SPE-200450-MS

CO2 Foam Field Pilot Design and Initial Results

Downloaded from http://onepetro.org/SPEIOR/proceedings-pdf/20IOR/1-20IOR/D011S007R003/2369009/spe-200450-ms.pdf by University of Bergen user on 11 January 2022

Zachary Paul Alcorn, University of Bergen; Tore Føyen, University of Bergen/SINTEF Industry; Leilei Zhang, Rice
University; Metin Karakas, University of Bergen; Sibani Lisa Biswal and George Hirasaki, Rice University; Arne
Graue, University of Bergen

Copyright 2020, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Improved Oil Recovery Conference originally scheduled to be held in Tulsa, OK, USA, 18 – 22 April 2020. Due
to COVID-19 the physical event was postponed until 31 August – 4 September 2020 and was changed to a virtual event. The official proceedings were published
online on 30 August 2020.

This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents
of the paper have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect
any position of the Society of Petroleum Engineers, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written
consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may
not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.

Abstract
This paper presents the field design, monitoring program, and initial results from a CO2 foam pilot in East
Seminole Field, Permian Basin, USA. Tertiary miscible CO2 injection has suffered from poor areal sweep
efficiency due to reservoir heterogeneity and an unfavorable mobility ratio between CO2 and reservoir
fluids. A surfactant-stabilized foam was selected to reduce CO2 mobility for increasing oil recovery and CO2
storage potential in an inverted 40-acre five spot well pattern. The foam system was designed to maximize
the success of foam generation through surfactant screening and optimizing surfactant concentration and
foam strength. Previous work identified a water-soluble, non-ionic surfactant at a concentration of 0.5
weight percent (wt%) and 70% foam quality for the pilot. A surfactant-alternating-gas (SAG) injection
strategy, consisting of 10 days of surfactant solution injection followed by 20 days of CO2, began in May
2019.
Baseline CO2 injection profiles, tracer tests, injection bottom hole pressures, and flow rates were collected
for comparison to pilot surveys. The pilot monitoring program included repeat injection profiles, tracer
tests, three-phase production monitoring, and collection of downhole pressure data for evaluation of
reservoir response to foam injection. Produced fluids were also collected for chemical analysis to determine
surfactant breakthrough time. A field injection unit was designed to meet the requirements of surfactant
delivery, mixing, and storing. A methodology was also established to effectively validate foam formulation
consistency in the field.
Initial results revealed that pilot CO2 injectivity was reduced by 70%, compared to baseline CO2
injection, indicating reduced CO2 mobility after each surfactant slug. Baseline and pilot injection profiles
show increased flow into the reservoir interval and potential blockage of a high permeability streak. The
baseline CO2 tracer test measured CO2 breakthrough in 22 days, in one of the pattern producers. Expected
breakthrough, based upon simulation, is 66 days during the pilot, which will be verified by a repeat tracer
test at the end of the pilot. Production response is not expected for another six to nine months due to the
volumes injected during the pilot. However, the early signs of sustained oil production despite less volume
injected during the pilot indicate an initial positive response to foam.
2 SPE-200450-MS

Introduction
A CO2 foam field pilot was implemented in East Seminole Field in the Permian Basin of west Texas. Tertiary
miscible CO2 enhanced oil recovery (EOR) performance suffered due to reservoir heterogeneity and the
unfavorable mobility ratio between CO2 and in-situ fluids, resulting in poor areal sweep efficiency and low
oil recoveries. A surfactant-stabilized foam was selected to mitigate CO2 injection challenges by reducing
CO2 mobility, in an effort to improve sweep efficiency for increased oil recovery and CO2 storage potential.
Combined CO2 EOR and CO2 storage has recently gained attention due to growing concerns regarding

Downloaded from http://onepetro.org/SPEIOR/proceedings-pdf/20IOR/1-20IOR/D011S007R003/2369009/spe-200450-ms.pdf by University of Bergen user on 11 January 2022

anthropogenic CO2 emissions and their impact on climate change. Utilizing CO2 to produce additional
energy while simultaneously storing CO2 can reduce emissions as part of carbon capture, utilization, and
storage (CCUS). The increased revenues from additional oil production incentivizes industry participation,
which is required for gigaton volume CO2 storage, and can offset the large costs of carbon capture and
storage (CCS) processes. This work is part of an international multidisciplinary collaboration between
academia and industry to advance the use of mobility control foams to mitigate CO2 injection challenges
during combined CO2 EOR and CO2 storage.
Foam injection is a laboratory verified and field-tested technique to reduce CO2 mobility for improved
conformance and in-depth mobility control (Heller 1966; Heller et al., 1985; Hoefner and Evans, 1995;
Shramm 1994; Enick, 2012). Foam is a dispersion of gas in liquid where the liquid is interconnected
and gas flow is impeded by stable liquid films called lamellae (Falls et al. 1988; Rossen, 1996). In EOR
applications, water-soluble surface-active agents (surfactants) which decrease interfacial tension are often
used to stabilize the lamellae. Surfactants must be thoroughly screened to ensure the success of a CO2
foam process. Foam can be generated in the reservoir by simultaneous injection of CO2 and surfactant
solution (coinjection) or in alternating slugs of CO2 and surfactant solution (SAG). Implementation of
foam at the field-scale must balance injectivity, gas-mobility reduction, and operational constraints. A
number of previous CO2 foam field tests have reported success by reducing gas mobility and improving
oil recovery, whereas others report injectivity issues and difficulty in attributing additional production
specifically to CO2 foam (Chou et al. 1992; Stephenson et al. 1993; Martin et al. 1992; 1995). Therefore,
a more thorough integration of optimized laboratory foam formulations and field-scale tests can provide
additional knowledge of size-dependent CO2 foam displacement mechanisms during CO2 EOR and CO2
storage. This work is part of an integrated upscaling approach, which characterized foam systems at the
pore- and core-scale and piloted the optimal foam formulation in the Permian Basin (Jian et al. 2016; Alcorn
et al. 2018; Rognmo et al. 2019).
East Seminole Field produces from the San Andres formation, a heterogeneous cyclical carbonate and
one of the most prolific producers in the Permian Basin. An inverted 40-acre five spot was selected to pilot
the foam technology, which includes a central injection well, and four surrounding producers. The well
pattern was selected based upon rapid CO2 breakthrough, high producing gas-oil-ratios (GOR), and close
well proximity. A rapid surfactant-alternating-gas (SAG) injection strategy was initiated in May 2019 to
generate foam in-situ for in-depth mobility control. The injection strategy was designed to limit injectivity
issues and maximize the chance of success for foam generation.
This paper describes the pilot design and initial reservoir response to foam injection. Baseline CO2
injection profiles, CO2 tracer tests, injection bottom hole pressures (BHP), and flow rates were collected for
comparison to pilot surveys. The pilot monitoring program included repeat injection profiles, tracer tests,
three-phase production monitoring, and collection of downhole-injection pressure data. Produced fluids
were also collected for chemical analysis to determine surfactant breakthrough time. A field injection unit
was designed to meet the requirements of surfactant delivery, mixing, and storing. A methodology was
established to effectively validate foam formulation consistency in the field. A simulation study was also
conducted to evaluate interwell connectivity and foam's impact on CO2 breakthrough at the field-scale. This
SPE-200450-MS 3

