S

SPE 1506
634

E
Experime
ental Stud
dy for Op
ptimizing Injected Surfactant Volum
me in a Miiscible
C
Carbon Dioxide
D Fllood
G
Gbolahan Afo
onja, Richard G. Hughes an
nd Shrinidhi Shetty,
S Louisi ana State Un
niversity

C
Copyright 2012, Society
y of Petroleum Enginee
ers

T
This paper was prepare
ed for presentation at the SPE EOR Conferen
nce at Oil and Gas We
est Asia held in Muscatt, Oman, 16–18 April 2
2012.

T
This paper was selected for presentation by an a SPE program comm mittee following review of information containned in an abstract submmitted by the author(s)). Contents of the papeer have not been
re
eviewed by the Society y of Petroleum Engine eers and are subject to o correction by the autthor(s). The material ddoes not necessarily reeflect any position of the Society of Petroleu
um Engineers, its
officers, or members. Electronic
E reproductionn, distribution, or stora
age of any part of thiss paper without the wrritten consent of the S Society of Petroleum E Engineers is prohibite
ed. Permission to
re
eproduce in print is res
stricted to an abstract of
o not more than 300 words;
w illustrations mayy not be copied. The abbstract must contain co
onspicuous acknowled dgment of SPE copyrig ght.

AAbstract
OOne major drawback in the implemen ntation of COO2/surfactant enhanced oiil recovery iss the high co ost of the
ssurfactant volume needed and the extra apolated amount of surfacttant required for use in the e field. Thereffore, there
eexists a need d for a methodology to eva aluate the mo ost economicc surfactant vvolume that w will lead to opptimum oil
pproduction.
This paper
p presentts the results of laboratory
y experimentss that were deesigned to stu udy the perforrmance of
fo
foam injectionn and to obta ain informatioon on the pa arameters thaat will be utilized in a future modeling g study to
ddetermine the e minimum bu ut most effecttive volume of
o surfactant n
needed for op ptimum foam generation a and hence
ooil recovery in
n a porous me edium. The efffect of reserv
voir pressure and surfactan nt concentratiion is studied.
Carboon dioxide an nd surfactant solution werre co-injectedd into Berea ssandstone co ore that was saturated
wwith brine andd oil after a waaterflood proc
cess. Some ofo the values oof parameterss utilized in th
he experimen nts include
ssurfactant conncentrations ofo 0.1 wt % anda 0.5 wt %,, average bacck pressure o of 2350 psi and total flowrrate of 0.4
ccm3/min, with h CO2 injecte ed at 0.3 cm m3/min and surfactant con ncentration aat 0.1 cm3/min, which ressults in a
vvolumetric floww ratio of 3:1 and foam quality of 75%.
Resullts show that at a higher surfactant con ncentration, m
more oil was rrecovered, an nd a steady in ncrease in
ppressure diffeerence was observed. This s indicates th
hat foam was formed, whicch was also sseen at the p production
eend of the sysstem. The foa am created du uring the expeeriments with
h the 0.1 wt % surfactant cconcentration,, on visual
innspection, apppeared to be weaker than that formed in i the case off the 0.5 wt % concentratio on.

Introduction
TThe applicatioon of carbon dioxide has been recognized as a succcessful meth hod for recovvering residua al oil after
pprimary and secondary
s recovery. The method is offten hampere ed by heterog geneity effectts and by the e viscosity
ccontrast between CO2 and d oil that leads to very highh mobility rattios. The use of surfactantts is one wayy to rectify
the drawbacks associated with CO2 floods as they help h to reducce the viscosity variation bbetween the d displacing
aand displaced d fluids and by
b restricting the flow of COC 2 through h high permeab bility zones. TThe economiccs of such
ffloods are dep pendent on thhe amount off surfactant necessary to p provide improoved sweep. Having a me ethodology
too scale laborratory studies
s to field need
ds is then one e of the princciple steps in deciding whe ether to imple
ement the
ssurfactant trea
atment.
Severral researchers have carrie ed out studiees that aim to o understand the process of foam gene eration via
ssurfactant solution and gas injection as s it pertains to
t oil recoverry. Liu et. al. (2005) studie ed various paarameters
that impact CO O2-foam stability. The varia
ables studiedd include presssure, temperrature, and su urfactant conccentration.
TThey found th hat generated foam was stable at all tes sted temperattures, 25°C to o 75°C, at surfactant concentrations
ggreater than 0.1
0 wt %. Foam stability alsoa occurredd at pressure s between 80 00 psi to 20000 psi when surfactant
cconcentration was above 0.025 wt %. % Yaghoobi (1994) utilizzed laborato ory experimen nts to investtigate the
mmechanism of o foam mobility and its efffect on flowra ate, flow quaality, type of ssurfactant, su
urfactant concentration
2 SPE 150634

