Report 1 PDF

Surfactant Based Enhanced Oil Recovery and Foam Mobility Control
Semi Annual Technical Report
Reporting Period Start Date: July 2004
Reporting Period End Date: June 2005
Principal Authors:
George J. Hirasaki, Rice University

Clarence A. Miller, Rice University
Gary A. Pope, The University of Texas
Date Report was Issued: July 2005
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

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.
1-2
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 combination of two surfactants was found to
be particularly effective for application in carbonate formations at low temperature. A
formulation has been designed for a particular field application.
The addition of an alkali such as sodium carbonate makes possible in situ generation of
surfactant and significant reduction of surfactant adsorption. In addition to reduction of
interfacial tension to ultra-low values, surfactants and alkali can be designed to alter
wettability to enhance oil recovery. The design of the process to maximize the region of
ultra-low IFT is more challenging since the ratio of soap to synthetic surfactant is a
parameter in the conditions for optimal salinity. Compositional simulation of the
displacement process demonstrates the interdependence of the various components for
oil recovery.
An alkaline surfactant process is designed to enhance spontaneous imbibition in

fractured, oil-wet, carbonate formations. It is able to recover oil from dolomite core
samples from which there was no oil recovery when placed in formation brine.
Mobility control is essential for surfactant EOR. Foam is evaluated to improve the
sweep efficiency of surfactant injected into fractured reservoirs.
UTCHEM is a reservoir simulator specially designed for surfactant EOR. It has been
modified to represent the effects of a change in wettability. Simulated case studies
demonstrate the effects of wettability.
1-3
TABLE OF CONTENTS
ABSTRACT .................................................................................................................. 1-3
TABLE OF CONTENTS ............................................................................................... 1-4
TABLE OF TABLES ..................................................................................................... 1-6
TABLE OF FIGURES ................................................................................................... 1-6
INTRODUCTION........................................................................................................ 1-16
EXECUTIVE SUMMARY............................................................................................ 1-17
Task 1 Improved surfactants and formulations........................................................... 1-19
Subtask 1.1 Identifying and synthesizing improved, cost-effective surfactants1-19
1.1.1 Surfactant Screening using Phase Behavior Experiments.......... 1-21
1.1.2 Results and Discussion for Subtask 1.1...................................... 1-21
Subtask 1.2. Surfactant Tailoring for Crude Oils and Phase Behavior ........... 1-24
1.2.1 Results of Phase Behavior Experiments..................................... 1-25
Effects of Electrolyte Concentration .......................................... 1-28
Effects of Sodium Carbonate .................................................... 1-30
Effects of sec-Butyl Alcohol (SBA) ............................................. 1-32
1.2.2 Results of Polymer Testing ......................................................... 1-33
1.2.3 Results of Experiments to Characterize Rock Cores .................. 1-36
Berea core preparation ............................................................. 1-36
Dolomite core testing and preparation ...................................... 1-36
Air permeability tests ................................................................. 1-36
High Resolution X-ray Computed Tomography (HRXCT) scans 1-37
Conservative tracer tests .......................................................... 1-39
Polymer retention and degradation ........................................... 1-40
Surfactant adsorption ................................................................ 1-42
1.2.4 Results of Core Flood Experiments to Evaluate Oil Recovery ... 1-42
Midland Farms Core Flood #1 ................................................... 1-45
Midland Farms Core Flood #2.................................................... 1-46
Task 2 Phase behavior, adsorption, and composition changes during displacement .. 2-1
Subtask 2.1 Surfactant Adsorption .................................................................... 2-1
Surfactant propagation velocity with adsorption...................................... 2-1
Alkali consumption on carbonate formation ............................................ 2-3
Potential determining ions for surfactant adsorption ............................... 2-4
Surfactant aggregates’ size .................................................................... 2-5
Surfactant adsorption on porous media with different surface area ........ 2-6
Adsorption isotherms for NEODOL-67-7PO:IOS (4:1) and threshold of
Na2CO3 .................................................................................................. 2-7
Conclusions ............................................................................................ 2-9
Subtask 2.2 Composition route for alkali-surfactant flooding ............................. 2-9
Assumptions and Models ..................................................................... 2-10
1-4
Partition................................................................................................. 2-10
Interfacial tensions ................................................................................ 2-12
Aqueous Phase Viscosity ..................................................................... 2-14
Fractional flow....................................................................................... 2-14
Surfactant Adsorption ........................................................................... 2-15
Equations and calculation procedure.................................................... 2-16
Calculation example.............................................................................. 2-17
Conclusions .......................................................................................... 2-23
Reference: ............................................................................................ 2-23
Subtask 2.5 Surfactant-enhanced spontaneous imbibition experiments ......... 2-24

Surfactant Identification ........................................................................ 2-24
Problems with the TC Blend ................................................................. 2-25
N67-3PO S and Blends......................................................................... 2-26
N67-7PO Sulfate and Blends................................................................ 2-28
N67-7PO S: IOS 1518 = 4:1 (wt) .......................................................... 2-30
Adsorption............................................................................................. 2-31
Phase Behavior and IFT ....................................................................... 2-31
Oil recovery by centrifuge imbibition ..................................................... 2-33
Conclusions .......................................................................................... 2-35
References ........................................................................................... 2-35
Task 3 Foam for Mobility Control ................................................................................. 3-1

3.1 Foam diversion in fracture network model ................................................... 3-1
Experimental technique .......................................................................... 3-1
Theory..................................................................................................... 3-3
Apparent viscosity................................................................................... 3-3
Experimental results ............................................................................... 3-5
Conclusions .......................................................................................... 3-11
3.2 Test of surfactant’s ability in generating foam............................................ 3-12
Task 4: Simulation of Field-Scale Processes .............................................................. 4-1

Subtask 4.2: Wettability alterations in naturally fractured reservoirs................. 4-1
Effect of Wettability on Residual Oil Saturation, Relative Permeability, and
Capillary Desaturation Curves ................................................................ 4-1
Simulations with Differing Wettability ...................................................... 4-2
Case #1: Strongly water-wet................................................................... 4-3
Case #2: Neutral-wet .............................................................................. 4-4
Case #3: Neutral-wet with a different CDC ............................................. 4-5
Case #4: Oil-wet ..................................................................................... 4-6
Discussion ............................................................................................. 4-6
Computing the Relative Permeability Curves as a Function of Wettability4-8
References ........................................................................................... 4-61
1-5
TABLE OF TABLES
Table 1.1-1. Surfactants Tested in Year 2.................................................................. 1-20

Table 1.2-1. Formulations Screened with Midland Farms Crude Oil at 38° C ............ 1-26
Table 1.2-2. Formulations Screened with Elk Hills Crude Oil, 100°C ......................... 1-27
Table 1.2-3. Effects of SBA and Sodium Carbonate on Phase Behavior ................... 1-33
Table 1.2-4. Summary of MF Core Permeability ........................................................ 1-37
Table 1.2-5. Summary of Outcrop Dolomite Rock Permeability ................................. 1-37
Table 1.2-6. Sulfate Measured in Produced Formation Brine..................................... 1-47
Table 1.2-7. Summary of Core Floods ....................................................................... 1-50
Table 2.2-1 Operating Parameters for the example ................................................... 2-17

Table 2.2-2 Other major parameters for the example ................................................ 2-17
Table 2.5-1 Surfactant and co-surfactant identification ............................................. 2-24

Table 2.5-2 Bond number at 6,000 RPM, Perm=6 md .............................................. 2-33
Table 2.5-3 Core properties, experimental conditions and results ............................. 2-35
Table 4.2.1a. Reservoir Properties and Simulated Well Conditions ........................... 4-11
Table 4.2.1b. Fluid Properties .................................................................................... 4-11
Table 4.2.2. Input Parameters for Each Test Case .................................................... 4-11
Table 4.2.3. Summary of Literature Search to Obtain Trapping Parameter Data....... 4-12
TABLE OF FIGURES
Figure 1.1-1. Neodol® N67- 7PO-sulfate (C16-17 -(PO)7-SO4) ................................... 1-22

Figure 1.1-2. C20-24 AOS............................................................................................ 1-23
Figure 1.1-3. C15-18 IOS ............................................................................................. 1-24
Figure 1.2-1. Salinity Scan with MF3 Crude Oil at 38°C............................................. 1-29
Figure 1.2-2. Salinity Scan with Elk Hills Crude Oil at 100°C .................................... 1-29
Figure 1.2-3. Effect of Sodium Carbonate on MF3 Crude Oil at a WOR 1:1 and 38°C
using a mixture of 0.75% N67-(PO)7-SO4, 0.25% C15-18 IOS, 2% SBA............ 1-30
Figure 1.2-4. Effect of Sodium Carbonate on Optimal Salinity with ELK Crude, WOR 1:1
and 100°C ....................................................................................................... 1-31
Figure 1.2-5. Effect of Sodium Carbonate on Equilibration Time with Elk Hills Crude,
WOR 1:1, 100°C, 2% C20-24 AOS, 4% SBA ..................................................... 1-32
Figure 1.2-6. Viscosity of HPAM Polymer (Flopaam 3330S) in Brine......................... 1-34
Figure 1.2-7. Viscosity for HPAM Polymer (Flopaam 3330S) in Brine........................ 1-35
Figure 1.2-8. Viscosity for HPAM Polymer (Flopaam 3330S) in Brine........................ 1-35
1-6
Figure 1.2-9. CT Scans of MF54 Dolomite Core ....................................................... 1-38
Figure 1.2-10. CT Scans of MF11 Dolomite Core ..................................................... 1-39
Figure 1.2-11. IPA Tracer Breakthrough Curve......................................................... 1-40

Figure 1.2-12. Viscosity for HPAM polymer Injectate and Effluent ............................ 1-41
Figure 1.2-13. Pressure Drop for HPAM Injection Test ............................................. 1-41
Figure 1.2-14. Surfactant Recovery and Concentration Curves ................................ 1-42
Figure 1.2-15. Pressure Drop for Berea Core #3 with MF2 Crude ............................ 1-44
Figure 1.2-16. Oil Recovery in Berea Core #3 with MF2 Crude ................................ 1-45
Figure 1.2-17. XRD Data from Crushed MFU Well #629 Rock Sample .................... 1-47
Figure 1.2-18. Oil Recovery for MF Core #2 with MF3 Crude Oil.............................. 1-48
Figure 1.2-19. Pressure Drop Data for MF Core #2 .................................................. 1-49
Figure 2.1-1 Relationships between effluent retardation and Langmuir parameters .... 2-2
Figure 2.1-2 Relationships between retardation and CaSO4 fraction in porous medium
(porosity=0.3) .................................................................................................... 2-3
Figure 2.1-3 Adsorption on powdered dolomite of blended surfactant with different
potential determining ion ................................................................................... 2-4
Figure 2.1-4 Particle size distribution of Stepan TDA-4PO........................................... 2-5
Figure 2.1-5 Relationship between appearance of scatter beam and dominant peak
particle size........................................................................................................ 2-6
Figure 2.1-6 Adsorption of CS330: TDA-4PO (1:1) on dolomite with different surface
area (I)............................................................................................................... 2-7
Figure 2.1-7 Adsorption of CS330: TDA-4PO (1:1) on dolomite with different surface
area (II) .............................................................................................................. 2-7
Figure 2.1-8 Adsorption on powdered calcite of N67:IOS (1:1) with different Na2CO3
concentration ..................................................................................................... 2-8
Figure 2.1-9 Test of threshold concentration of Na2CO3 for the adsorption ................. 2-9
Figure 2.2-1 Optimal Salinity vs. Soap-Synthetic Surfactant Ratio Curve for N67:IOS
(4:1) ................................................................................................................. 2-11
Figure 2.2-2 Contour of Average Partition Coefficient (log10(Kaver))............................ 2-12
Figure 2.2-3 Contour of interfacial tension (log10(IFT)) (IFT: dyne/cm)....................... 2-13
Figure 2.2-4 IFT (dyne/cm) vs Salinity for Soap Surfactant Ratio = 1 ........................ 2-13
Figure 2.2-5 Contour of Viscosity (cp) ....................................................................... 2-14
Figure 2.2-6 fractional flow changes with saturation at different IFT
(Aqueous phase viscosity = Oleic phase viscosity) ......................................... 2-15
Figure 2.2-7 Profiles of surfactant and soap at 0.5PV ............................................... 2-18
1-7
Figure 2.2-8 Profiles of IFT and soap-synthetic surfactant ratio at 0.5PV .................. 2-18
Figure2.2-9 Profile of oil saturation at 0.5PV.............................................................. 2-19
Figure 2.2-10 Effluent History of Synthetic surfactant and Soap ................................ 2-19
Figure 2.2-11 Effluent History of Oil ........................................................................... 2-20
Figure 2.2-12 Oil recoveries vs. injecting brine salinities (no soap, 1800ppm polymer)2-21
Figure 2.2-13 Oil recoveries vs. injecting brine salinities (with soap, 1800ppm polymer)2-21
Figure 2.2-14 Oil recoveries vs. injecting brine salinities (no soap, 2500ppm polymer)2-22
Figure 2.2-15 Oil recoveries vs. injecting brine salinities (with soap, 2500ppm polymer)2-22
Fig 2.5-1 Dependence of optimal salinity on soap/synthetic surfactant ratio ............. 2-25
Fig 2.5-2 TC Blend surfactant forms viscous emulsions........................................... 2-25
Fig. 2.5-3 Optimal salinity of N67-3PO S / 1% Na2CO3 / x% NaCl ......................... 2-26
Fig 2.5-4 IFT and viscosity of 3% N67-3PO S / 1%AOS C2024 / 8%SBA / 1% Na2CO3
/ NaCl .............................................................................................................. 2-26
Fig 2.5-5 Solubility parameters of 3% N67-3PO S and N67-7PO S, with n-C10 at
WOR=1:1, 78 °F .............................................................................................. 2-27
Fig 2.5-6 Phase behavior of 3% N67-7PO S with 1% Na2CO3 as a function of NaCl
concentration. .................................................................................................. 2-28
Fig 2.5-7 Phase behavior of 3% N67-7PO S / 8% secondary butyl alcohol, with 1%
Na2CO3 as a function of NaCl concentration. .................................................. 2-28
Fig 2.5-8 Salinity of phase separation increases when N67-7PO S and IOS 1518 are
blended compared with used alone. ................................................................ 2-29
Fig 2.5-9 Optimal salinity of 3% total surfactant at different N67-7PO S and IOS 1518
ratio. ................................................................................................................ 2-29
Fig 2.5-10 Phase behavior of 2.4% N67-7PO S / 0.6% IOS 1518 with 1% Na2CO3 as a
function of NaCl concentration......................................................................... 2-30
Fig 2.5-11 Soap/Synthetic Surfactant Mole Ratio Correlation ................................... 2-30
Fig 2.5-12 Adsorption of NI Blend on calcite powder ................................................ 2-31
Fig 2.5-13 Phase Behavior of 0.05% NI Blend / 1% Na2CO3 / x% NaCl ................... 2-32
Fig 2.5-14 IFT of 0.05% NI Blend / 1% Na2CO3 / x% NaCl ........................................ 2-32
Fig 2.5-15 Schematic set-up of centrifuge imbibition with formation brine of surfactant
solution. ........................................................................................................... 2-33
Fig 2.5-16 Correlation between residual oil saturation and Bond number. ................ 2-34
Fig 2.5-17 Phase behavior of 0.2% NI Blend / 1% Na2CO3 / x% NaCl at WOR = 1 and 32-34
Fig 2.5-18 Oil Recovery by centrifuge imbibition in either formation brine (water
flooding) or 0.2% NI Blend / 1% Na2CO3 / x% NaCl. ....................................... 2-35
1-8
Fig. 3.1-1. Detailed diagram of heterogeneous fracture model .................................... 3-1
Fig. 3.1-2. Set –up diagram for foam mobility control experiment in fracture model..... 3-2
Fig. 3.1-3. Mechanism of affecting apparent viscosity in fracture system .................... 3-3
Fig. 3.1-4. Apparent viscosity for fracture ratio of 0.05 mm/0.15 mm, bubble size = 0.4
mm and Re = 0.22 ............................................................................................. 3-6
Fig. 3.1-5. Apparent viscosity for fracture ratio of 0.1 mm/0.2 mm, bubble size = 0.4 mm
and Re = 0.22.................................................................................................... 3-6
and Re = 0.44.................................................................................................... 3-7
and Re = 0.22 Fig. 3.1-8. Surfactant solution sweeping heterogeneous fracture,
Re = 0.22, aperture ratio = 0.05 mm/0.15 mm................................................... 3-7
Fig. 3.1-8. Surfactant solution sweeping heterogeneous fracture, Re = 0.22, aperture
ratio = 0.05 mm (top)/0.15 mm (bottom) ................................................. 3-9
Fig. 3.1-9. Foam/surfactant solution sweeping heterogeneous fracture, Re = 0.22, fg =
0.9, aperture ratio = 0.05 mm/0.15 mm, dB = 0.4 mm........................................ 3-9
Fig. 3.1-10. Comparison between the calculation and the experiment results for
foam/surfactant sweep in heterogeneous fractures with 1:3 aperture ratio ..... 3-11
Fig. 3.2-1. Set –up diagram for foam strength test ..................................................... 3-12
Fig. 3.2-2 Foam strength at different surfactant composition .................................... 3-13
Fig. 3.2-3 Foam strength at different surfactant composition and salinity .................. 3-13
Figure 4.2.1. Effect of Amott-Harvey Wettability Index on Residual Oil Saturation in

Berea Sandstone............................................................................................. 4-13
Figure 4.2.2. Effect of Wettability on Oil/Water Relative Permeability Curves (Morrow et.
al. 1973) .......................................................................................................... 4-13
Figure 4.2.3. Example Water/Oil CDC for a Water Wet Berea Sandstone (Amaefule,
1982) ............................................................................................................... 4-14
Figure 4.2.4. Effect of Wettability on the CDC for a Berea Sandstone
(Mohanty, 1983) .............................................................................................. 4-15
Figure 4.2.5. Effect of Wettability on the CDC for Three Weakly Oil-Wet to Neutral-Wet
Carbonate Rocks ............................................................................................ 4-15
Figure 4.2.6. Permeability Field (md) Used in All Test Simulations ............................ 4-16
Figure 4.2.7. Relative Permeability Curves used in the Test Simulations .................. 4-16
Figure 4.2.8. Water and Oil CDC for Case #1 ............................................................ 4-17
Figure 4.2.9. Water and Oil CDC for Case #2 ............................................................ 4-17
1-9
Figure 4.2.10. Water and Oil CDC for Case #3 .......................................................... 4-18
Figure 4.2.11. Water and Oil CDC for Case #4 .......................................................... 4-18
Figure 4.2.12. Capillary Pressure Curves for Each Test Simulation........................... 4-19
Figure 4.2.13. Waterflood – Cumulative Oil Recovery for Each Test Simulation........ 4-19
Figure 4.2.14. Waterflood – Average Reservoir Pressure for Each Test Simulation .. 4-20
Figure 4.2.15. – Waterflood – Water Production Rate for Each Test Simulation........ 4-20
Figure 4.2.16. Waterflood – Oil Production Rate for Each Test Simulation ................ 4-21
Figure 4.2.17. Hypothetical 1-D Fractional Flow Curves for Case #1......................... 4-21
Figure 4.2.18. Hypothetical 1-D Fractional Flow Curves for Cases #2 and #3 ........... 4-22
Figure 4.2.19. Hypothetical 1-D Fractional Flow Curves for Case #4......................... 4-22
Figure 4.2.20. S/P Flood - Incremental Waterflood Oil Recovery for Each Test
Simulation........................................................................................................ 4-23
Figure 4.2.21. S/P Flood - Average Reservoir Pressure for Each Test Simulation .... 4-24
Figure 4.2.22. S/P Flood - Water Production Rate for Each Test Simulation ............. 4-24
Figure 4.2.23. S/P Flood - Oil Production Rate for Each Test Simulation .................. 4-25
Figure 4.2.24. Oil Saturation at 0.2 PV of Waterflood (Case #1)................................ 4-26
Figure 4.2.25. Oil Saturation at 0.9 PV of Waterflood (Case #1)................................ 4-26
Figure 4.2.26. Pressure at 0.2 PV of Waterflood (Case #1) ....................................... 4-26
Figure 4.2.27. Pressure at 0.9 PV of Waterflood (Case #1) ...................................... 4-26
Figure 4.2.28. Oil Sat. in Layer 3 at 0.2 PV of Waterflood (Case #1) ........................ 4-27
Figure 4.2.29. Oil Sat. in Layer 3 at 0.9 PV of Waterflood (Case #1) ........................ 4-27
Figure 4.2.30. Pressure in Layer 3 at 0.2 PV of Waterflood (Case #1) ...................... 4-27
Figure 4.2.31. Pressure in Layer 3 at 0.9 PV of Waterflood (Case #1) ...................... 4-27
Figure 4.2.32. Oil Sat. at 0.25 PV of Chemical Flood (Case #1) ................................ 4-28
Figure 4.2.33. Oil Sat. at 0.75 PV of Chemical Flood (Case #1) ............................... 4-28
Figure 4.2.35. Pressure at 0.25 PV of Chemical Flood (Case #1).............................. 4-28
Figure 4.2.41. Pressure at 0.25 PV of Chemical Flood (Case #1) ............................. 4-29
1-10
Figure 4.2.44. ME Sat. at 0.25 PV of Chem. Flood (Case #1 .................................... 4-30
Figure 4.2.45. ME Sat. at 0.75 PV of Chem. Flood (Case #1) ................................... 4-30
Figure 4.2.47. Poly Conc. (wt.%) at 0.25 PV of Chemical Flood (Case #1)................ 4-30
Figure 4.2.51. ME Sat. at 0.75 PV of Chem. Flood (Case #1) .................................. 4-31
Figure 4.2.56. Surf. Conc. (vol. frac.) at 0.25 PV of Chem. Flood (Case #1) ............. 4-32
Figure 4.2.59. IFT (log(dynes/cm)) at 0.25 PV of Chem. Flood (Case #1) ................. 4-32
Figure 4.2.62. Surf. Conc. (vol. frac.) at 0.25 PV of Chem. Flood (Case #1).............. 4-33
Figure 4.2.66. IFT (log(dynes/cm)) at 0.75 PV of Chem. Flood (Case #1) ................ 4-33
Figure 4.2.68. Oil Saturation at 1.0 PV of Waterflood (Cases #2,3) .......................... 4-34
Figure 4.2.69. Oil Saturation at 3.5 PV of Waterflood (Cases #2,3) .......................... 4-34
Figure 4.2.70. Pressure at 1.0 PV of Waterflood (Cases #2,3) .................................. 4-34
Figure 4.2.71. Pressure at 3.5 PV of Waterflood (Cases #2,3) .................................. 4-34
Figure 4.2.72. Oil Sat. in Layer 3 at 1.0 PV of Waterflood (Cases #2,3) ................... 4-35
Figure 4.2.73. Oil Sat. in Layer 3 at 3.5 PV of Waterflood (Cases #2,3) ................... 4-35
Figure 4.2.74. Pressure in Layer 3 at 1.0 PV of Waterflood (Cases #2,3).................. 4-35
1-11
Figure 4.2.75. Pressure in Layer 3 at 3.5 PV of Waterflood (Cases #2,3).................. 4-35
Figure 4.2.85. Pressure at 0.25 PV of Chemical Flood (Case #2) ............................. 4-37
Figure 4.2.87. Pressure at 2.25 PV of Chemical Flood (Case # 2)............................. 4-37
Figure 4.2.88. ME Sat. at 0.25 PV of Chem. Flood (Case #2) .................................. 4-38
Figure 4.2.100. Surf. Conc. (vol. frac.) at 0.25 PV of Chem Flood (Case #2) ............ 4-40
Figure 4.2.103. IFT (log(dynes/cm)) at 0.25 PV of Chem Flood (Case #2) ................ 4-40
Figure 4.2.104. IFT (log(dynes/cm)) at 0.75 PV of Chem Flood (Case #2) ............... 4-40
Figure 4.2.105. IFT (log(dynes/cm)) at 2.25 PV of Cheml Flood (Case #2) ............... 4-40
1-12
Figure 4.2.112. Oil Sat. at 0.25 PV of Chemical Flood (Case #3) ............................. 4-42
Figure 4.2.115. Pressure at 0.25 PV of Chemical Flood (Case #3)............................ 4-42
Figure 4.2.121. Pressure at 0.25 PV of Chemical Flood (Case #3) ........................... 4-43
Figure 4.2.124. ME Sat. at 0.25 PV of Chem. Flood (Case #3) ................................. 4-44
Figure 4.2.127. Poly Conc. (wt.%) at 0.25 PV of Chemical Flood (Case #3).............. 4-44
Figure 4.2.132. ME Sat. at 2.25 PV of Chem. Flood (Case #3) ................................ 4-45
Figure 4.2.133. Poly Conc. (wt.%) at 0.25 PV of Chem Flood (Case #3) ................... 4-45
Figure 4.2.134. Poly Conc. (wt.%) at 0.75 PV of Chem Flood (Case #3) .................. 4-45
Figure 4.2.135. Poly Conc. (wt.%) at 2.25 PV of Chem Flood (Case #3) ................... 4-45
1-13
Figure 4.2.142. Surf. Conc. (vol. frac.) at 0.25 PV of Chem Flood (Case #3)............. 4-47
Figure 4.2.143. Surf. Conc. (vol. frac.) at 0.75 PV of Chem Flood (Case #3)............. 4-47
Figure 4.2.148. Oil Saturation at 0.3 PV of Waterflood (Case #4) ............................. 4-48
Figure 4.2.149. Oil Saturation at 1.9 PV of Waterflood (Case #4) ............................. 4-48
Figure 4.2.150. Pressure at 0.3 PV of Waterflood (Case #4) ..................................... 4-48
Figure 4.2.151. Pressure at 1.9 PV of Waterflood (Case #4) ..................................... 4-48
Figure 4.2.152. Oil Sat in Layer 3 at 0.3 PV of Waterflood (Case #4) ....................... 4-49
Figure 4.2.153. Oil Sat in Layer 3 at 1.9 PV of Waterflood (Case #4) ....................... 4-49
Figure 4.2.154. Pressure in Layer 3 at 0.3 PV of Waterflood (Case #4)..................... 4-49
Figure 4.2.155. Pressure in Layer 3 at 1.9 PV of Waterflood (Case #4)..................... 4-49
Figure 4.2.165. Pressure at 0.25 PV of Chemical Flood (Case #4) ........................... 4-51
1-14
Figure 4.2.174. ME Sat. at 0.25 PV of Chem. Flood (Case #4 .................................. 4-53
Figure 4.2.183. Surf. Conc. (vol. frac.) at 2.25 PV of Chem. Flood (Case #4) ........... 4-54
Figure 4.2.189. Surf. Conc. (vol. frac.) at 2.25 PV of Chem. Flood (Case #4) ........... 4-55
1-15
INTRODUCTION
Oil recovery by primary depletion and waterflooding recovers only