paper will first review the foam formulation for the pilot and then present the field design and objectives.
Next, we discuss the data collection and pilot monitoring program, surface facility design, foam formulation
validation, and the simulation study. Finally, we present and discuss the initial results from the first foam
injection in East Seminole field.

Foam Formulation
The laboratory program aimed to determine the optimal foam formulation for the field test. This included

Downloaded from http://onepetro.org/SPEIOR/proceedings-pdf/20IOR/1-20IOR/D011S007R003/2369009/spe-200450-ms.pdf by University of Bergen user on 11 January 2022

surfactant-screening studies, evaluation of the optimal foam quality (gas fraction), and quantification of
CO2 EOR and CO2 storage potential of the optimized foam formulation at the core-scale. The individual
components of the laboratory program were recently published elsewhere and are included in this work for
completeness (Jian et al. 2016; Alcorn et al. 2019).
Surfactant screening studies identified the nonionic water-soluble Huntsman L24-22, a linear ethoxylated
alcohol (C12-14 EO22), for the field pilot based upon the loss to the formation due to adsorption, foam
strength, and chemical stability. The loss to the formation was quantified by static adsorption, measured
using high performance liquid chromatography (HPLC) combined with an evaporative light scattering
detector (ELSD). Low adsorption was measured between 0.1 mg/m2 and 0.3 mg/m2 on dolomite and calcite
in deionized water (DI) and field brine (Figure 1a). Adsorption was also measured in the presence of CO2,
which resulted in no observable impact (Figure 1b). The chemical stability was evaluated, concluding that
the nonionic surfactant can degrade at high temperatures (above 60⁰C) under aerobic conditions, i.e. in the
presence of oxygen. The reservoir temperature is 40⁰C; however, heat must be applied when mixing the
surfactant with injection brine in the field. Therefore, an oxygen scavenger was selected to be added to the
mixture. Additionally, a scale inhibitor was identified to be compatible with the surfactant. No detrimental
effect on foam stability was found when the oxygen scavenger and scale inhibitor was used in reservoir
brine with surfactant. The interested reader is referred to Jian et al. 2016 for additional information.

Figure 1—a) Static adsorption of Huntsman L24-22 in deionized (DI) water (open symbols) and field brine
(closed symbols) on calcite (blue symbols) and two different dolomites (orange and purple symbols). b) Static
adsorption of Huntsman L24-22 on dolomite in the presence of CO2 and air (modified from Jian et al. 2016).

Once the field-specific surfactant was selected, the foam formulation was designed by evaluating the
impact of surfactant concentration on foam strength in aged reservoir core material with residual oil present
(Sor). In coreflood experiments, foam strength and stability is evaluated by measuring foam apparent
viscosity at different foam qualities (gas fractions), where a higher apparent viscosity corresponds to a
4 SPE-200450-MS

stronger foam. Figure 2 shows apparent CO2 foam viscosity as a function of gas fraction (fg) for five foam
quality scans in reservoir core material at 40⁰C and 2500 psig (172 bar) at a rate of 1 ft/day. Apparent foam
viscosities of 20 centipoise (cP) and 30 cP were observed at a gas fraction of 0.70 with 0.5 weight percent (wt
%) and 1.0 wt% surfactant concentration, respectively. The slight difference in apparent viscosity between
0.5 wt% and 1.0 wt% surfactant solution did not justify the use of a more concentrated solution for the
pilot. Therefore, a surfactant concentration of 0.5 wt%, at a foam quality of 70%, was recommended for
the field test.

Downloaded from http://onepetro.org/SPEIOR/proceedings-pdf/20IOR/1-20IOR/D011S007R003/2369009/spe-200450-ms.pdf by University of Bergen user on 11 January 2022

Figure 2—Apparent CO2 foam viscosity versus gas fraction (fg) for five foam quality scans in
reservoir core material at 40⁰C and 2500 psig (172 bar) at a rate of 1 ft/day in the presence of
residual oil (Sor). The dashed line shows the recommend gas fraction. From Alcorn et al. 2019.