and rock permeability. Yaghoobi observed that foam was generated at all flowrates (0.5 cc/hr to 109.75 cc/hr);
foam mobility increased with injection flowrates and foam quality; foam flow performance is considerably affected
by surfactant concentration; and, with increasing liquid viscosity at a fixed gas flowrate, pressure drop increased
with a corresponding decrease on mobility.
Coombe et. al. (1997), Wassmuth et. al. (2000), and Vikingstad (2006) succesfully applied an empirical
model that was used in a commercial reservoir simulator (CMG-STARS). The presence of foam is indicated by
modifying gas relative permeability using a reduction factor (CMG, 2009). The reduction factor is dependent on
surfactant concentration, oil saturation, gas velocity, capillary number and a mobility reduction factor.
In this work, the CMG empirical technique is used to model the effect of varying the concentration of
surfactant, and modifying other parameters to obtain the similar values for the modified gas relative permeability
for both cases of low and high surfactant concentrations. The values required for use in the empirical model are
derived from coreflood experiments.

Laboratory Experiments
Experimental Apparatus
The experimental setup consists of four parts: the injection and fluid delivery unit, the core unit, the production
unit, and the data acquisition unit. A diagram of the equipment setup is shown in Figure 1.
The injection and fluid delivery unit transports fluids into the core system. It is made up of two Teledyne-
ISCO syringe pumps, transfer vessels, stainless steel flow tubes, and 2- and 3-way valves. The Teledyne-ISCO
pumps, A and B, are setup to operate independently. Each pump has a volume capacity of 266 cc, with a flowrate
range of 0.001 cm3/min to 107 cm3/min. The pumps have a pressure rating of 7500 psi. Pump A is used to inject
water to drive the pistons in the transfer vessels, and pump B is used to inject brine and surfactant directly into the
core. Decane, CO2, and cleaning chemicals (methylene chloride, toluene, and methanol) are held in five transfer
vessels that are connected to pump A. The flow lines and valves are made of stainless steel which is able to
withstand the corrosive effects of CO2, and have a working pressure of 5000 psi.
The core unit is made up of a core holder, a core, and stainless steel spacer tubes. A Hassler-type core
holder with a maximum pressure rating of 5000 psi was used in all experiments. The core holder is designed to
hold cores with a maximum length of 12 inches with 1 inch diameter. Five pressure-taps placed 2 inches apart at
the 2-, 4-, 6-, 8-, and 10-inch positions along the top of the core holder have pressure transducers attached to
them to measure pressure within the core. A Berea sandstone core is housed within a 0.25-inch thick Viton sleeve
that is placed inside the core holder. Stainless steel spacer bars of various lengths are available to fill-up empty
space within the Viton sleeve if shorter cores are used. In this study, a 3-inch core is used; therefore spacer tubes
of 7.5 and 1.5 inches are inserted at the upstream and downstream positions of the core. The pressure taps at the
8-in and 10-in locations are taken as the inlet and outlet ports for the analysis that follows to avoid end [Link]
properties of the core used are listed in Table 1.
The production unit is comprised of a backpressure regulator (BPR), a graduated burette, and a wet-test
meter. The backpressure regulator, manufactured by TESCOM, is set to maintain an average pressure of 2350
psi and is connected to the outlet of the core holder. A graduated burette is connected to the outlet end of the
BPR and is used to collect produced liquid. A system of 2-size tubings, 1/8-in and 1/4-in inner diameter, is set up
with the smaller tubing inserted within the bigger tubing. This enables all produced fluids to flow through the 1/4-in
ID tubing and when dense CO2 flashes and reverts to gaseous phase at room conditions, it flows through the
annulus between the tubings to the wet-test meter.
A personal computer (PC), pressure transducers, a data acquisition box, various software packages, and
a camera make up the data acquisition unit. The pressure transducers are connected to a data acquisition device
(DAQ) that is manufactured by National Instruments. The DAQ device converts analog input signals obtained
from the transducers to digital data that is transmitted to the PC. A digital video camera that is driven by the
software, LabVIEW 8.5, acquires continuous images from the production burette. The software distinguishes the
interface between produced fluids, measures the increase of each fluid, and converts the interface location
information to volumes. Different dye colors are applied to the fluids to allow the software to accurately discern the
fluid interface.
SPE 150634 3

