Final Report
Final Report
Final Report
Master Thesis
Spring semester, 2014
Study program/ Specialization:
Writer: ………………………………
Abdullah Tariq (Writer’s signatures)
Faculty supervisor: Mesfin Belayneh
Credits (ECTS): 30
Key words:
Shale Pages: 93
Collapse
Chemical +enclosure: 1
Activity
Stavanger, 16.06.2013
Acknowledgements
I would firstly like to thank Mesfin Belayneh; Professor at the University of Stavanger and Hans
Joakim Skadsem; External Supervisor from IRIS for their supervision and support throughout
my master thesis.
__________________________
i
Abstract
Well instability in shale formations has been a very major problem due to physiochemical
interactions between drilling fluid and formation. In this thesis, chemical, thermal and diffusion
effects on the well collapse strength are evaluated in order to investigate the dominating driving
forces. A case study on the designed Heidrun well program was also performed in order to study
the dynamics of the collapse pressure during drilling phase due to various driving forces.
ii
Table of Contents
ACKNOWLEDGEMENTS ..................................................................................................... I
ABSTRACT .............................................................................................................................. II
NOMENCLATURE ............................................................................................................... IX
1 INTRODUCTION ................................................................................................................ 1
1.1 Background .................................................................................................................. 1
1.2 Problem formulation .................................................................................................... 1
1.3 Objectives .................................................................................................................... 3
iii
3.7.3 Creep failure .................................................................................................... 24
3.7.4 Pore collapse or compaction failure ................................................................ 25
iv
6.2 Effect of different properties ..................................................................................... 63
6.2.1 Wellbore Wall Temperature ............................................................................ 63
6.2.2 Effect of activity values................................................................................... 64
6.2.3 Permeability .................................................................................................... 65
6.2.4 Thermal diffusivity.......................................................................................... 66
6.3 Effect of time ............................................................................................................. 67
6.4 Effect of the inclination of well ................................................................................. 69
6.4.1 Radial Stresses................................................................................................. 69
6.4.2 Hoop Stresses .................................................................................................. 70
6.4.3 Axial Stress ..................................................................................................... 71
6.4.4 Effective Collapse failure ................................................................................ 72
8 CONCLUSION ................................................................................................................. 90
REFERENCES ........................................................................................................................ 91
APPENDIX .............................................................................................................................. 93
v
List of figures
Figure 1: Laminated clay minerals in shale (Institute) ................................................................... 6
Figure 2: Laminated shale (Institute) .............................................................................................. 6
Figure 3: Different types of Shale colors (Geoscience) .................................................................. 7
Figure 4:Categories of Oil shale (Dyni 2006)................................................................................ 8
Figure 5:Permeability of different sedimentary rocks (Jefferson 2011) ....................................... 12
Figure 6:Petro-physical properties of shale along the depth (Okiongbo 2011) ............................ 13
Figure 7: Collapse in pressurized shale formation (SWACO)...................................................... 14
Figure 8: Compressive and tensile failure in shale formation (E. Fjær 2008). ............................ 15
Figure 9: Wellington Shale reduction in swelling with different salts (S.O. Osisanya 1996) ...... 16
Figure 10: Atoka Shale reduction in swelling with different salts (S.O. Osisanya 1996) ............ 16
Figure 11: Effect of pore pressure on brittle-ductile transition (E. Fjær 2008) ............................ 18
Figure 12: Three-dimensional stress state of a cube (Bernt S. Aadnøy 2011) ............................. 18
Figure 13: Different types of faults in the formations (E. Fjær 2008) .......................................... 20
Figure 14: Position of stresses around a wellbore in the rock formation (Bernt S. Aadnøy 2011)
....................................................................................................................................................... 22
Figure 15: Collapse of borehole wall (Mitchell, Miska et al. 2011). ............................................ 24
Figure 16: Principle sketch of stress vs. deformation in a uniaxial compression test (Fjær, Holt et
al. 2008). ....................................................................................................................................... 28
Figure 17: Cohesive Strength as a function of time (Properties taken from Table 6 to plot the
graph on Matlab) ........................................................................................................................... 29
Figure 18: Failure mode for laminated rocks ................................................................................ 37
Figure 19: Loading on laminated rocks ........................................................................................ 37
Figure 20: Comparisons of Arkansas Sandstone Data to Single Plane of Weakness Theory
(Gatlin 1965) ................................................................................................................................. 38
Figure 21: Osmosis process over a semi-permeable membrane (CFCF 2013) ............................ 39
Figure 22: Approximate analytical and implicit solution for the temperature and pore pressure
profiles (C. Chen 2001)................................................................................................................. 42
Figure 23:Pore pressure profile when water is being sucked out of shale formation (Mengjiao Yu
2003) ............................................................................................................................................. 44
Figure 24: Pore pressure profile when water is being sucked in the shale formation (Mengjiao Yu
2003) ............................................................................................................................................. 44
Figure 25: Mud weight effect on pore pressure ............................................................................ 50
Figure 26: Pore pressure profile at different times ....................................................................... 51
Figure 27: Pore Pressure graph for chemical effects .................................................................... 52
Figure 28: Effective Collapse Stress graph for chemical effects .................................................. 52
Figure 29: Effective Collapse Stress graph for no effects ............................................................ 53
Figure 30: Pore Pressure graph for thermal effects ...................................................................... 54
Figure 31: Effective Collapse Stress graph for thermal effects .................................................... 54
vi
Figure 32: Pore Pressure graph for diffusion effects .................................................................... 55
Figure 33: Effective Collapse Stress graph for diffusion effects .................................................. 56
Figure 34: Pore Pressure graph for thermal and diffusion effects ................................................ 57
Figure 35: Effective Collapse Stress graph for thermal and diffusion effects .............................. 57
Figure 36: Pore Pressure graph for chemical and thermal effects ................................................ 58
Figure 37: Effective Collapse Stress graph for chemical and thermal effects .............................. 58
Figure 38: Pore Pressure graph for chemical and diffusion effects .............................................. 59
Figure 39: Effective Collapse Stress graph for chemical and diffusion effects ............................ 59
Figure 40: Pore Pressure graph for chemical, thermal and diffusion effects ................................ 60
Figure 41: Effective Collapse Stress graph for chemical, thermal and diffusion effects ............. 60
Figure 42: Pore Pressure graph for all scenarios .......................................................................... 62
Figure 43: Effective Collapse Stress graph for all scenarios ........................................................ 62
Figure 44: Pore Pressure graph for wall temperature effects ........................................................ 63
Figure 45: Effective Collapse Stress graph for temperature effects ............................................. 64
Figure 46: Pore Pressure graph for activity effects ....................................................................... 64
Figure 47: Effective Collapse Stress graph for activity effects .................................................... 65
Figure 48: Pore Pressure graph for permeability effects .............................................................. 66
Figure 49: Effective Collapse Stress graph for permeability effects ............................................ 66
Figure 50: Pore Pressure graph for thermal diffusivity effects ..................................................... 67
Figure 51: Effective Collapse Stress graph for thermal diffusivity effects .................................. 67
Figure 52: Pore Pressure graph for time effects............................................................................ 68
Figure 53: Effective Collapse Stress graph for time effects ......................................................... 68
Figure 54: Radial stresses around a wellbore at different specific radius values for a vertical well
....................................................................................................................................................... 69
Figure 55: Radial stresses around a wellbore at different specific radius values for a horizontal
well ................................................................................................................................................ 70
Figure 56: Hoop stresses around a wellbore at different specific radius values for a vertical well
....................................................................................................................................................... 70
Figure 57: Hoop stresses around a wellbore at different specific radius values for a horizontal
well ................................................................................................................................................ 71
Figure 58: Axial stresses around a wellbore at different specific radius values for a vertical well
....................................................................................................................................................... 71
Figure 59: Axial stresses around a wellbore at different specific radius values for a horizontal
well ................................................................................................................................................ 72
Figure 60: Effective collapse stresses around a wellbore at different specific radius values for a
vertical well ................................................................................................................................... 72
Figure 61: Effective collapse stresses around a wellbore at different specific radius values for a
horizontal well .............................................................................................................................. 73
Figure 62: Effective collapse stress at different well inclination.................................................. 73
Figure 63: Field location fo Heidrun field (G. Stjern 2003) ......................................................... 74
vii
Figure 64: Prognosis stability plot for a typical Heidrun TLP well (G. Stjern 2003)................... 75
Figure 65: Collapse pressure graph for chemical effects .............................................................. 78
Figure 66: Collapse pressure graph for activity effects ................................................................ 79
Figure 67: Collapse pressure graph for thermal effects ................................................................ 80
Figure 68: Formation and wall temperature profile with respect to depth ................................... 81
Figure 69: Collapse pressure graph for combined thermal and chemical effects ......................... 82
Figure 70: Collapse pressure graph for Biot’s constant effects .................................................... 83
Figure 71: Collapse pressure graph for friction angle effects ....................................................... 84
Figure 72: Collapse pressure graph for Poisson’s ratio effects..................................................... 85
Figure 73: Collapse pressure graph for cohesive strength effects ................................................ 86
Figure 74: Collapse pressure graph for wall temperature effects ................................................. 87
Figure 75: Collapse Pressure graph for change in drilling mud activity ...................................... 89
Figure 76: Collapse pressure graph for different scenarios .......................................................... 89
List of tables
Table 1: Relative characteristics of sedimentary rocks (SWACO) ................................................ 4
Table 2: Mineral composition of a typical shale (Weaver 1965)&(S. Hillier 2006) ...................... 5
Table 3: Mechanical Properties of Shale at different bedding plane angles (Md. Aminul Islam
2013) ............................................................................................................................................. 10
Table 4: Unconfined Strength with respect of depositional environments (E. Eseme 2012) ....... 10
Table 5: Borehole failure criteria for Mohr-Coulomb (E. Fjær 2008) .......................................... 33
Table 6: Mogi-Coulomb equations for collapse failure (A.M. Al-Ajmi 2006) ............................ 35
Table 7: Mogi-Coulomb equations for fracture failure (A.M. Al-Ajmi 2006) ............................. 36
Table 8: List of input parameters for modelling (C. Chen 2001) ................................................. 49
Table 9: Formation and wall temperature data at different depths ............................................... 81
viii
List of abbreviations
EM – Scanning Electron Microscope
FI – Failure Index
ix
1 Introduction
This thesis deals with the well bore stability issues of shale formation with special focus on the
transient effect of pore pressure on the well collapse pressure and stress fields. The main driving
forces are chemical, hydraulic and thermal effects. For the analysis, a Heidrun well program case
study was considered (G. Stjern 2003).
