SPE 75207

A Guide to High Pressure Air Injection (HPAI) Based Oil Recovery

R.G. Moore SPE, S.A. Mehta SPE, and M.G. Ursenbach, University of Calgary

Copyright 2002, Society of Petroleum Engineers Inc.

this paper is to provide managers as well as engineers with a
This paper was prepared for presentation at the SPE/DOE Improved Oil Recovery Symposium reference that discusses several design and operational issues
held in Tulsa, Oklahoma U.S.A., 13–17 April 2002.
that are important to consider. It is based on the authors’
This paper was selected for presentation by an SPE Program Committee following review of
information contained in an abstract submitted by the author(s). Contents of the paper, as
nearly thirty years of research and consulting in the area of in
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to situ combustion and HPAI.
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at While the paper was not intended as a review of the large
SPE meetings are subject to publication review by Editorial Committees of the Society of
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
body of existing HPAI literature, there are several useful
for commercial purposes without the written consent of the Society of Petroleum Engineers is references that serve as a starting point for understanding the
prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300
words; illustrations may not be copied. The abstract must contain conspicuous process. The papers by Erickson, et al.1 and Kumar, et al.2
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O. provide good summaries of the HPAI projects that were
Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.
started by Koch Exploration Company in the Buffalo Red
River Units and the Medicine Pole Hills Unit located in the
Abstract Williston Basin; these projects continue to be operated,
In recent years, high pressure air injection (HPAI) has proven currently by Continental Resources, Inc. Watts, et al.3 describe
to be a valuable IOR process, especially in deep, high a newer project started in the nearby Horse Creek Field by
pressure, low permeability fields where other recovery Total Minatome Corporation, and a paper by Fassihi, et al.4
processes are uneconomic. discusses the light oil air injection process and addresses
This paper will provide engineers and engineering various recovery mechanisms based on field and laboratory
managers with a wide-ranging look at the key factors that data. Fassihi, et al.5 discuss the economics associated with
should be addressed when considering a high pressure air two separate air injection projects at West Hackberry,
injection (HPAI) based IOR process in a light oil reservoir. Louisana, and the Medicine Pole Hills Unit, N. Dakota.
The paper is based on many years of involvement of the Moore, et al.6,7 describe strategies for design and operation of
authors both in the laboratory and the field, as well as drawing successful air injection-based processes, and highlight the
on published literature. significant differences between air injection in light and heavy
The main focus is on key design and operating criteria oil reservoirs. Yannimaras and Tiffin8 and Tiffin and
that must be considered, including reservoir screening, air Yannimaras9 describe laboratory testing of the HPAI process
injection design, ignition, and monitoring. The benefits and at the Amoco Production Company laboratory using the
potential risks of HPAI are also discussed. combustion tube and the accelerating rate calorimeter, while
Along with the discussion of design and operating Greaves, et al.10 describe light oil recovery by air injection,
criteria, the paper contributes significantly in the comparison based on laboratory work at the University of Bath, U.K.
of oxidation/combustion kinetics for light oils versus heavy Field-specific numerical simulations of the HPAI process have
oils (HPAI versus in situ combustion), as well as in a been described by Kumar11, Glandt, et al.12, and Kuhlman13.
discussion of the oil mobilizing effects of a gas flood Finally, two classical air injection design papers by Nelson
compared to an advancing thermal (combustion) front. and McNeil14 and Gates and Ramey15 are important reading
for anyone involved in the design of air injection-based IOR
Introduction processes. As stated previously, this list does not represent the
In recent years, High Pressure Air Injection (HPAI) has complete literature on HPAI, but represents an excellent
received considerable attention as an effective IOR process, starting point in learning about the process.
based primarily on the success of several projects located
within the Williston Basin, and the vision and initiative by Process Overview
Amoco Production Company (now BP) in forming an High Pressure Air Injection (HPAI) is loosely defined as an
industrial consortium to develop a state-of-the-art HPAI IOR process in which compressed air is injected into a high
laboratory. This has led to increased interest from companies gravity, high pressure oil reservoir, with the expectation that
in exploring the suitability of the HPAI process for application the oxygen in the injected air will react with a fraction of the
in their onshore and offshore reservoir holdings. The intent of reservoir oil at an elevated temperature to produce carbon
2 R.G.MOORE, S.A. MEHTA, M.G. URSENBACH SPE 75207

dioxide. The resulting flue gas mixture, which is primarily quite readily, under a wide range of operating conditions.
