investigation of lost circulation materials impact on
TRANSCRIPT
Scholars' Mine Scholars' Mine
Doctoral Dissertations Student Theses and Dissertations
Fall 2015
Investigation of lost circulation materials impact on fracture Investigation of lost circulation materials impact on fracture
gradient gradient
Mortadha Turki Alsaba
Follow this and additional works at: https://scholarsmine.mst.edu/doctoral_dissertations
Part of the Petroleum Engineering Commons
Department: Geosciences and Geological and Petroleum Engineering Department: Geosciences and Geological and Petroleum Engineering
Recommended Citation Recommended Citation Alsaba, Mortadha Turki, "Investigation of lost circulation materials impact on fracture gradient" (2015). Doctoral Dissertations. 2437. https://scholarsmine.mst.edu/doctoral_dissertations/2437
This thesis is brought to you by Scholars' Mine, a service of the Missouri S&T Library and Learning Resources. This work is protected by U. S. Copyright Law. Unauthorized use including reproduction for redistribution requires the permission of the copyright holder. For more information, please contact [email protected].
INVESTIGATION OF LOST CIRCULATION MATERIALS IMPACT ON
FRACTURE GRADIENT
by
MORTADHA TURKI ALSABA
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
2015
Approved by:
Runar Nygaard, Advisor
Baojun Bai
Shari Dunn-Norman
Mingzhen Wei
Arild Saasen
3
2015
Mortadha T. Alsaba
All Rights Reserved
iii
ABSTRACT
Lost circulation is a challenging problem to be prevented or mitigated during
drilling. Lost circulation treatments are widely applied to mitigate losses using a
corrective approach or to prevent losses using preventive approaches, also known as
“wellbore strengthening”. The disagreement among the different wellbore strengthening
theories and the lack of understanding the strengthening mechanism resulted in the
absence of a standardized method to evaluate the effectiveness of lost circulation
materials (LCM) for wellbore strengthening application.
An extensive experimental investigation was performed by constructing a high
pressure LCM test apparatus to investigate the effects of different parameters on the
sealing efficiency of LCM treatments. In addition, hydraulic fracturing experiments,
which simulates downhole conditions, were carried out to evaluate the impact of LCM
addition on enhancing both; breakdown and re-opening pressure.
The results showed that the sealing efficiency of LCM treatments is highly
dependent on the fracture width and the particle size distribution (PSD). Carefully
selected LCM blends can seal fractures up to 2500 micron and certain unconventional
squeeze LCM can seal wider fractures. A particle size distribution selection criterion for
LCM treatments was developed based on a statistical analysis of the experimental results
states that D50 and D90 should be equal or greater than 3/10 and 6/5 the fracture width,
respectively. The addition of different LCM blends enhanced the breakdown pressure up
to 18% and the re-opening pressure up to 210%. Comparing the fractures created by the
experiments with analytical models, only one model estimated similar fracture widths.
iv
ACKNOWLEDGMENTS
First of all, thanks to Allah (God) for granting me the blessings, health and
strength to complete this research “Alhamdulillah”. Second, I would like to express my
sincere gratitude to my PhD advisor Dr. Runar Nygaard for his invaluable support
throughout my research. He has been my mentor, motivator, supporter, my older brother,
and more since the first day I joined his group. I would also like to thank my committee
members, Dr. Baojun Bai, Dr. Shari Dunn-Norman, Dr. Mingzhen Wei, and Dr. Arild
Saasen for their valuable advice and recommendations. Dr. Wisner, and Dr. Wronkiewicz
for their help with the microscope work, Dr. Galecki for his help with the PSD analysis.
Many thanks to Jeff Heniff, Mike Bassett, and John Tyler for their valuable assistance in
manufacturing the testing apparatuses. I am very grateful to our research group members,
especially Mohammed Al Dushaishi for his help in the statistical analysis, Reza Rahimi
for his help in the analytical modeling, and Montri Jeennakorn for his help in the lab.
I would also like to acknowledge the funding received from Det Norske
Oljeselskap ASA under research agreements #37709 and #45343.
A Special thanks to my family for their love, support, encouragement, and prayers
throughout my study. For my parents, Turki and Maryam, who brought me to this life and
raised me under their watchful eyes.
Last, but not least, I would like to thank my lovely wife Zahra for her love,
understanding, and the years of patience. I would never have been able to complete this
work without her support for the last 8 years. She was and will be always there for me.
v
TABLE OF CONTENTS
Page
ABSTRACT ....................................................................................................................... iii
ACKNOWLEDGMENTS ................................................................................................. iv
LIST OF ILLUSTRATIONS ............................................................................................. ix
LIST OF TABLES ........................................................................................................... xiv
NOMENCLATURE ........................................................................................................ xvi
UNIT CONVERSION ................................................................................................... xviii
SECTION
1. INTRODUCTION ...................................................................................................... 1
1.1. LOST CIRCULATION TREATMENTS ........................................................... 3
1.1.1. Corrective Treatments. ............................................................................. 4
1.1.2. Preventive Treatments .............................................................................. 5
2. LITERATURE STUDY ............................................................................................. 6
2.1. PREVIOUS EXPERIMENTAL INVESTIGATIONS ....................................... 6
2.1.1. LCM Performance Evaluation.................................................................. 6
2.1.2. Hydraulic Fracturing Experiments. .......................................................... 9
2.2. WELLBORE STRENGTHENING THEORIES .............................................. 11
2.3. ANALYTICAL MODELING OF FRACTURE WIDTH ................................ 13
2.4. REVIEW OF BRIDGING THEORIES ............................................................ 17
2.5. CRITICAL REVIEW OF THE LITERATURE ............................................... 18
2.6. RESEARCH OBJECTIVES ............................................................................. 23
3. METHODOLOGY ................................................................................................... 24
vi
3.1. LCM CHARACTERIZATION ........................................................................ 25
3.1.1. LCM Classification. ............................................................................... 26
3.1.2. Particle Size Distribution Analysis......................................................... 26
3.1.3. Optical Microscopy. ............................................................................... 27
3.1.4. Scanning Electron Microscopy. ............................................................. 28
3.1.5. Swelling. ................................................................................................. 28
3.2. LCM PERFORMANCE EVALUATION ........................................................ 29
3.2.1. Fluid Used. ............................................................................................. 30
3.2.2. Conventional LCM Used........................................................................ 30
3.2.3. Unconventional LCM Used.................................................................... 33
3.2.4. Fracture Width Variation........................................................................ 34
3.2.5. Low Pressure LCM Testing. .................................................................. 35
3.2.6. High Pressure LCM Testing. .................................................................. 36
3.3. HYDRAULIC FRACTURING TEST .............................................................. 40
3.3.1. Core Preparation. .................................................................................... 41
3.3.2. Test Procedures. ..................................................................................... 41
3.4. DATA ANALYSIS ........................................................................................... 42
3.4.1. Statistical Methods. ................................................................................ 42
3.4.2. Analytical Fracture Models. ................................................................... 45
4. RESULTS ................................................................................................................. 46
4.1. LCM CHARACTERIZATION ........................................................................ 46
4.1.1. Updated LCM Classification. ................................................................. 46
4.1.2. Particle Size Distribution Analysis......................................................... 49
vii
4.1.3. Optical Microscopy. ............................................................................... 50
4.1.3.1 LCM particles. ............................................................................50
4.1.3.2 Formed seals. ..............................................................................52
4.1.4. Scanning Electron Microscopy. ............................................................. 55
4.1.5. Swelling. ................................................................................................. 58
4.2. LOW PRESSURE LCM TESTING ................................................................. 59
4.3. HIGH PRESSURE CONVENTIONAL LCM TESTING ................................ 61
4.3.1. Effect of LCM Type and Fracture Width. .............................................. 64
4.3.2. Effect of Concentration. ......................................................................... 65
4.3.3. Effect of Temperature. ........................................................................... 66
4.3.4. Effect of Particle Size Distribution. ....................................................... 68
4.3.5. Effect of Base Fluid. ............................................................................... 72
4.3.6. Effect of Density. ................................................................................... 73
4.3.7. Effect of Weighting Material. ................................................................ 75
4.3.8. Effect of LCM Shape. ............................................................................ 75
4.4. HIGH PRESSURE UNCONVENTIONAL LCM TESTING .......................... 76
4.5. HYDRAULIC FRACTURING TEST .............................................................. 79
4.5.1. Test # 1 with EDC95-11 Base Fluid. ..................................................... 79
4.5.2. Test # 2 with 11 lb/gal OBM without LCM. .......................................... 80
4.5.3. Test # 3 with 11 lb/gal OBM + 20 ppb Graphite and Nutshells. ............ 82
4.5.4. Test # 4 with 11 lb/gal OBM + 30 ppb Graphite and Sized Calcium
Carbonate. .............................................................................................. 84
4.5.5. Test # 5 with 11 lb/gal OBM + 55 ppb Graphite, Sized Calcium
Carbonate, and Cellulosic Fiber............................................................. 85
viii
4.5.6. Hydraulic Fracturing Results Comparison. ............................................ 87
4.6. DATA ANALYSIS ........................................................................................... 88
4.6.1. Statistical Analysis. ................................................................................ 88
4.6.2. Analytical Fracture Models. ................................................................... 92
5. DISCUSSION .......................................................................................................... 95
5.1. LCM PERFORMANCE EVALUATION ........................................................ 95
5.2. HYDRAULIC FRACTURING EXPERIMENTS .......................................... 104
5.3. PARTICLE SIZE DISTRIBUTION SELECTION CRITERIA ..................... 109
5.3.1. Development of PSD Selection Method. ............................................. 111
6. CONCLUSIONS AND RECOMMENDATIONS ................................................. 115
6.1. CONCLUSIONS............................................................................................. 115
6.2. RECOMMENDATION AND FUTURE WORK ........................................... 118
APPENDICES
A. PREVIOUSLY REPORTED LABORATORY RESULTS .................................. 120
B. HYDRAULIC FRACTURING TEST PREPARATION AND
PROCEDURES ..................................................................................................... 124
C. CROSS-REFERENCING TABLES FOR AVAILABLE LCMS UNDER THE
NEW CLASSIFICATION ..................................................................................... 130
D. PARTICLE SIZE DISTRIBUTION ANALYSIS ................................................. 138
E. PREDICTIVE LINEAR MODEL ......................................................................... 149
F. PSD MODELS COMPARISON ........................................................................... 151
G. PUBLICATION FROM THIS DISSERTATION ................................................. 154
BIBLIOGRAPHY ........................................................................................................... 156
VITA .............................................................................................................................. 165
ix
LIST OF ILLUSTRATIONS
Figure Page
2.1. HPHT Fluid Loss Apparatus ........................................................................................ 7
2.2. Particle Plugging Apparatus (Fann Instrument Company, 2011) ................................ 7
2.3. Schematic of the Impermeable Fracture Test Apparatus (Van Oort et al. 2011) ........ 8
2.4. Corrugated Aluminum Platens (Hettema et al. 2007) .................................................. 8
2.5. Schematic of the Permeable Fracture Test (Hettema et al. 2007) ................................ 8
2.6. Cross Section View of the Fracture Cell (Kaageson-Loe al. 2009) ............................. 8
2.7. Symmetric Bi-wing Fracture around Wellbore.......................................................... 13
3.1. Dry Sieve Analysis Setup .......................................................................................... 27
3.2. Sphericity and Roundness Chart (Powers, 1953) ...................................................... 28
3.3. Schematic of the Swelling Test Apparatus ................................................................ 29
3.4. Wedge Foam of Three Sizes; small (S), medium (M), and coarse (C) ...................... 34
3.5. Schematic of Tapered Discs ....................................................................................... 35
3.6. Set of Tapered and Straight Slotted Discs ................................................................. 35
3.7. (a) Low Pressure LCM Testing Apparatus (LPA), (b) Snug-fit Spacer, (c) Tapered
Disc on the Spacer ..................................................................................................... 36
3.8 Schematic of the High Pressure Testing Apparatus .................................................... 37
3.9. Pressure vs. Time Plot Generated from HPA Tests for Conventional LCMs ........... 38
3.10. Pressure vs. Time Plot Generated from HPA Tests for Unconventional LCM ....... 39
3.11. Schematic of the Hydraulic Fracturing Apparatus ................................................... 40
4.1. Graphite Particles at 60x Magnification (left) and 200x Magnification (right) ......... 50
x
4.2. Calcium Carbonate Particles at 60x Magnification (left) and 200x Magnification
(right) ......................................................................................................................... 51
4.3. Nutshells Particles at 60x Magnification (left) and 200x Magnification (right) ....... 51
4.4. Cellulosic Fiber Particles at 60x Magnification (left) and 200x
Magnification (right) ................................................................................................. 51
4.5. Optical Microscopy Images of Graphite and Sized Calcium Carbonate (30 ppb)
Seal Retrieved from 1000 microns Fracture Width ................................................... 52
4.6. Optical Microscopy Images of Graphite and Nutshells (20 ppb) Seal Retrieved
from 1500 microns Fracture Width ........................................................................... 53
4.7. Optical Microscopy Images of Graphite and Nutshells (40 ppb) Seal Retrieved
from 2000 microns Fracture Width ........................................................................... 54
4.8. Optical Microscopy Images of Graphite, Sized Calcium Carbonate, and
Cellulosic Fiber (55 ppb) Seal Retrieved from 1500 microns Fracture Width .......... 55
4.9. Nutshells # 1 (50 ppb) SEM Images of Formed Seal Retrieved from
2000 microns Fracture Width .................................................................................... 56
4.10. Graphite and Sized Calcium Carbonate #1 (80 ppb) SEM Images of Formed
Seal Retrieved from 1500 microns Fracture Width ................................................. 57
4.11. Graphite and Nutshells # 1 (40 ppb) SEM Images of Formed Seal Retrieved
from 2000 microns Fracture Width ......................................................................... 57
4.12. Graphite, Sized Calcium Carbonate, and Cellulosic Fiber # 1 (55 ppb) SEM
Images of Formed Seal Retrieved from 1000 microns Fracture Width ................... 58
4.13. Nutshells Swelling (24 hr Water Soak) @ Room Temperature and 180 ˚F ............ 59
4.14. Effect of LCM Type on the Sealing Pressure for (a) Single LCM Treatments
and (b) LCM Mixtures ............................................................................................. 65
4.15. Effect of LCM Concentration on the Sealing Pressure for (a) Single LCM
Treatments and (b) LCM Mixtures .......................................................................... 66
4.16. Effect of Temperature on the Sealing Pressure........................................................ 67
4.17. Effect of PSD on the Sealing Pressure using 1000 microns Fracture
Width (TS1) for (a) Single LCM Treatments and (b) LCM Mixtures .................... 68
xi
4.18. Effect of PSD on the Sealing Pressure using 1500 microns Fracture
Width (TS2) for (a) Single LCM Treatments and (b) LCM Mixtures .................... 69
4.19. PSD vs Sealing Pressure for 1000 microns Fracture Width .................................... 70
4.20. PSD vs Sealing Pressure for 1500 microns Fracture Width .................................... 71
4.21. PSD vs Sealing Pressure for 2000 microns Fracture Width .................................... 72
4.22. Effect of Base Fluid on Low and High Concentration LCM Mixtures ................... 73
4.23. Effect of Increasing the Mud Density in OBM and WBM ...................................... 74
4.24. Effect of Weighting Materials Type in OBM and WBM ........................................ 75
4.25. Wedge Foam in WBM Formed Plug using 5 mm disc; (left) Side View,
(right) Bottom View ................................................................................................ 78
4.26. Wedge Foam in OBM Formed Plug using 1500 microns disc; (left) Side View,
(right) Bottom View ................................................................................................ 78
4.27. Damaged Slotted Disc.............................................................................................. 79
4.28. Pressure vs. Time for Test # 1 ................................................................................. 80
4.29. The Resulting Fracture under Microscope (200x Magnification) for Test # 1
Showing Fracture Widths Ranging between 15 – 22 microns ................................ 80
4.30. Pressure vs. Time for Test # 2 ................................................................................. 81
4.31. Fractured Core Showing Two Vertical Fractures and a Cross-sectional View of
the Fractured Core for Test # 2 ................................................................................ 81
4.32. The Resulting Fracture under Microscope (200x Magnification) for Test # 2
Showing Fracture Widths Ranging between 27 – 36 microns ................................ 82
4.33. Pressure vs. Time for Test # 3 ................................................................................. 82
4.34. Fractured Core Showing Two Vertical Fractures and a Cross-sectional View
of the Fractured Core for Test # 3 ........................................................................... 83
4.35. The Resulting Fracture under Microscope (200x Magnification) for Test # 3
Showing Fracture Widths Ranging between 40 – 85 microns ................................ 83
4.36. Pressure vs. Time for Test # 4 ................................................................................. 84
xii
4.37. Fractured Core Showing Two Vertical Fractures and a Cross-sectional View
of the Fractured Core for Test # 4 ........................................................................... 84
4.38. The Resulting Fracture under Microscope (200x Magnification) for Test # 4
Showing Fracture Widths Ranging between 40 – 145 microns .............................. 85
4.39. Pressure vs. Time for Test # 5 ................................................................................. 85
4.40. Fractured Core Showing Two Vertical Fractures for Test # 5 ................................. 86
4.41. The Resulting Fracture under Microscope (200x magnification) for Test # 5
Showing Fracture Widths Ranging between 50 – 140 microns .............................. 86
4.42. Leverage Plot Showing the Effect of Fracture Width on Sealing Pressure ............. 89
4.43. Leverage Plot Showing the Effect of Density on Sealing Pressure ......................... 89
4.44. Leverage Plot Showing the Effect of D90 on Sealing Pressure............................... 89
4.45. Leverage Plot Showing the Effect of LCM Blend on Sealing Pressure .................. 89
4.46. Leverage Plot Showing the Effect of D75 on Sealing Pressure............................... 90
4.47. Leverage Plot Showing the Effect of Base Fluid on Sealing Pressure .................... 90
4.48. Leverage Plot Showing the Effect of D25 on Sealing Pressure............................... 90
4.49. Leverage Plot Showing the Effect of D50 on Sealing Pressure............................... 90
4.50. Leverage Plot Showing the Effect of D10 on Sealing Pressure............................... 91
4.51. Leverage Plot Showing the Actual Sealing Pressure versus the Predicted
Sealing Pressure using the fit model ........................................................................ 91
4.52. Residual Plot of Sealing Pressure versus the Predicted Sealing Pressure ............... 91
5.1. Comparison between Fluid Loss Values from the LPA and the Calculated
Average Fluid Loss per Cycle from the HPA ............................................................ 95
5.2. High Concentration of Nutshells (Left) and Graphite (Right) Treatments Screened
out at the Fracture Surface ......................................................................................... 96
5.3. Plugged Fracture tip of a 2000-micron Tapered Disc showing Stuck Nutshells
Particles ...................................................................................................................... 97
xiii
5.4. Sealing Pressure vs. Average Fluid Loss per Cycle using 1000 microns Fracture
Width (TS1) ............................................................................................................... 98
5.5. Sealing Pressure vs. Average Fluid Loss per Cycle using 1500 microns Fracture
Width (TS2) ............................................................................................................... 98
5.6. Sealing Pressure vs. Average Fluid Loss per Cycle using 2000 microns Fracture
Width (TS3 and TS4) ................................................................................................. 99
5.7. Particle Size Distribution for the used Barite and Hematite using Laser
Diffraction Technique .............................................................................................. 102
5.8. Effect of Sieving Barite in OBM ............................................................................. 103
5.9. Pressure vs. Time for the Control Sample (CS) without Nanoparticles .................. 106
5.10. Pressure vs. Time for the Optimum Concentration of Iron-based Nanoparticles
Blend (DF1) ........................................................................................................... 107
5.11. Pressure vs. Time for the Optimum Concentration of Calcium-based
Nanoparticles Blend (DC6) ................................................................................... 107
5.12. (a) Cross-section of Fracture Plane near Wellbore, (b) Seal at Fracture Mouth,
and (c) Cross-section of Fracture Plane at the Fracture End
(Contreras et al. 2014b) ......................................................................................... 108
5.13. EDX of Seal Cross-section at 500x magnification where the Green Color
Indicates Iron Particles Distribution on the Formed Seal
(Contreras et al. 2014b) ......................................................................................... 108
5.14. Fractures at Core End (left) and 3D Representation of the Core End
Containing the Fractures (right) (Contreras et al. 2014b) ...................................... 109
5.15. Leverage Plot Showing the Actual Sealing Pressure versus the Predicted
Sealing Pressure using the Simple Linear Fit Model for the Subset Dataset ........ 112
xiv
LIST OF TABLES
Table Page
2.1. Summary of the Previous LCM Testing Apparatuses and their Limitations ............. 19
2.2. Summary of the Competing Wellbore Strengthening Theories ................................. 20
2.3. Summary of Current Selection Criteria ..................................................................... 22
3.1. Drilling Fluid Properties as Measured in Accordance with API Standards
(API- RP 13B) ........................................................................................................... 30
3.2. Single LCM Treatment Formulation ......................................................................... 31
3.3. LCM Mixture Formulation ........................................................................................ 32
3.4. WEDGE-SET Formulation for Specific Fracture Widths ......................................... 34
3.5. Tapered and Straight Slotted Discs Specifications .................................................... 35
4.1. Summary of the Particle Size Distribution Analysis ................................................. 49
4.2. Low Pressure Apparatus (LPA) Fluid Loss Results in (ml) for Single LCM
Treatments ................................................................................................................. 60
4.3. Low Pressure Apparatus (LPA) Fluid Loss Results in (ml) for LCM Mixtures ....... 61
4.4. Selected Blends for High Pressure Apparatus (HPA) Testing................................... 61
4.5. Summary of HPA Results for Single LCM Treatments using WBM ........................ 62
4.6. Summary of HPA Results for LCM Mixture Treatments using WBM ..................... 63
4.7. Summary of HPA Results for LCM Mixture Treatments using OBM ...................... 63
4.8. HPA Results for Foam Wedges Based System.......................................................... 77
4.9. Summary of the Hydraulic Fracturing Experiments for the Concrete Cores............. 88
4.10. Effect of Different Parameters on the Sealing Pressure and Model Fit ................... 92
4.11. Analytical Models Input Parameters ........................................................................ 93
xv
4.12. The Estimated Fracture Width from the Analytical Models and the Measured
Width under Microscope for the Five Fractured Core Samples .............................. 94
5.1. Particle Size Distribution Analysis Comparision for Barite and Hematite .............. 102
5.2. Description of the Different Fluid used in the Fracturing Experiments .................. 104
5.3. Summary of the Hydraulic Fracturing Experiments for the Shale Cores
(Contreras et al. 2014b) ........................................................................................... 105
5.4. Effect of Different Parameters on the Sealing Pressure and Model Fit for the
Subset Dataset .......................................................................................................... 111
5.5. Comparison of Percentage Match for the Different PSD Selection Criteria ........... 114
xvi
NOMENCLATURE
Symbol Description
LCM Lost Circulation Material
LOT Leak Off Test
XLOT Extended Leak Off Test
PSD Particle Size Distribution
LPM Loss Prevention Material
G Graphite
SCC Sized Calcium Carbonate
NS Nutshells
CF Cellulosic Fiber
HPA High Pressure Apparatus
LPA Low Pressure Apparatus
HPHT High Pressure High Temperature
PPA Particle Plugging Apparatus
NC Non-Controlled Fluid Loss
WBM Water-Base Mud
OBM Oil-Base Mud
SBM Synthetic-Base Mud
TS Tapered Slot Disc
SS Straight Slot Disc
NPs Nanoparticles
CS Control Sample
xvii
DC Blends Containing Calcium-Based Nanoparticles
DF Blends Containing Iron-Based Nanoparticles
PB Breakdown Pressure
T0 Tensile Strength
σH Maximum Horizontal Stress
σh Minimum Horizontal Stress
Wc Fracture Width
L Fracture Length
E Young’s Modulus
v Poisson’s Ratio
Pf Pressure along the fracture surface
pw Wellbore Pressure
rw Wellbore Radius
x Distance from the fracture mouth
p0 Pore Pressure
H0 Null Hypothesis
H1 Alternative Hypothesis
α Type I Error
Si Sample Variance
b0 Constrained Sealing Pressure under the Hypothesis
b Least Square
λ Lagrangian Multiplier
xviii
UNIT CONVERSION
To Convert From To Multiply By
bbl m3 0.1589873
micron mm 0.001
ppb kg/m3
2.85
psi Pascal 6894.76
lb/gal kg/m3
119.826
inches cm 2.54
1
1. INTRODUCTION
With the significant increase in oil demand, a huge number of conventional
hydrocarbon resources are being depleted. As a result, more challenging drilling
operations are required. When drilling challenging wells, such as extended reach wells or
deep water wells, the operational mud weight window narrows (Charlez, 1999). The
lower limit is increased due to higher collapse pressure in deviated wells (Aadnoy and
Chenevert, 1987). While the upper limit, controlled by the fracturing gradient, is reduced
due to higher equivalent circulation density (ECD) in extended reach wells, damaged
wellbores, natural fractures, and lower overburden gradient or as a result of wellbore
deviation (Whitfill and Wang, 2005; Fjaer et al. 2008).
Lost circulation events, defined as the loss of drilling fluids into the formation
(Messenger, 1981), are known to be one of the most challenging problems to be
prevented or mitigated during the drilling phase. The severity of the consequences varies
depending on the loss severity. The loss severity could be classified based on loss rate in
bbls/hr into; seepage (1-10 bbls/hr), partial (10 to 500 bbls/hr) and severe (>500 bbls/hr)
(Nayberg, 1987). By identifying the loss severity, proper remedial action can be taken to
mitigate or stop the losses.
In addition, lost circulation events, because they hinder further drilling, are well-
known for their contribution toward increased cost of drilling operations as a result of
non-productive time (NPT). Approximately 1.8 million bbls of drilling fluids are lost per
year (Marinescu, 2014). This huge number demonstrates the operational challenges
caused by losses while drilling.
2
Lost circulation events are often encountered when drilling into cavernous,
vugular, high permeability, naturally fractured formations or as a result of drilling
induced tensile failures. The loss mechanism differs for each candidate formation.
Drilling fluid losses into natural fractures, cavernous, vugular and high permeability
formations are initiated as soon as the drilling fluid pressure exceeds the pore pressure.
Losses into induced fractures are initiated when the drilling fluid pressure exceeds the
formation breakdown pressure (Howard and Scott, 1951) or in other words, when the
drilling fluid density (lb/gal) exceeds the formation breakdown pressure expressed in
equivalent density (lb/gal).
Assuming a linear poro-elastic stress distribution around the wellbore, the
formation breakdown pressure (PB) occurs when the tangential stress around the wellbore
exceeds the formation tensile strength (T0) (Eq. 1) (Fjaer et al. 2008).
𝑃𝐵 = 3𝜎ℎ − 𝜎𝐻 − 𝑃𝑃 + 𝑇0 (1)
Leak off tests (LOT) or extended leak off tests (XLOT) are often used to estimate
the formation breakdown pressure (fracture gradient) in order to construct the operational
mud weight window, which is very important to select mud weight limits and design
casing programs. However, different wellbore conditions affect the interpretation of
LOT/XLOT measurements (Nygaard and Salehi, 2011). Therefore, misinterpretation of
LOT/XLOT might result in inadequate estimation of formation breakdown pressure and
consequently improper mud weight selection, which could result in initiating fluid losses
into the formation.
3
Nygaard and Salehi (2011) provided LOT interpretation guidelines that takes into
consideration the effect of wellbore condition. In an intact wellbore, with no fractures,
fluid losses will occur whenever the drilling fluid pressure exceeds the formation
breakdown pressure defined in Eq. 1.
However, in the presence of a small fracture at the wellbore wall, losses will start
when the drilling fluid pressure surpasses the effective tangential stresses only (Eq. 2)
since the formation has already lost its tensile strength.
𝑃𝐵 = 3𝜎ℎ − 𝜎𝐻 − 𝑃𝑃 (2)
When a larger fracture exist, the effective tangential stresses goes to the minimum
horizontal stress that is perpendicular to fracture plane and the losses will occur as soon
as the drilling fluid pressure exceeds the minimum horizontal stress. In worst case
scenarios where a large fracture is intersecting with a network of fractures, losses will be
triggered when drilling fluid pressure exceeds the pore pressure.
One of the early efforts to cure losses by adding granular materials to the drilling
fluid was introduced by M T. Chapman in 1890 (White, 1956). Since then, lost
circulation materials (LCM) have been widely used to stop or mitigate drilling fluid
losses into the formation.
