SOCIETY OF PET~OLEUM ENGINEE~S OF AIME PAPER

NUMBE:R
s pE 5836

Preinjection of Polymers to Increase Reservoir

Flooding Efficiency

By
Mahmoud K. Dabbous and Lloyd E. Elkins, Members SPE-AIME, Amoco Production
@Copyright 1976
American Institute of Mining, Metallurgical, and Petroleum Engineers, Inc.
THIS PAPER IS SUBJECT TO CORRECTION
This paper was prepared for the Improved Oil Recovery Symposium of the Society
of Petroleum Engineers of AIME, to be held in Tulsa, Okla., March 22-24, 1976.
Permission to copy is restricted to an abstract of not more than 300 words. Illu·
strations may not be copied. The abstract should contain conspicuous acknowledg-
ment of where and by whom the paper is presented. Publication elsewhere after
publication in the JOURNAL OF PETROLEUM TECHNOLOGY or the SOCIETY OF PETROLEUM
ENGINEERS JOURNAL is usually granted upon request to the Editor of the appropri-
ate journal provided agreement to give proper credit is made.
Discussion of this paper is invited. Three copies of any discussion should be
sent to the Society of Petroleum Engineers office. Such discussion may be presen-
ted at the above meeting and with the paper, may be considered for publication in
one of the two SPE magazines.