CO2 and nitrogen, provides the mobilizing force to the oil Heavier oils, however, do not always enjoy the same
downstream of the reaction region, sweeping it to production predisposition for success. Faulty ignitions, insufficient air
wells. The gas-oil mixture may be immiscible, or partly or injection, and permeability restrictions due the rapid
completely miscible. In some situations, the elevated mobilization of oil into cooler, downstream parts of the
temperature reaction zone itself may provide a critical part of reservoir, can all lead to domination by the oxygen addition
the sweep mechanism in terms of incremental recovery. reactions. Indeed, the less than spectacular history of in-situ
In its simplest implementation, the process is initiated combustion in heavy oils arises from operation in the oxygen
simply by injecting air, which will spontaneously ignite the in- addition regime—often referred to as the “low temperature
place oil due to the high temperature and pressure conditions oxidation” mode.
in the reservoir. In situations where spontaneous ignition of
the native reservoir oil is not likely to occur—this can be Definitions. For the sake of definition, “in-situ combustion”
determined by laboratory testing—the ignition must be aided or “fireflooding” is used to refer to air injection based
with the injection of a chemical mixture capable of processes in heavy oil reservoirs, which require operation in
spontaneous ignition at the reservoir conditions, or by an the high temperature range (450°C plus), for successful
initial input of energy, usually provided by a downhole heater displacement of the oil by the oxidation zone. “High pressure
or a burner. Air injection is achieved using compressors that air injection” implies air injection into deep, light oil
are specifically designed for air at the pressure levels that are reservoirs, for which bond scission or combustion reactions
required to inject the desired volumes. are dominant in the 150 to 300°C or low temperature range.
Many candidate reservoirs for air injection processes show
Chemistry. The critical factor for success of HPAI is the oxidation characteristics that fall between those described
reaction between the oxygen and the hydrocarbons. Two above and it is for this reason that laboratory testing of the
possible reaction pathways exist. The first is referred to as oxidation characteristics is a highly recommended step in the
“bond scission” reactions, and represents the reactions where design of a new air injection process.
the oxygen breaks up the hydrocarbon molecules to
principally produce carbon dioxide and water; in other words, Air Injection Projects
these are the traditional combustion reactions. In many of the The notably successful air injection projects that are currently
high gravity, light oil reservoirs, these bond scission reactions active are primarily those being operated by Continental
occur in the 150 to 300°C range, while in heavier oils, bond Resources Inc. in the Williston Basin in South and North
scission reactions will not be the dominant oxidation reactions Dakota. The four original projects, three in the Buffalo Red
at temperatures below about 450°C. Bond scission reactions River Unit (BRRU) and one in the Medicine Pole Hills Unit
are critical to the success of HPAI based IOR processes. (MPHU), were started by Koch Exploration Company. The
The second possible set of reactions between oxygen and paper by Erikson, et al.1 describes the initial years of the
hydrocarbons are called “oxygen addition” reactions. In this projects up to 1992, and the MPHU project is further
scenario, oxygen atoms are chemically bound into the discussed by Kumar, et al.2 Both reservoirs were deep, high
molecular structure of the liquid hydrocarbons, producing pressure, thin dolomites with permeabilites in the 10 to 20 md
various oxygenated compounds such as hydroperoxides, range and high water cuts that made them poor candidates for
aldehydes, ketones and acids. The compounds tend to further waterflooding. Air injection was first started in BRRU in
react and polymerize with each other, forming heavier, less 1979, and by July 1992, 73 Bscf of air had been injected into
desirable oil fractions. To compound the problem, because the four projects. At this time cumulative incremental oil from
oxygen is being removed from the gas phase to the liquid the four projects was about 7 million barrels, for an overall
phase, and is not replaced with CO2, the pore pressure where air/oil ratio of 10.5 Mscf/bbl. While there is little published
the addition reactions occur will drop below that of the information in the last decade, public information indicates
flowing gas, which promotes trapping of the oil phase. that the projects are continuing successfully. Personnel from
In short, when oxygen addition reactions dominate, the Continental Resources indicate that horizontal wells are being
thermal zone becomes very ineffective at displacing oil and drilled and completed both as air injectors and producers in
the flowing gas stream undergoes significant shrinkage. BRRU and MPHU. The horizontal wells have successfully
Oxygen addition reactions appear to dominate at temperatures recovered new reserves and accelerated production. Injection
below about 150°C for lighter oils and are the dominant rates are about 3 to 4 times the injection rates of a typical
