wietsma, T.W.; Li, X.; Bonneville, A. Experimental Study of
Crossover from Capillary to Viscous Fingering for Supercritical CO2 –Water Displacement in a Homogeneous
Pore Network. Environ. Sci. Technol. 2012, 47, 212–218. [CrossRef]
11. Lenormand, R. Differents Mecanismes de Deplacements Visqueux et Cspillaires en Milieu Poreux: Diagramme De
Phase; Gauthier-Villars: Paris, France, 1985; Volume 301.
12. Schwille, F.; Pankow, J.F. Dense Chlorinated Solvents in Porous and Fractured Media-Model Experiments; Lewis
Publishers: Chelsea, MI, USA, 1988.
13. Changyong, Z.; Oostrom, M.; Grate, J.W.; Wietsma, T.W.; Warner, M.G. Liquid CO2 Displacement of Water in
a Dual-Permeability Pore Network Micromodel. Environ. Sci. Technol. 2011, 45, 7581–7588.
14. Chomsurin, C.; Werth, C.J. Analysis of Pore-Scale Nonaqueous Phase Liquid Dissolution in Etched Silicon
Pore Networks. Water Resour. Res. 2003, 39, 1265–1276. [CrossRef]
Energies 2019, 12, 1390 12 of 13

15. Werth, C.J.; Changyong, Z.; Brusseau, M.L.; Oostrom, M.; Baumann, T. A Review of Non-Invasive Imaging
Methods and Applications in Contaminant Hydrogeology Research. J. Contam. Hydrol. 2010, 113, 1–24.
[CrossRef]
16. Riazi, M.; Sohrabi, M.; Bernstone, C.; Jamiolahmady, M.; Ireland, S. Visualisation of Mechanisms Involved
in CO2 Injection and Storage in Hydrocarbon Reservoirsand Water-Bearing Aquifers. Chem. Eng. Res. Des.
2011, 89, 1827–1840. [CrossRef]
17. Kim, Y.; Wan, J.; Kneafsey, T.J.; Tokunaga, T.K. Dewetting of Silica Surfaces Upon Reactions with Supercritical
CO2 and Brine: Pore-Scale Studies in Micromodels. Environ. Sci. Technol. 2012, 46, 4228–4235. [CrossRef]
18. Liu, H.; Valocchi, A.J.; Kang, Q.; Werth, C. Pore-Scale Simulations of Gas Displacing Liquid in a Homogeneous
Pore Network Using the Lattice Boltzmann Method. Trans. Porous Media 2013, 99, 555–580. [CrossRef]
19. Ferer, M.; Ji, C.; Bromhal, G.S.; Cook, J.; Ahmadi, G.; Smith, D.H. Crossover from Capillary Fingering to
Viscous Fingering for Immiscible Unstable Flow: Experiment and Modeling. Phys. Rev. 2004, 70, 016303.
[CrossRef]
20. Rabbani, H.S.; Or, D.; Liu, Y.; Lai, C.-Y.; Lu, N.B.; Datta, S.S.; Stone, H.A.; Shokri, N. Suppressing Viscous
Fingering in Structured Porous Media. Proc. Natl. Acad. Sci. USA 2018, 115, 4833–4838. [CrossRef]
21. Cottin, C.; Bodiguel, H.; Colin, A. Drainage in Two-Dimensional Porous Media: From Capillary Fingering to
Viscous Flow. Phys. Rev. 2010, 82, 046315. [CrossRef]
22. Feng, G.; He, J.; Johnson, P.A.; Aryana, S.A. Stabilization of CO2 Foam Using by-Product Fly Ash and
Recyclable Iron Oxide Nanoparticles to Improve Carbon Utilization in Eor Processes. Sustain. Energy Fuels
2017, 1, 814–822.
23. Guo, F.; Aryana, S.A. An Experimental Investigation of Nanoparticle-Stabilized CO2 Foam Used in Enhanced
Oil Recovery. Fuel 2016, 186, 430–442. [CrossRef]
24. Alaskar, M. In-Situ Multifunctional Nanosensors for Fractured Reservoir Characterization. Ph.D. Thesis,
Stanford University, Stanford, CA, USA, 2013.
25. Alaskar, M.; Li, K.; Horne, R. Influence of Particle Size on Its Transport in Discrete Fractures: Pore-Scale
Visualization Using Micromodels. In Proceedings of the Thirty-Eighth Workshop on Geothermal Reservoir
Engineering, Stanford, CA, USA, 11–13 February 2013.
26. Aryana, S.A.; Kovscek, A.R. Experiments and Analysis of Drainage Displacement Processes Relevant to
Carbon Dioxide Injection. Phys. Rev. 2012, 86, 066310. [CrossRef]
27. Barenblatt, G.I.; Patzek, T.W.; Silin, D.B. The Mathematical Model of Nonequilibrium Effects in Water-Oil
Displacement. SPE J. 2003, 8, 409–416. [CrossRef]
28. Rapoport, L.A.; Leas, W.J. Properties of Linear Waterfloods. J. Pet. Technol. 1953, 5, 139–148. [CrossRef]
29. U.S. Energy Information Administration. Oil and Natural Gas Resource Categories Reflect Varying Degrees of
Certainty. Available online: www.eia.gov/todayinenergy/detail.php?id=17151. (accessed on 14 March 2019).
30. Liu, Y.; Pengfei, L.v.; Yao, L.; Lanlan, J.; Tetsuya, S.; Song, Y.; Wu, B.; Liu, S. CO2 /Water Two-Phase Flow
in a Two-Dimensional Micromodel of Heterogeneous Pores and Throats. RSC Adv. 2016, 6, 73897–73905.
[CrossRef]
31. Wang, Y.; Aryana, S.A.; Furtado, F.; Ginting, V. Analysis of Nonequilibrium Effects and Flow Instability in
Immiscible Two-Phase Flow in Porous Media. Adv. Water Resour. 2018, 122, 291–303. [CrossRef]
32. Aryana, S.A.; Kovscek, A.R. Nonequilibrium Effects and Multiphase Flow in Porous Media. Trans. Porous Media
2013, 97, 373–394. [CrossRef]
33. Guo, F.; Aryana, S.A. Improved Sweep Efficiency Due to Foam Flooding in a Heterogeneous Microfluidic
Device. J. Pet. Sci. Eng. 2018, 164, 155–163. [CrossRef]
34. Ren, G.; Rafiee, J.; Aryana, S.A.; Younis, R.M. A Bayesian Model Selection Analysis of Equilibrium and
Nonequilibrium Models for Multiphase Flow in Porous Media. Int. J. Multiph. Flow 2017, 89, 313–320.
[CrossRef]
35. Majhi, S.; Al-Otaibi, B.; Al-Omair, S.; Abdullah, M.B. Mitigating Water Recycling in a Heterogeneous
Carbonate Reservoir. In SPE Kuwait Oil and Gas Show and Conference; Society of Petroleum Engineers:
Mishref, Kuwait, 2015.
36. Lv, W.; Siyu, Y.; Yongzhi, Y.; Shenglai, Y.; Xiaohong, M.; Li, S.; Wu, C.; Hao, M. Produced Gas Utilisation
Strategy Study on Low Permeability Reservoir CO2 Eor Process. In SPE Asia Pacific Oil & Gas Conference and
Exhibition; Society of Petroleum Engineers: Perth, Australia, 2016.
Energies 2019, 12, 1390 13 of 13

