DOE-Surfactant-2006-Semi-Annual

Surfactant Based Enhanced Oil Recovery and Foam Mobility Control
3nd Semi-Annual Technical Report
Reporting Period Start Date: July 2005
Reporting Period End Date: December 2005
Principal Authors:
George J. Hirasaki, Rice University

Clarence A. Miller, Rice University
Gary A. Pope, The University of Texas
Date Report was Issued: February 2006
DE-FC26-03NT15406
Rice University
Department of Chemical Engineering, MS-362
6100 Main Street
Houston, TX 77005-1892
The University of Texas

Department of Petroleum Engineering
P.O. Box 7726
Austin, TX 78713-7726
INTERA, Inc.
9111A Research Blvd.
Austin, TX 78758
Surfactant Based Enhanced Oil Recovery and Foam Mobility Control, G.J. Hirasaki, C.A. Miller and G.A. Pope
DE-FC26-03NT15406, February 2006
DISCLAIMER
This report was prepared as an account of work sponsored by an

agency of the United States Government. Neither the United States Government
nor any agency thereof, nor any of their employees, make any warranty,
expressed or implied, or assumes any legal liability or responsibility for the
accuracy, completeness, or usefulness of any information, apparatus, product, or
process disclosed, or represents that its use would not infringe privately owned
rights. References herein to any specific commercial product, process, or
service by trade name, trademark, manufacturer, or otherwise does not
necessarily constitute or imply its endorsement, recommendation, or favoring by
the United States Government or any agency thereof. The views and opinions of
authors expressed herein do not necessarily state or reflect those of the United
States Government or and agency thereof.
2
ABSTRACT
Surfactant flooding has the potential to significantly increase

recovery over that of conventional waterflooding. The availability of a large
number of surfactant structures makes it possible to conduct a systematic study
of the relation between surfactant structure and its efficacy for oil recovery. A
mixture of two surfactants was found to be particularly effective for application in
carbonate formations at low temperature. The mixture is single phase for higher
salinity or calcium concentrations than that for either surfactant used alone. This
makes it possible to inject the surfactant slug with polymer close to optimal
conditions and yet be single phase. A formulation has been designed for a
particular field application. It uses partially hydrolyzed polyacrylamide for mobility
control.
The addition of an alkali such as sodium carbonate makes

possible in situ generation of naphthenic soap and significant reduction of
synthetic surfactant adsorption. The design of the process to maximize the
region of ultra-low IFT takes advantage of the observation that the ratio of soap
to synthetic surfactant is a parameter in the conditions for optimal salinity. Even
for a fixed ratio of soap to surfactant, the range of salinity for low IFT was wider
that that observed for surfactant systems in the literature. Low temperature,
forced displacement experiments in dolomite and silica sandpacks demonstrate
that greater than 95% recovery of the waterflood remaining oil is possible with
0.2% surfactant concentration, 0.5 PV surfactant slug, with no alcohol.
Compositional simulation of the displacement process demonstrates the role of
soap/surfactant ratio on passage of the profile through the ultralow IFT region,
the importance of a wide low IFT dependence on salinity, and the importance of
the viscosity of the surfactant slug.
Mobility control is essential for surfactant EOR. Foam is evaluated

to improve the sweep efficiency of surfactant injected into fractured reservoirs as
well as a drive fluid for ASP flooding.
UTCHEM is a reservoir simulator specially designed for surfactant

EOR. It has been modified to represent the effects of a change in wettability.
3
TABLE OF CONTENTS
DISCLAIMER...................................................................................................................2
ABSTRACT......................................................................................................................3
TABLE OF CONTENTS...................................................................................................4
TABLE OF TABLES.........................................................................................................5
TABLE OF FIGURES.......................................................................................................5
INTRODUCTION .............................................................................................................8
EXECUTIVE SUMMARY .................................................................................................9
Task 1 Improved surfactants and formulations ..............................................................11
1.1. Identifying and Synthesizing Improved, Cost-effective Surfactants............11
1.2. Surfactant Tailoring for Crude Oils and Phase Behavior.............................12
1.2.1 Phase Behavior Results ..................................................................12
1.2.2 Polymer Results ..............................................................................13
1.3 Propoxylated-sulfated surfactants ................................................................16
1.4 Calcium tolerance of NI surfactant blend ......................................................19
Task 2 Phase behavior, adsorption, and composition changes during
displacement..................................................................................................................20
2.1 Surfactant Adsorption....................................................................................20
Low surfactant adsorption domain............................................................20
Surfactant adsorption with different soaps................................................23
Conclusions..............................................................................................26
Reference:................................................................................................26
2.2 IFT measurement and ultra-low IFT region ..................................................27
Correlation between phase behavior and IFT .........................................37
2.3 Characteristics of Alkali-Surfactant-Polymer process ....................................42
2.4 Alkaline-Surfactant Polymer Forced Displacement .......................................48
Dolomite sand pack..................................................................................48
Silica sand pack........................................................................................53
Polymer/Surfactant phase separation.......................................................55
Conclusions..............................................................................................58
Task 3 Foam for Mobility Control ...................................................................................59
3.1 Bulk foam in fractures ......................................................................................59
Experimental technique ..........................................................................59
4
Theory ....................................................................................................61
Results and discussion ...........................................................................62
3.2 Foam drive for ASP process............................................................................63
3.3 Foam stability with the presence of residual oil................................................66
Reference ...............................................................................................68
Task 4: Simulation of Field-Scale Processes ................................................................69
4.2: Wettability alterations in naturally fractured reservoirs................................69
Wettability alteration model in UTCHEM ..................................................69
References ...............................................................................................71
TABLE OF TABLES
Table 1.1-1Surfactants Tested in Year 3 .......................................................................11
Table 1.2-1 Formulations Screened with Cedar Hills Crude Oil, 104°C .........................13
Table 2.3-1 Other major parameters for the example ....................................................43
Table 2.4-1 Formulation for the alkaline surfactant polymer solution.............................50
TABLE OF FIGURES
Figure 1.2-1. Viscosity of Flopaam®3330S Polymer.....................................................14
Figure 1.2-2. Polymer Permeability Reduction Factor vs. Permeability for Midland
Farms Core, 38oC ..........................................................................................................15
Figure 1.3-1. Test results of salinity scans for 3- and 7- propoxylated-sulfated
surfactants .....................................................................................................................17
Figure 1.3-2. Optimal salinity vs. Alkane Carbon Number (ACN)...................................18
Figure 1.1-1 Phase diagram of NI blend ........................................................................19
Figure 2.1-1 Adsorption of 4:1 N67:IOS on calcite in varying salinity and alkalinity.......21
Figure 2.1-2 Adsorption of 4:1 N67:IOS on calcite in varying alkalinity at 5% NaCl.......21
Figure 2.1-3 the contour of maximal adsorption for N57 IOS(4:1) .................................22
Figure 2.1-4 Adsorption of N67 on calcite powder .........................................................22
Figure 2.1-5 Adsorption of IOS on calcite powder..........................................................23
Figure 2.1-6 Adsorption of 4:1 N67:IOS on calcite with sodium oleate ..........................25
Figure 2.1-7 Adsorption of 4:1 N67:IOS on calcite with sodium naphthenates ..............25
Figure 2.2-1 Phase behavior of 0.2% NI blend / 1% Na2CO3 / 2% NaCl (23 days
settling) ..........................................................................................................................27
5
Figure 2.2-2 Spinning drop measurement of 0.2 % NI blend/1% Na2CO3/2%

NaCl, 4 hours’ settling. There are three layers...............................................................27
Figure 2.2-3 IFT of 0.2% NI blend / 1% Na2CO3 / 2% NaCl with different settling
time ................................................................................................................................28
Figure 2.2-4 Photos of spinning drop of IFT of 0.2% NI blend / 1% Na2CO3 / 2%
NaCl ..............................................................................................................................29
Figure 2.2-5 Photo of two different spinning drops of 0.2% NI blend / 1% Na2CO3
/ 2% NaCl ......................................................................................................................29
Figure 2.2-6 IFT of 0.2% NI blend / 1% Na2CO3 / 2% NaCl with different settling
time ................................................................................................................................30
Figure 2.2-7 Comparison of phase appearance of 0.2% NI / 1% Na2CO3 / x %
NaCl at different time .....................................................................................................31
Figure 2.2-8 Photos of spinning drop of 0.2% NI blend / 1% Na2CO3 / 3.4% NaCl
with all the oil-rich emulsion ...........................................................................................31
Figure 2.2-9 Photos of spinning drop of 0.2% NI blend / 1% Na2CO3 / 3.4%
NaCl; Remove most of the oil-rich emulsion (P=5.1) ....................................................32
Figure 2.2-10 Phase behavior of 0.2% NI blend/1% Na2 CO3; (24 hours mixing, 4
hours settling) ................................................................................................................34
Figure 2.2-11 IFT of 0.2% NI blend/1% Na 2CO3 /0%NaCl as a function of time.............35
Figure 2.2-16 IFT of 0.2% NI blend/1% Na2CO3 /5%NaCl as a function of time.............36
Figure 2.2-17 IFT change with salinity for 0.2NI-1%Na2CO3 /WOR=3............................37
Figure 2.2-18 Phase behavior after 2 months with 0.2% and 0.5% NI blend. 38
Figure 2.2-19 Solubility ratios of 0.2% NI blend/1% Na2CO3 /NaCl 38
Figure 2.2.20 Comparison of IFT measured by spinning drop measurements
using standard procedure and that estimated by Chun-Huh correlation of 0.2% NI
blend/1% Na2 CO3/NaCl. c=0.3..................................................................... .................39
Figure 2.2-21 Solubility ratios of 0.5% NI blend/1% Na2CO3 /NaCl ...............................40
Figure 2.2-22 IFT estimated by Chun-Huh correlation of 0.5% NI blend/1%
Na2CO3 /NaCl, c=0.3. .....................................................................................................40
Fig. 2.2-23 Phase behavior of 0.2% NI blend/1% Na2CO3/x% NaCl, 40 days of
settling. Thin creamed oil-rich layers exist at salinity of 2-3.4% NaCl. ...........................41
Figure 2.3-1 Contour of interfacial tension with wide low IFT region..............................42
6
Figure 2.3-2 Contour of interfacial tension with narrow low IFT region ..........................43
Figure 2.3-3 Comparison of profiles between wide and narrow low IFT regions............44
Figure 2.3-4 Oil recoveries vs. injecting brine salinities .................................................45
Figure 2.3-5 Comparison of profiles between varied injecting solution viscosities.........46
Figure 2.3-6 Oil Fractional Flow vs. Saturation at IFT=0.001dyne/cm ..........................47
Figure 2.4-1 Photos of oil flooding and water flooding...................................................49
Figure 2.4-2 Oil Recovery of Water Flooding in Dolomite Sand Pack ............................49
Figure 2.4-3 Photos of dolomite pack at different injecting pore volumes......................51
Figure 2.4-4 Oil Recovery of ASP Flooding in Dolomite Sand Pack ............................51
Figure 2.4-5 Effluent of ASP Flooding in Dolomite Sand Pack .....................................52
Figure 2.4-6 History of pressure drop for dolomite pack ................................................52
Figure 2.4-7 Comparison between simulation and experiments for dolomite pack ........52
Figure 2.4-8 Photos of silica sand pack at different injecting pore volumes...................53
Figure 2.4-9 Oil Recovery of ASP flooding in silica sand pack ......................................54
Figure 2.4-10 History of pressure drop for silica pack....................................................54
Figure 2.4-11 Comparison between simulation and experiments for silica pack............55
Figure 2.4-12 Photos of silica sand pack at different injecting pore volumes when
injecting salinity = 4.0% NaCl.........................................................................................56
Figure 2.4-13 Pressure difference vs. pore volume when polymer/surfactant
separation, 4% NaCl......................................................................................................56
Figure 2.4-14 Phase behaviors of different ASP solutions after one week ....................57
Figure 3.1-1 Detailed fracture model .............................................................................60
Figure 3.1-2 Diagram for experiment in fracture model..................................................61
Figure 3.1-3 Bulk foam apparent viscosity in fractures; measurement and
prediction. K=1.01 for Mooney and Krieger and Dougherty equations...........................63
Figure 3.2-1 Foam sweep of the sand pack presaturated with polymer/surfactant ........64
Figure 3.2-2 Apparent viscosity during the sweeping the sand pack presaturated
with polymer/surfactant ..................................................................................................65
Figure 3.3-1 Short sand pack.........................................................................................66
Figure 3.3-2 Comparison of foam strength with or without residual oil by different
surfactants .....................................................................................................................67
7
INTRODUCTION
Oil recovery by primary depletion and waterflooding, on the average,