1.1. LOST CIRCULATION TREATMENTS
The way that lost circulation treatments are applied could be classified based on
the time when these treatments are implemented. It can be either before (preventive) or
after (corrective) the occurrence of the lost circulation event:
4
1.1.1. Corrective Treatments. Corrective treatments can be defined as any
method that is applied after the occurrence of the losses (Kumar and Savari, 2011). In this
approach, lost circulation treatments are either added continuously to the drilling fluid or
spotted as a concentrated pill in order to mitigate the losses.
Conventional LCM’s which can be described as physical discrete particles
suspended in the drilling fluid to prevent losses, are often classified based on their
appearance or physical properties into four main categories; as fibrous, flaky, and
granular or a blend of all three (Howard and Scott, 1951; White, 1956; Canson, 1985).
Conventional LCM treatments are often used to mitigate seepage or partial losses;
however, there is no industry common approach to deal with severe losses. Different
solutions to prevent/mitigate severe losses have been introduced and can be categorized
in two main categories; LCM solutions and mechanical solutions.
LCM solutions includes but not limited to concentrated LCM pills, cement
(Messenger and McNiel, 1952; Messenger, 1981; Morita and Fuh, 1990; Fidan et al.
2004), chemically activated cross-linked pills (CACP) (Bruton et al. 2001; Caughron et
al. 2002), cross-linked cement (Mata and Veiga, 2004), deformable-viscous-cohesive
systems (DVC) (Whitfill and Wang, 2005; Wang et al. 2005; Wang et al. 2008), nano-
composite gel (Lecolier et al. 2005), gunk squeezes (Bruton et al. 2001; Collins et al.
2010), and concentrated sand slurries (Saasen et al. 2004; Saasen et al. 2011).
Mechanical solutions include expandable liners (Filippov et al. 1999; Lohoefer et
al. 2000; Perez-Roca et al. 2003; Abouelnaaj et al. 2013), and casing/liner drilling
(Shepard et al. 2002; Warren et al. 2004; Karimi et al. 2011). The mechanical solutions
often come at the end of a solution list due to the high cost associated as well as the
5
required logistics limitation. Therefore, LCM solutions are often applied in the first place.
Even though often these LCM treatments were successful, the underlying mechanisms
are not well quantified. As a result, what worked in one well might not work for other
wells.
1.1.2. Preventive Treatments. Preventive treatments can be defined as those
treatments/solutions which are applied prior to entering lost circulation zones in order to
prevent the occurrence of losses. The overall objective of this method is to strengthen the
wellbore (Whitfill, 2008). The concept of wellbore strengthening can be defined as “a set
of techniques used to efficiently plug and seal induced fractures while drilling to
deliberately enhance the fracture gradient and widen the operational window” (Salehi and
Nygaard, 2012).
The overall objective of preventive LCM treatments is to widen the mud weight
window or in other words, to enhance the fracture gradient. The strengthening effect by
means of LCM addition was previously investigated experimentally (Morita et al., 1990;
Van Oort, 2011; Mostafavi et al. 2011; Salehi, 2012). In the field, the addition of LCM to
drilling fluids has increased the fracture gradients effectively above previously drilled
wells in permeable rocks (Fuh et al. 1992; Dupriest et al., 2008). Extensive experimental
investigations took place in order to fully understand how these techniques could possibly
increase the fracture gradient. Different models that explain wellbore strengthening
phenomena have been also introduced (Fuh et al. 1992; Alberty and McLean, 2004;
Aadnoy and Belayneh, 2004; Dupriest, 2005; Van Oort et al. 2011).
6
2. LITERATURE STUDY
The main objective of this section is to review the previous experimental
methods, wellbore strengthening theories, bridging theories and analytical modeling
efforts used to estimate fracture widths followed by a critical review of the literature to
identify the current gaps and limitations.
2.1. PREVIOUS EXPERIMENTAL INVESTIGATIONS
2.1.1. LCM Performance Evaluation. Evaluating the performance of LCM’s is
an important step for an optimized LCM treatment prior to field application. Different
parameters are measured and used as an indication of the LCM performance.
One of the early testing apparatuses to evaluate the effectiveness of LCM was
introduced in the early 1980’s by Sandia National Laboratories (Kelsey, 1981). This test
evaluates the effectiveness of LCM in mitigating losses based on the maximum pressure
that an LCM plug will seal across different slot sizes. The apparatus can be operated at
temperatures up to 300 F˚ and pressures up to 1000 psi.
HPHT filter press (Figure 2.1) and Particle plugging apparatus (PPA) (Figure 2.2)
are currently the most common tests used to evaluate LCMs effectiveness in controlling
fluid loss through hardened filter paper, slotted (SS), and tapered discs (TS) that simulate
natural/induced fractures or ceramic discs that simulate a porous formation (Collins et al.
2010; Kumar and Savari, 2011; Savari et al. 2013). The main difference between PPA
and HPHT is that for the PPA, the pressure is applied from the bottom. The Drilling fluid
containing LCM is forced to pass through these discs under a constant pressures and
7
temperatures. The results are evaluated based on both the total fluid loss within 30
minutes and the time it takes to form a tight seal.
Figure 2.1. HPHT Fluid Loss Apparatus
Figure 2.2. Particle Plugging Apparatus
(Fann Instrument Company, 2011)
A testing apparatus was designed to investigate the efficiency of different LCMs
in sealing impermeable fractured formations (Hettema et al. 2007; Van Oort et al. 2011).
The apparatus (Figure 2.3) uses three pumps and two accumulators to control and
monitor both the fluid and the fracture tip pressure while recording the fluid loss volume
across the fracture tip. The impermeable fracture was simulated by an opposed piston that
uses two matched 2.5-in.-diameter uneven aluminum platens where the fracture width is
set using three screws (Figure 2.4).
Drilling fluids containing LCMs are pumped through the open fracture at a
constant flow rate while keeping both fracture tip and fracture closure pressures constant
8
and monitoring the sealing pressure. A similar testing apparatus (Figure 2.5 and Figure
2.6) was also developed to determine the sealing efficiency in permeable formations
(Hettema et al. 2007).
Figure 2.3. Schematic of the Impermeable
Fracture Test Apparatus (Van Oort et al. 2011)
Figure 2.4. Corrugated Aluminum
Platens (Hettema et al. 2007)
Figure 2.5. Schematic of the Permeable
Fracture Test (Hettema et al. 2007)
Figure 2.6. Cross Section View of the
Fracture Cell (Kaageson-Loe al. 2009)
9
A modified PPA, which can operate at a higher pressure rating, was recently
introduced by Mostafavi et al. (2011). This test measures the performance of LCMs in
terms of the maximum sealing pressure achieved before the seal fails.
Experimental investigations (Hinkebein, et al. 1983; Hettema et al. 2007; Tehrani
et al. 2007; Kaageson-Loe et al. 2009; Savari et al. 2013) were conducted using the
testing apparatuses discussed above to investigate the effect of different parameters on
either the amount of fluid loss or the sealing pressure. These parameters include LCM
types, LCM concentrations, fracture width, and base fluids.
2.1.2. Hydraulic Fracturing Experiments. The drilling engineering association
(DEA-13) fracturing experiments, conducted on 30 inch x 30 inch large blocks, were one
of the very first efforts to investigate the loss circulation phenomena (Morita et al., 1990).
The study revealed that lost circulation pressure is highly dependent on the formation
Young’s modulus, which is in agreement with linear elastic fracture mechanic theories
(Griffith, 1920; Irwin, 1954; Anderson, 1995), as well as wellbore size, and type of
drilling fluid. It was observed that drilling fluids (containing solids) tend to seal/bridge
narrow fractures, which will results in the development of a stabilized fracture and the
breakdown will not occur until the fluid starts to penetrate into these fractures again. The
existence of a narrow fracture tip zone that prevents drilling fluid invasion was observed
in addition to a dehydrated (de-fluidized) mud zone that forms a plug behind the fracture
tip which also helped in sealing the fracture pressure. The filled fracture tip, or in other
words the formed plug size is dependent on the spurt loss and fluid loss, which are
strongly related to the drilling fluid properties. In addition, it was observed that the
fracture reopening pressure depends on the amount of mud cake left on the wellbore wall
10
(Morita et al., 1990, Onyia, 1994 Morita et al., 1996a, and Morita et al., 1996b). It was
also observed that water-base muds (WBM) will have higher reopening pressures than
oil-base muds (OBM) due to the tendency of developing a larger mud cake in WBM. The
theory of fracture initiation and propagation around borehole, where drilling fluids
enhances the stability, has been developed and validated with experiments and field
evidence (Morita et al., 1990).
An increase in the fracture propagation pressure was observed from the
experiments carried out by Fuh et al. (1992) when lost circulation materials (LCM’s)
were used and compared with experiments with untreated drilling fluid. Field trials
showed an increase in the fracture propagation pressure in the range of 3-6 lbm/gal and it
was concluded that this was due to the isolation of the induced fracture tip (Fuh et al.
1992).
The Global Petroleum Research Institute (GPRI) Joint Industry Project (JIP) took
place in the late 1990's to replicate the DEA-13 experiments but on a smaller scale using
4 inch (diameter) x 6 inch (height) cylindrical cores (Dudley et al., 2001 and Van Oort,
2011). The main objective of this project was to investigate the effectiveness of different
LCM’s in increasing the fracturing pressure. It was observed that, by using a specific type
and size of synthetic graphite, the fracturing pressure could be increased significantly. It
was also concluded that WBM would yield a higher increase in the fracturing pressured
compared to synthetic-based muds (SBM). In addition, it was found that the existence of
hydraulically conductive fractures could lower the ideal fracture reopening pressure to
the confining pressure or the minimum horizontal stress.
11
To investigate the effect of including nano sized particles in the LCM blend, , a
set of hydraulic fracturing experiments on 5 7/8 inch (diameter) x 9 inch (height)
cylindrical cores, were conducted on both permeable and impermeable rocks to
investigate the feasibility of using nanoparticles (NP) for wellbore strengthening
applications (Nwaoji et al. 2013). Sandstone cores were used to simulate permeable
formation while concrete cores were used to simulate impermeable formation.
The results indicated that iron hydroxide NP combined with graphite increased
the fracture pressure in permeable rocks by 70% when used in WBM. In addition, the
calcium carbonate NP combined with graphite produced a 36% increase in OBM in
permeable rocks. When these NPs were tested on impermeable samples (concrete), iron
hydroxide NP combined with graphite achieved a 25% increase. The blend of calcium
carbonate NP and graphite increased the fracture pressure by 22%. The results showed
how the size of LCM could results in increasing the fracture pressure due to optimum
sealing of fractures.
2.2. WELLBORE STRENGTHENING THEORIES
The concept of wellbore strengthening can be defined as “a set of techniques used
to efficiently plug and seal induced fractures while drilling to deliberately enhance the
fracture gradient and widen the operational window” (Salehi and Nygaard, 2012). This
approach depends on propping or sealing the fractures using LCMs (Salehi and Nygaard,
2011).
Different models have been introduced to address and explain wellbore
strengthening. Based on a theoretical approach and field trials, Fuh et al. 1992 presented
12
an approach to inhibit the initiation and propagation of fractures while drilling by the
addition of “loss prevention materials” (LPM), or in other words LCMs. The fracture
pressure inhibitor model suggests an increase in both the formation breakdown and
fracture propagation pressures as a result of LCMs screen out at the fracture tip, which
will form a good seal that prevents further fracture growth.
Based on a linear elastic fracture mechanics model, the stress cage model was
presented by Alberty and Mclean (2004). This model suggests that the hoop stress around
the wellbore is increased as a result of propping the fracture by LCMs set at the fracture
mouth to form a seal.
Fracture Closure Stress (FCS) model was introduced by Dupriest (2005) to
explain how LCM’s could increase the fracture gradient (FG) when added to drilling
fluids as a remedial solution for loss circulation. FCS is defined as “the normal stress on
the fracture plane keeping the fracture faces in contact”. The increase in FCS is achieved
by widening the fracture and sealing the fracture tip thus, compressing the adjacent rock
which will result in changing the near wellbore hoop stresses.
The Elastic-Plastic Fracture model presented by Aadnoy and Belayneh (2004)
explains how the FG could be increased above the theoretical Kirsch model value by
means of fracture healing. This model suggests that the fracturing resistance could be
improved as a result of the mud cake plastic deformation that builds up an effective
barrier at the fracture mouth. In addition, it was observed that the fracturing pressure is
sensitive to the type of LCM and concentration used.
Fracture propagation resistance (FPR) is another explanation introduced by Van
Oort et al. 2011 that was based on the DEA-13 experiments (Morita et al., 1990). This
13
work suggested that the increase in the fracture gradient resulted from enhancing the
fracture propagation pressure by means of tip isolation using LCM.
Salehi (2012) investigated the hypothesis of wellbore hoop stress enhancement
when sealing fractures using a three-dimensional poro-elastic finite-element model to
simulate fracture initiation, propagation and sealing in the near wellbore region. The
simulation results contradict that sealing fractures will increase the hoop stresses above
the ideal state of an intact wellbore, which can also be defined as the Kirsch solution.
Based on the simulation results, it was concluded that effective fracture mouth sealing
could restore the hoop stresses around the wellbore.
2.3. ANALYTICAL MODELING OF FRACTURE WIDTH
Different analytical models (Hillerborg et al. 1976: Carbonell and Detournay,
1995; Alberty and McLean, 2004; Wang et al. 2008; Morita and Fuh, 2012) based on
linear elastic fracture mechanics are used to estimate both; the fracture width and the
stress intensity factor at the tip of fracture. These models assume a symmetric fracture
around the wellbore as shown in Figure 2.7.
Figure 2.7. Symmetric Bi-wing Fracture around Wellbore
14
The LCM seal/plug is assumed to either form at the mouth of fracture or at some
distance from the fracture mouth. A brief description of fracture analytical models used in
LCM analysis and fracture width estimation is given below.
An analytical solution (Eq. 3) to estimate the fracture width (Wc) was presented
by Hillerborg et al. (1976). The solution was developed by combining both; fracture
mechanics principal and the finite element method.
𝑊𝑐 =2√2𝐿(𝑝𝑓−𝜎ℎ)
√√(𝑝𝑓−𝜎ℎ)4
+𝐸2𝜎ℎ
2
𝜋2(1−𝜐2)2−(𝑝𝑓−𝜎ℎ)
2
(3)
Where (pf) is the pressure along the fracture, (σh) is the minimum horizontal
stress, and (L) is the fracture length. Assuming a linear crack, the crack propagation
decreases when the crack width increases. The model was developed for aggregated
material like concrete where the fracture width is almost equivalent to the maximum size
of aggregate particles in the order of few millimeters. The fracture width solution is
primarily a function of fracture length and the pressure along the fracture, which is
assumed to be constant.
Carbonell and Detournay (1995) developed a semi-analytical solution to estimate
the fracture propagation pressure using a dislocation based approach. Their solution was
based on the original singular solution of a finite edge dislocation on the boundary of
circular wellbore in an infinite elastic plane derived by Warren (1982). Both, the stress
intensity factor at the tip of fracture and the fracture width could be determined using
Guass-Chebyshev (Erdogan and Gupta, 1972) integration formula to the singular integral
15
of dislocation density function (Eq. 4) where m, h, and tk are Chebyshev polynomials and
Gauss-Chebyshev integration parameters.
𝑊𝑐 =𝜋(𝐿)
2𝑚
4(1−𝜐2)
E∑ ℎ(𝑡𝑘)𝑚
𝑖=0 (4)
The fracture propagation pressure could be calculated using the Newman’s
solution (1971) and the calculated stress intensity factor. Although, the model gives a
solution for the fracture width, fracture length is an input parameter. In addition, the
pressure inside the fracture is assumed to be constant.
A 2-D line crack solution (Eq. 5) was presented by Alberty and McLean (2004) to
compare the results of the fracture width with the result from a finite element analysis.
𝑊𝑐 =4(1−𝜐2)
E(𝑝𝑤 − 𝜎𝐻)√(𝐿 + 𝑟𝑤)2 − 𝑥2 (5)
Where (pw) is wellbore pressure and (rw) is the wellbore radius. The original
solution was for a small line crack (Continuum Fracture Mechanics). Alberty and
McLean (2004) modified the original analytical solution by including the wellbore radius
to make it applicable for wellbore analysis which they compared to the developed finite
element analysis (FEA). The fracture width could be estimated at any distance from the
fracture mouth; however, the estimation is dependent on the fracture length similar to
Hillerborg et al. (1976) and Carbonell and Detournay (1995).
Simplifying an earlier model (Deeg and Wang, 2004), which was based on three
distinct symmetrical pressurized regions to two pressurized regions, Wang et al. 2008
presented a semi-analytical solution for fracture length (Eq. 6).
16
𝑊𝑐 =8(1−𝜈2)𝐿
𝜋𝐸
{𝜋
2𝑎 (𝑝0−𝜎ℎ)+(𝑝𝑤−𝑝0)(𝑎 𝑏 + ∑ sin(2𝑛 𝑏)(2𝑎 cos(2𝑛 𝑏)+
𝑟𝑤 sin(2𝑛 𝑏)
𝐿𝑛)∞
𝑛=1 )}
(2𝑛−1)(2𝑛+1) (6)
Where:
𝑎 = √1 − (𝑟𝑤
𝐿)
2
(7)
And
𝑏 = sin−1 (𝑟𝑤
𝐿) (8)
The crack length is calculated by comparing the stress intensity factor of fracture
to the fracture toughness of rock using an analytical solution for the stress intensity
factor. Similar to the previous models, the fracture length is the dominant parameters to
determine the fracture width.
Based on a 2-D boundary element model and assuming a small fracture around
wellbore, a closed form solution was developed to estimate the fracture width (Morita
and Fuh, 2012). The analytical model (Eq. 9) determines the fracture width (Wc) as a
function of rock properties (E and v), in-situ stress (σh), the length of fracture (L), and the
wellbore (pw) and pore pressure (po), however, in the described workflow (Morita and
Fuh, 2012) to calculate fracture propagation pressure (Pf), the fracture width is an input
parameter to find the fracture length.
𝑊𝑐 =4(1−𝜐2)
𝐸𝐿(𝜎ℎ − 𝑃𝑓) (9)
The fracture propagation pressure came from the linear elastic fracture mechanics
of solids (Tada et al. 2000), which is based on the known toughness at the tip of the
17
crack. Since the plugging material (LCM) might seat somewhere along the fracture
(possibly close to tip of fracture) rather than mouth of fracture, the alternate closed form
solution was provided.
The main limitation of Morita and Fuh’s (2012) model is the use of fracture
geometry as an input parameter due to the fact that there is cumbersome to accurately
measure it. The model is also based on an assumed fracture length of 6 inches, which is
not a verified assumption. Overall, the model is highly sensitive to fracture geometry,
which might results in unrealistic estimated fracture propagated pressure according to the
authors.
2.4. REVIEW OF BRIDGING THEORIES
Bridging can be defined as the buildup of solids within the formation pore spaces
to restrict fluid losses into the formation. Different bridging theories are used as a
guideline to enhance the bridging capabilities of drill-in fluids (i.e. drilling through
reservoir sections), thus minimizing the formation damage caused by fluid losses into the
formation.
Two rules to select an optimum PSD and concentration that is required to
minimize formation damage caused by mud invasion into the formation were proposed
by Abrams in 1977. The first rule suggests that the average particle size (D50) of
bridging materials should be equal or slightly larger than one-third the formation average
pore size. The second rule suggests that the selected bridging materials concentration
should be at least 5% by volume of total solids in the mud formulation.
18
Another criterion for selecting PSD based on laboratory and field application was
suggested by Smith et al. 1996. It was later recommended by Hands et al. (1998) based
on field observations where formation damage was reduced and production rates were
increased. This rule suggests that the D90 of PSD of bridging materials should match the
pore size in order to minimize the fluid invasion into the formation.
To enhance the bridging effectiveness by having a wider range of optimized PSD,
two more target fractions were suggested by Vickers (2006). This method suggests that
both D25 and D75 should be included in the selection criteria, in addition to D10, D50,
and D90, based on a set of laboratory studies. The five D-values selection criteria is a
combination of both Abrams and Hands rules where the D90 should be equal to the
largest pore throat, D75 is less than two-third of the largest pore throat, D50 equal or
slightly larger than one-third, D25 equal to one-seventh of the mean pore throat and D10
greater than the smallest pore throat.
To overcome the uncertainty in the expected fracture widths and optimize PSD, a
new method, which optimize PSD based on fracture width rather than pore throat sizes,
was proposed by Whitfill (2008). This method proposed that the D50 value should be
equal to the fracture width to ensure the formation of an effective seal or plug.
2.5. CRITICAL REVIEW OF THE LITERATURE
The literature study of LCM performance evaluation revealed that there are
pressure, temperature, and fracture width limitations for the different testing apparatuses.
Some of the previous tests emphasized the amount of fluid loss (Savari et al. 2013) while
other tests focused on the sealing pressure (Hettema et al. 2007; Tehrani et al. 2007;
19
Kaageson-Loe et al. 2009; Van Oort et al. 2011; Mostafavi et al. 2011). HPHT filter press
and PPA tests might give a good indication of how much of the drilling fluid might be
lost into the formation. However, these tests were conducted at a constant pressure, which
is not always the case while drilling.
From the permeable and impermeable fracture test (Hettema et al. 2007; Tehrani
et al. 2007; Van Oort et al. 2011), particle size distribution was found to be a critical
parameter for fracture sealing. However, the maximum particle size that can be tested
using these two apparatuses must be less than 1 mm to avoid blocking the tubing in the
apparatus. This limitation resulted in using a maximum fracture width of 1000 microns. It
was also concluded that higher LCM concentration of larger particle sizes would improve
the sealing capabilities. The variation in base fluid showed an impact on the fracture
sealing pressure where higher sealing pressure and lower fluid loss values were obtained
when LCMs were added to OBM compared to WBM and SBM. A summary of the
previous LCM testing apparatuses and their limitation is tabulated in Table 2.1.
Table 2.1. Summary of the Previous LCM Testing Apparatuses and their Limitations
Apparatus Measured
Values
Max.
Press.
(psi)
Temp.
˚F
Filtration
Medium
Fracture
Width (mm)
LCM Tester Sealing
Efficiency 1000 300
API Slots
TS
1 - 5 mm (SS)
2 – 12.7 (TS)
HPHT Fluid Loss Filtration
Characteristic
1500-
2000
350-
500
Filter paper
Ceramic discs N/A
PPA Filtration/
Bridging
4000-
5000 500
Ceramic discs
API/ TS
2 mm (TS)
5 mm (SS)
Impermeable
Fracture Test
Sealing
Efficiency 1250 N/A
Uneven
aluminum
platens
0.3 - 1 mm
20
Table 2.1. Summary of the Previous LCM Testing Apparatuses and their Limitations
(Cont’d)
Permeable Fracture
Test
Sealing
Efficiency 6000 N/A Porous plates 0.25 – 1 mm
Modified PPA
Filtration/
Bridging
Characteristic
8700 N/A Fabricated
Steel Fractures 0.3 - 0.7 mm
Even though a large dataset was developed in these studies, only a limited amount
of the laboratory results were reported in these studies (Appendix A). There are still
limited published results on how fracture width, permeability of the rock, type of base
fluid, mud density, temperature, and particle size distribution could affect the
performance of different LCM’s.
Based on the hydraulic fracturing experiments and the strengthening theories
reviews, we can see that all the previous studies agreed on the effect of LCM addition in
terms of increasing the fracture gradient but no agreement has been achieved on how
LCM size and strength could affect the strengthening. Table 2.2 summarizes the
discussed wellbore strengthening theories and their emphasis on the material size and
strength.
Table 2.2. Summary of the Competing Wellbore Strengthening Theories
Wellbore Strengthening Model Material
Size
Material
Strength Authors
Fracture Pressure Inhibitor Important Important Fuh et al. 1992
Stress Cage Important Important Alberty and McLean,
2004
21
Table 2.2. Summary of the Competing Wellbore Strengthening Theories (Cont’d)
Fracture Closure Stress (FCS) Not
Important
Not
Important Dupriest, 2005
Fracture Healing Important Important Aadnoy and Belayneh,
2004
Fracture Propagation Resistance
(FPR) Important
Not
Important Van Oort et al. 2011
Fracture Mouth Sealing Important Important Salehi et al. 2015
The lack of understanding has resulted in a controversy about the models that
explains the strengthening mechanism. Some of these models (Alberty and Mclean,
2004) suggested that the strengthening effect could be achieved by sealing the fracture
mouth while the others (Fuh et al. 1992; Dupriest, 2005; Van Oort et al. 2011) suggested
sealing the fracture tip. In addition, some models (Alberty and Mclean, 2004; Dupriest,
2005) suggested that sealing fractures would result in hoop stress alteration while others
(Salehi and Nygaard, 2015) suggested that hoop stresses could only be restored and not
increased.
All the analytical fracture models considered in this study estimates the fracture
width primarily as a function of rock properties, fluid pressure within fracture, in-situ
stresses, and fracture length. One of the main limitations of analytical models is using the
length of fracture as input parameters since the measurement or estimation of the fracture
length could be not practical. Some models are assuming fracture length close to 6 in
(Carbonell and Detournay, 1995; Morita, et al. 2012) which might not be realistic for
field applications. Assuming a constant fluid pressure within the fracture is another
22
simplification, which might not be true due to the fact that the pressure beyond the
plugging material might decrease gradually for both permeable and non-permeable
(shale) formations. The shape of the induced or natural fractures might not necessarily be
a line crack and the width of fracture might change along the fracture.
The current bridging theories and their selection criteria (Table 2.3) were
developed to optimize PSD for drill-in fluids based on the pore size distribution, which
can be estimated from thin section analysis, scanning electron microscopy, mercury
injection, or derived from permeability measurements (Dick, 2000; He and Stephens,
2011). The pore sizes dimensions are relatively small compared to fracture widths and
therefore, using these criteria might not be applicable for fracture sealing. Whitfill (2008)
had the only proposed selection criteria based on fracture sealing application; however,
there were no experimental validation presented.
Table 2.3. Summary of Current Selection Criteria
Method Selection Criteria Authors
Abrams Rule D50 ≥ 1/3 the formation average pore size Abrams 1977
D90 Rule D90 = the formation pore size Smith et al. 1996
Hands et al. 1998
Vickers Method
D90 = largest pore throat
D75 < 2/3 the largest pore throat
D50 ≥ 1/3
D25 = 1/7 the mean pore throat
D10 > the smallest pore throat
Vickers et al. 2006
Halliburton Method D50 = fracture width Whitfill, 2008
23
2.6. RESEARCH OBJECTIVES
The literature review in the previous section shows that there is an agreement that
plugging existing fractures by means of LCM resulted in either restoring or increasing the
field experienced fracture gradient (i.e. when losses occur). It can also be concluded that
the sealing efficiency of different LCM’s have a strong effect on wellbore strengthening
by allowing drilling to continue after losses has occurred. However, no single study has
shown how the type and concentration of LCM’s could contribute toward the
enhancement of the fracture gradient. The mechanism of how different LCM’s
characteristics affect the fracture initiation or propagation pressure is still not well
understood.
The main objective of this dissertation is to investigate the feasibility of
effectively sealing wide fractures, with specific application to overburden shales, using
conventional LCM and potentially enhance the fracture gradient. The main objectives can
be broken down into four sub-objectives as follows:
1. Investigate how LCM properties such as shape, PSD, LCM concentration,
types and mixtures impact sealing efficiency.
2. Determine limitations of conventional LCM for sealing existing wide fractures.
3. Investigate the enhancement of the fracture gradient of intact wellbores by the
addition of LCMs.
4. Evaluate the ability to estimate fracture width analytically.
The ability to predict fracture widths and optimize PSD based on the predicted
widths should help in designing effective LCM treatments that would mitigate or prevent
losses, which will be very beneficial in both planning and drilling challenging wells.
24
3. METHODOLOGY
This chapter outlines the different research methodologies used to address the
research objectives listed in chapter 2. In addition, procedures, experimental setups used,
materials as well as the different methods of data analysis will be presented in details.
Due to the large number of up to date available LCM’s and their different
applications, LCM’s will be re-classified into different categories. Dry sieve analysis
techniques were used to analyze the particle size distribution (PSD) of LCM treatments in
order to understand how PSD could affect both the fracture sealing and the sealing
pressure for a known fracture width. PSD analyses should answer the following
questions; what is the role of PSD in effective fracture sealing? Did the effective blends
follow a specific selection criterion? What is the best criterion to select PSD with respect
to fracture width for an effective sealing?