1.1 Background
It is reported that shale makeup 75% of drilling formation where 90% of well instability occurs
(Ronald Steiger 1992). Well instability is a major cost factor for the industry. The wellbore
instability problems increase the overall drilling budget by about 10% (Aadnøy 2003). The
wellbore instability problems are basically well collapse and well fracturing. The problems are
caused by shear and tensile failure mechanisms respectively. Several well stability theoretical
and experimental studies have been done in the industry today. Despite the efforts and more
understanding about the subject, the industry is still facing this problem in shales. This is due to
the complex nature of this formation in terms of its transient phenomenon resulting from the
physiochemical interaction of rock and drilling fluid. Several models have been derived in order
to understand these transient effects in order to adjust the mud weight density accordingly. In a
formation where the drilling window is very small, a slight change in pore pressure may cause
well fracturing or well collapse. During a drilling operation, the change in pore pressure is due to
the physio-chemical interaction between the chemistry of drilling fluid, well pressure and
temperatures. This change in pore pressure and temperature causes a change in the stress
concentrations around the wellbore. Therefore the overall effect is modifying the original well
collapse and fracture strength. The hydraulic diffusion, chemical and thermal effects are transient
effects. Therefore, a good model is needed which is capable of predicting these effects. In this
thesis, the linear elastic and poro-elastic based models are reviewed in order to analyze these
transient effects.
1
formation is very high compared to that of a high permeable formation due to pore pressure not
able to dissipate that easily when in contact with the mud (Jianguo Zhang 2006). Also, for the
chemically-active shale, the water causes the shale to swell which results in a change in the
physical properties of the shale. Experimental studies show changes in strength and young’s
modulus during such chemical interactions (Jianguo Zhang 2006). Effects on the properties of
shale and the critical mud weight density due to temperature is also very important when taking
into account the time delayed or transient changes in the properties of the shale formation. In the
small analysis above, it can be seen that there are a lot of parameters which are taken into
consideration when drilling in the shale formations and these parameters are changing with
respect to time and the human controlled attributes such as the mud weight density are to be
adjusted accordingly when this happens. Also, since due to so much happening at the same time,
coupling all the changes is a troublesome process and an optimal solution is very intricate and
most of the times not possible.
The primary step before drilling is designing of well operation programs. The main objective of
well program is to drill safely without facing well instability problems such as well collapse and
well fracturing. The well pressure should be designed to be within the allowable safe operational
window. The physiochemical rock-fluid interaction effect tends to alter the well program.
This thesis addresses these issues with respect to the collapse pressure gradient and analyses the
following:
• Single effect of hydraulic diffusion, thermal and chemical effects around the wellbore
• Combined effect of hydraulic diffusion, thermal, and chemical around the wellbore
• Dominant driving forces
• Sensitivity analysis of the model parameters with respect to the driving forces
The overall study may assist design well program with respect to drilling fluid chemistry design.
In addition, it will improve understating of time dependent well program conditions.
2
1.3 Objectives
The main objective of this thesis is to analyze the transient wellbore instability caused by
different driving forces. The activities are:
3
2 Literature Study on Shale
2.1 Geology
Sedimentary rocks are a type of rocks that are formed by the deposition of the material on the
earth’s surface or beneath the water bodies. Shale comes under this category of rocks and is
formed by the compaction of silt and clay sized mineral particles commonly known as mud
(Geoscience). They come under the category of mudstones in the sedimentary rocks
classification. Table 1 shows some types of sedimentary rocks with some of their characteristics.
4
Shales are unique in nature compared to other mudstones due to them having a laminated
structure and that they are fissile. This laminated structure is made of thin layers or beds which
are less than one centimeter in thickness as illustrated in scanning electron microscope picture in
Figure 1 and 2.
Shale consists of clay minerals, which accounts for about 50-60% of shale composition and are
therefore the main reason for many problems when drilling through shale formation (Weaver
1965). Clay minerals consist of illite, kaolinite and smectite. Also, the particles which shale
consists are mainly quartz, chert and feldspar (Geoscience). Additional to the grains and these
minerals, shale structure consists of organic matter, carbonate minerals, iron oxide, sulfide
minerals and other heavy mineral constituents which are often present depending on the
environment in which the shales are being deposited (Geoscience). These minerals mostly decide
the color of the shales which they have. Figure 3 shows some of the different colors and laminar
like structure of shales.
The mineral composition of a typical shale is calculated by various authors is shown in Table 2.
Shaw and Weaver (1965) Hillier (2006)
Quartz 30.8 23.9
Feldspar 4.5 3.7 (K-spar)
2.4 (Plag.)
Carbonate 3.6 7.5 (Calcite)
1.3 (Dolomite)
0.5 (Siderite
Fe-oxides 0.5 0.8
Clay minerals 60.9 47.7 (Di-clay)
7.5 (Tri-clay)
Other minerals 2 0.5 (Pyrite)
Organic matter 1 Not determined
Table 2: Mineral composition of a typical shale (Weaver 1965)&(S. Hillier 2006)
5
Figure 1: Laminated clay minerals in shale (Institute)
6
Figure 3: Different types of Shale colors (Geoscience)
Among the different color of shales, black shales are a source of oil and gas. The black color
they get is from the organic matter they consist of. This organic matter after a certain amount of
time and provided that heat is provide to the rock through natural means changes into oil or gas.
Also, shales which usually yield oil and gas are supposed to contain at least 67% of clay minerals
(Geoscience). Other shales which do not come into this category can be broken into small pieces
to be used as a source of clay that can be used for several purposes such as making objects or the
use in cement as a main constituent.
7
A.C. Hutton categorized oil shales into three sections according to their depositional
environments in 1991 (Dyni 2006). Fluorescent microscopy was used to do so. These three
sections of oils shales were subcategorized into further headings. This tree can be seen in Figure
4. Terrestrial oil shales are composed of lipid-rich organic matter such as resin spores, waxy
cuticles, and corky tissue of roots and stems of vascular terrestrial pants found in coal forming
swamps and bogs. Lacustrine oil shales organic matter is derived from the algae found in the
freshwater or saline lakes. Marine oil shales organic matter is derived from the marine algae
(Dyni 2006). These categories as seen in the Figure 4 can are further divided into more
categories. They are named after the different geological places they were discovered at. For
example Torbanite is named after Torbon Hill in Scotland where it was discovered and so on
(Dyni 2006).
8
2.3 Shale Geomechanics
Shale is an anisotropic material and the mechanical properties of it vary in different directions of
the load applied to it. Shale as explained above has bedding planes and the properties parallel to
and perpendicular to this bedding plane are different. Hence it is very important to understand
and know the shale properties in different directions as this can alter the results dramatically
while drilling. Also, it will make a key area to look into when deciding the mud weight density
and the direction in which the well is being drilled. Table 3 shows some of the anisotropic
properties of one of the shale samples tested in the laboratory (Md. Aminul Islam 2013). It can
clearly be seen that the maximum Young’s modulus of the sample was when then load was at 90
degrees to the normal of the bedding plane that is parallel to the bedding plane. On the other
hand, lowest value is observed when this angle is 0 degrees. Note that it is the opposite with the
Poisson’s ratio. Poisson’s ratio is maximum when this angle is zero and minimum when it is at
90 degrees.
Another important factor to look into when looking into shale properties is the confinement
pressure due to the presence of pore fluid in the shale. It is observed through an experiment
where the Young’s modulus of the shale is almost 48% less for the drained sample compared to
that of an undrained one (Md. Aminul Islam 2013). Also the Poisson’s ratio is 40% lower in
value compared to the value of the undrained sample and hence is stiffer than the drained sample
(Md. Aminul Islam 2013). Note that these percentages are for a specific sample and the
percentages and the values will change depending on the sample but the general big decrease in
the values will be seen in the undrained and drained samples. Therefore, getting the values of the
mechanical properties at different bedding angles and also getting the undrained and drained
values is very essential when drilling into a formation. Table 4 below also shows some of the
shales and their unconfined strength with their depositional environments and carbon content
percentages with it.
9
Table 3: Mechanical Properties of Shale at different bedding plane angles (Md. Aminul Islam 2013)
10
2.4 Petro-physical properties
Shales have a very low permeability compared to other source rocks due to narrow pore sized in
the matrix. This can be seen in Figure 5 which shows the permeability of different rock types
(Jefferson 2011). It can easily be se seen in this figure that shale has the lowest permeability
among all of them. Also, shales tend to have low porosities too in general. Another interesting
factor when looking into the petro-physical properties of shale is the change in the permeability
and porosity with the depth. A study was conducted on several wells in North Sea and Figure 6
below shows the results (Okiongbo 2011). It can easily be concluded from Figure 6 that with the
increase in the depth, the porosity of the shale formation tends to decrease due to a large
overburden stress. In general, if porosity decreases, permeability decreases logarithmically with
it (Okiongbo 2011). But another factor which affects the permeability is the pore mean radius
and if the radius increases substantially, it will increase the permeability even if the porosity
decreases. The density of shales is in the range of 2.65-2.8 g/cm3. Note that the density, pore size
and the porosity can be found by different logging tools while drilling into a formation. Once this
data is available, permeability can be found through different models or equations which exist.
The most general equation which is used to find the permeability is the Darcy’s law which states
that the rate at which the fluids flow through a permeable material per unit area is equal to the
permeability. The equation is as below (Bernt S. Aadnøy 2011):
𝑢̇
𝑘=𝜇 2.1
∆𝑃
Where
‘k’ is the permeability, ∆𝑃 is the change in pressure per length, 𝜇 is the dynamic viscosity and 𝑢̇
is the fluid velocity.
Note that the above equation is only valid if the flow is laminar and the fluid is viscous. Also, if
the flow is in a different geometry like a circular borehole of a well, a geometry factor is also
needed for the equation. In order to overcome this problem and also for bedded planes structure
shale, Josef Kozeny and Philip C. Carman derived an equation. The equation known as the
Kozeny-Carman equation is as below (E. Eseme 2012):
11
𝑐𝐾𝐶∙Φ3
𝑘𝐾𝐶 = 2.2
𝑆 2 ∙ (1 − Φ)2
where S is the specific surface area (m²/kg) and Φ is the fractional porosity and 𝑘𝐾𝐶 is in 𝑚2 .
The Kozeny–Carman constant cKC including tortuosity and a generalized factor to account for
different pore shapes was taken as 2.064 × 10−13 m6kg−2.
Note that if the flow is not laminar or if the fluid is not non-Newtonian, the above equations will
not be accurate to calculate permeability and other equations and models are used in that case.
12
Figure 6:Petro-physical properties of shale along the depth (Okiongbo 2011)
• Mechanical stress.
13
• Chemical interactions with the drilling fluid.
Erosion
Hole instability is seen most often as sloughing and caving shale, resulting in hole enlargement,
bridges and fill. The most common consequences are stuck pipe, sidetracks, logging and
interpretation difficulties, and sidewall core recovery difficulties, difficulty running casing, poor
cement jobs and lost circulation.
All contribute to increased costs, the possibility of losing part of the hole or the entire well or
reduced production.
14
Figure 8: Compressive and tensile failure in shale formation (E. Fjær 2008).
It is not always possible to use oil based muds due to environmental and economic concerns as
oil base muds are more prone to environment damage and are also very expensive compared to
water based muds. Hence in order to use water based muds, additives or inhibitors are used to
stop the swelling in shales. Note that the swelling cannot be completely illuminated from the
formation as there is always some hydration in the formation due to contact with water (Oort
2003). Note that different salts can be used in water in order to reduce swelling. Each salt will
have different effect on the swell reduction depending on the diffusivity of the ions in the salt
and the difference in the activity of shale and water base fluid. Also, for different type of shales,
different type of salts will have different effects (S.O. Osisanya 1996). In Figure 9 and 10 below,
it can easily be seen that for different shales, a different salt is good for swell reduction.