Field Pilot Design

The main objective of the foam pilot was to achieve in-depth CO2 mobility control to reduce producing
gas-oil ratios (GOR) and improve CO2 sweep efficiency and oil recovery. The pilot area was an inverted 40
acre 5-spot pattern with a central injection well and four surrounding producers (Figure 3). The wells were
selected based upon rapid CO2 breakthrough, high producing GORs, and close proximity. The baseline data
collection and pilot monitoring program focused on production wells P1 and P4 because they experienced
the most rapid CO2 breakthrough time from tertiary CO2 injection. The reservoir interval has an average
permeability of 13 mD, pay thickness of 110 ft, and consists of six flow zones separated by impermeable
flow barriers. Composite logs from the pilot injection well indicated a six-foot thick high permeability
streak of 223 mD and baseline injection profiles showed that this zone has been taking the majority of the
injected CO2. Therefore, this high permeability zone was targeted since foam is capable of forming in high
permeability streaks and diverting flow to unswept regions of the reservoir with lower permeabilities. The
reservoir and fluid properties are shown in Table 1.
SPE-200450-MS 5

Downloaded from http://onepetro.org/SPEIOR/proceedings-pdf/20IOR/1-20IOR/D011S007R003/2369009/spe-200450-ms.pdf by University of Bergen user on 11 January 2022

Figure 3—Pilot pattern (shaded area) and surrounding wells in East Seminole Field.
The pilot injection well was I1 and the monitored producers were P1 through P4.

Table 1—Reservoir and fluid properties of the San Andres unit in East Seminole Field.

Reservoir Characteristic Value

Depth 5200 ft

Permeability 1 to 250 md (average: 13 md)

Porosity 3% to 28% (average: 12%)

Pay thickness 110 ft

Reservoir pressure (initial) 2500 psig

Reservoir pressure (current) 3400 psig

Fracture pressure 3900 psig

Reservoir temperature 104⁰F

Oil gravity 31 ⁰API

Formation brine salinity 70,000 ppm

Prior to the pilot, the reservoir was over pressurized to 3400 psig due to the injection of produced water
in a deeper interval in an offset pattern. Since foam can cause sharp increases in injection BHP and the
formation fracture pressure is 3900 psig, the injection strategy was designed to mitigate injectivity losses. In
addition, the injection strategy was designed to volumetrically target a 70% foam quality, as recommended
from the laboratory studies. A rapid surfactant-alternating-gas (SAG) injection strategy was selected for the
pilot test. Twelve cycles were to be injected, consisting of 10 days surfactant solution followed by 20 days
of CO2 at a target rate of 500 rb/day. This results in a total of 8% pattern hydrocarbon pore volume (HCPV)
injected by the end of the 12 month pilot. The injection well was operated at a constant tubing head pressure
(THP) at the target injection rate.
6 SPE-200450-MS

Data Collection and Monitoring Program

The data collection and pilot monitoring program aimed to obtain baseline in the pilot pattern and
monitor pilot performance to evaluate reservoir response to foam injection. Since foam is expected to
increase the injection pressure by reducing CO2 mobility, the baseline period focused on characterizing
interwell connectivity and injectivity of CO2 and water. The baseline data collection program consisted
of CO2 injection profile logs, an interwell CO2 tracer test (IWTT), collection of injection pressures and
temperatures, and flow rates for comparison to repeat surveys. The injection pressure and temperature was

Downloaded from http://onepetro.org/SPEIOR/proceedings-pdf/20IOR/1-20IOR/D011S007R003/2369009/spe-200450-ms.pdf by University of Bergen user on 11 January 2022

monitored by mounting a downhole pressure gauge (DHPG) in the pilot injection well. The gauge was
installed two weeks before the start of the pilot and pulled every other month during the pilot to retrieve the
pressure and temperature data. Table 2 shows the schedule of surveys and data to be collected. Produced
fluids were also collected, before the pilot and once a week during the pilot, for chemical analysis to
determine surfactant breakthrough time.

Table 2—Schedule of surveys and data collection during the baseline, pilot, and post-pilot periods.
Two SAG cycles are shown for simplicity, but surveys were identical for the subsequent cycles.

Stage Baseline Pilot (2 SAG cycles shown) Post-pilot

Slug Water CO2 Surfactant CO2 Surfactant CO2

IWTT x x*

Injection
x x x x x
Profiles

Produced
water x x x x x
collection

DHPG reading x x x

*denotes repeat CO2 tracer test (IWTT) in the last CO2 slug of the pilot.

The IWTT injected a nonradioactive CO2 tracer into the selected pilot injection well (I1) to determine
CO2 breakthrough time. All peripheral producers were monitored on a 10-month sampling schedule, first
focusing on two of the wells that had the most rapid CO2 breakthrough (P1 and P4). Samples from P1 and
P4 were collected twice a week for the first month with the frequency reduced to once a week, thereafter.
Breakthrough was observed in P1 and P4 in 22 days and 28 days, respectively (Figure 4). The tracer
migration rate was calculated to be 7 ft/day. Tracer breakthrough was not observed until 4 months in P2 and
P3. Foam is expected to increase the CO2 breakthrough time, indicating CO2 mobility reduction by foam
and improved reservoir sweep efficiency. A repeat CO2 IWTT will be conducted upon the conclusion of
the pilot to compare pilot response.
SPE-200450-MS 7

Downloaded from http://onepetro.org/SPEIOR/proceedings-pdf/20IOR/1-20IOR/D011S007R003/2369009/spe-200450-ms.pdf by University of Bergen user on 11 January 2022

Figure 4—Baseline interwell CO2 tracer test (IWTT) results. P1 and P4 correspond to pilot
pattern production wells 1 and 4, respectively. The pilot pattern is shown at right for reference.

Surface Facility Design

A field injection unit was designed for surfactant delivery, mixing, and storing. The surface facilities for
foam injection required minor modifications to the existing equipment at the injection well site. The injection
well site was outfitted with a water-alternating-gas (WAG) skid and standard injection wellhead assembly.
Brine was delivered via a subsurface fiberglass pipe at a tubing head pressure (THP) of 1450 psi and CO2 was
delivered via a subsurface steel pipe at THP of 1900 psi. All pattern production wells were on artificial lift.

