SPE 164078

Polymeric and Alkali-Surfactant Polymer Enhanced Oil Recovery Chemical

Treatment: Chemistry and Strategies Required After Breakthrough into the
Process
Jonathan J. Wylde, Jubal L. Slayer and Victor Barbu (Clariant Oil Services)

Copyright 2013, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE International Symposium on Oilfield Chemistry held in The Woodlands, Texas, USA, 8–10 April 2013.

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
The use of polymer flooding and alkali surfactant polymer (ASP) flooding techniques is becoming commonplace in the oil
industry, particularly in projects where heavier and more viscous crude oil is produced. While the efficacy of these enhanced
oil recovery (EOR) techniques cannot be disputed, as recovery factors (RF) can be doubled, often there is little consideration
given to the implications of these EOR chemicals breaking through into producer wells. The impact caused by EOR chemical
breakthrough can be varied, but most commonly the efficacy of oil/water separation is seriously affected. The contribution
that EOR chemicals can have on reservoir souring is often underestimated, as is the effect they can have upon standard
production chemicals such as scale and corrosion inhibitors.

This paper draws upon the experience of several different EOR operations from around the world and demonstrates the
different range of challenges that EOR chemical breakthrough has upon the production chemical strategy. Examples include
the effect on separation, scale inhibition and on integrity management - three essential components of any flow assurance
strategy. Two case histories are taken from Canada where the effect on separation was profound. A case history is also
provided where a scale inhibition / solids mitigation strategy had to be significantly amended in order to accommodate the
breakthrough of an ASP EOR flood.

The paper details several lessons learned that can be used to assist the front end engineering and design (FEED)
considerations of EOR strategies, including future-proofing facilities and production chemistry strategies for planned and
retrospective EOR flooding projects.

Introduction
The average RF from hydrocarbon reservoirs around the world from primary recovery alone is in the order of 30%, where oil
is produced by natural reservoir pressure together with artificial lift methods, e.g. pumping. With secondary recovery
processes (e.g. water flood) this can increase to 50%. Tertiary recovery methods (EOR) extend the field lifecycle and
production and it is now generally accepted that RF can be as high as 80% (Yee et al., 2007; Iwere et al., 2012). Typical
EOR methods include thermal recovery methods such as steam injection (Penney et al., 2005; Yongbin et al., 2010), miscible
gas injection such as CO2 injection (Bonis 2012; Sahin et al., 2012) and chemical injection methods such as polymer,
surfactant and alkali injection (Zhang et al., 2005; Srivastava et al., 2009).

Currently, there are an estimated 370 EOR projects around the world; the type and distribution can be seen in Figure 1. These
EOR projects were responsible for combined production of approximately 3.7 million barrels of oil per day in 2012
(Kootungal, 2012). A summary of these projects is as follows:

 Thermal methods comprise 48% of EOR projects where steam is used to reduce the viscosity of crude oil in order to
allow better transport through and out of the reservoir to surface. These methods tend to require higher than usual
levels of process chemicals like corrosion inhibitors for high temperature steam breakthrough in producing wells,
more water treatment chemicals due to the propensity of invert emulsion formation, more effective scale inhibitors
2 SPE 164078

due to silica based scales and more aggressive conventional scale formation driven by the increased kinetics of
formation due to higher temperatures.
 Miscible gas injection projects make up 40% of EOR projects. Gases such as CO2 are injected into the reservoir and
mix with crude oil, thereby decreasing viscosity and increasing recovery. Soluble surfactants can also be added to
water slugs, thus increasing viscosity of the injection front followed by gas injection thus improving sweep, e.g.
Foam Assisted Water Alternated Gas (FAWAG) (Blaker et al., 2002).
 Chemical EOR (cEOR) makes up 11% of EOR projects globally. There are many different types of cEOR and they
involve the individual or combined injection of:
o Surfactants that lower the surface interfacial tension (IFT) between the injected water and crude oil in the
reservoir and/or change wettability of the reservoir rock surface allowing ‘desorption’ of crude oil.
o Water-soluble polymers are added to injection water to increase viscosity, allowing more efficient
displacement of crude oil from the reservoir.
o Alkaline chemicals such as sodium hydroxide or sodium carbonate are added to the injected water and
saponify natural surfactants (in-situ surfactant generation) in crude oil, creating water-soluble soap that
increases pH and mobility of the crude oil.
 Other methods make up approximately 1% of EOR projects and include methods such as microbial EOR (mEOR).

This paper focuses on the challenges created by the injection of miscible gases and chemicals during EOR and cEOR
processes caused by breakthrough into producing wells.

Processing Challenges Caused by cEOR Methods

The injection of surfactants, alkaline chemicals and polymers can seriously affect the ability to process the produced fluids.
While the efficacy of cEOR programs on increasing RF cannot be disputed, the effect on the process upon breakthrough is
often underestimated and results in significant challenges to retrospectively install process equipment and/or production
chemical strategies.

