See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/266671268

Low-Salinity Polymer Flooding: Improving Polymer Flooding Technical

Feasibility and Economics by Using Low-Salinity Make-up Brine

Article · January 2014

DOI: 10.2523/17342-MS

CITATIONS READS

44 1,486

5 authors, including:

Shehadeh Masalmeh
Abu Dhabi National Oil Company
114 PUBLICATIONS 0 CITATIONS

SEE PROFILE

All content following this page was uploaded by Shehadeh Masalmeh on 02 February 2016.

The user has requested enhancement of the downloaded file.

IPTC 17342

Low-Salinity Polymer Flooding: Improving Polymer Flooding Technical

Feasibility and Economics by Using Low-Salinity Make-up Brine
Esther C.M. Vermolen, Monica Pingo-Almada, Bart M. Wassing, Dick J. Ligthelm and Shehadeh K. Masalmeh,
Shell Global Solutions International B.V.

Copyright 2014, International Petroleum Technology Conference

This paper was prepared for presentation at the International Petroleum Technology Conference held in Doha, Qatar, 20–22 January 2014.

This paper was selected for presentation by an IPTC Programme Committee following review of information contained in an abstract submitted by the author(s). Contents of the paper, as
presented, have not been reviewed by the International Petroleum Technology Conference and are subject to correction by the author(s). The material, as presented, does not necessarily
reflect any position of the International Petroleum Technology Conference, its officers, or members. Papers presented at IPTC are subject to publication review by Sponsor Society Committees
of IPTC. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the Internati onal Petroleum Technology Conference 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
where and by whom the paper was presented. Write Librarian, IPTC, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax +1-972-952-9435

Abstract
Polymer flooding is a mature EOR technique, which is successfully applied in both sandstone and carbonate reservoirs. In
ongoing polymer projects, make-up brine is either formation water, sea water or any available water sources like deep or
shallow aquifers. In this paper we focus on the use of low salinity water as the make -up brine. The objectives of combining
low salinity flooding (LSF) with polymer flooding are three-fold:
• Using low salinity brine reduces the amount of polymer required to obtain t he target viscosity, which may lead to
significant cost reduction.
• Combining the benefit of low salinity flooding with polymer flooding leads to higher oil recovery over conventional
polymer flooding.
• Enhancing the elasticity of polymers by using low s alinity brine which may lead to reduced Sorw and increased oil
recovery.

In addition to the objectives mentioned above, the use of a low-salinity make-up brine can give other benefits, such as better
polymer stability (especially at high temperatures), lower sensitivity to polymer shear degradation, lower polymer adsorption
and lower scaling and souring tendency.
The paper will present
1- Experimental procedures for investigating the potential benefits of low salinity polymer on both the required polymer
concentration and the oil recovery.
2- Experimental results for several field cases
3- De-risking activities that were undertaken to mitigate any potential negative impact of using low salinity polymer, in
the areas of clay swelling, polymer shear sensitivity, mixing and adsorption.
The paper concludes that low-salinity polymer flooding can significantly improve existing and anticipated polymer flooding
projects by reducing polymer volumes and/or increasing oil recovery. Low-salinity polymer flooding provides opportunities to
apply polymer flooding in high-salinity and high-temperature reservoirs, for which polymer flooding with produced or
formation water would be technically unfeasible or uneconomic.
2 IPTC 17342

Low salinity polymer flooding: the benefits

Both polymer flooding and low-salinity flooding are Enhanced Oil Recovery (EOR) techniques that are relatively easy to
apply in a wide range of oil fields. Polymer flooding is a well-known EOR method based on the sweep efficiency
improvement (over that of a conventional waterflood) achieved by viscosifying the water. The application envelope of
polymer flooding is mainly limited by the reservoir temperature and the make-up brine salinity and hardness due to chemical
degradation of the polymer, while the application of low-salinity flooding only depends on the availability or the option to
produce a low-salinity injection brine. The benefit of low-salinity flooding is dependent on the combination of the reservoir
mineralogy, the reservoir brine and crude composition.

Higher incremental oil recovery

By combining both techniques, i.e. by injecting a low-salinity polymer solution, the mutual benefits are amplified. First of all,
adding polymer to a low-salinity flood improves the sweep efficiency and it could mobilize part of the oil detached by the low-
salinity brine, which otherwise would have been trapped (Shaker Shiran 2013). In addition, by using a low-salinity brine to
make-up the polymer solution, a low-salinity effect could increase the incremental oil recovery of the polymer flood by
changing the wettability of the rock surface and releasing extra oil. The workflow for determining the incremental oil recovery
will be described in more detail in the last chapter.

Lower required polymer concentration

In addition to the combined incremental oil of the two EOR techniques , the polymer concentration needed to obtain a certain
viscosity is much lower. The extent of this gain becomes clear from Figure 1: In a low-salinity reservoir with an original
salinity of 7,000 ppm Total Dissolved Solids (TDS) the polymer concentration needed to obtain a viscosity of 60 cP at a shear
rate of 11.5/s can be halved when desalinating the make-up brine to 10 times its original salinity. This would be a significant
reduction in polymer OPEX, while the OPEX/CAPEX to desalinate such a low-salinity brine is usually much lower. The
reduction in polymer concentration is even larger when operating in a highly saline reservoir. The right graph in Figure 1 is an
example of a 260,000 ppm TDS reservoir, where a 3-4 times reduction in polymer concentration can be made by lowering the
make-up brine salinity to 1,500 ppm TDS. Desalinating the 260,000 ppm TDS produced water would n ot be technically
feasible option yet, so in this case acquiring lower salinity brine from a different source and handling the produced water reject
stream have to be taken into account as well.

Figure 1: Tw o exam ples of the gain in polym er concentration w hen low ering the m ake -up brine salinity. Both fields are low
tem perature (around 50°C) fields and all viscosities given are at reservoir tem perature for the sam e polym er (20 MDa HPAM).