Therefore, it is very important to have this type of data when designing a mud for shale
formations.
15
Figure 9: Wellington Shale reduction in swelling with different salts (S.O. Osisanya 1996)
Figure 10: Atoka Shale reduction in swelling with different salts (S.O. Osisanya 1996)
16
3 Theory of Rock Mechanics
𝐹𝑜𝑟𝑐𝑒 𝐹
𝑆𝑡𝑟𝑒𝑠𝑠 = =
𝐴𝑟𝑒𝑎 𝐴
𝐶ℎ𝑎𝑛𝑔𝑒 𝑖𝑛 𝐿𝑒𝑛𝑔𝑡ℎ ∆𝐿
𝑆𝑡𝑟𝑎𝑖𝑛 = =
𝐼𝑛𝑖𝑡𝑖𝑎𝑙 𝐿𝑒𝑛𝑔𝑡ℎ 𝐿
𝑆𝑡𝑟𝑒𝑠𝑠 𝐹𝐿
𝑌𝑜𝑢𝑛𝑔′ 𝑠 𝑀𝑜𝑑𝑢𝑙𝑢𝑠 = =
𝑆𝑡𝑟𝑎𝑖𝑛 𝐴∆𝐿
Any ductile material like steel under a stress will show an elastic behavior. That is the stress will
be directly proportional to the strain till the proportionality limit and on the removal of any load
applied, the material will return to its original position. This is also known as the Hooke’s law.
Once the transition from elastic to plastic takes place, the material goes under the maximum load
known as the ultimate tensile tress. This is the maximum load which can be applied to a material
before its fails. After this point, a material usually becomes weak and fracture. The stress at
which the transition from elastic to plastic region takes place is called the yield point and is
generally the reference point which is taken when noting the strength of the material because
even though ultimate tensile strength is more in magnitude, after this point the material changes
its shape and is not in its original position anymore which is not acceptable for most of the
applications. Figure 11 below shows such behavior of a ductile material. Note that if the material
is brittle such as glass, the material will not show any plastic behavior and will break away right
after the elastic behavior shown. Poisson’s ratio is the ratio of the change in length in the lateral
direction of the force applied to the original length in that direction.
Figure 11 is a typical pore pressure dependent stress-strain behavior of rock material. The curve
shows a complete transition from ductile to brittle behaviour as the pore pressure increased.
17
σ1
σ3
σ1
σ3
Figure 11: Effect of pore pressure on brittle-ductile transition (Bernt S. Aadnøy 2011)
When the object is at rest, that is it is not rotating in any direction, the shear stresses in opposite
directions become equal as shown below.
18
The stress matrix in three dimensional can be shown as below (Bernt S. Aadnøy 2011):
𝜎𝑥 𝜏𝑥𝑦 𝜏𝑥𝑧
𝜏
[𝜎] = � 𝑥𝑦 𝜎𝑦 𝜏𝑦𝑧 �
𝜏𝑥𝑧 𝜏𝑦𝑧 𝜎𝑧
The above matrix coordinate system can be rotated in a direction such that all the shear stresses
disappear. When this happens, only three stress vectors are left in the matrix. These are the three
normal stresses also known as the principal stresses. Principal stresses are very important for
evaluating rock failure since most shear failure criteria involve two or three principal stresses.
The directions of each of them are mutually orthogonal to each other. In an area without a fault
and heterogeneities in structure, the overburden is usually one of the principal stresses. The two
remaining principal stresses are therefore in the horizontal plane.
𝐷
𝜎𝑣 = � 𝜌(𝑧)𝑔𝑑𝑧 3.1
0
19
3.3.2 Horizontal Stresses
When an overburden stress exists in a formation, it will also push the sediment in the horizontal
direction in additional to the vertical squeezing. This will result in horizontal stresses acting on
the sediment too. Note that it is assumed that the rock is isotropic. Also, if the tectonic
movements are assumed to not exist, then the stresses in the horizontal direction due to them are
also excluded. In this case the horizontal in-situ stress is equal to the equation below (E. Fjær
2008):
𝑣
𝜎ℎ = �𝜎 − 𝛼𝑜 𝑃𝑝 � + 𝛼𝑜 𝑃𝑝 3.2
1−𝑣 𝑣
where 𝜎ℎ = minimum horizontal stress, 𝜎𝑣 = overburden, 𝑃𝑝 = pore pressure, 𝛼𝑜 = Biot –
coefficient (set as 1.0 for unconsolidated sands, and 0.9 in shale and consolidated sands). Figure
13 illustrates the schematic in-situ stress and the associated fault systems.
Figure 13: Different types of faults in the formations (E. Fjær 2008)
In stress relaxed environments, the two horizontal stress tensor components 𝜎ℎ and 𝜎𝐻 are equal.
They will change if tectonic movements are taking place. The stress contribution due to these
20
movement or the faults as shown above in Figure 13 are not known. Generally the effect of these
contributions exits and hence the two horizontal stresses are not equal. If all the in-situ stresses
are known, the type of fault which is occurring in the rock can be easily known from the data
above.
1 𝑎2 1 𝑎4 𝑎2
𝜎𝑟 = �𝜎𝑥 + 𝜎𝑦 � �1 − 2 � + �𝜎𝑥 − 𝜎𝑦 � �1 + 3 4 − 4 2 � 𝑐𝑜𝑠2𝜃
2 𝑟 2 𝑟 𝑟
𝑎4 𝑎2 𝑎2
+ 𝜏𝑥𝑦 �1 + 3 4 − 4 2 � sin 2𝜃 + 2 𝑃𝑤
𝑟 𝑟 𝑟
1 𝑎2 1 𝑎4
𝜎𝜃 = �𝜎𝑥 + 𝜎𝑦 � �1 + 2 � − �𝜎𝑥 − 𝜎𝑦 � �1 + 3 4 � cos 2𝜃
2 𝑟 2 𝑟
4 2
𝑎 𝑎
− 𝜏𝑥𝑦 �1 + 3 4 � sin 2𝜃 − 2 𝑃𝑤
𝑟 𝑟
3.3
2 2
𝑎 𝑎
𝜎𝑧 = 𝜎𝑧𝑧 − 2𝑣�𝜎𝑥 − 𝜎𝑦 � 𝑐𝑜𝑠2𝜃 − 4𝑣𝜏𝑥𝑦 2 𝑠𝑖𝑛2𝜃 → 𝑃𝑙𝑎𝑛𝑒 𝑠𝑡𝑟𝑎𝑖𝑛
𝑟2 𝑟
1 𝑎4 𝑎2
𝜏𝑟𝜃 = � �𝜎𝑥 − 𝜎𝑦 �𝑠𝑖𝑛2𝜃 + 𝜏𝑥𝑦 𝑐𝑜𝑠2𝜃� �1 − 3 4 + 2 2 �
2 𝑟 𝑟
2
𝑎
𝜏𝑟𝑧 = (𝜏𝑥𝑦 cos 𝜃 + 𝜏𝑦𝑧 𝑠𝑖𝑛𝜃) �1 − 2 �
𝑟
𝑎2
𝜏𝜃𝑧 = �−𝜏𝑥𝑧 𝑠𝑖𝑛𝜃 + 𝜏𝑦𝑧 𝑐𝑜𝑠𝜃� �1 + 2 �
𝑟
21
Figure 14: Position of stresses around a wellbore in the rock formation (Bernt S. Aadnøy 2011)
σ yy = σ h sin 2 ϕ + σ H cos 2 ϕ
1
τ yz = (σ H − σ h ) sin 2ϕ sin γ
2
1
τ xz = (σ h cos 2 ϕ + σ H sin 2 ϕ − σ v ) sin 2γ
2
22
3.6 Principal Stresses around a wellbore
If the normal and shear stresses are known for a well at the wall of the borehole, the principal
stresses can easily be calculated by the equations below. The stress tensor at the wall of the
wellbore where r = a is given as:
𝜎𝑟 0 0
[𝜎] = � 0 𝜎𝜃 𝜏𝜃𝑧 �
0 𝜏𝑧𝜃 𝜎𝑧𝑧
The equations below show the principal stresses are simply given as (Bernt S. Aadnøy 2011):
σ 1 = Pw
σ2 =
1
(σ θ + σ zz ) +
1
(σ θ − σ zz ) + 4τ θ2z
2
3.5
2 2
σ3 =
1
(σ θ + σ zz ) −
1
(σ θ − σ zz ) + 4τ θ2z
2
2 2
σ 3' = −σ T 3.6
23
3.7.2 Shear or collapse failure
When a rock undergoes a high compressive load, shear failure normally takes place. This failure
takes place due to shear stress along the plane exceeding its shear strength and letting a fault to
create along the plane. This will separate the two planes and slide them against each other. Note
that the shape of the borehole during compressive loading will depend on the loads. That is if the
stresses around the wellbore are uniform and equal, the shape will be like a circle but no changes
will be seen on the shape during deformation as seen in Figure 15. On the other hand, if the
compressive loading is polyaxial, then the shape will also change into an eclipse like shape.
These changes in the wellbore geometry can be observed with the help of the caliper logs.
Collapse failure will occur in both situations. In order to analyze the collapse and other failure
modes, analytical or numerical modeling is used. Two models that are widely used in the
petroleum industry for the collapse failure are Mohr-coulomb criteria and Mogi-coulomb criteria.
24
the deformation completely vanishes. Second state of creep is called the steady state creep.
During this stage, the same things happens as in transient creep but on removing the load, the
deformation does not vanishes completely and there remains a small permanent deformation.
The third state is the final stage known as the accelerating creep. During this stage, the
deformation will exceed rapidly on the exertion of stress and will tend to fracture and hence fail
the material instantly.
25
4 Collapse modeling
As explained above, whenever a material goes a shear failure, collapse occurs. In order to predict
the collapse pressure there exist several collapse failure criteria available in literature. One can
then determine a minimum mud weight to help ensure no well collapse during drilling. The
collapse pressure increases and the fracture pressure decreases as the well inclination increases.
σ h ( MPa) = 0.0053(D )1.145 + 0.46(Po − p fn ) For Depth ‘D’ < 3500m 4.1
p fn = ρ fl gD( water.depth )
3
V
Po = Pobs − ( Pobs − Phyd ) i 4.3
Vn
26
Phyd = Hydrostatic pressure (fluids)
Note that the Pp, Phyd and Vn are calculated by the empirical values data collected for each well.
‘Vi’ is calculated from the seismic data which is collected during well logging. The pore pressure
can be calculated from seismic data collected from well logging or calculated while drilling and
the mud weight can be adjusted while drilling depending on the pore pressure calculated. Also,
instead of putting in the values for velocity, time interval values can also be put in the above
equation to find the pore pressure. In this case, the above will equation will be expressed as
(Eaton 1975)
3
dt
Po = Pobs − ( Pobs − Phyd ) actual
4.4
dt normal
27
Figure 16: Principle sketch of stress vs. deformation in a uniaxial compression test (Fjær, Holt et al. 2008).