Surfactant Delivery and Mixing

Surfactant was delivered to the injection site in a heated tanker truck as a molten 100% product. The
surfactant was mixed in a covered holding tank with pre-heated injection brine to achieve a concentrated
25 wt% solution. The injection brine also contained the laboratory selected scale inhibitor and oxygen
scavenger to prevent surfactant degradation at high temperatures. The surfactant solution in the holding
tank was regularly monitored to ensure that adequate amounts of oxygen scavenger were present to remove
dissolved oxygen. Two static in-line mixers were included in the set-up to ensure homogeneous mixing of
heavily concentrated surfactant and injection brine. The concentrated 25 wt% solution and injection brine
were metered into a final dilution line to achieve a final concentration of 0.5 wt%. Sampling ports were
installed before the injection point and on the holding tank to analyze for concentration accuracy. Figure 5
shows a schematic of the field injection unit.
8 SPE-200450-MS

Downloaded from http://onepetro.org/SPEIOR/proceedings-pdf/20IOR/1-20IOR/D011S007R003/2369009/spe-200450-ms.pdf by University of Bergen user on 11 January 2022

Figure 5—Schematic of the field injection unit for the CO2 foam field pilot.

Foam Formulation Validation

It was important to keep the injected foam formulation consistent at each cycle. The target foam formulation
as determined in the laboratory consisted of a 0.5 wt% surfactant solution, 100 parts per million (ppm)
oxygen scavenger, and 50 ppm scaling inhibitor. A methodology to effectively validate foam formulation
consistency in the field was established.
A refractometer measured surfactant concentration in the holding tank and in the final dilution line.
A correlation was used to convert the refractive index (RI) reading to surfactant concentration in weight
percent (Figure 6). Surfactant solutions with known concentrations were first measured and used to calibrate
the readings. Surfactant samples were collected for measurement several days prior to injection and every
day during the 10-day surfactant slugs. The method proved to be effective in measuring and verifying the
foam formulation in the field.
SPE-200450-MS 9

Downloaded from http://onepetro.org/SPEIOR/proceedings-pdf/20IOR/1-20IOR/D011S007R003/2369009/spe-200450-ms.pdf by University of Bergen user on 11 January 2022

Figure 6—Correlation between refractive index (RI) and normalized surfactant
concentration in weight percent. Surfactant solutions were prepared in East Seminole
brine with 100 parts per million (ppm) oxygen scavenger and 50 ppm scaling inhibitor.

Foamability of the Field-Mixed Surfactant Solution

In order to verify the foaming capability of the field-mixed surfactant solution, a coreflood was conducted at
reservoir conditions. The objective was to determine the foamability of the field-mixed surfactant to exclude
any adverse chemistry impacts or contamination on the stability of the surfactant solution. Experimental
conditions were set to 42⁰C and 2600 psia, similar to conditions used in previous experimental work for
consistency. The experiment was conducted with a 1.5-inch diameter and 6-inch long Indiana limestone
core with a permeability of 796 mD. The surfactant solution had a concentration of 0.5 wt% and included
100 ppm oxygen scavenger and 50 ppm scale inhibitor, as used in the field. The core was first flooded with
three pore volumes (PV) of surfactant solution to satisfy any adsorption. Supercritical CO2 and surfactant
solution were then co-injected at 1 ft /day at 70% foam quality. Figure 7 shows the results of the in-situ
foamability test of the field-mixed surfactant sample.

Figure 7—Apparent foam viscosity as a function of pore volume injected

for the in-situ foamability test with the field-mixed surfactant sample.

Foam generation was observed within the first 0.5 PV injected based upon the increasing apparent
viscosity (pressure gradient). Apparent foam viscosity stabilized after 0.5 PV injected, reaching a steady-
state apparent viscosity of 97 cP. Therefore, the field-mixed surfactant solution has good foamability and
10 SPE-200450-MS

the pressure gradient is above the minimum required pressure gradient for foam generation at a flow rate of
1 ft/day. The results indicate that there were no adverse chemical issues influencing the foamability of the
field-mixed surfactant solution due to contamination during the field mixing procedure.

Initial Pilot Results

The reservoir response was evaluated by analyzing injection flow rates, injection BHP, injection profile logs,
and three-phase production rates. Baseline measurements were compared to pilot surveys to evaluate foam

Downloaded from http://onepetro.org/SPEIOR/proceedings-pdf/20IOR/1-20IOR/D011S007R003/2369009/spe-200450-ms.pdf by University of Bergen user on 11 January 2022

generation, CO2 mobility reduction, and increased reservoir sweep efficiency and oil production. Figure 8
shows the injection rates from two weeks before the pilot and for six complete SAG cycles.

Figure 8—Pilot injection rates of CO2 (red curve) and surfactant solution (blue curve) from two weeks
before the pilot and for six pilot SAG cycles. Injected foam quality (black solid points) for each SAG
cycle is also shown on the secondary x- and y-axis. Black dashed line represents the start of the pilot.

Injection rates during the pilot were 520 rb/day and 470 rb/day for surfactant solution and CO2,
respectively. The volumetric ratio of the cumulative injected CO2 relative to the total volume of CO2 and
surfactant injected was used to evaluate injected foam quality per cycle (Figure 8, black solid points). The
aim was to inject foam at 70% quality (0.70 gas fraction) per cycle as determined in the laboratory studies.
As shown in Figure 8, the injected foam qualities ranged from 61% to 71%, which was within the designed
target. After six complete SAG cycles, approximately 4% HCPV has been injected in the pilot pattern.