The most common classes of chemical injected for cEOR include:

 Sulfonates - secondary alkane sulfonates, α-ether sulfonates, petroleum sulfonates

 Ethersulfates - tributylphenol with ethylene oxide ether sulfate, alkylaryl ether sulfates with ethylene oxide
 Ethersulfonates - alkylaryl ether sulfonates with ethylene oxide
 Ether acetates - alkyl and alkylaryl ether acetates with ethylene oxide
 Polyacrylamide (PAM) and Hydrolyzed Polyacrylamide (hPAM)

The environmental profile of these cEOR chemicals can be poor. This is especially the case for PAM and hPAM where it is
very typical for operators to reuse the ‘contaminated’ produced water, adding fresh polymer before reinjecting the treated
water.

Oil and Water Separation

The formation of persistent and stable emulsions should come as no great surprise when one considers the chemistry of
surfactants and polymers injected for cEOR. As these injected chemistries break through into producing wells, the strong
chemical interaction of polymers,especially surfactants, attracts them to the oil/water interface, stabilizing emulsions and
preventing agglomeration, creaming and even settling. PAM and hPAM also introduce higher viscosity to the produced
fluids, which (according to Stoke’s Law) affects settling efficacy. It is clear that due to breakthrough of these cEOR products,
new demulsifier and water treatment chemistries are required. It is common for chemical consumption of emulsions
stabilized by cEOR breakthrough to be higher than those associated with primary or secondary recovery. A thorough review
of the mechanics and chemistry of emulsion formation and processing challenges has been given in the literature (You et al.,
2007; Zheng et al., 2011).

Water treatment is another critical process in the separation of oil and water. Polishing oil out of produced water is critical for
environmental impact (e.g. offshore discharge), for reinjection quality, reuse for cEOR methods and for fiscal reasons. The
presence of cEOR products in produced water negatively impacts the ability for produced water treatment. Experimentation
has shown that if PAM concentrations are in the order of 1g per litre (1,000 ppm/v), then it takes the same concentration of
clarifier chemical to achieve efficient deoiling of the produced water. Thorough studies are available in the literature
evaluating the effect of hPAM on the efficacy of mechanical water treatment processes such as hydrocyclones and IGFs
(Wang et al., 1999; Khan et al., 2009; Zheng et al., 2011).
SPE 164078 3

Northern Alberta Case History 1

This case history tracks the continual improvement cycle for the fluid separation process of a heavy oil/oil sands production
facility in Northern Alberta over a period of three years. During this time, a cEOR polymer flood pilot was running when
breakthrough into the producing wells occurred. This posed a major challenge for the operator of this 13 to 16° API crude oil
field.

Demulsifier was injected in two separate locations. The first was at a series of injection points upstream of the gathering
stationsin the field where ambient temperatures could reach sub zero conditions (<-40°C); the second injection location was
at the battery receiving facility where heating increased temperatures to approximately 100°C. The operator required a very
strict <0.5% BS&W on the crude exiting any of the four treater tanks. To further complicate issues, crude oil viscosity ranged
from 500 to 5,000 cP with the crude oil coming from a relatively shallow play, resulting in wellhead temperatures barely
reaching 20°C.

Fluids arrived from the field at an inclined, two-phase separator (SKUD) and from there passed through a heat exchanger to a
free water knockout vessel. Fluids then passed through four parallel heater treaters to sales. Various recycle and skimming
operations were in operation and have been indicated in Figure 2.

Injection of two different demulsifier products in the field and at the battery occurred for several years (both were mixtures of
EO/PO block polymers and alkoxylated resins) with trouble-free operation – the Lease Automatic Custody Transfer (LACT)
BS&W was below the 0.5% specification. However, in mid-2007, it was a requirement to not only move to a single
demulsifier product for application in the field and the battery, but also to decrease the BS&W to 0.2%. The reasons for these
changes were to 1) ease logistical and operational issues by moving to one product and 2) the BS&W specification was to
have more of a buffer between the actual BS&W of the LACT and the sales specification of 0.5%. Simultaneous with these
changes, the polymer pilot flood was ongoing and breakthrough of polymer into some of the producing wells was occurring.
By 2009, a significant amount of polymer was being measured in the production battery vessels - see Figure 3. The polymer
flood utilized three different products, ranging from a very high molecular weight 30% anionic charge PAM, to an hPAM of
20 to 22 million Daltons molecular weight with very little anionicity.

The separation challenge was exacerbated by the formation of large scale deposits on the surfaces of the heat exchange
devices (this is explained in more detail below in the sludge formation section), which resulted in less heat being transferred
to the fluids in the vessels. This meant more challenging separation and the requirement for any injected demulsifier to have
the ability to ‘cold treat’ the produced fluids.

In order to transition to a more suitable demulsifier product to cope with the challenge of cEOR polymer breakthrough, a
redesign of the bottle test was required to better simulate real field conditions and passage fluids through the multi-stage
process. Based on the results, a new, single demulsifier product was identified that was a blend of EO/PO block polymers and
an alkoxylated amine. The new product was very good for fast water drop, and the BS&W of the SKUD and FWKO were
much improved; however, BS&W was not as low as 0.2%. During this time there were significant operation issues with the
functionality of the heater treaters. In nearly all cases, at least one of the treaters ran colder than 100°C - this has been
explained below in more detail. In order for a single demulsifier product to be as effective as possible, the temperature and
operation of the treaters needed to be as smooth and constant as possible.