Experimental Fluids
N-decane was used as the oleic phase because it does not affect the wettability of the core (Kulkarni, 2003). A
2% by weight calcium chloride solution was prepared by mixing distilled water and CaCl2 salt. The use of brine
instead of fresh water is essential to minimize the swelling of clays in sandstone (Shetty, 2011). The cleaning
fluids include methylene chloride, methanol, and toluene. These chemicals were chosen because when applied in
the proper sequence during cleaning, the core was restored to a state that was close to the initially measured
properties, especially permeability. Carbon dioxide was contained in a transfer vessel and drawn into the core via
pump A. Dense CO2 was used in this study because pressure values were above 1000 psi and room temperature
was 24°C. This temperature is less than the critical temperature of 31°C. The surfactant used in this work was
Chevron’s CS-1045. This surfactant has been identified as an effective foaming agent based on reports by
Yaghoobi (1994), Heller (1994), and Moradi-Araghi et. al. (1997). This surfactant solution was formulated by
mixing brine and surfactant. Table 2 shows the properties of fluids used this study.

Experimental Procedure
The bulk volume of the core was measured using the diameter and length of the core. Moisture was removed
from the core by heating it in an oven for 5 hours at a temperature of 100°C. After the core cooled, it was placed
on a scale and the weight was measured. The core was then placed in a glass evaculation chamber that was
attached to a vacuum pump that was used to extract air. The chamber was filled with brine until the core was fully
submerged and allowed to soak for 30 minutes. The core was removed and weighed and the difference in mass
was used to obtain the pore volume.
The core was placed in the Viton sleeve, which was in turn placed in the core holder. The annulus between the
sleeve and core holder was filled with hydraulic oil and pressure was applied through a hand pump. The exerted
axial and radial pressure helps to keep the sleeve and core in good contact and prevents the flow of fluids along
the core surface.
A back pressure of approximately 2350 psi was applied to the system. At this pressure, the requirement
for miscibility between decane and CO2 is met. The minimum miscibility pressure for CO2 and decane was
estimated to be 1880 psi at 160°C and 1000 psi at 82°C (Kulkarni, 2003).
The core was completely saturated by injecting about 2 PV of brine at 2.0 cm3/min. The flowrate was
reduced to 1.5 cm3/min and the system was allowed to stabilize, and the pressure drop between the inlet and
outlet was recorded. Pressure drops at 1.0 cm3/min and 0.5 cm3/min were also recorded. Absolute permeability of
the core was measured using Darcy’s Law.
Decane was injected into the core at 1.5 cm3/min to displace brine. The oil flood was continued until there
was no more observed brine production. The volume of oil remaining in the core and the initial oil saturation were
obtained from the difference between the total volume of oil injected, the deadweight volume, and the volume of
oil produced. Material balance was used to calculate the volume of brine left in the core and the residual water
saturation.
A waterflood was carried out by injecting brine into the core at 1.4 cm3/min. This brought the saturation of
decane to residual oil saturation. After about 5 PV of brine was injected, there was no observed increase in oil
production.
In order to generate foam in the core, CO2 and surfactant solution were co-injected. Surfactant
concentrations of 0.5 wt % and 0.1 wt % were used. The 0.5 wt % concentration was injected at a rate of 0.1
cm3/min and CO2 at 0.3 cm3/min. The volumes of CO2, decane, and brine/surfactant solution produced and
corresponding pressure drops were recorded. The core was cleaned and the sequence of experiments was
repeated from permeability measurement to waterflood. The final step was the co-injection of 0.1 wt % surfactant
concentration, injected at a rate of 0.1 cm3/min, and CO2 at 0.3 cm3/min.

