SPE 65124

EOR Screening for Ekofisk

T.B. Jensen, K.J. Harpole, and A. sthus, Phillips Petroleum Company

Copyright 2000, Society of Petroleum Engineers Inc.

This paper was prepared for presentation at the SPE European Petroleum Conference held
in Paris, France, 2425 October 2000.
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
presented, have not been reviewed by the Society of Petroleum Engineers and are subject
to 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 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 for commercial purposes without the written consent of the Society of Petroleum
Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not
more than 300 words; illustrations may not be copied. The abstract must contain
conspicuous acknowledgment of where and by whom the paper was presented. Write
Librarian, SPE, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-9529435.

ABSTRACT
An investigation of alternative EOR processes having
potential application in the giant Ekofisk chalk field is
presented. Technical feasibility, process readiness, oil
recovery potential, and related uncertainties and risks of five
selected EOR processes, namely hydrocarbon (HC) WAG,
nitrogen (N2) WAG, carbon dioxide (CO2) WAG, air
injection and microbial EOR (MEOR), are assessed for
possible application at Ekofisk. The objective of the
screening study was to evaluate and rank the EOR
alternatives and to select the most attractive process(es) on
which to pursue further work toward possible field pilot
testing. The focus of the paper is on the technical assessment
of the relative oil recovery potential of each process, and on
the importance of identifying critical operational and
logistical considerations for implementation of an EOR
processes in the offshore North Sea operating environment.
Estimates of potential EOR incremental oil recovery for
the Ekofisk field can be quite significant. However, key
project development and implementation issues and
additional cost elements must be weighed equally with oil
recovery forecasts in any EOR process ranking. Some of
these issues (e.g. injection gas supply, facilities
requirements, and the impact of EOR on chalk compaction,
subsidence and wellbore integrity) may be significant
enough to eliminate a process from further consideration.
In addition, there are significant differences in the
quantity and quality of key laboratory and field data
supporting the viability of the various EOR processes being
considered. Only a limited amount of field-specific data are
available to calibrate the performance predictions for some
of the processes. There is also a wide variation in the
technical readiness of each process to begin field pilot
design studies. Table 1 summarizes the state of technical
readiness for field implementation of each process and
identifies some of the major risk elements and remaining
work required to progress these EOR processes at Ekofisk.

BACKGROUND
The Ekofisk Field is located in the Norwegian Sector of the
North Sea, Figure 1. The reservoir is an elongated anticline
with the major axis running North-South covering roughly
12,000 acres, Figure 2. It produces from two fractured chalk
horizons, the Ekofisk and Tor Formations, separated by a
tight zone. The overlying Ekofisk Formation has a depth of
about 9,600 feet and thickness varies from 350 - 500 feet
with porosities from less than 30% to 48%. The underlying
Tor Formation thickness varies from 250 - 500 feet with
porosities from less than 30% to 40%. About two thirds of
the 6.4 billion STB OOIP is in the Ekofisk Formation. The
initial reservoir pressure was 7135 psia at a depth of 10,400
feet. The field initially contained an undersaturated volatile
oil with a bubble point pressure of 5560 psia at the
temperature of 268F.
Ekofisk1 was discovered in 1969 and test production was
started in 1971 from the discovery well and three appraisal
wells. Commercial test rates prompted development of the
field from three platforms. Permanent facilities with 54 well
slots and 300,000 STB/D (design capacity) process facilities
were operational in May 1974. Development drilling was
started June 1974. Oil production peaked at 350,000 STB/D
in October 1976. Produced gas was reinjected2 until a gas
pipeline was installed to Emden, Germany, September 1977.

Figure 1. Map of North Sea with Ekofisk location.

T.B. JENSEN, K.J. HARPOLE, A. STHUS

Figure 2. Top structure map for Ekofisk field

Ekofisk went through the bubble point in 1976 and GOR
went from 1,500 to 9,000 SCF/STB by 1986. Primary
depletion recovery (with reinjection of gas in excess of
contract quantities) was initially estimated to be 18% of
OOIP of which 6-8% was produced above bubble point.
Improved recovery studies were initiated soon after start of
primary production.3 Favorable water imbibition results
prompted a decision in 1983 to waterflood the northern Tor
Formation using unheated, 42F, seawater. A 30 slot water
injection platform with 375,000 BWPD injection capacity
was approved. Water injection started in 1987 to develop
162 MMBOE. Favorable waterflood response prompted
fieldwide water-flood expansion4. Water injection capacity
was increased to 820,000 BWPD which added almost 300
MMBOE.
Reservoir compaction and seabed subsidence continue to
be a very important consideration in Ekofisk reservoir
management strategy. A 1992 field study5 concluded that the
optimum strategy for Ekofisk would be to minimize future
seabed subsidence by arresting pressure decline with
pressure maintenance. Due to the age of the Ekofisk
facilities and continued seabed subsidence, a decision was
made in 1994 to gradually phase out existing facilities and
install a new 50 slot wellhead platform and new process and
transportation facilities. Drilling from the new wellhead
platform started in October 1996. The new process and
transportation facilities were commissioned in August 1998.

