SPE 114336

New Developments in Production Technology for Polymer Flooding

Wang Yan, SPE, Wang Demin SPE, Wan Jun, Luo Jiangtao, Yu Runtao, and Dong Zengyou, Daqing Oil Company
Limited, Daqing Petroleum Institute
Copyright 2008, Society of Petroleum Engineers

This paper was prepared for presentation at the 2008 SPE/DOE Improved Oil Recovery Symposium held in Tulsa, Oklahoma, U.S.A., 1923April2008.

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
After large scale polymer flooding in the field, due to the high elasticity and viscosity of the fluid, the efficiency and
effective life-span of the producers (and injectors) decreased significantly, new technologies were developed.
After polymer is produced, severe eccentric wear on sucker rods occur. To solve the problem, on the basis of theoretical and
lab studies, techniques to reduce the normal force on sucker rods and control its eccentric wear were developed. Field practice
shows that this technology has increased the average life span of pumping wells by 3 times.
The effective time of conventional fracturing and acid stimulation of polymer injectors is less than 3 months. Therefore,
technologies of resin-coated sand tail-in fracturing and surfactant stimulation were developed. These 2 new technologies have
increased the effective time of injector stimulation by about 5 times.
The above techniques are all used in a mass scale in the field with significant results; they can and are also used on
conventional wells.

Introduction
Commercial polymer flooding has commenced in Daqing Oil Field since 1996, resulting in oil recovery over 50%OOIP. At
present, the annual oil output by polymer flooding has reached 1.210
8
t, accounting for a quarter of the total oil output of the
oilfield. With the wide application of polymer flood, there are also many problems that adversely impact the injection and
production of polymer flooding. These problems are as following:
1. Severe eccentric wear on sucker rod and tubing has occurred on a large number of pumping wells and the increase in
wells that have wear problems has reached 3000%, which has increased workover frequencies and production cost as well.
The problem of eccentric wear has greatly affected oil production.
2. On approximate one third of the polymer injection wells, injection pressure is close to the formation fracture pressure, the
injection rate is low also. The effective time by conventional fracturing treatment can only last about three months.
3. Acid fluids react with the injected polymer solution. Therefore, conventional acidizing cannot be used on polymer
injection wells.
Some novel techniques, including eccentric wear control, resin sand fracturing and surfactant stimulation were developed to
solve the above problems.

Technology for Controlling Eccentric Wear of Pumping Wells
Mechanism of eccentric wear of pumping rod. The constitutive equation describing viscoelastic fluids is set up and the
flowing equation under the condition of discrete state can be calculated, then velocity distribution of flow field can be
obtained. The distribution of velocity contour line and normal stress is shown in Fig1. As seen in Fig.1, when the distance
between the inner axis and outer tubing is small, its corresponding velocity gradient will be higher than that of larger distance.
And the resultant force of normal stress points to the side of which the distance between the inner axis and outer tubing is
smaller.

2 SPE 114336

a b


a. Velocity Contour Line
b. Normal Stress Distribution on Sucker Rod Surface

Fig. 1--- V/V max Contour Line of Static Sucker Rod and Normal Stress (Srr) Distribution

When the sucker rod is placed in the middle of the tubing, the fluid field is symmetrical, implying the values of normal
stress are equal in every direction, so the resultant force is zero at the moment. However, it is very hard to ensure that the
sucker rod is always concentric with the tubing in the oil well; therefore the velocity field must be asymmetrically distributed.
Normal stress is also asymmetric; the resultant force cannot be zero. Therefore, the higher the velocity gradient, the greater the
normal force.
The relationship of normal stress, the eccentricity and motion velocity of sucker rod is shown in Fig. 2 and Fig. 3.



D=0.06m, d=0.02m, 1=0.019s, =0.02Pas, w0=2m/s

Fig.2--- Relationship between Normal Stress and Eccentricity

The more severe the eccentricity, the greater the asymmetry of the flow field and the larger the normal force acting on
sucker rod. The opposite situation is true also. Therefore, once the suck rod begins to be slightly eccentric, the eccentricity will
be increased due to the action of the normal force. Meanwhile, with the eccentricity increased, the normal force will be further
simultaneously enlarged, resulting in dramatic rise of eccentricity. As a result, the sucker rod will continuously touch the
tubing and causing the problem of eccentric wear.
4


