Polymer Flooding
Polymer Flooding
Polymer Flooding
This paper was prepared for presentation at the SPE Heavy and Extra Heavy Oil Conference - Latin America held in Medellin, Colombia, 24 –26 September 2014.
This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents
of the paper have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect
any position of the Society of Petroleum Engineers, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written
consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may
not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.
Abstract
Thermal EOR has long been considered the sole Enhanced Oil Recovery method for heavy oil but this is
no longer the case; several heavy oil polymer floods have proven successful and more are in the planning
stages. In the US alone several billion barrels of oil could be targeted; in the rest of the world and in Latin
America in particular the potential target is also probably large but mostly unknown at this point. Even
though polymer flooding recovery is usually lower than with thermal methods, it is less capital intensive
and may be the only economical solution for instance in thin reservoirs.
As any EOR project, polymer flooding of heavy oil is done in stages – screening, feasibility study, pilot
preparation, pilot execution and eventually full field deployment. Each of these stages requires care and
attention to details and many pitfalls need to be avoided in order to reach the final stage of full
deployment.
This paper intends to provide guidelines on the whole process, based on practical experience and
illustrated with actual field cases. This should allow operators to benefit from a better understanding of
the challenges and potential of polymer flooding of heavy oil and open the door for more projects.
Introduction
Polymer flooding is becoming recognized as an efficient and attractive process to increase recovery in
heavy oil reservoirs, following the successes of pilots and even field scale expansions in medium to high
viscosity oil in several countries.
In Oman, a polymer flood is now implemented at a large scale (27 patterns) in the Marmul field in an
oil with a viscosity of 90 cp (Al-Saadi et al. 2012, Jaspers et al. 2013).
In China in one field in Bohai Bay where the oil viscosity varies between 30 to 450 cp, polymer is
injected into 13 wells and both the oil rate and water-cut seem to have improved compared to waterflood
forecast (Kang et al. 2011).
In Canada the largest heavy oil polymer flood in existence in Pelican Lake has hundreds of injection
wells in an oil with a live oil viscosity ranging from 800 to over 10,000 cp (Delamaide et al. 2013,
Delamaide et al. 2014c). In Seal after a successful pilot (Delamaide et al. 2014a) the injection has been
extended to 14 wells and the live oil viscosity varies from 3,000 to 7,000 cp (Murphy Oil Company Ltd.
2014).
2 SPE-171105-MS
Successful smaller scale pilots are also ongoing or have been completed in Argentina (6 injection and
3 production wells in an oil with a live oil viscosity of 160 – 300 cp, (Hryc et al. 2013)), Suriname (3
injection and 9 production wells in an oil with an apparent live oil viscosity of 300 to 600 cp, (Moe Soe
Let et al. 2012), the North Sea (Etebar 1997) and Canada (Wassmuth et al. 2009, Liu et al. 2012,
Delamaide et al. 2014a). The data for the documented field tests are summarized in Table 1 to Table 4.
Polymer is composed of water-soluble molecules which increase water viscosity; it is used for mobility
control i.e. improving the Mobility Ratio between the water and the oil and can improve sweep efficiency.
Polymer injection can increase reservoir pressure more effectively than waterflooding especially in
viscous oil because it reduces water-cut and the need to recycle numerous pore volumes through the
reservoir. This was evidenced in Pelican Lake by the comparison of the water-cut in the waterflood and
the polymer flood pilots (Delamaide et al. 2014c). For light to medium oil the target is usually to achieve
a Mobility Ratio of 1 (Green and Willhite 1998) but in heavy oil this is not always possible nor advisable.
Polymer flooding should be viewed as an improved waterflood.
Polymer flooding cannot compete with steam injection (steam flood or SAGD) in terms of ultimate
recovery and cannot be used when the oil is not mobile (such as in the case of bitumen) but it can be a
good alternative in cases where steam injection is not applicable or too expensive. In particular steam
injection cannot be used in thin or deep reservoirs because of the prohibitive heat losses that this entails
while polymer flood is not as restricted in that regard. Steam injection is also capital intensive because
SPE-171105-MS 3
clean water is required to go through the boilers for steam generation while polymer flooding can use
much lower quality water.
Polymer flooding of heavy oil is in many ways similar to polymer flooding in any kind of reservoir,
but it also differs in several key aspects. Although numerous papers have been published on individual
field tests or specific aspects of the process, there is no single paper addressing all the issues for a polymer
flood project in heavy oil from screening to full field extension and this is what this paper will provide.
All the existing heavy oil polymer flood field applications will not be reviewed in details but some of the
best documented field tests will be used throughout the paper to illustrate some specific points.
We will successively address the following aspects which constitute the workflow for a polymer
flooding project:
4 SPE-171105-MS
● Screening;
● Feasibility;
● Pilot design;
● Pilot implementation and surveillance;
● Field extension.
The screening phase consists in checking the reservoir properties and verifying that they are adequate
for polymer injection.
When some parameters are borderline, some laboratory tests or simplified reservoir simulations can be
performed during the Feasibility phase to clarify key uncertainties but this is not always required.
Pilot design consists in selecting the pilot area and preparing the test itself using laboratory as well as
reservoir simulations. This phase can include drilling new wells or logging existing ones or using tracers
to better establish flow paths between wells. Existing data are analyzed together with newly acquired ones
and reservoir simulations are often performed first to history match the pilot area and then to forecast the
future behavior of the pilot. At the same time field work is taking place – facilities construction in
particular for the polymer injection skids.
In the simplest cases the pilot design phase can be accomplished within a year; the timing is mostly
dictated by the field operations in particular the procurement of the polymer mixing facilities, regulatory
approval etc.
Screening
Screening is the first step on the long journey to full field extension, and is an important one. Selecting
the wrong reservoir can have dire consequences in terms of resources and capital spent.
Screening guidelines cover most EOR processes (Taber et al. 1997) but even in the most recent
guidelines published (Saleh et al. 2014) the maximum oil viscosity recommended for polymer flooding
is only 130 cp. This obviously falls far short of the mark. We will thus review the screening process and
the guidelines for polymer flooding in heavy oil. As with any EOR project, the use of guidelines should
not replace a proper screening study.
Screening tools
Screening tools from the simplest to the most complicated can be used by non-experts to perform EOR
screening and sometimes to forecast production and recovery (Zerafat et al. 2011, Parada et al. 2012,
Lefebvre et al. 2012, Bang 2013, Zijlstra et al. 2014). Such tools are sometimes based on standard
screening criteria (Taber et al. 1997) but can incorporate reservoir simulations or complex correlations.
They are also usually proprietary.
Even sophisticated tools do not handle nuances such as detailed water composition or complex geology
but rather focus on estimating the sweep efficiency and recovery using standard polymer properties. In
addition they usually consider parameters separately which is not adequate in many cases. For instance
hard water may not be a problem at low temperature but will be at higher temperature; high viscosity and
low permeability are both acceptable separately but not when they are combined. As such these screening
tools should only be used to assist in doing some basic scoping economics and should not replace an
expert’s opinion.
Other simple analytical tools based on Buckley-Leverett adapted to polymer may also be used at the
screening stage to have a rough estimate of potential improvement due to polymer as described by Green
and Willhite (Green and Willhite 1998); however as for the most complex tools their use is limited by the
need to use “standard” polymer data for the specific reservoir considered and by the simplicity of the
reservoir description (homogeneous or multi-layer).
SPE-171105-MS 5
Screening parameters
Temperature Although heavy oil is often found in relatively shallow and thus relatively low temperature
reservoirs (such as in Canada, California and some parts of Venezuela) this is not always the case in
particular in Colombia.
It is difficult to consider temperature in isolation because of the interactions between temperature and
water composition. Still, for guidelines purposes it must be noted that temperature can be a severe
limitation to polymer flooding because the most commonly used polymer – hydrolyzed polyacrylamide
or HPAM – is limited to temperatures below 70°C. At higher temperature, the functional groups of the
polymer molecules change in particular through hydrolysis which leads to degradation and reduction in
viscosity (Gaillard et al. 2010); this effect varies in magnitude depending on the temperature and water
composition. It is of course critical that polymer retains its viscosity during its transit in the reservoir,
which can take several months to several years. However it must be noted that HPAM can remain stable
at high temperature (⬎100°C) in the absence of divalent ions (Seright et al. 2010) and that at least one
successful polymer flood using HPAM has been reported at a temperature of 99°C (Greaves et al. 1984)
albeit with fresh injection water.
Special polymers combining acrylamide with acrylic acid (AA), acrylamide-methyl propane sulfonates
(AMPS), acrylamido-tertiary-butyl sulfonate (ATBS) and N-vinyl-pyrrolidone (NPV) have been devel-
oped for higher temperatures (Audibert et al. 1995, Gaillard et al. 2010, Vermolen et al. 2011, Gaillard
et al. 2014). These polymers are termed co- or ter-polymers depending on how many components are
used. Commercial products such as ter-polymers combining acrylamide, ATBS and NPV now exist for
temperatures up to 120°C although they are much more expensive than regular HPAM (Gaillard et al.
2014). We are however not aware of any field test of these polymers reported in the literature.
Biopolymers such as xanthan and scleroglucan have also long been considered to tackle slightly higher
temperatures and salinity (Rivenq et al. 1992, Audibert et al. 1993) and they have been found to exhibit
good resistance up to 90°C; however these polymers are very sensitive to biodegradation as evidenced by
some field tests (Bragg et al. 1983, Norton et al. 1986).
Another biopolymer called Schizophyllan is currently being tested on the Bockstedt field in Germany
(Leonhardt et al. 2014, Ogezi et al. 2014); the polymer is reported to have a high resistance to temperature
up to 135°C, high tolerance to salinity and good viscosifying properties. However the reservoir temper-
ature is only 54°C so the test will not provide much information on that aspect.
Water composition As mentioned above polymer is extremely sensitive to water composition, in
particular salinity and divalent cations. Both HPAM and xanthan are composed of long molecular chains
with charges located along the chain and these charges interact with ions in solution. The most important
polymer property – viscosity – is caused by the interaction between the different molecules as well as
between the molecules and the porous medium. In low salinity environment, the molecular chains expand
and thus have more interactions and as a result polymer viscosity increases; the reverse is also true. This
means that in higher salinity environments higher concentrations of polymer are necessary to achieve a
similar viscosity. HPAM is also affected by divalent cations and can precipitate at high temperature/high
concentrations of divalent cations (Moradi-Araghi et al. 1987, Levitt et al. 2011).
Biopolymers have more rigid structures and thus are less affected by salinity (Audibert et al. 1993); this
is why they are often considered for high salinity reservoirs.