Cohesive shear stress also known as the uniaxial compressive strength can be calculated by two
models derived by Horsud (2001) and Lal et al (1999). They used sonic logs to find the above
strength. The models were as below (Horsrud 2001):
2.93
304.8
Per Horsrud (2001): C o [MPa ] = 0.77
∆t (sonic)
4.6
304.8
Lal (1999) : C o [MPa ] = 10 − 1
∆t (sonic)
The cohesive strength also changes with respect to the time. It is as below (Lal 1999):
28
Figure 17: Cohesive Strength as a function of time (Properties taken from Table 6 to plot the graph on
Matlab)
σh
υ= 4.8
σv +σh
where σ h , v = σ h , v − Po
29
4.10 (Lal 1999). ‘ V p ’ below is the sonic wave velocity in km/s and cohesive strength is
measured in MPa.
Vp −1
sin φ = 4.9
Vp +1
5(Vp − 1)
So = 4.10
Vp
30
The above equation can also be written in terms of the two principal stresses which are the
maximum and minimum principal stresses. Also, if the pore pressure effect is considered, the
new poro-elastic model for Mohr-Coulomb failure criteria is given as in equation 4.13 (E. Fjær
2008).
σ 1 − α o Po = C o + (σ 3 − α o Po ) tan α 4.13
Where αo is to denote the Biot’s coefficient and Po is the pore pressure and Co is uniaxial
compression strength. The uniaxial compressive strength can be determined from inherent shear
strength and the internal friction angle φ as (Lianyang Zhang 2010):
2S o cos φ
Co = or Horsrud’s empirical model (Eq4.6) 4.14
1 − sin φ
Note that 4.14 is just another way of expressing equation 4.5 for the same property and either can
be used. Some relations are given below in equation 4.15 for different angles that can be used in
the model above (E. Fjær 2008).
1 + sin φ
tan 2 β =
1 − sin φ
4.15
∅
𝛽 = 45𝑜 +
2
𝑡𝑎𝑛2 𝛽 = 𝑡𝑎𝑛𝛼
It can be seen from the above equation 4.13 for this criterion that it does not take into account the
intermediate principal stress. This results in an underestimation of the rock strength which ends
up in a very conservative collapse pressure curve (A.M. Al-Ajmi 2006). It is also known from
31
different studies that another failure criterion known as Drucker- Prager does very well with a
material under triaxial loading where 𝜎1 ≠ 𝜎2 = 𝜎3 but when the criterion is applied for
polyaxial loading where 𝜎1 ≠ 𝜎2 ≠ 𝜎3 , the test is not that accurate and the results are not a true
representation of the critical mud weights (Lianyang Zhang 2010).
In order to get the well collapse and fracture equations from this model for well borehole, some
assumptions are made. Firstly, it is assumed that the shear stress in the Kirsch’s equations is zero
as it is usually very small and can be neglected compared to other normal stresses. From the
Equations 3.3 above, the normal and shear stresses are calculated at the wall of the borehole
where ‘a’ will be equal to ‘r’ in the equation and the angle 𝜃 is equal to 𝜋/2 to get maximum
values for the stresses. When these values are inserted in Equation 3.3, the stress components
will be reduced to:
σ r' = Pw − Po
4.16
σ 'θ = 3σ x − σ y − Pw − Po
σ ' z = σ z + 2υ (σ x − σ y ) − Po
τ rθ = τ rz = 0
τθz = −2τ xz
Now since all the stresses are known in terms of in-situ stresses, the principal stresses can be
calculated by inserting these stress equations in the equations 3.4 for principal stresses as shown
above.
Assuming a vertical well, isotropic stresses and negligible shear stresses, the principal stresses
can be given as (E. Fjær 2008):
σ r = Pw
σ θ = 2σ h − Pw 4.17
σz =σv
32
From the above equations, it can easily be concluded that if the pressure of the well will change,
it will have an effect on the radial and hoop stresses and the axial stress will have no effect at all
when the well pressure changes. Also, in order for a collapse to happen, hoop stress has to be
greater than the radial stress. That means there will only be three situations when this can
happen. They are when σ θ > 𝜎𝑟 > 𝜎𝑧 , 𝜎𝑧 > σ θ > 𝜎𝑟 and σ θ > 𝜎𝑧 > 𝜎𝑟 (A.M. Al-Ajmi 2006).
Therefore taking these three situations and putting the values of maximum and minimum
principal stresses in the above model, the equations below were derived for well collapse
pressure. Also, for the fracture to happen, the radial stress has to be greater than the hoop stress.
This will again result in only three possible outcomes. A table was formed for all the six
equations.
σθ ≥ σ z > σ r 2(σ h − Po ) − C o
Pw ≤ Po +
1 + tan 2 β 4.18
σ z ≥ σθ > σ r σ v − Po − C o
Pw ≤ Po +
tan 2 β 4.19
σ z ≥ σ r > σθ σ v − Po − C o 4.20
Pw ≥ Po + 2(σ h − Po ) −
tan 2 β
σ r ≥ σθ > σ z Pw ≥ Po + (σ v − Po ) tan 2 β + C o
4.22
33
4.2.2 Mogi-Coulomb criteria
Mogi-Coulomb is said to most accurate model for the failure criteria for sedimentary rocks like
shales. It takes into account all the principal stresses, cohesions strength and the angle of friction.
From the studies and experiment conducted on this criterion, it was shown that the intermediate
principal stress has indeed a strengthening effect on the strength of the rock (A.M. Al-Ajmi
2006). It is also observed that this failure criteria is very accurate for polyaxial loadings too
which is realistically a more common situations for the in-situ stresses. This criterion takes into
account the octahedral shear stress and the effective mean stress. It is expressed as below (A.M.
Al-Ajmi 2006):
′
𝜏𝑜𝑐𝑡 = 𝑎 + 𝑏𝜎𝑚,2 4.24
The above values in the equation can be found by the equations below (A.M. Al-Ajmi 2006):
′
𝜎1′ + 𝜎3′
𝐸𝑓𝑓𝑒𝑐𝑡𝑖𝑣𝑒 𝑀𝑒𝑎𝑛 𝑆𝑡𝑟𝑒𝑠𝑠 = 𝜎𝑚,2 =
2
𝑂𝑐𝑡𝑎ℎ𝑒𝑑𝑒𝑟𝑎𝑙 𝑆𝑡𝑟𝑒𝑠𝑠 = 𝜏𝑜𝑐𝑡
1
= �(𝜎1′ − 𝜎2′ )2 + (𝜎2′ − 𝜎3′ )2 + (𝜎3′ − 𝜎1′ )2
3
4.25
2√2
𝑎= 𝑐 cos ∅
3
2√2
𝑏= sin ∅
3
From the above equations and the equations for the principal stresses, the two tables below are
generated the same way as for Mohr-coulomb criteria. The tables for collapse and fracture are as
below:
34
Borehole failure model (Collapse)
σθ > σ z > σ r
2
1
6
[ 2
]
Pwb = A − 12 (a ' + b ' (A − 2Po ) − 3.(A − 2B) 2
1
4.27
Where A = 3σ x − σ y , (
B = σ z + 2υ σ x − σ y ,) H = A 2 (4b '2 − 3) + (B 2 − AB)(4b '2 − 12)
σ r > σ z > σθ
2
1
6
[ 2
]
Pwb = D + 12 (a ' + b ' (D − 2Po ) − 3.(D − 2E) 2
1
4.30
35
Where, D = 3σ x − σ y , ( )
B = σ z + 2υ σ x − σ y , J = D 2 (4b '2 − 3) + (E 2 − DE)(4b '2 − 12)
𝑒𝑓
4.32
�𝐽2 = 𝐴𝐽1 + 𝐵
where
Failure occurs when the effective collapse stress or the failure index becomes negative
(Mengjiao Yu 2003).
36
4.2.4 Weakness of plane model
Failure model for laminated rocks
Figure 18 shows laminated shale with an angle β, where the weak plane is along this. The shear
collapse failure criteria for laminated layer are different from the non-laminated shales. Figure 19
is the loading on the given laminated rock specimen which consists of plane of weakness
inclined at an angle β from the vertical plane and η is the anisotropic of a rock. The problem of
sliding along a preexisting plane of weakness is illustrated as Figure 18 and 19 (Gatlin 1965).
37
If Sw is the inherent shear strength of the plane of weakness and µw is the coefficient of internal
friction along those planes, then the condition of sliding along these planes can be written as
(J.C. Jaeger 2007):
2(S w + µ wσ 3 )
σ1 = σ 3 +
(1 − cot β .µ w ) sin 2β 4.35
Figure 20: Comparisons of Arkansas Sandstone Data to Single Plane of Weakness Theory (Gatlin 1965)
38
5 Chemical and thermal effects on Collapse Model
Chemical and thermal effects are of immense importance when dealing with shale formations
and they tend to change the collapse and tensile strength significantly when the effects take
place. In the early experiments performed by Mody & Hale (C. Chen 2001), it was seen that the
initial pore pressure was altered due to these effects. The chemical effects were caused due to
osmosis which affects the pore pressure. Osmosis here is the net movement of the solvent into
the area of high solute concentration region through a partially permeable membrane. The
direction in which the solvent moves is the direction of high solute concentration region such
that it tends to equalize the solute concentration between two regions separated by the membrane
(CFCF 2013). Figure 21 below shows a visual illustration of such a process where the container
on left shows the unbalance position and the container on the right shows the concentration of
the solute to be equal on both sides.
Osmosis effects are function of two main attributes which are the membrane efficiency and the
activity of water and formation. Chemical effect due to the difference between the shale water
activity and the drilling mud water activity is to be treated as equivalent to the hydraulic potential
in the system (C. Chen 2001). Also the membrane efficiency is calculated by taking the ratio of
39
the observed osmotic pressure to that of the theoretical osmotic pressure. Osmotic pressure here
is the force per unit area required to prevent the water to pass through the semi permeable
membrane into the area of high solute concentration (CFCF 2013). Note that due to the effect on
the pore pressure, these chemical effects therefore have an influence on the wellbore stresses too
due to wellbore stresses dependent on the pore pressure.
Thermal stresses can also be induced in the wellbore due to thermal changes in the well. These
temperature changes can be due to the injection or production of well where fluids with different
temperatures in the annulus tend to change the all temperature at the wellbore. The pore pressure
can also change because of the volumetric expansion in the rock matric and the pore fluid and is
transient in nature for a low permeable such as shale (Ewy 2005). The thermal and chemical
effects can be coupled into thermoporoelastic equations for a radial system such as a wellbore.
The details of these equations can be seen below in the report.