SPE 65124

The current ultimate oil recovery estimate from Ekofisk

is 38% of OOIP. The increase in recovery from the initial
estimate of 18% is due primarily to the extensive water
flooding, implementation of effective well monitoring and
stimulation techniques, compaction drive, aggressive infill
drilling, the Ekofisk redevelopment and extended field life,
and overall field optimization.
Current plans for Ekofisk include continued water
injection until 2012. At that point water injection will cease,
and the field will be depressurized to the end of the license
period in 2028. Opportunities to optimize the waterflood
with continued infill drilling exist. Also, a major
recompletion/sidetrack program is being evaluated for a
number of existing water injection wells. This ongoing
optimization of the Ekofisk waterflood has the potential for
significant incremental oil recovery. Opportunities also exist
for optimization of reserves during blowdown, being
primarily associated with acceleration of reserves with
additional production wells and enhanced compaction drive
due to pressure depletion. Enhanced recovery methods also
continue to be investigated as means of significantly
increasing reserves and accelerating recovery.
Several enhanced oil recovery (EOR) methods have been
considered for improving oil recovery beyond the Ekofisk
waterflood scenario. Recognizing that 1% enhanced oil
recovery represents about 64 million barrels of oil on
Ekofisk, significant resources have been used in studying
various EOR processes. Crestal gas injection, mainly for
operational swing gas purposes, has been ongoing since
1975. A total of 1.3 TSCF of gas, around 21% of the
cumulative produced gas, has been reinjected in Ekofisk as
of June 2000. Analysis of early production behavior above
the bubble point suggest that no immediate production of
free gas was experienced. Less than 1% of the injected
tracers were recovered following gas tracer injection tests in
1986 to 1988 which led to the conclusion that the injected
gas has covered a large area. Review of well performance
below the bubble point pressure has provided little clear
evidence of oil production response to variation in gas
injection rates. Based on numerous laboratory evaluations of
oil phase continuity and field performance the assumption of
capillary continuity has been used in modeling work to
evaluate gas injection processes.
A major crestal nitrogen injection study6 was completed
in the late 1980s. Project economics were not favorable,
due mainly to the significant costs of nitrogen rejection from
the gas sales stream. Efforts were also put into a study of
enhanced imbibition processes in the mid 1990s in order to
determine the potential for increased oil recovery by
injection of low concentration surfactants. The project was
terminated in 1997 due to laboratory measurements of high
surfactant adsorption, especially at the reduced reservoir
temperatures around existing Ekofisk water injection wells.
WAG processes have been evaluated in the laboratory,
through mechanistic simulations, and in one limited field
injection test at Ekofisk. A single-well WAG pilot test was
initiated in 1996 to test hydrocarbon gas injectivity into an
Ekofisk well which had injected a total of 44 million barrels
of water since 1991. Gas injection rates dropped to zero in a
matter of hours. Subsequent analysis of the injection test
showed that hydrate formation in the cold region around the

SPE 65124

EOR SCREENING FOR EKOFISK

injector was the most likely reason for the loss of gas
injectivity. Various remedial measures are under
considerations for gas injection in a possible future
hydrocarbon WAG scenario.
APPROACH AND PREMISES
In 1998 the decision was made to conduct a broad
scoping study to assess enhanced oil recovery potential for
the Ekofisk Field. The objective was to evaluate and rank
various EOR alternatives and to select the most attractive
process(es) on which to pursue further work toward possible
field pilot testing. Two underlying premises were adopted in
the approach to evaluating EOR alternatives for Ekofisk.
First, in the absence of definitive data to the contrary, any
minor technical or logistical uncertainties were assumed to
be favorably resolved prior to process implementation.
Second, in the absence of sufficient Ekofisk-specific
laboratory or field performance data, EOR process
performance in the Ekofisk reservoir was assumed to be
consistent with typical performance demonstrated in other
field applications or calculated based on industry field and
laboratory performance data. One consequence of these
premises is that those less mature, emerging technologies
(i.e. MEOR and air injection) may tend to be represented
more favorably relative to the more extensively studied and
tested technologies (HC WAG) where process performance
is more constrained by Ekofisk-specific lab testing and/or
field performance data.
Injectant Contact with Waterflood Residual Oil
One of the key issues determining incremental oil
recovery by any EOR process being considered for
application at Ekofisk is the ability of the EOR injectant to
contact and mobilize waterflood residual oil in the fractured
chalk. The premise for the process performance modeling
used to generate the forecasts in this screening study is that
injected gas will be able to move into the chalk matrix to
contact and viscously displace water flood residual oil. This
modeling premise is based on performance of the
imbibition/viscous displacement of oil by water under
current water flood operations as represented in the current
full-field Ekofisk model.
However, any gas injection EOR process following
waterflood at Ekofisk will operate under a different
(dominantly gas/water) flow system than that which is
dominant under water flooding. If gas preferentially flows
through the fracture system and its contact with residual oil
in the matrix blocks is either prevented by capillary
threshold entry pressures or is severely rate-limited by a
relatively slow process such as gas-water diffusion, then the
incremental oil recovery could be dramatically lower than
forecasted. The field simulation model forecasts and
laboratory experimental work to date were all conducted
under the implicit premise that injection gas will be able to
contact residual oil in the chalk matrix.
Although studied in the past,2 additional study of gas
flow and displacement mechanisms in fractured Ekofisk
chalk is needed to validate this premise. This involves
studies to understand the relative importance of viscous,
capillary and diffusion dominated mechanisms in fractured
Ekofisk chalk gas injection processes. While this issue may

