Journal of Natural Gas Science and Engineering 26 (2015) 72e98

Contents lists available at ScienceDirect

Journal of Natural Gas Science and Engineering

journal homepage: www.elsevier.com/locate/jngse

A wellbore stability analysis model with chemical-mechanical

coupling for shale gas reservoirs
Tianshou Ma*, Ping Chen
State Key Laboratory of Oil and Gas Reservoir Geology and Exploitation, Southwest Petroleum University, Chengdu, Sichuan 610500, PR China

a r t i c l e i n f o a b s t r a c t

Article history: Chemical and mechanical coupling are the most important factors affecting wellbore instability. The
Received 3 February 2015 main objective of this research is to propose a wellbore stability analysis model for shale gas reser-
Received in revised form voirs. A mathematical model was proposed to analyze wellbore stability based on a quantitative so-
20 May 2015
lution for stress induced by mechanical, hydraulic and chemical effects, and the effective stress tensor
Accepted 21 May 2015
around the borehole in a cylindrical coordinate system was also obtained. Anisotropic mechanical
Available online 14 June 2015
properties and changes in the strength of shale rocks were collected from tri-axial compression ex-
periments, direct shear tests and the literature. To examine for shear failure along the weak plane and
Keywords:
Shale gas
across the weak plane, the effective stress tensor in a cylindrical coordinate system was transformed
Wellbore stability into the weak plane's local coordinate system and integrated into the strength criteria of the weak
Chemical-mechanical coupling plane. In addition, the failure regions around a horizontal well were simulated at different drilling
Weak plane times and for different drilling directions, and the real causes of wellbore instability for well X201-H1
Anisotropic strength in the Sichuan basin were analyzed. The results indicate that the nonlinear evolution equation for the
strength parameters obeyed the logistic model; the strength parameters decreased drastically as the
soaking time increased over the first five days, after which the strength parameters decreased slowly.
In addition, pore pressure increased and solute concentration decreased under the condition Cm < C0,
while pore pressure decreased and solute concentration increased under the condition Cm > C0. The
decrease in strength and the increase in pore pressure have significant impacts on the stability of
wellbores within shale gas reservoirs. Pore pressure propagation changes the effective normal stress
on the weak plane of the wellbore and results in the stress concentration exceeding the strength
envelope. In traditional models, failure regions occur only on the surface of a borehole; however, in
the new model, failure regions can also occur in the interior of a formation, and they can occur within
four zones around a wellbore's circumference. The decrease of shale strength and the increase of pore
pressure under the condition of water-based mud (WBM) has a greater impact than in oil-based mud
(OBM), which help define the critical equivalent mud weight (CEMW) requirements at which the rate
of collapse increases rapidly. To maintain borehole stability, a series of approaches must be adopted,
including reasonable mud weight (MW), mud system, well path, physical plugging, etc. This model can
be used to analyze the failure regions around boreholes and calculate the CEMW needed to maintain
wellbore stability at different times. This model is different from, and more practical than, the
traditional model.
© 2015 Elsevier B.V. All rights reserved.

1. Introduction years (Zou et al., 2010; Wang et al., 2013). A new official report from
the EIA (2014) has indicated that shale gas resources are abundant
Shale gas, an unconventional natural gas produced from around the world, and shale gas has the potential to play a critical
organic-rich shale, has been paid increasing attention in recent and decisive role in the global energy market in the future (Yuan
et al., 2014). Shale gas resources are also abundant in China and
have already been identified in certain areas. The technically
recoverable resources in China have been estimated between
* Corresponding author.
E-mail addresses: matianshou@126.com (T. Ma), chenping@swpu.edu.cn 10
1012 m3 and 15
1012 m3 (Wang et al., 2013). Because the
(P. Chen). native permeability of shale gas is extremely low, horizontal well

http://dx.doi.org/10.1016/j.jngse.2015.05.028
1875-5100/© 2015 Elsevier B.V. All rights reserved.
 T. Ma, P. Chen / Journal of Natural Gas Science and Engineering 26 (2015) 72e98
drilling and hydraulic fracturing are typically used to enhance the
recovery of these hydrocarbon resources (Wang et al., 2013; Yan
et al., 2014; Ma and Chen, 2014a). However, shale gas reservoirs
are often characterized by tight matrix, well-developed bedding
planes and micro-fissures, and they are rich in clay mineral, which
makes them robustly anisotropic and highly water-sensitive (Ma
and Chen, 2014a; Yan et al., 2014). The borehole instability prob-
lems that often occur in shale bedding planes rich in clay minerals
are often associated with equivalent mud weight (EMW, either
static or circulating equivalent density), in situ stress, wellbore
orientation, and chemical reactions, among other things. The single
most important cause of wellbore instability is incorrect EMW (Fjar
et al., 2008; Chen et al., 2008a; Zoback, 2012; Lee et al., 2012). Too
low of an EMW can cause shear failure around the borehole, which
can lead to cave-ins and hole collapses; too high an EMW can cause
tension failure around the borehole, which can lead to lost mud or
lost circulation.
Maintaining wellbore stability is an important issue in the oil
and gas industry because problems with borehole instability can
significantly increase drilling NPT (non-productive time), the
length of the drilling cycle, and drilling costs and can decrease the
efficiency of exploration and development (Aadnoy and Ong,
2003; Fjar et al., 2008; Chen et al., 2008a; Kang et al., 2009;
Zoback, 2012; Yan et al., 2014). With these factors in mind, the
purposes of researching wellbore stability are (1) to reduce well
construction costs by eliminating borehole instability related to
the NPT, (2) to optimize the design of a well's trajectory on the
basis of in situ stress and rock properties, and (3) to provide
guidelines for the design of the mud weight (MW) used to drill
each well in order to keep the borehole from failing during and
after drilling. Various researchers have proposed wellbore stability
analysis methods, all of which assume that rock is linear elastic
and has isotropic strength (Bradley, 1979; Cheatham, 1984; Zoback
et al., 1985; Zhou et al., 1994; Aadnoy and Ong, 2003; Fjar et al.,
2008; Chen et al., 2008a; Kang et al., 2009; Zoback, 2012; Lee
et al., 2012). However, because the anisotropic nature of shale
gas reservoirs, the propagation of pore pressure and the chemical
reactions therein are ignored, these methods may result in
incorrect wellbore stability results when applied to wellbores in
shale gas reservoirs. Aadnøy (1987) proposed a semi-analytical
model to study the effects of anisotropic strength, borehole
inclination, in situ stress, etc., to solve the issues of wellbore
instability that result from drilling in formations with anisotropic
rock strengths. The results of that study indicated that rocks
would fail along weak planes under certain conditions, which had
significant effects on wellbore stability. Following Aadnøy, various
researchers proposed new methods to determine wellbore sta-
bility by assuming that rocks were transversely isotropic materials
with anisotropic strength behaviors (Aadnøy and Chenevert, 1987;
Aadnøy, 1988; Ong and Roegiers, 1993, 1996; Ong, 1994; Okland
and Cook, 1998; Gupta and Zaman, 1999; Zhang, 2008, 2009;
Chen et al., 2008a,b; Pei, 2008; Al-Bazali et al., 2009; Younessi
and Rasouli, 2010; Tan et al., 2010; Jin et al., 2013; Hou et al., 2013;
Zhang, 2013). Their works indicated that neglecting the effects of
anisotropy could result in incorrect results. Despite these efforts to
modify the wellbore stability models, numerous accidents asso-
ciated with wellbore instability in shale gas reservoirs have been
reported (Yan et al., 2013, 2014; Ma and Chen, 2014a,b,c). Zoback
(2012) and Lee et al. (2012, 2013) found that the shapes and ori-
entations of failure regions are significantly affected by anisotropic
strength, and a few researchers have also studied the influence of
anisotropic strength on instability regions around the borehole
(Lee et al., 2012, 2013; Ma and Chen, 2014b). In addition, a few
researchers have investigated the influence of multiple weak
planes (Liu et al., 2014; Chen et al., 2014a). However, none of these
73
models have considered the impact of mechanical and chemical
coupling on wellbore stability. Some researchers have addressed
this issue; however, most of them have only focused on hydro-
mechanical coupling or thermo-hydro-mechanical coupling (Cui
et al., 1997; Abousleiman and Cui, 1998; Ekbote, 2002; Yuan
et al., 2013, 2014). And while a large number of researchers have
studied the influence of mechanical and chemical coupling on
wellbore stability related to hydration and interactions with dril-
ling mud (Chen et al., 2003a; Cheng et al., 2006; Lu et al., 2012,
2013; Li et al., 2012a; Yan et al., 2013; Ma and Chen, 2014a),
these researchers have still failed to consider the influence of
comprehensive factors such as wellbore inclination, anisotropic in
situ stress, anisotropic rock strength, mechanical and chemical
coupling, downhole pressure (or EMW), the changes in the shale
strength, etc., on the unstable zone around the borehole or the
critical collapse pressure. Therefore, the main objective of this
research is to propose a semi-analytical model and method of
wellbore stability analysis for shale gas formations, in which the
rock material is assumed to be poroelastic, have anisotropic
strength behaviors and have variable strength, while the multi-
field coupling is assumed to be both mechanical and chemical;
in addition, the anisotropic elasticity behavior and dynamic con-
ditions of wellbore pressure are ignored.
2. Chemical-mechanical collapse pressure model for shale
2.1. The propagation of pore pressure
Wellbore collapse occurs when exposure to mud causes the
propagation of pore pressure (Mody and Hale, 1993; Yu et al.,
2001; Yu, 2002; Van Oort, 2003; Yu et al., 2003; Chen et al.,
2003b, 2010; Zeynali, 2012). Mody and Hale (1993) assume that
shales can act as semi-permeable membranes, with hydraulic
pressure, osmotic pressure and electrical potential gradients
acting as the driving forces on those membranes. However, elec-
trical potential gradients are generally seldom taken into account
in pore pressure propagation models, leaving hydraulic pressure
and osmotic pressure gradients as the driving forces in these
models, where the equation of osmotic pressure is expressed as
(Mody and Hale, 1993),


RT a
p  p0 ¼ Im ln shale (1)
V amud
where p is pore pressure, MPa; Im is membrane efficiency; T is the
absolute temperature, K; R is the perfect gas constant; V is the
partial molar volume of water, l/mol; ashale is the activity of water
for pore fluid; amud is the activity of water for drilling mud; and p0 is
the original pore pressure, MPa.
2.1.1. General equation for water transport in shales
Solvent transport is driven by hydraulic pressure, osmotic
pressure gradients, as well as changes in solute concentration and
pore pressure. As a result, the solvent flux equation can be
expressed as (Yu et al., 2001; Yu, 2002),
Jv ¼ k1 Vp  nRTk2 VCs (2)
where Jv is the solvent flux, l; k1 is the hydraulic diffusion coeffi-
cient, m2/(Pa s); k2 is the membrane efficiency, m2/(Pa s); n is the
mole number of solute ions; and Cs is the solute concentration of
pore fluid, mol/l.
According to the principle of mass conservation, the conserva-
tion of mass applied to the solvent can be expressed as (Yu et al.,
2001; Yu, 2002),
74 T. Ma, P. Chen / Journal of Natural Gas Science and Engineering 26 (2015) 72e98

vr Cs ðr ¼ rw ; q; g; tÞ ¼ Cm
þ V$ðrJv Þ ¼ 0 (3) (11)
vt pðr ¼ rw ; q; g; tÞ ¼ pm

where Jv is the volumetric flux of the solvent; r is the fluid density,
g/cm3; and t is time, s.
Because the fluid in pore spaces is slightly compressible, we
must consider the elastic of fluid in the equation of state. As a result,
the density of fluid can be expressed as,
where rw is the radius of the borehole, m; Cm is the solute con-
centration of the drilling mud, mol/l; and pm is the downhole
pressure, MPa.
The external boundary of the borehole can be expressed as,

