Fuel 207 (2017) 226–239

Contents lists available at ScienceDirect

Fuel
journal homepage: www.elsevier.com/locate/fuel

Full Length Article

Numerical simulation study of factors affecting relative permeability

modification for water-shutoff treatments
Dheiaa K. Alfarge a, Mingzhen Wei b,⇑, Baojun Bai b
a
Iraqi Ministry of Oil, Missouri University of Science and Technology, United States
b
Missouri University of Science and Technology, United States

h i g h l i g h t s

 Identified eight parameters influencing RPM treatment performance.

 Provided the guidelines to select the best candidate well for RPM treatments.

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

Article history: Polymer solutions and gels are frequently used to control excessive water production in oil and gas wells
Received 18 January 2017 by reducing the permeability to water flow to a greater extent than to oil or gas flow (Relative
Received in revised form 6 June 2017 Permeability Modifiers, RPM, or disproportionate permeability reduction, DPR). RPM and DPR can be used
Accepted 8 June 2017
as synonymous in this study; however, specialists use RPM term for the cases in which permeability
reduction is less compare to DPR. The significance of RPM agents, chemicals using in water-shutoff treat-
ments, is that their placement does not require mechanical isolation. However, the performance of RPM
Keywords:
treatment is still poor in field applications. This study applied numerical simulation methods to investi-
Water-shutoff treatments
Gel treatments
gate the factors impacting RPM treatments on reservoir (Macroscopic) level. Furthermore, Design of
Relative permeability modifiers Experiments (DOE) was used to rank these factors based on their influence on RPM performance (water
Disproportionate permeability reduction cut reduction and oil recovery improvement).
Polymer gels property The results indicated that there are nine parameters which can enhance or downgrade the success of
DPR treatments, treatments at oil/gas wells when use of DPR polymer or gel results in Disproportionate
Permeability Reduction of treated formation. The performance of DPR treatment was more pronounced at
low oil density, low oil viscosity, high gel penetration depth, and at high permeability heterogeneity
among layers (linear flow is more dominated). However, the performance of DPR treatments was down-
graded if the treatments were applied at high production flowrate, low ratio of residual resistance factor
for water (Frrw) to residual resistance factor for oil (Frro), and high G-shape (crossflow indicator) values.
Moreover, when the capillary forces dominate the flow (capillary-viscous number > 10), RPM results
were not largely encouraging due to water blocking effect. On the other side, in the viscous dominated
flow, RPM performance was more pronounced. These factors which were studied in this work can pro-
mote a short-term successful remedy, a long-term successful treatment, or even a failed treatment.
Some of these factors can be controlled; the operator can choose the optimum level of the parameter, like
production flowrate. However, other factors cannot be controlled, but the value of this study is still
increase probability of success the treatment prior to field application.
Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction environmental damage, low oil recovery, and hydrostatic loading.

Excessive water production makes reservoir life shorter and
worse economically due to corrosion of tubulars, fines migration,
⇑ Corresponding author.
E-mail address: weim@mst.edu (M. Wei).
Hill et al. [56] estimated the total cost associated with separation,
treatment and disposal of produced water at $50 billion annually
which urges most specialists to find appropriate solutions for
excessive water production. Generally, different solutions for
water production control in oil and gas reservoirs were suggested
according to the source and reason of produced water in hydrocar-
bon reservoirs [30]. The reasons causing excessive water

http://dx.doi.org/10.1016/j.fuel.2017.06.041
0016-2361/Ó 2017 Elsevier Ltd. All rights reserved.
 D.K. Alfarge et al. / Fuel 207 (2017) 226–239
Nomenclature
ADMAX Maximum Adsorption Capacity, lbmole/ft3
DOE Design of Experiments
DOE Design of Experiments
DPR Disproportionate Permeability Reduction
Frro Residual Resistance Factor for Oil Phase
Frrw Residual Resistance Factor for Water Phase
Gshape Cross flow indicator
K Permeability
K1/K2 Ratio of High-Permeability Zone to Low-Permeability
Zone
production are commonly due to mobility issues, fractures, high-
permeability channels, or heterogeneous features which provide
preferential paths with least resistance to the fluid being injected
to sweep hydrocarbons, and lead to an early breakthrough for dis-
placing phase. The usual solution for these problems and to maxi-
mize the swept areas is to place sealants or blocking agents in the
least resistance paths. Polymer, gels and other types of confor-
mance materials are common examples of permeability-reducing
agents that can fill fractures and high-permeability channels
through injectors or producers to generate flow diversion and
increase sweep efficiency [54].
It is known that the gel treatments can be conducted in three
locations of hydrocarbon reservoirs: a) injection wells which is
called injection profile control, b) production wells which is called
water-shutoff, and c) in depth of reservoir which is called in-depth
diversion process. Each method has its advantages and disadvan-
tages. However, the main advantage of water-shutoff treatments
is the immediate response, while its disadvantages are low success
rate and the risk to damage oil zone [55].
One of the critical methods used in production wells as a water-
shutoff treatment is called Disproportionate Permeability Reduc-
tion (DPR) or Relative Permeability Modifier (RPM). This terminol-
ogy came from noticing the ability of polymers and some gels to
reduce the permeability to water flow (Krw) to a greater extent
than to oil or gas flow (Kro or Krg). The DPR property of gels and
polymer has a significant role in many of hydrocarbon reservoir
cases especially when mechanical isolation process is difficult to
be performed during gel placement process [41]. However, there
is a clear disagreement among the investigators about the main
mechanism beyond DPR behavior. This disagreement led to the
lack of understanding for this property and resulted in absence of
a standardized method to predict DPR performance in field appli-
cations. A brief description about these mechanisms would be
introduced since the discussion of the DPR mechanisms details
are out of this study scope. The objective from this study is to form
a prediction methodology for DPR success or failure depending on
reservoir and well candidate conditions. In addition, this study
would give details about the factors which impact DPR perfor-
mance on reservoir level (macroscopic level) and how to select
the best candidate well.
2. Disproportionate permeability reduction
DPR is the property which some polymers and weak gels cru-
cially have for reducing the permeability to water to a greater
extent than to oil or gas flow [9,10,60,44,65,28,11,41]. Water-
shutoff treatment by using DPR fluid is interestingly effective for
reducing water production in production wells which cannot be
generally treated with conventional methods like mechanical iso-
lation [21]. Although Seright [28] reported different types of gels
227
KH Horizontal Permeability
kV Vertical permeability
Perm-k Vertical Permeability
Qo Oil Flowrate (STB/day)
RPM Relative Permeability Modifiers
RRFT Residual Resistance Factor for the Adsorbing Compo-
nent in Rock Type
SCTR Sector
VE Vertical Equilibrium
which produce DPR property, the performance of DPR treatments
in field applications have a high ambiguity. However, [44] reported
significant successful jobs in field applications which used DPR flu-
ids in Mid-Continent Region for water-shutoff purposes. Therefore,
it is clear that there is a lack in understanding the behavior of DPR
treatments on macroscopic level. This study indicated that care-
fully selected-candidates, wells and reservoirs, are the critical
point to give a successful DPR treatment.
3. A critical review about DPR mechanisms
The ability of polymers and some gels to reduce the permeabil-
ity to water to a greater extent than to oil or gas flow urges differ-
ent researchers to investigate why these chemical agents produce
this behavior. Many previous investigators tried to explain differ-
ent mechanisms for DPR agents. However, our literature review
came with ten proposed mechanisms by different investigators,
but no agreement among the previous investigators on a unique
mechanism. Although some researchers think DPR resulted from
a combination of two or more from these ten mechanisms; another
opinion said that DPR property could be caused by hysteresis effect
because types of fluids are changing in the reservoir formation
before and during gel injection. However, [20] concluded that hys-
teresis has not effect to produce DPR behavior. The goal beyond
focusing on identifying the correct DPR mechanism is to help in
understanding and predicting the behavior of this treatment in
production wells. Also, if the correct mechanism has been known,
this would help to improve this treatment more by improving its
mechanisms. Table 1 summarized the proposed mechanisms with
their investigators, proposal of each mechanism, the opinions
which conflict with, and the weak points in each one. These results
regarding DPR mechanisms are solely based on the review and
analysis for different resources from lab works and field applica-
tions which have been reported by different investigators. It is
clear that the conditions which had been used by the investigators
are different from each other, but we tried to explain the strength
of each mechanism depending on their weak points and the
physically-based strength. Depending on the critical review con-
ducted, the segregated pathway mechanism is the most acceptable
one; therefore, this mechanism would be simulated in this study.
4. Numerical simulation methodology
STARS simulator ([53]), one of the CMG packages, was used to
simulate creating a DPR chemical agent. Since the most common
mechanism for DPR is segregated pathways mechanism
[44,19,39], this mechanism was represented in this work by pene-
trating the gels to the water zone deeper as compared to the oil
zone. The cases which were modeled in this work are heteroge-
228 D.K. Alfarge et al. / Fuel 207 (2017) 226–239