recovers only about one third of the original in place. The remaining oil can be
categorized into: (1) the residual oil in the regions swept by water and (2) the
movable oil in the regions unswept or poorly swept by water. This project uses
surfactants to reduce the residual oil saturation by both interfacial tension
reduction and wettability alteration, the latter in cases where wettability is
responsible for retaining oil in the matrix. A factor in the sweep efficiency of a
reservoir is the mobility ratio between the resident fluids and the injected fluids.
Polymer solution is the traditional method for mobility control in surfactant
flooding. This project will evaluate foam as an alternate or supplement to
polymer for mobility control. Our objective is to economically increase the
recovery efficiency beyond that achieved by waterflooding.
Both unfractured and fractured formations will be addressed in this project.
The driving force for displacement of oil in unfractured systems is primarily the
pressure gradient developed by displacing fluids from the injection well to the
production well. This pressure gradient may be only a small contributor in
fractured formations. In this case, spontaneous imbibition is needed to exchange
the injected fluid and oil between the fracture and matrix. The driving force for
spontaneous imbibition includes capillary pressure gradients and buoyancy, or
gravity drainage. The contribution due to capillary pressure gradients may be
diminished because of low interfacial tension.
Both sandstone and carbonate formations will be considered. Carbonate
formation usually tend to be more oil-wet and fractured compared to sandstone
formations. In either case, surfactant adsorption on the mineral surfaces must be
minimized. Sodium carbonate is used with anionic surfactants in carbonate
formations to reduce adsorption. The alkalinity of the sodium carbonate also
generates surfactants in situ by reacting with the naphthenic acids in the crude
oil.
Scale-up from the laboratory to the field is a necessary part of developing
an enhanced oil recovery process. The tool for this scale-up is the reservoir
simulator, UTCHEM.
8
EXECUTIVE SUMMARY
Twenty four surfactants are compared for their efficacy for oil recovery by
surfactant flooding. Surfactant structure – performance relationships are needed
for applications with a specified crude oil composition, brine salinity, reservoir
temperature, formation mineralogy, and recovery mechanism. The surfactants
are characterized by the optimal salinity for different pure hydrocarbon oils, the
solubilization ratio, which is an estimator of the level of interfacial tension at
optimal conditions, and whether it forms viscous gel or liquid crystalline phases
that cause slow emulsion coalescence. A combination of two surfactants, N67-
7PO-S and IOS 15-18, was found to be particularly effective. N67-7PO-S has a
moderately branched hydrophobe with 16-17 carbons, an average of 7 PO
(propylene oxide) groups, and is sulfated. IOS 15-18 is an internal olefin
sulfonate with 15-18 carbons. The location of the sulfonate in the IOS is
distributed along the carbon chain and thus the result is a twin-tailed or branched
hydrophobe. The branching reduces the tendency to form gels and viscous
emulsions at low temperatures. EO and PO groups impart tolerance to divalent
ions. PO is more lipophlic than the hydrophilic EO group and results in a lower
optimal salinity requirement. The sulfate has an ester linkage and is subject to
hydrolysis at high temperatures and low pH. Thermally stable sulfonates are
evaluated for high temperature applications.
A surfactant-polymer formulation is being developed for a West Texas
carbonate reservoir that has a pressure too low for CO2 flooding. The formation
has anhydrite, which will result in precipitation of sodium carbonate. The
formulation has recovered up to 95% of the oil remaining after waterflooding in
reservoir formation core material. The project team has met with the operator
and partners to plan for a field test.
An alkaline surfactant process is also being developed for enhanced
recovery in oil-wet, carbonate formations. The carbonate ion of sodium
carbonate is a potential determining ion in carbonate formations such as calcite
and dolomite. Alteration of the mineral surface to a negative charge aids in the
wettability alteration and makes a dramatic reduction in the adsorption of anionic
surfactants. Calcium ion concentration is sequestered because of the low
solubility product of calcium carbonate. Also the alkali raises the pH, which
results in saponification of naphthenic acids to naphthenic soap, a natural
surfactant. The naphthenic soap is usually too lipophilic by itself and addition of
a synthetic surfactant is needed. Ultra-low interfacial tensions are possible at
synthetic surfactant concentrations as low as 0.05%. However, the system is
complex because it is a mixture of naphthenic soap and synthetic surfactants
with very different properties. This results in optimal salinity that depends on the
water/oil ratio and surfactant concentration. However, these dependencies can
be correlated by the ratio of soap/synthetic surfactant. It was discovered that
even for a fixed soap/surfactant ratio the range of salinities over which low IFT (<
10-2 mN/m) occurs is much wider than that expected for conventional
akylarylsulfonate surfactants. Also, the changing ratio of soap/surfactant during
9
oil displacement can be utilized to have the composition pass through the low IFT
region at the displacement front. This results in a robust recovery process that is
not overly sensitive to salinity and surfactant concentration. ASP formulations
with 0.5 PV surfactant slug and low surfactant concentrations have recovered
over 95% of the waterflood remaining oil from dolomite (0.2% surfactant) and
silica (0.5% surfactant) sandpacks. This was possible at room temperature
without need for alcohol.
Surfactant retention by adsorption and phase trapping determines the
amount of surfactant required for a surfactant enhanced oil recovery process.
We show that the adsorption of anionic surfactants on calcite and dolomite can
be reduced by an order of magnitude by addition of sodium carbonate.
Mobility control is recognized as an essential element of surfactant EOR.
Surfactant injection into fractured formations imposes a severe challenge for
reservoir conformance or sweep efficiency. It was shown earlier that foam
improves sweep in fracture systems. The apparent viscosity of bulk foam in
fractures is evaluated in this report. The possibility of substituting the polymer
drive with a foam drive is examined.
The reservoir simulator, UTCHEM will be used as the tool to scale-up from
laboratory experiments to field design. An approach to model the change in
wettability is presented.
10
Task 1. Improved Surfactants and Formulations
Subtask 1.1. Identifying and Synthesizing Improved, Cost-effective

Surfactants
The research conducted during Year 1 and Year 2 identified several families of
propoxylated sulfate surfactants and also several sulfonate surfactants of interest
for enhanced oil recovery. Since the sulfate surfactants are not stable for long
time periods due to hydrolysis at temperatures above about 60o C, we tested
several sulfonate surfactants with crude oils from reservoirs at high temperature.
These surfactants included several branched alpha-olefin sulfonates (AOS)
provided by Stepan, a Shell internal olefin sulfonate (IOS), linear and branched
alkyl benzene sulfonates (ABS), a twin tailed sulfonate and a napthene sulfonate.
The sulfonate surfactants were also used in combination with cationic surfactants
and co-solvents to identify a suitable surfactant formulation. Table 1.1-1 lists the
surfactants tested during the first half of Year 3.
Table 1.1-1 Surfactants Tested in Year 3
Descriptive or Trade Name Abbreviated Chemical Formula

(Supplier) (b = branching in the carbon chain)
bC11-CH(OH)-CH2 -CH2 -SO3- (~75%)
C14 Alpha Olefin Sulfonate (AOS)
bC11-CH=CH-CH2-SO3- (~25%)
C16 o-Xylene Sulfonate C16-(C8 H12)-SO 3- where C8H12 = o-xylene
bC13-15-CH(OH)-CH2 -CH2-SO 3- (~75%)
C16-18 Alpha Olefin Sulfonate (AOS)
bC13-15-CH=CH-CH2-SO3- (~25%)
bC17-21-CH(OH)-CH2 -CH2-SO 3- (~75%)
C20-24 Alpha Olefin Sulfonate (AOS)
bC17-21-CH=CH-CH2-SO3- (~25%)
R-CH(OH)-CH2 -CH(SO3 -)-R’ (~75%)
C15-18 Internal Olefin Sulfonate (IOS) R-CH=CH-CH(SO3-)-R’ (~25%),
where R+R’ = C12-15
Polystep® A-16-22 (Branched Alkyl
bC11-13(C 6H5)-SO 3-
Benzene Sulfonate - ABS)
Linear Alkyl Benzene Sulfonate (L
C16H33(C6 H5 )-SO3 -
ABS)
BLANCOL® LIQUID (Sodium
C10H8 -SO3
Naphthalene Sulfonate)
Dowfax 3B2 Sulfonate (bis-n-decyl-
C10H21(C 6 H4 -SO3)-(C6 H4 -SO3 )C10H21
diphenyl-disulfonate)
11
Subtask 1.2. Surfactant Tailoring for Crude Oils and Phase Behavior
Surfactant tailoring for crude oils in Year 3 was conducted in the same
systematic way as reported in the Year 2 Annual Report. Phase behavior
experiments were used to narrow possible surfactant candidates in an efficient
and effective manner. These surfactants were selected based upon the design
criteria reported in Year 1 and 2. Polymers and mixtures of surfactants and
polymers were also tested to select suitable polymers for further testing. These
processes were integrated into ongoing core flood experiments to evaluate the
best surfactants for oil recovery for the particular crude oils provided by oil
companies.
1.2.1 Phase Behavior Results

The selection of a high-temperature surfactant formulation introduces new
challenges to the phase behavior experiments. The crude oil selected for these
experiments was from the Burlington Resources Cedar Hills (CH) field in
Montana and Wyoming. The reservoir temperature for Cedar Hills is 104oC. Due
to the high temperature a limited number of new surfactants were available for
screening with this crude oil.
Surfactant formulations were screened for several characteristics during the

phase behavior tests. Different mixtures are first tested for solubility and aqueous
stability. The surfactants are then mixed with brine and crude oil to observe their
phase behavior at reservoir temperature. In particular, the time required for the
samples to reach equilibrium is important. Also, the samples are observed to see
how viscous the phases are and whether macroemulsions form and how long it
takes for such macroemulsions to break. The solubilization ratios for the
equilibrium microemulsions are measured for those formulations that reach
equilibrium in a reasonable length of time. The viscosity of the equilibrium
microemulsion is measured for some of the most promising formulations. If
needed, more samples are made with higher alcohol co-solvent added to speed
up the equilibration and reduce microemulsion viscosity. The results of some of
the Cedar Hills phase behavior experiments are presented in Table 1.2-1.
12
Table 1.2-1. Formulations Screened with Cedar Hills Crude Oil, 104°C
Optimum
Optimum Equilibration
Salinity S*
Surfactant Formulation Sol. Ratio, Time at S*
(wt%
σ(cc/cc) (days)
NaCl)
1% C15-18 IOS, 1% C20-24 AOS, 4% SBA, 1%
5 5 1
Na2CO 3
1% C15-18 IOS, 1% C20-24 AOS, 4% SBA 6 4 1
2% C20-24 AOS, 4% SBA 1-2 3-5 1
1% C16 o-Xylene sulfonate , 1% C 20-24 AOS, 4%

1-2 5 1
SBA
2% C15-18 IOS, 4% SBA >10 2 1
2% C16-18 AOS, 4% SBA 10 2 1
Did not Did not Did not

1% C16 L ABS, 1% C15-18 IOS, 4% SBA
equilibrate equilibrate equilibrate
1% C20-24 AOS, 1% C11-13 ABS, 4% SBA >4 1 1
1% C20-24 AOS, 1% Dowfax 3B2, 4% SBA >4 0.5 1
1% C20-24 AOS, 1% Naphthalene Sulfonate, 4%