  Cs ðre /∞; q; g; tÞ ¼ C0
r ¼ r0 e cr ðpp0 Þ
zr0 1 þ cr ðp  p0 Þ (4) (12)
pðre /∞; q; g; tÞ ¼ p0

where p0 is atmospheric pressure, MPa; r0 is the density of fluid where re is the radius of the external boundary of the borehole, m;
under atmospheric pressure, g/cm3; and cr is fluid compressibility, C0 is the original solute concentration of the fluid in the pore space,
MPa1. mol/l; and p0 is the original pore pressure, MPa.
Substituting Eq. (4) into Eq. (3), we get,

vp 1 2.1.4. Initial conditions

þ VJv þ Jv Vp ¼ 0 (5) In the initial condition, the pore pressures and solute con-
vt cr
centrations within the test area are all of the initial pore pressures
Substituting Eq. (2) into Eq. (5), we get, and solute concentrations. So, the initial condition can be
expressed as,
vp 1
þ Vð  k1 Vp  nRTk2 VCs Þ þ ð  k1 Vp  nRTk2 VCs ÞVp ¼ 0

vt cr Cs ðr; q; g; t ¼ 0Þ ¼ C0
(13)
(6) pðr; q; g; t ¼ 0Þ ¼ p0

Because the magnitude of cr is approximately 105 and we as-

sume that V2p ≪ 1 and (VpVCs) ≪ 1 (Yu, 2002), the last item of Eq.
(6) can be ignored, and Eq. (6) can be simplified to,
vp 1  
þ  k1 V2 p  nRTk2 V2 Cs ¼ 0
vt cr
In general, pore pressure propagation around a borehole can be
simplified to a two dimensional plane; therefore, the general
equation for water transport in shales also can be transformed into
cylindrical coordinates and can be expressed as,
2.1.5. Solutions for pore pressure propagation
To simulate the pore pressure and solute concentration
around the borehole, we can use a quantitative method such as a
finite difference method (FDM) or a finite element method
(7)
(FEM). An FDM is typically used to solve differential equations,
and there are implicit FDM and explicit FDM. We selected an
implicit FDM to solve the general equation for water and ion
transport in shales. First, we completed the computational
model using iso-spacing grids, and the resulting equation can be
expressed as,


 
 ri ¼ rw þ ði  1Þh
vp k1 1 v vp nRTk2 1 v vCs (14)
¼ r þ r (8) tk ¼ kt
vt cr r vr vr cr r vr vr
where
2.1.2. General equation for ion transport in shales 8
> re  rw
The transport of solutes will also change the solute concentra- >
<h ¼ n
tion of the pore fluid. The solute concentration profile can be (15)
>
> t
simulated using the following diffusivity equation (Yu, 2002), :t ¼
m
vCs
 Deff V2 Cs ¼ 0 (9) where t is the time interval, s; h is the space interval, m; m is the
vt
number of time intervals; n is the number of space intervals; i is
where Cs is the solute concentration of the pore fluid, mol/l, and Deff the grid node of space, i ¼ 1,2,3,……,nþ1; and k is the grid node of
is the diffusion coefficient, m2/s. time, k ¼ 0,1,2,3,……,m. Consider a typical borehole which has a
In general, pore pressure propagation around a borehole can be borehole radius (rw) of 0.108 m and a boundary radius (re) of
simplified to a two dimensional plane; therefore, the general 1.08 m. If we assume a space interval of 0.02rw, then the number of
equation for ion transport in shales can be transformed into cy- space intervals (n) is 180. The number of time intervals (m) can be
lindrical coordinates and can be expressed as, arbitrary.

 The differential equations using Eqs. (8) and (10) were solved
vCs 1 v vCs with a backward temporal differentiation formula and a central
 Deff r ¼0 (10)
vt r vr vr spatial differentiation formula. In this scenario, the differential
equation for Eq. (10) can be expressed as,
Eqs. (7) and (9) are the general equations for water and ion


transport in shales, and Eqs. (8) and (10) are the general equa-
k1 Deff t k 2Deff t k
tions in cylindrical coordinates for water and ion transport in Cs;i ¼ r C
ðiþ0:5Þ s;ðiþ1Þ þ 1 þ Cs;ðiþ1Þ
ri h2 h2
shales. (16)
Deff t k
 rði0:5Þ Cs;ði1Þ
ri h2
2.1.3. Boundary conditions
The inner boundary of the borehole can be expressed as, We then combined the solution conditions to solve the
 T. Ma, P. Chen / Journal of Natural Gas Science and Engineering 26 (2015) 72e98 75

differential equation, in which the solution conditions must be 2.2. Effective stress tensors around a borehole
expressed as unit nodes,
8 2.2.1. Coordinate systems
> k
< Cs;1 ¼ Cm
> To propose a model for the effective stress tensor around a
k
Cs;ðnþ1Þ ¼ C0 (17) borehole, a number of coordinate transformations must be per-
>
> formed (Lee et al., 2012; Ma and Chen, 2014b). Three different
: C0 ¼ C
s;i 0 coordinate systems (CSs) are involved (Fig. 2) in the shale gas
reservoir borehole stability problem: (1) The global coordinate
Similarly, the differential equation for Eq. (8) can be expressed
system (GCS) (N,E,Z); (2) The local coordinate system of the
as,
selected borehole (BCS) (xb,yb,zb); and (3) The local coordinate
nRTk2 t h i
system of the weak plane (WCS) (xw,yw,zw). Once these coordinates
pk1
i þ 2
k
rðiþ0:5Þ Cðiþ1Þ  2ri Cik þ rði0:5Þ Cði1Þ
k
ri h cr are identified, rotation matrices, which transform the components

 of the stress tensor from one CS to the other, can be obtained (Lee
k1 t 2k t k t
¼  2 rði0:5Þ pkði1Þ þ 1 þ 21 pki  12 rðiþ0:5Þ pkðiþ1Þ et al., 2012).
ri h cr h cr ri h cr
As shown in Fig. 2(a), Z is the vertical axis, while N and E are
(18) horizontal axes in the GCS. The vertical stress (sv) generally acts in
The boundary conditions, initial conditions and the result of Eq. the vertical direction. Therefore, the azimuth angle (U) of the
(16) were then combined to solve the differential equation, where, maximum horizontal stress (sH) expresses the relationships be-
the boundary conditions and initial conditions must be expressed tween the local coordinates of in situ stress components and the
as unit nodes, GCS. As a result, the rotation matrix of the transformation of the in
situ stress tensor from the local coordinates to the GCS can be
8 k expressed as,
< ps;1 ¼ pm
>
2 3
pks;ðnþ1Þ ¼ p0 (19) cos U sin U 0
>
: 0 E ¼ 4 sin U cos U 05 (20)
ps;i ¼ p0
0 0 1
It is not difficult to see that pore pressure is coupled with water
and ion transport in shale (Yu, 2001). Therefore, to solve the pore where U is the azimuth angle of the maximum horizontal stress, ( ).
pressure profile, we can first solve Eq. (16) and then solve Eq. (18). As shown in Fig. 2(a), the local borehole frame (xb, yb, zb) is
The flowchart of the solution is illustrated in Fig. 1. A computer based on the orientation of the borehole, where zb defines the axis
program that can calculate the solute concentration profile and of the borehole, while xb and yb form the local normal cross-
pore pressure profile using Eqs. (16) and(18) was developed in section. The inclination angle of the borehole (ab) and the azi-
Matlab. muth angle of the borehole (bb) express the relationships between

Fig. 1. The solution flowchart for a pore pressure profile.

76 T. Ma, P. Chen / Journal of Natural Gas Science and Engineering 26 (2015) 72e98

Table 1
The degree of anisotropy for the elasticity modulus of different shales.

No. Type Eh/MPa Ev/MPa Eh/Ev Refs.

1 Green River shale 35.2 29.7 1.19 McLamore and Gray,

2 Permian shale 33.1 24.1 1.37 1967
3 X shale 17.3 10.4 1.66 Meier et al., 2013
4 Niutitang shale 20.8 19.8 1.05 Yang et al., 2013
5 Niutitang shale 19.2 10.2 1.88
6 Niutitang shale 20.4 10.9 1.87
7 Niutitang shale 18.2 11.9 1.53
8 Silurian shale 25.0 21.0 1.19 Yuan et al., 2013
9 Silurian shale 35.5 29.5 1.20 Yu et al., 2014
10 Silurian shale 35.5 25.9 1.37 Yan et al., 2014
11 Longmaxi shale 61.4 30.5 2.01 Yang et al., 2012
12 Longmaxi shale 32.6 23.7 1.38
13 Longmaxi shale 27.9 14.3 1.95 Jin et al., 2012
14 Longmaxi shale 24.6 14.1 1.74 Jia et al., 2013
15 Longmaxi shale 46.5 24.6 1.89
16 Longmaxi shale 51.8 22.9 2.26
17 Longmaxi shale 56.4 24.1 2.34
18 Longmaxi shale 27.6 25.8 1.07 Li et al., 2012a
19 Longmaxi shale 28.8 24.4 1.18
20 Longmaxi shale 29.3 24.1 1.22 Wang et al., 2012
21 Longmaxi shale 39.6 36.1 1.10
22 Longmaxi shale 13.3 10.6 1.25 Testing data
23 Longmaxi shale 12.9 10.4 1.24
24 Longmaxi shale 24.1 22.1 1.09
25 Longmaxi shale 24.5 25.3 0.97
26 Longmaxi shale 23.5 21.6 1.09
27 Longmaxi shale 30.2 29.3 1.03
28 Longmaxi shale 33.7 31.5 1.07
29 Average 1.46 /

coordinates to the BCS can be expressed as (Lee et al., 2012; Ma and

Chen, 2014),
2 3
cos q sin q 0
C ¼ 4 sin q cos q 05 (22)
0 0 1

where q is the point location angle, ( ).

