Updated EOR Screening Criteria and Modeling The Impacts of Water - Al Adasani 2012

Scholars' Mine
Doctoral Dissertations Student Research & Creative Works
Summer 2012
Updated EOR screening criteria and modeling the

impacts of water salinity changes on oil recovery
Ahmad Aladasani
Follow this and additional works at: http://scholarsmine.mst.edu/doctoral_dissertations

Department:
Recommended Citation
Aladasani, Ahmad, "Updated EOR screening criteria and modeling the impacts of water salinity changes on oil recovery" (2012).
Doctoral Dissertations. Paper 1963.
This Dissertation - Open Access is brought to you for free and open access by the Student Research & Creative Works at Scholars' Mine. It has been
accepted for inclusion in Doctoral Dissertations by an authorized administrator of Scholars' Mine. For more information, please contact
weaverjr@mst.edu.
i
i
UPDATED EOR SCREENING CRITERIA AND
MODELING THE IMPACTS OF WATER SALINITY CHANGES
ON OIL RECOVERY
By
AHMAD ALADASANI
A DISSERTATION
Presented to the Faculty of the Graduate School of the
MISSOURI UNIVERSITY OF SCIENCE AND TECHNOLOGY
In Partial Fulfillment of the Requirements for the Degree
DOCTOR OF PHILOSOPHY
in
PETROLEUM ENGINEERING
2012
Approved by
Dr. Baojun Bai, Advisor

Dr. Yu-Shu Wu
Dr. Ralph Flori
Dr. Shari Dunn-Norman
Dr. Runar Nygaard
ii
iii
ABSTRACT
Developing technologies for enhanced oil recovery (EOR) from existing oil fields
would supply the world’s energy needs for several decades. The application of EOR in
many major oil-producing countries remains in its conceptual stage. Thermal and gas
EOR methods achieve high incremental rates; however, their application range has not
broadened significantly as they have matured, while the rate at which new, promising
EOR methods such as Low-salinity water flooding (LSWF) are being implemented is
alarming. Despite, the potential of LSWF its development and application has been
hindered by the lack of consensus concerning its recovery mechanism(s). Every oil
reservoir has a unique ionic environment that changes naturally and by human
intervention, which makes it difficult to identify recovery mechanism(s) in EOR methods
such as LSWF. This study updates the EOR selection criteria and presents new EOR
screening tools based on dataset distribution, incremental recovery and deterministic
modeling. LSWF recovery mechanisms are investigated by statistical analysis and
numerical solutions. Furthermore, an up-scaled multi-dimensional model is developed
for LSWF under various reservoir wetting conditions. Finally, a risk analysis case study
is included. The results in this study include an incremental recovery prediction model for
miscible CO2 flooding. The use of statistical analysis and reservoir simulation identifies
different LSWF recovery mechanism(s) based on the initial and final wetting state in
conjunction with injection brine chemistry. Three dimensional models of LSWF outline
the importance of sweep efficiency and the potential incremental recovery in oil-wet
reservoirs. On a separate note, hydrogen sulfide (H2S) risk in anthropogenic CO2
transportation is highlighted.
iv
ACKNOWLEDGMENTS
At the outset, I would like to thank Allah for all his blessings.
Secondly, I would also like to express my gratitude to my committee members for
providing me with the opportunity to present this study.
Dr. Baojun Bai made it possible for me to work on a significant area of reservoir
engineering, focusing on strategic issues and using creative methods. In addition, Dr. Bai
encouraged me to network with the scientific community by participating in conferences
and collegial competitions, publishing my work and introducing me to experts in the field
of reservoir engineering. One of those experts is my external committee member Dr.
Yu-Shu Wu, who was instrumental in the success of my research. I would like to thank
Dr. Wu for accepting me as his student and spending countless weekends helping me
build my reservoir models. Dr. Runar Nygaard, Dr. Shari Dunn-Norman and Dr. Ralph
Flori played a vital role in teaching me the major petroleum engineering specialties.
Their articulate explanation of the most complex engineering problems was a source of
relief and encouragement. I am therefore, thankful for their leadership and guidance. I
would like to acknowledge Dr. Nygaard further for being my mentor and an impetuous
to my success and achievements.
The entire administrative staff in the Petroleum Engineering program spared no
effort in supporting my research and greeting me with the warmth of their smiles,
particularly, Mrs. Paula Cochran whom I am forever grateful. In addition, I would like
recognize Mr. Mohammed Al Dushaishi for supporting my work in preparing the thesis
for the graduate office and for the time we worked together on statistical analysis.
Dr. Saeed Salehi, was my friend and council during my years in the program. I
thank him for being the first person to call me when I completed my defense. I am
indebted to Dr. Salehi for his companionship and for putting a smile on face when
situations where dire. On a personnel note, I would like to acknowledge individuals
whom are dear to me. Mr. Qassem Dashti, Mr. Fouad Al-Shatti and Mr. Hesham Al-
Reshaid. The aforementioned gentlemen are a source of inspiration to me in the purse of
my Ph.D. In closing, my utmost appreciation goes to the scientists whose work made my
research possible.
v
TABLE OF CONTENTS
Page
ABSTRACT ....................................................................................................................... iii
ACKNOWLEDGMENTS ................................................................................................. iv
LIST OF ILLUSTRATIONS ............................................................................................. ix
LIST OF TABLES ............................................................................................................ xii
NOMENCLATURE ........................................................................................................ xiv
SECTION
1. INTRODUCTION ....................................................................................................... 1
2. LITERATURE REVIEW ............................................................................................ 3
2.1. ENERGY SUPPLY .................................................................................................. 3
2.1.1. World Energy Statistics. .................................................................................... 3
2.1.2. Future Energy Outlook. ..................................................................................... 3
2.1.3. Oil Reserves. ...................................................................................................... 6
2.1.4. Oil Supply. ......................................................................................................... 6
2.2. ENHANCED OIL RECOVERY .............................................................................. 7
2.2.1. Oil Production.................................................................................................... 8
2.2.2. Reservoir Flow Mechanics. ............................................................................... 8
2.2.3. Oil Production.................................................................................................. 10
2.2.4. Enhanced Oil Recovery Methods. ................................................................... 10
2.2.5. Enhanced Oil Recovery Selection Criteria. ..................................................... 12
2.3. WATER FLOODING ............................................................................................ 13
2.3.1. Low-Salinity Water Flooding. ......................................................................... 19
2.3.2. Low-Salinity Water-Flooding Experiments and Tests. ................................... 20
2.3.3. Reservoir Simulation. ...................................................................................... 23

vi
2.3.4. Recovery Mechanisms. .................................................................................... 23
3. ENHANCED OIL RECOVERY GUIDELINES AND SELECTION CRITERIA .. 28
3.1. UPDATING ENHANCED OIL RECOVERY SELECTION CRITERIA ............ 28
3.1.1. Enhanced Oil Recovery Database. .................................................................. 29
3.1.1.1. Database analysis. ..................................................................................... 29
3.1.1.2. Enhanced oil recovery selection criteria. .................................................. 30
3.1.1.3. New enhanced oil recovery selection criteria. .......................................... 31
3.2. UPDATED ENHANCED OIL RECOVERY SELECTION CRITERIA .............. 31
3.2.1. Databases Analysis. ......................................................................................... 32
3.2.2. Enhanced Oil Recovery Selection Criteria. ..................................................... 41
3.2.3. New Enhanced Oil Recovery Selection Criteria ............................................. 43
3.2.3.1. Dataset distributions.................................................................................. 43
3.2.3.2. Incremental recovery. ............................................................................... 51
3.3. UPDATED ENHANCED OIL RECOVERY GUIDELINES ............................... 53
3.3.1. Low-Salinity Water Flooding. ......................................................................... 59
3.3.2. Water-Alternating Gas. .................................................................................... 61
3.3.3. Surfactant Imbibition. ...................................................................................... 62
3.3.4. In-Depth Conformance Control. ...................................................................... 63
3.3.5. Microbial Enhanced Oil Recovery. ................................................................. 64
3.3.6. Steam-Assisted Gravity Drainage. .................................................................. 65
3.4. MISCIBLE CARBON DIOXIDE FLOODING .................................................... 66
3.5. NEW SELECTION CRITERIA FOR CO2 EOR .................................................. 67
4. MODIFIED WATER SALINITY ............................................................................. 75
4.1. LOW-SALINITY WATER-FLOODING MODEL ............................................... 75
4.1.1. Governing Equations. ...................................................................................... 76

vii
4.1.2. Numerical Formulation.................................................................................... 81
4.1.3. Handling Initial and Boundary Conditions. ..................................................... 85
4.1.4. Treatment of Fracture-Matrix Interaction........................................................ 85
4.1.5. Handling Immobile Water Zones. ................................................................... 86
4.1.6. Relative Permeability Functions. ..................................................................... 86
4.1.7. Capillary Permeability Functions. ................................................................... 88
4.1.8. Mesh Generation.............................................................................................. 94
4.1.9. Simulator Input File. ........................................................................................ 95
4.2. LOW-SALINITY WATER-FLOODING MODEL VALIDATION ..................... 96
4.2.1. Analytical Solution .......................................................................................... 96
4.2.1.1. One-dimensional analytical solution for solute transportation. ................ 96
4.2.1.2. Two-dimensional analytical solution for solute transportation. ............... 98
4.2.2. One-Dimensional Mass Component Transportation. .................................... 100
4.2.2.1. Analytical solution validation. ................................................................ 100
4.2.2.2. Comparison of analytical and numerical solutions. ................................ 102
4.2.3. Two-Dimensional Mass Component Transportation .................................... 104
4.3. EXPERIMENTAL VALIDATION ..................................................................... 106
4.3.1. Sandstone Reservoirs. ................................................................................ 107
4.3.2. Carbonate Reservoirs. ................................................................................ 120
5. SENSITIVITY ANALYSIS .................................................................................... 129
5.1. STATISTICAL ANALYSIS OF CORE-FLOODING EXPERIMENTS............ 129
5.1.1 Sandstone Reservoirs. ..................................................................................... 129
5.1.2 Carbonate Reservoirs. ..................................................................................... 130
6. CASE STUDIES...................................................................................................... 132
6.1. THREE-DIMENSIONAL IMMISCIBLE DISPLACEMENT ............................ 132

viii
6.1.1. Intermediate Wet Reservoirs. ........................................................................ 134
6.1.2 Water-Wet Reservoirs. ................................................................................... 135
6.1.3. Oil-Wet Reservoirs. ....................................................................................... 136
7. CONCLUSIONS......................................................................................................... 138
APPENDICES
A. ANALYTICAL PROGRAM FOR 1-DIMENSIONAL TRANSPORTATION
OF A CHEMICAL COMPONENT ........................................................................ 141
B. DISPERSION MODELING OF MISCIBLE CO2 RELEA.................................... 144
C. ANALYTICAL PROGRAM FOR 2-DIMENSIONAL TRANSPORTATION

OF A CHEMICAL COMPONENT........................................................................ 155
D. SIMULATION PROBLEM INPUT AND OUTPUT FILES.................................. 159
E. SUMMARY OF SANDSTONE COREFLOODING EXPERIMENTS ................. 206
F. SANDSTONE RESERVOIR CORE-FLOODING EXPERIMENT

CORRELATIONS ................................................................................................... 209
G. RESIDUAL OIL SATURATION PREDICTION MODEL
(EXCLUDING WETTABILITY AND CLAY CONTENT) .................................. 211
H. PREDICTION MODEL FOR MISCIBLE CARBON DIOXIDE FLOODING ..... 213
BIBLIOGRAPHY ........................................................................................................... 215
VITA ............................................................................................................................... 239

ix
LIST OF ILLUSTRATIONS
Figure Page
2.1. World Fuel Share (1973), Source: IEA, 2005.............................................................. 4
2.2. World Fuel Share (2003), Source: IEA, 2005.............................................................. 4
2.3. Forecast of World Energy Supply (2035), Source: IEA, 2010 .................................... 5
2.4. Forecast of World Energy Demand (2035), Source: IEA, 2010 .................................. 5
2.5. Enhanced Oil Recovery Potential, Source: IEA, 2010 ................................................ 7
2.6. The Role of Oil Discovery and Development, Source: IEA, 2010 ............................ 11
2.7. Porosity Illustrations .................................................................................................. 14
2.8. Reservoir Flow ........................................................................................................... 14
2.9. Reservoir Flow Behavior ........................................................................................... 15
2.10. Contact Angle and Wettability................................................................................. 16
2.11. Oil Expansions Due to Gas Entrapment and Displacement
of Oil from Larger Pores .......................................................................................... 17
2.12. Impact of Gas Trapping on Residual Oil Saturation ................................................ 17
2.13. Relative Permeability Curves .................................................................................. 18
2.14. Displacement Front Stability ................................................................................... 19
2.15. Summary of LSWF Experiments ............................................................................. 21
2.16. Clay Migration (Tang and Morrow, 1999) .............................................................. 24
2.17. Gouy Layer’s Expansion and Contraction ............................................................... 25
3.1. Worldwide EOR Project Categories (1959-2010) ..................................................... 32
3.2. Worldwide EOR Project Subcategories (1959-2010) ................................................ 33
3.3. EOR Methods and Country Distributions .................................................................. 34
3.4 EOR Projects and Enhanced Production Trends ........................................................ 35
3.5. Worldwide Enhanced Production Share .................................................................... 36
3.6. Oil Saturations and Enhanced Production Distributions ........................................... 37
3.7. EOR Methods and Formation Type Distributions ..................................................... 38
3.8. EOR Methods Versus Selected Average Fluid and Reservoir Properties ................. 39
3.9. EOR Methods – Selected Average Fluid and Reservoir Properties .......................... 40
3.10. API Gravity Distribution in Miscible EOR Projects................................................ 43
x
3.11. API Gravity Distribution Versus Selected EOR Methods ....................................... 44

3.12. Viscosity Distribution Versus Selected EOR Methods ........................................... 45
3.13. Saturation Distribution Versus Selected EOR Methods .......................................... 46
3.14. Permeability Distribution Versus Selected EOR Methods ...................................... 47
3.15. Porosity Distribution Versus Selected EOR Methods ............................................. 48
3.16. Depth Distribution Versus Selected EOR Methods ................................................. 49
3.17. Temperature Distribution Versus Selected EOR Methods ...................................... 50
3.18. Enhanced Oil Recovery Methods ............................................................................ 54
3.19. Miscible CO2 Projects and Enhanced Production Trends ....................................... 68
3.20. Distribution of Miscible CO2 Reservoir Properties................................................. 68
3.21. Initial Reservoir Property Correlations .................................................................... 70
3.22. Selected Reservoir Property Correlations ................................................................ 71
3.23. High and Low Correlations ...................................................................................... 72
3.24. Multiple Variable Curve and Prediction Model....................................................... 74
4.1. Low-Salinity Model Development ............................................................................ 75
4.2. Low Salinity Model Validation ................................................................................. 76
4.3. Grid Block X-Y Coordinates ..................................................................................... 82
4.4. Flow Paradigms ......................................................................................................... 85
4.5. High-Salinity Water-Flooding Permeability Curves ................................................. 90
4.6. High- to Low- Salinity Water-Flooding Permeability Curves ................................... 90
4.7. Low-Salinity Water-Flooding Permeability Curves .................................................. 91
4.8. Relative Permeability Transition during LSWF ........................................................ 92
4.9. Normalized Capillary Pressure Curves (Start of LSWF) ........................................... 92
4.10. Normalized Pc, Xc and θ Curves ............................................................................. 93
4.11. Radial and Rectangle Mesh Geometries .................................................................. 94
4.12. Horizontally-Layered Equal Radii and Vertically-Layered Unequal Radii Mesh... 94
4.13. Two-Dimensional Flow Assumptions ..................................................................... 99
4.14. Schematic for Numerical and Analytical Problems 2 and 3 .................................. 101
4.15. Analytical Versus Numerical Solution to Problem 2 ............................................. 103
4.16. Analytical Versus Numerical Solution to Problem 2 ............................................. 104
4.17. Problem 4 Schematic ............................................................................................. 105
xi
4.18. Schematic for Numerical Problem 6 ...................................................................... 108

4.19. Case 1 Simulation Recovery Curve ....................................................................... 111
4.27. Carbonate Reservoir Simulation Recovery Curve ................................................. 123
4.29. Simulation Recovery Curve ................................................................................... 125
4.30. Correlation of Residual Oil Saturation and Salt Concentration ............................. 126
4.31. Correlation of Contact Angle and Salt Concentration ........................................... 127
4.32. Correlation of IFT and Salt Concentration ............................................................ 127
5.1. Prediction Profiler Acid Number and IFT ............................................................... 131
5.2. Prediction Profiler (2nd and 3rd Stages) Injected Brine Anions ............................. 131
6.1. Schematic for Numerical Problem 9 ........................................................................ 134
6.2. Three-Dimensional, Five-Spot Well Pattern (Mixed Wettability Reservoir) .......... 135
6.3. Three-Dimensional, Five-Spot Well Pattern (Water-Wet Reservoir)...................... 136
6.4. Three-Dimensional, Five-Spot Well Pattern (Oil-Wet Reservoir) .......................... 137
xii
LIST OF TABLES
Table Page
2.1. North Slope Low-Salinity SWCTT Results (McGuire and colleagues 2005) ........... 22
3.1. EOR Selection Criteria .............................................................................................. 41
3.2. New EOR Criteria – Based on Project Distributions of Reservoir Properties ........... 51
3.3. New EOR Criteria Based on Incremental Recovery.................................................. 52
3.4. Gas-Based EOR Methods .......................................................................................... 55
3.5. Chemical-Based EOR Methods ................................................................................. 56
3.6. Thermal-Based EOR Methods ................................................................................... 58
3.7. Low-Salinity Water Flooding (LSWF) ...................................................................... 60
3.8. Water-Alternating Gas Flooding................................................................................ 61
3.9. Surfactant Imbibition ................................................................................................. 62
3.10. In-Depth Conformance Control ............................................................................... 64
3.11. Microbial Enhanced Oil Recovery (MEOR) ........................................................... 64
3.12. Steam-Assisted Gravity Drainage (SAGD) ............................................................. 66
3.13. Summary Distribution of Miscible CO2 Reservoir Properties ................................ 69
4.1. Rock Properties Input into the Simulator................................................................... 95
4.2. One-Dimensional Analytical Program Validation ................................................... 101
4.3. Two-Dimensional Analytical Program Validation .................................................. 105
4.4. Sandstone Core Plug Properties (Taken from Ashraf and colleagues 2010) .......... 109
4.5. Sandstone Core-Flooding Fluid Properties
(Taken from Ashraf and colleagues 2010)............................................................... 110
4.6. Sandstone Core (A) Case 1 Simulation Versus Core-Flooding Results .................. 111
4.10. Sandstone Core (B) Case 5 Simulation Versus Core-Flooding Results ................ 116
4.11. Sandstone Core (C) Case 6 Simulation Versus Core-Flooding Results ................ 118
4.12. Sandstone Core (D) Case 7 Simulation Versus Core-Flooding Results ................ 119
4.13. Carbonate Core Plug Properties (Taken from Yousef and colleagues 2010) ....... 121
xiii
4.14. Carbonate Core-Flooding Fluid Properties at 212 °F

(Taken from Yousef and colleagues 2010) ........................................................... 122
4.15. Problem 7 Carbonate Reservoir Simulation Versus Core-Flooding Results ......... 123
4.16. Core Plug Properties (Taken from Yousef and colleagues 2010)......................... 124
4.17. Problem 8 Carbonate Reservoir Simulation Versus Core-Flooding Results
(Taken from Yousef and colleagues 2010) ........................................................... 125
6.1. Reservoir Properties for Five-Spot Water-Flooding Pattern ................................... 133
xiv
NOMENCLATURE
Symbol Description
S Saturation
EOR Enhanced Oil Recovery
LSWF Low Salinity Water Flooding
LSSWF Low Salinity Seawater Flooding
IEA International Energy Association
SPE Society of Petroleum Engineers
ISO International Standards Association
DOE Department of Energy
OOIP Original Oil in Place
OandGJ Oil and Gas Journal
IFT Interfacial Tension
σ Interfacial Tension
m Micrometer
SAG Steam Assisted Gravity
SAGD Steam Assisted Gravity Drainage
Sg Gas Saturation
So Oil Saturation
Sw Water Saturation
Sgr Residual Gas Saturation
Sor Residual Oil Saturation
Swc Critical Water Saturation
Sorg Residual Gas Oil Saturation
Sgc Critical Gas Saturation
Soi Initial Oil Saturation
ED Displacement Efficiency
ES Sweep Efficiency
WAG Water Alternating Gas
k Permeability
xv
krβ Phase β Relative Permeability

krg Gas Relative Permeability
kro Oil Relative Permeability
krw Water Relative Permeability
Oil Relative Permeability at Critical Water Saturation
Oil Relative Permeability in 2-Phase Oil-Water System
Oil Relative Permeability in 2-Phase Oil-Gas System
Pg Gas Capillary Pressure
Po Oil Capillary Pressure
Pw Water Capillary Pressure
Pβ Phase Capillary Pressure
Pcgo Oil Gas Capillary Pressure
Pcow Water Oil Capillary Pressure
Pg Bubble Point Pressure
Initial Bubble Point Pressure
θ Theta (Contact Angle)
Bg Gas Formation Factor
Bo Oil Formation Factor
Bw Water Formation Factor
Bβ Phase β Formation Factor
Water Formation Factor at
Cw Water Phase Compressibility
Rs Solution Gas-Oil Ratio
STC Standard Tank Condition
E Oil Recovery
 Viscosity
β Phase B Viscosity
o Oil Viscosity
w Water Viscosity
M Mobility Ratio
λ Mobility Ratio
xvi
γ Transmissivity
ψ Potential
D Distance Between Elements
t Time
i Element Number
j Neighbor Elements
ij + ½ Averaging the Interface Between Two Elements i and j
Vi Volume of Element i
KCL Potassium Chloride
NaCl Sodium Chloride
ppm Parts Per Million
WF Water Flood
% Percentage
PV Pore Volume
SWCTT Single Well Chemical Tracer Tests
IOR Improved Oil Recovery
pH Universal Indicator
CaCO3 Calcium Carbonate
MIE Multiple Component Ion Exchange
Ca2+ Calcium Ion
2+
Mg Magnesium Ion
H+ Hydrogen Ion
DLVO Derjaguin, Landau, Verwey and Overbeek
Mg(OH)2 Magnesium Oxide
API American Petroleum Institute
CO2 Carbon Dioxide
 Gas
w Water
o Oil
ϕ Porosity
Xc Mass Fraction of Salt in the Water Phase
xvii
Xw Mass Fraction of Water Component in the Water Phase

ρg Gas Density
ρo Oil Density
ρw Water Density
ρdg Dissolved Gas Density
ρβ Phase Density
ρR Rock Grain Density
 Flux
Darcy Velocity for the Gas Phase
Darcy Velocity for the Oil Phase
Darcy Velocity for the Water Phase
qg Gas Flow Rate
qo Oil Flow Rate
qw Water Flow Rate
qx Salt Flow Rate
Kd Salt Distribution Coefficient Between Water Phase and Reservoir Rock
Dm Molecular Diffusion Coefficient
τ Formation Tortuosity
P Pressure
g Gravity Constant
d Surface Depth
V Volume of Elements
N Total Number of Elements
n Previous Time Level
n+1 Current Time Level
a Element 1
b Element 2
λ Phase Mobility Ratio
Δ Change
t Time
Da Distance from Center of Element a to the Interface Between Element a and
xviii
Db and b Distance from Center of Element b to the Interface Between Element a

da Depth of the Center of Element a
Qwa Sink or Source Term of Element a
Ф Potential
1
1. INTRODUCTION
Developing technologies for enhanced oil recovery (EOR) from existing oil fields
would supply the world’s energy needs for several decades. This alternative represents a
valuable option considering the current and future outlook of world energy supplies and
reserves. The most significant problems involve the stability of the oil supply, the
maturity of alternative sources of energy, the accuracy of oil reserve volumes, the
maximum oil production forecasts and increasing energy demands, especially in
developing nations.
The application of EOR in many major oil-producing countries remains in its
conceptual stage despite the implementation of hundreds of projects since 1959. The
most widely-cited EOR selection criteria on the Society of Petroleum Engineers (SPE)
website is more than 15 years old and has not been updated to incorporate new
technologies or field data. EOR projects having occurred since the concept’s inception
more than 60 years ago must be analyzed to encourage further EOR development and
implementation. Over 16 EOR methods are currently being applied worldwide, 3 of
which constitute 94% of enhanced production (Aladasani and Bai, 2010). Traditional
thermal and gas EOR methods achieve high incremental rates; however, their application
range has not broadened significantly as they have matured, while the rate at which new,
promising EOR methods are being implemented at reservoir scale is alarming.
Low-salinity water flooding (LSWF) in both sandstone and carbonate reservoir
has emerged as a promising EOR method, though its development and application has
been hindered by the lack of consensus concerning its recovery mechanism(s).
Numerous LSWF core-flooding experiments have been undertaken, and two LSWF
simulation models have been proposed. However, to date, core-flooding experiments
have not been correlated with simulation results, particularly when analyzing the
sensitivities of LSWF recovery mechanisms for core plugs with different petrologies.
Furthermore, all of the experiments and simulations presume one-dimensional, single-
layered and unfractured porous media.
2
Every oil reservoir has a unique ionic environment that changes naturally and by
human intervention, which makes it difficult to identify recovery mechanism(s) in EOR
methods such as LSWF.
This study systematically aims to promote EOR implementation. The EOR
selection criteria is updated to include new EOR methods and reservoir properties
reported in EOR projects from 1998 to 2010. An EOR database comprising of 652
projects is built to carry out statistical analysis. Two new EOR screening tools are
generated from statistical analysis of the EOR database. The first EOR screening tool is
based on the dataset distribution of the main EOR methods and their corresponding
reservoir properties. The second new EOR screening tool uses incremental recovery as a
basis for specifying reservoir properties for the main EOR methods. Guidelines on the
limitations of EOR methods are updated based on the recent advancements, which
expands the application range of the EOR method.
Further evaluation is conducted for miscible CO2 flooding and LSWF because of
rapid project growth and promising incremental recovery, respectively. A deterministic
prediction model is established for miscible CO2 flooding, this tool provides an efficient
surrogate to traditional resource intensive reservoir simulations and probabilistic models
that require a long learning cycle.
LSWF recovery mechanisms are investigated using statistical analysis and
numerical solutions. A core-flooding database comprising of more than 411 experiments
is built to carryout statistical analysis. The statistical analysis included correlations
amongst the experiment parameters and the influence of fluid and rock properties on
incremental recovery and residual oil saturation. A LSWF reservoir simulator has been
developed and validated. Contrasting core-flooding statistical analysis with reservoir
simulation made it possible to conclude LSWF recovery mechanism(s) and provide an
explanation as to the conclusions of other scientists.
An up-scaled multi-dimensional model is developed for LSWF under various
reservoir wetting conditions, this is intended to encourage LSWF field wide
implementation. The model reveals the importance of sweep efficiency in lieu of
reducing the residual oil saturation. Finally, in view of growing interest in carbon
dioxide (CO2) sequestration and storage projects, a risk analysis case study is included.
3
2. LITERATURE REVIEW
2.1. ENERGY SUPPLY

Energy serves as a pillar in any country’s policy and strategy. While energy
demand may fluctuate, it will always increase as economies grow in both developed and
undeveloped nations. Energy prices experience inherent volatility due to the contending
perspectives of suppliers and consumers. Consumers understand that economic growth is
on par with energy consumption; however, suppliers will only enhance production
capacity when energy prices are driven to record highs. Therefore, a shortage in the
capacity to supply energy will always exist, which will benefit nations that export major
fuels.
Existing technologies do not provide a comparable distribution of fuel types;
instead, a single category of fuels, fossil fuels, dominates the world’s energy supply, with
oil and gas contributing to more than half of the world’s energy consumption. This
situation motivates the establishment of cartels that control the world’s main source of
energy. The policies of such cartels and the stability of their member nations add to the
volatility of energy prices. Secondary issues that further influence this volatility include
a growing concern for environmental protection and the lack of transparency regarding
oil and gas reserves from major producers.
2.1.1. World Energy Statistics. Fossil fuels have served as the dominant source
of the world’s energy supply since the Industrial Revolution, a pattern forecasted to
continue for the next three decades. World energy statistics from 1973 to 2003 indicate
that more than three-quarters of the world’s energy supply comes from fossil fuels, as
shown in Figures 2.1 and 2.2 (International Energy Association (IEA), 2005).
2.1.2. Future Energy Outlook. A 24-year outlook of the world’s energy supply
under different policy scenarios does not significantly alter current distributions of the
world’s energy sources, as shown in Figure 2.3. The International Energy Association
(IEA) highlights the following potential scenarios in the evolution of the world’s energy
supply. Gas is expected to overcome coal, and the forecasted consumption of both gas
and coal up until 2030 will be much less than their proven reserves. The demand for
nuclear and hydro energy will plateau, with no significant change in market share. The
4
IEA 450 Scenario, which does not consider the Fukushima nuclear disaster that has
resulted in many countries revisiting their energy policies on nuclear power, provides the
only exception to these outcomes. Demands for oil, gas and coal will increase notably in
comparison, with moderate increases in other sources of energy, such as combustible
renewables, wastes and other renewable sources of energy, as shown in Figure 2.4 (IEA,
2010).
Natural
Coal
Gas
Hydro 24.40%
21.20%
2.20%
Nuclear
6.50%
Oil
34.40% Combustible
Renewables &
Waste
Others 10.80%
0.50%
Figure 2.1. World Fuel Share (1973), Source: IEA, 2005
Coal
Natural Gas 24.40%
Hydro 21.20%
2.20%
Nuclear
6.50%
Oil
34.40% Combustible
Renewables &
Others Waste
0.50% 10.80%
Figure 2.2. World Fuel Share (2003), Source: IEA, 2005

5
450 Scenario 2035
New Policies Scenario 2035
Current Policies Scenario 2035
2008
0%
20%
40%
60%
80%
100%
Coal Oil Gas Nuclear Hydro Biomass Other Renewables
Figure 2.3. Forecast of World Energy Supply (2035), Source: IEA, 2010
Figure 2.4. Forecast of World Energy Demand (2035), Source: IEA, 2010
6
2.1.3. Oil Reserves. The Security Exchange Commission recently decided to

utilize the SPE system for classifying oil reserves (Oil and Gas Journal (OandGJ) 2009),
notwithstanding the differences in the mechanism by which it classifies hydrocarbon
reserves as compared to systems used by other countries, such as the United Kingdom,
China and Canada (SPE, Oil and Gas Reserve Committee, 2005). The mechanism by
which oil reserves are classified in the Middle East, the location of the world’s largest oil
reserves, is unknown.
2.1.4. Oil Supply. In 2006, the world witnessed its peak oil production, 70
million barrels per day, a high never expected to be reached again (IEA, 2010). The
depletion of giant oil fields is evident (Ali and Thomas, 2000), and despite a huge
increase in oil demand prices in 2008, world oil production has not surpassed its previous
rates. Furthermore, the volatility of oil delivery can readily impact shortages in the
world’s energy supply. In this context, actual production trends published by the IEA
indicate a 6.3% drop in the Middle East’s share of oil production from 1973 to 2003
(IEA, Key World Energy Statistics, 2005), which is attributed primarily to the region’s
political instabilities.
A growing awareness of public health and safety concerns and the need to
preserve the environment could impact the oil supply from the Middle East. Over the
past decade, all oil and gas operators in the Middle East implemented a Health, Safety
and Environment management system. Several of those operators currently hold
International Standards Organization (ISO) certifications in environmental management
systems. By contrast, from 1934 to 1978, and as recently as 2000, most of the associated
gas in the Middle East was flared to produce oil. Today, the cost of gas given the
petrochemical industry and industrial consumption has provided further incentives to cut
oil production to avoid flaring.
The Middle East is an unstable region that continues to witness conflicts, revolts
and political transformations. The stability of the oil supply from the region of the world
with the largest oil reserves has always presented an issue. A constant stalemate exists
between the major oil and gas producers in the Persian Gulf.
7
2.2. ENHANCED OIL RECOVERY

Alternative or renewable energy is a long way from achieving the durability,
efficiency and feasibility of hydrocarbon fuels (Bennett, 2008). To operate efficiently,
several types of engines require a fuel with a very high calorific value in order to tolerate
energy losses attributed to heat and noise. Alternative organic-based fuel sources such as
ethanol effect world food supplies (University of Groningen, 2004), and renewable
energy sources that generate either electricity or hydrogen infringe on rural development
and produce significant risks (DOE, 2006). Therefore, world energy forecasts for
alternative or renewable energy sources are negligible for the next three decades (IEA,
2008; IEA, 2010).
EOR utilizes unconventional hydrocarbon recovery methods that target the
approximately two-thirds of oil volume remaining in reservoirs after conventional
recovery methods have been exhausted (Green and Wilhite, 1998). Applying EOR
provides operators with several advantages. EOR application does not require a
substantial capital investment because existing infrastructures can be used to develop
depleted hydrocarbon fields. The potential to tap into reserves from giant oil fields
without any discovery or drilling completion risks is beneficial. EOR has the potential to
secure the world’s needed energy supply for several decades, as demonstrated in Figure
2.5. An additional benefit is that this method deals with a proven energy source that is
highly efficient and familiar to the refining, petrochemical and transportation industries.
Figure 2.5. Enhanced Oil Recovery Potential, Source: IEA, 2010

8
2.2.1. Oil Production. Enhanced oil production is critical today when many
analysts are predicting that the world has already reached its peak production and that the
demand for oil continues to grow faster than the supply. Only 22 billion of the 649
billion barrels of oil remaining in reservoirs in the United States (US) are recoverable by
conventional means. However, EOR methods offer the prospect of recovering as much
as 200 billion barrels of oil from existing US reservoirs, a quantity of oil equivalent to the
cumulative oil production to date (DOE, 2005).
In the early 1980s, many researchers investigated EOR because oil prices were
rising unabated, and a dramatic need arose to extract oil from depleted reservoirs. During
this time, most major oil companies operated research centers and funded major
programs to develop new technologies. These programs resulted in the production of
more than 20,000 bbl/day as a result of chemical EOR in the US alone. However, oil
prices collapsed in 1986 and hovered around $20 per barrel from 1986 to 2003. Most
operators, concerned about the lower price of oil, simply did not invest in either new
EOR technologies or new ideas to extract incremental oil from existing reservoirs.
However, oil prices recently have reached new highs of $60 to even $140 per barrel, and
many analysts believe that oil prices may stabilize above $100 per barrel. In this new
price environment and under conditions of increasing worldwide oil demand, few
discoveries of new fields, and the rapid maturation of fields worldwide, EOR
technologies have drawn increased interest.
2.2.2. Reservoir Flow Mechanics. Crude oil is found in underground porous
sandstone and carbonate rock formations. In the primary stage of oil recovery, the oil is
displaced from the reservoir into the wellbore (production well) and up to the surface
under its own reservoir energy, such as gas drive, water drive, or gravity drainage. In the
second stage, an external fluid, such as water or gas, is injected into the reservoir through
injection wells located in the rock that have fluid communication with production wells.
The purpose of secondary oil recovery is to maintain reservoir pressure and displace
hydrocarbons towards the wellbore. The most common secondary recovery technique is
water flooding (Craig, 1971). Once the secondary oil recovery process has been
exhausted, about two-thirds of the original oil in place (OOIP) is left behind due to both
9
microscopic and macroscopic factors. EOR methods aim to recover the remaining OOIP
(Green and Wilhite, 1998).
Microscopic factors include the various effects of oil-water interfacial tension
(IFT) and rock-fluid interaction (wettability) that give rise to oil in pores and crevices;
this oil cannot be dislodged under even large applied pressures (Stegemeier, 1977;
Slattery, 1974). The reservoir pore size may be as small as 0.1 m or less; therefore, it is
not surprising that IFT influences oil mobilization. The oil left behind after a sweep is
called residual oil saturation, expressed as Sor.
Macroscopic factors include reservoir stratification, with some strata showing
varying degrees of permeability. Thus, the displacing fluid travels through the high-
permeability zones, leaving oil in the low-permeability zones unswept (Bai and
colleagues 2007a; Bai and colleagues 2007b). Even in a uniformly permeable reservoir,
uniform displacement can break down when the displacing fluid is less viscous than the
crude, a situation known as adverse mobility ratio. In places, the less viscous fluid
penetrates the oil, a situation known as viscous fingering.
Another important reason why oil remains unswept is the negative capillary force
in oil-wet formations; this force impedes water imbibition into pore spaces in the
reservoir rock. This situation often occurs in carbonate reservoirs, more than 80% of
which are said to be oil wet. Other factors, such as areal heterogeneity, permeability
anisotropy, and well patterns, also leave some oil unswept by water. The unswept oil is
called remaining oil, and its corresponding saturation is called remaining oil saturation.
Oil recovery is the product of displacement efficiency (ED) and sweep efficiency (ES).
EOR methods focus on increasing either displacement efficiency by reducing residual oil
saturation in swept regions or sweep efficiency by displacing the remaining oil in
unswept regions. Residual oil saturation is a function of the capillary number, which is
the ratio of viscous force to capillary force. Typically, the capillary number for water
flooding is confined to below 10-6, usually to 10-7. The capillary number increases during
effective EOR by three magnitudes to about 10-3 to 10-4.
The capillary number can be reduced significantly by either lowering the IFT or
altering the rock’s wettability to create a more water-wet surface. Although the capillary
number also can be reduced by increasing the viscous forces, the reservoir fracture
10
gradient and pressure drops across the wells are limiting factors in this method (Green
and Wilhite, 1998). Oil in unswept regions can be recovered by (1) increasing the
viscosity of the displacing fluid, (2) reducing oil viscosity, (3) modifying permeability,
and/or (4) altering wettability.
2.2.3. Oil Production. The variety of EOR methods provides flexibility in
applying them to oil fields with different petrologies and for different stages of oil and
gas production. Applying EOR to developed fields offers the advantage of utilizing
existing infrastructures.
However, as reservoirs are unique in terms of their characterization and
properties, each EOR method can serve as a candidate to reservoirs with a specific range
of rock and fluid properties. EOR can be applied in the first stage of oilfield
development in cases such as thermal flooding for heavy oil reservoirs in which natural
reservoir forces are inadequate to induce the flow of oil to producing wells. EOR also
has been adopted in the second stage to further augment production rates by promoting
oil flow and to realize favorable recovery conditions, such as hydrocarbon flooding.
Additionally, EOR methods often are used in the tertiary stage in cases in which
oil fields have high water cut and low oil production rates. Therefore, EOR has the
potential to reclassify unrecoverable and contingent reserves in amounts exceeding the
quantity of oil currently produced. Oil is predicted to dominate the world’s energy
supply for the next three decades, as shown in Figure 2.6. The development of
technologies that enhance oil recovery from existing oil fields would supply the world’s
energy needs for several decades. Therefore, it is more important than ever to understand
lessons learned from past EOR applications and to develop new technologies and
methods.
2.2.4. Enhanced Oil Recovery Methods. Gas EOR is subcategorized as
immiscible or miscible flooding using carbon dioxide and nitrogen gases, as well as
water-alternating hydrocarbon gas (WAG) flooding. In the case of immiscible gas
flooding, the gas is injected below its critical pressure, thereby enhancing the
macroscopic displacement efficiency by increasing reservoir pressure and causing oil to
swell. By contrast, miscible gas flooding involves injecting gas at a pressure high enough
to achieve miscibility with the oil. Oil gravity is inversely proportional to the minimum
11
miscibility pressure, whereas heavy gases have lower miscibility pressures. The injected
gas solution achieves miscibility with the oil through single or multiple contacts
(Ghomian and colleagues 2008). These contacts considerably reduce the IFT in the
miscible zone; thus, the residual oil saturation decreases, and oil is mobilized.
Additionally, when the miscible gas “evaporates” in oil (Vahidi and Zargar, 2007), the oil
viscosity decreases, and the oil swells. The increase in viscous forces improves the
macroscopic displacement efficiency. The improvement in both microscopic and
macroscopic displacement efficiencies serves as evidence of the ability of miscible gas
flooding generally to achieve greater effectiveness than immiscible flooding (Vahidi and
Zargar, 2007).
Figure 2.6. The Role of Oil Discovery and Development, Source: IEA, 2010
Thermal EOR methods include steam, combustion, and hot water flooding, all
three of which elevate the temperature inside the reservoir to reduce oil viscosity. In
addition, oil swelling and an increased reservoir pressure resulting from high
temperatures create favorable oil recovery conditions. Therefore, thermal EOR improves
both the macroscopic and microscopic displacement efficiencies by reducing viscous
12
forces and by reducing IFT, especially during steam distillation, respectively (Cadelle
and colleagues 1981).
Chemical EOR methods inject chemicals, such as soluble polymers, cross-linked
polymers, surfactants, alkalines and their combinations. Chemical EOR can improve
either microscopic or macroscopic efficiency, or both. Polymers are added to water
during flooding to achieve favorable mobility ratios in the displacing front. The
displacing water becomes more viscous as the under-riding water is mitigated, thus
improving the macroscopic displacement efficiency (Chang and colleagues 2006).
Surfactants are added to the water during flooding to improve the microscopic
displacement efficiency by generating an emulsion between the oil and water interface.
This emulsion significantly reduces the IFT and mobilizes the oil (Krumrine and
colleagues 1982). Surfactants also improve the microscopic displacement efficiency by
reducing the capillary force, which decreases the oil contact angle. Alkaline interacts
with some acid oils to generate surfactants, which reduce the IFT proportionally based on
the pH value (Smith, J.E., 1993). Therefore, alkaline is added to the water to minimize
the use of surfactants and reduce the capillary force. Polymer-based gels are used during
conformance control to block high-permeability zones, diverting the displacing medium
to areas where oil has not been swept (Bai and colleagues 2004).
Microbes can be utilized to improve oil recovery. Microbial EOR generates gases
under reservoir conditions, thus improving the macroscopic displacement efficiency by
increasing reservoir pressure and decreasing oil viscosity. The macroscopic displacement
efficiency also may improve when the absolute permeability increases due to acidic
dissolution; alternatively, microbes could block high-permeability zones, thereby
improving sweep efficiency. Microbes can generate bio-surfactants that could reduce the
IFT and favorably alter wettability. Wettability also could be altered favorably by some
microbes that decrease the population of sulfate-reducing bacteria (Dietrichm and
colleagues 1996). However, microbial EOR is difficult to control. Furthermore, the
adsorption of surfactants to the reservoir rock and the biodegradation of surfactants
adversely impact the performance of microbial EOR (Gray and colleagues 2008).
2.2.5. Enhanced Oil Recovery Selection Criteria. Taber et al. published the
first EOR selection criteria in 1982; these criteria were updated in 1996 in a paper that
13
became the most widely cited Society of Petroleum Engineers (SPE) paper. The EOR
selection criteria categorize EOR methods into gas, chemical and thermal and are based
on a range of reservoir properties listed for each of these methods. This range is based on
reported EOR projects and surveys. The publication also includes the limitations of each
EOR method based on the prevailing technologies at the time the paper was written.
Much has changed since the EOR selection criteria were published in 1996.
Firstly, numerous EOR projects have been implemented since 1996, out of which
several new EOR categories and subcategories have been introduced. Additionally,
technological advances have surpassed some of EOR’s previous limitations.
Furthermore, the EOR selection criteria were based on a range of reservoir properties
without considering incremental recovery or the project’s distribution scale. Despite the
implementation of over 600 EOR projects since 1959 (The Oil and Gas Journal,
(OandGJ), 1998-2008), the use of EOR remains limited worldwide. The development
and implementation of any recovery methodology, especially on a field-wide scale,
requires confidence in its efficacy. Establishing such confidence requires an in-depth
analysis of EOR projects that would provide updated and more concise EOR selection
criteria.
2.3. WATER FLOODING

