PETE5103 Enhanced Oil Recovery

1
PETE5103: ENHANCED OIL RECOVERY
(Dr Ngong + Mr Kiven)
General Course outline: Principles of displacement; rock properties fluid

properties in reservoir; phase behavior; displacement efficiencies. Gas methods;
miscible slug; enriched gas; high pressure lean gas; carbon dioxide; nitrogen and
other inserts.
Second Part of the Course Prepared By Mr. KIVEN Paul NSIIDZEYUF
Contents
General Introduction
I: Chemical methods:
1. Micellar-Polymer Methods.
2. Polymer Augmented Waterflood.
3. Permeability Alteration.
II: Thermal methods:
4. Steam Stimulation.
5. Steam Drive.
6. In-Situ Combustion.
PETE5103: Enhanced Oil Recovery Lecture Notes Prepared by Kiven P.

2
General Introduction
Definition of Enhanced Oil Recovery (EOR).
Oil recovery operations traditionally have been subdivided into three stages: primary, secondary,
and tertiary. Historically, these stages described the production from a reservoir in a chronological
sense. Primary production, the initial production stage, resulted from the displacement energy
naturally existing in a reservoir. Secondary recovery, the second stage of operations, usually was
implemented after primary production declined. Traditional secondary recovery processes are
waterflooding, pressure maintenance, and gas injection, although the term secondary recovery is
now almost synonymous with waterflooding. Tertiary recovery, the third stage of production, was
that obtained after waterflooding (or whatever secondary process was used). Tertiary processes
used miscible gases, chemicals, and/or thermal energy to displace additional oil after the secondary
recovery process became uneconomical.
The drawback to consideration of the three stages as a chronological sequence is that many
reservoir production operations are not conducted in the specified order. A well-known example
is production of the heavy oils that occur throughout much of the world. If the crude is sufficiently
viscous, it may not flow at economic rates under natural energy drives, so primary production
would be negligible. For such reservoirs, waterflooding would not be feasible; therefore, the use
of thermal energy might be the only way to recover a significant amount of oil. In this case, a
method considered to be a tertiary process in a normal, chronological depletion sequence would
be used as the first, and perhaps final, method of recovery.
In other situations, the so-called tertiary process might be applied as a secondary operation instead
of waterflooding. This action might be dictated by such factors as the nature of the tertiary process,
availability of injectants, and economics. For example, if a waterflood before application of the
tertiary process would diminish the overall effectiveness, then the waterflooding stage might
reasonably be bypassed.
Because of such situations, the term “tertiary recovery” fell into disfavor in petroleum engineering
literature and the designation of “enhanced oil recovery” (EOR) became more accepted. Another
descriptive designation commonly used is “improved oil recovery” (IOR), which includes EOR
but also encompasses a broader range of activities (e.g., reservoir characterization, improved

3
reservoir management, and infill drilling). Oil recovery processes now are classified as primary,
secondary, and EOR processes.
Primary recovery results from the use of natural energy present in a reservoir as the main source
of energy for the displacement of oil to producing wells. These natural energy sources are solution
gas drive, gas-cap drive, natural waterdrive, fluid and rock expansion, and gravity drainage. The
particular mechanism of lifting oil to the surface, once it is in the wellbore, is not a factor in the
classification scheme.
Secondary recovery results from the augmentation of natural energy through injection of water or
gas to displace oil toward producing wells. Gas injection, in this case, is either into a gas cap for
pressure maintenance and gas-cap expansion or into oil-column wells to displace oil immiscibly
according to relative permeability and volumetric sweepout considerations. Gas processes that are
based on other mechanisms, such as oil swelling, oil viscosity reduction, or favorable phase
behavior, are considered EOR processes. An immiscible gas displacement is not as efficient as a
waterflood and is used infrequently as a secondary recovery process today. Today, waterflooding
is almost synonymous with the secondary recovery classification.
EOR results principally from the injection of gases or liquid chemicals and/or the use of thermal
energy. A number of liquid chemicals are commonly used, including polymers, surfactants, and
hydrocarbon solvents. Thermal processes typically consist of the use of steam or hot water, or rely
on the in-situ generation of thermal energy through oil combustion in the reservoir rock.
EOR processes involve the injection of a fluid or fluids of some type into a reservoir. The injected
fluids and injection processes supplement the natural energy present in the reservoir to displace oil
to a producing well. In addition, the injected fluids interact with the reservoir rock/oil system to
create conditions favorable for oil recovery. These interactions might, for example, result in lower
interfacial-tensions (IFTs), oil swelling, oil viscosity reduction, wettability modification, or
favorable phase behavior. The interactions are attributable to physical and chemical mechanisms
and to the injection or production of thermal energy. Simple waterflooding and the injection of dry
gas for pressure maintenance or oil displacement are excluded from the definition.
EOR processes often involve the injection of more than one fluid. In a typical case, a relatively
small volume of an expensive chemical (primary slug) is injected to mobilize the oil. This primary

4
slug is displaced with a larger volume of a relatively inexpensive chemical (secondary slug). The
purpose of the secondary slug is to displace the primary slug efficiently with as little deterioration
as possible of the primary slug. In some cases, additional fluids of even lower unit cost are injected
after a secondary slug to reduce expenses. In such a case of multiple fluid injection, all injected
fluids are considered to be part of the EOR process, even though the final chemical slug might be
water or dry gas that is injected solely to displace volumetrically the fluids injected earlier in the
process. The intent of EOR is to increase the effectiveness of oil removal from pores of the rock
(displacement efficiency) and to increase the volume of rock contacted by injected fluids (sweep
efficiency)
Idealized Characteristics of an EOR Process
Efficient Microscopic and Macroscopic Displacement
The overall displacement efficiency of any oil recovery displacement process can be considered
conveniently as the product of microscopic and macroscopic displacement efficiencies. In equation
form,
E = ED EV,
Where E = overall displacement efficiency (oil recovered by process/oil in place at start of

process), ED = microscopic displacement efficiency expressed as a fraction, and EV = macroscopic
(volumetric) displacement efficiency expressed as a fraction.
Macroscopic displacement efficiency relates to the effectiveness of the displacing fluid(s) in

contacting the reservoir in a volumetric sense. Alternative terms conveying the same general
concept are sweep efficiency and conformance factor. EV is a measure of how effectively the
displacing fluid sweeps out the volume of a reservoir, both laterally and vertically, as well as how
effectively the displacing fluid moves the displaced oil toward production wells.
It is desirable in an EOR process that the values of ED and EV, and consequently E, approach 1.0.
An idealized EOR process would be one in which the primary displacing fluid (primary slug)
removed all oil from the pores contacted by the fluid (Sor → 0), and in which the displacing fluid
contacted the total reservoir volume and displaced oil to production wells.

5
Several physical/chemical interactions occur between the displacing fluid and oil that can lead to
efficient microscopic displacement (low Sor). These include miscibility between the fluids,
decreasing the IFT between the fluids, oil volume expansion, and reducing oil viscosity. The
maintenance of a favorable mobility ratio between displaced and displacing fluids also contributes
to better microscopic displacement efficiency. EOR processes are thus developed with
consideration of these factors.
Macroscopic displacement efficiency is improved by maintenance of favorable mobility ratios

between all displacing and displaced fluids throughout a process. Favorable ratios contribute to
improvement of both areal and vertical sweep efficiencies. An ideal EOR fluid then is one that
maintains a favorable mobility ratio with the fluid being displaced. Another factor important to
good macroscopic efficiency is the density difference between displacing and displaced fluids.
Large density differences can result in gravity segregation (i.e., the underriding or overriding of
the fluid being displaced). The effect is to bypass fluids at the top or bottom of a reservoir, reducing
EV. If density differences do exist between fluids, this might be used to advantage by flooding in
an updip or downdip direction. Reservoir geology, and in particular geologic heterogeneity, is an
important factor in the consideration of macroscopic displacement efficiency. The effects of
mobility and density differences can be amplified or diminished by the nature of the geology. An
ideal EOR fluid thus is one that has a favorable mobility ratio with the fluid(s) being displaced
and, further, maintains this favorable condition throughout the process. In addition, the density of
an ideal EOR fluid should be comparable with that of the displaced fluid unless flooding can be
performed in an updip or downdip direction.
Practical Considerations.
Fluids that possess the properties required for good microscopic and macroscopic displacement
efficiencies are certainly known or can be developed. A practical concern, however, is that the
fluids are expensive, or for the case of thermal processes, the cost of developing the thermal energy
that the fluids carry is high.
The size of the fluid slugs that are injected and their unit costs become major considerations in the
design of an EOR process. In fact, injected fluid cost and crude oil price (and instability in price)
are the two most important factors controlling the economic implementation of EOR processes.

6
Another consideration is the ease of handling an EOR fluid and its general compatibility with the
physical injection/production system. Highly toxic or corrosive fluids, or fluids that are not readily
injected, are not very amenable to use in EOR processes. While such fluids might be used with
installation of special equipment, the cost is usually prohibitive.
The availability of an EOR fluid is also a consideration. If an EOR process is applied in a major
reservoir, the fluid requirements for that single reservoir can be quite large. And if that process is
widely accepted for application across the country or the world, the volume requirements can
become an important limiting factor for its application.
The implementation and success or failure of an EOR process are always affected by reservoir
geology and reservoir geologic heterogeneities. Processes that are well-understood in a laboratory
environment and properly designed for the reservoir fluids may fail when implemented in the
reservoir because of geologic factors.
Geologic factors may lead to unexpected losses of chemicals or bypassing of fluids because of
channeling in high-permeability zones or fractures. Similarly, fluid movement may be very
nonuniform because of variations in rock properties. Unexpected chemical adsorption can
sometimes occur, causing a deterioration of fluid slugs. Factors of this type, unless properly
identified and understood before the start of a process, will likely cause a project failure. A number
of procedures exist that can be used before implementation of an EOR process in an attempt to
describe the reservoir geology. These procedures include geologic evaluations of well cores and
logs, single-well and well-to-well tracer tests, pressure-transient analysis, and seismic surveys.

7
PART I: CHEMICAL METHODS

CHAPTER 1: MICELLAR-POLYMERS METHOD
1.1 Introduction
Chemical processes involve the injection of specific liquid chemicals that effectively displace oil
because of their phase-behavior properties, which result in decreasing the IFT between the
displacing liquid and oil. In this process, the displacements are immiscible; that is, complete
solubility does not exist between the micellar solution and oil or between the micellar and polymer
solutions.
The micellar-polymer process uses a surfactant to lower the interfacial tension between the injected
fluid and the reservoir oil. The surfactant migrates to the interface between the oil and water phases
and helps make the two phases more miscible. Interfacial tensions can be reduced from
approximately 30 dynes/cm, found in typical waterflooding applications, to 10−4 dynes/cm with
the addition of as little as 0.1% to 5.0% w/w surfactant to water–oil systems. As the interfacial
tension between an oil phase and a water phase is reduced, the capacity of the aqueous phase to
displace the trapped oil phase from the pores of the rock matrix increases. The reduction of the
interfacial tension results in a shifting of the relative permeability such that the oil will flow much
more readily at lower oil saturations. The chemical system, usually called a micellar or
microemulsion solution, typically contains the following components: surfactant, cosurfactant
(which may be alcohol or another surfactant), hydrocarbon, water, and electrolytes. A water
soluble polymer is added to increase viscosity and to attain mobility control during the
displacement process. In the process, injection of the surfactant or micellar solution usually is
followed by injection of an aqueous solution to which polymer has been added to maintain mobility
control—thus the name micellar/polymer process. In most situations, the micellar/polymer process
is implemented as a tertiary displacement near the end of a waterflood. Fig. 1 shows a tertiary
process where the initial oil saturation is Sorw, ROS after waterflooding.

