SPE-180388-MS

Experimental and Simulation Study of Water Shutoff in Fractured Systems

Using Microgels
Ali Goudarzi, The University of Texas at Austin; Ali Alhuraishawy, Missouri University of Science and
Technology; Pongpak Taksaudom, and Kamy Sepehrnoori, The University of Texas at Austin; Baojun Bai,
and Abdulmohsin Imqam, Missouri University of Science and Technology; Mojdeh Delshad, The University of
Texas at Austin

Copyright 2016, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Western Regional Meeting held in Anchorage, Alaska, USA, 23–26 May 2016.

This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents
of the paper have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect
any position of the Society of Petroleum Engineers, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written
consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may
not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.

Abstract
Conformance control has long been a compelling subject in improving waterflood oil recovery. By
blocking the areas previously swept by water, subsequently injected water is allowed to access the
remaining unswept portions of the reservoir and thereby increase the ultimate oil recovery. One technique
that has recently received a great deal of attention in achieving the so-called ⬙in-depth water shut-off’ is
preformed gel injection. However, processing and predicting the performance of these gels in complex
petroleum reservoirs is extremely challenging. As target reservoirs for gel treatments are mainly those
with fractures or ultra-high permeability streaks, the ability to model the propagation of gels through a
fractured reservoir was considered as a new challenge for this research study.
The primary objectives of this work are to conduct laboratory work to understand the transport and
propagation of microgel through fractures and develop conformance control schemes using a reservoir
simulator to help in screening oil reservoir targets for effective particle gel applications to improve sweep
efficiency and reduce the water production. Fractured experiments using transparent apparatus were
performed to observe gel transport in matrix and fractures. The same set up was used to observe the effects
of gel strength, gel particle size, and fracture size on gel transport. Numerical simulation of fluid-flow in
fractured reservoirs can be computationally difficult and time consuming due to the large contrast between
matrix and fracture permeabilities and the extremely small fracture apertures and the need for using
unstructured gridding. In this work, a model that accurately represents the complex reservoir features,
chemical properties, and displacement mechanisms is developed.
The five-spot transparent fracture experiments allowed us to identify the transport mechanisms of
microgels through fractures-conduits and also the control variables. With an integration of comprehensive
gel transport modules and a novel Embedded Discrete Fracture Modeling (EDFM), gel rheological and
transport properties of shear thinning viscosity, adsorption, resistance factors, and residual resistance
factor, using multiple sets of fractures with dip angles and orientations were captured. The models were
validated against lab measurements and implemented into a reservoir simulator called UTGEL.
2 SPE-180388-MS

The mechanistic models and numerical tool developed will help to select future conformance control
candidates for a given field and to optimize the gel chemistry and treatment.
Introduction
Excess water production is becoming a critical problem in mature oil fields as the oil reservoir is subject
to long water flooding. Gel treatments if applied correctly can reduce excess water production and
improve conformance by filling water channels and fractures. In principle gel treatments can be divided
into in-situ gels and preformed gels. Traditionally in-situ gels were used for controlling water production
where a mixture of polymer and crosslinker which is called gelant is injected into the formation to form
gel at reservoir conditions for blocking the channels (Sydansk and Moore, 1992; Jain et al., 2005). The
disadvantages of this method are the effect of adsorption and formation water on the crosslinking reaction
and possible damages on the low permeability unswept oil zone (Bai, 2010). The new technology for gel
treatment is to form the gel at surface conditions and inject the preformed gel into the reservoir. The
process can overcome problems such as lack of control on gelation time and uncertainties due to the effect
of adsorption and shear degradation which usually occurs in traditional in- situ gel.
There are essentially four different types of microgels which can block the high permeability layers and
divert the injected water into low permeability unswept oil zones (Figure 1): Performed Particle Gel
(PPG), Thermally Active Polymer (TAP), Colloidal Dispersion Gel (CDG), and pH-sensitive microgel.
PPGs are dried superabsorbent crosslinked polymer powders that can swell up to 200 times their original
size (Bai et al., 2007a, 2007b; Muhammed, 2011). These particles are prepared by combining monomers,
controlled monomers, stable cross-linkers, initiators, and other agents in aqueous solution. This new
technology can be used for either conformance control or water shutoff or even both in some cases. This
technology has been successfully applied for more than 5000 wells (Bai et al., 2013). Bai et al. (1999)
used gel to divert injected water into un-swept oil zones of the formation and decrease the flow capacity
of channels or fractures. Seright (2000) showed that preformed gels have better efficiency through
fractures compared to in-situ gels and impose less damage in low permeability oil zones. Feng et al. (2003)
showed that microgels are good candidates for water shutoff and profile control without any problem of
plugging. Rousseau et al. (2005) used sandpack experiments to study the flow and transport of microgels.

Figure 1—Microgel adsorbs and retains on the rock surface in thief zones and diverts the water into low permeability areas.

