Ali Taghichian PDF

UNIVERSITY OF OKLAHOMA
GRADUATE COLLEGE

ON THE GEOMECHANICAL OPTIMIZATION OF HYDRAULIC FRACTURING
IN UNCONVENTIONAL SHALES

A THESIS

SUBMITTED TO THE GRADUATE FACULTY

in partial fulfillment of the requirements for the

Degree of

MASTER OF SCIENCE

By

ALI TAGHICHIAN
Norman, Oklahoma
2013

ON THE GEOMECHANICAL OPTIMIZATION OF HYDRAULIC FRACTURING
IN UNCONVENTIONAL SHALES

A THESIS APPROVED FOR THE
MEWBOURNE SCHOOL OF PETROLEUM AND GEOLOGICAL ENGINEERING

BY

______________________________
Dr. Musharraf Zaman, Chair

______________________________
Dr. Ahmad Ghassemi

______________________________
Dr. Deepak Devegowda

Copyright by ALI TAGHICHIAN 2013
All Rights Reserved.
iv
Acknowledgements
First of all, I would like to express my deep gratitude and sincere thanks to my helpful and
kind advisor, Dr. Musharraf Zaman, because of all his help and support, key advices, and
his special style of encouragement; without these I would never be able to complete this
research work.
I am deeply grateful of my parents and my dear siblings for all their help, support, and
inspirations regarding my education at each level.
I also want to express my appreciation to Mr. Timothy L. Beard, Manager-ETG
operations, at Chesapeake Energy Corporation, Dr. Ali Daneshy, Head of Daneshy
Consultant International, for allocating their valuable time to me, advising me regarding
completion of unconventional resources, and being so patient and helpful.
I am indeed thankful of Dr. Arul Britto, Emeritus Faculty at University of Cambridge, Dr.
Jean-Claude Roegiers, Emeritus Faculty at University of Oklahoma, Dr. Ahmad Ghassemi
and Dr. Deepak Devegowda, faculty members of Mewbourne School of Petroleum and
Geological Engineering, University of Oklahoma for their kindness and sympathetic helps
and advices.
My gratitude also goes to Mr. Amin Mousavi, PhD candidate of Petroleum Geomechanics,
Tarbiat Modares University and Dr. Nicolas Roussel, Reservoir Engineer at ConocoPhillips
for their useful discussions.
Final thanks are also given to Oklahoma Department of Transportation (ODOT) and
Oklahoma Transportation Center because of their financial support during the course of
this study.
v
Table of Contents
Acknowledgements .......................................................................................................................... iv
Table of Contents .............................................................................................................................. v
List of Tables ................................................................................................................................... viii
List of Figures ................................................................................................................................... ix
Abstract .............................................................................................................................................. xi
Chapter 1: Introduction .................................................................................................................... 1
1.1. Why unconventional resources ................................................................................... 1
1.2. Challenges and solutions in unconventional resources ........................................... 1
1.3. Critical questions in optimization of unconventional reservoirs ........................... 2
1.4. Approaches of fracing optimization in unconventional resources ........................ 3
1.4.1. Production approach ............................................................................................. 3
1.4.2. Geomechanics approach ....................................................................................... 3
1.5. Methods in geomechanical optimization of fracturing design ............................... 4
1.6. Strategies to control stress shadow and stress intensity factor ............................... 5
1.7. Contents of this thesis .................................................................................................. 6
Chapter 2: Literature Survey ............................................................................................................ 8
2.1. General ........................................................................................................................... 8
2.2. History of fracture mechanics ..................................................................................... 8
2.3. History of stress shadow around hydraulic fractures ............................................. 10
2.4. Methods of calculating stress field around cracks .................................................. 11
2.4.1. Stress field in the vicinity of crack tip ............................................................... 12
2.4.2. Stress field for the entire medium ...................................................................... 15
2.5. Calculation of stress shadow around hydraulic fractures ...................................... 17
2.5.1. Analytical determination of stress shadow size ............................................... 17
2.5.2. Numerical determination of stress shadow size .............................................. 18
2.6. Analytical calculation of crakc aperture ................................................................... 19
2.7. Need for a comprehensive study for optimization of fracing design .................. 20
Chapter 3: Methodology and Verification ................................................................................... 21
3.1. General ......................................................................................................................... 21
3.2. Significance of numerical methods for this problem ............................................. 21
3.3. Different scenarios for calculation of stress shadow size ..................................... 22
3.4. Major assumptions and data range in stress shadow analysis ............................... 24
3.4.1. Hydraulic fracture geometry ............................................................................... 25
3.4.2. Different boundary conditions for shadow analysis ....................................... 26
3.4.3. Assumptions ......................................................................................................... 26
3.4.4. Data range ............................................................................................................. 27
3.5. Shadow mechanisms ................................................................................................... 28
3.6. Method of studying stress intensity factor .............................................................. 29
3.6.1. Parameters related to SIF change ...................................................................... 29
3.6.2. Monitoring of SIF change along the fracture edge by a single value ............ 31
3.7. Different scenarios for calculation of SIF change.................................................. 33
3.8. Data range for the SIF change analysis .................................................................... 35
3.9. Verification of stress field for simple problems ..................................................... 35
Chapter 4: Prediction of shadow and aperture of hydraulic fractures ..................................... 38
4.1. General ......................................................................................................................... 38
vi
4.2. Prediction of aperture for a contained hydraulic fracture ..................................... 38
4.2.1. Effect of in-situ stresses on aperture ................................................................. 38
4.2.2. Mathematical equation for prediction of aperture of a single contained
fracture ................................................................................................................... 39
4.2.3. Reliability of the proposed function for aperture prediction ......................... 40
4.2.4. Visualization of aperture prediction function .................................................. 41
4.3. Prediction of shadow around a contained hydraulic fracture (T = S) ............ 42
4.3.1. Numerical results for stress shadow analysis ................................................... 42
4.3.2. Mathematical equation predicting stress shadow of a single contained
fracture ................................................................................................................... 43
4.3.3. Coefficients of the proposed function for stress shadow prediction ........... 44
4.3.4. Reliability of the proposed function for stress shadow prediction ............... 46
4.3.5. Visualization of stress shadow prediction function ........................................ 47
4.3.6. Discussion on the relationship between aperture and stress shadow ........... 48
4.4. Prediction of shadow around a contained hydraulic fracture (T > S) ............ 48
4.4.1. Mathematical equation for prediction of stress shadow with different
threshold angles .................................................................................................... 49
4.4.2. Reliability of the equation predicting stress shadow with different threshold
angles ...................................................................................................................... 51
4.5. Effect of uncontainment of the fracture (standalone well fracturing) ................ 52
4.5.1. Change of stress shadow by uncontainment of the fracture ......................... 53
4.5.2. Change of aperture by uncontainment of the fracture ................................... 54
4.5.3. Mathematical equations predicting aperture and shadow size in uncontained
single fractures ...................................................................................................... 54
4.5.4. Reliability of the proposed functions for uncontainment multiplier
determination ........................................................................................................ 56
4.5.5. Use of uncontainment multipliers (standalone fractures) .............................. 57
4.6. Effect of simultaneous fracturing on shadow size and aperture .......................... 57
4.6.1. Change of stress shadow size by simultaneous fracturing (0I = S) .......... 57
4.6.2. Mathematical equation for prediction of simultaneous multiplier ................ 59
4.6.3. Reliability of the proposed equation for simultaneous multiplier prediction
60
4.6.4. Change of stress shadow size by simultaneous fracturing (0I = 8u) ........ 61
4.7. Effective distance between simultaneous fractures of two parallel wells ........... 61
4.8. Prediction of shadow change for uncontained simultaneous fractures .............. 62
4.8.1. Mathematical equations for uncontainment multiplier in simultaneous
fracturing ............................................................................................................... 63
4.8.2. Reliability of uncontainment multiplier for simultaneous fracturing ............ 64
4.8.3. Use of uncontainment multipliers (simultaneous fractures) .......................... 64
Chapter 5: Prediction of Propagation potential in hydraulic fractures .................................... 66
5.1. General ......................................................................................................................... 66
5.2. Effect of aspect ratio of the fractures on the SIF (single fracture) ...................... 66
5.3. Effect of multistage fracturing on the SIF and aperture in standalone wells ..... 67
5.3.1. Qualitative description of multistage fracturing influence on SIF/aperture
of each fracture ..................................................................................................... 68
5.3.2. Mathematical equation for prediction of SIF and aperture change in
multistage fracturing ............................................................................................ 69
vii
5.3.3. Reliability of the proposed equation for SIF and aperture change in
multistage fracturing ............................................................................................ 70
5.3.4. Visualization of the proposed equation ............................................................ 71
5.4. Effect of simultaneous fracing on the SIF of single-stage fractures ................... 72
5.4.1. Qualitative description of SIF change via simultaneous fracturing .............. 72
5.4.2. Mathematical equation for prediction of SIF change in simultaneous
fracturing (single stage) ........................................................................................ 73
5.4.3. Reliability of the proposed function .................................................................. 74
5.5. Effect of simultaneous multistage fracturing on the SIF of the fractures .......... 75
5.5.1. Qualitative description of the SIF change by simultaneous multistage
hydraulic fracturing .............................................................................................. 75
5.5.2. Mathematical equation for prediction of SIF change in multistage
simultaneous fracturing ....................................................................................... 77
5.5.3. Reliability of the proposed equation .................................................................. 78
5.6. Effect of fracture offset on the SIF change in parallel wells ................................ 79
5.7. Use of the SIF change prediction functions .................................................................. 82
Chapter 6: Work flow for optimization of hydrofracing ........................................................... 83
6.1. General ......................................................................................................................... 83
6.2. Successive procedure of hydraulic fracturing optimization .................................. 83
6.2.1. Step one ................................................................................................................. 83
6.2.2. Step two ................................................................................................................. 84
6.2.3. Step three ............................................................................................................... 85
6.2.4. Step four ................................................................................................................ 85
Chapter 7: Concluding remarks and recommendations ............................................................ 86
7.1. Aperture analysis ......................................................................................................... 86
7.2. Stress shadow analysis ................................................................................................ 87
7.3. Analysis of propagation potential ............................................................................. 87
7.4. Optimized fracture network ...................................................................................... 88
7.5. Four steps in optimization of hydraulic fracturing ................................................ 89
References ......................................................................................................................................... 91
Appendix: Variable definition ........................................................................................................ 96

viii
List of Tables
Table 3. 1. Input variable range for the numerical simulation .................................................. 28
Table 3. 2. Input variable range for the numerical simulation .................................................. 35

Table 4. 1. Coefficients of the function in Eq. (4.7) ................................................................... 45
Table 4. 2. Shadow decrease by threshold angle ......................................................................... 50
Table 4. 4. Coefficients for Eqs. (4.12) and (4.13) for standalone fracturing ......................... 56
Table 4. 5. Coefficient of the function in Eq. (4.14) .................................................................. 60
Table 4. 6. Stress shadow change by simultaneous fracturing (0I = 8u) ............................ 61
Table 4. 7. Coefficients for Eqs. (4.12) and (4.13) for simultaneous fracturing ..................... 63

ix
List of Figures
Figure 2. 1. Configuration of stress field around a 2-D fracture ................................................ 9
Figure 2. 2. Stress field around a center crack with internal pressure in a 2-D medium....... 16

Figure 3. 1 Different scenarios for studying of stress shadow and aperture .......................... 22
Figure 3. 2. Geometry of a contained hydraulic fracture and stress shadow plane ............... 25
Figure 3. 3. Different stress shadow mechanisms by different hydraulic pressures .............. 29
Figure 3. 4. Fracture geometry together with stress change along its edges ........................... 31
Figure 3. 5. Stress variation along the fracture edges ................................................................. 33
Figure 3. 6. Different fracturing techniques/patterns in this study ......................................... 34
Figure 3. 7. Stress validation in the direction vertical away from the fracture center ........... 36

Figure 4. 1. Half-aperture prediction using Eqs. (4.1-4.3) ......................................................... 40
Figure 4. 2. Visualization of aperture determination function .................................................. 41
Figure 4. 3. Shadow change with Poissons ratio, numerical values and the predictions...... 43
Figure 4. 4. Shadow function coefficient change by Poissons ratio ........................................ 44
Figure 4. 5. Behavior of function coefficients with stress anisotropy (Kb) ............................ 46
Figure 4. 6. Shadow size prediction using Eqs. (4.4-4.7) ........................................................... 47
Figure 4. 7. Shadow around a hydraulic fracture ........................................................................ 47
Figure 4. 8. Distribution of SH_S
80
over SH_S
05
ratio (for all aspect ratios) .......................... 49
Figure 4. 9. Shadow difference with respect to varied threshold angles ................................. 50
Figure 4. 10. Shadow ratio prediction (threshold angle of 80) ................................................ 52
Figure 4. 11. Stress shadow change by fracture uncontainment ............................................... 53
Figure 4. 12. Aperture change as a result fracture uncontainment ........................................... 54
Figure 4. 13. Prediction of aperture and shadow size change by fracture uncontainment ... 56
x
Figure 4. 14. Shadow reduction by simultaneous fracturing ..................................................... 58
Figure 4. 15. Prediction of shadow size change by simultaneous hydraulic fracturing ......... 60
Figure 4. 16. Effective normalized distance between fracture tips .......................................... 62
Figure 4. 17. Prediction of shadow and aperture change by fracture uncontainment .......... 64

Figure 5. 1. SIF change along fracture edges by different aspect ratios .................................. 67
Figure 5. 2. SIF change along the edges of the fracture by multistage fracturing .................. 68
Figure 5. 3. Prediction of SIF and aperture change by multistage fractures ........................... 71
Figure 5. 4. Prediction of SIF and aperture change by multistage fracturing ......................... 72
Figure 5. 5. SIF change by simultaneous fracturing ................................................................... 73
Figure 5. 6. Precision of the predicted SIF ratios using the proposed Eq. (5.2) .................... 74
Figure 5. 7. Effect of multistage simultaneous fracturing on the SIF of fractures ................ 76
Figure 5. 8. Prediction of SIF ratio of simultaneous multistage fracturing ............................. 79
Figure 5. 9. Effect of distance and offset on the SIF of the fractures ..................................... 80
Figure 5. 10. Effect of fracture offset on SIF change ................................................................ 80
Figure 5. 11. SIF maximum line for different aspect ratios....................................................... 81

xi
Abstract
Hydraulic fracturing of unconventional resources is a way to effectively connect these
reservoirs to the wellbore and to increase their permeability. In order to have an efficient
hydraulic fracturing job, optimization of fracing design should be performed to get the best
production with the least possible cost. Since multistage simultaneous fracing is the method
by which unconventional resources are hydraulically fractured, a systematic strategy in
obtaining the optimal spacing/distance between the multistage/simultaneous fractures is of
paramount importance. From a production perspective, using a fluid flow simulator,
fracture geometry, spacing between fractures, and fracturing pattern can be determined
based on monitoring of the production. In this method, no significant change in
production by reducing the fracture spacing means that fractures with a lower spacing
would not be economical. Porosity, permeability, and fracture conductivity play key roles in
the determination of fracturing geometry/pattern. From geomechanics perspective, on the
other hand, stress shadow analysis and stress intensity factor are used to ensure straight
fractures with no deviation or collapse, and with the highest propagation potential in the
target zone. In this thesis, the two geomechanical aspects are studied in detail. The most
influencing variables on each of these strategies are included in the analysis and a
comprehensive set of equations are proposed for determination of stress shadow size and
propagation potential of the fractures. In addition, since aperture plays a decisive role in
determination of proppant type/size and conductivity of the fractures, this important
parameter is also quantified considering its own influencing variables. Finally, some
strategies are proposed for optimization of hydraulic fracturing for unconventional shales.
1
Chapter 1: Introduction
1.1. Why unconventional resources
Unconventional reservoirs have received more attention in recent years because of the
declining productivity of conventional reservoirs. Deployment of multilateral drilling
technologies together with simultaneous multistage hydraulic fracturing of parallel laterals
have also been much influencing in this regard. In 2009, almost 50% of domestic U.S. gas
production was attributed to unconventional resources and it is estimated to increase to
almost 75% by 2035 (EIA, 2011).
1.2. Challenges and solutions in unconventional resources
This growth in the development of unconventional reservoirs, primarily in shale gas/oil
reservoirs, has, however, encountered several challenges. The permeability of these shale
reservoirs typically ranges from a few to about one hundred nano-darcies. Consequently,
productivity from traditional well architecture and completion schemes was limited and
uneconomical. In early 2000s though, several key technological developments enabled the
unlocking of this vast resource potential (Andrew et al., 2009).
The issues surrounding connectivity to the productive regions of the reservoir was largely
mitigated by developments leading to maturation in horizontal and multilateral well drilling
and completion. Of course, the real growth in the development occurred as a result of a
combination of unconventional well architecture and massive hydraulic fracture treatments,
thereby connecting the horizontal laterals to increasingly larger volumes of the reservoir
(Warpinski et al., 1997). Eventually, single-stage fracture treatments evolved to multistage
stimulation treatments and fracing of standalone well was progressed to simultaneous
2
fracing of multilateral wells in order to increase reserves per well, enhance well
productivity, and improve project economics (King et al., 2008). Currently, industry is
practicing for horizontal drilling orientation along the direction of the minimum horizontal
stress. The aim of this practice is to create fractures oriented perpendicular to the axis of
the wellbore (Dasseault, 2011). This is because transverse hydraulic fractures provide better
coverage of the reservoir than do longitudinal fractures, and are more efficient in
producing gas/oil formations (Soliman et al., 1996). It was proposed that as the formation
permeability gets smaller, it would expectedly take a longer time to deplete the reservoir,
and it may be necessary to create more fractures to quickly deplete the reservoir (Soliman
et al., 1997).
1.3. Critical questions in optimization of unconventional reservoirs
According to the above-mentioned progresses in the field of hydraulic fracturing in
unconventional reservoirs, the critical questions would be How to design the fracing job
in such reservoirs? What is the optimized distance between hydraulic fractures? What
strategies should be used in order to determine fracturing pattern? What are the most
influencing variables in this regard? What is the share of each variable in defining fracturing
pattern in unconventional reservoirs? Finally and most importantly, is it possible to
quantify the effect of each variable to be used when designing fracturing treatment for such
reservoirs?
Any acceptable answer to any of these questions can be really helpful in defining fracture
treatment in unconventional reservoirs. In fact, addressing satisfactory responses for all the
above questions, one can do the fracturing design in an optimized way with minimum cost
and maximum production.
3
1.4. Approaches of fracing optimization in unconventional resources
There are two different approaches, by which fracing optimization is conducted; the first
one is production perspective, and the other is geomechanics perspective.
1.4.1. Production approach
Some researchers have studied the fracturing pattern only considering the distance between
fractures by which the highest production is obtained by the least cost. This means that
using reservoir simulators, they were able to start from a base fracture spacing, reduce that
spacing between fractures, and monitor production change. Negligible change in
production in this method means that placing closer fractures is not economical and not
contributing to production (see e.g., Yu and Sepehrnoori, 2013).
Many researchers, however, mentioned that accurate geomechanical information about the
rock and its variation through the shale is also important since stresses predominantly
control fracture initiation and development (Abousleiman, 2007; Barree, 2009; Britt, 2009;
Xu, 2009).
1.4.2. Geomechanics approach
The beginning step for this purpose is defining the term optimized fracture pattern by
which we can find solutions to each of the critical questions in Section 1.3. An optimized
fracture pattern should connect the most productive areas of the reservoir to the wellbore.
As multistage fractures perpendicular to the wellbore in horizontal laterals have been
proven to be the best general pattern in this field, the key point here is finding the
optimized spacing between fractures to have maximum production, with minimum cost. In
addition, propagation potential of hydraulic fractures should be considered which means by
4
having different fracturing patterns, how fracturing pressure of the reservoir is influenced
and whether it is required to apply more/less pressure to frac the rock. This means that the
energy required for fracturing of the reservoir is also an important matter which should be
considered as well.
According to these explanations, we can define the optimized fracture pattern as parallel
fractures, perpendicular to the wellbore axis, having the highest propagation potential in
the target formation, creating the highest production, with an optimized distance to
prevent any deviation/collapse of them.
Therefore, from a geomechanical perspective, controlling of fracture direction and its
propagation potential are of paramount importance in optimization of fracing design in
unconventional shales. This is because by determining these two factors, questions of
fracing design in unconventional shales can be addressed in a proper and systematic way.
Therefore, it is intended in this thesis to investigate designation of fracture pattern from a
geomechanical point of view.
1.5. Methods in geomechanical optimization of fracturing design
Controlling of fracture path is dominated by a phenomenon called stress shadow effect which
is a disturbed zone around any hydraulic fracture in which direction of principal stresses is
changed. Any other fracture in this region or fractures with overlapping stress shadow
zones will result deviation or collapse
1
(Fisher et al., 2004). Hence, this concept is used in
this study for determination of optimized distances by which fracture direction is
controlled in unconventional reservoirs.