Larger application envelope of polymer flooding (increased polymer stability)

Another benefit of using a different make-up brine for the polymer solution than the reservoir brine is the fact that it provides
opportunities to apply polymer flooding to high-salinity and/or high-temperature reservoirs that would otherwise be
uneconomic or impossible to treat with polymer flooding. The most widely (and cheapest) type of polymer used in EOR
projects is partially hydrolyzed polyacrylamide (HPAM). This polymer is electrically charged which results in a viscosifying
power that is sensitive to the water composition (e.g. salinity and hardness). Moreover the molecule can increase its charge
over time (by hydrolysis) thereby increasing the interaction with ions in the water. Above certa in temperature and
salinity/hardness levels the HPAM has poor viscosity-stability characteristics. In several studies, polymers were identified that
are stable in high-temperature or high-salinity reservoir brines (Levitt 2008) (Doe 1987)(Gaillard 2010)(Vermolen 2011).
However, due to the higher cost of these polymers combined with a larger amount required to achieve a target viscosity, the
project economics often become challenging. If the make-up brine would be desalinated (or more specifically, if the divalent
ions would be removed to a large extent), a range of much cheaper polymers will become applicable in the field (Seright, et al.
2009) improving project economics substantially.
IPTC 17342 3

Lower shear sensitivity

Viscosity loss by mechanical shearing of the polymers in pumps, valves, chokes, etc. has to be accounted and potentially
compensated for to obtain the required viscosity in the reservoir. As was shown by Zaitoun et al. (Zaitoun 2011), the shear
sensitivity of HPAM-related polymers decreases with decreasing brine salinity. The salinity range studied had a lower limit of
2000 ppm TDS, which is still above the salinities that are often aimed for in low-salinity flooding. In this paper, this salinity
range is extended to even lower salinity of 1120 ppm TDS for a 20 MDa HPAM polymer through a 0.4 mm orifice at different
flow rates. The degradation was calculated by:

(𝜇 𝑟0 − 𝜇 𝑟 )
𝐷𝑒𝑔𝑟𝑎𝑑𝑎𝑡𝑖𝑜𝑛 (% ) = 100
(𝜇 𝑟0 − 1)
where 𝜇 𝑟0 is the viscosity at a chosen shear rate of 2.2/s of the non-degraded polymer solution and 𝜇 𝑟 is the viscosity at a shear
rate of 2.2/s for the sheared polymer.
It was found that also for lower salinities, shear sensitivity still decreases with decreasing salinity (Figure 2). When decreasing
the salinity even further (not shown in the figure), no clear trend could be derived. The shear sensitivity seems to be flow rate
dependent in that regime.

Figure 2: Shear degradation w hen injecting an HPAM polym er through a 0.4 m m orifice at different flow rates.

Reduced production chemistry issues

Finally, combining low-salinity flooding and polymer flooding also reduces production chemis try issues at several stages of
the development scheme. Both scaling and souring are reduced when using low-salinity brine. The brine can be tailored to
avoid high scaling tendencies and the removal of sulphates (when desalinating for example seawater) will reduce the H2 S
formation. Also the negative effect of the polymer on the oil/water separation can be reduced by using a low-salinity polymer:
the concentration of polymer is much lower and the viscosity can easily be lowered by adding some of the high -saline reject
stream of the desalination system.

Potentially larger reduction of the residual oil saturation by the visco-elastic effect
Recently, researchers have claimed that polymer solutions with a higher visco -elasticity can reduce the residual oil saturation
(e.g. (Xia, et al. 2008) and references therein). The exact microscopic mechanism (cleaning dead ends, local pressure increase
or stabilizing oil threads) is still under discussion. In addition, there is no agreement yet on the extent of the effect (whether
present in both oil- and water-wet systems and whether the effect will be substantial on the field scale). Nevertheless, the
effect should be higher when the visco-elasticity of the polymer solution is higher. This parameter has been shown to increase
with polymer concentration and polymer molecular weight, but also decreasing salinity has a positive effect on the visco -
elasticity of the polymer solution. Actually, the salinity has a stronger effect on visco-elasticity over most of the range than the
concentration. As a result, the visco-elasticity increases with decreasing salinity, even when keeping the viscosity the same.
Dynamic (oscillatory) frequency sweep measurements were done to determine the G’ and G’’ relations for a range of polymer
solutions with varying salinity and concentration. The relaxation times of the polymer solutions can be compared by plotting
the inverse of the frequency of the cross -over point of G’ and G’’ as was done in Figure 3. In the same figure, the iso-viscosity
lines are drawn, which shows that the visco-elasticity increases when going from a higher salinity to a lower salinity while
4 IPTC 17342

keeping the viscosity constant (i.e. by adjusting the concentration). From this we can conclude that low-salinity polymer
flooding would in principle result in a higher visco-elastic effect than high-salinity flooding at the same viscosity.

Figure 3: Relaxation tim es of a large range of polymer solutions, all m ade up with the same polymer (a high MW HPAM) w ith varying
polym er concentration and brine salinity. The black lines indicate the iso -viscosity lines.
IPTC 17342 5

De-risking potential negative impacts of the use of low -salinity make-up brine for polymer flooding

As can be judged from the benefits listed above, even ignoring the incremental oil by the low-salinity effect, using a low-
salinity brine for making up the polymer solution for polymer flooding can give a substantial gain. However, injecting a brine
with a different composition than initially present can result in clay swelling and cation stripping. Also, since the polymer
concentration is much lower and it is on the steeper part of the viscosity vs. salinity curve (Figure 4), the polymer solution is
more sensitive to changes in the brine composition and to adsorption than in the high-salinity case. In the sections below, the
different de-risking activities are described that test the effect of cation exchange, mixing and adsorption on the polymer
viscosity and the project economics.