Table 1
The proposed mechanisms for DPR with their weak points.

DPR Mechanisms Proposal Investigated by Weak points Not supported by

1-Wall Effect and Gel 1-wall effect can explain DPR when the gelant is prepared Liang et al. [58] It would not explain DPR Al Sharji et al. [1],
Droplet from or match the wetting phase of the rock. 2. Gel droplet property happening at small oil Liang et al. (2000)
model explains DPR when the gelant is prepared from or residual saturation
match non-wetting phase of rock.
2-Gravity Effect The density of water soluble gel (usually 99% water) Liang et al. [21] 1-Frr is insensitive for change in White et al. [44],
= density of brine. Therefore; Gels would go to water rather direction and orientation 2- Nilsson et al. [25],
than oil. Then, gel would reduce Krw more than Kro. Different oil densities = Same Liang et al. [21]
Frro
3-Lubrication Effect The interface between oil and adsorbed polymer would Prado Paez et al. [59], DPR would happen even when Liang et al. [21],
lubricate path of oil rather than water. Liang et al. [21], Zai- water and oil have the same Nilsson et al. [25]
toun and Kohler [48] viscosity
4-Rock Wettability DPR is due to polymer adsorption on water-wet rock walls Zaitoun and Kohler DPR treatment is significant in Liang et al. [20],
Change and Water/ [48], Liang et al. [21], intermediate wet rocks not in Liang et al. [19]
Oil Pathways Seright et al. [31] water wet ones.
Constriction
5- Segregated Pathways The water based gel would flow through most parts of White et al. [44], In transparent micromodels, gel Al Sharji et al. [1]
Mechanism pores which are available to brine Nilsson et al. [25], Al goes for both oil and water
Sharji et al. [1] pathways
6- Capillary Forces and DPR resulted from the balance between capillary forces Liang et al. [19], Se- Change the confining pressure Liang et al. [19]
Gel Elastic Effect and gel elasticity right [63] and gel elasticity would not
support this theory
7-Polymer Leaching DPR due to polymer leaching from gel during water Liang et al. [19] Both of Frrw and Frro are Seright [61],
from Gel and injection and not leaching through oil injection decreasing with flowrate Willhite et al.
Reducing Brine following power law model [45], Yan et al.
Mobility Mechanism [46]
8- Gel Swelling in Water DPR due to water-based gel is shrinking in oil and swelling Alsharji et al. [1], No change in gel volume after Alsharji et al. [1],
and shrink in Oil in water Liang et al. [21] contact with oil and water by Liang et al. [21]
video monitoring
9- Polymer Adsorption Polymer layer would be formed on the crevices between Alsharji et al. [52], Why does DPR happen in oil wet Liang et al. [19]
Entanglement grains and handers only water Zitha et al. [51]. system?
10- Gel Deformation or Oil would deform and dehydrate the gels while water Krishnan et al. [57], Both of water and oil would Zaitoun et al.
Dehydration would not. Willhite et al. [45] deform the gel [49], Liang et al.
[19]

neous linear system of two layers with one injector and one pro-
ducer as shown in Fig. 1, and the second case is a heterogeneous
radial system of two layers with five-spot pattern as shown in
Fig. 2. The reservoir characterizations and fluid properties are
shown in Table 2. Although there are three types of gels producing
DPR property which are in situ gels, preformed gels and partially
preformed gel, in situ gel was used as the DPR chemical agent in
this simulator. The type of gel which was simulated in this study
is water based gel with concentrations illustrated in Table 3. The
reaction frequency factor between x-linker and polymer is 3240
which control how fast the cross-linker would interact with poly-
mer. The adsorption model was built by using Langmuir coeffi-
cients correlation. The main parameters for adsorption model are
shown in Table 4. The adsorption properties such as the compo-
nent adsorption and inaccessible pore volume are function of the
formation permeability. Reservoir heterogeneities would make
these properties of gels to vary largely in different parts of reser-
voir and that is a good representation for what is going on in the
field.
4.1. Linear system description
It is known that the crossflow from one layer to another
resulted from one or all the four driving forces which are capillary,
viscous, gravity and dispersion making the flow in porous media
[50]. Therefore; we used in this case a linear reservoir system con-
taining two layers having different permeability with crossflow
existing, with one producer and one injector to investigate the
effect of these forces on DPR performance. A geometric coordinate
system having three dimensions (i, j, k) was applied to represent
reservoir dimensions in this model. The number of grids in x, y, z
directions was 50, 1, and 6 respectively. The total number of grid
blocks was 300. The OOIP is 4.56E + 05 STB in undersaturated
reservoir with initial pressure of 5000 psi. The dimensions of the

Fig. 1. Reservoir model (linear system-two layers), (A) Permeability distribution (B) Adsorbed Gel amount distribution.
 D.K. Alfarge et al. / Fuel 207 (2017) 226–239 229

Fig. 2. 3-D View of radial system (five spot-pattern of two layers).