1
Stress shadow causes one fracture to collapse on the adjacent fracture and further growth is observed only
for one fracture.
5
Propagation potential of hydraulic fractures in the target zone, on the other hand, is
dominantly controlled by stress intensity factor and of course, toughness of the fractures.
Stress intensity factor (SIF) is an indicative of the stress field around the crack tip and its
corresponding strength variable in this regard is known as fracture toughness. As soon as
stress intensity factor reaches the fracture toughness, propagation occurs (Yew, 1997).
Thus, any increase/decrease in stress intensity factor directly results propagation to happen
with lower/higher hydraulic pressures, respectively. As a result, any change in stress
intensity factor is directly related to the amount of energy required for fracturing of the
formation and should be considered in different fracturing patterns as the concept to
examine fracturing potential change.
1.6. Strategies to control stress shadow and stress intensity factor
According to the fact that reducing stress shadow around hydraulic fractures without or at
least with the minimum decrease in propagation potential in the target zone brings the
opportunity to have more closely-spaced, non-deviating fractures, some technological
considerations have been reported in the literature in this regard. As stated previously,
simultaneous hydraulic fracturing is one of these methods performed in multi-lateral
horizontal wells. It is believed that the spacing between hydraulic fractures can be reduced
by utilizing this strategy. Mutalik and Gibson (2008) showed that simultaneously fractured
wells have 21-100% enhancement of initial production rates over the standalone horizontal
wells. King (2010) mentioned simultaneous fracturing as one effective strategy which
significantly raises the reservoir face-contact fractures. Waters et al. (2009) indicated
simultaneous fracturing as a powerful tool to reduce the spacing between fractures in
horizontal wells. Rafiee et al. (2012) studied simultaneous hydraulic fracturing of parallel
6
horizontal wells and modified the conventional approach of zipper frac to a more efficient
pattern called modified zipper frac. In this modified method, an alternating approach is
used for fracturing of parallel wells where far field path of hydraulic fracture is controlled
in a more acceptable way. Roussel and Sharma (2011b) suggested alternating fracturing
strategy to minimize the fracture spacing. In this method, stress shadow around outer
hydraulic fractures is used for making the middle fracture not to deviate. Of course, change
of propagation potential was not considered in their study.
Considering the most influencing variables in the size of stress shadow zone and stress
intensity factor of hydraulic fractures, aperture can also be predicted which is really useful
in term of proppant type/size and fracture conductivity determination.
Therefore, having the optimized distance between hydraulic fractures with no deviation or
collapse, having the propagation potential change in the acceptable range/direction, and
aperture of the fractures at hand, fracing crew can have an acceptable, optimized, and
desired fracturing design for an unconventional reservoir.
1.7. Contents of this thesis
In this thesis, it is tried to propose a work flow for optimization of hydraulic fracturing in
unconventional shales from a geomechanical point of view. Straight fractures without any
deviation/collapse, with the highest propagation potential in the target zone are the aim in
geomechanical optimization. This is done by utilizing two stress concepts called stress
shadow and stress intensity factor. In Chapter 2, the theoretical bases for these two
concepts are presented and the most important studies in these two fields regarding
optimization of hydraulic fracturing are also reported. In Chapter 3, the methodology of
7
determining stress shadow and stress intensity factor is presented. The most influencing
variables, the range of these variables, different boundary conditions, and different
mechanisms affecting the study are presented. Moreover, the need for investigating the
problem using numerical simulation is also reported and the used numerical technique is
verified by means of simulating simple crack problems and comparing the numerical with
analytical results. In Chapter 4, stress shadow and aperture of hydraulic fractures is studied
and two comprehensive sets of equations for estimation of stress shadow and aperture of
hydraulic fractures is proposed. In Chapter 5, stress intensity factor is investigated in a
comprehensive way and a complete set of equations for estimation of propagation
potential of hydraulic fractures having different geometries/patterns is proposed. In
Chapter 6, summing up the ideas regarding optimization of hydraulic fracturing in shales,
from production and geomechanics perspectives, a workflow is proposed for optimization
of hydraulic fracturing in shales. In Chapter 7, the summary and concluding remarks of
each step of the work is discussed and propositions are made in term of a systematic
method for optimization of hydraulic fracturing for this type of reservoirs.

8
Chapter 2: Literature Survey
2.1. General
In Chapter 1, the major factors in determining an optimized fracturing job in
unconventional reservoirs were introduced from a geomechanical point of view as stress
intensity factor and stress shadow effect. According to the fact that these two factors are
both indicative of stress field around hydraulic fractures, in this chapter, stress field around
hydraulic fractures is discussed in detail.
2.2. History of fracture mechanics
The birth point of obtaining stress field around cavities goes back to the work done by
Kirsch (1989), who succeeded to obtain stress field around a circular cavity in an infinite
two-dimensional (2-D) field. His closed form solution is still used for borehole stability
problems in elastic region. In this procedure, theory of elasticity together with complex
variable functions was utilized to obtain the stress function satisfying all the boundary
conditions. Then, Inglis (1913) obtained stress field around an elliptical cavity and proved
that by decreasing the curvature radius of the two ends of the ellipse, the increase in stress
concentration can be expressed by the following equation:
o
mux
o
= 1 +2
_
c
p
(2.1)
in which o is the major axis of the ellipse and p is the radius of curvature at the ellipse
sharp corners. As it is seen from Eq. (2.1), by tending the curvature of the corners to zero,
9
stress concentration approaches to infinity. This point was an eye-opening conclusion
which made engineers aware of the stress concentration at crack tips. Following this work,
a new field of mechanics was borne called fracture mechanics, in which stress field
around different fractures with different patterns and geometries were to be studied and
their propagation was to be pondered. Figure 2.1 shows an open fracture in a 2-D medium
with stress field around it.

Figure 2. 1. Configuration of stress field around a 2-D fracture
As it can be seen from Figure 2.1, a crack possessing two edges, have two half-lengths
called c. The plate having this fracture inside can be applied by far-field tensile stress
(o; perpendicular to the crack plane) or internal pressure (P
H
). By having any of these two
stress/pressure boundary conditions; a stress field is created around the crack which can be
obtained.

r2
r
1
r
X
Y
yy
xx
xy
yy
xx
Pu
c c
10
2.3. History of stress shadow around hydraulic fractures
Many researchers mentioned the importance of stress shadow around hydraulic fractures.
Fisher et al. (2004) demonstrated that creation of a hydraulic fracture generates a zone of
altered local stresses that may impact the orientation of subsequent fractures in a
phenomenon known as the stress shadowing effect. Cheng (2009) utilized boundary
element method for 2-D geomechanical modeling of hydraulic fractures and indicated that
the number and spacing of the fractures need to be carefully selected considering stress
change in order to create effective fractures with appropriate geometries. Wong et al.
(2013) studied the interaction between adjacent hydraulic fractures using analytical and
numerical methods in two dimensions. They observed the divergence of hydraulic fractures
outward or even collapse of inside fractures on the outside ones as a result of stress
shadow effect. Singh and Miskimins (2010) indicated that an increase in spacing between
the fractures induced less interference, and hence requires less breakdown pressure to
initiate a fracture.
Waters et al. (2009) also stated that shadow following a hydraulic fracture treatment
involves the creation of a localized region of high compressive stresses perpendicular to the
fracture face in the vicinity of the fracture center. This causes the direction of maximum
stress to be reoriented in the region of the stress shadow. By locating the next treatment in
this region, fracture growth is likely to deviate or even occur parallel to the borehole axis
and consequently, necessitates optimizing fracture spacing to obtain the maximum number
of fractures oriented perpendicular to the wellbore (see also Roussel and Sharma, 2011a;
Morrill and Miskimins, 2012).
11
In general, closely spaced hydraulic fractures lead to an increase in the stress in the
minimum stress direction with values higher than the original maximum horizontal stress.
Succeeding fractures, therefore, may tend to propagate in undesirable directions parallel to
the wellbore axis, thereby reducing the efficacy of the fracture treatments and
compromising well productivity.
Stress shadow effect is a useful concept not only for determining the spacing between
transverse hydraulic fractures but also for fracture mapping and distance between
tectonically created fractures. Daneshy et al. (2012) were able to observe the shadow effect
from readings of pressure gauges installed in observation wells and used that information
for determination of fracture orientation and extent, spacing between wells, and optimum
number and spacing between fracture stages. Fischer et al. (1995) utilized the concept of
stress shadow for prediction of distance between pressurized fractures in sedimentary
rocks.
It is evident from the above-mentioned reviews that it is important to study stress shadow
of hydraulic fractures for optimization purpose.
2.4. Methods of calculating stress field around cracks
There are two different ways of determining the stress field around hydraulic fractures. In
the first method, it is assumed that radius of stress calculation point in polar system is
much smaller than the half-length of the fracture (r
1
c; see Figure 2.1). Therefore, some
terms are neglected in the stress function, and a simple form of a complex variable
function is solved. In this method, stress field is obtained for only a small region around
the crack tip. According to the fact that propagation of a fracture is the main focus in
12
fracture mechanics and it mainly depends on stress field at the crack tip, this method
suffices for propagation potential of hydraulic fractures. In the second method, on the
other hand, no term is neglected from stress function and it is directly/numerically solved
for the entire field. This method is useful for determination of stress shadow around
hydraulic fractures.
2.4.1. Stress field in the vicinity of crack tip
Several researchers proposed the first solution for such problem as a 2-D crack embedded
in an infinite medium under far-field tensile stress or internal pressure. Among them,
works of Muskhelishvili (1933), Westergaard (1939), Sneddon and Elliot (1945), Paris
(1965), and Eshelby (1968) can be considered pioneering. They proposed that stress field
just around the crack tip can be obtained using Eq. (2.2), the notation of which
corresponds to the right tip (r
1
, 0
1
) in Figure 2.1.
_
o
x
o
x
_ =
K
I
2nr
1
cos _
0
1
2
]
`
1
1
1
11 -sin
0
1
2
sin_
S0
1
2
]
1 +sin
0
1
2
sin_
S0
1
2
]
sin
0
1
2
cos _
S0
1
2
]
1
1
1
1
1
(2.2)
in which K
I
is the stress intensity factor (SIF) of the crack for mode one fracturing
(opening). It is worth indicating that both of our loading boundary conditions (far-field
tensile stress or internal pressure) belong to mode one loading. In addition, the SIF of a
crack is the same for the cases of far-field tensile stress and internal pressure with the same
13
magnitude (superposition principle; see Janssen et al., 2009). The generalized form of stress
field around the crack tip can be written as:
o
]
= _
K
I
2nr
]
]
(0) (2.3)
It is important to notice that the trigonometric function,
]
(0), is always the same for one
stress component. It is evident from Eqs. (2.2) and (2.3) that having different geometries of
cracks under mode one loading, only the SIF is changing and all the remaining terms are
kept unchanged. Therefore, having the SIF of each crack geometry, one can have stress
field around the crack tip. As an example, the SIF for this particular problem for far-field
stress is given by Eq. (2.4a) and for internal pressure it is calculated using Eq. (2.4b):
K
I
= o2nc
(2.4a)
K
I
= P
H
2nc
(2.4b)
It is evident from Eqs. (2.4) that in case of equal magnitudes for far-field tensile stress and
internal pressure, the resultant SIF is the same. Moreover, it is seen that in the elastic
region, stress field is proportional to stress/pressure, square root of crack size, and
inversely related to the root distance from the tip.
Many 2-D crack geometries have been analytically solved and their SIFs have been
reported in the literature (e.g., Tada et al., 1973). However, there have also been some
problems in which the geometry of cracks were challenging and stress field for these
fractures were cumbersome to be analytically solved. SIF of such problems were defined by
14
utilizing Boundary Element (BE) and Finite Element (FE) methods (see e.g., Sih, 1973;
Tada et al., 1973; Murakami, 1987).
Regarding three-dimensional (3-D) crack problems, determination of stress distribution
near a flat crack, embedded in an infinitely extended homogeneous, isotropic solid, opened
up due to prescribed internal pressure, analytical solutions have been proposed by a
number of investigators for circular and elliptical geometries. Some of these solutions are
discussed by Keer (1964), Willis (1968), Sneddon and Lowengrub (1969), Shah and
Kobayashi (1971), and Guidera and Lardner (1975). Using the 2-D Fourier transform
method, Kassir (1981) succeeded in solving the SIFs around a rectangular crack. Moreover,
Mastrojannis et al. (1978) developed a method for determination of SIF of a general-
shaped crack with internal pressure in an infinite medium utilizing numerical integration.
By increasing of the internal pressure, SIF increases and reaches a critical value, called
fracture toughness, in which fracture tends to propagate and crack is elongated. It is worth
mentioning that different modes of fracturing have different fracture toughnesses.
For Mode I fracture toughness there exist ISRM (International Society of Rock Mechanics)
suggested methods to determine the fracture toughness (Ouchterlony, 1988; Fowell, 1995).
Several other methods have also been proposed in the literature (e.g., Evans, 1972; Barker,
1977; Atkinson et al., 1982; Chong and Kuruppu, 1984; Sun and Ouchterlony, 1986; Guo
et al., 1993; Chang et al., 2002, and Whittaker et al., 1992).
Based on the fact that SIF of fractures in three-dimensions is very challenging to be
analytically determined and in most of cases it is required to apply numerical integration,
15
therefore, researchers are using numerical techniques (e.g., FE and BE). As an example, in
hydraulic fracturing of unconventional shales with multistage simultaneous fracturing
strategy, it is important to consider the interaction of adjacent and meeting fractures on the
SIF. Numerical modeling is indeed a useful device in this regard because multi-driven
fractures in a medium may influence the SIF of each other and may change the
propagation potential. It is important to mention that propagation potential of a hydraulic
fracture is a relative term defined by the ratio of SIFs of multiple fractures over the SIF of
a single fracture. Thus, it gives a dimensionless number larger/smaller than unity because
of interacting crack tips effect.
2.4.2. Stress field for the entire medium
As explained in the previous section, Eqs. (2.2-2.4) only give stress field around the tip not
for the whole medium. This can be explained by considering a plate with a crack in the
center with far-field stress; if we put r
1
in Eqs. (2.2), it is evident that all the stresses
tend to zero far away from the tip, which is not the case. Therefore, in order to judge the
disturbed stress zone or stress shadow size, a complete formulation for the whole field is
required to have all the components of stress at every point in the medium. In the world of
linear elastic fracture mechanics (LEFM), this is not an easy task to determine the stress
field for the whole medium, specifically for 3-D crack problems. Sneddon and Elliot (1945)
proposed Eq. (2.5) that describes stress field for the whole medium of a 2-D plate with an
internally pressurized crack in the center.
16
1
2
(o
+ o
x
) = P
H
_
r
r
1
1
2
r
2
1
2
cos _0 -
1
2
0
1
-
1
2
0
2
] - 1_ (2.5a)
1
2
(o
- o
x
) = P
H
r sin 0
c
_
c
2
r
1
r
2
_
3
2
sin
S
2
(0
1
+ 0
2
)
(2.5b)
x
= P
H
r sin0
c
_
c
2
r
1
r
2
_
3
2
cos
S
2
(0
1
+ 0
2
)
(2.5c)
It is worth mentioning that the notation in Eqs. (2.5) corresponds to the notation in Figure
2.1. It is seen from Eqs. (2.5) that the stress field determined by six coordinate variables,
namely (r, 0), (r
1
, 0
1
), (r
2
, 0
2
), is very challenging when such determination involves. The
stress field (o
xx
, o
,
x
) for this 2-D medium (e.g., E=30 GPa; : = u.S) with an
imbedded center crack which is internally pressurized (e.g. P
H
= 1uu Po) is presented as
Figure 2.2.
Figure 2. 2. Stress field around a center crack with internal pressure in a 2-D
medium
It is seen from Figure 2.2 that stress concentration exists for all the components of stress at
crack tips. Shear stress is zero along the crack lips and on a line perpendicular to the crack
17
faces, passing through the crack center. This phenomenon happens because of the
symmetry in the model, which does not let the shear stress to exist on these symmetry
lines. The final point about Figure 2.2 is the contour shape of vertical stress (o
) around
the crack tips, which is a peanut-shaped inclined away from the crack tip. The plotted
contour shapes in Figure 2.2 exist for all the fracture geometries and can be used for
checking of the contour shapes in numerical simulations.
2.5. Calculation of stress shadow around hydraulic fractures
There are two different ways for determining the stress shadow size around hydraulic
fractures. In the first method, which is limited to some special cases of fracture geometries,
analytical stress field is used for determination of shadow size. In the second method,
numerical simulation is used for determination of stress field around hydraulic fractures
with any geometry and boundary conditions. Then, stress shadow is calculated from the
stress field. In the following sections both of these methods are presented.
2.5.1. Analytical determination of stress shadow size
Jo (2012) proposed an analytical method for prediction of stress shadow based on
analytical stress distribution around plane strain and penny-shaped fractures (see Green
and Sneddon, 1950). This method can only be used for plane strain and penny-shaped
cracks. Equations (2.6) show the proposed equations for predicting shadow size around
these two limiting cases of hydraulic fractures.
I
E
=
`
1
1
1
1
1
1
_
v
2(S -2v)

foi plane - stiain fiactuies (infinite length)
_
(1 +v)
4(S -v)
foi penny - shapeu fiactuies (AR = unity)

(2.6)
18
in which, I is the distance from the center of a hydraulic fracture to a point where
maximum stress contrast between o
xx
and o
is observed, v is the Poissons ratio, and E

is height of the hydraulic fracture. From Eq. (2.6) it is seen that the only influencing
variables in the shadow size are the Poissons ratio and fracture height. In addition, Eq.
(2.6) accounts for the aspect ratio but it only addresses stress shadow size for two limiting
cases of plane strain and penny-shaped cracks. Effect of internal pressure, stress
anisotropy, and aspect ratios of the hydraulic fracture, however, have also been found to
influence stress shadow size (Roussel and Sharma, 2011a,b; Morrill and Miskimins, 2012).
2.5.2. Numerical determination of stress shadow size
In this method, having the total stress field from numerical simulation, the change in
magnitude and direction of the principal stresses can be obtained from Eq. (2.7) and Eq.
(2.8), respectively.
S
22,33
=
o
xx
+o
2
_
_
_
o
xx
+o
2
]
2
+
x
2