Figure 4: Viscosity vs. salinity curve for a normal HPAM (~20 MDa) at a shear rate of 6/s. The reservoir brines are usually in the lo w er
plateau range, w hile the low-salinity range is more on the steep part of the curve, w hich increases the sensitivity of the viscosity to
salinity changes. Note that the last point is slightly higher than the point before, how ever taking into account the experim ental
uncertainty, these viscosities are the sam e.

De-risking: Cation exchange and clay swelling

The usage of low-salinity water as a make-up water for the polymer solution introduces several risks because the low-salinity
water is not initially in equilibrium with the reservoir rock and hence equilibrium needs to be established while the injected
water moves through the reservoir. Furthermore, the polymer itself is a charged molecule that has to equilibrate with the
porous media and the changing brine as it is injected. The re-equilibration process can lead to cation exchange between the
brine and porous media with the risk of clay swelling potentially resulting in reduced injectivity or mineral dissolution which
may increase the multivalent cations in solution affecting the viscosifying power of the polymer.

The rock and fluids interactions need to be considered at the pore scale. For sandstone rocks, the pore walls are normally
negatively charged surfaces mainly due to the presence of clays. Clays have a sophisticated interlayer structure and their net
surface charge varies depending on the type of clay and media conditions (e.g. pH). The surface charges in contact with the
liquids are electrostatically screened by a layer of counter-ions, cations in this case, in the electrolyte known as the diffuse
double layer. The concentration of cations in the brine near the pore walls will be higher than in the bulk and vice versa. If the
electrolyte concentration varies (i.e. salinity) a re-accommodation of the charges will occur (Mysels 1967). The double layer
will expand with a reduction in salt concentration which can lead to some interference between the clay sheets. This
destabilization varies according to the balancing forces between the sheets and can cause swelling and eventually clay
destabilization and/or deflocculation if the sheet binding forces are exceeded. This destabilization of clays could eventually
cause pore plugging yielding permeability reduction and in some cases injectivity problems.

A criterion for the analysis of the clay stability has been devised in the 90’s by Scheuerman (Scheuerman 1988)(Scheuerman
1989) after experimental work carried out on pure clay material without oil presen t based on the role of the brine divalent
cations for maintaining clay stability. The theory established diagrams indicating zones of clay stability per clay type
according to brine cation concentration and fraction of divalent cations in the brine. However, these diagrams may not be
applicable in the presence of oil, as will be discussed below.

The clay is out of equilibrium as soon as the Mass Action Ratio (MAR) of the contacting brine is changed. The MAR is the
square of the monovalent cation concentration divided by that of the divalent cations, both in milli-equivalents per liter:
MAR=(C+)2 /C2+. As a fluid with a lower salinity (a diluted brine) is injected into a rock material, the clays will absorb divalent
cations from the brine and exchange it for monovalent cations (Appelo 1994). The re-equilibration process is completed when
6 IPTC 17342

the MAR of the brine is equal to the MAR of the clay. The volume of brine (i.e. the number of pore volumes) needed for re-
equilibration depends on the cation exchange capacity (CEC) of the rock. Furthermore, the usage of an anionic polymer
(HPAM), a negatively charged molecule adds an extra complication to the system. The polymer can be thought of having a
role as exchanger comparable to that of clays.

In order to de-risk the effect of cation exchange during low-salinity polymer injection, core-flooding experiments were
performed with brines of varying salinities and polymer solutions, monitoring the cation exchange and/or mineral dissolution
(and its effect on the polymer viscosity) by measurement of differential pressure over the core plug, effluent elemental analysis
by Inductively Coupled Plasma (ICP) and effluent viscosity. The core flooding experiments were performed both at residual
oil saturation (in aged cores as well as in un-aged samples) and in core plugs without oil. Other auxiliary tools have been used
to interpret the data (e.g. cation exchange capacity (CEC) measurement and Scanning Electron Microscope (SEM)). The
investigation was done separately with low-salinity brine injection and low-salinity polymer flooding, to decouple the polymer
complexity (e.g. rheology) from the low-salinity effect.

Comparison of the experiments which tested formation damage boundaries with and without residual oil in the core plug
showed that the usage of unaged plugs may be slightly more pessimistic on predicting formation damage due to clay
destabilization (Mungan 1965)(Barkman, et al. 1975). This is probably caused by the fact that in a clean core plug more
surface area is exposed to the low-salinity brine. Since there will always be oil present in the field, the experiments at residual
oil should give a more realistic result. The experiments described below were therefore done at residual oil saturation.

Figure 5 (a) Scheuerman diagram displaying the measured effluent compositions low -salinity brine injection in a 2.9 D core plug at
residual oil saturation. Flooding sequence: reservoir brine (5600 ppm TDS)- 2× diluted reservoir brine- 4× diluted reservoir brine-10×
diluted reservoir brine- reservoir brine (5600 ppm TDS)- 10x diluted reservoir brine.
(b) Scheuerman diagram displaying the measured effluent compositions of a low -salinity brine/polymer injection in a 2.4 D core plug
at residual oil saturation. Flooding sequence: reservoir brine (5600 ppm TDS) - 10×diluted reservoir water - reservoir brine (5600 ppm
TDS) - Low salinity polym er (125 cP in 10× diluted reservoir brine).
IPTC 17342 7

Figure 5a shows the effluent composition on top of a Scheuerman diagram for an experiment in which a core plug was flooded
sequentially with:

 reservoir brine (5600 ppm TDS)

 2× diluted reservoir brine
 4× diluted reservoir brine
 10× diluted reservoir brine
 reservoir brine (5600 ppm TDS)
 10× diluted reservoir brine

This core plug had a high permeability of 2.9 D and was aged for 4 weeks with reservoir crude.