Table 2 first 5 blocks which are closest to the producer has the following
Input data of fluids and reservoir properties. dimensions 20 feet in the x direction, 100 feet in the y direction,
Property Value and 10 feet in the z direction while the rest of grids have the
Reservoir temperature (F) 140
dimensions 50 ft, 100 ft, 10 ft in x, y, z respectively. The reason
Water density (lb/ft^3) 62.4 beyond making the first ten grids which are closest to the producer
Oil density (lb/ft^3) 50 are smaller than the rest of grids is to monitor the small changes in
Oil viscosity (cP) 1 saturation, gel penetration, and adsorption in these cells. Fig. 1
Water viscosity (cP) 0.5
explains the 3D view of linear system model which has two wells,
Reservoir Pressure (PSI) 5000
Top of reservoir (ft) 9000 one producer located in block (1, 1, 1:6), and one injector located in
Number of layer 2 block (50, 1, 1:6). The injection flow rate and production flow rate
KH1 (md) 1000 are equal to 600 STB/D. The other fluid and reservoir properties are
KH2 (md) 10000 listed in Table 2.
kV (md) 0.1*KH
Porosity 1 0.20
Porosity 2 0.25 4.2. Radial system

In this case, radial reservoir system contains two layers with

Table 3 one producer and four injectors. These two layers divided to six
Gelant component concentrations. grids vertically to see any change in fluid saturation or any other
properties. The permeability of the top three grids is equal to
Component Mole Fraction%
1000 md (horizontal permeability) and 10,000 md for the bottom
Water 0.999863404
three grids. The number of grids in x, y, z directions was 97, 97,
Polymer 4.8839e-006
X-linker 0.000131712 and 6 respectively. The total number of grid blocks was 56,454.
Total 1 The OOIP is 1.87E + 06 STB in undersaturated reservoir with initial
pressure of 5000 psi. Fig. 2 explains the 3-D view of radial system

Table 4
Adsorption model parameters.

Parameters Value Unit Rock type

Langmuir Isothermal Coefficients Adsorption isothermal parameter A 11.46 lbmole/ft^3 1&2
salt effect parameter 0 dimensionless 1&2
Adsorption isothermal parameter B 5540000 lbmole/ft^3 1&2
Rock dependent parameters Max. adsorption capacity (ADMAXT) 0.00000259 lbmole/ft^3 1
Residual Adsorption Level (ADRT) 0.00000259 lbmole/ft^3 1
Accessible pore volume (PORFT) 0.01 dimensionless 1
Accessible Resistance Factor 20000 dimensionless 1
Max. adsorption capacity (ADMAXT) 0.00000459 lbmole/ft^3 2
Residual Adsorption Level (ADRT) 0.00000459 lbmole/ft^3 2
Accessible pore volume (PORFT) 1 dimensionless 2
Accessible Resistance Factor 80000 dimensionless 2
230 D.K. Alfarge et al. / Fuel 207 (2017) 226–239
model. It contains five wells, one producer located in block (49, 49,
1:6), and four injectors located in block (1, 1, 1:6), (97, 97, 1:6), (1,
97, 1:6), and (97, 1, 1:6) respectively. The same operating parame-
ters, fluid properties and reservoir criteria which were described
for linear system were used in radial system. Also, the rock and
fluid data is the same for both linear system and the radial system.
Two cases for both of linear system and radial system were used to
investigate the crossflow effect on DPR performance. One case is
with crossflow by setting vertical permeability to one tenth of hor-
izontal permeability and another case is without crossflow by rep-
resenting shale barrier conditions.
5. Results and discussion
5.1. Oil viscosity effect
DPR treatment was applied at different oil viscosity values
which are 1, 10, 20, 30, 40, and 50 cP. The results indicated that
DPR could give better results in both water cut reduction and oil
recovery improvement when it is applied in low-viscous oil rather
than high-viscous oil with the same other conditions as shown in
Fig. 3. For example, in linear with cross flow case, water cut
reduced in these different oil viscosities values by 26%, 20%, 17%,
15%, 14%, and 13% respectively. Accordingly, oil recovery factor
improved by 9%, 7%, 6%, 5%, 3%, 2.5% respectively. The other three
cases have approximately the same trend, but with different values
where we noticed the rank from the best scenario to the worse sce-
nario as linear without cross flow, linear with cross flow, radial
without cross flow, and the radial system with cross flow respec-
tively. The reasoning beyond DPR success at low-viscous oil is
more pronounced as compared with high-viscous oil is due to dif-
ferent scenarios in which water invades oil zone after DPR treat-
ment performed. In high-viscous oil reservoirs, water would
cross to the oil zone in fingering style (leaky piston displacement)
which increases water blocking problem. However, in low-viscous
oil reservoirs, water would cross to the oil zone as piston like dis-
placement which reduces water blocking effect. To sum up, DPR
treatment at low viscous oil gives better results than in high-
viscous oils for the same other conditions.
Fig. 3. Effect of oil viscosity on DPR performance (water cut reduction) in different systems.
5.2. Gel penetration depth effect
For all cases, the DPR treatment was applied at different gel
penetration depth of 10, 15, 20, 25, 30, and 35 ft. Fig. 4 showed that
DPR could give better results in both water cut reduction and oil
recovery improvement when the DPR fluid penetrates deeper in
the formation. However, this conclusion conflicts with Stavland
[40] conclusion regarding the most successful DPR would happen
at high ratio of Frrw/Frro with low gel volume injected (low gel
penetration depth). He reasoned his conclusion depending on the
understanding for that low-penetrated gel would not reduce sig-
nificantly oil productivity index. However, our conclusion is con-
sistence with [44]. DPR treatment could give better results at
deep gel penetration because the deepest gel penetration has the
most efficient filtration process for water in oil-water pathway
since the DPR fluid can reduce the permeability to water to greater
extent than to oil, so the long filter (DPR fluid volume) would give
more screening and blocking for water molecules. Second, if the
cross flow exists; the sweep efficiency would be improved just
on extent of gel penetration depth [27,36]. Also, we found a good
analog from production engineering principles that supports this
conclusion. Basically, we want to create skin for water pathways
in the reservoir region which is closed to the production well.
According to Hawkins formula, the skin is function of permeability
impairment (K/Ks) which has the same meaning to Frrw assuming
Frro equals to one, and also function of damage penetration (rs)
which has the same physical meaning to gel penetration depth.
We well understood as far as the radius of damage zone is longer
the production flowrate is lower for the same other conditions. If
the gel penetration depth increased, it would increase the skin
for water flow. To sum up, these results give a clear indication
and support for that the increasing in gel penetration depth would
increase DPR treatment performance when Frro is very low (close
to the ideal case).
5.3. Frrw/Frro effect
Different values of (Frrw/Frro) for different gels properties were
simulated to investigate the effect of Frrw/Frro (gel type criteria)
on DPR performance. The highest values of (Frrw/Frro) which were
 D.K. Alfarge et al. / Fuel 207 (2017) 226–239 231