(2.7)
tan20
p
=
2
x
|o
xx
-o
|
(2.8)
in which o
xx
, o
,
x
are induced stresses in the shadow region, S
22
and S
33
are changed
principal stresses in this region, and 0
p
is the direction of principal stresses. Deviation
angle of principal stresses (0
) can also be obtained based on the following criteria:

19
0
= _
0
p
o
xx
> o
9u -0
p
o
xx
< o
4S o
xx
= o

(2.9)
It is worth mentioning that maximum principal stress deviation, depending on the
magnitude of internal pressure, can be near the crack tip or crack center and it vanishes
away from the crack. Therefore, by defining a threshold angle (0
1
) the area around the
crack with stress deviations more than this threshold angle can be determined. Considering
Eq.(2.8), it is also evident that maximizing the stress contrast means lowering the deviation
angle. Using this numerical method, there is no limitation on the fracture geometry and
boundary condition. Thus, a wide range of problems can be solved satisfactorily using this
approach.
2.6. Analytical calculation of crakc aperture
Sneddon and Elliot (1946) also documented the analytical solution for crack opening
displacement field. This analytic expression for half-aperture of a 2-D hydraulic fracture
was derived as:
w(z)
c
=
w
mux
c
_
1 -[
z
c
2
;
w
mux
c
=
2(1 -v
2
)P
H
E
n

(2.10)
in which v is the Poissons ratio, E
n
is the Youngs modulus of net play, P
H
is the
hydraulic pressure applied inside the fracture, c is the half length of the fracture, w(z) is
the half aperture, w
mux
is the maximum half-aperture (located at the center of the fracture
(z = u), and z is defined on the z axis originated at crack center in the direction of fracture
20
height. From Eq. (2.10), we observe that the displacement of crack tips (at z = c) is zero
and displacement of the edges increases from the tips to the center of the fracture where
maximum displacement occurs. The equation also highlights the inverse relationship of
fracture aperture with Youngs modulus and Poissons ratio of the rock. Furthermore, the
equation describes an elliptical shape for the aperture of the fracture. It is important to
note that for a 3-D hydraulic fracture, maximum aperture is located at half-height and half-
length of it. So, an equation similar to Eq. (2.9) is written in x direction, along the length of
the hydraulic fracture. Therefore, having the maximum aperture of the fracture, one can
use Eq. (2.9) and a similar one for x direction for determination of aperture distribution on
the entire crack face.
2.7. Need for a comprehensive study for optimization of fracing design
According to the above-mentioned studies in the field of stress shadow and propagation
potential of hydraulic fractures, it is realized that most of these studies have been
performed in two dimensions or are more or less descriptive. Therefore, it can be seen that
there is a lack of comprehensive 3-D studies of stress shadow with propagation potential.
In this thesis, we plan to comprehensively study the stress shadow size and propagation
potential of hydraulic fractures together. In the following chapter, the methodology of
solving the present problem together with verification of the method is presented.

21
Chapter 3: Methodology and Verification
3.1. General
In Chapter 2, basic concepts of stress distribution around hydraulic fractures with
particular attention to propagation potential and stress shadow effect in the target medium
were explained using a 2-D infinite medium with an internally pressurized fracture in the
middle. In addition, a clear perspective from which optimization of hydraulic fracture
should be seen, was introduced by explaining the two concepts of stress intensity factor
(SIF) and stress shadow effect. In this chapter, the method by which we are going to
investigate the problem of optimization of hydraulic fractures in unconventional shale
reservoirs is introduced. The presented method is verified by solving a number of single
fractures in 2-D/3-D media with simple geometries and making a comparison between the
obtained results and analytical solutions reported in the literature for these problems.
3.2. Significance of numerical methods for this problem
Obtaining the stress field around a fracture with its own geometry and boundary
conditions, in a 3-D medium is a challenging matter. Although there are some analytical
solutions available in the literature for some cracks with simple geometries, but finally, a
numerical integral should be solved to obtain the stress field for these problems. As a
result, because of the limited flexibility in analytical solutions, a numerical method is
utilized in this thesis to study the problem of optimization of hydraulic fractures. Finite
Element Method (FEM) which is a powerful tool in solving governing equations of
problems in solid/fluid or coupled solid-fluid mechanics, is chosen for this purpose.
22
ABAQUS CAE 6.12, which is a general purpose FEM-based software package, is selected
for the modeling purposes. The procedure of modeling is designed in such a way to first
give us the opportunity to obtain stress shadow size which is indicative of hydraulic
fracture distance, based on different influencing variables. Then, propagation potential of
these hydraulic fractures is studied in detail to examine which geometry/pattern has what
effect on the propagation potential of the fractures. Finally, according to the constructed
numerical models, aperture of the hydraulic fractures which is highly important in the
selection of proppant type and size and determination of fracture conductivity is estimated
and its change with the defined variables is studied.
3.3. Different scenarios for calculation of stress shadow size
Four different scenarios are defined in this section, based on which stress shadow and
aperture analysis of hydraulic fractures are performed (Figure 3.1).
Figure 3. 1 Different scenarios for studying of stress shadow and aperture
(pictures show fracture plane)
Single Containeu BF
stanualone well
Simultaneous Containeu BF
Paiallel wells
Single 0ncontaineu BF
stanualone well
Simultaneous 0ncontaineu BF
Paiallel wells
S
1 2
4
23
In the first scenario (No. 1 in Figure 3.1), a 3-D hydraulic fracture, with different aspect
ratios is contained in a medium with different stress regimes, moduli, and net pressures. In
Chapter 2, theoretical background of stress shadow was presented and it was shown that
assuming different threshold angles, different shadow sizes are obtained (see Chapter 2,
Section 2.5.1). According to the fact that in the literature (see Roussel and Sharma, 2011),
region of 90 reorientation of maximum horizontal stress is assumed as the region of
minimum shadow effect, hence, it is tried to predict the shadow with all the threshold
angles.
In the second scenario (No. 2 in Figure 3.1), it is assumed that the hydraulic fracture under
consideration is not contained. Since shale net plays are located in between bounding layers
with different moduli from the net play, it is important to consider any fracture penetration
inside the bounding layers. This is done by assuming different Youngs modulus between
the net play and bounding layers, and different penetration extent to the bounding layers.
In the third and fourth scenarios (Nos. 3 and 4 in Figure 3.1), stress shadow change as a
result of simultaneous hydraulic fracturing of laterals is studied in detail. Effect of distance
between hydraulic fracture tips in parallel wells is investigated on the shadow size. In this
part of the work, effective length between fracture tips is also obtained for fractures with
different aspect ratios. It is worth mentioning that effective distance is a distance between
fracture tips beyond which no shadow size change is observed. In addition, effect of
penetration of the simultaneous hydraulic fractures inside the bounding layers (uncontained
fractures) is studied and the shadow change is investigated.
24
Using the above mentioned scenarios, stress shadow and aperture of a hydraulic fracture
can be predicted satisfactorily. First, a set of equations is proposed for a single contained
hydraulic fracture (Scenario 1). After that, in Scenario 2, effect of bounding layers for the
hydraulic fracture is presented as a ratio to be multiplied with the result of the first scenario
to result the value for an uncontained hydraulic fracture. Likewise, in the third scenario, a
ratio is obtained which is the effect of simultaneous fracturing for contained fractures. This
ratio should also be multiplied with the result of the first scenario to provide the result for
the case of simultaneous contained fractures. Finally, for the fourth scenario, effect of
uncontainment of simultaneous fractures is obtained as a ratio to be multiplied with the
result of the third scenario to give the results for simultaneous uncontained fractures.
As an example, in case one is interested in obtaining the shadow size of uncontained
simultaneous fractures between two parallel wells, it is first required to have the shadow for
the basic case; contained hydraulic fracture in a standalone well (Scenario 1). Then, effect
of simultaneous fracturing of parallel laterals is obtained for a contained fracture (Scenario
3) as a ratio to be multiplied with the result of the basic case. Finally, effect of
uncontainment of simultaneous fractures (Scenario 4) is obtained as a ratio to be multiplied
with the resultant value in the previous step.
3.4. Major assumptions and data range in stress shadow analysis
Geometry of the hydraulic fracture with respect to the net play and bounding layers,
different boundary conditions, the most influencing variables and their varying range, and
assumptions of the analyses are defined in this section.
25
3.4.1. Hydraulic fracture geometry
Geometry of the hydraulic fracture with respect to the wellbore, net play, and bounding
layers is shown in Figure 3.2a. As it can be seen from this figure, a rectangular hydraulic
fracture located in x -z plane is made perpendicular to the wellbore. According to the fact
that maximum stress change is observed in the center of the hydraulic fracture (see Waters
et al., 2009), a plane perpendicular to the height axis (z direction) passing through the half-
height of the fracture is used for stress shadow, and maximum aperture analysis. Q plane in
Figure 3.2b shows the plane of aperture and stress shadow calculation. Length, height, and
width (aperture) of the hydraulic fracture are also shown in Figure 3.2b. It is necessary to
indicate that in this analysis, w
mux
(half of the maximum aperture; see Eq. (2.10)) is
numerically calculated and can be predicted using the proposed equations in the following
sections. Therefore, aperture of the hydraulic fracture can simply be obtained by doubling
the value of w
mux
.
Figure 3. 2. Geometry of a contained hydraulic fracture and stress shadow plane
Bounuing layei
wellboie
c
z
y x
B
wiuth
L
e
n
g
t
h
: 2
c
B
e
i
g
h
t
:

B
Apeituie: 2 Wmax
Bounuing layei
Net Play
Q
(a) (b)
26
3.4.2. Different boundary conditions for shadow analysis
In general, modeling approaches for evaluating the stress shadow effect adopt either one of
two boundary conditions for their case studies. These are:
1. Choice of a fixed hydraulic pressure inside the fracture (P
H
= ctc): This implies that
the fracture aperture is changed by varying values of the moduli of the rock and
hydraulic pressure inside the fracture.
2. Choice of a fixed maximum fracture aperture (w
mux
= ctc): With a fixed fracture
aperture, the influence of other parameters on the stress shadowing may be
investigated.
The assumption of a constant pressure within the fracture mentioned as the first boundary
condition is a more reasonable approximation of reality and therefore, in this study, we
employ this boundary condition.
3.4.3. Assumptions
The key assumptions employed in the numerical simulation are:
The 3-D domain is assumed to be a completely elastic medium without any
plasticity-based constitutive laws;
Propagation of the fracture is not considered in this work; instead the hydraulic
fracture is assumed to have been created;
27
The aspect ratio defined as the ratio of height to the length of the fracture is
equal/less than unity;
The threshold angle (0
1
) is varied from 5-80;
The aperture of the fracture is = 2 w
mux
and therefore is twice the w
mux
;
In case shadow analyses are carried out for a threshold angle of 80 (SE_S
80

and SE_
80
), only analyses with hydraulic pressures larger than maximum
horizontal stress (P
H
> S
hmux
) are considered (see shadow mechanism part;
Section 3.5).
3.4.4. Data range
In this section, the adopted data range for the massive numerical simulation is presented.
Since stress shadowing is most challenging when we have lower stress anisotropies (see
Morrill and Miskimins, 2012), the ratio of minimum horizontal stress over maximum
horizontal stress is assumed to be in the range of K
h
= 0.95-0.99. Moreover, previous
studies in shale have indicated that Youngs modulus of shale changes from 7 to 77 GPa
with an average value of 26.9 GPa (Hay and Sondergeld, 2011). Therefore, Youngs
modulus of the net play is assumed in the range of E
n
= 10-70 GPa. For studying of the
effect of bounding layers, modulus of bounding layers is assumed as 0.25-4.0 times of that
of net play and 0.1-0.3 of the fracture height is assumed to penetrate inside the bounding
layers. For the case of simultaneous hydraulic fracturing of horizontal wells, the distance
between parallel wells is varied with similar fracture length and aspect ratios for both of the
wells. The input variables and their ranges are shown in Table 3.1.

28
Table 3. 1. Input variable range for the numerical simulation
Parameter Range Unit
Fracture aspect ratio 1.00-0.20 -
In-situ stress anisotropy 0.95-0.99 -
Poissons ratio 0.00-0.40 -
Youngs Modulus 10-70 GPa
Excess pressure 0.0-3.0 MPa
Youngs modulus ratio
1
0.25-4.0 -
Penetration
2
0.1H-0.3H m
Distance between tips
3c
-
3.5. Shadow mechanisms
In order to study the stress shadow, different mechanisms of shadow around a hydraulic
fracture should be recognized. The mechanisms of stress shadowing generally belong to
two categories which will be investigated in this study:
1. The pressure applied lies between the minimum and maximum horizontal
stress (S
hmn
< P
H
< S
hmux
). This results in only a marginal deviation of
the in-situ stresses around the crack tips (0
mux
< 45) and consequently,
does not lead to a principal stress reversal (90 rotation). Figure 3.3a
illustrates a marginal deviation of 40 for this scenario.
2. In cases where the applied pressure is larger than the maximum horizontal
stress (P
H
> S
hmux
), a reversal of stress occurs (0
mux
> 45) and this
phenomenon is illustrated in Figure 3.3b.
It should be noted that principal stress deviations plotted in Figs. 3.3 are from the
maximum deviation plane (plane Q defined in Figure 3.3b). The pink lines show the
fracture plane.

1
Youngs modulus ratio is the ratio of Youngs modulus of the bounding layer over that of the net play
2 Penetration is the part of the fracture height penetrating the bounding layers
Figure 3.
Consequentl
fracture and
mechanism
shadow. It i
only be ana
zero for thre
3.6. Met
In this secti
edge. In thi
studied alon
3.6.1. Param
According t
knowing the
3. Different
ly, stress sha
may cause a
difference s
is important
lyzed for th
eshold angles
hod of stud
on, the tech
is thesis, as
g length and
meters related t
o Irwin (195
e SIF of the c

t stress shad
adow is seen
a diverse ran
should be k
to notice th
e second me
s of larger tha
ying stress i
nique for inv
a result of
height of th
to SIF change
57), stress fie
crack. Consid
(a)
29
dow mechan
n to be a fun
ge of deviati
ept in mind
hat stress sh
echanism be
an 45.
intensity fac
vestigating o
assuming re
e fractures.
eld around th
dering Eq. (2
D
e
v
i
a
t
i
o
n

a
n
g
l
e
,

nisms by dif
nction of the
ions in the p
d when inter
hadow for th
ecause shado
ctor
of SIF chang
ectangular fr
he crack tip
2.2), we have:
fferent hydra
e hydraulic p
principal stres
rpreting the
hreshold angl
ow for the fi
ge is present
ractures, SIF
can be dete
:

aulic pressu
pressure insid
ss directions.
results of
les of 80 sh
irst mechani
ed for each
F change is t
rmined simp
(b)
ures
de the
. This
stress
hould
ism is
crack
to be
ply by
D
e
v
i
a
t
i
o
n

a
n
g
l
e
,

30
o
K
I

(3.1)
in which K
I
is linearly related to the internal pressure (P
H
), directly related to a function of
the geometry of the fracture (G), and a function incorporating the effect of boundaries on
the fracture (E). Effect of boundary represents itself as the effect of having cracked media
with finite dimensions, or having multiple adjacent fractures. Therefore, we can write:
K
I
= P
H
G E
(3.2)
As an example, for a 2-D fracture in an infinite medium under internal pressure (see
Chapter 2), G = no and E = 1. Determining G, E for different crack geometries and
different boundary conditions is of paramount importance. This is because by having these
functions, one can determine the SIF of a specific crack geometry, thereby having the
stress field around the crack tip and consequently, its potential for propagation.
Combining Eqs. (3.1) and (3.2), we can write:
o
P
H
G E
(3.3)
Therefore, ratio of a stress component over the internal pressure can be assumed as
indicative of the effect of crack geometry and its boundary conditions on the SIF.
According to the fact that in this thesis, we intend to change the geometry of the hydraulic
fracture (aspect ratio) and consider different fracture patterns (simultaneous and multistage
fracturing), all the changes in the SIF can be considered as a result of changes in these two
functions. Therefore, by the above-mentioned changes, the ratio of stress over the internal
31
pressure will result the SIF change as a result of geometry and boundary condition
functions (GE).
3.6.2. Monitoring of SIF change along the fracture edge by a single value
Crack tip of a 3-D hydraulic fracture is a closed line in space which may be an irregular
shape depending on many parameters including in-situ stress, hydraulic pressure change,
moduli variation, containment, heterogeneity of the medium, etc. In this study, this closed
loop has been simplified to a rectangle along which the SIF is changing by position (Figure
3.4).

Figure 3. 4. Fracture geometry together with stress change along its edges
Therefore, according to the prescribed mesh size of the FE model, a large number of data
(stress components) are associated with each case of study. In order to make a comparison
much simpler between different geometries/fracturing patterns, a function is fit to the data
and the resultant coefficients of the function are compared instead. The proposed function
which gives satisfactory fit on the numerical stress values is as given in Eq. (3.4).
Length, 2c
B
e
i
g
h
t
,

2
b
x
y
zz
zz
32
o
zz
-o
zz0
P
H
= o
1
[1 -cxp [-o
2
x
c

(3.4a)
o
zz
-o
zz0
P
H
= o
1
[1 -cxp [-o
2
y
b

(3.4b)
in which P
H
is the applied hydraulic pressure, o
zz0
is the normal stress at the corner, o
zz
is
normal stress along the length or height of the fracture, x axis is in the direction of length
originated from the corner, c is the fracture half-length, y is the axis in the direction of
fracture height originated from the corner, b is the half-height of the fracture, and o
are
coefficients of the function (i = 1,2). Investigating the proposed function, it is revealed
that stress change is highly controlled by o
1
rather than o
2
. This is because the negative
sign of o
2
in an exponential format makes a small value to be much smaller. Since this
exponential value is subtracted from unity, no noticeable change is observed for stress in
the coefficient range we are dealing with in this regard.
Figure 3.5 shows a typical example of normal stress change along the crack edge. As shown
in the figure, behavior of stress change along crack length or height has satisfactorily been
predicted using Eqs. (3.4). It is also evident from the figure that normal stress is maximum
in the middle point of the crack length and it is reduced to the minimum value at the crack
corner. Since normal stress is normalized by being subtracted by stress at the corner and
divided by the internal pressure, the values at the corners are zero. In addition, stress
change with height or length of fractures with aspect ratio of unity is similar as expected
but for cracks with aspect ratio lower than unity, stress decreases with both length and
height of the fracture, however, the magnitude of decrease is higher in fracture height
33
compared to that of the length. The conspicuous point is that despite the fact that length
of the fracture is not changing and aspect ratio is being reduced by decreasing the fracture
height, still stress along the crack length is reduced.