Fig. 2. The relationships between different CSs.
the BCS and GCS, in which ab is defined as the angle between the
axis of the borehole (zb) and the Z axis, and bb is defined as the
angle between the projection of zb onto the NE plane (xb) and the N
axis. Therefore, the transformation matrix of the stress tensor
from the BCS to the GCS can be expressed as (Lee et al., 2012; Ma
and Chen, 2014),
2 3
cos bb cos ab sin bb cos ab sin ab
B ¼ 4 sin bb cos bb 0 5
cos bb sin ab sin bb sin ab cos ab
where ab is the inclination angle of the borehole, ( ), and bb is the
azimuth angle of the borehole, ( ).
In addition, the stress components around the borehole are
usually expressed in cylindrical coordinates. Therefore, the trans-
formation matrix of the stress components from local cylindrical
As shown in Fig. 2(b), the local CS of the weak plane (xw, yw, zw)
is based on the orientation of the weak plane, where xw is normal
to the weak plane and xw and yw form the weak plane. The dip
angle (aw) and dip direction (bw) of the weak plane express the
relationships between the WCS and GCS, in which aw is defined as
the angle between the projection of zw onto the NE plane (hori-
zontal plane) and the zw axis, and bw is defined as the angle be-
tween the projection of zw onto the NE plane (horizontal plane)
and the N axis. Therefore, the transformation matrix of the stress
tensor from WCS to GCS can be expressed as (Lee et al., 2012; Ma
and Chen, 2014),
2 3
cos bw sin aw sin bw sin aw cos aw
W ¼ 4 sin aw cos bw 0 5 (23)
cos bw cos aw sin bw cos aw sin aw
where aw is the dip angle of the weak plane, ( ), and bw is the dip
direction of the weak plane, ( ).
(21)
2.2.2. Stress induced by in situ stress and the hydraulic effect
The anisotropic properties of rock are a very important topic
in the research and practice of rock engineering. Natural shale
rocks are more or less anisotropic, and the anisotropic elasticity
behavior of shale has been investigated by many researchers.
The mechanical properties of shale as collected from the liter-
ature and tri-axial compression experiments are listed in
Table 1. The table indicates that the average degree of anisotropy
 T. Ma, P. Chen / Journal of Natural Gas Science and Engineering 26 (2015) 72e98 77

is approximately 1.46, while the maximum degree of anisotropy 2.2.3. Stress induced by chemical reactions
derived from tri-axial compression experiments is approxi- Local stress induced by chemical reactions can be expressed as
mately 1.25. (Yu et al., 2001; Yu, 2002),
Jin et al. (2012) and Ong (1994) researched the effect of 8
anisotropic elasticity behavior on stress distribution around >
> Zr
>
> 0 ap ð1  2vÞ 1
boreholes, and their results showed that anisotropic elasticity >
> sr ¼ ½pðr; tÞ  p0 rdr
>
> 1v r2
behavior has a relatively small impact, especially when the >
>
>
>
rw
degree of anisotropy is low (<1.25). Therefore, we can ignore >
< 8 9
Zr
the influence of anisotropic elasticity behavior on stress dis- 0 ap ð1  2vÞ < 1 =
>
> sq ¼  ½pðr; tÞ  p0 rdr  ½pðr; tÞ  p0 
tribution. As shown in Fig. 2, the mechanical model of an in- >
> 1v : r 2 ;
>
>
clined well is assumed to be under a plain strain condition, >
> rw
>
>
which means there is no displacement along the z b axis. >
> a ð1  2vÞ
>
: s0z ¼ p ½pðr; tÞ  p0 
Therefore, the quantitative solutions for stress induced by in 1v
situ stress and the hydraulic effect are as follows (Bradley, (28)
1979; Chen et al., 2008a; Zoback, 2012; Lee et al., 2012; Ma
and Chen, 2014b), where s0r ; s0q ; s0z are the radial stress, hoop stress and axial stress

8

2
2
>
> sxx þ syy r2 sxx  syy 4
rw rw 4
rw rw r2
>
> sr ¼ 1 w þ 1 þ 3  4 cos 2 q þ txy 1 þ 3  4 sin 2 q þ w pm
>
> 2 r 2 2 r 4
r 2
r 4
r 2
r2
>
>
>

>
> 2
4
4
>
> s þ syy rw sxx  syy rw rw r2
> sq ¼ xx
> 1 þ  1 þ 3 cos 2 q  t 1 þ 3 sin 2 q  w pm
>
> 2 r 2 2 r 4 xy
r 4
r2
>
>
>
>
>
>   rw2 2
>
> rw
< sz ¼ szz  2v sxx  syy 2 cos 2 q  4vtxy 2 sin 2 q
>
r r

(24)
>
> 2 4
2 4
> s  s r r rw rw
>
> t ¼ 
xx yy
1 þ 2 w
 3 w
sin 2 q þ t 1 þ 2  3 cos 2 q
>
>
rq
2 r2 r4
xy
r2 r4
>
>
>
>


>
> 2
>
> r2 rw
>
> trz ¼ txz 1  w cos q þ tyz 1  sin q
>
> r2 r2
>
>
>
>


>
> r2 r2
>
: tqz ¼ tyz 1 þ w cos q  txz 1 þ w sin q
2 2
r r

The stress components for the local wellbore coordinates can be induced by chemical reactions, respectively, MPa; and ap is Biot's
expressed as, constant, dimensionless.
2 3
sxx txy txz 2.2.4. Effective stress tensor model
sinsituBCS ¼ 4 txy syy tyz 5 ¼ ВЕT sinsitu EBT (25) The stress tensors around the borehole are induced by me-
txz tyz szz chanical, hydraulic and chemical effects. Therefore, considering
2 3 Biot's effective stress law and combining Eqs. (24) and (28), the
sH 0 0 local stress components around the borehole can be obtained
sinsitu ¼ 40 sh 0 5 (26) through the following equation,
0 0 sv
8
Substituting Eqs. (20), (21) and (26) into Eq. (25), the following >
> sr ðt Þ ¼ sr þ s0r  ap pðr; t Þ
>
>
>
> s ðt Þ ¼ sq þ s0q  ap pðr; t Þ
equation is obtained, < q
sz ðt Þ ¼ sz þ s0z  ap pðr; t Þ
8 (29)
> sxx ¼ sH cos2 ab cos2 bb þ sh cos2 ab sin2 bb þ sv sin2 ab > trq ðt Þ ¼ tqz
>
>
>
> >
>
> t ðt Þ ¼ trz
>
> : rz
>
> syy ¼ sH sin2 bb þ sh cos2 bb tqz ðt Þ ¼ tqz
>
>
< szz ¼ sH sin2 ab cos2 bb þ sh sin2 ab sin2 bb þ sv cos2 ab
It also can be expressed as,
txy ¼ sH cosab cosbb sinbb þ sh cosab cosbb sinbb (27)
>
>
>
> tyz ¼ sH sinab cosbb sinbb þ sh sinab cosbb sinbb where sr(t), sq(t), sz(t) are effective radial stress, effective hoop
>
>
>
> t ¼ sH cosab sinab cos2 bb þ sh cos ab sinab sin2 bb
stress and effective axial stress at different times, respectively, MPa;
>
: xz and tqz(t), trq(t), trz(t) are three components of the effective shear
sv sin ab cos ab stress at different times, MPa.
The effective stress tensor around the borehole in a cylindrical
where sr, sq, sz are the radial stress, hoop stress and axial stress,
coordinate system can be expressed as,
respectively, MPa; tqz, trq, trz are three components of the shear
stress, MPa; n is Poisson's ratio, dimensionless; sxx, syy, szz, txy, tyz, 2 3
sr ðtÞ trq ðtÞ trz ðtÞ
txz are the stress components of the local wellbore coordinates, sccs ðtÞ ¼ 4 trq ðtÞ sq ðtÞ tqz ðtÞ 5 (31)
MPa; and sH, sh, sv are the maximum horizontal stress, minimum trz ðtÞ tqz ðtÞ sz ðtÞ
horizontal stress and vertical stress, respectively, MPa.
78 T. Ma, P. Chen / Journal of Natural Gas Science and Engineering 26 (2015) 72e98

8
>
>
2
2
2
>
> r2 sxx þ syy rw sxx  syy 4
rw rw 4
rw rw ap ð1  2vÞ 1
>
>
sr ðtÞ ¼ w p m þ 1  þ 1 þ 3  4 cos 2 q þ t xy 1 þ 3  4 sin 2 q þ
1v
>
> r 2 2 r 2 2 r 4
r 2
r 4
r 2
r2
>
> Z
>
>
r
>
>
½pðr; tÞ  p0 rdr  ap pðr; tÞ
>
>
>
> r
>
w

>
>
2
4
4
>
>
2
rw sxx þ syy rw sxx  syy rw rw ap ð1  2vÞ
>
>
sq ðtÞ ¼  p m þ 1 þ  1 þ 3 cos 2 q  txy 1 þ 3 sin 2 q 
1v
>
> r 2 2 r 2 2 r 4
r 4

>
>
8
<1 Z
9
>
<
r
=

½pðr; tÞ  p rdr  ½pðr; tÞ  p   ap pðr; tÞ
0 0
:r 2 ; (30)
>
>
rw
>
>   r2
>
> sz ðtÞ ¼ szz  2v sxx  syy w cos 2 q  4vtxy w
r2
sin 2 q þ
ap ð1  2vÞ
½pðr; tÞ  p0   ap pðr; tÞ
>
> 2 2 1v
>
>
r r
>
>
2 4
2 4
>
> t ðtÞ ¼ 
sxx  syy
þ
rw

rw
q þ t þ
rw

rw
cos 2 q
>
>
rq
2
1 2
r2
3
r4
sin 2 xy 1 2
r2
3
r4
>
>
>
>
2
2
>
> t ðtÞ ¼ t 
rw
q þ t 
rw
sin q
>
>
rz xz 1
r2
cos yz 1
r2
>
>
>
>
2
2
>
> t ðtÞ ¼ t þ
rw
q  t þ
rw
sin q
: qz yz 1
r2
cos xz 1
r2

If we calculate the maximum and minimum principal stresses
and apply them to the strength criteria, the collapse pressure
needed to prevent wellbore collapse can be obtained. Generally, the
collapse pressure can be converted to an equivalent mud weight in
engineering, which is called the critical equivalent mud weight
(CEMW).
2.3. The anisotropic strength of shale and changes in strength with
exposure to mud
Shale gas reservoirs are characterized by tight matrix, well-
developed bedding planes (weak planes) and are rich in clay
mineral, which makes them strongly anisotropic and highly water-
sensitive. To understand this anisotropy and changes in strength
that occur with exposure to mud, we identified a large number of
mechanical rock strength properties of shale from the literature,
and we tested the mechanical rock strength properties of the
Longmaxi shale.
2.3.1. The anisotropic strength of shale rock
The anisotropic properties of rock are a very important topic in
the research and practice of rock engineering. Natural shale rocks
are more or less anisotropic. To study the influence of weak planes
on shale mechanical properties, standard core samples of
F25 mm
50 mm were obtained, as shown in Fig. 3, j is the angle
between the surface of the weak plane and the axis of the stan-
dard core sample. Complete stressestrain curves for different
angles (j) of Longmaxi shale under uniaxial compression are
shown in Fig. 4. In addition, the anisotropy of shale strength has
also been tested by a large number of researchers (Jaeger, 1960;
Chenevert and Gatlin, 1965; Pinto, 1970; Jaeger, 1971; Vernik
and Nur, 1992; Aoki et al., 1993; Ong, 1994; Vernik and Liu,
1997; Niandou et al., 1997; Willson et al., 1999, 2007; Pei, 2008;
Jaeger and Cook, 2009; Sondergeld and Rai, 2011;
Kanitpanyacharoen et al., 2011; Wang et al., 2012; Li et al.,

Fig. 3. Schematic diagram of shale coring. Fig. 4. Complete stressestrain curves of the Longmaxi shale.
 T. Ma, P. Chen / Journal of Natural Gas Science and Engineering 26 (2015) 72e98 79

Table 2
The strength anisotropy of different shales.