3


2


1
0.4 0.5 0.6 0.7 0.8 0.9
Eccentricity
Normal
Stress
(N/m)
SPE 114336 3

Motion
Velocity of sucker rod m/s
D=0.06m, d=0.02m, 1=0.019s, =0.02Pas, =0.8

Fig. 3--- Relationship between Normal Force and Motion Velocity of Rod

The positive values in the abscissa represent the process of rod downstroke, while the negative ones describe upstroke
process in Fig.3. The higher the motion velocity, the greater the normal force. The motion velocity for sucker rod is zero when
the rod is situated at the center, where the normal force differences along the sides of sucker rod are so small that the rod
comes back to its initial state from eccentric wearing conditions. During the process of one stroke, due to the normal force, the
rod will move 4 times in a horizontal direction. For a well with a pumping speed of 9 times/m, the horizontal movement of a
sucker rod within the tubing will be 1810
6
times every year, which leads to eccentric wear and causes the probability of snap
failure of sucker rod to be drastically increased.
From the above analysis, it can be known that, for polymer flood wells, the normal force acting on sucker rods is caused by
the elasticity of production fluids, and the magnitude of elasticity will directly affect the value of normal force. In the field
centralizer installation and other measures have been taken, which can not only prevent the rod and the tubing from rubbing
each other, but also can decrease the eccentricity of the sucker rod to reduce the effect of normal force. Meanwhile, the
centralizer is helpful to the elimination of rod movement in a horizontal direction, 1810
6
times every year, which results in
snap failure of the sucker rod to be greatly decreased.

Lab studies and measures to prevent eccentric wear. The normal force acting on the sucker rod will increase when the
velocity of the fluid, polymer concentration and eccentricity between rod and tubing increase. The experiment results are
shown in Fig. 4.

The minimum uniform load, resulting in eccentric wear of the rod string with different lengths, can be calculatedSeen in
Table 1.




0
0. 1
0. 2
0. 3
0. 4
0. 5
0. 6
0. 7
0. 8
0 0. 2 0. 4 0. 6 0. 8 1 1. 2 1. 4
Flow Velocity m/s)
N
o
r
m
a
l

F
o
r
c
e



N
/
m
)

100ppm
200ppm
400ppm
600ppm
800ppm
1000ppm
Fig.4--- Normal Force Acting on Sucker Rod (Eccentricity 0.3)
Normal Force
N
4 SPE 114336
Table 1- Minimum Uniform Load for Eccentric Wear Vs the Length of Rod String
Length of rod string m

Minimum values of uniform
load for eccentric wearN/m

Length of rod string
m
Minimum values of uniform load
for eccentric wearN/m
8 1.37 328400 4.73 x10
-7
2.14 x10
-7

16 8.36 x10
-2
408480 1.98 x10
-7
1.03 x10
-7

2480 1.65x10
-2
1.37x10
-4
488560 9.66 x10
-8
5.67 x10
-8

88160 9.13 x10
-5
8.36 x10
-6
568640 5.26 x10
-8
3.26 x10
-8

168240 6.87 x10
-6
1.65 x10
-6
648720 3.11 x10
-8
2.04 x10
-8

248320 1.45 x10
-6
5.22 x10
-7
728800 1.95 x10
-8
1.33 x10
-8


With the increase in length of the rod string, the minimum uniform load to cause eccentric wear will be decreased. Although
a normal force of 0.1N/m is not powerful enough to cause a sucker rod of 8m to have eccentric wear, it is sufficient for a rod
of 16m to have eccentric wear. For sucker rods with a length over 488m, there will be in eccentric wear when the normal force
is 10
-8
N/m, which is 10
-7
times less than real measured value. So it is necessary to centralize the sucker rod to overcome the
influence of normal force acting on the sucker rod.
Based on the above studies, some matching measures to prevent the occurrence of eccentric wear were taken, including
centralizing sucker rod, using rods with uniform diameter and pumps with large channels, and also regularly rotating rod
string.
Field practice proves that these measures are very effective to prevent eccentric wear and easy to operate in the field,
beneficial to popularization and application of this technique.

Field application results. 8341 wells have taken the above measures to control eccentric wear in Daqing Oil Field by the end
of 2007. After running for 1365days, only 7.39% of the wells suffered the problem of eccentric wear. Compared with the wells
without taking these measures (34.22%), the decrease of eccentric wear is 78.4% (Seen in Fig.5).




