The co- and ter-polymers described above are also resistant to high salinity and high hardness (Gaillard
et al. 2014) however at high salinity (70 g/L in the cited paper) high polymer concentration of 2,000 to
3,000 ppm are still required to achieve moderate polymer viscosity (10 cp) at low temperature. Thus the
economics of polymer flooding at high temperature and high salinity remain questionable.
HPAM is also highly sensitive to the presence of iron (Fe2⫹) in water in the presence of oxygen. A
recent paper (Seright et al. 2014) shows that when dissolved oxygen concentration is over 200 ppb, even
6 SPE-171105-MS
a few ppm of Fe2⫹ can lead to loss of viscosity even at low temperature. When no Fe2⫹ is present, the
presence of oxygen is not an issue at low temperature but can lead to polymer degradation at higher
temperature.
Finally HPAM can also be degraded by H2S when oxygen is present (Yang et al. 1985, Gaillard et al.
2014).
Associative polymers are another family of polymer composed of hydrolyzed polyacrylamide with
small amounts of hydrophobic monomers which can associate with each other. These polymers are known
to have good rheological properties in high salinity (Dupuis et al. 2011) but high adsorption. Their
behavior is not yet well understood (Dupuis et al. 2012, Seright et al. 2011). They have seen little use in
field projects so far but in heavy oil they have been used successfully in the Mooney field in Canada
(Watson et al. 2014), in Phase 2 at Bodo (Renouf 2014) and in Bohai Bay (Kang et al. 2011).
Oil viscosity As mentioned above, existing screening guidelines need to be updated to take into account
the latest successes obtained in particular in Canada. Field results obtained to date suggest that maximum
acceptable live oil viscosity could be as high as 5,000 cp (Delamaide et al. 2013, Delamaide et al. 2014c,
Renouf 2014). Polymer injection has been successfully tested in viscosities as high as 10,000 cp
(Delamaide et al. 2014c) however the efficiency of the process seems to diminish as viscosity increases
and in particular water-cut can increase to high values even with polymer.
Cenovus is also testing the use of hot water in combination with polymer to extend the range of suitable
viscosity (Cenovus Energy 2013) and the production response appears positive but there is no information
on the economics of the process available.
Permeability Permeability is an important screening criteria in particular with heavy oil. Although
polymer can be successfully injected into low permeability reservoirs (Delamaide et al. 2014b), for high
viscosity oil, a high, darcy-level permeability is required in order to ensure sufficient injectivity. That is
the case in all the heavy oil polymer flood field tests reported in the literature (Table 2).
Lithology To the best of our knowledge, all heavy oil polymer floods have been performed in sandstone
formations because of the high permeability required although high permeability carbonates could also be
potential targets provided they are not fractured.
The presence of clays increases polymer adsorption thus can be detrimental to economics however this
is not considered to be a killing factor at the screening stage.
The presence of swelling clay is usually not an issue provided that formation water is used for injection
or that injection water salinity is not too low; however it is always recommended to ensure the
compatibility of injected water with both formation water and rock when planning to use low salinity
water (Vermolen et al. 2014a).
Geology/heterogeneity Heterogeneity plays a major role in flooding processes because it affects the
sweep efficiency: it is well known in reservoir engineering that sweep efficiency decreases when
heterogeneity increases (Craig. Jr. 1993). Polymer flooding is known to improve sweep efficiency by
reducing the mobility of the water.
It goes without saying that as for any type of flood, good continuity is required between injection and
production wells so reservoirs with high degrees of faulting for instance are generally not good candidates.
Polymer flooding can work in single or multi-layer reservoirs but will require adapted well architec-
tures. Horizontal wells are well adapted to thin, mono-layer reservoirs and vertical wells are better suited
to multi-layer reservoirs.
In the case of fractured carbonates containing heavy oil in matrix blocks, the situation is quite complex
and depends on the matrix and fracture properties. Although polymer should help increase viscous forces
and thus increase recovery from the matrix, this incremental recovery would probably be quite low.
Polymers with low molecular weight could also enter the matrix but high concentrations of polymer would
SPE-171105-MS 7
be required to generate sufficient viscosity to displace the oil. In addition natural imbibition appears lower
for polymer flood compared to waterflood (Doorwar et al. 2014). There has been no field test in heavy
oil so far and thus polymer flooding of heavy oil in fractured carbonates is not recommended at this point.
Depth There is no depth limitation for polymer flooding, unlike with steam injection. However
injectivity can be limited in shallow reservoirs due to limitations in maximum injection pressure while
high temperature in deep reservoirs is not favorable for polymer.
Reservoir thickness Reservoir thickness is not a limiting criteria for polymer flooding; polymer flood
has proven highly successful in thin reservoirs such as Pelican Lake (5 m) and there is no reason why they
would not perform equally well in much thicker reservoirs although such reservoirs are usually more
suited to thermal methods such as steamflood or Steam Assisted Gravity Drainage (SAGD) which are
expected to achieve higher recovery. However polymer flooding is much less capital intensive than steam
injection because water quality requirements are not as strict thus polymer flooding may remain an
attractive option for thick reservoirs.
Mobile oil saturation Oil saturation itself is not important; what matters is the amount of mobile oil
present in the reservoir – that is the oil saturation above the residual oil saturation Sor.
There is currently a great deal of controversy among experts as to whether polymer flooding can reduce
Sor. Since Stegmeier’s work (Stegemeier 1977) it has generally been accepted that reducing oil saturation
below Sor can only happen by increasing the capillary number by several orders of magnitude. The
capillary number can be defined in various ways but here we will define it as:
With:
⫽ velocity (m/s)
⫽ viscosity (cp)
⫽ interfacial tension (mN/m)
As velocity and viscosity of the injected fluid can only be increased by so much in practice, a
significant reduction of the Sor typically cannot happen unless the interfacial tension can be reduced
through the use of surfactant.
However some of the significant work performed for the large polymer flood in Daqing (China)
suggests that polymer can reduce oil saturation below Sor due to its viscoelastic properties (Wang et al.
2000, Wang et al. 2001a, Wang et al. 2001b) at least for low viscosity oil (9 cp for Daqing).
A recent paper (Vermolen et al. 2014b) summarizing the work of researchers on this issue and
presenting some additional lab experiments suggests that with high viscosity oil residual oil saturation
cannot be decreased.
In this paper we consider that polymer does not reduce oil saturation below Sor in viscous oil and that
this effect should not be relied upon when deciding whether or not to implement a polymer flood. If it
exists it will be an added bonus.
As a consequence polymer flooding requires a high mobile oil saturation to be economic, in most cases
over 30%.
Remark: It is often said that polymer flooding increases recovery compared to waterflooding, but in
fact it is more proper to say that polymer flooding accelerates recovery. As residual oil saturation is not
reduced, an equivalent recovery could in principle be achieved through the injection of large numbers of
pore volumes of water, although in most cases this cannot be achieved within a reasonable time scale.
Presence of bottom aquifer The presence of a bottom aquifer is a key detrimental factor for polymer
flooding and chemical EOR in general. The issue is that polymer will be lost into the aquifer and oil can
also be pushed into the aquifer with a significant negative impact on the economics of the process.
8 SPE-171105-MS
Several authors have studied this issue when evaluating pilots in reservoirs with bottom aquifer and
have come to various conclusions.
Al Azri (Al Azri et al. 2010) studied polymer injection in the Nimr field in Oman, in a 30 m to 50 m
thick heavy oil reservoir (viscosity 250 – 500 cp) underlain by an active aquifer. They found that
horizontal wells in combination with polymer injection could improve recovery but we suspect that this
is largely influenced by the very short spacing (43 m) between injection and production wells. A field trial
appears to be in progress (Al Abri et al. 2014) but results have not been published.
Mukherjee (Mukherjee et al. 2014) discussed a thin (22 m) heavy oil reservoir (291 cp) underlain by
a large aquifer. Using reservoir simulations they found limited incremental recovery using horizontal
wells and polymer (less than 5% OOIP) and a loss of over 60% of the polymer injected into the aquifer
along with between 5 to 10% of the oil present in the reservoir (investigated spacing was 50 m to 150 m).
They also suggested that better results could be achieved by a combination of hot water injection together
with polymer.
Li (Li et al. 2014) describes some laboratory tests and reservoir simulations for the Bentley field
offshore UK, which contains a high viscosity (1,500 cp) oil with a 40 m oil column and a large bottom
aquifer. Using horizontal producers and injectors separated by 60 m only (laterally) they found that
polymer injection could result in some incremental recovery of 48% of primary production which seems
to correspond to 2.1% OOIP.
Most of these works have been performed on simple homogeneous models however the performances
of the process are probably quite dependent on the reservoir heterogeneity (presence of barriers, vertical
permeability etc.)
As there has been no successful pilot implemented in such a reservoir to date we consider that the
presence of a bottom aquifer adds significant risk to a project and requires some modelling work to be
performed during the Feasibility phase in order to ascertain the economics of the process.
Presence of gas cap Although it is not a killing factor, the presence of a gas-cap is generally not
beneficial to chemical EOR. One of the main benefits of polymer flooding – as of waterflooding – is the
effect of pressure support on production; because of the high gas compressibility a gas-cap will reduce the
pressure effect resulting from the chemical injection and thus delay the response.
Reservoir dip In heavy oil reservoirs the density of oil and water are close, thus the effect of gravity is
very limited. As a result reservoir dip is not expected to have a significant impact on polymer flooding.
Waterflooded reservoirs The existence of a prior waterflood is not an issue for polymer flooding; indeed
most polymer floods occur in waterflooded reservoirs. The more important question is that of the mobile
oil saturation (already discussed). In Pelican Lake the polymer flood was implemented in some areas of
the reservoir that had already been waterflooded with no apparent detrimental effect on the recovery,
although the process is less efficient in terms of volumes of water injected (Delamaide et al. 2014c). The
polymer flood in Marmul (Jaspers et al. 2013) also followed an extensive waterflood period.
In heavy oil reservoirs it is also not uncommon to have interstitial water production very early due to
the mobility contrast between oil and water (as in the case in Pelican Lake and Tambaredjo). This is not
a problem either.
Special case of offshore operations Additional issues such as availability of platform space for the
mixing equipment and challenges in polymer supply and storage need to be addressed for offshore
operations. These issues can be addressed either at the screening or at the feasibility stage.
Summary In summary the key criteria for polymer flooding in heavy oil are the oil viscosity, the
reservoir temperature, the water composition and hardness, the absence of bottom aquifer, mobile oil and
the lithology.