3.5 5 1
SBA
Several phase behavior experiments provided comparable results in terms of

equilibration time and solubilization ratio. The best results were obtained with a
formulation containing the C20-24 AOS, C15-18 IOS and SBA. In spite of the
elevated temperatures used in this experiment, the use of SBA as a co-solvent
was necessary to achieve rapid coalescence. The addition of Na2CO3 improved
the phase behavior indicating the crude oil is reactive. Linear ABS surfactant
formulations did not reach equilibrium whereas the branched ABS formulations
reached equilibrium but had unacceptably low solubilization ratios.
1.2.2 Polymer Results
The Second Annual Report presented information on two types of commercially

available water-soluble polymers for EOR: xanthan gum and partially hydrolyzed
polyacrylamide (HPAM). In this report, we present additional data on
Flopaam®3330S (3330S), a HPAM manufactured by SNF Floerger. This product
is 25-30% hydrolyzed and has a molecular weight of approximately 8 million
Daltons.
13
A major focus of the work performed during the first half of Year 3 has been the
application of Flopaam®3330S in a heterogeneous, high salinity environment in
the presence of anhydrite (CaSO4 ). HPAMs are particularly sensitive to divalent
cations such as Ca ++ and Mg++. Figure 1.2-1 shows an example of this
dependence for a particular field water called SSR with and without CaCl2 added
to it and a comparison with deionized water with NaCl added to it.
Viscosity of Flopaam 3330 at various salinities

Shear rate = 3.23 sec-1 unless stated, 38 C
30.0
SSR
25.0
SSR + 1375 ppm CaCl2
20.0 4% NaCl in DI (69s^-1) SSR:
Viscosity (cp)
10 ppm Ca++
15.0 340 ppm Mg++
270 ppm Na+
39 ppm CO3--
10.0
398 ppm HCO3-
900 ppm SO4--
5.0
497 ppm Cl-
0.0
0 500 1000 1500 2000 2500 3000
Polymer Concentration (ppm)
Figure 1.2-1. Viscosity of Flopaam®3330S Polymer
For an efficient chemical flood, the mobility ratio should be less than one. The
mobility ratio of a surfactant-polymer slug displacing an oil bank can be
approximated by the equation:
 1 
 
 w RF 
M 
k ro k rw 
  
o w  min
where the resistance factor is given by


RF Rk p
w
The resistance factor can be calculated from the measured pressure drop across
a core for brine (Pw) and then the surfactant-polymer slug (Pp) at the same
flow rate (q) and water saturation (Sw).
14
Rf = (Pp /Pw)q, Sw
The permeability reduction factor (Rk) can then be calculated using the viscosity
data as follows:
Rk = Rf / (p/ w)
The value of the permeability reduction factor Rk for a given polymer and rock
type is a function of permeability among other variables. The permeability
reduction factor has been measured for Flopaam®3330S in a very
heterogeneous dolomite reservoir rock and is plotted in Figure 1.2-2. Such data
are very important for designing a stable surfactant-polymer flood (one with a
mobility ratio less than one) and are essential input for simulators such as
UTCHEM. In UTCHEM, the permeability reduction factor is modeled as a
function of permeability and other variables using a correlation with two
adjustable parameters. The model curve for this case is shown in Figure 1.2-2.
Most data in the literature are for sandstones, so it was very important to
measure the data for this reservoir dolomite and determine that the model was
valid for this reservoir rock, which is very different and much more
heterogeneous than typical sandstone.
100
Lab Data
UTCHEM Correlation
Flopaam 3330S
Temp = 38 C
Rk
10
1
0 100 200 300 400 500 600 700 800
Permeability (md)
Figure 1.2-2. Polymer Permeability Reduction Factor vs. Permeability for

Midland Farms Core, 38oC
15
Subtask 1.3 Propoxylated-sulfated surfactants
Figure 1.3-1 depicts the test results of salinity scans without

alcohol addition for 3- and 7- propoxylated-sulfated surfactants made from a
C16/17 controlled methyl-substituted alcohol1, (Neodol 67 or N67). These
surfactants are called N67-3PO and N67-7PO.
Tests were conducted at WOR=1 in borosilicate 10-cc pipettes.

Each pipette contained 3% surfactant, pure n-alkanes and brines of 1% NaCO3
and X% NaCl . The pipettes were sealed with a torch and samples were rotated
end-to-end for at least 24 hours and then left to stand quiescent. Samples
reported as studied at 78°F were kept in a laboratory room without strict
temperature control. Equilibration at 78°F was achieved after 3½ months.
Addition of secondary butyl alcohol could accelerate time of equilibration.
However, alcohol addition would reduce oil solubilization, thus impeding realistic
evaluation of surfactant potential.
Test results depicted in Figure 1.3-1 indicate that N67-7PO is

more hydrophobic and solubilized more n-decane oil than N67-3PO. Volumes of
oil and water solubilized per volume of surfactant (V/Vs) at the optimum was ~19
for 7PO and ~10 for 3PO. Because of these results, the N67-7PO was selected
as the surfactant with more potential for application.
1
JOURNAL OF SURFACTANTS AND DETERGENTS, VOL. 7, NO. 4 (OCTOBER 2004), page 319-
328
16
35 35
30 30
25 25
20 20
Vw/Vs
Vo/Vs
PO7
15 15
PO3
10 10
5 5
0 0
1 2 3 4 5 6 7 8 9
% NaCl +1% Na 2 CO3
Figure 1.3-1. Test results of salinity scans for 3- and

7- propoxylated-sulfated surfactants made from a
C16/17 controlled methyl-substituted alcohol.
3%N67-7PO and –3PO, WOR=1, n-C10, 78°F.
Figure 1.3-2 is a plot of optimal salinity vs. Alkane Carbon Number

(ACN) for the two surfactants; the number in parenthesis indicates solubilization
parameter at optimal conditions. What Figure 2 depicts is typical for optimal
salinities in a homologous series. Optimal salinities of many surfactants start to
converge for low molecular weight n-alkane oils2 . It was observed that PO3
produced a birefringent middle phase against n-decane while some birefringence
was present in PO7 against n-dodecane (V/Vs ~ 10). We have investigated PO7
samples at higher temperatures up to 140°F and failed to observe any
birefringence. When increasing temperature, we were able to clearly determine
from well-equilibrated samples that the optimal salinity decreased, or lipophilicity
increased. For example, an under-optimum microemulsion at 3.7 % NaCl and
78°F became an over-optimum microemulsion at 140°F. Because of lack of
complete separation of phases present at equilibrium after two months of
monitoring samples, we were unable to precisely determine optimal conditions at
2
Puerto Maura C., Reed Ronald L, “A Three Parameter Representation of Surfactant-
Oil-Brine Interaction”, Society Of Petroleum Engineers Journal , May 1990, page 198-
204.
17
140°F. We hypothesize that the appearance of birefringence as the oil molecular

weight increased is a manifestation of the surfactant lipophile not being long-
enough to solubilize enough oil to have all the surfactant exist as microemulsion.
For PO-3, birefringence started at n-decane, whereas for PO-7 it started at n–
dodecane. Because the propoxylated surfactants become more lipophilic when
raising test temperature, which should lead to greater solubilization, the absence
of birefringence at140°F is consistent with validate this hypothesis. This matter
needs further investigation.
7
(~10)
Optimal Salinity
6
5
4 (~10)
3
(~19)
2
PO 3
1 PO7
0
3 4 5 6 7 8 9 10 11 12 13
ACN
Figure 1.3-2. Optimal salinity vs. Alkane Carbon Number (ACN). 3%N67-
7PO and –3PO, WOR=1
18
Subtask 1.4 Calcium tolerance of NI surfactant blend
Blends of N67-7PO and IOS15-18 (called NI blend) can not only provide
good oil recovery as discussed in subtask 2.4, but also improve the tolerance of
surfactant to calcium.
Figure 1.4-1 shows that when N67-7PO and IOS were used separately, the
concentration of calcium chloride for phase separation occurred at 0.5% and
0.1%, respectively. The phase separation behaviors of the two surfactants were
different: IOS was precipitated by calcium, while N67-7PO formed cloudy
solutions, which ultimately separate into a viscous surfactant-rich phase and
brine, the former being more dense. However, when they were mixed at different
weight ratios, they stayed in the single-phase region over a much wider calcium
range than for IOS or N67-7PO alone. For 4:1 ratio which we used for forced
displacement experiments described later, the upper limit concentration of CaCl2
with clear solutions is 1.0%.
0.5% N67-7PO&IOS, 2% NaCl

2.5%
Phase Separation
Precipitation
2.0%
Clear
Concentration of CaCl2
1.5%
Multi Phase Region
1.0%
0.5%
Single Phase Region
0.0%
IOS 4:1 2:1 1:1 1:2 1:4 1:9 N67-7PO
Figure 1.4-1 Phase diagram of NI blend

Conclusion
1. Mixtures of N67-7PO and IOS have higher calcium tolerance than either used
separately.
19
Task 2 Phase behavior, adsorption, and composition changes during

displacement
Subtask 2.1 Surfactant Adsorption

Surfactant adsorption is crucial to surfactant recovery processes including the
alkali-surfactant process. Both alkali concentration and salinity influence the
surfactant adsorption significantly. The presence of Na2CO3 can reduce the
surfactant adsorption. However, higher salinity causes higher surfactant
adsorption and counteracts the adsorption reduction by alkali. Thus, it is better
to implement the alkali-surfactant process at lower salinity range. The effect of
soap on synthetic surfactant adsorption was also tested. No synthetic surfactant
adsorption reduction was found by using sodium oleate or sodium naphthenates
from Fisher Scientific.
Low surfactant adsorption domain

Figure 2.1-1 shows the adsorption of synthetic surfactant (4:1 N67:IOS; also
called NI blend) on calcite powder for different salinity and alkalinity. At same
alkalinity, we can find the adsorption increases with the salinity. By adding
Na2CO3 , the adsorption can be reduced. However, this adsorption reduction
effect will be weakened under higher salinity. At 5% NaCl, the adsorption is
reduced to 2.5*10-3 mmol/m 2 by adding 1.21%Na2 CO3, while the adsorption at 0%
NaCl is 0.2*10-3 mmol/m2 with the presence of 1% Na2 CO3. These results
indicate that it’s better to implement the alkali-surfactant process at low salinity
(<3%) because low surfactant adsorption can be achieved at this condition.
Figure 2.1-2 illustrates the adsorption isotherms at 5% NaCl. It shows that only
0.178% Na2 CO3 is needed for the adsorption reduction effects. Adding Na 2CO3
more than 1.2% Na2CO3 does not further reduce adsorption. As shown in the
previous annual report, only 0.1% Na2 CO3 is needed for the adsorption reduction
at 0% NaCl. Figure 2.1-3 is the contour of maximal adsorption for N67:IOS (4:1)
by summarizing all the adsorption data. This plot shows the domain of low
surfactant adsorption is: [Na 2 CO3]>0.2% and [NaCl]<3%. The alkali-surfactant
process with the current surfactant should be performed in this concentration
domain to have low surfactant adsorption.
Figure 2.1-1, 2.1-2 and 2.1-3 are for the 4:1 N67 and IOS mixture. N67 and
IOS were also tested individually to verify the adsorption reduction effect by
Na2CO3 , as Figure 2.1-4 and 2.1-5 show. These two figures show that the
presence of Na2CO3 can reduce the adsorption for N67 and IOS respectively
significantly. The adsorption for N67 with Na2CO3 is 1/10 of that without Na2CO3,
while the adsorption for IOS with Na2CO3 is 1/7 of that without Na 2CO3.
20
Adsorption of 4:1 N67:IOS on Calcite in Varying Salinity and Alkalinity

5.0
4.5
2
Adsorption Density, 10 mmol/m
4.0
3.5
-3
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0.0 0.5 1.0 1.5 2.0
Residual Surfactant Concentration (mmol/L)
5% NaCl, 0% Na2CO3 5% NaCl with 1.21% Na2CO3