It was found that by using the same chemistries but with a dilution of the product, there were some improvements to the
BS&W and treatments. Just after application of a diluted product (30% active instead of 50% active), BS&W results became
significantly better. It was also proposed that better penetration into the oil was occurring via the increased solvent content of
the demulsifier, which allowed for more effective treatment of the crude oil even at lower temperatures and significant
polymer breakthrough.

Some general plant trends have been plotted in Figure 4 that shows the evolution of the LACT BS&W with total oil and
water flow rates. Figure 4 illustrates the relationship between the LACT BS&W and the volume of water the field produced
and handled at the battery andit can be seen that as the produced water rates increased, the LACT BS&W also increased. This
would suggest that the water handling limitations of the battery are being exceeded, and that the plant has become more
challenging to operate with the increased volume of water. Residence times would also significantly decrease as the produced
water rates have increased from 8,000 to 16,000 m3/day (50,500 to 101,000 bbl/day). While the EOR polymer injected was
not measured frequently during this time period (by comparison with Figure 3), it can be seen that the significant up-kick in
BS&W during 2007 is contemporaneous with the onset of EOR polymer breakthrough.

Figure 5 shows that optimization of the original demulsifier combination injection rate had occurred from 2005 to 2007, and
it was possible to reduce the overall demulsifier consumption from 150 to 90 ppm/v (based on oil). During this time, the
4 SPE 164078

BS&W trends remained constant at approximately 0.3%. The big increase in demulsifier injection rate in the second half of
2007 was because of a change to a more dilute product, in an attempt to gain better penetration into the crude oil of
demulsifier to enable cold treating. Figure 4 shows a slow increasing trend in the LACT BS&W from early 2005 to the end of
2006, and this could be explained by increasing water production rates and generally decreasing demulsifier injection rates.
At the beginning of 2007, an increasing trend was experienced in the LACT BS&W; this was commensurate with a marked
increase in the produced water rate and cEOR polymer breakthrough. By June of 2007, the new demulsifier product was
introduced. Even though further increases in produced water rates occurred, as well as significant increases in cEOR polymer
breakthrough, the increasing trend in BS&W was halted where it has maintained at an average of 0.4%. Therefore, arguably,
whilst on face value the change in demulsifier did not decrease the BS&W to the target of 0.2%, the change was positive
because it ensured faster water drop and more efficient separation at the plant even with massive increases in gross fluid
throughput and cEOR polymer residuals. This resulted in maintaining a constant BS&W.

Northern Alberta Case History 2

This case history concerns a conventional heavy oil property located in northern Alberta. Initial field production utilized
primary techniques, but in 2011 a field-wide ASP flood was initiated after a very successful pilot trial a few years earlier of a
polymer flood alone. It was from this pilot flood that much direct experience was gained on the influence that this cEOR
technique had on fluid separation. This field produces just over 1,000 barrels of 16° API crude oil per day from the Bluesky
Formation.

The initial polymer flood was run for almost two years, after which the RF was determined to have increased from 3 to 5%
up to 18% of the STOIIP. The ASP flood is expected to increase oil production to >3,000 barrels of oil per day, with future
developments targeting a production rate of 10,000 barrels of oil per day.

Before the initial proof of concept polymer flood, typical demulsifier chemistry for this region was injected for several
trouble-free years. Soon after polymer breakthrough however, production issues highlighted the need for a more effective
chemistry. The problems were manifested as an interface build-up in tanks and FWKO vessels; this even led to wet export oil
in some circumstances. The objective was to seek a new product capable of giving dryer oil cuts and, if possible, reduce the
interface pad seen in the field process vessels.

Produced fluids travelled from surrounding horizontal wells to the main processing battery via four main in-field trunk
pipelines. The fluids commingled at the battery and were processed via two parallel separation vessels, where they resided for
around two hours, separating the bulk of the free water. It should be noted that these could be operated at very different
temperatures from each other, ranging from 50 to 90ºC. Following the vessels, partially dehydrated fluids flowed to a series
of treatment tanks. Crude resided here for approximately 24 hours, further treating the oil before being pumped for final
treatment at a polishing treatment facility. Treated crude exiting the battery, destined for the polishing facility, under normal
conditions had a BS&W of 1%. From the polishing facility, the partially dehydrated crude oil had to travel 1 km from the
battery and enter a first stage separator. It was in this vessel that many of interface issues and stable emulsion pads were
observed. Crude resided in this vessel at 80ºC for two hours before entering a series of treatment tanks for a further 24 hours
at 70ºC. Finally, treated crude entered one of two shipping tanks for LACT.

Demulsifier was injected in the field wells and flowlines, a long way upstream of the battery between 150 and 200 ppm/v
depending on field conditions. Further demulsifier injection took place before crude reached the battery at 40 ppm/v.

Focus for testing was directed at improving crude dehydration and interface. Initial ratio testing showed reasonable treatment
to be attained from top and mid cuts when injecting 200 ppm/v of the incumbent product (a mixture of diepoxide, amine
polyester and amine block-polymer with RSNs ranging from 4 to 19). A series of blended demulsifiers and individual bases
was assessed with many treating as well as the incumbent product. Therefore, it was deemed necessary to perform mixed cuts
in conjunction with top or mid cuts to help better differentiate crude dehydration efficacy between blends.