ABSTRACT in micellar floods as a result of decreased

mobility of fluids was moderate.
Preinjection of high-molecular-weight
polyacrylamide polymers in advance of a mis- Because laboratory models cannot be
cible slug is proposed as a means of improving scaled to reproduce the complex heterogeneities
reservoir sweep efficiency by reducing the encountered in reservoirs, extensive field
degree of "permeability" contrast between testing of polymer preinjection in a real
wells in a moderately heterogeneous system. environment and under carefully controlled
The use of polymers in a preinjected mode is test conditions is suggested. This is highly
radically different from their "conventional" desirable since the timing of development is
use to provide a favorable mobility condition an important factor if the full value of
between in-place and injected fluids and so having available sweep improvement methods is
maximize areal sweep even in a flood of a to be realized.
"homogeneous" sand.
INTRODUCTION
Theory and model flow tests were employed
in an investigation of polymer preinjection in While considerable attention has been
advance of a miscible flood. Flood tests in given to the chemistry of miscible flooding,
physical models of heterogeneous porous media little published work is available which
showed that preinjection of polymers could discusses effects of variations in formation
result in better flooding efficiency because rock texture on reservoir flooding efficiency
of increased volumetric sweep. This was con- by these improved recovery methods. Reservoir
cluded from tracer performance data and oil rock heterogeneity can result in an injected
recovery response. Studies of the interactions fluid bypassing significant portions of a
between the preinjected polymer and a subse- reservoir under flood. Circulation of large
quent micellar flood indicated that prior volumes of water to achieve higher ultimate
polymer flow had no adverse effects on oil dis- sweep in a waterflood is tolerable because of
placement efficiency by a micellar fluid and the low cost of injected water. In miscible
appeared to decrease surfactant loss to the flooding, however, the reservoir is treated
rock. Additional mobility control requirement with only a small size slug, usually less than
10 percent pore volume, of an expensive oil
References and illustrations at end of paper.
t____ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __.___ _ _
506 PREINJECTION OF POLYMERS TO INCREASE RESERVOIR FLOODING EFFICIENCY SPE 5836
recovery fluid such as the micellar fluid. reducing interwell permeability contrast.
Here, more uniform contacting of the reservoir Recently published results of a polymer flood
by the injected miscible fluid is critical, and in the North Burbank sand 5 demonstrate that
even moderate variations in permeability can polyacrylamide polymers can be successfully
cause significant reduction in expected oil used as diverting agents to subsequently
recovery. Mobility control behind a miscible injected fluids, improve interwell sweep in a
slug may have limited effect on causing the heterogeneous reservoir (k = 0.87) and recover
slug to avoid flowing through the high permea- more oil in a mature water¥lood. High-molecu-
bility, watered-out channels in reservoirs, lar-weight polyacrylamides were used to improve
thus bypassing much of the in-place oil that volumetric sweep through permeability "uni-
could be recovered. It is the opinion of the fication" in three Almy sand reservoirs in the
authors that development and implementation of Green River Basin. 6 The reservoirs had permea-
various sweep improvement methods to maximize bility variations ranging from 0.65 to 0.85,
reservoir flooding efficiency is critical at a and results indicated reduced permeability
time when the petroleum industry is launching variation and improvement in volumetric sweep
into an ambitious program of enhanced oil following polymer injection. Injection of
recovery. microemulsions is planned after the conclusion
of the polymer floods.
POLYMER PREINJECTION
There is only one known field test 7 where
Preinjection of polymers in advance of a a polymer slug was injected as a preflush ahead
miscible flood is proposed as a means of of a soluble oil slug (oil-external micellar
improving reservoir sweep efficiency by reduc- fluid) to improve reservoir contacting and
ing the degree of interwell permeability sweep by the miscible slug. Although tertiary
contrast in a moderately heterogeneous system. oil was mobi·lized and recovered by the process,
A reservoir with a Dykstra-Parsons permeability it is not clear from the results of the cited
variation coefficient k between 0.6 and 0.85, test how much of the tertiary oil was due to
with no fractures or un~sual nonuniformities, the polymer preflush or what effect the pre-
is considered moderately heterogeneous. The injected polymer had on reservoir sweep. No
use of polymers in a preinjected mode is radi- computations or performance predictions were
cally different from their "conventional" use· presented to show that preinjection of the
to provide a favorable mobility condition polymer would increase reservoir flooding effi-
between the in-place and injected fluids and so ciency by the miscible flood. Only li~ited
maximize sweep even in a flood of a homogeneous experimental data were published in a U.S.
sand or noncommunicating layered sands. In patent 8 to support claims that preceding a
addition to the potential of improving volu- micellar slug with polymer injection resulted
metric sweep in heterogeneous reservoirs, prior in better oil recovery performance in a labora-
polymer flooding could precondition the reser- tory model of a heterogeneous sand.
voir, chemically, for a subsequent miscible
flood. This is achieved through a more effi- The present study was undertaken to inves-
cient displacement of formation brines than tigate experimentally the effects of polymer
would be obtained with a water preflush. High- preinjection on sweep to water and a subsequent
salinity and multivalent ion formation waters miscible-type waterflood in heterogeneous porous
are known to be incompatible with many of the media. Interactions between the preinjected
present micellar fluid formulations. polymer and the subsequent micellar flood were
also studied and will be detailed in a future
The polymer flood should be carefully paper. Only a brief discussion is provided for
tailored to each specific reservoir application some interactions that are relevant to the
in order to achieve the desired changes in mqterial presented in this paper.
interwell sweep. Polymers particularly suit-
able for this type of application are the high- LABORATORY STUDIES
molecular-weight polyacrylamides which appear
to have a natural selectivity to flow in high- Laboratory studies were made in physical
permeability and high-water-saturation channels, models of heterogeneous porous media in both
thereby reducing the degree of permeability linear and five-spot geometries. Consolidated
contrast within a heterogeneous formation.l-4 and unconsolidated, natural and synthetic
When displaced by water, preferably containing porous media were used in building the flow
a trace concentration of the polymer, poly- models. Capillary contact between sections of
acrylamides leave a residual resistance that different permeabilities was provided in order
causes permanent reduction in water permea- to permit crossflow of fluids. Crossflow is
bility of the rock precontacted with the polymer thought to be an important mechanism control-
ling interwell sweep in many reservoirs that do
The literature reports a number of polymer not have no-flow barriers between zones of
flood projects designed to improve sweep by different permeabilities.
SPE 5836 MAHMOUD K. DABBOUS AND LLOYD E. ELKINS 507
Linear Flow Models saturated with oil and connate water until oil
production had practically ceased, injecting a
Tests were made in two types of linear polymer slug chased by water, and following
models of heterogeneous porous media. The with a micellar flood with Amoco's water-
first type was constructed by drilling out the external micellar system. 10 All fluids used in
center of cylindrical epoxy-wrapped Berea cores the flooding sequence are described in Table 2.
that varied in length from 1 to 4 ft. The Dow Pusher polymers were used in the initial
cores had absolute permeability to water (0.2N polymer floods except for tests conducted in
sodium chloride brine) in the range 400-750 md. the five-spot filtros models. Because of the
The bored-out section was repacked with uncon- high permeability of filtros, a very high-
solidated sands that gave packs with permea- molecular-weight (20 million) experimental
bility to water of 2 to 3.5 darcies. This polyacrylamide polymer was used as the preinjec-
resulted in heterogeneous models having a tion polymer to maintain a more or less similar
radial permeability contrast of 3 to 7 in a relationship between polymer molecular size and
cross section normal to the axial direction of pore size as with Pusher polymers in the-lower