Figure 3. 5. Stress variation along the fracture edges
Considering the dominant coefficient of the function proposed in Eqs. (3.4), one can have
a good comparison between SIF of different crack geometries and multi-fracture
configurations. As it can be seen from Eqs. (3.4), this value is multiplied with the
exponential function and as a result, any difference in this coefficient means the same
change in normal stress and consequently, the SIF of the crack.
3.7. Different scenarios for calculation of SIF change
There are four different scenarios that are considered for SIF change analysis in this thesis.
Figure 3.6 shows the adopted scenarios for the hydraulic fracturing pattern/technique.
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
N
o
r
m
a
l
i
z
e
d
n
o
r
m
a
l
s
t
r
e
s
s
,
z
z
z
z
0
)
/
P
H
Normalizedtipposition,(x/c,y/b)
Szz_Height(AR=1.0)
Szz_Length(AR=1.0)
Szz_Height(AR=0.2)
Szz_Length(AR=0.2)
1.000.0 1.00
34

Figure 3. 6. Different fracturing techniques/patterns in this study
It is evident from Figure 3.6 that Scenario 1 is the basic scenario in which we have a single
stage fracture from a standalone well. Scenario 2 shows multistage hydraulic fracturing in a
standalone well. In order to compare propagation potential of the second scenario
(multistage fracturing) with the basic case (Scenario 1), the numerically obtained SIF of the
second scenario is divided by the SIF of the first scenario. Any value lower/higher than
unity means that propagation potential for the scenario under study is smaller/larger than
that in the basic case. Scenario 3 considers the effect of simultaneous single stage fracturing
of the medium between two parallel wells. The obtained SIF for this scenario should also
be compared with that in scenario one in order to quantify the effect of simultaneous
fracturing in a single stage mode. Finally, Scenario 4 is studied in which the effect of
simultaneous multistage hydraulic fracturing is considered between two parallel wells.
Scenario 4 has two influencing variables acting simultaneously on the SIF of the fractures.
The resultant SIF from this scenario should be compared with results of Scenario 1 in
order to have the effect of simultaneous multistage fracturing pattern on propagation
potential of the fractures.
1 2
S 4 Ls
2Lp
35
Therefore, changing the spacing between fractures in multistage fracturing and distance
between the meeting fractures in simultaneous fracturing brings the opportunity of
quantifying the SIF change for each scenario. Of course, the effect of fracture geometry
(aspect ratio) on the SIF is also considered.
The above-mentioned scenarios are established to quantify the SIF change according to the
geometry and pattern of the fractures. One qualitative analysis is also done on the effect of
offset between meeting fractures on the SIF change of the fractures. The term offset in this
regard means the distance between two planes on which hydraulic fractures are located in a
simultaneous fracturing mode.
3.8. Data range for the SIF change analysis
In this study, an effective range has been assumed for all the studies. The input variables
and their range are defined in Table 3.2.
Table 3. 2. Input variable range for the numerical simulation
No. Parameter Range
1 Fracture aspect ratio 0.2-1.0
2 Fracture spacing
1
0.1 c-7.0 c
3 Fracture distance
2
0.1 c-3.0 c
4 Fracture offset 0.1 c-1.0 c
Table 3.2 shows that an extensive number of numerical models are required to be built and
analyzed to predict the SIF change via simultaneous multistage fracing job.
3.9. Verification of stress field for simple problems
In this section, the modeling strategy used in ABAQUS is examined by numerically solving
simple crack geometries, the analytical solutions of which are available in the literature.

1
The term spacing in this thesis is only used as a measure of the interval between the adjacent fractures in multistage fracturing mode.
2
The term distance in this thesis only refers to the distance between meeting fracture tips in simultaneous fracturing mode.
36
Semi-infinite and penny-shaped cracks with internal pressure are chosen for this purpose
(Sneddon, 1945). In this step of the work, appropriate size of the model is defined in order
to minimize the boundary effects to the model response.
Cracks in a material are associated with three geometrical parameters; length, width (called
aperture), and height. The ratio of any two of these can be considered as the aspect ratio.
In this paper, the ratio of height over length of the fracture is called aspect ratio (AR =
bcigbtlcngtb). We numerically simulate two cracks with different geometries, one with
an infinite height (plane strain crack; AR = ), and the other with aspect ratio of unity
(penny-shaped crack; AR = 1.u), both with an internal pressure. The results of these
numerical simulations are compared with the analytical solutions.
The validation approach described herein compares the horizontal and vertical stresses
around the simulated crack with the analytical results presented by Sneddon and Elliot
(1946). These are shown in Figure 3.7 for horizontal and vertical stresses along a line
perpendicular to the face of the hydraulic fracture as shown in the inset.
Figure 3. 7. Stress validation in the direction vertical away from the fracture center
1.20
1.00
0.80
0.60
0.40
0.20
0.00
0.20
0 2 4 6 8 10
N
o
r
m
a
l
i
z
e
d
S
t
r
e
s
s
,
S
/
P
H
Normalizeddistance,x/c
planestraincrack
SXX/PH(NUMERICAL)
SYY/PH(NUMERICAL)
SXX/PH(ANALYTICAL)
SYY/PH(ANALYTICAL)
xx
xx
yy yy
1.2
1
0.8
0.6
0.4
0.2
0
0.2
0 1 2 3 4 5
N
o
r
m
a
l
i
z
e
d
S
t
r
e
s
s
,
/
P
H
Normalizeddistance,x/c
Pennyshapedcrack
Syy/P(ANALYTICAL)
Sxx/P(ANALYTICAL)
Syy/PH(NUMERICAL)
Sxx/PH(NUMERICAL)
xx
xx
yy yy
37
The figure demonstrates excellent agreement between the analytic and numeric simulation
results for plane-strain and penny-shaped cracks and, therefore, the numerical modeling
strategy adopted here can be considered valid for modeling of hydraulic fractures and stress
shadowing effects. As a result, according to the used numerical method and all the
conditions/assumptions we can run numerical simulations to study the influence of the
considered variables on stress shadow size and stress intensity factor. In Chapters 4 and 5,
all the results of numerical simulations are presented.

38
Chapter 4: Prediction of shadow and aperture of hydraulic fractures
4.1. General
In Chapter 3, the methodology of investigating stress shadow and stress intensity factor
(SIF) of a hydraulic fracture was proposed. All the applied boundary conditions and major
assumptions were also presented. In this chapter, based on explanations in Chapter 3, the
results of all the analyses conduced for stress shadow of hydraulic fractures are reported
and a detailed discussion on the stress shadow is also presented.
4.2. Prediction of aperture for a contained hydraulic fracture
First, aperture of hydraulic fractures is measured according to the numerical results and it is
predicted by proposing an equation. This is done first because aperture of hydraulic
fractures is of crucial significance in proppant size/type and estimating fracture
conductivity.
4.2.1. Effect of in-situ stresses on aperture
As mentioned in Chapter 2, Eq. (2.10) describes half-aperture of a hydraulic fracture in a 2-
D medium as a function of the applied hydraulic pressure, relative distance from fracture
center, and rock moduli. The formulation of Eq. (2.10) ignores the impact of any far-field
stress. However, for the case of aperture determination, when in-situ stresses are present,
the excess hydraulic pressure (P
N
= P
H
-S
hmn
) should be considered instead. This is
because according to the numerical results, maximum horizontal stress (applied parallel to
the fracture face) has negligible effect on the aperture of the internally pressurized fracture
(see also Jeffrey, 1989) and therefore, a simple subtraction of minimum horizontal stress
from the applied pressure and neglecting far-field stresses suffices to estimate the aperture
39
similar to the case mentioned in Eq. (2.10). Of course, this elimination is only valid for the
case of aperture, where stress anisotropy is not an influencing variable.
4.2.2. Mathematical equation for prediction of aperture of a single contained fracture
In this section, aperture of a contained hydraulic fracture in a standalone well is predicted
considering different mechanical properties for the shale, variant stress anisotropies,
different internal pressures, and fracture aspect ratios (see Scenario 1, Figure 3.1; Chapter
3).
First, it was observed that there is similar effect of Youngs modulus in aperture
determination of 2-D and 3-D fractures. Influence of Poissons ratio, on the other hand,
was observed to be different from the 2-D fractures in such a way that aperture of a 2-D
fracture shows a quadratic dependence on Poissons ratio (see Eq. (2.10)), while for a 3-D
hydraulic fracture, there is a third order polynomial dependence between aperture and
Poissons ratio. In addition, aperture of 3-D hydraulic fractures shows a quadratic
dependence on the fracture aspect ratio as well. Following the above explanations, aperture
of a contained, 3-D hydraulic fracture can be determined using Eq. (4.1).
w
mux
AR<1
(P
N
, v, E
n
)
c
= RS
AR
w
F(v)
P
N
E
n

(4.1)
in which F(v) is a third order polynomial of the Poissons ratio, multiplier RS
AR
w
is a
quadratic function of aspect ratio, w
mux
AR<1
is maximum half-aperture, and P
N
, E
n
, c are
excess pressure, net play Youngs modulus, and half length of the hydraulic fracture,
respectively. These functions are defined in Eqs. (4.2) and (4.3):
40
F(v) = (o
1
+o
2
v +o
3
v
2
+o
4
v
3
) ;
o
1
= +1.1286 o
3
= -u.8919
o
2
= +u.1494 o
4
= -2.u286
(4.2)
RS
AR
w
= -u.27S7(AR)
2
+1.Su92(AR) -u.uS16 (4.3)
This equation underscores the independence between the stress anisotropy and the half-
aperture of hydraulic fractures, implying that aperture only depends on the minimum
horizontal stress (P
N
= P
H
-S
hmn
).
4.2.3. Reliability of the proposed function for aperture prediction
A comparison of the numerical results and the predictions for Eq. (4.1) are shown in
Figure 4.1 which demonstrates excellent agreement between numerically obtained values
and the predicted ones.

Figure 4. 1. Half-aperture prediction using Eqs. (4.1-4.3)
Therefore, employing Eqs. (4.1-4.3), one can have a satisfactory estimate of the aperture of
a hydraulic fracture by giving the aspect ratio, Poissons ratio, excess pressure, and Youngs
modulus. Excess pressure of a hydraulic fracture is determined via subtraction of minimum
horizontal stress from hydraulic pressure inside the fracture. Therefore, minimum
w'
max
=0.9991w
max
R=0.9999
0.00E+00
1.00E04
2.00E04
3.00E04
4.00E04
0.00E+00 1.00E04 2.00E04 3.00E04 4.00E04
P
r
e
d
i
c
t
e
d
h
a
l
f
a
p
e
r
t
u
r
e
r
a
t
i
o
,
w
'
m
a
x
/
c
Numericalhalfapertureratio,w
max
/c
41
horizontal stress should also be considered as an influencing variable in determination of
aperture of hydraulic fractures.
4.2.4. Visualization of aperture prediction function
The proposed equation for prediction of aperture in 3-D fractures has four inputs
(Poissons ratio, aspect ratio, Youngs modulus, and excess pressure) and aperture as the
function of these inputs. Visualization of this function is only possible when plotting is
done by the ratios, and having different plots for different aspect ratios.
The expression for the fracture half-aperture in Eq. (4.1) is re-drawn in Figure 4.2 for
different Poissons ratios, excess hydraulic pressures, two aspect ratios, and two different
values of Youngs moduli.
Figure 4. 2. Visualization of aperture determination function
The key observation from Figure 4.2 is the relationship between Youngs modulus and
fracture half-aperture. More compliant rocks characterized by a lower Youngs modulus are
associated with a larger fracture aperture. Increases in excess pressure also lead to a
widening of the fracture and so do decreases in the value of Poissons ratio. In order to
42
justify the behavior of aperture with respect to moduli of the rock, consider elastic moduli
relationship. Bulk compressibility of rock is obtained as C
b
= S(1 -2v)E
n
. This means
by decreasing of Poissons ratio and Youngs modulus, compressibility of rock increases,
thereby raising the aperture of the hydraulic fracture.
4.3. Prediction of shadow around a contained hydraulic fracture (6
T
= 5)
This section focuses on the description of the stress shadowing around a single contained
hydraulic fracture (see Figure 3.1; Scenario 1 in Chapter 3). The input parameters were
varied according to Table 3.1 to quantify the shadow size around a hydraulic fracture. It is
worth mentioning that because of using pressure boundary condition, Youngs modulus of
the medium does not show any effect on stress shadow.
4.3.1. Numerical results for stress shadow analysis
In order to study the stress shadow, excess pressure (P
N
) is normalized with ten times of
the atmospheric pressure (1uP
u
) and shadow length is normalized with the fracture half
length (c). The normalized shadow lengths predicted as a function of the normalized
excess pressures are typically shown in Figure 4.3 for an aspect ratio of 1.0, horizontal
stress anisotropy of 0.95 and varying Poissons ratios.
It can be seen that there is a nonlinear direct relationship between the shadow size and
internal pressure. This means that by raising the internal pressure, shadow size increases
but with a decreasing gradient. In Figure 4.3, we also describe the variations in the stress
shadow effect with Poissons ratio which shows that shadow size increases with Poissons
ratio.

43
Figure 4. 3. Shadow change with Poissons ratio, numerical values and the
predictions
4.3.2. Mathematical equation predicting stress shadow of a single contained fracture
We use all the numerical results of a single contained hydraulic fracture, like the ones
typically shown in Figs. 4.3, to develop an expression for predicting the stress shadow size,
as shown in Eq. (4.4):
SE_S
05
AR=1
c
= RS
AR
SH05
A(v, K
h
) [
P
N
1uP
u
1 +B(v, K
h
) [
P
N
1uP
u

(4.4)
The proposed Eq. (4.4) contains two coefficients as A(v, K
h
) and B(v, K
h
) which are
functions of Poissons ratio and stress anisotropy. It also contains multiplier RS
AR
SH05
, a
quadratic function of aspect ratio, which applies the effect of hydraulic fracture geometry
using Eq. (4.5).
RS
AR
SH05
= -u.2S96(AR)
2
+1.2888(AR) -u.u28S (4.5)
0
0.5
1
1.5
2
2.5
3
0 0.5 1 1.5 2 2.5 3
N
o
r
m
a
l
i
z
e
d
S
h
a
d
o
w
,
S
H
_
S
0
5
c
NormalizedExcessPressure,P
N
/(10P
0
)
Poisson'sRatio=0.00
Poisson'sRatio=0.40
Stressratio=0.95
Aspectratio=1.00
0
0.5
1
1.5
2
2.5
3
0 0.5 1 1.5 2 2.5 3
N
o
r
m
a
l
i
z
e
d
S
h
a
d
o
w
,
S
H
_
S
0
5
c
NormalizedExcessPressure,P
N
/(10P
0
)
Poisson'sRatio=0.00
Poisson'sRatio=0.05
Poisson'sRatio=0.10
Poisson'sRatio=0.15
Poisson'sRatio=0.20
Poisson'sRatio=0.25
Poisson'sRatio=0.30
Poisson'sRatio=0.35
Poisson'sRatio=0.40
Stressratio=0.95
Aspectratio=1.00
44
Any aspect ratio, between 1.0 and 0.0 results a value for multiplier RS
AR
SH05
to be used in
Eq. (4.4) for stress shadow size prediction. It is seen that aspect ratio of the fractures are
really important in determination of stress shadow size. In addition, effect of Poissons
ratio and stress anisotropy is present in the coefficients of Eq. (4.4).
4.3.3. Coefficients of the proposed function for stress shadow prediction
Coefficients of the function, Eq. (4.4), can be obtained from mathematically relating stress
shadow size with Poissons ratio and stress anisotropy. The first observation was that the
coefficients of the proposed function are linear functions of Poissons ratio. Having
obtained the values of the linear regression between function coefficients and Poissons
ratio, one can ponder the dependency on the stress anisotropy afterwards. The linear
relationship between the coefficients of the function and Poissons ratio is typically shown
for two different stress anisotropies (K
h
= 0.95, 0.99) in Figure 4.4.
Figure 4. 4. Shadow function coefficient change by Poissons ratio
The coefficients of A(v, K
h
) and B(v, K
h
) are then expressed as:
A=0.302v +3.6115
B=0.736v +1.2408
0
1
2
3
4
0 0.1 0.2 0.3 0.4
S
h
a
d
o
w
f
u
n
c
t
i
o
n
c
o
e
f
f
i
c
i
e
n
t
s
,
A
,
B
Poisson'sratio,v
STRESSANISOTROPY=0.95
A
B
A=1.7331v +10.698
B=0.9431v +2.7923
0
2
4
6
8
10
12
0 0.1 0.2 0.3 0.4
S
h
a
d
o
w
f
u
n
c
t
i
o
n
c
o
e
f
f
i
c
i
e
n
t
s
,
A
,
B
Poisson'sratio,v
STRESSANISOTROPY=0.99
A
B
45
A(v, K
h
) = H
A
(K
h
)v +N
A
(K
h
)

(4.6a)
B(v, K
h
) = H
B
(K
h
)v +N
B
(K
h
) (4.6b)
As it is shown in Eqs. (4.6), there is a linear relationship between coefficients of the
proposed function and Poissons ratio. Additionally, effect of stress anisotropy shows itself
in the linear regression coefficients H, N in Eqs. (4.6). According to the numerical results,
there is a nonlinear relationship between H and N coefficients and stress anisotropy (K
h
)
which are given as:
H
A
=
o
M
A
+b
M
A
K
h
1 +c
M
A
K
h
+J
M
A
K
h
2
; N
A
=
o
N
A
+b
N
A
K
h
1 +c
N
A
K
h
+J
N
A
K
h
2

H
B
=
o
M
B
+b
M
B
K
h
1 +c
M
B
K
h
+J
M
B
K
h
2
; N
B
=
o
N
B
+b
N
B
K
h
1 +c
N
B
K
h
+J
N
B
K
h
2

(4.7a)
(4.7b)
in which the values of the defined coefficients are listed in Table 4.1:
Table 4. 1. Coefficients of the function in Eq. (4.7)
coefficient value coefficient value
o
M
A