8
Figure 1: Micellar/polymer displacement process (Poettmann 1974).
In some cases, a preflush is injected ahead of the micellar solution to adjust brine salinity or pH.
The preflush solution may contain a sacrificial adsorbent (i.e., a relatively inexpensive chemical
that will be adsorbed on the rock and occupy adsorption sites). The purpose is to reduce adsorption
and loss of the surfactant contained in the micellar solution that will follow. In pilot tests, the
volume of the preflush may be as large as 100% of the flood-pattern PV. This amount is not
practical in a full-scale field application where much smaller volumes must be used. Field
experience has shown that preflushes are often not effective in controlling brine salinity and
divalent-ion content. Thus, a micellar solution generally must be designed to tolerate resident
brine.
A micellar solution is relatively expensive, so a limited volume, or slug size, is used. The micellar
slug must be displaced with a less expensive fluid. Water is not a suitable displacing fluid because
an unfavorable mobility ratio usually exists between water (brine) and a micellar solution.
Therefore, a mobility buffer slug that consists of a solution of polymer in water is used. The
mobility buffer slug, which may range up to 100% of the flood-pattern PV, is then followed by

9
water. The polymer mobility buffer is such an integral concern that the overall process usually is
called a micellar/polymer flood, as indicated in Fig. 1 When the polymer and micellar solutions
are not miscible, there should be a low IFT between the liquids. Some of the micellar solution will
be trapped as a residual phase and lost from the primary micellar slug. Physical processes, such as
adsorption of surfactant and other constituents, degrade or diminish slug volume as the process
proceeds.
In laboratory corefloods, recovery efficiency can be quite high, approaching 90% or more of the
oil present at the start of displacement. Fig. 2 shows a recovery curve from a laboratory tertiary
displacement.
Figure 2: Cumulative recovery curve, laboratory micellar/polymer displacement test (Davis and Jones 1968).
1.2 Surfactants
Surface-active agents, or surfactants, are chemical substances that adsorb on or concentrate at a

surface or fluid/fluid interface when present at low concentrations in a system. They alter the
interfacial properties significantly; in particular, they decrease the surface tension, or IFT. In their
most common form, surfactants consist of a hydrocarbon portion (nonpolar) and a polar, or ionic,
portion. Fig. 3 is a simplified sketch of the molecule. The hydrocarbon portion is often called the
“tail” and the ionic portion the “head” of the molecule.

10
Figure 3: Schematic of surface-active molecule (Ottewill 1984).
The hydrocarbon portion interacts very weakly with water molecules in aqueous solution. In fact,
the water molecules act to try to “squeeze” the hydrocarbon out of the water. Thus, the tail is called
hydrophobic. On the other hand, the head or polar portion interacts strongly with water molecules,
undergoing solvation. This part of the surfactant is hydrophilic.
Basically, it is the balance between the hydrophilic and hydrophobic parts of a surfactant that gives
it the characteristics associated with a surface-active agent. When water and oil or water and gas
are in contact, surfactant molecules tend to partition to the oil/water or gas/water interface and
reduce the interfacial tension of the interface. When surfactants are mixed above a critical
saturation in a water-oil system, the result is a stable micellar solution which is made up of
microemulsions that are homogeneous, transparent, and stable to phase separation.
1.2.1 Classification and Structure of Surfactants.
Surfactants are classified into four types that are distinguished by the chemistry of the surfactant
molecule’s polar group.
• Anionics—The polar group of an anionic surfactant is a salt (or possibly an acid) where the polar
anionic group is directly attached to the surfactant molecule and the counter and surfaceinactive
cation (often sodium) is strongly partitioned into the aqueous side of an oil/water or gas/ water
interface. Anionic surfactants are often used in oilfield foams because they are relatively good
surfactants, generally resistant to retention, quite chemically stable, available on a commercial
scale, and fairly inexpensive.

11
• Cationics—The polar group of a cationic surfactant is a salt where the polar cationic group is
directly attached to the surfactant molecule and the counter and surface-inactive anion is strongly
partitioned into the aqueous side of an oil/water or gas/water interface. Cationic surfactants are
infrequently used in oilfield foams because they tend to strongly adsorb on the surfaces of clays
and sand and are relatively expensive.
• Nonionics—The polar group of a nonionic surfactant is not a salt, but rather a chemical specie,
such as an alcohol, ether, or epoxy group, which promotes surfactant properties by imposing
electronegativity contrast. Nonionic surfactants are less sensitive to high salinities and can be
relatively inexpensive.
• Amphoterics—Amphoteric surfactants contain two or more characteristics of the previously

listed chemical types of surfactants
Fig. 4 illustrates the chemical structure of selected surfactants. Within any of the surfactant types,
there can be substantial variations in their chemistries and performance
Figure 4: Surfactant chemistry types
When using foam in conjunction with steam flooding or any elevated-reservoir-temperature

application, it is important to choose a surfactant that will be thermally stable over the needed life
of the foam in the reservoir. Historically, alpha-olefin surfactants and petroleum sulfonate

12
surfactants have been most widely used in foams applied to high-temperature (> 170°F) reservoirs.
Sulfate surfactants have been used at times in low-temperature (< 120°F) reservoirs.
Alpha-olefin sulfonates have emerged to be one of the most popular and widely employed
surfactant chemistries for use in foams. This has resulted in large part because of their combined
good foaming characteristics, relatively good salt tolerance, good thermal stability, availability,
and relatively low cost. Mixtures of different surfactant chemistries have been suggested to provide
advantages when formulating conformance foams.
1.3 Micelles and Microemulsion
When a surfactant is added to a solvent at very low concentrations, the dissolved surfactant
molecules are dispersed as monomers, as Fig. 4shows. As the concentration of surfactant increases,
the molecules tend to aggregate. Above a specific concentration, called the critical micelle
concentration (CMC), further addition of surfactant results in the formation of micelles. The
concentration of surfactant as monomers essentially remains constant above the CMC. That is,
surfactant added at concentrations above the CMC results in formation of additional micelles but
relatively little change in monomer concentration.
Figure 5: Formation of micelles.

13
If the solvent is water, the micelles form with the tail portion directed inward and the head (polar)
portion outward. Water would be the continuous phase, as shown in the lower right side of Fig. 5.
For a hydrocarbon solvent, the orientation of the surfactant molecules is reversed, as indicated in
the upper right side of Fig. 7.9, and the hydrocarbon would be the continuous phase.
Micelles that have solubilized a phase that is immiscible with the solvent often are called swollen
micelles or the solutions are called microemulsions. One definition of a microemulsion used in
EOR processes is “...a stable, translucent micellar solution of oil and water that may contain
electrolytes and one or more amphiphilic compounds” .The swollen micelles typically range from
10 to 200 μm, much larger than micelles that have not solubilized a phase. A microemulsion
consists of external and internal phases sandwiched around one or more layers of surfactant
molecules. The external phase can be either aqueous or hydrocarbon in nature, as can the internal
phase. The microemulsions can exist in several shapes, which depend on the concentrations of
surfactant, oil, water, and other constituents. Spherical microemulsions have a typical size ranging
from 10−6 to 10−4 mm. Fig. 6 shows some structures
Figure 6: structure of microemulsion (Healy and Reed 1974).
There are, in general, two types of micellar–polymer processes: (1) one which uses a low-
concentration surfactant solution (<2.5% w/w) but a large injected volume (up to 50% pore
volume); and (2) another which involves use of a high-concentration surfactant solution (5% to
12% w/w) and a small injected volume (5% to 15% v/v of the pore volume). Either type of process
has the potential of achieving low interfacial tensions with a wide variety of brine–crude oil
systems. Whichever system is selected, the system is made up of several components. The
multicomponent facet leads to an optimization problem, since many different combinations could
be chosen. Because of this, a detailed laboratory screening procedure is necessary and typically

14
involves three types of tests for: (1) phase behavior; (2) interfacial tension; and (3) oil
displacement.
Phase behavior tests are typically conducted in small (up to 100 mL) vials in order to determine
what type, if any, of microemulsion is formed with a given micellar– crude oil system. The salinity
of the micellar solution is usually varied around the salt concentration of the field brine where the
process will be applied. Besides the microemulsion type, other factors examined could be oil
uptake into the microemulsion, ease with which the oil and aqueous phases mix, viscosity of the
microemulsion, and phase stability of the microemulsion. Interfacial tension tests are conducted
with various concentrations of micellar solution components to determine the optimal
concentration ranges. The oil displacement tests are usually conducted using two or more types of
porous media. Often initial screening experiments are conducted in unconsolidated sand packs and
then in Berea sandstone. The last step in the sequence is to conduct the oil displacement
experiments in actual cored samples of reservoir rock. Frequently, core samples are placed end to
end in order to obtain a core of reasonable length since the individual core samples may be only 5
to 7 inches long.
1.4 Phase Behavior of Microemulsions
Microemulsion systems can be designed that have ultralow IFT values with either aqueous or
hydrocarbon phase (approximately 10–3 dynes/cm). This property makes micellar solutions, or
microemulsions, attractive for use as oil recovery agents. Ultralow IFTs correlate with high
solubilization of oil and water by the microemulsion system. Thus, regions of low IFT are usually
found by studying the phase behavior of microemulsion systems to locate regions of high
solubilization.
The phase behavior of microemulsions is complex and dependent on a number of parameters,

including the types and concentrations of the surfactants, cosurfactants, hydrocarbons, and brine;
temperature; and to a much lesser degree, pressure. There are no universal equations of state for
even simple microemulsions. Thus, phase behavior for a particular system has to be measured
experimentally, and results typically are presented in graphical form. Where desired, results can
be put in equation form for application in a mathematical model.

15
1.4.1 Phase-Behavior Representations on a Ternary Diagram.
A microemulsion usually is composed of at least five components: a surfactant, cosurfactant,

hydrocarbon, water, and NaCl. Cosurfactant and NaCl are not essential. To study phase behavior
rigorously, the effects of each component would need to be determined over a wide range of
compositions. Time and economic constraints prohibit extensive phase-behavior studies of each
system. Consequently, the number of components must be reduced by combining one or more
components into pseudocomponents. For example, water and NaCl are commonly represented by
the brine pseudocomponent and the hydrocarbon phase may be a mixture of hydrocarbons. When
brine is a pseudocomponent, the microemulsion system has four components and phase behavior
can be represented on a quaternary diagram, as depicted in Fig. 7. In this diagram, the apexes are
100% surfactant, cosurfactant, brine, and hydrocarbon. All points within the diagram represent
overall compositions of the four components.
Figure 7: Quaternary diagram with pseudoternary diagram at constant sulfonate/alcohol ratio.
Phase boundaries are drawn on ternary diagrams, implying that the actual compositions of
equilibrium phases lie on the ternary diagram. For some systems, this is a good assumption. For
most systems, however, some partitioning of the components that make up a pseudocomponent
occurs. NaCl, cosurfactant, and surfactant do not always partition equally between phases. In spite
of this, pseudoternary diagrams are one tool for the study of phase behavior of surfactant systems.

16
Fig. 7 shows a general representation of “ideal” phase behavior of a microemulsion on a ternary

diagram. The apex locations on the equilateral triangle represent surfactant, brine, and oil, the three
components of the solution. Concentrations may be expressed on a mass or volume basis.
At high water concentrations, as Point S1 on Fig. 7, the microemulsion would be a water-external

system with oil solubilized in the cores of the micelles.
Figure 8: Pseudoternary phase diagram for a micellar (microemulsion) solution.
A microemulsion with high oil concentration, such as Point S2, would be an oil-external system
with water solubilized in the cores of the micelles. The dashed line connecting Points S1 and S2
represents all mixtures of these two compositions. Although the mixtures remain single phase and
thermodynamically stable, the microemulsion structure changes through a series of intermediate
states (Bourrel and Schechter 1988). The structures of these intermediate states are not well-
understood. However, the solutions are thermodynamically stable and isotropic.

17
1.4.2 Effect of Brine Salinity on Phase Behavior.
In general, increasing salinity of an aqueous phase (brine) decreases the solubility of an ionic
surfactant. Surfactant is driven out of a brine as the electrolyte concentration increases. Thus, brine
salinity has a significant effect on phase behavior.
For a “simple” or ideal microemulsion system, the effect of the brine salinity on the phase behavior
is that the multiphase behavior divides into three basic classes illustrated in Fig. 9.
Two-phase lobes exist to the upper right and upper left of the three-phase triangle. Systems with
overall concentrations within these lobes separate into two equilibrium phases, as indicated by tie-
lines in Fig. 9. The microemulsion concentration point, Point M*, on the three-phase triangle will
shift toward the water or oil apex as salinity is increased or decreased around the optimal salinity
value.
Figure 9: Effect of salinity on microemulsion phase behavior (Healy et al. 1976).