From this analysis it was seen that a blend of diepoxide, amine block-polymer and amine polyester worked the best,
consistently leaving far less emulsion in the mixed cuts. In general, interface qualities in samples treated with this blend were
shown to be similar to those treated with the incumbent. All tested bases seemed to leave a bag at the interface that generally
broke up on a gentle swirl/twist of the bottle, but did show slight variations in size from blend to blend.

The new blend was absent of any wetting agent. This was of some concern, as the long-standing incumbent contained a high
percentage of amine block-polymer and to remove a wetting agent completely may have had adverse affects in the field not
visible in the bottle tests. To address this, several blends were tested adding a series of high RSN chemistries to the base
blend. This gave a final formulation of diepoxide, amine block-polymer, amine polyester and a nonyl acid catalyzed resin.
This product gave a very dry top oil, dryer than the incumbent and the new blend in the absence of resin, as well a giving a
SPE 164078 5

sharper interface. It also still showed the dryer mixed cuts and gave faster separation at low doses.

Foamed Surfactant Case History

The use of surfactants in combination with gas and water injection to create foams has been covered well in the literature
(Hoefner et al., 1992; Krause et al., 1992; Hanssen et al., 1994; Aarra et al., 1996, Svorstøl et al., 1997; Aarra and Skauge,
2000; Blaker et al., 2002). Essentially, injection of water and alternating gas is prone to premature gas breakthrough due to
fingering of the gas through the water. Injecting surfactant with the water followed by gas injection creates foam. Foam
reduces the gas relative mobility and leaves the water relative mobility unaltered. The result is that this method forces
injected gas to give piston like displacement and, therefore, better sweep through the reservoir by blocking the high
permeable zones and changing the flooding pattern in the reservoir.

Injection of several hundred cubic meters of diluted surfactant (sodium α-olefin sulfonate) was injected into the reservoir via
two wells. After only a few weeks, gas breakthrough was observed and the operation was deemed unsuccessful and further
plans to inject more surfactant solutions were cancelled. As the gas breakthrough occurred, so too did the surfactant-loaded
fluid which began to significantly affect the efficacy of the separation train in the surface processing equipment.

The water quality of two particular producing wells (connected directly to one of the surfactant-loaded injectors) became
increasingly poor. The overall effect was to have a significant impact on the overall discharge quality of the oil in water on
the field. A plot of oil in water from the field has been given in Figure 6. The different trends in oil-in-water evolution have
been designated letters (A through F) and explanation of the trends is as follows:

A) Demulsifier A was injected during this time giving an oil-in-water average of 28 mg/L.
B) A new demulsifier product was deployed (Demulsifier B) that gave a significant improvement in the oil-in-water
discharge, giving an average of 12 to 13 mg/L.
C) The sodium α-olefin sulfonate was injected at approximately 1.500,000 barrels of cumulative water production, and very
soon after this injection the oil-in-water climbed steadily towards 50 to 60 mg/L. Attempts were made to determine
concentration of sulfonate in the produced water using ICP to detect residual sulfur, but it was not possible to determine
this due to the high background of sulfur from sulfate coming from the injected seawater.
D) The increasing oil-in-water trends had to be addressed due to discharge concerns so a combined strategy was
implemented after cumulative water production of 2,250,000 barrels. The hydrocyclone units were reconditioned,
flocculent (acrylic copolymer) was injected and demulsifier injection optimized. It can be seen that this had a positive
effect on reducing the oil-in-water (c.20 mg/L) but not to the degree required. It necessitated choking back the wells that
produced the most surfactant breakthrough, which immediately gave excellent water quality at the pre-surfactant
breakthrough levels of c.10 mg/L.
E) This period saw even more surfactant breakthrough on additional producer wells, and again the oil-in-water readings
steadily increased above 40 mg/L. Again, the strategy after this rapid increase in oil-in-water was to optimize the
demulsifier and flocculent injection, which only had a limited effect as by this time there was a significant amount of
surfactant breaking through into the process system from several different wells.
F) The decline after 4,000,000 barrels of cumulative water production was only realized, again, after choking back the wells
that were experiencing significant surfactant breakthrough.

Overall, it can be seen from this case history that the breakthrough of surfactant chemical, in this case a sulfonate, seriously
affected the separation process, in particular the quality of the produced water. Chemical strategies involving demulsifier and
flocculant/deoiler only had a limited effect; such was the severity of surfactant interaction with the produced water quality.

Sludge and Mineral Scale Formation

The effect of organic-based sludge deposits in oil and water separation and processing systems has been reported (Wylde et
al., 2011; Zheng et al., 2011) that can be attributed to the breakthrough of cEOR polymer breakthrough. This is particularly
challenging in areas where heat exchangers are used to process the crude oil, as these surfaces are particularly prone to
organic fouling due to the decomposition temperature of hPAM and PAM.