about one third of the original in place, on the average. 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 in the
reservoir simulator, UTCHEM.
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 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 being developed for enhanced spontaneous
imbibition in a fractured, oil-wet, carbonate formation. 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 sponification 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 two surfactants with very different properties.
This results in the optimal salinity that depends on the water/oil ratio and
surfactant concentration. However, these dependencies can be correlated by the
ratio of natural surfactant/synthetic surfactant.
Surfactant retention by adsorption and phase trapping determine 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.
1-17
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. Foam has the potential to improve
the liquid distribution in fractured systems. The volumetric flow rate of a
Newtonian liquid into a set of parallel fractures is proportional to the third power
of the fracture width. This will result in large fractures acting as “thief zones” and
small fractures being bypassed. For bubbles of the same size, the apparent
viscosity is higher in larger fractures compared to smaller fractures. We have
verified that the model developed for foam in capillaries can be extended to flow
between parallel plates. We demonstrated that better sweep efficiency is
possible with the use of foam.
The reservoir simulator, UTCHEM will be used as the tool to scale-up from
laboratory experiments to field design. 3-D simulations with UTCHEM
demonstrate the effect of changing wettability on oil recovery.
1-18
1.0 Task 1. Improved Surfactants and Formulations
The following series of steps were taken to identify promising surfactants and
surfactant EOR formulations:
• Use knowledge of surfactant chemistry and commercial surfactant
production capabilities to identify prospective surfactant test candidates
(Subtask 1.1)
• Acquire samples from surfactant companies and screen surfactants using
phase behavior experiments with alkanes (Subtask 1.1)
• Use phase behavior experiments to optimize formulations of surfactants,
co-surfactants and co-solvents with crude oils (Subtask 1.2)
• Conduct laboratory core flood experiments using the best surfactant
formulations from the phase behavior screening data (Subtask 1.2)
Based upon the surfactant selection and phase behavior screening procedure
discussed in this report, a core flood was performed using a very heterogeneous
dolomite field core and crude oil from the Midland Farms reservoir. The oil
recovery was 95% and the final oil saturation only 0.01. The results of this core
flood experiment indicate that the surfactant selection procedure used in this
research is based upon sound scientific principles and leads to excellent
performance as well as being very efficient.
1.1Subtask 1.1. Identifying and Synthesizing Improved, Cost-effective
Surfactants
The research in Year 1 identified several families of propoxylated sulfate

surfactants that were suitable for enhanced oil recovery both from standpoints of
both performance and the ability to be commercially manufactured. Specifically,
branched alcohol propoxy sulfate surfactants were synthesized for screening with
both pure hydrocarbons and crude oils. The isotridecanol (C13 Exxal® TDA) was
considered to be most suitable hydrophobe in terms of phase behavior for pure
hydrocarbons ranging between hexane and decane. Additionally, the TDA is a
low-cost commodity alcohol that would minimize total surfactant cost.
Formulations with this C13PO-sulfate exhibited high optimal salinities and
thus would work well in high salinity environments. Another property inherent to
surfactants with sulfate ester hydrophilic groups is their susceptibility to
hydrolysis at elevated temperatures. Therefore, additional surfactant
development and screening were undertaken to find candidates for high-
temperature and low-salinity reservoirs. This involved the testing of alcohol
propoxy sulfate surfactants with larger hydrophobic groups (to lower the optimal
salinities) and the screening of temperature-stable sulfonate surfactants
containing desirable branched hydrophobes. This follow-on screening for
surfactants with lower optimal salinities plus resistance to high temperature
degradation utilized surfactants identified in Year 1 in addition to ones that were
1-19
specially designed in Year 2. Table 1.1-1 below lists the surfactants used in Year
2 screening.
Table 1.1-1 Surfactants Tested in Year 2

Descriptive or Trade Name Abbreviated Chemical Formula
(supplier) (b = branching in the carbon chain)
Amphosol 810-B (octyl&decyl-
amidopropyl betaine) (Stepan) C7-9CO-NH-(CH2)3-N+(CH3)2-CH2COO-
Amphosol HCG (coco-amidopropyl
betaine) (Stepan) C7-11CO-NH-(CH2)3-N+(CH3)2-CH2COO-
C14 Alpha Olefin Sulfonate (AOS) bC11-CH(OH)-CH2-CH2-SO3- (~75%)
(Stepan) bC11-CH=CH-CH2-SO3- (~25%)
C16-(C8H12)-SO3-
C16 o-Xylene Sulfonate (Shell) where C8H12 = o-xylene
C16-18 Alpha Olefin Sulfonate (AOS) bC13-15-CH(OH)-CH2-CH2-SO3- (~75%)
(Stepan) bC13-15-CH=CH-CH2-SO3- (~25%)
C20-24 Alpha Olefin Sulfonate (AOS) bC17-21-CH(OH)-CH2-CH2-SO3- (~75%)
(Stepan) bC17-21-CH=CH-CH2-SO3- (~25%)
C16-18 Alcohol Propoxy Sulfate
(HN1618-5.7POS) (Harcros) C16-18(CH3-CH-CH2-O-)5.7-SO3-
Hostapur Secondary Alkane R-CH(SO3-)-R’
Sulfonate (SAS-60) (Clariant) where R + R’ = C14-C17
R-CH(OH)-CH2-CH(SO3-)-R’ (~75%)
C15-18 Internal Olefin Sulfonate (IOS) R-CH=CH-CH(SO3-)-R’ (~25%),
(Shell) where R+R’ = C12-15
Neodol® N67-3POS (C16-17 (PO)3
APS) Alcohol Propoxy Sulfate bC16-17(CH3-CH-CH2-O)3-SO3-
(Stepan)
(Stepan)
(Stepan)
ORS-62HF (Oil Chem Tech.) Unknown (not provided)
ORS-66HF (Oil Chem Tech.) Unknown (not provided)
Petrostep® SB (coco amidopropyl C7-11CO-NH-(CH2)3-N+(CH3)2-
hydroxyl sultaine (Stepan) CH2CH(OH)-CH2-SO3-
Polystep® A-16-22 (Branched Alkyl
Benzene Sulfonate) (Stepan) bC11-13(C6H5)-SO3-
XSA 1416 (Oil Chem Tech.) Unknown (not provided)
XSA 2024 (Oil Chem Tech.) Unknown (not provided)
1-20
1.1.1 Surfactant Screening using Phase Behavior Experiments
Surfactant screening in Year 2 followed the same process of phase behavior

testing as in Year 1 and involved the quantification of the low IFT microemulsions
formed from test oils and the surfactants (either individually or in combinations).
Promising surfactant formulations were tested with pure alkane hydrocarbons as
a first step, followed by a more detailed study of the better-performing surfactants
with selected crude oils. The surfactant screening tests for Year 2 can be divided
into two groups; one for surfactant formulation testing below 50°C and the
second set for testing above 50°C. The temperature of the phase behavior
testing reflected the actual temperatures of the reservoirs from which the crude
oils were obtained. Another issue that was addressed in the Year 2 testing was
the requirement for low viscosity phases, particularly at low temperatures. This
requirement pointed to the need for highly branched hydrophobes or the mixture
of hydrophobic structures such that the tendency to form liquid crystals and gels
was minimized.
1.1.2 Results and Discussion for Subtask 1.1

In order to reduce optimum salinity the size of the hydrophobic group
needed to be larger than the C13 TDA used in previous alcohol propoxylate
sulfate surfactants. A new Shell alcohol, Neodol® 67, (slightly branched C16-C17
primary alcohol) was identified, and a set of N67-(PO)n-SO4 (where n = 3, 5, or 7)
surfactants were prepared by Stepan Company. These three new surfactants
were designated N67-3POS, N67-5POS and N67-7POS or generically as C16-17-
(PO)3 -SO4, C16-17-(PO)5-SO4 and C16-17 -(PO)7-SO4 in this report. A space-filling,
molecular model of C16-17 -(PO)7-SO4 without its hydrogen atoms is shown in
figure 1.1-1.
1-21
Figure 1.1-1. Neodol® N67- 7PO-sulfate (C16-17 -(PO)7-SO4)
The minimal energy image shows that the molecule is very compact because the
large propylene oxide (PO) part of the molecule folds into a spiral. The
hydrophobe here is shown with 16 carbon atoms and 3 methyl branches. Both
the PO and the branching result in a molecule that is very different from the
chemically similar, linear alcohol ethoxy sulfate surfactants used as laundry
detergents. Linear hydrophobes tend to form ordered structures that lead to
undesirable liquid crystals and gels. Also, POs, in contrast to EOs, are larger
and more balanced at the interface between the water and oil. This leads to
lower IFT. Changing the number of POs alters optimum salinity and thus can be
used to tailor the surfactant's performance. Phase behavior testing with pure
hydrocarbons predictably has shown that the optimum salinity is lower than those
with the C13 TDA hydrophobe. Finally, the presence of propoxyl groups in the
surfactant molecule enhances calcium tolerance. The only negative aspect of
this molecule is the sulfate hydrophilic group, which is susceptible to hydrolysis at
the elevated temperatures of many reservoirs. This surfactant is expected to cost
approximately $2/lb (100% active), which is very reasonable for a high
performance surfactant.
The Year 2 work also examined sulfonate surfactants for higher temperature
reservoirs. Three custom-made alpha-olefin sulfonate (AOS) surfactants were
screened: (1) C14-16 AOS, (2) C16-18 AOS and (3) C20-24 AOS. A model of the C20-
24 AOS is shown below.
1-22
Figure 1.1-2. C20-24 AOS
The image shows a 22-carbon alpha olefin sulfonate with some typical methyl
branching. The actual surfactant has 20 to 24 carbon atoms and a variable
number and positioning of methyl branches and either a double bond or a
hydroxyl group (as shown in figure 1.1-2) at a position located two carbon atoms
from where the sulfonate group is located. If linear, it would make a poor
surfactant for EOR.
Additionally, a C15-18 internal-olefin sulfonate (C15-18 IOS) surfactant made by

Shell was identified as a potential co-surfactant for use with both high
temperature AOS surfactants and low temperature alcohol propoxylate sulfate
surfactants. A model of the hydroxyl form of the molecule is shown in figure 1.1-
3. This IOS tends not to form liquid crystals and gels, thereby reducing the need
for alcohol (co-solvent) addition. The figure below depicts a structure with “twin
hydrophobe tails”. However, this is just one of the many sulfonate positional
isomers that can exist depending on the location of the internal double bond of
the starting internal olefin feedstock. This has the same overall effect as with the
branched alcohol-PO sulfate surfactants by negating the orderly alignment
among molecules to form liquid crystals and gels.
1-23
Figure 1.1-3. C15-18 IOS
Both the AOS and IOS surfactants are inexpensive in part because they are
made in a one step synthesis. Both surfactants are estimated to cost only about
$1 per lb (100% active), which is very low for a high performance EOR
surfactant.
Surfactants or surfactant blends also have been provided by Oil Chem

Technologies (OCT) for laboratory testing (see Table 1.1-1). These are custom
surfactants manufactured for EOR projects. No detailed surfactant-structure
information was provided by OCT other than that shown in Table 1.1-1.
Progress for Year 2 can be summarized as focused efforts to find surfactants

with the desired phase behavior. The synthesis and testing of a group of alcohol
propoxy sulfates with the larger Neodol® 67 alcohol resulted in the desired
behavior for low-temperature EOR applications. The synthesis and testing of
novel AOS and IOS surfactants resulted in the desired behavior for high
temperature EOR applications.
1.2 Subtask 1.2. Surfactant Tailoring for Crude Oils and Phase
Behavior
Surfactant tailoring for crude oils in Year 2 was conducted in a series of steps.
First, the phase behavior experiments were used to narrow possible surfactant
candidates in an efficient and effective manner. Second, polymers and mixtures
1-24
of surfactants and polymers were tested to select suitable polymers for core
flooding tests. Next, experiments were conducted to characterize reservoir rock
cores prior to core flood experiments. Finally, these processes were integrated
into core flood experiments to evaluate the best surfactants for oil recovery for
the particular crude oils provided by oil companies.
1.2.1 Results of Phase Behavior Experiments
One crude oil was from the Midland Farms (MF) reservoir in Andrews County TX
and was provided by Oxy Permian, the operator of the reservoir. Surfactants
were evaluated by measuring their aqueous solubility, solubilization parameters,
coalescence time and the viscosity of the aqueous and microemulsion phases.
This process includes determining how much alcohol co-solvent is needed for
acceptable phase behavior and compatibility with polymers. Suitable formulations
were then evaluated in core floods to assess oil recovery and surfactant
adsorption on the rock. Importantly, pressure was monitored in the core flood
experiments since high pressure gradients are not practical in reservoirs and
would indicate problems with the use of the surfactant or polymer.
Surfactant formulations were screened at the actual 38oC reservoir temperature.

Table 1.2-1 lists the various formulations evaluated with the MF crude oil. A
formulation that showed excellent middle phase microemulsions in phase
behavior experiments with octane and decane was tested with the MF crude oil.
This formulation contained 2 wt% isotridecyl-3PO-sulfate, 2 wt% branched-
C12(Guerbet)-3PO-sulfate and 1 wt% sodium dihexyl-sulfosuccinate (Aerosol®
MA-80). However, a high electrolyte concentration on the order of 12 wt% brine
was required to induce middle phase microemulsions. Such a high optimum
salinity necessitated the switch to the hydrophobic C20-24 alpha olefin sulfonates
and the Neodol® N67 propoxylate sulfate surfactants.
1-25
Table 1.2-1. Formulations Screened with Midland Farms Crude Oil at
38° C
Optimum Equilibration
Optimum Salinity
Surfactant Formulation Sol. Ratio, Time at S*
S* (wt% NaCl)
σ (cc/cc) (days)
2% bC12-(PO)3-SO4, 2% TDA-(PO)3-SO4,
7.5 3.5 <2
1% dihexyl-sulfosuccinate
4% C20-24 AOS, 8% SBA 2.5 6 5
4% N67-(PO)3-SO4 (NH4), 8% SBA >3 5 9
4% N67-(PO)5-SO4 (NH4), 8% SBA 1.9 7 2
4% N67-(PO)7-SO4 (NH4), 8% SBA 1.75 7 2
0.75 % N67-(PO)7-SO4, 0.25% C20-24
>2 5 5
AOS, 2% SBA
1.5 % N67-(PO)7-SO4 , 0.5% C15-18 IOS,
3.8 6 1
1% SBA, 1% Na2CO3
1.5 % N67-(PO)7-SO4, 0.5% C15-18 IOS,

3.5 12 2
1% Na2CO3
1.5 % N67-(PO)7-SO4, 0.5% C20-24 AOS,

1.6 6.1 1
4% SBA, 1% Na2CO3
1.5 % N67-(PO)7-SO4, 0.5% C15-18 IOS gels gels gels
0.325 %N67-(PO)7-SO4, 0.125% C15-18 Did not Did not
Did not equilibrate
IOS, 0.5% SBA equilibrate equilibrate
1.5 % N67-(PO)7-SO4, 0.5% C20-24 AOS,
1.7 11 1
4% SBA, 1% Na2CO3, 0.125% HPAM
0.75 % N67-(PO)7-SO4, 0.25% C15-18 IOS,
4.5 9 7
2% SBA
0.75 % N67-(PO)7-SO4, 0.25% C15-18 IOS,
3.4 12 16
2% SBA, 1% Na2CO3
The most favorable phase behavior was created with the Neodol® N67
propoxylated sulfate surfactants containing 7 moles PO. These experiments
showed that increasing the number of POs results in higher solubilization and a
lower optimum salinity. Gels were observed when SBA was greatly reduced or
eliminated from the formulation. The addition of the C15-18 internal olefin
sulfonate co-surfactant allowed for the reduction in SBA. Also, the C15-18 IOS
performed better than the C20-24 AOS when diluted to 1 wt% total surfactant.
The best phase behavior in terms of coalescence and high solubilization ratio at
optimum salinity (an indicator of low IFT) was observed with formulations
containing the N67-7PO-SO4, C15-18 IOS and SBA. These experiments
demonstrated that the formulation containing 0.75% Neodol67-7PO-SO4, 0.25%
C15-18 IOS and 2% SBA produced rapid coalescence and high solubilization
1-26
ratios. This surfactant formulation was chosen for flooding of using Midland
Farms reservoir core and is presented in the “Results of Core Flood
Experiments” section below.
Surfactants were also evaluated with crude oil from the Elk Hills (ELK) reservoir
in California, which is operated by OXY Elk Hills. Surfactant formulations were
screened at the 100°C reservoir temperature, and the results are presented in
Table 1.2-2. Candidate surfactant formulations were tested in the same way as
with MF crude by measuring aqueous solubility, solubilization parameters,
coalescence times and viscosities of the aqueous and microemulsion phases.
Table 1.2-2. Formulations Screened with Elk Hills Crude Oil, 100°C
Optimum
Optimum Equilibration
Salinity S*
Surfactant Formulation Sol. Ratio, Time at S*
(wt%
σ (cc/cc) (days)
NaCl)
1.5% C15-18 IOS, 0.5% C20-24 AOS, 2% SBA 6-9 4 1
0.75% C15-18 IOS, 0.25% C20-24 AOS 9 - 11 4 4
1.5% SAS-60, 0.5% C20-24 AOS 5-7 3 3, until gel
2% C20-24 AOS, 4% SBA 2.1 8.5 1
1% C20-24 AOS, 1% C15-18 IOS 6-8 3 5
Did not Did not Did not

1.5% C16 o-Xylene sulfonate , 0.5% C20-24 AOS
equilibrate equilibrate equilibrate
0.5% C20-24 AOS, 1% SBA 2.1 9 1
0.5% C20-24 AOS, 1% SBA, 1% Na2CO3 1.2 14 1
Did not Did not Did not

0.05% C20-24 AOS, 0.05% C15-18 IOS, 0.2% SBA
equilibrate equilibrate equilibrate
2% C20-24 AOS, 4% SBA, 1% Na2CO3 1.4 12 1
The sulfate surfactants are not stable at 100°C, so sulfonate surfactants were
used for the Elk Hills study. Alcohol was added to decrease the equilibration
time and microemulsion viscosity. The experiments using only C20-24 AOS and
SBA were the most promising because they showed the highest solubilization
without gelling after short equilibration times and with sharp and distinct
interfaces between the microemulsion and the excess water and excess oil.
1-27
Effects of Electrolyte Concentration. Phase behavior testing seeks to establish
the salinity where the Winsor Type III middle phase microemulsions are largest
(i.e. maximum volume oil solubilized per volume surfactant). If Vo is the volume of
solubilized oil in the middle phase, Vw the volume of solubilized water and Vs the
volume of neat surfactant, then the solubilization ratios Vo/Vs and Vw/Vs are the
volume of oil and water per unit volume of surfactant in the microemulsion phase.
The ratio Vo/Vs increases and the ratio Vw/Vs decreases with salinity. When
these ratios are plotted, the intersection point within the Type III salinity range is
the optimum solubilization ratio at the optimum salinity as shown in Tables1.2-1
and 1.2-2. This optimum solubilization ratio corresponds to the lowest IFT, which
is the desired condition for mobilizing oil in EOR. Thus, phase behavior testing
also indicates the electrolyte concentrations needed to produce the lowest IFT
and maximum oil recovery for EOR. Optimum solubilization ratios for specified
oils will vary for different surfactants and their mixtures. High values are
noteworthy because they denote surfactants and conditions that produce ultra-
low IFT. However, as pointed out earlier in this report, a high optimum
solubilization ratio is not sufficient for acceptable behavior and high oil recovery.
The absence of viscous phases such as gels, liquid crystals and macroemulsions
and short equilibration times are equally important. The behavior of neighboring
Type I and II regions also are evaluated for viscosity of the aqueous and
microemulsion phases. This information is particularly important in developing
the salinity gradient used for core flood experiments and later application in the
field.
Figure 1.2-1 shows the salinity scan for a mixture containing N67-7PO-SO4, C15-
18 IOS and SBA with MF crude at 38°C. The surfactant mixture included 0.02
wt% sodium carbonate as a buffer to keep the stock solution at about pH=9.
The optimum salinity recorded 21 days after mixing occurred at 1.65 wt% Na+
concentration in the water with a solubilization ratio of 12. Figure 1.2-2 shows the
optimum solubilization ratio for the C20-24 AOS (2%) at 100oC with Elk Hills crude
oil was 9 at 0.85% Na+ concentration in the water. Generally speaking, optimum
solubilization ratios that are ~10 or higher indicate the preferred surfactant(s) and
chemical conditions for use in core floods. These figures show just some of
hundreds of phase behavior experiments that have been conducted to screen
surfactants and to establish electrolyte conditions for the core flood experiments.
1-28
24
0.75% C16-17 (PO)7 SO4, 0.25% C15-18 IOS, 2% SBA, 0.02% Na2CO3
20
Oil
16 Water
Solublization Ratio
MF3 Crude Oil

12 WOR = 1
21 days after mixing
0
0.0 0.5 1.0 1.5 2.0 2.5 3.0
Sodium Ion Concentration (wt%)
Figure 1.2-1. Salinity Scan with MF3 Crude Oil at 38°C
20
2% C20-24 AOS, 4% SBA
18
16
Oil
14
Water
Solublization Ratio
12
Elk Hills Crude
10 WOR = 1
3 days after mixing
8
0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
Sodium Ion Concentration (wt%)
Figure 1.2-2. Salinity Scan with Elk Hills Crude Oil at 100°C
1-29
Effects of Sodium Carbonate. Crude oils contain napthenic acids that may
saponify upon the addition of sodium carbonate and generate an in-situ soap
(surfactant) and is the basis for the so called ASP process. The addition of
sodium carbonate is also desirable because it reduces anionic surfactant
adsorption in reservoir rocks, allowing a smaller mass of surfactant to be
injected. Additionally, the higher pH resulting from the addition of sodium
carbonate helps maintain the stability of some of the surfactants being used.
However, the interaction between each crude oil and sodium carbonate is unique
and must be tested. Some crude oils have a low acid number and may not form a
significant amount of soap, but the other effects may still be significant.
However, even if the sodium carbonate has a beneficial effect on the phase
behavior, it must still be tested with the reservoir core because of chemical
reactions with the minerals in the rock.
A formulation using N67-(PO)7 SO4, C15-18 IOS and SBA with MF3 crude oil was
evaluated by testing with 0.02%, 1.0% and 2.0% sodium carbonate. The
measurements of oil and water solubilization with respect to time are shown in
figure 1.2-3.
24
22 Oil (0.02% Na2CO3, 1.6% Na+) Water (0.02% Na2CO3, 1.6% Na+)
20 Oil (1.0% Na2CO3, 1.9% Na+) Water (1.0% Na2CO3, 1.9% Na+)
Oil (2.0% Na2CO3, 2.0% Na+) Water (2.0% Na2CO3, 2.0% Na+)
18
16
Solubilization Ratio
14
12
10
8
6
4
2
0
0 2 4 6 8 10 12 14 16 18 20 22 24
Time (days)
Figure 1.2-3. Effect of Sodium Carbonate on MF3 Crude Oil at a WOR 1:1
and 38°C using a mixture of 0.75% N67-(PO)7-SO4, 0.25% C15-18 IOS, 2% SBA
All the phase behavior tubes were observed just below optimum salinity. A
comparison of the three experiments near optimum shows that the optimum
1-30
solubilization ratio remains about 12 for all three cases. This means the addition
of more than 0.02% sodium carbonate is unnecessary for the MF3 crude oil. The
figure does show a quicker equilibration for the samples at 1% and 2% sodium
carbonate, indicating that the phase behavior equilibration time is reduced at
higher carbonate concentrations. Overall, the sodium carbonate has no
substantial effect on the solubilization of Midland Farms crude, but may be used
to help lower surfactant adsorption and speed up equilibration times.
A formulation using C20-24 AOS and SBA with Elk Hills (ELK) crude oil was
evaluated at both 0.02% and 1.02% sodium carbonate. The observed effect was
a change in the optimum salinity, solubilization ratio and equilibration time.
Figure 1.2-4 shows the salinity scan for a C20-24 AOS and SBA formulation with
ELK crude at 100°C and the shift resulting from adding sodium carbonate.
25
Oil Sol Ratio After 21 Days
(0.02% Na2CO3)
20 Water Sol Ratio After 21