There are two main equations which have been introduced which take into account the chemical
and temperature effects for a radial system for permeable and non-permeable formations. They
are as below (C. Chen 2001):
𝜕𝑇 𝜕 2 𝑇 1 𝜕𝑇 ′
𝜕𝑇 𝜕𝑃 𝜕 2 𝑃 1 𝜕𝑃
= 𝑐𝑜 � 2 + � + 𝑐𝑜 � +𝑇� 2 + �� 5.1
𝜕𝑡 𝜕𝑟 𝑟 𝜕𝑟 𝜕𝑟 𝜕𝑟 𝜕𝑟 𝑟 𝜕𝑟
𝜕𝑃 𝜕 2 𝑃 1 𝜕𝑃 𝜕𝑇
= 𝑐� 2 + � + 𝑐′
𝜕𝑡 𝜕𝑟 𝑟 𝜕𝑟 𝜕𝑡 5.2
Where c= hydraulic fluid diffusivity; 𝑐𝑜 = thermal diffusivity of the porous medium; 𝑐𝑜′ =
coupling coefficient; 𝑐 ′ = coupling coefficient; T= temperature, P= pore pressure
40
Hydraulic and thermal diffusivity can also be expressed by the equations below (Venkanna
2010) & (Cosse 1993).
𝑘
𝑐=
𝜑𝜇𝑐𝑡 5.3
Where k is the permeability, ϕ is the porosity, µ is the dynamic viscosity and ct is the
compressibility of fluid.
𝑘𝑜
𝑐𝑜 =
𝜌𝑐𝑝 5.4
Where 𝑘𝑜 is the thermal conductivity, 𝜌 is the density and 𝑐𝑝 is the specific heat capacity.
In the equations above, the first terms with the 𝑐𝑜 in equation 5.1 is the temperature change due
to heat conduction. The first term with 𝑐𝑜′ shows the temperature changes due to heat convention
and the second term show the temperature change due to pressure diffusion. In equation 5.2, the
terms with c shows the pore pressure diffusivity due to net hydraulic force and the term with 𝑐 ′
shows the pore pressure change due to temperature variation (Ewy 2005).
In order to generate a less complicated solution to the above equation, some assumptions were
taken to get a simplified solution. Firstly, for low permeable formations such as Shale, the
convection term above in equation 5.1 can be neglected as it is very small. Also, the coupling
coefficient in 5.1 is very small too compared to the thermal diffusivity and hence the whole term
with this coupling coefficient can be neglected (C. Chen 2001). In this case the above equations
can be partially decoupled and can be expressed as equations 5.5 and 5.6. In order to ensure the
validity of the above assumption, a numerically solved solution was plotted for the equations 5.1
and 5.2 and it was compared with their analytical. Figure 22 below shows the visual illustration
41
of numerical and approximated analytical solution generated by Chen & Chenevert (C. Chen
2001). It was seen that the results had an error of only 1% ensuring that the assumptions taken
the heat convection effect and coupling coefficient term equal to zero (C. Chen 2001).
𝜕𝑇 𝜕 2 𝑇 1 𝜕𝑇
= 𝑐𝑜 � 2 + � 5.5
𝜕𝑡 𝜕𝑟 𝑟 𝜕𝑟
𝜕𝑃 𝜕 2 𝑃 1 𝜕𝑃 𝜕𝑇
= 𝑐� 2 + � + 𝑐′ 5.6
𝜕𝑡 𝜕𝑟 𝑟 𝜕𝑟 𝜕𝑡
Figure 22: Approximate analytical and implicit solution for the temperature and pore pressure profiles (C.
Chen 2001)
42
5.1 Chemical Potential
In order to find the chemical potential or the osmotic pressure increase in a system, Mody &
Hale came up with a model which is as below in Equation 5.7 (C. Chen 2001). Note that the
membrane efficiency below is coupled in the equation since shale does not have a permeable
membrane during the interaction with water. Therefore, in order to take into account the partial
membrane of shale, the membrane efficiency is used. The range of 𝐼𝑚 is between 0.01-0.1 for
shale. Note that the direction of water movement in or out of the shale depends on the activity of
the two systems in the well. These are the mud weight water activity and the shale water activity.
The activity here is the ratio of vapor pressure of a given liquid divided by the vapor pressure of
pure water (Soroush 2013). Activity of any fluid can be controlled by the amount of salt content
being put in it which is the salinity. The higher the amount of salt concentration is in a fluid, the
lower will be the activity of it (Soroush 2013). The flow of the water will always will be in the
direction of the higher activity to low activity and hence the values of activity decide on whether
there will be an increase or decrease in pore pressure at the wellbore wall due to the osmotic
effect.
If there is an increase in pore pressure, that will decrease the effective stresses in the wellbore
which will result in reduced collapse strength and if the pore pressure will decrease the result
will be the opposite. Also, if there is an increase in pore pressure due to the chemical effects,
then the pore pressure will tend to increase if moved away from the wellbore but it will
eventually reduce if the pore pressure profile is observed away from the wellbore wall. This is
due to the balancing of the hydraulic effect and the osmotic effect and then the hydraulic effect
surpassing this effect resulting in a decrease in pore pressure (Mengjiao Yu 2003). This combine
effect of hydraulic and osmotic effect can be seen in Figure 23 and 24 below. It can be seen that
when the water is flowing into the formation due to osmotic effect in Figure 24, the pore pressure
tends to increase till a certain point after which it starts decreasing again and then when the water
is coming out, there is just a decrease in pore pressure when moved away from wellbore wall.
Also note that the pore pressure is increased at the wall whenever the mud weight density is
increased due to a higher hydraulic force on the wall due to the difference in the initial pore
pressure and the well pressure.
43
𝑅𝑇 𝑎𝑤𝑚
𝑃𝜋 = −𝐼𝑚 𝑙𝑛 5.7
𝑉 𝑎𝑤𝑠ℎ
Where 𝑃𝜋 is the chemical potential, 𝐼𝑚 is the membrane efficiency, 𝑅 is the gas constant, T is
temperature of the well wall, V is the partial molar volume of water, 𝑎𝑤𝑚 is the activity of
drilling mud and 𝑎𝑤𝑠ℎ is the activity of shale.
Figure 23:Pore pressure profile when water is being sucked out of shale formation (Mengjiao Yu 2003)
Figure 24: Pore pressure profile when water is being sucked in the shale formation (Mengjiao Yu 2003)
44
5.2 Rock Temperature
In order to get a solution for equation 5.5, Carslaw & Jaeger (1959) came up with an analytical
solution for this equation for short time and distance. The solution was as below (C. Chen 2001):
𝑟𝑤 𝑟 − 𝑟𝑤
𝑇(𝑟, 𝑡) = 𝑇𝑜 + (𝑇𝑤 − 𝑇𝑜 )� 𝑒𝑟𝑓𝑐 � � 5.8
𝑟 2�𝑐𝑜 𝑡
Where erfc is the complementary error function, 𝑇𝑜 is the formation temperature, 𝑇𝑤 is the well
wall temperature, 𝑟𝑤 is the well radius, r is the distance away from the wellbore center, t is the
time and 𝑐𝑜 is the thermal diffusivity.
With the use of the above equation, the temperature profile could be known with respect the
wellbore radius and specific radius and at any time for small distances and time. These
temperature changes will have an effect on the effective stresses and hence the collapse strength.
45
𝑟𝑤 𝑟 − 𝑟𝑤
𝑝(𝑟, 𝑡) = 𝑝𝑜 + (𝑝𝑤 − 𝑝𝜋 − 𝑝𝑜 )� 𝑒𝑟𝑓𝑐 � �
𝑟 2√𝑐𝑡
𝑐 ′ (𝑇𝑤 − 𝑇𝑜 ) 𝑟𝑤 𝑟 − 𝑟𝑤
− � �𝑒𝑟𝑓𝑐 � �
1 − 𝑐/𝑐𝑜 𝑟 2√𝑐𝑡 5.9
𝑟 − 𝑟𝑤
− 𝑒𝑟𝑓𝑐 � ��
2�𝑐𝑜 𝑡
Where 𝑝𝑤 is the well pressure, 𝑝𝜋 is the chemical potential and 𝑝𝑜 is the initial pore pressure.
46
𝑟 𝑟
𝛼(1 − 2𝑣) 1 𝐸𝛼𝑚 1
𝜎𝑟𝑟 = 2
� 𝑝 𝑓 (𝑟, 𝑡)𝑟𝑑𝑟 + � 𝑇𝑓 (𝑟, 𝑡)𝑟𝑑𝑟
1−𝑣 𝑟 3(1 − 𝑣) 𝑟 2
𝑟𝑤 𝑟𝑤
𝑟𝑤2
+ 2 𝑃𝑤
𝑟
𝑟
𝛼(1 − 2𝑣) 1 5.10
𝜎𝜃𝜃 =− � 2 � 𝑝 𝑓 (𝑟, 𝑡)𝑟𝑑𝑟 − 𝑝 𝑓 (𝑟, 𝑡)�
1−𝑣 𝑟
𝑟𝑤
𝑟
𝐸𝛼𝑚 1 𝑓 (𝑟, 𝑓 (𝑟,
𝑟𝑤2
− � �𝑇 𝑡)𝑟𝑑𝑟 − 𝑇 𝑡)� − 2 𝑃𝑤
3(1 − 𝑣) 𝑟 2 𝑟
𝑟𝑤
𝜎𝑟𝑟 = 𝑝𝑤
47
6 Simulation of different scenarios for Well Instability
Since the new pore pressure and stresses around the wellbore can be known from the equations
above, any collapse failure model from above can be used with these new stresses and pore
pressures in order to get a new collapse pressure profile and observe the effects of the hydraulic
diffusion, thermal and chemical changes. For the purpose of this report, Drucker-Prager Criteria
is used. The main objective of the simulation study is to learn the effect of various driving forces
on the failure envelope and around the near wellbore about 1.3 times the radius of the well.
During the study, a single and the combined effects of the driving forces will be evaluated.
In order to see the effects of chemical, diffusion and temperature, the pore pressure variation is
observed away from the wellbore to see how these conditions effects the pore pressure and hence
the failure index or the effective collapse strength as shown in equation 4.33. The data to perform
simulation can be seen in Table 8. The time for all the simulations below was taken to be 10
hours and all the changes in pore pressure are known at this specific time with the help of
equation 5.9 above. Equation 5.9 is a complete solution for the chemical, thermal and diffusivity
effects. It can be divided into three different parts in order to get the solutions for only one effect
or combination of two or all three effects. Also, note that the cohesive strength is to be taken as
8.0695 MPa instead of 8.736 MPa for all simulations at time 10 hours. This is to take into
account the change in cohesive strength with respect to time as explained in equation 4.7 above.
The initial pore pressure is taken as 66.94 MPa and both horizontal in-situ stresses are assumed
to be equal. The mud weight pressure for all simulations is taken to be 78.345 MPa. The well is
also assumed to be vertical for the simulations. The effect on the collapse strength due to change
of inclination will be shown later in the report. All the pore pressure profiles are calculated using
‘Matlab’ and the data is put into excel to make relevant graphs. Matlab codes are attached to the
appendix of this report.
The models presented in the section above for pore pressure and temperature doesn’t take into
account the horizontal stresses. In this thesis, these models as explained above are coupled with
the Kirsch’s equations and the results are generated.
48
Table 8: List of input parameters for modeling (C. Chen 2001)
49
6.1 Chemical, Diffusion and Temperature effects
Simulations in Figure 25 and 26 takes into account all effects which are the chemical, thermal
and diffusion. Initially, the mud weight is increased from the initial mud or well pressure of 78.4
MPa and the effect on the pore pressure at the wall and away from it is observed. It can be seen
that pore pressure at the wall increases as the well pressure is increased and decreases in an
exponential manner until it reaches its initial state of 66.94 MPa at some distance away from the
wellbore wall. The increase in the pore pressure due to well pressure was equal to the increase in
mud well pressure from its initial state.