to some extent be examined in laboratory experiments, it is a

field-scale issue and one of the central questions to be
addressed in a field pilot test of any gas injection EOR
process. Additional mechanistic modeling work would be
useful in further examination of this question and will be
needed to help design and interpret any field pilot test.
The Incremental Impact of Compaction
Any incremental impact (positive or negative) of a
specific EOR process on the mechanical properties of the
chalk, have not been incorporated into the process
performance modeling forecasts or project implementation
scenarios presented in this screening study. Such
incremental compaction impacts on recovery and/or
subsequent impacts on seabed subsidence may be caused by
chalk dissolution, alteration or compaction.
Modeling Approach Used To Generate EOR Forecasts
The SENSOR7 compositional simulator was used to
model gas injection phase behavior and the ability of HC
and non-HC injection gases to mobilize and recover
incremental oil in water flooded portions of Ekofisk. A
sector model having representative Ekofisk layer properties
and well spacing was used for the EOR process simulations.
The sector model was a 7x7x14 five-spot configuration with
layer properties taken from the central water flood area of
Ekofisk. The residual oil saturation under water-oil
displacement was 30% and the residual oil saturation under
gas-oil displacement was 25%. The EOR process
simulations used a 15-component EOS tuned to available
laboratory PVT data on Ekofisk fluids and the specific
injection gas.
All EOR simulation modeling used a single porosity
model with effective properties and viscous displacement
characteristics tuned to match the primary depletion and
waterflood performance in the history-matched full-field
Ekofisk model. This provides the EOR injectant unrestricted
access to water flooded residual oil in the chalk matrix. If
gas preferentially flows through the fracture system and its
access to residual oil in the matrix blocks is limited, then the
incremental oil recovery could be dramatically lower.
However, the effect should be similar for all the gas
injection processes.
The air injection oil displacement process involves first a
combustion effluent gas front (N2/CO2) which moves ahead
of the thermal reaction front, followed by a thermal reaction
zone with a complex of possible mechanisms including
steam distillation and low or high-temperature oxidation
reactions, depending on local reservoir conditions, O2 flux
rates, etc. Air injection was modeled in Sensor using a
hybrid approach by first forecasting the effect of the
combustion reaction effluent gas front. Following this, the
effect of the thermal front was simulated using a firstcontact miscible gas flood with displacement parameters set
to model those of the thermal displacement front (as
determined from prior industry combustion tube
experiments with North Sea oils). The velocity of the
miscible front was calibrated to the air utilization
requirement and velocity of the thermal front, and the fuel

T.B. JENSEN, K.J. HARPOLE, A. STHUS

requirement of the thermal reaction was modeled using a

miscible residual oil saturation of 5% PV.
Detailed modeling of the thermal reaction mechanisms
require laboratory data from reservoir core material and
fluids (combustion tube experiments using Ekofisk oil and
chalk) and a rigorous-physics thermal simulator, neither of
which were available or within the scope of this preliminary
study. Subsequent simulation modeling using the STARS
thermal simulator produced incremental oil recovery results
very close to those obtained using the hybrid modeling
approach.
Microbial (MEOR) oil recovery forecasts were generated
by up-scaling a composite of typical field project results. It
was necessary to use this empirical approach to estimate
MEOR performance as the exact mechanisms of microbial
oil recovery are still poorly understood and consequently
there are no simulation models available to make reliable
MEOR field performance forecasts.
Scaling of forecasts from sector model to full field
The various EOR production forecasts and relative
injection and production volumes were scaled up from the
individual sector model process simulations for comparison
and ranking in a hypothetical full-field EOR project
development. Scaling factors were chosen such that the final
upscaled incremental cumulative oil, cumulative gas
injection, and cumulative gas production for the HC WAG
process sector model matched earlier full-field model HC
WAG simulation forecasts for a premised full-field project
development having a total injection rate of 350 MMSCF/D.
All the other EOR process sector model forecasts were then
upscaled using the identical scaling factors such that key
process-related differences in the magnitude and timing of
production responses were preserved and would be properly
reflected in any subsequent comparative analyses. Obviously
such a scaling procedure is not rigorous, and the resulting
incremental EOR production forecasts are hypothetical and
intended for comparative ranking purposes only.
The final comparative process performance forecasts for
a hypothetical full-field implementation of each of the five
EOR processes are shown in Figure 3. Keep in mind that the
character of the upscaled, full-field response retains the
profile of the underlying mechanistic sector model and
would generally not be expected to be the same as if a
comprehensive full-field modeling study had been
conducted for each individual EOR process. The most
notable difference is that the EOR response is much sharper
in the sector model than would be expected from a full field
model, even though the final incremental recoveries should
be very close. The scaled-up forecasts provided realistic
full-field volumes and relative injection/production rates to
allow gas supply and facilities needs to be assessed for the
various EOR alternatives.
The EOR performance forecasts shown in Figure 3 all
premise the same hypothetical, full-field EOR development
plan. These forecasts honor both the relative process
performance characteristics from the mechanistic sector
modeling and existing field facilities constraints. WAG
cycle sizes and operating schedules were adjusted slightly as
needed to keep upscaled field production rates within
existing facilities capacity limits. Also there was no

SPE 65124

optimization performed for any individual EOR process.

Each EOR process is premised to be implemented beginning
in 2004. EOR injection continues for 13 years, with
blowdown of the reservoir beginning in 2017. The
hypothetical full-field WAG development plan premises 30
current water injection wells being converted to EOR
(WAG) service. A gas injection capacity of 350 MMSCF/D
is premised for all WAG processes based on facility
concerns.
Air injection is premised as continuous gas injection into
30 injection wells at an injection capacity of 650
MMSCF/D. MEOR premises treatment of 30 injection wells
with nominal capacity of 800,000 BWIPD.
S c re e n in g E O R : In c re m e n ta l C u m u la tiv e O il E q u iv a le n ts
600
HC W AG
500
Increm ental Cum ulative BO E , M M BO E

CO2 W A G
N2 W A G

400

A ir Injec tion
M E OR

300

200

100

-100
2004

2006

2008

2010

2012

2014

2016
YEAR

2018

2020

2022

2024

2026

2028

Figure 3. Comparative full-field cumulative incremental oil

equivalent forecasts for the five EOR processes.

COMPARATIVE EOR PROCESS EVALUATIONS

The key technical and logistical issues associated with
each EOR process are addressed in the individual process
discussions which follow. Each discussion includes: (1) a
summary of major process mechanisms and the sources and
quality of data available to characterize the process, (2)
examination of key premises and issues regarding
implementation of the EOR process at Ekofisk, and (3) a
discussion of major cost/benefit elements associated with
each process. If not resolved, some of the issues raised may
be significant enough to eliminate an EOR process from
further consideration.
HC WAG
Summary of Major Process Mechanisms, Data Sources,
and Data Quality
The initial displacement of water flood residual oil by
injected HC gas takes place as a viscous displacement of the
oil by the injected gas. The residual oil saturation following
WAG displacement is lower than the residual oil saturation
to water flooding. This viscous displacement mechanism is
volumetrically relatively minor over the project life, and
progressive vaporization of intermediate oil components by
injection gas becomes the dominant oil recovery mechanism
as cumulative injected gas volume increases.8 Laboratory
experiments have shown that HC gas efficiently vaporizes
intermediate Ekofisk oil components in the C5 to C13 range
under field operating conditions.