flow. permeability porous media. Residual resistance
factors obtained following the displacement of
The second type of linear heterogeneous the polymer in various filtros grades saturated
model was made with packs of three different with water and residual oil are given in
mixtures of unconsolidated sands. These were Table 1. Comparison of screen factor and
packed in communicating layers in a 1 x 4-3/4 viscosity of a 1,000 ppm solution of the poly-
x 25-3/4 in. Lucite flow model. The three mer and a similar solution of Dow Pusher 700 is
sands, identified as A, B and C, consisted of provided in Table 3.
mixtures of No. 16 Ottawa sand and increasing
percentages of fine silica flour. The wide SWEEP STUDIES IN HETEROGENEOUS POROUS MEDIA
variation in grain size distribution for each
mixture resulted in considerable heterogene- Oil recovery and tracer performance were
ities within a section of each mixture in the primary tools used to evaluate the effect
addition to the macroscopic heterogeneity of polymer preinjection on the sweep in physi-
between gross layers that were packed in cal models of heterogeneous porous media.
capillary contact. Composition of the sand
mixtures and their waterflood characteristics Oil Recovery Performance
are given in Table 1. It is observed that
packs of unconsolidated sands with permeabil- Table 5 summarizes the results of two
ities below 100 md were obtained by increasing flood tests made in two similar five-spot
the percentage of the fines (350 mesh) in the models of biparous filtros. Characteristics of
sand mixture. Listed also in Table 1 is the the two models, fluid, and tracer injection
residual resistance factor (Ry) following the sequence for each test are listed in Table 4.
displacement of solutions of Dow Pusher 700 Results in Table 5 show that the original oil
polymer in packs of the sands containing saturations and waterflood residual oil satura~
waterflood residual oil saturations. tions are quite similar in both tests. Water
breakthrough during waterflooding occurred
Five-Spot Models after injection of 18 percent and 20 percent
pore volumes respectively, indicating almost
Five-spot models of heterogeneous porous identical volumetric sweep efficiency at break-
media were made from biparous filtros slabs through for both heterogeneous models. A
12 x 12 x 1-1/2 in. This synthetic porous micellar flood in the first test reduced the
medium had two layers of different permeabil- average residual oil saturation from 43 percent
ities in capillary contact and parallel to the pore volume at the end of the waterflood to 16
12 x 12 in. base. Heterogeneity of the model percent pore volume at the conclusion of the
was due to a contrast in the range 2 to 5 fold micellar flood, thus recovering approximately
in permeability of the two layers. Flow 63 percent of the in-place oil. A 20 percent
properties and waterflood characteristics of pore volume slug of the high-molecular-weight
various filtros grades used in the models are polyacrylamide polymer was injected and dis-
given in Table 1. The biparous slabs were placed by water in the second model which
encapsuled in epoxy, and injection and pro- contained an average waterflood residual oil
duction wells were drilled at the corners and saturation of 46 percent pore volume. The
in the center. The models were flooded from polymer slug mobilized and recovered tertiary
corner to corner, thus simulating flow in an oil, bringing the average residual oil satura-
isolated quadrant of a five-spot pattern. tion in the heterogeneous model down to 34
percent pore volume. A micellar flood similar
Procedure to that of the first test was then conducted in
the model. In this case, an average residual
Procedure common to most tests was water- oil saturation of only 8.7 percent pore volume
flooding the models that were initially was left behind at the conclusion of the flood
508 PREINJECTION OF POLYMERS TO INCREASE RESERVOIR FLOODING EFFICIENCY SPE 5836
sequence, which recovered a total of 82 percent Table 6 lists volumetric sweep at break-
of the in-place oil by polymer and micellar through as determined from tracer data for the
flooding. fluid banks injec.ted in the five-spot model
tests. There is a consistent agreement between
The micellar flood in the second test the volumetric sweep at waterflood breakthrough
alone recovered 75 percent of. the oil in the and the sweep determined by the tracer injected
model at the commencement of the flood, com- at the conclusion of the waterflood. In Test
pared to 63 percent in the first test. This is MF-1, no polymer was preinjected ahead of the
interpreted as an improvement in volumetric micellar slug. Initial sweep to water is 21
sweep by the micellar flood in the test where percent pore volume and sweep at breakthrough
the flood was .preceded.by polymer preinjection. of the micellar bank tracer is 28 percent. In
Oil recovery performance after waterflooding is Test MF-2 in a similar five-spot model, a slug
shown for both tests in Figs. 1 and 2. of the high-molecular-weight polymer was injecte
in the waterflooded model and displaced by
Injection of the high-molecular-weight water. Breakthrough sweep of the. water injected
polyacrylamide polymer into the five-spot following the polymer displacement occurred at
models through the producing well was also 33-37 percent pore volume compared to an initial
found to result in better flooding efficiency sweep to water ahead of the polymer of 19
by a subsequent micellar flood conducted in the percent. Thus, the polymer appears to have
normal flooding direction. This was indicated increased the sweep to subsequently injected
by higher ultimate recoveries (rvl7 percent water by approximately 14 percent pore volume,
increase) and high oil cuts (rv40 percent i.e. , an improvement in volumetric sweep of
increase) during initial oil production by 74 percent. In this case, the tracer in the
micellar flooding (Fig. 3). micellar bank indicates breakthrough after 38
percent pore volume compared to 28 percent in
Tracer Performance Test MF-1 where no polymer was preinjected. In
tests where a bank of a biopolymer was dis-
Analysis of tracer elution curves provided placed through the model after waterflooding,
information on the changes in sweep during the .. the tracer breakthrough data before and after
flood tests made in heterogeneous laboratory polymer injection indicate little improvement
models. to sweep by water. This is not surprising, as
the biopolymers do not affect rock permeability
Typically, a tracer was injected with the and therefore have little effect on sweep of
water at the tail end of the waterflood to subsequently injected fluids.
determine the "initial" or "reference" break-
through sweep to water. Then, tracers were Comparison of Oil Recovery and Tracer
added to the polymer bank, the water behind the Performance
polymer bank, and the micellar preflush water
to determine their respective sweeps. The Fig. 6 compares tracer elution curves
volumetric sweeps by the polymer bank and sub- before and after injection of the polyacryla-
sequently injected water are compared with the mide polymer in the waterflooded five-spot
sweep of water prior to polymer injection. The filtros model MF-2 described in Table 4, with
micellar flood and its polymer mobility buffer results summarized in Tables 5 and 6. Tracer
bank are also traced, and their respective data show an improvement in sweep to water of
sweeps are compared with the initial sweep to 14 percent pore volume after polymer injection
water. Thus, the effect of polymer preinjec- (Table 6) and oil recovery was increased by 12
tion on the volumetric sweep by water and a percent pore volume following polymer flooding
subsequent micellar flood could be determined. (Table 5). Thus, the improvement in sweep
indicated by the tracers was accompanied by an
Tracer production curves for the five-spot increase in oil recovery. However, this.may
model tests described in Tables 4 and 5 are not always be true.
given in Figs. 4 and 5. The pore volumes shown
on the figures start at the injection of the The breakthrough sweep, which is critical
first traced water at the tail end of the for a subsequent miscible flood, can be improved
waterflood. There was little or no oil pro- without additional oil being recovered during
duction during this late stage of the water- the sweep improvement treatment. This was
flood. Injection of subsequent traced banks is demonstrated by the results of the test shown
indicated on the figures by the vertical lines in Figs. 7 and 8. This flood was conducted in
and arrows. Respective tracer production a three-layered linear heterogeneous model made
curves are identified with the appropriate of sand mixtures A, B and C. Characteristics
labels. Oil production response to polymer of the model and test sequence are given in
injection and subsequent micellar flooding is Table 7. The model was extensively waterflooded
shown on the same figures to facilitate com- to a low average residual oil saturation of 25
parison with tracer performance. percent pore volume. Following the waterflood,