-0.06415
o
N
A

0.81730
b
M
A

0.06631
b
N
A

-0.52469
c
M
A
-1.97737 c
N
A
-0.44572
J
M
A

0.97792
J
N
A

-0.54166
o
M
B
-25.97845 o
N
B
0.61366
b
M
B

25.92616
b
N
B

-0.50759
c
M
B

39.27341
c
N
B

-0.25782
J
M
B

-40.35336
J
N
B

-0.71923
Substituting values of Table 4.1 inside Eqs. (4.7), one can draw H and N coefficients with
respect to K
h
. The resulting curves are shown in Figure 4.5. It is seen from Figure 4.5a that
there is a nonlinear relationship between the coefficients (H, N) and K
h
. From Eq. (4.6),
46
one can have Figure 4.5b for coefficients (A, B) assuming a value for the Poissons ratio
(v = u.2 for this case). As it can be seen from Figure 4.5b, gradient of the curve (A -K
h
)
is higher than that of (B -K
h
). Based on the fact that coefficient A is divided by
coefficient B in stress shadow determination (see Eq. (4.4)), their ratio can be indicative of
the behavior of shadow versus stress anisotropy. Figure 4.5c shows the ratio of AB for
the special case of v = u.2. As it can be seen from Figure 4.5c, there is a nonlinear
relationship with increasing gradient between shadow size and stress anisotropy.
Figure 4. 5. Behavior of function coefficients with stress anisotropy (K
h
)
4.3.4. Reliability of the proposed function for stress shadow prediction
Equations (4.4-4.7) now can be employed to describe the shadow size (SE_S
05
) around a
contained hydraulic fracture for any given aspect ratio. The comparison between the
predicted values and numerical results is shown in Figure 4.6 which demonstrates excellent
agreement between the predicted and numerically obtained stress shadow sizes for
threshold angles of 5 and any specified aspect ratio.
2
0
2
4
6
8
10
12
0.94 0.96 0.98 1
C
o
e
f
f
i
c
i
e
n
t
v
a
l
u
e
(
M
,
N
)
Horizontalstressanisotropy,K
h
MA
NA
MB
NB
(a)
0
2
4
6
8
10
12
0.94 0.96 0.98 1
C
o
e
f
f
i
c
i
e
n
t
v
a
l
u
e
(
A
,
B
)
h
A
B
(b)
v=0.2
3.00
3.25
3.50
3.75
4.00
4.25
4.50
0.94 0.96 0.98 1
C
o
e
f
f
i
c
i
e
n
t
r
a
t
i
o
(
A
/
B
)
h
A/B
(c)
47
Figure 4. 6. Shadow size prediction using Eqs. (4.4-4.7)
4.3.5. Visualization of stress shadow prediction function
As in the case of aperture predicting function, there are four input variables, namely as
stress anisotropy, Poissons ratio, excess pressure, and aspect ratio and one output function
in case of stress shadow size. Equations (4.4-4.7) may be replotted as shown in Figure 4.7
which expresses that shadow size has a direct relationship with excess pressure, stress
anisotropy, and Poissons ratio.
Figure 4. 7. Shadow around a hydraulic fracture
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
P
r
e
d
i
c
t
e
d
s
h
a
d
o
w
,
S
H
_
S
0
5
/
c
Numericalshadow,SH_S
05
/c
AR=1.0
AR=0.8
AR=0.6
AR=0.4
AR=0.2
(SH05/c)
Pre
=0.9866(SH05/c)
Num
R=0.9966
48
4.3.6. Discussion on the relationship between aperture and stress shadow
As indicated in Section 4.2, stress anisotropy is not an influencing variable on the aperture
(despite of its noticeable effect on shadow size) and Youngs modulus does not have any
significant effect on the shadow size (even though it inversely controls the aperture). In
addition, Poissons ratio inversely controls the aperture of the fracture, while it directly
influences the shadow size. As a result, although there is some direct relationship between
aperture and stress shadow size because of the effect of excess pressure, stress shadowing
of a hydraulic fracture cannot be assumed entirely dependent on its aperture.
Based on the numerical results, more stress contrast is observed for lower Poissons ratios
(see also Roussel and Sharma, 2011). Larger stress contrast (o
x
-o
) means that we are

getting closer to the original state of stress. This concept was used by Jo (2012) to calculate
the shadow size around hydraulic fractures in an analytical approach. Considering Eq. (2.8),
it is also induced that larger stress contrast means smaller deviation angles which confirms
more closeness to the original state of stress. This means that lower Poissons ratios will
result the original state of stress to be at a smaller distance to the fracture face compared to
the case of larger Poissons ratios. Furthermore, from analytical Eqs. (2.6), it is implied that
there is a direct relationship between shadow size and Poissons ratio which confirms our
numerical results.
4.4. Prediction of shadow around a contained hydraulic fracture (6
T
> 5)
This section focuses on the variations of the shadow size around a contained hydraulic
fracture with different threshold angles. Again considering Figure 3.3, it is evident that by
adopting different threshold angles, the size of stress shadow will change. In the literature,
49
sometimes complete reorientation of in-situ stresses (90) meaning SE_S
80
has been called
half of minimum fracture spacing (see Roussel and Sharma, 2011). More importantly,
having a decreasing trend of stress shadow size with respect to the threshold angle, one can
have a better insight of shadow size gradient with threshold angle. A detailed study of the
numerical results showed that SE_S
80
cannot be satisfactorily predicted using an equation
like Eq. (4.4). Figure 4.8 also shows that the ratio of shadow with threshold angle of 80
over that of 5 (SE_S
80
SE_S
05
) for all the analyses varies from 0.13 to 0.80. The mean
value using a normal distribution is u.S662 _u.11S2. Because of the high variation of the
values, a more acceptable method is applied to have a satisfactory prediction of shadow for
higher threshold angles.

Figure 4. 8. Distribution of SH_S
80
over SH_S
05
ratio (for all aspect ratios)
4.4.1. Mathematical equation for prediction of stress shadow with different threshold angles
A precise investigation of the numerical results showed that the difference between
SE_S
05
and SE_S
10<0
T
<90
is approximately a fixed value changing with threshold angle
and Poissons ratio. This means that the curve of shadow decrease with threshold angle
(assuming a fixed Poissons ratio) is almost a fixed function which is only shifted. The
0.8 0.7 0.6 0.5 0.4 0.3 0.2
160
140
120
100
80
60
40
20
0
SH_S 80/ SH_S 05
F
r
e
q
u
e
n
c
y
Mean 0.5662
StDev 0.1152
N 1855
50
mean differences (ASE_S
) are shown in Table 4.2 for different aspect ratios. It is worth

indicating that these values are obtained in an iterative approach which results the least
difference between the calculated and numerical shadow results (SE_S
05
AR<1
-
ASE_S
05-0
T
AR<1
= SE_S
0
T
AR<1
).
In order to observe the behavior of shadow difference with respect to the threshold angle,
the values in Table 4.2 are also plotted in Figure 4.9. As it is evident from Figure 4.9, there
is a large difference between SE_S
05
and other shadows with threshold angles up to 30
and the difference for the rest of the threshold angles is gently sloped.
Table 4. 2. Shadow decrease by threshold angle
Threshold
angle, 0
1

_
1
c
] ASE_S
05-0
T
AR=1.0
_
1
c
] ASE_S
05-0
T
AR=0.8
_
1
c
] ASE_S
05-0
T
AR=0.6
_
1
c
] ASE_S
05-0
T
AR=0.4
_
1
c
] ASE_S
05-0
T
AR=0.2

5 0.000 0.000 0.000 0.000 0.000
10 0.426 0.352 0.279 0.202 0.096
15 0.632 0.527 0.424 0.303 0.146
20 0.758 0.637 0.510 0.366 0.175
30 0.903 0.759 0.610 0.432 0.211
40 0.963 0.817 0.658 0.468 0.230
50 0.981 0.831 0.674 0.484 0.239
60 1.002 0.840 0.681 0.490 0.244
70 1.019 0.859 0.690 0.495 0.247
80 1.064 0.893 0.710 0.502 0.250

Figure 4. 9. Shadow difference with respect to varied threshold angles
0.000
0.200
0.400
0.600
0.800
1.000
1.200
0 10 20 30 40 50 60 70 80 90
S
h
a
d
o
w
d
i
f
f
e
r
e
n
c
e
,
(
S
H
_
S
0
5
S
H
_
S
)
/
c
Thresholdangle, (Degree)
AR=1.0
AR=0.8
AR=0.6
AR=0.4
AR=0.2
51
Using these values (Table 4.2), one can predict the values of shadow in different threshold
angles. However, the results showed some scattering when using these values. This
observation shows that Poissons ratio also contributes in shadow change versus threshold
angle. As a result, the formulation for shadow for any threshold angle can be written as Eq.
(4.8) for different aspect ratios.
_
1
c
] SE_S
0
T
AR
= _
1
c
] SE_S
05
AR
-_
1
c
] ASE_S
05-0
T
AR
-(o
1
v
2
+o
2
v +o
3
)
(4.8)
The values of o
1
, o
2
, o
3
are given for different aspect ratios in Table 4.3.
Aspect ratio (AR) o
1
o
2
o
3

1.0 +u.782S +u.4927 -u.1S6u
0.8 +2.2uS8 -u.u41S -u.u6S6
0.6 +u.9264 +u.SSSS -u.u27u
0.4 +S.1194 +u.u119 -u.u774
0.2 +u.8824 +u.148u -u.u284
Therefore, by having Poissons ratio of rock and coefficients mentioned in Table 4.3, the
shadow difference (see Table 4.2) and shadow for threshold angle of 5 (see Eqs. (4.4-4.7)),
one can have stress shadow for any threshold angle (SE_S
0
T
).
4.4.2. Reliability of the equation predicting stress shadow with different threshold angles
In this section, shadow size with complete reorientation (SE_S
80
) is predicted using Eq.
(4.8) and the resultant values are compared with the numerical results. Figure 4.10 shows
the reliability of predicted data for this analysis. It is important to notice that this
formulation is only valid for the cases that excess pressure is more than the maximum
horizontal stress (see Figure 3.3; second shadow mechanism).
52
Figure 4. 10. Shadow ratio prediction (threshold angle of 80)
From Figure 4.10, one can see that minimum shadow is predicted using Tables 4.2, 4.3 and
Poissons ratio in a fairly accurate manner. As a result, stress shadow around a contained
hydraulic fracture can be predicted for different aspect ratios and for different threshold
angles. It is worth mentioning that in case aspect ratio is in-between any of the mentioned
aspect ratios (See Eq. (4.8); Table 4.3), stress shadow for the bounding aspect ratios should
be determined and then, by a linear interpolation, stress shadow of the hydraulic fracture is
determined for the desired aspect ratio.
4.5. Effect of uncontainment of the fracture (standalone well fracturing)
All the calculations of stress shadow and aperture of a single hydraulic fracture in the
previous sections were done for a contained hydraulic fracture. This means that hydraulic
fracture is entirely located inside the net play without any penetration into the bounding
layers. In this section, we are intended to investigate the effect uncontainment of the
hydraulic fracture. As it is shown in Table 3.1, Youngs modulus of the bounding layers is
assumed 0.25-4.0 times of that in the net play to investigate the shadow size and aperture
of the uncontained hydraulic fracture. Moreover, penetration extent of the hydraulic
0
0.5
1
1.5
2
2.5
3
0 0.5 1 1.5 2 2.5 3
P
r
e
d
i
c
t
e
d
s
h
a
d
o
w
(
S
H
_
S
8
0
/
c
)
N
u
m
Numericalshadow(SH_S
80
/c)
pre
(SH
80
/c)
Num
=1.0031(SH
80
/c)
pre
0.0584
R=0.9832
53
fracture into the bounding layers has been assumed to be between 0.1-0.3 of the hydraulic
fracture height. It is important to mention that for modeling of hydraulic fracture
penetration into the bounding layers, fracture geometry was not changed, but the net play
thickness was assumed to be smaller (see Figure 3.1). Therefore, bounding layers were
becoming effective on the upper and lower tips of the hydraulic fracture.
4.5.1. Change of stress shadow by uncontainment of the fracture
The shadow sizes of this study (uncontained fracture) were compared with those of the
contained hydraulic fracture by an investigation on the shadow size ratio of uncontained
over those of the contained hydraulic fracture. The generated graphs for this comparison
are shown in Figure 4.11 for two selected threshold angles of 5 and 80.
Figure 4. 11. Stress shadow change by fracture uncontainment
As it can be seen from Figure 4.11, penetration of a hydraulic fracture inside the bounding
layers may result significant shadow size change depending on the ratio of Youngs moduli
of the bounding layers and the net play, and the penetration extent. For the case of lower
Youngs modulus for bounding layers, shadow size increases and the increase is more
significant for higher penetrations. For the circumstance of higher Youngs modulus of
54
bounding layers, on the other hand, shadow size decreases and the decrease is likewise
more significant for higher penetrations. It is important to note that both of the shadow
sizes obtained based on different threshold angles (0
1
= S, 8u) show approximately the
same change. It is worth mentioning that shadow multipliers mentioned in Figure 4.11
work for all the cases of single fracture mentioned in Table 3.1.
4.5.2. Change of aperture by uncontainment of the fracture
In addition to the shadow size change according to the Youngs modulus ratio and
penetration extent, aperture is also influenced by this uncontainment in the same way.
Figure 4.12 shows the change in aperture as a result of penetration of the hydraulic fracture
in the bounding layers.

Figure 4. 12. Aperture change as a result fracture uncontainment
As it can be seen from Figure 4.12, similar to the shadow size change (see Figure 4.11),
aperture multiplier also shows the same behavior and even approximately the same values.
4.5.3. Mathematical equations predicting aperture and shadow size in uncontained single fractures
In this section, we propose a set of equations for prediction of aperture and stress shadow
multipliers as a result of uncontainment of the fractures. The following set of equations can
55
be used for prediction of aperture and shadow multipliers as a result of hydraulic fracture
penetration into the bounding layers with an extent of EP = (E
p
E). EP is the ratio of
penetrated height (E
p
) of the fracture over its total height (E) and it is assumed to be in
the range of 0.1-0.3.
RS
BL
w
=
w
mux
AR<1
(E
n
= E
b
, EP)
w
mux
AR<1
(E
n
= E
b
)
=
C +|E
b
E
n
]
1 +(C + -1)|E
b
E
n
]

(4.9)
RS
BL
SH05
=
SE_S
05
AR<1
(E
n
= E
b
, EP)
SE_S
05
AR<1
(E
n
= E
b
)
=
C +|E
b
E
n
]
1 +(C + -1)|E
b
E
n
]
(4.10)
RS
BL
SH80
=
SE_S
80
AR<1
(E
n
= E
b
, EP)
SE_S
80
AR<1
(E
n
= E
b
)
=
C +|E
b
E
n
]
1 +(C + -1)|E
b
E
n
]
(4.11)
in which RS
BL
w
is uncontainment aperture multiplier, , RS
BL
SH05
, RS
BL
SH80
are uncontainment
shadow multipliers for threshold angles of 5 and 80, and C and are functions of
penetration extent (EP) given by:
C =
1
o
1
+o
2
(EP)
u
3
(4.12)
=
1
o
4
+o
5
(EP)
u
6
(4.13)
The values of coefficients o
1
-o
6
in Eqs. (4.12) and (4.13) are given in Table 4.4 for
aperture, and shadow multipliers with threshold angles of 5 and 80. These sets of values
56
(Table 4.4) together with Eqs. (4.9-4.13) can be used to predict aperture and shadow
multipliers
1
of hydraulic fractures as a result of penetration into the bounding layers.
Table 4. 4. Coefficients for Eqs. (4.12) and (4.13) for standalone fracturing
parameter o
1
o
2
o
3
o
4
o
5
o
6
Aperture 0.45553 -3.06420 3.49308 -0.14382 0.72369 0.52965
Shadow (0
1
= uS) 0.42801 -0.84289 3.41533 -2.39090 2.72864 0.04526
Shadow (0
1
= 8u) 3.18372 -2.89633 0.04991 0.08886 0.81315 1.17125
Based on these numerical results, it is implied that shadow and aperture of a single
hydraulic fracture is highly influenced by its containment.
4.5.4. Reliability of the proposed functions for uncontainment multiplier determination
In order to check the precision of the predicted uncontainment multipliers with respect to
the numerical ones, Figs. 4.13 are plotted which show the reliability of predictions using
these equations.
Figure 4. 13. Prediction of aperture and shadow size change by fracture
uncontainment

1
The term multiplier in this thesis means that it only takes the effect of the extra variable (e.g. uncontainment, or
simultaneous/multistage fracing) into account and it should be multiplied to the basic case (prediction without this variable) to have this
extra effect incorporated in prediction of the parameter under study (aperture, shadow, or propagation potential).
y=1.0006x+0.0011
R=0.9997
0.8
1
1.2
1.4
1.6
0.8 1 1.2 1.4 1.6
P
r
e
d
i
c
t
e
d
a
p
e
r
t
u
r
e
r
a
t
i
o
Numericalapertureratio
y=0.923x+0.0812
R=0.9972
0.8
1
1.2
1.4
1.6
0.8 1 1.2 1.4 1.6
P
r
e
d
i
c
t
e
d
s
h
a
d
o
w
r
a
t
i
o
Numericalshadowratio(
T
=5)
y=1.0015x+0.0009
R=0.9992
0.8
1
1.2
1.4
1.6
0.8 1 1.2 1.4 1.6
P
r
e
d
i
c
t
e
d
s
h
a
d
o
w
r
a
t
i
o
Numericalshadowratio(
T
=80)
57
As it can be seen from Figure 4.13, the ratio of aperture and shadow size of an
uncontained hydraulic fracture (penetrating into the bounding layers) over those of the
contained hydraulic fracture can be predicted using Eqs. (4.9-4.13).
4.5.5. Use of uncontainment multipliers (standalone fractures)
The aperture multiplier obtained from section 4.5.3 (RS
BL
w
) can be simply multiplied with
Eq. (4.1) for aperture prediction of a single uncontained hydraulic fracture. Similarly, for
shadow size change, the obtained multipliers of shadow in Section 4.5.3 (RS
BL
SH05

and RS
BL
SH80
) should be multiplied with Eqs. (4.4,4.8) respectively for prediction of Shadow
size for minimum and maximum threshold angles (0
1
= S, 8u).
4.6. Effect of simultaneous fracturing on shadow size and aperture
Simultaneous hydraulic fracturing of multilateral wells has been proven to be an efficient
way of having a better fracture network for hydrocarbon flow in unconventional shales. In
this part, two contained hydraulic fractures are assumed from two parallel horizontal wells
(see Figure 3.1; Scenario 3). It is assumed that there is no offset between the crack tips,
which means that the two fractures are aligned on a single plane and their distance is only
changed. The distance change has been assumed by changing of the distance between the
wells, not the aspect ratio or length of the fractures. Referring again to the presented
numerical scheme in Table 3.1, one can see that an extensive number of numerical analyses
are required for this purpose.
4.6.1. Change of stress shadow size by simultaneous fracturing (0
1
= S)
According to the fact that stress shadow has been satisfactorily predicted for a single
contained hydraulic fracture, the best approach was determined to merely compare the
58
shadow size of contained simultaneous fractures with that of a single contained fracture
(comparing Scenario 3 with Scenario 1; see Figure 3.1). Therefore, multiplying a multiplier
to the single fracture equation will result the shadow for simultaneous fractures.
In order to do so, the ratios of SE_
05
AR
SE_S
05
AR
and SE_
80
AR
SE_S
80
AR
were investigated
in detail in order to come up with the multiplying factors to be multiplied with the shadow
size around a contained fracture in a standalone well to predict the shadow size around
simultaneous contained fractures. It is worth indicating that SE_ defines the shadow size
for simultaneous fractures, while SE_S is the shadow size for a single fracture.
According to the numerical results, it was observed that hydraulic fracture aspect ratio
again plays an important role in the magnitude of multiplying factor for reduction of
shadow size around simultaneous fractures. From a statistical analysis on the ratio of
shadow size of simultaneous over single fractures, it was evident that this ratio is only
changing by the aspect ratio and distance between hydraulic fractures (for threshold angle
of (0
1
= S)). Figure 4.14 shows the results of this analysis for simultaneous fractures with
crack tip distances ranging from 0.07 c to 3.0 c.
Figure 4. 14. Shadow reduction by simultaneous fracturing
0
0.2
0.4
0.6
0.8
1
1.2
0 0.5 1 1.5 2 2.5 3
M
e
a
n
S
h
a
d
o
w
r
a
t
i
o
(
S
H
_
D
/
S
H
_
S
)
Normalizedtipdistance,(x/c)
(AR=1.0)
(AR=0.8)
(AR=0.6)
(AR=0.4)
(AR=0.2)
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0 0.5 1 1.5 2 2.5 3
S
D
o
f
S
h
a
d
o
w
r
a
t
i
o
(
S
H
_
D
/
S
H
_
S
)
Normalizedtipdistance,(x/c)
(AR=1.0)
(AR=0.8)
(AR=0.6)
(AR=0.4)
(AR=0.2)
59
As it can be seen from Figure 4.14, shadow ratio of (SE_)(SE_S) starts from 0.69 for a
normalized distance of 0.07 (aspect ratio of unity) and it becomes larger by decreasing the
aspect ratio of hydraulic fractures. As an example, shadow ratio starts from 0.83 for
normalized distance of 0.07 (aspect ratio of 0.2). Another important point about this
analysis is that approaching of shadow ratio to unity for hydraulic fractures with lower
aspect ratios is seen in lower distances between the tips. This means that the lower the
aspect ratio is, the closer the fracture tips must be together to have reduction effect of
stress shadow on each other.
Since these values are determined by a statistical analysis, it is beneficial to also report the
associated standard deviations (SD) for the analyses. As it is evident from Figure 14.4b,
SDs of the analyses are in between 0.015-0.060 and show a more or less decreasing trend
by increasing the distance between the hydraulic fractures.
4.6.2. Mathematical equation for prediction of simultaneous multiplier
After these interpretations, it is also required to have a robust equation for shadow
reduction ratio when we are dealing with simultaneous contained hydraulic fractures. It was
feasible to develop a function to predict the simultaneous multiplier with the numerical
results. Eq. (4.14) shows that the multiplier can be determined for all aspect ratios.
RS
P
SH05
=
SE_
05
AR
SE_S
05
AR
= |1 -o
1
c
-u
2
(IS1c)
]
(4.14)
It can be seen from Eq.(4.14) that there are two coefficients for this function, which can be
obtained for different aspect ratios from Table 4.5.
60
It is worth indicating that the coefficients for any aspect ratio can be determined via a
linear interpolation between those of the boundary aspect ratios.
Table 4. 5. Coefficient of the function in Eq. (4.14)
AR o
1
o
2
1.0 0.33522 1.05700
0.8 0.31354 1.40230
0.6 0.29420 1.99715
0.4 0.27786 3.41253
0.2 0.27772 7.38754