SPE-171105-MS 9
Feasibility
Depending on reservoir conditions, a Feasibility phase may or may not be necessary. As its name
indicates, the purpose of this phase is to identify key uncertainties and challenges and ascertain whether
they can be overcome in order for the project to be feasible. Typical uncertainties are reservoir
heterogeneity, mobile oil remaining in the reservoir, presence of swelling clays or simply economics.
In cases where reservoir conditions (temperature, water composition) are difficult the bulk of the work
will consist in identifying what polymer could be used and at what concentration in order to perform
simple economics. When reservoir conditions are mild but the reservoir structure or heterogeneity is
challenging – such as when a bottom aquifer is present – the work should focus on reservoir simulations
using standard polymer properties in order to identify whether recovery and polymer losses to the aquifer
would be reasonable and the economics of the project still justify moving forward.
Reservoir simulations can be performed either on simplified phenomenological models, on cross
sections incorporating key reservoir features or on sector models extracted from larger full field models
if they exist (Lefebvre et al. 2012). Production data should be history matched if available.
In simple cases – mild reservoir conditions, limited heterogeneity – the feasibility phase may be
skipped altogether.
In some cases further field testing such as tracer tests – single well to ascertain oil saturation or
interwell to verify communication – or injectivity tests may be recommended.
In heavy oil reservoirs particularly when the viscosity is very high polymer injection is often
considered before waterflood has been implemented and in that case important parameters such as
interwell connectivity and sweep efficiency are still unknown. In such cases tracer tests are not feasible
and history matching production will not be very useful thus the pilot will be the main source of
information.
Injectivity
One of the main questions that arises when evaluating the feasibility of a polymer flood is the issue of
injectivity. Because of the high viscosity of polymer compared with that of water, it would appear
reasonable to expect a severe reduction of injectivity. However this is not what has been observed in field
operations.
In Pelican Lake injectivity decline from waterflood to polymer flood is approximately 25% (the wells
are horizontal) and there has been no reported injectivity issue. Newly drilled injection wells are usually
produced for a few months prior to initiating injection in order to reduce the pressure around the wellbore
and improve injectivity (Delamaide et al. 2014c).
In Tambaredjo where the wells are vertical injectivity decreased by 30% to 40% until hydraulic
fractures were initiated by the injection of the polymer solution (Manichand et al. 2013).
In Bodo, Suffield and Corcobo Norte, injectivity also declined by 20% to 30% with polymer compared
to waterflood (Delamaide 2014, Liu et al. 2012, Hryc et al. 2013).
In Marmul no injectivity issue has been reported (Jaspers et al. 2013) but most injection wells are
fractured (Al-Saadi et al. 2012).
As discussed earlier, polymer flooding of heavy oil is usually conducted in high permeability reservoirs
which helps injectivity. Horizontal wells – used in some of the field tests – also allow higher injectivity
compared to vertical wells. In vertical wells, injectivity can be enhanced by fractures or pseudofractures
(Fletcher et al. 1992, Seright et al. 2009, Khodaverdian et al. 2010, Clemens et al. 2013) as in Tambaredjo;
the risk however is that fractures could extend too far and connect injection and production wells. The risk
of creating a filter cake which could result in a reduction of injectivity and/or mechanical degradation of
the polymer also exists; thus the good filterability of the polymer needs to be checked in the lab.
In summary injectivity is a question that needs to be considered but field experience suggests that this
has not been a significant issue.
10 SPE-171105-MS
Pilot design
Pilot design starts with the selection of the pilot pattern and usually consists of three parts: lab, reservoir
simulation and field operations. Before addressing the pilot design a few common issues need to be
discussed.
Pilot economics
Although it is not always easy to convince management, the pilot phase is not the time to scrimp on
expenses; doing a pilot “on the cheap” almost guarantees that the full objectives of a pilot – allowing to
save money in the future field extension – will not be reached. A failed pilot or a pilot which cannot be
interpreted properly by lack of necessary data could lead to rejecting a valid EOR method by error.
A pilot is not designed to make money although it does not mean that it should be designed to lose
money! One must think of a pilot as some kind of exploration well: most exploration wells are dry – this
is a risk accepted by management – and exploration wells seldom make money; the return on investment
comes from the future field development.
That being said, the better the economics of a pilot the easier it will be to justify extending the process
to the rest of the field.
Secondary or tertiary
A question that often arises – in particular for polymer floods in heavy oil – in fields which have not been
waterflooded is whether it is better to inject polymer immediately or start by a period of waterflood.
Displacement of viscous oil by water is not very efficient and could lead to early water breakthrough;
on the other hand establishing a waterflood baseline would allow to better estimate the benefits of polymer
injection.
Some researchers have suggested based on coreflood experiments that polymer flooding viscous oil
can result in higher recovery in tertiary conditions (after waterflood) than in secondary conditions (Fabbri
et al. 2014, Doorwar et al. 2014). However these results need to be explained and confirmed by further
work.
Actual field experience in heavy oil is not very conclusive either. As described by Delamaide et al
(Delamaide et al. 2014c) in Pelican Lake polymer injection has usually been conducted as a secondary or
even as a primary process; however some wells were converted to polymer injection following a period
of water injection. Comparing the cumulative oil production in the three cases does not reveal any
significant difference except for the fact that recovery is faster for secondary polymer flood compared to
tertiary (Figure 1). In Marmul the field had also been waterflooded prior to the injection of polymer and
yet the process was successful.
We can conclude that establishing a baseline of water injection in the pilot can allow to better quantify
the economic benefits of polymer injection and does not appear detrimental to the recovery; however it
is by no means indispensable.
Another solution if capital is available would be to operate simultaneously two pilots nearby, one for
waterflood and one for polymer flood in order to compare.
SPE-171105-MS 11
Figure 1—Comparing primary, secondary and tertiary polymer injection (modified from Delamaide et al. 2014c)
Choice of pattern
The choice of the pilot pattern is an important one and needs to be considered carefully. A number of
criteria need to be taken into account. One thing that should be obvious from the Screening section is that
polymer injection wells should not be placed in an aquifer, contrary to the usual practice for waterfloods.
Representativity One of the main criteria for pattern selection is that it should be representative of the
reservoir. It is an obvious point which may however sometimes be overlooked. As stated earlier, the aims
of a pilot are:
1. To verify by a field test the viability of the process;
2. To reduce risks;
3. To gather sufficient data in order to improve the process for the full field deployment.
If the pilot is performed in an area which is very different from the rest of the field these aims may not
be reached.
Local knowledge of reservoir It goes without saying that it is recommended to select a pattern where
the geology is well understood and where there is enough well data available, in particular cores and logs
as well as a long production history. This will improve the interpretation of the pilot.
Well type The well type usually depends on how the field has been developed but the pilot may test a
different well configuration. Depending on the geology and thickness of the reservoir vertical or
horizontal wells or a combination could be used. The main criteria are to ensure sufficient injectivity and
also a good sweep efficiency (vertical in particular).
In the case of heavy oil density difference between oil and water is small so the impact of gravity will
be minor in particular when the well spacing is small and/or when the vertical permeability is low.
12 SPE-171105-MS
Geology also plays a role: the use of horizontal wells may be difficult in highly stratified reservoirs
with non communicating layers.
Spacing Well spacing should not be too large in order to reduce the response time. The distance between
injection and production wells used in most heavy oil field polymer floods (Table 4) varies between 70
to 300 m. To ensure a quick response in the pilot we recommend to remain below 150 m if possible.
Pattern type The number of wells can vary but in general it is better to have at least two injection wells
and several production wells in order to reduce the risks of something going wrong in one of the wells
and introducing some bias in the pilot. It is also better to have at least one confined well because this is
more representative of the future field extension and will allow more accurate recovery calculations.
With horizontal wells, five parallel wells (2 injectors and 3 producers) such as used in Pelican Lake,
Mooney and Seal (Delamaide et al. 2014a) are a good pattern layout. For vertical wells, regular 5-spot
patterns for instance have the advantage of a confined central producer which allows to estimate recovery.
On the other hand a significant part of the polymer will be lost to the outside of the pattern thus it would
be useful to monitor the surrounding producers if any. An inverted 5-spot pattern is unconfined thus
recovery is more difficult to estimate but polymer losses will be reduced. It is better in those two cases
to use several patterns such as was done in Tambaredjo (Moe Soe Let et al. 2012). The reader is referred
to Teletzke et al. (Teletzke et al. 2010) for a more detailed discussion on pilot best practices.
Infrastructure The presence or absence of infrastructure should not be the main criteria when selecting
a pilot location but nevertheless it needs to be taken into account to reduce capital expenditures. Polymer
flooding requires water thus the proximity of a source of water is clearly an advantage.
The presence of potential monitoring wells (suspended wells for instance) can be used to the pilot’s
advantage to monitor the advance of the polymer front and the oil bank, although this is not indispensable.
Laboratory work
The aim of the laboratory work – either during the Feasibility or the pilot Design phase – is mostly to
determine the most suitable polymer, and also to obtain data on polymer behavior in the reservoir for the
reservoir simulation.
The selected polymer needs to remain stable at reservoir conditions (temperature, hardness) for long
periods and also to develop required viscosity at a minimum cost without plugging the formation. The type
of polymer is thus selected depending on reservoir temperature and water composition as discussed in the
Screening section. Its molecular weight is selected depending on the reservoir permeability (higher
permeability allows for higher molecular weight); the polymer with the highest molecular weight that will
not plug the formation is usually selected as it will develop the highest viscosity at the lowest cost. A
compromise is often required.
Polymer properties are different in bulk and in cores because of the interaction between the long
polymer molecules and the porous medium thus the rheology of the polymer solution is determined in
bulk and in corefloods.
Water and oil samples Because of the sensitivity of polymer to the water composition, the water used
for the work needs to be the same as the one that will be used for the pilot. When a water other than
formation water is used for the preparation of the polymer solution the performances of the polymer will
have to be evaluated with both the injection water and a blend of injection and reservoir water. It is also
necessary to check the compatibility of this water with the reservoir rocks for potential issues such as clay
swelling.
It is extremely important to obtain representative water samples; one common issue that happened both
in the first polymer flood pilot in Pelican Lake (Delamaide et al. 2013) and in Bodo (Wassmuth et al.
2009) is that if the water samples get in contact with oxygen and if iron is present in the water, it can
precipitate and remain undiscovered. As discussed in the Screening section, polymer is highly sensitive
SPE-171105-MS 13
Figure 2—Example of viscosity vs. shear rate plot – Reproduced from (Thomas et al. 2012)
to iron and that needs to be taken into account when designing the facility. In both Pelican Lake and Bodo
the issue was discovered during the pilot start-up and then changes had to be made on the spot to ensure
that acceptable polymer solutions could be produced. Thus it is highly recommended to ensure that water
samples are collected in containers filled to the brim and sealed properly so that they never get into contact
with air.