The full lines in the graph are the theoretical paths and the dots are the experimental measurement points. The sections of the
paths that go under the kaolinite (purple) and illite (green) lines of the Scheuerman diagrams are zones of risk of formation
damage in Scheuerman theory.

Point 1 is the starting reservoir brine composition. Point 2 is the cation stripped brine composition at the low salinity (2×
dilution) front. Point 3 is the injected low salinity brine composition (2× dilution). Point 4 is the cation stripped brine
composition at the low salinity (4× dilution) front. Point 5 is injected low salinity brine composition (4× dilution). Point 6 is
the cation stripped brine composition at the low salinity (10× dilution) front. Point 7 is injected low salinity brine composition
(10× dilution).

After all low-salinity brine injection steps, the core was flooded with reservoir brine, therefore the path went back to point 1.
The final step was a 10× dilution from reservoir brine. Point 8 is the cation stripped brine composition at the low salinity (10×
dilution) front in one step. The cation stripping is more severe when the dilution factor is larger. The final point in the f looding
is again point 7, after total re-equilibration of rock and fluid. This experiment clearly demonstrated the cation stripping effect
upon decreasing brine salinity.

Figure 5b shows effluent composition on top of the Scheuerman diagram for a core flooding experiment in which a core plug
was flooded sequentially with:

 reservoir brine (5600 ppm TDS)

 10× diluted reservoir brine
 reservoir brine (5600 ppm TDS)
 polymer solution prepared in 10× diluted reservoir brine (125 cp at 50°C and a shear rate of 2 sec -1 ).

This core plug also had a high permeability (2.4 D) and after initialization with a crude oil it was aged for 4 weeks. The full
black lines are theoretical paths and the dots are experimental measurement points. Point 1 is the starting reservoir brine
composition. Point 2 is the brine composition at the low salinity front during the retardation with cation stripping. Point 3 is
the final (injected) low salinity brine composition.

Experimental data on total cation concentration and fraction of divalent cations in the effluent of the brine low salinity
flooding (black dots) shows that the fraction of divalent cations varied less than theoretically anticipated thus there is less
cation stripping observed than predicted theoretically . Moreover, the data of the low salinity polymer flooding (red dots)
shows that the fraction of divalent cations in the effluent follows roughly the stripping trend (a decreasing divalent cation
concentration followed by an increase) while the starting and final divalent cation fractions were higher than the original level
in the reservoir. The polymer appears to preferentially attract Ca2+ and Mg 2+ ions from the core. The affinity of the divalent
ions to the polymer has also been observed in other in -house polymer flooding experiments and in Kozaki’s recent work
(Kozaki 2012). Because of the higher Ca2+ and Mg 2+ concentrations, the brine composition during the low-salinity polymer
flooding is totally outside the formation damage zone (assuming that these ions are freely available in solution); however, it is
unclear how this effect competes with the stabilizing role of the divalent ions to keep the clays stable. Another point of
consideration here is the increase of divalent cations in solution. An increase in Ca 2+ and Mg 2+ can decrease the polymer
viscosifying power, which has to be studied separately.

In this particular experiment a decrease of up to 30% of injected polymer viscosity was measured in the effluent. Uncertainty
remains as to whether the decrease in viscosity could be attributed to the increase in divalent cations in the electrolyte or to a
combination of factors, including some polymer adsorption and shearing in the core and/or core flooding setup. This could be
8 IPTC 17342

studied further by analyzing polymer concentration, hydrolysis degree and molecular weight distribution.

In both experiments described above as well as in the experiments performed without the presence of oil, no injectivity issues
were observed, not even where the brine composition entered the boundaries of formation damage in the Scheuerman
diagrams. This is probably due to the fact that a permeability reduction is not easy to observe at these high permeabilities (low
differential pressures).

In conclusion, a feasibility study for low-salinity (polymer) flooding should always contain field -specific brine/polymer
compatibility tests at Sor. Permeability impairment by potential clay swelling and deflocculation (if present) could be observed
by differential pressure measurement over the core plug. Elemental analysis on the effluent yields insight on the cation
exchange processes and the effect of the polymer as an additional cation exchanger. Finally, also the effluent viscosity (and
possibly other effluent polymer parameters) should be monitored to make sure that the polymer viscosity is maintained.

De-risking: mixing of low-salinity polymer with reservoir brine

While in principle the mixing of the low-salinity polymer injection slug with the reservoir brine could pose a threat to the
polymer slug viscosity, it will be shown here that because of the polymer adsorption, the polymer slug will not be in contact
with the high-salinity brine: a low-salinity buffering zone without polymer will form in between.

The performance of a low-salinity polymer flood is significantly determined by the degree of depletion of polymeric material
by adsorption to reservoir rock at the polymer front. The depleted volume Aads in Pore Volumes (PVs) is given by (Pope 1980):

𝐶𝑠 (𝐶𝑝 ) 1 − 𝜑 𝜌𝑠
𝐴𝑎𝑑𝑠 = {1}
𝐶𝑝 𝜑 𝜌𝑤
where
Cs (Cp ) = polymer adsorption as a function of the polymer concentration
Cp = bulk polymer concentration in ppm
ρs = rock density
φ = porosity
ρw = density polymer solution

For low-salinity polymer flooding, the polymer concentration Cp required is lower by typically a factor 4 compared to
conventional polymer flooding. Sparse data on the effect of salinity on adsorption (Martin 1983) indicates an approximate
halving of the adsorption when reducing salinity from 2%wt NaCl to 0.1%wt NaCl. This will vary however with polymer and
rock type. These two factors combined would yield an increase in Aads by a factor 2.