37. Belhaj, H. Management of Injected Nitrogen into a Gas Condensate Reservoir. Ngeniería Investig. 2016, 36,
52–61. [CrossRef]
38. Liu, H.; Yonghao, Z.; Valocchi, A.J. Lattice Boltzmann Simulation of Immiscible Fluid Displacement in Porous
Media: Homogeneous Versus Heterogeneous Pore Network. Phys. Fluids 2015, 27, 052103. [CrossRef]
39. Yamabe, H.; Takeshi, T.; Liang, Y.; Matsuoka, T. Influence of Fluid Displacement Patterns on Seismic Velocity
During Supercritical CO2 Injection: Simulation Study for Evaluation of the Relationship between Seismic
Velocity and CO2 Saturation. Int. J. Greenh. Gas Control 2016, 46, 197–204. [CrossRef]
40. Tsakiroglou, C.D.; Theodoropoulou, M.; Karoutsos, V.; Papanicolaou, D.; Sygouni, V. Experimental Study
of the Immiscible Displacement of Shear-Thinning Fluids in Pore Networks. J. Coll. Interf. Sci. 2003, 267,
217–232. [CrossRef]
41. Chang, L.-C.; Tsai, J.-P.; Shan, H.-Y.; Chen, H.-H. Experimental Study on Imbibition Displacement
Mechanisms of Two-Phase Fluid Using Micro Model. Environ. Earth Sci. 2009, 59, 901. [CrossRef]
42. Lenormand, R.; Zarcone, C. Role of Roughness and Edges During Imbibition in Square Capillaries. In SPE
Annual Technical Conference and Exhibition; Society of Petroleum Engineers: Houston, TX, USA, 1984.
43. Falconer, K. Fractal Geometry: Mathematical Foundations and Applications; John Wiley & Sons: West Sussex,
UK, 2004.

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).