18
1.4.3 Solubilization Parameters.
The volume of oil and brine that can be solubilized by a microemulsion is of interest in
characterizing a surfactant system. Solubilization parameters are defined as follows.
Vo Volume of oil in microemulsion phase

Po = =
Vs Volume of oil in microemulsion phase
Vw Volume of water in microemulsion phase

Pw = =
Vs volume of surfactant in microemulsion
Vs is the volume of surfactant in the system and includes no cosurfactant. The surfactant is assumed
to be in the microemulsion phase and not in the excess-oil or -water phase.
1.4.4 Phase Behavior and IFT
A strong correlation exists between phase behavior of a microemulsion system and the IFT
between the equilibrium phases. Because the phase behavior progresses in an orderly way as brine
salinity is increased, a reasonable conclusion is that IFT also would be a function of salinity.
1.4.5 IFT as a Function of Salinity
Fig. 10 is a typical plot of IFT between equilibrium phases as a function of salinity. The quantity
σ mo is the IFT between the microemulsion phase and the excess-oil phase, while σ mw is the IFT
between the microemulsion and excess-brine phases. σ mo decreases significantly as salinity
increases as phase behavior progresses from a lowerphase system [Type II(–)], into a middle-phase
system (Type III), and toward an upper-phase system [Type II(+)]. Change in this direction
corresponds to increasing solubilization of oil from the excess phase into the microemulsion
yielding a lower-density microemulsion.

19
Figure 10: IFT as a function of salinity (Healy et al. 1976).
The IFT values shown in Fig. 10. are typical values measured in EOR systems. Recall that the IFT
for an oil/water system (without surfactant) is typically approximately 30 dynes/cm. Thus, the
values shown are three and four orders of magnitude below values encountered in a standard
waterflood. In a flood with systems of the type shown in Fig. 10, the capillary number could be
dramatically increased above that in a waterflood and trapped oil would be mobilized.
The value of salinity at which 𝝈mo = 𝝈mw is called the optimal salinity for IFT. This salinity is
usually very close to the optimal salinity for phase behavior previously defined as the salinity for
which Vo /Vs = Vw /Vs, as is shown in Fig. 11.

20
Figure 11: Optimal salinities for IFT and phase behavior (Healy et al. 1976).
1.4.6 Correlation of IFT and Solubilization Parameters.
The IFTs, 𝝈mo and 𝝈mw, are related to the solubilization parameters. Fig. 11.a plots IFT vs.
solubilization parameter for the same data in Fig. 11.b. The data can be correlated empirically in
the form
′ )
𝑎
log(𝜎𝑚𝑜 ⁄𝜎𝑚𝑜 =
𝑚𝑜 (𝑉𝑜 ⁄𝑉𝑠 ) + 1
𝑏
and log(σmo ⁄σ′mo ) =
𝑚𝑜 (𝑉𝑜 ⁄𝑉𝑠 )+1
Where a, b, mo, mw are constants and log 𝝈'mw and log 𝝈′mo are intercept values obtained from
experimental data. Healy and Reed (1977) reported the following values for the constants and

21
intercepts: a = 6.285, mo = 0.04477, log 𝝈′mo = –7.058, b = 12.167, mw = 0.01280, and log
𝝈′mw = –12,856.
A similar equation has been proposed by Nelson (1982) to be more general
4.80
log 𝜎𝑚𝑜,𝑚𝑤 = −5.40
1.0.210(𝑉𝑜,𝑤 ⁄𝑉𝑠′ )
1.4.7 Variables Affecting Phase Behavior and IFT
A number of variables affect phase behavior and solubilization parameters, and thus IFT, including
temperature; types of ions in the brine phase, alcohol, and oil; water/oil ratio (WOR); surfactant
structure; addition of polymer to the solution; and pressure.
Effect of Oil Type: When this aromatic fraction is replaced by additional paraffin oil, optimal
salinity is increased, for this system, increasing aromaticity causes optimal salinity and the IFT
value at optimal salinity to decrease. If the aromaticity is decreased while holding salinity constant,
the phase behavior shifts from lower- to middle- to upper-phase microemulsions. Values of Ϭmo
and Vw / Vs decrease while σ mw and Vo / Vs increase.
Effect of Cosurfactant Type: Water-soluble alcohols (lower molecular weight) make a

microemulsion more hydrophilic (i.e., to increase in capacity to solubilize water but decrease in
capacity to solubilize oil). Alcohols with low water solubility (higher molecular weight), such as
pentanol and hexanol, have the opposite effect: oil solubilization increases and water solubilization
decreases.
Effect of Temperature: An increase in temperature causes solubilization parameters Vo / Vs and

Vw / Vs to decrease at optimal salinity, increasing the IFT and shifting the optimal salinity for a
given system to a higher value.
Effect of Divalent Ions: Oilfield brines typically contain divalent ions Ca++ and Mg++. These
ions contribute to the hardness of a brine (i.e., the tendency to precipitate and an increased
incompatibility with a surfactant). Divalent ions are present in the porous matrices of many
reservoir rocks. The presence of divalent ions shifts the optimal salinity to a lower value.

22
Effect of Surfactant Structure: The molecular structure of the surfactant affects phase behavior,
solubilization parameters, and consequently, IFTs. When the carbon number of the side chain on
the surfactant is increased, optimal salinity is reduced, with a corresponding reduction in IFT and
increase in the optimal solubilization parameters.
Effect of Pressure: increasing the pressure caused a shift in phase behavior toward a lower-phase
microemulsion. For a given system, optimal salinity increased as pressure on the system increased.
Effect of Polymer Addition: Behavior is complex, but it has been shown that polymer can be
added to a micellar slug under controlled conditions to increase slug viscosity without adversely
affecting phase behavior or IFT.
Tutorials refer TD paper

23
CHAPTER 2: POLYMER AUGMENTED WATER FLOOD.

2.1 Polymers
A polymer is substance which has a molecular structure built up chiefly or completely from a large
number of similar units bonded together. Early application of polymers for use during oilfield
conformance-improvement operations or EOR was focused on improving volumetric sweep
efficiency of waterfloods. More recently, polymers have been used extensively in Disproportionate
permeability reduction (DPR) and relative permeability modification (RPM) treatments for water
shutoff and in conformance improvement polymer-gel treatments.
2.1.1 Polymer Fundamentals.
Polymers are large molecules and chemical entities referred to as macromolecules. Polymer
molecules are the resultant chemical specie when a large number of relatively small and repeating
molecular entities, called monomers, are joined together chemically. The chemical process of
joining together the monomers and forming polymer molecules is referred to as the polymerization
reaction process. Polymers, both natural and man-made, have numerous beneficial uses and
applications in modern society (everything from wood, to plastics, to man-made thickening agents
added to milk shakes). Polymers can come in pure solid and liquid forms. Some polymers can be
dissolved in liquids.
2.2. Physical and Chemical Characteristics of Polymers.

2.2.1 Basic Oilfield Polymer Types.
There are two fundamentally different types of water-soluble and viscosity-enhancing polymer
chemistries that have been used during polymer waterflooding and conformance-improvement
treatments. The first type is biopolymers, such as Xanthan gum polymer which is a biopolymer
produced commercially by microbial action of the organism Xanthomonas campestris on a
carbohydrate feed stock. Fig.12.The second type is man-made synthetic polymers, such as
acrylamide-based polymers Fig. 13. Polyacrylamides can be manufactured by polymerization of
the acrylamide monomer, shown in the brackets in Fig. 13 (Willhite and Dominguez 1977), to
produce a polymer.

24
Figure 12: Structure of xanthan biopolymer (Xanthan Gum/Keltrol/Kelzan 1974).
Figure 13: Molecular structure of high-molecular-weight polyacrylamides (Willhite and Dominguez 1977).

25
Biopolymers of the type used in conformance improvement are polysaccharides (poly sugars) in
which the monomer chemical linkages of the polymer backbone are glycoside linkages, involving
carbon-oxygen-carbon chemical bonds. In aqueous solution, the multistranded molecular
complexes of xanthan polymer are fairly rigid molecular species, causing the polymer molecules
to take on an extended molecular conformation.
Synthetic polymers of the type used in conformance improvement are usually highly flexible
molecules in which the polymer backbone consists of a relatively chemically stable carbon
molecular chain with single and flexible carbon-carbon bonds. Pendant water-soluble chemical
groups (e.g., amide groups) on the molecule render the polymer molecule to be soluble in water.
Synthetic polymers have emerged to become the predominant and preferred polymer type for use
in commercial oilfield conformance-improvement operations because of the inherent chemical and
biological stability of synthetic polymers, along with injectivity and cost issues.
Historically, the two types of biopolymers primarily used in polymer waterflooding have been
xanthan and scleroglucan polymers. The only synthetic polymers that have been used extensively
during polymer waterflooding (and in polymer-gel conformance treatments) are those based on
acrylamide-polymer chemistry.
Other polymers have been developed for particular applications. For example,
polyvinylpyrrolidone was developed for high-temperature applications in harsh environments
where polyacrylamides and biopolymers were found not to be applicable.
2.2.2 Polymer Stability.
The property that makes polymers useful for EOR applications is that small concentrations of
polymer, on the order of a few hundred to a few thousand ppm (by weight), increase the viscosity
of an aqueous solution significantly.
For a polymer to be useful in EOR applications, it must be stable at reservoir conditions for the
expected residence time in the reservoir rock. Because polymers can degrade under certain
conditions, short-time laboratory tests can be misleading. Polymer stability at reservoir
temperature and in the presence of reservoir brine is essential to EOR applications.
There are several methods of evaluating polymer stability:
It is well-established that both polyacrylamides and biopolymers are susceptible to oxidative attack
by dissolved oxygen in the injected water. Degradation is detected by the loss of solution viscosity

26
with time. At low temperatures, the reaction rate is slow and can go undetected in short tests. The
degradation rate increases as temperature increases, which is consistent with chemical reaction
kinetics. The oxidative degradation reaction is catalyzed by dissolved metal ions, such as Fe+++
.Degradation by oxidative attack can be prevented or minimized by reducing the oxygen content
of the water or brine to less than a few parts per billion.
Oxygen scavengers or deaeration are mostly used. Sodium dithonite is used to stabilize
polyacrylamides. A mixture of thiourea, isopropyl alcohol, and sodium bisulfate is found to retard
oxidative attack on xanthan biopolymers at temperatures up to 207°F.
Oxygen scavengers are not typically used in field applications because most reservoirs have a
reducing environment and because dissolved oxygen is consumed rapidly after the injected fluid
containing oxygen enters the reservoir.
The thermal stability of polymers (i.e., stability at higher temperatures) is a second
important consideration. Laboratory tests indicate that the carbon/carbon backbone of
polyacrylamides is stable in the absence of oxygen and divalent ions to temperatures up to 194°F.
The xanthan molecule has a helical structure that appears to take on different configurations,
depending upon salinity, divalent-ion concentrations, and temperature. Changes in structural
configuration can be correlated in terms of a transition temperature. The transition temperature
increases with salinity and divalent-ion content. Xanthan solution could maintain at least half its
original viscosity for 5 years if the temperature does not exceed 167 to 176°F. This estimate is
consistent with current practice, in which most xanthan biopolymers are limited to applications
where the temperature is ≤140°F.
It is usually necessary to prepare the injected polymer solution in reservoir brine. Reservoir
brines often contain high concentrations of divalent cations, in particular Ca++ and Mg++. The
solution viscosity of each polymer is affected by the presence of divalent cations. Both
polyacrylamides and biopolymers are stable in high concentrations of divalent cations at low
reservoir temperatures. Ferric ion (Fe+++) will cause gelation of polyacrylamide and must be
excluded or chelated within reservoir brines. The presence of divalent cations causes stability
problems for polyacrylamides at elevated temperatures. As the degree of hydrolysis increases as a
result of polyacrylamide reaction with water, the solubility of the polymer decreases in the
presence of calcium and magnesium.