Fig. 4. Effect of gel penetration depth on DPR performance for different systems of oil reservoirs.

used in this model could be practical since it has been reported
that new gel formulation gives Frrw greater than 2000 and Frro
equal to 2 or less [29]. We noticed that DPR treatment gives better
results in both water cut reduction and oil recovery improvement
when Frrw is very high and Frro is very low as shown in Fig. 5 and
Fig. 6. Also, our conclusion is consistent with Stavland [40] conclu-
sion regarding the most successful DPR would occur at high Frrw/
Frro. Gels with high (Frrw/Frro) value have the highest change
chance to be the real DPR-chemical agents according to the DPR
definition. When the ratio of Frrw/Frro increased, the chance to
get more connected oil channels would be increased after DPR
treatment performed, so the DPR performance increased. Also, if
the (Frrw/Frro) is high, this would reduce the concerns regarding
the damage which gel might cause in the oil zone because the
residual resistance factor increased with decreasing permeability
since the oil zone usually has the lower permeability
[17,15,42,48,34,62]).
5.4. Production rate effect
Two different production flow rates are simulated to under-
stand the DPR performance at each one of them. Production flow-

Fig. 5. Effect of Frrw/Frro on water cut reduction in radial system with cross flow.
232 D.K. Alfarge et al. / Fuel 207 (2017) 226–239

Fig. 6. Effect of Frrw/Frro on water cut reduction in radial system without cross flow.

rates of 1000 STB/D and 10,000 STB/D were used to investigate
how we can produce after DPR treatment performed. We noticed
that DPR gives better results in both water cut reduction and oil
recovery improvement when production rate is low as shown in
Fig. 7. We noticed, in linear with cross flow case, that water cut
reduced by 40% at production flowrate equals to 1000 STB/D and
oil recovery increased by 5%. On the other hand, the water cut
reduced by 20% and oil recovery increased by 2% at 10,000 STB/
D. There are two reasons explaining this behavior. The first one is
as far as the flowrate increased; the Frrw could be decreased as
reported by many investigators in their lab work
[21,2,7,12,24,40]. The reduction in Frrw with increasing in produc-
tion flowrate is clear in Fig. 8. However, other investigators said
that both of Frrw and Frro decrease with increasing in production
flowrate according to power law model ([61,45,46]. The second
reason is that increasing production flow rate would result in
increasing of viscous forces, so the water crossing to oil zone would
increase at high production rate. If water crossed to the oil zone,
the water would build up in oil zone and creates what is called a
water blocking effect. To sum up, as the production rate after
DPR treatment get lowered, the DPR treatment is more successful.
5.5. Cross flow effect
DPR treatment was applied at different values of cross flow to
investigate the effect of cross flow on DPR performance. The best
translation of cross flow values is by G-shape values [50,39]. The
G-shape includes not only the ratio of vertical permeability to hor-
izontal permeability but also the reservoir aspect ratio which is the
ratio of length of reservoir to the thickness of that reservoir
[5,50,39,47]. Also, we found that if we use the effect of the cross
flow through the ratio of vertical permeability to horizontal per-

Fig. 7. Effect of Production rate on DPR performance (water cut reduction).

 D.K. Alfarge et al. / Fuel 207 (2017) 226–239 233

Fig. 8. How Frrw changes with changing production flowrate, (A) Production flowrate equals to 1000 STB/D, (B) Production Flowrate = 10,000 STB/D.

meability only, the impact of cross flow on DPR performance can
increase or decrease as we increase or decrease the thickness of
reservoir consequently. Therefore, using G-shape factor to quantify
the effect of cross flow on DPR performance is crucial. We noticed
that DPR gives better results in both water cut reduction and oil
recovery improvement when it is applied at very low values of
G-shape, shale barrier conditions, as shown in Fig. 9. From Fig. 9,
we notice that there are two different trends in all curves for plot-
ting of water cut reduction or oil recovery factor improvement ver-
sus G-shape. We think this happens because the behavior of DPR
performance is different as in before Vertical Equilibrium (VE) con-
ditions as compared with after Vertical Equilibrium (VE) condi-
tions. Vertical Equilibrium (VE) has been discussed extensively
by many investigators [14,43,16,18,66]. DPR treatment at low G-
shape values is more successful because there is no water crossing
to oil zone which would create water blocking effect. Before verti-
cal equilibrium is achieved, the oil zone is less affected by water
zone. However, when vertical equilibrium is achieved, the oil zone
would be more affected by the water zone because the horizontal
pressure drop is the same at any point vertically. Second, during
gel placement process in a reservoir with cross flow, the gelant
would go to oil zone in larger amount as compared with barrier-
existing case [4,38,37]. To sum up, DPR treatment application in
reservoir with small cross flow values or barrier existed conditions
would give better results than if it is applied in a reservoir having
cross flow. This conclusion is also consistent with simulation work
of Gao [13] when he concluded to use polymer flooding in reservoir
with high ratio of (kV/Kh) and use the gel treatment in reservoirs
with low ratio of (kV/Kh).
5.6. Oil gravity effect
For all cases, the DPR treatment was applied at different oil
specific gravity values which were 0.65, 0.75, 0.85, 0.95, and 1. This
work was to understand where DPR can be successfully applied in
heavy oil reservoir or light oil reservoir with the same other condi-
tions. The study indicated that DPR could give better results in both
water cut reduction and oil recovery improvement when it is
applied at reservoir with light oils rather than heavy oils. For
example, in linear system with cross flow case, water cut could
be reduced in different oil-density values by 28%, 25%, 22%, 19%,
and 17.5% respectively. And, the oil recovery factor could be

Fig. 9. Effect of cross flow on DPR performance in radial system.