Optical microscopic analysis were undertaken for LCM particles in order to
investigate how LCM shape, in terms of sphericity and roundness, could affect the
performance of LCM in forming an effective seal. Formed seals within the fractures were
examined using a conventional scanning electron microscope (SEM) to investigate how
the particles shape affected the permeability of the formed seal.
Evaluating the performance of LCMs consisted of two steps using two developed
apparatuses. First, the low pressure apparatus (LPA) was used as a quick indicative
measurement that served as a screening phase of the different conventional treatments,
and based on the results; further evaluation of LCM performance at higher pressure was
conducted using the high pressure apparatus (HPA).
25
The effects of varying LCM type, formulation, concentration, fracture width
particle size distribution, base fluid, density, weighting material, and temperature were
studied with respect to differential pressure and fluid loss volume. The objectives are to
establish a better understanding of how these parameters could affect the sealing
efficiency of LCMs and identify their limitations in sealing fractures.
Hydraulic fracturing experiments were carried out using concrete core sample.
These experiments were used to: a) measure the fracture breakdown and re-opening
pressures for OBM, b) investigate the strengthening effect as a result of adding LCM to
OBM, c) investigate whether the sealing pressure measured using the high pressure LCM
tester correlate with the enhancement in breakdown or re-opening pressure.
Statistical analysis of the LCM evaluation results was performed to define the
parameters with the highest effect on the sealing pressure. The knowledge of the
dominant parameter will help in improving in designing an effective LCM treatment.
Finally, the analytical models were used to estimate the fracture widths and the
results were compared to the fracture widths measured under optical microscopy in order
to determine the applicability of the analytical models in estimating fracture widths.
3.1. LCM CHARACTERIZATION
Optical microscopy was used to look at both LCM and the formed plug while a
Scanning Electron Microscopy (SEM) was used to examine the formed plugs. LCMs
PSD were analyzed using dry sieve analysis techniques. In addition, the swelling of one
LCM was quantified at high temperature.
26
3.1.1. LCM Classification. Conventional LCM’s has been classified based on
their appearance as fibrous, flaky, and granular or a blend of all three (Canson, 1985).
LCM’s have different physical and chemical properties and therefore a proper LCM
selection is a key factor for a successful lost circulation treatment.
Howard et al. 1951 classified LCM’s based on their physical properties into four
groups: fibrous, granular, lamellated and dehydratable. Robert J White (1956) modified
the previous classification by replacing the dehydratable category with mixture of LCM
category.
A review of available LCM’s that are currently used in drilling operations was
conducted in order to re-classify LCM into more representative classification (Alsaba et
al. 2014a).
3.1.2. Particle Size Distribution Analysis. Two methods were considered to
determine particle size distribution (PSD); sieve analysis and laser diffraction. However,
since the provided D50 values (from manufacture) for some of the used LCMs in this
study are 2000 microns and due to the limitation of the laser diffraction techniques in
measuring particles larger than 2000 microns (Zhang et al 2015), dry sieve analysis was
used instead to analyze PSD for LCMs. For the weighting materials, the particle size
distribution was measured with MicroTrac S3500 laser diffraction particle size
distribution analyzer.
Samples of LCM treatments were sieved through a series of stacked sieves using
a Gilson mechanical tapping sieve shaker (Figure 3.1). The cumulative weight percent
retained for each sieve size was calculated from the measured weight retained in each
sieve and then plotted versus the sieve sizes.
27
The main five parameters of interest for the different blends that sealed different
fractures width, which were obtained from the resulting plot, are the D10, D25, D50,
D75, and D90, measured in mm and converted to microns where 1 mm = 1000 microns.
In addition, random blends that failed in creating a seal or resulted in a high fluid loss
during the screening phase were analyzed to understand the reason behind their failure in
forming a seal within the fracture.
Figure 3.1. Dry Sieve Analysis Setup
3.1.3. Optical Microscopy. Optical microscopic images were taken for LCM
particles using the INSIZE digital measuring microscope (ISM-PM200) with
magnification up to 200X. The formed plugs were examined under a polarized light
microscopy (Nikon Eclipse E400 POL) that was connected to a digital camera.
28
The sphericity and roundness chart presented by Powers (Figure 3.2), was used to
provide a qualitative visual inspection of LCM particles.
Figure 3.2. Sphericity and Roundness Chart (Powers, 1953)
3.1.4. Scanning Electron Microscopy. Formed seals within the fractures were
analyzed under scanning electron microscope (SEM) using a Hitachi S-570 SEM at the
S&T’s advanced materials characterization laboratory (AMCL). The formed seal samples
were prepared by removing the seal gently from the tapered discs after running the high
pressure apparatus test. After that, the seals were left to dry prior to inspection.
3.1.5. Swelling. A swelling test apparatus (Figure 3.3) using a linear variable
differential transformer (LVDT) was developed following the design and procedure used
to quantify the swelling of shale (Fann, 2015). The apparatus consist of 5 main
components labeled on Figure 3.3 as follows:
1- Adjustable stand to hold LVDT
2- LVDT
3- Mold to hold the sample for compaction and then running the test
29
4- Plate inserted on top of the sample to ensure distributed swelling measurement
5- Data logging (swelling versus time)
The selected LCM sample was first sieved to remove larger particles in order to
have a well-sorted sample. Then the sample was compressed inside the sample holder (3)
and covered with a thin plate (4). The LVDT (2) is lowered and the test was run for 24
hours and the change in the sample height represents the percentage of swelling. To
quantify the effect of temperature on swelling, two tests were conducted; first at room
temperature and secondly at 180 ˚F.
Figure 3.3. Schematic of the Swelling Test Apparatus
3.2. LCM PERFORMANCE EVALUATION
This section presents the preparation and formulation of the evaluated LCM
treatments as well as the drilling fluids used in this study.
(1)
(3)
(5)
(4)
(2)
30
3.2.1. Fluid Used. Unweighted water-based mud (WBM) containing 7%
bentonite was used to eliminate negative or positive effects, if any, of drilling fluid
additives such as weighting materials and fluid loss reducers on LCM performance. In
addition, a pre-mixed low-toxicity mineral oil-based mud (OBM) was used to investigate
the effect base fluid on the performance of LCM. Rheological parameters for the two
fluids used are summarized in Table 3.1.
The effect of density was investigated by repeating some tests at different fluid
densities up to 16.5 lb/gal. In addition, the effect of weighting material was studied by
repeating tests using fluids weighted with barite and hematite.
Table 3.1. Drilling Fluid Properties as Measured in Accordance with API Standards
(API- RP 13B)
Drilling Fluid Properties WBM OBM
Density (lb/gal) 8.6 11
Gels 10s (lb/100 ft2) 18 7
Gels 10m (lb/100 ft2) 35 10
Plastic Viscosity (cP) 10 36
Yield Point (lb/100 ft2) 19 14
3.2.2. Conventional LCM Used. Four commonly used conventional LCMs
having different physical properties, including graphite (G), cellulosic fiber (CF), sized
calcium carbonate (SCC), and nutshells (NS), were used in this investigation to formulate
LCM treatments at different concentrations.
31
To represent background treatments, a total concentration of 15 ppb was used to
formulate single LCM blends. For single LCM concentrated pills, a total concentration of
50 ppb was used. Table 3.2 shows the percentage of total concentration of each LCM to
formulate any individual LCM blend.
Table 3.2. Single LCM Treatment Formulation
Blend Name G Blends SCC Blends NS Blends CF Blends
Blend # 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3
LCM
Type
D(50)
microns % of Total Concentration if used Individually
Gra
ph
ite (
G) 50 20 14 0 0 -- -- -- -- -- -- -- -- -- -- --
100 20 20 20 0 -- -- -- -- -- -- -- -- -- -- --
400 30 26 40 50 -- -- -- -- -- -- -- -- -- -- --
1000 30 40 40 50 -- -- -- -- -- -- -- -- -- -- --
Siz
ed
Ca
lciu
m
Ca
rb
on
ate
(S
CC
)
5 -- -- -- -- 16 6 0 0 -- -- -- -- -- -- --
25 -- -- -- -- 16 6 0 0 -- -- -- -- -- -- --
50 -- -- -- -- 16 13 0 0 -- -- -- -- -- -- --
400 -- -- -- -- 16 21 33 20 -- -- -- -- -- -- --
600 -- -- -- -- 18 27 33 27 -- -- -- -- -- -- --
1200 -- -- -- -- 18 27 34 53 -- -- -- -- -- -- --
Nu
t S
hell
s
(NS
)
620 -- -- -- -- -- -- -- -- 33.3 0 0 0 -- -- --
1450 -- -- -- -- -- -- -- -- 33.3 50 100 0 -- -- --
2300 -- -- -- -- -- -- -- -- 33.3 50 0 100 -- -- --
Cell
ulo
sic
Fib
er
(CF
)
312 -- -- -- -- -- -- -- -- -- -- -- -- 50 100 0
1060 -- -- -- -- -- -- -- -- -- -- -- -- 50 0 100
For example, to mix 1 bbl. Of drilling fluid containing a 15 ppb Graphite Blend
#2 (Table 3.2), the formulation will be; 15 ppb x 14% = 2.1 ppb (D50 = 50 microns), 15
32
ppb x 20% = 3 ppb (D50 = 100 microns), 15 ppb x 26% = 3.9 ppb (D50 = 400 microns),
and 15 ppb x 40% = 6 ppb (D50 = 1000 microns).
Three recommended LCM mixtures (Aston et al. 2004; Hettema et al. 2007,
Kumar et al. 2011) were used in this study to benchmark the results (Table 3.3). Graphite
and sized calcium carbonate blends # 1 and 2 were investigated at two different
concentrations, 30 ppb and 80 ppb to follow the recommendations by Aston et al. (2004).
Graphite and nutshells blends were tested at the concentrations of 20 ppb and 40 ppb as
suggested by Hettema et al. (2007). A 55-ppb blend combining graphite, sized calcium
carbonate, and cellulosic fiber was investigated following Kumar et al. (2011). Graphite
and sized calcium carbonate blends # 3-9 were evaluated at high concentration of 105
ppb as recommended by Det Norske Oljeselskap ASA for field applications (A. Saasen,
2015, Pers. Comm.).
Table 3.3. LCM Mixture Formulation
Blend Name G & SCC Blends G, SCC &
CF Blends
G & NS
Blends
Blend # 1 2 3 4 5 6 7 8 9 1 2 1 2
LCM
Type
D(50)
microns % of Total Concentration if used in combinations
Gra
ph
ite (
G) 50 10 6.7 -- -- -- -- -- -- -- 3.6 2.4 10 6.5
100 10 10 -- -- -- -- -- -- -- 3.6 3.6 10 10
400 15 13.3 33.3 -- -- -- -- 33.3 -- 5.5 4.8 15 13.5
1000 15 20 -- -- -- -- -- -- -- 5.5 7.3 15 20
Siz
ed
Ca
lciu
m
Ca
rb
on
ate
(S
CC
)
5 3.3 -- -- -- -- -- -- -- -- 4.4 0 -- --
25 3.3 -- -- -- -- -- -- -- -- 4.4 0 -- --
40 -- -- 33.3 -- -- -- -- 16.7 --
50 6.7 -- -- -- -- -- -- -- -- 9.5 0 -- --
400 10 16.5 -- -- -- -- -- -- -- 15.3 24 -- --
600 13.3 16.5 -- -- -- -- -- -- -- 19.6 24 -- --
33
Table 3.3. LCM Mixture Formulation (Cont’d)
Siz
ed
Ca
lciu
m
Ca
rb
on
ate
(SC
C)
1200 13.3 17 -- -- -- -- -- -- -- 19.6 24.7 -- --
1400 -- -- 33.3 -- 33.3 50.0 33.3 16.7 37.5 -- -- -- --
2400 -- -- -- -- -- -- 33.3 33.3 37.5 -- -- -- --
Nu
t
Sh
ell
s
620 -- -- -- -- -- -- -- -- -- -- 16.5 16.5
1450 -- -- -- -- -- -- -- -- -- -- 16.5 16.5
2300 -- -- -- -- -- -- -- -- -- -- 17 17
Cell
ulo
sic
Fib
er
(CF
)
312 -- -- -- -- -- -- -- -- 4.5 4.5 -- --
1060 -- -- -- -- -- -- -- -- 4.5 4.5 -- --
Pro
prie
tary
G&
SC
C
Ble
nd
500 -- -- 100 66.7 50 33.3 -- 25 -- -- -- --
3.2.3. Unconventional LCM Used. A wedge foam based system was included in
the study as an unconventional LCM. This system is used to address fracture size and
shape uncertainties when experiencing losses (Wang 2011). This system consists of two
main components: foam wedges at different sizes that are highly deformable (Figure 3.4)
and micron-sized particles.
This high deformability allows the foam wedges to be compressed and forced into
openings of different sizes and shapes. Once the openings are filled with the foam, they
will form a highly permeable filtration bridge for the second component. The second
component consists of high fluid loss fine particles that will form a solid plug on the
filtration bridge.
34
Figure 3.4. Wedge Foam of Three Sizes; small (S), medium (M), and coarse (C)
(Taken from Wang, 2011)
Table 3.4 shows the formulation used to mix unweighted pill for specific fracture
widths. The coarse WEDGE-FOAM was only used with a 10000-micron fracture width
while all tests were mixed with WEDGE-FOAM (M).
Table 3.4. WEDGE-SET Formulation for Specific Fracture Widths
Material
Fracture Width
Up to 8000 microns 10000 micron
Fracture Width
WEDGE-SET 100 ppb 100
WEDGE-FOAM (M) 0.25 ppb 0.5 ppb
WEDGE-FOAM (C) -- 1 ppb
3.2.4. Fracture Width Variation. A set of tapered disc that have a larger
opening (fracture aperture) and tapers down to a smaller opening (fracture tip) as shown
in Figure 3.5 were manufactured at Missouri S&T to simulate sealable fracture widths. In
addition, two commercially available straight slotted discs were used. The tapered and
straight slotted discs (Figure 3.6) specifications are tabulated in Table 3.5.
35
Figure 3.5. Schematic of Tapered Discs
Figure 3.6. Set of Tapered and Straight
Slotted Discs
Table 3.5. Tapered and Straight Slotted Discs Specifications
Disc
#
Diameter
(inches)
Thickness
(inches)
Fracture Aperture
(microns)
Fracture Tip
(microns)
Fracture
Angle
(Degrees)
TS1 2.5 0.25 2500 1000 6.73
TS2 2.5 0.25 3000 1500 6.73
TS3 2.5 0.25 4000 2000 8.95
TS4 2.5 0.25 5000 2000 13.3
TS5 2.5 0.25 4500 2500 8.95
TS6 2.5 1.0 5000 3000 8.95
TS7 2.5 1.0 7000 5000 8.95
TS8 2.5 1.0 18000 10000 8.95
SS1 2.5 0.25 3000 3000 0
SS2 2.5 1.0 5000 5000 0
NOTE: Tapered Discs (TS) Straight Slot (SS)
3.2.5. Low Pressure LCM Testing. The low pressure apparatus (LPA) shown in
Figure 3.7 (a) was manufactured based on the standard API filter press design (API RP
Fracture Aperture
Fracture Tip
36
13B). A snug-fit spacer, shown in Figure 3.7 (b), was manufactured to serve as a holder
for the tapered discs (Figure 3.7 (c)) to prevent plugging the cell outlet.
The test is run by filling the cell with the simple WBM containing LCM’s and
then slowly apply a 100 psi to force the fluid containing LCM to flow through the tapered
disc until no more fluid is coming out. The most important parameter here is the fluid
loss volume. The fluid loss volume could tell if the concentration and PSD are able to
seal a specific fracture width.
Figure 3.7. (a) Low Pressure LCM Testing Apparatus (LPA), (b) Snug-fit Spacer, (c)
Tapered Disc on the Spacer
3.2.6. High Pressure LCM Testing. The effectiveness of LCM treatments was
investigated using a high-pressure LCM testing apparatus, which was designed and
(c)
(b)
(a)
37
manufactured to evaluate the formed seal integrity. The seal integrity, or in other words
the sealing pressure is defined here as the maximum pressure at which the formed seal
breaks and fluid loss resumes. Figure 3.8 shows a schematic drawing of the experimental
setup.
A plastic accumulator (1) used to transfer the drilling fluids to the metal
accumulator (2) prior to pressurizing the fluids containing LCM treatments inside the
testing cell (3) through the tapered discs (4) using an IscoTM
pump (DX100) (5) to
provide injection pressure, which was connected to a computer to record pressure versus
time. The testing cell could also be heated up to 300˚F using an insulated heating blanket,
which was used to investigate the effect of temperature by repeating some tests at a
higher temperature (180˚F) and compare it with the results at ambient temperature.
Figure 3.8 Schematic of the High Pressure Testing Apparatus
(4)
(2)
(5) (3)
(1)
38
The conventional LCM test is run by injecting fluids containing LCMs at a
constant flow rate of 25 ml/min until a rapid increase in the injection pressure is observed
(Figure 3.9). This increase indicates the formation of a seal. Once the seal has formed, the
drilling fluid containing no LCMs is injected continuously until the maximum sealing
efficiency is reached. A significant drop in the pressure is observed as a result of breaking
the seal. However, the test is repeated as cycles until no further seal can be formed. For
tests at high temperature, fluids containing LCM were poured into the high pressure cell
and heated up to 180oF before running the test.
Figure 3.9. Pressure vs. Time Plot Generated from HPA Tests for Conventional LCMs
To evaluate the unconventional LCM, a hesitation squeeze technique was used.
The main objective of this technique is to ensure placing the LCMs (WEDGE-FOAM) in
the fracture and forming the seal within the fracture. Continuous cycles of injection,
0
100
200
300
400
500
100 120 140 160 180 200 220 240 260
Se
ali
ng
Pre
ss
ure
(p
si)
Time (s)
Seal Broken Seal Formation
39
followed by a shut in time, were applied during the test to simulate the hesitation
squeeze. Continuous injection with no shut in period caused the foam wedges to pass
through the fractures before forming a seal. Once the injection stops, the remaining
pressure in the system will push the fluid into the fracture but at a reduced pressure.
Different injection rates were used to determine the optimum injection rate that would
form the seal.
The testing procedures can be summarized into three main steps as follows: (a)
seal initiation, (b) plug development, and (c) plug integrity evaluation (Figure 3.10). The
seal is initiated by injecting fluid at a constant flow rate for 40 seconds and then injection
stopped for 60 seconds. The step is repeated until pressure starts to increase. This process
is continued to develop the plug until significant pressure increase up to 100 psi is
observed. Finally, the plug is further developed and its integrity is tested by injecting
fluid until the pressure goes up to 500 psi and then injection stops for 60 seconds. This
step is repeated and pressure is increased by 500 psi each time until the seal breaks.
Figure 3.10. Pressure vs. Time Plot Generated from HPA Tests for Unconventional LCM
0
500
1000
1500
2000
2500
3000
100 1100 2100 3100 4100 5100 6100Seali
ng
Pre
ssu
re (
psi)
Time (s)
(a) (b) (c)
40
3.3. HYDRAULIC FRACTURING TEST
Hydraulic fracturing experiments were performed on five concrete cores using
five different drilling fluid formulations (OBM). The tests were conducted using a triaxial
pressure cell, which was designed and manufactured at Missouri S&T.
Figure 3.11 shows a schematic drawing of the hydraulic fracturing apparatus.
Two pumps are used to apply confining and injection pressure. Hydraulic hand pump is
used to apply overburden stress on the core sample. A metal accumulator was used to
inject fluids into the core. Injection pressures was recorded using LabVIEW© software.
Figure 3.11. Schematic of the Hydraulic Fracturing Apparatus
41
3.3.1. Core Preparation. Cement core samples were prepared using Portland
cement (Class H) to simulate impermeable formations. The cement core preparation steps
are summarized as follows:
1- Class H cement was mixed with API recommended water requirements of 38%
by weight of cement in a large batch following the standard mixing procedures to
ensure the same physical properties of the fractured cores.
2- The cement mixture was poured into 5 7/8 inch (diameter) x 9 inch (height)
molds and left to cure for at least 7 days.
3- A ½ inch wellbore was drilled in the cement cores using a drill press.
4- Top and bottom caps was attached to the cement cores using epoxy.
3.3.2. Test Procedures. After preparing the cement cores, the hydraulic
fracturing test procedures can be summarized as follows:
1- The core was placed in the fracturing cell
2- The pressure lines were connected to both the cell and the core sample
3- 400 psi overburden and 100 psi confining pressure were applied
4- Filled up accumulator with drilling fluid (without LCM)
5- Wellbore was filled up with drilling fluid containing LCM
6- Injection was started to build pressure inside the wellbore until reaching the
breakdown pressure where an increase in the confining pressure will be observed
due to fluid pushing against the rubber sleeve.
7- The injection was stopped to allow for 10 minutes before running the re-opening
cycle.
8- The test was stopped after the second cycle when an increase in the confining
42
pressure is observed, indicating that fluid already propagated through the fracture.
Detailed steps for the core preparation and test procedures can be found in Appendix B.
3.4. DATA ANALYSIS
3.4.1. Statistical Methods. A statistical analysis was conducted using JMPTM
statistical analysis software to understand the relationship between the different
investigated parameters such as fracture width, LCM type/blend, base fluid, and PSD on
the performance of LCM in terms of the sealing pressure.
First, a regression analysis was conducted by performing a multiple linear
regression analysis to model a relationship between 9 explanatory variables and the
sealing pressure response. The 9 variables used are fracture width, LCM type/blend, base
fluid, fluid density, and the five D-values obtained from PSD analysis; D10, D25, D50,
D75, and D90.
The probability test (F-test) was performed to test the influence of each parameter
on sealing pressure. The F-test provides a P-value, where the P-value is basically a
statistical probability that the predicted value (in this case the F-value) is similar or very
different from the measured value, assuming a true null hypothesis (H0) that proposes no
influence of a specific variable on the sealing pressure (Montgomery, 2001). With a
confidence interval of 95% and type I error (α) of 0.05, P-values less than 0.05 suggests
rejecting the null hypothesis (H0) and accepting an alternative hypothesis (H1). The
alternative hypothesis suggests that the sealing pressure is influenced by a specific
variable. The F-test is calculated based on the variance of the data as:
43
𝐹 =𝑆1
2
𝑆22 (10)
Where 𝑆12 is the variance of the first sample and 𝑆2
2 is the variance of the second
sample. The variance can be defined as the average squared difference from the mean.
Leverage plots for general linear hypothesis, introduced by Sall (1990), were
plotted for each of the 9 variables (predictors) to show their contribution to the predicted
sealing pressure. The Effect Leverage plot is used to characterize the hypothesis by
plotting points where the distance between each point to the fit line shows the
unconstrained residual while the distance to the x-axis shows the constrained residual by
the hypothesis. The constrained sealing pressure for each parameter under the hypothesis
can be written as:
𝑏0 = 𝑏 − (𝑋′𝑋)−1𝐿′𝜆 (11)
Where b is the least square, (𝑋′𝑋) is the inverse matrix (the transpose of the
matrix data being D-values and other parameter, used to enforce orthogonality), and λ is
the Lagrangian multiplier for the hypothesis constraint (L). The residual constrained by
the hypothesis (r0) is:
𝑟0 = 𝑟 + 𝑋(𝑋′𝑋)−1𝐿′𝜆 (12)
Where the Lagrangian multiplier is defined as:
𝜆 = (𝐿(𝑋′𝑋)−1𝐿′)−1𝐿𝑏 (13)
44
The residuals unconstrained by the hypothesis (r) are the least squares residuals
defined as:
𝑟 = �́� − 𝑋𝑏 (14)
The Leverage plot is constructed by plotting vx on the x-axis (Eq. 15) versus vy on
the y-axis (Eq. 16). vx values represents the difference in the residuals caused by the
hypothesis, which is the distance from the model fit line to the x-axis while vy values are
vx plus the unconstrained residuals.
𝑣𝑥 = 𝑋(𝑋′𝑋)−1𝐿′𝜆 (15)
𝑣𝑦 = 𝑟 + 𝑣𝑥 (16)
The sealing pressure residuals are regressed on all predictors except for the
variable of interest while the x-residuals (variable of interest) are regressed on all other
predictors in the model. The mean of the sealing pressure, without the effect of variable
of interest, is plotted as well as a least square fit line and confidence interval for easier
interpretation of the results. The upper and lower confidence interval could be plotted
using Eq. 17 and Eq. 18, respectively. The least squares fit line slope is a measure of how
the tested variable affects the sealing pressure i.e. a non-zero slope implies that the tested
variable will affect the sealing pressure (Sall, 1990).
𝑈𝑝𝑝𝑒𝑟 (𝑥) = 𝑥𝑏 + 𝑡𝛼/2𝑠√𝑥(𝑋′𝑋)−1𝑥′ (17)
45
𝐿𝑜𝑤𝑒𝑟 (𝑥) = 𝑥𝑏 − 𝑡𝛼/2𝑠√𝑥(𝑋′𝑋)−1𝑥′ (18)
Where x = 1 x is the 2-vector of regressors.
3.4.2. Analytical Fracture Models. The test parameters used for the hydraulic
experiments were applied as the input values for the five analytical fracture models
presented in chapter 2.3. The analytical models were used to predict the fracture width.
The estimated fracture widths were compared to the fracture width measured under
optical microscopy after the sample was retrieved from the fracturing apparatus.
46
4. RESULTS
4.1. LCM CHARACTERIZATION
4.1.1. Updated LCM Classification. A new classification of LCM was
performed including new LCM technologies, based on both the physical, chemical
properties, and their application. The physical properties include the shape and the size of
these particles while the chemical properties include material solubility in acids,
swellability, and reactivity with other chemicals to activate the blend. LCM’s are re-
classified into eight categories: granular (1), flaky (2), fibrous (3), LCM’s mixture (4),
acid soluble (5), high fluid loss LCM’s squeezes (6), swellable/hydratable LCM’s (7),
and nanoparticles (8).
1. Granular. Granular materials are defined as additives that are capable of forming a
seal at the formation face or within the fracture to prevent the losses into the formation
(Howard and Scott, 1951; Nayberg, 1987). They are available in a wide particle size
distribution. Due to their rigidity, this type of materials is used often for wellbore
strengthening applications. Granular materials have higher crushing resistance than other
LCM types and some of them could be classified as granular and at the same time acid
soluble such as calcium carbonate. Granular materials include graphite, nutshells, sized
calcium carbonate, glisonite, course bentonite, asphalt, and perlite.
2. Flaky. Flaky materials are defined as “A type of LCM that is thin and flat in shape,
with a large surface area” (SLB Oil Field Glossary). This type may or may not have any
degree of stiffness and they are capable of forming a mat over the permeable formation
face (Howard and Scott, 1951; Nayberg, 1987). Flaky materials include cellophane, mica,
cottonseed hulls, vermiculite, corn cobs, and flaked calcium carbonate.
47
3. Fibrous. Fibrous materials can be defined as “A type of LCM that is long, slender and
flexible and occurs in various sizes and lengths of fiber” (SLB Oil Field Glossary). This
type of materials may have low stiffness and will form a ‘’mat-like’’ bridge when used to
reduce the losses into fractures or vugular formations (Howard and Scott, 1951). The
ability to form a “mat-like” bridge serves as a filtration medium for smaller particles in
the drilling fluids to deposit and form a seal (Nayberg, 1987). Fibrous materials comes in
a wide range of grades/sizes and some types of fibrous material are acid soluble such as
Magma Fiber. Fibrous materials are often used in both WBM and OBM but some of
these materials have some limitations when used in OBM. Fibrous materials include
cellulose fibers, nylon fibers, mineral fibers, saw dust, and shredded paper.
4. Mixture of LCM’s. Several studies have shown that mixing two or more LCM’s
together will yield a better performance in mitigating losses due to the different
properties and particle sizes of the mixed blend compared to individual loss circulation
materials (Nayberg, 1987; Whitfill, 2008; Van Oort et al. 2011; Kumar and Savari, 2011).
Engineered LCM’s blends are available for different lost circulation scenarios. These
blends contain optimized types and particle size distributions that have been proved
experimentally to seal a wide range of fracture sizes.
5. Acid soluble/water soluble. Conventional LCM’s have the disadvantage of damaging
the formation when used in the reservoir section; as a result the development of non-
damaging LCM’s has increased (Sweatman and Scoggins, 1988). Acid/water soluble
LCM’s are considered as non-damaging LCM’s that could be used to cure losses in
reservoir sections. Acid soluble materials include calcium carbonate and mineral fibers.