27
Polymer degradation also results from bacterial attack. Biopolymers are susceptible to
biological attack resulting in the loss of solution viscosity from the destruction of the carbohydrate
backbone, which can be rapid. For this reason, the polymer broth usually contains a bactericide,
such as formaldehyde, to control bacterial growth. At one time, it was thought that bacterial attack
would not be a problem in petroleum reservoirs because the organisms could not be transported
through the porous rock. Unfortunately, bacterial attack has been observed in at least two field
tests. Specimens of bacteria were recovered from core material from the Loudon reservoir,
demonstrating that bacteria could be transported appreciable distances through porous rock. In the
case of the Loudon field test, high concentrations of formaldehyde (1,500 ppm) successfully
controlled the bacterial attack. When H2S is present, no known bactericide is effective.
2.2.3 Rheological Properties.
Effect of Shear Rate: Polymers are of interest in EOR applications because of their rheological
properties in dilute solutions. Aqueous solutions of polyacrylamides and xanthan biopolymers
often exhibit non-Newtonian rheological behavior. A Newtonian fluid has a linear relationship
between shear stress and shear rate given in the equation below. The proportionality constant in
this relationship is the viscosity of the fluid.
τ = µγ
Where τ = shear stress, μ = solution viscosity, and γ = shear rate. A more general expression
relating shear stress to shear rate is
τ=Kγn
Where K and n are constants that characterize the fluid. If n ≠ 1.0, then shear stress does not vary
linearly with shear rate and the fluid is non-Newtonian. For this case, an apparent viscosity, μa,
may be defined by
τ = µa γ s
The apparent viscosity varies with shear rate. The subscript on μa is usually dropped when
discussing non-Newtonian fluids.
Normally, the apparent viscosity of polymer solutions used in EOR processes decreases as
shear rate increases. Fluids with this rheological characteristic are said to be shear thinning. The
apparent viscosity decreases because the polymer molecules are able to align themselves with the
shear field to reduce internal friction. Often it is possible to represent the rheological properties of
a shear-thinning fluid by the power-law model given by the equation below, which results from

28
combining the last two equations above. The constants K and n depend on the concentration of the
polymer.
µ = K γ (n− 1)
Where μ = apparent viscosity, K = power-law constant, n = power-law exponent, γ = shear rate,
and consistent units should be used.
2.3 Effect of Salinity.
The rheological behavior of polymer solutions also may be affected by salinity and divalent-ion
content. The effects are specific to polymer type, and the largest effects are observed with
polyacrylamides.
Because it is not possible to produce unhydrolyzed polyacrylamide commercially, the
discussion concerning polyacrylamides refers to partially hydrolyzed polyacrylamides. Hydrolysis
of polyacrylamide introduces negative charges on the backbone of the polymer chain that have a
large effect on the rheological properties of the polymer solution. At low salinities, the negative
charges on the polymer backbone repel each other and cause the polymer chain to stretch. Each
polymer molecule occupies more space in solution, and the apparent viscosity of a dilute solution
increases accordingly.
When an electrolyte, such as NaCl, is added to a polymer solution, the repulsive forces are
screened by a double layer of electrolytes and extension is reduced. As the electrolyte
concentration increases, the extension of the polymer chain decreases and the solution viscosity
declines. Fig. 5.15 (Mungan 1969a) illustrates the effect of salinity on the relative viscosity
(apparent solution viscosity/solvent viscosity) of 250-ppm solutions of Pusher 700, a partially
hydrolyzed polyacrylamide.

29
Figure 14: Variation in relative viscosity with salinity and shear rate for 250-ppm Pusher 700
solutions at 25°C (Mungan 1969a).
2.4 Flow of Polymers through Porous Media.

Polymer solutions used in waterfloods must be able to be transported successfully and effectively
through the reservoir. Thus, the manner in which polymer solution flows through porous rock and
the associated polymer interaction with the pore walls of matrix reservoir rock are important
aspects regarding the attainment of the technical and economic success of a polymer flood.
2.4.1 Polymer Retention.
Polymer retention will often profoundly affect the technical and economic success of a polymer-
flooding project. The amount of oil that will be recovered per pound of polymer injected is
inversely related to polymer retention.
Retention for a given polymer during a polymer flood increases as the permeability decreases,
increases as the polymer molecular weight increases, increases as the clay content in the reservoir
rock increases, usually decreases as oil wetness increases, tends to increase in sand and sandstone
reservoirs with decreasing anionic charge and increasing cationic charge of the polymer’s pendant
groups, and has been reported to increase at times in the presence of crude oil.

30
Polymer retention should be determined carefully, or at least estimated carefully, before initiating
a polymer waterflood. Polymer retention for a given polymer flood is normally best estimated by
conducting flooding experiments in reservoir rock with reservoir fluids at reservoir temperature.
“Field-measured values of retention range from 7 to 150 μg of polymer per cm3 of bulk volume,
with a desirable retention level being less than approximately 20 μg/cm3.”46 Laboratory
measurements of polymer retention in reservoir rock are usually reported as mass of polymer
adsorbed per unit mass of rock, Γ, and is usually reported as μg/g of polymer adsorbed onto
reservoir rock. Frequently, it is preferred to have polymer retention reported in terms of mass of
polymer adsorbed per unit volume of reservoir rock, Γv, or, more specifically, in terms of pounds
of polymer adsorbed per acre-foot of reservoir, lbm/acre-ft. To convert from Γ to Γv,
Γv = 2.7194Γ (1−ϕ) ρRG,
where ϕ is porosity, ρRG is the density of the reservoir rock grains (no pore space included), Γv is
in units of lbm/acre-ft, and Γ is in units of μg/g.46 Polymer retention, as measured during field
projects, has been reported to range from 20 to 400 lbm polymer/acre-ft bulk volume, with
desirable retention reported to be less than 50 lbm/acre-ft.
2.4.2 Inaccessible and Excluded Pore Volume (PV)
Polymer molecules are larger than water molecules and are large relative to some pores in a porous
rock. Because of this, polymers do not flow through all the pore space contacted by the brine. The
fraction of the pore space not contacted by the polymer solution is called the IPV.
Two explanations for, and contributions to, the IPV phenomenon have been reported. The first
IPV explanation is that the large size of the polymer molecules prevents entry into smaller and
dead-end pores. This promotes propagation of the polymer molecules faster than an inert chemical
tracer because the polymer flows only through the larger-pore flow paths.
The second IPV explanation is the wall-exclusion effect. It is hypothesized that polymer molecules
flow and concentrate in the center of the pore-level flow channels of matrix reservoir rock because
the polymer molecule flow and the free tumbling of the polymer molecules are excluded from the
near-surface volume of the pore walls. Such flow behavior accelerates the rate of polymer
propagation through the porous media relative to the rate of propagation of an inert chemical tracer.

31
IPV has been observed in all types of porous media for both polyacrylamides and biopolymers and
is considered to be a general characteristic of polymer flow in porous media. The magnitude of the
IPV can range from 1 to 2% to as much as 25 to 30%, depending on the polymer and porous
medium. The impact of IPV on polymer transport in porous rocks often is concealed by polymer
retention.
2.4.3 Flow Characteristics. Permeability Reduction

Polymer flow through reservoir matrix rock can cause permeability reduction. A measure of the
polymer-induced permeability reduction is the residual resistance factor, Rrf
𝑘b
𝑅𝑟𝑓 =
𝑘a
where kb is brine permeability measured before polymer flooding, and ka is brine permeability
measured after polymer flooding.
Permeability reduction induced by polymers tends to be greater in lower permeability reservoir

rock. This is, in general, counterproductive. Polymers, especially HPAMs, that undergo even a
small amount of mechanical shear degradation often lose much of their permeability reduction
propensity because the relatively small number of exceptionally large molecules of a given
polymer MW distribution (especially for many HPAMs) are the first polymer molecules to be
shear degraded. These large molecules contribute disproportionately to permeability reduction.
2.4.4 Mobility Reduction.

When polymer in solution flows through reservoir matrix rock, it imposes a mobility reduction
that is the primary conformance-improvement benefit of polymer waterflooding. The mobility
reduction can be imparted by one of two distinctly different mechanisms. First, the polymer can
cause an increase in the viscosity of the brine being flooded through the porous media. This is
normally the desired effect when flooding with polymer solutions for mobility control. The second
mechanism reduces the permeability of the reservoir matrix rock. One measure of mobility
reduction imparted by polymer-solution flow is the resistance factor, Rf, which is defined as
λw
𝑅𝑓 =
λp

32
where λw is the mobility of the solvent of the polymer solution, and λp is the mobility of the
polymer solution. When the polymer solution imparts no permeability reduction and for
measurements made at ambient temperature,
𝑅𝑓 = μeff
where μeff is the effective viscosity of the polymer solution as it flows through the reservoir matrix
rock. Alternatively, for a single-phase polymer solution flowing through matrix reservoir rock at
a given temperature and there is no imparted permeability reduction,
μeff
𝑅𝑓 =
μw
where μw is the viscosity of the brine in which the polymer is dissolved.
2.4.5 Extensional Flow
As mentioned in the discussion on rheology, when flexible, coiled, highMW polymers, such as
HPAM, are forced to flow through matrix reservoir rock at exceptionally high rates and experience
exceptionally high-flow shear fields, the polymer can enter extensional and elongational flow at
which point the polymer solution’s apparent viscosity can rise rapidly. In this flow regime, the
polymer is also often mechanically shear degraded. Solutions of well-designed polymer floods are
likely to experience extensional flow of noticeable consequence only in certain instances in the
very near-wellbore region adjacent to the injection or production well.
2.4.6 Adsorption.
Polymer adsorption results primarily from physical adsorption and not chemisorption. Polymer
adsorption is often the major cause of polymer retention.
2.4.7 Mechanical Entrapment.
Mechanical entrapment of polymer during propagation through reservoir porous media results
from the larger polymer molecules becoming lodged in narrow flow channels (e.g., pore throats).
There are several significant consequences of mechanical entrapment: permeability reduction, loss
of the entrapped polymer’s favorable viscosity enhancing functionality beyond the point of
entrapment, loss of the largest of the polymer molecules first has a disproportionately large
negative impact during the remainder of the polymer flood on viscosity and mobility-control
properties, and loss of a disproportionately large portion of its viscosity and mobility control
functionality relatively soon after the polymer solution is injected into a reservoir.

33
2.4.8 Precipitation.
Polymer precipitation from solution, especially in the presence of high reservoir brine salinity, is
another source of polymer retention. Precipitation is especially problematic when flooding with
HPAM in high-temperature reservoirs with formation waters containing hardness divalent cations.
2.5 Estimation of Injection Rates/Pressure Drop for Polymer Flooding.

Polymer-solution injectivity is an important consideration for several reasons. First, the rate at
which the polymer solution can be injected directly impacts the economics of a polymer-flood
project. Second, routine injection-well cleanup jobs may be required if polymer or polymer-
microgel damages injectivity. These cleanup jobs can detract from the polymer flood’s economics
and effectiveness. Injectivity decreases as polymer MW increases. Polymer-solution injectivity is
more favorable when the polymer solution exhibits shear-thinning viscosity behavior.
When injection begins at a constant rate, the leading edge of the polymer zone moves radially
away from the wellbore and the pressure drop changes continuously with time. The location of the
polymer front can be determined by material balance if dispersion and IPV are neglected. The
equation below gives the radius of the polymer front, rp, after Wp barrels of polymer have been
injected into a formation of thickness h and porosity ϕ
where rp = radius of polymer front, ft; ρp = density of polymer solution, fbm/ft3; ρr = grain
density of rock, lbm/ft3; C = polymer concentration, lbm polymer /lbm solution; Ĉ = polymer
retention, lbm polymer /retained lbm rock or g pol /g rock h = formation thickness, ft; and Wp =
cumulative volume of polymer solution injected, bbl. The polymer displaces a Newtonian fluid
with mobility λN. Flow is assumed to be radial from the injection wellbore to an effective radius
re, where the reservoir pressure is maintained constant at pe.