Days (0.02% Na2CO3)
Oil Sol Ratio After 13 Days
(1.02% Na2CO3)
15
Water Sol Ratio After 13
Days (1.02% Na2CO3)
10 2% C20-24 AOS, 4% SBA
0
0.0 0.5 1.0 1.5 2.0
Sodium Concentration (wt %)
Figure 1.2-4. Effect of Sodium Carbonate on Optimal Salinity with ELK

Crude, WOR 1:1 and 100°C
These phase behavior tests were measured at their optimum salinities. The
recorded measurements of oil and water solubilization ratios with respect to time
are shown in Figure 1.2-5. A comparison of the two results shows that the
solubilization ratio increases from about 9 to 12 with the addition of 1.02%
sodium carbonate. The addition of sodium carbonate also helps reduce the
equilibration time of the sample from about 90 days to about 12 days. These
results show that the addition of sodium carbonate to the ELK crude oil could
lead to a higher oil solubilization ratio and lower interfacial tension and quicker
1-31
equilibration times. It would very likely also result in decreased surfactant
adsorption on the reservoir rock, but adsorption tests have not been done.
20
Oil (0.0% Na2CO3, 0.8% Na+)
Oil (1.02% Na2CO3, 1.0% Na+)
Water (0.0% Na2CO3, 0.8% Na+)
15
water (1.02% Na2CO3, 1.0% Na+)
Elk Hills Crude

Temp. = 100 °C
10
0
0 20 40 60 80 100
Time (days)
Figure 1.2-5. Effect of Sodium Carbonate on Equilibration Time with Elk

Hills Crude, WOR 1:1, 100°C, 2% C20-24 AOS, 4% SBA
Effects of sec-Butyl Alcohol (SBA). Phase behavior experiments done with

varying concentrations of sec-butyl alcohol (SBA) in order to observe its effects
on separation time, optimal salinity and optimum solubilization ratio. The results
are presented below in Table 1.2-3. N67-7PO-SO4 and C15-18 IOS were selected
as surfactant and co-surfactant respectively based on good results from previous
phase behavior experiments with Midland Farms crude oil batch #3 at 38°C.
1-32
Table 1.2-3. Effects of SBA and Sodium Carbonate on Phase Behavior
Optimum Optimum
Surf Co-surf SBA Na2CO3 Separation
Salinity, S* Sol. Ratio,
(wt%) (wt%) (wt%) (wt%) Time (days)
(wt% Na+) σ* (cc/cc)
1.50 0.50 0 0 gel d.n.e. d.n.e.
1.50 0.50 2 0 1 1.6 12
0.75 0.25 2 0 2 1.7 12
1.50 0.50 0 1 4 1.8 13
0.75 0.25 2 1 2 1.7 13
1.50 0.50 4 1 1 1.9 5
0.75 0.25 2 2 2 2.0 13
d.n.e. = Did not equilibrate; WOR = 1:1
The phase behavior results in this table illustrate the trade-offs between phase
separation time and optimum solubilization ratio when alcohol is added to the
formulation. The addition of alcohol prevented gel formation and shortened
separation times to 1 or 2 days. As SBA concentration increased to 4%,
optimum solubilization ratios dropped significantly from 12 to 5. These and other
results show that adding SBA reduces equilibration time and viscosity but at the
expense of reducing the solubilization ratio at optimum salinity. The addition of
up 2% SBA to the surfactant mixture with N67-7PO-SO4 and C15-18 IOS produced
reasonable equilibration times while maintaining a sufficiently high solubilization
ratio of about 12. Also, alcohols such as SBA can make the formulation more
robust by giving low IFT across a wider range of electrolyte concentrations and is
sometimes needed as well to prevent incompatibility between the polymer and
surfactant.
1.2.2 Results of Polymer Testing
Mobility control is needed for an efficient surfactant flood to prevent fingering,

reduce the effects of channeling caused by permeability layering and also to
improve displacement sweep efficiency and lower surfactant retention. Either
polymer must be added to the surfactant solution or foam must be used for
mobility control. Both approaches are being investigated as part of this research.
In this section, the results of polymer tests are presented. The only two
commercially available, low-cost water-soluble polymers currently being used in
EOR are xanthan gum and partially hydrolyzed polyacrylamide (HPAM). Each of
these polymers has well known advantages and limitations. We report both a
core flood experiment using xanthan gum polymer in Berea sandstone and a
core flood experiment using HPAM in a dolomite core from the MF reservoir.
Preliminary experiments were conducted to verify performance of HPAM for use

with a surfactant formulation consisting of 0.75% N67-7PO-SO4, 0.25% C15-18
IOS and 2% SBA. There was no substantial effect on the microemulsion phase
behavior at 2000 mg/L HPAM. Next, it was necessary to evaluate viscosity as a
1-33
function of salinity. Figure 1.2-6 shows the relationship between viscosity and
HPAM concentration at 4% NaCl, 69.5 sec-1 shear rate and 38 C. Figure 1.2-7
shows the relationship between viscosity of the HPAM polymer and salinity.
Increasing salinity causes the viscosity to decrease rapidly until about 3% NaCl
after which it decreases only slowly. The range of interest for Midland Farms is 3
to 5 wt% NaCl and the viscosity change was small in this range of salinity. At
1500 ppm polymer concentration, the viscosity of an aqueous solution in this
range of salinity is about 5 cp, which was deemed adequate for the purposes of
testing the candidate surfactant formulation in MF cores since some permeability
reduction was also expected and observed. The viscosity is also a function of
shear rate. Figure 1.2-8 shows that this polymer under these conditions is only
mildly shear thinning and taking a typical reservoir or core flooding shear rate of
69.5 sec-1 as a benchmark value is more than adequate to characterize it for core
flooding.
12
Salinity = 4% NaCl
10 Temperature = 38 °C
-1
Shear Rate = 69.5 sec
8
Viscosity (cp)
0
0 500 1000 1500 2000 2500
HPAM Concentration (ppm)
Figure 1.2-6. Viscosity of HPAM Polymer (Flopaam 3330S) in Brine
1-34
40
1500 ppm Flopaam 3330S
35 Temperature = 38 °C
-1
Shear Rate = 69.5 sec
30
25
Viscosity (cp)
20
15
10
0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
Salinity, wt% NaCl
Figure 1.2-7. Viscosity for HPAM Polymer (Flopaam 3330S) in Brine
100
Salinity = 4% NaCl
Temperature = 38 °C
Viscosity (cp)
10
1
0.01 0.1 1 10 100 1000
-1
Shear Rate (sec )
Figure 1.2-8. Viscosity for HPAM Polymer (Flopaam 3330S) in Brine
1-35
1.2.3 Results of Experiments to Characterize Rock Cores
Berea rock is an outcrop sandstone commonly used for EOR core flooding and
serves both as a useful benchmark because of the extensive amount of
experimental data available for it, including previous studies by this team, and
because it is serves as a surrogate for medium permeability sandstone reservoir
rock. Furthermore, it can be easily obtained in large quantities and longer
samples can be used for testing than with reservoir cores. For these and other
reasons, it was used to evaluate one of the more promising surfactant
formulations as a first step before reservoir cores were used.
OXY Permian provided core samples from the Midland Farms reservoir for us to
evaluate surfactants in core floods. The MF reservoir is a dolomite formation
with characteristics very different from Berea sandstone so considerable
characterization measurements were needed before it could be used in
surfactant core flooding. Using dolomite cores was much more difficult and
complicated than testing in sandstones. Very little surfactant EOR data have
been reported for carbonate or dolomite reservoir rocks, so this study also
represented a more significant innovation. Also, the cooperation of OXY
geologists and engineers was needed and very significant in this study.
Berea core preparation. Uniform Berea blocks were drilled to make two-inch
diameter cylindrical cores and typically one foot in length. Due to previous
experience with our stock of Berea blocks no further characterization tests were
performed to examine heterogeneity. Each core was fitted with specially
machined, plastic end pieces with 1/32 in. headspaces and then potted in epoxy
contained within a Lexan outer sleeve. Each core was saturated with brine under
vacuum and then flooded with synthetic formation brine. The pressure drop
across the core was measured to obtain a baseline value, and pressure drop
readings vs flow rate were used to calculate the brine permeability for the core.
Dolomite core testing and preparation. MF core samples were potted in epoxy
resin by the same procedures as for Berea sandstone cores. Carbonate and
dolomite rocks often have local, small-scale irregularities such as vugs, fractures
and highly cemented, low permeability nodules that affect flow. These features
do not significantly impact large-scale processes where fluids can move around
them, but in a core plug any of these features potentially can result in failed
experiments. It was therefore important to screen the rocks for properties
affecting their permeability and pore volume.
Air permeability tests. The MF core samples received from OXY were tested
with a mini air permeameter as a first step in a core selection process. Cores
with verifiable and consistent air permeability values were selected for further
testing. A summary of selected data are presented in Table 1.2-4.
1-36
Table 1.2-4. Summary of MF Core Permeability
Location Air Permeability
(Sample I.D.) Average (md) Standard Deviation (md)
MFU629/4823A 41 33
MFU629/4824A 195 109
MFU637/4834D 224 120
MFU637/4838B 61 23
MFU637/4831A 15 10
The core flood experiments were originally intended to use outcrop dolomite
samples provided by OXY so that larger samples could be used. When air
permeability measurements were performed on these outcrop samples, very low
permeability values that were not representative of the oil reservoir were
observed (see Table 1.2-5). Therefore, subsequent core flood experiments used
actual field cores rather than outcrop samples.
Table 1.2-5. Summary of Outcrop Dolomite Rock Permeability

Location Air Permeability
(Sample I.D.) Average (md) Standard Deviation (md)
Boulder 1 3.7 0.7
Stone Canyon 2 3.2 1.3
Stone Canyon 3 1.2 0.2
High Resolution X-ray Computed Tomography (HRXCT) scans. CT scans were

conducted on MF core plugs that passed the screening for sufficiently high air
permeability. HRXCT scans were performed by Dr. Richard Ketchum at the UT
CT laboratory. Heterogeneity created from vugs, fractures and low permeability
nodules were noted in several cores. The CT data was used to select reservoir
core plugs to use in core flooding experiments. Figure 1.2-9 displays a few of the
many images of a 6 cm long x 3.8 cm diameter core plug. In these images,
lighter areas indicate denser materials with less porosity whereas vugs and high
porosity appear as dark spots or darkened areas. Generally permeability and
porosity are correlated although in a very complicated way that varies
significantly from one rock type to another. Nevertheless, these images were
very useful in understanding the rock and selecting samples for core flooding.
1-37
Figure 1.2-9. CT Scans of MF54 Dolomite Core
Low permeability nodules can be seen in images A, B, and D, but these are
isolated and do not propagate through the entire length of the core. The overall
cross-sectional area of these nodules is not large and probably would not greatly
affect core permeability.
Images from a CT scan for a second core are shown in Figure 1.2-10. This core
was scanned both vertically and horizontally allowing us to see changes along
the entire length of the core. The images show this core to have more
permeable ends and a less permeable middle. The top of the core (Image A)
also appears to have lots of vugs. Core plugs with less extreme heterogeneities
were marked for use in further tests leading up to a surfactant core flood.
1-38
Figure 1.2-10. CT Scans of MF11 Dolomite Core
Conservative tracer tests. Core plugs were saturated with a synthetic reservoir
brine and baseline pressure drop and brine permeability were measured. A
conservative tracer, isopropyl alcohol (IPA), was added to the water and injected
into the core. Tracer data are a useful indication of heterogeneity such as
secondary porosity or dead end pores among other features. Figure 1.2-11
shows the tracer breakthrough data for one of the core plugs tested. The pore
volume was estimated based upon the first temporal moment of the tracer data.
1-39
1
Normalized Tracer Concentration
0.1
0.01
0.001 IPA Concentration History for MF4 (Dead volume ~ 7 mls)

First Moment = 62.5 mls
Injected tracer concentration ~ 2000 ppm (for 14.25 mls)
0.0001
0 25 50 75 100 125 150 175 200
Cumulative Volume Produced (mL)
Figure 1.2-11. IPA Tracer Breakthrough Curve
Polymer retention and degradation. An 18 md core plug from well MFU#623 was
prepared to study the polymer transport and specifically to verify that the polymer
did not plug the core. The core was flooded with more than 4 pore volumes of
1000 ppm HPAM (Alcomer 60RD) containing 1.0% Na+, 0.88% Cl-, and 0.56%
CO3-2. The effluent from the core was collected and its viscosity tested. Figure
1.2-12 presents viscosity data at varying shear rates and shows that there was
no measurable difference in viscosity between the injected polymer solution and
produced polymer effluent. The viscosity data indicate the polymer transport is
acceptable and that there were no problems with polymer degradation or
excessive retention in this low permeability core. The pressure drop data for the
polymer flood are presented in figure 1.2-13.
1-40
100
1000ppm Alcomer 60RD, 1% Na2CO3 Injectate

1000ppm Alcomer 60RD, 1% Na2CO3 Effluent from Core
10
Viscosity (cp)
0.1
0.1 1 10 100 1000
-1
Shear Rate (s )
Figure 1.2-12. Viscosity for HPAM polymer Injectate and Effluent
10
8
Pressure Drop, psi
Pressure Drop Across Core

6 Pressure Drop to Pressure Tap
Pressure Drop Across Core During Waterflood
Pressure Drop to Pressure Tap During Waterflood
0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
Pore Volume
Figure 1.2-13. Pressure Drop for HPAM Injection Test
1-41
Surfactant adsorption. The adsorption of surfactant in the dolomite rock was
expected to be high. Concurrent experiments by team members at Rice
University had shown that the addition of sodium carbonate could minimize
surfactant adsorption. An experiment was devised to quantify the level of
adsorption in a MF core. An 18 md MF core was prepared and injected with a
0.6 PV slug of surfactant-polymer solution consisting of 1.5% N67-(PO)7-SO4,
0.5% C20-24 AOS, 4% SBA and 2000 ppm Alcomer 60 RD HPAM polymer with
1% Na2CO3. The surfactant slug was followed by a polymer drive solution with a
lower electrolyte concentration to create a salinity gradient across the core. The
effluent samples produced during the core flood were collected and analyzed for
surfactant concentration using an anionic surfactant-specific electrode and
Hyamine™ titration assays. The surfactant breakthrough and recovery curves
are shown in figure 1.2-14.
100% 1.50
Cumulative Recovery
90% 1.35
Surfactant Concentration
80% 1.20
Surfactant Concentration (wt%)

% Surfactant Recovered
70% 1.05
60% 0.90
50% 0.75
40% 0.60
30% 0.45
20% 0.30
10% 0.15
0% 0.00
0.0 0.5 1.0 1.5 2.0 2.5 3.0
Pore Volumes
Figure 1.2-14. Surfactant Recovery and Concentration Curves
The surfactant recovery data show that 88 mg of the injected surfactant were
retained by the core during the flood. This correlates to a surfactant adsorption
value of 0.54 mg of surfactant per gram of rock.
1.2.4 Results of Core Flood Experiments to Evaluate Oil Recovery
Core flooding experiments in this project required an approach to minimize the

use of valuable and scarce field cores. Each experiment is considered a building
1-42
block for subsequent work. The results and lessons learned from each
experiment were incorporated into subsequent tests in a fine-tuning process.
Once suitable surfactant formulations had been identified and tested in the phase
behavior tests, a limited number of core floods were done. As described in the
previous section, the rock properties of sandstone and dolomite cores were
evaluated before conducting any core floods. The first two core flooding
experiments were conducted in Berea sandstone cores with octane as the
surrogate oil. These experiments were performed at 60oC as they were not
reservoir specific.
The first core flood was performed in a 448 md Berea core (#1) with a waterflood
residual oil saturation of 0.38. A 0.10 PV slug of 4% surfactant/8% SBA solution
was injected into the core followed by a 1.9 PV polymer drive. The oil recovery
from this core flood was about 68% with a final oil saturation of 11%. Due to
surfactant retention, no surfactant was recovered in the effluent. A surfactant
retention value of 0.39 mg/g of rock was measured. The data collected during
this experiment indicated the need for a larger surfactant slug and for the use of
sodium carbonate in both the slug and the polymer drive to help minimize
surfactant retention in the core. The results also showed the need for tighter
control on the salinity of the injected fluids.
A second core flood was performed in a 400 md Berea core (#2) with a
waterflood residual oil saturation of 0.31. A larger surfactant slug of 0.20 PV
containing 4% surfactant/8% SBA solution was injected into the core. The
salinity of the surfactant slug and the polymer drive solution was maintained
closer to optimum design parameters throughout the injection period. This core
flood resulted in a final oil saturation of 0.05 with an oil recovery of 83%.
Surfactant adsorption was decreased to 0.36 mg/g of rock due to the addition of
sodium carbonate. The lower surfactant retention and larger surfactant slug size
and the better salinity control helped increase oil recovery.
These two experiments were followed by a Berea sandstone core flood (#3)
containing MF2 crude oil with a waterflood residual oil saturation was 0.35. This
experiment introduced additional difficulties to the process by reducing the
temperature to 38oC (the actual MF reservoir temperature) and by using a
synthetic reservoir brine to saturate the core. A salinity gradient was used in this
core flood to improve its performance. The initial brine permeability (100% water
saturation) was 338 md and the brine permeability at residual oil saturation was
28 md. The initial brine salinity was 6.1 wt% total dissolved solids (TDS), which is
the same as the reservoir brine. Figure 1.2-15 shows the pressure gradient
across the Berea core approximately one foot in length.
A 0.20 PV surfactant slug with 3% N67-(PO)7-SO4, 1% C20-24 AOS, 8% SBA with

0.02% sodium carbonate and 1000 ppm xanthan gum polymer was injected into
the core. This was followed by a 2.7 PV polymer drive with sodium carbonate
1-43
added to the polymer solution. The surfactant retention was 0.41 mg of
surfactant per gram of rock. This core flood recovered 92% of the oil in the core,
reducing the oil to a final saturation of 0.027. The cumulative oil recovery and
the fractional flow of oil are shown on figure 1.2-16. The pressure gradient during
the surfactant/polymer slug and the polymer drive was about the same as the
water flood pressure gradient of 2.3 psi/ft (as shown in figure 1.2-15). The low
pressure gradient indicates there were no problems with plugging or
surfactant/polymer transport during this core flood.
4
Pressure Gradient (psi/ft)
3
Water flood pressure drop (psi)
0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
Pore Volumes
Figure 1.2-15. Pressure Drop for Berea Core #3 with MF2 Crude
1-44
100% 1.0
90% 0.9
80% 0.8
70% 0.7
Oil Recovered (%)
60% Cumulative Oil Recovery 0.6
Oil Cut
50% Fraction of Oil 0.5
40% 0.4
30% 0.3
20% 0.2
10% 0.1
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
Figure 1.2-16. Oil Recovery in Berea Core #3 with MF2 Crude
The results of this set of experiments indicated that the selected surfactant
formulation worked well in the Berea sandstone cores with both surrogate oil and
field crude oil. These results also confirm the value and validity of the phase
behavior screening approach to core flooding as the research team was able to
achieve greater than 90% oil recovery with just a few core flooding experiments.
Midland Farms Core Flood #1. After successfully demonstrating a sandstone

core flood using a field crude oil, a dolomite reservoir core from the MF reservoir
was selected for the next core flood. The dolomite core (#1) was saturated with
a synthetic formation brine and brine flooded to a residual oil saturation of 0.23.
The core had an initial brine permeability of 92 md and a brine permeability of 8
md at residual oil saturation.
Due to additional phase behavior observations, the surfactant formulation from

the last Berea sandstone core flood was modified. The surfactant formulation
substituted C15-18 IOS for C20-24 AOS to help reduce the need for SBA and
improve the behavior at low surfactant concentrations. The change in the co-
surfactant increased the optimum salinity of the slug making it closer to the
formation salinity of 6.1 wt%. The surfactant slug contained 2000 ppm of a
hydrolyzed polyacrylamide (HPAM) polymer for mobility control. A 0.40 PV slug
of 2% surfactant/4% SBA solution was used in this test. This was followed by a
2000 ppm Alcomer 60RD polymer drive solution. The surfactant slug and the
1-45
polymer drive both contained 1% sodium carbonate to reduce the amount of
surfactant adsorption.
This core flood yielded an oil recovery of approximately 70% leaving a chemical
residual oil saturation of 6.9%. Most of this oil was recovered as free oil, with
approximately 15% of it in an emulsion. The oil in the emulsion was separated
via centrifugation so it could be measured. Approximately 0.10 PV of emulsion
remained after centrifugation. This volume of oil was not included in the oil
recovery results, so the actual oil recovery is slightly higher than 70%. Surfactant
breakthrough occurred at 0.12 PV, which was earlier than expected and is likely
due mostly to inadequate mobility control in this core flood and the adverse
effects of channeling and fingering. 68% of the injected surfactant was
recovered, which gives a surfactant retention of 0.21 mg/g of rock. This is
significantly lower than retention values measured in previous experiments.
Lower than expected pressure gradients indicated that mobility control was
unfavorable. This may explain the early surfactant breakthrough, low surfactant
retention, and lower than expected oil recovery. As a result a higher polymer
resistance factor was targeted for the next flood by using a higher molecular
weight HPAM polymer.
Midland Farms Core Flood #2. A second carbonate core flood (#2) was
designed to address the problems identified in the previous one. Core material
was crushed and analyzed for anhydrite using X-ray Diffraction as shown figure
1.2-17. The magnitude of the peak corresponding to anhydrite indicated this
sample of rock had roughly 1.2% anhydrite in it. Sulfate was measured in the
produced brine from the core to test for dissolved anhydrite. Results reported in
Table 1.2-6 show that sulfate is present in quantities consistent with anhydrite
dissolution. These levels of sulfate correspond to high concentrations of
dissolved calcium and this was verified by some qualitative measurements of
Ca++ in the produced brine on the order of 1200 ppm. Therefore, sodium
carbonate was not used due to concerns about precipitation of calcium
carbonate. It is unlikely that sodium carbonate could be used in a rock with
anhydrite present, but some further investigation is needed to better understand
the aqueous chemistry and verify this conclusion.
1-46
Figure 1.2-17. XRD Data from Crushed MFU Well #629 Rock Sample
Courtesy of Dr. Steve Swinnea, University of Texas at Austin Materials Engineering Department.
Table 1.2-6. Sulfate Measured in Produced Formation Brine

Effluent Sample Sulfate Concentration [mg/L]
Tube #8 – analysis #1 1238
Tube #8 – analysis # 2 1686
The MF core used for core flood #2 was saturated with the synthetic formation
brine containing about 2400 ppm Ca++ and oil flooded with MF3 crude oil. The
initial brine permeability was 166 md and the brine permeability at a residual oil
saturation of 0.39 was 5 md. A surfactant slug of 0.8 PV consisting of 0.75%
N67-(PO)7-SO4, 0.25% C15-18 IOS, 2% SBA, and 1500 ppm Flopaam 3330S
polymer was injected into the core. This was followed by a polymer drive of 2.5
PV of 1200 ppm Flopaam 3330S solution. No surfactant breakthrough was
observed prior to the start of the polymer drive injection.
Surfactant breakthrough occurred at 1.3 PV and the peak concentration was

reached at 1.5 PV. The injected surfactant mass was 640 mg and 322 mg of
surfactant was retained by the core or 0.43 mg surfactant per gram of rock. The
cumulative oil recovery and the oil cut for the core is presented in figure 1.2-18.
1-47
The flood recovered 95% of the oil in the core. Approximately 90% of the oil was
recovered in a clean oil bank and 5% in the form of an emulsion.
100% 1.0
Cumulative Oil Recovery
90% 0.9
Fraction of Oil
80% 0.8
70% MF Core (166 md) 0.7

% Oil Recovered
MF3 Crude Oil at 38 °C

60% 0.75% N67 (PO)7SO4 0.6
Oil Cut
0.25% C15-17 IOS
50% 2% SBA, 0.02% Na2CO3 0.5
40% 0.4
30% 0.3
20% 0.2
10% 0.1
0% 0.0
0.0 0.4 0.8 1.2 1.6 2.0 2.4
Pore Volumes
Figure 1.2-18. Oil Recovery for MF Core #2 with MF3 Crude Oil
The pressure drop data are shown in figure 1.2-19. The pressure drop increased
during the flood but there were no indications of plugging.
1-48
30
Oil Breakthrough Polymer Drive Surfactant
Breakthrough
25
20
Pressure (psi)
15
10
0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
Pore Volume
Figure 1.2-19. Pressure Drop Data for MF Core #2
The mobility reduction caused by the polymer was good from the point of view of
mobility control and stability, but it was larger than expected and higher than
desirable for a field process. The viscosity of the polymer solution was
reasonable and known in advance, but the permeability reduction factor of about
7 was higher than expected for a core of this permeability. Subsequent core
floods will be done with a polymer of lower molecular weight to reduce the
pressure gradient to values closer to field values.
In summary:
• 95% of the waterflood residual oil saturation was recovered.