Another type of particles gels called nanoparticles have drawn a great deal of attention from the oil
industry for their EOR potential (Hendraningrat et al., 2013), namely, their ability to modify certain
factors in reservoir formations. Kong and Ohadi (2010) emphasized that nano-agents may significantly
increase oil recovery by modifying surface tension. Hendraningrat et al. (2013) studied the disjoining
pressure as a displacement mechanism due to the existence of nanoparticles in various permeability core
SPE-180388-MS 3

plugs. The results show that hydrophilic nanoparticles have capability to decrease the contact angle of the
aqueous phase and increase water-wetness. Ogolo et al. (2012) investigated the effect of some nanopar-
ticles (oxides of aluminum, zinc, magnesium, iron, zirconium, nickel, tin and silicon) on oil recovery. The
coreflood tests on transport behavior of nanogel through sandpack showed that both the resistance factor
and the residual resistance factor decrease with an increase in the size of the swollen particles, because
the larger particles were weaker than the small particles, indicating that the gel strength is more important
than the particle size for nanogel particle injectivity and permeability reduction (Almohsin et al., 2014).
Imqam et al. (2014) investigated PPG injection and placement mechanisms through conduits or large
channels, under conditions where the channel opening size was larger than, equal to, or smaller than the
swollen PPG size. Their results indicated that PPG strength affected injectivity more significantly than did
the particle opening ratio. Additionally, the size of the PPGs decreased during their transport though the
channels due to both dehydration and breakdown. Subsequent work conducted by Imqam and Bai (2015a)
to further understand the resistance of PPGs to water flow. They designed a large transparent channel to
investigate factors such as PPG strength, PPG size, and load pressure on the PPG placement behavior.
They reported that the PPG did not fully block the channel but rather formed gel pack permeability along
the channel model. When gel pack permeability decreased means PPG blocking efficiency to water flow
increased. Results showed that the gel pack permeability decreased when the gel strength, particle size,
and load pressure increased. Additionally, the gel strength impacted the gel injection pressure more than
the particle opening ratio did (Imqam, 2015a, 2015b).
PPG strength is an essential parameter for designing gel treatment in conformance control processes.
There are several measurements for gel strength (Sydansk, 1990). One such measurement is the elastic
strength which can be defined as the resistance to physical deformation that a gel will exhibit while
extruding through a restriction in its flow path, such as the restriction in a fracture flow path. Another
measurement is the yield strength. This gel strength is measured by placing a gel sample in a large
container having a small orifice and then increasing the pressure in the container until the gel flows
through the orifice. The yield strength of a gel is often much larger than its elastic strength (Sydansk,
1990).

Experimental Procedure

Materials
Preformed Particle Gel (PPG): A super absorbent polymer (20-30) mesh size was used as the
preformed particle gel for this study. The particle was synthesized by a free radical process using
acrylamide, acrylic acid, and N,N’-methylene- bisacrylamide. Most PPGs reach full swelling in half an
hour, but a field operation usually takes a few hours to a few months, so we used fully swelling particles
in our experiments.
Brine: Sodium chloride (1.0 wt.% NaCl) was used to prepare the swollen gels.
Oil: A heavy mineral oil was used with a viscosity of 195 cp.
Sandstone rock: Berea sandstone was used with 21 cm length x 21 cm width x 2 cm height shown in
Figure 2.
4 SPE-180388-MS

Figure 2—The image of the Berea sandstone core used for gel conformance control study.

Experimental Setup
The setup was made with transparent acrylic board to provide a transparent window to observe fluid
flow and gel transport. Two acrylic plates were used to construct the fracture model. Square pocket (22
cm long, 22 cm wide, and 2.2 cm thickness) was drilled in the center of one side of the acrylic plates. The
top of sandstone core was covered by sealing material. There are 5 inlets/outlets on the model. During the
water flooding and gel treatment, only two ports were used and the others were shut in. These two ports
are set at corners along the diagonal line to simulate 5-spot scenario as shown in Figure 3. The rock used
for experiment was sandstone and it was placed into the model and sealed using sealing material. The
schematic of the setup is shown in Figure 4.

Figure 3—Schematic diagram of the 5-spot pattern in the fracture experiment.

SPE-180388-MS 5

Figure 4 —Schematic diagram of the apparatus used in the fracture experiment.

Fracture Design (Straight connection between injector and producer)

At least one fracture connects injector and producer directly. The fracture model (Figure 5) was used
to test gel transportation in the model with fractures parallel to the direction of fluid flow.

Figure 5—Fractures parallel to the direction of fluid flow with one fracture connects injector and producer directly.

The model was vacuumed and saturated with brine (1.0 wt.% NaCl) and the pore volume was
calculated. Heavy oil (195 cp) was injected to reach irreducible water saturation (Swi⫽ 32% and Soi⫽ 68
%). Afterwards, the fracture (2.5 mm fracture width) was created in the core. A summary of fracture
properties is given in Table 1.
6 SPE-180388-MS

Table 1—Fracture properties used in microgel experiment.

Length (cm) Width (cm) Height (cm) Total Volume (cm3)

Core 21 21 2 Number of Fractures SS2

Length Fracture 19.5 0.25 2 3 29.25

Medium Fracture 9 0.25 2 2 9
Short Fracture 3 0.25 2 2 3
Total Fractures Vol. 41.25
Total Pore Vol. 231.25

Experimental Procedure
Pieces of sandstone were fixed into the fractured model using sealing material. The top of core was
covered by sealing material to prevent the flow between the pieces of sandstone and the upper acrylic
plate. After vacuumed, saturated, the connate water saturation and initial oil saturation were calculated,
the following steps were performed:
1. Waterflooding
We kept port 1 (inlet) and 3 (outlet) open and shut in ports 2, 4, and 5. The brine was injected from
port 1 and fluid from port 3 was produced. The injection was continued at the flow rate of 1 cc/min
until no oil was produced from the outlet and the injection pressure was recorded using a computer.
After no additional oil was produced, we changed flow rates to 2, 4, and 6 cc/min to measure the
water effective permeability. The oil recovery, water cut, and effective permeability are also
measured during the experiment. A summary of rock and fluid properties is presented in Table 2.