No. Type Confining Compressive strength under different j (s1－s3)/MPa Max./MPa Min./MPa max./min. Data sources
pressure/MPa
0 15 30 45 60 75 90

1 Green River I 6.90 252.09 207.31 160.04 218.27 221.89 216.82 229.13 252.09 160.04 1.58 McLamore and Gray,
2 Green River I 34.48 345.45 321.14 260.23 299.22 304.08 305.84 321.87 345.45 260.23 1.33 1967
3 Green River I 68.97 438.88 421.40 370.42 405.68 422.95 420.37 427.09 438.88 370.42 1.18
4 Green River I 103.45 538.52 518.56 471.30 516.49 512.66 519.38 528.59 538.52 471.30 1.14
5 Green River I 172.41 716.69 683.70 664.37 667.36 685.25 693.22 693.73 716.69 664.37 1.08
6 Green River II 6.90 148.66 / 93.51 / 119.86 / 128.81 148.66 93.51 1.59
7 Green River II 34.48 203.07 / 148.27 / 177.41 / 188.09 203.07 148.27 1.37
8 Green River II 68.97 282.48 / 211.31 / 252.64 / 262.63 282.48 211.31 1.34
9 Green River II 103.45 356.32 / 278.88 / 320.91 / 330.55 356.32 278.88 1.28
10 Green River II 172.41 493.21 / 404.27 / 442.12 / 464.65 493.21 404.27 1.22
11 Tournemire 1.00 35.60 / 20.17 23.27 25.50 31.63 47.06 47.06 20.17 2.33 Niandou et al., 1997
12 Tournemire 5.00 60.14 / 35.63 41.11 44.86 / 58.62 60.14 35.63 1.69
13 Tournemire 20.00 105.19 / 83.93 77.94 108.61 / 102.81 108.61 77.94 1.39
14 Tournemire 40.00 144.00 / 113.01 109.84 / / 136.43 144.00 109.84 1.31
15 Tournemire 50.00 159.84 / 129.71 130.00 / / 154.87 159.84 129.71 1.23
16 Niutitang 0 66.90 / / / / / 69.10 69.10 66.90 1.03 Yang et al., 2013
17 Niutitang 10 125.38 / / / / / 105.63 125.38 105.63 1.19
18 Niutitang 20 121.89 / / / / / 126.27 126.27 121.89 1.04
19 Niutitang 40 145.35 / / / / / 99.845 145.35 99.85 1.46
20 Niutitang 0 102.5 / 11.43 21.24 37.23 / 92.23 102.50 11.43 8.97 Chen et al., 2014
21 Niutitang 10 150.78 / 97.17 76.47 87.21 / 127.84 150.78 76.47 1.97
22 Niutitang 20 178.74 / 123.2 106.21 137.49 / 152.79 178.74 106.21 1.68
23 Niutitang 30 222.92 / 163.52 214.35 215.05 / 201.6 222.92 163.52 1.36
24 Niutitang 40 225.86 / 151.71 186.7 191.79 / 186.77 225.86 151.71 1.49
25 Niutitang 50 235.9 / 217.05 230.49 242.47 / 234.36 242.47 217.05 1.12
26 Silurian 0 69.15 58.68 39.54 23.83 32.77 48.78 64.68 69.15 23.83 2.90 Yuan et al., 2013
27 Silurian 0 103.5 74.73 52.16 75.35 111.07 121.24 143.07 143.07 52.16 2.74 Yu et al., 2014
28 Silurian 20 168.37 121.74 115.54 130.54 168.99 192.81 215.88 215.88 115.54 1.87
29 Silurian 0 103.8 72.9 52.1 74.7 111.4 121.8 143.2 143.20 52.10 2.75 Yan et al., 2014
30 Haynesville 50 143.73 / / / / / / 143.73 143.73 / Li et al., 2012b
31 Haynesville 60 171.92 / / / / / / 171.92 171.92 /
32 Haynesville 90 209.19 / / / / / / 209.19 209.19 /
33 Eagle Ford 30 39.16 / / / / / / 39.16 39.16 /
34 Eagle Ford 50 / / / / / / 214.09 214.09 214.09 /
35 Barnett 20 / / / 40.56 / / / 40.56 40.56 /
36 Barnett 60 / / / / / / 391.25 391.25 391.25 /
37 Longmaxi 0 69.18 / / / / / 151.93 151.93 69.18 2.20 Yang et al., 2013
38 Longmaxi 53 / / / / / / 261.47 261.47 261.47 /
39 Longmaxi 54 / / / / / / 270.03 270.03 270.03 /
40 Longmaxi 40 106.81 / / / / / 175.35 175.35 106.81 1.64 Jin et al., 2012
42 Longmaxi 0 124.26 / 47.22 / 108.76 / 121.64 124.26 47.22 2.63 Jia et al., 2013
43 Longmaxi 10 148.78 / 79.88 / 143.18 / 143.78 148.78 79.88 1.86
44 Longmaxi 20 205.65 / 102.65 / 151.65 / 173.34 205.65 102.65 2.00
45 Longmaxi 30 229.14 / 143.3 / 178.83 / 192.06 229.14 143.30 1.60
46 Longmaxi 18 120.5 / 26.3 11.3 31.7 / 198.95 198.95 11.30 17.61 Li et al., 2012a
47 Longmaxi 31 185.5 / 83.7 75.9 50 / 216.6 216.60 50.00 4.33
48 Longmaxi 0 100.34 80.61 53.46 77.52 115.75 120.13 126.89 126.89 53.46 2.37 Wang et al., 2012
49 Longmaxi 0 153.3 100.3 96.1 191.8 182.5 218.5 216.7 218.50 96.10 2.27 Testing data
50 Longmaxi 30 264.1 168.4 220 284.5 275.8 350.7 292.7 350.70 168.40 2.08
51 Longmaxi 60 362 272.6 286.6 354.8 348.6 391.3 314.7 391.30 272.60 1.44
52 Average 212.06 155.59 2.31

2012a, 2012b; Josh et al., 2012; Cho et al., 2012; Kim et al., 2012; j ¼ 90 , and minimum strength when j is between 10 and 40
Vasin et al., 2013; Horne, 2013; Das, 2013; Ma and Chen, 2014b; (Lee et al., 2012). The experiment results in this study showed that
Tang, 2014; Chen et al., 2014; Heng et al., 2014). The data on the the angle (j) of minimum strength is approximately 30 .
anisotropic strengths of different shales collected from tri-axial
compression experiments and the literature are listed in Table 2.
2.3.2. The main failure modes of shale rock
These data indicate that the behavior of anisotropic rock is
The results of the tri-axial compression experiments showed
different in different directions, which means the strength of
that there are multiple modes for shale rock failure, as shown in
shale changes with the angle (j). These studies were also per-
Fig. 5. Three different failure modes are involved under low
formed to characterize the anisotropy of strength. The ratio of
confining pressure: (1) Splitting failure along the axis, typically at
maximum and minimum strength is approximately 1.03e17.61,
0 j  15 ; (2) Shear sliding failure along the weak plane,
and the average ratio is approximately 2.31. There are great dif-
typically at 15 j  60 ; and (3) Shear failure across the weak
ferences between slippage along and shear across the weak plane.
plane, typically between 65 j  90 . Two different failure
Generally, shale rock attains maximum strength either at j ¼ 0 or
modes are involved under high confining pressure, shear sliding
80 T. Ma, P. Chen / Journal of Natural Gas Science and Engineering 26 (2015) 72e98

failure along the weak plane, typically at 15 j  60 , and shear

Chenevert and Gatlin,

failure across the weak plane, typically 0 j  15 or

Willson et al., 1999

Willson et al., 2007
Niando et al., 1997
65 j  90 . The anisotropic strength of shale is controlled by

Heng et al., 2014

Aoki et al., 1993

Yan et al., 2013

Lu et al., 2012
the failure modes of shale rock; and the failure of weak planes is

Data sources

Zhang, 2009

Testing data
Testing data
a key factor.

Liu, 2009
1965
2.3.3. The shear strength of intact rock and a weak plane
To understand the shear strength of shale rocks, standard core

60 MPa
samples of 50 mm
50 mm
50 mm were cored, and direct shear

1.39

1.33
1.17
1.32
1.30
2.66
1.09
1.96
1.71
1.21
2.44
1.86
tests were conducted on Longmaxi shale. In addition to these tests,

The strength ratio of between

intact rock and a weak plane
strength data for intact rock and weak planes of shale were

40 MPa
collected from the literature. These data and the data from the

1.46

1.36
1.25
1.38
1.31
2.93
1.14
2.06
1.89
1.26
2.45
1.96
direct shear tests are listed in Table 3. To distinguish the strength
difference between intact rock and weak planes, we calculated the

20 MPa
compressive strength of intact rock and weak planes under

1.61

1.44
1.42
1.55
1.33
3.57
1.30
2.31
2.35
1.36
2.47
2.23
different confining pressures. We also drew a cross-plot of the
compressive strength between the intact rock and a weak plane, as

0 MPa
shown in Fig. 6.

2.03

2.12
2.45
4.32
1.61
7.06
4.75
3.93
5.81
1.89
2.64
3.98
The relationship between the compressive strength of intact
rock and the weak plane can be expressed as,

60 MPa

251.70

265.18
148.91
163.81
125.55
85.11
370.55
127.43
168.94
244.66
171.11
141.43
 

Strength of weak plane under different

ssw ¼ 0:71794ss0  18:65710 R2 ¼ 0:80069 (32)

confining pressures (s1－s3)a/MPa

The results showed that the strength ratio of between the

40 MPa

189.33

183.91
105.28
111.44
86.56
60.44
250.82
89.54
117.88
174.33
119.10
99.39
intact rock and the weak planes is approximately 1.61e7.06 at
s3 ¼ 0 MPa, 1.30e3.57 at s3 ¼ 20 MPa, 1.14e2.93 at s3 ¼ 40 MPa,
and 1.09e2.66 at s3 ¼ 60 MPa. In other words, the strength ratio

20 MPa
gradually decreases with the confining pressure. There is a great

126.95

102.64
61.65
59.07
47.56
35.78
131.09
51.65
66.82
104.01
67.10
57.35
difference between shear failure along and across the weak
plane.
0 MPa

64.57

21.37
18.02
6.70
8.57
11.11
11.35
13.76
15.75
33.68
15.09
15.31
2.3.4. Changes in shale strength with exposure to mud
The shale strength changes when a shale is exposed to mud, and
4w/( )

various researchers have explored these changes. However, very

30.96

37.23
21.80
26.57
18.78
6.00
45.54
18.00
25.92
33.86
26.39
20.81
Weak planes

few researchers have studied the changes in the strength of the

intact rock matrix and weak planes, respectively. Yan et al. (2013)
cw/MPa

researched the changes in shale strength with exposure to

18.28

5.30
6.10
2.07
3.07
5.00
2.32
5.00
4.93
8.98
4.68
5.28
different muds (water-based and oil-based), and the experimental
results from the literature are listed in Table 4. We used a nonlinear
curve fit method, a logistic model, to fit the experimental data, as
60 MPa

349.15

352.23
174.96
216.07
162.63
226.32
403.59
249.40
288.64
297.13
416.86
262.39

The compressive strength data were calculated using the Mohr-Coulomb criteria.
shown in Fig. 7.
Strength of intact rock under different

Based on the fitting results, we derived evolution equations of

confining pressures (s1－s3)a/MPa

40 MPa

276.56

249.90
131.33
153.69
113.03
177.04
287.02
184.30
222.92
219.36
291.21
195.22
20 MPa

203.97

147.57
87.70
91.31
63.43
127.76
170.45
119.21
157.21
141.59
165.56
128.05
131.38

45.24
44.07
28.93
13.83
78.48
53.89
54.12
91.50
63.81
39.90
60.88
0 MPa
Shear strength parameters for different shales.

Intact rock matrix

40/( )

34.61

42.30
21.80
30.96
25.17
25.00
45.00
32.00
32.23
36.22
46.50
32.76
c0/MPa

34.48

10.00
14.92
8.19
4.39
25.00
11.16
15.00
25.24
16.18
7.96
16.61
Tournemire

Longmaxi 1
Pedernales

Longmaxi
Longmaxi
Trafalgar

DB shale
Permian

Shangta

M shale
Pierre 1
Chang
Type
Table 3

No.