SPE 65124

EOR SCREENING FOR EKOFISK

Specific laboratory PVT data were available to describe

the phase behavior and fluid properties of mixtures of
prospective HC injection gas and Ekofisk crude. These data
were used to tune the EOS in the compositional simulation
model. Gas injection core flood data from imbibed core
vaporization experiments were available to validate the
displacement mechanisms of HC gas displacing water flood
residual oil in chalk matrix.

Major Cost Elements

Some of the additional cost elements for the HC WAG
process in large-scale application at Ekofisk are:
Potential Incremental Costs
a) Acquiring HC gas for injection (product costs &
pipelines),
b) Upgrading of existing compression facilities,
c) Additional CO2 tax for added compression/operations,
d) Upgrading of injection system and wells for high
pressure gas injection,
e) Incremental operating costs (including costs to mitigate
gas hydrate problem), and
f) Opportunity costs produced gas capacity constraints.
Potential Benefits
a) Reduced costs for gas lift. Injection of HC gas, with the
resulting decrease water injection relative to the base
forecast will cause reduced water production over time.
Figure 4 shows a comparison of water production at
Ekofisk with and without HC WAG. A significant
reduction in produced water with HC WAG injection is
observed which will result in the need for less gas-lift
gas. This in turn reduces operating costs and will allow
for increased gas sales in periods when gas processing
is plateaued at facility capacity.
b) Reduced future costs/penalties associated with produced
water handling. The decreased amount of produced
water as a result of gas injection will reduce costs
associated with cleaning and disposing and/or
reinjection of produced water under a National Zero
Discharge Order as expected to be in effect from 2005.
c) Increased capacity for oil. Currently, Ekofisk is total

liquid capacity limited. If this capacity constraint

continues, we should see an increased capacity for oil
with a reduction of produced water. This will allow
further acceleration of oil recovery as long as the base
forecast is total liquid constrained.
An n u a l W a te r P ro d u ctio n R ate
2 0 0 0 00

W ater Pro d u c tio n R a te, B W /D

Major Process Implementation Issues and Premises

Successful implementation of any HC WAG project requires
a source of HC gas for injection. The project performance
forecasts premise that a supply of HC gas will be available
to be taken from the gas sales line at Ekofisk. Note that this
EOR process involves the injection of HC gas and thus
deferment of net HC gas sales.
A major technical issue remaining for implementation of
HC WAG is resolution of injectivity problems in cooled
regions around Ekofisk water injectors where hydrates are
expected to form based on thermodynamic considerations.
This study premises that an effective solution or work
around will be found for the gas hydrate problem.
The performance forecast does not include any impact of
or on chalk mechanical compaction due to the HC gas WAG
process. This premise should be evaluated further if this
process is selected to be progressed towards a field pilot.

1 8 0 0 00

W a te r in je c tio n c as e

1 6 0 0 00

W A G c as e

1 4 0 0 00
1 2 0 0 00
1 0 0 0 00
80000
60000
40000
20000
0
1996

2000

2004

2008

2012

2016

2020

2024

2028

Ye ar

Figure 4. Comparison of Water Production for Water Injection

Case Versus HC WAG Injection Case.

N2 / FLUE GAS WAG

Summary of Major Process Mechanisms, Data Sources,
and Data Quality
The displacement of water flood residual oil by injected N2
is much the same as that described for HC WAG, although
N2 (or flue gas) is less efficient than HC gas in vaporizing
intermediate Ekofisk oil components. Laboratory
experiments have shown that N2 is able to efficiently
vaporize intermediate Ekofisk oil components only up to
about C8 under field operating conditions. N2 injection is
volumetrically more efficient (in terms of voidage
replacement) than other injection gases being considered for
EOR application at Ekofisk. Furthermore, no injectivity
problems (hydrates) are expected with N2 gas injection
based on thermodynamic considerations.
Specific laboratory PVT data were available to describe
the phase behavior and fluid properties of mixtures of N2
with Ekofisk crude. PVT results and mechanistic simulation
modeling of N2 injection as a secondary recovery IOR
injectant7 was available to guide the work in this study.
These data were used to tune the equation of state (EOS) in
the compositional simulation model. Injected N2 interacts
with and dissolves in Ekofisk oil, replacing light HCs and
forming a methane-rich gas front at the leading edge of the
displacement. This results in significant shrinkage of the
remaining oil volume and reduction in the oil formation
volume factor. The remaining N2-contacted oil is less mobile
and more difficult to displace due to higher oil viscosity and
higher interfacial tension.
Compositional simulations with flue gas (N2/CO2
mixture) indicated that adding up to 20% CO2 provided only
marginal improvement in oil recovery performance.
Major Process Implementation Issues and Premises
N2 supply would be obtained using a cryogenic (or other) air
separation process. Size and weight requirements for air
separation facilitates and associated compression require an

T.B. JENSEN, K.J. HARPOLE, A. STHUS

additional process platform with related pipeline tie-ins, etc.