• Final chemical flood residual oil saturation was 0.01.
• Surfactant retention was 0.43 mg/g.
• No plugging occurred.
• Sulfate in produced brine before the surfactant was injected showed
the presence of anhydrite; therefore no sodium carbonate was used in
the slug or drive in this experiment.
• Mobility control with polymer was good, but the polymer resistance
factor was higher than needed.
1-49
The results of this core flood experiment indicate that the surfactant selection
procedure used in this research is based upon sound scientific principles and
leads to excellent performance as well as being very efficient.
Table 1.2-7. Summary of Core Floods
Rock Type Berea #1 Berea #2 Berea #3 Carbonate #1 Carbonate #2
o
Temperature, C 60 60 38 38 38
3
Pore Volume, cm 148 157 157.4 37.2 59.7
Initial Permeability,
448.7 404.4 338.4 91.9 166
md
Midland Farms Midland Farms Midland Farms 3
Oil Octane Octane
2 Crude 2 Crude Crude
Residual Oil
38.0% 31.5% 34.9% 23.0% 39.0%
Saturation, %
Permeability at
41.7 52.0 27.5 7.6 5
Sorw, md
Formation Brine, wt 1.6% (9:1 2.5% (9:1 6.1 % (8:1 6.1 % (8:1 6.1 % (8:1
% NaCl/CaCl2) NaCl/CaCl2) NaCl/CaCl2) NaCl/CaCl2) NaCl/CaCl2)
1.5% C16-17 0.75% C16-17
Surfactant/Solvent 3%C16-17 (PO)7
4%C16-17 (PO)7 4%C16-17 (PO)3 (PO)7 SO4, (PO)7 SO4,
concentrations, SO4, 1% C20-24
SO4, 8%SBA SO4, 8%SBA 0.5% C15-18 IOS, 0.25% C15-18 IOS,
wt% AOS, 8% SBA
4% SBA 2% SBA
1.2% (9:1 1.7% (9:1 1.7% (9:1

Slug Salinity, wt % 3.8% NaCl 4.45% NaCl
NaCl/CaCl2) NaCl/CaCl2) NaCl/CaCl2)
Polymer
1000 ppm 1000 ppm 1000 ppm 2000 ppm 1500 ppm
Concentration in
xanthan gum xanthan gum xanthan gum Alcomer 60RD Flopaam 3330S
slug
Na2CO3
concentration in none 0.02% 0.02% 1.00% 0.02%
slug, wt %
Surfactant Slug
0.107 0.2 0.2 0.4 0.8
size, PV:
Polymer
1000 ppm 1000 ppm 1000 ppm 2000 ppm 1200 ppm
concentration in
xanthan gum xanthan gum xanthan gum Alcomer 60RD Flopaam 3330S
polymer drive
Polymer drive 1.1% TDS (9:1
1.0% NaCl 1.0% NaCl 1.6% NaCl 1.93% NaCl
salinity, wt % NaCl/CaCl2)
Na2CO3
Concentration in none 0.02% 0.02% 1.00% 0.01%
polymer drive, wt %
Polymer Drive, PV 1.9 2.0 2.7 1.0 2.5
Oil Recovery 67.7% 83.2% 92.4% 70.0% 95.0%

Final Oil Saturation 10.9% 5.3% 2.7% 6.9% 1.0%
Oil Breakthrough,
0.2 0.2 0.3 0.1 0.3
PV
Surfactant
NA 0.7 0.8 0.12 1.3
Breakthrough, PV
Oil Production
1.1 1.1 1.4 1.4 1.7
ceases, PV
Surfactant
0.39 0.36 0.41 0.21 0.43
Retention, mg/g
1-50
Task 2 Phase behavior, adsorption, and composition changes during
displacement
Subtask 2.1 Surfactant Adsorption
The surfactant adsorption and alkali consumption are crucial for the
alkaline-surfactant process. By using the solubility calculation, alkali
consumptions were calculated for carbonate formations when CaSO4 exists.
Substantial retardation of Na2CO3 was found due to precipitation of CaCO3.
Two other potential determining ions, hydroxyl ion and sulfate ion were tested
to see whether they could reduce the surfactant adsorption as found previously
for carbonate ion. Light scattering could easily determine the sizes of
surfactant aggregates that can also significantly influence the propagation
velocity of the injected surfactant solution in a porous medium. By using
porous media with different surface area, we found it was the surface area, not
the weight, of the porous media that determine the adsorption amount. The
threshold concentration of potential determining ion (CO32-) required to
achieve significant reductions in surfactant adsorption was investigated.
Surfactant propagation velocity with adsorption

The adsorption of surfactant can alter the propagation velocity. For
one-phase flow, we have
∂ (c + c s ) ∂c
+v =0
∂t ∂x (1)
where cs is the adsorbed concentration of surfactant. And cs is a function

of local concentration c, i.e., cs=f(c). v is the interstitial velocity.
We assume Langmuir-type isotherm, where cmax is the maximum
adsorption value
c max c
cs =
k +c (2)
dc s c k c /k
= max 2 = max
dc ( k + c ) (1 + c / k ) 2 (3)
After some manipulations, we can obtain that the surfactant velocity vs is
1
vs = v
dc
1+ s
dc (4)
Since the isotherms of surfactant are Langmuir-type, dcs/dc will decrease

as c increases from (3) and hence vs will increase from (4). From the rule of
2-1
wave theory, waves originating from the same point must have non-decreasing
velocities in the direction of flow. If slower waves from compositions close to
the initial conditions originate ahead of faster waves as is the case here, a
shock will form as the faster waves overtake the slower waves. To determine
the shock velocity, we must apply a mass balance across the shock and gain
the result given by equation (5)
⎛ dx ⎞ c 2 − c1 Δc 1
⎜ ⎟ =v =v =v
⎝ dt ⎠ Δc c 2 − c1 + c s 2 − c s1 Δc + Δc s Δc
1+ s
Δc (5)
where Δcs/Δc is the chord slope of the isotherm between the

concentrations at each side of the shock.
2
c max /k =2, c max =2
1.8
1.6 c max /k= 1, c max =1
Retardation Time (PV)
1.4 c max /k =1, c max =0.5

1.2
1
0.8
0.6
0.4
0.2
0
0 0.5 1 1.5 2
Injection Concentration
Figure 2.1-1 Relationships between effluent retardation and Langmuir parameters
Figure 2.1-1 shows the relationship between break-through retardation

time and Langmuir parameters. The break-through retardation time is defined
as the difference between the required injection pore volume (PV) to see the
surfactant break-through and unit pore volume. Zero retardation time implies
unit propagation velocity and no adsorption. Larger retardation means slower
propagation velocity. From this plot, we can generalize that increasing the
injection concentration reduces the retardation, i.e., increases the propagation
velocity. Furthermore, the Langmuir parameters can significantly change the
retardation degree, i.e. the propagation velocity. At very low concentration, the
initial slope of Langmuir isotherm (cmax/k) determines the velocity. And lower
propagation velocity or larger surfactant retardation occurs at lower injection
concentration. At higher concentration, the surfactant propagation velocities
depend on the maximum adsorption amount cmax only. In our process, the
2-2
surfactant concentrations we apply are above the CMC so that the adsorption
isotherms already reach the plateau. Thus, the maximum adsorption amount is
the determining factor for the surfactant propagation.
Alkali consumption on carbonate formation

Pure calcite does not consume much alkali. However, the consumption of
alkali in carbonate reservoir may be a crucial problem because of the
precipitation reaction of alkali with gypsum and anhydrite impurities. Because
the solubility products of CaCO3 and CaSO4 are 4.96*10-9 and 7.10*10–5
respectively (CRC Handbook 68th Edition), it is a serious problem to apply
Na2CO3 as our alkali candidate because of the precipitation reaction shown as
equation (6).
2− 2−
CO3 + CaSO4 → CaCO3 ↓ + SO4 (6)
Figure 2.1-2 shows the retardation for a porous medium with porosity 0.3.
It illustrates that the retardation is significant. For a 0.1M (~1%) Na2CO3, the
concentration usually being considered for oil recovery processes, the
retardation is around 0.7 PV for the condition that 0.1% of the porous medium
is CaSO4. Although we still can reduce the retardation or enhance the
propagation velocity by increasing the injection alkali concentration, the total
amount of alkali consumption will not change. It is impractical to solve this
problem by increasing the sulfate ion concentration through adding Na2SO4
because of the tremendous difference between the two solubility products.
All the medium is CaSO4
1.E+04
10% of the medium is CaSO4
1% of the medium is CaSO4
1.E+03
0.1% of the medium is CaSO4
Retardation Time (PV)
0.01% of the medium is CaSO4

1.E+02
1.E+01
1.E+00
1.E-01
1.E-02
1.E-03
0 0.2 0.4 0.6 0.8 1
Na2CO3 injection concentration (mol/L)
Figure 2.1-2 Relationships between retardation and CaSO4 fraction in porous medium
(porosity=0.3)
NaHCO3 with Na2SO4 may be a potential choice for the situation with
CaSO4. NaHCO3 has much lower carbonate ion concentration and additional
sulfate ions can decrease calcium ion concentration in the solution. However,
2-3
this method is not applicable again. The concentration of CO32- is around the
one hundredth the concentration of NaHCO3 for a NaHCO3 solution so that we
still need large amount Na2SO4 to avoid precipitation of CaCO3. For a 0.1M
NaHCO3 solution, the carbonate ion concentration is around 1.0*10-3 M, and
we need 14 M Na2SO4 to restrain the precipitation of CaCO3.
The other alkali candidate is NaOH with Na2SO4 since the solubility
product of Ca(OH)2 is 4.68*10-6. The reaction between NaOH with Na2SO4 is
shown as equation (7). The minimum Na2SO4 concentration that restrains the
Ca(OH)2 precipitation can be calculated by equation (8). For a 0.1 M NaOH
solution, 0.15 M Na2SO4 is needed to suppress the calcium ion concentration
so that no Ca(OH)2 will precipitate. And higher Na2SO4 is necessary with
higher NaOH concentration. Also, the surfactant adsorption isotherm will
change at this condition and we need to measure it.
2 NaOH + CaSO4 ⇔ Ca(OH ) 2 + Na 2 SO4 (7)
2− K spCaSO4
[ SO4 ] = [OH − ] 2 = 15 * [OH − ] 2 (8)
K spCa (OH )2
Therefore, Na2CO3 is our first choice for the carbonate formation reservoir
if minimal CaSO4 exists. Otherwise, NaOH with Na2SO4 will be our next
candidate.
Potential determining ions for surfactant adsorption

1.2
Without alkali
1.0
Adsorption Density (mg/m )
2
0.8
With 0.1 M NaOH

0.6
With 0.1M NaOH
+0.15M Na2SO4
0.4
With Na2 CO3 (0.2M,0.3M,0.4M)

0.2
0.0
0.00 0.02 0.04 0.06 0.08 0.10 0.12
Residual Surfactant Concentration (Wt%)
Figure 2.1-3 Adsorption on powdered dolomite of TDA-4PO:CS330(1:1) with different

potential determining ion
As discussed in previous part, we can suppress the alkali consumption by
using NaOH with Na2SO4. However, we do not know whether this alkali
2-4
formulation could restrain the surfactant adsorption as well as Na2CO3 does.
Several experiments were performed and we found that this approach could
not decrease the surfactant adsorption. As figure 2.1-3 shows, the adsorption
amount on dolomite surface with these ions is the same as that without any
potential determining ion. Thus, we need to use other surfactant and alkali
when there is a large amount of CaSO4 in the reservoir.
Surfactant aggregates’ size

The sizes of surfactant aggregates can also significantly influence the
propagation velocity of the injected surfactant solution in a porous medium.
Small micelles are equilibrium aggregates and cause no problems. However,
phase separation occurs if aggregates’ sizes become large enough or the
shape and interaction between aggregates change. In this situation, the
surfactant may be trapped in the pores and not flow with the aqueous phase.
We found that the surfactant type and the alkali concentration determined the
aggregates’ sizes by using light scattering experiments for different surfactants
with same concentration (0.05%Wt). For the samples where bulk phase
separation occurred yielding a thin layer of surfactant-rich liquid at the air/water
interface (phase separation in figure 2.1-5), we measured the clear bulk phase
in the light scattering experiments.
10000 Very Faint
Clear Peak1
faint beam
Beam
beam Peak2
1000
particle size (nm)
100
10
1
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Na2CO3(Wt% )
Figure 2.1-4 Particle size distribution of Stepan TDA-4PO
Figure 2.1-4 is an example of particle size distribution for different

surfactants. In this plot, the area of each point indicates the intensity of that
particle size. And the total area of peaks for each sample is fixed. Thus, we
can know how many peaks the sample has and which peak of the sample is
the dominant one. It seems that the drop size distributions are different with
different surfactant. However, the size of the dominant peak increases with the
sodium carbonate concentration for all the surfactant solutions. Also, we found
an easy method to estimate the dominant particle size instead of the complex
2-5
light scatter experiments. Before the light scattering experiments, we record
the appearance of scattered light. Light from optical fiber and laser pointer light
will give us the same appearance. Figure 2.1-5 indicates this relationship. For
a very faint beam the dominant particle size is in 11±5 nm. Faint beam implies
the dominant particle size is in 30±17nm. When we see strong beam, the drop
size of the primary peak should be 69±37 nm. The beam strength is increased
with the dominant particle size. When we see phase separation, we may
conclude that the dominant particle size must be larger than 180 nm. We can
use this feature to estimate the dominant particle size in our future samples.
1000
dominant particle size (nm)
100
10
1
very faint faint beam phase
separation
Figure 2.1-5 Relationship between appearance of scatter beam and dominant peak
particle size
In order to propagate the surfactant, we need a suitable surfactant with

appropriate alkali concentration to avoid the phase separation.
Surfactant adsorption on porous media with different surface area

Surface area of the porous media has remarkable effect on the surfactant
adsorption. We tested two dolomite samples with different surface area. One is
powdered dolomite, whose surface area is 1.7m2/gram; the other is dolomite
sand, whose surface area is 0.3m2/gram. Figure 2.1-6 shows that the
adsorptions between the two samples are comparable if we calculate the
adsorption by using mg per surface area. If we calculate the adsorption by
using mg per weight as in figure 2.1-7, we can find the adsorption on sand
without alkali is about one sixth of the adsorption on powder. These results
imply that we should use the surface area, not the weight, of the porous media
to compare the adsorption.
2-6
Adsorption of CS330 &TDA-4PO (1:1) blend on dolomite
Adsorption Density (mg/m )

1.2
2 1.0
0.8
0.6
0.4
0.2
0.0
0.00 0.02 0.04 0.06 0.08 0.10 0.12

2
Adsorption of dolomite powder (1.7m /g) without alkali
2
Adsorption of dolomite sand (0.3m /g) without alkali
Figure 2.1-6 Adsorption of CS330: TDA-4PO (1:1) on dolomite with different surface area
(I)
Adsorption of CS330 &TDA-4PO (1:1) blend on dolomite

2.5
Adsorption density (mg/g)
2.0
1.5
1.0
0.5
0.0
0.00 0.02 0.04 0.06 0.08 0.10 0.12

2
Adsorption of dolomite powder (1.7m /g) without alkali
2
Adsorption of dolomite sand (0.3m /g) without alkali
Figure 2.1-7 Adsorption of CS330: TDA-4PO (1:1) on dolomite with different surface area
(II)
Adsorption isotherms for NEODOL-67-7PO:IOS (4:1) and threshold of

Na2CO3
We found the previous surfactant formulation (CS330&TDA-4PO) could
generate viscous phases and its optimal salinity is very high (12~13%) so that
it is not our choice anymore. We tested our new surfactant formulation
(NEODOL-67-7PO: IOS 4:1) adsorption on the calcite surface as figure 2.1-8
shows. This new formulation is selected because both surfactants in the new
2-7
mixture have branched hydrophobic chain while one surfactant in earlier
mixture is straight chain. We find that adsorption isotherms of this formulation
are similar in shape to the previous blend. These results at 0 and 1% Na2CO3
are consistent with Fig. 2.3-12 of this report. From figure 2.1-8, it is seen that
adsorption increases only slightly when Na2CO3 concentration is reduced to
0.08%.
We also tested the threshold of Na2CO3 concentration at which the
adsorption reduction effect occurs as shown in figure 2.1-9. We used same
amount of surfactant solution mixed with same amount calcite powder. We
fixed the initial surfactant concentration but changed Na2CO3 concentration.
And we found that the adsorption reduction effect seemed the same when the
Na2CO3 concentration was higher than 0.1 %, i.e., enhancing the Na2CO3
concentration would not decrease the surfactant adsorption when Na2CO3
concentration was higher than 0.1 %. However, the adsorption will increase as
the Na2CO3 concentration deceases below 0.1%. From this result, the lowest
Na2CO3 concentration for ASP process should be higher than 0.1%.
Adsorption Isotherm of 4:1(N67:IOS) on calcite powder

2.5
Adsorption Density(mg/m )
2
2.0
1.5
1.0
0.5
0.0
0.00 0.02 0.04 0.06 0.08 0.10 0.12
Residual Surfactant Concentration (Wt %)
Adsorption on calcite without alkali
Adsorption on calcite with 0.08%Na2CO3
Adsorption on calcite with 1%Na2CO3
Figure 2.1-8 Adsorption on powdered calcite of N67:IOS (1:1) with different Na2CO3
concentration
2-8
Adsorption threshold of Na2CO3 for N67:IOS(4:1) on calcite powder
0.30
0.25
Adsorption Density(mg/m )
2
0.20
0.15
0.10
0.05
0.00
0.0 0.2 0.4 0.6 0.8 1.0 1.2
Na2CO3 (%)
Figure 2.1-9 Test of threshold concentration of Na2CO3 for the adsorption
Conclusions
1. For alkaline surfactant process, the maximum adsorption amount is the
determining factor for the surfactant propagation.
2. NaOH with Na2SO4 could be our next candidate for the reservoir containing
CaSO4 if excessive precipitation occurs when Na2CO3 is injected.
3. Only the carbonate ion can reduce the surfactant adsorption on carbonate
formation. Other potential determining ions (sulfate ion and hydroxyl ion)
could not decrease the surfactant adsorption.
4. We can use lamp light scatter to estimate the surfactant aggregates’ size,
and we need to avoid large surfactant aggregates when we make the
surfactant solution.
5. It is the surface area, not the weight, of the porous media that determine
the surfactant adsorption.
6. The threshold Na2CO3 concentration for adsorption reduction is around
0.1%Wt.
Task 2 Phase behavior, adsorption, and composition changes during

displacement
Subtask 2.2 Composition route for alkali-surfactant flooding
In order to understand and describe the ASP process, a one-dimensional,
two phase, multi-componential simulator was developed to calculate the
profiles and oil recovery as a function of process variables. In this simulator,
2-9
the main relationships are: phase behavior and interfacial tension are the
functions of electrolyte (NaCl) concentration, surfactant and natural soap
concentration; fractional flow is a function of interfacial tension, aqueous phase
saturation and viscosity; surfactant adsorption is a function of surfactant
concentration in aqueous phase and alkali concentration; natural soap is
generated by the naphthenic acid contacted by the alkali; physical dispersion
is described by convective dispersion; aqueous phase viscosity is a function of
electrolyte (NaCl) and polymer concentration; alkali consumption is due to the
precipitation reactions between Na2CO3 that we inject and the CaSO4 that
resides in the reservoir and the calcium ions in the formation brine, and to the
reaction between alkali and clays.
Assumptions and Models

The basic assumptions of the model are as follow.
1. The system is one-dimensional and homogeneous in permeability and
porosity.
2. Local equilibrium exists everywhere.
3. Capillary pressure is negligible.
4. The system is one-dimensional and horizontal. Thus there is no gravity
effect.
5. The system has eight components. They are water(1), oil(2), synthetic
surfactant or surfactant(3), natural soap(4), electrolyte(NaCl)(5), alkali(6),
polymer(7), naphthenic Acid(8). The numbers behind the chemicals are the
index we will use for our future discussion.
6. Two mobile phases are: Aqueous(1), Oleic(2).
7. All the chemicals except water and oil are assumed to occupy negligible
volume and are treated as tracers.
Additional assumptions also required for our model are discussed below.
1 Partition
Since our model is a two-phase model, the partitioning of the chemicals is
very important. The partition coefficient of the component i is defined as
equation (1);
ci 2
K Ci = (1)
ci1
where ci1 is the concentration in aqueous phase for component i
ci2 is the concentration in oleic phase for component i
We assume that electrolyte (NaCl), alkali and polymer are totally in the
aqueous phase, i.e. KC5=0, KC6=0 and KC7=0; while the naphthenic acid is
entirely in the oleic phase, i.e. KC8→∞.
However, the partitioning for the surfactant and soap is not that simple and
depends on the salinity (concentration of electrolyte) and the mole ratio
2-10
between the soap and surfactant. The basis of this partition assumption comes
from the phase behavior experiments of Leslie Zhang as in discussed in
section 2.3. According to her discovery, the optimal salinity of the brine/crude
oil system with soap and surfactant is as in figure 2.2-1. From this curve, we
know that the optimal salinity is equal to that of soap when soap is dominant,
and vice versa. And we assume that at the optimal condition, the average
partition of soap and surfactant are unity. The average partition of soap and
surfactant Kaver is defined as equation (2). Figure 2.2-2 shows the Kaver contour.
For a specific salinity, surfactant concentration and soap concentration, we can
obtain the average partition.
⎛ C3 ⎞ ⎛ C4 ⎞
K aver = K C 3 ⎜⎜ ⎟⎟ + K C 4 ⎜⎜ ⎟⎟ (2)
⎝ C3 + C 4 ⎠ ⎝ C3 + C 4 ⎠
where C3 is the overall concentration of surfactant,
C4 is the overall concentration of soap
Figure 2.2-1 Optimal Salinity vs. Soap-Synthetic Surfactant Ratio Curve for N67:IOS
(4:1)
2-11
Figure 2.2-2 Contour of Average Partition Coefficient (log10(Kaver))
From figure 2.2-2, we can only find Kaver, not KC3 and KC4. To find KC3 and
KC4, we introduce another parameter k4k3 which is defined as:
KC4
k 4k 3 = (3)
KC3
In most cases, we assume that k4k3=1, i.e., the surfactant and the soap
have same partition coefficients. Such joint partitioning is not unreasonable for
ionic surfactants, which have low solubility as monomers in the oil phase. If
necessary, we can change the value of k4k3 to represent the difference
between surfactant and soap. By following these steps, we can calculate
partitioning of all the components.
2 Interfacial tensions
The interfacial tension is a crucial factor for ASP process. Without low
tension, the ASP process could not be a practical process. The basis of the
interfacial tension also comes from experiments of Leslie Zhang’s as figure
2.2-1. We assume that the interfacial tension could be as low as 10-3 dyne/cm
at optimal condition. And the IFT contour is generated as figure 2.2-3. The
difference between the optimal salinity and local salinity determine the IFT.
When the salinity is far away from the optimal salinity, the IFT will be higher.
Figure 2.2-4 displays how the IFT changes with salinity for soap surfactant
ratio equal to unity. These calculations are for the condition that the soap and
surfactant total concentration is greater than the CMC (Critical Micelle
2-12
Concentration). And when the soap and surfactant total concentration is less
than CMC, we calculate IFT by interpolation between the IFT calculated by
figure 2.2-3 and the IFT of crude oil and brine without any surface active
materials.
Figure 2.2-3 Contour of interfacial tension (log10(IFT)) (IFT: dyne/cm)
Figure 2.2-4 IFT (dyne/cm) vs Salinity for Soap Surfactant Ratio = 1

(According to Leslie Zhang’s experiment, the optimal IFT can be as low as 10-3 dyne/cm.
The width of low IFT region is base on reference [1])
2-13
3 Aqueous Phase Viscosity
Figure 2.2-5 Contour of Viscosity (cp)

The viscosity of aqueous phase is very important for ASP process and is a
function of polymer concentration and salinity. According to experimental data
for Flopaam, the aqueous phase viscosity is shown as figure 2.2-5. In this
figure, c71 represent the concentration of polymer. We find that the viscosity
increases as polymer concentration increases. When the salinity is less than
1%, lower salinity would lead to high viscosity; but when the salinity is larger
than 1 %, there is not much change on the viscosity by changing salinity. The
lines that salinity equals 4% and polymer concentration equals 1500ppm in
figure 2.2-5 are consistent with experimental data of figure 1.2-6 and 1.2-7.
Currently, we do not count the permeability reduction by adding the polymer.
But the permeability reduction due to the polymer is very important and we will
calculate it in our future work.
4 Fractional flow
The fractional flow curve is a function of residual saturations and relative
permeabilities, which are determined by IFT. For residual saturations, we
calculate them as equation (4).
S1r =0, S2r =0 when IFT<0.005dyne/cm
S1r = 0.3 ∗ [1 + (log10 ( IFT ) / 2.3)]
when 0.005 dyne/cm<IFT<1dyne/cm (4)
S 2 r = 0.3 ∗ [1 + (log10 ( IFT ) / 2.3)]
S1r =0.3, S2r =0.3 when IFT>1dyne/cm
where S1r is the residual saturation of aqueous phase.
S2r is the residual saturation of oleic phase.
2-14
For the relative permeability, we calculate as follow:
(S − S jr )
S jR =
j
j=1,2 (5)
(1 − S1r − S 2 r )
k rj = k rjo S Ej
jR j=1,2 (6)
k ro1 = k rw
o
+ (1 − k rw
o
)(0.3 − S 2 r ) / 0.3 o
k rw is end point permeability of water (7)
k ro2 = k roo + (1 − k roo )(0.3 − S1r ) / 0.3 k roo is end point permeability of oil (8)
Ej = 1.0 when IFT<0.005dyne/cm
Ej = 1.5 + 1 / 6 * log10 ( IFT ) when 0.005 dyne/cm<IFT<1dyne/cm (9)
Ej = 1.5 when IFT>1dyne/cm
At high IFT, the data we used come from the water-wet system.
Figure 2.2-6 shows the fractional flow changes with saturation at different IFT
by assuming aqueous phase and oleic phase have same viscosity.
Figure 2.2-6 fractional flow changes with saturation at different IFT