Oil samples will also be required; ideally these samples should be free of additives (production
chemicals) and the oil should be filtered to remove particles and impurities before being used for the
corefloods.
Bulk tests The bulk tests aim at determining the properties of various selected polymers in the relevant
(reservoir, injection, mixture) brines at various concentrations and shear rates. Polymer solutions are
non-Newtonian fluids and as such their viscosity varies with the shear rate. The tests are done at reservoir
temperature.
Typical results include plots such as the one in Figure 2.
This preliminary work allows:
1. To select the polymer which has the best viscosifying properties in the selected brines;
2. To determine the target concentrations to inject in the corefloods; this is done by estimating the
shear rate in the cores corresponding to standard displacement velocities in the reservoir (typically
1 ft/d) and the core properties (porosity, permeability) through the use of a capillary bundle model.
For more information on that aspect the reader is referred to the work of Chauveteau (Chauveteau
1981).
Other tests can also be performed: filterability tests on membranes to evaluate the risks of formation
damage, thermal and chemical stability of the polymer solution (done in particular for high temperature/
high hardness conditions).
Coreflood data Corefloods are performed with the selected polymers to obtain data on their behavior in
the reservoir. Polymer solutions are non-Newtonian fluids, meaning that their rheology depends on the
14 SPE-171105-MS
shear rate thus on the porosity and permeability of the reservoir rocks. Thus corefloods have to be
performed in representative reservoir plugs; when this is not possible for instance in unconsolidated
formations where no core is available, testing can be performed on sandpacks of equivalent properties
(grain size, mineralogy, porosity and permeability). Ideally plug samples should be selected to represent
the high, medium and low range of permeability in the reservoir in order to evaluate the polymer behavior
and suitability in various conditions. A high molecular weight polymer could be well adapted to the
highest permeability but not to the average or low permeability of the reservoir. The aim of a polymer
flood is not for the polymer to enter only in the high permeability intervals in order to plug them (this is
what conformance control is more about) but rather to control the mobility of the water; to do so it must
be able to penetrate in most of the reservoir.
As discussed earlier polymer does not reduce Sor thus does not increase recovery; incremental recovery
observed in corefloods is due to the improvement of the mobility ratio thus measuring it in the lab is not
necessary.
The parameters that need to be measured in the corefloods are:
● Water-oil relative permeabilities;
● Resistance Factor RF (also called Mobility Reduction Rm) as a function of polymer concentration
and shear rate;
● Residual Resistance Factor RRF (also called Permeability Reduction Rk) as a function of polymer
concentration;
● Polymer retention;
● Inaccessible Pore Volume (IPV).
Resistance Factor The Resistance Factor is an adimensional parameter that corresponds to the
apparent viscosity of the polymer solution in the porous medium relative to that of water. It is calculated
as the ratio of the pressure drop ⌬Pp during the injection of the polymer solution divided by the pressure
drop ⌬Pw during the injection of water, before the injection of polymer:
RF depends on the characteristics of the polymer – molecular weight – as well as of the reservoir rock
– permeability, pore size – thus RF and the viscosity of the polymer solution can differ significantly. Two
polymers of equal viscosity but different molecular weight could have vastly different RF. RF also
depends on the shear rate: in porous media polymer solutions can have a shear-thinning behavior
(viscosity decreases as shear increases) at low shear rate as well as a shear-thickening behavior (viscosity
increases as shear increases) at high shear rate (Green and Willhite 1998).
The Resistance Factor is obviously an important parameter which will allow to decide at which
concentration the polymer solution needs to be injected in order to achieve the target mobility reduction.
Residual Resistance Factor The Residual Resistance Factor corresponds to the permeability reduc-
tion caused by the adsorption of the polymer on the rock which results in a smaller cross-section for the
fluids to pass through. It is measured as the ratio of the pressure drop in the core before the polymer
injection has taken place to the pressure drop after. In high permeability formations RRF can be low (close
to 1); values in the 2-4 range are common but higher values suggest that the polymer molecules may be
too large for the formation.
Figure 3 presents a typical plot of RF and RRF vs. shear rate.
Polymer retention There are mainly 3 mechanisms responsible for the retention of polymer in porous
media (Sorbie, 1991, Green and Willhite 1998):
● Adsorption;
● Mechanical entrapment;
SPE-171105-MS 15
Figure 3—Example of RF (Rm) and RRF (Rk) plot – Reproduced from (Delamaide et al. 2014c)
● Hydrodynamic retention.
Adsorption, the most prevalent mechanism in polymer retention, is caused by physical interactions
between polymer chains and rock surfaces, such as Van der Waals attraction, electrostatic interaction, and
hydrogen bonding. For the main polymers used in polymer flooding applications, adsorption is generally
considered as an irreversible process. It is also an important parameter because adsorbed polymer will not
contribute to the recovery. Adsorption can be measured in static conditions (on a volume of crushed rock)
but the values obtained are usually different from those measured in corefloods (dynamic adsorption).
Polymer adsorption is higher at high concentration (Zhang et al. 2014) and depends mostly on the
mineralogy of the reservoir because polymer adsorbs preferentially on clay materials. In corefloods values
vary from less than 10 g/g to up to 200 g/g (Green and Willhite 1998, Manichand et al. 2014) but
everything over 50 g/g is considered to be high. Manichand et al. (Manichand et al. 2014) provide an
excellent review of polymer adsorption and compare the field values estimated for the Tambaredjo pilot
to those measured in the lab. Surprisingly the field estimates are exceedingly high (up to 250 g/g) but
this could be due to the high kaolinite content in the reservoir or to a problem with the calculation method.
Mechanical trapping occurs when polymer molecules are trapped in small pore throats and is normally
low when reservoir permeability is high.
Hydrodynamic retention is similar to mechanical entrapment, but is a flow rate dependent effect,
meaning that polymer retention increases with the velocity and it is reversible. This mechanism is not well
understood and is not believed to be a large contributor in field-scale polymer floods (Sorbie, 1991, Green
and Willhite, 1998).
Polymer retention will impact the economics and also has the practical effect of delaying the
progression of the polymer and the oil bank into the reservoir.
Inaccessible Pore Volume Large polymer molecules cannot enter into the smaller pores of the
reservoir; the percentage of pores polymer cannot enter is called the Inaccessible Pore Volume (IPV). IPV
usually varies between 10% and 50% (Manichand et al. 2014). In practice the IPV accelerates the
progression of the polymer front (compared to a tracer for instance).
The views on whether the IPV is beneficial or not differ; on one hand it can be argued that IPV is
beneficial because it compensates the adsorption and accelerates the propagation of the polymer – and thus
the oil bank – in the reservoir; on the other hand, if IPV is such that the polymer solution bypasses most
16 SPE-171105-MS
Figure 4 —Measurement of IPV and adsorption - Reproduced from (Gaillard, Giovannetti, et al. 2014)
of the reservoir, one must wonder what the benefit of the polymer is in the first place. We consider that
an IPV over 30% is too high and in such a case a polymer of lower molecular weight should be selected.
Coreflood procedure Corefloods are performed in reservoir conditions (pressure, temperature). A typ-
ical coreflood procedure is as follows:
1. Age the oil-saturated core to restore wettability;
2. Measure porosity, pore volume, permeability to water, water-oil relative permeabilities; at the end
of the phase the core should be at Sor;
3. Inject polymer at constant rate to measure adsorption;
4. Inject same water as in 2. until all polymer has been displaced; measure the Residual Resistance
Factor RRF;
5. Inject polymer at various concentrations and various rates to measure the Resistance Factor RF at
various concentrations and various shear rates and the IPV; water without polymer is also injected
between each step to check the Residual Resistance Factor;
Injection rate is controlled and pressure is measured at the inlet and outlet of the core (and sometimes
at various points along the core in case of long cores). A pressure that keeps increasing instead of
stabilizing is a sign of plugging which could be due to the polymer, to clay swelling or to fine migration.
There are several ways to measure adsorption and IPV (Manichand et al. 2014); the material balance
method is not very accurate and thus is not recommended. The most commonly used method described
by Osterloh et al. (Osterloh et al. 1998) among others consists in injecting a tracer together with the first
slug of polymer; this first polymer slug must be sufficient to satisfy adsorption completely. The difference
in breakthrough time between the first and the second front of polymer corresponds to the adsorption; the
difference between the second front of polymer and the tracer corresponds to the IPV. The method is
illustrated in Figure 4 reproduced from Gaillard (Gaillard et al. 2014).
Coreflood simulations It must be stressed that polymer adsorption and other parameters are actually
measured in the lab, not obtained from history matching the experiment. Thus it is usually not necessary
to simulate the coreflood experiments.
SPE-171105-MS 17
Reservoir simulations
Reservoir simulations are used to history match the pattern selected for the pilot and then to forecast pilot
behavior. Before that a detailed geological model is usually built, encompassing all available information.
Once the history match is obtained then the model can be used to design the injection sequence using
the lab data generated for the water and polymer selected.
The main pitfall is the injectivity which is usually poorly accounted for in the simulator. Polymer
viscosity depends on shear rate but in general grid cells in the model are large thus the shear rate will be
calculated improperly and injectivity will be severely limited especially at high polymer concentration.
One too common outcome is that when the viscosity of the injected polymer is high the maximum pressure
constraint is reached and the injection rate is limited, which leads to lower recovery and selection of a
lower viscosity as the optimum.
One potential way of dealing with the issue is to build a separate, radial model to test the relation
between injection rate and pressure separately (Delamaide et al. 1994). Another solution is to use local
grid refinement around the injection wells. As a guideline as discussed earlier it can be expected that
polymer injection will reduce injectivity by 20% to 30%.
Sensitivities are usually performed on polymer viscosity and slug size; by experience it is not necessary
to perform too many sensitivities at this stage because of the uncertainties surrounding the pilot but
typically the plans call for the injection of 0.3 PV to 0.5 PV of polymer solution followed by chase water.
As injectivity is usually low in heavy oil, injecting those volumes could take several years and the results
of the pilot should be obvious before that. In Pelican Lake for instance it took almost 2 years to inject 0.1
PV and after nine years only 0.35 PV of polymer have been injected (Delamaide et al. 2014a). In Seal it
took a little over 3 years to inject 0.1 PV (Delamaide et al. 2014a).
The recommendations of the model in terms of polymer viscosity to be injected need to be taken with
a grain of salt. Simulations may recommend either too low viscosity – if injectivity was not properly taken
into account – or too high to be realistic. Thus the model recommendations should be analyzed carefully
and reviewed in light of practical field experience.