reaction mode for heavy oils at temperatures of less than vertical well. Continental has recently begun HPAI in the
300°C. Avoiding operation in an oxygen addition reaction West Medicine Pole Hills Unit in North Dakota, where the
mode is highly desirable for essentially all air injection-based field was developed with horizontal wells. They are also in
processes. Stated another way, the key to success in HPAI is the process of applying HPAI in the North Cedar Hills Unit
to design the process such that it ensures operation in the bond (also developed with horizontal wells) and are actively
scission mode. Fortunately, for many high gravity, high seeking other fields in which to apply the technology. Current
pressure, light oils, the bond scission reactions are the favored air injection into projects operated by Continental Resources
reaction path, and air injection processes tend to operate there exceeds 53 MMscfd.16
SPE 75207 A GUIDE TO HIGH PRESSURE AIR INJECTION (HPAI) BASED OIL RECOVERY 3

Why Consider HPAI? vertical sweep of the pay zone, so long as the oxidation
The success of the projects operating in North and South kinetics can be predominantly maintained in the bond scission
Dakota, which were pioneered by Koch and Amoco and are mode. The overall conformance downstream of the
now being expanded by Continental, show that HPAI is a combustion zone will, of course, be subject to gas over-ride, as
viable process which can lead to sustained economic oil for any other gas injection process.
production in reservoirs which are not considered good
candidates for waterflood. Air is available everywhere and the Keys for Successful Project Design
air compression technology is now very reliable. As with any The following are some important areas that must be
gas injection process, significant volumes of gas can be considered when designing a high pressure air injection
injected into tight reservoirs, and the lifting costs for operation process. It is not a complete list of considerations, but rather,
in deep reservoirs is not impacted by the need to produce and addresses those which are unique to HPAI.
treat large volumes of water. A further benefit associated with
the gas injection is the reservoir repressurization, which Air Compressors. In times past, air compressors were
manifests itself in some production wells returning to natural unreliable and often subject to mechanical breakdown, and
flow, thus eliminating the expenses associated with pumping. thus were often a weak link in in situ combustion projects;
While not a concern in the reservoirs which we would fortunately, however, most of these earlier problems have
normally associate with high pressure air injection, the high been overcome with improved technology, including synthetic
theoretical efficiency associated with in situ generation of lubricants. Air compressors are the heart and soul of an air
energy becomes a more important consideration as we look for injection process, so care is required in their selection,
alternatives to steam in an environment of increasing natural particularly in the selection of pressure and volume ratings of
gas prices, and increasing competition for water sources. the units. There are a variety of options available for
All air injection processes benefit from the fact that the obtaining compressed air, ranging from owning and operation
reservoir is an excellent scrubber for the sulfur compounds of compressors, to buying compressed air at the plant gate or
formed during the oxidation reactions and the operating wellhead. Further optimizations can be obtained in selecting a
temperatures are generally below those that result in the power source for the compressor; either natural gas or
generation of nitrogen oxides. Sequestration of the carbon electricity, depending on the local circumstances. The
dioxide can be achieved by re-injection of the product gas possibility of only injecting air during off-peak hours in a
either by co-injection with the air or injection into dedicated mature project may add to the suitability of the electricity
flue gas injectors. option, and may have favorable tax implications.
Air injection processes can maintain stable oxidation Successful ignition, which achieves the temperature range
zones at very low initial oil saturations, hence it is possible to corresponding to the establishment of a combustion zone will
institute this process into reservoir which have been result in a significant rise in the injection pressure. This leads
waterflooded to the economic limit or which are still to an increased compressor discharge pressure as the
undergoing water flood. In the latter case, the prime goal is to mobilized fluids bank up and form a restriction to the flow of
provide flue gas, which, along with the water, will increase the gas. Many operators select the compressor discharge too low,
vertical sweep of the combined flood. based on the prior reservoir pressure; hence they are faced
with restricted air injection rates during the critical period
Pore Scale Efficiency of a Combustion Zone. Because bond following ignition. While it adds to the cost of the units,
scission or carbon oxide forming reactions are normally selection of the compressor frame for discharge pressures that
referred to as combustion reactions, the terminology are just below the fracture pressure is recommended,
“combustion or burn zone” will be used to describe those particularly for pilots or other early operations.