Fig. 5. Schematic of the main failure modes of shale rock (modified from Niandou
10
11
12
1

2
3
4
5
6
7
8
9

et al., 1997).
 T. Ma, P. Chen / Journal of Natural Gas Science and Engineering 26 (2015) 72e98 81

Fig. 6. The relationship of compressive strength between intact rock and a weak plane.

strength parameters for the intact rock and weak planes of shale
rocks. The evolution equations of strength parameters are listed in
Table 5. These equations define the changes in shale strength with
exposure to mud.

2.4. Strength criteria of shale

Two different failure modes, shear failure along and across the
weak plane, are involved in the in situ test. Therefore, the strength
criteria of shale rock can be characterized by different factors based
on the primary failure modes.

2.4.1. Single plane of weakness model for weak plane

Jaeger (1960) proposed a strength criterion to characterize the
failure of transversely isotropic rocks under tri-axial compression
conditions, the so called single plane of weakness model. This
model describes the failure mode of slippage along weak planes
and can be expressed as (Lee et al., 2012),

tw ¼ cw þ snw tan 4w (33)

where
8 w
< snw ¼r
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
xx
 2   (34)
: tw ¼ twxy þ tw xz
2

Table 4
Changes in strength parameters with exposure to mud.

No. Strength Soaked time/days Soaked

parameters liquids
0 1 2 4 6 8 10

1 c0(t)/MPa 25.24 25.04 24.73 24.10 23.75 23.60 23.61 Oil-based

2 40(t)/( ) 32.23 32.03 31.82 31.24 30.83 30.71 30.63 mud (OBM)
3 cw(t)/MPa 4.93 4.54 4.12 3.60 3.20 3.09 2.95
4 4w(t)/( ) 25.92 25.22 24.99 23.68 22.64 22.36 22.31

5 c0(t)/MPa 25.24 24.58 23.92 21.28 19.93 19.49 19.29 Water-based

6 40(t)/( ) 32.23 31.68 31.06 28.81 27.61 27.48 27.45 mud (WBM)
Fig. 7. The evolution law of strength parameters.
7 cw(t)/MPa 4.93 4.07 3.25 1.77 1.01 0.71 0.59
8 4w(t)/( ) 25.92 23.43 20.03 14.91 11.72 10.98 10.71
82 T. Ma, P. Chen / Journal of Natural Gas Science and Engineering 26 (2015) 72e98

Table 5 Generally, Eq. (39) can be rewritten in the form of principal

Evolution equations of strength parameters. stress for the convenience of computing,
No. Evolution equation of strength Goodness Soaked liquids
parameters of fit (R2)
2c0 ðtÞcos 40 ðtÞ 1 þ sin 40 ðtÞ
1 c0 ðtÞ ¼ 23:4374 þ 1:7820 0.9957 Oil-Based Mud s1 ¼ þ s (40)
1þðt=3:0732Þ2:1639
(OBM)
1  sin 40 ðtÞ 1  sin 40 ðtÞ 3
2 40 ðtÞ ¼ 30:4208 þ 1:7779 0.9941
1þðt=3:6106Þ2:0834
3 cw ðtÞ ¼ 2:5246 þ 2:4037 0.9974 where
1þðt=3:2920Þ1:3886
8 
2 pffiffiffiffi p 1
4 4w ðtÞ ¼ 21:6982 þ 4:0786 0.9738
1þðt=3:7135Þ2:0135 >
>
5 6:1320 0.9945 Water-Based
< s1 ¼ pffiffiffi
>
3
J2 cos 6 
3
þ I1
3
c0 ðtÞ ¼ 18:9373 þ
1þðt=3:3370Þ2:6636 (41)
6 40 ðtÞ ¼ 27:0281 þ 5:0289 0.9906
Mud (WBM) >
> 2 p ffiffiffiffi 1
1þðt=3:0872Þ2:9965 >
: s3 ¼ pffiffiffi J2 cos 6 þ I1
7 cw ðtÞ ¼ 0:0233 þ 4:9191 0.9965 3 3
1þðt=2:8655Þ1:6640
8 4w ðtÞ ¼ 9:0163 þ 16:7994 0.9959 !
1þðt=2:7737Þ1:8729 pffiffiffi
1 3 3 J3
6 ¼ cos1  pffiffiffiffi (42)
3 2 J2 J3

2 3 where s1 is maximum principal stress, MPa; s3 is minimum prin-

swxx tw
xy tw
xz
cipal stress, MPa; I1 is the first invariant of the stress tensor, MPa; J2
sccswcs ¼ 4 t w
xy sw
yy tw
yz
5 ¼ WBT CT sccs ðtÞCBWT (35)
is the second invariant of the stress deviator, MPa; and J3 is the third
twxz tw
yz sw
zz invariant of the stress deviator, MPa.
We also defined an index of additional cohesion to determine
where tw is the resultant shear stress load on the surface of the the shear failure across the intact rock, which is expressed as,
weak plane, MPa; snw is the resultant effective normal stress load
on the surface of the weak plane, MPa; cw is the cohesion of the s1 ½1  sin 40 ðtÞ  s3 ½1 þ sin 40 ðtÞ
c0a ¼  c0 ðtÞ (43)
weak plane, MPa; and 4w is the internal friction angle of the weak 2 cos 40 ðtÞ
plane, ( ).
If we consider the influence of the strength changes when they If the index of additional cohesion is greater than 0, the shear
are exposed to mud, Eq. (33) can be expressed as, failure occurs across the intact rock matrix. A higher value of the
index of additional cohesion indicates a higher risk of collapse, and
tw ¼ cw ðtÞ þ snw tan 4w ðtÞ (36)
a lower value of the index of additional cohesion indicates a lower
We defined an index for additional cohesion of the weak plane risk of collapse.
to determine the shear failure along the weak plane. The resulting
equations can be expressed as, 2.5. Solution method of collapse pressure model
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 2 The solution flowchart for the collapse pressure model is
2 w
cwa ¼ twxy þ ðtw xz Þ  sxx tan 4w ðtÞ  cw ðtÞ (37)
shown in Fig. 8. First, the effective stress tensor around the bore-
hole must be identified. Then, if the maximum and minimum
If an index of additional cohesion of weak planes is greater principal stresses are calculated and integrated into the Mohr-
than 0, the shear failure occurs along the weak plane. A higher Coulomb criteria, the collapse pressure at which wellbore
value for the index of additional cohesion of weak planes in- collapse can be prevented can be obtained for shale rock failure
dicates a higher risk of collapse, and a lower value of the index of across the weak plane (or failure across the intact rock).
additional cohesion of weak planes indicates a lower risk of Conversely, if the effective stress tensor around the borehole is
collapse. transformed into the WCS, the collapse pressure needed to prevent
wellbore collapse if the shale rock fails along the weak plane can
2.4.2. Mohr-Coulomb criteria for intact rock be obtained. In general, the maximum collapse pressure is regar-
The most commonly used criterion in wellbore stability analyses ded as the result of the collapse pressure calculation, and the
is the Mohr-Coulomb criterion (Zoback, 2012). We also used the collapse pressure is converted to the equivalent mud weight in
Mohr-Coulomb criterion to solve wellbore stability problems for engineering, or the so called critical equivalent mud weight
shear failure across a weak plane. The Mohr-Coulomb criterion can (CEMW). A computer program to calculate the failure regions and
be expressed as, the CEMW has been developed in Matlab. Therefore, we can
calculate the failure regions and the CEMW at different times to
t0 ¼ c0 þ sn0 tan 40 (38) maintain wellbore stability for different inclinations and azimuth
angles of the borehole.
where t0 is the resultant shear stress acting on the failure surface,
MPa; sn0 is the resultant effective normal stress acting on the 3. Wellbore stability analysis and discussions
failure surface, MPa; c0 is the cohesion of the intact rock, MPa; and
40 is the internal friction angle of the intact rock, ( ). 3.1. Basic data for a field case
If we consider the influence of strength changes on shales
exposed to mud, Eq. (38) can be expressed as, To illustrate this process, the input data for shale rock, chemical,
and in situ stress properties from a test site are summarized in
t0 ¼ c0 ðtÞ þ sn0 tan 40 ðtÞ (39) Table 6.
 T. Ma, P. Chen / Journal of Natural Gas Science and Engineering 26 (2015) 72e98 83