Water flooding is the most common and effective secondary-stage recovery
method because it improves or maintains water drive or depletion drive; in addition,
water flooding improves the displacement efficiency by manipulating relative
permeability and sweep efficiency. In hydrocarbon reservoirs, fluids are stored in porous
rocks. The capacity of the porous rocks to store fluids is termed porosity and is illustrated
in Figure 2.7.
Total porosity refers to the bulk volume of fluids occupying the porous media.
Effective porosity refers to the bulk volume of pores that are connected through the
porous media. The effective porosity may fluctuate due to mineral dissolution.
There are three main hydrocarbon phases that may flow within a porous media,
namely, oil, gas and water, as shown in Figures 2.8 and 2.9. The reservoir temperature
14
and pressure impact the number and quantity of phases present in the porous media. The
specific amount of a phase in the porous media is called saturation. Hydrocarbon phases
in the rock media interact with each other to form relative permeabilities and capillary
pressures.
Figure 2.7. Porosity Illustrations

Source:
http://www3.gov.ab.ca/env/water/GWSW/quantity/learn/evaluation/EV3_pumptest.html
Figure 2.8. Reservoir Flow
Source: http://www.co2crc.com.au/imagelibrary3/storage.php?screen=3
15
Figure 2.9. Reservoir Flow Behavior
Source: http://www.aps.org/meetings/march/vpr/2010/imagegallery/drainage.cfm
The ease with which a phase is dislodged from the porous rock depends on its
capillary pressure between the other phases. The ease with which a phase passes through
the porous media depends on its relative permeability. The capillary pressure is a
function of wettability and IFT. The phase predominantly responsible for wetting the
surface of the porous rocks classifies the reservoir’s wettability, as illustrated in Figure
2.10. Dislodging oil in an oil-wet reservoir proves difficult due to the imposed capillary
pressure.
It is less effective to mobilize oil when it forms high IFT against water. The other
aspect of hydrocarbon recovery involves displacing the oil into the producing well. Oil
displacement is a function of oil viscosity and flow velocity. The ability to displace
water into the producing wellbore also depends on the conductivity (permeability
variance, K variance) and connectivity (fractures/cross beddings) of the oil-bearing
formations. In typical hydrocarbon reservoirs, initial or primary oil recovery often
16
involves naturally-occurring forces that drive the oil to the wellbore. Such forces may
include dissolved gas, gravity, water and gas drives. Oil recovery can be represented as
the product of the microscopic displacement efficiency and the macroscopic displacement
efficiency, as indicated in Equation 1.
(1)
Water flooding improves oil recovery because the residual oil saturation
decreases, due to either oil expansion resulting from gas entrapment or gas displacement
of oil droplets from larger pores (Cole, 1969), as shown in Figure 2.11. Consequently, the
residual oil saturation decreases, as illustrated in Figure 2.12.
Figure 2.10. Contact Angle and Wettability

Source: http://www.spec2000.net/09-wettability.htm
17
Figure 2.11. Oil Expansions Due to Gas Entrapment and Displacement of Oil from
Larger Pores
Figure 2.12. Impact of Gas Trapping on Residual Oil Saturation

18
The water-flooding technique also is used to manipulate the reservoir oil’s

relative permeability in water-wet reservoirs by invoking the drainage process or by
mobilizing the non-wetting phase, as shown in Figure 2.13. In water flooding, the
correlation between wettability and relative permeability indicates an increase in the oil’s
relative permeability when water wettability increases. In other words, oil desorbs from
the reservoir rock when wettability approaches a more water-wet state.
Water production increases as oil fields mature. Therefore, oil exhibits very low
relative permeability in oil fields with high water cuts. The remaining oil saturation is
much higher than the optimal residual oil saturation due to oil or intermediate wetting
conditions. When the reservoir’s wettability is altered favorably to approach a more
water-wet state, the residual oil saturation decreases, and more oil is recovered
(Anderson, 1987).
Figure 2.13. Relative Permeability Curves

Source: http://www.petrocenter.com/reservoir/re02.htm
Water flooding typically is initiated when the reservoir pressure falls below the
bubble point. The resulting significant decrease in viscosity improves the oil’s mobility
19
ratio, as indicated in Equation 2. An unfavorable mobility ratio destabilizes the

displacement front and results in viscous fingering, which, under severe circumstances,
initiates under-riding or over-riding effects. The destabilization of a displacement front is
shown in Figure 2.14 as the mobility ratio becomes unfavorable and vice versa.
( ⁄ )
( ⁄ )
(2)
Unfavorable Mobility Ratio Favorable
Figure 2.14. Displacement Front Stability

Source: http://ulb.ac.be
2.3.1. Low-Salinity Water Flooding. Over the past decade, water-based EOR
methods have been developed to reduce IFT, increase water-wetness, improve the
mobility ratio and control conformance. This implies that water-based EOR methods can
improve microscopic and macroscopic displacement efficiency. More than 50% of the
world’s oil reservoirs are water flooded (Betty, 2003), and more than 50% of the world’s
oil reserves are stored in carbonate reservoirs (Okasha and colleagues 2009). Therefore,
low-salinity water flooding (LSWF) has emerged as a promising water-based EOR
20
method because it applies to both sandstone and carbonate reservoirs and has the
potential to achieve significant incremental recovery factors. In addition, LSWF is
relatively low cost and environmentally friendly.
Similar to alkaline flooding, LSWF spurred controversy amongst the scientific
community because of the variability involving its recovery mechanism(s), which
resulted from heterogeneous rock properties and different ionic environments. Another
similarity between alkaline flooding and LSWF is the potential for clay flocculation and
the consequent formation damage.
Although it has been suggested (Clementz, 1982) that a tradeoff between
microscopic sweep efficiency and macroscopic displacement efficiency will be
necessary, EOR does not demand such a compromise. However, considering the
heterogeneity amongst reservoirs’ chemical species, organic complexes and changes in
the ionic environments (Reed, 1967), each reservoir may be subject to core-flooding
experiments and simulation prior to selecting an optimal chemical design.
2.3.2. Low-Salinity Water-Flooding Experiments and Tests. Data from
laboratory core-flooding experiments have shown that LSWF in sandstone reservoirs
(Tang and Morrow, 1997), as well as low-salinity seawater flooding (LSSWF) in
carbonate reservoirs (Yousef and colleagues 2010), could result in a substantial
improvement in oil recovery (up to 25% in sandstone reservoirs and 17% in carbonate
reservoirs) over traditional water flooding. LSWF imbibition experiments conducted by
Tang and Morrow (1997) indicate significant incremental recovery rates of about 20%
and 33% when the salinity (mainly KCl) is reduced by 10:1 and 100:1, respectively.
LSWF core-flooding experiments for both outcrop sandstone and reservoir cores indicate
incremental recovery rates of about 13% and 25% for reductions in the salinity
concentration (mainly NaCl) of 10:1 and 100:1, respectively (Tang and Morrow, 1997;
McGuire and colleagues 2005; Zhang and colleagues 2007; Ashraf and colleagues
2010). LSSWF imbibition experiments conducted by Bagci and colleagues in 2001 and
Webb and colleagues in 2005 concluded favorable wettability modifications. Core-
flooding experiments also have shown that LSSWF improves oil recovery up to 17% in
carbonate cores when diluted to a ratio of 20:1 (Yousef and colleagues 2010). A
summary of LSWF experiments is illustrated in Figure 2.15.
21
35%
30%
Incremental Recovery 25%
20%
15%
10%
5%
0%
0.1000 0.2000 0.0100
Salinity
Imbibition (Sandstone Cores-KCL) Core Flooding (Sandstone Cores-NaCl)

Core Flooding (Carbonate Cores-NaCl)
Figure 2.15. Summary of LSWF Experiments
Single-well chemical tracer tests indicate incremental recovery rates from 8% to

19% for four different wells (McGuire and colleagues 2005). The reservoir conditions
are outlined in Table 1.1. The well with the lowest incremental recovery rate has been
flooded with comparably higher salinity water (7000ppm, compared to other wells at
>5000ppm). Zhang and colleagues (2007) also reported this observation when no
incremental recovery was observed for core flooding with water having a salinity of
8000ppm. Recovery rates for the remaining three wells range from 15% to 19%. The
well with the highest incremental recovery experiences a reduction in residual oil
saturation from 43% to 34%, which may indicate the weightage of the microscopic sweep
efficiency in LSWF modeling. The greatest reduction in residual oil saturation is from
21% to 13%; however, the incremental recovery rate is 18%, as summarized in Table 2.1
(McGuire and colleagues 2005), which could indicate the weightage of the macroscopic
22
displacement efficiency in LSWF modeling and the probable need to increase the low-
salinity water viscosity to mitigate viscous fingering (Ligthelm and colleagues 2009).
Table 2.1. North Slope Low-Salinity SWCTT Results (McGuire and colleagues 2005)
N-
Well L-1 L-122 3-39A
01A
Soi (%) 45 70 65 90
Sor-WF (%) 19 21 21 43
Sor-LSWF (%) 15 17 13 34
Water flood, PV (%) 26 49 44 47
EOR, PV (%) 4 4 8 9
EOR, OOIP (%) 9 6 12 10
Increase over water flooding (%) 15 8 18 19
Published oil production figures for a pilot well (Seccombe and colleagues 2010)
suggest a 10% incremental recovery rate from June 2008 through April 2009. The
salinity was lowered from approximately 27500ppm to approximately 13000ppm. The
oil production rate does not increase with a decrease in water salinity; however, water
production figures indicate a clear decrease after the start of LSWF.
A field-wide scale application of LSWF as a secondary recovery method was
inadvertently implemented in Syria because the only available source of water was river
water (1991-2004). After injecting 0.6 PV of low-salinity water in 2004, produced water
was injected thereafter. As of 2009, 0.6PV of produced water had been injected. Studies
conclude that wettability alteration has resulted in LSWF’s incremental recovery rate of
10-15% (Vledder and colleagues 2010).
23
2.3.3. Reservoir Simulation. The petroleum industry has witnessed a growing

interest in LSWF and LSSWF studies. However, most of the relevant research has
focused on the extent to which low-salinity water improves oil recovery and the
underlining mechanism of wettability alteration. By comparison, few quantitative studies
exist that investigate the flow and transport behavior of LSWF and LSSWF as improved
oil recovery (IOR) processes. Jerauld and colleagues (2006) modeled LSWF as a
secondary and tertiary recovery process in a one-dimensional model using salinity-
dependent oil/water relative permeability functions resulting from wettability alteration.
Tripathi et al. (2008) studied the flow instability associated with wettability alteration
using a Buckley-Leveret type analytical model in one dimension.
A general numerical model for LSWF and LSSWF in multi-dimensional, porous
or fractured reservoirs is required. The model’s formulation should incorporate known
IOR mechanisms for simulating LSWF and LSSWF processes. Homogenous models are
required to validate the mathematical formulation, contrast simulation results with
experimental results and conduct sensitivity analyses on recovery mechanisms.
Furthermore, multi-dimensional models are required, including fracture models for both
dual-porosity and dual-permeability models, to demonstrate the use of the proposed
modeling approach in simulating LSWF and LSSWF.
2.3.4. Recovery Mechanisms. Manipulating the reservoir’s ionic environment is
a guiding principle in LSWF and LSSWF. Understanding the rock’s mineralogy and
formation chemistry is crucial to achieving an optimal brine composition that would
promote multiple favorable recovery conditions (Zhang and colleagues 2007). Possible
mechanisms by which LSWF improves oil recovery could include: (1) the wettability
change towards water wetness as a result of clay migration (Tang and Morrow, 1999),
shown in Figure 2.16; (2) the pH increase as a result of CaCO3 dissolution, which
increases oil recovery through several mechanisms, including altering wettability,
generating surfactants, and reducing IFT (McGuire and colleagues 2005); and (3)
multiple-component ion exchange (MIE) between clay mineral surfaces and the injected
brine (Lager and colleagues 2006), illustrated in Figure 2.17. In general, the
improvement in oil recovery during LSWF in sandstone reservoirs depends on MIE, clay
content, formation water composition (Ca2+, Mg2+) and oil composition.
24
Figure 2.16. Clay Migration (Tang and Morrow, 1999)
Possible mechanisms by which LSSWF improves oil recovery could include: (1)
the wettability change towards water wetness as a result of sulphate content (Austad and
colleagues 2005) or sulphate content in combination with excess calcium close to the
rock’s surface (Zhang, and colleagues 2007); or (2) a decrease in IFT as brine
concentration decreases (Okasha and colleagues 2009). Generally, the improvement in
oil recovery during LSSWF in carbonate reservoirs depends on sulphate content,
formation water composition (Ca2+, Mg2+) and oil composition.
Clays have four characteristics. Firstly, they behave like colloids in aqueous
solution; therefore, their settling behavior is determined by electrostatic forces (Van
Olphen, 1964). Secondly, clays have an intrinsic negative charge caused by the
replacement of single-valent by multivalent cations. Thirdly, a positive or negative
charge is located on the edge of the clay grains and generated by an H+ reaction. In the
case of Kaolinite, the charge extends over the entire surface. There is little replacement
of single-valent cations at the aforementioned clay sites. Fourthly, absorbed ions are
exchangeable, and ion exchange is measured by the cation-exchange capacity (Fairchild
and colleagues 1988).
In sandstone reservoirs, both the silica surface and the oil (Buckley and
colleagues 1989) are assumed to be negatively charged, implying that the silica is water
wet and free of oil (Dubey and Doe, 1993). However, clay particles may also have a
25
positive charge (Clementz, 1976) due to contaminants. Carbonate reservoirs typically

maintain a positive charge; therefore, an anion exchange capacity exists for anion
adsorption.
The mechanism of oil adsorption and desorption to reservoir rocks depends on the
brine formation’s electrolytic content. In the context of lattice substitution in clays, if
clay is considered to have a negative charge on its surface, a double-layer ionic system is
generated. The first layer is considered synonymous with the surface and is followed by
a diffuse, or Gouy, layer that balances the surface’s negative charge with positive ions.
When the electrolytic content is lowered, the fixed layer contracts, and the Gouy layer
expands. Layer movement reverses when the electrolytic content is high. The expansion
and contraction of the Gouy layer dictates the predominant charge of free ions. The
contraction of the fixed layer does not alter the Gouy layer’s negative ion charge.
However, if the fixed layer expands due to multivalent cationic bonds, the Gouy layer
changes polarity. Therefore, when the brine formation’s electrolytic content is high, the
Gouy layer’s charge changes to consist predominantly of negative ions, which is
indicative of clay particles wetted by oil. On the other hand, if the brine formation’s
electrolytic content is low, the Gouy layer’s charge is predominantly positive, which is
indicative of clay particles wetted by water (Fairchild and colleagues 1988). The
expansion and contraction of the Gouy layer is illustrated in Figure 2.17.
Figure 2.17. Gouy Layer’s Expansion and Contraction

26
Experiments have shown notable incremental recovery in clay-free cores when

the injection’s brine composition is modified (Lager and colleagues 2007). Austad and
colleagues suggested that fluids such as seawater that contain sulphate can change the
reservoir’s wettability to a more water-wet state (Austad and colleagues 2005), and
studies have concluded that wettability modification in carbonate reservoirs occurs due to
sulphate adsorption in combination with excess calcium close to the rock’s surface
(Zhang and colleagues 2007). It follows that wettability modification in carbonate
reservoirs is not influenced by the expansion of the double Gouy layer due to a decrease
in the ionic charge of the injection fluid (Ligthelm and colleagues 2009). Although
sandstone core experiments suggest that decreasing the salinity increases the IFT
(Vijapurapum and Rao, 2004; Alotaibi and Nasr-El-din, 2009), carbonate core
experiments conclude that the IFT decreases for both dead and recombined oil as the
brine concentration decreases (Okasha and colleagues 2009). Reservoir core imbibition
experiments using seawater mixed with formation water (Webb and colleagues 2005)
and core-flooding experiments using seawater mixed with de-ionized water (Yousef and
colleagues 2010) indicate an increase in incremental recovery. Imbibition experiments
conducted for outcrop limestone cores using formation water and low-salinity water
produced similar results (Fjelde and colleagues 2008).
The potential of LSWF was first realized by Morrow and his and colleagues
(Jadhunandan and Morrow, 1991, 1995; Yildiz, Valat and Morrow, 1996; Tang and
Morrow, 1997, 2002); their published core-flooding and spontaneous imbibition
experiments concluded the attainability of incremental recovery from reservoir sandstone
cores. Later, several core-flooding experiments were conducted to identify LSWF’s
recovery mechanism(s). Tang and Morrow (1999a, 1999b) attributed LSWF’s recovery
mechanism to the partial stripping of mixed-water fines. McGuire et al. (2005) indicated
that LSWF has similar effects as alkaline on oil-water IFT due to increased pH values.
The idea that multivalent cations bridge the negatively-charged oil to the clay minerals
was introduced while researching rock wettability alterations (Anderson, 1986; Buckley
and colleagues 1998). However, in the context of LSWF, Lager (2006) advocated that
multi-component ionic exchange (MIE) results in the desorption of oil when low
electrolyte water is used for water flooding, especially Mg2+ exchange, which was
27
confirmed by measuring the magnesium content of the produced water (Lager, 2007;
Alotaibi and colleagues 2010). Lager et al. (2008) also attributed the diffuse layer’s
expansion and the associated increase in water wetting to the stripping of divalent cations
behind the LSWF displacement front, which has been confirmed to occur after the
breakthrough of the salinity front.
Although a consensus has been reached that LSWF ultimately reduces residual oil
saturation due to capillary desorbtion, contention continues regarding the root cause for
capillary desorbtion, whether it occurs due to the migration of fine grains (DLVO
theory), the reduction in IFT, the expansion of the Gouy layer caused by the stripping of
multivalent cations, or the importance of the mineral Kaolinite in clays. Core-flooding
experiments conducted by Zhang et al. (2007) showed no evidence of clay content in the
production stream or the oil/brine interface, which raised doubt about the role of fine-
grain migration as the recovery mechanism of LSWF. Similarly, Lager et al. (2006)
demonstrated LSWF’s incremental recovery in brine with a pH < 7, which raised doubt
about the suggestion that LSWF’s recovery was similar to alkaline flooding. As for the
MIE theory advocated by Lager et al. (2006), it was suggested that polar oil components
also can adsorb onto clay minerals without bridging divalent cations and that a reduction
in Magnesium content can be caused by precipitation as Mg(OH)2, especially at increased
pH levels during LSWF (Austad and colleagues 2010).
Furthermore, Ligthelm and colleagues (2009) concluded that cation stripping does
not play an essential role in wettability modification. Jerauld and colleagues (2006) and
Seccombe and colleagues (2008) suggested a relationship between the mineral content
Kaolinite in clays and LSWF’s incremental recovery. However, Cissokho (2009) found
substantial LSWF incremental recovery in cores free of Kaolinite. The literature
evidently assumes that LSWF’s recovery should be exclusive to a single mechanism.
However, it is more than likely that LSWF can create multiple favorable recovery
conditions (Austad and colleagues 2010) that are variably present; this would explain (a)
the varying recovery rates, and (b) the varying reductions in the ionic strengths required
for LSWF, especially when the heterogeneity of reservoir fluids and rock properties are
considered.
28
3. ENHANCED OIL RECOVERY GUIDELINES AND SELECTION CRITERIA
Enhanced oil recovery (EOR) technologies can augment the production of

hydrocarbons and therefore are key in achieving the ultimate goal of increasing recovery
volumes, which, as Chapter 1 has shown, is critical given the world’s predicted energy
needs and current supply. A review of the existing EOR criteria is presented here,
revealing the need for updated criteria because of their datedness and their emphasis on
minimum and maximum average values that do not represent a sound basis for the
selection of candidate reservoirs for EOR. Updated criteria that provide a more
representative understanding of selection values are necessary if EOR technologies are to
be implemented to their full potential.
Two new EOR selection criteria are proposed in this chapter after a thorough
analysis of all EOR projects, including those reported from 1998 through 2010. These
criteria also consider new EOR methods and the addition of reservoir properties.
The creation of the first new EOR criterion was motivated by the inherent risks of
using average values of reservoir properties for each EOR method. Alternatively, a data
distribution, as presented here, delineates ranges within which the majority of projects
fall, thus providing a much clearer picture of the reservoir properties for each EOR
method (Aladasani and Bai, 2010). The second proposed EOR criterion is based on
incremental recovery (Aladasani and Bai, 2011). The reservoir properties that achieve the
highest production gains are highlighted here. To ensure accuracy, these new criteria are
proposed to apply only to EOR methods associated with sufficient projects and datasets.
Chapter 2 provides further contrast to Taber’s (1983) consensus of previous work by
representing both an inductive and deductive approach to further offset the limitations of
developing EOR criteria.
3.1. UPDATING ENHANCED OIL RECOVERY SELECTION CRITERIA

The guidance available for selecting EOR methods (Taber and colleagues 1996) is
widely cited but does not include several EOR methods and projects that have been
reported in the past 15 years. Furthermore, the current EOR selection criteria serve only
29
as an initial selection tool because they are based on ranges of reservoir properties rather
than the distribution of reservoir properties in reported EOR projects.
3.1.1. Enhanced Oil Recovery Database. The EOR database was built based on
the EOR survey reports biennial published by The Oil and Gas Journal (from 1998
through 2010) and SPE publications. The Oil and Gas Journal has issued a consolidated
biennial EOR survey report since 1974. It includes, where available, details such as
operator, project location, start date, reservoir characteristics, production data, etc. SPE
publications serve as another source of EOR project information, which, though
fragmented, provide operators’ experiences in implementing EOR projects. Consistent
EOR project reporting remains a challenge. In several cases, critical details are not
included. The database excludes EOR projects that report no reservoir properties.
Furthermore, single EOR project entries (e.g., acid gas and combined nitrogen and
hydrocarbon flooding) are not included in the database analysis due to a lack of datasets.
The amount of data available, however, is sufficient to establish an EOR project database.
Microsoft Access was used to construct the EOR project database. Table fields
include oil properties such as gravity, viscosity and temperature, as well as reservoir
properties such as formation type, porosity, start and end oil saturations, permeability and
depth. They also include country, field name, project start date and production details.
The database includes 652 projects, of which 613 were reported in The Oil and Gas
Journal (from 1998 through 2010) and 39 by the SPE. The database covers all four
categories of EOR methods: gas, thermal, chemical and microbial. The database is
published at http://www.eorcriteria.com.
3.1.1.1. Database analysis. Microsoft Access can generate graphs with multiple
series (variables) on both the x and y axes. In addition, the datasets for each series can be
shown as mathematical functions, including sums, standard deviations, counts, etc. This
feature in Microsoft Access provides a means by which to correlate reservoir fluid
properties with rock characteristics and/or enhanced production. Initially, the reservoir
formation type is plotted against each EOR method, including the number of projects
using each method. Next, various figures are generated that show the entire range of
reservoir properties against each EOR method, project count and, finally, enhanced
production rates.
30
The first step in analyzing the EOR project database is to construct a profile of
worldwide EOR projects. This is followed by the categorization and sub-categorization
of EOR methods and their respective share of projects. Next, each county’s share of
EOR projects is represented. Project trends then can be correlated to oil prices and
cumulative enhanced recovery. To establish a baseline, cumulative enhanced recovery is
included only for the EOR projects reported in 2008; the enhanced production rates that
year should not be considered as the initial production rate. Subsequently, worldwide
cumulative enhanced production is expanded to illustrate the production share of each
EOR method and the geographical location of the project. In addition, correlations
between reservoir formation type, EOR methods and the corresponding number of EOR
projects are illustrated. Finally, the enhanced production versus EOR methods, viscosity,
permeability, American Petroleum Institute (API) gravity, depth, porosity, temperature
and oil saturations are shown.
3.1.1.2. Enhanced oil recovery selection criteria. One of the most widely cited
publications in the field of petroleum engineering is the EOR criteria published by Taber
and colleagues in 1996. These criteria consist of 12 EOR methods tabulated against 9
reservoir properties. The reservoir properties are based on minimum, maximum and
average values published by The Oil and Gas Journal of EOR surveys from 1974 to
1996.
The EOR criteria published by Taber and colleagues (1996) are updated here to
include reported EOR projects from 1998 to 2010, as well as new EOR categories,
subcategories and project details. Newly-added EOR categories include microbial EOR,
miscible WAG, and hot water flooding.
New subcategories also are added under the category of immiscible flooding and
include CO2, nitrogen and WAG methods of EOR. The reservoir properties also have
been expanded to include porosity, number of EOR projects for each EOR method,
permeability and depth ranges for both miscible and immiscible gas EOR methods. The
EOR criteria were constructed and updated in the following manner. Oil property and
reservoir characteristic fields were queried to determine the range of each reservoir
property for each EOR method. An average for each reservoir property was then derived.
The EOR selection criteria are not intended to present threshold limits because such
31
limits should be developed scientifically. The consolidation of 652 EOR projects into the
screening criteria stands as a testimony to the work of Taber and colleagues (1996).
3.1.1.3. New enhanced oil recovery selection criteria. EOR projects are better
represented through dataset distribution. The number of EOR projects (datasets) should
be evaluated to indicate where EOR projects are concentrated for each reservoir range.
Extreme minimum and maximum values could adversely impact the EOR criteria, even
when averages are established; therefore, box charts are used to illustrate the reservoir
property distributions for the main EOR methods. The generated figures represent the
range in which the majority of EOR projects are located plotted against selected reservoir
properties. The minimum and maximum values for each reservoir property are identified.
Five EOR methods were selected to ensure an adequate number of data-sets. Legends
include the minimum and maximum range and the average value; more significantly, the
number of projects for each value was determined from the minimum to maximum API
range. Subsequently, the highest percentage concentration of project clusters within the
reservoir property range was established. The project clusters and the reservoir property
dataset distributions are more indicative of EOR selection criteria than the minimum,
maximum and average values, similar to the data-set distribution of reservoir properties
reported in EOR projects.
Enhanced production, rather than project count, is used as an EOR selection
criterion to establish key reservoir properties and their corresponding ranges. Two new
approaches are proposed to identify candidate reservoirs for EOR methods. The first
criterion correlates reservoir properties with enhanced production, and the second
criterion correlates the number of data-set distributions.
3.2. UPDATED ENHANCED OIL RECOVERY SELECTION CRITERIA

The Enhanced Oil Recovery Database can be accessed at
http://www.eorcriteria.com, as well as the following mirror sites:
http://www.eorcriteria.net and http://www.eorcriteria.org. In addition, a soft copy of the
database is attached in Appendix B.
32
3.2.1. Databases Analysis. The first step in analyzing the data stored in the EOR
project database is to construct a profile of worldwide EOR projects. The EOR projects
are classified into four main categories, namely, thermal, gas, chemical and microbial
methods. The worldwide use of each of these main categories is shown in Figure 3.1.
The main EOR categories are then subcategorized, as shown in Figure 3.2, to provide a
further breakdown of worldwide EOR projects.
Microbial
0.61%
Chemical
10.89%
Gas
40.80%
Thermal
47.70%
Figure 3.1. Worldwide EOR Project Categories (1959-2010)

Data adopted from Anonymous, The Oil and Gas Journal (1998, 2000, 2002, 2006), Awan et al.
(2006), Cadelle et al. (1981), Demin and colleagues (1999), Demin et al. (2001), Hongfu et al. (2003),
Koottungal (2008), Koottungal (2010), Mortis (2004), Taber (1996).
Figure 3.1 indicates that thermal methods are the leading methods used worldwide
for EOR projects, followed by gas methods. More specifically, steam flooding is the
leading thermal EOR method, followed by miscible gas injection in the gas methods
33
category, as shown in Figure 3.2. While thermal EOR continues to dominate (Figure
3.1), the adoption of miscible flooding methods has increased gas EOR projects to 41%
(Figure 3.2), and since 2006, gas EOR methods in the United States (US) have accounted
for the majority of enhanced oil production at 53% (Koottungal, L., 2008). The second
step is to represent each country’s share of EOR projects and to break down the EOR
methods implemented by each corresponding country, as shown in Figure 3.3.
Therm-Combustion Therm-Hot
4% Water
2%
Gas Miscible
35%
Therm-Steam
42%
Gas Immiscible
6%
Chem-ASP
3% Chem-Polymer
8%
Figure 3.2. Worldwide EOR Project Subcategories (1959-2010)

(2006), Cadelle et al. (1981), Demin et al. (1999), Demin et al. (2001), Hongfu et al. (2003), Koottungal
(2008), Koottungal (2010), Mortis (2004), and Taber (1996).
34
Figure 3.3. EOR Methods and Country Distributions

35
As Figure 3.3 demonstrates, the US, Canada and China lead the world in EOR
project implementation. The US and Venezuela conduct the majority of steam-flooding
EOR projects. Miscible flooding is led by the US and Canada, while China leads the
world in chemical EOR projects. To further examine worldwide EOR project
implementation trends, the numbers of EOR projects implemented, as well as enhanced
oil production and crude oil prices, are cross-plotted, as shown in Figure 3.4, and the
production share of the main EOR methods is shown in Figure 3.5. To establish a
baseline, Figure 3.4 includes only EOR projects reported in 2010; the enhanced
production rates that year should not be considered as the initial production rate.
2000000 700
1800000
600
1600000
Enhanced Production (BOPD)
1400000 500
Number of EOR Projects

1200000
400
1000000
300
800000
600000 200
400000
100
200000
0 0
1970
2004
1959
1962
1964
1966
1968
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2006
2008
2010
Fiscal Year
Enhanced Production Number of EOR Projects Crude Price
Figure 3.4 EOR Projects and Enhanced Production Trends

Data adopted from Koottungal (2010)
36
The number of EOR projects has increased dramatically since 1959 when the first
project was undertaken, most notably during the early 1980s and late 1990s (Figure 3.4).
Despite increasing enhanced production rates and oil prices, the number of EOR projects
remained relatively constant from 2006 through 2010 (Figure 3.4), a pattern that could be
attributed to incomplete reporting of EOR projects.
Chemical
1%
Gas
32%
Thermal
67%
Figure 3.5. Worldwide Enhanced Production Share

Data adopted from Koottungal (2010)
Thermal EOR accounts for the majority of EOR (Figure 3.5); however, because
EOR can be applied as a primary, secondary, or tertiary recovery stage, a new illustration
is required to demonstrate the recovery stage of the main EOR methods. This is achieved
by cross-plotting start and end oil saturations and enhanced production against the main
EOR methods (Figure 3.6).
37
0.70 10000000
0.60 1000000
Enhanced Production (MBOPD)

0.50 100000
Oil Saturation
0.40 10000
0.30 1000
0.20 100
0.10 10
0.00 1
Chemical Thermal Gas
EOR Methods
Oil Saturation (Start) Oil Saturation (End) Enhanced Production
Figure 3.6. Oil Saturations and Enhanced Production Distributions

It is evident from Figure 3.6 that thermal EOR is applied over a wide range of oil
saturation levels because it is used also in the primary and secondary oil recovery stages
in heavy and medium-gravity oil recovery, respectively. Similarly, gas EOR also is used
as a secondary recovery method; thus, a wider oil saturation range is observed in gas than
in chemical EOR. Chemical EOR usually is employed after water flooding is well
underway. Figure 3.6 illustrates the benefits of initiating chemical EOR at the start of
secondary recovery to improve overall recovery efficiency.
38
The third step in analyzing the EOR database is to link reservoir formations with
EOR methods. Initially, the number of EOR projects was plotted against EOR methods
and formation types (Figure 3.7).
Figure 3.7. EOR Methods and Formation Type Distributions

39
Figure 3.7 indicates that sandstone reservoirs rely almost exclusively on thermal
(steam) methods, immiscible gas methods and chemical (polymer) methods. The
possible importance of permeability for the aforementioned EOR methods is highlighted.
To verify this observation, the range of reservoir properties for the selected EOR methods
is illustrated in Figures 3.8 and 3.9.
40
35
30
25
20
15
10
0
Porosity
API
API Porosity
Figure 3.8. EOR Methods Versus Selected Average Fluid and Reservoir Properties
40
8000
7000
6000
Permeability (mD)
5000
Viscosity (cP)
Depth (feet)
4000
3000
2000
1000
0
Chemical Thermal Thermal Hot Miscible Gas Immiscible Microbial
Combustion Water Gas
Viscosity Permeability Depth
Figure 3.9. EOR Methods – Selected Average Fluid and Reservoir Properties
It is concluded that EOR methods can be functions of one or more reservoir

properties. For example, sandstone reservoirs, which are typically characterized by high
permeability, rely almost exclusively on thermal steam, immiscible gas and chemical
polymer methods. Similarly, API gravity and depth are functions of miscible gas
flooding; this is to ensure that the minimum miscibility pressure (MMP) is achievable
and that the MMP does not fracture the formation.
41
This relationship is demonstrated in Figures 3.8 and 3.9 by examining API gravity
and depth ranges for miscible gas methods. Effective EOR criteria should consider all
pertinent reservoir properties that influence each EOR method.
3.2.2. Enhanced Oil Recovery Selection Criteria. The EOR criteria published
by Taber and colleagues (1996) was updated to include EOR survey reports submitted
from 1998 through 2010, as shown in Table 3.1. The updates to the EOR criteria include
the addition of the entire range of oil and reservoir properties for all EOR methods, a
reservoir fluid property, namely, porosity, and permeability and depth ranges for miscible
and immiscible gas EOR methods because of their importance, which was shown in
Figures 3.7, 3.8 and 3.9.
New categories and subcategories of EOR methods also were added to the EOR
criteria, including the categories of microbial EOR, miscible WAG, and hot water
flooding, as well as the immiscible gas flooding subcategories of CO2, nitrogen and
WAG. Furthermore, the new criteria include the number of EOR projects (the number of
datasets) to provide an impression of the confidence level used for each EOR method to
derive the EOR selection criteria.
As a result, the majority of the reservoir properties were updated, and the number
of EOR methods has been expanded from 12 to 16. To illustrate the contributions in
updating the EOR criteria, box figures represent values adopted from Taber and
colleagues (1996).
Table 3.1. EOR Selection Criteria