We accept these arguments but consider that this approach may be counterproductive, for the following
reasons:
● It appears to us that the increase in recovery does not justify the increase in injected concentration:
ⴰ In Tambaredjo, increasing the injected viscosity from 40 to 160 cp requires increasing the
polymer concentration by a factor , however incremental recovery will increase by
a smaller factor according to the estimates (Moe Soe Let et al. 2012). This remains true even
when accounting for the improved sweep efficiency, at least in Tambaredjo.
ⴰ In Pelican Lake, injection of a 20 cp viscosity resulted in an incremental recovery of 5-25%
OOIP; however it is unrealistic to expect that increasing the concentration by a factor of
would increase the incremental recovery to 14-70% OOIP.
● Increasing polymer viscosity increases the length of hydraulic fractures (Khodaverdian et al. 2010,
Seright et al. 2009). Hydraulic fractures that extend to large distances between wells can have a
potential negative impact on the sweep efficiency depending on their orientation (Bargas et al.
1988, van den Hoek 2004, Souza et al. 2005). In some ways, establishing a direct communication
between injection and production well may not be the worst in that regard because the issue will
be identified easily and the fracture can be closed by reducing the injection rate (Moe Soe Let et
al. 2012); on the contrary a fracture that extends a long distance towards the well but not all the
way could result in a poor sweep efficiency but the problem may be more difficult to identify. In
the worst case scenario that could also result in casing failure or loss of caprock integrity
(Khodaverdian et al. 2010). Thus near-wellbore fractures are good but long fractures are not.
● Increasing polymer viscosity will cause a larger pressure gradient between the injection and the
production well even if there is no impact on instantaneous injectivity. This larger gradient may
not always be sustainable in the field.
● In Pelican Lake injected solution viscosity in the first pilot was 100 cp and this was probably one
of the causes for the failure of the pilot (Delamaide et al. 2013).
● For most field cases summarized in Table 4, injected polymer viscosity was in the 20 – 50 cp range
even when oil viscosity was higher and in similar reservoir settings (free crossflow with variable
permeability).
In summary we consider that the benefits of injecting at very high viscosity (over 100 cp) are far from
obvious while the disadvantages are clear. Selecting the right viscosity is a challenge but in general we
recommend injecting at no more than 50 cp for future field projects.
Field operations
While the laboratory and simulation studies are under way many things need to happen on the field in
order to prepare for the pilot. These operations are of three kinds:
1. Reservoir
2. Wells
3. Surface facilities
Reservoir-related operations Reservoir-related operations take place in order to improve the under-
standing of the pilot area. Typical operations include tracer or interference tests to ascertain communi-
cation between wells, drilling or conversion of observation wells which will allow to monitor the progress
of the polymer front and of the oil bank, as well as injectivity tests.
These tests are not always performed; in heavy oil in particular if the pattern is not under waterflood
tracer tests are hardly possible. If injectivity is a concern, injectivity tests should have been performed
during the Feasibility phase.
Wells New wells may have to be drilled for the pilot, in particular if the field spacing is too large. That
raises the issue of representativity for the future extension but this is commonly done in order to reduce
the response time. New wells should of course be logged and incorporated into the existing geological and
simulation models as soon as possible. Water source wells may also have to be drilled.
Completion of injection wells should be designed to reduce or prevent mechanical degradation of the
polymer. In particular the number of perforations should be maximized in order to reduce the flow rate
through individual perforations (SNF Floerger 2011). Slotted liners have been used successfully in most
pilots with horizontal wells in Canada.
Surface facilities The surface facilities work consist in building and connecting the polymer mixing
facilities – where the polymer solution is going to be prepared – as well as the water source.
Polymer preparation is usually done in skids; a schematic of the process is presented in Figure 5 and
more complete descriptions can be found in (Thomas et al. 2012) and (Chang et al. 2013).
Polymer is usually delivered on site in powder form because it is the most cost effective way.
The water used for the preparation of the polymer solution needs to be filtered to eliminate solids.
The first step in the preparation of the solution is the slicing of the polymer grains in much smaller
particles in order to reduce the hydration time. This is done is special slicing units.
The second step is the hydration; during this phase the polymer in contact with water begins to expand.
Then the polymer is transferred to maturation tanks where it will complete the dissolution process; the
20 SPE-171105-MS
corresponding solution (so called “mother solution”) will be at concentrations ranging from 5,000 to
10,000 ppm.
Most polymers are sensitive to shear and can be severely degraded in surface facilities if no precaution
is taken. In particular non-shearing pumps have to be used and angles as well as flow restrictions in the
pipes have to be minimized. Centrifugal pumps in particular should not be used. Chokes can also induce
significant shearing and should be removed or left open if possible.
Polymer can be degraded by Fe2⫹ in the presence of oxygen so precautions need to be taken to prevent
that from happening. Seright (Seright et al. 2014) presents a detailed discussion on these aspects and here
we will only outline the main points.
When there is no Fe2⫹ in the water no precautions need to be taken at low temperature; at high
temperature as discussed earlier special polymers need to be used and they can be sensitive to high levels
of oxygen.
When Fe2⫹ is present in the water even at very low levels (over a few ppm), then oxygen content must
be kept below 200 ppb (parts per billion) otherwise polymer will be degraded. Degradation increases with
temperature and Fe2⫹ content.
Two solutions are usually used in the field:
● Precipitate the Fe2⫹ by aeration before it contacts the polymer as was done for instance during the
first pilot in Pelican Lake and in Bodo; this solution was employed because the presence of Fe2⫹
was discovered during the start-up phase and this was the only practical solution that could be
implemented onsite;
● Prevent oxygen from entering into the system by operating under an inert gas blanket (usually N2);
this is a better option although the skids are more expensive.
Oxygen scavengers are often used in combination with the above methods as a precaution. Biocide is
also used to prevent biological degradation of the polymer by bacteria; biopolymers are especially
sensitive but HPAM also – to a lesser extent. Some biocides are known to be incompatible with polymer
so care must be taken in the biocide selection.
The polymer solution is filtered before injection to remove gels or fish-eyes.
The concentration of polymer in the produced fluids also needs to be monitored periodically. Initially
this needs to be done only weekly but should be done more frequently after the polymer breakthrough.
Simple tests such as clay flocculation can be used to detect the presence of polymer and once the test is
positive then samples should be taken more frequently and the concentration should be measured.
Monitoring for the polymer breakthrough should start immediately and not around the time expected from
the reservoir simulations.
Start-up of the facilities
The start-up of the facilities is probably the most crucial moment of the whole project. It is critical to
ensure that the quality of the polymer solution is what was expected and that there is no degradation
between the outlet of the polymer preparation skid and the wellhead. Any issue must be troubleshooted
on the spot. Variations in source water composition or polymer quality can cause problems and require
adjusting the polymer concentration to reach the target viscosity.
At the same time the operators need to be trained in order to ensure that they will be able to perform
the required tasks: how to take polymer samples without degrading them, how to calibrate and use the
viscometer, how to make sure that a consistent solution viscosity is obtained etc.
Operators need to know what to expect and when to alert the head office that something unexpected
is happening so that the issue can be investigated promptly and corrective measures taken.
Pilot execution
This is the phase when all the work finally comes to fruition. Polymer preparation and injection take place
and rates, pressures, compositions etc. are carefully monitored. The main things to watch for are the
injectivity as well as potential water and/or polymer breakthrough and obviously the production response.
The field data need to be compared with the reservoir simulation forecasts in order to detect any
deviation from the plan. If that happens then the causes need to be identified and corrective action need
to be taken – if at all possible. In the second Pelican Lake pilot for instance the injection pressure was
increasing faster than expected so this was integrated into the reservoir simulation model; a new history
match was performed and as a result it was decided to reduce the molecular weight of the polymer
(Delamaide et al. 2013).
Pilot response
In most field cases analyzed the pilot response occurred within a few months of the beginning of injection
(Table 4). The response time depends on the injection rates and on the pattern volume. When the response
occurs it is clearly noticeable by an increase in oil rate. In Pelican Lake the increase in oil rate is usually
preceded by a decrease in gas production which signals that the reservoir is pressuring up. When there was
little to no water production prior to the polymer injection (such as in Pelican Lake, Mooney, Seal) the
water-cut increases as water then polymer start being produced; when the well was already producing
water (Suffield, Marmul) a reduction in water-cut can first be observed.
Pump changes may be required to accommodate the increase in total produced fluids.
Pilot interpretation
Once the pilot is completed or as soon as sufficient data is available the reservoir simulation model needs
to be updated and the results interpreted. Key elements such as the water breakthrough as well as the
increase in oil rate and pressure need to be correctly history matched along with the produced polymer
concentration. If the composition of the injection water is different from the reservoir water, the changes
in water composition in the producers can also be history matched if the reservoir simulator used has this
option.
This is not a simple task in particular because simulators do not account correctly for some parameters
such as adsorption as pointed out by Zhang (Zhang et al. 2014) and also for the degradation of the polymer
which takes place in the reservoir: when the temperature is high the polymer gets hydrolyzed and thus its
22 SPE-171105-MS
viscosity changes with time. In addition viscous fingering which is common for high viscosity oil is very
difficult to simulate.
The updated model can then be used to estimate incremental recovery and provide some data for the
evaluation of the field extension. Sensitivity studies on the size and concentration of the polymer slug can
also be performed in order to optimize the whole process. It is also possible to forecast the behavior of
a polymer flood in other parts of the field where the geology differs from that of the pilot.
Failed pilots
Sometimes pilots fail but even failed pilots can bring very useful information. In case of failure it is critical
to understand what happened: whether it was due to the process itself or to the operations; and what if
anything can be done to remedy the situation.
In Pelican Lake the first pilot failed but the analysis of the results revealed that the problem was due
to operational issues; at the same time it was realized that the injected polymer viscosity was too high.
Those learnings were exploited years later to design the second pilot which was successful and paved the
way to the full field deployment (Delamaide et al. 2013).
Incremental recovery
Incremental recoveries (over primary production) achieved in the projects listed in Table 4 vary between
7.5 to 25% OOIP. Incremental recovery depends on viscosity, heterogeneity and other factors and is of
course case dependent; however these recovery numbers can give a first indication as to what to expect
for polymer flooding of viscous oil and can also be used to calibrate reservoir simulations. Simulations
suggesting an incremental recovery much above these numbers should be analyzed carefully to ensure
their validity.
Field extension
At the moment field extensions have taken place in Marmul where 27 pads are being flooded (Jaspers et
al. 2013), in Pelican Lake where hundreds of wells are currently injecting polymer (Delamaide et al.
2014), in Bohai Bay and in Seal with 11 injectors and 13 producers (Murphy Oil Company Ltd. 2014).