(Aqueous phase viscosity = Oleic phase viscosity)
5 Surfactant adsorption
The assumption for surfactant adsorption is Langmuir-type adsorption with two
2-15
parameters (cmax, K) as equation (10).
c max c31
C3ads = (10)
c max
+ c31
K
where cmax is the maximal adsorption amount, K is the initial slope of the
isotherm.
Since the adsorption amount can be significant reduced with the present of
Na2CO3, we let cmax be a function of Na2CO3. When the Na2CO3 concentration
is larger than 0.1%, the value of cmax is one tenth of cmax in the absence of
Na2CO3.
6 Equations and calculation procedure

Given the above assumptions, the continuity equations for each component i
are
∂ (Ci + Ciads ) q ∂Fi ∂ 2 Ci
+ = Kl (11)
∂t Aφ ∂x ∂x 2
By using the following dimensionless variables, we can make the equation
dimensionless.
x
xD = (12)
L
q
tD = t (13)
AφL
Lq
Pe = Peclet Number (14)
K l Aφ
Equation (11) becomes

∂ (C i + C iads ) ∂Fi 1 ∂ 2 Ci
+ = (15)
∂t D ∂x D Pe ∂x D 2
where Ci is the overall concentration of component i. C i = ci1 * S1 + ci 2 * S 2
Ciads is the adsorbed concentration of component i

Fi is the overall fractional flow of component i. Fi = ci1 * f1 + ci 2 * f 2
These equations are solved numerically by explicit finite difference in both tD

and xD. The computational procedure is as follows:
1. For a given Ci, we can calculate the cij in each grid block by partition
calculation if the component i can not be adsorbed. For adsorbing
component, it should be solved by Newton-Raphson iteration with the
relationships of partition and adsorption. For the first time step, we start
with initial conditions.
2-16
2. With cij calculated in step 1, we can calculate IFT of each grid block.
3. With IFT, the residual saturation and relative permeability are solved. Thus
the overall fractional flow can be calculated.
4. With Fi, we can use equation (15) to calculate the concentration in the next
time step.
5. Steps 1 to 5 are repeated.
2 Calculation example
We will present one example of ASP process by using this simulator to
show what an ASP process will be and what operating parameters we could
use for an optimal strategy. The example shown here is an illustration of oil
recovery by injecting chemical (alkali, surfactant and polymer) slugs. The initial
condition of the example is that the reservoir has already been water flooded,
i.e., the oil saturation is the residual oil saturation after water flooding.
Table 2.2-1 shows the operating parameters. A surfactant slug of 0.4 PV
consisting of 0.1% (4:1NEODOL67:IOS),0.5%Na2CO3,1800ppm Flopaam, 5%
NaCl is injected. This is followed by 0.2 PV polymer drive consisting of
1800ppm Flopaam and 5%NaCl. After that, a 0.4 PV brine drive (5% NaCl)
continues to be injected. Finally, we inject the formation brine (1%NaCl).
The other major parameters for the calculation are listed in table 2.2-2. NX
presents the number of grid block in the simulation; dt/dx presents ration of the
time step and grid block size. For example, if NX equals 100, and dt/dx equals
0.05, the time step dt will be 0.0005 (=1/NX*dt/dx=1/100*0.05=0.0005). And
the Peclet number presents the longitudinal dispersion. Equation (14) indicates
small Peclet number implies large longitudinal dispersion.
Chemicals Concentration Total Slug Size (PV)

Alkali (Na2CO3) 0.5% 0.4
Surfactant (4:1 NEODOL 67:IOS ) 0.1% 0.4
Polymer (Flopaam 3330S) 1800ppm 0.6
NaCl 5% 1.0*
*After 1.0 PV, NaCl injecting concentration is 1.0%
Table 2.2-1 Operating Parameters for the example
Initial Oil Formation Acid No. of Crude oil NX (Grid block No.) dt/dx
Saturation Brine
0.3 1.0% 0.2g KOH/g 100 0.05
Peclet No Polymer Surfactant Alkali consumption Oil viscosity
adsorption Adsorption
500 20 μg/g 0.2mg/g 1.4mg/g 8 cp
Table 2.2-2 Other major parameters for the example
Figure 2.2-7 shows the profiles of surfactant and soap at 0.5 PV. Figure 2.2-8
displays the profiles of IFT and soap-synthetic surfactant ratio at 0.5 PV.
2-17
Figure 2.2-9 presents the profile of oil at 0.5 PV.
Figure 2.2-7 Profiles of surfactant and soap at 0.5PV
Figure 2.2-8 Profiles of IFT and soap-synthetic surfactant ratio at 0.5PV

From figure 2.2-7, we can find that alkali front is ahead of the surfactant
front so that the soap front is ahead of the surfactant front. Thus, there is the
type II (over-optimal region) ahead of the surfactant front because soap is
dominant here. There is the type I (under-optimal region) far behind the
surfactant front because surfactant is dominant and the drive salinity is only
1.0%. And an optimal region is between type II and type I. This ASP process
will have following merits: (1) it must pass through the optimal region; (2) it can
retard surfactant that moves fast into the soap-dominant region ahead of the
major surfactant bank, (3) it does not trap surfactant behind the surfactant
2-18
bank since the surfactant is in the aqueous phase. This process design will
make surfactant accumulate and oppose the dispersion effect. We can even
see the surfactant concentration is a little bit larger than what we inject due to
the accumulation of surfactants in figure 2.2-7. Figure 2.2-8 shows the
ultra-low tension region locates at the place where soap-synthetic ratio less
than 0.01. This is because we inject at the optimal salinity of the surfactant.
And figure 2.2-9 tells us that nearly all the oil has been recovered after the
low-tension region.
Figure2.2-9 Profile of oil saturation at 0.5PV
Figure 2.2-10 Effluent History of Synthetic surfactant and Soap
2-19
Figure 2.2-11 Effluent History of Oil
Figure 2.2-10 and 2.2-11 show the effluent history of surfactant, soap and
oil. We can find most oil comes out before the surfactant breaks through. And
the recovery efficiency of oil is 99.99%, i.e., nearly all the oil has been
recovered at time 2 PV. Since the soap and surfactant break through at 1.0
PV, the oil will be in the emulsion after 1.0 PV. And the clean oil recovery
efficiency is 74.6%.
This base case seems similar to the surfactant flooding. What are the
advantages that introducing the soap in the ASP process? Figure 2.2-12,
2.2-13 show one benefit that soap brings. Figure 2.2-12 is the surfactant
flooding process without soap. We find that the oil recovery is sensitive to the
injected brine salinity. At optimal condition (5% salinity), the total recovery
efficiency can be as high as 100%. However, the total recovery drops to 70%
when the salinity goes to 4.5%. At 1% salinity, no oil is recovered. For the ASP
process, we can recovery 35% total oil and 32% clean oil even at 1% salinity.
This result implies the ASP process is more flexible and robust than the
surfactant flooding.
If we increase the polymer concentration, i.e. increase the injecting
aqueous phase viscosity, this effect will be more significant. The conditions in
figure 2.2-14 are almost the same as the figure 2.2-12. The only difference of
the two plots is the polymer concentration. The results show that, in the
absence of soap, increasing polymer concentration without soap will not
increase the total oil recovery but will increase the clean oil recovery a little bit.
At 1% salinity, the total oil recovery will still be zero. But for ASP process,
increasing polymer concentration will significantly enhance the total oil
recovery and clean oil recovery. At 1% salinity, the total recovery and clean oil
recovery will be 51% and 49.5% respectively. At 5% salinity, the total
2-20
recoveries are 100% for both 1800ppm and 2500ppm. However, the clean oil
recovery increases from 70% to 85%.
Recovery vs Injecting Brine Salinity

(no soap, 1800ppm polymer)
100%
80%
Recovery
60%
40%
20% Total oil recovery

Clean oil recovery
0%
1 2 3 4 5 6
Injecting Brine Salinity(%)
Figure 2.2-12 Oil recoveries vs. injecting brine salinities (no soap, 1800ppm polymer)

(with soap, 1800ppm polymer)
100%
80%
Recovery
60%
40%
Total oil recovery
20%
clean oil recovery
0%
1 2 3 4 5 6
Figure 2.2-13 Oil recoveries vs. injecting brine salinities (with soap, 1800ppm polymer)
2-21
(no soap, 2500ppm polymer)
100%
80%
Recovery
60%
40%
Total oil recovery
20%
Clean oil recovery
0%
1 2 3 4 5 6
Figure 2.2-14 Oil recoveries vs. injecting brine salinities (no soap, 2500ppm polymer)

(With soap, 2500ppm polymer)
100%
80%
Recovery
60%
40%
Total oil recovery
20%
Clean oil recovery
0%
1 2 3 4 5 6
Figure 2.2-15 Oil recoveries vs. injecting brine salinities (with soap, 2500ppm polymer)
The other advantage of soap is the possibility that the presence of soap
may decrease the surfactant adsorption. Because soap is also negatively
charged and the carboxylate group desires to attach with alkali earth porous
media, the soap may occupy the surface that can be used for adsorbing the
surfactant. If so, the soap could be a sacrificial agent to save surfactant usage
as the soap will travel ahead of the surfactant. Current simulator just assumes
that surfactant adsorption is related only with the surfactant concentration. If
the soap has effect on the surfactant adsorption, we need to change this
assumption. This will be clarified in our future work.
2-22
Conclusions
1. A one-dimensional, two phase, multi-componential simulator was
developed. It can provide the profiles of all the components at any time and
predict the oil recovery.
2. All of the oil can be recovered with salinity near optimum and adequate
aqueous phase viscosity.
3. Soap generated by the alkali results in the ASP process being more robust
to salinity changes than a conventional surfactant process. In particular oil
recovery is substantially greater at salinities slightly below optimal for the
injected surfactant in alkaline processes where soap forms as a second
surfactant
4. The polymer concentration will significantly increase the total and clean oil
recovery in ASP process
Reference:
[1] C. A. Miller and P. Neogi, in Interfacial Phenomena, Marcel Dekker, New
York,1985
2-23
Subtask 2.5 Surfactant-enhanced spontaneous imbibition experiments
While the TC blend consisting of equal weight amount of TDA-4PO and CS-
330 showed promising results, as discussed in the first annual report, it also has
some drawbacks, such as optimal salinity increasing significantly with soap /
synthetic surfactant ratio, and the formation of viscous phases, possibly due to
liquid crystals. Therefore, this year’s work has been dedicated to finding a much
less hydrophilic surfactant or surfactant blend without viscous emulsions. N67-
3PO sulfate and N67-7PO sulfate were tested alone or with co-surfactants/co-
solvents.
2.5.1. Surfactant Identification

The surfactants and co-surfactants evaluated are identified in Table 2.5-1.
Table 2.5-1 Surfactant and co-surfactant identification
Surfactant
Trade Name Structure Name
Type
CS-330 Sodium dodecyl 3EO sulfate
TDA-4PO S Ammonium iso-tridecyl 4PO sulfate
TC Blend Blend of equal weight amount of TDA-4PO S and CS-330
N67-3PO S Ammonium NEODOL 67-3PO sulfate
N67-3EO S Sodium NEODOL 67-3EO sulfate

Anionic
N67-7 PO S Ammonium NEODOL 67-7PO sulfate
Surfactant
IOS 1518 Sodium C15-18 internal olefin sulfonate
NI Blend Blend of 4:1 weight ratio of N67-7PO sulfate and IOS 1518
AOS C1618 Sodium C16-18 alpha olefin sulfonate
AOS C2024 Sodium C20-24 alpha olefin sulfonate
Oil-Chem 4-22 C14-16 alkylsulfonate of phenol with 7EO
Triton X-100 Octyl phenol ethoxylate

Non-ionic
Tergitol 15-S-7 C11-15-7EO secondary alcohol
2-24
2.5.2. Problems with the TC Blend
14
TC Blend: TDA-4PO S/CS-330=1:1 WOR=1
Optimal NaCl Conc., %
12
WOR=3
10
WOR=10
8
2 Formation brine salinity

0
1E-02 1E-01 1E+00 1E+01
Soap/Synthetic Surfactant Mole Ratio
Fig 2.5-1 Dependence of optimal salinity on soap/synthetic surfactant ratio
Fig. 2.5-1 shows the dependence of the optimal salinity of the TC Blend on
the soap / synthetic surfactant ratio. The optimal salinity at high surfactant
concentration is about 12% NaCl (with 1% Na2CO3), while the reservoir salinity is
only 1% NaCl. The drawback of optimal salinity increasing significantly with soap
/ synthetic surfactant ratio is that it will result in the large dependence of IFT on
WOR.
3% TC Blend/ 1%Na2CO3 / x% NaCl @ WOR=1:1
0.05% TC Blend / 1%Na2CO3 / x% NaCl @ WOR=10:1 (7 weeks after settling at 30 C in the environmental room)
(27 weeks after settling at 30 C in the environmental room)
0 2 5 7 9 10 11 12 13 14 15 16
2 4 6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8 9
viscous emulsions/liquid crystals Gel/viscous emulsion
Fig 2.5-2 TC Blend surfactant forms viscous emulsions
2-25
Another problem of the TC Blend surfactant is that it coalesces slowly at low
surfactant concentration and forms viscous emulsions and gels at high surfactant
concentrations (see Fig 2.5-2). The reason is that one component in the blend,
CS-330, has a linear hydrophobe, which is known to form viscous phases such
as liquid crystals at high salinity and low temperature.
Therefore, a surfactant or surfactant blend is needed which has a lower
optimal salinity at high surfactant concentration, and does not form viscous
phases. N67-3PO sulfate and N67-7PO sulfate were tested alone or with co-
surfactants/co-solvent.
2.5.3. N67-3PO S and Blends

4.0
N67-3PO S used at both 0.05 0.05 wt% N67-3PO S
Optimal NaCl Conc, %

3.5
and 3%, along with 1% Na2CO3 / 3 wt% N67-3PO S
3.0
NaCl, was mixed with crude oil
MY3 at different WORs. The 2.5
optimal NaCl concentration was 2.0
estimated from phase behavior 1.5
after a month, and plotted against 1.0

soap/synthetic surfactant mole ratio 0.5
(Fig. 2.5-3). The optimal salinity 0
plateaus at around 3% NaCl plus 1.E-02 1.E-01 1.E+00 1.E+01
1% Na2CO3. Without 1% Na2CO3, Soap/N67-3PO Ratio
i.e., with NaCl as the only
electrolyte, the optimal salinity is Fig. 2.5-3 Optimal salinity of N67-3PO S
around 4% NaCl. When the oil is / 1% Na2CO3 / x% NaCl
changed to alkanes, the optimal
1.E+00 6
IFT of ME vs Oil
IFT of ME vs Aq. 5
Visc of ME
Visc of Aq.
Viscosity, cp
IFT, dyne/cm
1.E-01 3
1.E-02 0
0 1 2
NaCl, %
Fig 2.5-4 IFT and viscosity of 3% N67-3PO S / 1%AOS C2024 / 8%SBA / 1%

Na2CO3 / NaCl
2-26
salinity with n-C5 is around 2% NaCl plus 1% Na2CO3; and with n-C7 the optimal
salinity increased to about 3% NaCl plus 1% Na2CO3. That the optimal salinity is
lower with n-C5 than with n-C7, i.e.,with lower alkane number, is consistent with
existing literature on microemulsions and EOR.
When N67-3PO S is used by itself, slow coalescence of emulsions in the
aqueous phase occurs. Therefore, several co-surfactants - N67-3EO S, AOS
C1618, AOS C2024, Oil-Chem 4-22, Triton X-100, Tergitol 15-S-7, or alcohol -
second butyl alcohol (SBA) were used with the N67-3PO S to test their ability to
enhance emulsion coalescence rate. It was found that at high N67-3PO S
concentration (3%), the addition of 8% SBA enhanced emulsion coalescence
behavior, but nothing else had significant effect. However, at low N67-3PO S
concentration (0.05%), up to 1% SBA does not help emulsion coalescence.
IFT and viscosity of 3% N67-3PO S / 1%AOS C2024 / 8%SBA / 1%
Na2CO3 / NaCl were measured (Fig 2.5-4). Tensions were found to all be above
0.01 dyne/cm, and viscosity below 5 cp. The high tension was caused by the
addition of alcohol. But alcohol prevented the formation of viscous emulsion.
N67-3PO S was not further researched because N67-7PO S has lower optimal
salinity and higher solubility parameter at optimal condition, obtained at 3%
surfactant with n-C10 (normal decane) as shown in Fig 2.5-5. The work in the
rest of this report is done with N67-7PO S.
35 35
30 30
25 25
Vw/Vs
Vo/Vs
20 20
7PO
15 15
3PO
10 10
5 5
0 0
1 2 3 4 5 6 7 8 9
%NaCl +1% Na 2CO3
Fig 2.5-5 Solubility parameters of 3% N67-3PO S and N67-7PO S, with n-C10

at WOR=1:1, 78 °F
2-27
2.5.4. N67-7PO Sulfate and Blends
Phase behavior of 3% N67-7PO S with crude oil MY3 at WOR=3:1 is shown
in Fig 2.5-6. The optimal salinity at this condition is around 2.2% NaCl, which
indicates that N67-7PO S is a much less hydrophilic surfactant than most
surfactants tested so far. The picture also shows that it forms liquid crystals
around the optimal condition (e.g. the iridescent region at 2.2% NaCl).
3%N67-7PO S / 1% Na2CO3 / x% NaCl @ WOR=3:1

170 days after settling at room temperature
x= 0 1 2 2.2 2.4 2.6 2.8 3
Fig 2.5-6 Phase behavior of 3% N67-7PO S with 1% Na2CO3 as a function of NaCl

concentration.
One way to prevent the formation of liquid crystals is to add alcohol. So 8%

SBA was added to 3% N67-7PO S, and the phase behavior is shown in Fig 2.5-
7. A middle phase was formed at 0.6-1.0% NaCl (with 1% Na2CO3). Compared
with 3% N67-7PO S by itself, the optimal salinity after adding SBA has been
reduced from 2.2% to lower than 1.5% NaCl (To actually decide the optimal
salinity, a few more phase behavior tubes should be prepared between 1.0 and
1.5%)
3%N67-7PO S / 8%SBA / 1% Na2CO3 / x%NaCl @ WOR=3:1

X= 0 0.5 0.6 0.7 0.8 0.9 1.0 1.5
3 phase region
Fig 2.5-7 Phase behavior of 3% N67-7PO S / 8% secondary butyl alcohol, with
1% Na2CO3 as a function of NaCl concentration.
2-28
10
Mixtures of N67-7PO + IOS-15/18

9
Salinity of Phase Separation

2-Phase Region
With 1% Na2CO3 / X % NaCl

8
7
6
5
4 1-Phase Region
3
2
1
0
1:1 4:1 9:1
IOS 1518 N67-7PO S
N67-7PO S:IOS 1518 (w/w)
Fig 2.5-8 Salinity of phase separation increases when N67-7PO S and IOS
1518 are blended compared with used alone.
As shown in the last section, adding alcohol will increase IFT, which is not
desired in the low tension induced, gravity dominated process. The alternative is
to add a branched surfactant. Abe et al (Abe et al, 1986) found when a linear
surfactant was used aloneat room temperature, a large amount of alcohol was
needed to have microemulsion free of liquid crystals; but when twin-tailed
surfactants were used, less, or even no alcohol was needed, depending on the
location of the hydrophilic group. Based on this finding, IOS 1518, which is a
twin-tailed surfactant with the
sulfonate group distributed along 12
the hydrocarbon chain, was 10
Optimal Salinity
blended with N67-7PO S. 8

When N67-7PO S or IOS 6
1518 are used separately, the 4
salinity of phase separation
2
occured at 4 and 3% NaCl (with
1% Na2CO3) respectively, as 0
1:1 4:1 9:1
shown in Fig 2.5-8. But when N67-7PO S
they are mixed at 1:1, 4:1, and N67-7PO S : IOS 1518 (w/w)
9:1 weight ratio, they stay in the Fig 2.5-9 Optimal salinity of 3% total
single-phase region over a much surfactant at different N67-7PO S and IOS
wider salinity range. 1518 ratio.
2-29
2.4% N67-7PO S / 0.6% IOS-15/18 / 1% Na2CO3 / x%NaCl @ WOR=3:1
X= 4 5 5.2 5.4 5.6 5.8 6 7
Fig 2.5-10 Phase behavior of 2.4% N67-7PO S / 0.6% IOS 1518 with 1% Na2CO3
as a function of NaCl concentration.
Optimal salinities at 3% total surfactant concentration of N67-7PO S and IOS

1518 mixed at the three 14
different weight ratios
Optimal NaCl Conc., %
12
mentioned above are TDA-4PO S : CS-330=1:1
summarized in Fig 2.5-9. The 10
optimal salinity at 1:1 ratio is too 8
high to be acceptable, and the
6 N67-7PO S : IOS-1518=4:1
emulsion coalescence rate at
9:1 ratio is not acceptable. At 4
the 4:1 ratio (Fig 2.5-10), when 2 Formation brine salinity
the vials were tilted, all
0
interfaces moved smoothly with
no evidence of viscous phases. 1E-02 1E-01 1E+00 1E+01
At this ratio, the optimal salinity Soap/Synthetic Surfactant Mole Ratio
is closer to that of the formation
brine (Fig 2.5-11). Therefore the Fig 2.5-11 Soap/Synthetic Surfactant Mole
4:1 ratio was chosen to test its Ratio Correlation
ability to lower IFT and improve
oil recovery.
2.5.5. N67-7PO S: IOS 1518 = 4:1 (wt)

The mixture of 4:1 weight ratio of N67-7PO S and IOS 1518 (NI Blend) with
1% Na2CO3 was examined as to its adsorption on calcite powder, and its ability
to lower IFT and improve oil recovery.
2-30
2.5.5.1. Adsorption
Surfactant adsorption was determined by potentiometric titration with
hyamine. The initial surfactant concentration was fixed, while surfactant solution
and calcite powder were mixed at varied weight ratios. The equilibrium surfactant
concentration was determined by titration, while the calcite powder surface area
determined by BET adsorption. With thus information, surfactant adsorption
density was calculated.
The adsorption of NI Blend without or with sodium carbonate is shown in Fig
2.5-12. Two initial surfactant concentrations were used: 0.05 wt% (0.76 mmol/l)
and 0.1 wt% (1.52 mmol/l). The equilibrium adsorption density of the surfactant
without sodium carbonate is 1.5-2 mg/m2. But after adding 1% Na2CO3, the
adsorption is reduced several fold to about 0.1 mg/m2.
Adsorption Density, mg/m2
2.5
1.5
0.5
0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
Residual Surf. Conc., mmol/l
0.05% NI Blend 0.1% NI Blend

0.05% NI Blend / 1% Na2CO3 0.1% NI Blend / 1% Na2CO3
Original 0.05% NI Blend. Original 0.1% NI Blend.
Fig 2.5-12 Adsorption of NI Blend on calcite powder
2.5.5.2. Phase Behavior and IFT