The above “Northern Alberta Case History 1” shows the catastrophic influence that polymer breakthrough can have on heat
exchange surfaces, which in this case were fire tubes. Deposit build-up on the surfaces was analyzed intensively over a long
period of time. Great detail was placed on several of the samples that were high in organic content using FTIR in
combination with various wet chemistry extraction techniques to isolate the water soluble, hexane and pentane soluble
portions – Figure 7. The analysis showed weak amide peaks present in the hexane extraction. Water extraction showed a very
clear presence of EOR polymer as the sample had high water solubility but also strong IR peaks indicating acrylamide and
cellulose. Overall, this suggested a high concentration of EOR polymer. It was concluded that a majority of the organic and
volatile components were in fact EOR polymer and associated breakdown products.
6 SPE 164078

There were several different pieces of process trend monitoring evidence that showed a direct correlation to the amount of
EOR polymer breakthrough detected at the battery (Figure 3) and the commencement of failure, as well as frequency of
failure of the fire tubes (Figure 8). This also can be attributed to increasing mineral scale propensity, with the EOR polymer
coating or partially coating the fire tubes, resulting in much higher skin temperatures and a more challenging scaling regime.
The fire tube solid deposits analyzed showed less and less inorganic scale content and more organic content (Figure 9),
proving the scale inhibitor program to be effective. Furthermore, the inorganic content contained decreasingly less mineral
scale and more evaporite minerals and silica content. There were, therefore, two challenges with the fire tubes: mineral scale
and cEOR polymer deposition. However the two were very interlinked. Once polymer breakthrough occurred, it adhered to
the surface of the fire tubes because PAM breaks down much over 100°C. The ‘normal’ skin temperature of the fire tubes
was 400 to 450°C, and the only way PAM will not adhere to the tubes is if they are turned off; however, the hotter the tubes
the faster the build-up of polymer. Once the tubes became coated, they had to operate at even higher temperatures to yield the
same bulk temperature in the treaters; therefore, the skin temperature increased even further and the adsorption of polymer
became even more problematic. So, too, did mineral scale and the incumbent phosphonate product simply stopped working
because the average skin temperatures resulted in molecular scission of the inhibitor. Once the fire tubes got above 500°C,
evaporation was occurring on the fire tube surface of water (even at the operating pressure) and evaporate scales formed. The
mineral scale situation was addressed by changing to a polymeric product that served to increase in the lifetime of the fire
tubes, i.e. mineral scale aspects and failures are related only to cEOR polymer build up. The next point for mitigation of
cEOR polymer on the fire tubes was to get a high performance demulsifier so that ‘cold’ treating of the process was possible.
Cold meant below 100°C, i.e. get the bulk temperature below the PAM thermal stability and significantly lower the skin
temperature of the tubes, as well as dehydrating the crude oil more efficiently so that residual water did not contact the fire
tubes and lead to evaporite and exotic scale deposition. It has been previously reported in the literature that cEOR programs
can exacerbate mineral scaling due to a variety of mechanisms (e.g. Krumrine et al., 1985).

Processing Challenges Caused by CO2 Injection

The experience detailed in this case history comes from treating one of the largest CO2 floods in operation in SE
Saskatchewan, Canada. The injection of CO2 lowers crude oil viscosity but can give very quick breakthrough into producing
wells and can bring processing challenges. The crude oil in place is typically 27.5° API, thoughbenefits from CO2 injection
can be found in some wells with an API as high as 31 to 33°. The benefits from this particular method of EOR have become
very clear over the years that it has been in operation. In 2000, production from the field was approximately 2,500 to 2,600m3
per day. In 2012 with the CO2 flood in full effect, the production rate increased to approximately 4,500m3 per day.

There are more than 320 wells experiencing breakthrough of significant CO2. Of these wells, approximately half are ESP
pumped wells and half rod pumped. As soon as CO2 breaks through into a production well, the concentration quickly climbs
to an average of 30%; on some wells it can be as high as 90%. All gas is recycled and reinjected, thus topping up the neat
imported CO2.

The major flow assurance and production chemistry challenges are asphaltenes, corrosion and mineral scale deposition.
Generally, most ESP pumped wells are on a scale and corrosion inhibition program with some on an asphaltene inhibitor
program. A majority of rod pumped wells are on an asphaltene and corrosion inhibition program with only some on a scale
program.

Another challenge is one of surveillance; the breakthrough pattern and rate of CO2 is very unpredictable, and the flow
assurance implications can be as swift as the CO2 breakthrough itself. Not detecting the breakthrough in only a number of
weeks can result in a significant production decline. An aggressive approach has been successfully adopted where the gas oil
ratio (GOR) is used as an indicator of CO2 breakthrough (30 is typical and approaching 300 indicates significant CO2
breakthrough) and mitigation strategies are then put into effect.

The ability to introduce production chemicals to the source of the flow assurance/integrity problem is problematic. Wells are
typically very long, horizontals that (if at all) only have capillary strings to the submersible pump - the horizontal section is
sometimes impossible to treat. To complicate matters even further, the wells typically are multilateral and can have two or
sometimes three horizontal legs.

The most common type of fouling mechanism is deposition of asphaltenes. As a result, the wells are produced with a high
fluid level to maintain backpressure on the wellbore to minimize pressure drop up the tubing, thereby lowering the propensity
for asphaltene precipitation.

Location of asphaltene deposition can be variable and ranges from the build section to the wellhead. More asphaltene
SPE 164078 7

deposition is seen on rod pumped wells compared to ESP wells due to the higher pressure drop. The operator evaluated co-
rod pumps without couplers and this did sometimes help the situation, as did moving to bigger tubing (2⅞” to 3¼”).