In Mooney polymer flood has been abandoned and ASP is instead implemented at a large scale (Watson
et al. 2014). Field extension has also apparently been approved in Tambaredjo for 36 new injection wells
(Staatsolie 2014).
Pilots are usually done in “simple” areas of the field, far from features such as fault or edge aquifer;
in simple well patterns and also in the lower range of viscosity encountered in the field. The water source
for the pilot is also usually of good quality or at least as best as can be. All of this is going to change for
the field extension and this is why it is far from being a simple phase. The issues can be separated into
two categories: 1) reservoir and 2) surface facilities.
Reservoir challenges
Dealing with a different source of water The water used for the field extension is frequently different
from the water used for the pilot; the major reason for that is that regulatory bodies usually restrict fresh
water usage and require operators to use their produced water for re-injection. Additional laboratory work
may be required to evaluate the impact of this water on the polymer properties and the concentration to
use in order to achieve the target viscosity.
Dealing with larger well spacing Small spacings such as those usually used in pilots may not be
sustainable for the field extension. With larger spacing the production of the oil bank will be delayed. A
larger spacing will also increase the residence time of the polymer and thus its hydrolysis degree with a
potential impact on its viscosity; this can also be checked in the lab.
Dealing with initial mobile water It is not uncommon to be confronted to wells producing at high
water-cut, either due to an ongoing waterflood (as in Marmul or Suffield) or due to the production of
SPE-171105-MS 23
interstitial water (as in Pelican Lake, Mooney, Tambaredjo). This is not an issue provided that there is no
bottom aquifer. In such cases in Pelican Lake typically water-cut begins by increasing before the oil bank
starts being produced; when the oil bank arrives at the producer the water-cut goes down.
Dealing with variable viscosity oil This can be a challenge and may require local adjustments of
injected polymer concentration; higher viscosity usually would require slightly higher polymer viscosity
but the changes should not be drastic. This adjustment can be made using the reservoir simulation model
once it has been properly calibrated. If viscosity varies significantly in the field (such as in Pelican Lake)
then the adjustment is more difficult. In Pelican Lake the pilot was done in 1,500 cp oil but the polymer
flood has been extended to some areas where oil viscosity is over 10,000 cp (Delamaide et al. 2014a,
Delamaide et al. 2014c). Oil rate response is usually lower and can take longer because of the reduced
injectivity. To compensate a shorter well spacing was tested with no obvious disadvantage except for the
extra capital required.
Dealing with high permeability thief zones Although polymer is used for mobility control and can
reduce the effect of reservoir heterogeneity, its efficiency is limited in the case of large permeability
contrasts. In the Mooney polymer pilot for instance the oil rate increased but the water breakthrough was
very severe; this is probably due to heterogeneity or to the high injection rate required by the regulator
(Watson et al. 2014, Delamaide et al. 2014a). In Pelican Lake some wells also have conformance issues.
There is no ideal solution to deal with this issue. Increasing the polymer concentration can sometimes
help as does reducing the injection rate (Cenovus Energy 2013). Conformance control treatments have
also been tested with some measure of success (Buciak et al. 2013) but they are not well adapted to
reservoirs where the high permeability thief zone is connected to the rest of the pay.
Oil-water separation Oil-water separation can be difficult when the oil is heavy and the temperature is
low due to the small difference in density between oil and water and the high viscosity of the oil. In
addition the heavy fractions of the oil tend to stabilize emulsions (Wylde et al. 2010). Unfortunately
polymer makes the situation worse.
Chemical demulsifiers are used routinely to break emulsions however when emulsions are really tight
they may not prove sufficient and other methods need to be applied. Polymer may also interfere with the
demulsifiers used.
In Marmul separation is done using induced gas flotation for primary separation followed by nutshell
filters to allow reusing water for the polymer preparation (Al-Saadi et al. 2012). The separation issues do
not seem as severe in Marmul because of the lower viscosity and the generally high surface temperature.
In Pelican Lake oil viscosity is high (typically in the 1,500 – 5,000 cp range) and both reservoir and
surface temperatures are low. Due to the very large number of wells under polymer flood (relative to the
total number of wells in the field) the concentration of polymer produced in the facility is high and has
made separation issues more difficult (Wylde et al. 2011).
Heater treaters are commonly used for oil-water separation but the presence of polymer can cause
fouling of the treaters and failure of some of their components such as fire tubes (Zheng et al. 2011).
Polymer precipitates on the fire tubes and causes hot spots which eventually lead to the tube failure (Bartz
et al. 2014); these failures can be frequent and increase operating costs. Wylde et al. (Wylde et al. 2011)
correlated increased fire tubes temperatures to the increase in polymer concentration present in the
effluents in Pelican Lake.
The development of new scale inhibitors for fire tubes (Wylde et al. 2011) and modifications in fire
tubes design (Bartz et al. 2014) have successfully been used to alleviate these issues.
Conclusions
The paper has discussed many aspects of polymer flooding in heavy oil reservoirs including reservoir and
surface issues. The main conclusions are as follows:
● Several pilots and several full field commercial developments have taken place or are ongoing,
thus confirming that it is a viable EOR process that should be evaluated as an alternative to thermal
methods.
● The key screening criteria are oil viscosity (lower than 5,000 cp), low reservoir temperature (less
than 80°C preferred), no bottom aquifer, high permeability, low salinity and low hardness water,
sandstone reservoirs, mobile oil saturation (⬎30%).
● Injected polymer viscosity is 20-50 cp in most field projects and going above these values is
generally not recommended.
● Well established laboratory protocols exist and allow good project designs.
● In the field cases examined injectivity has usually not been an issue.
● Start-up of the surface facilities for the pilot is critical as is the training of the field operators.
● The extension of the process to the field brings new challenges both on a reservoir and surface
facilities standpoint. The most difficult are probably the need to use produced water for reinjection
and the oil-water separation but these issues are manageable.
● In spite of these challenges the process can be very successful and result in incremental recoveries
ranging from 7.5 to 25% OOIP.
Acknowledgements
The author thanks David Rousseau and René Tabary (IFP Energies nouvelles) for many fruitful
discussions.
SPE-171105-MS 25
Nomenclature
IPV ⫽ Inaccessible Pore Volume
Nc ⫽ Capillary number
RF ⫽ Resistance Factor (Rm)
Rk ⫽ Permeability Reduction (RRF)
Rm ⫽ Mobility Reduction (RF)
RRF ⫽ Residual Resistance Factor (Rk)
Sor ⫽ Residual oil saturation
⌬Pp ⫽ Pressure drop caused by injection of polymer solution in core
⌬PW ⫽ Pressure drop caused by injection of water in core
⫽ velocity (m/s)
⫽ viscosity (cp)
⫽ interfacial tension (mN/m)
References
Al Abri, K., R. Al-Mjeni, N. Al-Bulushi, K. Awan, N. Al-Azri, O. Al-Riyami, S. Al-Rajhi, et alet al.
2014. “Reducing Key Uncertainities Prior to a Polymer Injection Trial in a Heavy Oil Reservoir in Oman.”
Paper SPE 169686 presented at the SPE EOR Conference at Oil and Gas West Asia. Muscat, Oman, 31
March-2 April.
Al Azri, N., E. Jamal, A. Murshidi, A. Al Mahrouqi, I. Al Busaidi, A. Kazzaz, W. Al Ajmi, et alet al.
2010. “Polymer Injection in a Heavy Oil Reservoir under Strong Bottom Water Drive.” Paper SPE 129177
presented at the SPE EOR Conference at Oil & Gas West Asia. Muscat, Oman, 11-13 April.
Al-Saadi, F.S., B.A. Al-amri, S.M. Al Nofli, J.S.M. Van Wunnik, H.F. Jaspers, S. Al Harthi, K.
Shuaili, P.K. Cherukupalli, and R. Chakravarthi. 2012. “Polymer Flooding in a Large Field in South Oman
- Initial Results and Future Plans.” Paper SPE 154665 presented at the SPE EOR Conference at Oil and
Gas West Asia. Muscat, Oman, 16-18 April.
Audibert, A., and J.F. Argillier. 1995. “Thermal stability of sulfonated polymers.” Paper SPE 28953
presented at the SPE International Symposium on Oilfield Chemistry. San Antonio, Texas, 14-17
February.
Audibert, A., C. Noik, and J. Lecourtier. 1993. “Behaviour of Polysaccharides Under Harsh Condi-
tions.” Journal of Canadian Petroleum Technology 32(7): 53–58.
Bang, V. 2013. “A New Screening Model for Gas and Water Based EOR Processes.” Paper SPE
165217 presented at the SPE Enhanced Oil Recovery Conference. Kuala Lumpur, Malaysia, 2-4 July.
Bargas, C.L., and J.L. Yanosik. 1988. “The Effects of Vertical Fractures on Areal Sweep Efficiency
in Adverse Mobility Ratio Floods.” Paper SPE 17609 presented at the International Meeting on Petroleum
Engineering. Tianjin, China, 1-4 November.
Bartz, D., and J. Gotterba. 2014. “Results of Field Operation of a Distributed Flux Burner in a Heater
Treater in a Northern Canada Heavy Oil Field; Thermal Performance and Firetube Life.” Paper SPE
170172 presented at the SPE Heavy Oil Conference-Canada. Calgary, Alberta, Canada, 10-12 June.
Bondino, I., E. Santanach-Carreras, D. Levitt, S. Jouenne, and M. Bourrel. 2013. “Polymer Flooding
of Heavy Oil Under Adverse Mobility Conditions.” Paper SPE 165267 presented at the SPE Enhanced Oil
Recovery Conference. Kuala Lumpur, Malaysia, 2-4 July.
Bragg, J.R., S.D. Maruca, W.W. Gale, L.S. Gall, W.C. Wernau, D. Beck, I.M. Goldman, A.I. Laskin,
and L.A. Naslund. 1983. “Control of Xanthan-Degrading Organisms in the Loudon Pilot: Approach,
Methodology, and Results.” Paper SPE 11989 presented at the SPE Annual Technical Conference and
Exhibition. San Francisco, California, USA, 5-8 October.
26 SPE-171105-MS
Buciak, J., Fondevilla Sancet, G. and Del Pozo, L. 2013. “Polymer Flooding Pilot Learning Curve: 5⫹
Years Experience to Reduce Cost Per Incremental Oil Barrel.” Paper SPE 166265 presented at the SPE
Annual Technical Conference and Exhibition. New Orleans, Louisiana, USA, 30 September-2 October.
Buell, R.S., H. Kazemi, and F.H. Poettmann. 1990. “Analyzing Injectivity of Polymer Solutions With
the Hall Plot.” SPE Reservoir Engineering 5(01): 41–46.