Oil Properties Reservoir Characteristics
Porosit
# Gravit Oil Net
S EOR Viscosit y Formatio Permeabilit Temperatur
y Saturatio Thicknes Depth (ft)
N Method Project y (cp) n Type y (md) e (⁰F)
(⁰API) (%) n (% PV) s
s
Miscible Gas Injection
Sandston 1500-
22-45 35-0 Mar-37 15-89 1.5-4500 82-257
e or [Wide 13365
1 CO2 153
Avg. Avg. Avg. Carbonat Avg. Range] Avg. Avg.
Avg. 46
37 2.08 15.15 e 209.73 6230.17 138.10
18000- Sandston 4040[4000]
23-57 4.25-45 30-98 0.1-5000 [Thin 85-329
Hydrocarbo 0.04 e or -15900
2 67 unless
n Avg. Avg. Avg. Carbonat Avg.
Avg. 71 Avg. 726.2 dipping] Avg. 202.2
38.3 286.1 14.5 e 8343.6
42
Table 3.1 EOR Selection Criteria (Continued)

24- 130-
33-39 0.3-0.9 7545-8887 194-253
Nov 1000
3 WAG 3 Sandstone
Avg. Avg. Avg. Avg.
Avg. 0.6 NC Avg. 8216.8
35.6 18.3 1043.3 229.4
38[35]- 7.5- 0.76[0.4]- 10000[6000]-
0.2-0c Sandstone 0.2-35 [Thin 190-325
54 14 0.8 18500
4 Nitrogen 3 or unless
Avg. Avg. Avg. Avg. Carbonate Avg. dipping] Avg.
Avg. 14633.3
47.6 0.07 11.2 0.78 15.0 266.6
Immiscible Gas Injection
28-
16-54 18000-0 47-98.5 Mar-00 82-325
Nov 1700-18500
5 Nitrogen 8 Sandstone
Avg. Avg. Avg. Avg. Avg. 7914.2 Avg.
Avg. 71
34.6 2256.8 19.46 1041.7 173.1
Nov-
592-0.6 17-32 42-78 Sandstone 30-1000 1150-8500 82-198
35
6 CO2 16 or
Avg. Avg. Avg. Avg.
Avg. 56 Carbonate Avg. 3385 Avg. 124
22.6 65.5 26.3 217
22-
22-48 4-0.25 75-83 40-1000 6000-7000 170-180
May
7 Hydrocarbon 2 Sandstone
Avg. Avg. Avg.
Avg. 2.1 Avg. 79 Avg. 6500 Avg. 175
35 13.5 520
16000 - 18- 100-
9.3-41 Sandstone 2650 -9199 131-267
Hydrocarbon 0.17 31.9 6600
8 14 Avg. 88 or
+ WAG Avg. Avg. Avg. Avg. Avg.
Carbonate Avg. 7218.71
31 3948.2 25.09 2392 198.7
Chemical Methods
13- 10.4- 1.8 -
4000-0.4 34-82 9460-700 237.2-74
42.5 33 5500
9 Polymer 53 Sandstone [NC]
Avg. Avg. Avg. Avg.
Avg. 64 Avg. 4221.9 Avg. 167
26.5 123.2 22.5 834.1
g 68[35]- 3900[9000]- 158[200]-
23[20]- 6500 -11 26-32
74.8 2723 118 [80]
Alkaline Avg. Avg. Avg. Avg.
Surfactant 34[35] 596[10]- Avg. 2984.5
10 13 875.8 26.6 73.7 Sandstone [NC] 121.6
Polymer Avg. 1520
(ASP) 32.6
15.6- 14-
22-39 43.5-53 50-60 5300-625 155-122
Surfactant + 2.63 16.8
11 4 Sandstone [NC]
P/A Avg. Avg. Avg. Avg. Avg.
Avg. 49 Avg. 3406.25
31.75 7.08 15.6 56.67 126.33
Thermal/Mechanical Methods
2770- Sandstone 10 -
Oct-38 14-35 50-94 or 400-11300 64.4-230
1.44 15000
12 Combustion 27 Carbonate [>10]
Avg. Avg. Avg. [Preferably Avg. Avg.
Avg. 67 Avg. 5569.6
23.6 504.8 23.3 Carbonate] 1981.5 175.5
Aug- Dec- 1i-
5E6-3h 35-90 200-9000 10-350
33 65 15001
13 Steam 274 Sandstone [>20]
Avg. Avg. Avg. Avg. Avg.
Avg. 66 Avg. 1647.42
14.61 32594.96 32.2 2669.70 105.91
8000- 900-
25-Dec 25-37 15-85 500-2950 75-135
170 6000
14 Hot Water 10 Sandstone -
Avg. Avg. Avg. Avg.
Avg. 58.5 Avg. 1942 Avg. 98.5
18.6 2002 31.2 3346
[Zero [>8 wt% [> 3:1
[Surface [7] – [Mineable
15 - cold [NC] [NC] [>10] overburden [NC]
Mining] [11] tar sand]
flow] Sand] to sand ratio]
Microbial
26-
Dec-33 8900-1.7 55-65 180-200 1572-3464 86-90
Dec
Avg. Avg. Avg. Avg.
16 Microbial 4 Avg. 60 Sandstone - Avg. 2445.3 Avg. 88
26.6 2977.5 19 190
43
3.2.3. New Enhanced Oil Recovery Selection Criteria

3.2.3.1. Dataset distributions. The fourth stage of analysis requires representing
the distribution of EOR projects against the reservoir properties to determine where EOR
projects are concentrated for each reservoir range. As an example, Figure 3.10 represents
API gravity. Extreme minimum and maximum values could adversely impact the EOR
criterion, even when averages are established; therefore, box charts are used to illustrate
reservoir property distributions for the main EOR methods.
Figures 3.10 through 3.16 represent the range within which the majority of EOR
projects are located, plotted against selected reservoir properties. As an example, the
minimum and maximum API gravity values were identified for each of the five EOR
methods outlined in Figure 3.11 (with a red box and a purple cross indicating the
minimum and maximum values, respectively). The average API value then was
determined for each of the EOR methods and highlighted as a green triangle. (This was
the basis for J.J. Taber’s establishment of the EOR selection criteria in 1995) The next
step was to identify the number of projects for each API value from the minimum to the
maximum API value. Finally, the API range with the most datasets or projects was
identified from r1 (blue diamond) to r2 (sky-blue asterisk); therefore, r1 - r2 represents an
API range within which the majority of miscible flooding projects have been
implemented. Table 3.2 summarizes Figures 3.10 through 3.17 to represent a new
approach for developing EOR criteria.
25
20
Number of Projects
15
10
0
0.00 10.00 20.00 30.00 40.00 50.00 60.00
API Gravity
Figure 3.10. API Gravity Distribution in Miscible EOR Projects

44
60
50
73%
40
52%
API Gravity
r1
58%
30 min
avg
max
20 78% r2
50%
10
0
Miscible Immiscible Steam Flooding Combustion Chemical EOR
Flooding Flooding
Figure 3.11. API Gravity Distribution Versus Selected EOR Methods

The results in Figure 3.11 indicate that the average API values for both miscible
and immiscible EOR projects are a good indication of where most of the datasets lie and
hence serve as appropriate criteria for API gravity. However, the average API value for
steam flooding is close to the upper boundary of steam flooding project distribution. The
average API values for both combustion and chemical EOR methods lie outside the
boundaries of the box charts. The average API value for the combustion method
45
exaggerates the EOR criteria for API gravity, whereas the average API value for the
chemical EOR method understates the EOR criteria for API gravity. The use of average
values for EOR criteria can be misleading.
Miscible
Flooding Immiscible Steam Flooding Combustion Chemical EOR
(#226) Flooding (#40) (#274) (#27) (#70)
10000000
1000000
100000
10000
r1
1000
Viscosity (cP)
min
100 51%
avg
69%
10 67% max
r2
1
0.1
64% 58%
0.01
0.001
0.0001
(Number of Projects)
Figure 3.12. Viscosity Distribution Versus Selected EOR Methods

The results in Figure 3.12 highlight the risk of using average values in EOR
criteria because the average viscosity values for all of the selected EOR methods lie
outside the boundaries of the box chart that indicates where the majority of EOR projects
are implemented for the corresponding EOR method.
46
0.9
0.8
65%
0.7
Saturation (Fraction)
0.6 70%
64%
r1
67%
0.5 min
62% avg
0.4 max
r2
0.3
0.2
0.1
0
Flooding (#226) Flooding (#40) (#274) (#27) (#70)
(Numer of Projects)
Figure 3.13. Saturation Distribution Versus Selected EOR Methods

In Figure 3.13, the average viscosity values lie at the boundaries of the box charts;
more specifically, they lie at the upper limits for miscible, immiscible, steam flooding
and combustion methods and at the lower limit for chemical EOR methods. The results
in Figure 3.13 reiterate the inherent limitations of using average reservoir properties in
EOR criteria.
47

100000
10000
56%
1000
r1
60%
min
Permeability (mD)
100 53% avg

52% max
10 r2
64%
0.1
Figure 3.14. Permeability Distribution Versus Selected EOR Methods

(2008), Koottungal (2010), Mortis (2004), Taber (1996).
In Figure 3.14, the average permeability values for miscible, immiscible and
combustion EOR methods extend well beyond the box charts where the majority of
implemented EOR projects lie. This is significant because numerous candidate reservoirs
may be disqualified based on an EOR criterion that uses average values, especially when
the baseline consists of hundreds of projects (datasets). The average permeability values
for the steam and chemical EOR methods lie close to the upper limit of the box chart
boundaries. Similar to miscible flooding, this representation adversely impacts the
48
validity of the EOR criteria that use average values due to the number of steam EOR
projects.
70
60
50
Porosity (%)
40
r1
76%
min
30 avg
69% 67%
max
55% r2
20
62%
10
0
Miscible Immiscible Steam Combustion Chemical EOR
Flooding Flooding (#40) Flooding (#27) (#70)
(#226) (#274)
Figure 3.15. Porosity Distribution Versus Selected EOR Methods

Figure 3.15 indicates that the average porosity values lie at the upper limits of the
box charts for the miscible flooding and combustion EOR methods and at the lower limits
49
of the box charts for the immiscible, steam flooding and chemical EOR methods.
Clearly, average porosity values do not represent where the majority of EOR projects lie.
20000
18000
16000
14000
12000
Depth (ft)
10000 r1
min
8000 avg
max
6000
55% r2
4000 51%
48% 48%
64%
2000
0
Miscible Immiscible Steam Combustion Chemical EOR
Flooding Flooding (#40) Flooding (#27) (#70)
(#226) (#274)
Figure 3.16. Depth Distribution Versus Selected EOR Methods

Figure 3.16 indicates that the average depth for all of the EOR methods in
question is exaggerated, especially for miscible flooding. Overstating the EOR criteria
could cause future EOR project performance to suffer and good candidate reservoirs to be
disqualified. The magnitude of improper screening based on 226 datasets of miscible
50
flooding projects could potentially impede EOR implementation and success; this is a
detriment to EOR development and a deterrent to operators investing in EOR
technologies.
400
350
300
Temperature (⁰F)
250
r1
64%
200 min
avg
150 max
68%
52% 65% r2
100 77%
50
0
Figure 3.17. Temperature Distribution Versus Selected EOR Methods

In Figure 3.17, the average temperature values for immiscible and steam flooding
fall well within the box chart where the majority of implemented EOR projects lie.
Average temperature values are exaggerated for miscible and chemical EOR methods but
understated for combustion methods. The majority of miscible flooding projects lie
51
below the average temperature value, which may disqualify candidate reservoirs with
reservoir temperatures between 90 and 140 ⁰F. Table 3.2 summarizes Figures 3.11
through 3.17.
Table 3.2. New EOR Criteria – Based on Project Distributions of Reservoir Properties
Data adopted from Anonymous, The Oil and Gas Journal (1998, 2000, 2002, 2006), Awan et al. (2006),
Cadelle et al. (1981), Demin et al. (1999), Demin et al. (2001), Hongfu et al. (2003), Koottungal (2008),
Koottungal (2010), Mortis (2004), and Taber (1996).
EOR Reservoir Properties

No.
Start Oil Permeabilit Temperatur
Method Project Viscosit Porosity Depth
API Saturatio y e
s y (cP) (%)
n (mD) (ft) (○F)
34- 4200-
Miscible 0-1 0.33-0.55 0.1-100 16-Jul 95-160
226 44 6700
Flooding
73% 64% 62% 64% 62% 55% 52%
19- 1970-
Immiscible 0-10.5 0.42-0.62 30-300 22-32 120-194
40 36 5708
Flooding
66% 58% 67% 53% 69% 51% 68%
16- 800-
Steam Mar-00 0.50-0.70 1000-3000 30-38.8 80-130
274 Oct 1800
Flooding
78% 51% 64% 56% 76% 64% 77%
19- 1575-
1.44-2 0.50-0.70 Oct-85 17-25 185-230
Combustion 27 27 5000
50% 67% 70% 52% 55% 48% 64%
32- 2723-
Chemical Sep-75 0.65-0.82 173-875 21-33 108-158
42.5 3921
70
(Mainly
52% 69% 65% 60% 67% 48% 65%
Polymer)
Note: Percentages represent dataset (project) distributions
3.2.3.2. Incremental recovery. In the fifth and final stage, another new EOR
criterion is established that is based on reservoir properties that achieve the highest
incremental recovery for each EOR method. Table 3.3 focuses on three EOR methods
that have the largest dataset, or number of projects implemented. Therefore, an adequate
amount of data is provided to capture incremental recovery and correlate it with reservoir
properties, as shown in Table 3.3.
52
Table 3.3. New EOR Criteria Based on Incremental Recovery
Reservoir Properties
Reservoir
Properties Miscible
Miscible CO2 Thermal (Steam)
Hydrocarbon
6-12 (327,182), 12-18

30-36 (137,413) 24-30 (116,500)
API (846,065)
36-42 (112,117) 36-42 (144,088) 18-24 (264,804)
Viscosity 242-484 (202,692), 3872-

0-10 (264,304) 0-10, (375,174)
(cP) 4114 (263,996)
0.5-0.6 (477,540), 0.6-0.7
0.3-0.4 (66,352) 0.8-0.9
Start Oil (602,737)
Saturation 0.7-0.8 (147,848) , 0.8-0.9
0.4-0.5 (88,415) -204,483
(197,083)
1500-2000 (445,451), 2000-
Permeability 2500 (226,337)
0-20 (180,979) 1000-1020, (128,400)
(mD) 3000-3500 (117,184), 4000-
4500 (264,406 )
25-30 (123,203), 30-35
Porosity (%) 10-15 (141,771) 20-25 (239,676) (915,595)
35-40 (368,345)
8000-10000
0-2000 (1,137,316)
Depth (ft) 4000-6000 (169,770) (113,593)
10000 > (187,623) 2000-4000 (258,601)
Note: Figures in parentheses represent enhanced production in BPD
Tables 3.1 and 3.2 are comparable and generally indicate good agreement. The
approach of providing both single and double-layer criteria increases the confidence level
in EOR implementation. The first layer involves an EOR criterion based on project
distributions for each of the reservoir properties and for each EOR method. This is then
contrasted with reservoir properties that achieved the highest incremental recovery.
EOR criteria based on project distributions provide a better representation than
average values, which can be misleading in various instances, as Figures 3.10 through
53
3.17 revealed. Furthermore, basing EOR criteria on incremental recovery further refines
them, thus helping to ensure EOR development and implementation.
3.3. UPDATED ENHANCED OIL RECOVERY GUIDELINES
Traditionally, EOR methods have been classified into four major categories: gas,
thermal, chemical, and other (Green and Wilhite, 1998). A new classification of
principle EOR categories is proposed, which includes gas-based, water-based, thermal-
based, other, and combination methods. This new classification introduces the water-
based methods to replace the chemical methods because of two promising technologies
included in this category, LSWF and wettability alteration. It also introduces
combination methods that involve any two major EOR methods because such methods
break through the limitations of single-mechanism methods. Figure 3.18 illustrates the
various EOR methods.
Tables 3.3 through 3.5 summarize advances in EOR technologies primarily based
on SPE conference proceedings from 2007 through 2009. They list EOR limitations
reported by Taber et al. in 1996 and developments in EOR technologies that either break
through previous limitations or result in favorable oil recovery conditions. An overview
of a new EOR method is provided, followed by a tabulation that describes the new
method, as well as its mechanism(s) and limitations/challenges, as shown in Tables 3.6
through 3.11.
The proposed reclassification of EOR methods shown in Figure 3.18 is intended
to capture new EOR technologies and to encourage future EOR project reporting. In
addition, EOR methods can be used to combinations to overcome the limitation(s) of any
one EOR method, improve incremental recovery and/or reduce cost. Recent
developments in EOR technologies summarized in Tables 3.6 through 3.11 are intended
to outline the increase in EOR reservoir candidacy. Therefore, operators considering
EOR implementation are now informed as to new EOR technologies, combination of
EOR methods and recent developments in EOR research. The aforementioned EOR
guidelines provide an update to Taber et al. 1996 work and when used in conjunction
with the updated EOR selection criteria and/or new EOR screening tools provides an
impetus for augmenting EOR implementation.
54
Heavy Oil
Carbonate Reservoir CO2
Heavy Oil and Carbonate Reservoir Acidic Gases Sour
Miscible Hydrocarbon Gases Gas
Nitrogen
Gas EOR CO2
Acidic Gases Sour
Hydrocarbon Gases Gas
Immiscible
Nitrogen
Alkaline
Flue Gas
Surfactant
Gel Treatment
Polymer
Polymer Flooding
Water-Based EOR Micellar
Low Salinity
Water
Gravity Drive
Imbibition
Counter-Current
Cyclic
Steam Drive
SAGD
In-Fill Drilling EOR Methods
Dry
Thermal EOR Wet

Forward
Air Injection
In Situ Combustion Reverse
Catalyst
Hot Water Air Injection
Electrical Heating
MEOR
Others
Modified Enzyme
WAG SWAG
CO2 + WAG
Alkaline Surfactant Polymer
(ASP)
Alkaline + Surfactant (AS)
Combination EOR Miscible CO2 + SAGD
ASP + CO2
Conformance Control + CO2
Foam
Foam + Miscible CO2
Conformance Control + ASP
Figure 3.18. Enhanced Oil Recovery Methods

55
Table 3.4. Gas-Based EOR Methods

Limitation(s): (Taber and colleagues 1996)
A steep-dipping reservoir is preferred to permit some stabilization of the displacing front.
Reservoir Application
Property Studies Pilot Commercial
SN Advances in EOR Technologies
(Oil /
Lithology)
In-fill drilling extends the production
plateau and improves existing
(miscible CO2 flooding) and future
SPE
(miscible WAG) EOR methods
Light / 108060 SPE
because the displacing front remains
1 Sandstone SPE 114199
stabilized in short distances (Lopez
106575
and colleagues 2007; Sahin and
colleagues 2007; Xu and colleagues
2008).
Combinations of water-based EOR methods with gas EOR methods to overcome volumetric
sweep efficiency limitations
WAG is used to overcome the
inherently unfavorable gas injection
Light and
mobility ratios, which result in poor SPE
Heavy /
sweep efficiencies (Ning and 89353 SPE
2 Sandstone SPE 113933
colleagues 2004; Surguchev and SPE 106575
and
colleagues 1992; Sahin and 25075
Carbonate
colleagues 2007; Shi and colleagues
2008).
Modified WAG methods include
simultaneous water and gas injection
SPE
(SWAG) and a modified SWAG
105071
3 technique in which water is injected
SPE
on top of the reservoir and gas is
124197
injected at the bottom to improve
sweep efficiency.
SAG foam is injected into the
reservoir by alternating slugs of SPE
surfactant solution and gas injection, 114800
thus improving the mobility ratio and Light / SPE
sweep efficiency by decreasing the Sandstone 110408
4
gas velocity and plugging high- and SPE
permeability zones (Liu and Carbonate 113370
colleagues 2008; Renkema and OTC
Rossen, 2007; Le and colleagues 19787
2008; Yin and colleagues 2009).
ASPs are co-injected with CO2 to
SPE
5 enhance WAG flooding (Behzadi
123866
and Towler, 2009).
56
Table 3.4. Gas-Based EOR Methods (Continued)

Conformance control is achieved by
applying gel treatment to improve
SPE
6 CO2 flooding sweep efficiency SPE 35361
35379
(Asghari and colleagues 1996;
Fullbright and colleagues 1996).
Table 3.5. Chemical-Based EOR Methods

Limitation(s) : (Taber and colleagues 1996)
(a) An areal sweep efficiency of at least 50% for water flooding is desired.
(b) A relatively homogenous formation is preferred.
(c) Formation chlorides should be < 20,000 ppm, and divalent ions (Ca++ and Mg++) should be
< 500 ppm.
(d) Where the rock’s permeability is < 50 md, the polymer may only sweep fractures effectively
when the molecular weight is reduced.
Property Studies Pilot Commercial
(Oil /
Lithology)
Polymers
Polymer flooding is applied successfully
SPE
in heterogeneous and low-permeability
Light / 107727
1 reservoirs when combined with in-fill
Sandstone SPE
drilling (da Silva and colleagues 2007;
108661
Cheng and colleagues 2007).
High-molecular-weight polymers (18-20
million daltons) exhibit high viscosities
at salinities up to 170,000 ppm. For high SPE
2
concentrations of calcium, copolymers 113845
and AMPS can be considered (Levitt and
Pope, 2008).
Viscous oil is displaced by associative
SPE
3 polymer solutions (Buchgraber and
122400
colleagues 2009).
SPE
Hydrophobically associating polymer is
Heavy/ 104432
4 tolerant to salts (Han and colleagues
Sandstone IPTC
2006; Wei and colleagues 2007).
11635
The visco-elastic property of a polymer
SPE
5 can reduce the residual oil saturation
106005
(Lakatos and colleagues 2007).
Gel treatment
57
Table 3.5. Chemical-Based EOR Methods (Continued)

Apply an organically cross-linked
polymer that can withstand high SPE
6
temperature conformance treatment up to 121143
350○F (Mercado and colleagues 2009).
In the absence of divalent cations,
HPAM can remain stable with at least
SPE
7 half the initial viscosity for over 7 years
121460
at 100○C and about 2 years at 120○C
(Seright and colleagues 2009).
In-depth conformance control can be
achieved by injecting low-viscosity, pH-
Sandstone
triggered polymers into the reservoir to SPE
8 and
block swept fractures and high- 124773
Carbonate
permeability zones (Lalehrokh and
Bryant, 2009).
Liu and
Preformed particle gel (PPG) is used for
colleagues
9 large-volume conformance-control Sandstone
2010
treatments.
Surfactants
Super- and viscoelastic surfactants
Light /
provide both IFT reduction and mobility SPE
Sandstone
control over a wide temperature and 106005
10 and
pressure range (Lakatos and colleagues
Carbonate
2007).
The cost of chemical surfactant always
has been a drawback in actual field Light /
application. The use of agricultural Sandstone SPE
11
effluent to generate biological surfactants and 106078
provides a possible low-cost alternative Carbonate
(Johnson and colleagues 2007).
Alkaline, Alkaline-Surfactant (AS), ASP
Apply alkaline-surfactant (AS) flooding
in a high-temperature (119 °C) and high- Light / SPE
12
salinity environment (Othman and Sandstone 109033
colleagues 2007).
Adverse effects of alkaline injection are
SPE
mitigated by using organic alkaline,
107776
ceramic coatings on progressing cavity
Light / SPE SPE
13 pumps (PCP), and weak ASP systems
Sandstone 109165 104416
(Gang and colleagues 2007; Guerra and
SPE
colleagues 2007; Xinde and colleagues
114348
2006; Li and colleagues 2008).
Olefin sulfonates, when used with
appropriate co-surfactants, co-solvents,
Heavy / SPE
14 and alkali, yield results required for near-
Sandstone 113432
100% oil recovery in cores (Zhao and
colleagues 2008).
58
Table 3.6. Thermal-Based EOR Methods

Limitation(s): (Taber and colleagues 1996)
(a) Combustion sustainability is a limitation.
(b) Porosity must be high to minimize heat losses in the rock matrix.
(c) Sweep efficiency is poor in thick formations.
(d) Steam injection is limited to shallow reservoirs with thick (20 ft) pay zones to limit heat loss.
(e) Steam injection has a high cost per incremental barrel and thus is not used for carbonate
reservoirs.
Property
(Oil / Studies Pilot Commercial
Lithology)
Steam-assisted gravity drainage (SAGD)
follow-up to cyclic steam stimulation
improves the daily oil production capacity Heavy / SPE
1 of a single horizontal well from the initial Sandstone 104406
20-40 t/d to 70-80 t/d (Li-qiang and
colleagues 2006).
Redevelopment of an abandoned oil field
with SAGD. The power plant generates
steam and delivers surplus electricity to
the national power grid. Wastewater from Light / IPTC
2
a nearby sewage plant is used to produce Sandstone 11700
boiler feed water, and SAGD is expected
to deliver more than 100 million bbls of
oil (Jelgersma and colleagues 2007).
Property
(Oil / Studies Pilot Commercial
Lithology)
Steam injection is used to improve
recovery of a mature water-flooding
reservoir. Steam overriding can improve Light / SPE
3
vertical sweep efficiency and therefore Sandstone 116549
enhance recovery to 50% from 14%
(Shuhong and colleagues 2008).
Combinations of water-based EOR methods with thermal EOR methods to overcome

volumetric sweep efficiency limitations
A thermo reversible gel-forming system

improves the efficiency of cyclic steam
treatments. Steam injection conformance
is achieved because gelation occurs at Heavy / SPE
4
high temperatures. During oil drainage, Sandstone 104330
the reservoir’s temperature decreases, and
the gel converts into liquid (Altunina and
colleagues 2006).
59
Table 3.6. Thermal-Based EOR Methods (Continued)

A high-temperature slag-blocking agent
(with silicate as the main component) is
Heavy / SPE
5 developed for steam injection to plug gas-
Sandstone 104426
channeling paths and improve sweep
efficiency (Pan and colleagues 2006).
Chemical additives, including “SEPA,”
are incorporated into the steam to improve SPE
the efficiency of the stimulation process 108398
6 Heavy
and increase oil recovery factors SPE
(Belandria and Balza, 2006; Jianjun and 104404
colleagues 2006).
Catalysts are used for in-situ combustion
in carbonate reservoirs to generate a faster
Heavy / SPE
7 combustion front, higher combustion
Carbonate 107946
efficiency and higher initial temperatures
(Ramirez-Garnica and colleagues 2007).
Applied reaction technology uses
molybdenum oleate in steam stimulation Heavy / SPE
8
to effectively reduce oil viscosity Sandstone 106180
(Shoubin and colleagues 2007).
The vaporized solvent, when co-injected
with steam, condenses and mixes with oil,
SPE
9 creating a zone of low viscosity between Heavy
122078
the steam and heavy oil (Galvo and
colleagues 2009).
Combinations of gas-based EOR methods with thermal EOR methods to overcome steam
generation drawbacks and improve recovery
Non-thermal processes involving CO2
flooding are used in combination with
SPE
10 steam for oil recovery to limit the Heavy
113234
drawbacks of steam generation (Bagci and
colleagues 2008).
SAGD performance is improved by air SPE
injection and gas-assisted gravity drainage 106901
11 Heavy
(GAGD) (Belgrave and colleague 2007; SPE
Mahmoud and Rao, 2007). 110132
3.3.1. Low-Salinity Water Flooding. Low-salinity water flooding (LSWF) is a

new technology developed about a decade ago to improve oil recovery. Experiments
conducted by Tang and colleagues (1997) concluded that a decrease in salinity favorably
altered wettability and improved spontaneous imbibition and oil recovery by water
flooding. Alotaibi and colleagues (2009) concluded that “optimal salinity should be
60
maintained to maximize oil recovery” (p. 6). This concept was previously highlighted by
Surkalo (1990), who stated that “the effectiveness in reducing the interfacial tension
depends on where the surfactant forms. If the salinity is high or low the surfactant forms
away from the oil-water interface” (p. 6).
The mechanism of LSWF oil recovery remains unclear, despite several theories
(Boussour and colleagues 2009). Karoussi and Hamouda (2007) argue that oil recovery
by spontaneous imbibition does not depend exclusively on the imbibing fluid
composition, but relies also on the composition of the initial reservoir fluid. Akin and
colleagues (2009) note that favorable wettability alterations have been associated with
increased recovery temperatures, whereas Strandand colleagues (2005) associated
favorable wettability alteration in chalks with sulfate concentrations in the injection fluid
and with temperature. Using an injection fluid high in sulfate ions, such as seawater, to
favorably alter wettability may adversely affect the reservoir’s permeability if the
formation water contains high concentrations of barium and strontium ions. Carageorgos
and colleagues (2009) explained the adverse effects on permeability of injection fluid that
is incompatible with the formation water.
Table 3.7. Low-Salinity Water Flooding (LSWF)

Description
Decreasing the injection water’s salinity by reducing the total suspended solids (TSD) has been
proved to increase oil recovery.
Mechanisms
 Favorable wettability alteration in sandstone cores occurs when the injection water’s TSD is
reduced below 6,000 ppm (Tang and Morrow, 1997; British Petroleum PLC, 2009).
 IFT is reduced in carbonate cores when the injection water’s TSD is reduced from 214,943
ppm to 52,346 ppm (Okasha and colleagues 2009).
Limitations and Challenges

 The mechanism of LSWF oil recovery remains unclear, despite several theories (Boussour and
colleagues 2009).
 The availability of low-salinity water sources is a limiting factor in LSWF’s application.
 Maximized oil recovery during LSFW requires optimal salinity (Alotaibi and Nasr-El-Din,
2009; Surkalo, 1990) to effectively alter wettability without decreasing reservoir permeability
(British Petroleum PLC, 2009; Tang and Morrow, 1997).
61
3.3.2. Water-Alternating Gas. WAG alternates water flooding and gas flooding
to stabilize the displacement front. The breakdown of the water-alternating gas interface
primarily due to gravity segregation or low injection pressures offsets the favorable
mobility ratio and degrades the sweep efficiency. The critical design parameters in WAG
are timing and the water-to-gas ratio. If excessive water is used or flooding is prolonged,
capillary trapping occurs, and solvent-oil banks are broken. In the opposite case in which
inadequate quantities of water and short alternating durations are used, gas channeling
occurs, and an unfavorable mobility ratio degrades sweep efficiency. Therefore, well
spacing, injection pressure, and reservoir permeability variations are key WAG candidate
selection criteria. Reservoir simulation should be used to determine the optimal WAG
design parameters.
Table 3.8. Water-Alternating Gas Flooding

Description
WAG is a process of injecting gas as a slug alternately with a water slug to overcome the
inherently unfavorable gas injection mobility ratios (Dong and colleagues 2002).
Mechanisms
Ning and McGuire (2004) state, “Immiscible WAG flooding in saturated or near-saturated
reservoirs results in incremental recovery mainly due to an improvement in sweep efficiency. By
contrast, immiscible WAG flooding in under-saturated reservoirs results in incremental recovery
due to a reduction in oil viscosity and oil swelling” (p. 3). Miscible WAG flooding in suitable
candidate reservoirs results in incremental recovery due to a reduction in IFT and improvement in
sweep efficiency (Surguchev and colleagues 1992).
Applied Parameter Ranges
Miscible WAG Immiscible WAG

Number Of Projects Reported: 3 Number of Projects Reported: 11
Oil Properties Oil Properties
API Gravity Range: 33-39 API Gravity Range: 9.3 – 41
Oil Viscosity (cP): 0.3-0.9 Oil Viscosity (cP): 0.17- 16000
Reservoir Properties Reservoir Properties

Porosity: 11-24% Porosity: 18-31.9 %
Permeability (md): 130-2000 Permeability (md): 100-6600
Depth (feet): 7545-8887 Depth (feet): 2650-9090
62
Table 3.8. Water-Alternating Gas Flooding (Continued)

Stone (1982) states, “WAG is often limited by vertical gravity segregation, which causes the
injected gas to rise and the injected water to migrate to the bottom of the formation” (p. 2). This
limitation can be mitigated by using high injection rates or reduced well spacing (Stone, 1982).
Gorell (1988) suggests that “Mobile water may shield in place oil from contact with the injected
solvent” (p. 227). Therefore, the water slug’s size is critical in order to maintain an optimum
balance between reducing the oil’s IFT and improving the sweep efficiency (Gorell, 1988).
3.3.3. Surfactant Imbibition. Surfactants can be used to lower IFT during water
flooding (Calhoun and colleagues 1951; Dunning and Hsiso, 1953). However, recent
work has uncovered that surfactants favorably alter wettability in oil-wet reservoirs.
Tests reported by Flumerfelt and colleagues (1993) regarding the surfactant-based
imbibition/solution drive process for single-well treatment in low-permeability, fractured
environments demonstrate that “the surfactant appears to alter the wetting state of the
rock and promote imbibition significantly beyond that possible with water alone or water
with dissolved CO2” (p. 67).
Babadagli (2003) conducted an analysis of oil recovery by spontaneous imbibition
of surfactant solution on a variety of rock types, including sandstone, limestone,
dolomitic limestone and chalk; he concluded that “for some rock samples the imbibition
recovery by surfactant solution was strictly controlled by the surfactant concentration” (p.
1). However, the difference in the recovery rate and ultimate recovery rate between high
and low IFT samples can also be affected by wettability alteration and adsorption that can
vary with rock type.
Table 3.9. Surfactant Imbibition
Description
Oil is recovered from fractured carbonate reservoirs by wettability alteration with surfactants
(Gupta and Mohanty, 2008).
63
Table 3.9. Surfactant Imbibition (Continued)

Mechanisms
 Cationic surfactants can recover oil from chalk cores by spontaneous counter-current
imbibition (Austad and Milter, 1997).
 Anionic and non-ionic surfactants at low concentrations (<0.1 wt%) can improve oil recovery
up to 60% OOIP in carbonate cores. This is a gravity drainage process (Gupta and colleagues
2008).
 Incremental recovery by a counter-current imbibition process depends on brine pH
(Takahashi and Kovscek, 2009).
 Spontaneous imbibition in carbonate reservoirs is very slow because it is a gravity-driven
process (Gupta and colleagues 2008).
 Surfactant treatment is difficult in carbonate reservoirs due to poor volumetric sweep
efficiency (Wang and colleagues 2008).
 Surfactants represent a significant cost.
3.3.4. In-Depth Conformance Control. In-depth conformance control began in

the late 1990s, when most oilfields had become mature and contained less remaining oil
near the wellbore, and an interlayer heterogeneity conflict dominated. A major drawback
of near-well-bore gel treatments is the necessity of displacing the fluid bypass and
reverse flow in the high-permeability zone; this effect occurs especially when small
quantities of plugging agents are used. As a result, the gel treatment has little effect on
water production and incremental oil recovery.
Conformance-control treatments improve sweep efficiency much more effectively
in large volumes, especially if the agents can be placed between wells. Chemicals
typically used in large-volume conformance-control treatments are polymer-crosslinker-
retarder weak gel systems, colloid dispersion gels (CDGs) and particle gels. The latter
include micro- to millimeter-sized preformed particle gels (PPGs), microgels (Zaitoun
and colleagues 2007) and the submicro-sized particle gel Bright water® (Frampton and
colleagues 2004). Polymer concentrations usually range from 800 to 2,000 mg/l for weak
gels and from 400 to 1,000 mg/l for CDGs. China has accumulated much experience in
the application of large-volume PPG treatments, which have been used for more than
2,000 wells (Liu and colleagues 2010).
64
Table 3.10. In-Depth Conformance Control

Description
Plugging agents such as weak bulk gels, colloid dispersion gels, and particle gels are injected
deep into the reservoir to divert injection water to un-swept hydrocarbon zones/areas. This
improves oil recovery and reduces water production.
Mechanisms
Correct the severe heterogeneity of a reservoir with cross-flow between layers to redistribute
water flow and thus improve sweep efficiency.
 Delivering a bulk gel or particle gel to target locations is a challenge (Levitt and Pope, 2008).
 Gel properties designed for in-situ crosslinking systems are difficult to control in formation
due to shearing, dilution and the chromatography effect of chemicals.
 Selecting a particle gel size and strength appropriate for a specific formation is a challenge
because no proper technologies exist to identify the size of channels/streaks, a process
necessary for an in-depth conformance-control design.
 A cost-effective gel system for a high-temperature and high-salinity reservoir is not available.
3.3.5. Microbial Enhanced Oil Recovery. Microbes can be used to improve oil
recovery. Microbial EOR (MEOR) has always been an attractive EOR method due to its
low cost and potential to improve both microscopic and macroscopic displacement
efficiencies. However, the uncertainties, sensitivities and time constraints of biological
agents have always limited their success and application potential. Nevertheless, MEOR
has introduced the use of organic substitutes for chemical EOR methods; these include
alkaline (Guerra and colleagues 2007), surfactants (Kurawle and colleagues 2009) and
polymers (Jiecheng and colleagues 2007; Sugai and colleagues 2007). In addition,
MEOR continues to achieve success in some field applications (Town and colleagues
2009).
Table 3.11. Microbial Enhanced Oil Recovery (MEOR)

Description
Microorganisms and nutrients are injected into the reservoir, where the microorganism(s)
multiply and their metabolic products, such as polymers, surfactants, gases, and acids, improve
oil recovery (Sugai and colleagues 2007).
65
Table 3.11. Microbial Enhanced Oil Recovery (MEOR) (Continued)

Mechanisms
 Increased reservoir pressure as a result of microbial gas generation.
 Reduced oil viscosity.
 Modified permeability due to acidic dissolution or plugging.
 Decreased IFT resulting from microbial biosurfactant generation and a reduced population of
sulfate-reducing bacteria.
Bacteria Functions in MEOR (Sen, 2008)
IFT Reducers Conformance Viscosity Reducers Permeability Paraffin
(Surfactant) (Polymer) (Gas) (Solvent) Modifiers Deposition
(Acid) Reducers
Acinetobacter Bacillus Clostridium Clostridium Clostridium Pseudomonas
Arthrobacter Leuconostoc Enterobacter Zymomonas Enterobacter Arthrobacter
Bacillus Xanthomonas Desulfovibrio Klebsiella
Pseudomonas
Biosurfactants (Reported IFT Measurements Reference
in mN/m)
Mixed Culture 0.020 (Kowalewski and colleagues 2005)
Rhamnolipid 0.006 (Hung and Shreve, 2001)
Lipopeptide 0.080 (Makkar and Cameotra, 1999)
Surfactin
 The majority of successful MEOR projects have been applied to reservoirs with temperatures
below 55○ C (Goa and colleagues 2009).
 MEOR projects are suited for low-production-rate and high-water-cut reservoirs.
 In the past ten years, the success rate of MEOR projects has been about 60% (Goa and
colleagues 2009).
 Surfactant adsorption to the reservoir rock and biodegradation adversely impact MEOR’s
performance (Yeung and colleagues 2008).
3.3.6. Steam-Assisted Gravity Drainage. Horizontal wells achieved commercial

viability in the late 1980s (King, 1993). This milestone was preceded by the development
of steam-assisted gravity drainage (SAGD), which consists of two parallel horizontal
wells. The shallower well is injected with steam and, at times, solvent to mobilize the oil.
Gravity drains (Table 3.12) the oil to the bottom well for production. SAGD was
originally discovered by Dr. Roger Butler and proved commercially successful in 1992
(Edmunds, and colleagues 1994) Recent developments in SAGD include the use of
solvents (Galvo and colleagues 2009) and air (Belgrave and colleagues 2007) to
enhance oil recovery.
66
Table 3.12. Steam-Assisted Gravity Drainage (SAGD)

Description
Reis (1992) states, “Steam is injected into the formation through a horizontal well, and oil drains
into a separate, parallel, horizontal well located below the injection well” (p. 14).
Mechanisms
Steam injection reduces the oil’s viscosity and causes the oil to swell. The macroscopic
displacement is further improved by the density difference between the steam and the oil,
depending on flow regimes (Reis, 1992). The oil’s IFT could also decrease as a result of steam
distillation (Cadelle and colleagues 1981).