34
2.6 Polymer-Augmented Water Flood

Aqueous solutions of high-molecular-weight polymers are used widely to augment waterflooding.
Polymer- augmented waterflooding can be divided into two broad classifications. When the
mobility ratio of a waterflood is unfavorable, continuous injection of polymer solution increases
the microscopic displacement efficiency at a particular WOR and increases volumetric sweep
efficiency in the reservoir. Even when the mobility ratio is favorable, reservoir heterogeneity
primarily in the vertical direction may cause poor volumetric sweep. In this case, polymer-
augmented waterflooding is used to reduce the water mobility in the high-permeability layers, so
that oil can be displaced from lower-permeability layers.
When conducting a polymer waterflood, a high-molecularweight and viscosity-enhancing polymer

is added to the water of the waterflood to decrease the mobility of the flood water and, as a
consequence, improve the sweep efficiency of the waterflood. The primary purpose of adding
polymer to most polymer waterfloods is to increase the viscosity of the flood water; however,
polymer addition to the flood water in many instances also imparts a secondary permeability-
reduction component. Polymer waterflooding is normally applied when the waterflood mobility
ratio is high or the heterogeneity of the reservoir is high.
2.6.1 Slug Injection. As stated above continuous injection of polymer during a polymer-
augmented waterflood would not be economically attractive under many conditions. In most cases,
the polymer is not injected at a constant concentration but rather in a staged sequence of
concentration reduction. The polymer slug is displaced by drive water.
The objectives of this concentration sequence are to reduce the total amount of polymer used and
to prevent, or at least to reduce, viscous fingering of low-concentration fluid into regions of higher
concentrations. Viscous fingering occurs because each reduction in polymer concentration is
accompanied by a reduction in the apparent solution viscosity. It is assumed that mixing in the
reservoir will dampen out the viscous fingers. Fig. 15 shows the polymer waterflooding process.

35
Figure 15: Schematic of the polymer waterflooding process.
2.6.2 How Polymer Flooding Improves Recovery.

Polymer waterflooding promotes improved sweep efficiency by improving the mobility ratio.
Improved sweep efficiency imparted during polymer flooding is primarily accomplished by
increasing the viscosity of the waterflood drive fluid.
2.6.3 Design Considerations.
The first step in the design process is to determine whether a reservoir is a candidate for polymer-
augmented waterflooding. Reservoirs that are good candidates can be recognized by poor
volumetric sweep efficiency of the waterflood, low waterflood recovery compared with similar
reservoirs, rapid breakthrough of injected water into production wells, and high WOR throughout
the flood life.
A mobility-control problem can be recognized at the microscopic level by determining the mobility
ratio of the waterflood. If the mobility ratio is much larger than unity, both microscopic and
volumetric sweep efficiency decrease. At high mobility ratios, viscous fingering occurs, leading
to poor sweep efficiency. Effects of reservoir heterogeneity on volumetric sweep efficiency can
be assessed by determining the Dykstra-Parsons coefficients from core analysis and from

36
displacement calculations to predict whether the volumetric displacement efficiency is likely to be

low.
The critical question that must be addressed in design is whether polymer flooding in a particular
reservoir is economically attractive. In most cases, the economics of current operations is
compared with economics based on projections of the displacement performance.
2.6.4 Polymer Solutions.
The principal beneficial property of polymer solutions for use in flooding oil reservoirs is the
aqueous solution’s enhanced viscosity. Aqueous polymer solutions that are used for conformance-
improvement flooding normally exhibit non-Newtonian viscosity properties.
Viscosity of Polymer Solutions: The viscosity of a polymer solution is a measure of how “thick”
a fluid is. For example, molasses is characterized as being “thicker” and more viscous than water.
The viscosity of a fluid or solution may, in general terms, be defined as the solution’s resistance
to being sheared or as the resistance of a fluid mass to change its form. Fluid viscosity, μ, is defined
as
τ
μ=
γ
where τ is shear stress and γ is shear rate.
Many common fluids, such as water and motor oils, exhibit Newtonian viscosity. For fluids with
Newtonian viscosities, the fluid’s viscosity is independent of the shear rate that the fluid is
experiencing. That is, the value of the viscosity of a Newtonian fluid at a given temperature is a
single value that is independent of shear rate.
The viscosity-enhancing power of a polymer is related to the size and extension of the polymer
molecule in a particular aqueous solution.For a number of reasons, the viscosity of a polymer
solution that is measured in a viscometer and the effective viscosity of the polymer solution that
is measured during flow through porous reservoir matrix rock often have different values.
To predict the viscosity-enhancing power of a polymer in a given solution, the polymer’s

intrinsic viscosity, [η], can be measured by
η−ηs
[η] = Limc→0
𝑐ηs

37
where c is polymer concentration, η is polymer solution viscosity, and ηs is solvent viscosity.
Intrinsic viscosity is obtained by determining the value Limc→0 (η-ηs)/cηs) that is obtained
from the plot of (η–ηs)/cηs) vs.

2.6.5 Viscosity Enhancement and Permeability Reduction.
Water-soluble polymers used in conformance-improvement operations operate by reducing fluid

mobility by increasing viscosity of the oil-recovery drive fluid (primarily the flood water) and/or
by reducing permeability, by which the polymers, directly or indirectly, act as a fluid-flow
blocking agent. Reducing permeability is the conformance-improvement mechanism by which
conformance treatments operate when polymers and polymer gels are used for imparting DPR.
Polymers used in waterfloods often have a secondary component in their conformance-
improvement mechanism that involves permeability reduction within the flooded volume of a
matrix rock reservoir.
2.6.6 Polymer-Augmented Waterflood in a Layered Reservoir.

Reservoirs are usually heterogeneous, particularly with respect to permeability. The largest
changes in permeability typically occur in the vertical direction. When there is a wide range of
permeabilities in the vertical cross section, the injected water flows through the most-permeable
layers where the displacement process occurs rapidly. Water breakthrough from these layers may
cause high WORs and abandonment of the waterflood while substantial portions of the low-
permeability zones are being waterflooded at a slower rate.
Injecting polymer into a linear system slows the flow of injected fluids in all zones contacted by
the polymer. The amount of polymer flowing into each zone of a uniform, layered reservoir will
be proportional to the permeability of the layer. It is tempting to assume that polymer injection
will change the vertical distribution of fluids flowing in the system and thus change the
displacement efficiency. To be an effective mobility-control agent, however, the polymer solution
must not be slowed as much in low-permeability zones as in high-permeability zones.
2.6.7 Polymer-Augmented Waterflood in Patterns or Other Arrangements of Production

and Injection Wells. Estimation of polymer flood performance in pattern floods or other
flooding programs where the flow is not linear requires the use of more-complex models or

38
reservoir simulators. Most reservoir software companies have polymer flood simulators of
various degrees of complexity.
2.6.8 When and Where Applicable.
The following screening guidelines can be used to determine where polymer waterflooding is
most applicable, in terms of reservoir properties.
• Oil viscosity < 150 cp (preferably < 100 and > 10 cp) and API gravity > 15.
• Matrix-rock permeability > 10 md, with no maximum.
• Reservoir temperature: low temperatures are best (best at < 176°F; maximum of approximately
210°F).
• Water injectivity should be good with some spare capacity (hydraulic fracturing of injection
wells may help).
• Reservoir clay content should be low.
• Low salinity of the injection and reservoir brines are preferable.
Polymer waterflooding has been conducted successfully in sandstone and carbonate matrixrock
reservoirs, fractured reservoirs, and in water-wet, mixed-wetting, and oil-wet reservoirs.

39
CHAPTER 3: IN-SITU PERMEABILITY MODIFICATION

3.1 Introduction
The amount of oil that is recoverable from a reservoir by a displacement process depends on (1)
the effectiveness with which the injected fluid displaces oil from the pores in the rock (microscopic
displacement efficiency) and (2) the volumetric fraction of the reservoir contacted by the injected
fluid (macroscopic sweep efficiency). This latter efficiency is governed by the mobility ratio but
also in large measure by the geologic heterogeneity of the reservoir rock. Permeabilities vary both
areally and vertically, and large changes typically occur in the vertical direction in a single well.
When a displacement process is implemented in a reservoir with large variations in vertical

permeability, injected fluid tends to flow through the zones with the highest permeabilities; thus,
low-permeability zones may receive only a small fraction of the injected fluid.
Bypassing part of the reservoir by the injected fluid can lead to production of relatively large
volumes of injected fluid per barrel of recovered oil. The result can be that the displacement
process reaches the economic limit at a time when a large volume of oil remains in the bypassed
or unswept regions of the reservoir.
The objective of an in-situ permeability-modification process is to treat the reservoir in such a way
that the effective permeability of the high-permeability zones is significantly reduced.
Conceptually, it would be desirable to reduce permeability across the entire reservoir (i.e., from
injection wells to production wells). Practically, treatment tends to be limited to the region
relatively near the wellbore. Also, provisions must be made to prevent the gelling agents from
entering and damaging the lower-permeability zones.
Gels have found broad application as oilfield fluid-flow blocking agents because gels are often an
exceptionally cost-effective plugging and/or permeability-reducing agent for use in a number of
different conformance-improvement treatments
Various methods of permeability modification have been applied. These include crosslinked
polymer (gelled polymer), microbial-based processes, and precipitation processes. In-situ
permeability modification is not strictly a mobility-control process in that volumetric sweep
efficiency is improved through modification of rock permeability rather than through mobility

40
adjustment of an injected fluid. The processes are related, however, because of the use of polymer
systems that are similar and because the result (i.e., improved volumetric sweep) is similar.
3.2 General Description of the Process.

In the crosslinked, or gelled, polymer process, an aqueous solution or solutions of moderate
viscosity containing polymer and a crosslinking agent are injected into high-permeability or
fractured zones. Polymer and the crosslinker react to form a viscous gel (i.e., a fluid with a very
high apparent viscosity). The gel is essentially immobile and thus acts to reduce the apparent
permeability of the rock matrix or fracture.
Depending on the chemical system, the procedure used to mix the reacting chemicals is different.
In some applications, the reacting chemicals are mixed at the surface by passing the different
solutions through an in-line mixer before they enter the wellbore. Reaction to form a gel occurs in
the reservoir. Another approach, often used with biopolymers, is to mix the solutions in tanks at
the surface before injection. Reaction starts at the surface and continues after injection into the
reservoir. Yet another procedure is to inject the different solutions as slugs. In this case, mixing
occurs in the reservoir as a result of dispersion and chromatographic transport of the different
chemical species.
Permeability modification has been applied most commonly to injection wells but also is
applicable to production wells. Production well applications typically involve use of a gelled
polymer to shut off or to reduce water production where, for example, water coning is occurring.
3.3 Gelation Time and Gel Strength.
The time required for gelation to occur is an important design variable for the Cr(III)/
polyacrylamide system. With this system, chemicals are mixed at the surface and injected as a
viscous solution. The solution is designed to react and gel at a designated time after being placed
in the formation. Gel time is an ill-defined variable because gelation is a continuous kinetic process
and there is no clearly defined point at which one can state that a gel has been formed. There are
many techniques for the measurement of gel time, and many provide a consistent measurement
within the technique, but results do not necessarily correlate with other methods.

41
One method is based on the measurement of viscosity. In this procedure, the gel solution is mixed
in a thermally jacketed beaker and then the viscometer is turned on with a specified spindle at a
set shear rate. Apparent viscosity is measured as a function of time and typically yields a curve
like that shown in Fig. 16. There is no apparent increase in viscosity for a period of time, and then
a period of rapid increase occurs. Gel time may be arbitrarily defined as the time at which viscosity
starts to increase, the time at which the extensions of the two approximately straight lines intersect,
or the time at which the apparent viscosity reaches a specified value. This method has been found
to give reproducible results for gel times of up to 10 days or more.
Figure 16: Viscosity of a polyacrylamide gel during gelation (Jordan et al. 1982).
A number of factors affect gel time and strength such as; Polymer concentration and type,
Crosslinker concentration, Reducing agent strength and concentration, Temperature, Salinity, pH
and Contaminants (Fe+2 Fe+3 H2S Microorganisms Divalent ions, Ca++/Mg++).
Gel time is shorter at high temperatures. For example, for the redox-reaction-based process gel
time is dependent on temperature. Gel time decreases with increasing temperature.
System pH also affects gelation time, gel time increases significantly as pH was increases. Partially
hydrolyzed polyacrylamide/Cr(III) carboxylate gels are less sensitive to pH than gels made by
reducing Cr(VI) to Cr(III). Stable gels formed over a pH range of 4.0 to 12.5. Gelation rate
increases slightly with pH for the Cr(III) carboxylate gels up to a pH of 10.5.