Cenovus Energy. 2013. “Performance Review of In Situ Oil Sands Scheme Approval 9404T.” In Situ
Progress Presentation, Calgary. Accessed July 08, 2014. http://www.aer.ca/data-and-publications/activity-
and-data/in-situ-performance-presentations.
Chang, H.L., H. Hou, F. Wu, and Y. Gao. 2013. “Chemical EOR Injection Facilities - From Pilot Test
to Field-Wide Expansion.” Paper SPE 165308 presented at the SPE Enhanced Oil Recovery Conference.
Kuala Lumpur, Malaysia, 2-4 July.
Chauveteau, G. 1981. “Molecular Interpretation of Several Different Properties of Flow of Coiled
Polymer Solutions Through Porous Media in Oil Recovery Conditions.” Paper SPE 10060 presented at the
SPE Annual Technical Conference and Exhibition. San Antonio, Texas, USA, 4-7 October.
Clemens, T., M. Deckers, M. Kornberger, T. Gumpenberger, and M. Zechner. 2013. “Polymer
Solution Injection - Near Wellbore Dynamics and Displacement Efficiency, Pilot Test Results, Matzen
Field, Austria.” Paper SPE 164904 presented at the EAGE Annual Conference & Exhibition incorporating
SPE Europec. London, UK, 10-13 June.
Craig. Jr., F.F. 1993. The Reservoir Engineering Aspects of Waterflooding. Richardson, TX, USA:
Society of Petroleum Engineers.
Delamaide, E. 2014. Geowebworks - Public production database for Bodo. Accessed July 1, 2014.
http://www.geowebworks.com/.
Delamaide, E., P. Corlay, and D. Wang. 1994. “Daqing Oil Field: The Success of Two Pilots Initiates
First Extension of Polymer Injection in a Giant Oil Field.” Paper SPE 27819 presented at the SPE/DOE
Improved Oil Recovery Symposium. Tulsa, Oklahoma, USA, 17-20 April.
Delamaide, E., A. Zaitoun, G. Renard, and R. Tabary. 2013. “Pelican Lake Field: First Successful
Application of Polymer Flooding in a Heavy Oil Reservoir.” SPE 165234 presented at the SPE Enhanced
Oil Recovery Conference. Kuala Lumpur, Malaysia, July 2-4.
Delamaide, E., B. Bazin, D. Rousseau, and G. Degre. 2014a. “Chemical EOR for Heavy Oil: The
Canadian Experience.” Paper SPE 169715 presented at the SPE EOR Conference at Oil and Gas West
Asia. Muscat, Oman, 31 March-2 April.
Delamaide, E., R. Tabary, and D. Rousseau. 2014b. “Chemical EOR in Low Permeability Reservoirs.”
Paper SPE 169673 to be presented at the SPE EOR at Oil & Gas Conference West Asia. Muscat, Oman,
March 31-April 2.
Delamaide, E., R. Tabary, G. Renard, and P. Dwyer. 2014c. “Field Scale Polymer Flooding of Heavy
Oil: the Pelican Lake Story.” Paper presented at the 21st World Petroleum Congress. Moscow, Russia,
June 15-19.
Doorwar, S., and K.K. Mohanty. 2014. “Polymer Flood of Viscous Oils in Complex Carbonates.”
Paper SPE 169162 presented at the SPE Improved Oil Recovery Symposium. Tulsa, Oklahoma, USA,
12-16 April.
Dupas, A., I. Henaut, D. Rousseau, P. Poulain, R. Tabary, J.F. Argillier, and T. Aubry. 2013. “Impact
of Polymer Mechanical Degradation on Shear and Extensional Viscosities: Toward Better Injectivity
Forecasts in Polymer Flooding Operations.” Paper SPE 164083 presented at the SPE International
Symposium on Oilfield Chemistry. The Woodlands, Texas, USA, 8-10 April.
Dupuis, G., D. Rousseau, R. Tabary, and B. Grassi. 2012. “Flow of Hydrophobically Modified
Water-Soluble Polymers in Porous Media: Controlled Resistance Factors vs. Flow-Induced Gelation in the
Semidilute Regime.” SPE Journal 17(04): 1196 –1206.
SPE-171105-MS 27
Dupuis, G., D. Rousseau, R. Tabary, and B. Grassl. 2011. “Flow of Hydrophobically Modified
Water-Soluble-Polymer Solutions in Porous Media: New Experimental Insights in the Diluted Regime.”
SPE Journal 16(01): 43–54.
Etebar, S. 1997. “Captain Field Development Project Overview.” Paper OTC 8507 presented at the
Offshore Technology Conference. Houston, Texas, United States, 5 May.
Fabbri, C., C. Cottin, J. Jimenez, M. Nguyen, S. Hourcq, M. Bourgeois, and G. Hamon. 2014.
“Secondary and Tertiary Polymer Flooding in Extra-Heavy Oil: Reservoir Conditions Measurements -
Performance Comparison.” Paper IPTC-17703 presented at the International Petroleum Technology
Conference. Doha, Qatar, 19-22 January.
Fletcher, A.J.P., S.P. Lamb, and P.J. Clifford. 1992. “Formation Damage From Polymer Solutions:
Factors Governing Injectivity.” SPE Reservoir Engineering 7(02): 237–246.
Gaillard, N., B. Giovannetti, C. Favero, J.-P. Caritey, G. Dupuis, and A. Zaitoun. 2014. “New Water
Soluble Anionic NVP Acrylamide Terpolymers for Use in Harsh EOR Conditions.” Paper SPE 169108
presented at the SPE Improved Oil Recovery Symposium. Tulsa, Oklahoma, USA, 12-16 April.
Gaillard, N., D.B. Sanders, and C. Favero. 2010. “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, USA, 24-28
April.
Greaves, B.L., R.N. Marshall, and J.H. Thompson. 1984. “Hitts Lake Unit Polymer Project.” Paper
SPE 13123 presented at the SPE Annual Technical Conference and Exhibition. Houston, Texas, 16-19
September.
Green, D.W., and G.P. Willhite. 1998. Enhanced Oil Recovery, SPE Textbook Series Vol. 6.
Richardson, TX: Society of Petroleum Engineers.
Hryc, A., F. Hochenfellner, H. Paponi, R. Puliti, and T. Gerlero. 2013. “Design and Execution of a
Polymer Injection Pilot in Argentina.” Paper SPE 166078 presented at the SPE Annual Technical
Conference and Exhibition. New Orleans, Louisiana, USA, 30 September-2 October.
Jaspers, H.F., M.S. Al-Amri, K.A. Al-Saqri, K. Zuhaimi, K.H. Al-Hashmi, and C. Thakuria. 2013.
“Performance Review of Polymer Flooding in a Major Brown Oil Field of Sultanate of Oman.” Paper SPE
165262 presented at the SPE Enhanced Oil Recovery Conference. Kuala Lumpur, Malaysia, 2-4 July.
Kang, X., J. Zhang, F. Sun, F. Zhang, G. Feng, J.R. Yang, X.S. Zhang, and W.T. Xiang. 2011. “A
Review of Polymer EOR on Offshore Heavy Oil Field in Bohai Bay, China.” paper SPE 144932 presented
at the SPE Enhanced Oil Recovery Conference. Kuala Lumpur, Malaysia, 19-21 July.
Khodaverdian, M., T. Sorop, S. Postif, and P. Van den Hoek. 2010. “Polymer Flooding in Uncon-
solidated-Sand Formations: Fracturing and Geomechanical Considerations.” SPE Production & Opera-
tions 25(02): 211–222.
Lefebvre, C., P. Lemouzy, D. Sorin, G. Roy, and S. Serbutoviez. 2012. “Building a Roadmap for
Enhanced Oil Recovery Prefeasibility Study.” Paper SPE 159264 presented at the SPE Russian Oil and
Gas Exploration and Production Technical Conference and Exhibition. Moscow, Russia, 16-18 October.
Leonhardt, B., B. Ernst, S. Reimann, A. Steigerwald, and F. Lehr. 2014. “Field Testing The
Polysaccharide Schizophyllan: Results of The First Year.” Paper SPE 169032 presented at the SPE
Improved Oil Recovery Symposium. Tulsa, Oklahoma, USA, 12-16 April.
Levitt, D.B., G.A. Pope, and S. Jouenne. 2011. “Chemical Degradation of Polyacrylamide Polymers
Under Alkaline Conditions.” SPE Reservoir Evaluation & Engineering 14(03): 281–286.
Li, Z., M. Delshad, M. Lotfollahi, H. Koh, H. Luo, H.L. Chang, J. Zhang, P. Dempsey, C.
Lucas-Clement, and B. Brennan. 2014. “Polymer Flooding of a Heavy Oil Reservoir with an Active
Aquifer.” Paper SPE 169149 presented at the SPE Improved Oil Recovery Symposium. Tulsa, Oklahoma,
USA, 12-16 April.
28 SPE-171105-MS
Liu, J., K. Adegbesan, and J. Bai. 2012. “Suffield Area, Alberta, Canada - Caen Polymer Flood Pilot
Project.” Paper SPE 157796 presented at the SPE Heavy Oil Conference Canada. Calgary, Alberta,
Canada, 12-14 June.
Manichand, R.N., and R.S. Seright. 2014. “Field vs. Laboratory Polymer Retention Values for a
Polymer Flood in the Tambaredjo Field.” Paper SPE 169027 presented at the SPE Improved Oil Recovery
Symposium. Oklahoma, USA, 12-16 April, Tulsa.
Manichand, R.N., K.P. Moe Soe Let, L. Gil, B. Quillien, and R.S. Seright. 2013. “Effective
Propagation of HPAM Solutions Through the Tambaredjo Reservoir During a Polymer Flood.” SPE
Production & Operations 28(04): 358 –368.
Moe Soe Let, K.P., R.N. Manichand, and R.S. Seright. 2012. “Polymer Flooding a ~500-cp Oil.” paper
SPE 154567 presented at the SPE Improved Oil Recovery Symposium. Tulsa, Oklahoma, USA, 14-18
April.
Moradi-Araghi, A., and P.H. Doe. 1987. “Hydrolysis and Precipitation of Polyacrylamides in Hard
Brines at Elevated Temperatures.” SPE Reservoir Engineering 2(02): 189 –198.
Mukherjee, S., and P. Suss. 2014. “Performance Evaluation and Design Aspects of Polymer and Hot
Water Co-Injection in an Offshore Heavy Oil Field under Challenging Conditions.” Paper SPE 169711
presented at the SPE EOR Conference at Oil and Gas West Asia. Muscat, Oman, 31 March-2 April.