--·· - --· ····---------------------·------·-----------------'

 SPE 5836 MAHMOUD K. DABBOUS AND LLOYD E. ELKINS 509
a slug of Dow Pusher 700 polymer was injected are tailored to match that of the oil-water
and displaced by water. Fig. 7 shows no oil bank which builds up ahead of the micellar
response during the polymer flood. Tracer fluid. The oil-water bank is usually the zone
data, however, show a 100 percent improvement of highest flow resistance, and its total
in volumetric sweep at breakthrough of the relative mobility is the basis for mobility
polymer chase water, as indicated by the com- control design for a micellar flood. 9
parison of tracer production curves before and
after polymer injection (Fig. 8) and the data Theory and experiments were considered in
in Table 8. the analysis of the flow of oil and water in
the oil-water bank of a micellar flood and
Improvement in breakthrough sweep can effects of preinjection of polyacrylamide
result from readjustment of fluid velocities, polymers thereupon. Theory was based on the
i.e., mobilities, in flow channels within immiscible displacement relationships for flow
different permeability zones. This may not of oil and water utilizing rock-fluid proper-
result in added oil recovery during the re- ties and interactions. Experiments were made
adjustment or sweep control treatment if the in long Berea sandstone cores. Based on the
various permeability zones contain immobile results of the analysis, an estimate was made
oil. However, this improvement in breakthrough of the additional polymer concentration
sweep could result in an increase in oil recov- required to maintain mobility control in the
ery by a subsequent flood in which only a small micellar flood as a result of increased flow
size slug of an oil recovery miscible fluid is resistance following initial displacement of
injected into the reservoir. polymers. This represents an added concen-
tration (~C) in the polymer mobility buffer
Further demonstration of the improvement bank above the concentration (C 0 ) required for
in breakthrough sweep without oil production a similar flood without a polymer preflush.
following polymer displacement in heterogeneous Fig. 9 is a plot of the percentage increase in
flow models is indicated by the data in Table 9 polymer concentration (~C/C 0 ) in the polymer
summarizing the results of polymer displacement drive bank of the micellar flood as a function
tests in linear repacked Berea cores. Improve- of the residual resistance factor (Rr) follow-
ment in volumetric sweep following the dis- ing the initial polymer displacement. The
placement of polymer in the linear heterogene- plot shows a moderate increase in mobility
ous cores varied between 13 and 50 percent, control requirement as a result of initial
with the larger improvement in the more hetero- polymer flooding. It should be emphasized,
geneous system. however, that this is not a unique correlation
as it depends on the characteristics and
INTERACTIONS WITH A SUBSEQUENT MICELLAR FLOOD interactions of the rock-fluid-polymer system
and the rheological properties of the drive
Following the displacement of polyacryla- polymer used in the micellar flood.
mide polymers through porous media, some of the
polymer is retained by virtue of complex mech- Oil Displacement
anisms of adsorption, entrapment and mechanical
plugging. These mechanisms contribute sepa- Oil displacement efficiency by micellar
rately or jointly to the residual resistance flooding was not adversely affected by prior
that causes reduction in effective permeability polymer displacement through the rock. Slugs
to subsequently injected aqueous fluids. The of Dow Pusher polymers 500 and 700, ranging in
interactions between the retained or pread- size from a fraction of a pore volume to
sorbed polymer and the characteristic fluid several volumes, were preinjected into water-
banks in a subsequent micellar flood were flooded Berea cores and displaced by water
subject to study and will be detailed in a prior to conducting a micellar flood. Large
future paper. Emphasis in the interaction polymer banks preinjected in some core flood
studies was on the effect of the preinjected tests were to simulate prolonged polymer flow
polymer on mobility control requirement for the in reservoir elements near the wellbore. The
micellar flood and oil displacement character- data in Table 10 showing final oil saturations
istics by micellar flooding. below 5 percent pore volume at the conclusion
of the micellar floods indicated that the high
Mobility Control oil displacement efficiency by the micellar
fluid is maintained following prior polymer
Micellar floods are designed to have flooding. In some cases, preinjection of
favorable mobility displacement conditions polymers appeared to enhance a high oil dis-
except for the displacement of the polymer placement efficiency by the subsequent micellar
drive bank by the chase water. 9 , 10 In order flood.
to achieve a unit mobility displacement in the
process, effective viscosities of the micellar
and polymer drive banks in the reservoir rock

1-----------------------------------l---------------------------------------~
510 PREINJECTION OF POLYMERS TO INCREASE RESERVOIR FLOODING EFFICIENCY SPE 5836