Table 2—Core and fluid properties used in microgel experi-

ment.
Porosity 21.85%
Dry Weight 2010.5 gr
Wet Weight 2203.2 gr
Oil Viscosity 195 cp
Swi 32%
Soi 68%

2. Gel Treatment
We injected 5000 ppm preformed particle gel (PPG) into the core and collect fluid from the outlet
and then, injection pressure was recorded. The gel injection was stopped when gel production from
the outlet was detected. Finally, we calculated oil recovery improvement based on the collected
fluid from the outlet.
3. Post Waterflooding
Brine was injected from port 1 at flow rate of 1 cc/min and fluid from the outlet was collected. We
recorded injection pressure data and the flow rate was increased to 2, 4, and 6 cc/min consequently
to measure permeability after gel treatment. Finally we calculated overall oil recovery and water
residual resistance factor.
Experimental Results and Analysis
Oil Recovery Factor
After 0.7 pore volume (PV) of 1.0 wt.% NaCl was injected and injection pressure was stabilized, there was
no oil recovery during first water flooding because injected brine flows through the fracture easily.
SPE-180388-MS 7

However, the oil recovery factor increased to 21.28 % during 5000 ppm PPG injection. Finally, the oil
recovery factor reached 24.38 % during post water flooding as shown in Figure 6a. The flow rate was
increased to 2, 4, and 6 cc/min and no significant effects in the oil recovery and injection pressure was
observed.

Figure 6 —Oil recovery (a) and water cut (b) for the fracture 5-spot experiment.

Water Cut
After 0.2 pore volume injection, water production started and water cut reached 100% with no change
during first water flooding. During PPG (5000 ppm) injection, the water cut reduced to 88.54% and post
water flooding leads to water cut of 92.84% as shown in Figure 6b.

Injection Pressure
During the first water flooding, there was no increase in injection pressure and the injection pressure was
very low (0.1 psi) since injected brine flows through the fracture easily and produced at the outlet. During
PPG injection, the injection pressure increased till 15 psi and then the pressure started to decline when the
PPG was produced from the outlet. During the post water flooding, the injection pressure was stabilized
at 1 psi as shown in Figure 7a. When the flow rate was increased to 2, 4, and 6 cc/min, the water residual
resistance factor increased as shown in Figure 7b.
8 SPE-180388-MS

Figure 7—Injection pressure (a) and residual resistance factor (b) for the fracture 5-spot experiment.

UTGEL Reservoir Simulator Development Description

The UTGEL is a finite difference three-dimensional multiphase multi-component chemical composition
reservoir simulator that can be used for modeling chemical enhanced oil recovery (EOR) processes. Its
main applications are water flooding, controlling excess water production using different types of gels,
and polymer flooding. The UTGEL can be used to model both laboratory and field scales. A compre-
hensive module is available for polymer and gel rheological and transport properties, such as shear
thinning viscosity, adsorption, resistance factor, and inaccessible pore volume.

PPG Transport Model

There are many conditions that affect how particles flow and transport through porous media. The
viscosity, resistance factor, and residual resistance factor are most important properties for modeling PPG
flow in porous media. The resistance factor and residual resistance factor are functions of the salinity and
flow rate based on the laboratory results. To model the flow of PPG particles through pore throats, the
swelling ratio and subsequent size of the swelled particles need to be calculated.
The average pore throat radius was calculated using porosity and permeability:
(1)

where the average permeability, , was approximated using the following:

(2)

where Kx, Ky, and Kz are directional permeabilities, u x1, u y1, u z1 are water fluxes in each direction
and u is the aqueous phase flux.
In each gridblock, the pore throat size was calculated based on Eq. 1, and then permeability and
porosity were assigned to each cell. The PPG will move out of a gridblock depending on the size of the
particles in comparison to the pore throat diameter. If the PPG cannot flow out of the gridblock, the
resistance factor is calculated and the aqueous viscosity is increased accordingly (Goudarzi et al., 2013,
2014). The conditions for passing PPG particle through the pore throat for weak and strong PPG particles
(Bai et al., 2004) are as follows:
● For weak PPG particles: if the PPG particle diameter is less than 5.7 x dp.
● For strong PPG particles: if the PPG particle diameter is less than 1.3 x dp.
SPE-180388-MS 9

where dp is the average pore throat diameter. Using the above criteria for weak and strong gels, the
PPG particles would pass through the pore throat. If the above conditions for a specific gridlock hold and
the PPG can pass through the pore throat, then the gel particles will enter that specific grid block and the
PPG will cause resistance to the water flow. The PPG will increase the viscosity of the aqueous phase and
a new effective viscosity for water will be calculated as defined below:
(3)

(4)