The matching techniques for controlling eccentric wear, not only can inhibit the occurrence of eccentric wear, but also can
increase the rigidity of rod string and decrease the horizontal bending influence of the normal force, acting on the sucker rod
1810
6
times per year, which can effectively reduce the failure rate of the sucker rod. Therefore for wells with the measures,
its running life is prolonged and the workover rate is decreased; the running life lasted 1365d and the workover ratio was
55.46%. Compared with the wells not taking the above measures (83.01%), the declining of the workover rate was 33.2% and
the average running time improved from 325d to 1035d (Seen in Fig.6).
1. 01
3. 09
5. 29
6. 10
6. 74
7. 39 6. 57
19. 28
24. 40
28. 18
32. 46
34. 22
0
5
10
15
20
25
30
35
40
15 105 195 285 375 465 555 645 735 825 915 1005 1095 1185 1275 1365
Runi ng t i me day
R
a
t
e
%

Fig.5--- Comparison Curves Before and After Measures

Wear Ratio for the Wells with Measures
Wear Ratio for the Wells without Measures
SPE 114336 5
Wellbore
Quartz Sand
Resin Sand
4.20
20.85
30.98
34.96
45.21
50.11
54.62 55.46
28.69
52.99
61.96
65.88
73.58
76.23
81.64
83.01
0
10
20
30
40
50
60
70
80
90
15 135 255 375 495 615 735 855 975 1095 1215 1335
Running Time ( day )
W
o
r
k
o
v
e
r
R
a
t
e

(

%
)

Workover Rate for the Wells with Measures
Workover Rate for the Wells without Measures
4.20
20.85
30.98
34.96
45.21
50.11
54.62 55.46
28.69
52.99
61.96
65.88
73.58
76.23
81.64
83.01
0
10
20
30
40
50
60
70
80
90
15 135 255 375 495 615 735 855 975 1095 1215 1335
Running Time ( day )
W
o
r
k
o
v
e
r
R
a
t
e

(

%
)

Workover Rate for the Wells with Measures
Workover Rate for the Wells without Measures
Wellbore
Quartz Sand
Resin Sand
4.20
20.85
30.98
34.96
45.21
50.11
54.62 55.46
28.69
52.99
61.96
65.88
73.58
76.23
81.64
83.01
0
10
20
30
40
50
60
70
80
90
15 135 255 375 495 615 735 855 975 1095 1215 1335
Running Time ( day )
W
o
r
k
o
v
e
r
R
a
t
e

(

%
)

Workover Rate for the Wells with Measures
Workover Rate for the Wells without Measures
4.20
20.85
30.98
34.96
45.21
50.11
54.62 55.46
28.69
52.99
61.96
65.88
73.58
76.23
81.64
83.01
0
10
20
30
40
50
60
70
80
90
15 135 255 375 495 615 735 855 975 1095 1215 1335
Running Time ( day )
W
o
r
k
o
v
e
r
R
a
t
e

(

%
)

Workover Rate for the Wells with Measures
Workover Rate for the Wells without Measures

Fig. 6--- Workover Rate Comparisons Before and After Measure

Hydraulic Fracturing of Polymer Injection Wells
High injection pressure and poor injectivity are very common for polymer injection wells in Daqing Oil Field. The fracture
treatment is a major way to reduce injection pressure and increase injection rate. But the previous fracturing technology can
not satisfy the requirements of polymer injection, it has poor fracturing results, low efficiency and short effective time (less
than 3 months on average). The statistic data, collected from 35 polymer injection wells with conventional fracturing before
2001, showed that the average effective time was 1.5 months and only 12 wells reached 3 months.

The reasons for fracturing failure and experiment results. From analysis and studies, the main reasons for poor fracturing
efficiency and short effective running time are due to high injection pressure and low overburden pressure that acts on the
proppants within the fractures. In addition, polymer solution with higher sand-carrying capacity will push the proppants to
flow into deep formation, which will cause the fractures near the wellbore to close(heal), leading to fracture failure, as
illustrated in Fig.7.

Fig.7--- Sketch Map of Fracture Failure

The amount of quartz sand that migrate, obtained from laboratory studies, shows that the proppants within the fractures will
migrate under different kinds of test conditions, and with the increase in viscosity of the displacing fluid, the amount of the
proppants that migrate will increase (Seen in Fig.8).