Water soluble LCM’s include sized salts.
48
6. High fluid loss LCM’s squeeze. This type of LCM’s combination is often used to
cure severe losses when encountering fractured or highly permeable formations. The
filtration process will form a plug that seals the losses zone. These treatments often
require special procedures in order to squeeze them into the losses zone and it is usually
performed as a “hesitation squeeze” (Lost Circulation Guide, 2014).
7. Swellable/hydratable LCM’s combinations. Settable/Hydratable treatments are
basically a blend of LCM’s with a highly reactive material such as polymers. These
treatments will be activated either by chemical reagents or whenever they contact drilling
fluids or formation fluids; as a result, a plug will be formed within the losses zone. These
types of treatments often require special placement procedures (Lost Circulation Guide,
2014).
8. Nanoparticles. Current applied nanoparticles (NPs) include silica, iron hydroxide and
calcium carbonate (Friedheim et al. 2012; Hoelscher, et al. 2012; Nwaoji et al. 2013;
Contreras et al. 2014a). NPs can be prepared by either ex-situ or in-situ procedures. Ex-
situ stands for the preparation of NPs that occurs into an aqueous solution that is later
added to the mud. In-situ involves the addition of the precursors that forms the NPs
directly to the mud. Other types of NPs obtained from aluminum and titanium have been
investigated for permeability reduction in presence of WBM (Chenevert and Sharma,
2009). Carbon black NPs have been used to reduce mud cake thickness to mitigate
differential pipe sticking (Paiaman and Al-Anazi, 2008). Ballard and Massam (2008)
investigated the use of barium sulfate NPs as a weighting agent. Zinc NPs application for
lubricity improvement was reported by Griffo and Keshavan (2007) when used as a
drilling bit lubricator in the presence of other additives including silica gel.
49
A comprehensive summary that includes most of the available LCMs are
tabulated in cross-referencing table using the new classification to be used as a reference
table for operators and drilling engineers (Appendix C). The tables list the generic name,
trade name and the recommended application for each LCM. Majority of LCM comes in
different grades; fine, medium, and coarse to suit different losses scenarios.
4.1.2. Particle Size Distribution Analysis. Table 4.1 shows a summary of the
blends particle size distribution, which was analyzed using the dry sieve analysis
technique. Graphs of the cumulative % of finer particles passing versus the logarithmic
sieve size in mm for the tested blends can be found in Appendix D.
Table 4.1. Summary of the Particle Size Distribution Analysis
LCM Blend Total Concn.
(ppb)
Particle Size Distribution (microns)
D10 D25 D50 D75 D90
G # 1 15 60 85 320 800 1300
G # 1 50 60 95 340 800 1300
NS # 1 15 180 400 1000 1600 2000
NS # 1 50 180 400 1000 1600 2400
SCC # 3 50 250 360 680 950 1200
CF # 1 15 90 140 220 700 1400
CF # 1 50 90 150 220 800 1400
G & SCC # 1 30 80 100 460 900 1300
G & SCC # 1 80 80 120 480 900 1300
G & SCC # 3 105 65 90 420 1100 1400
G & SCC # 4 105 60 150 500 700 900
G & SCC # 5 105 90 400 700 1200 1400
G & SCC # 6 105 100 500 900 1250 1400
G & SCC # 7 105 170 650 1300 1900 2600
G & SCC # 8 105 100 250 1000 1800 2400
G & SCC # 9 105 300 800 1400 1800 2200
50
Table 4.1. Summary of the Particle Size Distribution Analysis (Cont’d)
G & NS # 1 20 65 180 500 1300 1900
G & NS # 1 40 80 180 580 1400 2000
G,SCC, & CF # 1 55 55 100 450 850 1200
NOTE: Graphite (G) Nutshells (NS) Sized Calcium Carbonate (SCC) Cellulosic Fiber (CF)
4.1.3. Optical Microscopy.
4.1.3.1 LCM particles. Figure 4.1 - Figure 4.4 show the optical microscopy
images for four of the LCMs used in this study. The images were analyzed using the
sphericity and roundness chart by Powers (1953). The graphite consist of (Figure 4.1) of
mostly sub-rounded particles with high sphericity. Calcium carbonate particles can be
classified as sub-angular with high sphericity (Figure 4.2). Nutshell particles (Figure 4.3)
could be described as angular with low sphericity. The cellulosic fiber (Figure 4.4) can be
described as nonspherical particles with some elongated and flat particles.
Figure 4.1. Graphite Particles at 60x Magnification (left) and 200x Magnification (right)
1 mm 1 mm
51
Figure 4.2. Calcium Carbonate Particles at 60x Magnification (left) and 200x
Magnification (right)
Figure 4.3. Nutshells Particles at 60x Magnification (left) and 200x Magnification (right)
Figure 4.4. Cellulosic Fiber Particles at 60x Magnification (left) and 200x Magnification
(right)
1 mm 1 mm
1 mm 1 mm
1 mm 1 mm
52
4.1.3.2 Formed seals. Figure 4.5 - Figure 4.8 show the formed seal of different
blends. The formed seals were retrieved from the tapered discs carefully after the high
pressure testing and left to dry prior to examination under microscopy.
Figure 4.5 shows the formed seal using a 30 ppb graphite and sized calcium
carbonate blend retrieved from a 1000 microns fracture width test. The image shows that
the sized calcium carbonate particles are relatively large compared to the graphite
particles. Despite the low average fluid loss per cycle (4.1 ml/cycle) caused by the
presence of very fine particles, which contributed in sealing smaller voids between the
larger particles, the seal broke at relatively low sealing pressure of 487 psi. High
sphericity and roundness of both graphite and calcium carbonate particles (Figure 4.1 and
Figure 4.2) may have resulted in lower friction between the particles and the tapered slot.
Thus, less pressure was required to force these particles through the fracture tip, which
can explain the low sealing pressure of this blend.
Figure 4.5. Optical Microscopy Images of Graphite and Sized Calcium Carbonate (30
ppb) Seal Retrieved from 1000 microns Fracture Width
40X 20X
Calcium carbonate
Fine graphite
Calcium carbonate
Fine graphite
53
Figure 4.6 shows the formed seal of 20 ppb graphite and nutshells blend at two
different magnifications, which was retrieved from a 1500 microns tapered slot test. The
image at 20x magnification shows more presence of the coarse nutshells particles while
graphite particles were very small compared to nutshells. It can be also seen that the
irregularity of the nutshells resulted in a better arrangement of the particles inside the
tapered disc. This seal was able to sustain higher sealing pressure of 805 psi at wider
fracture compared to the graphite and calcium carbonate seal (Figure 4.5), which was
tested with a 1000-micron fracture width. The higher sealing pressure can be explained
by the irregularity of nutshell particles, which resulted in higher friction between the
particles that required higher pressure to break the seal.
Figure 4.6. Optical Microscopy Images of Graphite and Nutshells (20 ppb) Seal
Retrieved from 1500 microns Fracture Width
Figure 4.7 shows the retrieved formed seal using 40 ppb graphite and nutshells
when tested at 2000 microns fracture width. The image shows the presence of fine
graphite particles among the formed seal with a very wide distribution (poor sorting) of
60X 20X Coarse nutshells
Irregular shape of nutshells
Fine graphite
54
nutshells particles. The formed seal was only able to hold a maximum pressure of 295 psi
before fluid flow through the tapered disc, which resulted in higher average fluid loss per
cycle (32 ml/cycle). The lack of large particles, seen in both the optical microscope
picture (Figure 4.7) and the PSD analysis (Table 4.1), may have contributed to the low
sealing pressure of this blend.
Figure 4.7. Optical Microscopy Images of Graphite and Nutshells (40 ppb) Seal
Retrieved from 2000 microns Fracture Width
Figure 4.8 shows the retrieved seal using 55 ppb graphite, sized calcium
carbonate, and cellulosic fiber when tested with a 1500 microns fracture width. The
image shows clearly the presence of fine graphite particles, relatively larger calcium
carbonate particles, and less presence of poorly sorted cellulosic fibers. The seal was
able to withstand a maximum pressure of 391 psi with an average fluid loss per cycle of
14.1 ml/cycle. The low sealing pressure is attributed to the lack of particles that are large
enough to bridge inside the tapered disc, where the D90 obtained for this blend from the
60X 20X
Fine graphite Wide PSD of nutshells Wide PSD of nutshells
Fine graphite
55
dry sieve analysis (1200 microns) relatively small compared to the fracture tip (1500
microns).
Figure 4.8. Optical Microscopy Images of Graphite, Sized Calcium Carbonate, and
Cellulosic Fiber (55 ppb) Seal Retrieved from 1500 microns Fracture Width
In general, it was observed that the finer particles helped in sealing the voids
between other coarser particles. The low particles sphericity and angularity of nutshells
improved the formation of a tighter seal higher sealing pressures for nutshell blends.
4.1.4. Scanning Electron Microscopy. Figure 4.9 - Figure 4.12 show SEM
images of the formed seals. When analyzing SEM images, the light is dense while dark is
less dense so the dark spots represent void spaces between the particles.
Figure 4.9 shows the formed seal of a 50 ppb nutshells retrieved from 2000
microns fracture width test. This seal was sustained a maximum sealing pressure of 755
psi with an average fluid loss per cycle of 18.3 ml/cycle. The SEM image shows that the
low sphericity and the irregular shapes nutshells particles (30x magnifications) with the
rough surface (100x magnifications) resulted in a good packing inside the tapered disc
40X 20X
Fine graphite Cellulosic fiber
Calcium carbonate
Cellulosic fiber
Fine graphite
Calcium carbonate
56
with higher internal friction and thus higher sealing pressure. This blend has a wide PSD
(poor sorting) of nutshell blends, with D10 of 180 microns and D90 of 2400 microns
(Table 4.1). The wide PSD resulted in sealing the 2000 microns fracture with low fluid
loss due to the presence of fine nutshell particle that filled void spaces between larger
particles.
Figure 4.9. Nutshells # 1 (50 ppb) SEM Images of Formed Seal Retrieved from 2000
microns Fracture Width
Figure 4.10 shows formed seal using an 80 ppb graphite and calcium carbonate,
which was retrieved from 1500 microns fracture width. This seal sustained lower sealing
pressure (224 psi) at a smaller fracture width compared to the nutshells seal (Figure 4.9).
The low sealing pressure is attributed to the high sphericity of both graphite and calcium
carbonate particles (Figure 4.1 and Figure 4.2), which resulted in more void spaces within
the formed seal.
100X 30X
Fine particles
Surface Roughness of Nutshells
57
Figure 4.10. Graphite and Sized Calcium Carbonate #1 (80 ppb) SEM Images of Formed
Seal Retrieved from 1500 microns Fracture Width
Figure 4.11 shows the 40 ppb graphite and nutshells seal retrieved from a 2000
microns fracture width test. This seal was able to hold a maximum sealing pressure of
242 psi, which is relatively low. The image (30x magnification) shows clearly the
irregular shape of nutshell particles as well as the presence of fine particles between the
larger particles. The image at higher magnification (100 x) illustrates the difference in the
surface roughness between graphite (lighter) and nutshells (darker). It can be seen that the
nutshell particles has relatively rougher surfaces compared to graphite particles.
Figure 4.11. Graphite and Nutshells # 1 (40 ppb) SEM Images of Formed Seal Retrieved
from 2000 microns Fracture Width
300X 180X
Void Spaces Void Spaces
100X 30X
Nutshells Graphite
Irregular shapes Void space
Void space
58
Figure 4.12 shows the formed seal using graphite, calcium carbonate, and
cellulosic fiber blend retrieved from 1000 microns fracture width. The formed seal was
capable of holding a maximum pressure of 1011 psi. The SEM image at 35x
magnification shows fine particles filling the gaps between larger particles. At 450x
magnification, the surface roughness of cellulosic fiber could be clearly seen, which
enhanced the sealing pressure due to the higher friction among the particles.
Figure 4.12. Graphite, Sized Calcium Carbonate, and Cellulosic Fiber # 1 (55 ppb) SEM
Images of Formed Seal Retrieved from 1000 microns Fracture Width
4.1.5. Swelling. Swelling test was conducted for nutshells to correlate their
sealing pressure enhancement when tested at higher temperature. The increase in the
sealing pressure for nutshells blend was experimentally confirmed to be due to the
swelling of nutshell particles at higher temperature. Figure 4.13 shows the swelling of
nutshells for 24 hours at room temperature (dotted line) and at 180 ˚ F (dashed line). The
solid line represents the temperature profile.
35X 450X
Fine particles Cellulosic fiber
59
Figure 4.13. Nutshells Swelling (24 hr Water Soak) @ Room Temperature and 180 ˚F
The results show a 3% swelling (water soak) over 24 hours at room temperature.
At higher temperature (180˚F), nutshells exhibited approximately 13% swelling as shown
in Figure 4.13. Based on these tests, temperature is another factor that could improve, or
reduce, the seal integrity for some LCM treatments. Therefore, it is recommended to
evaluate the seal integrity of LCM treatments at higher temperatures.
4.2. LOW PRESSURE LCM TESTING
A total of 160 tests were conducted using the low pressure apparatus (LPA) for
LCM blends at different concentrations. A cut off value of 100 ml was used in all the
testing, i.e. if fluid loss value exceeds 100 ml, the fluid loss is considered as non-
controlled (NC). All screened LCM treatments failed to establish a seal in fractures wider
than 2000 microns. Fluid loss values for single LCM treatments are tabulated in Table
4.2. For LCM mixtures, the results are tabulated in Table 4.3.
0
20
40
60
80
100
120
140
160
180
200
0
2
4
6
8
10
12
14
0 2 4 6 8 10 12 14 16 18 20 22 24
Tem
p (
F)
Sw
ellin
g %
Time (hr)
Swelling % @ Room Temp
Swelling % @ 180 Deg F
Temp F
60
From the screening tests results, blends were selected for further evaluation using
the HPA based on the fluid loss. Table 4.4 shows the selected blends and the fracture
width to be tested using the HPA. Nutshells blends were the only LCM capable of sealing
the 2000 microns fracture width (TS3 and TS4) when used individually and therefore, all
the other individual LCM treatments were excluded from being tested using the disks
TS3 and TS4. Because none of the LCM’s mixtures were able to seal the disks TS3 and
TS4 except the one with nutshells at 40 ppb, all other mixtures were also excluded from
being evaluated using the disks TS3 and TS4.
Table 4.2. Low Pressure Apparatus (LPA) Fluid Loss Results in (ml) for Single LCM
Treatments
LCM
Type
Blend # 1 2 3 4
Concn.
(ppb) 15 50 15 50 15 50 15 50
Gra
phit
e (G
) 1000 (TS1) 9 7 11 6 19 5 35 NC
1500 (TS2) 41 35 69 19 38 6 52 NC
2000 (TS3) NC NC NC NC NC NC NC NC
2000 (TS4) NC NC NC NC NC NC NC NC
Nu
tsh
ells
(N
S)
1000 (TS1) 4 2 15 2 12 3 19 33
1500 (TS2) 12 7 21 4 21 17 58 NC
2000 (TS3) NC 31 45 15 NC NC NC NC
2000 (TS4) NC 30 38 15 NC NC NC NC
Siz
ed C
aCO
3
(SC
C)
1000 (TS1) NC 64 NC 10 48 13 29 19
1500 (TS2) NC NC NC NC NC 13 NC 26
2000 (TS3) NC NC NC NC NC NC NC NC
2000 (TS4) NC NC NC NC NC NC NC NC
Cel
lulo
sic
Fib
er
(CF
)
1000 (TS1) 12 0 67 23 34 58 -- --
1500 (TS2) 39 4 NC NC 32 37 -- --
2000 (TS3) NC NC NC NC NC NC -- --
2000 (TS4) NC NC NC NC NC NC -- --
61
Table 4.3. Low Pressure Apparatus (LPA) Fluid Loss Results in (ml) for LCM Mixtures
LCM mixtures G & SCC
(Aston et al. 2004)
G & NS
(Hettema et al. 2007)
G, SCC, & CF
(Kumar et al.
2011)
Blend # 1 2 1 2 1 2
Concn. (ppb) 30 80 30 80 20 40 20 40 55 55
Fracture
Width
(micron)
1000 (TS1) 8 2 5 4 4 5 4 4 7 2
1500 (TS2) NC 22 68 8 8 7 11 7 13 19
2000 (TS3) NC NC NC NC 87 56 NC 54 NC NC
2000 (TS4) NC NC NC NC NC 63 NC 33 NC NC
Table 4.4. Selected Blends for High Pressure Apparatus (HPA) Testing
LCM
Type/Mixture
Blend
#
Concn.
(ppb)
Fracture Width Used (microns)
1000
(TS1)
1500
(TS2)
2000
(TS3)
2000
(TS4)
G 1 15 x x x
G 1 50 x x x
NS 1 15
NS 1 50
SCC 3 50 x x
CF 1 15 x x x
CF 1 50 x x
G & SCC 1 30 x x x
G & SCC 1 80 x x
G & NS 1 20 x x
G & NS 1 40
G, SCC & CF 1 55 x x
4.3. HIGH PRESSURE CONVENTIONAL LCM TESTING
A total of 75 tests were conducted using conventional LCM at different testing
conditions where 45 tests were done with WBM systems and 30 tests with OBM systems.
62
The results are summarized in Table 4.5 - Table 4.7. The tables list the following
for each test: base fluid, fluid density in lb/gal, type of weighting material, LCM blend
type and number, total LCM concentration in ppb, tapered disc used, the maximum
measured sealing pressure in psi and the average fluid loss per cycle. The fluid loss per
cycle was calculated as the total fluid loss collected divided by the number of cycles. A
cycle is here defined as any drop of the pressure that is equal to or greater than 100 psi.
Some tests noted with (R) were repeated to evaluate results repeatability. Results noted
with (HT) are tests that were repeated at higher temperature.
Table 4.5. Summary of HPA Results for Single LCM Treatments using WBM
Test # Base Fluid Density
(lb/gal)
Weighing
Material LCM Blend
Total
Concn.
(ppb)
Fracture
Width
(microns)
Sealing
Pressure
(psi)
Avg. FL
(ml/cycle)
1 WBM 8.6 None G # 1 15 1000 (TS1) 414 8.3
2 WBM 8.6 None G # 1 50 1000 (TS1) 449 7.6
3 WBM 8.6 None NS # 1 15 1000 (TS1) 984 3.5
4 WBM 8.6 None NS # 1 50 1000 (TS1) 2202 2
5 WBM (R) 8.6 None NS # 1 50 1000 (TS1) 2010 27
6 WBM 8.6 None NS # 1 15 1500 (TS2) 1754 11.8
7 WBM 8.6 None NS # 1 50 1500 (TS2) 2027 19.2
8 WBM (R) 8.6 None NS # 1 50 1500 (TS2) 2266 27.6
9 WBM 8.6 None NS # 1 15 2000 (TS3) 347 31
10 WBM 8.6 None NS # 1 15 2000 (TS4) 441 24.5
11 WBM 8.6 None NS # 1 50 2000 (TS4) 755 18.3
12 WBM (HT) 8.6 None NS # 1 50 2000 (TS4) 1164 17.3
13 WBM 8.6 None NS # 1 50 2000 (TS3) 2237 95
14 WBM (R) 8.6 None NS # 1 50 2000 (TS3) 2242 92
15 WBM 8.6 None SCC # 3 50 1000 (TS1) 589 13.3
16 WBM 8.6 None SCC # 3 50 1500 (TS2) 162 31.3
17 WBM 8.6 None CF # 1 15 1000 (TS1) 800 11.8
18 WBM 8.6 None CF # 1 50 1000 (TS1) 2160 0.5
19 WBM 8.6 None CF # 1 50 1500 (TS2) 1129 4.4
63
Table 4.6. Summary of HPA Results for LCM Mixture Treatments using WBM
Test # Base Fluid Density
(lb/gal)
Weighing
Material LCM Blend
Total
Concn.
(ppb)
Fracture
Width
(microns)
Sealing
Pressure
(psi)
Avg. FL
(ml/cycle)
20 WBM 8.6 None G & SCC # 1 30 1000 (TS1) 487 4.1
21 WBM 9.5 Barite G & SCC # 1 30 1000 (TS1) 1205 3.8
22 WBM 11 Barite G & SCC # 1 30 1000 (TS1) 901 7.6
23 WBM 12.5 Barite G & SCC # 1 30 1000 (TS1) 912 8.2
24 WBM 14.5 Barite G & SCC # 1 30 1000 (TS1) 1037 4..4
25 WBM 16.5 Barite G & SCC # 1 30 1000 (TS1) 1344 8.3
26 WBM 9.5 Hematite G & SCC # 1 30 1000 (TS1) 861 4.6
27 WBM 12.5 Hematite G & SCC # 1 30 1000 (TS1) 1507 5
28 WBM 14.5 Hematite G & SCC # 1 30 1000 (TS1) 1334 7.8
29 WBM 16.5 Hematite G & SCC # 1 30 1000 (TS1) 1842 7.9
30 WBM 8.6 None G & SCC # 1 80 1000 (TS1) 584 2.5
31 WBM (HT) 8.6 None G & SCC # 1 80 1000 (TS1) 537 2.9
32 WBM 11 Barite G & SCC # 3 105 1000 (TS1) 2050 1.9
33 WBM 11 Barite G & SCC # 5 105 1000 (TS1) 2859 5.5
34 WBM 8.6 None G & SCC # 1 80 1500 (TS2) 224 8.4
35 WBM 11 Barite G & SCC # 7 105 1500 (TS2) 2571 12.8
36 WBM 8.6 None G & NS # 1 20 1000 (TS1) 2372 3.1
37 WBM 8.6 None G & NS # 1 40 1000 (TS1) 2892 2.4
38 WBM 8.6 None G & NS # 1 20 1500 (TS2) 805 10.1
39 WBM (HT) 8.6 None G & NS # 1 20 1500 (TS2) 718 7
40 WBM 8.6 None G & NS # 1 40 1500 (TS2) 1713 3
41 WBM (HT) 8.6 None G & NS # 1 40 1500 (TS2) 1190 3
42 WBM 8.6 None G & NS # 1 40 2000 (TS3) 295 32
43 WBM 8.6 None G & NS # 1 40 2000 (TS4) 242 14.2
44 WBM 8.6 None G,SCC, & CF # 1 55 1000 (TS1) 1011 4.7
45 WBM 8.6 None G,SCC, & CF # 1 55 1500 (TS2) 391 14.2
Table 4.7. Summary of HPA Results for LCM Mixture Treatments using OBM
Test # Base Fluid Density
(lb/gal)
Weighing
Material LCM Blend
Total
Concn.
(ppb)
Fracture
Width
(microns)
Sealing
Pressure
(psi)
Avg. FL
(ml/cycle)
46 OBM 8.6 Barite G & NS # 1 20 1000 2398 6.8
47 OBM 8.6 Barite G,SCC, & CF # 1 55 1000 1505 20
48 OBM 8.6 Barite G & SCC # 1 30 1000 737 18.8
49 OBM 9.5 Barite G & SCC # 1 30 1000 1049 6.9
50 OBM 11 Barite G & SCC # 1 30 1000 1050 4
51 OBM 12.5 Barite G & SCC # 1 30 1000 1191 3.3
64
Table 4.7. Summary of HPA Results for LCM Mixture Treatments using OBM (Cont’d)
52 OBM 14.5 Barite G & SCC # 1 30 1000 1214 8.75
53 OBM 16.5 Barite G & SCC # 1 30 1000 1238 5.7
54 OBM 12.5 Hematite G & SCC # 1 30 1000 1269 3.5
55 OBM 14.5 Hematite G & SCC # 1 30 1000 1305 4
56 OBM 16.5 Hematite G & SCC # 1 30 1000 1283 6.7
57 OBM 12.5 Sieved
Barite (C) G & SCC # 1 30 1000 1073 4.4
58 OBM 12.5 Sieved
Barite (M) G & SCC # 1 30 1000 1134 6.4
59 OBM 12.5 Sieved
Barite (F) G & SCC # 1 30 1000 1249 4.4
60 OBM 8.6 Barite G & SCC # 1 80 1500 776 120
61 OBM 11 Barite G & SCC # 3 105 1000 1569 4
62 OBM 11 Barite G & SCC # 3 105 1500 205 47
63 OBM 11 Barite G & SCC # 4 105 1000 117 NC
64 OBM 11 Barite G & SCC # 4 105 1500 0 NC
65 OBM 11 Barite G & SCC # 5 105 1000 1489 4.4
66 OBM 11 Barite G & SCC # 6 105 1500 164 NC
67 OBM 11 Barite G & SCC # 7 105 1500 1708 13.5
68 OBM 11 Barite G & SCC # 7 105 2000 1669 15
69 OBM 14.5 Barite G & SCC # 7 105 2000 2471 6.2
70 OBM 16.5 Barite G & SCC # 7 105 2000 2751 4.4
71 OBM 12.5 Barite G & SCC # 7 105 2000 3097 11.4
72 OBM 12.5 Barite G & SCC # 7 105 2000 3132 11
73 OBM 11 Barite G & SCC # 7 105 2500 1534 17.5
74 OBM 11 Barite G & SCC # 8 105 3000 110 NC
75 OBM 11 Barite G & SCC # 9 105 3000 0 NC
4.3.1. Effect of LCM Type and Fracture Width. To investigate the effect each
individual parameter has on the sealing pressure, the results are plotted and described
with respect to one variable at a time. Four colors are used to define the tapered discs:
blue for 1000 microns (TS1), red for 1500 microns (TS2), orange for 2000 microns (TS3)
with a fracture mouth of 4000 microns, and purple for 2000 microns (TS4) with a fracture
mouth of 5000 microns for Figure 4.14 - Figure 4.18. From the plot of the three high
concentrations (50 ppb) single LCM treatments (Figure 4.14a), nutshells treatments
65
resulted in higher sealing pressures at the TS1 and TS2 fracture widths (up to 2200 psi)
compared to graphite (449 psi) and sized calcium carbonate (589 psi).
Figure 4.14. Effect of LCM Type on the Sealing Pressure for (a) Single LCM Treatments
and (b) LCM Mixtures
For LCM mixtures (Figure 4.14b), treatments containing nutshells and graphite
(40 ppb) exhibited higher sealing pressures up to 2800 psi compared to the other blend
containing graphite and sized calcium carbonate at a higher concentration (80 ppb) using
the two different fracture widths (589 psi and 224 psi for TS1 and TS2, respectively). In
general, increasing the fracture width for both single and LCM mixtures resulted in lower
sealing pressures.
4.3.2. Effect of Concentration. Figure 4.15 gives nutshells LCM concentration
versus sealing pressure for nutshells and the graphite nutshells blend for different fracture
widths. Increasing the concentration of nutshells from 15 ppb to 50 ppb increased the
sealing pressure up to 500% (Figure 4.15a). For graphite and nutshells blends (Figure
(a) (b)
66
4.15b), the sealing pressures were increased when tested using the TS1 (1000 microns)
and the TS2 (1500 microns) by 22% and 113%, respectively. Figure 4.15 shows that
increasing the LCM concentration is increasing the sealing pressure for both single LCMs
and mixtures.
Figure 4.15. Effect of LCM Concentration on the Sealing Pressure for (a) Single LCM
Treatments and (b) LCM Mixtures
4.3.3. Effect of Temperature. Figure 4.16 shows a comparison between four
tests, which were conducted at both room temperature and at an elevated temperature
(HT) of 180 ˚F.
The sealing pressure of an 80 ppb graphite and sized calcium carbonate # 1 was
not affected significantly by the elevated temperature, where a negligible reduction from
584 psi to 537 psi (8%) was observed. For both tests using a 20 and a 40 ppb graphite and
nutshells (G&NS # 1 blend), a reduction in the seal integrity was observed at 180 ˚F by
(b) (a)
67
11% and 30.5%, respectively. However, when a mud with 50 ppb nutshells added was
tested at 180 ˚F, an increase in the seal integrity from 755 psi to 1164 psi (54%) was
observed (Figure 4.16).
Figure 4.16. Effect of Temperature on the Sealing Pressure
The variation in the sealing pressure of the different blends when tested at higher
temperature suggests that temperature has a significant effect on LCM performance. The
reduction in the sealing pressure for graphite and nutshell blends could be attributed to
the increased graphite lubricity at high temperature, since graphite is known for acting as
a solid lubricant (Alleman et al. 1998; Savari et al. 2012).