The resistance factor (RF) is used during PPG injection. The residual resistance factor (RRF) is
calculated for post water injection. This increase in water viscosity leads to reduction of water phase
mobility, improvement in the mobility ratio, and subsequently delayed water production.
Swelling Ratio Model
The swelling ratio can be defined as the ratio of the PPG particle volume before and after swelling. Bai
(2010) and Imqam et al. (2014) reported a relationship for the swelling ratio as a function of the salinity
based on laboratory measurements. Both studies showed that PPG can swell very quickly within 60
minutes and that final swelling ratio depends on the salt concentration. The higher the salt concentration
the lower the swelling ratio. This result is presumably due to the static electric repulsive force and charge
balance. At low salt concentrations, the electric repulsive force will separate the gel molecules and create
more space for water to enter (Bai et al., 2004; Imqam et al., 2014).
The following empirical correlation was developed to estimate the swelling ratio vs. the effective
salinity to fit the laboratory data:
(5)

where a p and np are model parameters, SF is the swelling ratio, and CSEP is the effective salinity (cSEP
⫽ cs ⫹ ␤p C6) in meq/ml, which takes into account the combined effect of anions (C5) and divalent cations
(C6) on the swelling ratio. pP is the input parameter and is obtained based on lab data. The effect of pH
is not considered in this model.
PPG Viscosity Model
Viscosity of an aqueous solution containing gel is modeled as a function of the gel concentration and the
water viscosity as given below (Thurston et al. 1987):
(6)

where C ppg,1 is the PPG concentration in aqueous phase, ␮w is the water viscosity, and A ppg,1 and A
ppg,2 are model parameters.

PPG Resistance Factor Model with Salinity Effect

Gel can reduce the water effective permeability, where the degree of permeability reduction depends on
the gel type, salinity, water hardness, shear effects, and rock properties. The resistance factor (RF) is
determined by the ratio of the differential pressure for the PPG injection to that of the initial water
injection expressed as
(7)

where kw, kPPG are effective permeabilities during waterflood and PPG injection, and ␮w, ␮PPG are
water and PPG viscosities, and ⌬Pw, ⌬PppG are the pressure drop during waterflood and PPG injection.
It is clear that the resistance factor decreases as the flow rate increases indicating the shear thinning
behavior of PPGs (Zhang and Bai, 2010; Imqam et al., 2014). The viscoelastic behavior of the PPG relates
10 SPE-180388-MS

to the coil structure of the polyacrylamide molecules which produce a flexible nature (Green and Willhite
1998). The resistance factor is sensitive to the water hardness (i.e., calcium and magnesium concentra-
tions). The following correlation was proposed but additional laboratory data is required to validate it
(Goudarzi et al. 2015):
(8)

The resistance factor is expressed as

(9)

(10)

where a11, a12, and b1 are model parameters, is the equivalent shear rate, and CSEP is the effective
salinity, ␥c is the shear rate correction, is magnitude of water flux, and Krl is water relative
permeability. The proposed model considers the effect of shear rate and salinity on resistance factor.

Residual Resistance Factor with Salinity Effect Model

The residual resistance factor, RRF, is defined as the ratio of the pressure drop during post-water injection
to the pressure drop during the initial water flood as follows:
(11)

where ⌬PBaseWaler, ⌬PPostWater are the pressure drop during the initial water and post-water injections.
Similar to the resistance factor, the residual resistance factor decreases as the flow rate increases
(Zhang and Bai, 2010; Imqam et al., 2014). The residual resistance factor was sensitive to the brine
hardness (i.e., calcium and magnesium concentrations). The following correlation was used but needs
additional laboratory data for validation:
(12)

Accordingly, the final developed model for the residual resistance factor is expressed as follows:
(13)

where a2l, a22 and b2 are model parameters and is the equivalent shear rate. The new proposed
model considers combined effects of shear rate and salinity on residual resistance factor.

PPG Retention Model

A new model was developed and implemented into the simulator to consider the PPG retention. The
UTGEL uses the Langmuir isotherm for PPG retention and includes the PPG concentration and salinity
as shown below:
(14)

(15)

where CPPG1 is the PPG concentration in the aqueous phase 1 and the parameters a14,1, a14,2, and b14
are input parameters.
SPE-180388-MS 11

Embedded Discrete Fracture Model (EDFM)

The conformance control processes including PPG treatment are usually performed in mature water-
flooded reservoirs which typically contain fractures or very high permeability streaks. Modeling the
propagation of PPG through these fractures and conduits were considered as new challenges for this
research study. Numerical simulation of fluid-flow in fractured reservoirs is complex due to the large
contrast between matrix and fracture permeabilities, the extremely small size of fracture apertures, and the
unstructured grid.
Several approaches have been proposed to model fracture networks that can be classified as two major
classes of models: Dual continuum (Dual Porosity/Dual Permeability, DPDP) and Discrete Fracture
Models (DFM). The Dual continuum models provide an efficient approach to describe highly heteroge-
neous fractured formations using two domains, one for fracture system and other one for rock matrix.
However, they suffer from high degree of simplification in a way that they cannot consider the effect of
each fracture explicitly. On the other side, discrete fracture models are limited by unstructured gridding
algorithms and simulation times even though they are more accurate. Unstructured gridding imposes more
complexity for field-scale simulations (Figure 8).

Figure 8 —Discrete fracture model using unstructured grid.