Surfactant Loss especially for injected fluids in banks of

small fractional pore volumes. This is
The micellar fluid used in this study con- accomplished by reducing the degree of
tained 5 percent (by weight) Mahogany AA sul- interwell permeability contrast in mod-
fonate as the surfactant, 3 percent isopropyl erately heterogeneous systems.
alcohol as a cosurfactant, and 92 percent 0.23N
sodium chloride brine. Surfactant loss to the 2. Higher oil recoveries by subsequent micel-
rock was calculated for the core flood tests lar floods were obtained following dis-
from the effluent profiles of both surfactant placement of polymer slugs in laboratory
and cosurfactant. Adsorption loss values models of heterogeneous porous media than
determined for micellar floods that followed are obtained without prepolymer injection.
injection and displacement of polyacrylamide
polymers ranged from 0.22 to 0.51 lbs active 3. Injection of polymer slugs during a mature
sulfonate per barrel pore space (Table 11). stage of waterflood in heterogeneous
The average adsorption loss figure for the same laboratory models produced extra oil.
surfactant in the same rock without prior poly- Similar observations have been reported in
mer flow is approximately 1 lb active sulfonate field applications.
per barrel pore space. It appeared that the
preadsorbed polymer reduced the loss of sur- 4. Prior polymer flow had no adverse effects
factant to the rock by a factor of 40 percent on oil displacement characteristics by
or more. This reduction in loss could help subsequent micellar floods.
preserve the integrity of the micellar fluid
when displaced through a reservoir. However, 5. Preadsorbed polyacrylamide polymers
it should be realized that this is only a side appeared to reduce surfactant loss to the
benefit, as the reduced surfactant loss may be rock from a subsequently injected water-
limited to certain polymer-micellar systems and external micellar fluid.
would occur only in sections of the formation
precontacted with polymer. 6. Extensive field testing under carefully
controlled conditions is necessary for
FIELD TESTING providing more definitive answers to the
quantitative capability of polymers to
Because of the difficulties involved in improve sweep by reducing interwell
attempting to scale complex reservoir hetero- permeability contrasts in the reservoir.
geneities in simple laboratory physical models
of heterogeneous porous media, the direct NOMENCLATURE
applicability of results obtained in such
models to real systems is limited. Field A Area of flow, ft 2
testing in different environments and under h Thickness, ft
carefully controlled test conditions is essen- i Layer index: 1, 2, ... n
tial to quantify the capabilities of polymers K Absolute permeability, millidarcies
to improve sweep and increase oil recovery. As or darcies
many test evaluation tools as possible should Permeability variation coefficient
be used. This includes tracers, oil recovery, Permeability to water (at waterflood
injection profiles, and production logs, with residual oil), millidarcies or darcies
particular emphasis on extensive interwell Length, ft
tracer programs. Minitests utilizing single- Pore volume, ml
and two-well test arrangements might be used as Resistance factor
a substitute for more elaborate multiwell pilot Residual resistance factor
field tests to provide some quick answers to a Initial oil saturation, % pore volume
pressing problem. The level of current knowl- Residual oil saturation (after water-
edge is considered insufficient to provide flooding), %pore volume
widespread use of polymers to improve sweep, Final oil saturation (after micellar
particularly for preinjection in micellar flooding), %pore volume
floods. Proposed field testing is expected to Connate water saturation, % pore
move technology up the learning curve in an volume
expeditious manner. This is highly desirable sv Vertical sweep number (proportional.
since the timing of development is an important' to layer properties, saturations and
factor if full value of having available sweep volume injected at breakthrough)
improvement methods is to be realized. <P Porosity, fraction
OIP Oil-in-place (after waterflooding),%
SUMMARY pore volume
OOIP Original oil-in-place (prior to
1. Preinjection of specially selected poly~ waterflooding),% pore volume
mers should improve volumetric sweep, f' TOR Tertiary oil recovery, % OIP
or
'----------------------------------'--~---- --------
 SPE 5836 MAHMOUD K. DABBOUS AND LLOYD E. ELK_I_N_S_ _ _ _ _ _ _ _ _ _ _ _ _ _ _S--.11

Concentration required for polymer 5. Clampitt, R. L. and Reid, T. B.: "An

drive bank Economic Polymer Flood in the North Burbank
t:.C Additional concentration for polymer Unit, Osage County, Oklahoma," SPE 5552,
drive bank needed in micellar floods 50th Annual Fall Meeting of SPE-AIME~
preceded by a polymer preflush Dallas, Texas, Sept. 28-0ct. 1, 1975.
N Normality
6. Nugent, G. E. et al: "Sensitive Clay
REFERENCES Sands Respond to Chemical Oil Recovery,"
Petroleum Engineer, Nov. 1975.
1. Sloat, B.: "Choosing the Right Floods for
Polymer Treatment," Petroleum Engineer, 7. Knight, R. K. and Baer, P. J.: "A Field
May, 1971. Test of Uniflood at Higgs Unit," SPE 3807,
Improved Oil Recovery Symposium, Tulsa,
2. White, J. L. et al: "Use of Polymers to Okla., April 16-19, 1972.
Control Water Production in Oil Wells,"
SPE 3783, Symposium on Improved Oil 8. Sarem, A. M. and Holm, L. W.: "U. S.
Recovery, Tulsa, Okla., April 16-19, 1972. Patent 3,704,990 (Dec. 5, 1972).