Well bore Proppant
Formation
Injection well
Fig.8--- Accumulative Sand Production Rate Vs Flow RateOverburden Pressure 3MPaFracture width 3mm
0
10
20
30
40
50
60
70
80
0 0. 1 0. 2 0. 3 0. 4 0. 5
A
c
c
u
m
u
l
a
t
i
v
e

S
a
n
d

P
r
o
d
u
c
t
i
o
n

R
a
t
e
g

60mPa. s
40mPa. s
30mPa. s

Flow Ratem
3
/h
Polymer Viscosity 60mPa.s
Polymer Viscosity 40mPa.s
Polymer Viscosity 30mPa.s
Fresh Water
6 SPE 114336
Core tests with resin coated sands; walnut shell and carbon fiber have been performed to identify their respective capacity
for preventing the proppants from migration. The experiment results show that under different testing conditions, the resin
sand does not migrate within the fractures; therefore resin sand is the best material to prevent the proppants from moving
(Shown in Fig.9). And since resin sand has high consolidation strength to form an integrated sand pancake around the
wellbore, proppant migration can be effectively controlled.
Fig.10--- Accumulative Sand Production Rate Vs Polymer Viscosity 30mPas
0
5
10
15
20
25
30
35
40
0 0.05 0.1 0.15 0.2 0.25

+
+

Flow Rate
A
c
c
u
m
u
l
a
t
i
v
e

S
a
n
d

P
r
o
d
u
c
t
i
o
n

R
a
t
e

(
g
)
carbon fiber + quartz sand
walnut shell + quartz sand
resin coated sand
Fig.9 --- Accumulative Sand Production Rate Vs Polymer Viscosity 30mPas
0
5
10
15
20
25
30
35
40
0 0.05 0.1 0.15 0.2 0.25

+
+

Flow Rate
A
c
c
u
m
u
l
a
t
i
v
e

S
a
n
d

P
r
o
d
u
c
t
i
o
n

R
a
t
e

(
g
)
carbon fiber + quartz sand
walnut shell + quartz sand
resin coated sand
Fig.10--- Accumulative Sand Production Rate Vs Polymer Viscosity 30mPas
0
5
10
15
20
25
30
35
40
0 0.05 0.1 0.15 0.2 0.25

+
+

0
5
10
15
20
25
30
35
40
0 0.05 0.1 0.15 0.2 0.25

+
+

Flow Rate
A
c
c
u
m
u
l
a
t
i
v
e

S
a
n
d

P
r
o
d
u
c
t
i
o
n

R
a
t
e

(
g
)
carbon fiber + quartz sand
walnut shell + quartz sand
resin coated sand
Fig.9 --- Accumulative Sand Production Rate Vs Polymer Viscosity 30mPas
0
5
10
15
20
25
30
35
40
0 0.05 0.1 0.15 0.2 0.25

+
+

0
5
10
15
20
25
30
35
40
0 0.05 0.1 0.15 0.2 0.25

+
+

Flow Rate
A
c
c
u
m
u
l
a
t
i
v
e

S
a
n
d

P
r
o
d
u
c
t
i
o
n

R
a
t
e

(
g
)
carbon fiber + quartz sand
walnut shell + quartz sand
resin coated sand

The test results indicate that resin sand has a good migration-resistant capacity and its conductivity can meet the
requirements of stimulation. Therefore, the resin sand is selected to be the fracturing proppant.

Determine the radius of resin coated sand tail-in. Resin coated sand can effectively prevent the proppants from migration
and the technology of resin sand tail-in is adopted in the field (Seen in Fig.10, 11).
Wellbore Resin Sand Quartz Sand
Formation
Wellbore Resin Sand Quartz Sand
Formation


Wellbore
Quartz Sand
Resin Sand
Wellbore
Quartz Sand
Resin Sand



Fig.10--- Sketch Map of Resin Sand Tail-in Fracturing
Fig.11---Platform of Resin Sand Tail-in Fracturing
SPE 114336 7
For the proppants to be in a stable state in the entire fracture, it is necessary to determine the required radius of resin coated
sand tail-in, which can be obtained from the quartz sand migration experiment in the fracture. The experiment results can
provide different critical flow rates, matched to various fracture widths and polymer viscosities, at which the quartz sands
begin to migrate.
The experiment results show that in artificial simulated fractures, when the polymer flow rate reaches 0.0025m
3
/h (fluid
velocity in corresponding fractures is 0.0058m/s), the quartz sands begin to move within the fractures. Therefore, it is
recommended that when the fluid velocity within the fractures of polymer injection well is higher than 0.0058m/s, the process
of tail-in resin sand fracturing should be applied.