To overcome problems associated with unstructured gridding, a new model has been developed called
Embedded Discrete Fracture Model (EDFM). First, Li and Lee (2008) adopted a hierarchical modeling
approach to represent fractures with different length scales. Later, Moinfar et al. (2013) employed this
model to represent fractures with different dip and orientations in GPAS (in-house fully implicit parallel
compositional reservoir simulator). A 3D synthetic illustration of fracture inclination which comprises
eight fractures is shown in Figure 9.
12 SPE-180388-MS

Figure 9 —A synthetic 3D fractured reservoir with eight inclined macrofractures (Moinfar et al., 2012).

In this model, fracture planes are discretized by cell boundaries. In fact, for flow in rock matrix, the
structured grid is used and unstructured grid is used to model flow in fracture network. The fracture
control volumes are considered as nonneighboring connections (NNC). A preprocessing step is developed
to locate the fractures and to calculate the transmissibility factors among non-neighboring connections
(Cavalcante Filho et al., 2015). Since the fracture control volumes are introduced inside the matrix grid
domain, three new connections are defined based on non-neighboring connections. For each of these new
connections, a transmissibility factor is calculated as a preprocessing step explained briefly in the
following:
a. For matrix-fracture connection (Connection type I)
(16)

where A is the area of fracture cell inside the grid block, k is the harmonic average of permeability,
and d is the normal distance between center of matrix gridblock and fracture cell.
b. For fracture-fracture intersection (Connection type II)
(17)

(18)

(19)

where k is the fracture permeability, ␻ is the fracture aperture, L is the length of intersection line
between two fractures bounded in a gridblock, and the subscripts f1 and f2 represent the intersected
fracture number 1 and number 2.
c. For fracture-fracture connection of the same fracture plane (Connection type III)
(20)

where k is the fracture permeability, A is the length of intersection times the aperture, and d is the
distance between center of two segments.
SPE-180388-MS 13

The EDFM approach was implemented into UTGEL to provide efficient and robust tool to study the
flow of gels in complex fracture system (Shakiba, 2014; Shakiba and Sepehrnoori, 2015). The EDFM
implementations created a more realistic environment to study the behavior of fractured reservoirs and aid
in designing gel injection through fractures and conduits. Taksaudom (2014) investigated the effect of
PPG for a complex fracture conduit model which contains many fracture streaks with different dip angles
and the results showed that there was approximately 7% improvement in oil recovery with PPG treatment
compared to waterflood.

Simulation Results and Discussions

Slanted Fracture Plane Model

To substantiate the benefits of EDFM in gel transport simulation, we performed PPG simulations in a
synthetic model where a ⫾ 50° slanted fracture plane was placed through the reservoir between the
injector and producer pair. With the preprocessor, the slanted fracture plane was created and embedded
precisely in the reservoir model, and the transmissibility factors between non-neighboring connections
were calculated. The 3-dimensional views of the slanted fracture plane generated are shown in Figure 10.
The parameters used in the simulation are given in Table 3. The PPG treatments used in this case study
consisted of 0.8 PV of water injection and 0.2 PV of PPG suspension injection. However, to investigate
the effect of PPG injection timing, four different simulations were carried out in this study:

Figure 10 —3D views of the slanted fracture plane generated by the preprocessor for EDFM.
14 SPE-180388-MS

Table 3—Input parameters for the slanted fracture plane model.

Model 3-Dimentional Cartesian
Number of matrix gridblocks 50 ⫻ 50 ⫻ 15
Number of fracture gridblocks (NNC) 884
⌬x, ⌬y, ⌬z 2, 2, 2 ft
Fracture aperture 0.15 ft
Porosity 0.25
Permeability 100 mD
Fracture permeability 50,000 mD
Ratio of Kv/Kh 0.1
Oil viscosity 5 cp
Water viscosity 1 cp
Initial reservoir pressure 1100 psi
Production bottomhole pressure constraint 1000 psi
Injection rate 1000 ft3/day
Total injection period 1 PV
Injection PPG concentration 1000 ppm

➢ Waterflood: 1.0 PV of water injection.

➢ Early PPG injection: 0.1 PV of pre-treatment water injection, 0.2 PV of PPG suspension injection,
and 0.7 PV of post-treatment water injection.
➢ Intermediate PPG injection: 0.3 PV of pre-treatment water injection, 0.2 PV of PPG suspension
injection, and 0.5 PV of post-treatment water injection.
➢ Late PPG injection: 0.5 PV of pre-treatment water injection, 0.2 PV of PPG suspension injection,
and 0.3 PV of post-treatment water injection.
To demonstrate the behavior of the injectant (water) inside the fracture plane model created using
EDFM approach, snapshots of water saturation profile by grid block and a sector model with a proper cut
plane were generated. Figure 11 shows the water saturation profile of the base case simulation of the
slanted fracture plane model at 0.5 PV, along with a sector model cut by a plane at a 50 ° slanted angle.
It can be observed in Figure 11 that the injected water propagated faster through the slanted fracture plane
creating an abnormally slanted shape of higher water saturation at the producer end.

Figure 11—Water saturation profile for waterflood in a slanted fracture plane model at an output time of 0.5 PV.
SPE-180388-MS 15

The simulation results of all 4 cases are summarized in Table 4. The incremental oil recoveries varied
between 6% and 13% due to different PPG injection timing design. Figure 12 and Figure 13 illustrate
the oil recovery and water cut profiles from all cases, respectively. The earlier PPG injection resulted in
higher incremental oil recovery for this case study. The water cuts plotted in Figure 14 reveal that the
improvement in water cut became less significant from the early PPG treatment (approximately 35%
water cut reduction) to the late treatment (approximately 20% water cut reduction).