3. Goddar, J. E. et al: "Improved Control of 9. Gogarty, W. B. et al: "Mobility Control

Water in Producing Wells and Injection for Miscible-Type Waterfloods Using
Wells," SPE 5402, SPE Regional Meeting, Micellar Solutions," Journal of Petroleum
Oklahoma City, Okla., March 24-25, 1975. Technology, February, 1970. --

4. Sloat, B.: "Increasing Oil Recovery by 10. Trushenski, S. P., Dauben, D. L. and
Chemical Control of Producing Water-Oil Parrish, D. R.: "Micellar Flooding-
Ratios," SPE 5341, SPE Rocky Mountain Fluid Propagation, Interaction and Mobil-
Regional Meeting, Denver, Colo., April 7-9, ity, SPE Journal, Vol. 14, No. 6, Dec.,
1975. 1974.

L-----------------------------------'-----·- ------
TABLE 1 - POROUS MEDIA AND RESIDUAL RESISTANCE FACTOR
¢ k Swc 8 oi Sor kw(Sor)
Porous Medium Description R
Percent Darcy Percent Percent Percent Millidarcy r Polymer
Unconsolidated:

Sand A 97% No. 16 Ottawa 37 1.20 15 85 39 220 4. Pusher 700

3% Silica Flour

Sand B 85% No. 16 Ottawa 36 0.25 22 78 33 60 7. II

15% Silica Flour

Sand C 75% No. 16 Ottawa 35 0.07 36 64 26 25 + II

25% Silica Flour

Consolidated:

Berea Natural Sandstone 21 0.4- 39 61 34 35- 5- II

0.75 100 12
Filtros Synthetic 39 5. 19 81 33 740 9.2 ICI 2060
Grade Q-5

Filtros Synthetic 39 10. 17 83 31 1530 4.4 II

Grade Q-10

Filtros Synthetic 39 20. 18 82 35 3060 3.1 II

Grade Q-20

+Excessive pressure gradients during polymer injection.

 TABLE 2 - FLUIDS
Initial Saturation

0.2N NaCl Brine (O.OSN for tests in Filtros)

Second Wall Creek crude:

Viscosity at ll0°F* 4 cp
90°F 6.5 cp
Density at 90°F 0.84

Waterfloods

0.2N NaCl Brine (0.05N for Filtros)

Preinjection Polymer

Dow Pusher 700 solutions in .2N NaCl brine

ICI 2060 polyacrylamide solutions in .2N NaCl brine

0.2N NaCl brine chase water (0.05N for Filtros)

Micellar Preflush

0.23N NaCl brine

Micellar Fluid

Composition: 8% by wt micellar concentrate (5/3 Mahogany AA

sulfonate and IPA)
92% by wt 0.23N brine

Polymer Drive 1200-1500 ppm Kelzan MF biopolymer solutions

in 0.05N NaCl brine

Chase Water 0.05N NaCl brine

*Test temperature ll0°F (Second Wall Creek reservoir temperature).

 TABLE 3 - VISCOSITY AND SCREEN FACTOR OF PUSHER 700 AND ICI POLYMERS
ICI Pusher
2060 700

Molecular Weight, million rv20 4-5

Brookfield Viscosity (ll0°F), LVT Model with UL Adapter

12 rpm 8.6 cp 6.0 cp
60 rpm 5.75 cp 4.8 cp

Screen Factor (72°F)

Filtered+ 44 28
Unfiltered Plugs 29
Screens

*1000 ppm in 0.2N NaCl solution.

+Filtered through 200 mesh screen.
TABLE 4 - FLUID AND TRACER INJECTION SEQUENCE FOR FLOOD TESTS IN TWO-LAYER,
FIVE-SPOT MODELS
Model Characteristics
(MF-1 and MF-2)

h. K Permeability
l.
ct> (Air) kw(Sor) Contrast
Layer I:h.
__l._ % d md At Water Floodout
1 0.5 39 11 3,060
2 0.5 39 4.5 760 4:1

Fluid Injection Seguence

Test MF-1

3.30 PV Preflush 2.77 PV brfne (.05N)

0.78 PV brine (.23N) traced with 100 ppm NH4N03
0.09 PV Micellar 8% by wt micellar concentrate
92% by wt 0.23N brine
0.58 PV Polymer 1250 ppm Kelzan MF in 0.05N brine
1.39 PV Chase Water 0.05N brine

Test MF-2

3.06 PV Preflush 0.96 PV brine (.05N)

0.63 PV brine (.05) traced w/200 ppm NH4N03
0.20 PV ICI 2060 polymer (400 ppm in .05N
brine) w/1% ETOH
0.63 PV chase brine (.05N) traced w/.5% IPA
0.32 PV brine (.23N +250 ppm Kelzan) traced
w/190 ppm NH4CNS
0.32 PV chase brine (.23N) -No tracer
0.13 PV Micellar 8% by wt micellar concentrate
92% by wt 0.23N brine thickened with 250 ppm
Kelzan MF polymer
0.63 PV Polymer 1500 ppm Kelzan MF in 0.05N brine
1.46 PV Chase Water .05N brine
 TABLE 5 - OIL RECOVERY PERFORMANCE FOR FLOOD TESTS IN TWO-LAYERED,
FIVE-SPOT MODELS
MF-1 MF-2
Initial Saturation

8 PV 84.9 84
oi' %
swi' % PV 15.1 16

Waterflood

Water Breakthrough, % PV 18 20

S , % PV 43 46
or
Recovery, % OOIP 49 45

Polymer Flood None Yes*

s or' % PV 43 34

TOR, % OIP 26

Micellar Flood

S , % PV 16 8.7
or
Total TOR, % OIP+ 63 82

Micellar TOR, % OIP++ 63 75

TOR: Tertiary Oil Recovery.