 Reservoir depth (Taber and Seright, 1996).
 Formation net thickness (Taber and Seright, 1996).
 Pay zone net thickness should be deep enough to drill two parallel horizontal wells, one
above the other.
 Albahlani and Babadagli (2008) state that SAGD is challenged by the “high vertical
permeability requirement and high energy consumption” (p. 1).
3.4. MISCIBLE CARBON DIOXIDE FLOODING

This section focuses primarily on selection criteria for CO2 Enhanced Oil
Recovery (EOR) and the dispersion modeling of high-pressure CO2 release, which is
presented in Appendix B, as these are critical in offsetting capital investments and
managing legal liabilities.
The latest EOR survey published by The Oil and Gas Journal (2010) indicates a
total of 153 active miscible CO2 projects worldwide. The US houses the majority (139)
of these EOR projects (http://www.eorcriteria.com; Aladasani and Bai, 2011). The
number of implemented miscible CO2 projects has grown steadily in the past three
decades, as outlined by the following project timeline: 1968-1980 (7 projects), 1980-1990
(52 projects), 1990-2000 (98 projects) and 2000-2009 (153 projects).
Enhanced production remained below 37,000 BPD from 1968 through 1982.
During 1983, enhanced production increased by 55.8% to 93,250 BPD and thereafter
steadily increased during the 1980s to 166,807 BPD. This production figure corresponds
to a total of 51 miscible CO2 projects, a number that continued to increase during the
67
1990s to 95. However, enhanced production only increased by 25.8% to 209,892 BPD.
Similarly, from 2000 to 2009, the number of miscible CO2 projects increased to 153, and
enhanced production increased by 40.51% to 294,924 BPD (Figure 3.19).
3.5. NEW SELECTION CRITERIA FOR CO2 EOR

The methodology used to establish the selection criteria for miscible carbon
dioxide flooding consists of four stages. The first stage involves representing the
distribution of reservoir properties for the 153 projects reported by The Oil and Gas
Journal (2010). The second stage involves examining correlations between the reservoir
properties, and in the third stage, a prediction model is generated for enhanced oil
production. Finally, a prediction model for incremental recovery in miscible CO2
projects is presented.
The dataset was distributed using JMP statistical software. JMP calculates
quantiles with integer products and fractional products using Equations 3 and 4,
respectively. Aladasani and Bai (2010) adopted the reservoir property distribution shown
in Figure 3.20 to derive the EOR selection criteria. Dataset distribution allows for the
identification of various quartiles and dataset clusters, thus highlighting the frequency
range of each reservoir property, shown in Table 3.13, in the implemented EOR projects.
( ) (3)
( ) ( ) (4)
Where:
γn dataset values
p percent value
n total number of entered values
f fractional value
68
Sum Of Projects Sum Of Enhanced Production (BPD)

Year Project Initiated
1984
1998
1968
1980
1982
1983
1985
1986
1988
1989
1993
1994
1995
1996
1996
1997
2000
2001
2004
2004
2005
2006
2007
2008
2009
2009
170 325000
160 300000
150
140 275000
130 250000
Enhanced Production (BPD)

120 225000
Number Of Projects
110
100 200000
90 175000
80 150000
70 125000
60
50 100000
40 75000
30 50000
20
10 25000
0 0
Figure 3.19. Miscible CO2 Projects and Enhanced Production Trends

Source: Aladasani and Bai (2011) http://www.eorcriteria.com
Figure 3.20. Distribution of Miscible CO2 Reservoir Properties

69
Table 3.13. Summary Distribution of Miscible CO2 Reservoir Properties
Oil Oil
Quantile Gravit Viscosit Porosit Permeabilit Dept Temperatu
Saturatio Saturatio
s (%) y (No.) y (cP) y (%) y (mD) h (ft) re (F)
n (Start) n (End)
1040
90 42 2.31 25 0.6 0.428 296.7 227.7
0
75 40 2 20 0.52 0.36 100 7290 163.25
50 38 1 13 0.45 0.28 30 5475 111.5
25 34 0.995 10 0.38 0.23 5 4900 104
10 31.3 0.509 9 0.35 0.13 4 4060 98.1
The range of reservoir properties from the 25% to 75% quartiles indicates where
50% of the project datasets lie (highlighted in yellow), and the range of reservoir
properties from the 10% to 90% quantiles indicates where 80% of the project datasets lie
(highlighted in green). These values outline the lower and upper limits for each reservoir
property. Dataset distributions also identify outliers or skewed datasets, which is
important in generating a prediction model.
The second stage of the statistical analysis is to evaluate correlations amongst the
reservoir properties. Correlation measurements are required to screen reservoir
properties pertinent to miscible CO2 projects, which is critical for improving the accuracy
of the regression curve and the confidence level of the prediction model. The Restricted
Maximum Likelihood (REML) method was used to examine the relationships between
the reservoir properties because it uses the entire dataset, even missing cells. The entire
database, consisting of 153 miscible CO2 projects, is fed into JMP, and the correlations
are generated, as shown in Figure 3.21.
70
Figure 3.21. Initial Reservoir Property Correlations
The results in Figures 3.22 and 3.23 indicate that the highest correlation values
are for depth/temperature, porosity/permeability and porosity/temperature, while
viscosity generally has the lowest correlations, specifically against permeability, API and
recovery. Therefore, viscosity is removed from the database in view of its negative
impact on the prediction model. In addition, total oil production and enhanced oil
production initially were removed from the database because it is difficult to benchmark
project production figures without considering the number of wells and production
history. Instead, recovery as a percentage of OOIP is used based on the start and end oil
saturations, leaving the PV of injected CO2 unknown.
In this study, it is assumed that the number of wells and the quantity of injected
CO2 are proportional, which is a limitation that cannot be avoided at this stage. The
71
number of projects or datasets decreases from 153 to 94 due to the non-availability of

start or end oil saturations.
Figure 3.22. Selected Reservoir Property Correlations

72
Figure 3.23. High and Low Correlations

73
The third stage of the work is to generate a prediction model for miscible CO2
projects. The quality of the prediction model is based on the accuracy of the multiple-
variable regression curve and the confidence level. Generally, a tradeoff is required
between these two values because the regression curve improves when the number of
reservoir properties and the factorial degree increases, whereas the confidence limits
decrease with outliers/skewed datasets and reservoir properties with low correlations.
Nevertheless, the prediction model can be optimized by removing skewed datasets from
reservoir properties with the highest correlations and lowering the factorial degree to
decrease the standard error proportionally. The accuracy of the regression curve is
measured by the term RSquare in JMP.
The model is generated based on effects and roles; in this case, reservoir variables
represent the model effects, and recovery represents the model outcome. The small
number of datasets in this example limits the factorial degree. In addition, the weak
correlations of some reservoir properties, as shown in Figure 3.22, also serve as a limiting
factor in the model effects because weak correlations increase the magnitude of standard
error in the constructed model. The model effect should emphasize reservoir properties
with strong correlations, especially when higher-degree factorials are used and
outliers/skewed datasets are removed. The multivariable regression curve and prediction
model for miscible CO2 projects is presented in Figure 3.24.
Recovery (%OOIP) in Figure 3.24 represents the percentage of OOIP based on
the difference between the end and start oil saturations. The results in Figure 3.24
indicate that the best confidence limits are achieved for a recovery (OOIP%) of 22.40 that
corresponds to an API of 35.55, a porosity of 11.83, a start oil saturation of 0.5453, a
permeability of 20 mD, a depth of 5780 ft and a temperature of 116.3 oF. The recovery
rate is proportional to API, porosity, start oil saturation and depth. Based on existing CO2
project data, temperature and permeability are inversely proportional to recovery rates.
The prediction expression is shown in Appendix H. The following points should be
noted about the prediction model presented in Figure 3.24: (1) It is based on the reported
miscible CO2 projects; (2) It does not consider the number of wells or the injected PV of
CO2; and (3) The factorial degree is limited to 2, thus implying that reservoir properties
(model effects) are only considered for two variables at a time. This was necessary to
74
avoid poor confidence limits. Nevertheless, the EOR prediction model shows potential
and could be improved in the future.
Figure 3.24. Multiple Variable Curve and Prediction Model

75
4. MODIFIED WATER SALINITY
In previous research, compositional numerical simulators have been introduced to

simulate various EOR methods because they have the ability to consider mass
components for each of the three phases in the continuum equation. The challenge that
remains is how best to formulate fluid relative permeability functions. Existing
correlations have inherent limitations, which are further amplified by the fact that the
initial residual saturations are not constant because LSWF recovery is driven by
wettability modification. Overcoming these limitations to successfully formulate fluid
relative permeability functions requires a simulator that is not only validated analytically
for the transportation of mass components in the liquid phase but that is validated for all
wetting conditions in order to prove that the numerical solution is able to deliver
satisfactory results.
This chapter presents the development of a numerical model for LSWF and
LSSWF in multi-dimensional, porous and fractured reservoirs. The model is validated,
and its supporting equations are presented.
4.1. LOW-SALINITY WATER-FLOODING MODEL

A numerical model for LSWF and LSSWF in multi-dimensional, porous and
fractured reservoirs is presented. The model’s development involves six stages, as shown
in Figure 4.1. The model’s formulation incorporates known improved oil recovery (IOR)
mechanisms for simulating LSWF and LSSWF processes.
3.2.4 3.2.5
3.2.1 3.2.2 3.2.3 3.2.6
Initial and Matrix and
Governing Constitutive Numerical Immobile
Boundary Fracture
Equations Relations Formulation Water Zone
Conditions Interations
Figure 4.1. Low-Salinity Model Development

76
The numerical model requires the three validation stages shown in Figure 4.2.
Initially, the transportation of a mass component in an aqueous phase is validated with
published analytical solutions. Subsequently, published core-flooding experiments are
used to validate the model’s formulation for known IOR mechanisms. The third stage of
validation involves up-scaling the model to three dimensions and comparing published
five-spot well pattern results with both conventional water flooding and LSWF
simulation runs.
1-Dimension Sandstone Water Flooding
Analytical Core Flooding

Scaling-Up
Solution Experiments
2-Dimenions Carbonate LSWF
Figure 4.2. Low Salinity Model Validation
4.1.1. Governing Equations. Reservoir simulation is based on the law of

conservation, constitutive equations and equations of state. The reservoir is considered a
controlled volume containing three phases and various mass components. The saturation
occupied by each phase in the porous media represents the fractional phase volume.
Therefore, using material balance equations, the mass component in the gas, oil and water
phases can be derived. The fluid flow in a reservoir can be expressed as shown in
Equation 5. Constitutive equations are needed to determine the phase pressure and
relative permeability, which is achieved by relating the phase, saturations and mass
components (Equations 12 and 13). As a result, it is possible to derive the capillary
pressure and relative permeability expressions as a function of phase saturations and mass
component fractions (Equations 14 through 20). The equation of state describes phase
77
density or viscosity as a function of temperature and pressure; this is represented by the

phase formation factor (Equations 21 through 28).
( ) ( ) ( )
( ) ( )
  ( )
Expanding Equation 1 to represent the oil phase produces the following flow
equation:
( ) ( ) ( )
The oil phase is present only in its associative state, whereas the gas phase is present in
both its associative state and when dissolved in oil. Therefore, gas volume is a function of both
gas and oil saturation, in addition to gas density and dissolved gas density, respectively.
[( ) ( )] [( ) ( )] ( )
The water phase has two mass components, water and salt. To account only for
the water component in the water phase, the following expression is generated (Equation
10). The constitutive equation mandates that the mass components of the entire phase
equal unity.
( ) ( ) ( )
78
In LSWF, salt is considered a mass component in the water phase, which is

expressed by the product of the reservoir’s porosity, water saturation, water density and
salt mass component; as such, salt is transported by advection. Additionally, because the
salt mass component in the water phase is transported by diffusion and because, in
sandstone reservoirs, cations are prone to adsorption on the reservoir rock, an expression
is required to differentiate the fate of adsorbed salt and salt transported by diffusion
(based on Equation 13). A tortuosity term is added to account for increases in the
distance that molecules must travel in a porous media.
[( ) ( )]
[( ) ( ) ] ( )
Constitutive equations are needed to determine the phase pressures, saturations

and phase relative permeabilities; this is achieved by relating the phase, saturations and
mass components. The sum of the saturations of hydrocarbon phases equals unity, as
does the sum of the mass components in any phase.
( )
( )
The phase pressure is, by definition, the difference between the non-wetting phase
and the wetting phase. The non-wetting phase always has a higher pressure than the
wetting phase, and gas is always the non-wetting phase in hydrocarbon reservoir rocks
(Satter and colleagues 2008). The three-phase capillary pressures between the oil and
gas interface and between the water and oil interface are shown in Equations 14 and 15,
respectively. The water phase consists of two mass components, so both mass fractions
are a function of water-oil capillary pressure. This relationship makes it possible to
79
consider the effects of LSWF on capillary pressure. In addition, capillary pressure

correlations, such as in Parker et al. (1987), do not consider IFT parameters in the
capillary function. Therefore, a J-function can be used to relate both IFT and contact
angle changes occurring as a result of LSWF, as shown in Equation 16.
Po  Pg  Pcgo S w , S o  (14)
Po  Pw  Pcow (Sw , So , Xc ) (15)
Pcow    X  cos  ( X ) Pcow

0
(S w , So ) (16)
By definition, the relative permeabilities are functions of the saturations

occupying the porous media and also should include the phase mass components, as
shown in Equations 17 through 19. The Stone correlation, method II (Aziz and Settari,
1979), can be used if no three-phase relative permeability data are available, as shown in
Equation 20. This correlation provides three-phase relative permeability data based on
two sets of two-phase flow relative permeabilities.
k rg k rg Sg , Xc  (17)
k ro k ro Sw , Sg , Xc  (18)
k rw k rw Sw , Xc  (19)
 k wo  k og  
k r o k *r owo  r o *wo  k r w  r o *wo  k r g   k r w  k r g  (20)
 k ro  k ro  
80
The equation of state describes phase density as a function of temperature and

pressure; this is represented by the phase formation factor shown in Equations 21 and 23.
The water phase density is a function of temperature, pressure and the salt mass
component, as shown in Equation 26. Gas and oil viscosities are treated as functions of
phase pressure only, and the water phase viscosity is a function of the salt mass component,
as shown in Equations 28 and 29. The water phase viscosity is a function of the salt mass
component used to evaluate the mobility ratio during LSWF.
 
g 
g STC
Bg
(21)
Where,
( )
o 
1
Bo

 o STC  Rs  g STC  (23)
Where,
[ ( )] ( )
R s  R s (Po , Pb ) (25)
 w  X c STC
w  (26)
Bw
Where,
Bwo
Bw 

1  Cw Pw  Pbo  (27)
81
   P  (28)
 w   w  Xc  (29)
4.1.2. Numerical Formulation. The developed flow, Equations 8 through 11

consist of continuous, nonlinear, three-dimensional and time derivatives for each
hydrocarbon phase and mass component. The continuity of the flow equations correctly
represents the reservoir because creating discontinuities impacts the accuracy of the
changes in phase saturations and pressures. Paradoxically, such discontinuities are
required in a deterministic approach; the alternative would be a countless number of
solutions.
Discretizing space and time provides a means by which to solve the partial
derivatives in the flow equations. This is possible because, with defined boundaries, the
derivatives can be integrated, and the specified initial and boundary conditions provide
explicit expressions to the integrated derivatives. The discretized space is represented by
grid blocks, and the finite difference method is used to approximate the solution of the
partial derivatives (Figure 4.3). The grid blocks also describe different portions of the
reservoir that have heterogeneous fluids and/or rock properties. In addition, they specify
the spatial phase flow at various paradigms within the reservoirs, including the wellbore,
fractures and matrix, as well as the interactions amongst those continuums (Ertekin and
colleagues 2001).
The flow equations (8 through 11) are expanded to include the reservoir’s fluid
and rock properties (Equations 22, 24, 27, 28 and 29. The flow equations also are
multiplied by the reservoir unit’s bulk volume to include the cross-sectional area
perpendicular to the flow regime. Transmissibility terms for advection and diffusion, in
addition to the mobility ratio, are added. The flow equations then are discretized in a
three-dimensional plane, as developed by Wu and Bai (2009).
82
i ,j + 3/2
Δ j + 1/2
Δ i + 1/2
i + 1/2 ,j
+1
i ,j + 1 i + 1, j + 1/2
Δ j + 1/2 Δ j + 1/2
Δ i – 1/2 Δ i + 1/2
i – 1/2, j i + 1/2, j
i,j
i + 1, j - 1/2
Figure 4.3. Grid Block X-Y Coordinates

Rag ,n1   So  dg   S g  g    S    S   Vt
n1
a o dg g
n
g a
a

   dg o  nab11/ 2 ab  onb1       
n1
oa   g g
n1
ab1/ 2 ab
n1
gb  gna1  Qgna1  (30)
ja ja

Rao,n1   So o  na1  So o  na Vt     
a n1
o o ab1 / 2 ab   onb1  ona1  Qoan1
ja
(31)

Riw,n 1   S w X w  w  na 1  S w X w  w  na Vt a

   w X w w  nab11 / 2 ab  wnb1   wna1  Qwna1  (32)
j a
83

Ric, n1   S w X c  w  1     R  w X c K d  na1  S w X c  w  1     R  w X c K d  na Vt
a
  
   w X c w  inj11 / 2 ab  wnb1  wna1    w S w  nab11 / 2 iab
D

X cnb1  X cna1  Qcna1
ja ja
(33)
Where,
k r
  (34)

Aab kab1 / 2
 ab  (35)
Da  Db
Aab   Dm ab1/ 2
 abD  (36)
Da  Db
 na1 Pna1  n ,ab1 1 / 2 g da (37)
Qn a1 qn a1Va (38)
The Newton-Raphson iteration is used to solve Equations 30 through 38 of a flow

system, representing 4×N coupled non-linear equations, which include four equations at
each element for the four mass balance equations of gas, water, oil and salt, respectively.
Four primary variables (x1, x2, x3, x4) are selected for each element; these are oil pressure,
oil saturation, saturation pressure (or gas saturation) and the mass fraction of salt. The
primary variables are selected in a manner similar to that of a black-oil reservoir
simulator. An automatic variable switching scheme handles the transition of free gas
appearing and disappearing during simulation studies of oil production with oil, gas and
84
water three-phase flow conditions. Three of the four primary variables are fixed, and the
third variable depends on the phase condition at a node. If there is no free gas, a node is
said to be under-saturated or above the bubble point, and the saturation pressure, Ps, is
used as the third primary variable. When free gas is present, a node is said to be saturated
or below the bubble point, in which case gas saturation, Sg, is the third primary variable.
This variable switching scheme very rigorously and efficiently handles variable bubble-
point problems, which often are encountered in reservoir simulations (Thomas and
Lumpkin, 1976). Numerical experiments show that the choice of different primary
variables makes a difference in numerical performance during nonlinear iterations of
solving a three-phase flow problem, and the best combination for handling phase
transitions under different capillary/phase conditions is the mixed formulation.
The Newtonian iteration process continues until the residuals, R kn , n 1 , or changes
in the primary variables,  x m,p1 , over an iteration fall below preset convergence
tolerances. Numerical methods are used to construct the Jacobian matrix for Equation
38, as outlined in Forsyth et al. (1995). At each Newtonian iteration, Equation 30 through
33 represents a set of 4×N linear equations for 4×N unknowns of xk,p+1 with sparse
unsymmetrical matrices that are solved by a linear iterative equation solver.
 Ra ,n1 xk , p 

R  ,n1
a x   R
k , p 1
 ,n1
a xk , p     x xk , p1  xk , p   0 (39)
k k
 Ra ,n1 xk , p 

k  xk
xk , p1    Ra ,n1 xk , p  (40)
Where,
x k ,p1  x k ,p1  x k ,p (41)

85
4.1.3. Handling Initial and Boundary Conditions. Starting a transient

simulation requires a set of initial conditions. In other words, a complete set of primary
variables must be specified for every grid block or node. A commonly used procedure
for specifying initial conditions is based on the gravity-capillary equilibrium calculation
initially and on the restart option for the subsequent simulations, in which a complete set
of initial conditions or primary unknowns is generated in a previous simulation, with
proper boundary conditions described.
Using a block-centered grid, first-type or Dirichlet boundary conditions are
treated with the inactive cell or big-volume method, as typically used in the TOUGH2
code (Pruess, 1991). In this method, a constant pressure/saturation/ concentration node is
specified as an inactive cell or with a huge volume, while all the other geometric
properties of the mesh remain unchanged. For flux-type boundary conditions or for more
general types of flux or mixed boundaries, such as multi-layered wells, general handling
procedures are implemented, as discussed by Wu and colleagues (1996; 2000).
4.1.4. Treatment of Fracture-Matrix Interaction. The mathematical model
presented used a dual-continuum methodology (Warren and Root, 1963; Pruess and
Narasimhan, 1985; Wu and Pruess, 1988) to handle multiphase flow in the fractured
porous medium. The dual-continuum method primarily distinguishes the matrix flow
from the fracture flow, and, depending on the fracture’s orientation, the flow interaction
between the matrix and fracture is decided. Flow paradigms are illustrated in Figure 4.4.
Dual Porosity/Permeability Dual Permeability

Dual Porosity
Matrix-matrix flow is permitted Global matrix-matrix flow
Global flow occurs only through the
only in the vertical axis occurs in all directions
fracture continuum
Figure 4.4. Flow Paradigms

86
Using the dual-continuum concept, Equations 8 through 11 can be used to

describe multiphase flow both in fractures and inside matrix blocks, as well as the
fracture-matrix interaction or flow in discrete fractures. However, special attention must
be paid to treating the fracture-matrix flow or transport. When handling flow and
transport through a fractured rock using the generalized numerical formation, fractured
media (including explicit fracture, dual, or multiple-continuum models) can be
considered special cases of unstructured grids (Pruess, 1991).
A large portion of the work of modeling the flow in fractured rock consists of
generating a mesh that represents both the fracture system and the matrix system under
consideration. Several fracture and matrix subgridding schemes exist for designing
different meshes for different fracture-matrix conceptual models (Pruess and Narasimhan,
1985; Pruess, 1983). Once a proper grid of a fracture-matrix system is generated,
fracture and matrix blocks are identified to represent fracture and matrix domains,
separately. Formally, they are treated identically for the solution in the model simulation.
However, physically consistent fracture and matrix properties, parameter weighting
schemes, and modeling conditions must be specified appropriately for both fracture and
matrix systems.
4.1.5. Handling Immobile Water Zones. As an example of applying the
generalized multi-continuum concept discussed above, immobile or residual water zones
of in-situ brine within porous pores can be handled as separate domains containing
immobile water only, such as “dead” pores, acting as additional continuums with zero
permeability. The salt within the immobile zones will interact with mobile water zones
by diffusion only. This diffusion process is described by the same governing equations
and numerical formulations discussed above as a special no-flow case.
4.1.6. Relative Permeability Functions. The model considered two relative
permeability and capillary pressure formulations. The Brooks-Corey function
(Honarpour and colleagues 1986) is used with the following modifications: (1) a
decrease in the relative permeability of the water phase as salinity decreases, and (2) an
increase in the relative permeability of the oil phase as salinity decreases (Equations 42
and 43). The Brooks-Corey exponential index  (Corey, 1954) is adopted, and two
normalized fluid saturations are described in Equations 44 and 45. Published IOR
87
mechanisms for LSWF include decreasing the residual oil saturation and modifying the
contact angle and IFT. Therefore, relative permeability and capillary pressure functions
are modified accordingly to include the effects of salinity. Jerauld and colleagues (2008)
proposed a linear relationship between salt concentration and residual oil saturation
(Equation 46); a second approach is based on correlations from sandstone core-flooding
experiments (Tang and Morrow, 1997). The latter formulation is extrapolated from core-
flooding experiments that use multiple salinity concentrations for sandstone reservoirs
(Equation 47).
  S ( X )
krw  S w
2
w c (42)
S  1  S  
2 
kro o w (43)
S w  S wr
Sw  (44)
1  S wr
S o  S or ( X c )
So  (45)
1  S wr
Xc  Xc1
Sor ( Xc )  Sor1  Sor1  Sor2 
Xc1  Xc 2 (46)
The residual oil saturation is considered a function of salinity in the aqueous

phase and, hence, a function of water’s relative permeability. Jerauld and colleagues
(2008) first proposed a linear relationship between the salt mass component and residual
oil saturation and treated salt mass concentration as a function of both oil and water’s
relative permeability, as defined in Equation 46. In this equation, Sor1 is the maximum
residual oil saturation at high salt mass fraction Xc1, and Sor2 is the minimum residual oil
saturation at low salt mass fraction Xc2.
(0.1083)( X c ) 2  (1.244)( X c )  (4.694e  8)

S or ( X c )  S or1  S or1  S or2  (47)
( X c )  0.1353
88
Carbonate core-flooding experiments using various salinity concentrations

(Yousef and colleagues 2010) suggest the following correlation between salinity and
residual oil saturation (Equation 48), salinity and contact angle (Equation 49) and salinity
and IFT (Equation 50):
(0.6494)( X c ) 2  (0.02624)( X c )  (7.486e  10)

S or ( X c )  S or1  S or1  S or2  (48)
( X c ) 2  (0.7334)( X c )  (0.3566)
(0.4106)( ) 2  (0.8696)( )  (0.006548)

 ( X c )   or1   or1   or2  (49)
( )  0.2763
(82.44)( IFT ) 2  (738.4)( IFT )  (24.55)

IFT ( X c )  IFTor1  IFTor1  IFTor2  (50)
( IFT )  843.8
4.1.7. Capillary Permeability Functions. Capillary pressure functions are

modified to include the effects of salinity. A linear relationship with residual oil
saturation is introduced between the salt mass fraction and contact angle so that a
decrease in the salt mass fraction would favorably alter wettability to intermediate
wetting conditions, as shown in Equation 51. In this equation, or1 is the contact angle at
high salt mass fraction Xc1, and or2 is the contact angle at low salt mass fraction Xc2.
The capillary pressure function from van Genuchten (1980) and Parker and
colleagues (1987) is used for the oil-water system, with the addition of the cosine of
contact angles of the oil and water phases on the rock’s surface to include the effect of
low salinity on the contact angle, as shown in Equation 52, where vG,  and  are
parameters of the van Genuchten functions (van Genuchten, 1980), with  = 1 – 1 / .
(Wu and Bai, 2009)
Xc  Xc1
( Xc )  or1  or1  or2  (51)
Xc1  Xc 2
89
Pcow 
cos  w  g

cos  w o   vG


 1  S w )


1 / 

1
1/ 
(52)
In cases in which the initial and final IFT values are known, the j-functional in
Equation 53 can be used.
( ) ( ) ( ) ⁄
{ } { } {( ̅ ) ⁄
} ( )
( )
A representation of relative permeability curves and capillary pressure curves is

required to visualize wettability, IFT and changes in residual saturations as a function of
salt concentration and water saturation during water flooding. Two key issues exist
concerning the generation of relative permeability curves and capillary pressure curves:
(1) the relative permeability and capillary pressure curves are snapshots of the
displacement front along the porous medium, and (2) the curve’s start and end points are
based on the initial and final wetting states. A water-wet reservoir is considered under
the following conditions. Initially, the reservoir is flooded with high-salinity water
(snapshot occurs after breakthrough). Figure 4.5 depicts the relative permeability curves.
Subsequently, the reservoir is flooded with low-salinity water. The inlet and outlet
represent the minimum and maximum concentrations, respectively, while the middle
region of the reservoir is assumed to have an average salinity value. Figure 4.6 shows the
relative permeability curves.
Finally, the reservoir is flooded entirely with low-salinity water (with both the
inlet and outlet at the minimum salt concentration). Figure 4.7 shows the relative
permeability curves.
90
0.45 0.12
Oil's Relative Permeability (Kro)
Water's Relative Permeability (Krw)

0.40
0.10
0.35
0.30 0.08
0.25
0.06
0.20
0.15 0.04
0.10
0.02
0.05
0.00 0.00
0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80
Water Saturation Kro Krw
Figure 4.5. High-Salinity Water-Flooding Permeability Curves
0.70 0.12

0.60
0.10
0.50
0.08
0.40
0.06
0.30
0.04
0.20
0.02
0.10
0.00 0.00
0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80
Water Saturation Kro Krw
Figure 4.6. High- to Low- Salinity Water-Flooding Permeability Curves

91
0.70 0.07
0.60 0.06

0.50 0.05
0.40 0.04
0.30 0.03
0.20 0.02
0.10 0.01
0.00 0.00
0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80
Water Saturation
Kro
Figure 4.7. Low-Salinity Water-Flooding Permeability Curves
Figures 4.6 and 4.7 indicate that the oil’s relative permeability increases rapidly
when LSWF is introduced to the water-flooding program, while the water’s relative
permeability remains unchanged despite the reduction in the reservoir’s salinity. Water
wettability decreases because Kro = Krw at 0.54 Sw compared to 0.56 Sw. Figure 4.7
indicates a significant decrease in water’s relative permeability by approximately 50%
when LSWF is exclusive, and water wetness increases so that Kro = Krw at 0.59 Sw.
Figure 4.8 shows the transition in fluid’s relative permeability during LSWF.
92
0.70 0.12
0.60
0.10
Water Relative Permeability (Krw)

0.50
0.08
Oil Relative Permeability (Kro)
0.40
0.06
0.30
0.04
0.20
0.02
0.10
0.00 0.00
0.30 0.40 0.50 0.60 0.70 0.80
Water Saturation
Kro(HSWF) Kro(LSWF) Krw(HSWF) Krw(LSWF)
Figure 4.8. Relative Permeability Transition during LSWF
An intermediately-wetted reservoir is considered, and it is assumed that the

wettability changes as a function of salinity, where the minimum salinity is assumed at
the intermediate wetting state. Figure 4.9 illustrates normalized imbibition and drainage
curves at the start of LSWF.
1.00
0.80
Normalized Capillary
Pressure (Pa)
0.60
0.40
0.20
0.00
0 0.2 0.4 0.6 0.8 1
Water Saturation Imbibition Drainage
Figure 4.9. Normalized Capillary Pressure Curves (Start of LSWF)

93
An oil-wet reservoir is considered, and it is assumed that the wettability changes

as a function of salinity, where the minimum salinity is assumed at the intermediate
wetting state. Figure 4.10 illustrates the changes in capillary pressure as the wettability
changes from oil-wet and water-wet to intermediate wetting as a result of LSWF.
Figure 4.10. Normalized Pc, Xc and θ Curves
Figure 4.10 indicates the change in capillary pressure as the contact angle
increases from a strong water-wet to a strong oil-wet state (zero to π) and as the salinity
increases from intermediate to strong wetting conditions (π/2 to 0) or strong oil-wet
conditions (π/2 to π). The normalized water saturation is used in a water-wet system to
determine the capillary pressure. When the wettability changes to oil-wet, the normalized
oil saturation is used to determine the capillary pressure.
94
4.1.8. Mesh Generation. The first step in conducting a numerical simulation is

to generate a mesh. The mesh can be designed with one- or two- dimensional radial
geometries and multi-dimensional rectangular coordinate systems, as shown in Figure
4.11.
Figure 4.11. Radial and Rectangle Mesh Geometries
In addition, the number and spacing of grid blocks is specified for each axis.
Radial meshes can be layered vertically or horizontally and have equal or unequal radii,
as shown in Figure 4.12. In the case of fractured models, the mesh is post processed to
generate fracture planes (shown previously in Figure 4.4) in the matrix. The
heterogeneity of the constructed model can be based on layered grid blocks in any two
planes. It is important to consider heterogeneities in the model’s properties prior to
constructing the mesh. Furthermore, the number of grid blocks should be increased to
compensate for core plugs and reservoir heterogeneities.
Figure 4.12. Horizontally-Layered Equal Radii and Vertically-Layered Unequal Radii

Mesh
95
4.1.9. Simulator Input File. The second step in conducting numerical

simulations is to populate the data file. The simulator data file consists of eight sections,
the first of which assigns rock properties to the generated mesh and specifies the type of
relative permeability and capillary pressure functions, as shown in Table 4.1. The second
section specifies the phase properties, notably, the density and viscosity values. The
change in water phase viscosity and density as a function of the mass component (salt) is
included. LSWF reduces the viscosity of the displacing fluid, which impacts horizontal
displacement efficiency. The third section of the data file relates to phase volume,
temperature data at various pressures.
Table 4.1. Rock Properties Input into the Simulator

Permeability
Initial Final Mass Final
Contact Porosity x-axis y-axis z-axis Fraction Contact
Angle Concentration Angle
Initial Mass
Diffusion
Fraction
Coefficient
Concentration
Relative
Final
Permeability Residual Residual Initial
Residual Gas Exponential Residual
Correlation Water Oil Residual Oil
Saturation Index Oil
Equations Concentration Saturation Saturation
Saturation
26 and 27
Capillary Van
Residual Maximum NAPL
Pressure Genuchten
Water Capillary Threshold
Correlation Parameter
Concentration Pressure Saturation
Equation 33 “β”
The fourth section contains three subsections that relate to initial conditions, the
solutions to non-linear equations and time discretization. The initial conditions of the
grid blocks can be specified in terms of pressure, oil saturation and mass component
concentration. Newtonian iterations are used to solve the nonlinear equations. To
achieve convergence, several Newtonian iterations may be required. This process should
be managed to optimize simulator operations.
96
To control the iterative process, the maximum number of iterations allowed prior
to reducing the time step is specified, along with the maximum number of time steps
tolerated.
Accordingly, time discretization parameters must be specified to determine the
start of the time discretization, the minimum number of time steps in case the Newtonian
iteration does not converge, the pressure and saturation tolerances permissible for
convergence to be declared and the maximum allowable number of time steps prior to
reaching the end of the simulation run.
The fifth section of the simulator data file involves time points that require
reporting and the tolerances of numerical iterations. Typically, pore volume (PV) is
calculated based on the injected rate, and the time step is identified to report phase
saturation, pressure, velocity, flow-rate and mass component concentration.
The sixth section in the data file determines the number of injectors or producers
and the type of recovery (natural, artificial, water-flooding, EOR). The seventh section
lists the elements, such as the mesh with assigned petrological properties. The final
section specifies the initial conditions individually for each grid block and can be
reloaded from successive reruns to mimic primary, secondary or tertiary oil recovery.
4.2. LOW-SALINITY WATER-FLOODING MODEL VALIDATION

Law salinity water flooding model will be validated analytically and numerically
in the next subsections.
4.2.1. Analytical Solution. First one-dimensional analytical solution will be
validated, than the validation will also cover two-dimensional model.
4.2.1.1. One-dimensional analytical solution for solute transportation.
Javandel et al. (1982) presented the following explanation to van Genuchten and Alves’
(1982) analytical solution for one-dimensional solute transportation in a homogenous and
isotropic porous medium. A solute at concentration (Co) is injected at time (To) and
point (x=0) with a uniform rate into an infinitely long, homogenous and isotropic porous
medium. The solute’s concentration changes with time as a result of decay and
97
adsorption. Equations 54 through 57 represent the concentration of the solute at a given

time and distance while considering decay and adsorption.
( )
Because the porous medium was initially free of the solute:

( ) (55)
Therefore, the solute’s gradient at the end of the porous medium remains
constant:
[ ] ( ) (56)
The injected solute’s concentration varies as an exponential function of time,

decay and adsorption.
( ) ( ) (57)
Following the principle of mass conservation, the mass flux at the injection point
is equal at any moment in time to the accumulated mass flux of the solute as it is
transported by diffusion and advection. Van Genuchten (1982) solved Equations 54
through 57 using Laplace transformation, as shown in Equation 59.
( ) ( ) (58)
( ) ( ) ( ) ( ) (59)
98
Where,
( ) ( ) ( ) (60)
( ) ( ) ( ) (61)
Therefore,
( )
( ) [ ] [ ]
√ ( )
( )
[ ] [ ]
√ ( )
[ ( ) ] [ ] ( )
( ) √ ( )
Where,
√[ ( )] ( )
( )
( ) [ ] √ [ ]
√ ( )
( ) ( ) [ ] ( )
√ ( )
4.2.1.2. Two-dimensional analytical solution for solute transportation. Similar

to the one-dimensional mode, it is assumed that the porous media is homogenous and
isotropic and has a unidirectional, steady flow in the x-axis and longitudinal dispersion in
the direction of the flow along the axis, as illustrated in Figure 4.13. The two-
dimensional model also considers the transverse dispersion, which is orthogonal to the x-
axis; this additional term is added to Equation 54, as shown in Equation 65.
99
( )
v
s
x
s DT
DL
Figure 4.13. Two-Dimensional Flow Assumptions
The initial conditions assume that the porous medium is free of any chemical
component. At a specified time, a source of chemical release (2s) is introduced, as shown
in Figure 4.13. Therefore, the initial conditions can be represented as follows (Equations
66 and 67):
( ) ( )
( ) ( )
Similar to the one-dimensional dispersion mode, if it is assumed that the

concentration of the chemical component diminishes exponentially with time, the
boundary conditions can be represented as shown in Equations 68 and 69.
100
( )
( )
Cleary and Ungs (1978) presented the following analytical solution for the model
explained in Equations 67 through 70.
⁄
( ) [ ] ∫ [ { }
√( )
⁄
] [ { } { }] ( )
√ √
4.2.2. One-Dimensional Mass Component Transportation. In this section, a

validation of mass component transportation will be validated analytically and
numerically.
4.2.2.1. Analytical solution validation. This section is designed to examine the
accuracy of the model’s formation and numerical implementation in simulating salt
transport in the aqueous phase. The following details the validation process:
(a) Validate the analytical program’s output.
(b) Solve the transport of a chemical component using analytical and numerical
solutions in a one-dimensional domain.
The analytical solution for the one-dimensional transportation of a chemical
component is described in Appendix A. The program’s output is validated against results
published by Javandel and colleagues (p. 134, 1984). Problem 1 of the analytical
program considers the one-dimensional transportation of a mass component in an aquifer
with uniform flow, as illustrated in Figure 4.14.
101
1 meter
1 meter
Problem 2 - 50 meters
Problem 3 – 10 meters
Problem 4 – 100 meters
Figure 4.14. Schematic for Numerical and Analytical Problems 2 and 3