Results and Discussion

The core was thoroughly cleaned after the first sets of experiments were carried out and permeability was
measured. There was a decrease in permeability value from 13.15 mD to 11.32 mD. The graph shown in Figure 2
compares oil recovery (RF) for the waterflood carried out before the foam injection for the 0.5 wt % and 0.1 wt %
surfactant concentrations. In both cases, about 56% of the initial oil saturation was recovered. These results
indicate that despite the drop in permeability value, approximately equal amounts of oil were recovered. The
4 SPE 150634

simultaneous injection of CO2 and surfactant solution at 0.5 wt % led to the recovery of about 56% of oil left after
waterflood (Sowf). For the 0.1 wt % concentration, the recovery factor was about 23%. A comparison of oil
recovery for the co-injection cases is shown in Figure 3.
At a higher surfactant concentration, the pressure difference along the core increased significantly, and
denser and stronger foam was generated, as observed visually at the outlet of the core. The foam was able to
prevent the CO2 from channeling through and allowing it to contact more oil. As injected CO2 pore volume
increased, an increase in pressure difference was also observed. This effect occurred because with further
simultaneous injection, more foam was formed at the outlet end of the core which decelerated the flow of CO2 and
surfactant solution, leading to a backup at the inlet. The blocking effect of foam also significantly reduced the
amount of CO2 produced (Figure 4). The proportional increase in pressure difference was not observed in the
instance of the injection of the lower surfactant concentration. Figure 5a shows the relationship between pressure
drop and pore volume of injected CO2 using raw data. The fluctuations in the data occurred due to the sensitivity
of the back pressure regulator. A simple moving average (SMA) over a period of 150 seconds, which corresponds
to a cycle of approximately 0.11 PV of injected CO2, was applied to obtain a smoother plot (Figure 5b).

Conclusions
As expected, a higher surfactant concentration led to a reduction of residual oil saturation. This can be attributed
to the increase in pressure difference between the inlet and outlet of the system. Surfactants also help to reduce
the interfacial tension between the the oleic and aqueous phases (Behzadi & Towler, 2009). The effect of
increasing the pressure difference and reducing interfacial tension leads to an increase in capillary number. The
capillary number is computed based on permeability, pressure drop, IFT, and effective length. The relationship is:
∆
= (1)
Capillary number is used in an empirical model introduced by CMG. The model (Equation 6) is based on
applying a factor to modify gas relative permeability in the presence of foam (CMG, 2009). This factor,
Modification by Foam (FM), depends on a mobility reduction factor (MRF) which is a ratio of the pressure drop
observed across a core during the simultaneous injection of surfactant solution and CO2, and that obtained during
the co-injection of brine and CO2; surfactant concentration; oil saturation; water saturation; and capillary number.
( ) = ( ) (2)

= (3)

The factors f1, f2, f3, and f4 depend on the following:
f1 Surfactant concentration
f2 Oil saturation
f3 Water saturation
f4 Capillary number
Therefore, the next step in this study is to develop a methodology that will allow the injection of a lower
surfactant concentration to achieve a significant modification in gas relative permeability in the presence of foam
that is comparable to that obtainable in a higher surfactant concentration. This will be executed by modifying and
controling parameters that can be changed within the CMG model.