Fig. 8. The solution flowchart for the chemical-mechanical wellbore instability model.
84 T. Ma, P. Chen / Journal of Natural Gas Science and Engineering 26 (2015) 72e98
Table 6
Input parameters.
No. Parameter/unit
1 Measured depth (MD)/m
2 True vertical depth (TVD)/m
3 Borehole size/m
4 Maximum horizontal stress/MPa
5 Azimuth angle of maximum horizontal stress/( )
6 Minimum horizontal stress/MPa
7 Vertical stress/MPa
8 Initial pore press/MPa
9 Mud weight/(g∙cm3)
10 Solute concentration of pore fluid/(mol/L)
11 Solute concentration of WBM/(mol/L)
12 Solute concentration of OBM/(mol/L)
13 Diffusion coefficient of solute/(m2/s)
14 Diffusion coefficient of hydraulic impact/(m2∙Pa1∙s1)
15 Membrane efficiency/(m2/(Pa.s))
16 Compressibility of pore fluids/MPa1
17 Formation temperature/K
18 Porosity/%
19 Poisson's ratio
20 Biot's effective stress coefficient
21 Initial cohesion of intact rock/MPa
22 Initial internal friction angle of intact rock/( )
23 Initial cohesion of weak plane/MPa
24 Initial internal friction angle of weak plane/( )
25 Dip angle of weak plane/( )
26 Dip direction of weak plane/( )
27 Inclination angle of borehole/( )
28 Azimuth angle of borehole/( )
29 Time/hours
3.2. Pore pressure propagation around a wellbore
Fig. 9 shows the solute concentration and pore pressure pro-
files for a case study at six different times. These figures reveal
the effect of WBM (Cm < C0). A solute concentration difference
(0.25 mol/L) between WBM and the shale formation significantly
increases the pore pressure and decreases the solute concentra-
tion near the wellbore. However, the influence of water and ion
transport in and out of the shale will result in an increase in the
pore pressure in the region near the borehole (compared to the
poroelastic solution), and a decrease in the solute concentration
will lead to changes in rock strength and other properties. As a
result, wellbore stability decreases compared to the initial state.
The fundamental theory of the effect of pore pressure on well-
bore stability is based on the effective stress law and Mohr's
stress circle. As shown in Fig. 10, the Mohr's stress circle gradually
shifts to the left with an increase in pore pressure, and as a result,
the stress around the borehole more easily exceeds the strength
envelope and shear failure across or along the weak plane.
Fig. 11 shows the solute concentration and pore pressure profiles
at six different times in order to reveal the impact of OBM (Cm > C0)
on wellbore stability. A solute concentration difference (0.25 mol/L)
between a shale formation and OBM significantly decreases the
pore pressure and increases the solute concentration near the
borehole. However, the influence of water and ion transport into
and out of the shale leads to a decrease in the pore pressure in the
region near the borehole (compared to the poroelastic solution). As
a result, wellbore stability improved compared to the initial con-
dition. The functional mechanism of pore pressure on wellbore
stability runs opposite the mechanism shown in Fig. 10. Mohr's
stress circle gradually shifts to the right with a decrease in pore
pressure, and as a result, the stress around the borehole more easily
stays under the strength envelope.
3.3. Prediction of failure regions around a wellbore
Value To study the effect of the azimuth angle of a borehole on its
2420
stability, the failure regions around the horizontal well were
1610 simulated under drilling conditions along the direction of
0.216 maximum horizontal stress (sH), the direction of minimum hori-
85.43 zontal stress (sh), the dip direction of the weak plane, and the strike
N109 W
of the weak plane; the calculated results are shown in Figs. 12e15.
45.98
62.44 The failure regions around the horizontal well when drilling along
26.65 sh (19 ) are shown in Fig. 12, the failure regions around the hori-
1.80 zontal well when drilling along sH (109 ) are shown in Fig. 13, the
1.00
failure regions around the horizontal well when drilling along the
0.75
1.25
strike of the weak plane (42 ) are shown in Fig. 14, and the failure
4.9E-10 regions around the horizontal well when drilling along the dip
8.41E-18 direction of weak plane (132 ) are shown in Fig. 15. In these figures,
4.97E-17 the solid color regions are stable and the other colors are unstable.
1.45E-4
We can draw the following conclusions from the data presented in
323.15
5.00 Figs. 12e15:
0.22
0.80 (1) The failure regions around the horizontal well did not occur
25.24 in the commonly seen “dog ear” shape. Instead, the weak
32.23
4.93
planes led to failure regions that formed in a square shape,
25.92 which mirrored typical physical experimental results
7.00 (Okland and Cook, 1998; Li et al., 2012; Lee et al., 2012; Ma
N132 E and Chen, 2014b). The “dog ear” shape occurred in
0e100
Fig. 12(a) such that the failure regions occur within two
0e360
0e720 ranges (338 e22 , 158 e202 ) around the wellbore circum-
ference in the shale gas formation. The square shape was
seen in Figs. 12e15(b)e(h), and these figures reveal that
failure regions occur within four ranges around the wellbore
circumference in the shale gas formation. In other words, if
weak planes in a shale reservoir are significantly developed,
the formation around the borehole may shear more easily
along the weak planes. The effect of anisotropic strength is
detrimental to wellbore stability, which increases the risk of
collapse in horizontal wells.
(2) The failure regions around the horizontal well gradually
enlarge at different times, with the enlargement rate of the
failure regions at a maximum in the early stage after drilling
(especially the first five days), as shown in Figs. 12e15. In
particular, the evolution law of strength parameters reveals
that water and ion transport in to or out of shale reduces its
compressive strength, and the early period of exposure to
mud is the main period when changes in shale strength
occur. The impact of the strength parameters decreases
drastically as the soaking time increases in the first five days;
after that, the strength parameters' effect decreases slowly.
In addition, pore pressure propagation also affects wellbore
stability. Pore pressure build-up will result in poorer well-
bore stability later in drilling, while a reduction in pore
pressure will result in better wellbore stability as the time
increases. Therefore, a decrease in shale strength and an in-
crease in pore pressure are detrimental to wellbore stability.
In other words, changes in shale strength and pore pressure
propagation are key factors that induce wellbore instability.
In the first five days of drilling, the effect of the decrease in
shale strength is larger than the impact of the increase in
pore pressure; after that, the effect of pore pressure increases
may be larger than the impacts of a decrease of shale
strength.
(3) The failure regions around the horizontal well gradually
enlarge under the conditions of OBM and WBM; however,
the enlargement rate of the failure regions under the
 T. Ma, P. Chen / Journal of Natural Gas Science and Engineering 26 (2015) 72e98 85

Fig. 9. Solute concentration and pore pressure profiles around the borehole under the condition of WBM.

Fig. 10. The fundamental theory of the influence of pore pressure on wellbore stability.

condition of WBM is greater than that of OBM at the same
drilling time, as shown in Figs. 12e15(c)-(h). In particular,
pore pressure increased and strength decreased when drilled
by WBM, while both pore pressure and strength decreased
when drilled by OBM. In addition, the amplitude of the
decrease in shale strength under WBM conditions was larger
than under OBM conditions. Therefore, the OBM wellbore is
more stable than the WBM wellbore.
(4) The failure regions in the traditional model occur only on the
surface of borehole. However, in the new model, failure
86 T. Ma, P. Chen / Journal of Natural Gas Science and Engineering 26 (2015) 72e98
Fig. 11. Solute concentration and pore pressure profiles around the borehole under the condition of OBM.
regions may also occur in the interior of the formation, as
shown in Figs. 12e15. These results agree with typical results
from the literature (Yu, 2002). The real cause of failure re-
gions on the wellbore surface and inside the formation is
pore pressure propagation. In other words, water and ion
transport into or out of the shale increases the pore pressure
around the borehole. Therefore, the surface and inside of the
wellbore must be considered in order to calculate the CEMW
requirements to avoid collapse.
(5) Figs. 12e15 also show the wellbore stability condition after
the azimuth angle of the borehole was changed to orient it
along the path of the minimum horizontal stress sh (19 ), the
path of maximum horizontal stress sH (109 ), the strike of
the weak plane (42 ), and the dip of the weak plane (132 ).
These figures reveal that the horizontal borehole become
more stable when the azimuth angle of the borehole is
changed to follow the path of maximum horizontal stress sH
(109 ), while the horizontal borehole becomes more unsta-
ble when the azimuth angle of the borehole is changed to
follow the path of minimum horizontal stress sh (19 ).
(6) Wellbore instability is generally caused by in situ stress, pore
pressure propagation, anisotropic strength, and changes in
shale strength. The impacts of pore pressure propagation and
the changes in shale strength depend on time, so wellbore
stability is also a function of time.
3.4. CEMW and failure regions in the field case
(1) Figs. 16 and 17 show the CEMW requirements to prevent
collapse when drilling a stable horizontal well at different
times under WBM and OBM conditions. The CEMW re-
quirements for collapse at different times are plotted as a
function of the azimuth angle of the borehole. When time
increases from 0 day to 5 days, and from 5 to 10 days, the
CEMW requirements for collapse increase rapidly under the
condition of WBM; and the amplification of the CEMW is
approximately 0.19e0.26 g/cm3 after 5 days of drilling and
0.31e0.43 g/cm3 after 10 days of drilling. Conversely, the
CEMW requirements for collapse increase slowly under the
T. Ma, P. Chen / Journal of Natural Gas Science and Engineering 26 (2015) 72e98 87

Fig. 12. The failure regions at different times when drilling at 19 .
88 T. Ma, P. Chen / Journal of Natural Gas Science and Engineering 26 (2015) 72e98

Fig. 13. The failure regions at different times when drilling at 109 .
T. Ma, P. Chen / Journal of Natural Gas Science and Engineering 26 (2015) 72e98 89

Fig. 14. The failure regions at different times when drilling at 42 .
90 T. Ma, P. Chen / Journal of Natural Gas Science and Engineering 26 (2015) 72e98

Fig. 15. The failure regions at different times when drilling at 132 .
 T. Ma, P. Chen / Journal of Natural Gas Science and Engineering 26 (2015) 72e98
(3) The failure regions around the horizontal well during
Fig. 16. CEMW at different times vs. azimuth of horizontal well under WBM
conditions.
(4) The wellbore stability analysis results for horizontal well
Fig. 17. CEMW at different times vs. azimuth of horizontal well under OBM conditions.
OBM condition, and the amplification of the CEMW is
approximately 0.03e0.04 g/cm3 after 5 drilling days and
0.04e0.05 g/cm3 after 10 drilling days. The amplification of
the CEMW in the first five days is larger than the amplifica-
tion in the latter 5 days. In particular, the effect of the
decrease in shale strength and pore pressure propagation
under WBM conditions is larger than under OBM conditions,
lead to a rapid increase in the CEMW requirements for
collapse.
(2) The well X201-H1 was drilled in the Longmaxi shale at a 25
angle, as shown in Figs. 16 and 17. The initial CEMW required
to prevent borehole collapse from shear failure (breakouts) is
~1.72 g/cm3 for the Longmaxi shale in this well. To prevent
borehole collapse caused by weak bedding plane failure
under WBM conditions, a CEMW of ~2.00 g/cm3 is required
at t ¼ 0 days, a CEMW of ~2.21 g/cm3 is required at t ¼ 5 days,
91
and a CEMW of ~2.32 g/cm3 is required at t ¼ 10 days. To
prevent borehole collapse caused by weak bedding plane
failure under OBM conditions, a CEMW of ~2.00 g/cm3 is
required at t ¼ 0 days, a CEMW of ~2.03 g/cm3 is required at
t ¼ 5 days, and a CEMW of ~2.05 g/cm3 is required at t ¼ 10
days.
drilling along the azimuth angle of a real borehole (X201-
H1), both at different times and under WBM and OBM
conditions, are shown in Figs. 18 and 19, respectively.
These figures reveal that the failure regions of the test
borehole occur within four ranges around the wellbore
circumference in the shale gas formation when the new
model is used; but the failure regions occur in two ranges
when the traditional model is used. This indicates that
anisotropic strength has a detrimental effect on wellbore
stability. The failure regions around the horizontal well
gradually enlarge at later times, and the enlargement rate
of the failure regions was at a maximum in the early stage
after drilling began (especially in the first five days). In
addition, the enlargement rate of the failure regions under
WBM conditions was greater than under OBM conditions
for the same drilling times; the failure regions occurred on
the surface of the borehole and in the interior of the for-
mation, as shown in Fig. 18(c)e(h) and Fig. 19(c)e(h).
Therefore, based on the case study, wellbore instability
relates to various factors, including in situ stress, pore
pressure propagation, anisotropic strength, changes of
shale strength, etc.
X201-H1 in Longmaxi shale are shown in Fig. 20. To pre-
vent borehole collapse caused by weak bedding plane
failure under OBM conditions using the traditional model,
a low CEMW of approximately 1.72 g/cm3 is needed for
Longmaxi shale. When calculated using the model pro-
posed in this study, the lower CEMW of Longmaxi shale is
approximately 2.05 g/cm3. The real EMW was approxi-
mately 1.32e1.89 g/cm3 at a depth of 1500e2720 m from
1/20/2011 to 2/27/2011. The real equivalent mud weight
(EMW) is lower than the CEMW, and led to numerous
accidents, such as cave-ins, hole collapses, a stuck pipe,
reaming and so on. When repeated circulation and clean-
ing of the hole with high mud weight (MW) 2.19e2.32 g/
cm3 occurred from 3/12/2011 to 3/23/2011, there were
15e18 m3 cave-ins (maximum size 10 cm) at the surface.
The borehole stability analysis results were basically in
accord with the actual situation at the 1500e2720 m sec-
tion of well X201-H1, further verifying the accuracy of this
new model. In general, the single most important cause of
borehole instability is incorrect EMW (either static or
circulating equivalent density); too low an EMW can cause
compressive (shear) failure around the borehole, as shown
in Figs. 21 and 22, while the failure regions are smaller
under the condition of higher EMW. Therefore, if we
design a reasonable EMW (stress support), use an appro-
priate mud system (chemical inhibition), optimize the
design of the well path, and use the necessary plugging
approaches (superfine calcium carbonate, graphite, fiber,
or asphalt plugging agent; physical plugging), we can keep
the borehole from failing during drilling and after
completion, even in shale formations, which would result
in reduced NPT and well construction costs.
92 T. Ma, P. Chen / Journal of Natural Gas Science and Engineering 26 (2015) 72e98