The problem is characterized with the following parameters: a pore flow velocity
of 0.01 m/d, a molecular diffusion factor of 0.50 m2/d, a period of discharge over 60
years, no retardation (adsorption) and both the decay constant and decay factors of the
source considered zero. Published analytical results of Javandel and colleagues (p. 98,
1984) are used to validate the computer program’s output. The results of the analytical
program’s output are compared to these published results, as shown in Table 4.2.
Table 4.2. One-Dimensional Analytical Program Validation
Distance (meter)
X = 20 (meters) X = 50 (meters)
Time Normalized Concentrations
(Years) Program Standard Program Standard
Published Published
Generated Deviation Generated Deviation
1 0.6476e-01 0.064758 0.000001 0.1634e-02 0.001634 0.000000
2 0.1533e+00 0.152166 0.000802 0.1988e-01 0.019885 0.000004
3 0.2245e+00 0.224483 0.000012 0.5315e-01 0.053146 0.000003
4 0.2849e+00 0.284903 0.000002 0.9253e-01 0.092529 0.000001
5 0.3365e+00 0.336499 0.000001 0.1334e+00 0.133419 0.000013
6 0.3814e+00 0.381358 0.000030 0.1737e+00 0.173721 0.000015
7 0.4209e+00 0.420910 0.000007 0.2125e+00 0.212508 0.000006
The standard deviation between the published analytical results and the program-
generated results is negligible for both the selected distances of 20 meters and 50 meters
and over the discharge duration of 7 years. Therefore, the computer program is
considered valid. A comparison of the salt concentrations along the rock column from
102
the numerical and analytical solutions for t=10, 20 and 60 days, respectively, is shown in
Figure 4.15.
4.2.2.2. Comparison of analytical and numerical solutions. Problem 2
considers the one-dimensional transport of a chemical component in a homogeneous,
water-saturated, porous medium that is 10 meters long, similar to the one used by Wu and
colleagues (1996), as shown previously in Figure 4.14. It has a steady-state flow field
with a 0.1 m/day velocity. A chemical component is introduced at the inlet (x=0) with a
constant concentration, and transport starts at t=0 by advection and diffusion. This
problem is solved numerically by specifying both the inlet and outlet boundary elements
with constant pressures, which give rise to a steady-state flow field with a 0.1 m/day pore
velocity. The constant pressures are determined by specifying the following reservoir
properties: permeability of 0.898x10-12 m2, viscosity of 0.898x10-3 Pa.s and a 10-meter-
long domain with a unit cross-sectional area. The numerical simulation begins with the
generation of a one-dimensional, uniform, linear grid of 1,000 elements for the 10-meter
domain. In order to eliminate the effects of three-phase flow, only single-phase water is
specified.
The properties used in the comparison are as follows: porosity of  = 1, tortuosity
of  = 1, and an effective molecular diffusion coefficient of Dm = 1.157  10-8 m2/s. The
initial and boundary conditions are as follows: initially, no salt exists in the system; Xsalt=
1.010-5 at the inlet boundary (x = 0); and Xsalt = 0 at the outlet boundary at all times.
The analytical solution to Problem 2 is generated by a computer program based on the
analytical solution reached by Javandel and colleagues (1984). A comparison of the salt
concentrations along the rock column from the numerical and analytical solutions is
shown in Figure 4.15. for t=10, 20 and 60 days, respectively.
Problem 3 is another example that considers the one-dimensional transport of a
chemical component in a homogeneous, water-saturated, porous medium. However,
compared to Problem 2, the length of the porous medium has been raised to 100 meters,
as shown in Figure 4.15, and the diffusion coefficient exponent has been lowered to
evaluate the variation against the analytical solution. The medium has a steady-state flow
field with a 0.1 m/day velocity. A chemical component is introduced at the inlet (x=0)
with a constant concentration, and transport starts at t=0 by advection and diffusion. This
103
problem is solved numerically by specifying both inlet and outlet boundary elements with
constant pressures, which give rise to a steady-state flow field with a 0.1 m/day pore
velocity. The constant pressures are determined by specifying the following reservoir
properties: permeability of 0.820x10-6 m2, viscosity of 0.820x10-3 Pa.s and a 100-meter-
long domain with a unit cross-sectional area. In the numerical simulation, a one-
dimensional, uniform, linear grid of 2,000 elements was generated for the 100-meter
domain. In order to eliminate the effects of three-phase flow, only single-phase water is
specified.
1 Analytical Solution (t = 10 days or

0.1PV)
0.9 Numerical Solution (t =10 days or 0.1PV)
0.8 Analytical Solution ( t = 20 days or
Normalized Concentration
0.7 0.2PV)
Numerical Solution (t = 20 days or
0.6 0.2PV)
Analytical Solution ( t = 60 dats or
0.5 0.6PV)
Numerical Solution (t = 60 days or
0.4 0.6PV)
0.3
0.2
0.1
0
0 1 2 3 4 5 6 7 8 9 10
Distance (meters)
Figure 4.15. Analytical Versus Numerical Solution to Problem 2
The properties used in the comparison are as follows: porosity of  = 1, tortuosity

of  = 1, and an effective molecular diffusion coefficient of Dm = 1.0  10-9 m2/s. The
initial and boundary conditions are as follows: initially, no salt exists in the system; Xsalt=
1.010-4 at the inlet boundary (x = 0); and Xsalt = 0 at the outlet boundary at all times.
The analytical solution to Problem 3 is generated by a computer program based on the
analytical solution reached by Javandel and colleagues (1984). A comparison of the salt
104
concentrations along the rock column from the numerical and analytical solutions is
shown in Figure 4.16 for t=100, 200 and 400 days, respectively.
Figure 4.16 confirms that the analytical solution has inherent inaccuracies in
handling the boundary conditions, especially with high diffusion coefficients and
discharge durations. The mathematical model used in the numerical solution is valid and
able to produce accurate results. In addition, the numerical simulator is vigorous when
handling boundary conditions due to its ability to control time and space discretization
and to employ both implicit and explicit approaches in measuring saturation flux.
1
0.9
Analytical Solution (1PV)
0.8
Numerical Solution (1PV)
0.7
Normalized Concentration

0.6
0.5
0.4
0.3
0.2
0.1
0
0 20 40 60 80 100
Distance (meters)
Figure 4.16. Analytical Versus Numerical Solution to Problem 2
4.2.3. Two-Dimensional Mass Component Transportation. This section is

designed to examine the accuracy of the model’s formation and numerical
implementation in simulating salt transport in the aqueous phase. The following details
the validation process:
1. Validate the analytical program’s output.
105
The analytical solution for the two-dimensional transportation of a chemical

component is described in Appendix A. The program’s output is validated against results
published by Javandel and colleagues (p. 134, 1984). Problem 4 considers the two-
dimensional transportation of a chemical component in an aquifer with uniform flow, as
illustrated in Figure 4.17.
50
meters 40
meters
1
125 meter
meters
Figure 4.17. Problem 4 Schematic
The problem is characterized with the following parameters: a pore flow velocity
of 0.1 m/d, a longitudinal dispersion coefficient of 1.00 m2/d, a transverse dispersion
coefficient of 0.10 m2/d, a period of discharge over 100 days, a 50-meter source length,
no retardation (adsorption) and both the decay constant and decay factors considered
zero. The results of the analytical program’s output are compared to the published results
of Javandel and colleagues (1984), as shown in Table 4.3.
Table 4.3. Two-Dimensional Analytical Program Validation
Y = 5 (meters) Y = 40 (meters)
X Normalized Concentrations
(meters) Program Standard Program Standard
Published Published
Generated Deviation Generated Deviation
10 0.71379 0.713792 0.0000010 0.71270 0.712710 0.0000050
20 0.36498 0.364976 0.0000020 0.36361 0.363607 0.0000015
106
Table 4.3. Two-Dimensional Analytical Program Validation (Continued)

30 0.12563 0.125627 0.0000015 0.12489 0.124887 0.0000015
35 0.06277 0.062769 0.0000005 0.06277 0.062343 0.0002135
45 0.11190 0.111930 0.0000150 0.01110 0.011101 0.0000050
60 0.00035 0.000353 0.0000015 0.00035 0.000350 0.0000000
125 0.00000 0.000000 0.0000000 0.00000 0.000000 0.0000000
The standard deviation between the published analytical results and the program-
generated results is negligible for both the selected x and y coordinates. Therefore, the
computer program is considered valid.
4.3. EXPERIMENTAL VALIDATION

Over the past several decades, LSWF has become positioned as an attractive
option for enhancing oil recovery for a number of reasons. The costs associated with
chemicals, gas, steam, or heated water injection do not pertain to LSWF, thereby making
it relatively more economical. It far surpasses the application range of other EOR
methods that are able to achieve high incremental recovery rates, such as steam flooding
and miscible CO2 flooding, given its potential for application to more than 50% of the
world’s hydrocarbon reservoirs (Betty, 2003). This wide application range is due to
LSWF’s less stringent requirements regarding such factors as reservoir depth and fracture
gradient.
The percentage of incremental recovery possible with LSWF ranges from 2% to
40%. While the high end of this range highlights the enormous potential of LSWF, the
fact that the range is so large suggests uncertainty regarding its application. The relevant
literature and experiments also demonstrate such inconsistencies. For example, several
researchers have emphasized that LSWF is driven by wettability modification, yet only a
few core-flooding experiments have measured contact angles. The lack of sufficient data
and crucial boundary definitions for LSWF recovery mechanisms currently stunts the
ability to realize the full potential of this technique to improve oil recovery.
107
The intent of this chapter, then, is to broaden the knowledge of LSWF, examining
its recovery mechanisms by conducting parametric studies and statistical analyses. A
total of 411 core-flooding experiments on sandstone reservoirs are analyzed. These
experiments, which are conducted across wetting conditions, are matched with reservoir
simulations in order to provide a more definitive interpretation of LSWF recovery
mechanisms.
Another side of this research will focus on applying this methodology in
carbonate reservoirs, which provide a much more attractive opportunity for augmenting
oil production for three important reasons. Firstly, most of the world’s oil reserves are
found in carbonate reservoirs (Okasha and Al-Shiwaish, 2009). Secondly, the dilution of
injected brine in carbonate reservoirs is of a much lesser magnitude than in sandstone
reservoirs. Thirdly, polyatomic anions must only be increased until intermediate wetting
conditions are achieved, as this decreases the chances of precipitates forming and
damaging the formation’s permeability. However, despite this enormous potential, only a
few LSWF core-flooding experiments have been conducted in carbonate reservoirs.
This chapter details guidelines that should be followed in order to achieve
maximum recovery from carbonate reservoirs using LSWF. The importance of
measuring the initial wetting conditions is reiterated; past failure to take such
measurements could be why the scientific community has been unable to reach a
consensus on LSWF recovery mechanisms in carbonate reservoirs as well. LSWF
simulations in carbonate reservoirs are conducted and contrasted with core-flooding
experiments.
4.3.1. Sandstone Reservoirs. This section is designed to examine the accuracy of
the model’s formation and numerical implementation in simulating one-dimensional
immiscible displacement, in which oil in a one-dimensional linear rock column is
displaced by water. The reservoir rock’s wettability and injected water salinity are
modified to examine the impact on oil recovery. Published core-flooding experiments
will be compared with simulation results. The flow domain in Problem 6 consists of 12
one-dimensional, horizontal, homogeneous, and isotropic porous media that are 5
centimeters long with diameters of 3.8 centimeters, as illustrated in Figure 4.18. The
one-dimensional radial domain is represented by 100 uniform grid blocks, each with a
108
cross-sectional area of 11.34 cm2 and uniform mesh spacing (x = 0.05 cm). The
problem sets consider four types of crude with varying wettabilities and three cores for
each crude type with a slight variation in reservoir properties, as shown in Table 4.4.
The system initially is saturated with oil and water, the latter of which is at its
irreducible saturation. Water with three different salinities, as shown in Table 4.5, is
injected as a displacing fluid at the inlet to drive oil out of the porous medium domain at
a constant rate of 6 ml/minute (0.5 cubic centimeters per minute). The recovery rates for
water flooding with the three different salinities are compared for each crude type
(wettability).
V = 0.5 cc/minute
3.8 cm
5 centimeters
(cm)
Figure 4.18. Schematic for Numerical Problem 6
109
Table 4.4. Sandstone Core Plug Properties (Taken from Ashraf and colleagues 2010)
Injection
Permeability (md)
Brine
Contact Swi
Wettability Initial IFT Core Porosity
Angle (%
Type (assumed) # (ϕ)
(assumed) PV)
(% Brine
Air
Salinity) (assumed)
A2 100% 18.2 82 54 32
Water-Wet
(IAH =
0.63)
25▫ A3 10% 18.2 78 51 34
A4 1% 18 77 50 31
B2 100% 19.3 185 122 17

Neutral-
Wet (IAH
= 0.12) 70▫ B3 10% 19.3 178 117 19
30
Dynes/cm B4 1% 19 167 110 18
C2 100% 18 66 43 18
Neutral-
Wet (IAH
= -0.27) 117▫ C3 10% 19.2 86 56 21
C4 1% 19.2 78 51 23
D2 100% 19.1 82 54 19
Oil-Wet
(IAH = -
0.57) 141▫ D3 10% 19.1 78 51 21
D4 1% 19 72 47 21
110
Table 4.5. Sandstone Core-Flooding Fluid Properties (Taken from Ashraf and colleagues
2010)
Water Type TDS (ppm) Density (kg/m3) Viscosity (mPa*s)
Connate Water 38,522 1031 (assumed) 1.083 (assumed)
Synthetic Brine
24,951 1019 1.052
(100%)
Synthetic Brine (10%) 2,495 1001 1.008
Synthetic Brine (1%) 249 999 1.010
For Cores A2-A4, four cases are simulated. In Case 1, it is assumed that water’s
relative permeability remains constant. The intent is to examine how well the simulation
results match those of the core-flooding experiments when the salinity concentration is
considered solely as a function of oil’s relative permeability (Wu and Bai, 2009). In Case
2, zero capillary pressure conditions are assumed. The intent in this case is to rule out
any false assumptions related to the IFT value or contact angle value by examining
whether or not the recovery variances of the core-flooding experiments improve.
In Case 3, it is assumed that, similar to oil, water’s relative permeability is also a
function of salinity concentration. The intent is to validate the mathematical model
formulation related to relative permeability curves presented by Jerauld and colleagues
(2008). Case 4 is intended to determine the underlying recovery mechanism of LSWF in
water-wet reservoirs.
Case 1, water-wet cores, and Cores A2 – A4, as described in Table 4.6, are
examined. A comparison of oil recovery rates between core-flooding experiments and
the numerical simulator are shown in Table 4.3, and detailed results are shown in Figure
4.19. A sample of the simulation’s input and output file is shown in Appendix C.
The results summarized in Table 4.3 indicate some variances between the
simulation and experimental results. The variances are inversely proportional to salinity;
this is evident for Core A4, in which the variances for breakthrough and final recovery
are 4.4% OOIP and 3.7% OOIP, compared to Core A2, in which the variances are 2%
OOIP and 1.2% OOIP, respectively. In addition, the final core-flooding recovery rates
111
are all higher than the simulation results. Two possible explanations exist for these
variances. IFT may have been assumed too high, or the irreducible water saturation (Swr)
may increase with LSWF. To further evaluate the results in Table 4.6, a new set of
simulation runs (Case 2) is conducted assuming no capillary pressure effects. The
simulation results are shown in Table 4.7, and the recovery curves are shown in Figure
4.20. A sample of the simulation’s input and output file is shown in Appendix C.
Table 4.6. Sandstone Core (A) Case 1 Simulation Versus Core-Flooding Results
Core-Flooding Experiment
(Taken from Ashraf and colleagues 2010) Numerical Simulator
(No Change in Capillary Pressure)
Core Breakthrough Final Recovery Sor Contact Breakthrough Final Recovery
# Recovery %OOIP (% Angle Recovery %OOIP
%OOIP PV) %OOIP
A2 43 49 35 45.0 47.8
A3 50 56 29 25▫ 50.0 54.7
A4 61 69 21 55.6 65.3
0.7
0.6
Recovery Fraction
0.5
0.4
100% Brine
0.3 Concentration
10% Brine
0.2 Concentration
0.1
0
0 1 2 3 4 5 6 7 8
Pore Volume (PV)
Figure 4.19. Case 1 Simulation Recovery Curve

112
0.8
0.7
0.6
Recovery Fraction
0.5
0.4
100% Brine
Concentration
0.3 10% Brine Concentration
0.2 1% Brine Concentration
0.1
0
0 1 2 3 4 5 6 7 8
Pore Volume (PV)
Core-Flooding Experiment Numerical Simulator
(Taken from Ashraf and colleagues 2010) (Capillary Pressure Zero)
Core Breakthrough Final Recovery Sor Contact Breakthrough Final Recovery

# Recovery %OOIP at PV6 (% Angle Recovery %OOIP at
%OOIP at PV) %OOIP at PV1 PV30
PV1
A2 43 49 35 45.2 48
A3 50 56 29 25▫ 50.7 55.2
A4 61 69 21 58.8 67.7
The results summarized in Table 4.7 indicate that the variances decrease when no
capillary pressure conditions exist. The variances between the core-flooding experiment
and simulation results for Cores A2, A3 and A4 are 1%, 0.8% and 1.3% OOIP,
respectively. However, assuming a zero capillary pressure condition is unrealistic and
concludes that IFT does not contribute to this variance because the recovery factors
without capillary pressure are still lower than the core-flooding results. Furthermore, the
113
water-wet state does not justify a significant decrease in IFT. The other possible
explanation is that the irreducible water saturation is inversely proportional to the residual
oil saturation. To evaluate this assumption, additional simulations (Case 3) are conducted
considering a decrease in water’s relative permeability as the displacing water’s salinity
is decreased. The simulation results are shown in Table 4.8, and the recovery curves are
shown in Figure 4.21. A sample of the simulation’s input and output file is shown in
Appendix C.
0.8
0.7
0.6
Recovery Fraction
0.5
0.4
100% Brine
0.3 Concentration
0.2 10% Brine

Concentration
0.1
0.0
0 1 2 3 4 5 6 7 8
Pore Volume (PV)
( No Change in Capillary Pressure)
Core Breakthrough Final Sor Contact Breakthrough Final Swr (% PV)
# Recovery Recovery (% PV) Angle Recovery Recovery Final
%OOIP %OOIP %OOIP %OOIP
A2 43 49 35 45.0 47.8 36
A3 50 56 29 25▫ 50.3 54.7 40
A4 61 69 21 60.2 66.8 39
114
The results summarized in Table 4.8 indicate good agreement between the
numerical simulator and core-flooding experiments and suggest that the irreducible water
saturation increases during LSWF. Additional simulations (Case 4) are required to
examine the impact of capillary pressure on oil recovery versus the fluid’s relative
permeability. It is assumed that capillary pressure is zero and S wr is inversely
proportional to Sor. The simulation results are shown in Table 4.9, and the recovery
curves are shown in Figure 4.22. A sample of the simulation’s input and output file is
shown in Appendix C.
0.8
0.7
Recovery Fraction
0.6
0.5
0.4
100% Brine
Concentration
0.3 10% Brine

Concentration
1% Brine Concentration
0.2
0.1
0.0
0 1 2 3 Pore Volume
4 (PV)5 6 7 8

115

(Taken from Ashraf and colleagues 2010) (No Capillary Pressure)
Breakthrough Final Breakthrough Final

Core Sor Contact Swr (% PV)
Recovery Recovery Recovery Recovery
# (% PV) Angle Final
A2 43 49 35 45.1 48.0 36
A3 50 56 29 50.6 55.2 40
25▫
A4 61 69 21 60.7 67.7 39
The following is suggested regarding LSWF in strong water-wet reservoirs: (1)

The irreducible water saturation increases during LSWF; (2) The underlying recovery
mechanism in LSWF is the increase in oil’s relative permeability, which accounts for
incremental recovery rates up to 19% OOIP; (3) The reduction in capillary pressure
accounts for incremental recovery of about 0.9% OOIP; (4) there is negligible difference
in the simulation incremental recovery between the linear and nonlinear expressions of
the residual oil saturation and the mass component salt shown in equations 45 and 47.
The second set of core-flooding experiments involves neutrally-wet core systems,
as described in Table 4.1 (Case 5). An IFT of 30 dynes/cm and a contact angle of 70▫ are
assumed. To establish a baseline, the contact angle and IFT are held constant for all of
the water-flooded cores so that the influence of relative permeability on recovery can be
examined. The simulation results are shown in Table 4.10, and the recovery curves are
shown in Figure 4.23. A sample of the simulation’s input and output file is shown in
Appendix C.
116
0.8
Recovery Fraction (%OOIP) 0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0 1 2 3 4 5 6 7 8 9
Pore Volume
100% Brine Salinity 10% Brine Salinity
1% Brine Salinity
Table 4.10. Sandstone Core (B) Case 5 Simulation Versus Core-Flooding Results
Core Breakthrough Final Sor Contact Breakthrough Final Swr (% PV)
# Recovery Recovery (% PV) Angle Recovery Recovery Final
B2 60 63 31 57.8 61.2 21
B3 60 67 27 70▫ 58.3 67.1 23
B4 61 72 23 58.8 70.6 22
117
The results in Table 4.10 indicate very good agreement for both breakthrough and
final recovery between the numerical simulation and the experimental results. The
breakthrough recoveries are comparable for all salinities and higher than for the strong
water-wet cores. This implies that in weak water-wet systems, LSWF recovery is
governed by the low capillary pressure.
In the third set of core-flooding experiments, Cores C, neutral wet cores that
could be classified as a weak oil-wet system, as described in Table 4.11 (Case 6), are
examined. An IFT of 30 dynes/cm, an initial contact angle of 117▫ and a final contact
angle of 91▫ are assumed. To establish a baseline, the contact angle and IFT are held
constant for all the water-flooded cores so that the influence of relative permeability on
recovery can be examined. The simulation results are shown in Table 4.8, and the
recovery curves are shown in Figure 4.24. A sample of the simulation’s input and output
file is shown in Appendix C.
0.7
0.6
Recovery Fraction (%OOIP)
0.5
0.4
0.3
0.2
0.1
0
0 1 2 3 4 5 6 7 8 9
Pore Volume
100% Salinity 10% Salinity

118
Table 4.11. Sandstone Core (C) Case 6 Simulation Versus Core-Flooding Results

(Taken from Ashraf and colleagues 2010) ( No Change in Capillary Pressure)
Breakthrough Final Breakthrough Final Swr (%

Core Sor Contact
Recovery Recovery Recovery Recovery PV)
# (% PV) Angle
%OOIP %OOIP %OOIP %OOIP Final
C2 44 51 40 42.0 49.2 22
C3 49 58 34 43.3 54.6 27
117▫
C4 45 66 28 46.7 60.1 29
The simulation results in Table 4.11 indicate that the final recovery agrees well
with the experimental results, except for Core C4. The results suggest that in weak oil-
wet systems, LSWF recovery is influenced by the increase in oil’s relative permeability
(13.4% OOIP), followed by the decrease in capillary pressure when oil becomes the non-
wetting phase (about 6% OOIP).
In the fourth set of core-flooding experiments, Cores D, the oil wet cores, as
described in Table 4.12 (Case 7), are examined. An IFT of 30 dynes/cm and a contact
angle of 141▫ are assumed. To establish a baseline, the contact angle and IFT are held
constant for all of the water-flooded cores so that the influence of relative permeability on
recovery can be examined. The simulation results are shown in Table 4.9, and the
recovery curves are shown in Figure 4.25. A sample of the simulation’s input and output
file is shown in Appendix C.
The results presented in Table 4.12 indicate very good agreement between the
simulation and the experimental results. It is suggested that in oil-wet systems, the
increase in oil’s relative permeability is the underlying recovery mechanism. The
variance in the breakthrough recovery is subject to the selection of the initial contact
angle.
119
0.7
0.6
Recovery Fraction (%OOIP)
0.5
0.4
0.3
0.2
0.1
0
0 1 2 3 4 5 6 7 8 9
Pore Volume
100% Salinity 10% Salinity 1% Salinity
Table 4.12. Sandstone Core (D) Case 7 Simulation Versus Core-Flooding Results
Core Breakthrough Final Sor Contact Breakthrough Final Swr (%
# Recovery Recovery (% PV) Angle Recovery Recovery PV)
%OOIP %OOIP %OOIP %OOIP Final
D2 46 54 37 50.2 53.4 23
D3 53 56 36 141▫ 50.4 53.4 22
D4 57 61 30 55.8 62.0 27
Several points must be considered prior to contrasting numerical simulations with

core-flooding experiments. The major challenge is rock homogeneity; once a rock type is
declared in a numerical simulator and assigned oil and geological characteristics, those
reservoir properties are considered uniform. However, in actual reservoirs, oil saturations
are not distributed evenly across the length of the core. Consequently, numerical
120
simulation recovery rates for core-flooding experiments will vary. If the oil saturation is
located closer to the injection point, then the breakthrough recovery will be higher; if the
oil saturation is located close to the discharge end, then the final recovery will be higher.
Therefore, it is imperative to include a generous number of elements to control the
variances in breakthrough and final recovery. Another issue is air permeability, which
typically is reported as 50% higher than brine permeability. In cases in which only air
permeability is reported, about 66% represents brine permeability in the rock matrix.
Reservoir simulators discretize time to measure the flux between grid blocks. The
user specifies the maximum time-step value. Smaller time steps equate to more accurate
pore volume measurements and a longer processing time. Because core plugs are short in
length (generally 5 cm long) and core-flooding experiments have a very low injection
rate (as low as 0.1 cubic centimeters per minute as in Problem 6), it is essential to specify
a very low value for the maximum time-step interval (as low as 10 seconds in Problem 6)
in order to measure saturations at concise intervals. Several core-flooding experiments
consider 100% synthetic brine flooding as the baseline for incremental recovery and
continue with diluted concentrations of the synthetic brine to study the effects of low
salinity on incremental recovery. In several cases, the synthetic brine has a lower salinity
than the connate (formation) water; therefore, it should be used as the initial LSWF test
rather than as a baseline. As the baseline for incremental recovery, core-flooding
experiments should begin with connate water or brines that have ionic properties similar
to the connate (formation) water.
Unless a clear intent exists to study LSWF as a secondary recovery method in
comparison to conventional water flooding, it is difficult to understand to what extent
incremental recovery is attributed to LSWF as a secondary or tertiary recovery if 100%
synthetic brines have lower salinity than the connate water. Low-salinity water has lower
viscosity and density than high-salinity water, which implies that salinity concentration
also should be treated as a function of density and viscosity. The simulations address
these points in order to minimize variances with published core-flooding experimental
results.
4.3.2. Carbonate Reservoirs. Problems 7 and 8 are designed to validate the
numerical model’s ability to simulate low-salinity seawater injection in carbonate
121
reservoirs. The objective is to construct the core plug’s petro-physical properties

published by Yousef and colleagues (2010) and to contrast the recovery factors with the
simulation results. The flow domain consists of 2 one-dimensional, horizontal,
homogeneous, and isotropic porous media.
The first case (Problem 7) involves a series of core plugs that are 16.25
centimeters long with diameters of 3.8 centimeters, as shown in Figure 4.26. The core
plugs’ properties are shown in Table 4.13. The flow domain is represented by a one-
dimensional radial mesh comprised of 400 uniform grid blocks, each of which is assigned
a cross-sectional area of 11.34 cm2 and a uniform mesh spacing of x = 0.0406 cm.
Core 13 Core 74 Core 73 Core 10 3.8 cm
16.25 cm
Table 4.13. Carbonate Core Plug Properties (Taken from Yousef and colleagues 2010)
Sample # Length Diameter Brine (mD) Porosity Swi

(cm) (cm) Permeability (% OOIP) (% OOIP)
Case 1 (Core Plugs in Series 13-74-73-10)
13 4.25 3.81 53.80 20.80 13.5
74 3.93 3.80 30.55 28.70 8.60
73 4.02 3.80 45.71 28.90 6.80
10 4.04 3.81 35.00 22.10 14.3
The system initially is saturated with oil and water, the latter of which is at its
irreducible saturation. Seawater with five different salinities is injected as a displacing
122
fluid into the inlet to drive oil out of the porous medium. The published experiments
begin the injection at a constant rate of 1 cubic centimeter (cc) per minute until no oil is
produced. Subsequently, the injection rate is increased to 2 cc and then to 4cc per minute
to ensure that all of the mobile oil is recovered. This study adopts a different injection
scheme in which the 1 cc injection rate is maintained until all the mobile oil is recovered
in order to ensure an accurate pore volume count.
Table 4.14 presents the connate water and seawater properties. The water-
flooding scheme in this case is successive and represents both secondary and tertiary
stage recovery. Therefore, the saved primary thermodynamic variables for all of the grid
blocks are used to define the initial boundary conditions for the subsequent floods. The
residual oil saturation, contact angle and IFT are all functions of salinity, so the oil
saturation, contact angle, salinity, residual oil and irreducible water range must be
revised, in addition to the IFT value for each simulation run. In addition, because the
salinity range will vary for each simulation run based on the maximum and minimum
concentrations, the corresponding viscosity and density for each salinity concentration
should also be defined. The published core-flooding experiments have assigned
different connate water concentrations for each core plug; therefore, the Excel sheet used
to derive oil production should account for the different oil saturations available in each
core plug.
Table 4.14. Carbonate Core-Flooding Fluid Properties at 212 °F (Taken from Yousef and
colleagues 2010)
Property Connate Seawater Diluted Seawater

Water 50% 10% 5% 1%
TDS
213,734 57,670
(ppm)
Density
1108.3 1015.2 995.9 981.2 978.2 977.9
(kg/m3)
Viscosity
0.476 0.272 0.242 0.232 0.212 0.193
(cp)
Mean IFT
39.3 33.7 32.8 32.2 31.85 31.5
(Dynes/cm)
Contact
90▫ 90▫ 80.9▫ 69.0▫ 63.0▫ 62.2▫
Angle (▫)
123
The published incremental recovery and the simulation results for Problem 7 are
shown in Table 4.15. The recovery curves for the simulation results are shown in Figure
4.27. The results indicate good agreement for each seawater salinity concentration
injected. The aggregate value of the variance is 6.81%.
Table 4.15. Problem 7 Carbonate Reservoir Simulation Versus Core-Flooding Results
Injected Seawater Incremental Recovery (%) Incremental Recovery

(TDS) (Taken from Yousef and (Simulation)
colleagues 2010)
57,670 67.04 67.65
28,835 6.99 4.67
5,767 9.12 5.68
2,883.5 1.63 1.97
5,767 0.00 1.03
- 84.97 78.16
0.8
0.7
0.6
Recovery Factor (%OOIP)
0.5
0.4
0.3
0.2
0.1
0
0 20 40 60 80 100
Pore Volume
Seawater Seawater 50% Seawater 20% Seawater 10% Seawater 1%
Figure 4.27. Carbonate Reservoir Simulation Recovery Curve

124
Problem 8 involves a series of core plugs that are 23.65 centimeters long with
diameters of 3.8 centimeters, as shown in Figure 4.28. The core plugs’ properties are
shown in Table 4.16. The flow domain is represented by a one-dimensional radial mesh
comprised of 600 uniform grid blocks; each grid block is assigned a cross-sectional area
of 11.34 cm2 and a uniform mesh spacing of x = 0.0394 cm.
Core 159 Core 55 Core 91 Core 66 Core 61 Core 128 3.8 cm
23.65 cm
Table 4.16. Core Plug Properties (Taken from Yousef and colleagues 2010)
Sample # Length Diameter Brine (mD) Porosity Swi

(cm) (cm) Permeability (% OOIP) (% OOIP)
Case 2 (Core Plugs in Series 159-55-91-66-61-128)
159 3.94 3.81 74.34 22.57 12.6
55 4.16 3.81 59.44 27.73 14.7
91 3.83 3.81 73.26 24.97 6.60
66 3.77 3.81 64.51 25.65 19.0
61 4.02 3.81 73.25 26.60 17.6
128 3.93 3.81 65.26 20.36 15.7
The published incremental recovery and the simulation results for Problem 8 are
shown in Table 4.17. The recovery curves for the simulation results are shown in Figure
4.29. The results indicate good agreement for each seawater salinity concentration
injected. The aggregate value of the variance is 10.95%. There are two possible reasons
125
for the variances: (1) an unknown pore volume count for each corresponding injection
rate in the core-flooding experiment; (2) a non-linear relationship between the salt
concentration versus residual oil saturation, IFT and contact angle. Therefore, the
linearity of residual oil saturation, IFT and contact angle is examined in the proceeding
section.
Table 4.17. Problem 8 Carbonate Reservoir Simulation Versus Core-Flooding Results

(Taken from Yousef and colleagues 2010)
Injected Seawater Incremental Recovery (%) Incremental Recovery

(TDS) (Taken from Yousef and (Simulation)
colleagues 2010)
57,670 74.12 66.0
28,835 8.48 5.20
5,767 9.95 5.90
2,883.5 0.95 3.50
5,767 0.00 1.60
- 93.65 82.70
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0 20 40 60 80 100 120
Pore Volume
Seawater Seawater 50% Seawater 20%
Figure 4.29. Simulation Recovery Curve

126
A linear relationship may not exist between the salt concentration and the residual
oil saturation, contact angle and IFT, which would render the relative permeability and
capillary pressure formulation incorrect. Therefore, it is essential to examine core-
flooding experiments. It was assumed that the relative permeability and capillary
pressure have a linear relationship with the salt concentration; in carbonate reservoirs,
this assumption requires validation, which is achieved by reviewing changes in the
residual oil saturation, contact angle and IFT values in core-flooding experiments. The
flooding experiments conducted by Yousef and colleagues (2010) are used for validation
because they include IFT and contact angle measurements for six salinity concentrations
and residual oil saturation for five seawater salinities. In addition, up to six core plugs
are used in a series for the core-flooding experiment. The residual oil saturation, contact
angle and IFT for each salinity concentration are examined, as shown in Figures 4.30,
4.31 and 4.32, respectively.
1
y = -0.514x2 + 1.5616x - 0.0462
R² = 0.9976
Normalized Residual Oil Saturation
0.8
0.6
0.4
0.2
0
0 0.2 0.4 0.6 0.8 1
Normalized Salt Concentration
Figure 4.30. Correlation of Residual Oil Saturation and Salt Concentration

127
1
y = -0.768x2 + 1.7495x + 0.0144
R² = 0.9857
0.8
Normalized Contact Angle
0.6
0.4
0.2
0
0 0.2 0.4 0.6 0.8 1
Figure 4.31. Correlation of Contact Angle and Salt Concentration
1
y = 0.0923x2 + 0.8783x + 0.0292
R² = 0.9979
Normalized Contacct Angle
0.8
0.6
0.4
0.2
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Figure 4.32. Correlation of IFT and Salt Concentration
These results indicate that a nearly linear relationship exists between the salt
concentration and residual oil saturation, contact angle and IFT; therefore there is
negligible difference in the simulation incremental recovery between the linear and
128
nonlinear expressions of the residual oil saturation, contact angle and IFT as a function of
the mass component salt shown in equations 45 and 48, 49 and 51 in addition to 50 and
53, respectively. It is thought that the unknown pore volume count for each
corresponding injection rate in the core-flooding experiment is creating the variance
between the experiment and simulation results.
129
5. SENSITIVITY ANALYSIS
5.1. STATISTICAL ANALYSIS OF CORE-FLOODING EXPERIMENTS

5.1.1 Sandstone Reservoirs. A sandstone core-flooding experiment database
was built based on published journal and conference papers. The database consists of 411
LSWF experiments, of which 223 are secondary mode recovery and 188 are tertiary
mode recovery. In addition, reported fluid and core properties were included, such as
irreducible water saturation, wettability, IFT, clay content, aging and test temperatures, as
presented in Table E.1, which appears in Appendix E. A statistical representation of the
core-flooding database will be provided at http://www.eorcriteria.com. The summary of
core-flooding experiments highlights the extent and consistency of reporting boundary
conditions. It is evident that capillary pressure variables, such as wettability and IFT, are
reported infrequently, having a total of only 78 and 22 entries, respectively, out of 411 in
the core-flooding database. Similarly, clay content and the weight percentages of chlorite
and kaolinite are reported 66, 48 and 48 times out of 411, respectively, in the database.
The statistical analysis conducted for the low-salinity core-flooding database comprises
two stages. In the first stage, correlations are evaluated for the reported variables in the
core-flooding experiments. Correlation measurements are required in order to screen
sensitivities of various core-flooding variables versus the intended outcome, residual oil
saturation. Evaluating key variables in LSWF is critical for generating a prediction
model because strong correlations will improve the accuracy of the multivariable
regression curve.
The restricted maximum likelihood (REML) method was used to examine the
relationships between the variables in the core-flooding experiments, show in Equation
71. The entire database, consisting of 411 low-salinity core-flooding experiments, is fed
into the JMP statistical software, and one-to-one correlations are generated, as presented
in Figure F.1, which appears in Appendix F.
∑ ( )
∑ (71)
⁄
130
The results in Figure F.1 indicate strong correlations between the Sor and chlorite
(0.7891) and to a lesser extent kaolinite (0.4399) contents, in addition to the wettability
index (0.3890); however, none of these strong correlations can be used because the
majority of LSWF core-flooding experiments fail to report clay content or wettability.
Without strong correlations, the prediction model will have poor accuracy and confidence
limits, as demonstrated by generating a prediction model without the previously-
mentioned variables that effect capillary conditions, as shown in Figure F.2. The
multivariable regression curve and the confidence level both exhibit poor accuracy, and
as a result, the impact of each core-flooding variable on Sor cannot be examined.
However, the results in Figure F.2 indicate, in order of influence, that the oil aging time,
brine cation concentration at Swi and divalent ion concentration in the injected brine
strongly influence Sor, which emphasizes the possible role of wettability modification in
LSWF.
5.1.2 Carbonate Reservoirs. A database containing 18 core-flooding
experiments has been constructed. These experiments were taken from Yousef and
colleagues (2010) and Gupta and colleagues (2011). While the dataset is not adequate for
prediction modeling, the prediction profiler in JMP (statistical software) is used to
examine incremental recovery for the following variables: (a) acid number and IFT
sensitivities, as shown in Figure 5.1, and (b) 2nd and 3rd stage injected brine anion
content, as shown in Figure 5.2.
In Figure 5.1, the increase in water wetness improves secondary-stage oil
recovery; however, an opposite effect is seen at the tertiary stage. This suggests that a
reduction in the sulfate concentration continues to increase water wetness, thereby
increasing the capillary pressure. In contrast to the former observation, decreasing the
anion concentration in the injected brine improves oil recovery, as Figure 5.2 depicts. The
only exception is in tertiary recovery, where increasing the anion concentration in the
injected brine improves oil recovery because the capillary pressure would decrease as the
wettability is favorably modified to intermediate wetting conditions.
131
Figure 5.1. Prediction Profiler Acid Number and IFT
Figure 5.2. Prediction Profiler (2nd and 3rd Stages) Injected Brine Anions
132
6. CASE STUDIES
In this chapter, 3D models are constructed to examine LSWF recovery under

various wetting classifications. Such up-scaled models are required to mimic or represent
real reservoirs to an acceptable approximation. The role of sweep efficiency in oil
recovery, especially in three-dimensional space, as well as heterogeneities and reservoir
anisotropy, will be examined. The chapter is organized into three different case studies.
The first case study involves the modeling of various wetting conditions. In order
to understand LSWF recovery mechanisms, up-scaling beyond simple one-dimensional
models is required. In this way, the viability of LSWF and the impact of sweep
efficiency on LSWF recovery can be examined. Therefore, three up-scaled models are
constructed to depict water-wet conditions, intermediate wetting conditions and oil-wet
conditions. The validated fluid relative permeability functions and capillary pressure
functions are adopted from Chapters 4.
The work presented in the second case study extends beyond typical EOR
selection criteria to include statistical analysis. A prediction model is generated for
miscible CO2 flooding based on 153 reported EOR projects. This novel approach
provides a third criterion for EOR selection beyond those presented in Chapter 2.
Finally, the third case study grows out of the benefits of miscible CO2 flooding
and carbon dioxide sequestration coupled with the social responsibility to limit the risks
associated with the bulk transportation of known asphyxiant and acidic gas. The HSE
risks associated with the bulk transportation of CO2 are presented in terms of exposure
thresholds and the dispersion modeling of toxic substances.
6.1. THREE-DIMENSIONAL IMMISCIBLE DISPLACEMENT