Injectivity Analysis
The first response to foam injection was anticipated to be observed in the injection well (I1) in the form of
reduced CO2 injectivity. Therefore, baseline pressure and temperature readings were taken with a DHPG
two weeks before starting the pilot. During the pilot, the DHPG was pulled every other month from the well
to retrieve the data for analysis. The injectivity index was calculated for the baseline CO2 period and for
each of the CO2 slugs during the pilot (Figure 9).
SPE-200450-MS 11

Downloaded from http://onepetro.org/SPEIOR/proceedings-pdf/20IOR/1-20IOR/D011S007R003/2369009/spe-200450-ms.pdf by University of Bergen user on 11 January 2022

Figure 9—Injectivity index during the baseline CO2 period (black curve) and during each CO2 slug after surfactant slug
injection (red, green, purple, and blue curves). The rate of change (slope) for each slug is shown in the right inset.

Baseline CO2 injectivity was on average 6.45 Mscf/psi (Figure 9, black curve), whereas CO2 injectivity
directly after surfactant slugs was on average 1.96 Mscf/psi. Therefore, initial CO2 injectivity was reduced
by nearly 70%, indicating reduced CO2 mobility after each surfactant slug when compared with the baseline
CO2 injectivity. Foam dry-out and increased pathways for CO2 flow may have caused the CO2 injectivity to
slowly increase during each CO2 slug. The observed reduction in CO2 injectivity is difficult to distinguish
from potential effects on CO2 relative permeability at the high water saturations after each surfactant slug.
Therefore, a WAG injection is planned at the end of the pilot to reconcile this issue.
Transient analysis was conducted to analyze unsteady-state flow at the injection well during the pilot.
Injection BHP and temperature data from the DHPG was used to evaluate the differential pressure (dP)
and differential temperature (dT) through time for the first five SAG cycles (Figure 10). As shown in
Figure 10, the temperature responses are quite similar during the first five cycles. On the other hand,
differential pressure consistently increases during periods of surfactant injection and decreases during the
subsequent CO2 injection. The pressure increase (buildup) during surfactant injection is due to a decrease
in mobility, showing development of a mobility bank in the reservoir. The mobility reduction could be due
foam development in the formation and/or to relative permeability and viscosity effects. History matching
of this transient pressure data is in progress and is expected to shed some light to these possible mobility
effects.
12 SPE-200450-MS

Downloaded from http://onepetro.org/SPEIOR/proceedings-pdf/20IOR/1-20IOR/D011S007R003/2369009/spe-200450-ms.pdf by University of Bergen user on 11 January 2022

Figure 10—Plot of transient bottom hole pressure (dP) and temperature (dT) versus time for the first
five SAG Cycles. dP1 corresponds to slug 1, dP2 to slug 2, and so on. Differential pressure increases
during each consecutive surfactant injection and relaxes during the each following CO2 injection.

Radioactive tracer injection profile logs were also collected to determine zones of injection during the
baseline period and during the pilot (Figure 11). The vertical injection profiles may reveal the impacts of
in-depth CO2 mobility reduction, if flow has been diverted from high permeability streaks into the lower
permeable, unswept regions of the reservoir. For the injection profiles collected prior to the pilot (May
2019) and shortly after starting the pilot (July 2019), a high permeability streak between depths of 5465 ft
and 5471 ft took most of the flow for both injected phases (56% mean). Thus, no change in zonal injection
by foam was observed after two complete SAG cycles in July 2019. Since the profile from July 2019 was
ran at the end of the CO2 slug, the majority of the surfactant solution was likely far from the well and the
profile is indicative of normal CO2 injection. Therefore, after 20 days of CO2 injection, the injection profile
of CO2 has returned to baseline measurements. Subsequent CO2 profiles were scheduled earlier in the CO2
slugs and in the surfactant slugs to better monitor fluid diversion near the well, when both CO2 and foaming
agent are present.
SPE-200450-MS 13

Downloaded from http://onepetro.org/SPEIOR/proceedings-pdf/20IOR/1-20IOR/D011S007R003/2369009/spe-200450-ms.pdf by University of Bergen user on 11 January 2022

Figure 11—Injection profiles from before the pilot (May 2019), at the end of second CO2 slug (July
2019), at the end of the fifth and seventh surfactant slugs (Oct 2019, Dec 2019), and at the start of the
seventh CO2 slug (Dec 2019). Blue profiles indicate water (surfactant solution) injection, whereas red
indicates CO2. Green points on first log indicate perforated intervals. A permeability log for the pilot
injection well is also shown at far right. Note high permeability streak at depths of 5465 ft to 5471 ft.

The water profile taken at the end of the fifth surfactant slug in October 2019 indicated a slight downward
shift in flow, compared to earlier measurements, with 57% of flow between 5468 ft and 5488 ft. For the next
set of profiles, an effort was made to capture foam (if any) near the well by collecting the injection profiles
in rapid succession. Therefore, a profile was collected at the end of the seventh surfactant slug and at the
start of the subsequent CO2 slug in December 2019. These profiles revealed that flow has increased into
the reservoir intervals adjacent to the high permeability streak. The analysis of the profiles through time
indicates increased flow into the reservoir interval and potential blockage of the high permeability streak
from 5465 ft to 5471 ft. This may be due to foam diverting CO2 flow from the high permeability streak
into the adjacent upper and lower reservoir intervals. However, the impact of high water saturation on CO2
relative permeability could also explain the change in vertical injection profile. Since the foam treatment
was designed for in depth mobility control, rather than near well conformance, its impact on the injection
profile is expected to be minor. However, the observed change in the injection profile could indicate foam
generation further from the well.