Injection wells would need to be upgraded for high-pressure
WAG injection service. Significant additional facilities
would also be required on the production side to handle
rejection of N2 contaminated produced gas and recovery of
associated NGLs.
The N2 WAG injection performance forecasts do not
include any impact of or on chalk mechanical compaction
due to a N2/flue gas WAG process. This premise should be
evaluated further if this process is selected to be progressed
towards a field pilot. Flue gas, with its higher CO2 content,
would be of most concern in this regard.
Major Cost Elements
Some additional cost elements for the N2 WAG process in
large-scale application at Ekofisk are:
Potential Incremental Costs
a) Additional platform with air separation, compression,
and rejection facilities,
b) Additional CO2 tax for added compression/operations,
c) Pipelines to/from N2 facilities,
d) Upgrading of injection system and wells for high
pressure gas injection, and
e) Incremental operating costs for WAG
injection/production.
Potential Benefits
a) Reduced costs for gas lift,
b) Reduced future costs/penalties associated with handling
of produced water,
c) Increased capacity for oil during liquid constrained
periods,
d) Acceleration of HC gas recovery, and
e) No gas hydrate problems expected.
CO2 WAG
Summary of Major Process Mechanisms, Data Sources,
and Data Quality
Minimum miscibility pressure correlations and phase
behavior predictions indicate that CO2 will likely be multicontact miscible or near-miscible with Ekofisk crude over
significant portions of the reservoir. Miscible displacement
conditions are more likely in the low temperature regions
surrounding existing water injectors. An initial, minor
viscous displacement period will be followed by progressive
vaporization of intermediate oil components with continued
gas injection. CO2 should be significantly more efficient
than HC gas in vaporizing intermediate Ekofisk oil
components. While no specific CO2-oil PVT data has been
measured for Ekofisk, significant extraction of oil
components from C5 up to C20 could be expected based on
industry experience. This behavior is supported by
compositional simulation forecasts using an untuned
equation of state.
The basic oil recovery mechanisms by which CO2 is able
to mobilize water flood residual oil include swelling of the
residual oil volume and reduction in oil viscosity as CO2
dissolves in the oil, strong vaporization and extraction of
intermediate components in the oil into the CO2-rich gas

SPE 65124

phase, and reduction in the interfacial tension forces

allowing mobilization of the water flood residual oil. Both
increasing temperature and the presence of contaminants
(particularly N2) in the CO2 injection gas will have a
negative impact on oil recovery efficiency.
Major Process Implementation Issues and Premises
Implementation of any CO2 WAG project requires a CO2
supply source for injection. The project performance
forecasts premise that the required delivery rate of CO2 will
be available for the project. However, it is important to note
that no supply source of CO2 for any Ekofisk project has
been identified at this time. A project evaluation at the
Grane Field9 premises that an onshore CO2 supply of
comparable delivery capacity will come from a HC gas fired
power plant. A similar onshore CO2 supply scenario for
Ekofisk may be the most likely possibility, however the
costs to deliver CO2 from these types of sources could be
quite high. The only natural CO2 supply source identified in
the North Sea is from the Sleipner Gas Field,10 however it
appears that only about 60-70 MMSCF/D would be
available (which is currently reinjected at Sleipner). A CO2
supply of the requisite purity, volume, and deliverablity
must be identified for Ekofisk if this process is selected as
one to be progressed toward a field pilot.
Implementation of a CO2 project at Ekofisk would
require a supply pipeline to deliver CO2 from the source to
the field. CO2 is generally transported as a supercritical
fluid, typically at pressures of 1500-2000 psi as it can be
pumped much like a liquid at these conditions. Booster
compression would be required at the field to bring the CO2
to the required injection pressure. Rejection of produced
CO2 from produced fluids will be a major issue and will
require extensive CO2 separation, compression, and NGL
recovery facilities. The size and weight requirements for the
premised compression and recycle facilities will require an
additional CO2 facilities platform with related pipeline tieins, etc. Injection wells will need to be upgraded with
special wellheads for high-pressure CO2 WAG injection
service. Additional expense for increased corrosion treating
will also be required. Problems with injectivity due to CO2hydrate formation in cooled regions (below about 60F)
around existing water injectors may be anticipated based on
thermodynamic calculations.
The CO2 WAG injection performance forecasts do not
include any impact of or on chalk mechanical compaction.
Laboratory testing conducted to date indicates that
introducing CO2charged injection water into Ekofisk chalk
samples results in an immediate and vigorous dissolution
reaction with large axial strains and high strain rates.
The additional compaction would provide additional
drive energy in the system, the high local strain rates may
result in well failures and in ultimate sea bed subsidence
beyond acceptable limits. The impact of CO2 on compaction
must be evaluated extensively if this process is selected to
be progressed further. Also, reaction of CO2-rich injection
water with calcium carbonate in the rock matrix will
consume some (potentially significant) fraction of the total
injected CO2 volume, thus making it unavailable to
participate in the oil recovery process. The impact of this
reduced effective CO2 slug size has not been quantified or

SPE 65124

EOR SCREENING FOR EKOFISK

included in the screening forecasts and would need to be

evaluated in any future work on application of this process.
Major Cost Elements
Some additional cost elements for the CO2 WAG process in
large-scale application at Ekofisk are:
Potential Incremental Costs
a) CO2 supply gas acquisition costs, CO2 supply pipeline
and transportation costs,
b) Additional platform with CO2 compression and
rejection/recycle facilities,
c) Additional CO2 tax for added compression/operations,
d) Pipelines to/from CO2 facilities,
e) Upgrading of injection system and wells/wellheads for
high pressure CO2 service,
f) Incremental operating costs (incl. corrosion) for CO2
WAG injection/production, and
g) Possible increased well failures & subsidence
mitigation costs.
Potential Benefits
a) Reduced costs for gas lift,
b) Reduced future costs/penalties associated with handling
of produced water,
c) Increased capacity for oil,
d) Acceleration of HC gas recovery, and
e) Possible increased reserves due to enhanced compaction
drive.
AIR INJECTION
Summary of Major Process Mechanisms, Data Sources,
and Data Quality
The air injection process involves continuous injection of air
resulting in an exothermic reaction of oxygen with residual
oil and generation of a thermal reaction front that propagates
through the reservoir. Laboratory accelerating rate
calorimetry (ARC) tests conducted with Ekofisk crude
showed favorable reaction profiles indicating that a thermal
reaction front could be generated and maintained in the
reservoir. It is likely that an enriched air (30% O2) process
could be used to improve the process performance
characteristics at Ekofisk and to provide spontaneous
ignition around cooled water injectors. Specific laboratory
combustion tube (CT) tests have not been conducted with
Ekofisk chalk core material to date, however a significant
amount of ARC and CT data are available for similar
systems. These industry laboratory data, along with oil
recovery and air utilization data from industry field
projects,11-12 formed the database for the air injection
performance forecasts for Ekofisk.
Major Process Implementation Issues and Premises
Air compression/enrichment facilities capable of injecting
up to 650 MMCFPD were premised for the field-scale
project performance forecasts. Additional facilities would be
required on the production side to handle rejection of
produced gas contaminated with N2/CO2/CO/etc reaction
products and for recovery of a significant volume of
associated NGLs in the produced gas stream. The size and