oxidation zones in which carbon dioxide is the primary Selection of the air injection capacity is primarily
oxidation product. The actual temperature within the dependent on the volume of reservoir that the compressor is
oxidation zone may correspond to the low temperature range expected to service. From a pore level perspective, the
for light oils (150 to 300°C) or the high temperature range concept of “extinction radius” as introduced by Nelson and
(450 to 600°C) for heavy oils. McNeil14 is valid and, while an exact value of the minimum
air flux required to sustain the bond scission or combustion
Laboratory studies and cores taken from post-burn reactions is not known with any degree of certainty, heat
regions of fields show a complete absence of hydrocarbons in transfer considerations alone indicate that there will be a finite
the regions swept by a combustion zone. Approximately 5 to limit. Based on the rules of thumb proposed by Nelson and
10 percent of the in-place oil is consumed as fuel for the McNeil, this minimum flux is in the range of 39 scf/ft2d. This
process; the rest is mobilized and can be available for capture. should not be considered a firm value, as the minimum air flux
One of the unique features of air injection is the self- will be very dependent on whether the reservoir in question
correcting nature of the combustion zone, owing to the rapid must operate in the low or high temperature range to achieve
mobilization of oil into the downstream pores, which the desired carbon oxide forming reactions. An argument that
temporarily reduces the gas permeability and redirects the air is often heard is that HPAI only requires the oxidation
flow. This continual redistribution of the air promotes good reactions to remove oxygen from the injected gas, as oil
4 R.G.MOORE, S.A. MEHTA, M.G. URSENBACH SPE 75207

displaced by the combustion zone will never reach the improved oil recovery process), the ability to deliver air to the
production well. This may be true for a time, but the presence reservoir, and the ability of the oil to spontaneously ignite and
of residual hydrocarbon within the swept zone may cause the to sustain stable oxygen uptake reactions over the life of the
reaction zone to spread out, which will greatly increase the injectors. The ability of the oil to re-ignite at reservoir
relative levels of heat losses from the oxidation zone. The conditions is also important, if the reservoir is to be operated
increased relative heat load will ultimately cause the reaction at pressure levels that enable miscible or near miscible
temperatures to fall below the levels required for sustained operation.
operation in the reaction mode where the bulk of the oxygen The reactivity of a given oil is reservoir specific, and no
can be consumed. screening guides have yet been published which predict the
From a design viewpoint, the task of translating a oxidation characteristics of a specific reservoir. This results
“minimum flux” to the air injection rate at a given injector, is from the fact that the oxidation kinetics depend on the
complicated by the fact that the shape of the reaction zone is composition of the oil as well as that of the solid core matrix
not known with any high degree of certainty. For this reason, and, most probably the connate brine. While the burning
an operator has to specify both the volume of reservoir they characteristics of a heavy oil as observed within a one-
desire to service with a given injector and the operating life dimensional combustion tube are often reported to be
they desire for that injector. The “operating life” of an relatively pressure insensitive, this observation arises from the
injector may be totally dependent on the air injectivity of that fact that combustion tube tests are often forced to operate in
well, and, if this is the case, the operator needs to the high temperature mode (450°C plus), where the burning
acknowledge that the useful injector life is limited unless rate is controlled by the rate of delivery of oxygen to the
additional injectors are added to service the same reaction reaction front. Oxygen uptake rates are sensitive to pressure
zone. The operating life of an injector can be extended by the for high gravity oils which operate in the low temperature
use of horizontal wells, but this gives rise to other concerns, range (150 to 300 °C). This may be due to an oxygen partial
such as mal-distribution of the injected air and the method of pressure effect, or it may result from distillation effects that
ignition if the reservoir temperature is too low to enable are related to the total pressure. In any event, pressure is
spontaneous ignition within a reasonable period of time. known to affect the relative balance of total energy generation
Once the pore level and field design constraints are between the low and high temperature ranges, and it is very
evaluated, the operator must address the question of air important to evaluate oil reactivity at pressures which are
injection capacity in terms of the unit volume cost of the anticipated in the field project.