Phase behavior at low surfactant concentration, 0.05% NI Blend / 1% Na2CO3
/ x% NaCl, is shown in Fig 2.5-13. Traditionally, IFT was measured between
microemulsion and excess brine or microemulsion and excess oil phases. But at
very low surfactant concentration, there is not enough surfactant to form a large
enough microemulsion phase for IFT measurements. In these cases it is often
difficult to get reproducible measurements of IFT as discussed in last year’s
report. A new procedure has been developed to overcome this problem.
Most obvious difference between samples at 7 days and 1 month is that as
brine phase at 2.0% NaCl cleared up, IFT also changed with time, as shown in
Fig 2.5-14, because the surfactant rich microemulsion phase has settled to the
2-31
0.05 % NI Blend / 1% Na 2 CO 3 / x% NaCl @ WOR=3:1, Room Temperature
X= 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 3.0
Settling time: 7 days
Settling time: 1 month
Fig 2.5-13 Phase Behavior of 0.05% NI Blend / 1% Na2CO3 / x% NaCl
interface. This phenomenon was observed with CS-330, TDA-4PO S, and their
blend, and reported previously (Zhang et al, 2004). It was found that if a small
amount of the middle layer was added, IFT could be lowered to ultra-low values
again. But because the microemulsion layer was so thin, the result was difficult to
reproduce.
A new sample of 1.8% NaCl was prepared. After settling for 6 days, excess
brine and oil phases were taken for IFT measurements, but the aqueous phase
was too dark to see the oil phase in the spinning drop apparatus. Therefore the
lower phase was separated from the oil phase, and centrifuged at 6,000 RPM for
1E+01
1E+00
IFT, dyne/cm
1E-01
7-19 days settling 26 days settling
Bottom of the centrifuged lower phase (clear)
1E-02
M iddle of the centrifuged lower phase
80% botthom + 20% topm ost of the
centrifuged lower phase
1E-03
0 1 2
NaCl, %
Fig 2.5-14 IFT of 0.05% NI Blend / 1% Na2CO3 / x% NaCl
2-32
30 minutes. There was a turbidity gradient in the centrifuged solution - the bottom
is clear while the top is the most turbid. IFT was measured between the
equilibrated oil phase and aqueous phase sampled from different position of the
centrifuged lower phase. When sampled from the bottom where it was clear, IFT
was close to 0.1 dyne/cm. When sampled from the middle, where it was turbid,
IFT was still high – close to 0.1 dyne/cm. But when mixing 80% bottom clear
solution with 20% topmost turbid layer (about top 2% of the centrifuged lower
phase), the IFT was low – 0.003 dyne/cm. This indicates that turbidity of the
lower phase is a necessary but not sufficient condition for low IFT.
This last result also suggests an effective way of measuring IFT: A system
should be settled only until enough oil coalesces for IFT measurement. Then IFT
should be measured. If the lower phase is too dark to see the oil drop, the
solution should be centrifuged. The clear bottom and upmost turbid layer should
be mixed at a certain ratio, e.g. 4:1, as the aqueous phase for the spinning drop
IFT measurements.
2.5.5.3. Oil recovery by centrifuge imbibition

Centrifuge imbibition experiments were conducted instead of spontaneous
imbibition experiments because imbibition in centrifuge took much less time and
allowed multiple experiments – 6 cores can be processed at the same time.
1” diameter by 1” length Texas Cream limestone cores were used in the
Axis
x
x
x
x
Stainless
x
x
Formation
x Core Brine or
Screen x
x rotor
x
x
Surf. Sol.
x
x
Fig 2.5-15 Schematic set-up of centrifuge imbibition with formation brine of

surfactant solution.
centrifuge imbibition experiments. The cores
jacketed with heat shrinkable Teflon sleeve Table 2.5-2 Bond number
were first subjected to vacuum for over 8 at 6,000 RPM, Perm=6 md
hours, then brine vacuum-saturated for 8 hours
before being pressure-saturated for another 8 IFT, mN/m NB
hours. The brine saturated cores were 30 1.6*10-6
centrifuged in crude oil at 15,000 RPM, which
1 4.7*10-5
corresponds to about 80 psi, for 12 hours, then
the cores were turned upside down to 0.1 4.7*10-4
centrifuge at the same speed for another 12 0.01 4.7*10-3
hours. After the oil-saturated cores were aged
0.001 4.7*10-2
2-33
NB
Fig 2.5-16 Correlation between residual oil saturation and Bond

number. Dots are marked according to Table 2.5-2
at 80 °C for 48 hours, they were centrifuged at 6,000 RPM, which corresponds to

about 27 psi, in either formation brine or surfactant solution. The schematic set-
up is shown in Fig 2.5-15. The stainless steel screen was used to allow the
aqueous phase to be present at the end of the core and co-current flow allowed
to occur. 0.2% NI Blend / 1% Na CO / x% NaCl
2 3
(9 days after settling at room temperature)
Alkaline-surfactant solution improves oil
WOR=1:1 WOR=3:1
recovery by altering wettability and lowering x= 2 3 4
x= 1 1.5 2
interfacial tension. Bond number at 6,000
RPM is calculated as a function of IFT (Table
2.5-2), and marked as red (water-wet) and
green (oil-wet) dots on the correlation
between residual oil saturation and Bond
number (Fig 2.5-16). The figure suggests that
if IFT can be lowered to 10-3 dyne/cm
(mN/m), ideally, all oil can be recovered.
The oil recovery with 0.05% NI Blend is
only marginally better than with formation
brine, because 0.05% is close to the CMC of
the blend surfactant, and when the surfactant
solutions entered cores and mixed with
formation brine, they were diluted and their
concentrations dropped below CMC which
may cause the system not to have low
tension any more. Therefore, 0.2% NI Blend Fig 2.5-17 Phase behavior of
was used instead. 0.2% NI Blend at WOR = 1 and 3
2-34
Table 2.5-3 Core properties, experimental conditions and results
NaCl, kair, Porosity PV, ml OOIP, So Centrifuge OR, Sor Rec.,
% md ,% ml time, days ml OOIP%
4.933 21.2 2.661 2.11 0.79 4 1.45 0.25 69

Forma-
tion ~1 6.821 22.4 2.897 2.36 0.81 4 1.53 0.28 65
brine
8.873 24.2 2.972 2.35 0.79 3 1.60 0.25 68
1 9.485 24.3 3.048 2.41 0.77 3 2.02 0.13 84
1.5 4.642 21.8 2.77 2.00 0.72 6 1.64 0.13 82
1.8 4.281 20.6 2.687 1.94 0.72 6 1.61 0.12 83
0.2% 2 8.748 25.1 3.415 2.68 0.79 3 2.29 0.11 86

NI
blend / 2.2 4.349 22.1 2.843 2.01 0.71 6 1.79 0.08 89
1%
Na2CO 2.4 4.445 20.2 2.698 1.97 0.73 6 1.60 0.14 81
3 / x%
NaCl 2.5 4.581 22.3 2.880 2.07 0.72 6 1.92 0.05 93
2.7 4.284 21.1 2.840 2.02 0.71 6 1.76 0.09 87
3 5.228 21.6 2.853 2.22 0.77 5 1.79 0.15 81
4 7.705 24.3 3.108 2.48 0.79 3 2.02 0.15 81
5 9.210 24.4 3.231 2.59 0.8 3 1.96 0.19 76
Phase behavior of 0.2% NI Blend / 1% Na2CO3 / x% NaCl at WOR of one

and three is shown in Fig 2.5-17. The optimal salinity at WOR = 1:1 is between 1
and 1.5% NaCl, while that at WOR = 3:1 is between 3 and 4% NaCl.
The core info and experimental conditions and results are listed in Table 2.5-
95
0.2% new blend/1% Na2CO3/NaCl
Recovery, %OOIP
90
85
80
75
70 Water flooding
65
1 2 3 4 5
NaCl, %
Fig 2.5-18 Oil Recovery by centrifuge imbibition in either formation brine

(water flooding) or 0.2% NI Blend / 1% Na2CO3 / x% NaCl.
2-35
3. Centrifuge time is inversely proportional to permeability (Hirasaki et al, 1995),
which is the reason for the difference in the centrifuge time
The recovery is plotted in Fig 2.5-18. The oil recovery by formation brine is
67% on average, while the oil recovery by 0.2% NI Blend / 1% Na2CO3 / x% NaCl
is much higher-76-93%. Further experiments are planned at NaCl concentrations
between 2 and 3% to establish whether higher recovery can be achieved there
as suggested by some of the data points.
Conclusions
1. Blend of N67-7PO S and IOS 1518 at 4:1 weight ratio has phase behavior
free of viscous phases and all interfaces can move smoothly. Its optimal
salinity is closer to that of the formation brine than the previous TC blend.
2. Sodium carbonate reduces severalfold the adsorption of NI Blend on calcite.
3. When emulsions are present near optimal condition at low surfactant
concentration, one sometimes observes a rather turbid lower phase that
becomes clear with time. The IFT of the excess oil with the clear lower phase
was always high. Turbidity of the lower phase is a necessary but not
sufficient condition for ultra-low IFT. A procedure to capture enough
microemulsion for low IFT is introduced.
4. 0.2% NI Blend increases oil recovery from 67% of water flood to at least 85%
and perhaps higher.
Reference
1. Abe, M, Schechter, D., Schechter, R.S., Wade, W.H., Weerasooriya, U., and
Yiv, S., “Microemulsion Formation with Branched Tail Polyoxyethylene
Sulfonate Surfactants”, JCIS 114 (2), 1986
2. Hirasaki, G.J., Rohan, J.A., Dudley, J.W., “Interpretation of Oil-Water Relative
Permeabilities from Centrifuge Experiments”, SPE 24879 in SPE Advanced
Technology Series, 3(1), 66-75, March 1995.
3. Zhang, D.L., Liu, S., Puerto, M., Miller, C., and Hirasaki, G.J., “Wettability
Alteration and Spontaneous Imbibition in Oil-Wet Carbonate Formations”,
presented at the 8th International Symposium on Evaluation of Wettability and
Its Effect on Oil Recovery, Houston, TX, May 16-18, 2004. Accepted by J.
Pet. Sci.& Eng.
2-36
Task 3 Foam for Mobility Control
3.1 Foam diversion in fracture network model

The main purpose of this research is to investigate the foam flow and foam
diversion effect in heterogeneous fracture system. From our previous work [DOE
annual report 2003-2004, OSTI ID: 829648], we found two contributions to the
apparent viscosity when foam flows in homogeneous fracture. Here the theory
was extended to heterogeneous fractures, and the parameters affecting the
diversion by foam were studied.
Experimental technique
The fracture model has been established as in Fig. 3.1-1.The fracture model
Variable thickness
0.1 mm
30.48 cm
35.56 cm
15.24 cm 20.32 cm
25.40 cm
Fig. 3.1-1. Detailed diagram of heterogeneous fracture model
consists mainly of two parallel glass plates. Changing the gasket thickness
between the two parallel plates can change the aperture of the fracture. The
method of making the heterogeneous fracture is to adhere a 0.1mm thickness
glass to one of the glass plates (shaded region in Fig. 3.1-1). The procedure is:
1. Adhere the 0.1mm glass to the thick glass by Norland optical adhesive; 2. Roll
the thin glass on the thick glass to squeeze out any air and excess adhesive
between them; 3. Cure the adhesive by a high intensity ultraviolet lamp (Model
Spectroline SB-100P) for over 48 hours. Two spacers are put at the two ends of
3-1
the model between the different apertures to ensure the uniform aperture at each
side.
Transducer
HP digital camera
Surfactant solution
syringe pump
Computer for recording
pressure difference
frit
Fracture model
Air Foam generator
Air mass flow controller
Fig. 3.1-2. Set –up diagram for foam mobility control experiment in fracture
model
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 surfactant solution are introduced on opposite sides of the frit in
the foam generator. Choosing frits with different pore sizes can generate different
sizes of bubbles. Two grooves were made along the inlet and outlet of the
fracture model to assure uniform pressures at the inlet and outlet.
The surfactant solution in the experiments was 0.5% C13-4PO and 0.5% CS330.
STEOL C13-4PO is from Harcros Company and its chemical description is
propoxylated C13 alcohol ether sulfate, ammonium salt. STEOL CS330 is from
Stepan Company and its chemical description is C12-3EO sulfate. The salinity
was 0.23% NaCl, 0.07% CaCl2 and 0.04% MgCl2. The bubble diameters in the
experiment were from 0.4 to 1.0mm. The aperture is 0.1 mm or 0.2 mm for
homogeneous fracture experiments and the combinations of 0.1 mm/0.2 mm or
0.05 mm/0.15 mm for heterogeneous fracture experiments. The gas fractional
flow range was from 0.0 to 0.9. The values used for the viscosity of solution and
surface tension were 1.0 mPa.s and 28 mN/m from measurement.
3-2
Theory
Apparent viscosity
The theory has been described in our DOE annual report 2003-2004 [OSTI ID:
829648]. But we repeat it here because it is applied to explain our new
experimental results. From our previous research, the most important variable
affecting foam apparent viscosity in uniform, smooth capillaries is foam texture.
The principal factors affecting apparent viscosity of foam in uniform fracture are
dynamic changes at gas/liquid interfaces. The apparent viscosity is the sum of
two contributions as in Fig. 3.1-3:
1. Slugs of liquid between gas bubbles resist flow.
2. Viscous and capillary forces result in interface deformation against the
restoring force of surface tension.
Fig. 3.1- 3. Mechanism of affecting apparent viscosity in fracture system
The apparent viscosity from the contribution of foam bubble deformation in

uniform fracture can be predicted from the Plane-Poiseuille flow:
μ app
shape
= n L Δ pb 2 /( 12 U ) = 0 . 57 ( μ liq n L b )( 3 μ liq U / σ ) − 1 / 3 /( rc / b ) (1)
where nL is the number of equivalent lamellae per unit length and b is the
aperture.
n L and rc are important parameters in determining the value of apparent

viscosity. By assuming that individual bubbles are distributed uniformly in the
fracture, the number of lamellae per unit length can be expressed in terms of the
aperture and the equivalent bubble radius:
[
n L = 3 f g b /(4πrB )
3
]
1/ 2
(2)
3-3
where f g is fractional flow, b is the fracture aperture and rB is the equivalent
bubble radius. The equivalent bubble diameter is bigger than the fracture
thickness. By assuming the bubbles are normally not in contact, the radius of
curvature, rc , is equal to the half aperture of the fracture.
The contribution to apparent viscosity from liquid slugs in uniform fracture can be
predicted from the contribution of liquid viscosity in the total fluid. That is:
μapp
liq
= (1 − f g ) μ liq (3)
where μ liq is the viscosity of pure liquid and f g is the gas fractional flow.
The total apparent viscosity from theory when foam bubbles flow in uniform
fracture is the sun of the two above contributions.
μ app = μ app
liq
+ μ app
shape
(4)
The total apparent viscosity can be obtained from measuring the pressure
difference across the model. That is from Plane-Poiseuille law:
μ app = b 2 ∇p /(12U ) (5)
where ∇p is the pressure gradient.
The velocities in both aperture regions of the heterogeneous fracture need to be

determined to get the apparent viscosity either by theory or measurement. If only
water or surfactant solution is injected, the apparent viscosity is around 1.0 cp.
By assuming the pressure gradient is the same, the velocity ratio in different
aperture regions can be derived from equation (5) as below.
U 1 / U 2 = (b1 / b2 ) 2 (6)
For foam case, when contribution to apparent viscosity from bubble deformation
dominates, by setting equation (1) equal to equation (5), assuming that bubble
size and fractional flow are equal and that the pressure gradient is the same, the
velocity ratio in different aperture portions of fracture can be derived as a function
of the ratio of apertures as in equation (7).
U 1 / U 2 = (b1 / b2 ) 3 / 4 (7)
Then from overall material balance, the velocities in different aperture regions of
fracture can be estimated.
3-4
Combining equation (1) with equation (7), the apparent viscosity from
deformation contribution increases with the 5/4 power of aperture ratio, when the
other conditions are the same. This result predicts liquid diversion into thinner
fracture region because the flow encounters more resistance in higher thickness
region.
Experimental results
All the experiments were done at Reynolds number less than 1.0, and the
average bubble diameters were bigger than the aperture to meet the conditions
of our theory from previous research. The Reynolds numbers were Re=0.22 and
Re=0.44. The bubble diameters were 0.4mm and 0.6mm. Gas fractional flow, fg
was 0.0, 0.15, 0.25, 0.33, 0.67 and 0.9. Two different aperture ratios were used
in the experiments: 0.10 mm/0.20 mm and 0.05 mm/0.15 mm or 1:2 and 1:3.
1. Apparent viscosity
The apparent viscosity in each region can be calculated using either equation
(4) or equation (5). It is restricted to the case in which both regions of fracture
are fully filled with foam so that there are minimal lateral pressure gradients
and no cross flow. Also, we assume the fractional flow is the same in thinner
and thicker apertures of the heterogeneous fracture. But by either method,
the flow velocity in each layer is needed. They can be obtained from equation
(7) and material balance.
The results of apparent viscosity from theory and measurement of pressure

difference are shown in the Fig. 3.1-4 ~ 3.1-7. The apparent viscosity from
theory is from equation (4), which includes the contributions from both bubble
deformation and liquid slugs between bubbles. In the scope of our
experiments, at aperture ratios 0.05mm/0.15mm and 0.1mm/0.2mm, bubble
size 0.4mm and 0.6mm, Reynolds number 0.22 and 0.44, the apparent
viscosity from measurement of pressure difference fits well with that from
theory prediction. This agreement between theory and experiment confirms
the assumption of same fractional flow in both apertures.
Another assumption is that the contribution from foam bubble deformation

dominates in the apparent viscosity. This assumption is valid for medium to
high gas fractional flow. For example, at fg = 0.9, the contribution from liquid
slug is only 0.1 cp and the contribution from foam bubble deformation is
above 3 cp in the scope of all our experiments. At low gas fractional flow,
because the contribution from liquid slug to apparent viscosity can’t be
neglected, equation (7) may not reflect the real velocity ratio in different
layers. The reason may be that at low apparent viscosity, for example fg =
0.15, the contribution from liquid slug is 0.85 cp. If the contribution from foam
bubble deformation is 2.0 cp, the liquid slug contribution can be about 30% of
the apparent viscosity. So one might expect deviation of the data from
3-5
experiment from theory prediction at low gas fractional flow and indeed some
deviation is seen in the plot.
aperture = 0.05mm/0.15mm,DB=0.4mm,
Re=0.22
0.05mm aperture region from measurement

0.05mm aperture region from theory
100.00 0.15mm aperture region from measurement
Apparent viscosity (cp)
10.00
1.00
0 0.2 0.4 0.6 0.8 1
Fractional flow
Fig. 3.1- 4. Apparent viscosity for aperture ratio 0f 0.05 mm/0.15 mm,
bubble size = 0.4mm, and Re = 0.22
aperture = 0.1mm/0.2mm,DB=0.4mm
Re=0.22
0.1mm aperture region measurement
100.00 0.1mm aperture region from theory

Apparent viscosity(cp)
10.00
1.00
0 0.2 0.4 0.6 0.8 1
Fractional flow
Fig. 3.1- 5. Apparent viscosity for aperture ratio of 0.1 mm/0.2 mm, bubble
size = 0.4mm and Re = 0.22
3-6
aperture = 0.10mm/0.20mm,DB=0.4mm,
Re=0.44
100.00
10.00
1.00
0 0.2 0.4 0.6 0.8 1
Fractional flow
size = 0.4 mm and Re = 0.44
fracture thickness = 0.10mm/0.20mm,DB=0.6mm,

Re=0.22
100.00
0.2mm apertureregion from measurement
10.00
1.00
0 0.2 0.4 0.6 0.8 1
Fractional flow
size = 0.6 mm and Re = 0.22
From Fig. 3.1-4~7, it is obvious that the gas fractional flow is an important
factor in determining the apparent viscosity of foam flow. By comparing Fig.
3.1-4 and Fig 3.1-5, the difference of apparent viscosity between the thin and
thick layers is bigger for the aperture ratio 0.05 mm/ 0.15 mm or 1:3 than for
aperture ratio 0.1 mm/0.2 mm or 1:2. But from Fig 3.1-5, Fig 3.1-6 and Fig
3.1-7, the different bubble size (0.4mm and 0.6mm) and different velocity (Re
= 0.22 and 0.44) doesn’t affect much the difference of apparent viscosity
between the thin and thick layers. So we may expect distinct diversion effect
3-7
or sweep efficiency improvement for high aperture ratio by high gas fractional
flow foam sweep.
2. Sweep efficiency
Pictures were taken during experiments to investigate the sweep by foam and
surfactant solution in heterogeneous fracture. An example of foam/surfactant
solution sweep is shown in Fig. 3.1-9. For comparison, the picture of sweep
by surfactant solution alone is shown in Fig. 3.1-8.
The injection of foam bubbles together with surfactant solution improves the
sweep over the injection of surfactant solution only. From Fig. 3.1-8 for
0.05mm/0.15mm fracture, at Reynolds number 0.22, it takes about 6.6 pore
volumes of surfactant solution to sweep the 0.05 mm aperture region, while
from theory prediction 7.0 pore volume is needed. With gas fractional flow 0.9
and bubble size at 0.4 mm in diameter, only 0.15 pore volume of surfactant
solution is needed, which is shown in Fig. 3.1-9. The amount of surfactant
solution needed to sweep both regions is reduced by a factor of more than
40.
By using the similar derivation as Lake [Enhanced Oil Recovery, Prentice

Hall, 1989, Page 201-205], the volume of surfactant solution required for
sweep or dimensionless time can be obtained. The derivation is based on the
assumptions:
1. No crossflow;
2. Equal pressure difference across different layers along the flow direction;
3. Plug flow;
4. The foam apparent viscosity is determined by the velocity of foam flow at
steady state, i.e., the velocity after both layers have been swept.
The foam front position in each layer may be determined from Darcy’s law
dxl k Δp
= vl = −( ) l λrel , l=1, 2 (8)
dt ΔS L
where ΔS is the water saturation change, vl is the interstitial velocity in layer l,

and λrel is the relative mobility in layer l defined by
xl (1 − xl )
λ rel = [ + ] −1 for xl < 1 (9)
λ0
r1 λ 0
r2
or λ rel = λ0r1 for xl > 1
3-8
0.41 PV 0.82 PV 1.23 PV 1.64 PV
2.05PV 6.56 PV
Fig. 3.1-8. Surfactant solution sweeping heterogeneous fracture, Re = 0.22,

aperture ratio = 0.05 mm (top)/0.15 mm (bottom) (Note: PV refers to pore
volume of the entire heterogeneous fractures)
0.04 PV 0.08 PV 0.12 PV 0.16 PV
0.04 PV 0.08 PV 0.12 PV 0.16 PV

Fig. 3.1-9. Foam/surfactant solution sweeping heterogeneous fracture, Re =
0.22, fg = 0.9, aperture ratio = 0.05 mm/0.15 mm, dB = 0.4 mm (Note: PV is B
the injected liquid pore volumes)
Taking the ratio of the interstitial velocities in the two layers will eliminate time
and the pressure drop since both layers experience the same Δp . Thus before
breakthrough ( xl < 1), we have
dx1 k1 μ12 x 2 + (1 − x 2 ) M 2
= (10)
dx 2 k 2 μ11 x1 + (1 − x1 ) M 1
where M1, M2 are the mobility ratios, e.g. M = λ0r1 / λ0r 2 , k1, k2 are the
permeabilities, k = b 2 / 12 in fractures where b is the aperture, and M 1 = μ 0 / μ11 ,
μ12 are the foam apparent viscosities in thinner and thicker layers respectively,
M 1 = μ 0 / μ11 .and M 2 = μ 0 / μ12 where μ 0 is the viscosity of water.
∫ k2 μ 11[ x1 + (1 − x1 ) M 1 ]dx1 = ∫ 0x 2 k1μ12[ x2 + (1 − x2 ) M ]dx2

x1
0 (11)
2
3-9
(1 − M 1 ) 2 k μ (1 − M 2 ) 2
x1 + M 1 x1 = 1 12 [ x2 + M 2 x2 ] (12)
2 k 2 μi1 2
When the foam front reaches the outlet of the thicker layer (x2 = 1.0), the
dimensionless front position at the thinner layer is
1/ 2
⎧ ⎫
⎪⎪ (1 − M 1 )(1 + M 2 ) ⎪⎪
⎨ 1M 2
+ ⎬ − M1
⎪ k μ ⎪
( 2 11 )
⎪⎩ k1 μ12 ⎪⎭
x1 = (13)
1− M1
Then the dimensionless time in the liquid pore volume needed for the sweep of
thicker region is
x1b1 + b2
Dimensionless time = (1 − f g ) (14)
b1 + b2
When the thinner layer was swept, the dimensionless front at the thicker layer
(outside the fracture) can be obtained from
dx2 k 2 μ11
= [ x1 + (1 − x1 ) M 1 ] (15)
dx1 k1 μ12
x2 x1 k 2 μ11
∫1
dx2 = ∫ 0
x1 k1 μ12
[ x1 + (1 − x1 ) M 1 ]dx1 (16)
k 2 μ11 (1 − M 1 ) 2
x2 = 1 + [ ( x1 − ( x10 ) 2 ) + M 1 ( x1 − x10 )] (17)
k1 μ12 2
1/ 2
⎧ ⎫
⎪⎪ (1 − M 1 )(1 + M 2 ) ⎪⎪
⎨ 1M 2
+ ⎬ − M1
kμ
⎪ ( 2 11 ) ⎪
⎪⎩ k1μ12 ⎪⎭
Setting x1 = 1.0 , and x 1 =
0
1 − M1
3-10
k 2 μ11 (1 + M 1 ) (1 + M 2 )
x2 = 1 + [ − ] (18)
k1 μ12 2 k 2 μ11
2( )
k1 μ12
And the dimensionless time or the liquid pore volume needed to sweep the
thinner layer is
x2b2 + b1
Dimensionless time = (1 − f g ) (19)
b2 + b1
Fig. 3.1-10 shows the comparison between the theory prediction and experiment
results. The theory prediction well matches the data from experiment. This
validates our assumptions for the theory calculation. Also, it demonstrates the
value of foam in greatly reducing the amount of surfactant required to sweep a
heterogeneous fracture.
10
DB=0.4mm, Re=0.22
Aperture 0.05mm/0.15mm
Liquid pore volume
0.05mm aperture region from experiment