Nomenclature
A Cross-sectional area of core
cs Surfactant concentration
FM Modification by foam
k Permeability
krg(f) Gas relative permeability in the presence of foam
krg(nf) Gas relative permeability with no foam
L Length of core
MRF Mobility reduction factor
Nca Capillary number
SPE 150634 5

∆p Pressure difference
∆pf Pressure difference for foam injection
∆pnf Pressure difference for no-foam injection
so Oil saturation
sw Water saturation
σ Interfacial tension
µ Viscosity of injected fluid

References
Behzadi, S., & Towler, B. (2009). A New EOR Method. SPE Annual Technical Conference and Exhibition. New
Orleans, LA: SPE.
CMG. (2009). User's Guide STARS. Calgary, Canada: Computer Modelling Group.
Coombe, D., Shrivasta, V., Singhal, A., & Belgrave, J. (1997). Numerical Simulation of Foam Flooding for Sweep
Improvement. Annual Technical Meeting. Calgary, Alberta: Petroleum Society of Canada.
Heller, J. (1994). CO2 Foams in Enhanced Oil Recovery. Advances in Chemistry, 242.
Kulkarni, M. (2003). Immiscible and Miscible Gas-Oil Displacement in Porous Media. Master's Thesis, Louisiana
State University, Baton Rouge, LA.
Liu, Y., Grigg, R., & Svec, R. (2005). CO2 Foam Behavior: Influence of Temperature, Pressure and Concentration
of Surfactant. SPE Production and Operations Symposium. Oklahoma City, OK: Society of Petroleum
Engineers.
Moradi-Araghi, A., Johnston, E., Zomes, D., & Harpole, K. (1997). Laboratory Evaluation of Surfactants for CO2-
Foam Applications at the South Cowden Unit. International Symposium on Oilfield Chemistry. Houston,
TX: Society of Petroleum Engineers.
Shetty, S. (2011). Evaluation of Simultaneous Water and Gas Injection Using CO2. Master's Thesis, Louisiana
State University, Baton Rouge, LA.
Vikingstad, A. K. (2006). Static and Dynamic Studies of Foam and Foam-Oil Interactions. PhD Dissertation,
University of Bergen, Bergen, Norway.
Wassmuth, F., Schramm, L. L., Mannhardt, K., & Hodgins, L. (2000). Scale-up Evaluations and Simulations of
Mobility Control Foams for Improved Oil Recovery. In L. L. Schramm, Surfactants: Fundamentals and
Applications in the Petroleum Industry (pp. 251–292).
Yaghoobi, H. (1994). Laboratory Investigation of Parameters Affecting CO2-Foam Mobility in Sandstone at
Reservoir Conditions. SPE Eastern Regional Meeting. Charleston, WV: Society of Petroleum Engineers.

Appendix

Table 1: Properties of core used in experiments

Length, cm 7.65
Diameter, cm 2.56

Permeability, mD (Measured
before first set of experiments) 13.15
Permeability, mD (Measured
before second set of
experiments) 11.32
Porosity, % 17.4
Pore Volume, cm3 6.84
6 SPE 150634

Table 2: Properties of fluids

Fluid Viscosity, cp Density, g/cc Density, g/cc
24°C, 2500 psi 24°C, 2500 psi 24°C, 14.7 psi
Decane 1.0256 0.7388 0.7273

2% CaCl2 Brine 1.1 1.0124 1.009

CO2 0.0901 0.8948 0.0018

CS-1045 Surfactant 200 1.07

Table 3: Fluid saturation remaining in core

0.5 wt % 0.1 wt %
Initial oil saturation 0.76 0.74
Connate water saturation 0.24 0.26
Residual oil saturation after waterflood 0.33 0.33
Water saturation after waterflood 0.67 0.67
Residual oil saturation after co-injection 0.14 0.25
Surfactant solution + Brine saturation after co-
injection 0.37 0.57
CO2 saturation after co-injection 0.49 0.18
S
SPE 150634 7

Figure 1: Schematic
S of Laboratory
y Equipment Setup

Figure 2:
2 Compariso
on of oil reco
overy before
e CO2-surfacttant co-injec
ction
8 S
SPE 150634

Figure 3: Comparison
n of oil recovery for differrent surfacta
ant concentra
ations

Figure 4: Comparison
C of CO2 recov
very for diffe
erent surfacta
ant concentrrations
S
SPE 150634 9

Figure 5a: Comparison of

o raw pressure differenc ce between tthe 8-in (0.5-iin from core inlet) and 10
0-in (0.5-
in from core outlet) for diifferent surfa
actant conce
entrations

n of pressure
Figure 5b: Comparison e difference after applyin
ng moving av
verage