Polymer front propagation is not only affected by adsorption but also by hydrodynamic acceleration. Hydrodynamic
acceleration (Teeuw 1980) refers to the phenomenon that on average polymer molecules travel faster than the molecules of the
water in which they are dissolved. As a result of the polymer molecule size, its centers of gravity are a certain distance away
from the pore wall and the molecules will move in the inner part of the pore flow channels where the velocity will be higher.
Hydrodynamic acceleration counteracts the effect of adso rption (term AHA ) in that it will accelerate the propagation of the
polymer front. The retardation factor therefore changes to

𝐶𝑠 (𝐶𝑝) 1 − 𝜑 𝜌𝑠 1
𝐴𝑎𝑑𝑠 + 𝐴𝐻𝐴 = + ( −1 + ) (1 − 𝑆𝑜𝑟 ) {2}
𝐶𝑝 𝜑 𝜌𝑤 𝐻
Where
H = velocity enhancement factor (𝐻 ≥ 1)
Sor = remaining oil saturation behind polymer front

The correction term for hydrodynamic acceleration is typically about 0.15 PV. The magnitude of the hydrodynamic
acceleration is expected to increase with reduced salinity as this causes an increase in the h ydrodynamic volume of the
polymer molecule pushing the center of gravity further away from the pore wall towards the center of the pore channel.

For a conventional polymer flood with low net depletion Aads +AHA =0.05, the net depletion for the low salinity polymer flood
would be A=2×(0.05+0.15)-0.15=0.25. Hence, in a low salinity polymer flood the depleted volume is larger than in a
conventional polymer flood. This leads to retardation of the polymer front and a zone of low sa linity brine ahead of the
polymer front (Figure 6). There may be significant smearing-out of the front between displaced high salinity formation brine
and displacing low salinity brine by mixing due to diffusion and dispersion, whereas the polymer front remains relatively
sharp as a result of suppression of mixing at the polymer front due to high p olymer/brine viscosity ratio (Perkins 1963).
IPTC 17342 9

The overall result is that increased depletion by adsorption for a low salinity polymer flood creates a buffer zone of low
salinity brine ahead of the polymer front that is sufficiently large to protect low salinity polymer fluid from displaced hig h
salinity formation brine. The retardation of the polymer front has also an adverse effect . It will cause some delay in the
incremental oil recovery, which is described hereafter.
The high polymer/brine viscosity ratio also makes, that follow-up of a polymer slug by brine should be avoided or should be
done with a carefully designed grading scheme as it could yield early brine breakthrough as a result of viscous fingering of t he
brine through the polymer (Todd 1972). Hence a polymer flood must be designed such that incoming aquifer water is pushed
back.

1.0 1.0
buffer zone of
0.9 low salinity Normalized salinity 0.9
brine, being
0.8 depleted from 0.8
polymer
Aqueous Phase Saturation

0.7 0.7

Normalized Salinity
0.6 0.6

0.5 0.5
low salinity
0.4 polymer-flooded 0.4
zone
0.3 0.3

0.2 Polymer front 0.2

Aqueous phase saturation
0.1 0.1

0.0 0.0
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Dimensionless Distance
Figure 6: 1D sim ulation of low salinity polym er flood w ith depletion volum e A = 0.25 PV.

Technical feasibility versus economic viability

Technical feasibility of the use of low salinity (i.e. fresher and/or softer) injection water aimed at reduction of polymer OPEX
requires demonstration that the lack of chemical equilibrium bet ween rock and low salinity injection water does not lead to
unwanted effects such as injectivity loss through formation damage (clay swelling) or loss of polymer viscosity due to change s
to the injection water composition while in the reservoir (through processes as mineral dissolution, cation exchange or mixing
between injection water and formation brine). These aspects have been treated in detail in preceding sections.

However, the decision to apply polymer with low salinity brine is in the end an econ omic decision. Even in the case that it has
been demonstrated that injectivity and polymer viscosity are safeguarded, this decision is not purely based on the offset of the
polymer cost with the cost to provide the low salinity injection water, but is als o impacted by the effect of the lower polymer
concentration on the oil production function.

In the case of non-zero polymer adsorption (which is not fully compensated by the hydrodynamic acceleration) the oil
recovery will be delayed, inherent to the use of a lower polymer concentration. This can be understood intuitively: a reduced
polymer concentration will require a larger slug volume to satisfy the adsorption potential of the rock. Hence, a larger volume
needs to be injected before achieving polymer breakthrough and therefore the oil recovery is delayed (not necessarily lost!).
Even when the ultimate recovery is not affected by the recovery delay, it will result in a reduction of economic value of the oil
production for a non-zero discount-rate (discounted cash flow methodology) and erode the value of cost reduction aimed at by
the use of polymer under low salinity.
The fractional flow analysis for polymer flooding introduces a retardation factor representing the delay of the polymer front
advancement expressed in pore-volumes (Pope 1980) due to adsorption which is dictated by the ratio of polymer adsorption
over polymer concentration as given in Eq. 1.

Adsorption isotherms may be so steep that the use of lower polymer concentration hardly impacts the equilibrium adsorption
level for the same adsorption isotherm. Indeed under these conditions the reduction of the polymer concentration associated
with use of the low salinity water has the largest possible impact on polymer front advancement and hence the oil recovery
delay as can been seen from Eq.1: Cs constant, reduction of Cp results in a larger retardation factor implying a delay in polymer
propagation and oil recovery .
However the use of lower salinity water will result in a reduced maximum adsorption level (i.e. different isotherm than the
10 IPTC 17342