Production Analysis
Three-phase production data was collected from the four surrounding producers in the pilot pattern and
for the peripheral production wells. The first sign of CO2 mobility reduction was expected to be seen as
a decrease in gas production compared to historical measurements. In addition, the foam treatment aimed
to increase CO2 reservoir sweep efficiency, which may be observed through increased liquid production
(water and oil) compared to historical values. However, production response is not expected to be observed
until several months after the 12-month pilot based upon earlier simulation results and the relatively small
volumes injected during the pilot. Figure 12 shows the three-phase production rates for the pilot pattern
from two weeks before pilot until after six complete SAG cycles.
14 SPE-200450-MS

Downloaded from http://onepetro.org/SPEIOR/proceedings-pdf/20IOR/1-20IOR/D011S007R003/2369009/spe-200450-ms.pdf by University of Bergen user on 11 January 2022

Figure 12—Three-phase production rates for the four surrounding production wells in the pilot pattern from two
weeks before the pilot until after six complete SAG cycles. The black dashed lines represents the start of the pilot.

To date, the pattern production response has been limited. Nevertheless, a decrease in gas production was
observed after four SAG cycles (Figure 12, red curve). This may be due to reduced CO2 mobility by foam
and/or a relative permeability effect with water and CO2 present. It is difficult to separate the two issues with
the currently available data. After six complete SAG cycles, there has been no observable increase in oil
production (Figure 12, green curve). However, sustained oil production despite less volume injected during
the pilot compared to historical values has been observed. Therefore, the foam treatment has maintained
the current level of oil production and has reduced CO2 purchasing costs. Additional production analysis
is ongoing as data becomes available and a follow-up WAG injection will be conducted to improve the
understanding of the reservoir response to foam injection.

Simulation Study
To provide additional information on interwell connectivity and foam's impact on CO2 breakthrough in the
field, a conventional finite-difference compositional model was set-up. The objective was to calibrate the
model to the observed breakthrough time of CO2 from the baseline IWTT. The model was then used to
evaluate foam's impact on CO2 breakthrough and guide decisions for the repeat IWTT to be conducted at
the end of the pilot. A high-resolution two-dimensional cross-section model was set-up, which included
the porosity and permeability distribution from a validated sector-scale model of the pilot pattern and
surrounding producers. Additional information on the sector-scale model and the history match can be found
in Sharma et al. 2017.
The model included one injector (I1) and one producer (P1) to simulate the historical (baseline) CO2
injection period and the 12-cycle SAG pilot period. The grid contained 54 layers, which were refined from
the base sector model. Individual grid cells were 10 ft in the x-direction and an average of 2.5 ft in the z-
direction. Two water components were used; one for surfactant solution and one for water. Laboratory foam
scans were used to derive foam model parameters for the implicit-texture local-equilibrium foam model to
capture the effects of foam mobility, foam viscosity as a function of velocity, surfactant concentration, and
water and oil saturations (Law et al. 1992; Cheng et al. 2000; Farajzadeh et al. 2012). To account for the
SPE-200450-MS 15

decrease in gas mobility during foam floods, the gas relative permeability for floods without foam is scaled
by a mobility reduction factor, whereas the water relative permeability remains unchanged.
The effect of permeability on foam behavior was included by dividing the model into three regions
depending upon the grid cell permeability. Region 1 had permeability less than 10 mD, Region 2 had a
permeability range from 10 to 50 mD, and Region 3 had permeability greater than 50 mD. These regions
were assigned different foam model parameters (fmmob, fmdry, and epdry). Most of the experiments in the
laboratory were performed with reservoir cores having permeability in range of Region 2 (20 to 30 mD).
Therefore, the base values for foam model parameters for Region 2 were obtained by performing regression

Downloaded from http://onepetro.org/SPEIOR/proceedings-pdf/20IOR/1-20IOR/D011S007R003/2369009/spe-200450-ms.pdf by University of Bergen user on 11 January 2022

on the quality scan data to fit the empirical foam model (Zheng et al. 2016; Sharma et al. 2017). In absence
of cores from Region 1 and Region 3, assumptions were made based upon earlier observations about the
foam model parameters.
Figure 13 shows the permeability distribution in the model and the foam quality scan that was used to
derive the foam model parameters for permeability Region 2. The surfactant selected for the pilot was found
to have very low adsorption reservoir material in the laboratory so it was excluded from the simulation study.
In order to model near well foam dry-out during SAG, the cells connected to the injector were assigned a
reference gas mobility reduction (fmmob) of zero for foam collapse around the injector (Leeftink et al. 2015).

Figure 13—Permeability distribution in the cross-section model (left). The foam quality
scan from which the Region 2 foam model parameters (green curve) were derived is
shown at upper right along with the foam model parameters for each permeability region.

To determine CO2 breakthrough time in the model, a CO2 tracer was injected during the baseline CO2
period. The aim was to evaluate the model's response and compare it to the observed CO2 breakthrough
time of 22 days in the field. Figure 14 shows the model results of the baseline CO2 tracer test.
16 SPE-200450-MS

Downloaded from http://onepetro.org/SPEIOR/proceedings-pdf/20IOR/1-20IOR/D011S007R003/2369009/spe-200450-ms.pdf by University of Bergen user on 11 January 2022

Figure 14—Baseline CO2 tracer injection (blue line) and production (red curve) as a function
of time in the cross-section model. Time (t) is given in days since the tracer was injected.
CO2 breakthrough (BT) was observed in 22 days, consistent with the field CO2 tracer results.

The CO2 tracer was produced in P1 in 22 days, which is the same as the tracer breakthrough time from
the IWTT in the field. This provided confidence in the model for representing CO2 migration rates in the
reservoir, providing a basis for evaluating the impact of foam on CO2 mobility. In order to evaluate foam's
impact on CO2 mobility, a separate CO2 tracer was injected in the first CO2 slug (after the first surfactant
solution slug) of the pilot rapid SAG injection scheme. Figure 15 shows the pilot CO2 tracer injection and
production results as a function of time, given in days since the tracer was injected. The pilot CO2 tracer
was first produced in P1 in 66 days. Thus, the CO2 breakthrough time was increased by 44 days during the
foam pilot compared to the baseline CO2 tracer.