42
Salinity also affects gelation rate. Polyacrylamide gels produced with the Cr(VI) reduction have
been reported to gel faster with increasing salinity at low salt concentrations but to gel slower with
increasing salinity at high salt concentrations.
Contaminants (Fe+2 Fe+3 H2S Microorganisms Divalent ions, Ca++/Mg++), affect the gelation
rate. H2S, which is often present in oilfield brines, is a strong reducing agent and will accelerate
the reduction of Cr(VI) to Cr(III). H2S is not a severe problem for the Cr(III) carboxylate gels or
for gels made by crosslinking biopolymers with Cr(III). Iron and divalent cations (Ca++ and
Mg++) can interfere in the process because of ionic effects and promotion of crosslinking.
Biopolymers, and to a lesser degree polyacrylamide, are susceptible to destruction by
microorganisms.
Gel strength, swelling, and syneresis are dependent on the same parameters that affect gel time.
Xanthan/Cr(III) gels have been shown to synerese at low pH (< approximately 4.0) and swell at
higher pH. Both xanthan and polyacrylamide gels have been shown to synerese rather rapidly at
194°F. Polyacrylamide/ Cr(III) carboxylate gels have been reported to be stable for long periods
at temperatures as high as 255°F. Gel strength increases with an increase in concentrations of
polymer and crosslinker. However, at high concentration ratios of crosslinking agent to polymer,
syneresis is promoted. In this case, the gel apparently is “overcrosslinked” as the reaction continues
and solvent is forced out of the gel.
3.4 Behavior of Gel Systems in Porous Media.
The crosslinked polymer systems mixed at the surface are injected into a formation, where reaction
occurs to form a gel. Ideally, once formed, the gel has sufficient strength or viscosity to be
immobile. The effective permeability of the rock matrix or fractures in which the gel resides is
reduced or, in some cases, essentially eliminated. For processes based on injection of slugs in
which the different reactants reside, mixing followed by reaction occurs within the rock. The
objective here also is to form a gel that is not mobile.
3.5 Design Considerations for Gel Treatments.
Design of a gel treatment consists of two parts: analysis of the reservoir problem and design of the
gel system to be applied. The reservoir problem is typically analyzed with standard reservoir
engineering calculations combined with field testing. Injectivity profiles conducted to determine

43
the zones in which fluids are leaving the wellbore in an injection well are useful. Radioactive
tracers, spinner surveys, and shut-in temperature profiles are commonly used. These do not define
flow in the region away from a wellbore, however. Chemical or radioactive tracer tests can be used
to obtain information on flow patterns over a wider region of a reservoir. Depending on flow
conditions, tracer tests may require several months for completion of field monitoring and analysis.

44
PART II: THERMAL METHODS

Introduction
Thermal recovery processes rely on the use of thermal energy in some form both to increase the
reservoir temperature, thereby reducing oil viscosity, and to displace oil to a producing well. Four
processes have evolved over the past 50 years to the point of commercial application. These are
cyclic steam stimulation (CSS), steam-assisted gravity drainage (SAGD), steamdrive, and forward
in-situ combustion.
The motivation for developing thermal recovery processes was the existence of major reservoirs
all over the world that were known to contain billions of barrels of heavy oil and tar sands that
could not be produced with conventional techniques. In many reservoirs, the oil viscosity was so
high that primary recovery on the order of a few percent of original oil in place (OOIP) was
common. In some reservoirs, primary recovery was negligible.
Thermal recovery processes are the most advanced enhanced-oil-recovery (EOR) processes and
contribute significant amounts of oil to daily production. Most thermal oil production is the result
of cyclic steam injection and steamdrive.
There is a screening criteria for thermal recovery processes to be applied in EOR. The table below
summarizes the various screening criteria.

45
Table 1: Screening criteria for thermal recovery processes
Thermal Recovery
Screening Parameters Steam In-Situ
Combustion
Oil gravity (°API) 10 to 34 10 to 35
ln-situ oil viscosity, μ (cp) ≤ 15,000 ≤ 5,000
Depth, D (ft) ≤ 3,000 ≤ 11,500
Pay-zone thickness, h (ft) ≥ 20 ≥ 20
Reservoir temperature, Tr (°F) — —
Porosity, ϕ (fraction) ≥ 0.20 ≥ 0.20
Average permeability, k (md) 250 35
Transmissibility, kh/μ (md-ft/cp) ≥5 ≥5
Reservoir pressure, pr (psi) ≤ 1,500 ≤ 2,000
Minimum oil content at start of process, So × ϕ ≥ 0.10 ≥ 0.08
(fraction)
Salinity of formation brine, total dissolved solids — —
(ppm)
Rock type Sandstone or Sandstone or
carbonate carbonate

46
CHAPTER 4: CYCLIC STEAM STIMULATION

4.1 Introduction
Historically, primary production from heavy-oil reservoirs before the development of thermal
recovery techniques was 5% of OOIP or less. Production rates were low, declining with time as
the reservoir energy was depleted. In some wells, production declines were known to be caused
by deposition of solids, paraffins, or asphalts in the region around the wellbore. Hot-oil squeezes
or swabbing the well with a diluent, such as kerosene, would increase the production rate,
sometimes by a factor of two or more, and the well would resume its decline. Attempts to use
downhole heaters to increase oil production have been attempted and largely discarded.
CSS was discovered by accident in the Mene Grande field in Venezuela in 1959 when steam broke
out behind casing in a steam-injection well. This well, which had produced no oil previously,
flowed oil at rates of 100 to 200 B/D when the well was blown down. The discovery that steam
injection into a heavy-oil reservoir could increase production rates by factors of 5 to 10 was a
historic point in the development of thermal recovery techniques.
In CSS, steam is injected into a production well for a period of 2 to 4 weeks. The well is shut in
and allowed to “soak” before returning to production. The initial oil rate is high because of the
reduced oil viscosities at the increased reservoir temperatures. There is also some acceleration
from increased reservoir pressure near the wellbore. With time, the heated-zone temperature
declines as a result of heat removed with the produced fluids and conduction losses to over- and
underlying formations. Oil rates decline as the heated-zone temperature decreases and oil viscosity
increases. When the production declines to a predetermined level, another cycle of steam injection
is initiated. In some reservoirs, up to 20 cycles have been carried out.
CSS is a precursor to steamdrive in most reservoirs. When the pattern spacing is small, interference
between wells may be noticed during steam injection. This is an indication that the heated regions
from adjacent wells may overlap or that the steam-slug volume is too large. If interference occurs
after several cycles have been conducted, consideration of conversion to steamdrive may be
appropriate. CSS is preferred when the natural reservoir energy has not been depleted. Steamdrive
is used when the reservoir energy is depleted.
4.2 Production Mechanisms—CSS.
Heavy-oil reservoirs are characterized by oil viscosities at reservoir temperatures on the order of
100 to 10,000 cp. When a reservoir has a source of natural energy to displace oil from the reservoir

47
to the production wells, the oil rate is controlled primarily by the flow resistance in the immediate
vicinity of the wellbore. Cyclic steam injection heats the reservoir rock around the wellbore and
permits this region to remain at an elevated temperature for long periods of time.
The essential requirement for successful CSS is a source of natural reservoir energy. Reservoir
energy may be available in the form of (1) fluid expansion by solution-gas drive or reduction in
reservoir pressure, (2) natural waterdrive, (3) gravity drainage, or (4) compaction.
CSS is examined by first considering a depletion-type reservoir. Consider a heavy-oil primary
production rate from a reservoir. The reservoir is developed on 10-acre spacing. Permeability is 3
darcies, and thickness is 100 ft. The oil rate at reservoir temperature can be estimated provided
that there is sufficient reservoir energy to maintain the reservoir pressure at 500 psi at the effective
drainage radius of the well, defined by
𝑟𝑒 = √(𝐴/𝜋)(43,560𝑓𝑡 2 ∕ 𝑎𝑐𝑟𝑒)
where A = well spacing (10 acres) and re = effective drainage radius (372 ft). The oil production
rate when the wellbore pressure is maintained at pw is given by the equation below, where oilfield
units (stock-tank barrels, days, darcies, feet, centipoise, and psi) are used.
7.082𝑘𝑜𝑐 ℎ(𝑝𝑒 − 𝑝𝑤 )
𝑞𝑜𝑐 =
𝐵𝑜𝑐 𝜇𝑜𝑐 ln(𝑟𝑒 ∕ 𝑟𝑤 )
If the reservoir pressure was 500 psi at discovery and the viscosity at reservoir temperature is 1,000
cp, the initial oil rate is computed as
(7.082)(3.0𝑑𝑎𝑟𝑐𝑖𝑒𝑠)(100𝑓𝑡)(500 − 100)𝑝𝑠𝑖
𝑞𝑜𝑐 =
(1.1𝑏𝑏𝑙⁄𝑆𝑇𝐵)(1.000𝑐𝑝) ln(372⁄0.5)
= 116.6 STB/D
As this reservoir is depleted, the reservoir pressure falls and the production rate declines. For
example, when the effective reservoir pressure at the boundary is 50 psi and the bottomhole
pressure (BHP) in the production well is maintained at 10 psi, the production rate will be

48
𝑞𝑜𝑐 = 116.6[(50 − 10)⁄(500 − 100)]
= 11.7STB/D
4.3 Estimating the Radius of the Heated Zone—The Marx and Langenheim Model.
The radius of the heated region can be estimated using a model developed by Marx and
Langenheim.
Energy Requirements for Heating Porous Rock: The amount of energy required to increase the
temperature of a porous rock is easily calculated from thermodynamic tables and from heat
capacity data at constant pressure. Eq. 8.49 gives the energy (in Btu) required to increase the
temperature of 1 ft3 of reservoir rock from an initial value, Tr, to a higher value, T (in °F).
Q = M (T - Tr)
M is the average volumetric heat capacity of the fluid-saturated rock over the temperature range
Tr to T. If T is the temperature of injected steam, Ts, then,
𝑀 = (1 − ∅)𝜌𝑟 𝐶𝑟 + ∅𝑆𝑜 𝜌𝑜 𝐶𝑜 + ∅𝑆𝑤 𝑝𝑤 𝐶𝑤 + ∅𝑆𝑔 𝜌𝑠 𝐶𝑠
where Cr, Co, Cw, and Cs are mean heat capacities at constant pressure (Btu/lbm-°F) for the rock,
oil, water and steam if present. The mean heat capacities of each component are based on the
temperature difference, Ts – Tr.