injected air. Too low an air injection capacity will lead to high
per unit volume air costs due to non-optimized labour costs. Laboratory Screening of Candidate Reservoirs. Laboratory
While air costs are very site specific, reported numbers vary testing of potential HPAI candidates, under conditions of
between $0.35 to $1.20 per Mscf. It is obvious that operators reservoir pressures and temperatures, serves several important
will be looking at alternatives such as carbon dioxide injection functions. First, it provides an indication of whether the oil
if the unit volume cost of air exceeds that of the CO2. will react in the desired oxidation mode at reservoir
An important caution when purchasing air compressors is conditions. Second, it can provide parameters, such as air and
that converted natural gas compressors are not directly fuel requirements (how much air is needed to sweep a unit
adaptable to air service due to inherent design differences. volume of reservoir, and how much fuel will be consumed in
Compressors suppliers must be consulted before purchasing that process), and air/oil ratios that can be used in conducting
used natural gas units for installation in an air injection preliminary economic assessments of field projects. Third, it
project. can supply kinetic data that can be used for numerical
simulations of the HPAI process. Fourth, it can provide
Condensate Production. The sweeping of light oil reservoirs produced gas compositions, water compositions and other
with large volumes of injected gas has the potential to strip the production related data that have been shown to correlate well
oil of light ends. It may be worth gathering and chilling the with the field, and can be used to monitor field performance.
vent gases, particularly in the early portion of the project, and Laboratory screening tests on candidate reservoirs should
recovering this valuable product. include combustion tube test(s), performed at the design
pressure for reservoir operation, supported by tests that are
Screening Air Injection Prospects. Screening parameters designed to evaluate the effect of temperature on the oxidation
published for in situ combustion projects are not directly kinetics. Combustion tube tests provide a direct observation
applicable to HPAI operations. The projects operated by of the ability of the oxidation reactions to maintain a stable
Continental in the Dakotas violate many of the screening process. Gas compositions (and the resulting gas phase
parameters which were primarily developed for heavy oil “in combustion parameters) corresponding to stabilized operation
situ” combustion projects. within the combustion tube are very similar to those expected
For HPAI projects which are to be initiated by for stable operation in the field. The one exception is CO,
spontaneous ignition, the key parameters are the amount of oil which shows very low concentrations in field projects, except
in place at the start of air injection (essentially the porosity when the combustion zone is very near to the production well.
times oil saturation or Φ·Soi product, as for any other
SPE 75207 A GUIDE TO HIGH PRESSURE AIR INJECTION (HPAI) BASED OIL RECOVERY 5

Accelerating rate calorimeter (ARC) tests and pressurized based on the observation that wells outside of those which
differential scanning calorimeter (PDSC) tests provide handle the bulk of the vent gas are generally the best
information on the energy generation rate associated with the producers. Where possible, the initial injectors should be
oxidation reactions as a function of temperature. When these placed in the most up-dip locations, with the selection of
units are configured for evolved gas analysis, it is possible to future sites being dependent on the performance of the project.
evaluate the product gas compositions as a function of At least two injectors should be installed at the start of the
temperature. These data are very useful for future field project, so that the operator has the opportunity to vary the
operation, as they enable the operator to interpret the average rates at a given well without having to shut in the compressors
reaction temperature for the oxidation zone (or zones) that is or vent excess air.
in communication with a given producer. Ramped
temperature oxidation (RTO) tests provide a direct measure of Keys for Successful Project Operation
the oxygen uptake rates as functions of temperature under air The above discussion has addressed keys for successful
flux and core packing conditions, which normally correspond project design. The following section examines unique issues
to those in the combustion tube. Post-test core analyses of that must be addressed during operation of a high pressure air
RTO tests identify the residual hydrocarbon that remains in injection project.
the swept zone following operation at different oxidation
conditions. These data provide important insights as to the Monitoring HPAI Projects. In situ combustion has been
nature of the hydrocarbon participating in the oxidation reported to be an engineering intensive process; however, this
reactions that occur within a given temperature range. should not be the case so long as the field project is operating
in the temperature range corresponding to stable combustion
Numerical Modeling. Numerical modeling of potential IOR reactions. Air injection processes require monitoring of the
projects is recognized as a valuable and cost-effective method product gas vent rates and composition, but the frequency can
of screening processes and optimizing their implementation. be monthly once the initial combustion zone is established.