Problem 9 aims to validate the numerical simulator’s ability to model three-
dimensional immiscible displacement. Published experiments will be used to validate the
numerical solution. LSWF’s performance in a stratified, 20-acre, 5-spot pattern was
modeled by Gaucher and Lindley (1960); a total of 6 reservoir models were scaled to
133
examine the impacts of permeability, oil viscosity, mobility ratios and injection rate on
water-flooding performance. The properties of Reservoir 4 will be adopted, as shown in
Table 6.1, to construct a 20-acre, 5-spot water-flooding well pattern, as illustrated in
Figure 6.1.
The injection wells lie at each corner of the four quadrants and diagonal to the
producer, which is positioned in the center. The injection and production rates are
assumed constant, and the formations are assumed homogenous and isotropic. The
producing zone’s length and width are assumed identical. A three-dimensional mesh is
generated with 10 grid blocks in the x and y axes and 5 in the z axis, which is equivalent
to 500 elements for a single quadrant. Therefore, the spacing between the grid blocks is
as follows: x and y = 14.22 m and z = 1.212 m. In addition, both the injection and
producing wells will be considered fully penetrating.
Table 6.1. Reservoir Properties for Five-Spot Water-Flooding Pattern
Areal Pattern Units Reservoir 5 (20-Acre, 5-Spot)

Top-Layer Thickness, ha Feet 10
Bottom-Layer Thickness, hb Feet 10
Top Layer, Absolute Permeability, ka md 16
Bottom Layer, Absolute Permeability,
md 16
kb
kowr, a md 13
kowr, b md 13
kwor, a md 9
kwor, b md 9
Top-Layer Porosity, ϕa % 20
Bottom-Layer Porosity, ϕb % 20
1-Sor-Swr, a % 0.50
1-Sor-Swr, b % 0.50
Oil Viscosity, μo cp 2.17
Water Viscosity, μw cp 0.50
Oil-Water Density Difference, Δρ gm/cc 0.20
Interfacial Tension, σ dynes/cm 25
Water Injection Rate, iw B/D 44
134
Injectors Producer Injectors
Layer a
3.03 meters
Layer b
3.03 meters
285.4 meters
6.1.1. Intermediate Wet Reservoirs. Problem 9 examined metrological

properties identical to Gaucher and Lindley (1960), and the result of the simulation is
compared with their experimental results. Two additional problems are considered,
Problem 10 for a water-wet reservoir and Problem 11 for an oil-wet reservoir; in both
cases, HSWF and LSWF are simulated.
The oil recovery factor provided by Gaucher and Lindley’s (1960) experiment for
Reservoir 4 is 90.4% after injecting 3 pore volumes. Additionally, because the sum of oil
and water’s residual saturations is 0.5, the residual oil saturation must equal 0.048.
Therefore, Reservoir 4 can be considered to have intermediate wetting conditions. The
result of the simulation in Problem 9 is compared with the experimental results from
Gaucher and Lindley (1960), as shown in Figure 6.2.
135
1
0.9
0.8
0.7
0.6
MSFlow
0.5
Gaucher and Lindley (1960)
0.4
0.3
0.2
0.1
0
0 1 2 3 4 5 6 7 8
Pore Volume
Figure 6.2. Three-Dimensional, Five-Spot Well Pattern (Mixed Wettability Reservoir)
The breakthrough recovery is identical for both the simulation and experimental
results. The final recovery results agree well; the simulation final recovery is 85.5
%OOIP, and the experimental final recovery is 90.4%. However, the variance between
the simulation and experimental results can be attributed to oil relative permeability
correlations. Buckley and Leveret’s (1942) correlation overestimates the fluid’s relative
permeability, and Corey’s (1954) underestimates it.
6.1.2 Water-Wet Reservoirs. Problem 9 underlines the importance of the initial
wetting conditions on LSWF recovery. When the residual oil saturation is very low due
to intermediate wetting conditions, there is little or no incremental recovery
(Skrettingland colleagues 2011). Therefore, Problems 10 and 11 consider water-wet
conditions with a 60° contact angle and oil-wet conditions with a 120° contact angle, as
shown in Appendix C. Simulation results for Problem 10 consider HSWF and LSWF. In
LSWF, the contact angle shifts to intermediate wetting conditions as the salinity
decreases, whereas in HSWF, the contact angle remains constant. The recovery curves
are shown in Figure 6.3.
136
0.9
0.8
0.7
0.6
0.5
HSWF
0.4
LSWF
0.3
0.2
0.1
0
0 1 2 3 4 5 6 7 8
Pore Volume
Figure 6.3. Three-Dimensional, Five-Spot Well Pattern (Water-Wet Reservoir)
The breakthrough recovery of HSWF is 47.5 %OOIP, in contrast to LSWF, which

is 45.7 %OOIP. This difference is due to an increase in the mobility ratio that occurs
because the injected brine viscosity is proportional to salinity concentrations. However,
Lemon andcolleagues, 2011 suggested that desorption and migration of fine particles
during LSWF may improve sweep efficiency. After breakthrough recovery as wettability
is modified, the capillary pressure decreases, and the incremental recovery of LSWF
increases as the contact angle shifts to intermediate wetting conditions, as shown in
Figure 6.3.
6.1.3. Oil-Wet Reservoirs. Problem 11 considers HSWF and LSWF in a three-

dimensional, five-spot well pattern in an oil-wet reservoir. The contact angle is assumed
to remain at a constant value of 120° for HSWF but to decrease linearly with salinity
during LSWF, as shown in Figure 6.4. The results in Figure 6.4 underline the impact of
wettability modification in oil-wet conditions. LSWF’s incremental recovery is more
rapid and overshadows an adverse mobility ratio resulting from a decrease in injected
brine salinity.
137
0.9
0.8
0.7
0.6
0.5
HSWF
0.4
LSWF
0.3
0.2
0.1
0.0
0 1 2 3 4 5 6 7 8
Pore Volume
Figure 6.4. Three-Dimensional, Five-Spot Well Pattern (Oil-Wet Reservoir)
The initial wetting state is the single most important criterion for LSWF. The
incremental recovery in oil-wet systems, or “typical carbonate reservoirs,” is more rapid
and thus requires less dilution of the injected brine salinity. This has been proven by
core-flooding experiments and up-scaled simulation results. Water-wet systems, or
“typical sandstone reservoirs,” can achieve slightly higher incremental recovery than
“oil-wet” reservoirs; however, it take higher salinity dilution ratios and higher injection
pore volumes to achieve the final recovery. There is little benefit from LSWF in
reservoirs with intermediate wetting conditions. In addition, simulation results indicate
substantial un-swept quantities of available oil saturation, as has been validated by three-
dimensional simulations.
138
7. CONCLUSIONS
This work has constructed an EOR database based on 652 reported EOR projects.
The database lists reservoir rock and fluid properties in addition to project attributes,
which yields a profile of worldwide EOR trends. The EOR screening criteria published
by Taber and colleagues in 1996 (SPE 35385) is updated, and the database analysis
presented here illustrates the relationship between EOR project distributions and key
reservoir properties. The in-depth analysis of EOR projects presented in this paper has
the potential to support EOR selection, implementation and development and to
encourage continual improvement. Furthermore, updating EOR criteria encourages
research advancements that would broaden the range of EOR applications and introduce
innovative technologies capable of reclassifying unrecoverable and contingent
hydrocarbon reserves.
A reservoir simulator has been developed for LSWF. The developed relative
permeability and capillary pressure functions have been illustrated, along with capillary
pressure desaturation curves, and the LSWF reservoir simulator has been validated
analytically for the transportation of a mass component in the aqueous phase. Reservoir
simulations conducted to examine LSWF recovery sensitivities conclude that LSWF
recovery mechanisms are governed based on capillary conditions. In strong water-wet
conditions, the increase in oil’s relative permeability is the underlying recovery
mechanism. In weak water-wet conditions, LSWF’s incremental recovery is driven by
low capillary pressures. In weak oil-wet conditions, the primary LSWF recovery
mechanism is the increase in oil’s relative permeability, and the secondary LSWF
recovery mechanism is the change of the non-wetting phase to oil.
In strong oil-wet conditions, on the other hand, the underlying LSWF recovery
mechanism is the increase in oil’s relative permeability. In all cases, an appreciable
decrease in interfacial tension (IFT) is realized at the breakthrough recovery; however,
that is rapidly overshadowed by the increase in oil’s relative permeability and the
decrease in contact angle.
The simulator and recovery mechanisms presented by Aladasani and colleagues
(2012) are used, and their suitability and validity to low-salinity water flooding in
139
carbonate reservoirs has been confirmed by comparing simulated LSWF secondary and
tertiary recoveries with published core-flooding experiments. Simulation and statistical
analysis suggest that, under intermediate wetting conditions, the incremental recovery of
LSWF is driven by low capillary pressures, and the primary LSWF recovery mechanism
is the increase in oil’s relative permeability. Therefore, it is ideal to modify wettability
by shifting and then maintaining the wetting state from oil-wet or water-wet to
intermediate wetting conditions irrespective of the salinity dilution. Furthermore, if the
wettability is shifted to a strong water-wet state, it becomes more favorable to use brine
with added anions to shift the wettability back to an intermediate wetting state. IFT has a
bigger impact on LSWF in carbonate reservoirs; however, the contact angle is more
significant to ultimate oil recovery.
LSWF core-flooding experiments will conclude different results depending on the
initial wetting state and final wetting state. Future core-flooding experiments should
focus on measuring the contact angle and IFT before and after each injected brine
modification. Low salinity does not always improve oil recovery, especially when the
initial wetting condition is intermediate.
The summary of 411 core-flooding experiments highlights the extent and
consistency of reporting boundary conditions, with the following two implications for
statistical analysis: (1) Even though statistical correlations of the residual oil saturation to
chlorite (0.7891) and kaolinite (0.4399) contents, as well as to the wettability index
(0.3890), are comparably strong, the majority of dataset entries are missing, and no
prediction model can be generated; (2) If a prediction model is generated without clay
content and a wettability index, even though LSWF emphasizes wettability modification
by virtue of the strong influence on Sor of oil aging time, brine cation and divalent ion
concentration, the prediction model regression curve and confidence level will be poor.
A database containing 18 core-flooding experiments has been constructed. An
increase in water wetness improves secondary-stage oil recovery; however, an opposite
effect is seen at the tertiary stage. This suggests that a reduction in the sulfate
concentration continues to increase water wetness, thereby increasing the capillary
pressure. In contrast to the former observation, decreasing the anion concentration in the
injected brine improves oil recovery. The only exception is in tertiary recovery, where
140
increasing the anion concentration in the injected brine improves oil recovery, because
the capillary pressure decreases as the wettability is favorably modified to intermediate
wetting conditions.
Several screening tools have been presented in this paper, namely, the distribution
of reservoir properties in miscible CO2 projects, correlations of reservoir properties in
miscible CO2 projects and a prediction model for miscible CO2 recovery. These
screening tools will provide insight into miscible CO2 reservoir selection beyond
conventional selection criteria. An overview of the frequency of hydrocarbon and CO2
pipeline failures has been outlined, including wet sour service risks in pipelines
manufactured prior to 1975 and/or 1984. In particular, attention has been drawn to the
risks of third-party damage of transit pipelines and the associated size of the failure.
Finally, a quick screening tool for CO2 releases has been presented, and the toxicity risk
of H2S has been highlighted in anthropogenic CO2 streams.
141
APPENDIX A.
A. ANALYTICAL PROGRAM FOR 1-DIMENSIONAL TRANSPORTATION OF

A CHEMICAL COMPONENT
142
Based on Van Genuchten (1982), Javandel et al. (1982) published the following program
called (ODAST) to analytically solve one dimensional transportation of a chemical
component.
DOUBLE PRECISION X,T,CD,PI,D1,V1,R,T0,ALAM1,ALFA1

1 ,VD,ALAM,ALFA,T1,X1,DSQRT2,U,C1,C2,B1,B2,B3,C3,
2 A1,A,BB1,BB2,DD2,CC1,CC2,DD,DD1,A1,ARG
DIMENSION CD(6),X(100),T(100)
DATA ISE,MINUS/2HX=,1H-/
PI=4*DATAN(1.D0)
READ(5,510) NUMX,NUMT
READ(5,520) (X(I),I=1,NUMX)
READ(5,520) (T(I),I=1,NUMT)
10 READ(5,520) D1,V1,R,T0,ALAM1,ALFA1
IF (D1.EQ.O) STOP
V=V1*365.25
D=D1*365.25
ALAM=ALAM1*365.25
ALFA=ALFA1*365.25
DO 90 IX=1,NUMX,6
LX=MIN0(IX+5,NUMX)
IP=MAX0(2-IX,0)
WRITE(6,610) IP,V1,D1,R,ALAM1,ALFA1,T0,(IXE,X(I),I=IX,LX)
WRITE(6,620) (MINUX,I=IX,LX)
DO 80 II=1,NUMT
T1=T(II)
K1=0
DO 70 KK =IX,LX
K1=K1+1
X1=X(KK)
IFLAG=0
20 IF (AFLA.EQ.ALAM) GO TO 30
DSQRT2=2*DSQRT(D*R*T1)
U=DSQRT(V*V+4*D*R*(ALAM-ALFA)
C1=X1*(V-U)/(2*D)
C2=X1*(V+U)/(2*D)
B1=(R*X1-U*T1)/DSQRT2
B2=(R*X1+U*T1)/DSQRT2
B3=(R*X1+V*T1)/DSQRT2
C3=(V*X1/D)+(ALFA-ALAM)*T1
A1=(V/(V+U))*EXER(C1,B1)+(V/(V-U))*EXER(C2,B2)+
1 (V*V/(2*D*R*(ALAM-ALFA)))*EXER(C3,B3)
A=A1
GO TO 40
30 DSQRT2=2*DSQRT(D*R*T1)
BB1=(R*X1-V*T1)/DSQRT2
BB2=V*X1/D
DD2=DSQRT(V*V*T1/PI*D*R))
CC1=BB1*BB1
CC2=(R*X1+V*T1)/DSQRT2
DD=1. +(V*X1/D)+(V*V*T1/(D*R))
143
DD1=0.0
A2=0.5*EXER(DD1,BB1)+DD2*EXPD(-CC1)-0.5*DD*EXER(BB2,CC2)
A=A2
40 IF 9IFLAG.EQ.1) GO TO 50
ARG=-T1*ALFA
CD(K1)=EXPD(ARG)*A
IF (T1.LE.T0) GO TO 60
T1 = T1-T0
IFLAG=1
GO TO 20
50 CD(K1)=CD(K1)-A*EXPD(-ALFA*T(II))
T1=T(II)
60 IF(CD(K1).LT.1.E-14)CD(K1)=0.
70 CONTINUE
WRITE(6,630) T(II),(CD(KK),KK=1,K1)
80 CONTINUE
90 CONTINUE
GO TO 10
510 FORMAT(2I5)
520 FORMAT(6D10.4)
610 FORMAT(//I1,20X,36HDIMENSIONLESS CONCENTRATION C/CO FOR
1 /1H0,7X,2HV=,F6.2,2X,2HD=,F6.2,2X,2HR=,F3.1,2X,7HLAMDA=
2 ,F5,3,2X,6HALPHA=,F5.3,
3 2X,3HT0=,F5.1/8X,61(1H-)/,9H T(YEARS),6(4X,A2,F5.0))
620 FORMAT(1X,10(1H-),6(A1,10(1H-)))
630 FORMAT(3X,F5.1,2X,6D11.4)
END
DOUBLE PRECISION A,B,C,X,T,Y

EXER=0.0
IF((DABS(A).GT.170.).AND.B.LE.O.) RETURN
IF(B.NE.O.O)GO TO 100
EXER=EXPD(A)
RETURN
100 C=A-B*B
IF((DABS(C).GT.170.).AND.(B.GT.0))RETURN
IF(C.LT.-170.) GO TO 130
X=DABS(B)
IF(X.GT.3.0)GO TO 110
T=1./1.+.3275911*X)
Y=T*(.2548296-T*(.2844967-T*(1.421414-T*(1.453152-1.061405*T))))
GO TO 120
110 Y=.5641896/(X+.5/(X+1./(X+1.5/(X+2./(X+2.5/(X+1.))))))
120 EXER=Y*EXPD(C)
130 IF(B.LT.0.0)EXER=2*EXPD(A)-EXER
RETURN
END
DOUBLE PRECISION X
EXPD=0.0
IF(X.LT.-170) RETURN
EXPD=DEXP(X)
RETURN
END
144
APPENDIX B.
B. DISPERSION MODELING OF MISCIBLE CO2 RELEA

145
Carbon dioxide is a hazardous substance that contributes to the dimensions of risk,

likelihood and consequence of failure. Carbon steel pipelines are prone to failure due to
wet CO2 corrosion when the partial pressure of CO2 exceeds 7 psig (Craig, 2003); this
threshold is exceeded readily when CO2 constitutes the main flow stream and when it is
handled above its critical condition. When a CO2 pipeline fails, asphyxiation becomes a
hazard, and CO2 toxicity grows exponentially with higher concentrations because of the
manner in which oxygen depletion impacts living beings. Therefore, the consequence of
CO2 asphyxiation increases considerably in confined spaces and in areas where fires may
start. An area requiring attention is the corrosiveness and, more significantly, the
toxicity of impurities in CO2 streams. Although hydrogen sulfide (H2S) concentrations
are lower in anthropogenic CO2 streams than in naturally-occurring CO2 reservoirs (e.g.,
Canyon Reef Carriers’ H2S concentration is 1500 ppm versus 0.9% (9,000 ppm) H2S in
the Weyburn Pipeline (Doctor and colleagues 2005)), these concentrations classify both
service streams as sour. Sour service has two critical implications, design suitability and
consequence modeling.
NACE standards pertinent to testing the susceptibility of materials to sulfide
stress cracking (SSC) and hydrogen-induced cracking (HIC) were first published in 1977
and 1984, respectively; this implies that pipelines constructed prior to these years contain
materials that may not conform to proven resistance against SSC, HIC or both. The
consequence modeling of CO2 pipeline bulk transportation is limited and focuses
primarily on CO2 rather than H2S, which is lighter and about 120 times more toxic .
Several hydrocarbon gas and liquid streams are hydrated to avoid hydride
formation and corrosion; however, even if corrosion can be mitigated effectively, the
leading cause of pipeline failure is third-party damage, which often results in large holes
and/or ruptures (Conservation of Clean Air and Water in Europe (CONCAWE), 2011;
European Gas Pipeline Incident Data Group (EGIG), 2011). The bulk transportation of
carbon dioxide over long distances and above the critical condition poses both an
integrity management and emergency preparedness challenge. The US contains about
3,950 miles of CO2 pipeline currently in service for EOR operations (N.V. Nederlandse
Gasunie, 2009). The main transient CO2 pipeline network spans about 2,798 miles and
comprises 20 pipelines of various diameters, lengths and ages, ranging from 8” to 30” in
146
diameter, 40 to 502 miles in length and having offered 1 to 50 years of service life (Table
B.1).
Table B.1. CO2 Pipelines (Source: Kinder Morgan (2011), URS (2009), Denbury,
(2011))
Pipelines Constructed Diameter Length Operating Capacity Operator
(inches) (miles) Pressure (BCFD)
(psig)
Cortez 1984 30 502 1900 1.30 Kinder
Morgan
McElmo Creek 8 40 - 0.06 Resolute
Bravo Line 1984 20 218 1800-1900 0.382 BP
Transpetco/Bravo - 12 ¾ 120 - 0.175 Transpetco

Sheep Mountain - 20 184 2050 0.330 Occidental
Permian
Rosebud - 24 224 2050 0.480 Occidental
Permian
Central Basin - 26/16 - 0.600 Kinder
Pipeline Morgan CO2
Company,
L.P. (owner)
Este Pipeline - 12/14 119 - 0.250/0.150 Occidental
Permian
Slaughter - 12 40 - 0.160 Occidental
Pipeline Permian
West Texas - 12/8 127 - 0.100 Trinity
Pipeline Pipeline L.P.
Llano lateral - 12/8 53 - 0.100 Trinity
Pipeline L.P.
CRC pipeline 1972 16 140 - 0.270 Kinder
Morgan CO2
Company,
L.P. (owner)
Val Verde 1998 10 82 - 0.096 Petrosource
Weyburn 1999 12/10 205 - 0.192 DSC
NEJD 1986 20 183 - 0.441 Denbury
Free State 2006 20 86 - 0.257 Denbury
Delta 2008/2009 24 31/68 - 0.301 Denbury
Gwinville 1963 - 51 - 0.112 Denbury
Pipeline
Green Pipeline 2011 24 325 - 0.725 Denbury
147
A review of CO2 pipeline incidents from 1986 to 2008 based on DOT records was
prepared by URS (2009). The review highlights 13 incidents involving CO2 pipelines
operated by 8 different organizations, with a total failure rate of 1.69E-4 per mile/year.
The failure modes are listed in Table B.2. For comparison, this table also includes the
failure frequencies and failure modes of European oil and gas pipelines.
Table B.2. Pipeline Failure Statistics
Failure Mode US CO2 EU Gas Pipelines

EU Oil Pipelines
Pipeline (1971-2004)
(1971-2010)
(1986-2008) EGIG Report,
CONCAWE, 2011
URS, 2009 2011
Equipment Failure 46% 26.1% 16.7%
Corrosion 15.5% 27.4% 16.1%
Operation Error 15.5% 6.4% 4.8%
Unknown 23% - 6.6%
Natural - 3.1% 7.4 %
3rd Party - 36.8% 48.4%
Failure Frequency
1.69E-4 5.49E-4 2.18E-4
(per mile/year)
Failure size and frequency

The HSE published guidelines on failure rates and event data for use in land use
planning risk assessments (HSE, 2010). These guidelines classify the pipelines used for
the bulk transportation of gas into aboveground and underground constructions. In
addition, the HSE guidelines categorize gas pipeline failure sizes into four groups, along
with their corresponding failure rates (Table B.3). Similarly, CONCAWE published a
report on the performance of European cross-country oil pipelines (1971-2010). This
report classifies oil gas pipeline failure sizes into six classifications with their
corresponding failure rates (Table B.4).
148
Table B.3. Gas Pipeline Failure Sizes and Rates
Failure Category AboveGround Underground Pipelines

Pipelines HSE, 2009 (Mile/Year) @t 32barg
HSE, 2010 (Mile/Year) (PIPIN)
Rupture > 1/3rd
Diameter 1.05E-05 4.41E-06
Large Hole = 1/3rd
Diameter 5.31E-05 1.61E-06
Small Hole = 5mm to
25mm 1.08E-04 1.51E-05
Pin Hole < 5 mm 2.57E-04 2.14E-04
Table B.4. Oil Pipeline Failure Sizes and Rates (CONCAWE, 2010)
Failure Category Failure Rate Average Spill

(Mile/Year) Volume
(Barrels/Event)
Loss of Pressure Containment,
equipment other than the pipeline 1.405E-5 308
Pinhole less than 2mm x 2mm 3.576E-5 365
Fissure 2 to 75 mm long x 10% max wide 5.109E-5 1723
Hole 2 to 75mm long x 10% min wide 1.071E-4 566
Split 75 to 1000 mm long x 10% max wide 6.258E-5 1541
Rupture = > 75mm x 10% min wide 7.152E-5 4195
The performance history of the European oil and gas pipelines shown in Tables
B.3 and B.4 indicates that aboveground pipelines are more prone to failure than
underground pipelines, with the exception of pinhole leaks that share the same
magnitude. Oil pipeline failure frequencies are comparable for almost all of the failure
categories. What is most striking is the failure frequency of ruptures, splits and large
holes; this is indicative of incidents involving heavy earth equipment. By contrast,
equipment failure in CO2 pipelines in the US is more than double the rate reported by
European hydrocarbon pipelines, and unknown causes of failures stagger at 23%
compared to the EGIG’s 6.6%.
149
Carbon dioxide pipeline release scenarios

Carbon dioxide release scenarios will be studied for two pipelines, Pipeline 1 and
Pipeline 2. The failure frequencies will be based on the CONCAWE 40-year
performance history, and the failure size considered for the pipelines will be a hole and
rupture, respectively, as shown in Table B.5. The only human effect considered will be
toxic exposure due to CO2 and H2S, both of which substances cause harm without the
occurrence of a thermal or overpressure event. However, increases in the toxicity level of
CO2 due to secondary causes, such as a fire, is beyond the scope of the present study.
Two methods exist by which to determine the level of lethality of a toxic release as a
function of concentration and time.
The first method uses probit functions, and the alternative method is based on the
Specified Level of Toxicity (SLOT equivalent to 1-5% fatality rate) and Significant
Likelihood of Death (SLOD equivalent to 50% fatality rate). In this study, SLOT and
SLOD, as described by Turner and Fairhurst (1993) and Franks, Harper and Bilio (1996),
respectively, and calculated using Equation 72, will be used, as shown in Table B.5.
Pipeline release durations typically depend on the time required to isolate and
depressurize or clamp the pipeline. Choking effects resulting from ice formation are not
considered. The studied release scenarios will use 60-minute release durations for both
pipelines.
n
C t=A (72)
Where C represents the concentration (ppm), t represents the exposure time (minutes),
and A represents the dangerous toxic load.
150
Table B.5. CO2 and H2S SLOT and SLOD Values
Substance SLOT SLOD

(Lethal Dose 1-5 %) (Lethal Dose 50%)
Carbon Dioxide (n=8) 1.5E40 1.5E41
Hydrogen Sulfide (n=4) 2E12 1.5E13
Pipeline 1 is a 30” diameter pipe that operates at 1900 psig. The pipeline’s stream
is a natural stream of CO2 => 95%. Although the pipeline should be classified as sour
service due to the partial pressure and concentrations of H2S (1500 ppm), the absence of
water is a strongly migratory measure against hydrogen damage. The CO2 release
scenario will assume a hole measuring 40 mm in diameter located aboveground;
however, because only 23% of the failure incidents occur aboveground, the failure rate in
Table 6.5 will be adjusted to 2.46E-5.
Pipeline 2 is a 24” diameter pipe that operates at 2500 psig. The pipeline’s stream
is an anthropogenic CO2. It will be assumed that this pipeline is designed for sour service
with a wet H2S stream of 9,000 ppm. The CO2 release scenario will assume a rupture
measuring 120 mm in diameter that creates a crater located 1.5 meters underground.
However, the rupture could be oriented away from the crater’s opening; therefore, to
maintain a high vapor discharge rate, the failure frequency in Table 6.5 will be adjusted
by a probability factor of 0.4 that the jet orientation is pointing towards the crater’s
opening (HSE, 2009) to 2.86E-5.
Release rates
The release of CO2 is considered a two-phase flow. A quantity of CO2 will flash
to the atmosphere; this will be modeled as continuous release. The remaining quantity of
CO2 that freezes and then sublimates will have release durations in the order of hours;
however, because the sublimation release rate is considerably lower, this aspect will not
be included in this study. The first step is to calculate the quantity of CO2 that would
151
flash using the isenthalpic analysis presented by Crowl and Louvar (1990), who
developed the flash fraction equation (Equation 73).
Fv = Cp (T - Tb / Hfg) (73)
Where:
Fv flash fraction
Cp average liquid heat capacity at (T - Tb)/2 and Cp = hb – h / Tb – T
T initial temperature (20 °C)
Tb final temperature (CO2 atmospheric boiling point -78.5 °C)
hfg heat of vaporization at Tb
hb enthalpy of liquid at the final temperature Tb = 239 kJ / kg
h enthalpy of vapor at the initial temperature T = 424 kJ / kg
A spreadsheet was created to calculate the liquid and gas release rates
corresponding to the operating pressures and leak sizes of Pipelines 1 and 2, as shown in
Table B.6. Regarding its dispersion, carbon dioxide is treated as a dense gas based on
Britter and McQuaid’s (1988) model. The first step, adopted from Britter and McQuaid
(1988), is to adjust the concentration for a non-isothermal release using Equation 74. The
second step is to calculate the volumetric discharge rate using Equation 75. The third
step is to derive the dimensionless group using the Britter-McQuaid correlation (Equation
77). The extent of SLOT concentration downstream of the CO2 leak’s location is
summarized in Table B.7.
152
Table B.6. Carbon Dioxide Pipeline Release Rates and Lethal Limits
Pipeline Pipeline 1 Pipeline 2

Hole Size (mm) 10 40 80 120 150
Total Release Rate (kg/s) 7.30 116.83 466.90 1205.03 1882.86
Liquid Quantity (kg/s) 4.95 79.21 316.56 817.01 1276.58
Release Duration (minutes) 60
Exposure Time (minutes) 10
SLOT Concentration (ppm) CO2 - 78,888 H2S – 668
Flash Fraction (%) 32.2% (Typical Estimates are 25% GPSA, 2004)
(74)
( )( ⁄ )
Where:
C* SLOT concentration
T initial temperature (Kelvin)
Tb final temperature (Kelvin)
(75)
Where:
qo initial jet volumetric flow rate
qs liquid spill rate (m3/s)
ρl liquid density (kg/m3)
ρo vapor density (kg/m3)
153
( ) (76)
Where:
go initial buoyancy factor (m3/s)
ρa atmospheric density (assumed as 1.224 kg/m3)
( ) ( )
Where:
µ wind speed (assumed as 5 m/s)
( )
⁄
( )
Where:
x downwind distance of concern
154
Table B.7. Carbon Dioxide Plume SLOT Concentration
Pipeline Pipeline 1 Pipeline 2

Hole Size (mm) 10 40 80 120 150
Effective Concentration, (ppm), (Eq. 7) 0.054
Spill Rate (m3/s) , (Eq.8) 1.27 20.40 81.61 210.65 329.14
Dimensionless Group (Eq.10) 0.16 0.28 0.36 0.44 0.48
Britter-McQuaid Correlation (Eq.11) 69 69 88 85 95
Downstream Distance (meters) 35 140 356 552 771
In this case, it is assumed that H2S vapor discharges at the same rate as CO2 from
an 80mm hole (Pipeline 1) and has a concentration of 9,000 ppm. After adjusting the
Cm/Co ratio in the Britter-McQuaid model, the H2S SLOT distance would be 410 meters,
about 15% more than CO2. A complete risk assessment study is beyond the scope of this
paper. Several probabilities that should be considered include wind speed, wind
direction, incident occurring close to a populated area, time of day of incident occurrence
and the number of people exposed. Furthermore, the Britter-McQuaid model is only
intended as a benchmark (Hanna and Drivas, 1996). Nevertheless, consequence analysis
indicates harmful levels of CO2 comparable to methane, except CO2 is hazardous without
an ignition source. A more serious danger in the bulk transportation of CO2 is H2S,
especially at trace concentrations above ~7,000 ppm, levels that typically occur in
anthropogenic CO2 streams aiming to reduce scrubbing costs.
155
APPENDIX C.
C. ANALYTICAL PROGRAM FOR 2-DIMENSIONAL TRANSPORTATION OF

A CHEMICAL COMPONENT
156
Based on Cleary and Ungs, 1978, , Javandel, and colleagues 1982, published the
following program called (TDAST) to analytically solve two dimensional transportation
of a chemical component.
C THIS PROGRAM EVALUATES THE TWO DIMENSIONAL ANALYTICAL

C SOLUTE TRANSPORT SOLUTION CONSIDERING CONVECTION,
C DISPERSION, DECAY AND ADSORPTION.
DIMENSION CD(100,20),X(100),Y(20),T(100),C(100,20)
COMMON NS(20)
COMMON/FAT/NNS
COMMON/DAT/DL,DT,V,A
COMMON/BAT/ALFA,ALAM,R
COMMON/CAT/XX,YY,TT,TT0
C ***************************************
C READ INPUT PARAMETERS
C NUMX = NUMBER OF X POSITIONS
C NUMY = NUMBER OF Y POSITIONS
C NUMT = NUMBER OF TIME POINTS
C NNS = NUMBER OF INTEGRATIONS TO ACHIEVE CONVERGENCE
C X = X COORDINATES OF THE POINTS
C Y = Y COORDINATES OF THE POINTS
C T = TIME ELAPSED SINCE THE BEGINNING OF OPERATION
C DL = LONGITUDINAL DISPERSION COEF M**2/DAY
C DT = TRANSVERSE DISPERSION COEF M**2/DAY
C V = PORE WATER VELOCITY M/DAY
C A = HALF LENGTH OF SOURCE M
C ALAM = DECAY FACTOR OF THE SOLUTE 1/DAY
C R = RETARDATION FACTOR
C ALFA = DECAY FACTOR OF THE SOURCE 1/DAY
C NS = NUMBER OF POINTS FOR INTEGRATION
C ***************************************
10 READ(5,510) NUMX,NUMY,NUMT,NNS
IF(NUMX.LT.1)STOP
READ(5,520) (X(I),I=1,NUMX)
READ(5,520) (Y(I),I=1,NUMY)
READ(5,520) (T(I),I=1,NUMT)
READ(5,530) DL,DT,V,A
READ(5,540) ALAM,R,ALFA
READ(5,550) (NS(I),I=1,NNS)
C ***************************************
C WRITE INPUT PARAMETERS
C ***************************************
WRITE(6,610) V,DL,DT,C0,A
WRITE(6,620) ALAM,R,ALFA
WRITE(6,630) NUMX,NUMY,NUMT
C ***************************************
DO 30 KK=1,NUMX
DO 20 JJ=1,NUMY
CD(KK,JJ)=0.0
20 CONTINUE
30 CONTINUE
TT0=0.0
DO 80 I=1,NUMT
157
TT=T(I)/R
DO 50 J=1,NUMX
XX=X(J)
DO 40 K=1,NUMY
YY=Y(K)
CALL CONC(CC0)
C(J,K)=CC0+C(J,K)
CD(J,K)=C(J,K)/EXP(ALFA*T(1))
40 CONTINUE
50 CONTINUE
WRITE(6,640) T(I),V,A,DL,DT,R,ALFA,ALAM
DO 70 IX=1,NUMY,6
LX=MIN0(IX+5,NUMY)
WRITE(6,650)(Y(L),L=IX,LX)
DO 60 J=1,NUMX
WRITE(6,660) X(J),(CD(J,K),K=IX,LX)
60 CONTINUE
70 CONTINUE
TT0=TT
80 CONTINUE
GO TO 10
C ***************************************
C FORMAT STATEMENTS
C ***************************************
510 FORMAT(4I5)
520 FORMAT(8F10.3)
530 FORMAT(4F10.4)
540 FORMAT(3F10.3)
550 FORMAT(10I5)
C
610 FORMAT(1H1,4X,21H*CONTROL INFORMATION*',//
1 43H VELOCITY(M/DAY)------------------------=F10.4/
2 43H LONGITUDINAL DISPERSION COEF.(M*M/DAY)-=F10.4/
3 43H TRANSVERSE DISPERSION COEF.(M*M/DAY)---=F10.4/
4 43H HALF LENGTH OF SOURCE (M)--------------=F10.4/)
620 FORMAT(1H,/
1 43H RADIOACTIVE DECAY CONSTANT(1/DAY)------=F10.4/
2 43H RETARDATION FACTOR---------------------=F10.4/
3 43H SOURCE DECAY FACTOR(1/DAY)-------------=F10.4/)
630 FORMAT(1H ,42H TOTAL NUMBER OF X POSITIONS------------ =I5/
1 43H'TOTAL NUMBER OF Y POSITIONS------------= I5/
2 43H'TOTAL NUMBER OF X POSITIONS------------= I5//)
640 FORMAT(1H1,10X,39HVALUES OF CONCENTRATION (C/C0) AT TIME T=
1 ,F7.1,6H DAYS/10X,27(2H**)//9X,2HV=,F5.3,5X,2HA=,F6.1,5X
2 ,3HDL=,F5.2,5X,3HDT=,F5.2,5X,2HR=,F5.2// 18X,5HALFA=,F6.4,
3 10X,7HLAMBDA=,F6.4)
650 FORMAT(//6X,1HX,7X,6(2HY=,F5.1,3X)/)
660 FORMAT(3X,F6.1,2X,6F10.5)
END
SUBROUTINE CONC(CC0)
COMMON NS(20)
COMMON/FAT/NNS
158
EXTERNAL D01BAZ,FUN1
PI=4*ATAN(1.)
I=1
ANS1=D01BAF(D01BAZ,TT0,TT,NS(I),FUN1,IFAIL)
90 I=I+1
IF(I.GT.NNS) GO TO 110
ANS2=D01BAF(D01BAZ,TT0,TT,NS(I),FUN1,IFAIL)
ANS=ABS(ANS1-ANS2)
ERR=ANS/AMAX1(ANS2,.1)
IF(ERR.LT.0.01) GO TO 100
ANS1=ANS2
GO TO 90
100 CONTINUE
CC0=ANS2
GO TO 120
110 CC0=ANS1
WRITE(6,670) TT,XX,YY
780 FORMAT(38H INTEGRAL DOES NOT CONVERGE AT TIME T=,F10.3,
1 3X,2HX=,F10.3,2HY=,F10.3)
120 RETURN
END
FUNCTION FUN1(X)
PI=4.ATAN(1.)
AA=(V**X/(2.*DL))-(ALAM*R-ALFA*R+(V**2)/(4.*DL))*X-
1 (XX**2)/(4.*DL*X))
AAA=EXP(AA)/SQRT(X**3)
BB=(A-YY)/(2.SQRT(DT*X))
CC=(-A-YY)/(2.SQRT(DT*X))
BBB=1.-ERF(BB)
CCC=1.-ERF(CC)
FUN1=AAA*(CCC-BBB)*(XXX/(4.SQRT(PI*DL)))
RETURN
END
FUNCTION ERF(X)
DIMENSION D(101)
N=100
N1=N+1
PI=4.*ATAN(1.)
C=2./SQRT(PI)
H=X/N
DO 130 I=1,N1
Y=(I-1)*H
130 D(1)=EXP(-Y*Y)
E1=0.0
DO 140 I=3,N1,2
140 E1=E1+(D(I-2) +4.*D(I-1)+D(I))*(H/3.)
ERF=C*E`
RETURN
END
159
APPENDIX D.
D. SIMULATION PROBLEM INPUT AND OUTPUT FILE