Another simulation was run on Matlab to see the effect of time. It can be seen once the time is
increased, the effects can be seen further away from the wall on the pore pressure and the initial
pore pressure is not even seen at a specific radius of 1.3 for long times as for the case when the
time was 1 minute. At 1 minute, the effects were only seen at the very near wellbore wall and the
pore pressure at any distance away was equal to the initial state.
50
Figure 26: Pore pressure profile at different times
6.1.1 Chemical
In order to observe the chemical effects, equation 5.7 is used. Note that the chemical effects are
only applicable to the wellbore wall and the effect can only be seen when specific radius ‘R/Rw’
is equal to one. This effect can be seen in Figure 27 below. A reduction in initial pore pressure is
seen only on the wall and if moved away from the wall, the pore pressure tends to go back to its
original initial position. Due to the pore pressure decrease at the wall, an increase in the effective
collapse stress is seen in Figure 28. Once the effect is not there anymore, effective collapse stress
graph tends to follow the same path as it had without any effects considered as seen in Figure 29.
Note that since the activity of the formation was higher than the activity of the mud, the
formation fluid was flown out of the formation which resulted in this decrease of the pore
pressure. If the activity would have been higher in the mud compared to that of the shale
formation, an increase in the pore pressure at the wall would have been observed. Such an
observation will be shown below in the report when discussing the effects for different properties
affecting the collapse stress results.
51
67.5
67
66.5
Pore Pressure (MPa)
66
65.5
65
64.5
64
1 1.05 1.1 1.15 1.2 1.25 1.3
R/Rw
15
Effective Collapse Stress (MPa)
14.5
14
13.5
13
12.5
1 1.05 1.1 1.15 1.2 1.25 1.3
R/Rw
52
14.4
14.2
Effective Collapse Stress (MPa)
14
13.8
13.6
13.4
13.2
13
1 1.05 1.1 1.15 1.2 1.25 1.3
R/Rw
6.1.2 Thermal
In order to evaluate the profile of pore pressure due to the temperature effects, equation 5.9 for
pore pressure is used. Note that only the second term in the equation 5.9 is used since it takes
into account the temperature effects. The pore pressure profile is seen in Figure 30. Near
wellbore temperature is lower than the temperature of the formation which results in the thermal
diffusivity of the pore pressure. This results in a decrease in the pore pressure as the position is
moved away from the wellbore wall. Note that there is no effect at the wall since that is the
starting point of the thermal diffusion effect to take place. That is the pore pressure at the wall is
66.94 MPa at all times. Some effects can be seen even till a specific radius of 1.3 as the pore
pressure at this point has still not reached its initial state showing the existence of still some
temperature effects. An increase in effective collapse stress is seen in the Figure 31 below due to
the decrease in the pore pressure till the specific radius of 1.1 after which the gradient of the
slope of effective collapse stress decreases due to a slight increase in pore pressure until the
specific radius of 1.3.
53
67.5
67
66.5
Pore Pressure (MPa)
66
65.5
65
64.5
64
63.5
1 1.05 1.1 1.15 1.2 1.25 1.3
R/Rw
16
15.5
Effective Collapse Stress (MPa)
15
14.5
14
13.5
13
12.5
12
11.5
1 1.05 1.1 1.15 1.2 1.25 1.3
R/Rw
54
6.1.3 Diffusion
In order to observe the hydraulic diffusion affect in the formation, the first part for the equation
5.9 only is taken into account which consist of the net hydraulic force. Since, the well pressure is
higher than the initial pore pressure; an increase in the pore pressure is seen at the wall due to
this hydraulic force. Note that when moved away from the wellbore wall, the pore pressure tends
to reduce exponentially until at a certain point where it equalizes to the initial pore pressure.
Further away from this point, there will be no more diffusivity effect. At 10 hours, this effect
tends to disappear at specific radius of around 1.1. Due to this decrease in pore pressure, the
effective collapse stress is hence increased until the point where this effect takes place after
which the effective collapse curve tends to follow the normal curve. Also, the effective collapse
stress is minimum at the wall due to the increase in pore pressure there and highest in magnitude
at the wall.
80
78
76
Pore Pressure (MPa)
74
72
70
68
66
1 1.05 1.1 1.15 1.2 1.25 1.3
R/Rw
55
16
14
Effective Collapse Stress (MPa)
12
10
0
1 1.05 1.1 1.15 1.2 1.25 1.3
R/Rw
56
80
78
76
74
Pore Pressure (MPa)
72
70
68
66
64
62
60
1 1.05 1.1 1.15 1.2 1.25 1.3
R/Rw
Figure 34: Pore Pressure graph for thermal and diffusion effects
18
16
Effective Collapse Stress (MPa)
14
12
10
8
6
4
2
0
1 1.05 1.1 1.15 1.2 1.25 1.3
R/Rw
Figure 35: Effective Collapse Stress graph for thermal and diffusion effects
57
64.5
64
63.5
Pore Pressure (MPa)
63
62.5
62
61.5
61
1 1.05 1.1 1.15 1.2 1.25 1.3
R/Rw
Figure 36: Pore Pressure graph for chemical and thermal effects
17.5
17
Effective Collapse Stress (MPa)
16.5
16
15.5
15
14.5
14
13.5
13
1 1.05 1.1 1.15 1.2 1.25 1.3
R/Rw
Figure 37: Effective Collapse Stress graph for chemical and thermal effects
58
thermal effects that might be taking place more significantly further away in the wall. This is due
to the very low value for hydraulic diffusivity compared to thermal diffusivity resulting in a very
slow propagation of the pressure into the formation.
77
76
75
74
Pore Pressure (MPa)
73
72
71
70
69
68
67
66
1 1.05 1.1 1.15 1.2 1.25 1.3
R/Rw
Figure 38: Pore Pressure graph for chemical and diffusion effects
16
14
Effective Collpase Stress (MPa)
12
10
0
1 1.05 1.1 1.15 1.2 1.25 1.3
R/Rw
Figure 39: Effective Collapse Stress graph for chemical and diffusion effects
59
6.1.7 Thermal, chemical & diffusion
Below is a pore pressure and effective collapse curve if all the effects are taken into
consideration. They are very similar to the diffusion and temperature graph. The only difference
is a further reduction in pore pressure at the wellbore wall due to chemical change.
78
76
74
Pore Pressure (MPa)
72
70
68
66
64
62
1 1.05 1.1 1.15 1.2 1.25 1.3
R/Rw
Figure 40: Pore Pressure graph for chemical, thermal and diffusion effects
18
16
Effective Collpase Stress (MPa)
14
12
10
8
6
4
2
0
1 1.05 1.1 1.15 1.2 1.25 1.3
R/Rw
Figure 41: Effective Collapse Stress graph for chemical, thermal and diffusion effects
60
6.1.8 Discussion and Comparison of different scenarios
From the above results for pore pressure and effective collapse stress under different scenarios,
some observations were made at the wellbore wall. The pore pressure tends to decrease at the
wall only due to chemical change and no effects are seen away from the wall. For the diffusion
effect, an increase in pore pressure is seen due to the net hydraulic force at the wall due to
overbalance drilling mud conditions. This is due to assumption made by M. E. Chenevert that the
wall is completely penetrating and any increase in well pressure will increase by the magnitude
of the difference of the well pressure and the initial pore pressure (C. Chen 2001). In reality, this
is not the case and because of this the results might be overstating the pore pressure at this stage.
Also, there was no change seen due to temperature effects at the wall. On the other hand, the
effects were seen far away from the well even at specific radius of 1.3 due to the temperature
effects. This was not seen in diffusion or chemical scenarios where the effects stopped taking
place just near the wellbore as the pore pressure tend to return to its initial state. Different pore
pressure profiles for different scenarios can be seen in Figure 42 below.
It can easily be seen from Figure 42 and 43 below that the different effects have quite a
prevailing effect on the pore pressure and the effective collapse stress and hence cannot be
ignored when designing the mud weight window. It can be seen that the effects significantly
either reduce or increase the effective collapse strength depending on their respective effects. It
was seen that all the curves with the diffusion effect in it tend to reduce the effective collapse
strength due to increase in pore pressure due to hydraulic diffusion. Whereas temperature and
chemical effects tend to increase the effective collapse strength or decrease the pore pressure.
61
80
78
76
74 chemical
Pore Pressure (MPa)
72 thermal
70 diffusion
68 thermal+chemical
66 thermal+diffusion
64 thermal+chemical+diffusion
62 normal
60
1 1.05 1.1 1.15 1.2 1.25 1.3
R/Rw
19
17
Effective Collapse Stress (MPa)
15
chemical
13 thermal
diffusion
11 thermal+chemical
thermal+diffusion
9
thermal+chemical+diffusion
7 normal
5
1 1.05 1.1 1.15 1.2 1.25 1.3
R/Rw
62
6.2 Effect of different properties
For the analysis of the effect of different properties, time and inclination of well, all the effects
which are chemical, thermal and diffusion are taken into account when performing simulations.
The normal pore pressure or effective collapse stress is the one which is performed taking initial
parameters which can be seen in Table 8.
78
76
74
Pore Pressure (MPa)
72
70
Normal Pore Pressure
68
66 Increase in T
64 Decrease in T
62
60
1 1.1 1.2 1.3
R/Rw
63
18
14
12 Normal
10 Increase in T
Decrease in T
8
6
1 1.1 1.2 1.3
R/Rw
85
80
Pore Pressure
65
60
1 1.1 1.2 1.3
R/Rw
64
18
6.2.3 Permeability
When the permeability was increased 100 times in order to see the effect on pore pressure and
effective collapse strength, it was seen that the hydraulic diffusivity tends to be more prevailing
now compared to the initial permeability condition. It was seen that the pore pressure tends to
decrease in a smoother manner rather than the drastic change seen in the original curve. The
profile seems to be going a long way away from the wellbore showing that the penetration of this
effect is much higher now. This resulted in evener and reduced effective collapse strength in
nearby wellbore area. The smooth curve of pore pressure tends to increase the pore pressure due
to fluid flowing into formation more easily. Another thing to take into consideration is the
assumptions taken at the start of this report regarding the convection component in the pore
pressure equation 5.1 which was assumed to be zero as the permeability was very low. Since the
permeability is increasing, the convection component might not be very small anymore and
because of this there might be a small error in the analysis below.
65
78
76
74
68
Permeability Increase 5
66 times
64
62
1 1.1 1.2 1.3
R/Rw
16
14
12 Normal
10 Permeability Increase
5 times
8
6
1 1.1 1.2 1.3
R/Rw
66
78
76
68
Thermal Diffusivity
66 increase 5 times
64
62
1 1.1 1.2 1.3
R/Rw
16
Effective Collpase Stress
14
12 Normal
10 Thermal Diffusivity
increase 5 times
8
6
1 1.1 1.2 1.3
R/Rw
Figure 51: Effective Collapse Stress graph for thermal diffusivity effects
67
effect is due to the hydration of shales which weakens it. It can be calculated at different times
using equation 4.7. It can be seen that the change is at the wellbore wall which can be a worrying
factor when designing the mud weight window. Therefore it is very important to take into
account all the changes with time too in order to avoid any well collapse after days from the
initial point of drilling.