In the case of HPAI, numerical modeling can be an effective Gas phase composition data provide information on the
tool in situations where immiscible, or miscible gas flood is average operating temperature within the reaction zone(s).
considered to be the dominant recovery mechanism. In this Oxygen concentrations in the produced gas provide
situation, compositional simulators can provide adequate information on the combustion efficiency, and they are
accuracy. In situations where accurate representation of the required to ensure that the well is not operating under
reaction zone is desired, the modeling is much more difficult. flammable conditions. Most operators use 3 mole percent
Generally, thermal fronts are very narrow, relative to typical oxygen as the maximum safe level; however, this limit is
grid sizes, and have proven difficult to represent accurately. dependent on the gas phase composition and oxygen
Recent advances in front-tracking techniques may provide a concentration, and, along with the temperature of the well,
solution to this problem. An additional limitation for impact corrosion of the production well. Elevated oxygen
numerical models is the inability to accurately represent the concentrations at the producers indicates problems with the
complex chemical reactions between the oxygen and the air process, which may either arise from gross air bypassing in
(oxidation kinetics). Most current models carry a simple set of thief zones or from a lack of residence time within the
oxidation kinetics that presumes that bond scission and elevated temperature reaction zone. While geological
thermal cracking reactions are occurring, but these models fail considerations are often blamed for oxygen bypass, it often
to predict when a shift from bond scission to oxygen addition relates to the reaction temperatures being too low for effective
reactions may occur. Developing pseudo-kinetic models that oxygen consumption in either the reaction region or
will predict significant changes in the oxidation mechanisms, downstream portion of the reservoir.
which in turn will impact the effectiveness of the oil The apparent atomic hydrogen/carbon ratio or the fraction
displacement efficiency of the HPAI process, remains as a of reacted oxygen converted to carbon oxides, both of which
significant technical challenge. are essentially controlled by the carbon dioxide/nitrogen ratio,
provide a direct indication of the mean temperature within the
Selection of Injectors. Intelligent selection of injectors must reaction zone. Projects involving air injection only should
be made by the people who are familiar with the field under have apparent hydrogen/carbon ratios of less than 3.0, and
consideration. Most operators want to minimize the number reacted oxygen converted to carbon oxides of greater than
of new wells that must be drilled; hence, it is important to 50% if the oxidation kinetics are operating in the proper mode.
select the injectors so as to make the best use of the existing Vent gas rates are important, as they indicate the relative
infrastructure. If no new wells are to be installed, it is level of communication of a given producer with the oxidation
important to select an existing well that shows good zones. High vent rate producers provide the most useful
communication with the reservoir. This is generally the best information with respect to the state of the oxidation reactions;
producer, which tends to upset the field personnel, but the key however, the high vent wells are often the up-dip wells while
to all air injection processes is the ability to put air in the the best oil producers are generally down-dip of the injectors.
reservoir. Injection wells should be distributed around the Carbon dioxide concentrations in up-dip wells, are often
field, as opposed to ordered in patterns. This suggestion is higher than those in the down-dip direction, which suggests
6 R.G.MOORE, S.A. MEHTA, M.G. URSENBACH SPE 75207

secondary reactions, involving either gas phase CO or the operation in the desired oxidation mode, and it allows for
burning of oxidized liquid phase components. interference testing to evaluate the effectiveness of the
Product gas combustion parameters provide a different injectors. Even for a pilot, more than one air injector
thermometer for estimating the temperature within the is obviously required, and the cycling injection strategy
oxidation zone; however, the best indicator of the oxidation requires that the in service injectors can handle the total air
reactions not being in the correct mode is the lack of oil capacity. It also assumes that sanding of the shut in injectors
production response due to the combustion zone. When is not a significant problem.
operating in the correct mode, oil displacement by the thermal The concept of scheduled shut in of air injectors is
zone is closely linked to the volume of air injected. The consistent with the author’s belief that is preferable to inject
relatively close spacing associated with heavy oil projects into fewer injectors at higher rates than to operate a large
allows for relatively quick observation of production number of injectors at lower rates. Needless to say, this is not
responses; however, this is not the case for the large well a universal belief within the air injection community.
spacings associated with light oil projects.
Oil Displacement by Elevated Temperature Zone. Another
Operating Strategies. The previous section discussed how to source of argument within the air injection community is the
interpret an air injection process, but the question remains as role of the combustion zone in a HPAI process. There is
to what to do if the process is not performing according to general agreement that, at a minimum, the reaction zone needs
expectations (gas phase combustion parameters or oil to remove the bulk of the injected oxygen, and preferably, to
production response). Assuming the problem is not due to avoid excessive shrinkage of the gas phase by replacing the
insufficient air injection capacity, the best approach is to shut reacted oxygen with as much carbon dioxide as possible. The
in the injector that is believed to service the production well, state of the oxidation zone that achieves these conditions is the
or wells, in question. This may be easier said than done if a same as that required for effective oil displacement by the
large number of injectors are involved; however, the concept thermal front; hence, the argument concentrates on how far to
of using shut in periods is based partly on the reports for propagate the oxidation zone.