high salinity one). This will partially offset the effect of the lower polymer concentration on the oil recovery de lay.
Besides a lower adsorption, hydrodynamic acceleration will further offset the negative impact of lower polymer concentration
on the oil recovery delay.
For the quantification of the delay in oil recovery associated with a lower polymer concentratio n it is crucial to acquire data on
adsorption and hydrodynamic acceleration at original (high) salinity and the aspired low salinity brine composition.
Experimental procedures for dynamic measurement of adsorption and hydrodynamic acceleration are document ed by Hughes
et al. (Hughes 1990).
Below, an example is given that assesses the impact of reduced salinity on the economic value related to polymer cost saving
and oil recovery delay in a worst-case scenario. The example assumes that the same polymer (HPAM) can be used under
original reservoir conditions and with the low salinity injection water implying that the cost saving relates only to the polymer
concentration reduction. As a result of the large price level difference between oil and HPAM it is no surprise that the oil delay
has a dominating effect on the economics. This large price level difference is also the reason for the increase in polymer flood
projects (Pope 2011). If the original reservoir conditions would require a more costly polymer then the polymer cost saving
associated with the low salinity injection water would increase significantly.
In the example, oil forecasts (watercut limit at 98%) have been generat ed using fractional flow theory for the following three
cases with identical polymer viscosity:
 High concentration/high salinity
 Low concentration/low salinity with same level of adsorption as in the high-salinity case (worst case)
 Low concentration/low s alinity with reduced adsorption
Parameters determining the fractional flow curves are given in Figure 7. The polymer concentration estimates for the seawater
and low salinity make-up brines to achieve the indicated polymer viscosity were obtained from correlation. Only adsorption
has been included in the retardation factor, hydrodynamic acceleration is ignored.

Figure 7: Left: Param eters for fractional flow curves and retardation factors . Right: Profiles of recovery and w atercut for high
concentration (top), low concentration (m iddle) and low concentration w ith reduced adsorption ( bottom )

oil visc mo 2 cP endpoint w Krw 0.2

brine visc mw 0.5 cP endpoint o Kro 1
polymer visc mp 5 cP Corey exp w nw 2
Corey exp o no 3
3
rock density rs 2645 kg/m
3
High polymer
solution density r w 1000 kg/m
concentration
porosity f 0.25

High concentration/high salinity

polymer conc Cp 2370 ppm in seawater
max adsorption Cs 25 mg/g
Retardation Aads 0.084
Low concentration/low salinity
polymer conc Cp 310 ppm in 1000 ppm NaCl
max adsorption Cs 25 mg/g Low polymer
Retardation Aads 0.640 concentration
Low concentration/low salinity with reduced adsorption
polymer conc Cp 310 ppm in 1000 ppm NaCl
max adsorption Cs 12.5 mg/g
Retardation Aads 0.320

Low polymer
concentration
+ reduced adsorption
IPTC 17342 11

The resulting values for the retardation factor are given in the table. For the low concentration/low salinity with reduced
adsorption it has been assumed that the adsorption will reduce to 50% of its value at high salinity based on (Martin 1983).
Forecast profiles in terms of recovery factor and watercut for the three cases are shown in Figure 7. This clearly demonstrates
the delay in oil recovery due to adsorption when reducing the polymer concentration; reducing the adsorption level reduces the
delay.

The economic model uses the following assumptions:

 Market forces are neglected i.e. Estimate Date Money and Real Terms money are the same
 Oil price is fixed in RT at 100 $/bbl,
 Polymer is HPAM with fixed price in RT at 3 $/kg
 “Open Book” discount rate of 10% is used
 Only polymer OPEX and oil production affect the economics comparison, i.e. the impact of low concentration/low
salinity on polymer CAPEX, and CAPEX and OPEX required for low salinity water supply have not been considered
here.

For this economic model, the change in NPV resulting from the use of a lower polymer concentration are defined as follows
∆𝑁𝑃𝑉 = 𝑂𝑖𝑙𝑝𝑟𝑖𝑐𝑒 (𝑅𝑇)( 𝑃𝑉 𝑃𝑟𝑜𝑑 𝐿𝐶 − 𝑃𝑉 𝑃𝑟𝑜𝑑 𝐻𝐶 ) − ( 𝑃𝑉 𝑂𝑝𝑒𝑥 𝐿𝐶 − 𝑃𝑉 𝑂𝑝𝑒𝑥 𝐻𝐶 ) {3}
The difference between the NPV in a low concentration case relative to that in the high concentration case is split into
contribution from oil delay (negative) and polymer OPEX saving (positive). Results are shown in Figure 8.

Figure 8: Change in NPV resulting from the use of a low polym er concentration show ing the
contribution of oil delay and polym er cost to the total change in a w orst case scenario (no
hydrodynam ic acceleration, no cost saving due to the use of a cheaper polym er type)
12 IPTC 17342

Low-Salinity Polymer: experimental workflow and results for a field case

The purpose of the experimental workflow for a feasibility study for low-salinity polymer flooding is threefold. First, pre-
selection of a potentially suitable polymer type, brine composition and polymer concentration is made. Then, as de-risking
activities, the effect of cation exchange, clay swelling, adsorption and hydrodynamic acceleration are tested and quantified in a
core flooding tests with detailed effluent analysis . In the same experiment, the in-situ rheology of the polymer can be
measured. Finally, the incremental oil recovery is determined in a series of core flooding displacement tests.

The possible brine compositions are chosen based on available data from the field: original field brine composition,
availability of low-salinity brine sources, options to desalinate these brine sources and dispose of the high -salinity reject
stream. The lowest possible salinity would give the highest polymer OPEX reduction. Lowering the salinity while keeping the
MAR constant, ensures a constant brine composition during the low salinity flood.

Next, polymer stability tests need to be done in the available brines (reservoir brine, low salinity brine) to determine the
available polymer types for each brine type. These stability tests are typically done at reservoir temperature in an oxygen -free
and metal-free environment, to avoid radical-catalyzed degradation (see e.g.(Vermolen 2011)). For the stable polymers, the
polymer concentration to obtain the required viscosity is determined by rheology measurements. The optimal viscosity level is
derived for the specific field from reservoir simulation studies.