Figure 15—Foam pilot CO2 tracer injection (blue line) and production (red curve) as a function of time in the
cross-section model. Time (t) is given in days since the tracer was injected. Pilot CO2 breakthrough (BT) was
observed in 66 days, increasing CO2 breakthrough time by 44 days compared to the baseline CO2 tracer test.

The simulation results help to guide the repeat IWTT and provide insight for designing the sampling
schedule for producers in the field. Based upon these results, it is expected that the foam treatment
will increase CO2 breakthrough time by approximately 1.5 months. Additional simulation efforts aim to
investigate the injectivity response to foam injection. Therefore, a radial model is being constructed to
accurately represent injectivity and near well foam behavior in a SAG process.
SPE-200450-MS 17

Conclusions
A CO2 foam field pilot is underway in East Seminole Field in the Permian Basin of west Texas. A
surfactant-stabilized foam was selected to mitigate CO2 EOR challenges by reducing CO2 mobility, thereby
improving sweep efficiency, oil recovery, and CO2 storage potential. The surfactant system was designed
in the laboratory by measuring surfactant adsorption and verifying foam stability. A surfactant-alternating-
gas (SAG) injection strategy, with 10 days of surfactant solution followed by 20 days of CO2, began in
May 2019. The baseline data collection and pilot monitoring program aimed to establish baseline in the

Downloaded from http://onepetro.org/SPEIOR/proceedings-pdf/20IOR/1-20IOR/D011S007R003/2369009/spe-200450-ms.pdf by University of Bergen user on 11 January 2022

pilot pattern and monitor the pilot to evaluate reservoir response to foam injection. Surveys included CO2
injection profiles, CO2 tracer tests, collection of injection bottom hole pressures, and three-phase flow rates.
In addition, produced water samples were collected once a week for analysis of surfactant breakthrough
time.
Initial results reveal that pilot CO2 injectivity was reduced by 70%, compared to baseline values,
indicating reduced CO2 mobility after each surfactant slug. However, the reduction in CO2 injectivity could
also be related to a CO2 relative permeability reduction solely due to the presence of water. A water-
alternating-gas (WAG) injection at the end of the pilot would offer additional insight. Since the foam
treatment was designed for in-depth mobility control, rather than near well conformance, its impact on the
injection profile is expected to be minor. Analysis of injection profiles indicates increased flow into the
reservoir interval during the pilot and potential blockage of a high permeability streak, which has been
taking majority of the flow historically. It is expected from simulation studies that CO2 breakthrough may
be delayed by 1.5 months during the pilot. Production response is not expected to be observed for another
six to nine months due to the volumes injected during the pilot. However, the early signs of sustained oil
production despite less pore volume injected during the pilot are encouraging.
Ongoing works aims to investigate the foam propagation rate and flow velocities in the field. There
are also questions regarding the impact of foam and/or relative permeability on injectivity and mobility
reduction when switching between surfactant solution and CO2 in a SAG process. Therefore, a near injector
radial model is being set-up and calibrated with observed field data. Field data collection will continue for
several months after the pilot to monitor production response and the impacts of foam of CO2 mobility.
Three-phase production and injection rates at the end of the pilot will be used to assess CO2 retention in the
reservoir to evaluate the ability of foam to provide additional CO2 storage capacity during CO2 EOR.

SI Metric Conversion Factors

Acre × 4.046873 E+03 = m2

°API 141.5/(131.5+ °API) = g/cm3
bbl × 1.589873 E – 01 = m3
cp × 1.0 E – 03 = Pa∙s
°F (°F – 32)/1.8 = °C
ft × 3.048 E – 01 = m
psi × 6.894757 E + 00 = kPa

Acknowledgments
The authors wish to acknowledge the Norwegian Research Council CLIMIT program for financial support
under grant number 249742 and Gassnova project number 618069. The authors also acknowledge industry
18 SPE-200450-MS

partners; Shell Global Solutions, TOTAL E&P USA, Equinor ASA, and Occidental Petroleum. The authors
also thank the field operator.

Nomenclature
fg Gas fraction or foam quality
cP Centipoise
K Permeability

Downloaded from http://onepetro.org/SPEIOR/proceedings-pdf/20IOR/1-20IOR/D011S007R003/2369009/spe-200450-ms.pdf by University of Bergen user on 11 January 2022

mD Millidarcy
MPa Megapascal
Psig Pound per square inch, gauge
Mscf Thousand standard cubic feet
οAPI American Petroleum Institute gravity
rb/day Reservoir barrels per day
Mscf/day Thousand standard cubic feet per day
ppm Parts per million
Sor Residual oil saturation, fraction of pore volume
fmmob Foam model, maximum gas mobility reduction factor
fmdry Foam model parameter in Fwater
epdry Foam model parameter in Fwater
fmsurf Foam model parameter in Fsurf
epsurf Foam model parameter in Fsurf

Abbreviations
CCUS Carbon Capture, Utilization, and Storage
CCS Carbon Capture and Storage
DI Deionized water
EOR Enhanced Oil Recovery
SAG Surfactant Alternating Gas
WAG Water Alternating Gas
DHPG Downhole Pressure Gauge
BHP Botom Hole Pressure
BT Breakthrough
Wt % Weight Percent
GOR Gas-Oil Ratio
MRF Mobility Reduction Factor
IWTT Interwell CO2 tracer test
PV Pore Volume
HCPV Hydrocarbon Pore Volume
RI Refractive Index
HPLC High Performance Liquid Chromatography
ELSD Evaporative Light Scattering Detector

References
Alcorn, Z.P. Upscaling CO2 Foam for Enhanced Oil Recovery and CO2 Storage from Laboratory to Field Scale - An
Integrated Approach to Designing a Field Pilot Test. 2018. PhD Dissertation. University of Bergen, September 2018.
Alcorn, Z.P., Sharma, M., Fredriksen, S. B., Fernø, M.A., and Graue, A. 2019. An Integrated CO2 Foam EOR Pilot
Program with Combined CCUS in an Onshore Texas Heterogeneous Carbonate Field. SPE Reservoir Evaluation and
Engineering 22 (04): 1449–1466. https://doi.org/10.2118/190204-PA.
SPE-200450-MS 19