Fig. 18. The failure regions at different times under the WBM condition for well X201-H1.
 T. Ma, P. Chen / Journal of Natural Gas Science and Engineering 26 (2015) 72e98 93

Fig. 19. The failure regions at different time under the OBM condition for well X201-H1.
94 T. Ma, P. Chen / Journal of Natural Gas Science and Engineering 26 (2015) 72e98
Fig. 20. Horizontal wellbore stability analyses of well X201-H1 in Longmaxi shale.
4. Conclusions
(1) The anisotropic mechanical properties and the changes in
the strength of shale rocks were analyzed using data
collected from tri-axial compression experiments, direct
shear tests and the literature. The results indicated that the
influence of anisotropic elasticity behavior is relatively small.
Two different failure modes, shear failure along and across
the weak plane, typically at 15 j  60 and 65 j  15
(0 j  15 or 65 j  90 ), respectively, are involved in
wellbore instability in the in situ test. The strength ratio
between the intact rock and the weak plane is approximately
1.61e7.06 at s3 ¼ 0 MPa, 1.30e3.57 at s3 ¼ 20 MPa, 1.14e2.93
at s3 ¼ 40 MPa, and 1.09e2.66 at s3 ¼ 60 MPa; the strength
ratio gradually decreases with the confining pressure. The
nonlinear evolution equation for the strength parameters
obeyed the logistic model. Water and ion transport in to or
out of shale reduces its compressive strength. Exposure to
mud during the first days of drilling lead to the greatest
changes in shale strength, and the strength parameters
decrease drastically as the soaking time increases in the first
five days; after that, the impact of the strength parameters
decreases slowly.
(2) The chemical-mechanical collapse pressure model for
shale was proposed based on the analytical solution of
stress induced by mechanical, hydraulic, and chemical ef-
fects and the influence of the formation properties, the
fluid properties and the environment considered in the
model, such as the in situ stress, mud weight (or downhole
pressure), water transport, ions transport, pore pressure
propagation, anisotropic strength, changes in strength, and
weak planes. We obtained a semi-analytical method of
evaluating the collapse pressure for shale gas reservoirs.
The pore pressure propagation was solved using the finite
difference method, and the solution method of the
chemical-mechanical wellbore instability model was also
obtained using a numerical method. This method can be
used to analyze the failure regions around the borehole
and the CEMW requirements for collapse at different
times.
(3) The failure regions around the horizontal well were
simulated at different times when drilling along the di-
rection of the minimum horizontal stress, the direction of
the maximum horizontal stress, the strike of the weak
plane, the dip direction of the weak plane, etc. The results
showed that the failure regions could occur not only on the
wellbore surface but also inside the formation; conversely,
the failure regions of the traditional model occur only on
the wellbore surface. The failure regions occur within four
ranges around the wellbore circumference in the shale gas
formation.
(4) The real reasons for wellbore instability are mainly the in situ
stress, the pore pressure propagation, the anisotropic
strength, and changes in the shale strength. The pore pres-
sure propagation and the changes in the shale strength
depend on the time, so the wellbore stability condition also
depends on the time. The decrease in the shale strength and
the increase in the pore pressure are detrimental to the
wellbore stability. The fundamental theory of the effect of
pore pressure on wellbore stability is based on the effective
stress law and the Mohr's stress circle; the effect of pore
pressure on wellbore stability is to change the effective
normal stress on weak plane and leads to the stress state
more easily exceeding the strength envelope and shear fail-
ure occurring across along weak plane.
(5) The collapse pressure under conditions of OBM is better than
that under conditions of WBM at well X201-H1 in the
Sichuan basin. The effect of the decrease in the shale strength
and pore pressure propagation under WBM conditions is
larger than that under OBM conditions, which causes the
CEMW requirements for collapse to increase rapidly. There-
fore, to maintain borehole stability, we must prevent well-
bore collapse by a series of approaches, such as reasonable
MW, a mud system, a well path, and physical plugging. As a
result, both the NPT and the well construction costs will be
reduced.
 T. Ma, P. Chen / Journal of Natural Gas Science and Engineering 26 (2015) 72e98 95