0.1 0.05mm aperture region from theory
0.15mm aperture region from experiment
0.01
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Gas fractional flow
Fig. 3.1- 10. Comparison between the calculation and the experimental results
for foam/surfactant sweep in heterogeneous fractures with 1:3 aperture ratio
Conclusions
We investigated different factors’ effects on sweep efficiency by foam in

heterogeneous fracture and applied our theory for different contributions to foam
apparent viscosity. Some conclusions can be made accordingly.
1. The foam apparent viscosity in heterogeneous fracture is from two

contributions-bubble deformation and liquid slugs between bubbles, which
is the same as in homogeneous fracture.
3-11
2. The amount of surfactant required to sweep both regions of a
heterogeneous fracture can be decreased by more than a order of
magnitude.
3. Gas fractional flow, aperture ratio and bubble size can greatly affect the
sweep efficiency.
4. Predictions based on assuming the same fractional flow in each layer of a
heterogeneous fracture yield results in good agreement with experimental
results for the conditions studied here.
3.2 Test of surfactant’s ability in generating foam
From previous description, foam can greatly improve the sweep efficiency in
fracture network. But the success of the foam sweep depends strongly on the
strength and stability of foam, which is related to the surfactant composition and
salinity. The surfactant composition and salinity in the above research are from
an aquifer remediation project and were used in the fracture study because they
had been demonstrated to form strong foam in sand packs. The surfactant used
in the EOR process needs to be tested near its optimal salinity. We tested the
strength of foam by coinjection of surfactant solution and air into a horizontal
sand pack to generate foam and check the pressure difference across the sand
pack to get the apparent viscosity of foam. The experimental device is shown
schematically in Fig. 3.2-1. The surfactant tested is various mixtures of N67-
7POS and IOS1518 for total surfactant concentration of 0.5 wt% at the salinity
1% Na2CO3 and 2% NaCl. The sand pack is 1 foot long and its permeability is 40
darcy.
Transducer
Surfactant solution
Air Sand column

Syringe pump Computer for recording
pressure difference
Fig. 3.2-1. Set –up diagram for foam strength test
Fig. 3.2-2 shows the results of foam strength tests for different surfactant
compositions and different velocities at the same salinity. With total surfactant
concentration 0.5%, at mixing ratio N67-7POS with IOS1518 at 1:1 to 4:1, the
foam apparent viscosity is almost the same as N67-7POS itself at concentration
0.5%. The foam strength by the blend of N67-7POS&IOS1518 is weaker than
that by the old blend C13-4PO&CS330 at the same salinity. The other finding is
that IOS1518 alone is a strong foamer. At the superficial velocity around 1 ft/day,
3-12
the apparent viscosity of foam by IOS1518 can be close to 900 cp. One reason
may be that IOS1518 is hydrophilic, i.e., well below its optimal salinity with most
oils at the conditions studied, such surfactants are generally good foamers. Fig.
3.2-3 indicates for the mixture of N67-7POS and IOS1518, the foaming ability is
not affected by different salinity. So the conclusion is that the foaming ability of
the mixture of N67-7POS and IOS1518 is limited by the foaming ability of N67-
7POS. From our experience in 1-D to 3-D sand pack experiments in the aquifer
remediation project, the foam strength in 3-D sand pack by the same surfactant
and salinity can be 50 times weaker than in 1-D sand pack. But from the
preliminary experiments, at least for the few conditions studied so far, behavior of
new mixture of surfactants in fractures performs almost the same as for previous
mixture. Other possible surfactant mixtures for EOR will be tested in the future.
And the other future work is to test the foam stability at the presence of residual
oil.
Horizontal 1 foot sand colum n, 40 darcy, fg=0.5,no N67-7POS,0.5% IOS1518

1% Na2CO3, 2% NaCl fg=0.5,0.05% C13-4PO&CS330(1:1)
fg=0.5,0.5% N67-7POS&IOS1518(1:1)
fg=0.5,0.5% N67-7POS&IOS1518(4:1)
1000
fg=0.5,0.5% N67-7POS,no IOS1518
100 fg=0.0
Old surfactant blend from
10 aquifer remediation project,
0 05% C13 4PO&CS330(1 1)
0.1
0.1 1 10 100 1000
Superficial Velocity (ft/day)
Fig. 3.2- 2. Foam strength at different surfactant composition
Horizontal 1 foot sand colum n, 40 darcy, 0.5% N67-7POS&IOS1518(4:1),2%

gas fractional flow = 0.5 NaCl, 1% Na2CO3
0.5% N67-7POS&IOS1518(4:1),0.5%
100 NaCl, 1% Na2CO3
0.5% N67-7POS,No IOS1518,1%

NaCl, No Na2CO3
10
0.1
0.1 1 10 100 1000
Superficial velocity (ft/day)
Fig. 3.2- 3. Foam strength at different surfactant composition and salinity

3-13
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. An extensive literature search to gather data for residual oil
saturation, relative permeability curves, and capillary desaturation curves for
different wetting conditions has been performed. Based on these data, a
procedure was developed to compute the capillary desaturation and relative
permeability curves as a function of wettability. The next step is to implement the
procedure in UTCHEM.
Effect of Wettability on Residual Oil Saturation, Relative Permeability, and

Capillary Desaturation Curves
The wettability is the measure of the preference that a rock has for a
particular fluid. If a rock is water-wet, the water phase will occupy the small
pores and coat the remaining rock surfaces. The location of the fluids in an oil-
wet rock will reverse from the water-wet case. In some instances, a rock will not
have a strong preference for any fluid and is thought to have a neutral wettability.
In other cases, a rock will have a varying wettability throughout and is defined as
having mixed wettability (Anderson, 1986). Presently, three methods (USBM,
Amott-Harvey, and contact angle measurements) are used to calculate the
wettability of a rock. For this work, data obtained from the Amott-Harvey method
was used. The Amott-Harvey wettability test utilizes a series of spontaneous
imbibition and forced displacement processes to calculate the change in fluid
volumes expelled from the core as described in Amott, 1959. The Amott-Harvey
wettability index (Iw) can range from 1 (strongly water-wet) to -1 (strongly oil-wet).
The wettability of a rock is important because it controls the residual oil
saturation, relative permeability curves, and the capillary desaturation curves
(CDC).
The effect of wettability on the residual saturation is based on the location

and distribution of the fluids within the rock pores. The wetting phase, occupying
the small pores and coating the walls of the larger pores, is immobile due to high
capillary forces. The non-wetting phase, which is present as globules in the large
pores or across several pores, is held immobile by the interfacial tension
between the two phases. Several papers have documented the change in
residual oil saturation with changes in the Amott-Harvey Wettability Index (Figure
4.2.1). The common conclusion is that the residual oil saturation is lowest under
neutral-wet conditions. Data provided by Jerauld, 1997 indicates that the
residual water saturation also follows a similar tend. Another interesting finding
is the trend of the residual oil saturation under oil-wet conditions, which mirrors
the trend under water-wet conditions (Figure 4.2.1).
4-1
The water/oil relative permeability curves are also dependent upon the
wettability state of a rock. The primary changes to the relative permeability
curves are the location of the crossover point and the movement of the endpoint
relative permeabilities for water and oil. For a strongly water-wet rock, the
crossover point of the water/oil relative permeability curves will occur at a water
saturation greater than 0.5. This effect is due to the low water endpoint relative
permeability (~0.1 to 0.3) and high oil endpoint relative permeability (~1). For a
strongly oil-wet rock, the crossover point will occur at a water saturation less than
0.5. This is due to the higher water endpoint relative permeability (~0.4 to 0.7)
and reduced oil endpoint relative permeability. To a lesser extent, the relative
permeability curves are shifted due to the changes in residual phase saturations
of differing wettability conditions. An example of the change in relative
permeability with changes in wettability was obtained from Morrow, 1973 and
curve fit using Corey type parameters as shown in Figure 4.2.2. It is apparent
from this example that the primary effects of changes in wettability are the
increase in the water relative permeability curve and the decrease in the oil
relative permeability curve when shifting from water-wet to oil-wet. This effect is
shifting the crossover point to lower water saturations. The minor effects of
changes in residual oil saturation can also be seen.
The effect of wettability on capillary desaturation is also apparent. Local

phase trapping occurs until the viscous forces overcome the capillary forces. A
dimensionless constant to relate the viscous and capillary forces known as the
capillary number was first introduced by Brownell and Katz, 1949. A common
usage of the capillary number is to relate it to the residual phase saturations
using a CDC. An example CDC for a water-wet sandstone is shown in Figure
4.2.3. As expected, the wetting phase (water) requires a higher critical capillary
number (Ncc) to begin desaturation compared to the non-wetting phase (oil). The
effect of changes in wettability on the CDC of a Berea sandstone is shown in
Figure 4.2.4. This plot shows the variation of the oil Ncc and the residual oil
saturation at low capillary number for different wettability conditions. Changes in
the CDC for weakly oil-wet and neutral-wet carbonate rocks are shown in Figure
4.2.5. The key observation of this plot is the extremely low values of Ncc for each
CDC. The Ncc for the carbonate rocks in Figure 4.2.5 is three orders of
magnitude lower than the mixed-wet and oil-wet sandstone values shown in
Figure 4.2.4. A few possible explanations can explain the behavior of a
carbonate compared to a sandstone: differences in pore-size distribution,
permeability, porosity, and fluid distributions.
Simulations with Differing Wettability
In order to understand the effects of different wettabilities on

surfactant/polymer flooding, four test simulations were conducted. Each of the
four cases utilized the same reservoir model and well conditions, but had
differing relative permeability curves, CDCs, and capillary pressure curves to
mimic different wettability conditions. The model is based on physical and
4-2
chemical properties for a dolomite reservoir within the Midland Farms Unit in
West Texas (Table 4.2.1a). The permeability field for this model is shown in
Figures 4.2.6. The permeability field was generated with an average
permeability of 80 md and Dykstra Parsons coefficient of 0.75 and the x and y
correlation length of 20 ft and the vertical correlation length of 4 ft. The relative
permeability curves and CDCs for each case were developed using the data
presented in the previous section. Since the majority of the available data are for
sandstone rocks, all data were used for these test simulations regardless of rock
type. Typical capillary pressure curves for each wettability condition were used.
Table 4.2.2 provides the varying input parameters for each case. For
comparison purposes of each test case, a waterflood simulation was performed
in addition to a surfactant/polymer flood. The waterflood simulations were
stopped when a 98% water cut was being produced. Immediately after the
waterflood period a 0.25 pore volume surfactant slug of 2 vol% surfactant and
1000 ppm polymer was injected. This followed by 1 PV of 1000 ppm polymer
drive and one PV water post flush. Therefore, the only difference in the flood
design between each test case is the length in pore volumes of the initial
waterflood period. The surfactant was injected at an optimum salinity of 0.445
meq/ml as indicated in Table 4.2.1b. The surfactant/polymer flood simulations
were performed as a salinity gradient design where the initial salinity at the end
of the water injection is in Type II(+) as 1.026 meq/ml and the chemical slug at an
optimum salinity of 0.445 meq/ml in Type III and followed by polymer drive at a
salinity of 0.34 meq/ml in Type II(-). The surfactant phase behavior and
properties used in our simulations were based on 1.5% Neodol N67 -7 POS and
0.5% C2024 Alpha Olefin Sulfonate and 4% SBA and 1% Na2CO3 as described
in Task I of this report. Polymer solution properties were based on HPAM
(Alcomer 60 RD) described earlier in Task I.
Case #1: Strongly water-wet
The first test case is representative of a strongly water-wet system. The

relative permeability curves, CDCs, and capillary pressure for case #1 are shown
in Figures 4.2.7, 4.2.8, and 4.2.12, respectively. The first step was to conduct a
waterflood simulation for this test case. The waterflood ran for 0.9 pore volumes
until a 98% water cut was established. The production histories of this simulation
are shown in Figures 4.2.13 through 4.2.16. In addition, oil saturation and
pressure profiles are shown in Figures 4.2.24 through 4.2.31. The waterflood
cumulative oil recovery reached a plateau relatively quickly and terminated at
38% (Figure 4.2.13). The water breakthrough time for this case was relatively
fast. Figure 4.2.17 shows the hypothetical 1-D fractional flow curve for this
wettability condition. The slope of the water fractional flow curve between the
initial water saturation and the water frontal saturation is relatively high, which
explains the fast water breakthrough time. It is also important to note the high
average reservoir pressure at the end of the waterflood due to capillary pressure
and relative permeability effects (Figure 4.2.14). The final average oil saturation
was 39%, which is 2% higher than the waterflood residual oil saturation.
4-3
Next, a surfactant/polymer simulation was performed immediately after the
0.9 PV waterflood period. The production histories of this simulation are shown
in Figures 4.2.20 through 4.2.23. Oil saturation, pressure, microemulsion
saturation, polymer concentration, surfactant concentration, and interfacial
tension profiles are shown in Figures 4.2.32 through 4.2.67. The effectiveness of
surfactant in mobilizing remaining oil saturation is apparent in Figures 4.2.32
through 4.2.34. Oil saturation distributions are also shown for a vertical slice
through the middle layer as given in Figures 4.2.38 though 4.2.40. The profiles
of surfactant concentration and corresponding interfacial tension values are given
in Figures 4.256 through 4.2.61. The interfacial tension is reduced by several
orders of magnitude.
The effects of fractional flow in a water-wet system were supported by the
relatively early oil breakthrough as shown in Figure 4.2.17. The estimated
breakthrough time can be determined by computing the inverse slope of a line
connecting the initial waterflooded oil saturation and the expected chemical flood
oil bank saturation. As shown in Figure 4.2.17, the slope of this line is very high
indicating a very early oil breakthrough time. The final incremental oil recovery
was 43% (Figure 4.2.20) leaving an average oil saturation of about 0.22.
Case #2: Neutral-wet
The second test case is representative of a neutral-wet condition. The

relative permeability curves, CDCs, and capillary pressure curve for case #2 are
shown in Figures 4.2.7, 4.2.9, and 4.2.12, respectively. The first step was to
conduct a waterflood simulation for this test case. The waterflood ran for 3.5
pore volumes until a 98% water cut was established. The production histories of
this simulation are shown in Figures 4.2.13 through 4.2.16. In addition, oil
saturation and pressure profiles are shown in Figures 4.2.68 through 4.2.75. The
waterflood cumulative oil recovery increases slowly and terminates at a high
value of 62% (Figure 4.2.13).
It is also important to note the effect of low residual phase saturations,

relative permeabilities, and fractional flow on the length of time it took to achieve
a 98% water cut. The hypothetical neutral-wet 1-D fractional flow curve shown in
Figure 4.2.18 has a steadily decreasing slope at high water saturations. This is
the primary reason for the required waterflood period to reach 98% water cut.
The final average oil saturation was 0.32, which is 14% higher than the
waterflood residual oil saturation.
Next, a surfactant/polymer simulation with similar conditions as before was

performed immediately after the 3.5 pore volume waterflood period. The
production histories of this simulation are shown in Figures 4.2.20 through
4.2.23. Oil saturation, pressure, microemulsion saturation, polymer
concentration, surfactant concentration, and interfacial tension profiles are shown
in Figures 4.2.76 through 4.2.111. The oil breakthrough for this case occurred
4-4
later than water-wet case but had a slightly higher incremental oil recovery in the
end (Figure 4.2.20). The formation of oil bank due to the chemical injection and it
movement towards the producer is demonstrated in the vertical slices in layer 3
as a function of time in Figures 4.2.82 through 4.2.84. Polymer concentration
profiles in layer 3 (Figures 4.2.97 through 4.2.99) and surfactant concentration
profiles given in Figures 4.2.100 through 4.2.102 indicate the in sync movement
of surfactant and polymer in the chemical slug.
The late breakthrough time can be explained similar to the water-wet

case. The chemical flood fractional flow curve is shown in Figure 4.2.18. The
inverse slope of a line connecting the initial waterflooded oil saturation and the
expected chemical flood oil bank saturation is significantly lower than the water-
wet case. Therefore, a later oil breakthrough time is expected. The final
incremental waterflood oil recovery was 44% (Figure 4.2.20) leaving an average
oil saturation of 0.18.
Case #3: Neutral-wet with a different CDC
The third case is very similar to Case #2. The only difference between the
two test cases is the CDC. This case uses a CDC that allows an easier
mobilization of both water and oil phases than in Case #2. The relative
permeability curves, CDCs, and capillary pressure curve for case #3 are shown
in Figures 4.2.27, 4.2.10, and 4.2.12, respectively. The only difference in the
waterflood results between Case #2 and Case #3 is the reduction in average
reservoir pressure (Figure 4.2.14). This is primarily the effect of capillary
desaturation near the wellbore, which is causing larger negative capillary
pressures than Case #2. This effect did not cause differences in the cumulative
oil recovery and production rates because the desaturation near the wellbore
only slightly reduced the residual oil saturation. The final average oil saturation
was 0.32, which is 14% higher than the waterflood residual oil saturation.
The surfactant/polymer simulation was performed after the 3.5 pore

volume waterflood period. The production histories of this simulation are shown
tension profiles are shown in Figures 4.2.112 through 4.2.147. The oil
breakthrough for this case occurred later than water-wet case for the same
reason as in Case #2 but had a significantly higher incremental waterflood oil
recovery of 58% (Figure 4.2.20) leaving an average oil saturation of 0.13. The
increase in incremental recovery from Case #2 was solely due to difference in
the CDCs. The residual oil saturation was reduced by a greater amount for any
given value of interfacial tension (capillary number).
4-5
Case #4: Oil-wet
The fourth test case is representative of an oil-wet system. The relative

permeability curves, CDCs, and capillary pressure curve for this case are shown
in Figures 4.2.7, 4.2.11, and 4.2.12, respectively. The first step was to conduct a
waterflood simulation for this test case. The waterflood ran for 1.9 PV until a
98% water cut was established. The production histories of this simulation are
shown in Figures 4.2.13 through 4.2.16. In addition, oil saturation and pressure
profiles are shown in Figures 4.2.148 through 4.2.155. The waterflood
cumulative oil recovery plateaus relatively slowly and terminates at 34% (Figure
4.2.13).
The water breakthrough time for this case was faster than all previous
cases. Figure 4.2.19 shows the hypothetical 1-D fractional flow curve for this
wettability condition. The slope of the water fractional flow curve between the
initial water saturation and the water frontal saturation is significantly higher than
the water- and neutral-wet cases, which explains the faster water breakthrough
time. It is also important to note the low average reservoir pressure at the end of
the waterflood due to capillary pressure and relative permeability effects (Figure
4.2.14). The final average oil saturation was 0.43, which is 8% higher than the
waterflood residual oil saturation.
Next, a surfactant/polymer simulation was performed immediately after the

0.9 PV waterflood period. The production histories of this simulation are shown
tension profiles are shown in Figures 4.2.156 through 4.2.192. An oil bank forms
ahead of the chemical slug and is produced from the production well as shown in
Figures 4.2.156 through 4.2.158.
The effects of fractional flow in an oil-wet system were supported by the

later oil breakthrough compared to the water-wet case (Figure 4.2.19). The
inverse slope of a line connecting the initial waterflooded oil saturation and the
expected chemical flood oil bank saturation is lower than the water-wet case but
higher than the neutral-wet case. Therefore, a later oil breakthrough time is
expected compared to water-wet case. The final incremental waterflood oil
recovery was 37% (Figure 4.2.20) leaving an average oil saturation of 0.27.
Discussion
The results of the four test cases provide a good understanding of

recovering oil from reservoirs of different wettability during waterflood and
surfactant/polymer flooding. The residual phase saturations, relative
permeabilities, and fractional flow are the primary causes of differences in
waterflooded oil recovery for each test case. The oil-wet case has the most
unfavorable mobility ratio and has the fastest water breakthrough as expected.
4-6
The mobility ratio becomes more favorable and the breakthrough time increases
as the wettability condition is changed from oil-wet to neutral- and water-wet.
The waterflood oil recovery is the highest for the neutral-wet condition (Cases #2
and #3) due to the low residual oil saturation. However, the neutral-wet case
took the longest amount of time to reach 98% water cut. This is due primarily to
the fractional flow to water. As shown in Figure 4.2.18, the waterflood fractional
flow curve shows a steadily decreasing slope at higher water fractional flow
values. For the water- and oil-wet cases (Figures 4.2.17 and 4.2.19,
respectively), the slope of the fractional flow curve at high fractional flow does not
steadily decline.
For the surfactant/polymer flood, the oil breakthrough time for each test
case was affected by the relative permeabilities and fractional flow similar to the
waterflood simulations. The water-wet case had the fastest oil breakthrough time
and the neutral-wet cases had the slowest breakthrough time. This phenomenon
can be explained by comparing the expected fractional flow curves for each
wetting condition. However, the actual values on the fractional flow curves
presented cannot be used literally because fractional flow theory is valid only for
one dimensional flow. The following equation can be used to approximate the oil
breakthrough time for a surfactant/polymer flood:
SoB − S oI
tdb.t . = (1)
f oB − f oI
The values used to approximate the breakthrough times of each test case
are shown in Figure 4.2.17, 4.2.18, and 4.2.19. The water-wet case was
expected to have an oil breakthrough time of 0.3 pore volumes. Whereas, the
oil-wet and neutral-wet cases were expected to have oil breakthrough at 0.33
and 0.6 pore volumes, respectively. These values supported the actual
breakthrough times of each simulation by comparing their relative values.
The incremental waterflood oil recovery results are most dependent upon
the CDC. Case #3 has the most favorable CDC because it has the lowest critical
capillary number. In addition, Case #3 had the highest incremental waterflood oil
recovery. As expected, the oil-wet case (Case #4), which had the least favorable
CDC, had the lowest incremental waterflood oil recovery.
Based on the results of these test simulations, a series of events is

expected to occur during the wettability alteration process. If a rock is initially oil-
wet, waterflood and chemical flood oil recovery is affected significantly by
unfavorable mobility and CDCs. If the rock is changed to neutral wettability by
means of chemical alteration, several processes will take place. First, the
relative permeability curves will change giving a more favorable mobility ratio and
lower phase saturations. This change in relative permeability will alter the
fractional flow curve. For the fractional flow range of interest (high water
saturation), a less favorable fractional flow will occur in neutral wet conditions
4-7
and the chemical flood oil bank will "slow down". Second, the CDC will change
based on changes in low capillary number residual phase saturations and
wettability dependent trapping parameter. The CDC for a neutral-wet condition is
more favorable for displacing oil at high capillary numbers. If the rock is further
altered to a water-wet state, these processes will continue. The relative
permeability curves will change due to increases in the residual phase
saturations and changes in the endpoint relative permeabilities. The chemical
flood fractional flow and CDC will also become more favorable for oil recovery.
Computing the Relative Permeability Curves as a Function of Wettability

As a result of the extensive literature survey and test simulations, the
effects of wettability on residual oil saturation, relative permeability curves, and
CDCs are clear. However, a mathematical relationship to link the changes of
each property with changes in wettability due to surfactant injection is unclear. A
preliminary method has been established in an attempt to fulfill this complex task.
Prior to running a simulation the following tasks should be completed.
1) Establish relative permeability curves and CDCs for the initial wettability
condition and curves for the expected final condition.
2) Determine the changes in wettability index (Iw) due to surfactant injection. It is
possible that the amount of surfactant adsorbed to the pore wall or the
concentration of surfactant within a pore can be related to wettability
alteration. This relationship has not been completely studied at this time, but
is expected to provide promising results.
3) Develop a relationship between residual phase saturation (Sjr) and Iw. This
relationship was discussed previously.
The next step is to modify UTCHEM to model the wettability alteration
based on changes to the relative permeability, capillary pressure and capillary
desaturation curves. During the simulation, the surfactant adsorption (or
concentration) and capillary number is known for each gridblock during every
timestep. Using these data, UTCHEM can perform the wettability alteration by
following these next steps:
4) Determine Iw based on the relationship developed in Step 2.
5) Determine the low capillary number residual saturation for water and oil (Sjr)
from the relationship developed in Step 3.
6) Develop a new capillary desaturation curve for the altered wettability. The
equation that UTCHEM uses to generate the CDC is as follows:
 Slow high 
 high jr − S jr 
S jr = min  S j ,S jr +  (2)
τ
 1 + Tj Ncjj 
 