Table 4 —Simulation results for the slanted fracture plane model.

Simulation Cases Start time of PPG Injection (PV) Oil Recovery (%) PPG Incremental Recovery (%)

1. Waterflood – 55.77 –
2. Early PPG treatment 0.1 68.25 12.49
3. Intermediate PPG treatment 0.3 65.75 9.99
4. Late PPG treatment 0.5 62.55 6.78

Figure 12—Oil recovery vs. time for slanted fracture plane model.

Figure 13—Water cut vs. time for slanted fracture plane model.
16 SPE-180388-MS

Figure 14 —3D views of the fracture conduit generated by the preprocessor for EDFM.

Complex Fracture Conduit Model

Another scenario where EDFM can be useful is illustrated here with a complex fracture conduit model.
Many fracture streaks with different dip angles were generated to create an extensive fracture conduit in
this case study. The preprocessor was employed to embed the conduit at the precise coordinates of the
reservoir model and calculate the transmissibility factors between non-neighboring connections. The
3-dimensional views of the fracture conduit embedded are shown in Figure 14. For this study we aligned
the conduit to be in one vertical plane so that the injection and production wells could be placed directly
on the opposite sides of the conduit. Hence, it was convenient to visualize the fluid behavior around the
conduit and the wells.
Two simulations were performed for this case study; (1) the waterflood case consisted of 1 PV of water
injection, and (2) the PPG treatment case consisted of 0.3 PV of pre-treatment water injection, 0.2 PV of
PPG injection, and 0.5 PV of posttreatment water injection. The simulation parameters are given in Table
5.
SPE-180388-MS 17

Table 5—Input parameters for the complex fracture conduit model.

Model 3-Dimentional Cartesian

Number of matrix gridblocks 40 ⫻ 20 ⫻ 20

Number of fracture gridblocks (NNC) 85
⌬x, ⌬y, ⌬z 2, 2, 1.5 ft
Fracture aperture 0.25 ft
Porosity 0.25
Permeability 50 mD
Fracture permeability 80,000 mD
Ratio of Kv/Kh 0.25
Oil viscosity 2.5 cp
Water viscosity 1 cp
Initial reservoir pressure 1100 psi
Production bottomhole pressure constraint 800 psi
Injection rate 600 ft3/day
Total injection period 1 PV
Injection PPG concentration 2000 ppm

The behavior of the injectant (water) inside the fracture conduit model created with the EDFM
approach is demonstrated in Figure 15. It can be observed in the figure that the injected water propagated
faster through the complex fracture conduit creating an abnormal front of the injected water, which
resulted in partial water breakthrough at the producer at the time of 0.3 PV of water injection.

Figure 15—Water saturation profile for waterflood in a fracture conduit model at an output time of 0.3 PV.

The results of oil recovery versus pore volumes injected for the waterflood and the PPG treatment
simulations are shown in Figure 16. The results of the simulation suggested that a PPG treatment led to
an incremental oil recovery of approximately 7%. The water cut vs. PV was also plotted in Figure 17. For
18 SPE-180388-MS

the PPG case, the water cut reduction as high as 20% was observed after about 0.2 PV of PPG injection.
The simulation run times for the waterflood and PPG treatment were 17.5 and 19.7 CPU hrs, respectively.

Figure 16 —Oil recovery vs. time for fracture conduit model.

Figure 17—Water cut vs. time for fracture conduit model.

From the two cases studied, a slanted fracture plane and a complex fracture conduit with several dip
angles were successfully generated and implemented in the reservoir models using the EDFM prepro-
cessor. With the non-neighboring connections (NNC) concept of the EDFM approach, the simulations of
fluid and gel transport in reservoirs containing these uncharacteristic fracture passages were successfully
performed with representative results. The computational times used in the simulations were fairly
reasonable considering the number of gridblocks and the contrast between the size of the fracture aperture
and the matrix gridblocks. No local grid refinement was required and the fractures were easily placed in
a realistic manner in the simulation domain.
SPE-180388-MS 19

Summary and Conclusions

● Fractured coreflood experiments were conducted to understand the transport of PPG microgels and
their impact on flow conformance and reducing water production through fractured systems.
● The fractured coreflood results indicated that PPG has capability to block fracture pathways and
increase the incremental oil recovery by 24% over waterflood.
● We developed empirical correlations for resistance factor (RF) and residual resistance factor
(RRF) using different size conduits and for a wide range of flow rate, brine salinity and hardness.
Also, we have developed models for gel rheology, gel adsorption, swelling ratio, and gel viscosity.
The developed correlations are implemented in UTGEL reservoir simulator.
● The gel transport models were implemented in a reservoir simulator and validated against different
types of laboratory experiments.
● The numerical studies indicated that main PPG design variables are treatment size, PPG concen-
tration, permeability contrast between layers, reservoir heterogeneity, water cut, mobility ratio, and
the ratio of vertical to horizontal permeability.

Acknowledgments
We acknowledge the financial support from the RPSEA subcontract#11123-32 for water management in
the small producer program. We also acknowledge additional support of company members of the
Reservoir Simulation JIP in the Center for Petroleum and Geosystems Engineering (CPGE) at The
University of Texas at Austin.