3% NaCl, 0% Na2CO3 3% NaCl, 1.21% Na2CO3
1% NaCl, 0% Na2CO3 1% NaCl, with 1.0% Na2CO3
0% NaCl, 0% Na2CO3 0% NaCl, with 1.0%Na2CO3
Figure 2.1-1 Adsorption of 4:1 N67:IOS on calcite in varying salinity and alkalinity
Adsorption of 4:1 N67:IOS on Calcite at 5% NaCl

5.0
4.5
2
mmol/m
4.0
3.5
-3
Adsorption Density, 10
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
5% NaCl, 0% Na2CO3 5% NaCl with 0.178% Na2CO3

5% NaCl with 0.404% Na2CO3 5% NaCl with 1.21% Na2CO3
Figure 2.1-2 Adsorption of 4:1 N67:IOS on calcite in varying alkalinity at 5% NaCl
21
Figure 2.1-3 the contour of maximal adsorption for N67:IOS(4:1)
2
Adsorption of N67 on Calcite (17.851 m /g)
4.0
2
3.5
3.0
-3
2.5
2.0
1.5
1.0
0.5
0.0
0.0 0.5 1.0 1.5 2.0
0% NaCl, 0% Na2CO3 0% NaCl, 1% Na2CO3
Figure 2.1-4 Adsorption of N67 on calcite powder
22
2
Adsorption of IOS on Calcite (17.851 m /g)
10.0
2
9.0
8.0
7.0
-3
6.0
5.0
4.0
3.0
2.0
1.0
0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
Figure 2.1-5 Adsorption of IOS on calcite powder
Surfactant adsorption with different soaps
If soap is present, the surfactant adsorption may change. Several soaps were
tested. These tests were done without sodium carbonate.
The first soap that was tested is sodium oleate. Sodium oleate solution having
a concentration of 1.4 mmol/L and different amounts of calcite powder (0.2~8g)
were mixed together. After one-day mixing, the residual solutions were titrated.
However, no soap can be detected. Meanwhile, a white paste material was
found at the air-liquid interface and on the side of the tube. This material seems
hydrophobic because it adheres to the polypropylene tube surface even after
centrifuging. Since the solubility product of CaCO 3 is 3.8 x 10 -9 and the solubility
product of calcium oleate is 3.98 x 10 -13 (from reference [1]), the following reaction
must happen during the mixing. The white material is calcium oleate.
According to the solubility product, the minimum Na2CO3 concentration that
restrains the Ca(oleate)2 precipitation can be calculated by equation (3) below.
For a 0.1% sodium oleate solution (3.29mmol/L), 1.1% Na 2CO3 is needed to
suppress the calcium ion concentration so that no Ca(oleate)2 will precipitate.
The experiment shows that no soap was detected with 1% Na2 CO3 and white
23
materials were found inside the tube. Only after adding 5% Na2CO3, no white
material was formed. The relevant reactions and equilibrium relations are:
2 Na  Oleate  Ca 2   Ca 
Oleate 
2 2 Na  (1)
2 2
Ca CO3  CaCO3 (2)
K spCaCO3
[CO3 2 ] [Oleate ]2 9500 * [Oleate ] 2 (3)
K spCa ( Oleate ) 2
When the sodium oleate and synthetic surfactant (4:1 N67:IOS) were mixed
together, the white material was found in the presence of calcite. According to the
solubility product calculation, the sodium oleate would be totally consumed and
the titrated residual surfactant concentration should be the concentration of
synthetic surfactant. With this assumption, the adsorption of the synthetic
surfactant was calculated. The presence of sodium oleate did not reduce the
synthetic surfactant adsorption as Figure 2.1-6 shows. In this plot, the
soap/synthetic surfactant ratio is 1:2 (mole ratio). The adsorption plateau does
not change when sodium oleate is present.
The other soap tested was the sodium naphthenates from Fisher Scientific.
For these sodium naphthenates, a small amount of white precipitation was also
observed when mixed with calcite powder. However, the amount of this
precipitation was less than what was observed in the sodium oleate experiments,
suggesting that calcium naphthenates are more soluble than calcium oleate.
Before the adsorption experiments, two samples were titrated. One is 0.1%
N67:IOS (4:1), and the other is 0.1% N67:IOS (4:1) with 0.05% sodium
naphthenates. The titration results for these two samples were quite close to
each other. The first sample was1.500.02 mmol/L, the second sample was
1.490.03 mmol/L. Thus, the presence of sodium naphthenates does not
change the titration results to the extent that the synthetic surfactant adsorption
can be measured. The surfactant adsorption does not change much with the
presence of sodium naphthenates as shown in Figure 2.1-7.
24
Adsorption of 4:1 N67:IOS and Na Oleate on Calcite

4.5
Adsorption Density, 10-3 mmol/m 2

4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0.0 0.5 1.0 1.5 2.0
No Na Oleate, 0% NaCl (1:2 mol)(Na Oleate: Surfactant), 0% NaCl
Figure 2.1-6 Adsorption of 4:1 N67:IOS on calcite with sodium oleate
Adsorption of N67IOS with sodium naphthenates on Calcite

5.0
2
4.0
-3
3.0
2.0
1.0
0.0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
NI Blend only (1:2 weight) (sodium naphthenates:NI Blend
Figure 2.1-7 Adsorption of 4:1 N67:IOS on calcite with sodium naphthenates
25
Conclusions
1. Both alkali concentration and salinity influence the anionic surfactant

adsorption significantly. The presence of Na2 CO3 can reduce the surfactant
adsorption. However, higher salinity increases surfactant adsorption and
counteracts the adsorption reduction by alkali.
2. The concentration domain with low surfactant adsorption for the current
surfactant is: [Na2CO3 ]>0.2% and [NaCl]<3%.
3. The presence of Na2CO3 reduces the adsorption for N67 and IOS respectively
significantly.
4. The presence of sodium oleate and sodium naphthenates does not reduce the
synthetic surfactant adsorption on calcite, in the absence of sodium carbonate.
Reference:
[1] Beneventi D, Carre B, Gandini A., Precipitation and Solubility of Calcium
Soaps in Basic Aqueous Media. J Colloid Interface Sci. 2001 May
1;237(1):142-144
26
Subtask 2.2 IFT measurement and ultra-low IFT region

Interfacial tension (IFT) is one of
the key factors in the alkali-surfactant
process. However, it’s difficult to
obtain reproducible results in our IFT excess oil
measurement because there is not
enough surfactant to generate enough
middle phase microemulsion phase
for IFT measurements. Also, an oil-rich
oil-rich emulsion was found in the emulsion
under optimum salinity
alkali-surfactant system as shown in
Figure 2.2-1. This is a creamed layer lower-phase
of emulsion between the excess microemulsion
crude-oil and lower phase
microemulsion. Since the color of
this emulsion is darker than the lower
phase microemulsion and it is less Figure 2.2-1 Phase behavior of 0.2% NI blend
dense than the lower phase
/ 1% Na2CO3 / 2% NaCl (23 days settling)
microemulsion, the oil concentration in
this emulsion is expected to be higher
than that in the lower phase aqueous middle
microemulsion so that this layer is called layer
“oil-rich emulsion”. The presence of this oil
oil-rich emulsion significantly changes the
IFT results.
The creamed middle layer can also be Fig. 2.2-2 Spinning drop measurement
observed during spinning drop interfacial of 0.2 % NI blend/1% Na2 CO3 /2% NaCl, 4
tension measurements. Fig 2.2-2 shows hours’ settling. There are three layers.
that during spinning drop tension measurement of 0.2 % NI blend/1% Na 2CO3 /2%
NaCl, there are three regions: creamed middle layer, oil and aqueous phase. As
shown in Fig 2.2-1, the volume of the creamed middle layer is too small to
accurately measure the density. In calculation of the interfacial tension, the
diameter of the oil phase, and the density difference between oil and the aqueous
solution were used.
Several experiments were done for the system which contains 0.2% NI blend,
1% Na2 CO3 and 2.0% NaCl and MY4 crude oil at WOR equals 3
(soap/surfactant=0.35). The samples were first equilibrated for 24 hours by
27
rotation. The excess oil and aqueous phase were sampled at different settling
times, and the dynamic IFT plot is shown as Figure 2.2-3. This plot indicates the
dynamic IFT is a function of settling time. A photo of a spinning drop after 2
hours settling sample is shown in Figure 2.2-4. It is difficult to see the oil drop
when its diameter is less than that of the oil-rich emulsion because the latter is
opaque. It was found that oil-rich emulsion was the most important factor for the
low tension as Figure 2.2-5 shows. There are two drops with different diameters
in Figure 2.2-5. The two drops came from a single large drop. The one with
more oil-rich emulsion has smaller diameter than the other with less oil-rich
emulsion. The smaller diameter drop on the left of Figure 2.2-5 is obscured by
the opaque oil-rich emulsion. Figure 2.2-6 shows that equilibrium low IFT could
be reached more quickly if the oil-rich emulsion was added to the spinning drop
tube.
1E+00
1E-01
IFT, dyne/cm
1E-02
1E-03
0 100 200 300 400 500 600 700
Time, minutes
1 hour's settling 2 hours' settling 4 hours' settling
8 hours' settling 24 hours' settling
Figure 2.2-3 IFT of 0.2% NI blend / 1% Na2CO3 / 2% NaCl with different settling time
28
Figure 2.2-4 Photos of spinning drop for IFT of 0.2% NI blend / 1% Na2CO3 / 2% NaCl.
The time is after start of drop spinning. The oil drop is on left. The oil-rich emulsion is to
the right of the drop.
Figure 2.2-5 Photo of two different spinning drops of 0.2% NI blend / 1% Na 2 CO3 / 2% NaCl
29
1E+00
IFT, dyne/cm 1E-01
1E-02
1E-03
1 hour's settling 2 hours' settling 4 hours' settling
2 hours settling, clear aqueous + oil-rich emulsion
1E-04
0 100 200 300 400 500 600 700
Time, minutes
Figure 2.2-6 IFT of 0.2% NI blend / 1% Na2 CO3 / 2% NaCl with different settling time
It was found that the settling time for oil-rich emulsion is different at different
salinities as shown in Figure 2.2-7. For the samples with salinity less than 3.0%
NaCl, the oil-rich emulsion settled between the excess oil and lower phase
microemulsion after 23 days settling. However, the oil-rich emulsion did not
settle for the 3.4% NaCl sample even after 23 days. By observing this sample in
the spinning drop apparatus, the oil-rich emulsion was found to surround the oil
drop and the diameter would decrease as shown in Figure 2.2-8. (The spinning
drop speed was reduced in the lower two photos.) Thus, the oil-rich emulsion
density for the 3.4% NaCl sample may be close to that of the lower phase
microemulsion and is slow to settle out. For IFT measurement, the presence of
the oil-rich emulsion is still important no matter the density of oil-rich emulsion. If
most (but not all) of the oil-rich emulsion was removed by centrifuging before the
IFT experiment as in Figure 2.2-9, the time that was needed to reach the
equilibrium low IFT became longer.
30
NaCl%= 2.0 2.2 2.6 3.0 3.4 3.6 3.8 4.0
4 hours settling
23 days settling
Figure 2.2-7 Comparison of phase appearance of 0.2% NI / 1% Na2CO3 / x % NaCl at

different settling time.
0.2% NI blend /1% Na2CO 3 / 3.4% NaCl, 23 days settling

with oil-rich emulsion
P=5.1
3 min 20 min
P=13.4 (Slower)
60 min 102 min
Figure 2.2-8 Photos of spinning drop of 0.2% NI blend / 1% Na2CO3 / 3.4% NaCl with all
of the oil-rich emulsion. The lower photos are at slower speed.
31
0.2% NI blend /1% Na2CO3 / 3.4% NaCl, 23 days settling