SARA and pour point analysis was performed on many wells on this field but showed vastly variable and unpredictable
trends with no real patterns to assist with development of the flow assurance strategy. Typically, the crude contains 10%
asphaltenes and 2% wax with a pour point of up to 15°C.

Usually when a restriction is detected in a well, the first remedial action taken is application of an organic solvent to treat
asphaltene deposition. After the initial asphaltene deposit has been removed, a dispersant technology is deployed (ideally
through) capillary strings to prevent deposition rather than allow it and therefore more remediation. The inhibitor contains a
small amount of polymeric combination wax and asphaltene inhibitor with a larger component of asphaltene dispersant in a
winterized, freeze-proof solvent matrix. This product is generally applied at 200 ppm/v on the ESP wells (which are usually
completed with capillaries). For rod pumped wells, capillary strings are not normally installed; therefore, backside injection
is the method of deployment. For this method of injection, the asphaltene inhibitor differs slightly by being weighted with
glycol for reliable distribution down the backside of the tubing and into the produced fluids. It is more heavily loaded with
surfactants and impregnating agents to allow better penetration into the crude oil and access to the growing asphaltene
crystals. For this application, using a crystal modifier is not beneficial given that the point of injection is nearly always
downstream of the point of asphaltene flocculation, and injection of this product can range from 200 to 2,000 ppm/v.

Scale deposition is dominated by calcium sulfate scales (CaSO4, CaSO4.½H2O, CaSO4.2H2O) on breakthrough wells,
whereas the dominance on non-breakthrough wells is carbonate scale (CaCO3). This is driven by pH which itself is driven by
the degree of CO2 saturation in the produced water. The typical pH for natural produced water is approximately 6.5. On wells
experiencing >50% CO2 breakthrough, the pH can be as low as 4.5 due to the formation of carbonic acid due to the high
dissolved CO2 content. A typical water chemistry has been included in Table 1. It can be seen that the TDS is relatively high,
as is calcium, sulfate and bicarbonate ion concentrations.

Scale deposition most commonly occurs on the build section of the well; therefore, injection of scale inhibitor has to occur
upstream of the ESP as it costs in excess of CDN$150K to pull an ESP. Injection rates of scale inhibitor are between 20 to 25
ppm/v, and a novel phosphorous-based chemistry is neutralized to a non-corrosive pH and carried in a winterized solvent-
based matrix. While the lowering of pH only affects the CaSO4 scaling propensity slightly, it can have a dramatic effect on
the scale inhibitor efficacy. Lower pH tends to keep the scale inhibitor chemistries associated (increased pH dissociates scale
inhibitors) and therefore lowers the number of available active sites for complexation with the growing scale crystals. In the
past, trials have been performed where scale inhibitor was put into injected water of water alternative gas (WAG) wells.
There was little or no scale inhibitor detected in the producing wells, and certainly no optimization of the scale inhibitor
injected into the producing wells was possible; therefore, no benefit to continuing injection into the WAG wells were seen.
Acidic dissolvers are used for carbonate scale remediation, and if this does not work then a gypsum converter is deployed
that is composed of blends of organic acids and caustic converting agents.

Batch corrosion inhibitors are applied due to the challenges of protecting the tubing in the well using continuous injection,
which either has to be through annular injection or through capillary injection (which more often than not are either not
installed or are being used for asphaltene inhibitor injection). The highest performing product, determined through very
detailed corrosion inhibitor laboratory testing performed over several years of optimization, is winterized, oil-soluble, batch
corrosion inhibitor composed of alkyl pyridine quaternary ammonium compounds, imidazoline and synergists with glycol
weighting for better dispersion in the fluid column. Furthermore, a bespoke blend of demulsifier components is added to
ensure no dehydration issues occur upon returning the batch treatment from its circulation period.

Batch corrosion inhibitor is introduced by a pressurized tank truck, followed by a flush of produced water and a recirculation
period to distribute the corrosion inhibitor evenly over the whole tubing. The integrity management strategy taken for wells is
based upon where they are in their production cycle. A film 3 mil thick is applied (100 to 120 liters of corrosion inhibitor) to
newly drilled wells; thereafter a 0.5 mil thickness ‘maintenance’ film is applied (20 to 30 liter batches of corrosion inhibitor).
The frequency of application of maintenance batch treatments is production rate dependent. For wells making >60m3 per day
(gross), batch treatments are applied once per week. For wells making between 31 and 60 m3 per day (gross), treatments are
applied every two weeks. And for wells making <30m3 per day (gross), treatments are applied monthly. This is an average
taken from all the wells and monitoring techniques employed on each well to determine the actual and optimized frequency
of treatment. This includes the typical wet chemistry methods (iron counts and corrosion inhibitor residuals), as well as
corrosion coupon analysis and online LPR probe measurements.

Summary and Conclusions

The conclusions from these case histories is that EOR methods can often result in processing, flow assurance or integrity
8 SPE 164078

related challenges that otherwise would not have manifested. The case histories above exemplify this:
 Water and oil separation challenges due to polymer and surfactant stabilized emulsions;
 Oily water flocculation problems caused by surfactant breakthrough;
 Fouled heat exchange surfaces caused by degraded polymer deposition;
 Mineral scale deposition exacerbated by alkaline surfactant breakthrough;
 Asphaltene deposition increased corrosion and increased scale inhibitor requirements caused by injection and
breakthrough CO2.