The increase in the sealing pressure for nutshells blend is linked to the swelling of
nutshells at higher temperature, which was experimentally confirmed above (Figure
4.13).
0
500
1000
1500
2000
Sealin
g P
ressu
re (
psi)
LCM Type
68
4.3.4. Effect of Particle Size Distribution. In order to understand the effect of
PSD on the sealing pressure, the first step was to examine test results for individual
fracture widths separately. The sealing pressures for each fracture width are plotted on
the left y-axis versus the particle size distribution obtained from the dry sieve analysis on
the right y-axis as shown in Figure 4.17 and Figure 4.18.
Figure 4.17 shows the sealing pressure versus the particle size distribution for
single LCMs and LCM mixtures using the 1000 microns fracture width disc. For blends
containing graphite, calcium carbonate or a mixture of the two, the sealing pressures were
around 500 psi at both low and high concentration. For blends with nutshells, the
measured sealing pressures were 984 psi and 2200 psi for the 15 ppb and the 50 ppb
blends, respectively. When graphite was combined with nutshells (Figure 4.17b), the
sealing pressures were increased to 2800 psi.
Figure 4.17. Effect of PSD on the Sealing Pressure using 1000 microns Fracture Width
(TS1) for (a) Single LCM Treatments and (b) LCM Mixtures
(a) (b)
69
The particle size distribution of the different blends (Table 4.1) shows that
nutshells particles are coarser than both graphite and sized calcium carbonate particles.
The PSD of the single LCM nutshell treatments ranged between 180 – 2400 microns
while the PSD of graphite and sized calcium carbonate ranged between 60 – 1300
microns and 250 – 1200 microns, respectively.
Figure 4.18 shows the sealing pressures versus the particle size distribution for
single LCM treatments and LCM mixtures when tested at 1500 microns fracture width.
Both calcium carbonate blend and the mixture of graphite and sized calcium carbonate
resulted in sealing pressures lower than 250 psi. For blends containing nutshells only, the
sealing pressures were 1754 psi and 2027 psi for the 15 ppb and the 50 ppb blends,
respectively. The 20 ppb and the 40 ppb graphite and nutshell mixtures gave sealing
pressures of 805 psi and 1713 psi, respectively.
Figure 4.18. Effect of PSD on the Sealing Pressure using 1500 microns Fracture Width
(TS2) for (a) Single LCM Treatments and (b) LCM Mixtures
(a) (b)
70
The presented results in both Figure 4.17 and Figure 4.18 imply that both single
LCMs and LCM blends with coarser materials gave higher sealing pressure. In other
words, the wider the range of particle sizes increases the ability in sealing fractures and
improves the sealing pressure. The tests that contained nutshells exhibited higher sealing
pressures which indicate that the type of material is also influencing the results.
To qualitatively understand the overall trend of how PSD could affect the sealing
pressure with respect to fracture width, all sealing pressure results in chapter 4.3 were
plotted (black dots) regardless of base fluid, concentration, or LCM type and sorted with
the highest pressure to the right (Figure 4.19 - Figure 4.21). The black dashed line is
plotted as a reference line that represents the fracture width in microns. The quadratic
regression fit lines for the five D-values as well as the sealing pressure were plotted in
different colors, where D90 (red solid line), D75 (orange dashed line), D50 (blue dashed
line), D25 (green dashed line), and D10 (red dashed line).
Figure 4.19. PSD vs Sealing Pressure for 1000 microns Fracture Width
3000
2500
2000
1500
1000
500
0
3000
2500
2000
1500
1000
500
0
Se
ali
ng
Pre
ssu
re
(p
si)
Part
icle
Siz
e D
istr
ibu
tio
n(m
icro
ns)
Sealing Pressure (R 2̂ = 93.8%)
D10 (R 2̂ = 8.6%)
D25 (R 2̂ = 27.9%)
D50 (R 2̂ = 20%)
D75 (R 2̂ = 40%)
D90 (R 2̂ = 35%)
71
Figure 4.19 shows the sealing pressures (left y-axis) versus PSD (right y-axis) for
1000 microns fracture width. It is noticeable that higher sealing pressures are achieved
with wider PSD; however no clear correlation that suggests the effect of a specific D-
value could be observed.
Figure 4.20 and Figure 4.21 show the sealing pressure results versus PSD for
1500 microns and 2000 microns fracture widths, respectively. In general, the sealing
pressure results tend to increase as the particle size distribution gets coarser, i.e. wider
PSD. In other words, blends with higher D-values resulted in higher sealing pressures.
The D-values fit lines shows an upward increase with the pressure.
Even though the observed PSD fit lines shows an increase with the pressure, no
clear conclusion could be made on the significance of each D-value contribution toward
enhancing the sealing pressure from the results.
Figure 4.20. PSD vs Sealing Pressure for 1500 microns Fracture Width
3000
2500
2000
1500
1000
500
0
3000
2500
2000
1500
1000
500
0
Sea
lin
g P
res
su
re (
ps
i)
Pa
rtic
le S
ize
Dis
trib
uti
on
(mic
ron
s)
Sealing Pressure (R 2̂ = 98.6%)
D10 (R 2̂ = 39.6%)
D25 (R 2̂ = 44%)
D50 (R 2̂ = 56.1%)
D75 (R 2̂ = 64.3%)
D90 (R 2̂ = 78.1%)
72
Figure 4.21. PSD vs Sealing Pressure for 2000 microns Fracture Width
4.3.5. Effect of Base Fluid. In this subsection, different blends mixed with water
and oil base fluid were analyzed to investigate the effect of base fluids on LCM
performance. Figure 4.22 shows the effect of base fluid for different blends with a
variation in the total concentration between 20 – 105 ppb. All blends were tested at 1000
microns fracture width (TS1) using both; OBM and WBM except for graphite and sized
calcium carbonate # 7, which was tested at 1500 (TS2).
The sealing pressure of both blends; graphite, sized calcium carbonate, and
cellulosic fiber # 1 and graphite and sized calcium carbonate # 1 were increased by
approximately 50% when used in OBM. However, graphite and nutshells # 1 blend was
not affected by the change of base fluid. For the three LCM mixtures at relatively high
concentration (105 ppb), graphite and sized calcium carbonate blends # 3, 5, and 7,
higher sealing pressures were observed when used in WBM. An increase of 30%, 92%
and 50% was observed for graphite and sized calcium carbonate blends # 3, 5, and 7,
when using WBM.
3500
3000
2500
2000
1500
1000
500
0
3000
2500
2000
1500
1000
500
0
Se
ali
ng
Pre
ssu
re
(p
si)
Part
icle
Siz
e D
istr
ibu
tio
n(m
icro
ns)
Sealing Pressure (R 2̂ = 96.7%)
D10 (R 2̂ = 67.7%)
D25 (R 2̂ = 74.8%)
D50 (R 2̂ = 77%)
D75 (R 2̂ = 73%)
D90 (R 2̂ = 83.7%)
73
Figure 4.22. Effect of Base Fluid on Low and High Concentration LCM Mixtures
The difference in the effect of base fluid on low LCM mixtures versus the high
concentration LCM mixtures is attributed to the difference in rheological properties of
WBM and OBM. WBM tested has a yield point (YP) equal to 19 lb/100 ft2 (Table 3.1)
which makes is it more capable of suspending higher LCM concentrations. LCMs were
suspended in WBM and plugged the slot continuously. However, the OBM had a lower
YP of 14 lb/100 ft2 compared to WBM, which could have possibly resulted in LCM
sagging at the bottom of the cell.
4.3.6. Effect of Density. To understand the effect density, the 30 ppb graphite
and sized calcium carbonate blend # 1 was used in all tests for this step at different
densities. The density of drilling fluids was increased using barite for both WBM and
OBM. Since the pre-mixed OBM is already weighted to 11 lb/gal, the density was
reduced by adding the base fluid (EDC95-11).
Figure 4.23 shows the effect on the sealing pressure when varying the density for
the 30 ppb graphite and sized calcium carbonate blend # 1 when tested with the 1000
0
500
1000
1500
2000
2500
3000
3500
G & NS(20 ppb)
G,SCC,CF # 1(55 ppb)
G & SCC # 1(30 ppb)
G & SCC # 3(105 ppb)
G & SCC # 5(105 ppb)
G & SCC # 7(105 ppb)
Sealin
g P
ressu
re (
psi)
LCM Blend
WBM OBM
74
microns fracture disc using both WBM and OBM and for the 105 ppb graphite and sized
calcium carbonate blend # 7 when tested at 2000 microns fracture disc using OBM.
Figure 4.23. Effect of Increasing the Mud Density in OBM and WBM
For the 1000 microns fracture tests, the sealing pressure in OBM was increased by
approximately 68% from 737 psi to 1238 psi when the fluid density was increased from
8.6 lb/gal to 16.5 lb/gal. A higher increase of the sealing pressure by 175 % was observed
for the same blend mixed in WBM when the fluid was weighted from 8.6 lb/gal to 16.5
lb/gal.
For the 2000 microns using the graphite and sized calcium carbonate blend # 7 in
OBM, the sealing pressure was also increased by approximately 65% from 1669 psi for
the 11 lb/gal mud to 2751 psi when the fluid density was increased to 16.5 lb/gal.
Therefore, increasing the density of the drilling fluid caused the sealing pressure to
increase.
0
500
1000
1500
2000
2500
3000
3500
8 10 12 14 16 18
Sealin
g P
ressu
re (
ps
i)
Mud Density (lb/gal)
WBM (1000 microns)OBM (1000 microns)OBM (2000 microns)
75
4.3.7. Effect of Weighting Material. The 30 ppb graphite and sized calcium
carbonate blend #1 was tested using 1000 microns fracture width and with drilling fluids
weighing from 9.5 lb/gal to 16.5 lb/gal using either barite and hematite as weight
material.
Figure 4.24 shows the effect of the weighting material type on the sealing
pressure for OBM and WBM. The results show that increasing the density improved the
sealing pressure for both WBM and OBM. Using hematite as the weight material resulted
in higher sealing pressures compared to barite, especially for WBM.
Figure 4.24. Effect of Weighting Materials Type in OBM and WBM
4.3.8. Effect of LCM Shape. In this paragraph the variation in sealing pressure
results is evaluated based on the qualitative analysis of LCM particle shapes (Chapter
4.1.3.1). The angularity and low sphericity of nutshells resulted in higher sealing
pressures (up to 2800 psi) due to higher shear strength of the formed plug. Angularity and
0
200
400
600
800
1000
1200
1400
1600
1800
2000
8 10 12 14 16 18
Sealin
g P
ressu
re (
psi)
Mud Density (ppg)
Barite (OBM)
Hematite (OBM)
Barite (WBM)
Hematite (WBM)
76
sphericity was found to affect the angle of internal friction, thus affecting the degree of
mechanical interlocking among particles (Shinohara et al. 2000, Rong et al. 2013). The
higher angularity of nutshell particles seems to create higher mechanical interlocking
between the particles at the fracture tip, thus higher pressures were required to break the
formed seal. While the high sphericity of both graphite and calcium carbonate resulted in
lower sealing pressures at low concentrations due to the higher tendency of spherical
particles to flow through the fracture tip.
The variation in nutshell particle sizes enhanced the sealing capabilities compared
to graphite and calcium carbonate, which are moderately sorted. The asphericity of
cellulosic fiber also helped in a better sealing pressure compared to graphite and sized
calcium carbonate at the same testing conditions, i.e. fracture width and concentration
(See Test # 1, 2, 15, 17, 18, and 19 in Table 4.5.)
4.4. HIGH PRESSURE UNCONVENTIONAL LCM TESTING
A total of 10 tests were conducted using the unconventional LCM at different
testing condition where 8 tests with WBM and 2 tests with the low-toxicity mineral
OBM. The results are summarized in Table 4.8.
The tables list the following for each test: base fluid, fluid density in lb/gal, type
of weighting material, tapered disc used, injection rate used to initiate the seal, injection
rate used to evaluate the formed plug, the maximum measured sealing pressure in psi and
the total fluid loss collected. While conventional LCMs weren’t able to seal fractures
wider than 2500 microns, this unconventional system was able to seal fracture widths up
to 5000 microns.
77
Table 4.8. HPA Results for Foam Wedges Based System
Test
#
Base
Fluid
Density
(lb/gal)
Weighing
Material
Fracture
Width
(microns)
Injection
Rate
before
Seal
Initiation
(ml/min)
Injection
Rate
after
Seal
Initiation
(ml/min)
Sealing
Pressure
(psi)
Fluid
Loss
(ml)
1 Water Unweighted None 1500 (TS2) 5 10 2571 180
2 Water Unweighted None 2000 (TS4) 5 5 764 183
3 Water Unweighted None 2000 (TS4) 10 25 101 196
4 Water Unweighted None 2000 (TS3) 5 10 2657 184
5 Water Unweighted None 3000 (SS1) 25 25 340 190
6 Water Unweighted None 3000 (SS1) 5 10 2809 185
7 Water Unweighted None 5000 (SS2) 5 25 1077 187
8 Water Unweighted None 10000 (TS8) 5 10 101 375
9 EDC95-11 12.5 Barite 1500 (TS2) 5 10 78 325
10 EDC95-11 12.5 Barite 5000 (TS7) 5 10 69 350
The seal is formed by slow filtration of the carbonate particles through the wedge
foams trapped in the fracture. A slower process (i.e. hesitation squeeze) is required for
initiating a seal. A high flow rate can break the seal before it becomes strong enough to
hold an increasing differential pressure. Lower sealing pressures were observed for
injection rates higher than 5 ml/min (Table 4.8). Tests with injection rate of 5 ml/min
showed higher seal integrity and therefore, an injection rate of 5 ml/min was used for the
tests. The effect of the injection rate after initiating a good seal was not significant. There
is only a certain amount of fluid that can be squeezed out of the slurry for each test. The
lower fluid loss values at the lower injection rate indicates that the seal started to form
earlier within the fracture.
78
Figure 4.25 and Figure 4.26 show the formed seals when used in WBM and
OBM, respectively. The unconventional LCM performed good in WBM with sealing
pressures up to 2800 psi using 3000 microns straight slotted disc, which was damaged as
a result of the high pressure Figure 4.27. However, when EDC95-11 (Low-toxicity
mineral oil) was used as the base fluid, the system failed to seal both 1500 and 5000
microns fracture widths. For this reason, the system wasn’t evaluated at wider fracture
width with the OBM.
Figure 4.25. Wedge Foam in WBM Formed Plug using 5 mm disc; (left) Side View,
(right) Bottom View
Figure 4.26. Wedge Foam in OBM Formed Plug using 1500 microns disc; (left) Side
View, (right) Bottom View
79
Figure 4.27. Damaged Slotted Disc
4.5. HYDRAULIC FRACTURING TEST
Five concrete cores were fractured with different LCM blends and the pressure
versus time was recorded and breakdown and re-opening pressure interpreted. The first
test was conducted using EDC95-11, which is a solid free clear base fluid used to prepare
the pre-mixed OBM. The second test was conducted with the pre-mixed OBM without
LCM to serve as a control sample. Tests # 3, 4, and 5 were conducted using the pre-
mixed OBM in addition to 3 different LCM mixtures.
4.5.1. Test # 1 with EDC95-11 Base Fluid. Figure 4.28 shows the pressure
versus time plot for the first hydraulic fracturing test, which was conducted using a solid
free base fluid (EDC95-11) with no LCMs. The blue line represents the first injection
cycle, which is used to estimate the breakdown pressure. The red line represents the
second injection cycle, which is used to estimate the re-opening pressure after the 10
minutes fracture healing period. The breakdown pressure was found to be 843 psi while
the re-opening pressure was 571 psi. The fracture widths were measured under optical
microscope (Figure 4.29) and the width ranged between 15 - 22 microns.
80
Figure 4.28. Pressure vs. Time for Test # 1
Figure 4.29. The Resulting Fracture under Microscope (200x Magnification) for Test # 1
Showing Fracture Widths Ranging between 15 – 22 microns
4.5.2. Test # 2 with 11 lb/gal OBM without LCM. Test 2 was conducted with
the pre-mixed OBM and no LCMs were added. The breakdown pressure was 2008 psi
and the re-opening pressure was 592 psi (Figure 4.30). Two vertical fractures were
observed (Figure 4.31). The fractured core sample was split to take a cross-section view
of the core. The measured fracture widths ranged between 27 - 36 microns (Figure 4.32).
0
200
400
600
800
1000
1200
1400
1600
0 500 1000 1500 2000
Pre
ssu
re (
psi)
Time (s)
1st Cycle
2nd Cycle
Breakdown @ 843 psi
Re-Opening @ 571 psi
1 mm
Up Up
1 m
81
Figure 4.30. Pressure vs. Time for Test # 2
Figure 4.31. Fractured Core Showing Two Vertical Fractures and a Cross-sectional View
of the Fractured Core for Test # 2
0
500
1000
1500
2000
2500
0 500 1000 1500 2000
Pre
ssu
re (
psi)
Time (s)
1st Cycle
2nd CycleBreakdown @ 2008 psi
Re-Opening @ 592 psi
82
Figure 4.32. The Resulting Fracture under Microscope (200x Magnification) for Test # 2
Showing Fracture Widths Ranging between 27 – 36 microns
4.5.3. Test # 3 with 11 lb/gal OBM + 20 ppb Graphite and Nutshells. In this
test, 20 ppb graphite and nutshells # 1 was added to the pre-mixed OBM. The breakdown
and re-opening pressures were 2199 psi and 1334 psi, respectively (Figure 4.33).
Figure 4.33. Pressure vs. Time for Test # 3
0
500
1000
1500
2000
2500
0 500 1000 1500 2000
Pre
ssu
re (
psi)
Time (s)
1st Cycle
2nd CycleBreakdown @ 2199 psi
Re-Opening @ 1334 psi
1 mm
Up
83
Figure 4.34 shows the resulting two vertical fractures and a cross-section of the
fracture core. The measured fracture widths ranged between 40 - 85 microns (Figure
4.35).
Figure 4.34. Fractured Core Showing Two Vertical Fractures and a Cross-sectional View
of the Fractured Core for Test # 3
Figure 4.35. The Resulting Fracture under Microscope (200x Magnification) for Test # 3
Showing Fracture Widths Ranging between 40 – 85 microns
1 mm
Up
84
4.5.4. Test # 4 with 11 lb/gal OBM + 30 ppb Graphite and Sized Calcium
Carbonate. Figure 4.36 shows the pressure versus time for test # 4 where 30 ppb
graphite and calcium carbonate was added to the pre-mixed OBM. The breakdown
pressure was 2309 psi and the re-opening pressure was 1834 psi. Figure 4.37 shows the
resulting two vertical fractures and a cross-section of the fracture core in test # 4. The
measured fracture widths ranged between 40 - 85 microns (Figure 4.38).
Figure 4.36. Pressure vs. Time for Test # 4
Figure 4.37. Fractured Core Showing Two Vertical Fractures and a Cross-sectional View
of the Fractured Core for Test # 4
0
500
1000
1500
2000
2500
3000
-100 400 900 1400
Pre
ssu
re (
psi)
Time (s)
1st Cycle2nd Cycle
Breakdown @ 2309 psi
Re-Opening @ 1834 psi
85
Figure 4.38. The Resulting Fracture under Microscope (200x Magnification) for Test # 4
Showing Fracture Widths Ranging between 40 – 145 microns
4.5.5. Test # 5 with 11 lb/gal OBM + 55 ppb Graphite, Sized Calcium
Carbonate, and Cellulosic Fiber. Figure 4.39 shows the hydraulic fracturing results
using 55 ppb graphite, sized calcium carbonate, and cellulosic fiber # 1. The breakdown
and re-opening pressures were 2372 psi and 1717 psi, respectively.
Figure 4.39. Pressure vs. Time for Test # 5
0
500
1000
1500
2000
2500
3000
0 500 1000 1500
Pre
ssu
re (
psi)
Time (s)
1st Cycle
2nd CycleBreakdown @ 2372 psi
Re-Opening @ 1717 psi
Up Up
1 mm 1 mm
86
The fractured core is shown in Figure 4.40, where two vertical fractures were
observed. The measured fracture widths under the microscope (Figure 4.41) were ranging
between 50 – 140 microns.
Figure 4.40. Fractured Core Showing Two Vertical Fractures for Test # 5
Figure 4.41. The Resulting Fracture under Microscope (200x magnification) for Test # 5
Showing Fracture Widths Ranging between 50 – 140 microns
1 mm
Up Up
1 mm
87
4.5.6. Hydraulic Fracturing Results Comparison. The effect of adding LCMs
on the breakdown pressure and the re-opening pressure was calculated as percentage
increase based on the results obtained from the control sample of pre-mixed oil based
drilling fluid (Test # 2). The percent (%) increase in the breakdown and re-opening
pressures were calculated based on the control sample (Test # 2) breakdown pressure of
2008 psi and re-opening pressure of 592 psi as shown in Table 4.9.
When comparing the solid free oil base fluid (EDC95-11) from Test # 1 with the
pre-mixed OBM control sample (Test # 2), the breakdown pressure was reduced to 58%
and re-opening was reduced by 4%. The reduction in the breakdown pressure for the
solid free oil base fluid is caused by to the lack of formation of a filter cake around the
wellbore.
For Test # 3 with the 20 ppb graphite and nutshells blend enhanced the
breakdown pressure by approximately 10% while the re-opening pressure was enhanced
by 125% compared to the control sample (Test # 2). The enhancement in both the
breakdown and the re-opening pressure is attributed to the formation of a stronger filter
cake with the presence of LCMs. The increase suggests also that LCMs sealed the first
fracture, which occurred in the first cycle, and improved the wellbore integrity resulting
in higher re-opening pressure in the second injection cycle.
The 30 ppb graphite and sized calcium carbonate blend in Test # 4 resulted also in
enhancing both the breakdown and re-opening pressure by 15% and 210%, respectively.
For Test # 5 which contained 55 ppb graphite, sized calcium carbonate, and cellulosic
fiber resulted in the highest enhancement in the breakdown pressure by 18%. The re-
opening pressure increase also by 190% compared to the control sample (Test # 2).
88
Table 4.9. Summary of the Hydraulic Fracturing Experiments for the Concrete Cores
Test
# LCM Blend
Concn.
(ppb)
Breakdown
Pressure
(psi)
% Inc. in
Breakdown
Pressure
Re-opening
Pressure
(psi)
% Inc. in
Re-Opening
Pressure
1 None 0 843 -58 571 -4
2 None 0 2008 N/A 592 N/A
3 G&NS 20 2199 10 1334 125
4 G & SCC 30 2309 15 1834 210
5 G, SCC, & CF 55 2375 18 1717 190
4.6. DATA ANALYSIS
4.6.1. Statistical Analysis. Figure 4.42 - Figure 4.50 show the Effect Leverage
plots for each parameter with the resulting P value. The blue dashed line represents the
mean sealing pressure, the solid red line represents the fitted model, and the dashed red
line represents the confidence interval (5% confidence level). If the mean sealing
pressure is inside the confidence interval envelope the parameter does not have any
significant effect on sealing pressure. If the confidence interval crosses the mean pressure
at a high angle, it has a significant contribution to sealing pressure.
Figure 4.42 shows the effect of fracture width in the sealing pressure. The effect
of fracture width is very significant since the confidence interval curves crossed the mean
pressure with a high slope. The P-value of 0.0003 also indicates that the fracture width
influenced the predicted sealing pressure. The effect of fluid density (Figure 4.44) was
also pronounced since confidence curve crossed the horizontal line with a P-value that is
less than 0.05 suggesting a good correlation.
89
Figure 4.42. Leverage Plot Showing the
Effect of Fracture Width on Sealing
Pressure
Figure 4.43. Leverage Plot Showing the
Effect of Density on Sealing Pressure
The effect of D90 was the third significant parameter to affect the prediction of
sealing pressure. The Effect of D90 can be clearly seen (Figure 4.44) with a P-value that
is slightly larger than 0.05. The variation in LCM blends showed also a clear effect with
P-value 0.1147 (Figure 4.45)
Figure 4.44. Leverage Plot Showing the
Effect of D90 on Sealing Pressure
Figure 4.45. Leverage Plot Showing the
Effect of LCM Blend on Sealing Pressure
90
The Effect Leverage plots (Figure 4.46, 4.48, 4.49, 4.50, and 4.51) for D75, base
fluid, D25, D50, and D10, respectively shows less effect on the sealing pressure with P-
values ranging between 0.2863 and 0.9817. However, the less significance of these
parameters might be due to the outliers in the analyzed data set.
Figure 4.46. Leverage Plot Showing the
Effect of D75 on Sealing Pressure
Figure 4.47. Leverage Plot Showing the
Effect of Base Fluid on Sealing Pressure
Figure 4.48. Leverage Plot Showing the
Effect of D25 on Sealing Pressure
Figure 4.49. Leverage Plot Showing the
Effect of D50 on Sealing Pressure
91
Figure 4.50. Leverage Plot Showing the Effect of D10 on Sealing Pressure
A predictive linear fit model (equation given in Appendix E) is shown in Figure
4.51 shows a good correlation with an R2 of 80% and a P-value less than 0.05. The
residual plot (Figure 4.52) showed the data being randomly distributed around x-axis,
verifying that a linear model was appropriate for the collected data. Table 4.10
summarizes the P-values for the different variables as well as the model fit R2.
Figure 4.51. Leverage Plot Showing the
Actual Sealing Pressure versus the
Predicted Sealing Pressure using the fit
model
Figure 4.52. Residual Plot of Sealing
Pressure versus the Predicted Sealing
Pressure
92
Table 4.10. Effect of Different Parameters on the Sealing Pressure and Model Fit
Parameter Unit P-Values
Fracture Width (microns) 0.00028
Density (lb/gal) 0.00267
D90 (microns) 0.05957
LCM Blend N/A 0.11473
D75 (microns) 0.28628
Base Fluid (WBM/OBM) 0.30786
D25 (microns) 0.60427
D50 (microns) 0.75352
D10 (microns) 0.98169
Model Fit
R2 0.8
From the statistical analysis, the sealing pressure was highly dependent on the
different parameters in the following order: fracture width, fluid density, D90, LCM
blend/type, D75, base fluid, D25, D50, and D10. Out of these parameters, the fracture
width cannot be controlled and the fluid density should be designed based the mud
weight window.
4.6.2. Analytical Fracture Models. In this section, the five analytical fracture
models presented in subchapter 2.3 were used to estimate the fracture width based on the
hydraulic fracturing test results for the different blends and the mechanical properties of
concrete cores. The estimated fracture widths are then compared to the measured fracture
width from the microscopic images of the fractured cores.
Table 4.11 shows the input parameters used to estimate the fracture width from
the five analytical models. The applied confining pressure in the hydraulic fracturing test
was used for both the maximum and minimum horizontal stress. Both Young’s modulus
93
and Poisson’s ratio were measured previously (Weideman, 2014) for a cured class-H
Portland cement mixed and hardened following the procedure outlined in subchapter
3.3.1. The radius of the fractured core samples was used as the fracture length since the
fracture reached the end of the cores. The pressure along the fracture was set to zero
similar to previous studies (Morita et al. 2012).
Table 4.11. Analytical Models Input Parameters
Parameter Value Unit
Minimum horizontal stress σH 100 psi
Maximum horizontal stress σh 100 psi
Young's Modulus E 1450377 psi
Poisson’s ratio v 0.25
Fracture length L 2.65 inches
Wellbore radius rw 0.275 inches
Pressure along the fracture Pf 0 psi
The calculated fracture widths using the five analytical models and the range of
the measured fracture width under the microscope are tabulated in Table 4.12. It is clear
that both Hillerborg’s and Wang’s models gave unrealistically small fracture widths
compared to the actual measured widths. Alberty’s model gave relatively large fracture
width compared to the measure widths. Carbonell’s model gave fracture widths that are
comparable to the measured widths. The estimated fracture widths by Morita’s model
were constant for all different cases since the model is not a function of the wellbore
pressure.
94
Table 4.12. The Estimated Fracture Width from the Analytical Models and the Measured
Width under Microscope for the Five Fractured Core Samples
Fracture Estimation Method Fracture Width (microns)
Test # 1 Test # 2 Test # 3 Test # 4 Test # 5
Measured under Microscope 15 - 22 27 - 36 40 - 85 40 - 145 50 - 140
Hillerborg et al. (1976) 2.54 7.62 8.1 8.5 8.77
Carbonell and Detournay (1995) 22.86 29.46 30.48 31.31 31.7
Alberty and McLean (2004) 144.78 368.3 403.86 424.18 436.88
Wang et al. (2008) 0.082 0.2 0.21 0.21 0.23
Morita and Fuh (2012) 19.94 19.94 19.94 19.94 19.94
Based on the measured fracture width for the tests using the pre-mixed OBM
containing LCM with the PSD analysis of the tested blends (Chapter 4.1.2), we can see
that the fracture widths are within the range of the D10 and D25.