Murphy Oil Company Ltd. 2014. “Seal Polymer Project - Scheme Approval No. 11320B - Annual
AER Progress Presentation.” Presentation, Alberta Energy Regulator. Accessed 07 09, 2014. http://
www.aer.ca/documents/oilsands/insitu-presentations/2014PeaceRiverMurphySeal11320.pdf.
Norton, D.K., and D.L. Dauben. 1986. “Evaluation of the Storms Pool Improved Waterflood Project.”
US Department of Energy, Bartlesville, OK.
Ogezi, O., J. Strobel, D. Egbuniwe, and B. Leonhardt. 2014. “Operational Aspects of a Biopolymer
Flood in a Mature Oilfield.” Paper SPE 169158 presented at the SPE Improved Oil Recovery Symposium.
Tulsa, Oklahoma, USA, 12-16 April.
Osterloh, W.T., and E.J. Law. 1998. “Polymer Transport and Rheological Properties for Polymer
Flooding in the North Sea.” Paper SPE 39694 presented at the SPE/DOE Improved Oil Recovery
Symposium. Tulsa, Oklahomam USA, 19-22 April.
Parada, C.H., and T. Ertekin. 2012. “A New Screening Tool for Improved Oil Recovery Methods
Using Artificial Neural Networks.” Paper SPE 153321 presented at the SPE Western Regional Meeting.
Bakersfield, California, USA, 21-23 March.
Renouf, G. 2014. “A Survey of Polymer Flooding in Western Canada.” Paper SPE 169062 presented
at the SPE Improved Oil Recovery Symposium. Tulsa, Oklahoma, USA, 12-16 April.
Rivenq, R.C., A. Donche, and C. Noik. 1992. “Improved Scleroglucan for Polymer Flooding Under
Harsh Reservoir Conditions.” SPE Reservoir Engineering 7(1): 15–20.
Saleh, L.D., M. Wei, and B. Bai. 2014. “Data Analysis and Updated Screening Criteria for Polymer
Flooding Based on Oilfield Data.” SPE Reservoir Evaluation & Engineering Preprint.
Seright, R., A.R. Campbell, P.S. Mozley, and P. Han. 2010. “Stability of Partially Hydrolyzed
Polyacrylamides at Elevated Temperatures in the Absence of Divalent Cations.” SPE Journal 15(02):
341–348.
Seright, R., and I. Skjevrak. 2014. “Effect of Dissolved Iron and Oxygen on Stability of HPAM
Polymers.” Paper SPE 169030 presented at the SPE Improved Oil Recovery Symposium. Tulsa, Okla-
homa, USA, 12-16 April.
Seright, R.S. 2010. “Potential for Polymer Flooding Reservoirs with Viscous Oils.” SPE 129899
presented at the SPE Enhanced Oil Recovery Symposium. Tulsa, OK, USA, April 24-28.
Seright, R.S., M. Seheult, and T. Talashek. 2009. “Injectivity Characteristics of EOR Polymers.” SPE
Reservoir Evaluation & Engineering 12(05): 783–792.
SPE-171105-MS 29
Seright, R.S., T. Fan, K. Wavrick, H. Wan, N. Gaillard, and B. Grassi. 2011. “Rheology of a New
Sulfonic Associative Polymer in Porous Media.” SPE Reservoir Evaluation & Engineering 14(06):
726 –734.
Skauge, A., P.A. Ormehaug, T. Gurholt, Bartek Vik, I. Bondino, and G. Hamon. 2012. “2-D
Visualisation of Unstable Waterflood and Polymer Flood for Displacement of Heavy Oil.” Paper SPE
154592 presented at the SPE Improved Oil Recovery Symposium. Tulsa, Oklahoma, USA, 14-18 April.
SNF Floerger. 2011. www.force.org. May 30. http://www.force.org/Global/Seminars/2011/
EOR_Subsea/Kieffer.pdf.
Sorbie, K.S. 1991. Polymer-improved Oil Recovery. Blackie and Sons Tld./US and Canada: CRC
Press Inc.
Souza, A.L.S., P.D. Fernandes, R.A. Mendes, A.J. Rosa, and C.J.A. Furtado. 2005. “The Impact of
Fracture Propagation on Sweep Efficiency During a Waterflooding Process.” Paper SPE 94704 presented
at the SPE Latin American and Caribbean Petroleum Engineering Conference. Rio de Janeiro, Brazil,
20-23 June.
Staatsolie. 2014. “Annual Report 2013.” Accessed 07 09, 2014. http://www.staatsolie.com/.
Stegemeier, G.L. 1977. “Mechanisms of Entrapment and Mobilization of Oil in Porous Media.” In
Improved Oil Recovery by Surfactant and Polymer Flooding, 55–91. New York City: Academic Press.
Taber, J.J., F.D. Martin, and R.S. Seright. 1997. “EOR Screening Criteria Revisited - Part 1:
Introduction to Screening Criteria and Enhanced Recovery Field Projects.” SPE Reservoir Engineering
12(3): 189 –198.
Teletzke, G.F., R.C. Wattenbarger, and J.R. Wilkinson. 2010. “Enhanced Oil Recovery Pilot Testing
Best Practices.” SPE Reservoir Evaluation & Engineering 13(01): 143–154.
Thomas, A., N. Gaillard, and C. Favero. 2012. “Some Key Features to Consider When Studying
Acrylamide-Based Polymers for Chemical Enhanced Oil Recovery.” Oil & Gas Science and Technology
67(6): 887–902.
van den Hoek, P.J. 2004. “Impact of Induced Fractures on Sweep and Reservoir Management in
Pattern Floods.” Paper SPE 90968 presented at the SPE Annual Technical Conference and Exhibition.
Houston, Texas, 26-29 September.
Vermolen, E.C.M, M.J.T van Haasterecht, S.K. Masalmeh, M.J. Faber, D.M. Boersma, and M.
Gruenenfelder. 2011. “Pushing the envelope for polymer flooding towards high-temperature and high-
salinity reservoirs with polyacrylamide based ter-polymers.” Paper SPE 141497 presented at the SPE
Middle East Oil and Gas Show and Conference. Manama, Bahrain, 25-28 September.
Vermolen, E.C.M., M. Pingo-Almada, B. Wassing, D.J. Ligthelm, and S.K. Masalmeh. 2014a.
“Low-Salinity Polymer Flooding: Improving Polymer Flooding Technical Feasibility and Economics by
Using Low-Salinity Make-up Brine.” Paper IPTC 17342 presented at the International Petroleum
Technology Conference. Doha, Qatar, 19-22 January.
Vermolen, E.C.M., M.J.T. van Haasterecht, and S.K. Masalmeh. 2014b. “A Systematic Study of the
Polymer Visco-Elastic Effect on Residual Oil Saturation by Core Flooding.” Paper SPE 169681 presented
at the SPE EOR Conference at Oil and Gas West Asia. Muscat, Oman, 31 March-2 April.
Wang, D., H. Xia, Z. Liu, and Q. Yang. 2001a. “Study of the Mechanism of Polymer Solution With
Visco-Elastic Behavior Increasing Microscopic Oil Displacement Efficiency and the Forming of Steady
“Oil Thread” Flow Channels.” Paper SPE 68723 presented at the SPE Asia Pacific Oil and Gas
Conference and Exhibition. Jakarta, Indonesia, 17-19 April.
Wang, D., J. Cheng, H. Xia, Q. Li, and J. Shi. 2001b. “Viscous-Elastic Fluids Can Mobilize Oil
Remaining after Water-Flood by Force Parallel to the Oil-Water Interface.” Paper SPE 72123 presented
at the SPE Asia Pacific Improved Oil Recovery Conference. Kuala Lumpur, Malaysia, 6-9 October.
30 SPE-171105-MS
Wang, D., J. Cheng, Q. Yang, W. Gong, Q. Li, and F. Chen. 2000. “Viscous-Elastic Polymer Can
Increase Microscale Displacement Efficiency in Cores.” Paper SPE 63227 presented at the SPE Annual
Technical Conference and Exhibition. Dallas, Texas, USA, 1-4 October.
Wassmuth, F.R., W. Arnold, K. Green, and N. Cameron. 2009. “Polymer Flood Application to
Improve Heavy Oil Recovery at East Bodo.” Journal of Canadian Petroleum Technology 48(2).
Watson, A., G.A. Trahan, and W. Sorensen. 2014. “An Interim Case Study of an Alkaline-Surfactant-
Polymer Flood in the Mooney Field, Alberta, Canada.” Paper SPE 169154 presented at the SPE Improved
Oil Recovery Symposium. Tulsa, Oklahoma, USA, 12-16 April.
Wylde, J.J., K. Allan, J. McMahon, and S. Mayner. 2011. “Scale Inhibitor Application In Northern
Alberta: A Case History of an Ultra High Temperature Scale Inhibition Solution in Fire Tube Heater
Treaters.” Paper SPE 141100 presented at the SPE International Symposium on Oilfield Chemistry. The
Woodlands, Texas, USA, 11-13 April.
Wylde, J.J., S. Coscio, and V. Barbu. 2010. “A Case History of Heavy-Oil Separation in Northern
Alberta: A Singular Challenge of Demulsifier Optimization and Application.” SPE Production &
Operations 25(01): 19 –24.
Yang, S.H., and L.E. Treiber. 1985. “Chemical Stability of Polyacrylamide Under Simulated Field
Conditions.” Las Vegas, Nevada, USA, 22-26 September, Paper SPE 14232 presented at the SPE Annual
Technical Conference and Exhibition.
Zerafat, M.M., S. Ayatollahi, N. Mehranbod, and D. Bargezari. 2011. “Bayesian Network Analysis as
a Tool for Efficient EOR Screening.” Paper SPE 143282 presented at the SPE Enhanced Oil Recovery
Conference. Kuala Lumpur, Malaysia, 19-21 July.
Zhang, G., and R.S. Seright. 2014. “Effect of Concentration on HPAM Retention in Porous Media.”
SPE Journal 19(03): 373–380.
Zheng, F., P. Quiroga, and G.W. Sams. 2011. “Challenges in Processing Produced Emulsion from
Chemical Enhanced Oil Recovery - Polymer Flood Using Polyacrylamide.” Paper SPE 144322 presented
at the SPE Enhanced Oil Recovery Conference. Kuala Lumpur, Malaysia, 19-21 July.
Zijlstra, E., J. Van Wunnik, J. Van Doren, A. Al Bimani, and O. Wilson. 2014. “Accurate Tool for
IOR/EOR Screening and Estimating Target Recovery Factors.” Paper SPE 169653 presented at the SPE
EOR Conference at Oil and Gas West Asia. Muscat, Oman, 31 March-2 April.