It is possible to mitigate challenges induced by EOR applications but the solutions are often complex and require time to
identify and solve. Furthermore, breakthrough of chemicals or gases used in EOR is not predictable, nor is it a linear problem
that becomes more challenging with time. It is dynamic, unpredictable and requires an equally dynamic and robust
monitoring and mitigation strategy.

Acknowledgements
The authors would like to thank the Society of Petroleum Engineers for the opportunity to present this work and to Clariant
Oil Services for allowing this work to be published.

References

Aarra, M.G. and Skauge, A. (2000) Status of Foam Applications in North Sea Reservoirs. Paper 22 presented at the Annual International
Energy Agency Workshop and Symposium, Edinburgh, UK, 19-22 September

Aarra, M.G., Skauge, A., Sognesand, S., and Stenhaug, M. (1996) A foam Pilot Test Aimed at Reducing Gas Inflow in a Production Well
at the Oseberg Field. Petroleum Geoscience, 2, 125

Blaker, T., Aarra, M.G., Skauge, A., Rasmussen, L., Celius, H.K., Martinsen, H.A., and Vassenden, F. (2002) Foam for Gas Mobility
Control in the Snorre Field: The FAWAG Project. SPE 78824. SPE Reservoir Evaluation and Engineering 5(4), 317-323

Bonis, M. (2012) Managing the corrosion impact of dense phase CO2 injection for an EOR purpose. SPE 161207. Abu Dhabi International
Petroleum Conference and Exhibition, Abu Dhabi, UAE, 11-14 November

Hanssen, J.E., Holt, Torleif, Surguchev, L.M. (1994) Foam Processes: An Assessment of Their Potential in North Sea Reservoirs Based on
a Critical Evaluation of Current Field Experience. SPE 27768. SPE/DOE Improved Oil Recovery Symposium, Tulsa, Oklahoma, 17-20
April

Hoefner, M.L., Evans, E.M., Buckles, J.J., and Jones, T.A. (1995) CO2 Foam: Results from Four Developmental Field Trials. SPE 27787.
Reservoir Engineering, 10(4), 273-281

Iwere, F.O., Heim, R.N., and Cherian, B.V. (2012) Numerical Simulation of Enhanced Oil Recovery in the Middle Bakken and Upper
Three Forks Tight Oil Reservoirs of the Williston Basin. SPE 154937. SPE Americas Unconventional Resources Conference, Pittsburgh,
Pennsylvania, USA 5-7 June

Khan, R., Lee, J., Sams, G. (2009) Produced Water Viscosity Effects on IGF Sweeping Efficiency. Annual Produced Water Society
Seminar, 21-23 January, Houston, TX, USA

Koottungal, L. (2012) Worldwide EOR Survey. Oil and Gas Journal, 2 April

Krause, R.E., Lane, R.H., Kuehne, D.L., and Bain, G.F. (1992) Foam Treatment of Producing Wells to Increase Oil Production at Prudhoe
Bay. SPE 24191. SPE/DOE Enhanced Oil Recovery Symposium , Tulsa, Oklahoma, 22-24 April

Krumrine, P.H. Mayer, E.H., and Brock, G.F. (1985) Scale Formation During Alkaline Flooding. SPE 12671. Journal of Petroleum
Technology, 37(8), 1466-1474

Penney, R., Moosa, R., Shahin, G., Hadhrami, F., Kok, A., Engen, G., van Ravesteijn, O., Rawnsley, K., and Kharusi, B. (2005) Steam
Injection in Fractured Carbonate Reservoirs: Starting a New Trend in EOR. IPTC 10727. International Petroleum Technology Conference,
Doha, Qatar, 21-23 November

Sahin, S., Kalfa, U., and Celebioglu, D. (2012) Unique CO2-Injection Experience in the Bati Raman Field May Lead to a Proposal of
EOR/Sequestration CO2 Network in the Middle East. SPE 139616. SPE Economics & Management, 4(1), 42-50

Srivastava, M., Zhang, J., Nguyen, Q.P., and Pope, G.A. (2009) A Systematic Study of Alkali Surfactant Gas Injection as an Enhanced Oil
Recovery Technique. SPE 124752. SPE Annual Technical Conference and Exhibition, New Orleans, Louisiana, 4-7 October
SPE 164078 9

Svorstøl, I., Blaker, T., Tham, M.J., and Hjellen, A. (1997) A Production Well Foam Pilot in the North Sea Snorre Field-Application of
Foam to Control Premature Gas Breakthrough. Presented at the 9th European Symposium on Improved Oil Recovery, The Hague, 20-22
October

Wang, Z., Zhao, L., Li, F., He, J., and Chen, L. (1999) Characteristics Study of Hydrocyclone Used for Separating Polymer-Flood
Produced-Water. International Offshore and Polar Engineering Conference, Brest, France, May 30-June 4