Remove most oil-rich emulsion
2 min
135 min
210 min
Figure 2.2-9 Photos of spinning drop of 0.2% NI blend / 1% Na2CO3 / 3.4% NaCl
Most of the oil-rich emulsion has been removed. (P=5.1)
The oil-rich emulsion is very important for the IFT measurement. However,
the spinning drop can not be seen if the oil-rich emulsion surrounds the oil drop
and extends to the end of the tube. The oil drops in Figures 2.2-4 and 9 can be
seen because the amount of oil-rich emulsion is no more than the amount of oil
drop. The oil-rich emulsion needs time to occupy the oil-water surface. It is
better to let the oil drop and the oil-rich emulsion settle in the spinning tube for
some time before the spinning experiments. Otherwise, the phenomena in
Figures 2.2-5 and 8 will occur and a longer time is needed to reach equilibrium
IFT.
A standard method which can quickly provide reproducible equilibrium IFT is

introduced. The spinning drop IFT experiments was conducted as follows:
1. Mix the crude oil with the alkaline surfactant solutions containing 0.2% NI
blend and 1% Na2CO3 at WOR equals 3. These solutions have different salinity
(0%NaCl~5%NaCl).
2. Rotate the mixture for 24 hours to reach equilibrium.
32
3. After settling the mixture for 4 hours, oleic and aqueous phases were taken
out into different syringes. The phase appearances of these samples are shown
as Figure 2.2-10. If settling time is longer than 24 hours, the aqueous phase will
clear up due to separation of the oil-rich emulsion. Low IFT may not be observed
as discussed in previous annual report.
4. Since the samples in the syringes may continue to settle and the settling time
in the syringe may be different, they were shaken before the IFT spinning drop
measurement so that they can be considered as the same sample that was
obtained after 4 hours settling.
5. Before the spinning drop measurement, the aqueous phase was centrifuged
first to remove some of the excess oil-rich emulsion because the sample will be
too dark if too much oil-rich emulsion is left. The remaining oil-rich emulsion
should be less volume than the volume of the excess oil drop that is added into
the spinning drop tube.
6. Let the oil drop settle in the vertical tube for some time (~12 hours) so that
the oil-rich emulsion can equilibrate with the oil and the lower phase
microemulsion.
7. Begin the spinning drop IFT measurement.
Step 6 will be shown to be a necessary procedure in the following. Figure

2.2-11 - 16 show the dynamic IFT of the samples with different salinities
(0%~5%). For 0% NaCl, 1% NaCl, 2% NaCl and 4% NaCl samples, step 6
reduces the time that is needed to reach the equilibrium low tension. However,
there is no significant difference for 3% NaCl sample by using step 6. This can
be explained by phase behavior in Figure 2.2-10. The lower phase for 3% NaCl
is similar in density to the oil-rich emulsion phase so that the oil-rich emulsion was
probably dispersed throughout the lower phase. For 5% NaCl sample, step 6
was not used because there is no oil-rich emulsion in the aqueous phase as the
aqueous phase is clear. The oil-rich emulsion must be the upper phase
emulsion, and Figure2.2-10 shows the upper-phase emulsion for 5% NaCl
sample.
33
Figure 2.2-10 Phase behavior of 0.2% NI blend/1% Na2 CO3

(24 hours mixing, 4 hours settling)
34
Dynamic IFT of 0.2%NI-1%Na2CO3-0%NaCl

1.E+00
IFT, dyne/cm
1.E-01
1.E-02
1.E-03
0 50 100 150 200 250
Time, minutes
0% NaCl with step 6 0% NaCl without step 6
Figure 2.2-11 IFT of 0.2% NI blend/1% Na2 CO3/0%NaCl as a function of time

1.E+00
IFT, dyne/cm
1.E-01
1.E-02
1.E-03
0 50 100 150 200 250
Time, minutes

1.E+00
IFT, dyne/cm
1.E-01
1.E-02
1.E-03
0 100 200 300 400 500 600
Time, minutes
35

1.E+00
IFT, dyne/cm
1.E-01
1.E-02
1.E-03
0 50 100 150 200 250 300 350
Time, minutes

1.E+00
IFT, dyne/cm
1.E-01
1.E-02
1.E-03
0 50 100 150 200 250
Time, minutes
4% NaCl With step6 4% NaCl Without step 6

1.E+00
IFT, dyne/cm
1.E-01
1.E-02
1.E-03
0 20 40 60 80 100
Time, minutes
5% NaCl without step 6
36
Figure 2.2-17 shows the equilibrium IFT measured with different procedures.
It shows that the step 6 does not change the equilibrium IFT in most cases but the
equilibrium IFT is reached in much less time. The settling time of step 3 has
significant effect on the equilibrium IFT at 3.6% and 4% salinity. This is because
at these intermediate salinity and long settling time, the oil-rich emulsion is not
present in the lower phase to settle between the oil and aqueous phase as shown
in Figure 2.2- 7. Thus, the low tension could not be achieved since no oil-rich
emulsion left in the lower phase. However, at the salinity of 5% NaCl, the upper
phase was the oil-rich emulsion and steps 5 and 6 are not needed.
1.00E-01
IFT(dyne/cm)
1.00E-02
1.00E-03
0 1 2 3 4 5
Salinity(% NaCl)
IFT vs Salinity (standard procedure) 4 hours settling in step 3 & no step 6
23 days settling in step 3 & no step 6 40 days settling in step 3 & no step 6
Figure 2.2-17 IFT change with salinity for 0.2NI-1%Na2 CO3 /WOR=3
Correlation between phase behavior and IFT
The solubility ratios of 0.2% NI blend/1% Na2CO3 /NaCl is calculated from the
phase behavior shown in Fig 2.2-18 and plotted in Fig 2.2-19 and 21. When the
lower phase is colored, there often is a thin creamed layer of oil-rich emulsion
between the oil and lower phase. In calculation of the solubility ratios, the
creamed layer volume was counted towards the volume of the lower phase. The
error of reading the volumes of oil and aqueous phase was estimated to be 0.01
ml. The error of solubility ratios thus was calculated to be 1.3. The solubility
ratios of this system are very high. Even at a low salinity of 2% NaCl, the
solubility ratio of oil to surfactant is close to 7.
37
0.2% NI blend/1% Na 2CO3 /x% NaCl @ 0.5% NI blend/1% Na 2 CO3/x% NaCl @

soap/synthetic surfactant ratio of 0.35 (WOR=3:1) soap/synthetic surfactant ratio of 0.14 (WOR=3:1)
2 months settling at room temperature 2 months settling at room temperature
x= 2.0 3.0 3.2 3.4 3.6 3.8 4.0 x= 3.0 3.5 4.0 4.3 4.5 4.7 5.0
Figure 2.2-18 Phase behavior after 2 months with 0.2% (left) and 0.5% (right) NI blend.
r a t i o
1000
Vw/ Vs
100
S o l u b ilit y
Vo/ Vs
10
1
2 2. 5 3 3. 5 4
NaCl, %
Figure 2.2-19 Solubility ratios of 0.2% NI blend/1% Na2 CO3/NaCl
38
Interfacial tension can be estimated from Chun-Huh correlation:
c
l 3  2 (Huh, 1979) ........................................................................(5.1)
Rl 3
where l3 : interfacial tension between excess oil or aqueous phase and
middle phase,
Rl3: solubility ratio of oil or water by surfactant,
c: a constant with a typical value of 0.3.
Figure 2.2.20 Comparison of IFT measured by spinning drop measurements using

standard procedure and that estimated by Chun-Huh correlation of 0.2% NI blend/1%
Na2 CO3 /NaCl. c=0.3.
Interfacial tensions of 0.2% NI blend/1% Na2CO3 /NaCl estimated from Chun-Huh

correlation with c=0.3 are plotted in Fig 2.2-19. At salinity of 2% - 4% NaCl, IFT
measured by spinning drop method and that estimated from Chun-Huh correlation
is close.
The phase behavior, with increased concentration of 0.5% NI blend is

shown in Fig 2.2-18 (right). The solubility ratios and IFT estimated from
Chun-Huh correlation with c=0.3 is plotted in Fig 2.2-21 and 2.2-22, respectively.
Both the solubility ratios and the estimated IFT are similar to that of the 0.2% in
39
magnitude but the optimal salinity has increased with increased surfactant
concentration.
1000
r a t i o
Vw/ Vs
100
S o l u b ilit y
10 Vo/ Vs
1
3 3. 5 4 4. 5 5
NaCl, %
Figure 2.2-21 Solubility ratios of 0.5% NI blend/1% Na2 CO3/NaCl
0.1
Chun-Huh correlation
IFT, mN/m
0.01
0.001
0.0001
3 3.5 4 4.5 5
NaCl, %
Figure 2.2-22 IFT estimated by Chun-Huh correlation of 0.5% NI blend/1% Na 2CO3 /NaCl,
c=0.3.
40
The remarkable observation of the phase behavior (Fig. 2.2-18) and IFT
measurements (Fig. 2.2-20 & 222) is that low IFT (i.e., below 10-2 mN/m) appears
to not be limited to the three-phase or Winsor III region but also occur well into the
lower-phase microemulsion or Winsor I region. We interpret this by looking at
the close-up of the phase behavior in Fig. 2.2-1. A system that appears to be a
lower-phase microemulsion has a creamed, oil-rich emulsion at the top just
beneath the oil phase. This oil-rich emulsion appears to be a “middle” phase
between the lower-phase microemulsion and the excess oil phase. We
observed in the IFT measurements that the presence of this oil-rich emulsion was
necessary for low tensions in most cases. We speculate that this oil-rich
emulsion may be due to the naphthenic soaps from action of the alkali on the
naphthenic acids in the oil. If this is the case, we would expect more of this
oil-rich emulsion to be present if the amount of crude oil is increased. That
appears to be the case in Fig. 2.2-23. The left two tubes both have a salinity of
2% NaCl but have soap/surfactant ratio of 1:1 and 0.35:1 respectively. The left
tube with more crude oil does appear to have more oil-rich emulsion.
S/SR=1 Soap / Synthetic surfactant ratio = 0.35

x= 2.0 2.0 3.0 3.2 3.4 3.6 3.8 4.0
Fig. 2.2-23 Phase behavior of 0.2% NI blend/1% Na2 CO3 /x% NaCl, 40 days of
settling. Thin creamed oil-rich layers exist at salinity of 2-3.4% NaCl.
41
Subtask 2.3 Characteristics of Alkali-Surfactant-Polymer process

Some important characteristics of the ASP process are discussed using the
one-dimensional ASP simulator that was discussed in the previous annual report.
The contour plot of IFT was generated in Figure 2.3-1 based on the earlier
experimental data which had high IFT at 4% NaCl. The low IFT region will be
even wider with latest results. This plot is different from the narrow low IFT
region contour plot, Figure 2.3-2 that was used in the previous annual report.
Figure 2.3-1 Contour of interfacial tension with wide low IFT region, (log 10(IFT))
Log10 IFT
Over-optimum
Under -optimum
Figure 2.3-2 Contour of interfacial tension with narrow low IFT region
(log10 (IFT)) (IFT: dyne/cm) (The width of low IFT region is base on reference [1])
42
Initial Oil Formation Acid No. of Crude oil Injecting Na2 CO3 Injecting
Saturation Brine concentration Salinity
0.3 2.0% 0.2g KOH/g 1.0% 2.0% NaCl
Surfactant Surfactant Injecting Polymer Injecting solution Crude Oil

concentration Slug Size (flopaam3330S) viscosity: viscosity
0.5%(NI blend) 0.5 PV 5000ppm 40 cp 19 .7 cp
Polymer Surfactant NX (Grid block No.) dt/dx Peclet No

adsorption Adsorption
20 g/g 0.2mg/g 100 0.05 500
Table 2.3-1 Other major parameters for the example
These two different models of IFT behavior were compared with the ASP
simulator. Other major parameters are the same and shown in Table 2.3-1.
The simulation results show that the recovery is 95.6% with wide low IFT region
and only 62.3% with narrow low IFT region. This result can be explained by
Figure 2.3-3. The left two figures show the surfactant and soap, IFT, and oil
saturation profiles respectively at wide low IFT region when the dimensionless
time equals 0.5 PV. The right two figures show the same profiles at same time
but with narrow IFT region. The low tension region (<10-2 dyne/cm) in the left
profile is around 0.1 dimensionless distance, while low tension region in the right
profile is only 0.03 dimensionless distance. The oil saturation in the left profile is
much less than that in the right profile. Narrow low IFT region will have less
recovery because oil will be trapped again when the IFT increases. When the
low IFT region is wide, less oil is trapped after the low tension region and recovery
increases.
In the previous annual report, high simulated oil recovery was achieved only
close to the optimal salinity because of the narrow low IFT region assumption.
However, with the wide low IFT region, high oil recovery can be achieved in a
wide range of injected salinities. Figure 2.3-4 illustrates this phenomenon.
Thus, injected salinity with high surfactant adsorption can be avoided. Also the
ASP process is more robust because of its large operational salinity region.
43
Wide low IFT region Narrow low IFT region