49
CHAPTER 5: STEAM DRIVE (STEAM INJECTION)
5.1 Introduction
In CSS and steamdrive, reservoir heating is accomplished by the injection of steam. The steam
drive process (or processes) are considerably different in performance from hot-water drives due
to the presence and effects of the steam which causes low molecular weight constituents in the oil
to be distilled and carried along as hydrocarbon constituents in the gas phase. When and wherever
the steam condenses, the condensable hydrocarbon components also condense and reduce the
viscosity of the crude oil at the condensation front. Moreover, the condensing steam makes the
displacement process more efficient and improves the sweep efficiency. Thus, the net effect is that
recovery from steam drives is significantly higher than from hot-water drives.
An important additional phenomenon affecting displacement in steam drives is the steam

distillation of the relatively low-boiling fractions in the oil. Distillation causes the vapor phase to
be composed not only of steam but also of condensable hydrocarbon vapors. Some hydrocarbon
vapors will condense along with the steam, mixing with the original crude and increasing the
amount of relatively light fractions in the residual oil trapped by the advancing condensate water
ahead of the front. Dilution by the low-boiling constituents causes some of the trapped oil to be
displaced by the condensed water. The remainder of the oil is stripped by the steam of all the
remaining low boilers to leave the higher boiling residuum. The lower-boiling constituents help to
regenerate and maintain a solvent bank just downstream of the condensation front. During the
process, the composition of the produced oil generally does not change until the steam zone is
relatively near, at which point the volatile content increased markedly.
5.2 Mechanism of Steam Drive.
Steam drive injection (steam injection) has been commercially applied since the early 1960s. The
process occurs in two steps: (1) steam stimulation of production wells, that is, direct steam
stimulation; and (2) steam drive by steam injection to increase production from other wells
(indirect steam stimulation). Steam drive differs considerably in performance from hot-water
drive—the difference in performance is due to the presence and effects of the condensing vapor.
The presence of the gas phase causes low-boiling constituents present in the crude oil to be distilled
and carried along as hydrocarbon components in the gas phase. When the steam condenses, the

50
condensable hydrocarbon constituents of the vapor also condense and reduce the viscosity of the
crude oil at the condensation front. In addition, the condensing steam makes the displacement
process more efficient and improves the sweep efficiency and the net effect of the steam drive is
that crude oil recovery is significantly higher than the crude oil recovery from hot-water drives.
However, both steam and hot-water drives (1) improve oil mobility by reducing viscosity and (2)
reduce residual oil at high temperatures. They achieve those effects to such a degree that the
recovery of viscous crudes is better than could be attained by waterflooding. Moreover, all
phenomena encountered in hot-water displacement are also present in steam displacement
operations, not only because the arrival of the condensation front is preceded by a hot-water drive
but also because a hotwater phase is present along with steam in all cases of practical interest.
However, the magnitudes of these similar forces may be different as a result of the vapor phase.
The steam drive process is much like a conventional waterflood insofar as when a pattern
arrangement is established, steam is injected into several injection wells while the oil is produced
from other wells. This is different from the steam stimulation process, whereby the oil is produced
from the same well into which the steam is injected. As the steam is injected into the formation,
the thermal energy is used to heat the reservoir oil but, unfortunately, the energy also heats the
entire environment such as formation rock as well as water and some energy is also lost to the
underburden and overburden. Once the oil viscosity is reduced by the increased temperature, the
oil can flow more readily to the producing wells. The steam moves through the reservoir and comes
in contact with cold oil, rock, and water. As the steam comes in contact with the cold environment,
it condenses and a hot water bank is formed. This hot water bank acts as a waterflood and pushes
additional oil to the producing wells.
Typically, a hot-water drive involves the flow of only two phases— water and oil—whereas steam-
based processes and combustion processes, on the other hand, always involve a third phase—gas.
Thus, the hot water drive is a displacement process in which oil is displaced by both the immiscible
hot water and immiscible cold water. Except for temperature effects and the fact that they generally
are applied to relatively viscous crude oils, hot-water floods have many elements in common with
conventional waterfloods. Moreover, because of the presence of water in all petroleum reservoirs,
displacement by hot water must occur to some extent in all thermal recovery processes and is

51
known as one of the processes that contribute to the displacement of oil in the zones downstream
of both steam drives and combustion drives.
Several mechanisms have been identified that are responsible for the production of oil from a
steam drive. These include thermal expansion of the crude oil, viscosity reduction of the crude oil,
changes in surface forces as the reservoir temperature increases, and steam distillation of the
lighter portions of the crude oil. In fact, an important phenomenon affecting displacement in steam
drives is the steam distillation of the relatively light fractions in the crude. Distillation causes the
vapor phase to be composed not only of steam but also of condensable hydrocarbon vapors. Some
hydrocarbon vapors will condense along with the steam, mixing with the original crude and
increasing the amount of relatively light fractions in the residual oil trapped by the advancing
condensate water ahead of the front. Dilution by low-boiling constituents causes some of the
trapped oil to be displaced by the condensed water while the remainder is stripped by the steam of
all the remaining light ends, thus leaving less volatile residue. The lower-boiling constituents that
are stripped from the bypassed oil help to regenerate and maintain a solvent bank just downstream
of the condensation front. The produced crude oil may not change in composition until the steam
zone is relatively close, at which point the volatile content of the produced crude oil will increase.
In summary, the light ends in the vapor contribute to the high volatile content of the oil produced
after steam breakthrough.
When there is some natural reservoir energy, steam stimulation normally precedes steam drive. In
steam stimulation, heat is applied to the reservoir by the injection of high-quality steam into the
produce well. This cyclic process, also called huff and puff or steam soak, uses the same well for
both injection and production. The period of steam injection is followed by production of reduced
viscosity oil and condensed steam (water). One mechanism that aids production of the oil is the
flashing of hot water (originally condensed from steam injected under high pressure) back to steam
as pressure is lowered when a well is put back on production.
Mechanisms of oil recovery due to this process include (1) reduction of flow resistance near the
well bore by reducing the crude oil viscosity and (2) enhancement of the solution gas drive
mechanism by decreasing the gas solubility in an oil as temperature increases. Often, in heavy-oil
reservoirs, the steam stimulation process is applied to develop injectivity around an injection well
and, once injectivity has been established, the steam stimulation process is converted to a

52
continuous steam drive process. In some projects, producing wells are periodically steam
stimulated to maintain high production rates. Normally, stream drive projects are developed on
relatively close well spacing to achieve thermal communication between adjacent injection and
production wells. To date, steam methods have been applied almost exclusively in relatively thick
reservoirs containing viscous crude oil.
Reservoir Heating by Gravity Override.
The reservoir heating models described earlier in this chapter assume that the injected fluid is
uniformly distributed through any vertical cross section where fluids flow. This is a convenient
assumption for the development of simple heating models but is not the case in practice. The
densities of steam and water at the same temperature differ by factors of approximately 10 to 100
in the range of temperature and pressure used in most steamdrive projects. When there are no
impermeable barriers to flow that extend between wells, the injected steam will segregate by
gravity, with saturated vapor rising to the top of the formation and water flowing in the bottom of
the heated region. Reservoir heating occurs as the steam zone spreads horizontally across the top
of the reservoir and expands vertically. This process is called gravity override, gravity segregation,
or gravity overlay.
Fig. 8.50 compares a gravity-override (or bypass) model for reservoir heating with the frontal-
advance model. The bypass, or gravity-override, model is characterized by relatively rapid
segregation of the steam in the top of the reservoir, irrespective of where the steam is injected in
the formation. Thus, when steam is injected into a viscous-oil reservoir, the vapor phase will
migrate to the top of the reservoir, driven by gravity forces that promote the separation of vapor
and liquid phases. The rate of migration is controlled by the viscosity of the crude oil flowing
countercurrent to the rising steam and the vertical permeability in the strata.
Gravity segregation is a complex process. When gravity segregation is a critical factor, numerical
simulation is necessary to predict reservoir heating and oil-displacement performance.

53
Figure 17: Idealized temperature distribution in (a) frontaladvance model and (b)
bypass or override model (Prats 1982).

54
CHAPTER 6: IN-SITU COMBUSTION.

6.1 Introduction.
In-situ combustion is a displacement process in which an oxygen-containing gas is injected into a
reservoir where it reacts with the crude oil to create a high-temperature combustion front that is
propagated through the reservoir. In most cases, the injected gas is air, although the use of 100%
oxygen has been reported. The fuel consumed by the combustion front is a residuum produced by
a complex process of cracking, coking, and steam distillation that occurs ahead of the combustion
front. In-situ combustion is possible if the crude-oil/rock combination produces enough fuel to
sustain the combustion front. In-situ combustion field tests have been carried out in reservoirs
containing gravities from 9 to 40 °API.
6.2 Types of combustion.
Based on the respective directions of front propagation and air flow, the process can be forward
(when the combustion front advances in the same direction as the air flow) or reverse (when the
front moves against the air flow).
6.2.1 Reverse Combustion.
This process could be a useful way to produce very heavy oils with high viscosity. In brief, it has
not been successful economically for two major reasons.
First, combustion started at the producer results in hot produced fluids that often contain unreacted
oxygen. These conditions require special, high-cost tubulars to protect against high temperatures
and corrosion. More oxygen is required to propagate the front compared to forward combustion,
thus increasing the major cost of operating an ISC project.
Second, unreacted, coke-like heavy ends will remain in the burned portion of the reservoir. At
some time in the process, the coke will start to burn, and the process will revert to forward
combustion with considerable heat generation but little oil production. This has occurred even in
carefully controlled laboratory experiments.
In summary, reverse combustion has been found difficult to apply and economically unattractive.
6.2.2 Forward Combustion.
Because only forward combustion is practiced in the field, we will only consider this case. Forward
combustion can be further characterized as “dry” when only air or enriched air is injected or “wet”
when air and water are coinjected.
6.2.2.1 Dry Combustion.

55
The first step in dry forward ISC is to ignite the oil. In some cases, autoignition occurs when air
injection begins if the reservoir temperature is fairly high and the oil is reasonably reactive.
After ignition, the combustion front is propagated by a continuous flow of air. Rather than an
underground fire, the front is propagated as a glow similar to the hot zone of a burning cigarette
or hot coals in a barbecue. As the front progresses into the reservoir, several zones exist between
injector and producer as a result of heat, mass transport, and chemical reactions.
Fig. 16.14 is an idealized representation of the various zones and the resulting temperature and
fluid-saturation distributions. In the field, there are transitions between zones; however, the
concepts illustrated provide insight on the combustion process.
Figure 18: Schematic diagram of temperature and saturation profiles and zones in ISC.

56
6.2.2.2 Wet Combustion
A large amount of heat is stored in the burned zone during dry forward ISC, (Fig. 17) because the
low heat capacity of air cannot transfer that heat efficiently. Water injected with the air can capture
and advance more heat stored in the burned zone.
During the wet combustion process, injected water absorbs the heat from the burned zone,
vaporizes, moves through the burning front, and expands the steam plateau. As a result, the steam
plateau is extended. This results in faster heat movement and oil displacement.
Depending on the water/air ratio, wet combustion is classified as (1) incomplete when the water is
converted into superheated steam and recovers only part of the heat from the burned zone, (2)
normal when all the heat from the burned zone is recovered, and quenched or (3) super wet when
the front temperature declines as a result of the injected water.
When operated properly, water-assisted combustion reduces the amount of fuel needed, resulting
in increased oil recovery and decreased air requirements to heat a given volume of reservoir. Up
to a 25% improvement in process efficiency can be achieved. Determination of the optimum
water/air ratio is difficult because of reservoir heterogeneities and gravity override that can affect
fluid movement and saturation distributions. Injecting too much water can result in an inefficient
fire front, thus losing the benefits of the process.
Some authors recommend, as a best practice, injecting water at high rates to achieve “partially
quenched combustion.” This method has limited application. A high-temperature burn is preferred
but is difficult to achieve with oils that are not highly reactive. Injecting large amounts of water
can lower combustion temperatures, resulting in a greater fraction of oil burned and higher costs
for oxygen. At the same time, these types of burns only partially oxidize the oil. This partial
oxidation results in a much more viscous liquid, which in turn lowers the flow rate. So, in brief, if
water injection is used, great care should be taken to assure that liquid water never reaches the
high-temperature combustion front.