Wylde, J.J., Allan, K., McMahon, J., and Mayner, S. (2011) Scale Inhibitor Application in Northern Alberta - A Case History of an Ultra
High Temperature Scale Inhibitor Solution in Fire Tube Heater Treaters. SPE 141100. Presented at the International Symposium on
Oilfield Chemistry, The Woodlands, TX, USA, 11-13 April

Yee, H.V., Kechut, N.I., and Razak, W.N.A.W (2007) Enhanced Oil Recovery Potential of Heavy-Oil Fields in Africa. SPE 108513.
International Oil Conference and Exhibition, Veracruz, Mexico, 27-30 June

Yongbin, W., Desheng, M., Shangqi, L., Hongzhuang, W., and Xin, Z. (2010) EOR of Superheated Steam Injection in Shallow Heavy Oil
Reservoir: A Case Study. SPE 131224. International Oil and Gas Conference and Exhibition in China, Beijing, China, 8-10 June

You, Q., Zhao, F., Wang, Y., and Mu, L. (2007) Comparison of the properties of injected and released polyacrylamide in polymer flooding.
Journal of Beijing University of Chemical Technology (Natural Science Edition) 34(4): 414-417

Zhang, Y.P., Huang, S., and Dong, M. (2005) Determining the Most Profitable ASP Flood Strategy for Enhanced Oil Recovery. Journal of
Canadian Petroleum Technology, 44(2)

Zheng, F., Quiroga, P., and Sams, G.W. (2011) Challenges in Processing Produced Emulsion from Chemical Enhanced Oil Recovery -
Polymer Flood Using Polyacrylamide. SPE 144322. Presented at the SPE Enhanced Oil Recovery Conference, Kuala Lumpur, Malaysia,
19-21 July
10 SPE 164078

TABLE 1: PRODUCED WATER CHEMISTRY AND PROPERTIES FROM THE CO2 EOR CASE
HISTORY IN SOUTH EAST SASKATCHEWAN

Analysis Result Unit

Alkalinity Total 836 mg/L
Bicarbonate 982 mg/L
Carbonate <1 mg/L
Chloride 43,600 mg/L
pH 6.61
Barium <0.5 mg/L
Calcium 2,993 mg/L
Magnesium 918 mg/L
Potassium 400 mg/L
Sodium 25,315 mg/L
Strontium 79.4 mg/L
Sulfate 2,101 mg/L
Iron 0.23 mg/L
Manganese 0.04 mg/L
Specific Gravity 1.0588
Cation Sum 1,381.02 mEq/L
Anion Sum 1,457.02 mEq/L
Ion Balance (Cation/Anion) 0.95
Ion Balance (% Difference) -2.68
Total Dissolved Solids 82,811 mg/L
SPE 164078 11

FIGURE 1: TYPE AND DISTRIBUTION SUMMARY OF THE c.370 EOR PROJECTS CURRENTLY IN OPERATION
AROUND THE WORLD

FIGURE 2: BATTERY SCHEMATIC FROM NORTHERN ALBERTA CASE HISTORY 1

12 SPE 164078

FIGURE 3: POLMYER BREAKTHROUGH MEASURED IN THE BATTERY SEPARATION VESSELS OF NORTHERN

ALBERTA CASE HISTORY 1

FIGURE 4: CRUDE OIL AND PRODUCED WATER RATES VERSUS THE LACT (SALES) BS&W FOR NORTHERN
ALBERTA CASE HISTORY 1
SPE 164078 13

FIGURE 5: DEMULSIFIER USAGE VERSUS THE LACT BS&W FOR NORTHERN ALBERTA CASE HISTORY 1

FIGURE 6: OIL IN WATER QUALITY FROM THE FOAMED SURFACTANT CASE HISTORY
14 SPE 164078

FIGURE 7: FTIR ANALYSIS OF THE EXTRACTED FIRETUBE SOLID DEPOSITS. TOP LEFT: WATER SOLUBLE
EXTRACTION; TOP RIGHT: WATER INSOLUBLE EXTRACTION; MIDDLE LEFT: PENTANE SOLUBLE EXTRACTION;
MIDDLE RIGHT: PENTANE INSOLUBLE EXTRACTION; BOTTOM LEFT: HEXANE SOLUBLE EXTRACTION;
BOTTOM RIGHT: HEXANE INSOLUBLE EXTRACTION
SPE 164078 15

3:00 position
700
9:00 position
Treater 3 Fire Tube Skin Temperature (°C) 12:00 position
600

500

400

300

200

100
01/09 03/09 05/09 06/09 08/09 10/09 12/09 02/10 04/10 06/10 08/10 10/10

FIGURE 8: FIRE TUBE SKIN TEMPERATURE MEASUREMENTS OVER TIME - THE VERTICAL RED LINES SHOW THE
FIRETUBE FAILURES RESULTING FROM INCREASED cEOR POLYMER BREAKTHROUGH

100%
Organic
90% Inorganic

80%

70%

60%

50%

40%

30%

20%

10%

0%
01/2010 03/2010 05/2010 06/2010 08/2010 09/2010 11/2010

FIGURE 9: DEVELOPMENT OF ORGANIC AND INORGANIC PHASES ONCE NORMALIZED BY REMOVAL OF THE
AQUEOUS AND VOLATILE PHASE