In parallel to the stability tests, core-flooding experiments for de-risking the cation exchange effects are performed as
described extensively above. If the low-salinity brine of choice appears to give problems with clay swelling, mineral
dissolution or polymer viscosity loss, a higher salinity brine should be chosen and tested. Based on our experience, no
formation damage has been observed in core-floods at Sor, which is probably representative for the field. Therefore, the
stability of the polymer is the most important factor here.

The experimental procedure for determining the parameters of adsorption and hydrodynamic acceleration is described by
Hughes et al. (Hughes 1990). In a two-stage core flood, first the combined effect of adsorption and hydrodynamic acceleration
are measured and in the second stage, the hydrodynamic acceleration is measured separately, by which the ads orption term can
be derived. These parameters will feed into an economic evaluation for each brine/polymer type combination. While injecting
the polymer in this test, the polymer injectivity and in -situ rheology can be evaluated as well. Both are measured by
monitoring the differential pressure over the core (or preferably sections of the core) at varying flow rates. Since in the first
section of the core light plugging might occur by the high-end MW tail of the polymer solution (Seright, et al. 2010), the other
sections should give a more reliable value of the polymer in -situ rheology. The apparent viscosity of the polymer solution can
be derived from the differential pressure by Darcy’s law.

A few typical in-situ rheology curves are given in Figure 9 for four polymer solutions with different salinities at a reservoir
temperature of 50°C, which were injected sequentially in one reservoir core plug. The in-situ rheology measurement of the
high-salinity polymer was repeated in another core plug to test repeatability. The polymer solutions had different polymer
concentrations to match the bulk rheology at a shear rate of 2 sec-1 . As a general trend it can be seen that the low-salinity
polymer solutions give less shear thickening, which would be beneficial for injectivity (less viscosity at high shear rates). The
salinity-trend at lower shear rates is not so clear. Care has to be taken with the fact that the polymer solutions were injected in
series, so the permeability of the core probably reduced during the experiment (while the permeability at the start of the
experiment was taken to calculate the apparent viscosity). Also, the injection of low-salinity polymer after a high-salinity
polymer slug might give an extra permeability reduction because the dense layer of adsorbed high -salinity polymer swells in
the low-salinity brine.
IPTC 17342 13

Figure 9: In-situ rheology curves (at 50°C) for four polymer solutions with different salinities but similar bulk rheology (at a shear rate
of 2 sec-1) and one curve is a repeat of the high-salinity solution, m easured in a different field core.

In principle, the low-salinity polymer field modeling and economic evaluation of the polymer flood can be done without any
oil displacement experiment tests if the following is assumed: the polymer relative permeability is the same as the water
relative permeability (if the impact of the adsorbed polymer on the relative permeability is incorporated in the apparent
viscosity of the polymer) and no Sor reduction is expected by either a low-salinity effect or a visco-elastic effect. As input
parameters, the measured in-situ rheology, adsorption and hydrodynamic acceleration can be used. However, to check whether
there is a low-salinity effect or a visco-elastic effect in the field core material, oil displacement tes ts can be done. The
workflow for testing incremental oil recovery by the low-salinity effect is described in (Suijkerbuijk 2013). The incremental
oil recovery by low-salinity polymer with respect to a high-salinity polymer (the visco-elastic effect) can be determined by
injecting the high-salinity and low-salinity polymer solutions in series, making sure that the differential pressure over the core
stays the same (by lowering the polymer concentration). A constant (or decreasing) differential pressure is essential to
conclusively attribute the incremental recovery to the visco-elastic effect. An example of such a flood is given in Figure 10. In
this case, a reservoir core aged with reservoir crude was flooded (at reservoir tempe rature) with a high-salinity brine and
subsequently with a lower-salinity polymer solution and an even lower-salinity polymer solution of the same bulk viscosity.
All floods were done at the same rate. The incremental recovery in the third step (while keep ing the pressure the same) clearly
gives an indication that the low-salinity polymer does not only improve the polymer economics (by decreasing the required
polymer concentration), but also increases the recovery by either a low-salinity effect or a visco-elastic effect.
14 IPTC 17342

Figure 10: Pressure and production curve for a low -salinity polym er oil displacem ent experim ent in reservoir core m aterial.

Conclusions

Using a low-salinity make-up brine for polymer flooding will provide many benefits over traditional polymer flooding in the
high-salinity reservoir brine. Even when there will be no incremental oil recovery due to a wettability change, the project
economics could improve because the required polymer concentration decreases by a factor two to four. Also, the use of a
low-salinity make-up brine might allow the application of cheaper polymers, that normally would not have been stable in the
high-salinity brine. Additional benefits are the lower sensitivity of the low-salinity polymer to mechanical shear, less potential
production chemistry issues (scaling, souring, oil/water separation) and possibly an additional residual oil saturation reduc tion
by increasing the visco-elasticity of the polymer solution.

For successful application of low-salinity polymer flooding, several risks have been studied. The risk of cation exchange and
clay swelling needs to be tested in a core flood at residual oil saturation, because a core flood in a bare core (no oil) might
over-estimate the risk. By addition of the polymer to the low-salinity slug, the divalent cation fraction decreases less, which
lowers the risk of clay swelling.

Mixing of the low-salinity polymer slug with the high-salinity brine already present will not affect the polymer viscosity
because of a low-salinity slug without polymer will form in between the two fluids because of polymer adsorption and the high
salinity brine will be stably displaced by the low salinity polymer.

While polymer adsorption has a positive effect as a buffer against mixing, it could also cause a delay in oil recovery which
negatively affects the polymer flooding economics. For the quantification of the delay in oil recovery associated with a lower
polymer concentration it is crucial to acquire data on adsorption and hydrodynamic acceleration at original (high) salinity a nd
the aspired low salinity brine composition and do a full economics evaluation. In case the use of low salinity polymer would
lead to a delay in oil recovery, either the stability of the injected brine and/or the concentration of polymer can be optimized in
order to improve the economic viability of such a process.