Recovery=95.0% Recovery=62.3%
Figure 2.3-3 Comparison of profiles between wide low IFT region and narrow low IFT
region
44
Recovery vs Injecting Brine Salinity

100%
80%
Recovery
60%
40%
Wide low IFT region

20%
Narrow low IFT region
0%
1 2 3 4 5 6
Injecting Brine Salinity(%)
Figure 2.3-4 Oil recoveries vs. injecting brine salinities
The other important parameter is viscosity. The injected solution viscosity

can change the oil recovery significantly as Figure 2.3-5 shows. The wide low
IFT region was used in all the figures of Figure 2.3-5. All the other parameters
are the same as Table 2.3-1 except the injection solution viscosities. The left
figures are calculated using 40 cp as the injected solution viscosity, while the right
used 24 cp as the injected solution viscosity. The IFT profiles of these two
conditions are similar to each other. However, the oil saturation profiles are
different, thus the oil recoveries are different. This is because the injected
solution viscosity has significant effect on recovery because it is related to mobility
ratio. Because the oil fractional flow increases with the aqueous phase viscosity,
the oil in the low tension region can flow more easily. Figure 2.3-6 illustrates this
effect. Lower aqueous phase viscosity, i.e., higher mobility ratio, has lower oil
recovery even with wide low IFT region because it takes a greater distance to
displace the oil before the IFT increases. Thus, the injected solution viscosity
should be large enough to obtain high oil recovery.
45
injecting solution viscosity=40cp injecting solution viscosity=24cp

Mobility Ratio =0.54 Mobility Ratio = 0.91
Recovery=95.0% Recovery=86.1%
Figure 2.3-5 Comparison of profiles between varied injecting solution viscosities
46
(Oil Fractional Flow)
(Oil Saturation)
Figure 2.3-6 Oil Fractional Flow vs. Saturation at IFT=0.001dyne/cm

(Oil viscosity =19.7 cp)
47
Subtask 2.4 Alkaline-Surfactant Polymer Forced Displacement
1. Dolomite sand pack
An alkaline-surfactant-polymer process was performed on a one-dimensional,

35 darcy dolomite sand pack column. Before the alkaline-surfactant process, the
water-flooding was conducted by the following procedure:
1. The sand pack was saturated with CO2 so that there was no air left in the
sand pack.
2. Brine with 2% NaCl was injected at the flow rate of 0.5 ml/min (interstitial
velocity = 14 ft/day) until the column was saturated by the brine.
3. Crude oil (MY4) was injected at the flow rate of 0.5 ml/min (interstitial
velocity = 14 ft/day) until the oil broke through. The oil cut was 100%
immediately after breakthrough. The oil saturation after the oil flooding
was 0.98.
4. The oil saturated column was placed into a 60 

C oven for 60 hours. The
purpose of this aging procedure was to change the wettabilty of the
substrate (dolomite) to mixed-wet.
5. After aging, brine with 2 % NaCl was pumped into the column at velocity
0.5 ml/min (interstitial velocity = 14 ft/day) until there was no oil in the
effluent.
Figure 2.4-1 show the photos of oil flooding and water flooding. The left one is
the photo for oil flooding. The right five photos show the water flooding at
different pore volume injected. The flow is upward.
48
Oil Flooding Water Flooding
Aged in 60ºC for

60 hours
0.1PV 0.3PV 0.5PV 1.0 PV 2.0PV 3.0PV

Figure 2.4-1 Photos of oil flooding and water flooding
The cumulative oil recovery and the fractional flow of oil are shown on Figure
2.4-2. In this figure, the oil saturation is 0.177 (Sor=0.177) after 3.2 PV was
injected. Further injection of water would not be expected to recover much more
oil.
Water flooding
1.0 1.0
0.9 0.9
Recovery/ OOIP .
0.8 0.8
Cumulative Oil
0.7 0.7
0.6 0.6
Oil Cut
0.5 0.5
0.4 0.4
0.3 0.3
0.2 0.2
0.1 0.1
0.0 0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
Pore Volumes
Cumulative Oil Recovery Oil Cut

Figure 2.4-2 Oil Recovery of Water Flooding in Dolomite Sand Pack
(OOIP: Original Oil in Place)
49
After water-flooding, the alkaline-surfactant process was implemented on the

dolomite sand pack. A 0.50 PV slug of alkaline surfactant polymer solution was
injected into the sand pack followed by a 1.0 PV polymer drive. The formulation
of alkaline surfactant polymer solution is shown in Table 2.4-1. The viscosity of
this solution was 45.1 cp at the shear rate of 66 sec-1. The reason that such high
viscosity was used is that the crude oil viscosity was 19.4 cp. The polymer drive
consists of 5000 ppm polymer and 2.0% NaCl.
Chemicals Concentration
Alkali (Na2CO3 ) 1.0 %
NaCl 2.0%
Surfactant (4:1 NEODOL 67:IOS ) 0.2%
Polymer (Flopaam 3330S) 5000ppm
Table 2.4-1 formulation for the alkaline surfactant polymer solution
The photos in Figure 2.4-3 illustrate how the oil bank forms and propagates in
the ASP process. Figure 2.4-4 shows the cumulative oil recovery and the
fractional flow of oil. Figure 2.4-5 shows the effluent. The oil bank breaks
though at 0.8 PV. The surfactant breaks through at 0.99 PV because we can
find lower phase micro-emulsion in the effluent at 0.99 PV. The incremental oil
recovery is 98.1% and the clean oil recovery is 61.3%. There may be some oil in
the aqueous phase because the color of the aqueous phase of some effluent is
brown. The history of pressure drop is shown in Figure 2.4-6. The pressure
increased with the injection volume because the surfactant slug and polymer drive
were designed to have a favorable mobility ratio. The pressure became stable
and did not increase any more after the surfactant broke through because the
whole column was occupied by the surfactant slug and polymer drive. Further
injection of the polymer solution did not change the pressure of the system.
The one-dimensional simulator was used to calculate this experiment. Figure
2.4-7 compares the simulation results and the experimental results. This plot
shows the simulation result matches the experimental results quite well.
50
0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60 0.65 0.70 0.75 0.90 1.50
Injected Pore Volumes
Figure 2.4-3 Photos of dolomite pack at different injected pore volumes
Alkaline surfactant flooding after water flooding

1.0 1.0
0.9 0.9
Recovery/ROIP .
Incremental Oil .
0.8 0.8
0.7 0.7
0.6 0.6
Oil Cut
0.5 0.5
0.4 0.4
0.3 0.3
0.2 0.2
0.1 0.1
0.0 0.0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Pore Volumes
Cumulative Oil Recovery Oil Cut
Figure 2.4-4 Oil Recovery of ASP Flooding in Dolomite Sand Pack

(ROIP: Residual Oil in place)
51
Effluent Pore Volumes

0.09 0.27 0.45 0.63 0.81 0.99 1.17 1.35 1.53 1.71 1.89 2.07
0.18 0.36 0.54 0.72 0.90 1.08 1.26 1.44 1.62 1.80 1.98
Figure 2.4-5 Effluent of ASP Flooding in Dolomite Sand Pack
3.5
Polymer Drive
Pressure difference (psi) .
2.5 Surfactant Slug

Surfactant Breakthrough
2
1.5
0.5
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
Pore Volume (PV)
Figure 2.4-6 History of pressure drop for dolomite pack
Figure 2.4-7 Comparison between simulation and experiments for dolomite pack
52
2. Silica sand pack

An Alkaline-surfactant-polymer (ASP) flooding experiment was performed in a
silica sand pack. The sand pack’s permeability was 40 darcy and the
waterflooding remaining oil saturation was 0.25. The operation parameters in
this experiment are the same as in the previous experiment except the surfactant
concentration is 0.5% and the injection solution viscosity is 43 cp at the shear rate
of 66 sec-1. The photos in Figure 2.4-8 illustrate how the oil bank forms and
propagates in the silica pack. From Figure 2.4-9, it took about 1.3 pore volumes
to get the incremental recovery of 98% of the remaining oil after waterflooding.
The pressure drop history is shown as Figure 2.4-10, which is similar to Figure
2.4-6.
Figure 2.4-11 also shows that the simulation result matches the experimental
results with the exception of the early oil production. The early oil production in
this experiment is because the remaining oil saturation after waterflooding was
slightly greater than the residual oil saturation after water flooding.
Figure 2.4-8 Photos of silica sand pack at different injecting pore volumes
53
1
Surfactant Slug Polymer Drive
Accumulated Oil R ecovery Efficiency

0.9
0.8
0.7 Surfactant Breakthrough
0.6
0.5
0.4
0.3
0.2
0.1
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
Pore Volume
Figure 2.4-9 Oil Recovery of ASP flooding in silica sand pack
3.5
Polymer Drive
3
Pressure difference (psi)
2.5
Surfactant Slug
2
1.5
0.5
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
Pore Volume
Figure 2.4-10 History of pressure drop for silica pack
54
Figure 2.4-11 Comparison between simulation and experiments for silica pack
3. Polymer/Surfactant phase separation

An experiment was done at the same condition as in Figure 2.4-8 except the
injected salinity was 4.0% NaCl. This was thought to be optimal salinity based
on the phase behavior experiments without presence of polymer. Before
injection, the ASP solution (4% NaCl) looked slightly turbid while the 2% NaCl
ASP solution was clear. After surfactant injection, no oil bank was found, as
shown in Figure 2.4-12. It seems the surfactant is ahead of polymer front. The
mobilized oil appears to be behind the surfactant front as if there was lack of
mobility control. The pressure history in Figure 2.4-13 shows that the pressure
drop in this experiment is around 10 times that in Figures 2.4-6 and 2.4-10. This
implies that there is a more viscous phase in that column since the flow rates are
same in all experiments. After one week settling of the ASP solution, phase
separation can be observed on the left of Figure 2.4-14. There is a clear,
separated layer in that sample. The injected solution should be a clear,
one-phase system. Otherwise, phase separation may occur during the flooding
process. Phase separation can be avoided by injecting at a lower salinity.
55
0 PV 0 .1 7 PV 0 .3 3 PV 0 .5 PV 0 .6 7 PV 0 .8 3 PV 1 .0 PV 2 .0 PV
Figure 2.4-12 Photos of silica sand pack at different injecting pore volumes when injecting
salinity = 4.0% NaCl
30
25
Pressure Difference (psi)
20
Surfactant Slug Polymer Drive
15
10
0
0 0.5 1 1.5 2
Pore Volume
Figure 2.4-13 Pressure difference vs. pore volume when polymer/surfactant