160
SIMULATION PROBLEM 6, CASE 1 (100% INJECTED BRINE) SAMPLE INPUT

FILE AND OUTPUT FILE
161
162
SIMULATION PROBLEM 6, CASE 1 (10% INJECTED BRINE) SAMPLE

INPUT FILE AND OUTPUT FILE
163
164
SIMULATION PROBLEM 6, CASE 1 (1% INJECTED BRINE) SAMPLE INPUT FILE

AND OUTPUT FILE
165
166

167
168

169
170

171
172

173
174

175
176

177
178

179
180

181
182

183
184

185
186

187
188

189
190
SIMULATION PROBLEM 6, CASE 6 (100%) INJECTED BRINE SAMPLE

191
192

193
194

195
196

197
198

199
200

201
202
SIMULATION PROBLEM 7, (1% INJECTED SEAWATER) SAMPLE INPUT FILE

AND OUTPUT FILE
203
204
SIMULATION PROBLEM 7, (1% INJECTED SEAWATER) SAMPLE INPUT

205
206
APPENDIX E.
E. SUMMARY OF SANDSTONE COREFLOODING EXPERIMENTS

207
Table E.1. Summary of Sandstone Coreflooding Experiments
Num
Secon Tertia Irreduc Forma Inject
ber Agi Cla Calc
Paper dary ry ible tion ion Aging Test Kaoli
of Visco ng IA IF y ite
Refere Recov Recov Water Brine Fluid Temper Temper nite
Plugs sity Tim H T (wt (wt
nce ery ery Saturat ions Catio ature ature (wt%)
/Core e %) %)
Runs Runs ion ns
s
Yildz,
et al. 13 13 0 √ √ √ √ √ √ √ X X X X X
(1999)
Austad,
et al. 1 1 1 √ √ √ √ √ √ √ X X √ X X
(2010)
Tang
and
√
Morro 21 21 3 √ √ √ √ √ √ √ X X X X
w
*
(1996)
Bousso
ur, et
1 1 0 √ √ √ √ √ √ √ X X √ √ √
al.
(2009)
Agbala
ka, et
16 16 80 √ √ √ √ √ √ √ X X X X X
al.
(2006)
Ashraf,
et al. 12 12 0 √ √ √ √ √ √ √ √ X X X X
(2010)
Roberts
on 23 23 0 √ √ √ √ √ X √ X X X X X
(2010)
Bernard
15 14 20 √ √ √ X X X X X X X X X
(1967)
Zhang
and
Morro 34 34 2 √ √ √ √ √ √ √ X X X X X
w
(2006)
Zhang,
et al. 2 11 10 √ √ √ √ √ √ √ X X X X X
(2007)
Ligthel
m, et al. 1 2 2 X √ √ √ √ X √ X X X X X
(2009)
Pu, et
al. 9 9 9 √ √ √ √ √ √ √ X X X X X
(2010)
Hadia,
et al. 14 14 28 √ √ √ √ √ √ √ √* X √* √* √*
(2011)
Gamag
e and
12 12 12 √ √ √ √ √ √ √ X X X X X
Thyne,
(2011)
Nasrall
a, et al. 8 8 0 √ √ √ √ √ X √ X X X X X
(2011)
208
Table E.1 Summary of Sandstone Coreflooding Experiments (Continued)
Thyne and Gamage

4 4 2 √ √ √ √ √ X √ X X √ X X
(2011)
Nasralla, et al.
8 8 6 √ √ √ √ √ X √ X X √ √ √
(2011)
Rivet, et al. (2010) 17 8 11 X √ √ √ √ √ √ X X √* X X
Sharma and Filoo

X 12 2 X √ √ √ X X X X X X X X
(2000)
Total
211 214 188 374 411 411 365 397 308 397 78 22 66 48 48
209
APPENDIX F.
F. SANDSTONE RESERVOIR CORE-FLOODING EXPERIMENT

CORRELATIONS
210
Figure F.1. Sandstone reservoir core-flooding experiment correlations

211
APPENDIX G.
G. RESIDUAL OIL SATURATION PREDICTION MODEL (EXCLUDING

WETTABILITY AND CLAY CONTENT)
212
H.
Figure H.1. Residual oil saturation prediction and profiler models

213
APPENDIX H.
I. PREDICTION MODEL FOR MISCIBLE CARBON DIOXIDE FLOODING

214
215
BIBLIOGRAPHY
Agbalaka, C.C., Dandekar, S.L., Khataniar, S., and Hemsath, J.R. (2008). Coreflooding
Studies to Evaluate the Impact of Salinity and Wettability on Oil Recovery
Efficiency. Trans. Porous Med. DOI 10.1007/s11242-008-9235-7.
Akin, S., Celebioglu, D., Kalfa, U. (2009). Optimum Tertiary Steam-Injection Strategies
for Oil-Wet Fractured Carbonates. Proceedings of the 2009 SPE Western
Regional Meeting, 24-26 March, San Jose, California, U.S.A., SPE 120168-MS,
DOI 10.2118/120168-MS.
Aladasani, A. and Bai, B. (2010). Recent Developments and Updated Screening Criteria
of Enhanced Oil Recovery Techniques. Proceedings of the CPS/SPE International
Oil and Gas Conference and Exhibition, Beijing, China, 8-10 June, SPE 130726,
DOI 10.2118/130726-MS.
Aladasani, A. and Bai, B. (2011). Analysis of EOR Projects and Updated Screening
Criteria. Journal of Petroleum Science and Engineering, 79, 1-2, 10-24, DOI
10.1016/j.petrol.2011.07.005.
Aladasani, A. Bai, B. and Nygaard R. (2012a). A Selection Criterion for CO2-Enhanced

Oil Recovery and Dispersion Modeling of High-Pressure CO2 Release.
Proceedings of the SPE Western Regional Meeting, Bakersfield, California, 21-
23 March, SPE 152998, DOI 10.2118/152998-MS.
Aladasani, A., Bai, B. and Wu, Y. (2012b). Investigating Low Salinity Water Flooding
Recovery Mechanism(s) in Sandstone Reservoirs: A Parametric Study Using
Reservoir Simulation and Statistical Analysis. Proceedings of the 18th SPE
Improved Oil Recovery Symposium, Tulsa, Oklahoma, USA, 14-18 April, SPE
152997.
Aladasani, A., Bai, B. and Wu, Y. (2012c). Investigating Low-Salinity Water Flooding
Recovery Mechanism(s) in Carbonate Reservoirs. Proceedings of the 2012 SPE
EOR Conference at Oil and Gas West Asia, Muscat, Oman, 16-18 April, SPE
155560-PP.
Albahlani, A.M., SPE, Babadagli, T. (2008). A Critical Review of the Status of SAGD:
Where Are We and What Is Next? Proceedings of the 2008 SPE Western
Regional and Pacific Section AAPG Joint Meeting, 29 March-2, Bakersfield,
California, U.S.A., 113283-MS, DOI 10.2118/113283-MS.
216
Algharaib, M., Gharbi, R., Malallah, A., Al-Ghanim, W. (2007). Parametric

Investigations of a Modified SWAG Injection Technique. Proceedings of the
15th SPE Middle East Oil and Gas Show and Conference, 11-14 March, Bahrain
International Exhibition Centre, Kingdom of Bahrain, SPE 105071-PP, DOI
10.2118/105071-MS.
Ali, S.M.F. and Thomas, S. (2000). Enhanced Oil Recovery - What We Have Learned,
Journal of Canadian Petroleum Technology, Volume 39, Number 2.
Alotaibi, M.B. and Nasr-El-Din, H.A. (2009). Effect of Brine Salinity on Reservoir
Fluids Interfacial Tension. Proceedings of the SPE EUROPEC/EAGE Annual
Conference and Exhibition, Amsterdam, The Netherlands, 8-11 June, SPE
121569, DOI 10.2118/121569-MS.
Alotaibi, R.M., Azmy, R.M. and Nasr-El-Din, H.A. (2010). A Comprehensive EOR
Study Using Low Salinity Water in Sandstone Reservoirs. Proceedings of the SPE
Improved Oil Recovery Symposium, Tulsa, Oklahoma, U.S.A., 24-28 April, SPE
122976, DOI 10.2118/129976-MS.
Altunina, L., Kuvshinov, V., Stasyeva, L. (2006). Improved Cyclic-Steam Well

Treatment with Employing Thermoreversible Polymer Gels. Proceedings of the
2006 SPE Russian Oil and Gas Technical Conference and Exhibition, 3-6
October, Moscow, Russia, SPE 104330, DOI 10.2118/104330-RU.
Anderson, W.G. (1986). Wettability Literature Survey – Part 6: The Effects of

Wettability on Waterflooding. Journal of Petroleum Technology, December, pp.
1605-1622.
Anderson, W.G. (1987). Wettability Literature – Part 1: Rock/Oil/Brine Interactions and

the Effects of Core Handling On Wettability. Journal of Petroleum Technology,
38, 10, 1125-1144, DOI 10.2118/13932-PA.
Anonymous. (1998). 1998 Worldwide EOR Survey. The Oil and Gas Journal, 96, 16, pp.
60-77.
Anonymous. (2000). 2000 Worldwide EOR Survey. The Oil and Gas Journal, 98, 12, pp.
46-61.
Anonymous. (2002). 2002 Worldwide EOR Survey. The Oil and Gas Journal, 100, 15,
pp. 72-83.
Anonymous. (2006). Special Report: 2006 Worldwide EOR Survey. The Oil and Gas
Journal, 104, 15, pp. 46-57.
217
Anonymous. (2007). World Energy Outlook 2007. International Energy Agency, pp. 76-
93.
Ashraf, A., Hadia, N.J. and Torsaeter, O. (2010). Laboratory Investigation of Low
Salinity Waterflooding as Secondary Recovery Process: Effect of Wettability.
Proceedings of the SPE Oil and Gas India Conference and Exhibition, Mumbai,
India, 20-22 January, SPE 129012, DOI 10.2118/129012-MS.
Austad, T. and Milter, J.(1997). Spontaneous Imbibition of Water into Low Permeable
Chalk at Different Wettabilities Using Surfactants. Proceedings of the 1997
International Symposium on Oilfield Chemistry, 18-21 February, Houston, Texas,
U.S.A., SPE 37236-MS, DOI 10.2118/129767-MS.
Austad, T., RezaeiDoust, A. and Puntervold, T. (2010). Chemical Mechanisms of Low

Salinity Water Flooding in Sandstone Reservoirs. Proceedings of the SPE
Improved Oil Recovery Symposium, Tulsa, Oklahoma, U.S.A., 24-28 April, SPE
129767, DOI 10.2118/129767-MS.
Austad, T., Strand, S., Hognesen, E.J., and Zhang, P. (2005). Seawater as IOR Fluid in
Fractured Chalk. Proceedings of the SPE International Symposium on Oilfield
Chemistry, Texas, U.S.A. 2-4 February, SPE 93000, DOI 10.2118/93000-MS.
Awan, A.R., Teigland, R., and Kleppe, J. (2006). EOR Survey in the North Sea.
Proceedings of the SPE/DOE Symposium on Improved Oil Recovery, 22-26 April
2006, Tulsa, Oklahoma, U.S.A., SPE 99546, DOI 10.2118/99546-PA.
Aziz, K. and Settari, A. (1979). Petroleum Reservoir Simulation, Applied Science

Publishers LTD, London.
Babadagli, T. (2003). Analysis of Oil Recovery by Spontaneous Imbibition of Surfactant

Solution. Proceedings of the 2003 SPE International Improved Oil Recovery
Conference in Asia Pacific, 20-21 October, Kuala Lumpur, Malaysia, SPE 84866-
MS, DOI 10.2118/84866-MS.
Bai, B., Li, L., Liu, Y., Liu, H., Wang, Z., and You, C. (2004). Preformed Particle Gel for
Conformance Control: Factors Affecting its Properties and Applications.
Proceedings of the SPE/DOE Symposium on Improved Oil Recovery, 17-21 April
2004, Tulsa, Oklahoma, U.S.A., SPE 89389, DOI 10.2118/89389-PA.
Bai, B., Li, L., Liu, Y., Liu, H., Wang, Z., and You, C. (2007). Conformance control by
preformed particle gel: Factors affecting its properties and applications. SPE Res.
Eval. Eng. 10, 415.
218
Bai, B., Liu, Y., Coste, J.-P., and Li, L. (2007). Preformed particle gel for conformance
control: Transport mechanism through porous media. SPE Res. Eval. Eng. 10,
176.
Bagci, A.S., Olushola, S., and Mackay, E. (2008). Performance Analysis of SAGD Wind-
Down Process with CO2 Injection. Proceedings of the 2008 SPE Improved Oil
Recovery Symposium, 19-23 April, Tulsa, Oklahoma, U.S.A., SPE 113234, DOI
10.2118/113234-MS.
Bagci, S., Kok, M.V. and Turksoy, U. (2001). Effect of Brine Composition on Oil
Recovery by Waterflooding. Journal of Petroleum Science and Technology, 19,
3-4, 359-372, DOI 10.1081/LFT-100000769.
Behzadi, S.H. and Towler, B.F. (2009). ASPaM, A New EOR Method. Proceedings of
the 2009 Annual Technical Conference and Exhibition, 4-7 October, New
Orleans, Louisiana, U.S.A., SPE 123866-PP, DOI 10.2118/123866-MS.
Belandria, V. and Balza, A. (2006). Heavy oil and chemical additive interaction for EOR
thermal applications. Proceedings of the 2006 SPE Intelligent Energy Conference
and Exhibition, 11-13 April, Amsterdam, The Netherlands, SPE 108398.
Belgrave, J.D.M., Nzekwu, B., Chhina, H.S. (2007). SAGD Optimization with Air
Injection. Proceedings of the 2007 Latin American and Caribbean Petroleum
Engineering Conference, 15-18 April, Buenos Aires, Argentina, SPE 106901-MS,
DOI 10.2118/106901-MS.
Bernard, G.C., Holm, L.W., and Harvey, C.P. (1980). Use of Surfactant to Reduce CO2
Mobility in Oil Displacement. SPE Journal, 20, 4, pp. 281-292.
Boussour, S., Cissokho, M., Cordier, P., Bertin, H. and Hamon, G. (2009). Oil Recovery
by Low Salinity Brine Injection: Laboratory Results on Outcrop and Reservoir
Cores. Proceedings of the 2009 SPE Annual Technical Conference and
Exhibition, New Orleans, Louisiana, USA, 4-7 October, SPE 124277, DOI
10.2118/124277-MS.
BP Trading Limited (1979). Enhanced Oil Recovery by Displacement with Saline

Solutions. Gulf Publishing Company, Houston, Texas, ISBN 0-87201-263-8.
British Petroleum (2009). Less Salt More Oil. Frontiers, 25, 6-9.
Britter, R.E. and McQuaid, J. (1988). Workbook on the Dispersion of Dense Gases.
Research Report 17/1988, Health and Safety Executive.
http://www.hse.gov.uk/research/crr_pdf/1988/crr88017.pdf (accessed
01/21/2012).
219
Buchgraber, M., Clemens, T., Casanier, L.M., Jovscek, A.R. (2009). The Displacement
of Viscous Oil by Associative Polymer Solutions. Proceeding of the 2009 SPE
ATCE Annual Conference and Exhibition, 4-7 October, New Orleans, Louisiana,
U.S.A., SPE 122400.
Buckley, J.S., Takamura, K. and Morrow, N.R. (1989). Influence of Electrical Charges
on the Wetting Properties of Crude Oils, SPE Reservoir Engineering, 4, 3, 332-
340, DOI 10.2118/16964-PA.
Buckley, S.E. and Leverett, M.C. (1942). Mechanism of Fluid Displacement in Sands,
Trans. AIME, 146, pp. 107-116.
Cadelle, C.P., Burger, J.G., Bardon, C.P., Machedon, V., Carcoana, A., and Petcovici, V.
(1981). Heavy-Oil Recovery by In-Situ Combustion – Two Field Cases in
Romania. Proceedings of the SPE 50th California Regional Meeting, April 9-11,
1980, Los Angeles, California, U.S.A., SPE 8905, DOI 10.2118/8905-PA.
Calhoun, J.C., Jr., Stahl, C.D., Preston, F.W., and Nielson, R.F. (1951). A Review of
Laboratory Experiments on Wetting Agents for Waterflooding. Production
Monthly, 16, 15.
Carageorgos, T., Marotti, M., and Bredrikovetsky, P. (2008). A New Method to

Characterize Scaling Damage from Pressure Measurements. Proceedings of the
2008 SPE International Symposium and Exhibition on Formation Damage
Control, 13-15 February, Lafayette, Louisiana, USA, SPE 112500, DOI
10.2118/112500-MS.
Center for Chemical Process Safety (1999). Guidelines for Chemical Process
Quantitative Risk Analysis (2nd ed.). American Institute of Chemical Engineers,
New York, NY. ISBN 978-0-8169-0720-5.
Chang, H.L., Zhang, Z.Q., Wang, Q.M., Xu, Z.S., Guo, Z.D., Sun, H.Q., Cao, X.L., and
Qiao, Q., (2006). Advances in Polymer Flooding and Alkaline/Surfactant/Polymer
Processes as Developed and Applied in the People’s Republic of China. Journal
of Petroleum Technology, 58, 2, pp. 84-89.
Chattopadhyay, S., Jain, V. and Sharm, M.M. (2002). Effect of Capillary Pressure,
Salinity, and Aging on Wettability Alteration in Sandstones and Limestones.
Proceedings of the SPE/DOE Improved Oil Recovery Symposium, Tulsa,
Oklahoma, SPE 75189, DOI 10.2118/75189-MS.
Cheng Jiecheng, Sui Xinguang, Bai Wenguang, Luo Jiangtao, (2007). Case Studies on
Polymer Flooding for Medium Reservoirs in Daqing Oilfield. Proceedings of the
2007 SPE Asia Pacific Oil and Gas Conference and Exhibition, 30 October – 1
November, Jakarta, Indonesia, SPE 108661-PP, DOI 10.2118/108661-MS.
220
Cissokho, M., Boussour, S., Cordier, P., Bertin, H., and Hamon, G. (2009). Low salinity
oil recovery on clayey sandstone: Experimental study. Proceedings of the
International Symposium of the Society of Core Analysts, Noordwijk aan Zee,
The Netherlands, 27-30 September, SCA2009-05.
Clementz, D.M. (1976). Interaction of Petroleum Heavy Ends with Montmorillonite.

Clays and Clay Minerals, Volume 34, pp. 312-319.
Clementz, D.M. (1982). Alteration of Rock Properties by Adsorption of Petroleum Heavy

Ends: Implications for Enhanced Oil Recovery. Proceedings of the SPE/DOE
Third Joint Symposium on Enhanced Oil Recovery, Tulsa, Oklahoma, U.S.A., 4-7
April, SPE/DOE 10683, DOI 10.2118/10683-MS.
Coats, K.H., Nielsen, R.L., Terhune, M.H., and Weber, A.G. (1967). Simulation of
Three-Dimensional, Two-Phase Flow in Oil and Gas Reservoirs. SPE Journal,
Volume 7, Number 4, pp. 377 – 388.
Conservation of Clean Air and Water in Europe (2010). Performance of European

Crosscountry Oil Pipelines.
http://www.concawe.be/DocShareNoFrame/docs/1/NOBIJBPAJFHDCBJHHKG
CNIONVEVCWY9W9YBYB3BDWYG3/CEnet/docs/DLS/Rpt_11-8-2010-
00114-01-E.pdf (accessed 01/21/2012).
Corey A.T. (1954).The Interrelations Between Gas and Oil Relative Permeabilities.
Producers Monthly, 19, 38-41.
Craig, F.F., Jr. (1971). The Reservoir Aspects of Waterflooding, SPE Dallas, Texas,
U.S.A.
Craig, B.D. (2003). Sour Gas Design Considerations, Society of Petroleum Engineers,
ISBN 1555630448.
Crowl, D.A. and Louvar, J.F. (1990). Chemical Process Safety: Fundamentals with
Applications. Englewood Cliffs, NJ: Prentice Hall, ISBN 0-13-129701-5.
Demin, W., Jiecheng, C., Junzheng W., Zhenyu, Y., Yuming, Y., and Hongfu, L. (1999).
Pilot Tests of Alkaline/Surfactant/Polymer Flooding in Daqing Oil Field. SPE
Reservoir Engineering, 12, 4, pp. 229-233.
Demin, W., Jiecheng, C., Qun, L., Junzheng, W., Wenxiang, W., and Yanqing, Z. (2001).
Summary of ASP Pilots in Daqing Oil Field. Proceedings of the SPE Asia Pacific
Conference on Improved Oil Recovery, 25-26 October 1999, Kuala Lumpur,
Malaysia, SPE 57288, DOI 10.2118/57288-MS.
Denbury (2011). Operations CO2 Pipelines. http://www.denbury.com/index.php?id=18

(accessed 01/21/2012).
221
Department of Energy (2005). “Oil Exploration and Production Program, Enhanced Oil
Recovery.”
Dietrichm, F.L., Brown, F.G., Zhou, Z.H., and Maure, M.A. (1996). Microbial EOR
Technology Advancement: Case Studies of Successful Projects. Proceedings of
the SPE Annual Technical Conference and Exhibition, 6-9 October 1996, Denver,
Colorado, U.S.A., SPE 36746, DOI 10.2118/36746-MS.
DNV Technica (2001). Human Resistance Against Thermal Effects, Explosion Effects,
Toxic Effects and Obstruction of Vision. http://www.preventor.no/tol_lim.pdf
(accessed 01/21/2012).
Doctor, R., Palmer, A., Coleman, D., Favison, J., Hendriks, C., Kaarstad, O., Ozaki, M.
(2005). Transport of CO2. In: Metz, B., Davidson, O., de Coninck, H., Loos M.,
and Meyer, L. (Eds.), Special Report on Carbon Dioxide Capture and Storage.
Cambridge University Press, Cambridge, UK.
Dong, M., Foraie, J., Huang, S., and Chatzis, I. (2002). Analysis of Immiscible Water-
Alternating-Gas (WAG) Injection Using Micromodel. Proceedings of the
Canadian International Petroleum Conference, Jun 11 - 13, Calgary, Alberta,
Canada, 2002-158, DOI 10.2118/2002-158.
Dubey, S.T. and Doe, P.H. (1993). Base Number and Wetting Properties of Crude Oils.
SPE Reservoir Engineering, August, pp. 195-200.
Dunning, H.N. and Hsiso, L. (1953). Experiments with Detergents as Waterflooding

Additives. Production Monthly. 18, 24.
Edmunds, N.R., Kovalsky, J.A., Gittins, S.D., and Pennacchioli, E.D. (1994). Review of
Phase A Steam-Assisted Gravity-Drainage Test. SPE Reservoir Engineering
Journal, 9, 2, pp. 119-124.
E-ON (2010). Consequence Assessment of CO2 Pipeline Releases, KCP-ENT-SHE-

REP-0004, Revision 2. http://www.decc.gov.uk/assets/decc/11/ccs/chapter8/8.8-
consequence-assessment-of-co2-pipeline-releases.pdf (accessed 01/21/2012).
Ertekin, T., Abou-Kassem, J.H., and King, G.R. (2001). Basic Applied Reservoir
Simulation. Richardson, Texas: Society of Petroleum Engineers.
European Gas Incident Group (2011). 8th Report of the European Gas Pipeline Incident
Data Group. Document Number EGIG 11.R.0402 (Version 2). http://www.egig.eu
(accessed 01/21/2012).
Fairchild, I., Hendry, G., Quest, M. and Tucker, M. (1988). Geochemical Analysis of
Sedimentary Rocks. In: Techniques in Sedimentology (M. E. Tucker, Ed.).
Blackwells, Oxford 274-354.
222
Fairhurst, S. and Turner, R.M. (1993). Toxicological Assessments in Relation to Major

Hazards. Journal of Hazardous Materials, 33, 215-227.
Faisal, A., Bisdom, K., Zhumabek, B., Mojaddam Zadeh, A., and Rossen, W.R. (2009).
Injectivity and Gravity Segregation in WAG and SWAG Enhanced Oil Recovery.
Proceedings of the 2009 SPE Annual Technical Conference and Exhibition, 4-7
October, New Orleans, Louisiana, U.S.A., SPE 124197, DOI 10.2118/124197-
MS.
Fjelda, I. (2008). Low Salinity Water Flooding Experimental Experience and Challenges.
Force Workshop: Low Salinity Water Flooding, The Importance of Salt Content
in Injection Water, Stavanger, Norway, 15 May, IF 23.
Flumerfelt, R.W., Li, X., Cox, J.C., Hsu, W.-F. (1993). A Cyclic Surfactant-Based
Imbibition/Solution Gas Drive Process for Low-Permeability, Fractured
Reservoirs. Proceedings of the 1993 SPE Annual Technical Conference and
Exhibition, 3-6 October, Houston, Texas, U.S.A., SPE 26373-MS, DOI
10.2118/26373-MS.
Forsyth, P. A., Wu, Y.S., and Pruess, K. (1995). Robust Numerical Methods for
Saturated-Unsaturated Flow with Dry Initial Conditions in Heterogeneous Media,
Advances in Water Resources, Vol. 18, pp. 25-38.
Frampton, H., Morgan, J.C., Cheung, S.K., Munson, L., Chang, K.T., and Williams, D.
(2004). Development of a novel waterflood conformance control system.
Proceedings of the 2004 SPE/DOE Symposium on Improved Oil Recovery, 17-21
April, Tulsa, Oklahoma, U.S.A., SPE 89391-MS, DOI 10.2118/89391-MS.
Franks, A., Harper, P., and Bilio, M. (1996). The Relationship Between Risk of Death
and Risk of Dangerous Dose for Toxic Substances, Journal of Hazardous
Materials, 51, 11-26.
Fullbright, G.D., Hild, G.P., Korf, T.A., Myers, J.P., O'Toole, F.S., Wackowski, R.K.,
and Smith, M.E. (1996). Evolution of Conformance Improvement Efforts in a
Major CO2 WAG Injection Project. Proceedings of the 1996 SPE/DOE Improved
Oil Recovery Symposium, 21-24 April, Tulsa, Oklahoma, U.S.A., SPE 35361-
MS, DOI 10.2118/35361-MS.
Galvo, E.V.R.P., Rodrigues, M.A.F., Barillas, J.L.M., Dutra, T.V., Jr., and da Mata, W.
(2009). Optimization of Operational Parameters on Steamflooding with Solvent in
Heavy Oil Reservoirs. Proceedings of the 2009 SPE Latin American and
Caribbean Petroleum Engineering Conference, 31 May – 3 June, Cartagena,
Columbia, SPE 122078-PP, DOI 10.2118/122078-MS.
223
Gamage, P. and Thyne, G. (2011). Comparison of Oil Recovery of Low Salinity

Waterflooding in Secondary and Tertiary Recovery Modes. Proceedings of the
SPE Annual Technical Conference and Exhibition, Denver, Colorado, USA, 30
October-2 November 2011, SPE 147375, 10.2118/147375-MS, DOI
10.2118/147375-MS.
Gang, C., He, L., Guochen, S., Guoqing, W., Zhongwen, X., Huaifeng, R., and Ying, Z.
(2007). Technical Breakthrough in PCPs’ Scaling Issue of ASP Flooding in
Daqing Oil Field. Proceedings of the 2007 SPE Annual Technical Conference and
Exhibition, 11-14 November, Anaheim, California, U.S.A., SPE 109165-PP, DOI
10.2118/109165-MS.
Gaucher, D.H. and Lindley, D.C. (1960). Waterflood Performance in a Stratified, Five-
Spot Reservoir - A Scaled-Model Study. Petroleum Transactions, AIME, Volume
219, 1960, pp. 208-215.
Gasunie, N.V. Dedarlandse (2009). Towards a Transport Infrastructure for Large-Scale

CCS in Europe, CO2 Europe.
http://www.co2europipe.eu/Publications/D3.1.2%20-
%20Standards%20for%20CO2.pdf (accessed 01/21/2012).
Ghomian, Y., Pope, G.A., and Sepehrnoori, K. (2008). Development of a Response

Surface Based Model for Minimum Miscibility Pressure (MMP) Correlation of
CO2 Flooding. Proceedings of the SPE Annual Technical Conference and
Exhibition, 21–24 September 2008, Denver, Colorado, U.S.A., SPE 116719, DOI
10.2118/116719-MS.
Goa, C.H., Zekri, A., and El-Tarabily, K. (2009). Microbes Enhance Oil Recovery
Through Various Mechanisms. Oil and Gas Journal, 107, 31, 39-43.
Gonzalez da Silva, I.P., Aparecida de Melo, M., Luvizotto, J.M., and Lucas, E.F. (2007).
Polymer Flooding: A Sustainable Enhanced Oil Recovery in the Current Scenery.
Proceedings of the 2007 SPE Latin American and Caribbean Petroleum
Engineering Conference, 15-16 April, Buenos Aires, Argentina, SPE 107727.
Gorell, S.B. (1988). Modeling the Effects of Trapping and Water Alternate Gas (WAG)
Injection on Tertiary Miscible Displacements. Proceedings of the 1988 SPE
Enhanced Oil Recovery Symposium, 16-21 April, Tulsa, Oklahoma, U.S.A., SPE
17340-MS, DOI 10.2118/17340-MS.
Gozalpour, F., Ren, S.R., and Tohidi, B. (2005). CO2 EOR and Storage in Oil
Reservoirs. Oil and Gas Science and Technology, 60, 3, pp. 537-546.
224
Gray, M.R., Yeung, A., Foght, J.M., and Yarranton, H.W. (2008). Potential Microbial
Enhanced Oil Recovery Processes: A Critical Analysis. Proceedings of the SPE
Annual Technical Conference and Exhibition, 21-24 September 2008, Denver,
Colorado, U.S.A., SPE 114676, DOI 10.2118/114676-MS.
Green, D.W. and Wilhite, G.P. (1998) Enhanced Oil Recovery, SPE Dallas, Texas,
U.S.A.
Guerra, E., Valero, E., Rodriguez, D., Gutierrez, L., Castillo, M., Espinoza, J., and
Granja, G. (2007). Improved ASP Design Using Organic Compound-Surfactant-
Polymer (OCSP) for La Salina Field, Maracaibo Lake. Proceedings of the 2007
SPE Latin American and Caribbean Petroleum Engineering Conference, 15-18
April, Buenos Aires, Argentina, SPE 107776, DOI 10.2118/107776-MS.
Gupta, R. and Mohanty, K.K. (2008). Wettability Alteration of Fractured Carbonate

Reservoirs. Proceedings of the 2008 SPE/DOE Symposium on Improved Oil
Recovery, 20-23 April, Tulsa, Oklahoma, U.S.A., SPE 113407-MS, DOI
10.2118/113407-MS.
Gupta, R., Adibhatla, B., and Mohanty, K.K. (2008). Parametric Study to Enhance Oil
Recovery Rate from Fractured Oil-Wet Carbonate Reservoirs. Proceedings of the
2008 SPE Annual Technical Conference and Exhibition, 21-24 September,
Denver, Colorado, U.S.A., SPE 116485-MS, DOI 10.2118/116485-MS.
Gupta, R., Smith Jr.,P.G.,Hu, L., Willingham, T.W., Cascio, M. L., Shyeh, J.J. and
Harris, C.R. (2011). Enhanced Waterflood for Middle East Carbonate Cores –
Impact of Injection Water Composition. Proceedings of the SPE Middle East Oil
and Gas Show and Conference, Manama, Bahrain, 25-28 September, SPE
142668, DOI 10.2118/142668-MS.
Hadia, N., Lehne, H.H., Kumar, K. G., Selboe, K., Stensen, J. A. and Torsaeter, O.
(2011). Laboratory Investigation of Low Salinity Waterflooding on Reservoir
Rock Samples from the Froy Field. Proceedings of the SPE Middle East and Gas
Show and Conference, Manama, Bahrain, 25-28 September, SPE 141114, DOI
10.2118/141114-MS.
Han, M., Xiang, W., Zhang, J., Jiang, W., and Sun, F. (2006). Application of EOR
Technology by Means of Polymer Flooding in Bohai Oilfields. Proceedings of the
2006 SPE International Oil and Gas Conference and Exhibition, 5-7 December,
Beijing, China, SPE 104432, DOI 10.2118/104432-MS.
Hanna, S.R. and Drivas, D.G. (1996). Guidelines for Use of Vapor Cloud Dispersion
Models (2nd ed.). Wiley-American Institute of Chemical Engineers. ISBN 0-
8169-0702-1.
225
Health and Safety Executive (2008). Pipeline Design Codes and Standards for Use in UK
CO2 Storage and Sequestration Projects.
http://www.hse.gov.uk/pipelines/resources/designcodes.htm (accessed
01/21/2012).
Health and Safety Executive, (2009). Comparison of risks from carbon dioxide and
natural gas pipelines.
Research Report RR749. http://www.hse.gov.uk/research/rrpdf/rr749.pdf (accessed

01/21/2012).
Health and Safety Executive (2010). Failure Rate and Event Data for Use Within Land
Use Planning Risk Assessments. http://www.hse.gov.uk/landuseplanning/failure-
rates.pdf (accessed 01/21/2012).
Hognesen, E.J., Strand, S. and Austad, T. (2005). Waterflooding of Preferential Oil-Wet

Carbonates: Oil Recovery Related to Reservoir Temperature and Brine
Composition. Proceedings of the SPE Europec/EAGE Annual Conference,
Madrid, Spain, 13-16 June, SPE 94166, DOI 10.2118/94166-MS.
Honarpour, M., Koederitz, L., and Harvey, A.H. (1986). Relative Permeability of
Petroleum Reservoirs, CRC Press, Inc., Boca Raton, Florida, 1986.
Honarpour, M.M., McGee, K.R., Crocker, M.E., Maerefat, N.L. and Sharma, B. (1986).
Detailed Core Description of a Dolomite Sample From the Upper Madison
Limestone Group. Proceedings of the SPE Rocky Mountain Regional Meeting,
Billings, Montana, USA, 19-21 May, SPE15174, DOI 10.2118/15174-MS.
Hongfu, L., Guangzhi, L., Peihui, H., Zhenyu, Y., Xiaoline, W., Guangyu, C., Dianping,
X., and Peiqiang, J. (2003). Alkaline/Surfactant/Polymer (ASP) Commercial
Flooding Test in Central Xing2 Area of Daqing Oilfield. Proceedings of the SPE
Asia Pacific Conference on Improved Oil Recovery, 20-21 October 2003, Kuala
Lumpur, Malaysia, SPE 84896, DOI 10.2118/84896-MS.
Hung, H.C. and Shreve, G.S. (2001). Effect of the Hydrocarbon Phase on Interfacial and
Thermodynamic Properties of Two Anionic Glycolipid Biosurfactants in
Hydrocarbon/Water Systems. Journal of Physical Chemistry, B, 105, pp. 12596-
12600.
Israelachvili, J. N. (1991). Intermolecular and Surface Forces. Academic Press, 2nd Ed.,
San Diego, California, ISBN-10: 0123751810.
Jadhunandan, P.P. and Morrow, N.R. (1991). Spontaneous Imbibition of Water by Crude
Oil/Brine/Rock System. InSitu, Volume 15, No. 4, pp. 319-345.
226
Javandel, I., Doughty, C., and Tsang, C.F. (1984). Groundwater Transport Handbook of
Mathematical Models. American Geophysical Union, Water Resources
Monograph 10, Washington, D.C., 1984.
Jelgersma, F. (2007). Redevelopment of the Abandoned Dutch Onshore Schoonebeek

Oilfield with Gravity Assisted Steam Flooding. Proceedings of the International
Petroleum Technology Conference, 4-6 December, Dubai, U.A.E., IPTC 11700-
PP, DOI 10.2523/11700-MS.
Jerauld, G.R., Lin, C.Y., Webb, K.J. and Seccombe, J.C. (2006). Modeling Low-Salinity
Waterflooding. Proceedings of the SPE Annual Technical Conference and
Exhibition, San Antonio, Texas, 24-27 September, SPE 102239, DOI
10.2118/102239-MS.
Jianjun, X., Buqian, S., Jianhua, Z., Jiangli, Guosheng, L., and Yiming, Y. (2006). New
Technology for Raising the Production in Injecting Steam Well by Using SEPA.
Proceedings of the 2006 SPE International Oil and Gas Conference and
Exhibition, 5-7 December, Beijing, China, SPE 104404, DOI 10.2118/104404-
MS.
JMP® 9 (2010). SAS Institute Inc., Cary, NC, ISBN 978-1-60764-599-3.
JMP® 9 Modeling and Multivariate Methods (2010). SAS Institute Inc., Cary, NC, ISBN
978-1-60764-595-5.
JMP® 9 Basic Analysis and Graphing (2010). SAS Institute Inc., Cary, NC, ISBN 978-1-
60764-596-2.
Johnson, S.J., Salehi, M., Eisert, K.E., Liang, J.T., Bala, G., Fox, S.L. (2007).
Biosurfactants Produced from Agriculture Process Waste Streams to Improve Oil
Recovery in Fractured Carbonate Reservoirs. Proceedings of the 2007 SPE
International Symposium on Oilfield Chemistry, 28 February – 2 March,
Houston, Texas, U.S.A., SPE 106078, DOI 10.2118/106078-MS.
Karoussi, O. and Hamouda, A.A. (2008). Imbibition of Sulfate and Magnesium Ions into
Carbonate Rocks at Elevated Temperatures and Their Influence on Wettability
Alteration and Oil Recovery. Energy Fuels, 22, 3, 2129–2130.
Kinder Morgan (2011). Kinder Morgan CO2 Transportation.