* 20% PV slug of 400 ppm ICI 2060 polyacrylamide.
+ Based on oil-in-place after waterflooding.
++Based on oil-in-place after polymer flooding.

TABLE 6 - TRACER PERFORMANCE FOR FLOOD TESTS IN TWO-LAYERED,

FIVE-SPOT MODELS
Volumetric Sweep at Breakthrough
% PV

MF-1 MF-2
Waterflood Breakthrough 18 20
Flood Water Tracer+ 21 19
Polymer Bank* None 37
Injected

Chase Water 33

Micellar Preflush Water 21 35

Micellar Bank 28 38

Polymer Drive Bank 29 43

Percent Volumetric Sweep

Improvement of Micellar
Bank Over Sweep of Water 34 90

+Trac~r injected with water at tail end of waterflood.

*20% PV slug of 400 ppm ICI 2060 polyacrylamide.
 TABLE 7 - FLOOD TEST IN A LINEAR HETEROGENEOUS MODEL
Model Characteristics *

h.
Layer 1 K. kw. (S )
Eh. 1 1 or
1 md md

1 .42 350 65
2 .26 1400 250
3 .32 70 25

s oi 65% PV+

sor 25% PV+

\(

Fluid and Tracer Injection Sequence

1.73 PV Preflush 0.42 PV brine (0.2N)

0.19 PV brine (0.2N) with CNS tracer 100 ppm
0.16 PV brine (.2N)
0.11 PV 400 ppm Pusher 700 430 ppm IPA
0.27 PV (225 ppm) Pusher 700
0.29 PV brine (0.2N) w/75 ppm Pusher 700 and 1%
ETOH
0.29 (0.2N) brine with NH4N03 tracer (100 ppm)

0.12 PV Micellar 8% by wt micellar concentrate

92% by wt 0.23N brine

0. 58 PV Polymer 1500 ppm Kelzan MF in 0.05N brine

2. 86 PV Chase 0.05N brine

*Average values for· each layer (considerable variation within each layer).
+Average values for the model.

TABLE 8 - BREAKTHROUGH SWEEP IN A THREE-LAYER LINEAR

HETEROGENEOUS MODEL
Percent PV
Waterflood Breakthrough 28
Flood Water Tracer 22
Polymer Bank* 48
Chase Water+ 44
Micellar Preflush Water 32
Micellar Bank 53
Polymer Drive Bank 56

*Dow Pusher 700.

+70 ppm polymer added to water to maintain residual resistance.
 TABLE 9 - SWEEP TESTS WITH POLYMERS IN LINEAR HETEROGENEOUS MODELS-POROUS MEDIA
Model Characteristics Tracer Breakthrough Data SweeE ImErovement
(1) (2) (3)
Water
+
Displace- Kl K2 Initial Polymer After Percent *
PV :PV K :K ++ K A :K A
ment Rate Water Bank Bank Polymer Increase
Test ft/da~ L/h 1 2 1 2 1 1 2 2 ~1=~2 % PV % PV % PV in SweeE

L-1 1.0 19 1:2 4.8:1 1.4:1 2.8:1 28 42 50

L-2 2.3 10 1:2.4 3.3:1 1:1.3 1.9:1 52 71 61 17

L-2 2.3 19 1.2 3.8:1 1.1:1 2.1:1 37 51 42 13

~30% PV slug of Pusher 700.

* (3)-(l)
(1)
X 100

++Permeability Range: 400-3,500 md

TABLE 10 - SUMMARY OF RESIDUAL OIL SATURATIONS AND TERTIARY OIL RECOVERY BY MICELLAR
FLOODING IN BEREA CORES
swe s oi s or 8
of TOR
Run No. Preflush % PV % PV % PV % PV % OIP
8440-lo! Water 38.0 62.0 38.2 6.0 84.3
8439-05 Polymer 34.5 65.5 31. 2.7 91.3
8439-08 II
35.3 64.7 28. 3.5 87.5
II
8439-22 40.6 59.4 31.7 0.7 97.8
II
8440-24 44.5 55.5 31.3 0 100.
II
8439-30 41.9 58.1 34.7 3.7 89.3
II
8439-40 42.5 57.4 35.2 0.7 98.0
II
8568-01 35.1 64.9 37.9 2.6 93.0
II
8568-10 39.9 60.1 34.7 1.0 97.0

+Micellar slug size 0.10 PV.

* Micellar slug size 0.20-1.20 PV.
TABLE 11 - SURFACTANT LOSS FOR MICELLAR FLOODS WITH AMOCO 231 FOLLOWING FLOODS WITH DOW PUSHER POLYMERS
Percent
Polymer Micellar Sulfonate Loss * Reduction +
Run No. Polymer Slug Slug Size Slug Size (Active) in Sulfonate
PV PV lb/bbl PV Loss

8439-30 225 ppm 0.22 1.08 0.51 49

Pusher 700

8439-40 370 ppm 0.23 0.22 0.49 51

Pusher 500

8468-01· 400 ppm 1.18 0.23 0.38 62

Pusher 700

8568-10 400 ppm 1.26 0.25 0.22 78

Pusher 500

*Mahogany AA: Active Sulfonate ~ 60% of total sulfonate.