4.6.3. Reliability of the proposed equation for simultaneous multiplier prediction
In this section, the numerically obtained multipliers are compared with the mathematically
predicted ones using Eq. (4.14). The calculated simultaneous multipliers with the proposed
function are shown in Figure 4.15.
Figure 4. 15. Prediction of shadow size change by simultaneous hydraulic fracturing
It can be seen from Figure 4.15 that Eq. (4.14) predicts the shadow change satisfactorily. It
is important to notice that all these discussions are stress shadow change for threshold
angle of 5. In the following section, change of minimum shadow size is investigated.
y=1.0326x 0.0265
R=0.9969
0.6
0.7
0.8
0.9
1
0.6 0.7 0.8 0.9 1
P
r
e
d
i
c
t
e
d
s
h
a
d
o
w
c
h
a
n
g
e
NumericalShadowchange
61
4.6.4. Change of stress shadow size by simultaneous fracturing (0
1
= 8u)
According to the numerical results, it was observed that simultaneous hydraulic fracturing
of parallel laterals influences the minimum shadow size (0
1
= 8u) not as significantly as
that of the threshold angle of 5. In fact, shadow size change for threshold angle of 80 is
much smaller compared to that for threshold angle of 5. The results of this study are
shown in Table 4.6.
It is evident from Table 4.6 that the decrease in shadow is from 0.93 (for aspect ratio of
unity) to almost no change (for aspect ratio of 0.2).
It is worth mentioning that apertures of hydraulic fractures are not influenced by
simultaneous fracturing of parallel wells in case there is no overlap between the hydraulic
fractures. This means that maximum half-aperture of simultaneous fractures can be
determined via using the same equation for hydraulic fractures in standalone wells (see Eq.
(4.1)).
Table 4. 6. Stress shadow change by simultaneous fracturing (0
T
= 8)
Normalized
distance
Simultaneous multipliers (Mean and SD)
(AR=1.0) (AR=0.8) (AR=0.6) (AR=0.4) (AR=0.2)
Mean SD Mean SD Mean SD Mean SD Mean SD
0.07 0.93 0.10 0.93 0.09 0.93 0.08 0.96 0.05 0.99 0.04
0.33 0.91 0.07 0.91 0.08 0.94 0.06 0.97 0.05 1.00 0.03
0.67 0.91 0.05 0.93 0.06 0.94 0.04 0.98 0.04 1.00 0.04
1.00 0.91 0.07 0.94 0.05 0.94 0.05 0.98 0.04 1.00 0.04
1.33 0.92 0.05 0.94 0.05 0.95 0.04 0.97 0.08 1.00 0.04
1.67 0.93 0.04 0.95 0.04 0.97 0.04 0.99 0.04 1.01 0.03
2.00 0.94 0.04 0.96 0.04 0.96 0.04 0.98 0.08 1.00 0.04
2.33 0.93 0.04 0.95 0.04 0.96 0.04 0.97 0.08 1.00 0.03
2.67 0.94 0.04 0.95 0.04 0.96 0.04 0.98 0.08 1.00 0.03
4.7. Effective distance between simultaneous fractures of two parallel wells
In addition to the prediction of the effect of simultaneous hydraulic fracturing on shadow
change, it is also required to show the effective distance between hydraulic fracture tips,
62
beyond which there no effect for shadow change in case of simultaneous hydraulic
fracturing. This boundary distance was determined precisely for different aspect ratios,
which can be seen in Figure 4.16.
It can be seen from Figure 4.16 that there is a distance between the tips of two aligned
hydraulic fractures beyond which there is no effect in stress shadow reduction between
parallel wells.
Figure 4. 16. Effective normalized distance between fracture tips
The effective distance is highly dependent on the aspect ratio of the fracture, which means
by having a fracture with lower aspect ratio, the effective distance is smaller compared to
that of fractures with higher aspect ratios.
4.8. Prediction of shadow change for uncontained simultaneous fractures
In this section, shadow change as a result of having two aligned uncontained hydraulic
fractures penetrated in bounding layers is investigated. It is worth mentioning that this
change is similar to the change of shadow for a single hydraulic fracture penetrating inside
the bounding layers. However, the amount of change is not the same as a single hydraulic
(DIST/c)=1.2964AR
2
+2.0593AR 0.0258
R=0.9997
0
0.5
1
1.5
2
2.5
3
3.5
0 0.2 0.4 0.6 0.8 1 1.2
N
o
r
m
a
l
i
z
e
d
e
f
f
e
c
t
i
v
e
t
i
p
d
i
s
t
a
n
c
e
,
(
D
I
S
T
/
c
)
Fractureaspectratio,AR
63
fracture. This is because we have two shadow changing mechanisms, one because of
having aligned hydraulic fractures and the second is because of the uncontainment.
4.8.1. Mathematical equations for uncontainment multiplier in simultaneous fracturing
According to the numerical results, it was observed that the equations for obtaining
shadow multipliers are given by:
R
BL
SH05
=
SE_
05
AR<1
(E
n
= E
b
, EP)
SE_
05
AR<1
(E
n
= E
b
)
=
C +|E
b
E
n
]
1 +(C + -1)|E
b
E
n
]
(4.15)
R
BL
SH80
=
SE_
80
AR<1
(E
n
= E
b
, EP)
SE_
80
AR<1
(E
n
= E
b
)
=
C +|E
b
E
n
]
1 +(C + -1)|E
b
E
n
]
(4.16)
in which R
BL
SH05
and R
BL
SH80
are uncontainment multipliers of simultaneous fractures for
threshold angles of 5 and 80. The coefficients C and are also related to penetration
extent and can be calculated in the same way as mentioned for uncontainment of single
fractures (see Eq. (4.12) and Eq. (4.13)). It is seen from Eq. (4.12) and Eq. (4.13) that there
are 6 coefficients for C, to be obtained. These coefficients for uncontainment of
simultaneous hydraulic fractures are reported in Table 4.7.
Table 4. 7. Coefficients for Eqs. (4.12) and (4.13) for simultaneous fracturing
parameter o
1
o
2
o
3
o
4
o
5
o
6
Shadow (0
1
= uS) 1.72063 -1.28616 0.04982 0.04256 1.11211 1.22879
Shadow (0
1
= 8u) 0.64112 -1.27348 2.01901 -3.39967 3.96728 0.05015
Using the values reported in Table 4.7 in Eqs. (4.12) and (4.13), C, coefficients can be
determined and Eqs. (4.15) and (4.16) can be used for determination of uncontainment
multiplier of simultaneous fracturing.
64
4.8.2. Reliability of uncontainment multiplier for simultaneous fracturing
The precision of predicted uncontainment multiplier for simultaneous fracturing using Eq.
(4.15) and Eq. (4.16) can be seen in Figure 4.17. It can be seen from Figs. 4.17 that shadow
multiplier for incorporation of uncontainment of the hydraulic fractures in simultaneous
fracturing is predicted satisfactorily using Eq. (4.15) and Eq. (4.16).
Figure 4. 17. Prediction of shadow and aperture change by fracture uncontainment
4.8.3. Use of uncontainment multipliers (simultaneous fractures)
As a result, for the shadow size around hydraulic fractures in simultaneous fracing in which
hydraulic fractures penetrate inside the bounding layers, it is first required to calculate the
shadow size for the single hydraulic fracture (using Eqs. (4.4-4.7)). Then, multiplier of
simultaneous fracing (RS
P
SH05
) should be multiplied with it in order to take the
simultaneous fracing effect into account for calculation of shadow size (Using Eq. (4.14)).
Finally, the multiplier of uncontainment in simultaneous fracturing (R
BL
SH05
and R
BL
SH80
)
should be multiplied with the resultant value in order to have the effect of bounding layers
(using Eq. (4.15) and Eq. (4.16)).
y=0.9983x+0.0032
R=0.9997
0.6
0.8
1
1.2
1.4
1.6
0.6 0.8 1 1.2 1.4 1.6
P
r
e
d
i
c
t
e
d
s
h
a
d
o
w
c
h
a
n
g
e
y=1.0007x 0.005
R=0.9996
0.6
0.8
1
1.2
1.4
1.6
0.6 0.8 1 1.2 1.4 1.6
P
r
e
d
i
c
t
e
d
s
h
a
d
o
w
c
h
a
n
g
e
65
It is again important to indicate that aperture of simultaneous hydraulic fractures is the
same as proposed for single hydraulic fracture. This is because the effect of simultaneous
fracturing is negligible on the aperture of the fracture. Therefore, Eq. (4.1) and Eq. (4.9)
result the aperture of simultaneous uncontained hydraulic fractures as well.

66
Chapter 5: Prediction of Propagation potential in hydraulic fractures
5.1. General
In Chapter 4, a complete set of equations was presented for calculation of stress shadow
around single/simultaneous, contained/uncontained hydraulic fractures with different
aspect ratios, boundary conditions, moduli, and other geometrical considerations. In
addition, aperture of the fractures which plays a key role in proppant type/size
determination and conductivity of the fractures was also predicted for the same scenarios.
However, as mentioned in Chapters 1 and 2, stress shadow and stress intensity factor (SIF)
of hydraulic fractures should be considered together in order to have a satisfactory
judgment about hydraulic fracture optimization. Therefore, in this chapter, effect of
different fracture geometries and their relative positions are considered to study the effect
of fracturing pattern on propagation potential of the fractures. It is important to note that
the same method mentioned in Chapter 3 is used for the study of SIF.
5.2. Effect of aspect ratio of the fractures on the SIF (single fracture)
For the case of a single hydraulic fracture in a standalone well, by changing the aspect ratio,
the SIF along the length and height of the fracture changes. This observation has been
shown in Figure 5.1a. As it is evident from Figure 5.1a, the SIF of the fracture is reduced
by decreasing of aspect ratio in an asymmetric way. This means that the decrease along the
height is not equal to that in length of the fracture (lower change is observed for the
length). This change can also be shown in term of o
1
(defined in Chapter 3; Eq. (3.4)) as
shown in Figure 5.1b. As it is seen in Figure 5.1b, o
1
is the same for length and height of
the fracture when aspect ratio is unity but it starts to diverge to different values along the
67
length and height of the fracture by decreasing of the aspect ratio. The lower curve of o
1
is
indicative of the SIF behavior along the height of the fracture, while the upper curve is
indicative of the SIF along the length of the fracture (see Figure 5.1b).
According to the correspondence between the SIF change and o
1
, it is first induced that
the introduced coefficient suffices for showing the SIF change along edges of the hydraulic
fracture. Therefore, using the introduced coefficient, one can study the behavior of the SIF
along the edge of a fracture only by making a comparison between the fitted o
1
values.
Figure 5. 1. SIF change along fracture edges by different aspect ratios
5.3. Effect of multistage fracturing on the SIF and aperture in standalone wells
In multistage hydraulic fracturing of shales, perforations are created along an interval of the
horizontal wellbore with a pre-calculated spacing. In this section, we are intended to
investigate the effect of adjacent hydraulic fractures from a standalone well on the SIF and
aperture of each fracture. In order to do so, parallel hydraulic fractures with varying
spacing and aspect ratio are assumed and the SIF change with all the available
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
N
o
r
m
a
l
i
z
e
d
n
o
r
m
a
l
s
t
r
e
s
s
,
z
z
z
z
0
)
/
P
H
Normalizedtipposition,(x/c,y/b)
AR=1.0
AR=0.8
AR=0.6
AR=0.4
AR=0.2
1.000.0 1.00
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
0.0 0.2 0.4 0.6 0.8 1.0
C
e
f
f
i
c
i
e
n
t
o
f
t
h
e
f
i
t
t
i
n
g
f
u
n
c
t
i
o
n
,
(
a
1
)
Fractureaspectratio,(b/c)
AlongFractureHeight
AlongFractureLength
(b)
(a)
68
configurations is studied. It is worth indicating that the SIF of each fracture with a certain
aspect ratio with different spacing is compared with the SIF of a single-stage fracture with
the same aspect ratio.
5.3.1. Qualitative description of multistage fracturing influence on SIF/aperture of each fracture
According to the numerical results, it was observed that closely-spaced hydraulic fractures
influence the SIF and aperture in a negative manner, which means that aperture and SIF of
the fractures are lowered compared to the case of efficiently spaced fractures. Using the
defined coefficient in Chapter 3 (o
1
), we can show the change in the SIF along the height
and length of the fracture as shown in Figure 5.2.
Figure 5. 2. SIF change along the edges of the fracture by multistage fracturing
As it can be seen from Figure 5.2, the decrease in the SIF along the height and length of
the fractures shows the effect of spacing between multi-staged fractures on their
propagation potential. This means that having closely-spaced fractures causes the SIF to
decrease significantly and makes the propagation to occur with larger hydraulic pressures.
Having Figure 5.2 seen, one can see that aspect ratio of the fractures also plays an
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0
T
h
e
S
I
F
c
h
a
n
g
e
a
l
o
n
g
t
h
e
h
e
i
g
h
t
Normalizedhalfspacingbeweenfractures,L
p
/c
AR=1.0
AR=0.8
AR=0.6
AR=0.4
AR=0.2
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0
T
h
e
S
I
F
c
h
a
n
g
e
a
l
o
n
g
t
h
e
l
e
n
g
t
h
Normalizedhalfspacingbeweenfractures,L
p
/c
AR=1.0
AR=0.8
AR=0.6
AR=0.4
AR=0.2
69
important role in the SIF change of the adjacent fractures. It is observed that going back to
the original state of SIF (standalone fracture) occurs in a shorter distance by having
fractures with lower aspect ratios. In addition, comparing the SIF change along the height
and length of the hydraulic fracture, it is induced that the SIF along the fracture height is
back to the original state in a shorter distance between adjacent fractures, while this is done
for fracture length in a longer distance.
5.3.2. Mathematical equation for prediction of SIF and aperture change in multistage fracturing
In order to have a quantified SIF change by the spacing between adjacent fractures, the
following equation is proposed:
R
MSP
SIP,w
=
m
1
m
2
+m
3
_
I
p
c
]
m
4
m
2
+_
I
p
c
]
m
4
(5.1)
1
in which I
p
is the half-spacing between adjacent fractures, R
MSP
SIP
is the SIF reduction
multiplier along the height or length of the fracture, R
MSP
w
is the aperture reduction
multiplier, and m
are the coefficients of the function defined for SIF change along the
length/height or aperture of the fracture. Table 5.1 shows the values for the coefficients of
the function. Using the values mentioned in Table 5.1 together with Eq. (5.1), according to
the allowed decrease for the SIF or aperture, one can have a good estimate about the
spacing between adjacent hydraulic fractures. As it can be seen from Table 5.1, coefficients
of Eq. (5.1) are also dependent on the aspect ratio of the fracture. It is worth indicating
that in case aspect ratio of the fracture is different from the values mentioned in Table 5.1,

1
Note that in case R
MSP
is used for aperture, it is R
MSP
w
and in case it is used for SIF change, it becomes
R
MSP
SIP

70
spacing for two bounding aspect ratios are determined and, then by a simple interpolation
between the results, spacing for the desired aspect ratio is calculated.
Usage coefficients AR=1.0 AR=0.8 AR=0.6 AR=0.4 AR=0.2
SIF change along
the height
m
1
0.16447 0.16582 0.16785 0.12873 -0.20650
m
2
0.42449 0.31392 0.20294 0.10429 0.02090
m
3
1.01680 1.01308 1.00820 1.00309 1.00179
m
4
1.92373 1.93651 1.94993 1.87369 1.85405
SIF change along
the length
m
1
0.16447 0.17497 0.18725 0.19689 0.19505
m
2
0.42449 0.37417 0.29515 0.17889 0.05834
m
3
1.01680 1.01373 1.01024 1.00477 1.00275
m
4
1.92373 2.02108 2.12896 2.22962 2.28194
Aperture change
m
1
0.09109 0.09643 0.10355 0.10695 0.17979
m
2
0.46656 0.35414 0.23938 0.13539 0.03863
m
3
1.00792 1.00662 1.00542 1.00338 1.00603
m
4
2.41794 2.42928 2.40773 2.29029 2.35485
Referring to the concept of stress shadow, it was understood that there is always a
disturbed zone around hydraulic fractures which prevents any fracture to propagate in this
region. Therefore, really short spacing between adjacent fractures in this section (e.g. 0.17c)
should only be considered as a device for intuition and better understanding of the
concept. Likewise to the SIF, aperture of the adjacent fractures is also influenced by close
spacing of the fractures. The same procedure applied to calculate the SIF change of
hydraulic fractures was also utilized for estimation of aperture change of the fractures as a
result of having multistage fracturing. The last four rows of Table 5.1 together with the
same equation used for the SIF change (see Eq. (5.1)) can be used for prediction of
aperture change.
5.3.3. Reliability of the proposed equation for SIF and aperture change in multistage fracturing
In order to check the validity of the proposed Eq. (5.1), numerical results are compared
with the mathematically predicted values for the SIF and aperture change of the fractures
in multistage fracturing. The results of this comparison are shown in Figs. 5.3.
71
Figure 5. 3. Prediction of SIF and aperture change by multistage fractures
As it is evident from the figure, SIF and aperture change in multistage fracturing is
satisfactorily predicted using Eq. (5.1).
5.3.4. Visualization of the proposed equation
In this section, the proposed function for prediction of SIF change is visualized by using 3-
D plots. Since aspect ratio and spacing between fractures are input variables and SIF
decrease in percentage is the output function, Figure 5.4 is drawn for the SIF change along
height and length of the fractures. As it is seen from Figure 5.4, the highest decrease of SIF
belongs to the case with the highest aspect ratio and the lowest spacing between adjacent
fractures. Moreover, the SIF change along the height of the fracture is more influenced by
aspect ratio than that along the length of the fracture. One similar surface alike to those of
SIF change with aspect ratio and spacing between fractures also exists for aperture change.
y=1x
R=0.9992
0.0
0.2
0.4
0.6
0.8
1.0
0 0.2 0.4 0.6 0.8 1
P
r
e
d
i
c
t
e
d
S
I
F
m
u
l
t
i
p
l
i
e
r
NumericalSIFmultiplier
y=1x
R=0.9997
0.0
0.2
0.4
0.6
0.8
1.0
0.00 0.20 0.40 0.60 0.80 1.00
P
r
e
d
i
c
t
e
d
a
p
e
r
t
u
r
e
m
u
l
t
i
p
l
i
e
r
Numericalaperturemultiplier
72
Figure 5. 4. Prediction of SIF and aperture change by multistage fracturing
5.4. Effect of simultaneous fracing on the SIF of single-stage fractures
In this section, two wells are assumed to be placed parallel to each other and only one
hydraulic fracture exists for each well. It is also assumed that both of the hydraulic fracture
faces align on a single plane with a certain distance between the tips. The distance between
fracture tips is changed and SIF behavior with respect to the basic case (standalone
fracture) is studied. Likewise to the previous case, o
1
behavior is investigated as a
representative of the SIF change along the fracture height or length. Figure 5.5 shows the
results of this comparison. Of course, in this section, the ratio of SIF for simultaneous
fracture over standalone fracture is reported.
5.4.1. Qualitative description of SIF change via simultaneous fracturing
According to the numerical results, the first observed point is that the meeting edges
(heights) are influenced by each other, while other edges (lengths) do not show any change.
73
As it can be seen from Figure 5.5, the SIF along the height of simultaneous fractures is
controlled by two key variables; distance between the tips and aspect ratio.