95
5. DISCUSSION
In this chapter, the experimental results and observations are discussed in details.
The results were also compared with previous results. The discussion is divided into 3
subchapters as follows; LCM performance evaluation, hydraulic fracturing experiment
results, and PSD selection criteria.
5.1. LCM PERFORMANCE EVALUATION
Figure 5.1 shows the low pressure (LPA) fluid loss results (circle) of the 26 tests
using the selected blend (Table 4.4) plotted with the average fluid loss per cycle (dash)
values calculated for the same blends under high pressure.
Figure 5.1. Comparison between Fluid Loss Values from the LPA and the Calculated
Average Fluid Loss per Cycle from the HPA
0
10
20
30
40
50
60
70
80
90
100
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
Flu
id L
oss (
ml)
Test #
Fluid Loss (ml) @ Low Pressure
Fluid Loss (ml) Per Cycle @ High Pressure
96
The majority of the LPA results in Figure 5.1 were comparable to the average
fluid loss per cycle, which could be attributed to a fixed amount of fluid lost prior to the
formation of a seal within the fracture. However, some fluid losses per cycle values
(circled on Figure 5.1) were significantly higher than the LPA values. This is believed to
be due to the continuous injection of drilling fluid through a permeable seal that did not
seal the fracture effectively. The high seal permeability resulted from the formation of a
thick plug due to the high concentration of larger LCM particles, which screened out at
the fracture surface (Figure 5.2) rather than plugging the slot. This comparison also
highlights the effect of higher LCM concentrations on the fluid loss. This observation
suggests that selecting the right LCM concentration is critical to avoid screen out at the
fracture openings, however further investigation is required to define the critical
concentration of LCM.
Figure 5.2. High Concentration of Nutshells (Left) and Graphite (Right) Treatments
Screened out at the Fracture Surface
After running HPA tests and when cleaning the used tapered discs, it was hard to
remove the formed seals when nutshells were used because the particles were stuck at the
97
fracture tip (Figure 5.3). This indicates a degree of deformability and strength of nutshells
particles that resulted in higher sealing pressure compared to other LCM treatments at the
same concentration.
Figure 5.3. Plugged Fracture tip of a 2000-micron Tapered Disc showing Stuck Nutshells
Particles
The sealing pressures versus the average fluid loss per cycle for the 26 tests
mentioned above were plotted in Figure 5.4, 5.5, and Figure 5.6 for 1000 microns, 1500
microns and 3000 microns fracture width, respectively. The results are plotted and circled
in groups from high to low sealing pressure.
From Figure 5.4 for tests using 1000 microns fracture width (TS1), it can be
observed that the results were distributed in three distinguished regions. In region (a),
higher sealing pressures were observed at lower fluid loss values <5 ml at small fracture
widths. While test (b) shows an evidence of LCM screen out at the fracture mouth, which
resulted in higher fluid loss but with high sealing pressure. Region (c) shows lower
sealing pressures with relatively higher fluid loss values compared to region (a).
98
Figure 5.4. Sealing Pressure vs. Average Fluid Loss per Cycle using 1000 microns
Fracture Width (TS1)
Figure 5.5 shows the sealing pressures versus the average fluid loss per cycle
using the 1500 microns fracture width (TS2).
Figure 5.5. Sealing Pressure vs. Average Fluid Loss per Cycle using 1500 microns
Fracture Width (TS2)
0
500
1000
1500
2000
2500
3000
3500
4000
0 5 10 15 20 25 30
Sealin
g P
ressu
re (
psi)
Avg. Fluid Loss per Cycle (ml)
50 ppb CF 15 ppb CF 40 ppb G & NS20 ppb G & NS 50 ppb NS 15 ppb NS55 ppb G, SCC & CF 50 ppb SCC 80 ppb G & SCC30 ppb G & SCC 50 ppb G 15 ppb G
(b)
(a)
(c)
0
500
1000
1500
2000
2500
3000
0 5 10 15 20 25 30 35
Sealin
g P
ressu
re (
psi)
Avg. Fluid Loss per Cycle (ml)
50 ppb NS 15 ppb NS 40 ppb G & NS
50 ppb CF 20 ppb G & NS 55 ppb G, SCC & CF
80 ppb G & SCC 50 ppb SCC
(b)
(a)
(c)
(d)
99
The figure above (Figure 5.5) shows 4 distinguished regions, where region (a)
contains high sealing pressures in conjunction with low fluid loss values similar to the
one observed in Figure 5.4. Region (b) shows high sealing pressure but with high fluid
loss values that are higher than 10 ml/cycle, while region (c) exhibited moderate sealing
pressures and fluid loss values. Test (d) shows a very low sealing pressure as well as high
fluid loss per cycle.
Figure 5.6 shows the sealing pressure versus the average fluid loss per cycle for
tests conducted using 2000 microns fractures (TS3 and TS4). In this figure, only two
regions were observed. Region (a) containes moderate sealing pressures with higher fluid
loss values compared to the results at smaller fractures, while region (b) exhibted high
sealing pressures with a very high fluid loss values.
Figure 5.6. Sealing Pressure vs. Average Fluid Loss per Cycle using 2000 microns
Fracture Width (TS3 and TS4)
0
500
1000
1500
2000
2500
3000
0 20 40 60 80 100
Sealin
g P
ressu
re (
psi)
Avg. Fluid Loss per Cycle (ml)
50 ppb NS @TS3 50 ppb NS @TS4
15 ppb NS @ TS4 40 ppb G & NS @ TS4
15 ppb NS @ TS3 40 ppb G & NS @TS3
(b)
(a)
100
Based on these observation of sealing pressures versus fluid loss per cycle at the
different fracture widths, it can be concluded that lower fluid loss values will result in
higher sealing pressures at fractures up to 1000 microns. However, the higher fluid loss
values and higher sealing pressures at fractures wider than 1000 microns suggest that de-
fluidization of the LCM treatments was required to form a strong seal.
The results of graphite and sized calcium carbonate mixtures and graphite and
nutshells mixtures were compared with the results published in Aston et al. 2004 and
Hettema et al. 2007. Aston et al. (2004) showed that both the 30 ppb and 80 ppb were
successful for field application in terms of enhancing the fracture gradient. The 80 ppb
graphite and sized calcium carbonate enhanced the break down pressure by
approximately 850 psi when applied as a stress cage treatment in a shale formation
(Arkoma basin). Due to the limited information about their laboratory test methodology,
it is difficult to establish a comparison between their findings and this study.
From the results published by Hettema et al. (2007) for the permeable fracture test
using the 1000 microns open fracture tip (Table A.2), which is comparable to the test
conditions in this study, higher sealing pressure were observed up to 4420 psi. Only one
test (Test # 10 in Table A.2) using the 40 ppb graphite and nutshells LCM blend in a
synthetic based mud (SBM) showed a comparable sealing pressure (2037 psi) compared
to our results, where 2398 psi and 2372 psi were observed in OBM and WBM,
respectively (Test # 36 in Table 4.6 and Test # 46 in Table 4.7). For the 20 ppb graphite
and nutshells blend, no tests were conducted at 1000 microns fractures by Hetema et al.
(2007) and therefore the results cannot be compared.
101
Hettema et al. (2007) concluded based on two tests using WBM and synthetic-
base mud (SBM) (Test # 6 and 7 in Table A.2) that the base fluid is largely independent
factor that does not have an effect on the sealing pressure. Our investigation showed that
graphite and nutshells blend was not affected by the base fluid; however, the effect of
base fluid on the sealing pressure for blends using graphite and sized calcium carbonate
was significant (Figure 4.22).
The effect of density on the sealing pressure was also studied by Hettema et al.
(2007) where 3 tests were repeated at 3 different densities. It was concluded that higher
density will reduce both fluid loss and sealing pressure. However, a different effect was
observed in our study (Figure 4.23 and Figure 4.24) where higher densities resulted in
improving the sealing pressure.
The laboratory investigation by Kumar et al (2011) was conducted on the particles
plugging apparatus (PPA) using one tapered disc size (1000 microns) where LCM
performance was based on the ability of LCM blends in forming an impermeable seal in
the tapered disc. However, no information about the applied pressure was provided.
Therefore, no comparison for the sealing pressures could be established with their results.
It was concluded from their study that sized calcium carbonate and graphite blends were
not very successful in arresting fluid loss which is in agreement with what was observed
in this study for similar blends.
To investigate the reason behind the better performance of hematite compared to
barite (Subchapter 4.3.7), particle size distribution analysis was conducted for both barite
and hematite using laser diffraction technique to investigate the PSD differences.
102
Figure 5.7 shows the PSD analysis for barite (black) and hematite (red). It can be
seen that hematite has slightly coarser distribution compared to barite, which could have
resulted in improving the sealing pressure. Table 5.1 shows the PSD analysis of the used
weighting materials compared to the PSD conducted by Tehrani et al. 2014 for API size
barite and hematite, which is relatively comparable.
Figure 5.7. Particle Size Distribution for the used Barite and Hematite using Laser
Diffraction Technique
Table 5.1. Particle Size Distribution Analysis Comparision for Barite and Hematite
PSD (microns)
Weighting Material Barite Hematite
Measured
D10 1.8 6
D50 13 23
D90 50 70
Taken from
Tehrani et al.
2014
D10 3 3
D50 21 27
D90 64 68
1001010.1
100
90
80
70
60
50
40
30
20
10
0
Size (microns)
% F
ine
r
D 90
D 50
D 10
Barite
Hematite
103
To investigate the effect of sieving barite on the sealing pressure, a sample of
barite was sieved into three grades; fine (F), medium (M), and coarse (C). Cutoffs for
defining the three ranges of sizes were made up as follows:
a) Particles equal to or smaller than 50 microns were considered fine.
b) Particles > 50 microns < 90 microns were considered medium.
c) Particles larger than 90 microns were considered coarse.
The sieved samples were used to weight up OBM containing 30 ppb graphite and
sized calcium carbonate blend # 1 to 12.5 lb/gal. Figure 5.8 shows the sealing pressure
when using different grades of the sieved barite. The fine barite resulted in increasing the
sealing pressure by only 5% while using the cores and medium barite resulted in reducing
the sealing pressure by approximately 10% and 5%, respectively. The reduction in the
sealing pressure as a result of using course and medium grade (as per the definition
above) barite highlights the effect of finer particles in improving the sealing pressure.
Figure 5.8. Effect of Sieving Barite in OBM
950
1000
1050
1100
1150
1200
1250
1300
No Sieve Coarse Medium Fine
Sealin
g P
ressu
re (
psi)
Sieved Barite
104
5.2. HYDRAULIC FRACTURING EXPERIMENTS
From the hydraulic fracturing experiments on concrete cores, it was observed that
in general the addition of LCM to fracturing fluids enhanced both the breakdown and re-
opening pressures compared to the results obtained from the same fluid that has no LCMs
(Table 4.9). However, when comparing the results of the fractured core with the solid
free clear fluid to the one that has been pre-mixed and weighted with barite, only the
breakdown was significantly enhanced by 138% while the re-opening improved slightly
by 4%.
Similar hydraulic fracturing experiments were conducted previously to investigate
the effect of adding nanoparticles and graphite to an OBM on the breakdown and re-
opening pressure of shale cores (Contreras et al. 2014b). The same test condition such as
the overburden pressure, confining pressure, and injection rate were applied to fracture
Catoosa shale cores. Two injection cycles were also applied allowing 10 min for fracture
healing after the first cycle.
A summary of 7 hydraulic fracturing tests results, using different fluid
formulation (Table 5.2), with the % increase in breakdown and re-opening pressures as a
result of adding nanoparticles is tabulated in Table 5.3.
Table 5.2. Description of the Different Fluid used in the Fracturing Experiments
LCM
Blend Description
CS Control sample without NPs or graphite
G1 Graphite blend # 1
G2 Graphite blend # 2
105
Table 5.2. Description of the Different Fluid used in the Fracturing Experiments (Cont’d)
DF1 Graphite and Iron-based NPs blend # 1
DF2 Graphite and Iron-based NPs blend # 2
DC4 Graphite and Calcium-based NPs blend # 4
DC6 Graphite and Calcium-based NPs blend # 6
Table 5.3. Summary of the Hydraulic Fracturing Experiments for the Shale Cores
(Contreras et al. 2014b)
Oil Based Mud Results
LCM
Blend
Breakdown
Pressure
(psi)
% Inc. in
Breakdown
Pressure
Re-opening
Pressure
(psi)
% Inc. in
Re-Opening
Pressure
% Inc. in
Re-Opening
Pressure
Above the
Breakdown
Pressure
CS 520 N/A 498 N/A N/A
G1 522 0.4 533 7 3
G2 520 0 538 8 3.5
DF1 524 0.8 626 26 20
DF3 524 0.8 552 11 6
DC4 520 0 659 32 27
DC6 503 -3.3 674 35 30
The % increase in the reopening pressure using an oil based mud was calculated
based on the breakdown (520 psi) and re-opening (448 psi) pressures for the control
sample. In addition, the % increase in the re-opening pressure when compared to the
breakdown pressure was calculated.
106
The results in Table 5.3 showed a different enhancement behavior compared to
the results shown in Table 4.9. The breakdown pressures were not enhanced when
nanoparticles were added compared to the control sample. However, an increase in the
re-opening pressures not only above the control sample re-opening pressure but also
above the breakdown pressure of the control sample. An increase of up to 30% above the
original breakdown pressure was observed when calcium-based NPs blend (DC6) was
used. The same was observed with iron-based NPs blend (DF1) where the re-opening
pressures were enhanced by 20% above the original breakdown pressure.
The pressures vs. time plots obtained from the hydraulic fracturing experiments
with the optimum nanoparticles concentrations for both; iron- based and calcium-based
NPs are shown in Figure 5.9 - Figure 5.11.
Figure 5.9. Pressure vs. Time for the Control Sample (CS) without Nanoparticles
0
100
200
300
400
500
600
700
800
0 10 20 30 40 50
Pre
ssu
re (
psi)
Time (min)
1st Cycle (CS)
2nd Cycle (CS)
Breakdown @ 520 psi Re-Opening @ 498 psi
107
Figure 5.10. Pressure vs. Time for the Optimum Concentration of Iron-based
Nanoparticles Blend (DF1)
Figure 5.11. Pressure vs. Time for the Optimum Concentration of Calcium-based
Nanoparticles Blend (DC6)
The detailed analysis of the fractured shale cores under the SEM (Figure 5.12)
and the energy dispersive X-ray spectroscopy (EDX) (Figure 5.13) showed that the
fractures were completely sealed from wellbore to tip.
0
100
200
300
400
500
600
700
800
0 10 20 30 40 50
Pre
ssu
re (
psi)
Time (min)
1st Cycle (DF1)
2nd Cycle (DF1) Re-Opening @ 626 psi
Breakdown @ 524 psi
0
100
200
300
400
500
600
700
800
0 10 20 30 40 50
Pre
ssu
re (
psi)
Time (min)
1st Cycle (DC6)
2nd Cycle (DC6)
Breakdown @ 503 psi
Re-Opening @ 674 psi
108
Figure 5.12. (a) Cross-section of Fracture Plane near Wellbore, (b) Seal at Fracture
Mouth, and (c) Cross-section of Fracture Plane at the Fracture End (Contreras et al.
2014b)
Figure 5.13. EDX of Seal Cross-section at 500x magnification where the Green Color
Indicates Iron Particles Distribution on the Formed Seal (Contreras et al. 2014b)
This observation suggests that the nanoparticles penetrated through the resulting
fracture and formed the seal inside the fracture. However, the observed fractures in the
concrete cores were not sealed with LCMs. LCMs were observed within the formed filter
cake around the wellbore and present inside the fracture in the immediate vicinity close to
the fracture mouth. The calcium-based and iron-based nanoparticles have an average size
(a) (c) (b)
(b)
109
(D50) of 60 and 30 nm, respectively (Zakaria, 2013), which makes it easy for these
nanoparticles to penetrate inside the fractures while the D10 values of the conventional
LCM treatments used on the concrete cores ranged between 55 and 80 microns (Table
4.1).
The average fracture width measure after retrieving the shale cores form the
experiment was found to be approximately 20 microns (Figure 5.14), which can be
compared to the measured fracture width of the concrete core that was fractured with
clear fluid. With this small width and the size of nanoparticles, the nanoparticles
penetration inside the fracture plane is possible.
Figure 5.14. Fractures at Core End (left) and 3D Representation of the Core End
Containing the Fractures (right) (Contreras et al. 2014b)
5.3. PARTICLE SIZE DISTRIBUTION SELECTION CRITERIA
From the qualitative analysis of the effect particle size distribution (PSD) has on
sealing pressure (Chapter 4.3.4), the improved performance of treatments containing
(b) (c)
110
nutshells could be attributed to the fact that the D90 values of theses blends ranged
between 1900 – 2400 microns. In general, all LCM treatments having a D90 smaller than
the fracture width failed to establish a seal within the fracture. This suggests that the D90
value should be selected such that it matches the fracture width to ensure an effective
fracture sealing. In addition, the statistical analysis (Chapter 4.6.1) showed that the
sealing pressure is highly affected by the D90 and D75; hence proper selection of the D-
values is required to ensure effective fracture sealing. However, the current selection
methods were developed to enhance the bridging capabilities for drill-in fluid not to seal
fractures. Therefore, using these selection criteria might be somehow irrelevant for
fracture sealing applications.
In this section, the development of a method to select PSD of LCM treatments for
a known fracture width is presented. The method was developed based on a statistical
analysis of a subset dataset (Table F. 1) containing 31 tests to determine the relationship
between the effectiveness of different LCM treatments in terms of the sealing efficiency
and PSD.
The main reason for running the statistical analysis on a subset dataset is
eliminate the effect of other parameters such as the drilling fluid density and LCM
particles with highly irregular shapes. Since cellulosic fiber particles were described as
nonspherical particles with some elongated and flat particles, PSD analysis of blends
containing cellulosic fiber was found to be unrepresentative. For this reason it was
decided to exclude their results from the statistical analysis and limit it to granular
materials only. In addition, all the results for tests run with higher drilling fluid densities
111
were excluded since the density was found to have a strong effect on the sealing pressure
(Table 4.10).
5.3.1. Development of PSD Selection Method. In normal overbalanced drilling
operations (i.e. drilling fluid pressure higher than formation pressure), a minimum static
overbalance pressure of 150-300 psi is required to prevent formation fluid influx (Jahn et
al. 2008; Rehm et al. 2012; The Drilling Manual, 2015). Therefore, the formed seal
within the fracture should withstand at least the minimum overbalance pressure without
failing. In the statistical analysis, the sealing efficiency results of the different blends
were grouped into two categories; weak and strong seal. The seals that failed to withstand
an overbalance of 500 psi to (selected to account for the static and dynamic overbalance
pressure) are considered weak seal and the ones that survived 500 psi or more are
considered strong seal regardless of LCM type, concentration and fracture width.
A simple linear model was fitted using fracture width, fluid base, total LCM
concentration and the five D-values. The Leverage plot and P-test was performed on each
parameter individually to get an overall understanding of how each parameter affect the
sealing pressure (Table 5.4). The overall model fit resulted in a good correlation with R2
of 74% and a P-value less than 0.05 (Figure 5.15)
Table 5.4. Effect of Different Parameters on the Sealing Pressure and Model Fit for the
Subset Dataset
Parameter Unit P-Values
Fracture
Width (microns) 0.00001
D90 (microns) 0.08116
D50 (microns) 0.26377
D10 (microns) 0.44148
112
Table 5.4. Effect of Different Parameters on the Sealing Pressure and Model Fit for the
Subset Dataset (Cont’d)
D75 (microns) 0.53233
Total LCM (ppb) 0.53522
D25 (microns) 0.85566
Base Fluid (WBM/OBM) 0.90849
Model Summary Fit
R2 0.743
Figure 5.15. Leverage Plot Showing the Actual Sealing Pressure versus the Predicted
Sealing Pressure using the Simple Linear Fit Model for the Subset Dataset
In order to develop a generalized model that predicts the sealing pressure for
different fracture widths, statistical partition technique was used based on the obtained
information from the previous model. The statistical partition was performed using the
four significant parameters; D90, D75, D50 and D10 as a function of fracture width and
taking the total LCM as a weighted parameter. Sensitivity analysis was then performed
by first taking the three parameters (D90, D50 and D10) to predict sealing pressure and
113
second by taking only the first two parameters (D90 and D50). The model using D90 and
D50 to predict sealing pressure for a given fracture width gave a model fit similar (3%
difference) to the model with the four parameters. Thus, the model using D90 and D50
was chosen, where the model have a better fit than the previous model with an R2 of
74%. The statistical partitioning was then converted into a mathematical relationship that
predicts a good sealing pressure, which was defined earlier as any pressure >500 psi, as a
function of the selected PSD D-values. The obtained mathematical relationship suggests
the following:
D50 should be ≥ 3
10 the fracture width
D90 should be ≥ 6
5 the fracture width
In order to validate the proposed selection criteria and compare it with the
previous selection criteria discussed in subchapter 2.5 (Table 2.3), a comparison table
was constructed (Table F.2). The table lists the test number, the fracture width used, the
measured seal integrity, and the predicted seal integrity.
The measured seal integrity is based on the measured sealing pressure, i.e. if the
measured sealing pressure is larger than 500 psi, then the measured seal integrity is
considered strong. If the measured sealing pressure is less than 500 psi, then the
measured seal integrity is considered weak. In the comparison, it is assumed that
following the selection criteria should result in a strong seal that could withstand
pressures higher than 500 psi. The predicted seal integrity column is based on comparing
the suggested D-Value by the selection criteria with the actual measured PSD. If the
measured PSD is equal or larger than the suggested value by the selection criteria, the
114
predicted seal integrity is considered strong. If the measured PSD is less than the
suggested PSD, the predicted seal integrity is considered weak. The last step is to check
if the measured seal integrity matches with the predicted one and calculate the overall
match percentage. The overall match percentage was calculated for the selected 31 tests
using the previous selection criteria as well as the proposed criteria as shown in Table
5.5.
Table 5.5. Comparison of Percentage Match for the Different PSD Selection Criteria
Proposed
Method
Abrams
Rule (1977)
D90 Rule
(Smith et al. 1996,
Hands et al. 1998)
Halliburton
Method
(Whitfill, 2008)
Vickers’s
Method
(2006)
90% 68 % 77 % 55 % 45%
The verification showed a good match between the predicted seal integrity and the
measured seal integrity pressure with an overall match of 90% for the proposed selection
criteria. Vickers method showed the lowest percentage match due to the number of D-
values that needs to be satisfied. Halliburton’s method (Whitfill, 2008) also showed a low
predictive match. Abrams and D90 rules had higher percentage match compared to
Vickers and Halliburton’s methods. Since the validation was performed on the same
dataset, additional testing would be beneficial to independently validate the proposed
criteria. However, the result is in agreement with the observation in subchapter 4.3.4
were a wide particle size distribution gave higher sealing pressure since smaller particles
filled the voids between the larger particles in the formed plugs.
115
6. CONCLUSIONS AND RECOMMENDATIONS
6.1. CONCLUSIONS
In this dissertation, an experimental study of the factors affecting the effectiveness
of LCM treatments in sealing fracture was carried out with a characterization of the used
LCMs. Two apparatuses were developed to evaluate the capability of LCM treatments in
sealing fractures, as well as measuring the maximum pressure that can be achieved before
the seal/plug breaks. The investigated factors include; fracture width, LCM type, LCM
concentration, temperature, PSD, base fluid, fluid density, weighting material, and LCM
shape. Experimental results and statistical methods were used in this study to provide a
better understanding of the reasons behind the variation in LCM performance. Hydraulic
fracturing tests were conducted to investigate the effect of LCM addition to drilling fluid
for wellbores strengthening application and the results were used to validate how realistic
available analytical models in estimating fracture widths.
An updated classification of LCM into eight categories, based on their appearance
and field application, was presented to include the recent technologies. In addition, a
comprehensive list of available LCM, which includes the generic name and description of
each LCM as well as the trade name and it field application, was constructed to serve as a
cross-reference table for operators, service companies and drilling engineers.
Based on this work the following conclusions can be drawn;
Widening the fractures (increasing slot size) reduced sealing pressures.
Lower fluid loss values resulted in higher sealing pressures for fractures up to
1500 microns. For wider fractures, de-fluidization was required to initiate a strong
plug for fractures.
116
Blends containing nutshells showed the best performance in terms of sealing
fractures at both low (15 - 20 ppb) and high concentrations (40 – 50 ppb).
Graphite and sized calcium carbonate blends were only effective at a very high
concentration (105 ppb). This can be attributed to the effect of LCM particle
shape, in terms of sphericity and roundness, which showed a significant effect on
the overall sealing pressure where higher angularity resulted in higher sealing
pressure and higher sphericity resulted in lower sealing pressure.
The sealing pressure increased with higher LCM concentrations; however, there is
a critical maximum concentration at which LCM will not penetrate the fracture
and no tight seal will be formed.
The variation in the sealing pressure at high temperatures suggests that LCM
treatments should be evaluated at the expected formation temperature.
The increase in the sealing pressure for nutshells by 54% when used in single
LCM treatment at high temperature is attributed to the swelling of nutshell
particles, which was quantified experimentally.
An updated selection criterion for PSD selection for a known fracture width was
developed stating that D50 and D90 should be equal or greater than 3/10 and 6/5
the fracture width, respectively.
The proposed criterion was verified with the lab results and it gave a 90% match
between the actual and predicted performance.
There is no clear trend on how different base fluids affect the sealing pressure.
Increasing fluid density increased the sealing pressure for both WBM and OBM.
The finer the PSD of the weighting materials, the higher the sealing pressures.
117
Conventional LCMs used in this study were able to seal fractures up to 2500
microns. For fractures wider than 2500 microns, the unconventional LCM
treatments were able to seal fractures up to 5000 microns using WBM. However,
the unconventional LCM failed to initiate a strong seal at 1500 microns with
OBM.
The lowest breakdown and re-opening pressure for the fractured concrete samples
was observed with the solid free clear base fluid (EDC95-11).
The addition of different LCM blends enhanced the breakdown and re-opening
pressure up to 18% and 210%, respectively, compared to the control sample that
contained no LCM. Only tests with nanoparticles were able to increase reopening
pressure above breakdown pressure.
The fractured core using base fluid only had the smallest fracture widths
compared to the other cores, which were fractured with fluid containing solids.
Of the five analytical fracture models evaluated, only one model (Carbonell and
Detournay, 1995) estimated comparable fracture widths to the measured widths
on the fractured cores. For the other models, two models underestimated fracture
width, one model overestimated fracture width, and the last model had a constant
fracture width.
As a final note, this experimental study shows that carefully selected LCM
treatments can successfully seal drilling induced and natural fractures up to 2500 microns
for any base fluid. This result is higher than any previously reported studies. For larger
fractures there is still need for unconventional LCM materials or conventional methods
such as setting cement plugs. It is also seen that introducing LCM particles in the drilling
118
fluid is causing the fracture breakdown pressure and the fracture reopening pressure (i.e.
propagation pressure) to be increased. This shows that the experienced safe mudweight
window can be manipulated with including LCM materials in the drilling fluid.
6.2. RECOMMENDATION AND FUTURE WORK
Even though a large data set was developed from the high pressure LCM
evaluation, there are some experimental limitations with the current apparatus. The fluid
loss values were measure at the end of each test and the average fluid loss per cycle was
calculated based on the number of cycles for each test. The high pressure tests were
conducted at low injection rate, which does not simulates the flow rates seen in a
wellbore. Fluid loss values should be recorded with respect to time in order to provide an
accurate measurement of fluid loss through fractures. The testing apparatus could be
improved by installing a flow loop that circulates drilling fluids across the fractures.
Different factors that might have an effect on the sealing pressure were not
addressed in this study. Only one fracture length (i.e. slot thickness) was investigated for
conventional LCM evaluation. The designed tapered slots do not simulate the actual
formation roughness, which might have an effect on the overall sealing effectiveness. The
slot fracture angle also might have an effect on the packing of LCM inside the fracture. A
parametric study of how the variation in fracture length, roughness, and angle is
suggested to investigate how slot design will affect the results.