234 D.K. Alfarge et al. / Fuel 207 (2017) 226–239

Fig. 10. Oil recovery factor versus time; before and after DPR treatment (comparison between DPR performances when the high-k in lower zone versus when the high-k in
upper zone).

increased by 8%, 6%, 5%, 3%, 2% respectively. The other three cases
could have approximately the same trend, but with different val-
ues where we noticed the order from the best model to worse
model as linear system without cross flow, linear system with
cross flow, radial without cross flow, and the radial system with
cross flow respectively. The reasoning beyond applying DPR treat-
ment at reservoir with light oils is more successful rather than in
reservoir with heavy oils is due to gravity segregation help. When
the production process resumed after DPR treatment, the gravity
forces would help in preventing water from crossing to oil zone
and creating water blocking effect for limited viscous forces in light
oil reservoirs. While in the heavy oil, the density difference
between water and oil is not almost exist, so the water blocking
effect is significant. Also, the dragging force for heavy oil is higher
than light oil which might downgrade the gel resistance more, so
DPR performance would be downgraded in heavy oil reservoirs.
We also investigated another scenario in which the high perme-
ability layer located in the top layer. The results indicated that the
DPR performance would be much pronounced in the reservoirs
which have high streak in the bottom of the reservoir instead of
the top. To investigate that, at water cut of 80% in both models,
the DPR treatment was performed with the same parameters in
both models. We noticed the oil recovery factor for the reservoir
of high-k in the upper zone was improved by 0.5% while the oil
recovery factor in the reservoir with high-k in the lower zone
was improved by 5% as shown in Fig. 10. And, the water cut for
the reservoir of high-k in the upper zone was reduced by 4% while
the water cut in the reservoir of high-k in the lower zone was

Fig. 11. Effect of initial water cut on DPR performance (water cut reduction) in different systems.
 D.K. Alfarge et al. / Fuel 207 (2017) 226–239
reduced by 20%. The reasoning beyond this difference in DPR per-
formances is under two reasons. The first reason is that, in the
reservoir with high-k in the lower zone, the gravity helps the gel
to be segregated to water zone and block water movement because
both of gel and water have approximately the same density. There-
fore; the water cut reduction in the reservoir with high-k in lower
zone was better than the reservoir with high-k in upper zone. The
second reason is that, in the reservoir with high-k in lower zone,
there is gravity force which reduces the dominant viscous forces
resulting in less water crossing to the oil zone. On other hand,
the gravity force is not high enough, in the reservoir with high-k
in upper zone, to reduce water crossing to the oil zone, but in this
case, the gravity forces and viscous forces are in the same direction
which is downward to the oil zone. The late case makes water
crossing to oil zone in high rate causing significant water build
up effect.
To sum up, the chance of DPR success in light oils is better as
compared to the heavy oils. However, this conclusion is neither
consistent with Prado Paez et al. [59] lab work nor [21] lab work
because we think that the gravity effect is not significant in lab
work as compared on the reservoir scale.
5.7. Time of DPR treatment
For all cases, the DPR treatment applied at different water cut
values which are 50%, 60%, 70%, 80%, 90%, and 95%. This work is
to see when DPR can be successfully applied and how pre-
Table 5
Parameters with their range which were used in CMOST.
Parameters
Reservoir Thickness (ft)
Ratio (K zone1/Kzone2)
Ratio of (Vertical Permeability/Horizontal permeability)
Oil viscosity cP
Oil Density (lb/Ft^3)
Gel Volume (bbl)
Frrw/Frro
Fig. 12. Sobol approach for factors impacting water cut% on first 3 months after DPR treatment.
235
treatment water cut affects DPR performance. We concluded that
DPR gives better results in both water cut reduction and oil recov-
ery improvement when it is applied at lower water cut, which
means, as soon as DPR treatment performed is the better perfor-
mance would happen as shown in Fig. 11. For example, in linear
with cross flow case, the water cut could be reduced by 35%,
31%, 28%, 25%, 20%, 10% respectively. The oil recovery could be
increased by 7%, 5%, 3%, 2.5%, 2%, 1.5% respectively. The other three
cases could have the same trend, but with different values. We
noticed the arrangement from the best case to worse case as linear
without cross flow, linear with cross flow, radial without cross
flow, and the radial system with cross flow respectively. The rea-
soning for applying DPR treatment at low water cut value is more
successful than at high water cut is that the chance to get oil path-
ways are not continuous and not connected is so significant at
higher water cut. Since DPR treatment needs more oil channels
connected after treatment, the treatment would not be pro-
nounced in high water cut because the oil molecules would be
encapsulated by water molecules. To sum up, starting DPR treat-
ment earlier would give better results rather than starting it later.
6. Design of experiments for eight factors affecting DPR
performance using CMOST 2015
6.1. Sensitivity analysis
The purpose of Sensitivity Analysis is for determining how sen-
sitive an objective function to different parameters qualitatively
and quantitively. Identifying the parameters which have a high
impact on DPR performance would give a good prediction for
Range DPR success or failure depending on reservoir properties prior to
10–100 DPR field application. In this part, the objective functions which
1–10000 were used are water cut, cumulative oil production, and oil recov-
0–1 ery factor at three different time periods of DPR post-treatment
0.75–60 which were 3 months, 6 months, and one year. The parameters
30–62
500–5000
which were investigated and their range values are listed in Table 5.
1–173 The statistical methods which were used for parameters ranking
are as follow:
236 D.K. Alfarge et al. / Fuel 207 (2017) 226–239

Fig. 13. Morris method for each parameter on water cut after 6 months from DPR.