weight requirements for the premised air separation

facilities, produced gas processing facilities, and associated
compression would likely require an additional facilities
platform with related pipeline tie-ins, etc.
Two different scenarios were considered for the air
injection process. Scenario 1 involves no separation of
salable gas from flue gas. All produced gas will be
reinjected in this case. There will be a significant capital
cost saving in this case as the gas separation/rejection unit
will not be needed, but there will also be a significant loss of
revenue from loss of the salable gas. Scenario 2, which is
perceived as the more realistic, involves complete separation
of salable gas from flue gas with export and sales of sales
gas and reinjection of flue gas. Evaluations showed that
separation of sales gas from flue gas (i.e. reinjection of only
flue gas) is the preferable option.
Injection wells and lines would need to be upgraded for
high-pressure air injection service. Injection and production
wells would probably need to be upgraded to handle
exposure to the elevated temperatures. Additional expense
for increased corrosion treating should also be anticipated.
The air injection performance forecasts do not include
any impact of or on chalk mechanical compaction due to the
thermal effects of the high-temperature air injection process
and the effects of combustion gases on the chalk matrix.
Possible high-temperature calcination reactions may
substantially alter and weaken the rock. Preliminary data
from one combustion tube test on an outcrop chalk showed
that the rock matrix may be weakened by 25% through
exposure to this high-temperature process. The CO2 in the
combustion products generated by the thermal reactions
would likely have a significant impact on the mechanical
integrity of the matrix also, as discussed for the CO2 WAG
process. The impact of this process on mechanical integrity
of the Ekofisk chalk and resulting compaction/subsidence
effects must be evaluated more thoroughly if this process is
selected to be progressed further.
The air injection process might be considered an
alternative to blowdown as the final stage in the exploitation
process. Air injection can potentially be implemented
following any of the other processes.
Major Cost Elements
Some additional cost elements for the air injection process
in large-scale application at Ekofisk are:
Potential Incremental Costs
a) Additional platform with air compression and
enrichment facilities,
b) Additional CO2 tax for added compression/operations,
c) Upgrading of injection system and wells for high
pressure air injection,
d) Upgrading of injection & production wells for hightemperature service,
e) Incremental operating costs for high volume gas
handling and corrosion treating, and
f) Possible increased well failures & subsidence
mitigation costs.

T.B. JENSEN, K.J. HARPOLE, A. STHUS

Potential Benefits
a) Reduced costs for gas lift,
b) Reduced future costs/penalties associated with handling
of produced water,
c) Increased capacity for oil during liquid constrained
periods,
d) Acceleration of HC gas recovery, and
e) Possible increased reserves due to enhanced compaction
drive.
MICROBIAL EOR
Summary of Major Process Mechanisms, Data Sources,
and Data Quality
Microbial enhanced oil recovery (MEOR) is still an
emerging technology and the exact mechanisms by which
microbial agents are able to effect mobilization and recovery
of incremental oil in a given reservoir are not clearly
understood. MEOR processes likely act through a complex
of different mechanisms, including wettability alteration,
IFT reduction, oil viscosity reduction, and others. As a
consequence, there are no robust predictive models available
to effectively simulate MEOR process applications and oil
recovery potential at Ekofisk. Moreover, at this time, there
are no specific laboratory data to support or refute claims of
oil recovery potential by MEOR in the Ekofisk reservoir.
Estimates of recovery potential in this paper are made by
analogy under the premise that a viable MEOR system can
be developed for application at Ekofisk which will have
performance characteristics analogous to other field results
reported to date.
Major Process Implementation Issues and Premises
The microbial oil recovery forecasts were generated by upscaling a composite of typical field project results. It was
necessary to use this empirical approach to estimate MEOR
performance as the exact mechanisms of microbial oil
recovery are still poorly understood and consequently there
are no simulation models available to make reliable MEOR
field performance forecasts. The nature of microbial
transport in a fractured chalk reservoir has not been
evaluated. Reduced reservoir temperatures near the water
injection wells would probably be favorable for microbial
growth. Issues of the viability of microbes over the wide
range of temperatures existing in Ekofisk, and particularly
the impact of higher reservoir temperatures on the MEOR
process, needs to be investigated further.
Several of the possible mechanisms suggested for oil
mobilization in MEOR involve wettability alteration and/or
IFT reduction. These are mechanisms similar to those of
surfactant flooding, which has been considered previously
but is not being actively pursued as an EOR method at
Ekofisk because of low oil recovery forecasts. This raises
the question of how and why the MEOR process is different
from surfactant processes considered earlier and therefore
merits continued interest?
The MEOR process is not just a surfactant process the
mechanisms suggested as being important in MEOR are
varied and complex (e.g. oil viscosity reduction, gas
production, solvent production) and are not fully
understood. Another key difference between MEOR and