Figure 5. 5. SIF change by simultaneous fracturing
It is evident that SIF increases as a result of simultaneous fracturing where fracture tips
meet each other and their tip stress field is affected by that of the other fracture. In
addition, it is also observed that aspect ratio also plays an important role in this SIF
increase in such a way that higher aspect ratio results higher SIF change.
5.4.2. Mathematical equation for prediction of SIF change in simultaneous fracturing (single stage)
In order to have a quantifying equation for SIF ratio of simultaneous fractures versus
standalone fractures in a single staged mode, the following relationship is proposed:
R
SPM
SIP
= exp_
nc
I
s
] (5.2)
in which R
SPM
SIP
is the simultaneous fracing multiplier for SIF change, I
s
is the distance
between fracture tips, c is the fracture half-length, n is the coefficient varying with aspect
ratio of the fractures. Coefficient n can be determined regarding Table 5.2.
74
Aspect ratio
Function coefficient, n in
Eq. (5.2)
1.0 0.08337
0.8 0.07161
0.6 0.05899
0.4 0.04337
0.2 0.03642
Of course, all these values are only valid for single stage fractures without having any
fractures beside each other.
5.4.3. Reliability of the proposed function
The reliability of the proposed Eq. (5.2) for prediction of the SIF change along the meeting
edges is shown in Figure 5.6.
Figure 5. 6. Precision of the predicted SIF ratios using the proposed Eq. (5.2)
As it is seen from Figure 5.6, the predicted SIF changes coincide on the numerical values
satisfactorily. Therefore, the proposed equation can be considered as a reliable equation for
prediction of the SIF change along the meeting edges of rectangular hydraulic fractures.
y=0.9954x+0.0146
R=0.9965
1
1.1
1.2
1.3
1.4
1.5
1 1.1 1.2 1.3 1.4 1.5
P
r
e
d
i
c
t
e
d
S
I
F
r
a
t
i
o
SIFratiofromnumericalresults
75
5.5. Effect of simultaneous multistage fracturing on the SIF of the fractures
For the case of multistage hydraulic fracturing, existence of simultaneous fractures
influence the SIF in a similar way but with different magnitude. It is obvious that
multistage fracturing has negative influence on the SIF (decreasing effect), while
simultaneous fracturing has positive influence on the SIF (increasing effect).
5.5.1. Qualitative description of the SIF change by simultaneous multistage hydraulic fracturing
In order to investigate this effect, simultaneous multistage hydraulic fracturing is
performed and the effect of these two techniques is studied together. It is important to
mention that the ratio of the SIF for simultaneous multistage over single-stage hydraulic
fractures in a standalone well is calculated and used for this analysis. Figure 5.7 is drawn
typically for two aspect ratios of 1.0 and 0.4. The depicted surfaces shown in Figs. 5.7 are
for the SIF change along the height (5.7a, 5.7b) and along the length of the fractures (5.7c,
5.7d).
As it is evident from Figs. 5.7a and 5.7b, the SIF change has been drawn for different
distances between fracture tips and different fracture spacings. The SIF change surface
shows the same behavior for all the aspect ratios, however, we have only plotted for aspect
ratios of 1.0 and 0.4. As it is seen from Figs. 5.7a and 5.7b, the highest SIF increase belongs
to the smallest distance between the tips and the largest spacing. It is also observed from
Figs. 5.7a and 5.7b that as we reduce the spacing between the multi-staged fractures, the
SIF of the fractures is reduced by this decreased spacing. In fact, the decreasing mechanism
of fracture spacing (multistage fracturing) and the increasing mechanism of distance
between fracture tips (simultaneous fracturing) are fighting together, one for decreasing
76
and the other for increasing of the SIF. The results (Figs. 5.7a and 5.7b) show that in case
of having too closely-spaced fractures, simultaneous fracturing has no effect on
propagation of fractures from parallel wells. This means that spacing between hydraulic
fractures should be selected carefully in order to have the increasing effect of simultaneous
fracturing incorporated in the fracturing treatment.
Figure 5. 7. Effect of multistage simultaneous fracturing on the SIF of fractures
(a) (b)
(d) (c)
77
The boundary spacing below which the SIF starts to decrease (no matter how much the
distance between the tips of simultaneous fractures is) varies with aspect ratio of the
fractures in such a way that for lower aspect ratios, the boundary spacing region is shorter
compared to that for higher aspect ratios. As the aspect ratio of the fracture increases, this
boundary spacing for the SIF change becomes larger.
It is also seen from Figs. 5.7c, 5.7d that the SIF along the length of the fracture is not
influenced by the simultaneous fracturing. For the case of spacing between multistage
fractures, however, the SIF along the length of the fracture is reduced. Likewise to the
height of the fractures, the SIF decrease/the decreasing boundary along the length is
higher/larger for larger aspect ratios.
5.5.2. Mathematical equation for prediction of SIF change in multistage simultaneous fracturing
Likewise to the previous sections, it is important to quantify the SIF change according to
the distance between fracture tips and different spacing between fractures in simultaneous
multistage fracturing of the reservoir. For any spacing between the fractures, SIF change as
a result of simultaneous fracturing can be predicted using an appropriate equation for tip
distances up to three times of the half-length of the fracture (Sc) as follows:
R
SPM-MSP
SIP
=
q
1
q
2
+q
3
(I
p
c)
q
4
q
2
+(I
p
c)
q
4
(5.3)
in which R
SPM-MSP
SIP
is the ratio of the SIF for simultaneous multistage fracturing over
standalone fracturing and is estimated using four coefficients as q
for i = 1 -4. The

values of the coefficients for the proposed function are calculated for different tip
distances and are reported in Table 5.3. It is worth indicating that for any other in-range
78
distance, spacing can be calculated for two boundary distances and a simple linear
interpolation can be applied for the desired distance.
Tip distance, (I
s
c) coefficients AR=1.0 AR=0.8 AR=0.6 AR=0.4 AR=0.2
0.25
q
1
0.17201 0.18063 0.18685 0.20748 0.10236
q
2
0.90047 0.64194 0.39699 0.19147 0.05582
q
3
1.34113 1.27754 1.20242 1.12411 1.03763
q
4
1.98234 1.98857 1.96326 1.93938 1.70901
0.50
q
1
0.18998 0.19526 0.19393 0.19476 -0.24732
q
2
0.80502 0.57514 0.35572 0.16592 0.03463
q
3
1.16000 1.12493 1.08375 1.04449 1.00091
q
4
1.84573 1.83626 1.79059 1.77249 1.55430
0.75
q
1
0.17970 0.18134 0.17360 0.16485 -0.19061
q
2
0.69429 0.49660 0.30539 0.13928 0.02931
q
3
1.10209 1.07760 1.05131 1.02491 0.99304
q
4
1.70569 1.70689 1.69860 1.73495 1.66500
1.0
q
1
0.16554 0.16431 0.16126 0.15787 -0.74749
q
2
0.61076 0.43819 0.26749 0.12180 0.02059
q
3
1.07491 1.05268 1.03472 1.01397 0.99449
q
4
1.64114 1.65266 1.68927 1.77491 1.54847
2.0
q
1
0.15221 0.16056 0.17045 0.19478 -0.03603
q
2
0.46249 0.33927 0.21240 0.09934 0.02634
q
3
1.02857 1.02140 1.00862 1.00104 0.99125
q
4
1.69518 1.75410 1.83902 1.98255 1.83895
3.0
q
1
0.16155 0.17036 0.17857 0.21098 0.01983
q
2
0.43019 0.31744 0.20265 0.09400 0.02545
q
3
1.02019 1.01411 1.00731 1.00177 0.99930
q
4
1.78498 1.83897 1.91251 2.05990 1.91793
Table 5.3 shows the values of the function coefficients for different tip distances and
aspect ratios. One practical use of this Table and the proposed Eq. (3.5) is that for an
allowable decrease in SIF, one can have a satisfactory estimation of the spacing for the
multistage fractures in simultaneous multistage fracturing scenario.
5.5.3. Reliability of the proposed equation
Using the values of the coefficients reported in Table 5.3 in Eq. (5.3), one can have the
prediction of the numerical results for the SIF change as a result of simultaneous
multistage fracturing of rock. The reliability of Eq. (5.3) is shown in Figure 5.8.
79
Figure 5. 8. Prediction of SIF ratio of simultaneous multistage fracturing
Figure 5.8 shows that the SIF change as a result of simultaneous multistage fracturing is
satisfactorily predicted using the proposed function.
5.6. Effect of fracture offset on the SIF change in parallel wells
According to the fact that simultaneous fracturing between parallel wells may generate
propagating fractures in between the wells that their meeting tips may have some offset,
effect of offset between the tips should also be studied. Therefore, in this section, hydraulic
fractures with different aspect ratios in simultaneous fracturing of parallel wells are studied
with different distance and offset between the tips. Then, the SIF change is compared with
respect to single-stage hydraulic fractures. The results of this change is typically shown for
AR= 1.0, 0.4 in Figure 5.9.
As it is evident from Figure 5.9, the highest SIF change is when the distance between the
tips is the lowest and there is zero offset between the tips. By increasing the distance and
offset between the tips, SIF change reduces sharply and goes back to its original value in a
standalone well. The magnitude of change, however, is highly dependent on the aspect
y=1x 9E06
R=0.9993
1
1.1
1.2
1.3
1.4
1.5
1 1.1 1.2 1.3 1.4 1.5
P
r
e
d
i
c
t
e
d
S
I
F
r
a
t
i
o
SIFratiofromnumericalresults
80
ratio of the meeting fractures in such a way that by increasing the aspect ratio, the amount
of change is raised. Figs. 5.10 show the SIF change typically for aspect ratio of unity.
Figure 5. 9. Effect of distance and offset on the SIF of the fractures

Figure 5. 10. Effect of fracture offset on SIF change
As it is evident from Figure 5.10a, the highest decrease is for the shortest offsets. This
means that at the lowest values of distance between fracture tips, the highest SIF change
belongs to the ones with the shortest offset magnitude. It is also seen from Figure 5.10b
that by increasing the offset, maximum SIF change is no longer for the shortest distance
1
1.05
1.1
1.15
1.2
1.25
1.3
1.35
1.4
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
r
a
t
i
o
o
f
S
I
F
s
i
m
u
l
t
a
n
e
o
u
s
/
s
t
a
n
d
a
l
o
n
e
f
r
a
c
t
u
r
e
s
Distancebetweenthetips,Ls/c
Lo=0.1c
Lo=0.2c
Lo=0.3c
Lo=0.4c
Lo=0.5c
Lo=0.6c
Lo=0.7c
Lo=0.8c
Lo=0.9c
Lo=1.0c
1
1.02
1.04
1.06
1.08
1.1
1.12
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
S
I
F
r
a
t
i
o
o
f
s
i
m
u
l
t
a
n
e
o
u
s
/
s
t
a
n
d
a
l
o
n
e
f
r
a
c
t
u
r
e
s
Lo=0.4c
Lo=0.5c
Lo=0.6c
Lo=0.7c
Lo=0.8c
Lo=0.9c
Lo=1.0c
81
between the tips. This means that as the offset between the tips increases, maximum SIF
change is seen at larger distances between the tips. The red line on Figure 5.10b shows the
maximum SIF change for different offset values. As it is seen, the highest SIF change is
moved to higher distances between the tips as the offset increases. The highlighted redline
in Figure 5.10b, is slightly shifted to smaller distances between the tips for lower aspect
ratios. Figure 5.11 shows this shifting of maximum values to lower distances between the
tips.
As it is evident from Figure 5.11, some data points are missing for lower aspect ratios. This
is because maximum values of SIF change for these missing points are located at distances
smaller than 0.1c.

Figure 5. 11. SIF maximum line for different aspect ratios
In addition, the gradient of the SIF change after its maximum value is higher for shorter
offset magnitudes. As the offset increases, gradient of SIF change decreases and this causes
the SIF change to be more for higher offset values at higher distances.
1
1.02
1.04
1.06
1.08
1.1
1.12
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
S
I
F
r
a
t
i
o
o
f
s
i
m
u
l
t
a
n
e
o
u
s
/
s
t
a
n
d
a
l
o
n
e
f
r
a
c
t
u
r
e
s
AR=1.0
AR=0.4
AR=0.6
AR=0.8
82
5.7. Use of the SIF change prediction functions
In the preceding sections, a complete set of equations were proposed for prediction of any
change in SIF as a result of using different fracturing technique/pattern. Using these sets
of equations, one can determine the SIF behavior corresponding to the selected fracing
strategy/pattern. Therefore, according to the availability of equipment and maximum
applicable pressure, the allowable SIF change is obtained and based on that, spacing
between adjacent hydraulic fractures is determined.
83
Chapter 6: Work flow for optimization of hydrofracing
6.1. General
In Chapters 4 and 5, the results of a comprehensive study for stress shadow, aperture, and
propagation potential of hydraulic fractures were presented. The results were qualitatively
and quantitatively discussed and complete sets of equations were proposed for prediction
of stress shadow, aperture, and the SIF change in hydraulic fractures. The reliability of all
the equations was also verified using the numerical results. In this chapter, the final work
flow for geomechanical optimization of hydraulic fracturing is proposed according to the
equations proposed in the two preceding chapters.
6.2. Successive procedure of hydraulic fracturing optimization
As indicated in Chapter 1, optimization of hydraulic fractures is performed from different
perspectives/approaches. Any of these perspectives are of paramount importance in their
own scale and influence on production and the cost for completion. In fact, they are not
separated from each other but instead they should be investigated together for an
optimized hydraulic fracturing. Four important steps for the optimization of hydraulic
fractures in unconventional shales are given below:
6.2.1. Step one
In the first step, optimization of hydraulic fracturing should be investigated from a
production point of view. In this method, porosity, permeability, fracture conductivity,
reservoir geometry and its fluid boundary conditions are used in a reservoir simulator for
optimization purpose. In this step, horizontal well spacing, fracturing pattern, and fracture
84
geometry are investigated and an optimized fracturing pattern including fracture geometries
are defined.
However, according to the fact that aperture of hydraulic fractures plays a key role in
conductivity of the fractures, it is first required to have an estimate about the fracture
aperture according to its influencing variables. Aperture is influenced by fracture geometry
(length and aspect ratio), moduli of the medium, excess pressure, and the used fracturing
technique (multistage fracing). From all the above-mentioned influencing factors in
aperture determination, except for the geometry of the fracture which should come from
the flow simulator, all the other variables are geomechanical factors defined in Chapter 4
and 5. The general equation for prediction of aperture for a hydraulic fracture is
determined as:
w
mux
AR<1
(P
N
, v, E
n
)
c
= R
MSP
w
RS
BL
w
RS
AR
w
F(v)
P
N
E
n

(6.1)
Estimation of aperture for a hydraulic fracture is done using the above equation together
with Eqs. (4.2), (4.3), (4.9), (4.12), (4.13), (5.1) and Tables 4.4, 5.1. Then, from the aperture,
proppant size/type, and finally, fracture conductivity is estimated. Therefore, in each try
for production optimization by changing fracture geometry/pattern, hydraulic conductivity
of the fractures should also be updated based on the new aperture.
6.2.2. Step two
In this step, having fracture geometry, stress shadow analysis can be used for determination
of shadow zone around the hydraulic fractures according to the in-situ stress regime, rock
moduli, and excess pressure by using the sets of equations proposed in Chapter 4. The
85
general form of shadow prediction for threshold angle of 5 is written as Eq. (6.2) which is
used for stress shadow size of simultaneous uncontained hydraulic fractures.
SE_S
05
AR=1
c
= R
BL
SH05
RS
P
SH05
RS
AR
SH05

A(v, K
h
) [
P
N
1uP
u
1 +B(v, K
h
) [
P
N
1uP
u

(6.2)
In this equation, Eqs. (4.5), (4.6), (4.7), (4.14), (4.15), (4.7), (4.12), (4.13) and Tables 4.1, 4.5,
4.7 are used to give the value for shadow size of a hydraulic fracture under simultaneous
uncontained hydraulic fracturing.
6.2.3. Step three
In this step, using the fracture geometry from step one, the allowable propagation potential
decrease (decrease of the SIF in the target zone; mentioned in Chapter 5), and considering
the fracing technique used for the job (simultaneous, multistage), the spacing between
hydraulic fractures are determined. For the case of simultaneous multistage fracturing of
rock, Eq. (5.3) together with Table 5.3 can be used for investigation of propagation
potential of this fracing technique.
6.2.4. Step four
In this final step, according to the fracture geometry from step one, and proposed
fracturing patterns from steps one, two, and three, a final optimized fracturing pattern is
obtained by making decision between the three fracturing patterns in the three preceding
steps.