Fig. 14. Tornado plot, sensitivity plot, explains the qualitative and quantitative effect for the parameters investigated on cumulative oil (bbl) in the first 6 months from DPR
treatment. The positive sign means that parameter has a positive effect on the objective function. However, the negative sign means that parameter has a negative effect on
the objective function. The number for each parameter measures the quantitative effect for that parameter on the objective function.
 D.K. Alfarge et al. / Fuel 207 (2017) 226–239
 Sobol Method: The Sobol method is one of the variance-based
sensitivity analysis methods to quantify the amount of variance
that each input factor Xi contributes to the unconditional vari-
ance of output V(Y) (CMG). For example, a given case with 3
inputs and one output, if 50% of the output change would hap-
pen by changing the first input, 30% by changing the second
input, 10% by changing the third one, and 10% due to interac-
tions between the first two input parameters, these percentages
are clearly reflected as measures of sensitivity. For more infor-
mation about the basics and principles of this method, the ref-
erence, Sobol [64] can be reviewed.
 Morris Method: The Morris method (also named the Elemen-
tary Effects (EE) method) is one of the screening methods which
is used to specify the effect of input parameters on model out-
puts (CMG). Morris approach has two measures, mean and stan-
dard deviation, which are used together. Mean provides linear
influence of the input factor on the output while Standard devi-
ation reflects the nonlinear or interactions. For more informa-
tion about the basics and principles of this method, the
reference, Morris [23] can be reviewed.
 Tornado Plot: a visual tool provides a qualitative and quantitve
effect for input Parameters on output ones, with higher values
meaning more sensitive to that parameter change rather than
parameters with a low value (CMG). For more information
about the basics and principles of this method, CMG reference
number can be reviewed.
6.2. High impact parameters in the first 3 months of DPR post-
treatment
 Oil Viscosity: All of Sobol approach, Morris method, and Tor-
nado plot indicated that the most important factor which
affects water cut is the oil viscosity and the permeability
heterogeneity (linear flow or radial flow). We noticed that as
oil viscosity increases, water cut would increase, which means
that oil viscosity has a negative effect on DPR performance as
shown in Fig. 12. The interpretation which we think behind
this behavior is that increasing oil viscosity would increase
water blocking effect due to the fingering problems according
to fractional flow equations, so DPR performance would down-
grade. Also, oil viscosity has negative effect on both of cumu-
lative oil production and oil recovery factor but it has the
second rank. However, the permeability heterogeneity has a
positive effect on DPR performance. As Permeability hetero-
geneity increased (linear flow more likely), DPR performance
increased.
 Other Parameters (Frrw/Frro, Reservoir Thickness, Vertical
Permeability): Frrw/Frro, reservoir thickness, and vertical per-
meability were the third, fourth, and fifth rank-parameter con-
sequently. Frrw/Frro and reservoir thickness showed a positive
effect on DPR performance. Increasing Frrw/Frro would increase
DPR property, which in turn leads to increase DPR performance.
Increasing the thickness of reservoir has a positive effect on DPR
performances because increasing thickness of reservoir would
make the gravity force overcomes the viscous force (more con-
fining process), so the water blocking effect would reduce. How-
ever, vertical Permeability had a negative impact on DPR
performance because water blocking would be increased as ver-
tical permeability increased.
6.3. High impact parameters in the first 6 months and one year of DPR
Post-treatment
The ranking of the previous parameters approximately has not
been changed after 6 months and one year as in after 3 months
of DPR treatment as shown in Fig. 13 and Fig. 14. However, the
237
interaction effect among parameters could significantly increase
after 6 months as compared with after 3 months of the treatment.
We think the reason for this increasing in interaction effect is that
as far as the production process progresses, the depletion in reser-
voir increases, so this drainage would increase dynamic process
between fluids and reservoir characterizations which creates wide
interaction effect.
7. Conclusions
 For the cases which have been studied in this paper, DPR could
be applied successfully in thick, low viscous oil, light oil, and
very heterogeneous reservoirs rather than thin reservoirs, high
viscous, heavy oil, and homogenous reservoirs with the same
other conditions according to reasons and details in the discus-
sion part of this study.
 Design of experiments methods explained that oil viscosity and
reservoir permeability heterogeneity (linear flow or radial flow)
are the most important factors to predict DPR treatment
Success.
 DPR treatment is less pronounced in reservoirs with cross flow
conditions as compared without-cross flow reservoir due to
water blocking effect.
 DPR performance in the reservoirs which are not in Vertical
Equilibrium (VE) conditions is better than in the reservoirs with
VE conditions.
 The production flowrate after DPR treatment has an important
impact on DPR success because increasing production flowrate
would make the viscous forces overcomes gravity forces which
could create a significant water block problems.
Acknowledgment
We greatly acknowledge Higher Committee for Education
Development (HCED) in Iraq for their financial support.
References
[1] Al-Sharji HH, Grattoni CA, Dawe RA, Zimmerman RW. Pore-Scale Study of the
Flow of Oil and Water through Polymer Gels. Paper SPE 56738 presented at the
SPE Annual Technical Conference and Exhibition, Houston, 3–6 October; 1999.
DOI: 10.2118/56738-MS.
[2] Bryant SL, Rabaioli MR, Lockhart TP. Influences of synersis on permeability
reduction by polymer gels. SPE Prod Fac 1996;11(4):209–15. http://dx.doi.org/
10.2118/35446-PA. SPE-35446-PA.
[4] Craig Jr FF. The Reservoir Engineering Aspects of Waterflooding. Dallas: Society
of Petroleum Engineers; 1971. p. 62–76.
[5] Dake LP. Fundamentals of Reservoir Engineering. New York: Elsevier Scientific
Publishing Co.; 1978. p. 349–428.
[7] Di Lullo G, Rae P, Curtis J. New Insights into Water Control- A Review of the