Only one unconventional LCM treatment (WEDGE-SET) was used in this study
due to the difficulty in getting other unconventional treatments from services companies.
Further investigation on unconventional treatments is encouraged.
119
The effect of increasing the temperature was investigated at one temperature;
however further investigation using a wider range of temperatures is required to establish
an understanding of how LCM is affected.
Five hydraulic fracturing experiments were carried out in this study using OBM
on concrete core samples, where 3 tests were performed using LCMs. Future work should
include further investigation using permeable and impermeable rocks using WBM and
different LCM treatments to identify the relationship between the maximum sealing
pressure and the enhancement in the breakdown and re-opening pressure.
The fracture widths of the fractured cores were measured after removing the
applied overburden and confining stresses. Therefore, the measured width represent the
relaxed fracture width rather than the actual stressed width. For a better measurement,
future work should consider in-situ fracture width estimation.
120
APPENDIX A
A. PREVIOUSLY REPORTED LABORATORY RESULTS
121
Table A.1. Summary of Results from The Impermeable Test using Different LCMs and
Fracture Widths (Tehrani et al. 2007)
Test
# LCM Used
Concn
(ppb)
Base
Fluid
Fracture
Width
(microns)
Fluid
Loss
(ml)
Sealing
Pressure
(psi)
1
SCC (C/M) 20
WBM 530 12.8 73 SCC (F) 10
CF (F) 20
2
SCC (C/M) 20
WBM 848 3.3 998 SCC (F) 10
CF (F) 20
NS (F) 10
3 G 40
OBM 555 4.4 425 CF (F) 10
4
G 40
OBM 637 3.2 876 CF (F) 10
NS (F) 10
5
G 3.5
OBM 742 4.7 900 SCC (C/M) 11.5
SCC (F) 14.7
SCC (C) 10.2
6
G 20.5
OBM 650 3.5 685 SCC (M/F) 16
SCC (F) 3.5
7
G 7
OBM 555 14.9 491 SCC (M/F) 17.5
SCC (F) 17.5
Note. Coarse (C) Medium (M) Fine (F)
122
Table A.2. Summary of the Results from the Permeable Test using Different LCMs and
Fracture Widths (Hettema et al. 2007)
Test # LCM
Used
Concn.
(ppb)
Base
Fluid
Fracture
Width
(microns)
Fluid
Loss
(ml)
Sealing
Pressure
(psi)
1 None 0 OBM 250 66.5 679
2 None 0 SBM 250 78.1 1746
3 None 0 SBM 250 53.3 2733
4 SCC 27
SBM 500 45.5 988 G (A) 13
5 SCC 27
SBM 500 6.7 1547 G (A) 13
6 G (B) 20 SBM 500 13.9 1962
7 G (B) 20 WBM 500 35.14 2041
8 G (B) 20 OBM 500 8.8 3836
9 NS 10
SBM 500 19.5 6132 G (B) 10
10 NS 20
SBM 1000 11.7 2037 G (B) 20
11 NS 20
SBM 1000 2.5 3611 G (B) 20
12 NS 20
SBM 1000 1.8 4420 G (B) 20
123
Table A.3. Summary of Results from the Permeable Test using Different LCMs and
Fracture Widths (Kaageson-Loe et al. 2009)
Test # LCM
Used
Concn.
(ppb)
Base
Fluid
Fracture
Width
(microns)
Fluid
Loss
(ml)
Sealing
Pressure
(psi)
1 G (R)
* 20
SBM 250 N/A 1490**
NS (R)
* 20
2 G (R)
* 20
SBM 250 N/A 1710**
NS (R)
* 20
3 G (R)
* 10
SBM 250 N/A 1970 NS (R)
* 10
4 G (R)
* 20
SBM 500 N/A 1190 NS (R)
* 20
5 G (R)
* 20
SBM 500 N/A 1520**
NS (R)
* 20
6 G 20
SBM 500 N/A 3900 NS 20
7 G 20
SBM 500 N/A 6131 NS 20
8 G 20
SBM 500 N/A 6132 NS 20
9 G 20
SBM 1000 N/A 1560**
NS 20
10 G 20
SBM 1000 N/A 2037 NS 20
11 G 20
SBM 1000 N/A 2803 NS 20
12 G 20
SBM 1000 N/A 3611 NS 20
* Resized LCMs ** Test conducted at closed fracture tip
124
APPENDIX B
B. HYDRAULIC FRACTURING TEST PREPARATION AND PROCEDURES
125
CORE PREPARATION
1- Mix cement in a large container to ensure the consistency of the cores (Figure B.
1.1)
2- A 94 lbs sack of Portland cement Class-H was mixed with 38% of water (35 lbs
of water) following the recommended mixing procedures.
3- Apply grease or any kind of lubricants inside the molds to easily extract the
samples after the dry.
4- Pour the cement mixture into 5 7/8 inch (diameter) x 9 inch (height) molds
(Figure B. 1.2) while hitting on the molds outside to get rid of the air bubbles.
5- Let the concrete to cure for at least 7 days.
6- Drill ½ inch wellbore in the cement cores using the drill press (Figure B. 1.3)
7- Screw casings into the top and bottom cap.
8- Attach top and bottom caps to the cement cores using epoxy (Figure B. 1.4) and
let the epoxy dry for 24 hrs. Note that each cap needs to be attached on a different
day.
9- Grind the core sample or use sanding papers to remove excess dried epoxy and
ensure a smooth surface to avoid damaging the confining rubber sleeve.
10- Screw the injection nipple into the top cap (Figure B. 1. 5)
126
Figure B. 1. Concrete Core Sample Preparation Steps
(1) Mix Cement (2) Pour cement in molds
(4) Attach top and bottom cap (3) Drill wellbore
(5) Screw Injection Nibble
127
TEST PROCEDURES
1- Raise the pressure cell and remove the clevis pins after removing the safety cutter
pins.
2- Lower the pressure cell (Figure B. 2).
3- Place an O-ring inside the pressure cell at the bottom.
4- Place the prepared core in the fracturing cell carefully.
5- Place the three top spacers with O-rings between them on the core sample.
6- Raise the pressure cell and place the clevis pins back again with the safety cutter
pins.
7- Once the clevis pins are in place, drop the cell on the clevis until the hoist cables
are no longer in tension.
8- Close the cell exit valve.
9- Fill up wellbore with drilling fluid containing LCM
10- Connect the injection line into the injection nipple.
11- Connect the confining line to the cell and close the confining exit line.
12- Apply 400 psi overburden on the core, which reads 7000 psi in the upper gauge,
using the hydraulic hand pump.
13- Apply 100 psi confining pressure using the Isco Pump.
14- Fill up accumulator with drilling fluid (without LCM) using the upper plastic
accumulator.
15- Open the Isco pump software and assign a name to file and take the injection
pump into the remote control.
128
16- Change the flow rate to 5ml/min and start recording the data (Logging ON) and
hit the start button to start injection.
17- Continue injection until a dramatic decrease in the injection pressure is observed,
which indicates the sample is fractured. Note that the confining pump will show
an increase in the pressure above too.
18- Stop the injection and open the mud exit valve to relief wellbore pressure
19- Close the mud exit valve and start timing (10 minute for fracture healing) before
the second cycle.
20- Refill injection pump if needed.
21- Start the second injection cycle until another drop in the injection pressure is
observed.
22- Stop the pump and stop the data logging, at this point the test is complete.
23- Open the mud exit valve again to release the wellbore pressure.
24- Remove the overburden pressure and open the confining exit valve to remove
confining pressure.
25- Unscrew both injection and confining lines from the pressure cell.
26- Raise the pressure cell and remove the clevis pins after removing the safety cutter
pins.
27- Lower the pressure cell.
28- Remove the three top spacers.
29- Pull the core sample out carefully.
30- Clean the system, including the cell inside.
129
31- Clean both plastic and metal accumulator before the next test if different fluid will
be used.
Figure B. 2. Hydraulic Fracturing Apparatus; (a) Metal Accumulator, (b) Plastic
Accumulator, (c) Hydraulic Hand Pump, (d) Overburden Gauge, (e) Pressure Cell, (f)
Confining Pump, (g) Injection Pump, (h) Clevis Pin
a
b
c
d
e
f
g
h
130
APPENDIX C
C. CROSS-REFERENCING TABLES FOR AVAILABLE LCMS UNDER THE NEW
CLASSIFICATION
131
Table C. 1. Examples of Granular LCMs
Granular LCM's
Generic Name /
Description Trade Name Provider Application
Ground/Sized walnut
shells
WALL-NUT Halliburton Used as concentrated
pills or high filtration
squeeze. It is used to
cure seepage, partial and
total loss based on the
selected grade.
MIL-PLUG Baker Hughes
NewPlug NEWPARK
WALNUT HULLS GEO Drilling Fluids
NUTSHELL Anchor Drilling Fluids
MESUCO-PLUG Messina Chemicals
Resilient, angular,
dual-carbon based,
sized graphite
STEELSEAL Halliburton Can be used as a
background treatment or
as a concentrated pill
depending on the losses
rate.
Used as wellbore
strengthening material.
G-SEAL MI SWACO
C-SEAL MI SWACO
LC-LUBE Baker Hughes
NewSeal NEWPARK
A proprietary natural
loss prevention
material (LPM)
SURE-SEAL Drilling Specialties Can be used as a
preventive treatment or
as a concentrated pill.
TORQUE-SEAL Drilling Specialties
A blend of acid
soluble particulates EZ-PLUG Halliburton
Can be used as a
background treatment,
for seepage losses or
severe losses
Sized-ground marble
BARACARB Halliburton Used as a bridging agent
for lost circulation
problems.
Used as wellbore
strengthening material.
SAFE-CARB MI SWACO
NewCarb NEWPARK
FLOW-CARB Baker Hughes
MIL-CARB Baker Hughes
W. O. 30 Baker Hughes Used for severe losses.
132
Table C. 2. Examples of Flaky LCMs
Flaky
Generic Name /
Description Trade Name Provider Application
Cellophane
MILFLAKE Baker Hughes Used in conjunction with
other LCM's depending
on the severity of the
losses.
MESUCO-
FLAKE Messina Chemicals
Sized grade of Mica
MILMICA Baker Hughes Used as preventive
measures for seepage
losses. MESUCO-MICA Messina Chemicals
Flaked Calcium
Carbonate SOLUFLAKE Baker Hughes
Used for seepage or
severe losses based on
the selected grade.
133
Table C. 3. Examples of Fibrous LCMs
Fibrous
Generic Name /
Description Trade Name Provider Application
Natural cellulose fiber
BAROFIBRE Halliburton
Can be used as a
preventive treatment or
concentrated pills to cure
seepage to severe losses
based on the selected
grade.
M-I-X II MI SWACO
VINSEAL MI SWACO
CHEK-LOSS Baker Hughes
MESUCO-
FIBER Messina Chemicals
CyberSeal NEWPARK
FIBER SEAL GEO Drilling Fluids
A proprietary micro-
cellulosic fiber for use
in water base muds
DYNARED Drilling Specialties Used as normal
treatment to cure
seepage losses or as a
concentrated pill for loss
circulation.
A proprietary micro-
cellulosic fiber for use
in oil base muds
DYNA-SEAL Drilling Specialties
Shredded cedar fibers.
M-I CEDAR
FIBER MI SWACO Can be used as a
preventive treatment,
concentrated pill or high
fluid-loss squeezes.
FIBER PLUG Anchor Drilling Fluids
PLUG-GIT Halliburton
MIL-CEDAR Baker Hughes
Acid soluble extrusion
spun mineral fiber
N-SEAL Halliburton Can be used as a
background treatment or
as a concentrated pill
CAVI-SEAL-AS Messina Chemicals
MAGMA FIBER GEO Drilling Fluids/
Anchor Drilling Fluids
134
Table C. 4. Examples of LCMs Combinations
LCM's Combinations
Generic Name /
Description Trade Name Provider Application
A combination of
different LCM types
and wide range of
particle sizes.
STOPPIT Halliburton Used as a concentrated
pill. PRIMA SEAL GEO Drilling Fluids
STOP-FRAC S Halliburton
WELL-SEAL Drilling Specialties
Used to cure minor to
severe losses based on
the selected grade.
BARO-SEAL Halliburton Can be used as a
preventive treatment or
concentrated pills to cure
seepage to severe losses
based on the selected
grade.
STOP-FRAC D Halliburton
M-I SEAL MI SWACO
MIL-SEAL Baker Hughes
CHEM SEAL Anchor Drilling Fluids
KWIK-SEAL Messina Chemicals
MESUCO-SEAL Messina Chemicals
A blend of acid
soluble particulates. EZ-PLUG Halliburton
Can be used as a
background treatment,
for seepage losses or
severe losses.
A proprietary
particulate blend
designed to be used
with foam wedges.
QUIK-WEDGE Sharp-Rock
Technologies, Inc.
Can be used as a pre-
treatment or as a
concentrated pill.
A proprietary
particulate blends
that includes
modified natural
materials and other
additives.
STRESS-SHIELD Sharp-Rock
Technologies, Inc.
Can be used as a pre-
treatment or as a
concentrated pill
135
Table C. 5. Examples of Acid Soluble LCMs
Acid Soluble/ Biodegradable/Water Soluble
Generic Name /
Description Trade Name Provider Application
A blend of acid soluble
particulates EZ-PLUG Halliburton
Can be used as a
background treatment,
for seepage losses or
severe losses
A non-damaging, cross
linkable water soluble
polymer blended with
selected sized
cellulosic fibers.
N-SQUEEZE Halliburton Used as a settable
squeeze for severe losses
Acid soluble extrusion
spun mineral fiber
N-SEAL Halliburton Can be used as a
background treatment or
as a concentrated pill
CAVI-SEAL-AS Messina Chemicals
MAGMA FIBER GEO Drilling Fluids/
Anchor Drilling Fluids
Sized and treated salts BARAPLUG Halliburton
Use as a temporary seal
in high permeability
formations
Acid Soluble Sized-
Calcium Carbonate
BARACARB Halliburton
Used as a bridging agent
for lost circulation
problems
SAFE-CARB MI SWACO
NewCarb NEWPARK
FLOW-CARB Baker Hughes
MIL-CARB Baker Hughes
W. O. 30 Baker Hughes Used for severe losses.
Flaked calcium
carbonate SOLUFLAKE Baker Hughes
Used for seepage or
severe losses based on
the selected grade.
Nontoxic fibrous
powdered
polysaccharide,
biodegradable and acid
soluble lost circulation
material.
HOLE-SEAL-II Messina Chemicals
Can be used as a
pretreatment or as a
concentrated pill.
136
Table C. 6. Examples of High Fluid Loss LCMs Squeezes
High Fluid Loss LCM's Squeeze
Generic Name /
Description Trade Name Provider Application
High fluid loss squeeze GEO STOP
LOSS GEO Drilling Fluids
Used as a high-fluid loss
squeeze.
High-solids, high-
fluid-loss reactive lost
circulation squeeze
DIASEL M Drilling Specialties
A specially formulated
high-solids high fluid
loss squeeze.
DIAPLUG Messina Chemicals
A proprietary blend of
granular and fibrous
materials.
X-Prima NEWPARK
A blend of granular
and fibrous materials. NewBridge NEWPARK
Micro-sized cellulosic
fiber combined with a
blend of organic
polymers
ULTRA SEAL GEO Drilling Fluids Used as concentrated
pills.
A blend of fine
particles to promote
high fluid loss and
other additives in
addition to highly
compressible and
permeable foam rubber
chunks.
WEDGE-SET Sharp-Rock
Technologies, Inc.
Used as a high-fluid loss
squeeze.
A combination of both
resilient graphitic
carbon and malleable
components
DUO-SQUEEZE Halliburton
Can be used as a high
fluid loss squeeze or as
concentrated pill.
137
Table C. 7. Examples of Settable/Hydratable LCM’s
Table C. 8. Examples of Nanoparticles LCM’s
Nanoparticles
Generic Name /
Description Trade Name Provider Application
Iron Hydroxide NP Iron Hydroxide NP nFluids Inc. Used as a background
treatment to seal micro
fractures and wellbore
strengthening
applications.
Calcium Carbonate NP Calcium Carbonate NP nFluids Inc.
Settable/Hydratable LCM's Combinations
Generic Name /
Description Trade Name Provider Application
A combination of
swelling polymer along
with engineered
combinations of
resilient graphitic
carbon and other
materials
HYDRO-PLUG Halliburton
Used as a hydratable pill
to plug vugular,
fractured, and cavernous
formations.
Dry powdered/granular
material with synthetic
polymers, inorganic
minerals, chemical
reagents and stabilized
organic filler.
SUPER-STOP Messina Chemicals Used as a swellable pill
for sever losses.
A non-damaging, cross
linkable water soluble
polymer blended with
selected sized
cellulosic fibers.
N-SQUEEZE Halliburton Used as a settable
squeeze for sever losses.
138
APPENDIX D
D. PARTICLE SIZE DISTRIBUTION ANALYSIS
139
Figure D. 1. Particle Size Distribution for 15 ppb G # 1 Blend
Figure D. 2. Particle Size Distribution for 50 ppb G # 1 Blend
1010.1
100
90
80
70
60
50
40
30
20
10
Seive Size (mm)
% F
ine
r D 90
D 75
D 50
D 25
D 10
D90 = 1300 D75 = 800 D50 = 320 D25 = 85 D10 = 60
1010.1
100
90
80
70
60
50
40
30
20
10
Seive Size (mm)
% F
ine
r
D 10
D 25
D 50
D 75
D 90
D90 = 1300 D75 = 800 D50 = 340 D25 = 95 D10 = 60
140
Figure D. 3. Particle Size Distribution for 15 ppb NS # 1 Blend
Figure D. 4. Particle Size Distribution for 50 ppb NS # 1 Blend
1010.1
100
90
80
70
60
50
40
30
20
10
Seive Size (mm)
% F
ine
r D 90
D 75
D 50
D 25
D 10
D90 = 2000 D75 = 1600 D50 = 1000 D25 = 400 D10 = 180
1010.1
100
90
80
70
60
50
40
30
20
10
Seive Size (mm)
% F
ine
r
D 10
D 25
D 50
D 75
D 90
D90 = 2400 D75 = 1600 D50 = 1000 D25 = 400 D10 = 180
141
Figure D. 5. Particle Size Distribution for 50 ppb SCC # 3 Blend
Figure D. 6. Particle Size Distribution for 15 ppb CF # 1 Blend
1010.1
100
90
80
70
60
50
40
30
20
10
Seive Size (mm)
% F
ine
r D 90
D 75
D 50
D 25
D 10
D90 = 1200 D75 = 950 D50 = 680 D25 = 360 D10 = 250
1010.1
100
90
80
70
60
50
40
30
20
10
Seive Size (mm)
% F
ine
r
D 10
D 25
D 50
D 75
D 90
D90 = 1400 D75 = 700 D50 = 220 D25 = 140 D10 = 90
142
Figure D. 7. Particle Size Distribution for 50 ppb CF # 1 Blend
Figure D. 8. Particle Size Distribution for 30 ppb G & SCC # 1 Blend
1010.1
100
90
80
70
60
50
40
30
20
10
Seive Size (mm)
% F
ine
r D 90
D 75
D 50
D 25
D 10
D90 = 1400 D75 = 800 D50 = 220 D25 = 150 D10 = 90
1010.1
100
90
80
70
60
50
40
30
20
10
Seive Size (mm)
% F
ine
r
D 10
D 25
D 50
D 75
D 90
D90 = 1300 D75 = 900 D50 = 460 D25 = 100 D10 = 80
143
Figure D. 9. Particle Size Distribution for 80 ppb G & SCC # 1 Blend
Figure D. 10. Particle Size Distribution for 105 ppb G & SCC # 3 Blend
1010.1
100
90
80
70
60
50
40
30
20
10
Seive Size (mm)
% F
ine
r D 90
D 75
D 50
D 25
D 10
D90 = 1300 D75 = 900 D50 = 480 D25 = 120 D10 = 80
1010.1
100
90
80
70
60
50
40
30
20
10
Seive Size (mm)
% F
ine
r
D 90
D 75
D 50
D 25
D 10
D90 = 1400 D75 = 1100 D50 = 420 D25 = 90 D10 = 65
144
Figure D. 11. Particle Size Distribution for 105 ppb G & SCC # 4 Blend
Figure D. 12. Particle Size Distribution for 105 ppb G & SCC # 5 Blend
1010.1
100
90
80
70
60
50
40
30
20
10
Seive Size (mm)
% F
ine
r
D 10
D 25
D 50
D 75
D 90
D90 = 900 D75 = 700 D50 = 500 D25 = 150 D10 = 60
1010.1
100
90
80
70
60
50
40
30
20
10
Seive Size (mm)
% F
ine
r
D 90
D 75
D 50
D 25
D 10
D90 = 1400 D75 = 1200 D50 = 700 D25 = 400 D10 = 90
145
Figure D. 13. Particle Size Distribution for 105 ppb G & SCC # 6 Blend
Figure D. 14. Particle Size Distribution for 105 ppb G & SCC # 7 Blend
1010.1
100
90
80
70
60
50
40
30
20
10
Seive Size (mm)
% F
ine
r
D 10
D 25
D 50
D 75
D 90
D90 = 1400 D75 = 1250 D50 = 900 D25 = 500 D10 = 100
1010.1
100
90
80
70
60
50
40
30
20
10
Seive Size (mm)
% F
ine
r
D 90
D 75
D 50
D 25
D 10
D90 = 2600 D75 = 1900 D50 = 1300 D25 = 650 D10 = 170
146
Figure D. 15. Particle Size Distribution for 105 ppb G & SCC # 8 Blend
Figure D. 16. Particle Size Distribution for 105 ppb G & SCC # 9 Blend
1010.1
100
90
80
70
60
50
40
30
20
10
Seive Size (mm)
% F
ine
r
D 10
D 25
D 50
D 75
D 90
D90 = 2400 D75 = 1800 D50 = 1000 D25 = 250 D10 = 100
1010.1
100
90
80
70
60
50
40
30
20
10
Seive Size (mm)
% F
ine
r
D 90
D 75
D 50
D 25
D 10
D90 = 2200 D75 = 1800 D50 = 1400 D25 = 800 D10 = 300
147
Figure D. 17. Particle Size Distribution for 20 ppb G & NS # 1 Blend
Figure D. 18. Particle Size Distribution for 40 ppb G & NS # 1 Blend
1010.1
100
90
80
70
60
50
40
30
20
10
Seive Size (mm)
% F
ine
r
D 10
D 25
D 50
D 75
D 90
D90 = 1900 D75 = 1300 D50 = 500 D25 = 180 D10 = 65
1010.1
100
90
80
70
60
50
40
30
20
10
Seive Size (mm)
% F
ine
r
D 90
D 75
D 50
D 25
D 10
D90 = 2000 D75 = 1400 D50 = 580 D25 = 180 D10 = 80
148
Figure D. 19. Particle Size Distribution for 55 ppb G, SCC, & CF # 1 Blend
1010.1
100
90
80
70
60
50
40
30
20
10
Seive Size (mm)
% F
ine
r D 90
D 75
D 50
D 25
D 10
D90 = 1200 D75 = 850 D50 = 450 D25 = 100 D10 = 55
149
APPENDIX E
E. PREDICTIVE LINEAR MODEL
150
Equation E.1 gives the linear predictive model presented in subchapter 4.6.1.
𝑆𝑒𝑎𝑙𝑖𝑛𝑔 𝑃𝑟𝑒𝑠𝑠𝑢𝑟𝑒 (𝑝𝑠𝑖) = −12006.89 + 𝐹𝑙𝑢𝑖𝑑𝐶𝑜𝑒𝑓𝑓. + (122.4 × 𝜌𝐹𝑙𝑢𝑖𝑑 ) +
𝐿𝐶𝑀𝐶𝑜𝑒𝑓𝑓. + (−0.9 × 𝑊𝑐 ) + (−1.85 𝐷10) + (8.93 𝐷25) + (−7.28 𝐷50) +
(9.69 𝐷75) + (2.45 𝐷90)
Where:
Fluid Coefficient
OBM -87.5
WBM 87.5
LCM Blends Coefficient
CF # 1 2881.496
G # 1 2995.117
G, SCC, & CF # 1 4251.583
NS # 1 -3153.86
SCC # 3 2418.024
LCM Blends Coefficient
G & SCC # 1 3207.762
G & SCC # 3 1298.886
G & SCC # 4 5353.874
G & SCC # 5 364.3878
G & SCC # 6 -1011.73
G & SCC # 7 -6158.57
G & SCC # 8 -4532.94
G & SCC # 9 -5785.11
151
APPENDIX F
F. PSD MODELS COMPARISON
152
Table F. 1. Selected Subset Data for PSD Selection Criteria Development
Test No.
Sealing
Pressure
(psi)
Fracture
Width
D10
(microns)
D25
(microns)
D50
(microns)
D75
(microns)
D90
(microns)
Measured
Seal
Integrity
1 414 1000 60 85 320 800 1300 Poor
2 449 1000 60 95 340 800 1300 Poor
3 487 1000 78 100 460 900 1300 Poor
4 584 1000 80 120 480 900 1300 Good
5 224 1500 80 120 480 900 1300 Poor
6 1569 1000 65 90 420 1100 1400 Good
7 117 1000 60 150 500 700 900 Poor
8 1489 1000 90 400 700 1200 1400 Good
9 205 1500 65 90 420 1100 1400 Poor
10 0 1500 60 150 500 700 900 Poor
11 164 1500 100 500 900 1250 1400 Poor
12 1708 1500 170 650 1300 1900 2600 Good
13 1669 2000 170 650 1300 1900 2600 Good
14 110 3000 100 250 1000 1800 2400 Poor
15 0 3000 300 800 1400 1800 2200 Poor
16 589 1000 250 360 680 950 1200 Good
17 162 1500 250 360 680 950 1200 Poor
18 984 1000 180 400 1000 1600 2000 Good
19 1754 1500 180 400 1000 1600 2000 Good
20 347 2000 180 400 1000 1600 2000 Poor
21 441 2000 180 400 1000 1600 2000 Poor
22 2202 1000 180 400 1000 1600 2400 Good
23 2027 1500 180 400 1000 1600 2400 Good
24 2237 2000 180 400 1000 1600 2400 Good
25 755 2000 180 400 1000 1600 2400 Good
26 2372 1000 65 180 500 1300 1900 Good
27 805 1500 65 180 500 1300 1900 Good
28 2892 1000 80 180 580 1400 2000 Good
29 1713 1500 80 180 580 1400 2000 Good
30 295 2000 80 180 580 1400 2000 Poor
31 242 2000 80 180 580 1400 2000 Poor
153
Table F. 2. PSD Model Effectiveness Comparison
Proposed
Criterion Abrams
(1977) D90 Rule
(1996, 1998) Vickers
(2006) Halliburton
(2008)
Test
#
FW
(microns) Measured P M P M P M P M P M
1 1000 W S N W Y S N W Y W Y
2 1000 W S N S N S N W Y W Y
3 1000 W S N S N S N W Y W Y
4 1000 S S Y S Y S Y W N W N
5 1500 W W Y W Y W Y W Y W Y
6 1000 S S Y S Y S Y W N W N
7 1000 W W Y S N W Y W Y W Y
8 1000 S S Y S Y S Y W N W N
9 1500 W W Y W Y W Y W Y W Y
10 1500 W W Y S N W Y W Y W Y
11 1500 W W Y S N W Y W Y W Y
12 1500 S S Y S Y S Y W N W N
13 2000 S S Y S Y S Y W N W N
14 3000 W W Y S N W Y W Y W Y
15 3000 W W Y S N W Y W Y W Y
16 1000 S S Y S Y S Y W N W N
17 1500 W W Y S N W Y S N W Y
18 1000 S S Y S Y S Y W N S Y
19 1500 S S Y S Y S Y W N W N
20 2000 W W Y S N S N W Y W Y
21 2000 W W Y S N S N W Y W Y
22 1000 S S Y S Y S Y W N S Y
23 1500 S S Y S Y S Y W N W N
24 2000 S S Y S Y S Y W N W N
25 2000 S S Y S Y S Y W N W N
26 1000 S S Y S Y S Y W N W N
27 1500 S S Y S Y S Y W N W N
28 1000 S S Y S Y S Y W N W N
29 1500 S S Y S Y S Y W N W N
30 2000 W W Y W Y S N W Y W Y
31 2000 W W Y W Y S N W Y W Y
Overall Match % 90% 68% 77% 45% 55%
Note: Fracture width (FW) Predicted (P) Match (M) Strong (S) Weak (W) Yes (Y) No (N)
154
APPENDIX G
G. PUBLICATION FROM THIS DISSERTATION
155
PUBLISHED PAPERS
Alsaba, M., Nygaard, R., Hareland, G., and Contreras, O. 2014a. “Review of Lost
Circulation Materials and Treatments with an Updated Classification”. AADE-14-
FTCE-25, AADE Fluids Technical Conference and Exhibition, Houston, USA, 15-16
April.