heavy oil in situ combustion projects of increased oil Early production during a HPAI process is related to re-
production following termination of air injection, and partly pressurization and gas flood effects; hence, the influence of
on laboratory experience which has shown that it is possible to the thermal zone is secondary during the early life of an
change the oxidation mode from the low to the high injector. The subsequent impact of the oil displaced directly
temperature range by changing the oxygen partial pressure. In by the combustion zone will depend on the effectiveness of
the laboratory, this is achieved by injecting an inert gas the generated flue gas on oil displacement from outside of the
(nitrogen or helium), or by water injection. Water injection thermal region. If oil displacement from the regions not
can be used in a field situation; however, the application of directly affected by the thermal zone is high, the air/oil ratios
shut in periods is a more universal approach. associated with displacement of oil by the combustion zone
Starving the oxidation reactions of a fresh supply of will be so high that air injection to that portion of the reservoir
external oxygen will promote the decomposition of a portion will be stopped long before the combustion zone has advanced
of the oxidized liquid phase components that were formed due any appreciable distance within the reservoir. Reservoirs, such
to oxygen addition reactions. This is believed to result from as those which have previously undergone extensive
the lack of gas phase oxygen, causing the oxidation reactions waterflooding, will not likely be operated for a sufficiently
to draw on the internal supply of oxygen that is present in long time that the oil displaced directly by the thermal front is
some of the oxidized hydrocarbons. The change in oxidation important. Other reservoirs, such as those which have not been
mode is signified by elevated carbon dioxide concentrations in previously waterflooded, should sustain economic air/oil ratios
the product gas. The shut in period will also result in a for much longer times (i.e. larger volumes of injected air);
cooling off of the region swept by the oxidation reactions, hence, these are the reservoirs in which a greater volume of
which often leads to cessation of oxygen consumption by the the oil displaced by the thermal front will be captured.
residual hydrocarbon which are left if the oxidation reactions
were not operating in the bond scission or combustion mode. Operating Team. As with all IOR operations, the project can
If no change in the carbon dioxide/nitrogen ratios occurs be no better than the people involved. An important
following termination of air injection, it is a good assumption contributor to success is the inclusion of all levels of stake-
that the temperatures within the oxidation zone are below the holders from early on in the project design. This is not a novel
levels required for effective operation, and that air injection at concept, but if ignored, can lead to a general reluctance by the
the problem injector should be permanently terminated. staff to believe in the process.
While termination of air injection at specific injectors in
response to problems with the process is a viable operating Summary Comments
strategy, a preferred option is to swing the air injection The opinions expressed in this paper are based primarily on
capacity through the available injectors. This is not a novel laboratory observations using many types of devices and
concept, as it was used by Amoco Canada in their Morgan in involving a large number of reservoirs. The key observations
situ combustion project17. Cyclic operation promotes from laboratory work is that all oils do not burn in the same
SPE 75207 A GUIDE TO HIGH PRESSURE AIR INJECTION (HPAI) BASED OIL RECOVERY 7

fashion, but all oils show the same characteristic that effective 5. Fassihi, M.R., Yannimaras, D.V., Westfall, E.E., and Gillham,
oil displacement requires the oxidation reactions to be T.H., “Economics of Light Oil Air Injection Projects”, SPE
operating in the bond scission or combustion mode. Koch 35393, Presented at the 1996 SPE/DOE Tenth Symposium on
Exploration Company, and now Continental Resources Inc., Improved Oil Recovery, Tulsa, Oklahoma, 21-24 April 1996, 9
pp.
have shown that HPAI can be successfully operated in the
6. Moore, R.G., Mehta, S.A., Ursenbach, M.G., and Laureshen,
field. Their success confirms the suitability of HPAI as a C.J., “Strategies for Successful Air Injection-Based IOR
viable process for the recovery of light oils. Laboratory Processes”, Paper 235, Proceedings of the 7th UNITAR
testing suggests that there is no reason why Continental’s International Conference on Heavy Crude and Tar Sands,
success cannot be translated to many other light and medium Beijing, China, 27-30 October 1998, pp. 2171-2178
gravity reservoirs onshore and offshore. 7. Moore, R.G., Laureshen, C.J., Mehta, S.A., and Ursenbach,
The two key requirements for the successful M.G., “Observations and Design Considerations for In Situ
implementation of the HPAI process are that the oil must be Combustion Projects”, J. of Cdn. Petroleum Tech., 38(13)
able to sustain the combustion reactions, and that sufficient air (Special Edition 1999)
8. Yannimaras, D.V. and Tiffin, D.L., “Screening of Oils for In-
must be injected to maintain the oxidation reactions in the Situ Combustion at Reservoir Conditions via Accelerating Rate
bond scission or combustion mode. The former requirement Calorimetry”, SPE RE, 10(1) (Feb. 1995), pp. 36-39.