State of the Art- Part-II. Paper SPE Presented at the SPE International Thermal
Operations and Heavy Oil Symposium and International Horizontal Wells
Technology Conference, Calgary, 4–7 November; 2002. DOI: 10.2118/79012-
MS.
[9] Eoff L, Dalrymple D, Reddy BR, Everett DM. Structure and process optimization
for the use of a polymer relative permeability modifier in conformance control.
SPEJ 2003;8(3):92–8. http://dx.doi.org/10.2118/84951-PA. SPE-84951-PA.
[10] Eoff L, Dalrymple D, Reddy BR, Morgan J, Frampton H. Development of
Hydrophobically Modified Water-Soluble Polymer as a Selective Bullhead
System for Water-Production Problems. Paper SPE 80206 presented at the
SPE International Symposium on Oilfield Chemistry, Houston, 5–7 February;
2003. DOI: 10.2118/80206-MS.
[11] Faber MJ, Joosten GJP, Hashmi KA, Gruenenfelder M. Water Shut-off Field
Experience with a Relative Permeability Modification System in The Marmul
Field (Oman). Paper SPE 39633 presented at the 1998 SPE/DOE Improved Oil
Recovery Symposium, Oklahoma, 19–22 April; 1998. DOI: 10.2118/39633-
MS.
[12] Ganguly S, Willhite GP, Green DW, McCool CS. Effect of Flowrate on
Disproportionate Permeability Reduction. Paper SPE 80205 Presented at the
International Symposium on Oilfield Chemistry, Houston, 5–7 February;
2003. DOI: 10.2118/80205-MS.
238 D.K. Alfarge et al. / Fuel 207 (2017) 226–239
[13] Gao HW et al. Permeability modification simulator studies of polymer-gel-
treatment initiation time and crossflow effects on waterflood oil recovery.
SPERE 1993:221–7.
[14] Hiatt WN. Injected-f1uid coverage of multi-well reservoirs with permeability
stratification. Drill Prod Prac API 1958:165–94.
[15] Hirasaki GJ, Pope GA. Analysis of factors influencing mobility and adsorption in
the flow of polymer solution through porous media. SPEJ 1974;14(4):337–46.
http://dx.doi.org/10.2118/4026-PA. SPE-4026-PA.
[16] Jacks HH, Smith OJE, Mat1ax CC. The modeling of a three-dimensional
reservoir with a two-dimensional simulator with the use of dynamic pseudo
functions. Soc Pet Eng J 1973;13:175.
[17] Jennings RR, Rogers JH, West TJ. Factors influencing mobility control by
polymer solutions. JPT 1971;23(3):391–401. http://dx.doi.org/10.2118/2867-
PA. Trans., AIME, 251. SPE-2867-PA.
[18] Lake Larry W, Hirasaki GJ. Taylor’s Dispersion in Stratified Porous Media, SPE-
8436, Presented at 1979 SPE Fall Meeting, Las Vegas, Nevada.
[19] Liang J, Seright RS. Further Investigations of Why Gels Reduce kw More Than
ko. Paper SPE 37249 presented at the 1997 SPE International Symposium on
Oilfield Chemistry, Houston, Texas, 18–21 February; 1997.
[20] Liang J, Sun H, Seright RS. Reduction of Oil and Water Permeabilities Using
Gels. Paper SPE 24195 presented at SPE/DOE Symposium on Enhanced Oil
Recovery, Tulsa, 22–24 April; 1992.
[21] Liang J-T, Sun H, Seright RS. Why do gels reduce water permeability more than
oil permeability? SPERE 1995;10(4):282–6. http://dx.doi.org/10.2118/27829-
PA. Trans., AIME, 299. SPE-27829-PA.
[23] Morris MD. Factorial sampling plans for preliminary computational
experiments. Technometrics (American Statistical Association) 1991;33(2).
[24] Nguyen TQ, Green DW, Willhite GP, McCool CS. Effect of gelant composition
and pressure gradients of water and oil on disproportionate permeability
reduction of sandpacks treated with polycrylamide-chromium acetate
gels. SPE J 2006;11(2):145–57. http://dx.doi.org/10.2118/89404-PA. SPE-
89404-PA.
[25] Nilsson S, Stavland A, Jonsbraten HC. Mechanistic Study of Disproportionate
Permeability Reduction. Paper SPE 39635 presented at the 1998 SPE/DOE
Improved Oil Recovery Symposium, Oklahoma, 19–22 April; 1998. DOI:
10.2118/39635-MS.
[27] Root PJ, Skiba FF. Crossflow effects during an idealized displacement process in
a stratified reservoir. SPEJ 1965:229–37.
[28] Seright RS. Reduction of Gas and Water Permeabilities Using Gels. 1995.
[29] Seright RS. Disproportionate permeability reduction with pore-filling gels.
SPEJ 2009;14(1):5–13. http://dx.doi.org/10.2118/99443-PA. SPE-99443-PA.
[30] Seright RS, Lane RH, Sydansk RD. A strategy for attacking excess water
production. SPEPF 2003;18(3):158–69. http://dx.doi.org/10.2118/84966-PA.
SPE-84966-PA.
[31] Seright RS, Liang J, Lindquist WB, Dunsmuir JH. characterizing
disproportionate permeability reduction using synchrotron X-ray computed
microtomography. SPEREE 2002;5(5):355–64. http://dx.doi.org/10.2118/
79717-PA. SPE-79717-PA.
[34] Seright RS, Liang J, Sun H. Gel treatments in production wells with water
coning problems. In Situ 1993;17(3):243–72.
[36] Sorbie KS, Seright RS. Gel Placement in Heterogeneous Systems with
Crossflow, paper SPE 24192 presented at the 1992 SPE/DOE Symposium on
Enhanced Oil Recovery, Tulsa, April 22–24.
[37] Sorbie KS. et al. ‘‘Scaled Miscible Floods in Layred Beadpacks Investigating
Viscous Crossflow, the Effects of Gravity and the Dynamics of Viscous Slug
Breakdown”, paper SPE 20520 presented at the 1990 SPE Annual Fall
Conference, New Orleans, LA, Sept. 23-26.
[38] Sorbie KS. et al. ‘‘Miscible Displacements in Heterogeneous Core Systems:
Tomographic Confirmation of Flow Mechanisms,” paper SPE 18493 presented
at the 1989 SPE International Symposium on Oilfield Chemistry, Houston, Feb.
8–10.
[39] Stavland A, Nilsson S. Segregated Flow is the Governing Mechanism of
Disproportionate Permeability Reduction in Water and Gas Shutoff. Paper SPE
71510 presented at the SPE Annual Technical Conference and Exhibition,
Louisiana, 30 September–3 October; 1998. DOI: 10.2118/71510-MS.
[40] Stavland A. How to apply the flow velocity as a design criterion in RPM
treatments. SPEJ 2010;2(25):223–31. http://dx.doi.org/10.2118/121850-PA.
SPE-121850-PA.
[41] Sydansk RD et al. When and where relative permeability modification water-
shutoff treatments can be successfully applied. SPE Prod Facil 2007;22
(2):236–47. http://dx.doi.org/10.2118/99371-PA.
[42] Vela S, Peaceman DW, Sandvik EI. Evaluation of polymer flooding in a layered
reservoir with crossflow, retention, and degradation. SPEJ 1976;16(2):82–96.
http://dx.doi.org/10.2118/5102-PA. SPE-5102-PA.
[43] Warren JE, Cosgrove JJ. Prediction of waterflood performance in a stratified
system. Soc Pet Eng J 1964:149–57.
[44] White JL, Goddard JE, Phillips HM. Use of polymers to control water production
in oil wells. SPEJ 1973;25(2):143–50. http://dx.doi.org/10.2118/3672-PA. SPE-
3672-PA.
[45] Willhite GP. Mechanisms causing disproportionate permeability in porous
media treated with chromium acetate/HPAAM gels. SPEJ 2002;7(1):100–8.
http://dx.doi.org/10.2118/77185-PA. SPE-77185-PA.