Field application results. This hydraulic fracturing technique has been used in 382 wells. The initial average injection
pressure was 3.0 MPa lower than before fracturing treatment, and daily injection rate increased 44.6m
3
as well. On average,
accumulative incremental injection rate for individual well has reached 1.9910
4
m
3
and the effective running time has been
prolonged to 26 months, which is 22 months more than that of previous conventional fracturing technology.

Surfactant Injection Stimulations
Since acid fluid will react with polymer solution, the conventional acid treatment can not be used for polymer injection
wells. Therefore, a new stimulation method using non-ionic surfactant was developed to improve relative permeability of the
reservoir.
From the formula kw = Kkrw, it can be seen that the increase of either absolute permeability (K) or relative permeability
(krw) can make effective permeability (kw) improved, which can achieve the targets of reducing injection pressure and
increasing injection rate. For polymer injection wells, conventional acid treatment is to enlarge the absolute permeability,
while surfactants with ultra-low interfacial tension can change properties of the sandstone and can modify interfacial tension
between crude oil and injection solution, which will increase the relative permeability and cause the injection rate to increase
and injection pressure to decrease.
Laboratory studies on surfactant system and test results
Selection of non-ionic surfactant and additives. From lab tests, a type of non-ionic surfactant (FLZB) was developed,
which is mainly composed of alkylolamide and polyoxyethylene polyoxypropyleneamide. The interfacial tension of the
surfactant can be 10
-3
mN/m. The HLB value of the surfactant can be modified by altering the number of carbon atoms in
the chain of alkyl carbon and the structure of the surfactant to meet the needs of different reservoirs and to realize an
ultra-low interfacial tension.

Performance evaluation on the system of FLZB and additive A. The evaluation studies on dilution-resistance, stability and
salinity-resistance of the compound system have been carried out and the study results are seen as Fig. 12, 13 and 14.


























Fig. 13--- Stability of the System Fig. 14--- Salinity-resistance of the System
Fig. 12--- Dilution-resistance of the System
Nacl Concentration (mg/L)
I
n
t
e
r
f
a
c
i
a
l

t
e
n
s
i
o
n

(
1
0
-
3
m
N
/
m
)



I
n
t
e
r
f
a
c
i
a
l

t
e
n
s
i
o
n

(
m
N
/
m
)

8 SPE 114336

From the analysis, it is known that the system consisting of surfactant and additive A has quite good properties in terms of
stability, salinity-resistance and dilution-resistance.
To prove the effect of surfactant system on decrease of injection pressure, the core test has been implemented to check how
the injection pressure change and how to optimize the process of core flooding. The pressure change curves are shown in
Fig.15.












The test results indicate that after chlorine dioxide is used to remove polymer films adsorbed on sandstone porous surface,
the pressure drop, caused by the surfactant system with ultra-low interfacial tension of 10
-3
mN/m, can reach about 60%.

Field application results
The technology of surfactant injection stimulation has been used in 226 wells in the field. The average injection rate per well
increased 24m
3
/d and injection pressure decreased 1.46MPa. The accumulative incremental injection rate averaged at 0.7876
10
4
m
3
for each individual well and the effective running life has reached 18 months. After stimulation, the average apparent
injectivity index of each well increases 55%.


Conclusions
1. From field application results of 8341 polymer production wells, the technology of eccentric wear control has turned out
to be effective in solving the problem of severe eccentric wear of sucker rod and tubing in pumping wells and its economic
benefits are considerable.
2. The technique to prevent fracture closure of polymer injection wells and surfactant injection stimulation are new in the
field of the fracturing and acidizing treatment for polymer injection wells. Actual field practice in 411 wells indicates that
these techniques can result in significant increase in injection rate, which will be major alternative technologies for polymer
injection and have good application future.
3. These techniques can result in improved polymer flooding and further perfect the technology to greatly enhance oil
recovery rate and to provide technical support for the oilfields long term and stable development.

Fig. 15--- Pressure Change Curves
P
r
e
s
s
u
r
e

(
M
P
a
)

Displacemnt Pore Volume Multiple (PV)