57
6.3 Laboratory Studies:
ISC processes are largely a function of oil composition and rock mineralogy. The extent and nature
of the chemical reactions between crude oil and injected air, as well as the heat generated, depend
on the oil-matrix system. Laboratory studies, using crude and matrix from a prospective ISC
project, should be performed before designing any field operation.
6.3.1The Reactions.
The chemical reactions associated with ISC are complex and numerous. They occur over a broad
temperature range. They are grouped into three classes in ascending temperature ranges:
• Low-temperature oxidation (LTO)—heterogeneous gas/liquid reactions producing partially
oxygenated compounds and few carbon oxides.
• Medium-temperature reactions—cracking and pyrolisis of hydrocarbons to form fuel.
• High-temperature oxidation (HTO)—heterogeneous H/C bond breaking reactions in which the
fuel reacts with oxygen to form water and carbon oxides.
LTO can be described as oxygen addition to the crude oil. LTO yields water and oxygenated
hydrocarbons such as ketones, alcohols, and peroxides. LTO generally increases original oil
viscosity, boiling range, and density. LTO increases the amount of fuel and is promoted by low air
flux in the oxidation zone. Poor crude oxidation characteristics can also play a role. In heavy-oil
reservoirs (API gravity < 20°), LTO tends to be more pronounced when oxygen (rather than air)
is injected in the reservoir.
6.3.2 Kinetics.
Kinetics of combustion reactions can be defined by how fast the chemical reactions occur and how
much of the oil is affected.
Because crude oils contain hundreds of different compounds, it is impossible to accurately
represent all the reactions occurring during ISC. Even if it were possible to detail all the reactions,
the use of such information in numerical models would be impossible because of cost and
computer limitations.
6.4 Combining Material- and Heat-Balance Calculations.
Many useful and reasonably accurate calculations can be made on ISC to predict the behavior of
a proposed project.

58
6.4.1 First Assumptions.

Start by assuming that no combustion data are available to get an initial idea of the feasibility of a
project. This preliminary work gives the engineer a sound basis to decide whether further work
has economic promise.
Assume a sandstone formation with a porosity of 22%, a temperature of 100°F, a 24°API oil at a
saturation of 65%, and an injection pressure of 300 psia. Also assume the CO2/CO atomic ratio m
to be
𝐶𝑂2
𝑚= = 20
𝐶𝑂
This is a reasonable ratio to assume, based on both laboratory and field experience. Because
there are no tube run data, generalized correlation curves (Figs. 18 and 19) will be used to calculate
expected results.
Figure 19: H/C ratio vs. combustion

temperature Figure 20: Fuel availability vs. crude oil gravity

59
From Fig. 19, the fuel availability, W, for 24°API crude is
W = 0.95lbmC/100lbm rock
The apparent H/C atomic ratio (n) of the fuel is also needed. This is a function of the
combustionfront temperature, as shown in Fig. 18. Selected data from the graph are listed in
Table below. These data for 21.8°API crude are close enough to 24°API crude for initial
estimates.
6.5 Design Considerations.
Conditions favoring the use of ISC rather than steam include the following: (1) high reservoir
pressure at which steam is inefficient, (2) potential for severe wellbore heat losses (i.e., depth,
offshore, permafrost), (3) reservoir clay swelling in contact with fresh water, (4) limited water
supply, and (5) environmental regulations prohibiting steam generation.
Like any other injection process, the design of ISC projects must consider injection-pressure
limitations and reservoir flow resistance. These are especially important in heavy-oil reservoirs in
which combustion must occur in the high-temperature regime to be successful. The minimum air
flux needed to maintain high temperatures at the front is estimated to be 0.125 ft/D (0.04 m/d).
Because the burn-zone growth is directly proportional to the injected air, the maximum air-
injection rate determines the minimum lifetime of the project. Ways to increase the air-injection
rate are often needed, especially in heavy-oil reservoirs. They may include reduced well spacing,
cyclic steaming of injectors and producers, and an increase in injection pressure. These factors will
determine the compressor pressure and volume output.

60
Because it is often difficult to decide a priori which direction the major flow will take, operating
plans should remain flexible until field performance indicates which injection scheme best uses
the flow directions.
For the above reasons, many of the more successful ISC projects have been line-drive operations
that start near the top of the reservoir and move downdip. In such an operation, the direction of the
fire front is known. The operating engineers can then plan the completion and operating history in
a rational way that will mirror the front movement and breakthrough history.
6.6 Performance Prediction.
There are many methods used in predicting the production response to ISC. The easiest method is
essentially a tank balance. The oil and water produced are given by
Np =ϕVb(Soi −Sf)+0.4Soi(Vp −Vb)
and
Wp =Vb ϕ (Swi +Swf),
where Soi = initial oil saturation, fraction; Sf = oil saturation burned, fraction; Vb = volume burned,
m3; Np = oil produced, m3; Wp = water produced, m3; ϕ = porosity, fraction; Vp = volume of the
pattern, m3; and Swf = water saturation resulting from the combustion process, fraction.
If the volumes are in acre-ft and the production terms are in bbl, a multiplication factor of 7,758
must be used.
6.7 Screening Guidelines.
ISC is a complex process; it combines the effects of steamdrive, hydrocarbon miscible and
immiscible flooding, immiscible gas drive, and hot and cold waterflooding. Because of its
complexity, there is a misconception that combustion has a low probability of success. The truth
is that combustion is an economically attractive, proven recovery process, capable of economically
recovering a large fraction of the oil in place.
ISC can be applied to many different reservoirs. Some suggested screening guidelines are:

61
• Nature of the formation: The rock type is not important provided that the matrix/oil system is
reactive enough to sustain combustion. As in any drive process, high-permeability streaks are
detrimental. Swelling clays may be a problem in the steam-plateau area.
• Depth: The reservoir must be deep enough to ensure containment of the injected air in the
reservoir. There is no depth limit, except that this may affect the injection pressure.
• Pressure: This will affect the economics of the process but does not affect the technical aspects
of combustion.
• Temperature: Temperature will affect the propensity for autoignition but is otherwise not critical.
• Reservoir thickness: Thickness must be greater than 4 m (15 ft) to avoid excessive heat losses to
surrounding formations. Thick formations may present sweep-efficiency problems because of
gravity override.
• Permeability: This has to be sufficient to allow injection of air at the designed air flux. The air
injectivity is especially important for heavy-oil reservoirs. Conditions are favorable when kh/μ is
greater than 5md m/cp3.
• Porosity and oil saturation: These must be large enough to allow economic oil recovery. The
product, ϕSo, should be more than approximately 0.08 for combustion to be economically
successful.
• Oil gravity: This parameter is not critical. In-situ viscosity has to be small enough to allow air
injection and resulting oil production at the design rate.
• Oil nature: In heavy-oil projects, the oil should be readily oxidizable at reservoir and rock matrix
conditions. This relationship must be determined by laboratory experiments. The same laboratory
experiments can also determine the amount of air needed to burn a given reservoir volume. This
is key to the economics of the process.
ISC is applicable to a wide array of reservoirs. In fact, it is the only thermal method that can
presently be applied to deep reservoirs, though deep downhole steam generation is being tested. It
can be used at any stage of reservoir depletion; it can be used in special situations such as offshore
or in Arctic regions. Because of the lack of heat losses at the surface and in the injection wells, it
is the most thermally efficient thermal-recovery method. The injectant (air) is readily available.

62
Combustion allows wider well spacing than steam; economic results are comparable to those of
steam injection.
Several aspects of operating ISC projects must be considered. First is the large compression ratio
and associated costs required to inject air into the formation. Second is the planning and design
requirements for a combustion project; these are more difficult than for steam injection. Third is
extensive laboratory work to assess fuel availability, air requirements, and burning characteristics
of the crude that are required before designing ISC projects. Fourth is the high degree of technical
sophistication and the careful monitoring needed to ensure proper operation of a project. Fifth is
the limitation of numerical simulation and other techniques that makes predictions of recovery
more difficult than most other enhanced oil recovery methods.
Considerable improvements in the application of ISC have been made since the early projects.
New developments, such as application to light-oil reservoirs, and the use of horizontal wells are
reviving interest in ISC. This process deserves consideration for many reservoirs, including those
in hostile environments or those not amenable to other recovery methods.

PETE5103 Enhanced Oil Recovery

Uploaded by

Copyright:

Available Formats

PETE5103 Enhanced Oil Recovery

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

PETE5103 Enhanced Oil Recovery

Uploaded by

Copyright:

Available Formats

1

PETE5103: ENHANCED OIL RECOVERY

(Dr Ngong + Mr Kiven)

General Course outline: Principles of displacement; rock properties fluid

Second Part of the Course Prepared By Mr. KIVEN Paul NSIIDZEYUF

PETE5103: Enhanced Oil Recovery Lecture Notes Prepared by Kiven P.

Definition of Enhanced Oil Recovery (EOR).

PETE5103: Enhanced Oil Recovery Lecture Notes Prepared by Kiven P.

PETE5103: Enhanced Oil Recovery Lecture Notes Prepared by Kiven P.

Idealized Characteristics of an EOR Process

Efficient Microscopic and Macroscopic Displacement

Where E = overall displacement efficiency (oil recovered by process/oil in place at start of

Macroscopic displacement efficiency relates to the effectiveness of the displacing fluid(s) in

PETE5103: Enhanced Oil Recovery Lecture Notes Prepared by Kiven P.

Macroscopic displacement efficiency is improved by maintenance of favorable mobility ratios

PETE5103: Enhanced Oil Recovery Lecture Notes Prepared by Kiven P.

PETE5103: Enhanced Oil Recovery Lecture Notes Prepared by Kiven P.

PART I: CHEMICAL METHODS

PETE5103: Enhanced Oil Recovery Lecture Notes Prepared by Kiven P.

Figure 1: Micellar/polymer displacement process (Poettmann 1974).

PETE5103: Enhanced Oil Recovery Lecture Notes Prepared by Kiven P.

Surface-active agents, or surfactants, are chemical substances that adsorb on or concentrate at a

PETE5103: Enhanced Oil Recovery Lecture Notes Prepared by Kiven P.

Figure 3: Schematic of surface-active molecule (Ottewill 1984).

1.2.1 Classification and Structure of Surfactants.

PETE5103: Enhanced Oil Recovery Lecture Notes Prepared by Kiven P.

• Amphoterics—Amphoteric surfactants contain two or more characteristics of the previously

Figure 4: Surfactant chemistry types

When using foam in conjunction with steam flooding or any elevated-reservoir-temperature

PETE5103: Enhanced Oil Recovery Lecture Notes Prepared by Kiven P.

1.3 Micelles and Microemulsion

Figure 5: Formation of micelles.

PETE5103: Enhanced Oil Recovery Lecture Notes Prepared by Kiven P.

Figure 6: structure of microemulsion (Healy and Reed 1974).

PETE5103: Enhanced Oil Recovery Lecture Notes Prepared by Kiven P.

1.4 Phase Behavior of Microemulsions

The phase behavior of microemulsions is complex and dependent on a number of parameters,

PETE5103: Enhanced Oil Recovery Lecture Notes Prepared by Kiven P.

1.4.1 Phase-Behavior Representations on a Ternary Diagram.

A microemulsion usually is composed of at least five components: a surfactant, cosurfactant,

Figure 7: Quaternary diagram with pseudoternary diagram at constant sulfonate/alcohol ratio.

PETE5103: Enhanced Oil Recovery Lecture Notes Prepared by Kiven P.

Fig. 7 shows a general representation of “ideal” phase behavior of a microemulsion on a ternary

At high water concentrations, as Point S1 on Fig. 7, the microemulsion would be a water-external

Figure 8: Pseudoternary phase diagram for a micellar (microemulsion) solution.

PETE5103: Enhanced Oil Recovery Lecture Notes Prepared by Kiven P.

1.4.2 Effect of Brine Salinity on Phase Behavior.

Figure 9: Effect of salinity on microemulsion phase behavior (Healy et al. 1976).

PETE5103: Enhanced Oil Recovery Lecture Notes Prepared by Kiven P.

1.4.3 Solubilization Parameters.

Vo Volume of oil in microemulsion phase

Vw Volume of water in microemulsion phase

1.4.4 Phase Behavior and IFT

1.4.5 IFT as a Function of Salinity

PETE5103: Enhanced Oil Recovery Lecture Notes Prepared by Kiven P.

Figure 10: IFT as a function of salinity (Healy et al. 1976).

PETE5103: Enhanced Oil Recovery Lecture Notes Prepared by Kiven P.

1.4.6 Correlation of IFT and Solubilization Parameters.

PETE5103: Enhanced Oil Recovery Lecture Notes Prepared by Kiven P.

A similar equation has been proposed by Nelson (1982) to be more general