Nomenclature
Appg,1, Appg,2 Gel viscosity parameters in UTGEL
a11, a12, b1 Resistance factor model parameters
a21, a22, b2 Residual resistance factor model parameters
a14,1, a14,2, b14 PPG retention model parameters
ap, np Swelling ratio model parameters
PPG concentration in aqueous phase, Adsorbed concentration of PPG,
ppm
CSEP Effective salinity, meq/ml
dp,rh Average pore throat diameter, Pore throat radius, ft
d Distance between fracture and matrix
Permeability, Average permeability, md
kw, kPPG Effective permeability during water flood, Effective permeability dur-
ing PPG injection
kx, ky, kz Directional permeabilities, md
krᐉ Relative permeability for phase ᐉ
⌬PPPG PPG injection pressure drop, psi
⌬PBaseWater, ⌬PPostWater Initial water injection pressure drop, Post water injection pressure drop,
psi
RF, RRF Resistance factor, Residual Resistance factor
SF Swelling ratio
Sᐉ Saturation of phase ᐉ
u1 Aqueous phase flux
ux1, uy1, uz1 Components of darcy flux for aqueous phase, ft/Day
␮aqueous phase Water phase viscosity (cp)
20 SPE-180388-MS

␮1, ␮w, ␮PPG Aqueous solution containing gel viscosity, Water viscosity, PPG vis-
cosity, cp
T Fracture transmissibility
␥c Shear rate correction
␾ Porosity
␻ Fracture aperture

References
Almohsin, A., Bai, B., Imqam, A., Wei, M., Kang, W., Delshad, M., and Sepehmoori, K., 2014. ⬙Transport of Nanogel
through Porous Media and Its Resistance to Water Flow⬙ Paper SPE 169078, presented at the SPE Symposium on
Improved Oil Recovery, Tulsa, Oklahoma, April 12-16.
Bai, B., Li., Y., and Liu, X., 1999. ⬙New development of water Shutoff and profile control in oilfields in china,⬙ Journal
of Oil Drilling and Production Technology, 20: 3.
Bai, B., Li, L., Liu, Y., Coste, J., and Li, L., 2004, ⬙Preformed Particle Gel for Conformance Control: Transport Mechanism
Through Porous Media,⬙ Paper SPE 89468, presented at the SPE/DOE 14th Symposium on Improved Oil Recovery,
Tulsa, Oklahoma, April, 17-21.
Bai, B., Li, L., Liu, Y., Liu, H., Wang, Z., and You, C., 2007a. ⬙Conformance control by preformed particle gel: factors
affecting its properties and applications,⬙ SPE Reservoir Evaluation and Engineering Journal, 10: 415–421.
Bai, B., Liu, Y., Coste, J. P., and Li, L., 2007b. ⬙Preformed particle gel for conformance control: transport mechanism
through porous media,⬙ SPE Reservoir Evaluation and Engineering Journal, 10: 176 –184.
Bai, B., 2010, ⬙Preformed Particle Gel (PPG) for Conformance Control,⬙ RPSEA contract No: 07123-02, Midland, Texas,
February, 3.
Bai, B., Wei, M., and Liu, Y., 2013. ⬙Field and Lab Experience with a Successful Preformed Particle Gel Conformance
Control Technology,⬙ Paper SPE 164511, presented at the SPE Production and Operations Symposium, Oklahoma
City, Oklahoma, March 23-26.
Cavalcante Filho, J. S. A., Shakiba, M., Moinfar, M., and Sepehrnoori, K., 2015. ⬙Implementation of a Preprocessor for
Embedded Discrete Fracture Modeling in an IMPEC Compositional Reservoir Simulator,⬙ Paper SPE 173289,
Presented at the SPE Reservoir Simulation Symposium, Houston, Texas, February 23-25.
Feng, Y., Tabary, R., Renard, M., LeBon, C., Omari, A., Chauveteau, G., 2003. ⬙Characteristics of microgels designed for
water shutoff and profile control,⬙ Paper SPE 80203, presented at SPE International Symposium on Oilfield
Chemistry, Houston, Texas, February 5-7.
Goudarzi, A., Zhang, H., Varavei, A., Hu, Y., Delshad, M., Bai, B., and Sepehrnoori, K., 2013, ⬙Water Management in
Mature Oil Fields using Preformed Particle Gels,⬙ Paper SPE 165356, presented at the SPE Western Regional &
AAPG Pacific Section Meeting, Monterey, California, April, 19-25.
Goudarzi, A., Almohsin, A., Varavei, A., Delshad, M., Bai, B., and Sepehmoori, K., 2014. ⬙New Experiments and Models
for Conformance Control Microgels,⬙ Paper SPE 169159, presented at the SPE Symposium on Improved Oil
Recovery, Tulsa, Oklahoma, April 12-16.
Goudarzi, A., Zhang, H., Varavei, A., Taksaudom, P., Hu, Y., Delshad, M., Bai, B., and Sepehmoori, K., 2015. ⬙A
laboratory and simulation study of preformed particle gels for water Conformance Control,⬙ Fuel Journal, 140:
502–513.
Green, D. W., and Willhite, G. P. 1998, Enhanced Oil Recovery, AIME Memorial Fund, Society of Petroleum Engineers.
Hendraningrat, L., Li, S., and Torsaeter, O., 2013. ⬙A Coreflood Investigation of Nanofluid Enhanced Oil Recovery in
Low-Medium Permeability Berea Sandstone,⬙ Paper SPE 164106, presented at the SPE International Symposium on
Oilfield Chemistry, The Woodlands, Texas, April 8-10.
Imqam, A., Bai, B., Al-Ramadan, M., Wei, M., Delshad, M., and Sepehmoori, K., 2014. ⬙Preformed particle gel extrusion
through open conduits during conformance control treatments,⬙ SPE 169107 presented at SPE Improved Oil Recovery
Symposium, Tulsa, Oklahoma, April 12-16.
Imqam, A., and Bai, B., 2015a. ⬙Optimizing the strength and size of preformed particle gels for better conformance control
treatment,⬙ Fuel Journal, 148: 178 –185.
Imqam, A., 2015b. ⬙Particle Gel Propagation and Blocking Behavior Through High Permeability Streaks and Fractures,⬙
Ph.D. dissertation, Missouri University of Science and Technology, Rolla, Missouri.
Jain, R., McCool, C.S., Green, D.W., Willhite, G.P., and Michnick, M.J., 2005, ⬙Reaction Kinetics of the Uptake of
Chromium (III) Acetate by Polyacrylamide,⬙ SPE Journal 10:247–255.
SPE-180388-MS 21