An experimental workflow is described to de-risk the application of low-salinity polymer flooding and to derive all necessary
parameters for full field modeling and economic evaluation.

Acknowledgements
The authors would like to acknowledge all research technicians from the Shell Rijswijk Rock and Fluids team that contributed
to the work on Low-Salinity Polymer Flooding. Furthermore, we would like to thank Bart Suijkerbuijk for reviewing the
paper.
IPTC 17342 15

Bibliography
Appelo, C.A.J. "Multicomponent Ion exchange and Chromatography in Natural Systems." Reviews in Mineralogy, 1994: 192 -
227.
Barkman, J.H., A. Abrams, H.C.H. Darley, and H.J. and Hill. "An Oil Coating Process to Stabilize Clays in Fresh
Waterflooding Operations." SPE 4786, 1975.
Doe, P. H., Moradi-Araghi, A., Shaw, J. E., & Stahl, G. A. "Development an Evaluarion of EOR Polymers Suitable for Hostile
Environments - Part 1: Copolymers of Vinylpyrrolidone and Acrylamide." SPE Reservoir Engineering, SPE 14233, 1987:
461-467.
Gaillard, N., Giovannetti, B., Favero, C. "Improved Oil Recovery using Thermally and Chemically Protected Compositions
Based on co- and ter-polymers Containing Acrylamide." Paper SPE 129756 presented at the SPE Improved Oil Recovery
Symposium, Tulsa, Oklahoma, 24-28 April 2010 ., 2010.
Hughes, D.S., Teeuw, D., Cottrell, C.W. and Tollas, J.M. "Appraisal of the Use of Polymer Injection To Suppress Aquifer
Influx and To Improve Volumetric Sweep in a Viscous Oil Reservoir." SPE 17400, 1990.
Kozaki, Chie. "Efficiency of Low Salinity Polymer in Sandstone Cores." Master of Science Thesis, University of Texas,
Austin, 2012.
Levitt, D. B., Pope, G. A. "Selection and Screening of Polymers of Enhanced -Oil Recovery." Paper SPE 113845 presented at
the SPE/DOE Improved Oil Recovery Symposium, Tulsa, Texas, 19 -23 April 2008, 2008.
Martin, F.D., Hatch, M.J., Shepitka, J.S. and Ward, J.S. "Improved Water-Soluble Polymers for Enhanced Oil Recovery." SPE
11786, 1983.
Mungan, N. "Permeability Reduction Through Changes in pH and Salinity." Journal of Petroleum Technology Vol 17, Number
12, 1965: 1449-1453.
Mysels, K.J. Introduction to Colloid Chemistry. New York: Interscience Publishers, 1967.
Perkins, T.K., Johnston, O.C. "A review of diffusion and dispersion in porous media." Soc. Petr. Eng. J. Volume 3, Number 1,
1963: 70 – 84.
Pope, G.A. "Recent Developments and remaining Challenges of Enhanced Oil Recovery." Journal of Petroleum Technology,
July 2011: 65-68.
Pope, G.A. "The application of Fractional Flow Theory to Enhanced Oil Recovery." SPE 7660, 1980.
Scheuerman, R.F. and Bergersen, B.M. "Injection-Water Salinity, Formation Pretreatment, and Well-Operations Fluid-
Selection Guidelines." SPE 18461, 1989.
Scheuerman, R.F. and Bergersen, B.M. "Well Operations Brine Selection Based on Minimum Salinity Criteria." Report BRC
85-88, 1988.
Seright, R S, A Campbell, P Mozley, and R.S., Campbell, A., Mozley, P. Seright. "Stability of Partially hydrolyzed
Polyacrylamides at Elevated Temperatures in the Absence of Divalent Cations." Paper SPE 121460 presented at the SPE
International Symposium on Oilfield Chemistry, Woodlands, Texas, 20-22 April 2009 ., 2009.
Seright, R. S., T. Fan, K. Wavrik, and Carvalho Bala. "New Insights into Polymer Rheology in Porous Media." SPE 129200,
2010.
Shaker Shiran, B., Skauge, A. "Enhanced Oil Recovery (EOR) by Combined Low Salinity Water/ Polymer Flooding." Energy
& Fuels vol.27, 2013: 1223-1235.
Suijkerbuijk, B.M.J.M., Kuipers, H.P.C.E., van Kruijsdijk, C.P.J.W., Berg, S., van Winden, J.F., Ligthelm, D.J., Mahani, H.,
Pingo Almada, M., Van den Pol, E., Joekar Niasar, V., Romanuka, J., Vermolen, E.C.M., Al-Qarshubi, I.S.M. "The
Development of a Workflow to Improve Predictive Capability of Low Salinity Response." IPTC 17157, 2013.
Teeuw, D. and Hesselink, F.T. "Power-law Flow and Hydrodynamic Behaviour of Biopolymer Solutions in Porous Media."
SPE 8982 , 1980.
Todd, M.R., Longstaff, W.J.,. "The development, testing and application of a numerical simulator for predicting miscible floo d
performance." J. of Petr. Techn., 1972: 874 - 882.
Vermolen, E.C.M., van Haas terecht, M.J.T., Masalmeh, S.K., Faber, M.J., Boersma, D.M., Gruenenfelder, M. "Pushing the
envelope for polymer flooding towards high-temperature and high-salinity reservoirs with polyacrylamide based ter-
polymers." SPE141497, 2011.
Xia, H.F., D.M. Wang, G. Wang, W.G. Ma, and J. Liu. "Mechanism of the Effect of Micro-Forces on Residual Oil in
Chemical Flooding." SPE114335, 2008.
Zaitoun, A., Makakou, P., Blin, N., Al-Maamari, R.S., Al-Hasmi, A.R., Abdel-Goad, M., Al-Sharji, H.H. "Shear stability of
EOR Polymers." SPE 141113, 2011.

View publication stats