86
Chapter 7: Concluding remarks and recommendations
Unconventional resources, having low permeability, are considered as potential
hydrocarbon reserves because of unconventional architecture; new drilling and completion
technology. Hydraulic fracturing plays a decisive role in economy of these reservoirs since
it makes the best productive regions connected to the wellbore, thereby raising the
permeability of the reservoir. In fracturing treatment, however, it is required to optimize
the fracing job in an efficient way to reduce the cost and make the best optimized fracture
network.
7.1. Aperture analysis
In this part of the thesis, a comprehensive equation was proposed for prediction of
aperture of hydraulic fractures with in-situ stress anisotropy, rock moduli, net pressure,
fracture aspect ratio, containment, and using different fracturing patterns (simultaneous
and/or multistage fracturing) as input variables. It was observed that stress anisotropy does
not have any effect on the aperture of hydraulic fractures. It was also observed that
aperture is inversely related to the rock moduli and directly related to the aspect ratio of the
fracture. Uncontainment of the hydraulic fracture highly influences the aperture of the
fracture, while simultaneous fracing does not have significant effect on the aperture size. In
case of lower Youngs modulus for the bounding layers and more penetration extent,
aperture increases. On the other hand, for the case of higher Youngs modulus for the
bounding layers and higher penetration, a lower aperture is observed. Finally, it was also
shown that aperture of the hydraulic fractures is negatively influenced by multistage
fracturing. In Chapter 5, this decrease in aperture was quantified considering different
spacing between the fractures.
87
7.2. Stress shadow analysis
First, it was shown that shadow mechanism of hydraulic fractures is different depending on
the net pressure magnitude. In addition, shadow size around a hydraulic fracture was also
calculated assuming the same variables used in aperture determination. It was observed that
Youngs modulus of the rock (using constant pressure boundary condition) does not
change the shadow size. Stress anisotropy, Poissons ratio, and net pressure directly
increase the shadow size. Effect of simultaneous fracturing and uncontainment were also
calculated as multipliers to be multiplied with the case of single/contained fractures for
shadow determination for these scenarios. A comprehensive set of equations for shadow
size was also proposed by which shadow size of hydraulic fractures can be predicted
satisfactorily. These equations are really useful in hydraulic fracturing treatment and design
in term of perforation distance and proppant size/type for the fracture.
7.3. Analysis of propagation potential
Based on the consideration that stress intensity factor (SIF) defines the propagation
potential of a hydraulic fracture, a comprehensive numerical simulation framework was
designed in chapter 5 to investigate the interaction between fracture tips and the influence
on the SIF. Different scenarios were considered specifically, fracture aspect ratio, spacing
between multistage fractures, and distance and offset between the tips in simultaneous
fracturing mode. First, it was shown that multistage fracturing in a standalone well
dominantly reduces the SIF of the propagating fractures. The level of this decrease is lower
for the lower aspect ratios. The effect of spacing on the SIF was quantified using a fitting
equation with its proposed coefficients. Secondly, effect of two hydraulic fractures from
two parallel wells (single stage fracturing) was considered to study the effect of
88
simultaneous fracturing of parallel wells on the SIF change. It was observed that SIF of the
meeting hydraulic fractures increases noticeably as a result of simultaneous fracturing. The
magnitude of this change, however, is higher for higher aspect ratios. This effect was also
quantified proposing an equation and its coefficients. Thirdly, the effect of having
simultaneous multistage fracturing of parallel wells were compared with the case of single
stage fracturing of standalone wells in order to quantify the effect of simultaneous
fracturing when multistage fracturing technique is applied. Finally, in order to show the
effect of offset between the fracture tips, effect of offset/distance between the tips were
studied and the behavior of SIF as a result such pattern was also studied. It was observed
that existence of offset between fracture tips is a retarding factor in the SIF increase in
simultaneous fracturing. In addition, for simultaneous fractures with no offset, the lower
the distance between the tips, the higher the SIF increase. For the case of existing offset,
on the other hand, in case offset is more than one-fourth of the fracture length, the highest
SIF change no longer belongs to the least distance and there will be a certain distance in
which SIF is maximum. It is worth mentioning that the range of SIF change in this range
of offset (offset u.4c), is between 1.1 to 1.0 of a single stage fracture in a standalone well.
7.4. Optimized fracture network
An optimized fracture pattern was defined in Chapter 1 as parallel fractures having the
highest production, the highest propagation potential in the target formation, perpendicular
to the wellbore axis, with an optimized distance to prevent any deviation/collapse.
89
7.5. Four steps in optimization of hydraulic fracturing
In order to reach the goal of optimized fracturing pattern, three important aspects from
geomechanical and production perspectives should be considered simultaneously to come
up with the final decision on the fracturing pattern.
It has been stated in the literature that low permeability of the reservoir makes the
fracturing of it essential to produce at an economical rate. The lower the permeability, the
higher the number of fractures should be generated to have efficient production (Soliman
et al., 1997). On the other hand, fractures cannot be placed too close to each other because
of the geomechanical aspect of fracturing since propagation potential may substantially
decrease and fractures may not propagate in the direction perpendicular to the wellbore.
Therefore, the following four steps are suggested in this thesis for an efficient optimization
of hydraulic fractures considering the most influencing parameters into account.
1. In the first step, a primary fracturing pattern is assumed based on experience or
data from adjacent wells. Then, calculating the aperture of the fractures from the
equations proposed in Chapter 4, proppant type/size is selected and estimation on
the fracture conductivity is made. After that, fracturing pattern and fracture
geometry is optimized noting the fact that in each pattern new fracture conductivity
should be calculated based on the new aperture. Finally, in this step, an initial
estimate is given on fracture geometry and fracturing pattern.
2. In the second step, according to the geomechanical data and the obtained fracture
geometry from the first step, stress shadow size around hydraulic fractures with
their own fracing pattern is estimated using the set of equations proposed in
90
Chapter 4. In this step, the second estimate about fracturing pattern; distance
between parallel wells and spacing between multistage fractures is made.
3. In the third step, the estimated fracturing pattern in the second step is used for
determination of the change in propagation potential of the hydraulic fractures
compared to the case of a single staged fracture in a standalone well.
4. In this final step, according to the obtained propagation potential change in the
third step and considering the allowable propagation potential decrease, the best
spacing between adjacent fractures in multistage fracturing and the best distance
between parallel wells is estimated based on the two estimations made in the first
and second steps. Basically, maximum distance obtained based on economical
production (first step), no deviation or collapse (second step), and allowable
propagation potential decrease (third step) is selected as the optimum spacing
between the fractures.
In this thesis, a detailed qualitative and quantitative study of aperture of a hydraulic fracture
required for the first step, stress shadow size for the second step, and change in
propagation potential in the third steps were done in order to come up with a complete
framework for optimization of fracture treatment in unconventional shales.

91
References
1. Abousleiman Y., Tran M., Hoang S., Bobko C., Ortega A., Ulm G.J. 2007.
Geomechanics field and laboratory characterization of Woodford shale: The next gas
play. Paper SPE 110120, presented at SPE annual technical conference and exhibition,
Anheim, CA, USA. 11-14 November.

2. Andrews A., Folger P., Humphries M., Copeland C., Tiemann M., Meltz R., and
Brougher C. 2009. Unconventional Gas Shales: Development, Technology, and Policy
Issues. Congressional research service.

3. Atkinson, C., Smelser, R.E. & Sanchez, J. 1982. Combined mode fracture via the
cracked Brazilian disk test. Int. J. Fract.; 18: 279-291.

4. Barker, L.M. 1977. A simplified method for measuring plane strain fracture toughness.
Eng. Fract. Mech.; 9: 361-369.

5. Barree R.D., Gilbert JV., Conway M.W. 2009. Stress and rock property profiling for
unconventional reservoir stimulation. SPE 118703, presented at SPE hydraulic
fracturing technology conference, The Woodlands, TX, USA, 19-21 January.

6. Britt L.K., Schoeffler J. 2009. The geomechanics of a shale play: what makes a shale
prospective! Paper SPE 125525, presented at SPE Eastern regional meeting,
Charleston, WVA, USA, 23-25 September.

7. Broek D. 1982. Elementary Engineering Fracture Mechanics. Third Revised Edition.
Martinus Nijhoff Publishers, Boston, USA.

8. Chang, S.-H., Lee, C.-I. & Jeon, S. 2002. Measurement of rock fracture toughness
under modes I and II and mixed-mode conditions by using disc-type specimens. Eng.
Geol.; 66: 79-97.

9. Cheng Y. 2009. Boundary Element Analysis of the Stress Distribution around Multiple
Fractures: Implications for the Spacing of Perforation Clusters of Hydraulically
Fractured Horizontal Wells. SPE 125769.

10. Chong, K.P. & Kuruppu, M.D. 1984. New specimen for fracture toughness
determination of rock and other materials. Int. J. Fract.; 26: 59-62.

11. Daneshy A., Au-Yeung J., Thompson T., Tymko D. 2012. Fracture shadowing: A
direct method for determination of the reach and propagation pattern of hydraulic
fractures in horizontal wells. SPE 151980.

12. Dusseault M., McLennan J. 2011. Massive Multistage Hydraulic Fracturing: Where are
We? ARMA (American Rock Mechanics Association) e-Newsletter. Published,
01/2011.
92

13. Evans, A.G. 1972. A method for evaluating the time dependent failure characteristics
of brittle materials and its application to polycrystalline alumina. J. Mater. Sci.; 7: 1137-
1146.

14. Fischer M.P., Gross M.R., Engelder T., Greenfield R.J. 1995. Finite-element analysis of
the stress distribution around a pressurized crack in a layered elastic medium:
implications for the spacing of fluid-driven joints in bedded sedimentary rock.
Tectonophysics 247, 49-64.

15. Fisher, M.K., Heinze, J.R., Harris, C.D., Davidson, B.M, Wright, C.A., and Dunn, K.P.
2004. Optimizing Horizontal Completion Techniques in the Barnett Shale Using
Microseismic Fracture Mapping, paper SPE 90051 presented at the SPE Annual
Technical Conference and Exhibition, Houston, Texas, September 26-29.

16. Fowell, R.J. 1995. Suggested methods for determining Mode I fracture toughness using
cracked chevron notched Brazilian disc specimens. Int. J. Rock Mech. Min. Sci. &
Geomech. Abstr.; 32: 57-64.

17. Green, A.E., Sneddon, I.N. 1950. The distribution of stress in the neighborhood of a
flat elliptical crack in an elastic solid. Mathematical Proceedings of the Cambridge
Philosophical Society. 46(01): p. 159-163.

18. Guidera J.T., and Lardner R.W. 1975. Journal of Elasticity, 5, 59-73.

19. Guo, H., Aziz, N.I. & Schmidt, L.C. 1993. Rock fracture toughness determination by
the Brazilian test. Eng. Geol.; 33: 177-188.

20. Hay J, Sondergeld C.H. 2010. Mechanical Testing of Shale by Instrumented
Indentation. Application Note. Agilent Technologies, Inc. Printed in USA, May 5.
5990-5816EN.

21. Inglis, C.E. 1913. Stresses in a plate due to the presence of cracks and sharp corners.
Transactions of the Institution of Naval Architects, Vol. 55, pp. 219-230.

22. Irwin G. 1957. Analysis of stresses and strains near the end of a crack traversing a
plate, Journal of Applied Mechanics 24, 361364.

23. Janssen M., Zuidema J., Wanhill R. 2009. Fracture Mechanics: Fundamentals and
Applications. Taylor and Francis.

24. Jeffrey, R.G. 1989. The Combined Effect of Fluid Lag and Fracture Toughness on
Hydraulic Fracture Propagation. SPE 18957.

25. Jo H. 2012. Optimizing Fracture Spacing to Induce Complex Fractures in a
Hydraulically Fractured Horizontal Wellbore. SPE 154930.

93
26. Kassir, M. K. 1981. Stress-intensity factor for a three-dimensional rectangular crack.
ASME J. Appl. Mech. 48, 309-312.

27. Keer, L.M. 1964. Quarterly Journal of Mechanics and Applied Mathematics, 17, 423-
436.

28. King G.E. 2010. Thirty years of gas shale fracturing: what have we learned?. Presented
at the SPE annual technical conference and exhibition held in Florence, Italy, 19-22
September.

29. King G.E., Haile L., Shuss J., and Dobkins T.A. 2008. Increasing Fracture Path
Complexity and Controlling Downward Fracture Growth in the Barnett Shale. SPE
119896.

30. Kirsch. 1898. Die Theorie der Elastizitt und die Bedrfnisse der Festigkeitslehre.
Zeitschrift des Vereines deutscher Ingenieure, 42, 797807.

31. Mastrojannis E.N., Keer L.M., Mura, T. 1979. Stress intensity factor for a plane crack
under normal pressure. International Journal of Fracture, Vol. 15, No. 3, June.

32. Morrill, J.C. and Miskimins, J.L. 2012. Optimizing Hydraulic Fracture Spacing in
Unconventional Shales, SPE 152595, SPE Hydraulic Fracturing Technology
Conference, The Woodlands, Texas, February 5-8.

33. Murakami Y. 1987. Stress intensity factors handbook, Volume 1. Pergamon.

34. Muskhelishvili, N.I. 1953. Some basic problems of mathematical theory of elasticity
(1933). English translation, Noordhoff.

35. Mutalik P.N., Gibson B. 2008. Case history of sequential and simultaneous fracturing
of the Barnett shale in Parker county. Presented at the 2008 SPE annual technical
conference and exhibition held in Denver, USA, 21-24 September.

36. Nicolas P. Roussel, SPE, and Mukul M. Sharma. 2011a. Strategies to Minimize Frac
Spacing and Stimulate Natural Fractures in Horizontal Completions. SPE 146104.

37. Ouchterlony, F. 1988. Suggested methods for determining the fracture toughness of
rock. Int. J. Rock Mech. Min. Sci. & Geomech. Abstr.; 25: 71-96.

38. Paris P.C. and Sih G.C. 1965. Stress analysis of cracks. ASTM STP 381, pp. 30-81.

39. Eshelby, J.D. 1968. Stress analysis: Elasticity and fracture mechanics, ISI publications,
121, pp. 13-48.

40. Roussel N.P., and Sharma M. 2011b. Strategies to Minimize Frac Spacing and Stimulate
Natural Fractures in Horizontal Completions. SPE 146104.

94
41. Rafiee M., Soliman M.Y., Pirayesh E. 2012. Hydraulic fracturing design and
optimization: A modification to zipper frac. SPE 159786.

42. Shah R.C. and Kobayashi A.S. 1971. Engineering Fracture Mechanics, 3, 71-96.

43. Sih, G.C. 1973. Handbook of Stress-intensity Factors: Stress-intensity Factor :
Solutions and Formulas for References. Institute of Fracture and Solid Mechanics,
Lehigh University.

44. Singh I, Miskimins J.L. 2010. A Numerical Study of the Effects of Packer-Induced
Stresses and Stress Shadowing on Fracture Initiation and Stimulation of Horizontal
Wells. CSUG/SPE 136856. Presentation at the Canadian Unconventional Resources &
International Petroleum Conference held in Calgary, Alberta, Canada, 1921 October.

45. Sneddon, I.N. and Elliot, H.A. 1946. The Opening of a Griffith Crack under Internal
Pressure. Quarterly of Applied Mathematics 4 (3): 262267.

46. Sneddon I.N. and Lowengrub M. 1969. Crack Problems in the Classical Theory of
Elasticity, J. Wiley and Sons, New York, N.Y.

47. Soliman M. Y., Azari M., Hunt J. L., Chen C.C. 1996. Design and Analysis Of
Fractured Horizontal Wells in gas reservoirs. SPE 35343.

48. Soliman M.Y., Boonen P. 1997. Review of fractured horizontal wells technology. SPE
36289.

49. Sun, Z. & Ouchterlony, F. 1986. Fracture toughness of stripa granite cores. Int. J. Rock
Mech. Min. Sci. & Geomech. Abstr.; 23: 399-409.

50. Tada H., Paris P.C., Irwin G.R. 2000. The stress analysis of cracks handbook. Third
edition. ASME press, New York.

51. US Energy Information Administration. 2011. Annual energy outlook.
(http://www.eia.gov).

52. Warpinski, N.R., Branagan, P.T., Peterson, R.E., Fix, J.E., Uhl, J.E., Engler, B.P., and
Wilmer, R. 1997. Microseismic and Deformation Imaging of Hydraulic Fracture
Growth and Geometry in the C Sand Interval, GRI/DOE M-Site Project, paper SPE
38573 presented at the SPE Annual Technical Conference and Exhibition, San
Antonio, Texas, October 5-8.

53. Waters G., Dean B., and Downie R., Kerrihard K., Austbo L., and McPherson B. 2009.
Simultaneous Hydraulic Fracturing of Adjacent Horizontal Wells in the Woodford
Shale. SPE 119635.

54. Westergaard H.M. 1939. Bearing pressures and cracks. Journal of applied mechanics.
61, pp. A49-A53.

95
55. Whittaker, B.N., Singh, R.N. & Sun, G. 1992. Rock Fracture Mechanics, Principles,
Design and Applications. Developments in Geotechnical Engineering, 71. Elsevier,
Amsterdam.

56. Willis, L.R. 1968. International Journal of Engineering Science, 6, 253-263.

57. Wong S.W., Geilikman M., Xu G. 2013. Interaction of multiple hydraulic fractures in
horizontal wells. SPE 163982.

58. Xu W., Calvez J.L., Thiercelin M. 2009. Characterization of hydraulically-induced
fracture network using treatment and microseismic data in a tight gas formation: A
geomechanical approach Abstract SPE 125237 and slides, presented at SPE tight gas
completions conference San Antonio, TX, USA. 15-17 June.

59. Yew. C.H. 1997. Mechanics of Hydraulic fracturing. Gulf publishing company,
Houston, Texas.

60. Yu W. and Sepehrnoori K. 2013. Optimization of multiple hydraulically fractured
horizontal wells in unconventional gas reservoirs. SPE 164509.

96
Appendix: Variable definition
variable Definition
o
mux
Maximum induced stress
o Applied far-field stress
p Curvature radius
c Crack half-length
P
H
Hydraulic pressure
r, 0 Polar coordinate system
K
I
Stress intensity factor for mode one
o
xx
, o
,
x
Stress components
E Youngs modulus
v Poissons ratio
E Crack height
0
p
Principal stress reorientation
0
Deviation angle
0
1
Threshold angle
w
mux
Maximum half-aperture
w(z) Half-aperture
SE_S
0
Shadow size of a single fracture under threshold angle of 0
SE_
0
Shadow size of simultaneous fractures under threshold angle of 0
K
h
Horizontal stress ratio (stress anisotropy)
E
n
Youngs modulus of the net play
97
Variable Definition
E
b
Youngs modulus of bounding layers
P
N
Excess hydraulic pressure
EP Penetration extent (penetrated fracture height over total height)
S
hmn
Minimum horizontal stress
S
hmux
Maximum horizontal stress
0
mux
Maximum deviation angle
G Geometric function
E Boundary condition function
I
p
Multistage fracturing spacing
I
s
Distance between fracture tips in simultaneous fracturing
AR Aspect ratio of the fracture
RS
AR
w
Aspect ratio multiplier for aperture
C
b
Bulk compressibility
RS
AR
SH05
Aspect ratio multiplier for shadow size determination
A, B Stress shadow coefficients
H, N Stress anisotropy coefficients
ASE_S
05-0
T
AR<1

Stress shadow size difference of threshold angle of 05 from that of
0 for any aspect ratio
RS
BL
w
Uncontainment aperture multiplier for a single fracture
RS
BL
SH05

Uncontainment stress shadow multiplier for a single fracture
(threshold angle of 5)

98
variable Definition
C, Coefficient s of Penetration extent function
RS
BL
SH80

Uncontainment stress shadow multiplier for a single fracture
(threshold angle of 80)
RS
P
SH05
Simultaneous shadow multiplier
R
BL
SH05

Uncontainment multiplier for shadow size of simultaneous
fracturing (threshold angle of 5)
R
BL
SH80

Uncontainment multiplier for shadow size of simultaneous
fracturing (threshold angle of 80)
R
MSP
SIP,w

Multistage fracturing multiplier for aperture and stress intensity
factor change
R
SPM
SIP

Simultaneous fracturing multiplier for stress intensity factor
change
R
SPM-MSP
SIP

Simultaneous multistage fracturing multiplier for stress intensity
factor change
I
o
Offset distance

Ali Taghichian PDF

Enviado por

Dados do documento

Descrição original:

Título original

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Ali Taghichian PDF

Enviado por

Direitos autorais:

Formatos disponíveis

UNIVERSITY OF OKLAHOMA

is observed, v is the Poissons ratio, and E

) can also be obtained based on the following criteria:

) means that we are

) are shown in Table 4.2 for different aspect ratios. It is worth

for i = 1 -4. The

Você também pode gostar