http://www.kindermorgan.com/business/co2/transport.cfm (accessed 01/21/2012).
King, R.F. (1993). Drilling Sideways -- A Review of Horizontal Well Technology and Its
Domestic Application. Energy Information Administration, DOE/EIA-TR-0565.
Koottungal, L. (2008). 2008 Worldwide EOR Survey. The Oil and Gas Journal, 106, 15,
pp. 47-59.
227
Koottungal, L. (2010). 2010 Worldwide EOR Survey, The Oil and Gas Journal, 108, 14,
41-53.
Kowalewski, E., RueslÃ¥tten, I., Boassen, T., Sunde, E., Stensen, J.Ã., LillebÃ, B.L.P.,
BÃdtker, G., and Torsvik, T. (2005). Analyzing Microbial Improved Oil
Recovery Processes from Core Floods. Proceedings of the 2005 International
Petroleum Technology Conference, 21-23 November, Doha, Qatar, IPTC 10924,
DOI 10.2523/10924-MS.
Krumrine, P.H., Falcone, J.S., and Campbell, T.C. (1982). Surfactant Flooding 2*: The
Effect of Alkaline Additives on Permeability and Sweep Efficiency. SPE Journal,
22, 6, pp. 983-992.
Kurawle, I., Kaul, M., Mahalle, N., Nair, A., Kulkarni, N., Amin, Z., Carvalho, V.
(2009). Bio-Genetic Engineering, Future of Multi Microbial Culture EOR – A
Detailed Report. Proceedings of the 2009 SPE Offshore Europe Oil and Gas
Conference and Exhibition, 8-11 September, Aberdeen, U.K., SPE 123303, DOI
10.2118/123303-MS.
Lager, A., Webb, K.J., Black, C.J.J., Singleton, M. and Sorbie, K.S. (2006). Low Salinity
Oil Recovery - An Experimental Investigation. Proceedings of the International
Symposium of the Society of Core Analysts, Trondheim, Norway, 12-16
September, SCA 2006-36.
Lager, A., Webb, K.J., and Black, C.J.J. (2007). Impact of Brine Chemistry on Oil
Recovery. Proceedings of the 14th European Symposium on Improved Oil
Recovery, Cairo, Egypt, 22-24 April, Paper A24.
Lager, A., Webb, K.J., Collins, I.R. and Richmond, D.M. (2008a). LoSalTM Enhanced
Oil Recovery: Evidence of Enhanced Oil Recovery at the Reservoir Scale.
Proceedings of the SPE/DOE Symposium on Improved Oil Recovery, Tulsa,
Oklahoma, USA, 20-23 April, SPE 113976, DOI 10.2118/113976-MS.
Lager, A., Webb, K. J., Black, C. J. J., Singleton, M., Sorbie, K. S. (2008b). Low Salinity
Oil Recovery – An Experimental investigation. Petrophysics, 49, 1, 28-35, DOI:
2008-v49n1a2.
Lakatos, I., Toth, J., Bodi, T., Lakatos-Szabo, J., Berger, P.D., and Lee, C. (2007).
Application of Viscoelastic Surfactants as Mobility Control Agents in Low-
Tension Surfactant Floods. Proceedings of the 2007 SPE International
Symposium on Oilfield Chemistry, 28 February – 2 March, Houston, Texas,
U.S.A., SPE 106005, DOI 10.2118/106005-MS.
228
Lalehrokh, F. and Bryant, S.L. (2009). Application of pH-Triggered Polymers for Deep
Conformance Control in Fractured Reservoirs. Proceedings of the 2009 SPE
Annual Technical Conference and Exhibition, 4-7 October, New Orleans,
Louisiana, U.S.A., SPE 124773-MS, DOI 10.2118/124773-MS.
Lee, S. Y., Webb, K. J., Lager, A., Clarke, S. M., O’Sullivan, A. F. and Wang, X. (2010).
Proceedings of the 2010 SPE Improved Oil Recovery Symposium, Tulsa,
Oklahoma, 24-28 April, SPE 129722, DOI 10.2118/129722-MS.
Le, V.A., Nguyen, Q.P., and Sanders, A.W. (2008). A Novel Foam Concept with CO2
Dissolved Surfactants. Proceedings of the 2008 SPE Improved Oil Recovery
Symposium, 19-23 April, Tulsa, Oklahoma, U.S.A., SPE 113370, DOI
10.2118/113370-MS.
Lemon, P., Zeinijahromi, A., Bedrikovetsky, P. and Shahin, I. (2011). Effects of Injected
Water Sweep Efficiency via Induced Fines Migration. Proceeds of the SPE
International Symposium on Oilfield Chemistry, 11-13 April, The Woodlands,
Texas, U.S.A., SPE 140141, DOI 10.2118/140141-MS.
Levitt, D.B. and Pope, G.A. (2008). Selection and Screening of Polymers for Enhanced-
Oil Recovery. Proceedings of the 2008 SPE Improved Oil Recovery Symposium,
19-23 April, Tulsa, Oklahoma, U.S.A., SPE 113845, DOI 10.2118/113845-MS.
Li, H.-F., Xu D.-P., Jiang, J., Du, X.-W., Hong, J.-C., Jiang, Y., and Xu, Y.-S. (2008).
Performance and effect analysis of ASP commercial flooding in Centeral Xing2
Area of Daqing oil field. Proceedings of the 2008 SPE Improved Oil Recovery
10.2118/114348-MS.
Ligthelm, D.J., Gronsveld, J., Hofman, J. P., Brussee, N. J., Marcelis, F. and Van der
Linde, H.A. (2009). Novel Waterflooding Strategy by Manipulation of Injection
Brine Composition. Proceedings of the SPE EUROPEC/EAGE Annual
Conference and Exhibition, Amsterdam, The Netherlands, 8-11 June, SPE
119835, DOI 10.2118/119835-MS.
Li-qiang, Y., Da-sheng, Z., and Yu-huan, S. (2006). SAGD as follow-up to cyclic steam
stimulation in a medium deep and extra heavy oil reservoir. Proceedings of the
2006 SPE International Oil and Gas Conference and Exhibition, 5-7 December,
Beijing, China, SPE 104406-PP, DOI 10.2118/104406-MS.
Liu, R., Liu, H., Li, X., Fan, Z. (2008). The Reservoir Suitability Studies of Nitrogen
Foam Flooding in Shengli Oilfield. Proceedings of the 2008 SPE Asia Pacific Oil
and Gas Conference and Exhibition, 20-22 October, Perth, Australia, SPE
114800-PP, DOI 10.2118/114800-MS.
229
Liu, Y., Grigg, R. B. and Bai. B. (2005). Salinity, pH, and Surfactant Concentration
Effects on CO2-Foam. Proceedings of the 2005 SPE International Symposium on
Oilfield Chemistry, Houston, Texas, 2-4 February, SPE 93095, DOI
10.2118/93095-MS.
Liu, Y., Bai, B., and Wang, Y. (2010). Applied Technologies and Prospect of
Conformance Control Treatments in China. Oil and Gas Science and Technology
65 (1): 1-20. DOI: 10.2516/ogst/2009057.
Lopez L.G., Rivas, A., Bello, J., Cordoba, P. (2007). Speeding Up Oil Recovery by Infill
Drilling Prior to Considering Implementation of a Full-Field WAG Process – El
Furrial; Field, Venezuela. Proceedings of the 2007 SPE Latin American and
Caribbean Petroleum Engineering Conference, 15-18 April, Buenos Aires,
Argentina, SPE 108060, DOI 10.2118/108060-MS.
Mahmoud, T.N. and Rao, D.N. (2007). Mechanisms and Performance Demonstration of
the Gas-Assisted Gravity-Drainage Process Using Visual Models. Proceedings of
the 2007 SPE Annual Technical Conference and Exhibition, 11-14 November,
Anaheim, California, U.S.A., SPE 110132, DOI 10.2118/110132-MS.
Makkar, R.S. and Cameotra, S.S. (1999). Structural Characterization of a Biosurfactant

Produced by Bacillus Subtilis at 45 Degrees C. Journal of Surfactants and
Detergents, 2, pp. 367-372.
McGuire, P.L., Chatham, J.R., Paskvan, F.K., Sommer, D.M. and Carini, F.H. (2005).
Low Salinity Oil Recovery: An Exciting New EOR Opportunity for Alaska’s
North Slope. Proceedings of the SPE Western Regional Meeting, Irvine,
California, 30 Mar - 01 April, SPE 93903, DOI 10.2118/93903-MS
Mercado, M., Acuna, J.C., Vasquez, J., Cabellero, C., and Soriano, E. (2009). Successful
Field Application of a High-Temperature Conformance Polymer in Mexico.
Proceedings of the 8th European Formation Damage Conference, 27-29 May,
Scheveningen, The Netherlands, SPE 121143-MS, DOI 10.2118/121143-MS.
Morrow, N. and Buckley, J. (2011). Improved Oil Recovery by Low-Salinity

Waterflooding. Journal of Petroleum Technology, 63,v5, 106-112, DOI
10.2118/129421-MS.
Mortis, G. (2004). EOR Continues to Unlock Oil Resources. The Oil and Gas Journal,
102, 14, pp. 54-65.
Moritis, G. (2010). Special Report: EOR/Heavy Oil Survey: CO2 Miscible, Steam
Dominate Enhanced Oil Recovery Processes. The Oil and Gas Journal, 108, 14,
36-40.
230
Nasralla, R., Alotaibi, M. B. and Nasr-El-Din, H.A. (2011). Efficiency of Oil Recovery
by Low Salinity Water Flooding in Sandstone Reservoirs. Proceedings of the
SPE Western North American Regional Meeting, Anchorage, Alaska, USA, 7-11
May, SPE 144602, DOI 10.2118/144602-MS.
Nasralla, R. and Nasr-El-Din, H. A. (2011b). Coreflood Study of Low Salinity Water

Injection in Sandstone Reservoirs. Proceedings of the SPE/DGS Saudi Arabia
Section Technical Symposium and Exhibition, Al-Khobar, Saudi Arabia, 15-18
May, SPE 149077, DOI 10.2118/149077-MS.
Narasimhan, T.N. and Witherspoon, P.A. (1976). An Integrated Finite Difference Method
for Analyzing Fluid Flow in Porous Media, Water Resources Research, 12 (1), pp.
57–64.
Ning, S.X. and McGuire, P.L. (2004). Improved Oil Recovery in Under-Saturated
Reservoirs Using the US-WAG Process. Proceedings of the 2004 SPE/DOE
Symposium on Improved Oil Recovery, 17-21 April, Tulsa, Oklahoma, U.S.A.,
SPE 89353-MS, DOI 10.2118/89353-MS.
Okasha, T.M. and Al-Shiwaish, A. (2009). Effect of Brine Salinity on Interfacial Tension
in Arab-D Carbonate Reservoir, Saudi Arabia. Proceedings of the 2009 SPE
Middle East Oil and Gas Show and Conference, Bahrain, Kingdom of Bahrain,
15-18 March, SPE 119600, DOI 10.2118/119600-MS.
Othman, M., Chong, M.O., Sai, R.M., Zainal, S., Zakaria, M.S., and Yaacob, A.A.
(2007). Meeting the Challenges in Alkaline Surfactant Pilot Project
Implementation at Angsi Field, Offshore Malaysia. Proceedings of the Offshore
Europe 2007, 4-7 September, Aberdeen, Scotland, U.K., SPE 109033-PP, DOI
10.2118/109033-MS.
Pan, J., Deng, W., and Wang, Z. (2006). Studies and Application of Slag Blocking Agent
for Steam Injection Reservoir. Proceedings of the 2006 SPE International Oil and
Gas Conference and Exhibition, 5-7 December, Beijing, China, SPE 104426, DOI
10.2118/104426-MS.
Parker, J.C., Lenhard, R.J., and Kuppusamy, T. (1987). A Parametric Model for
Constitutive Properties Governing Multiphase Flow in Porous Media. Water
Resources Research, 23 (4), pp. 618-624.
Paul, G.W. and Froning, H.R. (1973). Salinity Effects of Micellar Flooding. Journal of
Petroleum Technology, 25, 8, 957-958, DOI 10.2118/4419-PA.
Pruess, K. (1983). GMINC. A Mesh Generator for Flow Simulations in Fractured

Reservoirs, Research Report No. LBL-15227, Lawrence Berkeley Laboratory,
Berkeley, CA, March 1983.
231
Pruess, K. (1991). TOUGH2: A General Purpose Numerical Simulator for Multiphase

Fluid and Heat Flow, Research Report No. LBL-29400, Lawrence Berkeley
Laboratory, Berkeley, CA, May 1991.
Pruess, K. and Narasimhan, T.N. (1985). A Practical Method for Modeling Fluid and
Heat Flow in Fractured Porous Media, Soc. Pet. Eng. J., 25 (1), pp. 14-26,
February 1985.
Pruess, K., Oldenburg, C., and Moridis, G. (1999). TOUGH2 User’s Guide, Version 2.0,
Report LBNL-43134, Berkeley, California: Lawrence Berkeley National
Laboratory.
Pu, H., Xie, X., Yin, P. and Morrow, N.R. (2010). Low Salinity Waterflooding and
Mineral Dissolution. Proceedings of the SPE Annual Technical Conference and
Exhibition, Florence, Italy, 19-22 September, SPE 134042, DOI 10.2118/134042-
MS.
Radler, M. and Bell, L. (2009). Weak Energy Demand to Persist Through 2009. The Oil
and Gas Journal, 107, 27, pp. 24.
Raje, M., Asghari, K., Vossoughi, S., Green, D.W., Willhite, G.P. (1996). Gel Systems
for Controlling CO2 Mobility in Carbon Dioxide Miscible Flooding. Proceedings
of the 1996 SPE/DOE Improved Oil Recovery Symposium, 21-24 April, Tulsa,
Oklahoma, U.S.A., SPE 35379-MS, DOI 10.2118/35379-MS.
Ramirez-Garnica, M.A., Mamora, D.D., Nares, H.R., Schacht-Hernandez, P.,

Mohammed, A.A., Cabrera-Reyes, M.C. (2007). Increase Heavy-Oil Production
in Combustion Tube Experiments Through the Use of Catalyst. Proceedings of
the 2007 SPE Latin American and Caribbean Petroleum Engineering Conference,
15-18 April, Buenos Aires, Argentina, SPE 107946, DOI 10.2118/107946-MS.
Reed, M.G. (1967). Retention of Crude Oil Bases by Clay-Containing Sandstone. Clays
and Clay Minerals, Volume 16, pp. 173-178.
Reis, J.C. (1992). A Steam-Assisted Gravity Drainage Model for Tar Sands: Linear
Geometry. Journal of Canadian Petroleum Technology, 31, 10, Paper No. 92-10-
01.
Renkema, W.J. and Rossen, W.R. (2007). Success of Foam SAG Processes in
Heterogeneous Reservoirs. Proceedings of the 2007 SPE Annual Technical
Conference and Exhibition, 11-14 November, Anaheim, California, U.S.A., SPE
110408-MS, DOI 10.2118/110408-MS.
232
Rivet, S. M., Lake, L.W. and Pope, G.A. (2010). A Coreflood Investigation of Low-
Salinity Enhanced Oil Recovery. Proceedings of the SPE Annual Technical
Conference and Exhibition, Florence, Italy, 19-22 September, SPE 134297, DOI
10.2118/134297-MS.
Robertson, E.P. (2010). Oil Recovery Increases by Low-Salinity Flooding: Minnelusa
and Green River Formations. Proceedings of the SPE Annual Technical
Conference and Exhibition, Florence, Italy, 19-22 September, SPE 132154, DOI
10.2118/132154-MS.
Sahin, S., Kalfa, U., and Celebioglu, D. (2007). Bati Raman Field Immiscible CO2
Application, Status Quo and Future Plans. Proceedings of the 2007 Latin
American and Caribbean Petroleum Engineering Conference, 15-18 April,
Buenos Aires, Argentina, SPE 106575-PP, DOI 10.2118/106575-PA.
Satter, A., Iqbal, G. M. and Buchwalter, J. L. (2008). Practical Enhanced Reservoir

Engineering: Assisted with Simulation Software. PennWell Corporation, Tulsa,
Oklahoma, USA. ISBN-13:978-1-59370-056-0.
Seccombe, J.C., Lager, A., Webb, K., Jerauld, G. and Fueg, E. (2008). Improving
Waterflood Recovery: LoSalTM EOR Field Evaluation. Proceedings of the
SPE/DOE Symposium on Improved Oil Recovery, 20-23 April 2008, Tulsa,
Oklahoma, USA, 113480, DOI 10.2118/113480-MS.
Seccombe, J., Lager, A., Jerauld, G., Jhaveri, B., Buikema, T., Bassler, S., Denis, J.,
Webb, K., Cockin, A. and Fueg, E. (2010). Demonstration of Low-Salinity EOR
at Interwell Scale, Endicott Field, Alaska. Proceedings of the SPE Improved Oil
Recovery Symposium, Tulsa, Oklahoma, U.S.A., 24-28 April, SPE 129692, ISBN
978-1-55563-289.
Sen, R. (2008). Biotechnology in Petroleum Recovery: The Microbial EOR. Progress in

Energy and Combustion Science, 34, pp. 714-24.
Seright, R.S., Campbell, A.R., and Mozley, P.S. (2009). Stability of Partially Hydrolyzed
Polyacrylamides at Elevated Temperatures in the Absence of Divalent Cations.
Proceedings of the 2009 SPE International Symposium on Oilfield Chemistry, 20-
22 April, The Woodlands, Texas, U.S.A., SPE 121460, DOI 10.2118/121460-MS.
Sharma, M.M. and Filoco, P.R. (2000). Effect of Brine Salinity and Crude-Oil Properties
on Oil Recovery and Residual Saturations. Society of Petroleum Engineers
Journal, 5(3), 293-300, 10.2118/65402-PA.
Shi, W., Corwith, J., Bouchard, A., Bone, R., and Reinbold, E. (2008). Kuparuk MWAG
Project after 20 years. Proceedings of the 2008 SPE Improved Oil Recovery
10.2118/113933-MS.
233
Shuhong, W., Wenlong, G., Desheng, M., Dehuang, S., Jinzhong, L., and Xiaojin, W.
(2008). Utilising Steam Injection to Improve Performance of Mature
Waterflooding Reservoir. Proceedings of the 2008 SPE Asia Pacific Oil and Gas
Conference and Exhibition, 20-22 October, Perth, Australia, SPE 116549, DOI
10.2118/116549-MS.
Skrettingland, K., Holt, T., Tweheyo, M.T. and Skjevrak, I. (2011). Snorre Low-Salinity-
Water-Injection-Coreflooding Experiments and Single-Well Field Pilot. SPE
Reservoir Evalulation and Engineering, 14,2,182-192, DOI 10.2118/129877-PA.
Slattery, J.C. (1974) Interfacial effects in the entrapment and displacement of residual oil.
AIChE J 20, 1145.
Smith, J.E. (1993). How to Rate Crude Oils for Alkaline Flooding Potential: A Study
Based on 239 Crude Oils. Proceedings of the International SPE Symposium on
Oilfield Chemistry, March 2-5 1993, New Orleans, Louisiana, U.S.A., SPE
25171, DOI 10.2118/25171-MS.
Stegemeier, G.L. (1977) Mechanisms of entrapment and mobilization of oil in porous

media. Improved Oil Recovery by Surfactant and Polymer Flooding, D.O. Shah
and R.S. Schechter, Eds., Academic Press, New York, p. 55.
Stone, H.L. (1982). Vertical Conformance in an Alternating Water-Miscible Gas Flood.

Proceedings of the 1982 SPE Annual Technical Conference and Exhibition, 26-29
September, New Orleans, Louisiana, U.S.A., SPE 11130-MS, DOI
10.2118/11130-MS.
Strand, S., Høgnesen, E.J., and Austad, T. (2005). Wettability alteration of carbonates—
Effects of potential determining ions (Ca2+ and SO42−) and temperature.
Colloids and Surface, 275, 1-3, pp. 1-10.
Sugai, Y., Chengxie, H., Chida, T., and Enomoto, H. (2007). Simulation Studies on the
Mechanisms and Performances of MEOR Using Polymer Producing
Microorganism Clostridium sp. TU-15A. Proceedings of the 2007 SPE Asia
Pacific Oil and Gas Conference and Exhibition, 30 October – 1 November,
Jakarta, Indonesia, SPE 110173-PP, DOI 10.2118/110173-MS.
Surguchev, L.M., Korbol, Ragnhild, Haugen, Sigurd, Krakstad, O.S., (1992). Screening
of WAG Injection Strategies for Heterogeneous Reservoirs. Proceedings of the
European Petroleum Conference, 16-18 November, Cannes, France, SPE 25075-
MS, DOI 10.2118/25075-MS.
Surkalo, H. (1990). Enhanced Alkaline Flooding. Journal of Petroleum Technology, 68,

42, pp. 6-7.
234
Sweatman, R.E., Parker, M.E., and Crookshank, S.L. (2009). Industry Experience sith
CO2-Enhanced Oil Recovery Technology. Proceedings at the 2009 SPE
International Conference on CO2 Capture, Storage, and Utilization, 2-4
November, San Diego, California, U.S.A., SPE 126446, DOI 10.2118/126446-
MS.
Taber, J.J. (1983). Technical Screening Guides for the Enhanced Recovery of Oil.
Proceedings of the SPE Annual Technical Conference and Exhibition, 5-8
October 1983, San Francisco, California, SPE 12069-MS, DOI 10.2118/12069-
MS.
Taber, J.J., Martin, F.D., and Seright, R.S. (1996). EOR Screening Criteria Revisited.
Proceedings of the SPE/DOE Tenth Symposium on Improved Oil Recovery, April
21-24, 1996, Tulsa, Oklahoma, U.S.A., SPE 35385, DOI 10.2118/35385-PA.
Takahashi, S. and Kovscek, A.R. (2009). Spontaneous Countercurrent Imbibition and

Forced Displacement Characteristics of Low-Permeability, Siliceous Shale Rocks.
Proceedings of the 2009 SPE Western Regional Meeting, 24-26 March, San Jose,
California, U.S.A., SPE 121354-MS, DOI 10.2118/121354-MS.
Tang, G. and Morrow, N.R. (1997). Salinity, Temperature, Oil Composition, and Oil
Recovery by Waterflooding. SPE Reservoir Engineering, 12, 4, pp. 269-276.
Tang, G. and Morrow, N.R. (1999a). Oil Recovery by Waterflooding and Imbibition –
Invading Brine Cation Valency and Salinity. Proceedings of the International
Symposium of the Society of Core Analysts, Golden, CO, August, SCA 9911.
Tang, G. and Morrow, N.R. (1999b). Influence of Brine Composition and Fines
Migration on Crude Oil/Brine/Rock Interactions and Oil Recovery. Journal of
Petroleum Science and Engineering, 24, 2, 99-111, DOI 10.1016/S0920-
4105(99)00034-0.
Tang, G. and Morrow, N.R. (2002). Injection of Dilute Brine and Crude Oil/Brine/Rock
Interactions, Environmental Mechanics: Water, Mass and Energy Transfer in the
Biosphere, Geophysical Monograph,129, pp. 171-179.
Thomas, L. K. and Lumpkin, W.B. (1976). Reservoir Simulation of Variable-Point

Problems, Soc. Pet. Eng. J., Trans., AIME, pp. 137-143.
Thyne, G. and Gamage, P. (2011). Evaluation of the Effect of Low Salinity Waterflood
for 26 Fields in Wyoming. Proceedings of the SPE Annual Technical Conference
and Exhibition, Denver, Colorado, USA, 30 October-2 November , SPE 147410,
DOI 10.2118/147410-MS.
235
Town, K., Sheehy, A.J., and Govreau, B.R. (2009). MEOR Success in Southern
Saskatchewan. Proceedings of the 2009 SPE Annual Technical Conference and
Exhibition, 4-7 October, New Orleans, Louisiana, U.S.A., SPE 124319, DOI
10.2118/124319-MS.
Tripathi, I. and Mohanty, K.K. (2007). Flow Instability Associated with Wettability
Alteration. Paper SPE 110202 presented at the SPE Annual Technical Conference
and Exhibition, 11-14 November 2007, Anaheim, California, U.S.A, SPE 110202-
MS, DOI 10.2118/110202-MS.
URS Corporation (2009). Appendix E Carbon Dioxide Pipeline Risk Analysis, HECA
Project Site,
http://www.energy.ca.gov/sitingcases/hydrogen_energy/documents/applicant/revi
sed_afc/Volume_II/Appendix%20E.pdf (accessed 01/21/2012).
Vahidi, A. and Zargar, G. (2007). Sensitivity of Important Parameters Affecting

Minimum Miscibility Pressure (MMP) Nitrogen Injection into Conventional Oil
Reservoirs. Proceedings of the SPE/EAGE Conference on Reservoir
Characterization and Simulation, 28-31 October 2007, Abu Dhabi, U.A.E., SPE
111411-PP, DOI 10.2118/111411-MS.
Van Genuchen, M. Th., (1980). A Closed-Form Equation for Predicting the hydraulic
Conductivity of Unsaturated Soils. Soil Science Society America Journal, 44,
892-898.
Van Olphen, H. (1963). An Introduction to Clay Colloid Chemistry. Interscience

Publishers, John Wiley and Sons, New York, U.S.A.
Vijapurapu, C.S. and Rao, D.N. (2004). Compositional Effects of Fluids on Spreading,
Adhesion and Wettability in Porous Media. Colloids and Surface, 241(1-3), pp.
343-349.
Vledder, P., Fonseca, J.C., Wells, T., Gonzalez, I. and Lighelm, D. (2010). Low Salinity
Water Flooding: Proof of Wettability Alteration on a Field Wide Scale.
Proceedings of the SPE Improved Oil Recovery Symposium, Tulsa, Oklahoma,
U.S.A., 24-28 April, SPE 129692, DOI 10.2118/129564-MS.
Yousef, A.A., Al-Saleh, S., Al-kaabi, A. and Al-Jawfi, M. (2010). Laboratory

Investigation of Novel Oil Recovery Method for Carbonate Reservoirs.
Proceedings of the Canadian Unconventional Reservoirs and International
Petroleum Conference, Calgary, Alberta, Canada, 19-21 October, CSUG/SPE
137634, DOI 10.2118/137634-PA.
Yildiz, H.O., Valat, M. and Morrow N.R. (1999). Effect of Brine Composition on
Wettability and Oil Recovery of a Prudhoe Bay Crude Oil. Journal of Canadian
Petroleum Technology, 38, 1, 26-31.
236
Wang, D., Han, P., Shao, Z., Chen, J., and Seright, R.S. (2008). Sweep Improvement
Options for the Daqing Oil Field. Proceedings of the 2006 SPE/DOE Symposium
on Improved Oil Recovery, 22-26 April, Tulsa, Oklahoma, U.S.A., SPE 99441,
DOI 10.2118/99441-MS.
Warren, J.E. and Root, P.J. (1963). The Behavior of Naturally Fractured Reservoirs, Soc.
Pet. Eng. J., Trans., AIME, 228, pp. 245-255, September 1963.
Webb, K.J., Black, C.J.J. and Al-Ajeel, H. (2004). Low Salinity Oil Recovery – Log-
Inject-Log. Proceedings of the SPE/DOE Symposium on Improved Oil Recovery,
Tulsa Oklahoma, 17-21 April, SPE 89379-MS, DOI 10.2118/89379-MS.
Webb, K.J., Black, C.J.J. and Tjetland, G. (2005). A Laboratory Study Investigating
Methods for Improving Oil Recovery in Carbonates. Proceedings of the
International Petroleum Technology Conference, Doha, Qatar, 21-23 November,
IPTC 10506, DOI 10.2523/10506-MS.
Wei, Z., Jian, Z., Ming, H., Wentao, X., Guozhi, F., Wei, J., Fujie, S., Shouwei, Z.,
Yongjun, G., Zhongbin, Y. (2007). Application of Hydrophobically Associating
Water-Soluble Polymer for Polymer Flooding in China Offshore Heavy Oilfield.
Proceedings of the International Petroleum Technology Conference, 4-6
December, Dubai, U.A.E., IPTC 11635, DOI 10.2523/11635-MS.
Wen, S., Zhao, Y., Liu, Y., and Hu, S. (2007). A Study of Catalytic Aquathermolysis of
Heavy Crude Oil During Steam Stimulation. 2007 SPE International Symposium
on Oilfield Chemistry, 28 February – 2 March, Houston, Texas, U.S.A., SPE
106180, DOI 10.2118/106180-MS.
Wu, Y.S. and Pruess, K. (1988). A multiple-porosity method for simulation of naturally
fractured petroleum reservoirs, SPE Reservoir Engineering, v. 3, pp. 327-336.
Wu, Y.S., Ahlers, C.F., Fraser, P., Simmons, A., and Pruess, K. (1996). Software
qualification of selected TOUGH2 modules, Report LBNL-39490, Lawrence
Berkeley National Laboratory, Berkeley, CA, 1996.
Wu, Y.S., Forsyth, P.A., and Jiang, H. (1996). A Consistent Approach for Applying
Numerical Boundary Conditions for Multiphase Subsurface Flow, Journal of
Contaminant Hydrology, 23, pp. 157-184.
Wu, Y.S. (2000). A virtual node method for handling wellbore boundary conditions in
modeling multiphase flow in porous and fractured media. Water Resources
Research, 36 (3) (2000), pp. 807-814.
237
Wu, Y. and Bai, B. (2009). Efficient Simulation of Low-Salinity Waterflooding in Porous

and Fractured Reservoirs. Proceedings of the 2009 SPE Reservoir Simulation
Symposium, The Woodlands, xas, 2-4 February, SPE 118830, DOI
10.2118/118830-MS.
Xinde, W., Mei, H., Huiyu, Y., Yingzhi, Z., Weidong, C., and Ershuang, G. (2006). Pilot
Test of Weak Alkaline System ASP Flooding in Secondary Layer with Small
Well Spacing. Proceedings of the 2006 SPE International Oil and Gas
Conference and Exhibition, 5-7 December, Beijing, China, SPE 104416, DOI
10.2118/104416-MS.
Xu, Y., Pu, H., Shi, L. (2008). An Integrated Study of Mature Low Permeability
Reservoir in Daqing Oil Field. Proceedings of the 2008 SPE Western and Pacific
Section AAPG Joint Meeting, 31 March – 2 April, Bakersfield, California,
U.S.A., SPE 114199, DOI 10.2118/114199-MS.
Yildiz, H.O., Valat, M., and Morrow, N.R. (1999). Effect of Brine Composition on
Wettability and Oil Recovery of a Prudhoe Bay Crude Oil. Journal of Canadian
Petroleum Technology, Volume 38, No. 1, pp. 27-31.
Yin, G., Grigg, R.B., Svec, Y. (2009). Oil Recovery and Surfactant Adsorption During
CO2 – Foam Flooding. Proceedings of the 2009 Offshore Technology
Conference, 4-7 May, Houston, Texas, U.S.A., OTC 19787.
Yousef, A.A., Al-Saleh, S., Al-kaabi, A., and Al-Jawfi, M. (2010). Laboratory
Investigation of Novel Oil Recovery Method for Carbonate Reservoirs.
Proceedings of the Canadian Unconventional Reservoirs and International
Petroleum Conference, Calgary, Alberta, Canada, 19-21 October, CSUG/SPE
137634, DOI 10.2118/137634-MS.
Zaitoun, A., Tabary, R., Rousseau, D., Pichery, T., Nouyoux, S., Mallo, P., and Braun, O.
(2007). Using Microgels to Shut Off Water in a Gas Storage Well. Proceedings of
the 2007 International Symposium on Oilfield Chemistry, 28 February-2 March,
Houston, Texas, U.S.A., SPE 106042-MS, DOI 10.2118/106042-MS.
Zhang, P. and Austad, T. (2005). Waterflooding in Chalk: Relationship Between Oil

Recovery, New Wettability Index, Brine Composition and Cationic Wettability
Modifier. Proceedings of the Europec/EAGE Annual Conference, Madrid, Spain,
13-16 June, SPE 94209, DOI 10.2118/94209-MS.
Zhang, Y. and Morrow, N.R. (2006). Comparison of Secondary and Tertiary Recovery
with Change in Injection Brine Composition for Crude Oil/Sandstone
Combinations. Proceedings of the 2006 SPE/DOE Symposium on Improved Oil
Recovery, Tulsa, Oklahoma, U.S.A., 22-26 April, SPE 99757, DOI
10.2118/99757-MS.
238
Zhang, Y., Xie, X. and Morrow, N. R. (2007). Waterflood Performance by Injection of

Brine with Different Salinity for Reservoir Cores. Proceedings of the SPE Annual
Technical Conference and Exhibition, Anaheim, California, U.S.A, 11-14
November, SPE 109849. DOI 10.2118/109849-MS.
Zhao, P., Jackson, A.C., Britton, C., Kim, D.H., Britton, L.N., Levitt, D.B., and Pope,
G.A. (2008). Development of High-Performance Surfactants for Difficult Oils.
Proceedings of the 2008 SPE Improved Oil Recovery Symposium, 19-23 April,
Tulsa, Oklahoma, U.S.A., SPE 113432, DOI 10.2118/113432-MS.
239
VITA
Ahmad Aladasani was born in Lebanon on November 15th, 1970. He earned his
high school degree from the University of London GCE Board in 1989. During the Gulf
War in 1989, he joined the United States (US) army, 501 Military Intelligence Battalion,
1st Armored Division, 7th Core. He served from November 1989 to April 1990
and received commendations from the US Army, Royal Saudi Armed Forces and Kuwaiti
Armed Forces. After the war, he completed a BSc in Mechanical Engineering in 1995
from Gannon University, Erie, Pennsylvania. During that period, he volunteered as a
tutor for the Juvenile Delinquents Court in the Commonwealth of Pennsylvania. In 1996,
he joined the Kuwait Oil Company (KOC). In 2002, he was promoted to Senior
Inspection Engineer, and he currently works there as a Specialist Operational Planner, a
position to which he was promoted in 2009. In addition, during his employment with
KOC, he was certified as a welding inspector, industrial radiographer, lead auditor in
Environmental Management Systems and Occupational Health and Safety
Assessor. Furthermore, he completed his MSc in Process Safety and Loss Prevention
from the University of Sheffield, UK, in 2005. After a major incident in North Kuwait in
2001, he was recognized by KOC for his role in surveying strategic pipelines, managing
the proceeding underwriter survey report in 2003, standardizing Inspection and Corrosion
requirements and coaching several subordinates who now hold leading technical
positions in KOC. Finally, during his leave to study in the US, he served as a mentor for
undergraduate international SPE students from Nigeria. In August 2012, he received his
PhD in Petroleum Engineering from Missouri University of Science and Technology.

1

Updated EOR Screening Criteria and Modeling The Impacts of Water - Al Adasani 2012

Transféré par

Informations du document

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Updated EOR Screening Criteria and Modeling The Impacts of Water - Al Adasani 2012

Transféré par

Droits d'auteur :

Formats disponibles

Scholars' Mine

Doctoral Dissertations Student Research & Creative Works

Updated EOR screening criteria and modeling the

Follow this and additional works at: http://scholarsmine.mst.edu/doctoral_dissertations

UPDATED EOR SCREENING CRITERIA AND

MODELING THE IMPACTS OF WATER SALINITY CHANGES

Presented to the Faculty of the Graduate School of the

MISSOURI UNIVERSITY OF SCIENCE AND TECHNOLOGY

In Partial Fulfillment of the Requirements for the Degree

Dr. Baojun Bai, Advisor

intervention, which makes it difficult to identify recovery mechanism(s) in EOR methods

screening tools based on dataset distribution, incremental recovery and deterministic

modeling. LSWF recovery mechanisms are investigated by statistical analysis and

numerical solutions. Furthermore, an up-scaled multi-dimensional model is developed

reservoirs. On a separate note, hydrogen sulfide (H2S) risk in anthropogenic CO2

ABSTRACT ....................................................................................................................... iii

LIST OF ILLUSTRATIONS ............................................................................................. ix

LIST OF TABLES ............................................................................................................ xii

NOMENCLATURE ........................................................................................................ xiv

2. LITERATURE REVIEW ............................................................................................ 3

2.1. ENERGY SUPPLY .................................................................................................. 3

2.1.1. World Energy Statistics. .................................................................................... 3

2.1.2. Future Energy Outlook. ..................................................................................... 3

2.1.3. Oil Reserves. ...................................................................................................... 6

2.1.4. Oil Supply. ......................................................................................................... 6

2.2. ENHANCED OIL RECOVERY .............................................................................. 7

2.2.1. Oil Production.................................................................................................... 8

2.2.2. Reservoir Flow Mechanics. ............................................................................... 8

2.2.3. Oil Production.................................................................................................. 10

2.2.4. Enhanced Oil Recovery Methods. ................................................................... 10

2.2.5. Enhanced Oil Recovery Selection Criteria. ..................................................... 12

2.3. WATER FLOODING ............................................................................................ 13

2.3.1. Low-Salinity Water Flooding. ......................................................................... 19

2.3.2. Low-Salinity Water-Flooding Experiments and Tests. ................................... 20

2.3.3. Reservoir Simulation. ...................................................................................... 23

2.3.4. Recovery Mechanisms. .................................................................................... 23

3. ENHANCED OIL RECOVERY GUIDELINES AND SELECTION CRITERIA .. 28

3.1. UPDATING ENHANCED OIL RECOVERY SELECTION CRITERIA ............ 28

3.1.1. Enhanced Oil Recovery Database. .................................................................. 29

3.1.1.1. Database analysis. ..................................................................................... 29

3.1.1.2. Enhanced oil recovery selection criteria. .................................................. 30

3.1.1.3. New enhanced oil recovery selection criteria. .......................................... 31

3.2. UPDATED ENHANCED OIL RECOVERY SELECTION CRITERIA .............. 31

3.2.1. Databases Analysis. ......................................................................................... 32

3.2.2. Enhanced Oil Recovery Selection Criteria. ..................................................... 41

3.2.3. New Enhanced Oil Recovery Selection Criteria ............................................. 43

3.2.3.1. Dataset distributions.................................................................................. 43

3.2.3.2. Incremental recovery. ............................................................................... 51

3.3. UPDATED ENHANCED OIL RECOVERY GUIDELINES ............................... 53

3.3.1. Low-Salinity Water Flooding. ......................................................................... 59

3.3.2. Water-Alternating Gas. .................................................................................... 61

3.3.3. Surfactant Imbibition. ...................................................................................... 62

3.3.4. In-Depth Conformance Control. ...................................................................... 63

3.3.5. Microbial Enhanced Oil Recovery. ................................................................. 64

3.3.6. Steam-Assisted Gravity Drainage. .................................................................. 65

3.4. MISCIBLE CARBON DIOXIDE FLOODING .................................................... 66

3.5. NEW SELECTION CRITERIA FOR CO2 EOR .................................................. 67

4. MODIFIED WATER SALINITY ............................................................................. 75

4.1. LOW-SALINITY WATER-FLOODING MODEL ............................................... 75

4.1.1. Governing Equations. ...................................................................................... 76

4.1.2. Numerical Formulation.................................................................................... 81

4.1.3. Handling Initial and Boundary Conditions. ..................................................... 85

4.1.4. Treatment of Fracture-Matrix Interaction........................................................ 85