SPE 65124

surfactant flooding lies in the envisioned method of

surfactant delivery to the point of contact with residual oil.
Microbes would tend to move preferentially through the
fracture system, being excluded from directly entering the
chalk matrix because of their size relative to the pore throat
size distribution of the chalk. Microbes have the ability to
seek out and adhere to any films of residual oil at the
fracture/matrix interface or residing in larger pores adjacent
to the fracture system. Thus, microbes may penetrate fairly
deep into a fractured chalk reservoir and generate surfactant
in situ, thereby avoiding much of the potential adsorption
losses expected with surfactant flooding. Since the microbes
would use residual oil as a food source, total injection
volumes and costs would be substantially reduced.
The MEOR performance forecasts represent an
extension of limited laboratory and field results to a fullscale MEOR project at Ekofisk. The oil recovery projections
are based on the following premises: (1) a viable microbial
system will be identified for application at Ekofisk; and (2)
the oil recovery potential of that process will be comparable
to that of successful MEOR processes seen in other
laboratory and field studies (with adjustments for reservoir
conditions at Ekofisk). Potential regulatory and
environmental issues/concerns with injecting microbes into
a reservoir in the North Sea have not been addressed. Also,
the potential for reservoir souring must be closely evaluated
before implementing injection of any microbes into Ekofisk.
The material and equipment requirements and logistics
of implementing a full-field microbial oil recovery project in
an offshore North Sea operating environment do not appear
to be a limiting factor. Equipment requirements are modest,
operating costs are fairly low, and there would appear to be
only minimal impact on existing operations. The real issue
with this process is the viability of the oil recovery process
itself under Ekofisk reservoir conditions. Additional proof
of concept laboratory work is required to establish the
viability of any claims for incremental oil recovery
potential. MEOR is the most uncertain of any of the EOR
processes being considered for application at Ekofisk.
Major Cost Elements
Some additional cost elements for the MEOR process in
large-scale application at Ekofisk:
Potential Incremental Costs
a) Microbial culture generation tanks (small, inexpensive)
b) Small, low-volume chemical injection pumps,
c) Laboratory facilities for quality control and process
monitoring, and
d) Possible environmental/regulatory costs or concerns.
Potential Benefits
a) Relatively easy and inexpensive process to implement
and operate,
b) Compatible with existing water injection facilities and
operations, and
c) Could possibly be used in combination with a gas
injection EOR process.

SPE 65124

EOR SCREENING FOR EKOFISK

EOR DEVELOPMENT & OPERATIONAL PREMISES

Injection Gas Availability
It is assumed the gas can be made available in sufficient
quantities for the gas injection processes. For air injection
this is a compressor sizing issue. For N2 injection, the gas
availability also premises separation (cryogenic or other) of
N2 from air.
HC gas availability on the Norwegian Continental Shelf
(NCS) is as always quite uncertain and variable. The
availability depends on off-take to the European market,
variations in price, and development of new gas reserves.
Gas for HC WAG will be taken from the sales gas stream at
Ekofisk. HC gas is available in sufficient quantities at
Ekofisk, but substitute gas for sales commitments may have
to be acquired through a different source and supplier. The
only effect accounted for with respect to HC gas injection
was the economic effect of deferring the gas sales as a result
of gas injection.
On the CO2 availability issue, there is not much known.
The only concept studied is the Grane field9 where CO2 was
premised to be separated from flue gas at an onshore gas
power plant before being piped offshore for injection into
the reservoir. The Grane operator has not been able to find
viability in such a scenario and the concept has been
shelved. The only known significant natural source of CO2
gas on the NCS is at the Sleipner field.10 However, the
volumes/rates at Sleipner are only one sixth of that required
for full field CO2 WAG at Ekofisk.
For CO2 injection, we assume that the CO2 can be made
available from a gas power plant onshore Norway and that a
pipeline has to be built to transport the gas to Ekofisk with
associated compression facilities. The price estimate for CO2
was taken from a SINTEF study.13 Price for CO2 was
estimated to range from $1.13/MCF to $1.69/MCF. Here,
we premise that the CO2 gas is available at an average price
of $1.5/MCF at the wellhead.
Injection and Well premises
Gas injection will be at the main water injection facility into
30 wells. The WAG scenarios (HC, N2 and CO2) premise 30
injection wells divided into 3 well groups. Gas is premised
to be injected in 6 month WAG cycles into 10 injection
wells at the time at a nominal injection rate of 350
MMSCF/D at 6000 psia wellhead pressure. The remaining
20 injection wells will be on water injection. For air
injection, we premise that all 30 injection wells will receive
continuous air injection at a nominal rate of 650 MMSCF/D
at a wellhead pressure of 6000 psia.
For HC and N2 WAG the injection well upgrade premise
redrill of 20 injection wells and tubing replacement in 10
injection wells. For CO2 WAG there will be a need to
sidetrack all 30 injectors and complete with chrome tubing.
For air injection, 30 injection wells and 30 production wells
need to be upgraded (sidetracked and chrome tubing
completed). Chrome tubing for the CO2 and air injection
scenarios is required as a corrosion prevention measure.
For MEOR, there will be no change in the injection
pattern or volumes. No wells will require workover beyond
that which is required in the base case. Microbe solution is
premised to be commingled with the current injection water

topside. No additional cost for sour service in the event of

reservoir souring was accounted for.
Hydrate Prevention
HC gas/water hydrate was thought to cause the failure of the
first WAG pilot performed at Ekofisk. This is believed to be
a general problem with any future HC WAG scenario.
Prolonged injection of cold sea-water results in reservoir
temperatures and pressures such that gas hydrates form if
HC gas is subsequently injected in the same well.
Several possible alternatives for hydrate prevention have
been discussed, including:
1.
2.
3.
4.
5.

Chemical heating of the near wellbore area prior to gas

injection,
Heating of the injection water to raise near wellbore
temperature outside of the hydrate envelope,
Sidetrack injectors to outside of the cooled zone prior to
gas injection,
Co-injection of water and gas, and
Chemical inhibition to prevent and/or retard the onset of
hydrate formation.