[46] Yan Z, McCool CS, Green DW, Willhite GP. ‘‘Modification of oil and water
permeabilities in Berea sandstone by a gel treatment”, SPE 50753 presented at
the 1999 SPE International Symposium on Oilfield Chemistry held in Houston,
Texas, 16–19 February 1999.
[47] Yortsos YC. ‘‘A Theoretical Analysis of Vertical Flow Equilibrium”, paper SPE
22612, presented at the 1991 SPE Annual Fall Conference, Dallas, TX, Oct. 6-9.
[48] Zaitoun A, Kohler N. Two-Phase Flow through Porous Media: Effect of an
Adsorbed Polymer Layer. Paper SPE 18085 presented at SPE Annual Technical
Conference and Exhibition, Houston, 2–5 October; 1988.
[49] Zaitoun A, Kohler N. Thin Polyacrylamide Gels for Water Control in High-
Permeability Production Wells. Paper SPE 22785 presented at SPE Annual
Technical Conference and Exhibition, Dallas, 6–9 October; 1991.
[50] Zapata VJ, Lake LW. ‘‘A Theoretical Analysis of Viscous Crossflow,” paper SPE
101 11 presented at the 1981 SPE Annual Technical Conference and Exhibition,
San Antonio, Oct. 5–7.
[51] Zitha PLJ, Vermolen FJ, Bruining H. Modification of Two Phase Flow Properties
by Adsorbed Polymers or Gels. Paper SPE54737 presented at the SPE European
Formation Damage Conference, The Hague, The Netherlands, 31 May 1 June;
1999. DOI: 10.2118/54737-MS.
[52] Al-Sharji HH, Grattoni CA, Dawe RA, Zimmerman RW. Disproportionate
permeability reduction due to polymer adsorption entanglement. Paper SPE
68972 SPE european formation damage conference, 21–22 May, The Hague,
Netherlands; 2001. DOI:10.2118/68972-MS.
[53] Computer Modeling Group, STARS Manual (2010 & 2015 versions); https://
www.CMG.Ca/ [accessed 2015].
[54] Crespo F, Reddy BR, Eoff L, Lewis C, Pascarella N. Development of a polymer gel
system for improved sweep efficiency and injection profile modification of
IOR/EOR treatments. International petroleum technology conference; 2014.
Doi:10.2523/IPTC-17226-MS.
[55] Han M, Alshehri AJ, Krinis D, Lyngra S. State-of-the-art of in-depth fluid
diversion technology: enhancing reservoir oil recovery by gel treatments.
Society of petroleum engineers; 2014. Doi:10.2118/172186-MS.
[56] Hill F, Monroe S, Mohanan R. Water management—an increasing trend in the
oil and gas industry. Presented at the SPE/EAGE european unconventional
resources conference and exhibition, Vienna, 20–22 March; 2012. SPE-
154720-MS. doi: 10.2118/154720-MS.
[57] Krishnan P, Asghari K, Willhite GP, McCool CS, Green DW, Vossoughi S.
Dehydration and permeability of gels used in in-situ permeability
modification treatments. Society of petroleum engineers; 2000. Doi:10.2118/
59347-MS.
[58] Liang J, Seright RS. Wall-effect/gel-droplet model of disproportionate
permeability reduction. Society of petroleum engineers; 2000. Doi:10.2118/
59344-MS.
[59] Prado Paez M, Rauseo O, Reyna M, Ferreira I. Evaluation of the effect of oil
viscosity on the disproportionate permeability reduction of a polymeric gel
used for controlling excess water production. Society of petroleum engineers;
2009. Doi:10.2118/121499-MS.
[60] Sandiford BB. Laboratory and field studies of water floods using polymer
solutions to increase oil recovery. JPT(August): 917–922; Trans., AIME, 231.
SPE-844-PA; 1964. DOI: 10.2118/844-PA.
[61] Seright RS. Polymer gel dehydration during extrusion through fractures.
Society of petroleum engineers; 1999. Doi:10.2118/56126-PA.
[62] Seright RS. Impact of permeability and lithology on gel performance. Paper SPE
24190 presented at the SPE/DOE enhanced oil recovery symposium, Tulsa, 22–
24, April; 1992. Doi: 10.2118/24190-PA.
[63] Seright RS. Optimizing disproportionate permeability reduction. Paper SPE
99443 SPE/DOE symposium on improved oil recovery, Tulsa, 22–26 April;
2006. Doi: 10.2118/99443-MS.
[64] Sobol I. Sensitivity Estimates for Nonlinear Mathematical Models.
Matematicheskoe Modelirovanie 1993;2:112–8.
[65] Weaver JD. A new water-oil ratio improvement material. Paper SPE 7574
presented at the SPE annual fall technical conference and exhibition, Dallas, 1–
3 October; 1978. Doi: 10.2118/7574-MS.
[66] Yokoyama Y, Lake LW. The Effects of Capillary Pressure on Immiscible
Displacements in Stratified Porous Media. Society of Petroleum Engineers
1981. http://dx.doi.org/10.2118/10109-MS.
Dheiaa Alfarge, SPE, is a PhD Candidate at Missouri University of Science and
Technology, specializing in petroleum engineering. Previously, he was a drilling
Engineer in Maysan Oil Company. He has about 2 years of experience working in oil
fields. His research interests include enhanced oil recovery (EOR) and unconven-
tional reservoirs. He holds a Bachelor degree in Petroleum engineering from
University of Bagdad and Master degree in petroleum engineering from Missouri
University of Science and Technology.
Mingzhen Wei, She is an assistant professor of petroleum engineering at Missouri
University of Science and Technology, where she has been since 2012. Her research
interests include unconventional oil and gas resources, EOR, reservoir simulation,
and data analytics and data mining applied in the oil and gas industry. Wei has
authored or coauthored more than 50 technical papers. She holds a PhD degree in
computer science from New Mexico Institute of Mining and Technology; an MS
degree in reservoir engineering from China University of Petroleum, Beijing; and a
BS degree in petroleum engineering from China University of Petroleum, Huadong.
Wei is a member of SPE.
Baojun Bai, He is the Lester R. Birbeck Endowed Chair professor of petroleum
engineering at Missouri University of Science and Technology. Previously, he was a
reservoir engineer and head of the conformance-control team at the Research
 D.K. Alfarge et al. / Fuel 207 (2017) 226–239 239

Institute of Petroleum Exploration and Development, Petro China. Bai was also a
post-doctoral scholar at the California Institute of Technology and a graduate
research assistant at the New Mexico Petroleum Recovery Research Center for EOR
projects. He has more than 20 years of experience in the area of EOR, especially in
gel treatments. Bai has published more than 130 papers in the area of EOR methods
and application. He holds a PhD degree in petroleum engineering from New Mexico
Institute of Mining and Technology and a PhD degree in Petroleum geology from
China University of Geoscience, Beijing. Bai served on the editorial committee of the
Journal of Petroleum Technology for the feature of EOR Performance and Modeling
during 2007–13. He is a technical editor for SPE Journal and SPE Reservoir Evalu-
ation & Engineering. Bai is a member of SPE.