Fig. 21. The failure regions of different EMW of well X201-H1 under the OBM condition.
96 T. Ma, P. Chen / Journal of Natural Gas Science and Engineering 26 (2015) 72e98
Fig. 22. The failure regions of different EMW of well X201-H1 under the WBM condition.
Acknowledgments
This work is supported by a grant from the Major National Basic
Research Development Program of China (973 Program) (Grant No.
2013CB228003), the Science and Technology Support Program of
Sichuan Province (Grant No. 2015SZ0003) and the Basic Research
Subject of the State Key Laboratory of Oil & Gas Reservoir Geology
and Exploitation (Southwest Petroleum University) (Grant No G3-
1).
References
Aadnøy, B.S., 1987. Continuum Mechanics Analysis of the Stability of Inclined
Boreholes in Anisotropic Rock Formations. Ph.D. Dissertation. Norwegian
Institute of Technology.
Aadnøy, B.S., 1988. Modeling of the stability of highly inclined boreholes in aniso-
tropic rock formations. SPE Drill. Eng. 3 (3), 259e268.
Aadnøy, B.S., Chenevert, M.E., 1987. Stability of highly inclined boreholes. SPE Drill.
Eng. 2 (4), 364e374.
Aadnoy, B.S., Ong, S., 2003. Introduction to special issue on borehole stability. J. Pet.
Sci. Eng. 38 (3), 79e82.
Abousleiman, Y., Cui, L., 1998. Poroelastic solutions in transversely isotropic media
for wellbore and cylinder. Int. J. Solids Struct. 35 (34), 4905e4929.
Al-Bazali, T.M., Zhang, J., Wolfe, C., Chenevert, M.E., Sharma, M.M., 2009. Wellbore
instability of directional wells in laminated and naturally fractured shales.
J. Porous Media 12 (2), 119e130.
Aoki, T., Tan, C.P., Bamford, W.E., 1993. Effects of deformation and strength anisot-
ropy on borehole failure in saturated shales. Int. J. Rock Mech. Min. Sci. Geo-
mech. Abstr. 30 (7), 1031e1034. Pergamon.
Bradley, W.B., 1979. Failure of inclined boreholes. J. Energy Resour. Technol. 101 (4),
232e239.
Cheatham, J.B., 1984. Wellbore stability. J. Pet. Technol. 36 (6), 889e896.
Chen, X., Tan, C.P., Detournay, C., 2003a. A study on wellbore stability in fractured
rock masses with impact of mud infiltration. J. Pet. Sci. Eng. 38 (3), 145e154.
Chen, G., Chenevert, M.E., Sharma, M.M., Yu, M., 2003b. A study of wellbore stability
in shales including poroelastic, chemical, and thermal effects. J. Pet. Sci. Eng. 38
(3), 167e176.
 T. Ma, P. Chen / Journal of Natural Gas Science and Engineering 26 (2015) 72e98
Chen, M., Jin, Y., Zhang, G.Q., 2008a. Petroleum Related Rock Mechanics. Science
Press.
Chen, X., Yang, Q., Qiu, K.B., Feng, J.L., 2008b. An anisotropic strength criterion for
jointed rock masses and its application in wellbore stability analyses. Int. J.
Numer. Anal. Methods Geomech. 32 (6), 607e631.
Chen, G., Ewy, R.T., Yu, M., 2010. Analytic solutions with ionic flow for a pressure
transmission test on shale. J. Pet. Sci. Eng. 72 (1), 158e165.
Chen, P., Ma, T.S., Xia, H.Q., 2014a. A collapse pressure prediction model of hori-
zontal shale gas wells with multiple weak planes. Nat. Gas. Ind. 34 (12), 87e93.
Chen, T.Y., Feng, X.T., Zhang, X.W., Cao, W.D., Fu, C.J., 2014b. Experimental study on
mechanical and anisotropic properties of black shale. Chin. J. Rock Mech. Eng.
33 (9), 1772e1779.
Chenevert, M.E., Gatlin, C., 1965. Mechanical anisotropies of laminated sedimentary
rocks. SPE J. 5 (1), 67e77.
Cheng, Y.F., Wang, J.Y., Zhao, T.Z., Shen, H.C., 2006. Strength analysis of formation
rock for porous media coupled with mechanical-chemical effects. Chin. J. Rock
Mech. Eng. 25 (9), 1912e1916.
Cho, J.W., Kim, H., Jeon, S., Min, K.B., 2012. Deformation and strength anisotropy of
Asan gneiss, Boryeong shale, and Yeoncheon schist. Int. J. Rock Mech. Min. Sci.
50, 158e169.
Cui, L., Abousleiman, Y., Cheng, A.H.D., 1997. Poroelastic solution for an inclined
borehole. J. Appl. Mech. 64 (1), 32e38.
Das, I., 2013. Long Period Long Duration Seismic Events during Hydraulic Stimu-
lation of Gas Reservoirs. Ph.D. Dissertation. Stanford University.
Ekbote, S.M., 2002. Anisotropic Poromechanics of the Wellbore Coupled with
Thermal and Chemical Gradients. Ph.D. Dissertation. The University of
Oklahoma.
Fjar, E., Holt, R.M., Raaen, A.M., Risnes, R., 2008. Petroleum Related Rock Mechanics,
second ed. Elsevier.
Gupta, D., Zaman, M., 1999. Stability of boreholes in a geologic medium including
the effects of anisotropy. Appl. Math. Mech. 20 (8), 837e866.
Heng, S., Yang, C.H., Zeng, Y.J., Zhang, B.P., Guo, Y.T., Wang, L., Wei, Y.L., 2014.
Anisotropy of shear strength of shale based on direct shear test. Chin. J. Rock
Mech. Eng. 33 (5), 874e883.
Horne, S.A., 2013. A statistical review of mudrock elastic anisotropy. Geophys.
Prospect. 61 (4), 817e826.
Hou, B., Chen, M., Wang, Z., Yuan, J.B., Liu, M., 2013. Hydraulic fracture initiation
theory for a horizontal well in a coal seam. Pet. Sci. 10 (2), 219e225.
Jaeger, J.C., 1960. Shear failure of anistropic rocks. Geol. Mag. 97 (1), 65e72.
Jaeger, J.C., 1971. Friction of rocks and stability of rock slopes. Geotechnique 21 (2),
97e134.
Jaeger, J.C., Cook, N.G.W., Zimmerman, R., 2009. Fundamentals of Rock Mechanics.
John Wiley & Sons.
Jia, C.G., Chen, J.H., Guo, Y.T., Yang, C.H., Xu, J.B., Wang, L., 2013. Research on me-
chanical behaviors and failure modes of layer shale. Rock Soil Mech. 34 (S2),
57e61.
Jin, Y., Yuan, J.B., Hou, B., Chen, M., Lu, Y.H., Li, S., Zou, Z.P., 2012. Analysis of the
vertical borehole stability in anisotropic rock formations. J. Pet. Explor. Prod.
Technol. 2 (4), 197e207.
Jin, Y., Qi, Z.L., Chen, M., Zhang, F.X., Lu, Y.H., 2013. Study on mechanisms of borehole
instability in naturally fractured reservoir during production test for horizontal
wells. Pet. Sci. Technol. 31 (8), 829e839.
Josh, M., Esteban, L., Delle Piane, C., Sarout, J., Dewhurst, D.N., Clennell, M.B., 2012.
Laboratory characterisation of shale properties. J. Pet. Sci. Eng. 88e89, 107e124.
Kang, Y., Yu, M.J., Miska, S.Z., Takach, N., 2009, October. Wellbore stability: a critical
review and introduction to DEM. In: The SPE Annual Technical Conference and
Exhibition. Society of Petroleum Engineers (SPE 124669).
Kanitpanyacharoen, W., Wenk, H.R., Kets, F., Lehr, C., Wirth, R., 2011. Texture and
anisotropy analysis of Qusaiba shales. Geophys. Prospect. 59 (3), 536e556.
Kim, H., Cho, J.W., Song, I., Min, K.B., 2012. Anisotropy of elastic moduli, P-wave
velocities, and thermal conductivities of Asan Gneiss, Boryeong Shale, and
Yeoncheon Schist in Korea. Eng. Geol. 147, 68e77.
Lee, H., Ong, S.H., Azeemuddin, M., Goodman, H., 2012. A wellbore stability model
for formations with anisotropic rock strengths. J. Pet. Sci. Eng. 96, 109e119.
Lee, H., Chang, C., Ong, S.H., Song, I., 2013. Effect of anisotropic borehole wall failures
when estimating in situ stresses: a case study in the Nankai accretionary
wedge. Mar. Pet. Geol. 48, 411e422.
Li, Y.F., Fu, Y.Q., Tang, G., She, C.Y., Guo, J.H., Zhang, J.Y., 2012a. Effect of weak bedding
planes on wellbore stability for shale gas wells. In: IADC/SPE Asia Pacific Drilling
Technology Conference and Exhibition. Society of Petroleum Engineers (SPE
155666).
Li, Q.H., Chen, M., Jin, Y., Wang, F.P., 2012b. Experimental research on failure modes
and mechanical behaviors of gas-bearing shale. Chin. J. Rock Mech. Eng. 31 (s2),
3763e3771.
Liu, H., 2009. Research on Collapse Pressure of Layered and Disintegrated Shale.
Ph.D. Dissertation. Southwest Petroleum University.
Liu, Z.Y., Chen, M., Jin, Y., Yang, X.T., Lu, Y.H., Xiong, Q.Q., 2014. Calculation model for
borehole collapse volume in horizontal openhole in formation with multiple
weak planes. Pet. Explor. Dev. 41 (1), 113e119.
Lu, Y.H., Chen, M., Jin, Y., Zhang, G.Q., 2012. A mechanical model of borehole stability
for weak plane formation under porous flow. Pet. Sci. Technol. 30 (15),
1629e1638.
Lu, Y.H., Chen, M., Jin, Y., Ge, W.F., An, S., Zhou, Z., 2013. Influence of porous flow on
wellbore stability for an inclined well with weak plane formation. Pet. Sci.
Technol. 31 (6), 616e624.
97
Ma, T.S., Chen, P., 2014a. Study of meso-damage characteristics of shale hydration
based on CT scanning technology. Pet. Explor. Dev. 41 (2), 249e256.
Ma, T.S., Chen, P., 2014b. Prediction method of shear instability region around the
borehole for horizontal wells in bedding shale. Pet. Drill. Tech. 42 (5), 26e36.
Ma, T.S., Chen, P., 2014c. Influence of shale bedding plane on wellbore stability for
horizontal wells. J. Southwest Pet. Univ. Sci. Technol. Ed. 36 (5), 97e104.
McLamore, R., Gray, K.E., 1967. The mechanical behavior of anisotropic sedimentary
rocks. J. Manuf. Sci. Eng. 89 (1), 62e73.
Meier, T., Rybacki, E., Reinicke, A., Dresen, G., 2013. Influence of borehole diameter
on the formation of borehole breakouts in black shale. Int. J. Rock Mech. Min.
Sci. 62, 74e85.
Mody, F.K., Hale, A.H., 1993. Borehole-stability model to couple the mechanics and
chemistry of drilling-fluid/shale interactions. J. Pet. Technol. 45 (11), 1093e1101.
Niandou, H., Shao, J.F., Henry, J.P., Fourmaintraux, D., 1997. Laboratory investigation
of the mechanical behaviour of Tournemire shale. Int. J. Rock Mech. Min. Sci. 34
(1), 3e16.
Okland, D., Cook, J.M., 1998. Bedding-related borehole instability in high-angle
wells. In: The SPE/ISRM Rock Mechanics in Petroleum Engineering. Society of
Petroleum Engineers (SPE 47285).
Ong, S.H., 1994. Borehole Stability. Ph.D. Dissertation. University of Oklahoma.
Ong, S.H., Roegiers, J.C., 1993. Influence of anisotropies in borehole stability. Int. J.
Rock Mech. Min. Sci. Geomech. Abstr. 30 (7), 1069e1075. Pergamon.
Ong, S.H., Roegiers, J.C., 1996. Fracture initiation from inclined wellbores in aniso-
tropic formations. J. Pet. Technol. 48 (7), 612e619.
Pei, J.Y., 2008. Strength of Transversely Isotropic Rocks. Ph.D. Dissertation. Massa-
chusetts Institute of Technology.
Pinto, J.L., 1970. Deformability of schistose rocks. In: The 2nd ISRM International
Congress on Rock Mechanics. International Society for Rock Mechanics.
Sondergeld, C.H., Rai, C.S., 2011. Elastic anisotropy of shales. Lead. Edge 30 (3),
324e331.
Tan, Q., Deng, J.G., Yu, B.H., 2010. Wellbore instability and countermeasures in
offshore bedding shale formation. Pet. Sci. Technol. 28 (16), 1712e1718.
Tang, J., 2014. Experimental study of static and dynamic moduli for anisotropic rock.
Chin. J. Rock Mech. Eng. 33 (S1), 3185e3191.
Van Oort, E., 2003. On the physical and chemical stability of shales. J. Pet. Sci. Eng.
38 (3), 213e235.
Vasin, R.N., Wenk, H.R., Kanitpanyacharoen, W., Matthies, S., Wirth, R., 2013. Elastic
anisotropy modeling of Kimmeridge shale. J. Geophys. Res. Solid Earth 118 (8),
3931e3956.
Vernik, L., Liu, X., 1997. Velocity anisotropy in shales: a petrophysical study.
Geophysics 62 (2), 521e532.
Vernik, L., Nur, A., 1992. Ultrasonic velocity and anisotropy of hydrocarbon source
rocks. Geophysics 57 (5), 727e735.
Wang, Q., Wang, P., Xiang, D.G., Feng, S.Y., 2012. Anisotropic property of mechanical
parameters of shales. Nat. Gas. Ind. 32 (12), 62e65.
Wang, H.Y., Liu, Y.Z., Dong, D.Z., Zhao, Q., Du, D., 2013. Scientific issues on effective
development of marine shale gas in southern China. Pet. Explor. Dev. 40 (5),
615e620.
Willson, S.M., Last, N.C., Zoback, M.D., Moos, D., 1999. Drilling in South America:
stability approach for complex geological conditions. In: The SPE Latin Amer-
ican & Caribbean Petroleum Engineering Conference. Society of Petroleum
Engineers (SPE 53940).
Willson, S.M., Edwards, S.T., Crook, A., Bere, A., Moos, D., Peska, P., Last, N., 2007.
Assuring stability in extended-reach wells e analyses, practices, and mitigation.
In: The SPE/IADC Drilling Conference. Society of Petroleum Engineers (SPE/IADC
105405).
Yan, C.L., Deng, J.G., Yu, B.H., Tan, Q., Deng, F.C., 2013. Research on collapsing
pressure of gas shale. Chin. J. Rock Mech. Eng. 32 (8), 1595e1602.
Yan, C.L., Deng, J.G., Yu, B.H., Liu, H.L., Deng, F.C., Chen, Z.J., Hu, L.B., Zhu, H.Y., Han, Q.,
2014. Wellbore stability analysis and its application in the Fergana basin, central
Asia. J. Geophys. Eng. 11 (1), 15001e15009.
Yang, J., Fu, Y.Q., Chen, H.F., Zeng, L.X., Li, J.H., 2012. Rock mechanical characteristics
of shale reservoirs. Nat. Gas. Ind. 32 (7), 12e14.
Yang, D., Liu, S.G., Shan, Y.M., Sun, W., Ran, B., Luo, C., Wan, H.C., 2013. Mechanical
property of Lower Cambrian Niutitang Formation black shale in upper Yangtze
region, China. J. Chengdu Univ. Technol. Sci. Technol. Ed. 40 (6), 677e687.
Younessi, A., Rasouli, V., 2010. A fracture sliding potential index for wellbore sta-
bility analysis. Int. J. Rock Mech. Min. Sci. 47 (6), 927e939.
Yu, M., 2002. Chemical and Thermal Effects on Wellbore Stability of Shale Forma-
tions. Ph.D. Dissertation. The University of Texus at Austin.
Yu, M., Chen, G., Chenevert, M.E., Sharma, M.M., 2001. Chemical and thermal effects
on wellbore stability of shale formations. In: The SPE Annual Technical Con-
ference and Exhibition. Society of Petroleum Engineers (SPE 71366).
Yu, M., Chenevert, M.E., Sharma, M.M., 2003. Chemicalemechanical wellbore
instability model for shales: accounting for solute diffusion. J. Pet. Sci. Eng. 38
(3), 131e143.
Yu, B.H., Yan, C.L., Deng, J.G., Li, W., Hu, L.B., 2014. Rock mechanical properties and
borehole stability of gas shale. Sci. Bull. Natl. Min. Univ. 1, 21e27.
Yuan, J.L., Deng, J.G., Tan, Q., Yu, B.H., Jin, X.C., 2013. Borehole stability analysis of
horizontal drilling in shale gas reservoirs. Rock Mech. Rock Eng. 46 (5),
1157e1164.
Yuan, J.L., Deng, J.G., Tan, Q., Yu, B.H., Xiao, K., 2014. Analysis of Horizontal Borehole
Stability in Silurian Shale Gas Reservoir, Sichuan Basin. ASME (in press). http://
journaltool.asme.org/PDF/JERT/12e1185/DRAFT-JERT-12-1185-0.pdf.
Zeynali, M.E., 2012. Mechanical and physico-chemical aspects of wellbore stability
98 T. Ma, P. Chen / Journal of Natural Gas Science and Engineering 26 (2015) 72e98
during drilling operations. J. Pet. Sci. Eng. 82, 120e124.
Zhang, G.Q., 2009. Rock failure with weak planes by self-locking concept. Int. J. Rock
Mech. Min. Sci. 46 (6), 974e982.
Zhang, J., 2013. Borehole stability analysis accounting for anisotropies in drilling to
weak bedding planes. Int. J. Rock Mech. Min. Sci. 60, 160e170.
Zhang, F., Zhang, S., Jiang, X., Lu, R., Chen, M., 2008. Borehole stability in naturally
fractured reservoirs during production tests. Pet. Sci. 5 (3), 247e250.
Zhou, S., Hillis, R., Sandiford, M., 1994. A study of the design of inclined wellbores
with regard to both mechanical stability and fracture intersection, and its
application to the Australian North West Shelf. J. Appl. Geophys. 32 (4),
293e304.
Zoback, M.D., 2012. Reservoir Geomechanics. Petroleum Industry Press.
Zoback, M.D., Moos, D., Mastin, L., Anderson, N., 1985. Wellbore breakouts and in
situ stress. J. Geophys. Res. Solid Earth (1978e2012) 90 (B7), 5523e5530.
Zou, C.N., Dong, D.Z., Wang, S.J., Li, J.Z., Li, X.J., Wang, Y.M., Li, D.H., Cheng, K.M.,
2010. Geological characteristics and resource potential of shale gas in China.
Pet. Explor. Dev. 37 (6), 641e653.