78
76
74
Pore Pressure (MPa)
72
70
Normal Pore Pressure
68
Time (10 days)
66
64
62
1 1.1 1.2 1.3
R/Rw
18
Effective Collapse Stress (MPa)
16
14
Normal
12
8
Cohesive Strength
6 effect
4
1 1.1 1.2 1.3
R/Rw
68
6.4 Effect of the inclination of well
Inclination of well has a very strong influence on the stresses around the wellbore as well as the
effective collapse stress. In order to observe the changes in principal stresses around a wellbore
and the effective collapse stress, some results were generated. Below in Figure 54, hoop stresses
for well at zero degrees to the vertical is shown with another one (Figure 55) shown at 90
degrees which will be a horizontal well. It can be seen that for a well at zero inclination, the
stresses around the wellbore at a certain radius is same around it at any angle whereas for a
horizontal well the stresses around the wellbore vary with respect to the angle. The angle here is
the angle with respect to the direction of maximum horizontal in-situ stress. All the stresses are
in MPa. All the simulations are done taking into account the chemical, diffusion and thermal
effects.
-180
17018020 -170-160
160 -150
150 -140
15
140 -130
130 10 -120 rd=1
120 -110 rd=1.05
110 5 -100 rd=1.1
100 0 -90 rd=1.15
90 -80
rd=1.2
80 -70
rd=1.25
70 -60
rd=1.3
60 -50
50 -40
40 -30
30 -20
20 10 0 -10
Figure 54: Radial stresses around a wellbore at different specific radius values for a vertical well
69
-180
170 18020 -170-160
160 -150
150 -140
15
140 -130 rd=1
130 10 -120 rd=1.05
120 -110 rd=1.1
110 5 -100 rd=1.15
100 -90 rd=1.2
0 rd=1.25
90 -80
rd=1.3
80 -70
70 -60
60 -50
50 -40
40 -30
30 -20
20 10 0 -10
Figure 55: Radial stresses around a wellbore at different specific radius values for a horizontal well
-180
170180
25 -170-160
160 -150
150 20 -140
140 -130 rd=1
130 15 -120 rd=1.05
120 10 -110
110 5 -100 rd=1.1
100 0 -90 rd=1.15
90 -80
80 -70 rd=1.2
70 -60 rd=1.25
60 -50
50 -40 rd=1.3
40 -30
30 20 -10 -20
10 0
Figure 56: Hoop stresses around a wellbore at different specific radius values for a vertical well
70
-180
180 30 -170
170 -160
160 -150
150 25 -140
140 20 -130
130 -120
rd=1
15
120
10
-110 rd=1.05
110
5
-100
rd=1.1
100 -90
0 rd=1.15
90 -80
rd=1.2
80 -70
rd=1.25
70 -60
60 -50 rd=1.3
50 -40
40 -30
30 -20
20 10 0 -10
Figure 57: Hoop stresses around a wellbore at different specific radius values for a horizontal well
-180
17018025 -170-160
160 -150
150 20 -140
140 -130
130 15 -120 rd=1
120 10 -110
rd=1.05
110 5 -100
rd=1.1
100 0 -90
rd=1.15
90 -80
rd=1.2
80 -70
rd=1.25
70 -60
60 -50 rd=1.3
50 -40
40 -30
30 20 -20
10 0 -10
Figure 58: Axial stresses around a wellbore at different specific radius values for a vertical well
71
-180
170180
20 -170-160
rd=1
160 -150
150 -140 rd=1.05
140 15 -130
130 -120 rd=1.1
10
120 -110 rd=1.15
110 5 -100 rd=1.2
100 0 -90 rd=1.25
90 -80 rd=1.3
80 -70
70 -60
60 -50
50 -40
40 -30
30 20 -20
10 0 -10
Figure 59: Axial stresses around a wellbore at different specific radius values for a horizontal well
-180
160 17020 -170-160
150 -150
140 15 -140
130 -130
rd=1
120 10 -120
110 -110 rd=1.05
5 rd=1.1
100 -100
90 0 -90 rd=1.15
80 -80 rd=1.2
70 -70 rd=1.25
60 -60
rd=1.3
50 -50
40 -40
30 -30
20 10 -10 -20
0
Figure 60: Effective collapse stresses around a wellbore at different specific radius values for a vertical well
72
-180
160170
20 -170-160
150 -150
140 15 -140
130 -130 rd=1
120 10 -120
rd=1.05
110 -110
5 rd=1.1
100 -100
90 0 -90 rd=1.15
80 -80 rd=1.2
70 -70
rd=1.25
60 -60
50 -50 rd=1.3
40 -40
30 -30
20 10 -10 -20
0
Figure 61: Effective collapse stresses around a wellbore at different specific radius values for a horizontal
well
Figure 62 below shows the effective collapse stress at all well inclination angles. It can easily be
concluded that for any specific radius away or at the wall, the effective collapse stress is a
maximum at well inclination of zero degrees and minimum at a well inclination for 90 degrees. It
tends to decrease slowly as the inclination of the well is increased to 90 degrees as seen below.
18
16
14
Effective Collapse Stress
rd=1
12
rd=1.05
10
rd=1.1
8
rd=1.15
6
rd=1.2
4
rd=1.25
2
rd=1.3
0
0 15 30 45 60 75 90
Inclination of Well (Degrees)
73
7 Case Study
7.1 Introduction
Heidrun field was discovered in 1985 in the Haltenbanken area offshore Mid-Norway (G. Stjern
2003). Figure 63 below shows the location of this field on a map. The field is located in a very
environmentally vulnerable area therefore the use of oil based mud is not possible (G. Stjern
2003). Water based muds are used on the fields which can be a problem for shale as water tend
to swell the shale formation which exits in this field. Therefore, inhibitive like KCl is used in the
water based mud which acted as reactive clay to drill wells on this field. All the wells were
drilled on a Tension Leg Platform (TLP). Most of wells on the field are extended reach wells
with inclination of wells in the range of 55 to 70 degrees (E. Fjær 2008). Also, the salt content in
the mud was also reduced as this helped drilling a stable borehole. Reducing the salt content
tends to increase the mud activity and a change in chemical effect on the pore pressure. A case
study is initiated in order to see the chemical, thermal and poro-elastic effects on the wellbore
wall on one of the wells on the Heidrun field.
74
Before drilling the primary step is the design of the drilling program. The drilling program is
used to avoid well instability problems. These well instabilities are collapse and well fracture.
Figure 64 shows the well program.
Figure 64: Prognosis stability plot for a typical Heidrun TLP well (G. Stjern 2003)
75
7.2 Model used for chemical and temperature
In order to investigate the poro-elastic, chemical and temperature effects on the well collapse
pressure, a model was derived. Initially, Mohr-Coulomb model is used as listed above in
equation 4.18 using stress transformations equation 3.4 to take into account the well inclination
at all depths. In order to take into account the effects, the set of equations in equation 5.11 were
coupled with the model which takes into account all the induced chemical and temperature
stresses. At the wall, there is no change in pore pressure due to temperature. In order to calculate
the change in pore pressure needed for the model, equation 5.9 was used to find the change in
pore pressure. The model derived from for these effects are below for chemical, temperature and
chemical and temperature.
3σ x − σ y − Co − α o P0 (1 − tan α ) + K
Pw ≤ 7.2
1 + tan α − 2η
Where
𝐸𝛼𝑚
𝐾= (𝑇 − 𝑇𝑜 )
3(1 − 𝑣) 𝑤 7.4
𝛼𝑜 (1 − 2𝑣)
𝜂=
1−𝑣 7.5
76
7.3 Sensitivity analysis
In order to perform sensitivity analysis and see which properties influence the well collapse
pressure the most, first the data in Figure 64 above was digitized and Mohr-Coulomb collapse
curve was generated using equation 4.18. The reason for choosing this equation out of the other
equations was that the stresses calculated from the digitized date in Figure 64 were in the order
σ θ ≥ σ z > σ r . Once that was achieved, the above poro-elastic models equations were used to
see the effects of chemical and temperature. All the properties for the analysis were taken from
Table 8 above unless stated otherwise. The activity and the temperature values were changed as
for a temperature, a temperature profile was used all along the well depth and for the activity the
values in Table 8 were very high and were giving unrealistic results. Note that the normal
‘Collapse’ line in all the figures below is the line plotted from the equation 4.18 without any
effects and is the reference line to see any effects. Activity was increased and decreased 5% in
order to see the effect of it on the collapse pressure. All other parameters were increased and
decreased by 10% to observe changes on the collapse pressure.
77
Pressure (sg)
0 0.5 1 1.5 2
0
500
Pore pressure
1000
Overburden
Collapse
Depth (m)
1500
2000
2500
78
Pressure (sg)
0 0.5 1 1.5 2
0
500
Pore pressure
Overburden
Collapse
1000
Collpase chemical (adf > ash)
Depth (m)
1500 Pfrac
2000
2500
79
formation reduces down the depth. This is due to the decrease in wall temperature gradient. For
the wall and formation temperature profile along the depth, a temperature profile was generated.
It can be seen in Figure 68. Digitized data used to generate the temperature profile can also be
seen in Table 9 below.
Pressure (sg)
0 0.5 1 1.5 2
0
500
Pore pressure
1000
Overburden
Collapse
Depth (m)
Thermal Collpase
Pfrac
1500
2000
2500
80
Temperature (C)
0 20 40 60 80 100
0
500
To
1000
Depth (m)
Tw
1500
2000
2500
Figure 68: Formation and wall temperature profile with respect to depth
Depth
(m) To (°) Tw (°)
500 24 50.895
600 28 53.635
700 32 56.235
800 36 58.695
900 40 61.015
1000 44 63.195
1100 48 65.235
1200 52 67.135
1300 56 68.895
1400 60 70.515
1500 64 71.995
1550 66 72.6825
1600 68 73.335
1700 72 74.535
1800 76 75.595
1900 80 76.515
2000 84 77.295
2100 88 77.935
2200 92 78.435
Table 9: Formation and wall temperature data at different depths
81
7.3.3 Temperature and chemical combined effect
When both results were combined, it can easily be concluded that the chemical effects dominate
the temperature effects at the wall. But in the upper part of the formation, the collapse pressure
exceeds even the fracture pressure due to very narrow mud weight window and high chemical
and thermal effects in that region. Therefore it is very vital to evaluate these changes more
thoroughly when generating the final results in the drill program.
Pressure (sg)
0 0.5 1 1.5 2
0
500
Pore pressure
Overburden
1000
Collapse
Depth (m)
Thermal + Chemical
Thermal + Chemical 2
Pfrac
1500
2000
2500
Figure 69: Collapse pressure graph for combined thermal and chemical effects
82
7.3.4 Biot’s constant
Biot’s constant value was increased and decreased by10 percent of its initial value to see the
effect of this property on the collapse curve. As seen below in Figure 70, Biot’s constant has a
very strong effect on the collapse curve.