separation, 4% NaCl
56
0.5% N67-7PO&IOS(4:1), 0.5% N67-7PO&IOS(4:1),

0.5% FLOPAM 3330S, 0.5% FLOPAM 3330S,
Separate layer
Figure 2.4-14 Phase behaviors of different ASP solutions after one week
57
Conclusions
1. The oil-rich emulsion plays an important role for low IFT in the
alkali-surfactant system.
2. A spinning drop IFT measurement procedure was introduced for
alkali-surfactant system which can quickly reach the equilibrium IFT.
3. The NI Blend-MY4-Na 2CO3 system has a wider low IFT region than normally
seen for single surfactant systems. The width of the low IFT region is a key
factor for recovery. Narrow low IFT region will have less recovery because
oil will be trapped again when the IFT increases. When the low IFT region is
wide enough, less oil will be trapped after the low tension region.
4. The injection solution viscosity has significant effect on recovery. Lower
aqueous phase viscosity, i.e., higher mobility ratio, has lower oil recovery
even with wide low IFT region. This is because the oil fractional flow
increases with the aqueous phase viscosity.
5. Experimental results show that the ASP process with only 0.2%, 0.5 PV
surfactant slug recovers 98% of the oil that is trapped after water-flooding.
Good recoveries (>95%) were obtained in both dolomite and silica sand
packs.
6. The simulation matches the experimental data for both dolomite and silica.
7. High salinity can cause the phase separation of ASP solutions. This may
result in loss of mobility control.
58
Task 3 Foam for Mobility Control
3.1 Bulk foam in fractures
In our previous work, we found the factors that can affect foam apparent
viscosity in fractures when the foam bubble diameter is larger than the fracture
aperture. Two mechanisms – liquid between bubbles and bubble deformation --
were found to contribute to the foam apparent viscosity in this case. Predictions
of a theory modeling these two mechanisms fit experimental data. But when the
foam bubble diameter is significantly less than the aperture, bulk foam exists
throughout the fracture. Further experiments were performed and a different
model used for this case as discussed below.
Experimental technique
The fracture model shown in Fig. 3.1-1 has been described in previous
reports. It mainly consists of two parallel plates. Changing the gasket thickness
between the plates can change the aperture of the fracture. The set-up diagram
of the equipment for the foam experiments is shown in Fig. 3.1-2. A Harvard
syringe infusion pump (Model 22) is used to inject surfactant solution, and a
Matheson mass flow controller (Model 8270) is used to inject air into the foam
generator. Relatively uniform size bubbles can be generated only when the air
and liquid are introduced on opposite sides of the frit in the foam generator.
Choosing frits with different pore size can generate different sizes of bubbles.
Two grooves were made along the inlet and outlet of the fracture model to
ensure a uniform pressure at the inlet and outlet.
59
Figure 3.1-1 Detailed fracture model
The surfactant solution in the experiments was 0.5% C13-4PO and 0.5%
STEOL CS330. C13-4PO is from Harcros Chemicals and its chemical
description is propoxylated C13 alcohol ether sulfate, ammonium salt. STEOL
CS330 is from Stepan and its chemical description is C12-3EO sulfate. The
salinity was 0.23% NaCl, 0.07% CaCl2 and 0.04% MgCl2 . The mean bubble
diameters in the experiment were 0.04 and 0.06 mm. The aperture is 0.2 mm or
0.4 mm for homogeneous fracture experiments. The gas fractional flow range
was from 0.0 to 0.67. The viscosity and surface tension were 1.0 mPa.s and 28
mN/m, respectively.
60
Figure 3.1-2 Diagram for experiment in fracture model
Theory
Princen [1982] developed a theory for rheology of foams and highly
concentrated emulsions. Hirasaki and Lawson [1989] developed a theory to
describe the apparent viscosity of bulk foam in a capillary tube with plug flow and
high gas fraction. But the theory can be applied only to bubbles with the shape
of pentagonal dodecahedrons, which are obtained only at quite high gas
fractional flow when all the bubbles in the system are closely-packed.
Many semi-empirical expressions are available for describing the shear

viscosity of concentrated dispersions of hard spheres. The most widely used is
the functional form suggested by Krieger and Dougherty [1959].
r (1 K) 2 .5 / K (1)
Where r is the relative viscosity, which is the ratio of the viscosity of emulsion to
the viscosity of water.  is the volume fraction of the dispersed phase of the
emulsion. K is the crowding factor and equal to the reciprocal of the dense
random packing limit volume fraction max , at which r diverges to infinity. For
random close packing of monodisperse hard spheres, they found max = 0.64 and
K =1.56.
Mooney [1951] developed another expression for the relative viscosity of

emulsions where the particles behave as rigid spheres.
61
2.5
r exp[ ] (2)
1 K
The crowding factor K in the above two equations can be smaller when the
particles are not uniformly distributed or the particles are deformable because
these factors can cause an increase of the dense random packing limit volume
fraction.
Pal [1992] studied the rheology of polymer-thickened emulsions and found

that the effect of internal circulation within the drops was sometimes important.
He suggested the following equation:
2.5
r 1 / KI exp[ ] (3)
1 K
where K I is a factor which takes into account internal circulation effects and is
given by
1 0.4(c / d )
K I [ ]
1 (c / d )
In the above equation, c is the viscosity of the continuous phase and d is the
viscosity of the dispersed phase. Pal [1992] also suggested a crowding factor
K = 1.04, which means the dense random packing limit volume fraction max =
0.96.
Results and discussion
The apparent viscosity for bulk foam flow in fractures was measured at
different aperture, flow velocity and bubble size as shown in Fig. 3.1-3. The
highest gas fractional flow is 0.67 because we found that bubbles began to
coalesce for gas fractional flows exceeding 0.67.
The predictions from Krieger and Dougherty equation, Mooney equation and
Pal’s model are also plotted in Fig. 3.1-3. For the first two equations max has
been increased to 0.99 and K decreased to 1.01 because we deal with
deformable bubbles, not hard spheres. Even with this change there are still
significant deviations between the predictions and experimental results although
agreement is much better than with the corresponding hard sphere values given
above. In contrast, the experimental measurements match well the prediction of
Pal’s model.
Because the viscosity of water is much larger than that of gas, K I is close
to 0.4. We still use 0.96 as the dense random packing limit volume fraction.
62
Then the crowding factor K is 1.04. Because the viscosity of the dispersed
phase is small in the foam case, internal circulation doesn't contribute
significantly to the viscosity. .
0.5% C13-4PO & 0.5% CS330

0.23% NaCl, 0.07% CaCl2, 0.04% MgCl2,
1000
u = 20 cm/min, aperture = 0.2mm, DB = 0.04mm
Apparent viscosity (cp)

Pal equation
100 Mooney equation
Krieger and Dougherty equation
10
Bubble Coalescence
1
0 0.2 0.4 0.6 0.8 1
Gas fractional flow
Figure 3.1-3 Bulk foam apparent viscosity in fractures; measurement and
prediction. K=1.01 for Mooney and Krieger and Dougherty equations.
3.2 Foam drive for ASP process
In normal alkaline/surfactant/polymer process, polymer is used as mobility

control agent to be injected together with surfactant and also as drive. However
polymer is expensive. We investigated the possibility of using foam to replace
polymer as drive.
The experiment was performed in a horizontal one-dimensional 1-foot long

sand pack. The sand pack’s permeability is 40 darcy. It was presaturated with a
polymer/surfactant solution (the same as in the ASP flood of Section 3.1 below)
to simulate a situation when residual crude oil saturation was zero behind the
surfactant/polymer slug. The polymer in this slug is 0.5% FLOPAM 3330S. The
surfactant is 0.5% N67-7PO:IOS15-18 (4:1). The salinity is 2% NaCl and 1%
Na2CO3 . By measurement, the viscosity of the polymer/surfactant solution is
43cp at 24C and shear rate 66s-1 .
The experiment was performed at constant pressure drop between 4.0 and
4.5psi. The foam was generated in the sand pack by co-injection of surfactant
63
solution and air at gas fractional flow 0.67. We tried three different surfactants
using the same salinity as in the slug: 1% IOS, 0.5% CS330, 0.5% C13-
4PO&CS330 (1:1). The surfactant blend N67/IOS was not included although it
has good performance in displacing oil. As described as in last report, the foam
produced by the blend of N67/IOS is weak. The reason may be that N67 is
hydrophobic with a branched hydrophobe that prevents formation of a compact
surfactant monolayer. The experimental results are shown in Fig. 3.2-1. About
0.1~0.2PV of co-injected fluid was needed to get the pressure drop up to 4.5psi.
Then by changing the flow rate, the pressure drop was kept between 4.0~4.5psi.
The foam produced by 1% IOS broke through around 1PV, 0.5% CS330 at
0.9PV and 0.5% C13-4PO&CS330 (1:1) at 0.8 PV. We also tried to use the foam
generated by 1% N67/IOS (4:1) and found the pressure drop never exceeded
1.5psi before foam breakthrough.
fg = 0.67, pressure difference = 4.5psi
2% NaCl, 1% Na2CO3
Presaturated with 0.5% polymer, 0.4% N67-7PO, 0.1% IOS, 2% NaCl, 1% Na2CO3
0.1~0.2 PV injection is needed for pressure difference to increase to 4.5 psi.
120
100 1% IOS
0.5% CS330
0.5% C13-4PO&CS330(1:1)
Sup erficial Velocity (ft/day)
80
60 foam breakthrough
40
20
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Pore Volume
Figure 3.2-1 Foam sweep of the sand pack presaturated with

polymer/surfactant
64
fg = 0.67, pressure difference = 4.5psi

2% NaCl, 1% Na2CO3
Presaturated with 0.5% polymer, 0.4% N67-7PO, 0.1% IOS
1% IOS
250
0.5% CS330
0.5% C13-4PO&CS330(1:1)
200
Apparent Viscosity (cp)
150
100
50
0
0 0.2 0.4 0.6 0.8 1 1.2
Pore Volume
Figure 3.2-2 Apparent viscosity during the sweeping the sand pack
presaturated with polymer/surfactant
Fig 3.2-2 shows the apparent viscosity during the sweeping the sand pack
by foam. At the time of foam breakthrough, the apparent viscosity is about 40cp
for 0.5% CS330 and 200cp for 1% IOS and 0.5% C13-4PO&CS330 (1:1). The
viscosity for the polymer/surfactant solution is 43cp. If we assume the apparent
viscosity of the system at the foam breakthrough is equal to the foam apparent
viscosity, the foam apparent viscosity is close to or higher than the viscosity of
the polymer/surfactant solution. But we did see the fingering at the interface of
the foam and polymer/surfactant. The reason may be that the foam front is not
as strong as the steady foam behind and the N67 in the system may weaken the
foam front.
The results in Fig. 3.2-1 show that it is possible to use foam instead of
polymer as drive in ASP process. Although the cost of surfactant is comparable
to polymer at the same concentration, lower surfactant concentrations is possible
and also less surfactant solution is required because a substantial portion of the
drive fluid would be the gas in the foam bubbles.
65
3.3 Foam stability with the presence of residual oil
The stability of foam by different surfactants in the presence of residual oil

was tested. A short sand pack is used to avoid the long periods of time that
would have been required to test various surfactants in the 1-ft long sand pack.
The set up for the experiment is shown in Fig. 3.3-1.
Figure 3.3-1 Short sand pack
The porous medium consists of a 1.5 inch long and 0.5 inch diameter
cylinder packed with 50-mesh sand. Pore volume for the short sand pack is
about 5ml. Foam is generated in the sand pack by co-injection of surfactant
solution and air. The surfactant is injected by an Isco pump, and air is injected
by an air flow controller. Two pressure transducers measure the pressure drop
across the tubing and the injection pressure to the sand pack. Foam quality is
measured by passing foam from the tubing through an inverted burette
containing IPA/water mixture, which breaks the foam. The ratio of the volume of
air collected to the amount of liquid displaced gives the quality of foam.
To test the foam stability in the presence of residual oil, first, the foam
strength without the presence of oil is measured for comparison at some
constant gas fractional flow. Second, the water-saturated pack is filled with
hexadecane. Then water is used to flush the sand pack to achieve residual oil
saturation. Foam is then generated at the same gas fractional flow. The steady
state pressure readings are recorded to compare the foam strength with or
without residual oil in the sand pack.
66
The results are shown in Fig.3.3-2. IOS seems to be the best foamer
among all the surfactant candidates. Even after contacting with residual crude
oil, the foam produced by IOS still has a higher apparent viscosity than that
formed by any other foamer. N67 and the blend of N67/IOS(4:1) has almost the
same apparent viscosity with or without residual crude oil. But that doesn’t
necessarily mean that foam generated with N67 or the N67/IOS(4:1) is stable
when the residual oil is present. Because the salinity we used is close to the
optimal salinity of N67-7PO and N67-7PO&IOS(4:1), almost all residual oil is
displaced by the initial surfactant injected so that little oil is present at the final
steady state conditions. And the foam of both N67 or N67/IOS(4:1) is weak
compared with that of IOS. Therefore IOS is the best candidate of the evaluated
surfactants as a foamer.
Short core, fg = 0.8,