The chemical heating option was selected for field

testing in 1998. The chemicals showed great promise, but
the test failed as the reservoir pressure was too high to be
overcome with the current top-side wellhead and pipeline
pressure limitations.
For screening purposes, we assumed that a combination
of chemical heating and heating of the injection water would
suffice to prevent hydrates in a full field WAG
implementation as follows:
A. Heating 60,000 bbl/d to 85-90C and mix with cold sea
water to achieve proper temperature level
Other heating options, e.g. incorporating waste heat
recovery, is an upside.
B) Chemical treatment (exothermic reactions).
This elevates the near wellbore temperature, for a
designed treatment depth, to a level above the
hydrate formation temperature.
The chemical treatment is once per well only for the
first cycle of water injection.
CONCLUSIONS
1. Incremental oil recovery forecasts for the five processes
at Ekofisk were as follows:
HC WAG:
3.3% OOIP
N2 WAG:
-2.2% OOIP
CO2 WAG:
5.6% OOIP
Air Injection:
6.5% OOIP
MEOR:
0.6% OOIP
2. Only Hydrocarbon WAG and Air Injection show
sufficient promise for application at Ekofisk to be
carried forward into further studies. Key conclusions
and recommendations regarding each process are as
follows:
N2 WAG injection:
Negative reserves potential.
Eliminated from further consideration.

10

T.B. JENSEN, K.J. HARPOLE, A. STHUS

CO2 WAG injection:

Large reserves potential.
Significant dissolution/compaction concern.
No CO2 source is available in the foreseeable
future.
Dropped from consideration until CO2 can be
secured at cost significantly less than $1.50/
MCF at the wellhead.
Microbial EOR:
Limited reserves potential.
The MEOR process lacks definitive proof of
technical ability to economically mobilize
waterflood residual oil in Ekofisk or similar
chalk.
Dropped from further consideration.
Air Injection
Large reserves potential.
Significant risk and uncertainty.
If air injection is to be employed at Ekofisk it
will most likely be late in the field life,
potentially after some other EOR process has
been employed. Hence, further studies of air
injection are not time-critical but should be
progressed in a timely manner in order to
understand the risk areas and show stoppers
associated with this process.
Recommended further study of risk areas (O2
breakthrough, compaction, and flue gas
separation & reinjection).
HC WAG injection
Significant reserves potential.
Most mature and technically ready process.
Recommended updating of full field WAG
development forecasts with progress towards
pilot feasibility if full field economics are
sufficient.

ACKNOWLEDGEMENTS
The authors acknowledge permission to publish the above
paper from Phillips Petroleum Company Norway and the
Ekofisk Coventurers, including Fina Exploration Norway
s.c.a., Norsk Agip A/S, Elf Petroleum Norge A/S, Norsk
Hydro Production as, TOTAL Norge A.S. and Den norsk
stats oljeselskap a.s.
REFERENCES
1. Hermansen, H., Thomas, L.K., Sylte, J.E. and Aasbe,
B.T., Twenty Five Years of Ekofisk Reservoir
Management, Paper SPE 38927 presented at the 1997

SPE 65124

SPE ATCE, San Antonio, Texas, 58 Oct. 1997.

2. Jakobsson, N. M. and Christian, T. M., "Historical
Performance of Gas Injection of Ekofisk," Paper SPE
28933 presented at the 1994 SPE ATCE, New Orleans,
Sept. 25-28, 1994.
3. Thomas, L.K., Dixon, T.N. and Pierson, R.G., "Fractured
Reservoir Simulation," SPEJ (Feb. 1983) 42-54.
4. Hallenbeck, L.D., Sylte, J.E., Ebbs, and D.J., Thomas,
L.K., "Implementation of the Ekofisk Waterflood," Paper
SPE 19838 presented at the 1989 SPE Annual Technical
Conference, San Antonio, October 8-11, 1989.
5. Christian, T.M., Currie, J.C., Lantz, T.G., Rismyhr, O.,
and Snow, S.E., "Reservoir Management at Ekofisk
Field," SPE 26623 presented at the 1993 SPE Annual
Technical Conference, Houston, October 3-6,1993.
6. Thomas, L. K., Dixon, T. N., Pierson, R. G. and
Hermansen, H., "Ekofisk Nitrogen Injection," Paper SPE
19839 presented at the 1989 SPE ATCE, San Antonio,
Oct. 8-11, 1989.
7. Coats, K., Thomas, L.K., Dixon, T.N. and Pierson, R.G.,
Compositional and Black Oil Reservoir Simulation,"
Paper SPE 29111 presented at the 13th SPE Symp. on
Res. Sim., San Antonio, TX, USA, 12-15 Feb., 1995.
8. Wegener, D.C., Thomas, L.K. and Sylte, J.E.,
Effectiveness of a Dry Gas Injection WAG Process in a
Previously Waterflooded Chalk Reservoir, Paper SPE
48950 presented at the 1998 SPE ATCE, New Orleans,
Louisiana, 27-30 September 1998.
9. Skotner, P., Evaluation of CO2 Flooding on the Grane
Field, presented at the 7th Conf. On Reservoir Mgmt
The New Petroleum Age, Bergen, Norway, Nov. 18-19,
1998.
10. Baklid, A., Kobol, R. and Owren, G., Sleipner Vest
CO2 Disposal, CO2 Injection into a Shallow
Underground Aquifer, Paper SPE 36600 presented at
the 1996 SPE ATCE, Denver, CO, USA, 6-9 Oct. 1996.
11. Kumar, V.K. and Fassihi, M.R., SPE, Case History and
Appraisal of the Medicine Pole Hills Unit Air Injection
Project, Paper SPE 27792 presented at the SPE/DOE
9th Symp. on IOR, Tulsa, OK, USA, 17-20 April 1994.
12. Watts, B.C., Hall, T.F. and Petri, D.J., The Horse Creek
Air Injection Project: An Overview, Paper SPE 38359
presented at the Rocky Mountain Regional Meeting,
Casper, WY, USA, 18-21 May, 1997.
13. Holt, T. and Lindeberg, E., Thermal Power - Without
Greenhouse Gases and With Improved Oil Recovery,
Energy Convers. Mgmt., Vol. 33, No. 5-8, pp. 595-602,
1992.

Table 1. Technical Readiness of EOR implementation at Ekofisk.