Alsaba, M., Nygaard, R., Saasen, A., and Nes, O. 2014b. “Lost Circulation Materials
Capability of Sealing Wide Fractures”. SPE 170285, SPE Deepwater Drilling and
Completions Conference, Galveston, TX, USA, 10–11 September.
Alsaba, M., Nygaard, R., Saasen, A., and Nes, O. 2014c. “Laboratory Evaluation of
Sealing Wide Fractures Using Conventional Lost Circulation Materials”. SPE
170576, SPE Annual Technical Conference and Exhibition, Amsterdam, the
Netherlands, 27–29 October.
Contreras, O., Hareland, G., Husein, M., Nygaard, R., and Alsaba, M. 2014b.
“Experimental Investigation on Wellbore Strengthening In Shales By Means of
Nanoparticle-Based Drilling Fluids”. SPE 170589, SPE Annual Technical
Conference and Exhibition, Amsterdam, the Netherlands, 27–29 October.
PAPERS IN REVIEW
Alsaba, M., Nygaard, R., Saasen, A., and Nes, O. “Experimental Investigation of Fracture
Width Limitations of Granular Lost Circulation Treatments”. Journal of Petroleum
Exploration and Production Technology. (Submitted)
Contreras, O., Hareland, G., Husein, M., Nygaard, R., and Alsaba, M. “Experimental
Investigation on Wellbore Strengthening In Shales By Means of Nanoparticle-Based
Drilling Fluids”. SPE Journal. (Submitted)
156
BIBLIOGRAPHY
Aadnoy, B. S. and Belayneh, M. 2004. Elasto-Plastic Fracturing Model for Wellbore
Stability Using Non-Penetrating Fluids. Journal of Petroleum Science and
Engineering, Vol. 45 (3), pp. 179-192.
Aadnoy, B. S., and Chenevert, M. E. 1987. Stability of Highly Inclined Boreholes. SPE
Drilling Engineering, Vol. 2 (4), pp. 364 374.
Abrams, A. 1977. Mud Design to Minimize Rock Impairment Due to Particle Invasion.
Journal of Petroleum Technology, Vol. 29 (5), pp.586-592.
Alberty, M. W. and McLean, M. R. 2004. A Physical Model for Stress Cages. SPE-
90493-MS, SPE Annual Technical Conference and Exhibition, Houston, USA, 26-29
September.
Alleman, J., Owen, B., Cornay, F., and Weintritt, D. J. 1998. Multi-functional Solid
Lubricant Reduces Friction/Prevents Mud Loss. World Oil, Vol. 219 (9), pp. 87-90.
Anchor Drilling Fluids USA, Inc. Products Data Sheets,
http://intranet.anchorusa.com/msds-sheets. Accessed 24 January, 2014.
Anderson, T. L. 1995. Fracture Mechanics: Fundamentals and Applications. Second
edition, CRC Publications, United States.
API Recommended Practice 13B, 1974. Standard Procedure for Testing Drilling Fluid.
American Petroleum Institute, Washington, D.C., Section 10.
Aston, M. S., Alberty, M. W., McLean, M. R., Jong, H., and Armagost, K. 2004. Drilling
Fluids for Wellbore Strengthening. SPE-87130-MS, IADC/SPE Drilling Conference,
Dallas, Texas, 2-4 March.
Baker Hughes. 2010. Specialty Products: Drilling Fluids Solution,
http://www.bakerhughes.com/news-and-media/resources/brochures/specialty-
products-brochure. Accessed 24 January, 2014.
Ballard, A., and Massam, J. 2008. Precipitated Weighting Agents for use in Wellbore
Fluids. WO2008033838A2, 20 March.
Bruton, J. R., Ivan, C. D., and Heinz, T. J. 2001. Lost Circulation Control: Evolving
Techniques and Strategies to Reduce Downhole Mud Losses. SPE-67735-MS,
SPE/IADC Drilling Conference, Amsterdam, the Netherlands, 27 February -1 March.
Canson, B. E. 1985. Lost Circulation Treatment for Naturally Fractured, Vugular, or
Cavernous Formations. SPE/IADC 13440, SPE/IADC Drilling Conference, New
Orleans, USA, 5-8 March.
157
Carbonell, R., and Detournay, E. 1995. Modeling Fracture Initiation and Propagation
from a Pressurized Hole: A Dislocation-Based Approach. ARMA-95-0465, the 35th
U.S. Symposium on Rock Mechanics (USRMS), Reno, Nevada, 5-7 June.
Caughron, D., Renfrow, D., Bruton, J., Ivan, C., Broussard, P., Bratton, T., and
Standifird, W. 2002. Unique Crosslinking Pill in Tandem with Fracture Prediction
Model Cures Circulation Losses in Deepwater Gulf of Mexico. SPE-74518-MS,
IADC/SPE Drilling Conference, Dallas, Texas, USA, 26-28 February.
Charlez, P. A. 1999. The Concept of Mud Weight Window Applied to Complex Drilling.
SPE-56758-MS, SPE Annual Technical Conference and Exhibition, Houston, TX, 3-
6 October.
Chenevert, M., and Sharma, M. 2009. Maintaining Shale Stability by Pore Plugging. US
Patent 0314549. 24 December.
Chevron Phillips Chemical Company. 2014. “Lost Circulation Guide”,
http://www.cpchem.com/bl/drilling/en-us/Documents/LostCirculationGuide.pdf.
Accessed 24 January, 2014.
Chevron Phillips Chemical Company. 2014. Products Data Sheets,
http://www.cpchem.com/bl/drilling/en-us/Pages/Products.aspx. Accessed 24 January,
2014.
Collins, N., Whitfill, D., Kharitonov, A., and Miller, M. 2010. Comprehensive Approach
to Severe Loss Circulation Problems in Russia. SPE-135704-MS, SPE Russian Oil
and Gas Conference and Exhibition, Moscow, Russia, 26-28 October.
Contreras, O., Hareland, G., Husein, M., Nygaard, R., and Alsaba, M. 2014a.
Application of In-House Prepared Nanoparticles as Filtration Control Additive to
Reduce Formation Damage. SPE-168116-MS, SPE International Symposium and
Exhibition on Formation Damage Control, Lafayette, USA, 26-28 February.
Contreras, O., Hareland, G., Husein, M., Nygaard, R., and Alsaba, M. 2014b.
Experimental Investigation on Wellbore Strengthening In Shales By Means of
Nanoparticle-Based Drilling Fluids. SPE-170589-MS, SPE Annual Technical
Conference and Exhibition, Amsterdam, the Netherlands, 27–29 October.
Cutler, R., Enniss, D., Jones, A., and Swanson, S.1985. Fracture Conductivity
Comparison of Ceramic Proppants. Society of Petroleum Engineers Journal. Vol. 25
(2), pp. 157-170.
Deeg, W., and Wang, H. 2004. Changing Borehole Geometry and Lost-Circulation
Control. ARMA-04-577, the 6th North America Rock Mechanics Symposium,
Houston, Texas, June 5-9, 2004.
158
Dick, M. A., Heinz, T. J., Svoboda, C. F., and Aston, M. 2000. Optimizing the Selection
of Bridging Particles for Reservoir Drilling Fluids. SPE-58793-MS, SPE
International Symposium on Formation Damage Control, Lafayette, Louisiana, 23-24
February.
Dudley, J., Fehler, D. F., and Zeilinger, S. 2001. Minimizing Lost Circulation Problems
with Synthetic Muds, GPRI Project 2000 DC 3, December.
Dupriest, F. E. 2005. Fracture Closure Stress (FCS) and Lost Returns Practices. SPE-
92192-MS, SPE/IADC Drilling Conference, Amsterdam, Netherlands, 23-25
February.
Dupriest, F. E., Smith, M. V., Zeilinger, C. S. and Shoykhet, I. N. 2008. Method to
Eliminate Lost Returns and Build Integrity Continuously with High-Filtration-Rate
Fluid. SPE-112656-MS, SPE/IADC Drilling Conference, Orlando, Florida, USA, 4-6
March.
Erdogan, F. and Gupta, G. D. 1972. On the Numerical Solution of Singular Integral
Equations. Quarterly of Applied Mathematics, Vol. 29, pp. 525–534.
Fann Instrument Company (2011) Particle Plugging Apparatus Product Information.
http://www.fann.com/public1/pubsdata/Brochures/PPA_Paper.pdf. Accessed 3
September 2015.
Fann Instrument Company (2015) Linear Swell Meter 2100 Instruction Manual.
http://www.fann.com/public1/pubsdata/Manuals/LSM_2100_Manual_102114531.pdf
. Accessed 31 July 2015.
Fidan, E., Babadagli, T., and Kuru, E. 2004. Use of Cement as Lost-Circulation Material:
Best Practices. PETSOC-2004-090, Canadian International Petroleum Conference,
Calgary, Alberta, 8-10 June.
Filippov, A., Mack, R., Cook, L., York, P., Ring, L., and McCoy, T. 1999. Expandable
Tubular Solutions. SPE-56500-MS, SPE Annual Technical Conference and
Exhibition, Houston, Texas, USA, 3-6 October.
Fjaer, E., Holt, R., Raaen, A., Risnes, R., and Horsud, P. 2008. Petroleum Related Rock
Mechanics, 2nd Edition. Elsevier, Oxford, UK.
Friedheim, J. E., Young, S., De Stefano, G. Lee, J., and Guo, Q. 2012. Nanotechnology
for Oilfield Applications – Hype or Reality?. SPE-157032-MS, SPE International
Oilfield Nanotechnology Conference and Exhibition, Noordwijk, Netherlands, 12-14
June.
Fuh, G. F., Morita, N., Byod, P. A., and McGoffin, S. J. 1992. A New Approach to
Preventing Lost Circulation While Drilling. SPE-24599-MS, SPE Annual Technical
Conference and Exhibition, Washington D.C., USA, 4-7 October.
159
GEO Drilling Fluids, Inc. 2010. Product Descriptions,
http://www.geodf.com/index.cfm/fuseaction/Pages.Page/id/302. Accessed 24
January, 2014.
Griffith, A. A., 1920. The Phenomena of Rupture and Flow In Solids. Philosophical
Transactions of the Royal Society of London, A Vol. 221 (4), pp. 163–198.
Griffo, A., and Keshavan, M. 2007. High Performance Rock Bit Greased. US Patent
20070254817 A1. November.
Halliburton. 2014. Products Data Sheet, http://www.halliburton.com/. Accessed 24
January, 2014.
Hands, N., Kowbel, K., Maikranz, S., and Nouris, R. 1998. Drill-in Fluid Reduces
Formation Damage, Increases Production Rates. Oil & Gas Journal, Vol. 96 (28), pp.
65-69.
He, W., and Stephens, M. P. 2011. Bridging Particle Size Distribution in Drilling Fluid
and Formation Damage. SPE-143497-MS, SPE European Formation Damage
Conference, Noordwijk, The Netherlands, 7-10 June.
Hettema, M., Horsrud, P., Taugbol, K., Friedheim, J., Huynh, H., Sanders, M. W., and
Young, S. 2007. Development of an Innovative High-Pressure Testing Device for the
Evaluation of Drilling Fluid Systems and Drilling Fluid Additives within Fractured
Permeable Zones. OMC-2007-082, Offshore Mediterranean Conference and
Exhibition, Ravenna, Italy, 28-30 March.
Hillerborg, A., Modeer, M., and Petersson, P. 1976. Analysis of Crack Formation and
Crack Growth in Concrete by Means of Fracture Mechanics and Finite Elements.
Division of Building Materials Lund Institute of Technology, Lund, Vol. 6, pp. 773–
782, Pergamum, Sweden.
Hinkebein, T., Behr, V., and Steve L. Wilde. 1983. Static Slot Testing of Conventional
Lost Circulation Materials, SAND82-1080, January.
Hoelscher, K. P., De Stefano, G., Riley, M. and Young, S. 2012. Application of
Nanotechnology in Drilling Fluids. SPE-157031-MS, SPE International Oilfield
Nanotechnology Conference and Exhibition, Noordwijk, Netherlands, 12-14 June.
Howard, G. C., and Scott, P. P. 1951. An Analysis and the Control of Lost Circulation.
Journal of Petroleum Technology, Vol. 3 (6), pp. 171-182.
Irwin, G. R., and Kies J. A. 1954. Critical Energy Rate Analysis of Fracture Strength.
Welding Journal, Vol. 33, pp. 193-198.
Jahn, F., Cook, M., and Graham, M. 2008. Hydrocarbon Exploration & Production, 2nd
Edition. Elsevier Science, Oxford, UK.
160
Kaageson-Loe, N. M., Sanders, M. W., Growcock, F., Taugbol, K., Horsrud, P.,
Singelstad, A. V., and Omland, T. H. 2009. Particulate-Based Loss-Prevention
Material - The Secrets of Fracture Sealing Revealed!. SPE Drilling & Completion,
Vol. 24(4), pp. 581-589.
Karimi, M., Petrie, S. W., Moellendick, E., and Holt, C. 2011. A Review of Casing
Drilling Advantages to Reduce Lost Circulation, Augment Wellbore Strengthening,
Improve Wellbore Stability, and Mitigate Drilling-induced Formation Damage. SPE-
148564-MS, SPE/IADC Middle East Drilling Technology Conference and
Exhibition, Muscat, Oman, 24-26 October.
Kelsey, J. R. 1981. Geothermal Drilling and Completion Technology Development
Program Quarterly Progress Report. SAND81-1020, June.
Kumar, A., and Savari, S. 2011. Lost Circulation Control and Wellbore Strengthening:
Looking Beyond Particle Size Distribution. AADE-11 NTCE-21, AADE National
Technical Conference and Exhibition, Houston, Texas, USA, 12-14 April.
Kumar, A., Savari, S., and Jamison, D. E., and Whitfill, D. 2011. Application of Fiber
Laden Pill for Controlling Lost Circulation in Natural Fractures. AADE-11 NTCE-
19, AADE National Technical Conference and Exhibition, Houston, Texas, USA, 12-
14 April.
Lecolier, E., Herzhaft, B., Rousseau, L., Neau, L., Quillien, B., and Kieffer, J. 2005.
Development of a Nanocomposite Gel for Lost Circulation Treatment. SPE-94686-
MS, SPE European Formation Damage Conference, Sheveningen, The Netherlands,
25-27 May.
Lohoefer, C. L., Mathis, B., Brisco, D., Waddell, K., Ring, L., and York, P. 2000.
Expandable Liner Hanger Provides Cost-Effective Alternative Solution. SPE-59151-
MS, IADC/SPE Drilling Conference, New Orleans, Louisiana, USA, 23-25
February.
Marinescu, Pavel. 2014. “Loss Circulation Challenges – Current Approach and What are
the Next Solutions?”. SPE Lost Circulation Workshop, Dubai, 20-22 May.
Mata, F., and Veiga, M. 2004. Crosslinked Cements Solve Lost Circulation Problems.
SPE-90496-MS, SPE Annual Technical Conference and Exhibition, Houston, Texas,
USA, 26-29 September.
Messenger, J. U. 1981. Lost Circulation. PennWell Corp, Tulsa, Oklahoma.
Messenger, J. U., and McNiel, J. S. 1952. Lost Circulation Corrective: Time-Setting Clay
Cement. Journal of Petroleum Technology, Vol. 4 (3), pp. 59-64.
Messina Chemicals. 2012. Products Data Sheet,
http://www.messinachemicals.com/index.php/products/drilling-chemicals/lost-
circulation-seepage-control-materials. Accessed 24 January, 2014.
161
Montgomery, D. 2001. Design and Analysis of Experiments, 5th Edition. John Wiley &
Sons, Inc., New York, NY.
Morita, N., and Fuh, G-F. 2012. Parametric Analysis of Wellbore-Strengthening Methods
from Basic Rock Mechanics. SPE Drilling & Completion, Vol. 27(2), pp. 315-327.
Morita, N., Black, A. D., and Fuh, G-F. 1990. Theory of Lost Circulation Pressure. SPE-
20409-MS, SPE Annual Technical Conference and Exhibition, New Orleans,
Louisiana, USA, 23-26 September.
Morita, N., Fuh, G-F., and Black, A. D. 1996a. Borehole Breakdown Pressure with
Drilling Fluids-I. Empirical Results. International Journal of Rock Mechanics and
Mining Sciences & Geomechanics, Vol. 33 (1), pp. 39-51.
Morita, N., Fuh, G-F., and Black, A. D. 1996b. Borehole Breakdown Pressure with
Drilling Fluids-II. Semi-Analytical Solution to Predict Borehole Breakdown Pressure.
International Journal of Rock Mechanics and Mining Sciences & Geomechanics, Vol.
33 (1), pp. 53-69.
Mostafavi, V., Hareland, G., Belayneh, M. and Aadnoy, B. S. 2011. Experimental and
Mechanistic Modeling of Fracture Sealing Resistance with Respect to Fluid and
Fracture Properties. ARMA 11-98, the 45th US Rock Mechanics Symposium, San
Francisco, USA, 26-29 June.
Nayberg, T. M. 1987. Laboratory Study of Lost Circulation Materials for use in Both Oil-
base and Water-base Drilling Muds. SPE Drilling Engineering, Vol. 2 (3), pp. 229-
236.
Newman, J. C. 1971. An Improved Method of Collocation for the Stress Analysis of
Cracked Plates with Various Shaped Boundaries. NASA Technical Notes. TN D-
6376. National Aeronautics and Space Administration. Washington D.C.
Newpark Drilling Fluids. 2012. Products Data Sheet,
http://www.newpark.com/capabilities/newpark-drilling-fluids/united-states/about-
us/product-bulletins. Accessed 24 January, 2014.
Nwaoji, C.O., Hareland, G., Husein, M., Nygaard, R., Zakaria, M.F. 2013. Wellbore
Strengthening-Nano-Particle Drilling Fluid Experimental Design Using Hydraulic
Fracture Apparatus. SPE-163434-MS, SPE/IADC Drilling Conference and
Exhibition, Amsterdam, the Netherlands, 5 – 7 March.
Nygaard, R. and Salehi, S. 2011. A Critical Review of Wellbore Strengthening: Physical
Model and Field Deployment. AADE-11-NTCE-24, AADE National Technical
Conference and Exhibition, Houston, USA, 12-14 April
Onyia, E. C. 1994. Experimental Data Analysis of Lost-Circulation Problems during
Drilling with Oil-Based Mud. SPE Drilling and Completion, Vol. 9 (1), pp. 25-31.
162
Paiaman, M., and Al-Anazi, D. 2008. Using Nanoparticles to Decrease Differential Pipe
Sticking and its feasibility in Iranian Oil Field. Oil and Gas Business.
Perez-Roca, E., Andrews, S., and Keel, D. 2003. Addressing Common Drilling
Challenges Using Solid Expandable Tubular Technology. SPE-80446-MS, SPE Asia
Pacific Oil and Gas Conference and Exhibition, Jakarta, Indonesia, 9-11 September.
Powers, M. C. 1953. A New Roundness Scale for Sedimentary Particles. Journal
Sedimentary Research, Vol 23 (2), pp. 117-119.
Rehm, B., Haghshenas, A., Paknejad A. S., Al-Yami, A., and Hughes, J., 2012.
Underbalanced Drilling: Limits and Extremes. Gulf Publishing Company, Houston,
TX.
Rong, G., Liu, G., Hou, D., and Zhou, C. 2013. Effect of Particle Shape on Mechanical
Behaviors of Rocks: A Numerical Study Using Clumped Particle Model. The
Scientific World Journal, Vol. 2013.
Saasen, A., Godøy, R., Breivik, D.H., Solvang, S.A., Svindland, A., Gausel, E., and
Frøyland, K. 2004. Concentrated Solid Suspension as an Alternative to Cements for
Temporary Abandonment Applications in Oil Wells. SPE-SA-34, SPE Technical
Symposium of Saudi Arabia Section, Dhahran, Saudi Arabia, 15-17 May.
Saasen, A., Wold, S., Ribesen, B.T., Tran, T.N., Huse, A., Rygg, V., Grannes, I., and
Svindland, A. 2011. Permanent Abandonment of a North Sea Well Using
Unconsolidated Well Plugging Material. SPE Drilling & Completion, Vol. 26 (3), pp.
371-375.
Salehi, S. 2012. Numerical Simulations of Fracture Propagation and Sealing: Implication
for Wellbore Strengthening. PhD dissertation, Department of Geological Sciences
and Engineering, Missouri University of Science and Technology.
Salehi, S. and Nygaard, R. 2011. Evaluation of New Drilling Approach for Widening
Operational Window: Implications for Wellbore Strengthening. SPE-140753-MS,
SPE Production and Operations Symposium, Oklahoma City, USA, 27-29 March.
Salehi, S. and Nygaard, R. 2012. Numerical Modeling of Induced Fracture Propagation:
A Novel Approach for Lost Circulation Materials (LCM) Design in Borehole
Strengthening Applications of Deep Offshore Drilling. SPE-135155-MS, SPE Annual
Technical Conference and Exhibition, San Antonio, USA, 8-10 October.
Salehi, S. and Nygaard, R. 2015. Developing a Full Fluid-Solid Finite-Element Model to
Simulate near Wellbore Fractures for Wellbore Strengthening Applications. ASME
Journal of Energy Resources Technology, Vol. 137 (1).
Sall, J. 1990. Leverage Plots for General Linear Hypotheses. The American Statistician,
Vol. 44 (4), pp. 308-315.
163
Savari, S., Kumar, A., Whitfill, D. L., Miller, M., Murphy, R. J., and Dale E. Jamison.
2013. Engineered LCM Design Yields Novel Activating Material for Potential
Application in Severe Lost Circulation Scenarios. SPE-164748-MS, SPE North
Africa Technical Conference and Exhibition, Cairo, Egypt, 15-17 April.
Savari, S., Whitfill, D. l., and Kumar, A. 2012. Resilient Lost Circulation (LCM): A
Significant Factor in Effective Wellbore Strengthening. SPE-153154-MS, SPE
Deepwater Drilling and Completions Conference, Galveston, Texas, USA, 20-21
June.
Schlumberger. 2013. Products Data Sheet,
http://www.slb.com/services/miswaco/services/drilling_fluid/df_products/lost_circula
tion_material.aspx. Accessed 24 January, 2014.
Schlumberger. 2014. Oil Field Glossary, http://www.glossary.oilfield.slb.com. Accessed
24 January, 2014.
Sharp-Rock Technologies, Inc. 2014. Product Data Sheet, www.sharp-rock.com.
Accessed 2 February, 2014.
Shepard, S. F., Reiley, R. H., and Warren, T. M. 2002. Casing Drilling: An Emerging
Technology. SPE Drilling & Completion, Vol. 17 (1), pp. 4-14.
Shinohara, K., Oida, M., and Golman, B. 2000. Effect of Particle Shape on Angle of
Internal Friction by Triaxial Compression Test. Powder Technology, Vol. 107 (2), pp.
131-136.
Smith, P. S., Browne, S. V., Heinz, T. J., and Wise, W. V. 1996. Drilling Fluid Design to
Prevent Formation Damage in High Permeability Quartz Arenite Sandstones. SPE-
36430-MS, SPE Annual Technical Conference and Exhibition, Denver, Colorado, 6-9
October.
Sweatman, R. E., and Scoggins, W. C. 1988. Acid-Soluble Magnesia Cement: New
Applications in Completion and Workover Operations. SPE-18031-MS, SPE Annual
Technical Conference and Exhibition, Houston, USA, 2-5 October.
Tada, H., Paris, P., Irwin, G. 2000. The Stress Analysis of Cracks Handbook, 3rd
Edition.
ASME Press, New York.
Tehrani, A., Cliffe, A., Hodder, M. H., Young, S., Lee, J., Stark, J., and Seale, S. 2014.
Alternative Drilling Fluid Weighting Agents: A Comprehensive Study on Ilmenite
and Hematite. SPE-167937-MS, IADC/SPE Drilling Conference and Exhibition, Fort
Worth, Texas, USA, 4-6 March
Tehrani, A., Friedheim, J., Cameron, J., and Reid, B. 2007. Designing Fluids for
Wellbore Strengthening – Is It an Art?. AADE-07-NTCE-75, AADE National
Technical Conference and Exhibition, Houston, Texas, USA, April 10-12.
164
The Drilling Manual, Fifth Edition, 2015. CRC Press, Boca Raton, FL.
Van Oort, E., Friedheim, J., Pierce, T., and Lee, J. 2011. Avoiding Losses in Depleted
and Weak Zones by Constantly Strengthening Wellbores. SPE Drilling &
Completion, Vol. 26 (4), pp. 519-530.
Vickers, S., Cowie, M., Jones, T., and Twynam, A. J. 2006. A New Methodology that
Surpasses Current Bridging Theories to Efficiently Seal a Varied Pore Throat
Distribution as Found in Natural Reservoir Formations. AADE-006-DF-HO-16.
AADE Fluids Conference and Exhibition, Houston, Texas, USA, 11-12 April.
Wang H, Sweatman R, Engelman B, Deeg W, Whitfill D, Soliman M, Towler BF. 2008.
Best Practice in Understanding and Managing Lost Circulation Challenges. SPE
Drilling & Completion, Vol. 23 (2), pp. 168-175.
Wang, H. 2011. Is It Really Possible to Efficiently Form A Strong Seal inside
Subterranean Openings without Knowing Their Shape and Size?. AADE-11-NTCE-
25, AADE National Technical Conference and Exhibition, Houston, USA, 12-14
April.
Warren, T., Tessari, R., and Houtchens, B. 2004. Casing Drilling with Retrievable
Drilling Assemblies. OTC-16564-MS, Offshore Technology Conference, Houston,
Texas, USA, 3-6 May.
Warren, W. E. 1982. The Quasi-Static Stress Field around a Fractured Wellbore.
International Journal of Fracture, Vol. 18 (2), pp. 113–124.
Weideman, B. 2014. Investigation of Cased Wellbore Integrity in the Wabamun Area
Sequestration Project. MSc Thesis, Department of Geological Sciences and
Engineering, Missouri University of Science and Technology.
White, R. J. 1956. Lost-Circulation Materials and Their Evaluation. API-56-352.
Drilling and Production Practice, New York, New York, USA, 1 January.
Whitfill, D. 2008. Lost Circulation Material Selection, Particle Size Distribution and
Fracture Modeling with Fracture Simulation Software. SPE-115039-MS, IADC/SPE
Asia Pacific Drilling Technology Conference and Exhibition, Jakarta, Indonesia, 25-
27 August.
Zakaria, M. 2013. Nanoparticle-Based Drilling Fluids with Improved Characteristics.
Ph.D. Dissertation. Department of Chemical and Petroleum Engineering. University
of Calgary, Calgary, AB, Canada.
Zhang, K., Chanpura, R. A., Mondal, S., Wu, C-H., Sharma, M. M., Ayoub, J. A., and
Parlar, M. 2015. Particle-Size-Distribution Measurement Techniques and Their
Relevance or Irrelevance to Wire-Wrap-Standalone-Screen Selection for Gradual-
Formation-Failure Conditions. SPE Drilling & Completion, Vol. 30 (2), pp. 164-174.
165
VITA
Mortadha Alsaba was born the 17th
of April 1985, in Dhahran, Saudi Arabia. He
received a BEng. Degree in Petroleum Engineering with first class honours from London
South Bank University, London, UK, in 2007. He also received a MSc. in Petroleum
Engineering from Missouri University of Science and Technology, Rolla, MO, in 2012
and a PhD in Petroleum Engineering from Missouri University of Science and
Technology in 2015. He has published and presented several papers on lost circulation
prevention and mitigation.
Alsaba is a member of several professional associations, including Society of
Petroleum Engineering (SPE), American Association of Drilling Engineers (AADE),
American Association of Petroleum Geologist (AAPG) and the American Society of
Mechanical Engineers (ASME). He has served as a secretary for the AADE student
chapter from 2013-2014.
His research interests are in prevention and mitigation of lost circulation events,
laboratory evaluation of wellbore strengthening techniques, drilling fluids design and
optimization.