Flow rate = 6ml/min
20
18
Without residual oil
16
Pressure difference (psi)
With residual oil

14
12
10
0
1% Na2CO3, 2% 1% Na2CO3, 5% 1% Na2CO3, 2% 1% Na2CO3, 2% 1% Na2CO3, 2%
NaCl NaCl NaCl NaCl NaCl
0.5% N67-7PO 0.5% N67- 0.5% N67- 0.5% N67- 0.5% IOS
7PO&IOS(4:1) 7PO&IOS(4:1) 7PO&IOS(1:1)
Figure 3.3-2 Comparison of foam strength with or without residual oil by

different surfactants
67
Reference
1. G.J. Hirasaki and J.B Lawson, Mechanisms of Foam Flow in Porous
Media: Apparent Viscosity in Smooth Capillaries, Society of Petroleum
Engineers Journal, 25 (2), 1985, 176-190.
2. H.M. Princen, Rheology of Foams and Highly Concentrated Emulsions, I.
Elastic Properties and Yield Stress of a Cylindrical Model System, Journal
of Colloid Interface Science (1983) 91, No. 1 160-175.
3. I.M. Krieger, Rheology of Monodisperse Latices, Advances in Colloid and
interface Science, 1972, 3, 111-136.
4. M.J. Mooney, The Viscosity of a Concentrated Suspension of Spherical
Particles, Journal of Colloid Science, 1951, 6, 162-170.
5. R. Pal, Rheology of Polymer-thickened Emulsions, Journal of Rheology,
36(7), 1992.
68
Task 4: Simulation of Field-Scale Processes
Subtask 4.2: Wettability alterations in naturally fractured reservoirs

The objective of this task was to adapt the existing chemical reservoir
simulator UTCHEM to model wettability alteration in oil reservoirs due to
surfactant injection. We have continued our effort to compile both laboratory
data and correlations for relative permeability and capillary pressure as a function
of wettability. We have developed and implemented in UTCHEM a linear
correlation to take into account the effect of wettability alteration on relative
permeabilities. The correlation takes into account the effect of surfactant on IFT
reduction and oil mobilization in addition to relative permeability changes due to
wettability alteration.
Wettability alteration model in UTCHEM
Corey-type relative permeabilities are calculated for each gridblock.
k r k orSnn (1)
where l  is either water, oil or microemulsion phases, k oris the relative

permeability endpoint for phase l 
, nℓis the Corey exponent of phase l 
and Snℓis
the normalized saturation of phase l 
calculated as follows:
S S
Sn  3 r (2)
1 Sr
1

where Sℓis the saturation of phase l and Sℓr is the residual saturation of phase l.
As mentioned before, in addition to the wettability alteration effect, surfactants
also reduce the interfacial tension between oil and aqueous phases and help in
the oil mobilization. This effect is modeled by means of a dimensionless group
called trapping number, which is a combination of capillary number and bond
number and can adequately model the combined effect of viscous, capillary, and
buoyancy forces in three dimensions (Delshad, 1990; Delshad et al., 1996; Jin,
1995, UTCHEM technical manual, 2000). As the surfactant enters a gridblock, it
reduces the interfacial tension and as a result, trapping number increases.
Interfacial tension reduction and oil mobilization effects of surfactantsproduces
changes in the residual phase saturations, endpoint relative permeabilities, and
exponents. Mobilization effect on residual phase saturations is modeled in
UTCHEM as follows (Delshad et al., 1986):
r 
Slow Shigh
Sr Shigh   r
(3)
1 TNT
r
69
where Sℓrhigh and Sℓ low

r are residual saturations of phase l 
at high and low
capillary numbers respectively (given as input parameters), Tℓ is the input
trapping parameter of phase l and NTℓ is trapping number of phase l.
Mobilization effects on endpoint relative permeabilities are modeled using the
following correlation (Delshad et al., 1986; UTCHEM technical manual, 2000).
low
k ork or  low
r 
Slow
Sr Sr
high

Sr o high low
kr  k or  (4)
k or and k or represent the endpoint relative permeability of phase l 

low high
at low and
high capillary numbers respectively. Equation 5 gives the relative permeability
exponents as a function of capillary number (Delshad et al., 1986; UTCHEM
technical manual, 2000).
r 
n n low
 
Slow
 Sr high

high n  n 
Sr Sr
low
low
 (5)
where nℓlow and nℓhigh represent the relative permeability exponents for low and
high capillary numbers respectively specified as input parameters.
The above equations (Equations 1-5) are solved once for initial reservoir
wettability condition for example oil-wet ( k original
r ) and once for the extreme altered
condition of water-wet ( k r ). Two sets of relative permeability ( k or, Sℓr, nℓ
final
) and
trapping parameters (Tℓ) are required as input corresponding to each wettability
state.
The relative permeability in each gridblock, k r, is then obtained by linear
interpolation between two relative permeabilities corresponding to different
wettability conditions, provided that the concentration of surfactant in the
gridblock is greater than the critical micelle concentration. Interpolation is made
based on the scaling factor .
k actual
r r 
k final 1 k initial
r (6)
where is the interpolation scaling factor, k final original

r and k r represent the relative
permeabilities corresponding to the two extreme wetting states, i.e. final and
initial wettability states, respectively. The scaling factor is related to the
concentration of surfactant adsorbed in each gridblock as follows:
70
ˆsurf
C
 (7)
Cˆ
surf
Csurf
where ĈSurf and CSurf represent the adsorbed and total concentration of surfactant
respectively. The historical maximum of  in each gridblock is used for the
interpolation.
The model has been implemented in UTCHEM. We are in the process of
validating the model and its implementation against the laboratory imbibition cell
experiments performed at Rice University (Hirasaki et al., 2005).
References
Delshad, Mojdeh, Delshad, Mohammad, Bhuyan, D., Pope, G. A. and Lake,
L.W.: "Effect of Capillary Number on the Residual Saturation of a Three-
Phase Micellar Solution," Paper SPE 14911, Proceedings of the SPE/DOE
Fifth Symposium on Enhanced Oil Recovery, Tulsa, OK, April 20-23,
1986.
Delshad, Mohammad: "Trapping of Micellar Fluids in Berea Sandstone," Ph.D.
dissertation, The University of Texas at Austin, 1990.
Delshad, M., Pope, G. A., and Sepehrnoori, K.: "A Compositional Simulator for
Modeling Surfactant Enhanced Aquifer Remediation, 1 Formulation,"
Journal of Contaminant Hydrology, 23, 303-327, 1996.
Hirasaki, G.J., C.A. Miller, and G.A. Pope: "Surfactant Based Enhanced Oil
Recovery and Foam Mobility Control," DOE annual report for the period of
July 2004 to July 2005. DOE contract number DE-FC26-03NT15406, July
2005.
Jin, M.: "A Study of Nonaqueous Phase Liquid Characterization and Surfactant
Remediation," Ph.D. dissertation, The University of Texas at Austin, 1995.
"Technical Documentation for UTCHEM-9.0 A three- Dimensional Chemical
Flood Simulator," prepared by Reservoir Engineering Research Center for
Petroleum and Geosystems Engineering, The University of Texas at
Austin, Austin, Texas, July 2000.
71

DOE-Surfactant-2006-Semi-Annual

Uploaded by

Copyright:

Available Formats

DOE-Surfactant-2006-Semi-Annual

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

DOE-Surfactant-2006-Semi-Annual

Uploaded by

Copyright:

Available Formats

Surfactant Based Enhanced Oil Recovery and Foam Mobility Control

3nd Semi-Annual Technical Report

Reporting Period Start Date: July 2005

Reporting Period End Date: December 2005

George J. Hirasaki, Rice University

Date Report was Issued: February 2006

The University of Texas

This report was prepared as an account of work sponsored by an

Surfactant flooding has the potential to significantly increase

The addition of an alkali such as sodium carbonate makes

Mobility control is essential for surfactant EOR. Foam is evaluated

UTCHEM is a reservoir simulator specially designed for surfactant

Figure 2.2-2 Spinning drop measurement of 0.2 % NI blend/1% Na2CO3/2%

Oil recovery by primary depletion and waterflooding, on the average,

Task 1. Improved Surfactants and Formulations

Subtask 1.1. Identifying and Synthesizing Improved, Cost-effective

Table 1.1-1 Surfactants Tested in Year 3

Descriptive or Trade Name Abbreviated Chemical Formula

1.2.1 Phase Behavior Results

Surfactant formulations were screened for several characteristics during the

1% C15-18 IOS, 1% C20-24 AOS, 4% SBA 6 4 1

2% C20-24 AOS, 4% SBA 1-2 3-5 1

1% C16 o-Xylene sulfonate , 1% C 20-24 AOS, 4%

2% C15-18 IOS, 4% SBA >10 2 1

2% C16-18 AOS, 4% SBA 10 2 1

Did not Did not Did not

1% C20-24 AOS, 1% C11-13 ABS, 4% SBA >4 1 1

1% C20-24 AOS, 1% Dowfax 3B2, 4% SBA >4 0.5 1

1% C20-24 AOS, 1% Naphthalene Sulfonate, 4%

Several phase behavior experiments provided comparable results in terms of

1.2.2 Polymer Results

The Second Annual Report presented information on two types of commercially

Viscosity of Flopaam 3330 at various salinities

Figure 1.2-1. Viscosity of Flopaam®3330S Polymer

where the resistance factor is given by

Figure 1.2-2. Polymer Permeability Reduction Factor vs. Permeability for

Subtask 1.3 Propoxylated-sulfated surfactants

Figure 1.3-1 depicts the test results of salinity scans without

Tests were conducted at WOR=1 in borosilicate 10-cc pipettes.

Test results depicted in Figure 1.3-1 indicate that N67-7PO is

Figure 1.3-1. Test results of salinity scans for 3- and

Figure 1.3-2 is a plot of optimal salinity vs. Alkane Carbon Number

140°F. We hypothesize that the appearance of birefringence as the oil molecular

Subtask 1.4 Calcium tolerance of NI surfactant blend

0.5% N67-7PO&IOS, 2% NaCl

Single Phase Region

Figure 1.4-1 Phase diagram of NI blend

Task 2 Phase behavior, adsorption, and composition changes during

Subtask 2.1 Surfactant Adsorption

Low surfactant adsorption domain

Adsorption of 4:1 N67:IOS on Calcite in Varying Salinity and Alkalinity

5% NaCl, 0% Na2CO3 5% NaCl with 1.21% Na2CO3

Adsorption of 4:1 N67:IOS on Calcite at 5% NaCl

5% NaCl, 0% Na2CO3 5% NaCl with 0.178% Na2CO3

Figure 2.1-2 Adsorption of 4:1 N67:IOS on calcite in varying alkalinity at 5% NaCl

Figure 2.1-3 the contour of maximal adsorption for N67:IOS(4:1)

Figure 2.1-4 Adsorption of N67 on calcite powder

0% NaCl, 0% Na2CO3 0% NaCl, 1% Na2CO3

Figure 2.1-5 Adsorption of IOS on calcite powder