Pressure (sg)
0 0.5 1 1.5 2
0
500
Pore pressure
1000 Overburden
Collapse
Depth (m)
2000
2500
83
7.3.5 Friction angle
The effect of friction angle was also seen in Figure 71. It can be seen that the change in friction
angle does not affect the collapse curve that much and is less vital property when evaluating the
collapse curve.
Pressure (sg)
0 0.5 1 1.5 2
0
500
Pore pressure
1000 Overburden
Collapse
Depth (m)
2000
2500
84
7.3.6 Poisson’s ratio
This property tends to show a little effect on the initial collapse curve. But in comparison to the
Biot’s constant, the effect is very small.
Pressure (sg)
0 0.5 1 1.5 2
0
500
Pore pressure
1000 Overburden
Collapse
Depth (m)
2000
2500
85
7.3.7 Cohesive strength
The change is cohesive strength also shows very minute alterations to the collapse curve.
Pressure (sg)
0 0.5 1 1.5 2
0
500
Pore pressure
1000 Overburden
Collapse
Depth (m)
1500
2000
2500
86
7.3.8 Wall temperature
Wall temperature of the well was increase and decreased by 20 degrees in order to see the effect
of it. In the upper depths, the increase and decrease had a significant effect in the curve whereas
down the formation, the effect is becoming more and more insignificant.
Pressure (sg)
0 0.5 1 1.5 2
0
500
Pore pressure
1000 Overburden
Thermal Collpase
Depth (m)
2000
2500
87
7.3.9 Discussion and Comparison
It can be seen from results in the case study that the activity of drilling fluid is the most
dominating of all properties in terms of the changes on the collapse pressure. It can be seen that
only a 5% change in this value changes the collapse pressure immensely. Figure 75 shows that if
the drilling fluid activity is decreased, the collapse pressure will tend to return to its original
position where there were no effects. This is because the difference of the formation and drilling
fluid activity is reduced. Once this difference is increased the change in the well collapse
pressure will be increased again. Therefore, even though it is the most dominating parameter, it
can be controlled by altering the activity of drilling mud.
For other parameters, Biot’s constant was proven to be the most dominating parameter as a
change of 10% in this value resulted in large changes in the well collapse pressure. This can be
seen below in Figure 76. Another important factor for the dominance of Biot’s constant is that
the horizontal in-situ stress was calculated from this value using equation 3.2. Therefore, the
results also included the Biot’s constant effect on the in-situ stresses. If the in-situ stresses are
known using other models in equation 4.1 and 4.2, the results can be more accurate and a better
representation of the Biot’s constant effect on the collapse pressure can be known. Temperature
is also a sensitive parameter in the upper depths of the formations where the change in collapse
pressure due to temperature was very high. Figure 76 also included the well collapse pressure
lines for other parameters too. The other properties showed very small changes and didn’t alter
the initial collapse pressure that much. But note that if all the values are overestimated or
underestimated, the combination of the effects even for the small changes can amend the results
greatly resulting in quite affective changes in well collapse pressure.
88
1.2
0
0 500 1000 1500 2000 2500
Depth (m)
Figure 75: Collapse Pressure graph for change in drilling mud activity
1.3
89
8 Conclusion
It was concluded from the simulations performed for different effects that the chemical and
hydraulic diffusion effects were quite dominating at the wall of the wellbore. Both of these
effects could be controlled by using right design for mud weight density for hydraulic diffusion
affects as the higher the well pressure will be the greater will be the increase in pore pressure.
Chemical effects can be controlled by changing the activity of the drilling fluid and can increase
or decrease the pore pressure depending on the formation and drilling fluid activity. Thermal
effects had no influence on pore pressure at the wall even though it was more dominating when
moved away from wellbore wall. Also, cohesive strength reduces with respect to time because of
which the effective collapse strength reduced significantly at the wall. An increase in pore
pressure tends to decrease the effective collapse stress and vice versa.
For the case study, activity of the drilling fluid activity was the most sensitive parameter for the
well collapse pressure. The second parameter that was found to be very sensitive to well collapse
pressure after the activity of drilling fluid was the Biot’s constant. A small increase in this value
changed the well collapse pressure dramatically. Temperature was quite sensitive in the upper
depths where the change in temperature between the formation and the wall was significantly
high. All the other properties analyzed had very small effects on the well collapse pressure. In
terms of chemical and thermal, the case study showed that chemical effects were far more
dominating than the thermal effects for this Heidrun well program.
90
References
A.M. Al-Ajmi, R. W. Z. (2006). "Stability Analysis of vertical boreholes using the Mogi-Coulomb."
International Jounal of Rock Mechanics and Mining Sciences 43(8): 1200-1211.
Aadnøy, B. S. (2003). "Introduction to special issue on wellbore stability." Journal of Petroleum Science
and Engineering 38: 79-82.
B.S. Aadnøy, M. E. C. (1987). "Stability of highly inclined boreholes." SPE Drilling Engineering 2(04): 364-
374.
Bernt S. Aadnøy, R. L. (2011). Petroleum Rock Mechanics. USA, Gulf Professional Publishing.
C. Chen, M. E. C., M.M. Sharma & M. Yu (2001). Poroelastic chemical, and thermal effects on wellbore
stability in shales. U.S. Symposium on Rock Mechanics. Washington D.C, American Rock Mechanics
Association.
Dyni, J. R. (2006). "Geology and Resources of Some World Oil-Shale Deposits." USGS
5294.
E. Eseme, B. M. K., R. Littke (2012). "Evolution of petrophysical properties of oil shales during high-
temperature compaction tests." Marine and Petroleum Geology 31(1): 110-124.
E. Fjær, R. M. H., P. Horsrud, A.M. Raaen & R.Risnes (2008). Petroleum Related Rock Mechanics.
Hungary, Elsevier publications.
Eaton, B. A. (1972). "Graphical method predicts geopressure worldwide." World Oil 182: 51-56.
Eaton, B. A. (1975). "The equation for geopressure prediction from well logs." SPE 5544.
F.J. Santarelli, S. C. (1995). "Do shale swell? A critical review of available evidence." SPE-29421-MS.
91
Fjær, E., et al. (2008). Petroleum related rock mechanics, Elsevier Science.
G. Stjern, A. A., P Horsrud (2003). "Local rock mechanical knowledge proves drilling performance in
fractured formations at the Heidrun field." Journal of Petroleum Science and Engineering 38(3-4): 83-96.
Horsrud, P. (2001). "Estimating Mechanical Properties of Shale from Empirical Correlations." SPE Drilling
& Completion 16(02): 68-73.
I.M. Breckels, H. A. M. v. E. (1982). "Relationship between horizontal stress and depth in sedimentary
rocks." Journal of Petroleum Technology 34(09): 2,191-192,199.
J.C. Jaeger, N. G. W. C., and R.W. Zimmerman (2007). Fundamentals of rock mechanics. Singapore,
Blackwell Publishing.
Jianguo Zhang, M. Y., T.M. Al-Bazali, Seehong Ong, M.E. Chenevert, M.M. Sharma (2006). "Maintaining
the stability of deviated and horizontal wells." SPE.
Lal, M. (1999). "Shale stability: Drilling fluid interaction and shale strength." SPE 54356.
Lianyang Zhang, P. C., K.C. Radha (2010). "Evaluation of rock strength criteria for wellbore stability
analysis." International Jounal of Rock Mechanics and Mining Sciences 47(8): 1304-1316.
Md. Aminul Islam, P. S. (2013). "An experimental investigation of shale mechanical properties through
drained and undrained test mechanisms." Rock Mechanics rock engineering 46: 1391-1413.
92
Mengjiao Yu, M. E. C., Mukul M. Sharma (2003). "Chemical-mechanical wellbore instability model
instability model for shales." Journal of Petroleum Science and Engineering 38: 131-143.
Mitchell, R. F., et al. (2011). Fundamentals of drilling engineering, Society of Petroleum Engineers.
Okiongbo, K. S. (2011). "Petrophysical Properties of North Sea Shales." Research Journal of Applied
Sciences 3(1): 46-52.
Oort, E. v. (2003). "On the physical and chemical stability of shales." Journal of Petroleum Science and
Engineering 38: 213-235.
Ronald Steiger, P. R., Leung, K. peter (1992). "Quantitative Determination of the Mechanical Properies of
Shales." SPE 7.
S. Hillier, G. R., I Sims and JC Cripps (2006). "Clay materials used in construction, A mineralogical and
chemical data." Engineering Geology speical publications 21: 449-459.
S.O. Osisanya, M. E. C. (1996). "Physico chemical modelling of wellbore stability in shale formations."
Journal of Canadian Petroleum Technology 35(02).
Soroush, D. H. (2013). Shale Gas Geomechanics. Stavanger, European Association of Geosciences &
Engineers
Venkanna, B. K. (2010). Fundamentals of heat and mass transfer. New Delhi, PHI Learning.
Appendix
M-file for the equation 5.9 to calculate the thermal, chemical and hydraulic diffusion
effects
function [porep]=porep;
93
po=0.01538*4352.5; % Pore pressure
ppi=-(0.1*0.462*375.7*log(0.78/0.915)); %chemical potential
rw=0.127; % radius of well
r=[0.127 0.13335 0.1397 0.14605 0.1524 0.15875 0.1651]%specific radius
c=3.4e-10; %hydraulic diffusivity
pw=0.0180*4352.5; % well pressure
k=0.124; %coupling coefficient
co=9.54e-7; %thermal diffusivity
Tw=350.7; %temperature of wall
To=375.7; %temperature of formation
t=864000; %time 10 hours
for m=1:length(t)
porep(m,:)=po+((pw-ppi-po).*(sqrt(rw./r).*(erfc((r-
rw)./(2.*sqrt(c.*t(m)))))))-((k*(Tw-To)/(1-(c/co))).*(sqrt(rw./r).*((erfc((r-
rw)./(2.*sqrt(c.*t(m)))))-(erfc((r-rw)./(2.*sqrt(co.*t(m))))))));
end
hold on
plot(r./rw,porep(1,:),'b');
hold off
M-file for the equation 5.9 to calculate the thermal, chemical and hydraulic diffusion
effects at different times
function [porep]=time;
po=0.01538*4352.5;
ppi=-(0.1*0.462*375.7*log(0.78/0.915));
rw=0.127;
r=[0.127:0.001:0.1651];
c=3.4e-10;
pw=0.0180*4352.5;
k=0.124;
co=9.54e-7;
Tw=350.7;
To=375.7;
t=[60 3600 36000 864000];
for m=1:length(t)
porep(m,:)=po+((pw-ppi-po).*(sqrt(rw./r).*(erfc((r-
rw)./(2.*sqrt(c.*t(m)))))))-((k*(Tw-To)/(1-(c/co))).*(sqrt(rw./r).*((erfc((r-
rw)./(2.*sqrt(c.*t(m)))))-(erfc((r-rw)./(2.*sqrt(co.*t(m))))))));
end
%
hold on
plot(r./rw,porep(1,:),'b');
plot(r./rw,porep(2,:),'r');
plot(r./rw,porep(3,:),'k');
plot(r./rw,porep(4,:),'g');
hold off
94