Where, S high
jr = Residual saturation of phase j at high capillary number
(Typically 0)
4-8
S low
jr = Residual saturation of phase j at low capillary number (Step 5)
T j = Trapping parameter for phase j
N cj = Capillary number for phase j
τ j = Trapping parameter exponent for phase j
The trapping parameter exponent (τj) is typically 1 and therefore, the only
unknown in this equation is Tj. The trapping parameter (Tj) is important because
it determines the shape of the CDC and the value of the critical capillary number
(Ncc). A CDC obtained using a high trapping parameter (~105 to 106) will have a
lower Ncc compared to a CDC with a low trapping parameter (~102 to 103). The
literature search was extended to attempt to determine a relationship between Tj
and Iw. CDC data were obtained from various sources and curve fit using
Equation 2. A summary of the results of this literature search is provided in
Table 4.2.3. In some instances, the wettability index was provided along with the
CDC for the given rock sample. In other cases, the wettability index was
obtained based on data provided by other sources that had consistent core
preparation techniques (i.e. core firing, cleaning, etc.), initial water saturation,
and brine properties. It is apparent that a clear correlation does not exist.
However, the lack of data for oil-wet and neutral-wet conditions is important.
Additionally, the laboratory experiments that were used to obtain each data set
were conducted differently. Three main differences arose from each experiment:
the use of wettability altering chemicals, differing flow rates to obtain the CDC,
and differing interfacial tensions to obtain the CDC.
Using a correlation between Tj and Iw, the modified trapping parameter

can be obtained and a new CDC can be generated.
7) Determine the current Sjr for each gridblock based on the modified CDC (Step
6) and the capillary number (Ncj).
8) Develop a new set of water/oil relative permeability curves using the current
Sjr from Step 7. Currently UTCHEM uses the following equations to modify
the Corey type relative permeability parameters such as endpoint ( k 0rj ) and
exponent (nj):
Slow
jr − S jr
o olow  high low 
k rj = k rj + *  k orj − k orj  (2)
high 
Slow
jr − S jr

low
S jr − S jr
n j = n low
j +
low
S jr − S jr
(high
* nj
high
) − n low
j (3)
Where, k high
jr = Endpoint relative permeability of phase j at high capillary number
k low
jr = Endpoint relative permeability of phase j at low capillary number
4-9
n high
jr = Relative permeability exponent of phase j at high capillary
number
n low
jr = Relative permeability exponent of phase j at low capillary number
4-10
Table 4.2.1a. Reservoir Properties and Simulated Well Conditions
LxWxH 660x660x30 ft
Grid Blocks 11x11x5
Depth to Pay 4700 ft
Initial Pressure 1975 psi
Reservoir Temperature 103 oF
Average Permeability (kx, ky) 80 md
Kz/kx 0.1
Porosity 0.16
Water Compressibility 3x10-6 psi-1
Oil Compressibility 1x10-5 psi-1
Water Density 62.43 lb/ft3
Oil Density 54.33 lb/ft3
Water Viscosity 0.7 cp
Oil Viscosity 5 cp
Water/Oil IFT 20 dynes/cm
Constant Injection Rate 250 bbl/day
Constant Production Pressure 300 psi
Table 4.2.1ba. Fluid Properties

Initial formation salinity, meq/ml 1.026
Surfactant adsorption, mg/g 0.3
Polymer adsorption, µg/g 10
Surfactant slug size (2 vol%) 0.25
Polymer concentration in slug, 0.1
wt%
Slug salinity, meq/ml 0.445
Polymer drive, PV 1
Polymer concentration in drive, 0.1
wt%
Drive salinity, meq/ml 0.34
Salinity in water post flush, 0.34
meq/ml
Table 4.2.2. Input Parameters for Each Test Case

Cas To Tw koro kwro no nw Mo
Iw Sorlow (1) Swrlow
e
1000 500 1 0.1 0.7
#1 1 0.37 0.37 2 2
0
#2 0 0.18 0.18 2000 2000 0.5 0.5 2 2 7.1
4000 40000 0.5 0.5 7.1
#3 0 0.18 0.18 2 2
00 0
#4 -0.5 0.35 0.35 500 10000 0.4 0.7 2 2 12.5
1: From the Jadhunandan and Morrow curve in Figure 1
4-11
Table 4.2.3. Summary of Literature Search to Obtain Trapping Parameter Data
low low
Source Rock Type Iw Sor Swr To Tw
Bhuyan, 1986 Berea Sandstone 0.57 0.33 0.39 17791 781
Bhuyan, 1986 Berea Sandstone -0.65 0.35 0.38 53452 3206
Bhuyan, 1986 Berea Sandstone -0.7 0.26 0.34 35000 55999
Kamath, 2001 Oolitic Limestone -0.25 0.6 NA 1000000 NA
Kamath, 2001 Limestone -0.19 0.5 NA 200000 NA
Kamath, 2001 Micritic Limestone -0.04 0.32 NA 400000 NA
Abrams, 1975 Gallup Sandstone NA 0.34 NA 1500 NA
Abrams, 1975 Dalton Sandstone NA 0.3 NA 1000 NA
Abrams, 1975 Dalton Sandstone NA 0.29 NA 2000 NA
Abrams, 1975 Paluxy Sandstone NA 0.28 NA 2500 NA
Abrams, 1975 Bandera Sandstone NA 0.37 NA 1500 NA
Abrams, 1975 Berea Sandstone NA 0.37 NA 3000 NA
Abrams, 1975 Indiana Limestone NA 0.34 NA 500 NA
1
Gupta, 1979 Berea Sandstone 0.95 0.25 0.42 8000 2500
1
Chatzis, 1981 Berea Sandstone 0.92 0.34 NA 2000 NA
1
1
1
1
Cottage Grove
Chatzis, 1981 NA 0.37 NA 2000 NA
Sandstone
1
Chatzis, 1981 Boise Sandstone 1 0.27 NA 3000 NA
Fountainbleau
Chatzis, 1981 NA 0.34 NA 3200 NA
Sandstone
1
Stegemeier, 1974 Berea Sandstone 0.84 0.4 NA 3000 NA
1
Amaefule, 1982 Berea Sandstone 0.91 0.2 0.4 1500 200
1
Mohanty, 1983 Berea Sandstone 0.79 0.36 NA 4000 NA
1
Mohanty, 1983 Berea Sandstone -0.75 0.48 NA 1000 NA
1
Mohanty, 1983 Berea Sandstone 0.07 0.26 NA 1000 NA
Fountainbleau
Bardon, 1980 NA 0.4 NA 3000 NA
Sandstone
Boom, 1995 Reservoir Sandstone NA 0.09 0.29 500 79
Boom, 1996 Reservoir Sandstone NA 0.26 NA 315 NA
Delshad, 1990 Berea Sandstone NA 0.4 NA 35000 NA
Henderson, 1998 Berea Sandstone NA 0.29 NA 100000 NA
Henderson, 1998 Berea Sandstone NA 0.27 NA 14854 NA
1
Garnes, 1990 Berea Sandstone 0.77 0.49 NA 5000 NA
1
1
Garnes, 1990 Tarbet Sandstone 0.3 0.27 NA 1600 NA
Garnes, 1990 Oregon Sandstone NA 0.45 NA 15000 NA
Garnes, 1990 Oregon Sandstone NA 0.32 NA 8000 NA
NA - Not Available
1: Iw was obtained from a different source based on the core treatment process, initial
water saturation, and brine.
4-12
0.45
Jadhunandan and Morrow, 1991
0.4 Owolabi and Watson, 1993
Chen et al, 2004
0.35
0.3
0.25
Sor
0.2
0.15
0.1
0.05
0
-1 -0.75 -0.5 -0.25 0 0.25 0.5 0.75 1
Iw
Figure 4.2.1. Effect of Amott-Harvey Wettability Index on Residual Oil Saturation in Berea Sandstone
1
o o
S w r S or krw kro n w no
0.9 Water Wet 0.12 0.25 0.26 1 3 1.3
Neutral Wet 0.12 0.2 0.38 0.94 1.8 1.7
0.8 Oil Wet 0.12 0.28 0.56 0.8 1.4 3.3
Water Wet
Relative Permeability
0.7
(Frac. of Abs. Perm.)
Mixed Wet
0.6 Oil Wet
0.5
0.4
0.3
0.2
0.1
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Water Saturation (Frac. of PV)
Figure 4.2.2. Effect of Wettability on Oil/Water Relative Permeability Curves (Morrow et. al. 1973)
4-13
0.5
Residual Phase Saturation (Frac. of PV)
0.45
Oil
0.4 Water
0.35
0.3
Water Ncc
0.25
0.2
0.15
0.1 Oil Ncc
0.05
0
1.0E-07 1.0E-06 1.0E-05 1.0E-04 1.0E-03 1.0E-02 1.0E-01 1.0E+00
Capillary Number
Figure 4.2.3. Example Water/Oil CDC for a Water Wet Berea Sandstone (Amaefule, 1982)
4-14
Residual Oil Saturation (Frac. of PV) 0.60
Oil Wet
Water Wet
0.50
Mixed Wet
0.40
0.30
0.20
0.10
0.00
1.0E-07 1.0E-06 1.0E-05 1.0E-04 1.0E-03 1.0E-02 1.0E-01 1.0E+00
Capillary Number
Figure 4.2.4. Effect of Wettability on the CDC for a Berea Sandstone (Mohanty, 1983)
0.9 Iw = -0.25
Residual Oil Saturation (Frac. of PV)
Iw = -0.19
0.8 Iw = -0.04
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
1.00E-09 1.00E-08 1.00E-07 1.00E-06 1.00E-05 1.00E-04 1.00E-03 1.00E-02
Capillary Number
Figure 4.2.5. Effect of Wettability on the CDC for Three Weakly Oil-Wet to Neutral-Wet Carbonate Rocks
low
(Kamath, 2001). *Based on the available data, Sor values are not shown.
4-15
Figure 4.2.6. Permeability Field (md) Used in All Test Simulations
0.9
Case #1
Case #4
0.8 Case #2 and #3
0.7
Relative Permeability
(Frac. of Abs. Perm.)
0.6
0.5
0.4
0.3
0.2
0.1
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Figure 4.2.7. Relative Permeability Curves used in the Test Simulations
4-16
0.5
0.45
0.4
Residual Saturation for Phase J
Oil
0.35 Water
0.3
0.25
0.2
0.15
0.1
0.05
0
1.0E-06 1.0E-05 1.0E-04 1.0E-03 1.0E-02 1.0E-01 1.0E+00
Capillary Number
Figure 4.2.8. Water and Oil CDC for Case #1
0.5
0.45
0.4 Oil and Water

0.35
0.3
0.25
0.2
0.15
0.1
0.05
0
1.00E-06 1.00E-05 1.00E-04 1.00E-03 1.00E-02 1.00E-01 1.00E+00
Capillary Number
4-17
0.5
0.45
Oil and Water
0.4
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0
1.0E-09 1.0E-08 1.0E-07 1.0E-06 1.0E-05 1.0E-04 1.0E-03 1.0E-02
Capillary Number
0.5
0.45
0.4 Oil
0.35
Water
0.3
0.25
0.2
0.15
0.1
0.05
0
1.0E-07 1.0E-06 1.0E-05 1.0E-04 1.0E-03 1.0E-02 1.0E-01 1.0E+00
Capillary Number
4-18
15.000
Case #1
10.000 Case #4
Case #2 and #3
Capillary Pressure (psi)
5.000
0.000
-5.000
-10.000
-15.000
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Figure 4.2.12. Capillary Pressure Curves for Each Test Simulation
70
Cumulative Oil Recovery (%OOIP)
60
50
40
30
20
Case #1
Case #2
10 Case #3
Case #4
0
0 0.25 0.5 0.75 1 1.25 1.5 1.75 2 2.25 2.5 2.75 3 3.25 3.5
Pore Volumes of Waterflood
Figure 4.2.13. Waterflood – Cumulative Oil Recovery for Each Test Simulation
4-19
5000
Case #1
4500
Average Reservoir Pressure (psi)
Case #2
4000 Case #3
Case #4
3500
3000
2500
2000
1500
1000
500
0
0 0.25 0.5 0.75 1 1.25 1.5 1.75 2 2.25 2.5 2.75 3 3.25 3.5
Figure 4.2.14. Waterflood – Average Reservoir Pressure for Each Test Simulation
1200
Water Production Rate (STB/D)
1000
800
600
400
Case #1
Case #2
200
Case #3
Case #4
0
0 0.25 0.5 0.75 1 1.25 1.5 1.75 2 2.25 2.5 2.75 3 3.25 3.5
Figure 4.2.15. – Waterflood – Water Production Rate for Each Test Simulation
4-20
1200
Case #1
Oil Production Rate (STB/D) 1000 Case #2
Case #3
Case #4
800
600
400
200
0
0 0.25 0.5 0.75 1 1.25 1.5 1.75 2 2.25 2.5 2.75 3 3.25 3.5
Figure 4.2.16. Waterflood – Oil Production Rate for Each Test Simulation
1
Waterflood foI=0.02
0.9
S/P Flood SoI=0.39
0.8
Fractional Flow of Water
0.7
foB=0.42
0.6
SoB=0.52
0.5
0.4
0.3
0.2
0.1
0
-0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Figure 4.2.17. Hypothetical 1-D Fractional Flow Curves for Case #1
4-21
1
foI=0.01
0.9
SoI=0.32
0.8
0.7
0.6 foB=0.54
0.5 SoB=0.66
0.4
0.3
0.2
Waterflood
0.1 S/P Flood
0
-0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Figure 4.2.18. Hypothetical 1-D Fractional Flow Curves for Cases #2 and #3
1
foI=0.02
0.9
SoI=0.43
0.8
0.7
0.6 foB=0.47
0.5 SoB=0.58
0.4
0.3
0.2
Waterflood
0.1 S/P Flood
0
-0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Figure 4.2.19. Hypothetical 1-D Fractional Flow Curves for Case #4
4-22
70%
Incremental Oil Recovery (%ROIP)

60%
50%
40%
30% Case #1
Case #2
20% Case #3
Case #4
10%
0%
0 0.25 0.5 0.75 1 1.25 1.5 1.75 2 2.25
Pore Volumes of Chemical Injection
Figure 4.2.20. S/P Flood - Incremental Waterflood Oil Recovery for Each Test Simulation
4-23
7000
Case #1
Average Reservoir Pressure (psi)

6000 Case #2
Case #3
5000 Case #4
4000
3000
2000
1000
0
0 0.25 0.5 0.75 1 1.25 1.5 1.75 2 2.25
Figure 4.2.21. S/P Flood - Average Reservoir Pressure for Each Test Simulation
1200
Water Production Rate (STB/D)
1000
800
600
400 Case #1
Case #2
200 Case #3
Case #4
0
0 0.25 0.5 0.75 1 1.25 1.5 1.75 2 2.25
Figure 4.2.22. S/P Flood - Water Production Rate for Each Test Simulation
4-24
350
Case #1
300 Case #2
Oil Production Rate (STB/D) Case #3
250 Case #4
200
150
100
50
0
0 0.25 0.5 0.75 1 1.25 1.5 1.75 2 2.25
Figure 4.2.23. S/P Flood - Oil Production Rate for Each Test Simulation
4-25
Figure 4.2.24. Oil Saturation at 0.2 PV of Waterflood (Case #1) Figure 4.2.26. Pressure at 0.2 PV of Waterflood (Case #1)
4-26
Figure 4.2.28. Oil Sat. in Layer 3 at 0.2 PV of Waterflood (Case #1) Figure 4.2.30. Pressure in Layer 3 at 0.2 PV of Waterflood (Case #1)
Figure 4.2.29. Oil Sat. in Layer 3 at 0.9 PV of Waterflood (Case #1) Figure 4.2.31. Pressure in Layer 3 at 0.9 PV of Waterflood (Case #1)
4-27
Figure 4.2.32. Oil Sat. at 0.25 PV of Chemical Flood (Case #1) Figure 4.2.35. Pressure at 0.25 PV of Chemical Flood (Case #1)
4-28
4-29
Figure 4.2.44. ME Sat. at 0.25 PV of Chem. Flood (Case #1) Figure 4.2.47. Poly Conc. (wt.%) at 0.25 PV of Chemical Flood (Case #1)
4-30
4-31
Figure 4.2.56. Surf. Conc. (vol. frac.) at 0.25 PV of Chem. Flood (Case #1) Figure 4.2.59. IFT (log(dynes/cm)) at 0.25 PV of Chem. Flood (Case #1)
4-32
4-33
Figure 4.2.68. Oil Saturation at 1.0 PV of Waterflood (Cases #2,3) Figure 4.2.70. Pressure at 1.0 PV of Waterflood (Cases #2,3)
Figure 4.2.69. Oil Saturation at 3.5 PV of Waterflood (Cases #2,3) Figure 4.2.71. Pressure at 3.5 PV of Waterflood (Cases #2,3)
4-34
Figure 4.2.72. Oil Sat. in Layer 3 at 1.0 PV of Waterflood (Cases #2,3) Figure 4.2.74. Pressure in Layer 3 at 1.0 PV of Waterflood (Cases #2,3)
Figure 4.2.73. Oil Sat. in Layer 3 at 3.5 PV of Waterflood (Cases #2,3) Figure 4.2.75. Pressure in Layer 3 at 3.5 PV of Waterflood (Cases #2,3)
4-35
4-36
4-37
4-38
4-39
Figure 4.2.100. Surf. Conc. (vol. frac.) at 0.25 PV of Chem Flood (Case #2) Figure 4.2.103. IFT (log(dynes/cm)) at 0.25 PV of Chem Flood (Case #2)
Figure 4.2.102. Surf. Conc. (vol. frac.) at 2.25 PV of Chem Flood (Case #2) Figure 4.2.105. IFT (log(dynes/cm)) at 2.25 PV of Cheml Flood (Case #2)
4-40
4-41
4-42
4-43
4-44
Figure 4.2.130. ME Sat. at 0.25 PV of Chem. Flood (Case #3) Figure 4.2.133. Poly Conc. (wt.%) at 0.25 PV of Chem Flood (Case #3)
4-45
4-46
4-47
4-48
Figure 4.2.152. Oil Sat in Layer 3 at 0.3 PV of Waterflood (Case #4) Figure 4.2.154. Pressure in Layer 3 at 0.3 PV of Waterflood (Case #4)
Figure 4.2.153. Oil Sat in Layer 3 at 1.9 PV of Waterflood (Case #4) Figure 4.2.155. Pressure in Layer 3 at 1.9 PV of Waterflood (Case #4)
4-49
4-50
4-51
4-52
4-53
Figure 4.2.183. Surf. Conc. (vol. frac.) at 2.25 PV of Chem. Flood (Case #4) Figure 4.2.186. IFT (log(dynes/cm)) at 2.25 PV of Chem Flood (Case #4)
4-54
Figure 4.2.189. Surf. Conc. (vol. frac.) at 2.25 PV of Chem. Flood (Case #4) Figure 4.2.192. IFT (log(dynes/cm)) at 2.25 PV of Chem Flood (Case #4)
4-55
References
Abrams, A. 1975 "The Influence of Fluid Viscosity, Interfacial Tension, and Flow
Velocity on Residual Oil Saturation Left by Waterflood". SPE 5050.
Amaefule, Jude O. and Handy, Lyman L. 1982. "The Effect of Interfacial
Tensions on Relative Oil/Water Permeabilities of Consolidated Porous
Media". SPE 9783.
Amott, E. 1959. "Observations Relating to the Wettability of Porous Rock". Trans.
AIME 216.
Anderson, William G. 1986. "Wettability Literature Survey - Part 1: Rock/Oil/Brine
Interactions and the Effects of Core Handling on Wettability". SPE 13932.
Bardon, Charles and Longeron, G.G. 1980. "Influence of Very Low Interfacial
Tensions on Relative Permeability". SPE 7609.
Bhujan, Debojit 1986. "Effect of Wettability on Capillary Desaturation Curves".
Masters Thesis.
Boom, W. et. al. 1995. "Experimental Evidence for Improved Condensate Mobility
at Near-Wellbore Flow Conditions". SPE 30766.
Boom, W. et. al. 1996. "On the Use of Model Experiments for Assessing
Improved Gas-Condensate Mobility Under Near-Wellbore Flow Conditions".
SPE 36714.
Brownell, L. E and D. L. Katz. 1949. "Flow of Fluids Through Porous Media, Part
II," Chemical Engineering Process, 43, 601-612.
Chatzis, Ioannis and Morrow, Norman R. 1984. "Correlation of Capillary Number
Relationships for Sandstone". SPE 10114.
Chen, Jiansheng et. al.2004. "Study of Wettability Alteration From NMR: Effect of
OBM on Wettability and NMR Responses". 8th International Symposium on
Reservoir Wettability.
Delshad, M. 1990. "Trapping of Micellar Fluids in Berea Sandstone," PhD
dissertation, The U. of Texas, Austin, Texas.
Gupta, Surendra P. and Trushenski, Scott P. 1979. "Micellar Flooding -
Compositional Effects on Oil Displacement". SPE 7063.
Henderson, G.D. et al. 1998 ‘‘Measurement and Correlation of Gas Condensate
Relative Permeability by the Steady-State Method,’’ SPERE April 1998.
Jadhunandan, P.P. and Morrow, N.R. 1991. "Effect of Wettability on Waterflood
Recovery for Crude-Oil/Brine/Rock Systems". SPE 22597.
Jerauld, G.R. and Rathmell, J.J. 1997. "Wettability and Relative Permeability of
Prudhoe Bay: A Case Study in Mixed-Wet Reservoirs". SPE 28576.
Kamath, Jairam and Meyer, Robert F. 2001. "Understanding Waterflood Residual
Oil Saturation of Four Carbonate Rock Types". SPE 71505.
4-56
Mohanty, Kishore K. and Salter, Stephen J. 1983. "Multiphase Flow in Porous
Media: III. Oil Mobilization, Transverse Dispersion, and Wettability". SPE
12127.
Morrow, N.R. et. al. 1973. "Displacement Studies in Dolomite with Wettability
Control by Octanoic Acid". SPE 3993.
Owolabi, O.O. and Watson, R.W. 1993. "Effects of Rock-Pore Characteristics on
Oil Recovery at Breakthrough and Ultimate Oil Recovery in Water-Wet
Sandstones". SPE 26935.
Stegemeier, George L. 1974. "Relationship of Trapped Oil Saturation to
Petrophysical Properties of Porous Media" American Institute of Mining,
Metallurgical, and Petroleum Engineers, Inc. SPE 4754.
4-57

Report 1 PDF

Uploaded by

Copyright:

Available Formats

Report 1 PDF

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

Report 1 PDF

Uploaded by

Copyright:

Available Formats

Surfactant Based Enhanced Oil Recovery and Foam Mobility Control

Semi Annual Technical Report

Reporting Period Start Date: July 2004

Reporting Period End Date: June 2005

George J. Hirasaki, Rice University

Date Report was Issued: July 2005

Rice University Department of Chemical Engineering, MS-362 6100 Main Street

The University of Texas Department of Petroleum Engineering P.O. Box 7726

INTERA, Inc. 9111A Research Blvd. Austin, TX 78758

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

An alkaline surfactant process is designed to enhance spontaneous imbibition in

Subtask 2.5 Surfactant-enhanced spontaneous imbibition experiments ......... 2-24

Task 3 Foam for Mobility Control ................................................................................. 3-1

Task 4: Simulation of Field-Scale Processes .............................................................. 4-1

Table 1.1-1. Surfactants Tested in Year 2.................................................................. 1-20

Table 2.2-1 Operating Parameters for the example ................................................... 2-17

Table 2.5-1 Surfactant and co-surfactant identification ............................................. 2-24

Figure 1.1-1. Neodol® N67- 7PO-sulfate (C16-17 -(PO)7-SO4) ................................... 1-22

Figure 1.2-11. IPA Tracer Breakthrough Curve......................................................... 1-40

Figure 4.2.1. Effect of Amott-Harvey Wettability Index on Residual Oil Saturation in

Oil recovery by primary depletion and waterflooding recovers only

The research in Year 1 identified several families of propoxylated sulfate

Table 1.1-1 Surfactants Tested in Year 2

Surfactant screening in Year 2 followed the same process of phase behavior

1.1.2 Results and Discussion for Subtask 1.1

Additionally, a C15-18 internal-olefin sulfonate (C15-18 IOS) surfactant made by

Surfactants or surfactant blends also have been provided by Oil Chem

Progress for Year 2 can be summarized as focused efforts to find surfactants

1.2.1 Results of Phase Behavior Experiments

Surfactant formulations were screened at the actual 38oC reservoir temperature.

1.5 % N67-(PO)7-SO4, 0.5% C15-18 IOS,

1.5 % N67-(PO)7-SO4, 0.5% C20-24 AOS,

1.5% C15-18 IOS, 0.5% C20-24 AOS, 2% SBA 6-9 4 1

0.75% C15-18 IOS, 0.25% C20-24 AOS 9 - 11 4 4

1.5% SAS-60, 0.5% C20-24 AOS 5-7 3 3, until gel

2% C20-24 AOS, 4% SBA 2.1 8.5 1

1% C20-24 AOS, 1% C15-18 IOS 6-8 3 5

Did not Did not Did not

0.5% C20-24 AOS, 1% SBA 2.1 9 1

0.5% C20-24 AOS, 1% SBA, 1% Na2CO3 1.2 14 1

Did not Did not Did not

2% C20-24 AOS, 4% SBA, 1% Na2CO3 1.4 12 1

MF3 Crude Oil

Figure 1.2-1. Salinity Scan with MF3 Crude Oil at 38°C

20 Water Sol Ratio After 21

10 2% C20-24 AOS, 4% SBA

Figure 1.2-4. Effect of Sodium Carbonate on Optimal Salinity with ELK

Elk Hills Crude

Figure 1.2-5. Effect of Sodium Carbonate on Equilibration Time with Elk

Effects of sec-Butyl Alcohol (SBA). Phase behavior experiments done with

1.2.2 Results of Polymer Testing

Mobility control is needed for an efficient surfactant flood to prevent fingering,

Preliminary experiments were conducted to verify performance of HPAM for use

Figure 1.2-6. Viscosity of HPAM Polymer (Flopaam 3330S) in Brine

Figure 1.2-7. Viscosity for HPAM Polymer (Flopaam 3330S) in Brine

Figure 1.2-8. Viscosity for HPAM Polymer (Flopaam 3330S) in Brine

Table 1.2-5. Summary of Outcrop Dolomite Rock Permeability

High Resolution X-ray Computed Tomography (HRXCT) scans. CT scans were

0.001 IPA Concentration History for MF4 (Dead volume ~ 7 mls)