can only be accessed through direct laboratory testing at the 9. Tiffin, D.L., and Yannimaras, D.V., “The In-Situ Combustion
conditions that are to be encountered in the field. The latter Performanace of Light Oils as a Function of Pressure (1000 to
requirement is addressed by installing sufficient air injection 6000 psig)”, In Situ, 21(1) (1997), pp. 47-64
capacity, and by adjusting the air injection strategy so as to 10. Greaves, M., Ren, S.R., Rathbone, R.R., Fishlock, T., and
promote the operating conditions which lead to effective oil Ireland, R., “Improved Residual Light Oil Revcovery by Air
displacement in both the thermal and downstream regions. Injection (LTO Process)”, J. Cdn. Petroleum Tech., 39(1) (Jan.
2000), pp. 57-61
11. Kumar, M., “Simulation of West Heidelberg In-Situ Combustion
Acknowledgments Project”, SPE 16724, Presented at the 62nd Annual Technical
The authors wish to acknowledge the support of the In Situ Conference and Exhibition, Dallas, Texas, 27-30 September
Combustion Research Group in the Department of Chemical 1987, pp. 283-294
and Petroleum Engineering at the University of Calgary. We 12. Glandt, C.A., Pieterson, R., Dombrowski, A. and Balzarini,
also acknowledge the support of industry and in particular M.A., “Coral Creek Field Study: A Comprehensive Assessment
Amoco Production Company (now BP) High Pressure Air of the Potential of High-Pressure Air Injection in a Mature
Injection research group. We also acknowledge the ongoing Waterflood Project”, SPE 52189, Presented at the SPE Mid-
support of the Canadian and Alberta governments. Continent Operations Symposium, Oklahoma City, Oklahoma,
28-31 March 1999, 11 pp.
13. Kuhlman, M.I., “Simulation of Light-Oil Air Injection
References IntoViscous-Dominated and Gravity-Stable Reservoirs”, SPE
1. Erikson, A., Legerski, J.R., and Steece, F.V., “An Appraisal of
59331, Presented at the 2000 SPE/DOE Improved Oil Recovery
High Pressure Air Injection (HPAI) on the In Situ Combusiton
Symposium, Tulsa, Oklahoma, 3-5 April 2000.
Results from Deep, High Temperature, High Gravity Oil
14. Nelson, T.W., and McNeil, J.S., “How to Engineer an In Situ
Reservoirs,” Presented at the 50th Anniversary Field Conference
Combustion Project”, The Oil and Gas J., June 6, 1961, pp. 58-
of the Wyoming Geological Association Guidebook, 1994
65
2. Kumar, V.K., Fassihi, M.R., and Yannimaras, D.V., “Case
15. Gates, C.F., and Ramey, H.J., “A Method for Engineering In-
History and Appraisal of the Medicine Pole Hills Unit Air
Situ Combustion Oil Recovery Projects”, JPT, February 1980,
Injection Project,” SPE RE, 10(3), (Aug. 1995), pp. 198-202
pp. 285-294
3. Watts, B.C., Hall, T.F., and Petri, D.J., “The Horse Creek Air
16. Taylor, A.R., Continental Resources Inc., Private
Injection Project: An Overview”, SPE 38359, Presented at the
Communication, April 2002
1997 SPE Rocky Mountain Regional Meeting, Casper,
17. Marjerrison, D.M., and Fassihi, M.R., “Performance of Morgan
Wyoming, 18-21 May 1997, pp. 143-154
Pressure Cycling In-Situ Combustion Project”, SPE/DOE
4. Fassihi, M.R., Yannimaras, D.V., and Kumar, V.K., “Estimation
27793, Presented at the SPE/DOE 9th Symposium on Improved
of Recovery Factor in Light Oil Air Injection Projects”, SPE RE, Oil Recovery, Tulsa, Oklahoma, 17-20 April 1994, pp. 109-117
12(3) (Aug. 1997), pp. 173-178.