+Based on ~1.0 lb/bbl sulfonate loss for micellar floods in Berea without prior polymer injection.

a.
0
~
WATER
POLYMER
. CHASE WATER
>-' MICELLAR
0:::
LLJ
~>
.. 0
1- u
::::> LLJ PV = 1383.0, ML
uo:::
SWC = 15. 1 , % PV
--
~~

oo SOR = 42.8 , % PV
>-
0:::
SOF = 16. 0 , % PV
::: OIL CUT TOR= 62.6, OIP
~ II I
LLJ
1-
I
I

PORE VOLUMES PRODUCED

F i g. I
0 i I recovery response to a mice I I ar
flood 1n a waterflooded two-layer, five-spot
mode I .
 c..
o1oo LL' ~u.
l!F. POLYMER CHASE MICELLAR POLYMER
>-"' WATER
0::
t;F.LU
>
~o
1-u
=>UJ
Uo::
::::!,....J
oc; 50 PV = 1327. 0 , Ml
SWC = 16. 0 , % PV
>-
0:: SOR = 46. 1 , % PV
:$ SOF = 8. 1 , % PV
~ TOR = 82.3, % OIP
UJ
1-

0
PORE VOLUMES PRODUCED
Fig. 2 - Oi I recovery res~onse to p61ymer pre-
injection followed by a micellar flood in a water-
flooded two-layer, five-spot model.

t:F. 100 _+ + POLYMER

MICELLAR

50 OIL CUT PV = 1285.0 I Ml

SWC = 17. 7 , % PV
SOR = 34. 7 , % PV
SOF = 8. 6. , % PV
TOR = 75. 1 , % 0 IP

0~~~~~~~=-L_~~~_u~~~L_~
0.75 1.00 1.25 1.50
PORE VOLUMES PRODUCED
Fig. 3- Oi I recovery response to a micellar flood
in a waterflooded, two-layer, five-spot model pre-
t r e a t e d wi t h p o I y me r f r o m p r o d u c i n g we I I •

z
0

~100 WA_!ERr SULF.

0::
1-
CHASE WATER
z +
UJ
l!F.~ N03 IPA
~O
~u SULF.
uc
UJ
....J
-u
1- PV = 1383.0, ML
0
~50 SWC = 15. 1 , % PV
z SOR = 42. 8 , % PV
c SOF = 16. 0 I % PV
UJ
u TOR= 62.6 , % OIP
=>
c
0
0::
Q..

PORE VOLUMES PRODUCED

Fig. 4- Effluent profiles for a micellar flood
test in a waterflooded two-.layer five-spot model.
 ~

z
0
1-
<(
100
.. .. .. .. .. ..
WATER
+
N03
WATER
+
IPA
POLYMER
+
ETOH
. CHASE WATER

0:::
1- POLYMER SULF.
z
LU + +
~u ETOH IPA
z
-o
su
Uc
.....II=!
- u 50
(OILCU
OLU
-.,
~
c
LU
u
:::::>
c
0
0:::
0..
0
0.50 1.00 1. 50 2.00 2.50 3.00 3.50 4.00
PORE VOLUMES PRODUCED
F i g. 5 - E f f I ue n t prof i I e s for a p o I y mer- m i c e I I a r
flood sequence in a waterflooded two-layer five-
spot mode I .
~ 100,....------------------------------,
z
2 90
<
~ 80
z
z~ 70
8 60
c
I=!
u 50
LU
-.,
z 40
c 30
LU
u
:::::>
c 20
0
0:::
0.. 10

.10 .20 .30 .40 .50 .60 .70 .80 . 90 1.0

PORE VOLUME INJECTED
Fig. 6- Tracer production curves before and after
polymer displacement in a two-layer five-spot model.

POLYMER CHASE WATER

+ + + +
IPA ETOH IPA ETOH
PV = 647.2 I ML
SWC = 35. L % PV
SOR = 25.2 1 % PV
SOF = 8.1 I % PV
TOR = 67. 7 I 0 Ip

c
LU
u
:::::>
c
0
0:::
0..
OL_~~~~~~~~~~~~~~~~LU-L_L~~~~~~
0.25 0.50
1.25 0.75
1.50 1.00
1.75 2.00 2.25 2.50 2.75 3.00
PORE VOLUMES PRODUCED
Fig. 7 - Effl'uent profiles for a polymer-mice! l2r
f I ood sequence in a I i near heterogeneous mode I of
unconso I i dated sands.
 ~ 100
:z
0
1-
90
<(
0::::
1- 80
:z
LLJ
u 70
:z
0
u ()()
c
I=! 50
u
.....,
LLJ
:z 40
BEFORE
c
LLJ
30
u
=>
c 20
0
0::::
0.. 10

.10 .20 .30 .40 .50 .60 . 70 .80 . 90 1.0

PORE VOLUME INJECTED
Fig. 8 - Tracer production pi''Of i I es before and after
p o I y me r d i s p I a c _em e n t i n a I i n e a r he t e r o g e n e o u s · mo d e I
of unconsolidated sands.

50

40
~

8~
.-I
30
><
ul o
<lu
20

10

1 2 3 4 5 6 7 8

Rr
Fig. 9 - Calculated increase in polymer
concentration required for mobi I i ty con-
tra I in the buffer bank of a mice I I ar
flood following initial flow of a resi-
dual resistance building polymer.