Kong, X., and Ohadi, M., 2010. ⬙Applications of Micro and Nano Technologies in the Oil and Gas Industry- An Overview
of the Recent Progress,⬙ Paper SPE 138241, presented at the Abu Dhabi International Petroleum Exhibition and
Conference, Abu Dhabi, UAE, November 1-4.
Li, L., and Lee, S. H., 2008. ⬙Efficient field-scale simulation of black oil in a naturally fractured reservoir through discrete
fracture networks and homogenized media,⬙ SPE Reservoir Evaluation & Engineering Journal, 11: 750 –758.
Moinfar, A., Varavei, A., Sepehrnoori, K., and Johns, R. T., 2012. ⬙Development of a Novel and Computationally-Efficient
Discrete- Fracture Model to Study IOR Processes in Naturally Fractured Reservoirs,⬙ Paper SPE 154246, presented
at the Eighteenth SPE Improved Oil Recovery Symposium, Tulsa, Oklahoma, April 14-18.
Moinfar, A., Varavei, A., Sepehrnoori, K., and Johns, R. T., 2013. ⬙Development of a Coupled Dual Continuum and
Discrete Frac ture Model for the Simulation of Unconventional Reservoirs,⬙ Paper SPE 163647, presented at the SPE
Reservoir Simulation Symposium, The Woodlands, Texas, February 18-20.
Muhammed, F. A., 2011. ⬙Experimental study of the interaction between surfactant and preformed particle gels,⬙ Master’s
Thesis, Missouri University of Science and Technology, Rolla, Missouri.
Ogolo, N. A., Olafuyi, O. A., and Onyekonwu, M. O., 2012. ⬙Enhanced Oil Recovery Using Nanoparticles,⬙ Paper SPE
160847, presented at the Saudi Arabia Section Technical Symposium and Exhibition, Al-Khobar, Saudi Arabia, April
1-4.
Rousseau, D., Chauveteau, G., Renard, M., Tabary, R., Zaitoun, A., Mallo, P., Braun, O., and Omari, A., 2005. ⬙Rheology
and transport in porous media of new water shutoff/conformance control microgels,⬙ Paper SPE 93254, presented at
SPE International Symposium on Oilfield Chemistry, Houston, Texas, February 2-4.
Seright, R. S., 2000. ⬙Gel propagation through fractures,⬙ Paper SPE 59316, presented at the SPE/DOE Symposium on
Improved Oil Recovery, Tulsa, Oklahoma, April 3-5.
Shakiba, M., 2014. ⬙Modeling and Simulation of Fluid Flow in Naturally and Hydraulically Fractured Reservoirs Using
Embedded Discrete Fracture Model (EDFM),⬙ Master’s Thesis, The University of Texas at Austin.
Shakiba, M., and Sepehrnoori, K., 2015. ⬙Using Embedded Discrete Fracture Model (EDFM) and Microseismic Moni-
toring Data to Characterize the Complex Hydraulic Fracture Networks,⬙ Paper SPE 175142, presented at SPE Annual
Technical Conference and Exhibition, Houston, Texas, September 28-30.
Sydansk, R. D., 1990. ⬙A Newly Developed Chromium (III) Gel Technology,⬙ SPE Reservoir Engineering Journal, 5:
346 –352.
Sydansk, R.D., and Moore, P.E., 1992, ⬙Gel Conformance Treatments Increase Oil Production in Wyoming,⬙ Oil and Gas
Journal 40 –45.
Taksaudom, P., 2014. ⬙Simulation Study of Preformed Particle Gel for Conformance Control,⬙ Master’s Thesis, The
University of Texas at Austin.
Zhang, H., and Bai, B., 2010, ⬙Preformed Particle Gel Transport through Open Fractures and its Effect on Water Flow,⬙
Paper SPE 129908, presented at the SPE Improved Oil Recovery Symposium, Tulsa, Oklahoma, April, 24-28.