Cheng, L., A. B. RemeD. Shan et al 2000. Simulating Foam Processes at High and Low Foam Qualities. Proc., SPE/DOE
Improved Oil Recovery Symposium, Tulsa, Oklahoma. SPE-59287-MS. https://doi.org/10.2118/59287-MS.
Chou, S.I., Vasicek, S.L., Pisio, D.L., Jasek, D.E., Goodgame, J.A. 1992. CO2 Foam Field Trial at North Ward Estes.
Presented at the 67th SPE Annual Technical Conference and Exhibition, Washington, D.C., October 4-7, 1992.
Enick, Robert Michael, David Kenneth Olsen, James Robert Ammer et al 2012. Mobility and Conformance Control
for CO2 EOR via Thickeners, Foams, and Gels – A Literature Review of 40 Years of Research and Pilot Tests.
Proc. SPE Improved Oil Recovery Symposium, 14-18 April, Tulsa, Oklahoma, USA. SPE-154122-MS. https://
doi.org/10.2118/154122-MS.
Falls, A.H., Hirasaki G.J., Patzek, T.W., Gauglitz., P.A., Miller, D.D. and Ratulowski, T. 1988. Development of a

Downloaded from http://onepetro.org/SPEIOR/proceedings-pdf/20IOR/1-20IOR/D011S007R003/2369009/spe-200450-ms.pdf by University of Bergen user on 11 January 2022

Mechanistic Foam Simulator: The Population Balance and Generation by Snap-Off. SPE Reservoir Engineering 1988,
3 (03): 884–892. https://doi.org/10.2118/14961-PA
Farajzadeh, R., A.Andrianov R. Krastev et al 2012. Foam–oil interaction in porous media: Implications for foam
assisted enhanced oil recovery. Advances in Colloid and Interface Science 183-184: 1–13. https://doi.org/10.1016/
j.cis.2012.07.002.
Heller, J.P. 1966. Onset of Instability Patterns Between Miscible Fluids in Porous Media. Journal of Applied Physics
1966, 37, 1566–1579.
Heller, J.P., Boone, D.A., and Watts, R.J. 1985. Field Test of CO2-Foam Mobility Control at Rock Creek. Presented at the
SPE 60th Annual Technical Conference and Exhibition, Las Vegas, Nevada USA. September 22-25, 1985. SPE-14395.
Hoefner, M.L. and Evans, E.M. 1995. CO2 Foam: Results from Four Developmental Fields Trials, SPE Reservoir
Engineering, November 1995, p. 273–281.
Jian, G., Puerto, M.C. Wehowsky, A. et al 2016. Static Adsorption of an Ethoxylated Nonionic Surfactant on Carbonate
Minerals. Langmuir 32 (40): 10244–10252.
Law, D. H. S., Z. M.Yang, T. W. Stone. 1992. Effect of the Presence of Oil on Foam Performance: A Field Simulation
Study. SPE Reservoir Engineering 7 (02). SPE-18421-PA. https://doi.org/10.2118/18421-PA.
Leeftink, T.N., Latooij, C.A. and Rossen, W.R. 2015. Injectivity errors in simulation of foam EOR. J Petrol Sci Eng 126,
26–34.
Martin, F.D., Heller., J.P., Weiss, W.W., and Tsau, J.S. 1992. CO2-Foam Field Verification Pilot Test at EVGSAU Injection
Project Phase 1: Project Planning and Initial Results. Presented at the SPE Improved Oil Recovery Conference, Tulsa,
Oklahoma, USA, 22-24 April, 1992. SPE/DOE-24176
Martin, F.D., Stevens, J.E. and Harpole, K.J. 1995. CO2 Foam Field Test at the East Vacuum Grayburg/San Andres Unit.
SPE Reservoir Engineering, November 1995, p. 266–272
Rognmo A., Fredriksen, S.B., Alcorn, Z.P., Sharma, M., Føyen T., Eide, Ø., Fernø M.A., and Graue, A. 2019. Pore-to-
Core EOR Upscaling for CO2 Foam for CCUS. SPE Journal Preprint July 2019. https://doi.org/10.2118/190869-PA
Rossen, W.R. 1996. Foams in Enhanced Oil Recovery. In Foams Theory, Measurements, and Applications. eds.
Prud'homme, R.K. and Khan, S.A. volume 57, ch. 11, p 414–464, Marcel Dekker, Inc. New York
Sharma, M., Z P.Alcorn S B. Fredriksen et al Numerical Modeling Study for Designing CO2-foam Field Pilot.
IOR 2017 - 19th European Symposium on Improved Oil Recovery Stavanger, Norway, 24 April. https://
doi.org/10.3997/2214-4609.201700339.
Schramm, L. L. 1994. Foams: Fundamentals and Applications in the Petroleum Industry. Washington, DC: American
Chemical Society.
Stephenson, D.J., Graham, A.G., Luhning, R.W. 1993. Mobility Control Experience in the Joffre Viking Miscible CO2
Flood. SPE Reservoir Engineering, August 1993, p. 183–188.
Zheng, Y., Muthuswamy, A., Ma, K., Wang, L., Farajzadeh, F., Puerto, M., Vincent-Bonnieu, S., Eftekhari, A.A., Wang,
Y., Da, C., Joyce, J.C., Biswal, S.L. and Hirasaki, G.J. 2016. Insights on Foam Transport from a Texture- Implicit
Local Equilibrium Model with an Improved Parameter Estimation Algorithm. Ind Eng Chem Res, 55 (28), 7819–7829.