Applied Clay Science 108 (2015) 165172

Contents lists available at ScienceDirect

Applied Clay Science

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

Research paper

Water activity characteristics of deep brittle shale from Southwest China

Hang Wen a,, Mian Chen a, Yan Jin a, Yayun Zhang a, Weifeng Ge b, Jinlong Du a, Chao Zeng a
a
b

State Key Laboratory of Petroleum Resource and Prospecting, China University of Petroleum-Beijing, #18, Fuxue Rd, Changping, Beijing 102249, PR China
CNOOC Safety Technology Service Co. Ltd., Tianjin 300456, PR China

a r t i c l e

i n f o

Article history:
Received 10 October 2014
Received in revised form 12 February 2015
Accepted 13 February 2015
Available online 26 February 2015
Keywords:
Water activity
Deep brittle shale
Mineralogical composition
Swelling ratio
Failure mechanism
Uniaxial compressive strength

a b s t r a c t
Understanding the effect of water activity on shale is of major importance for the exploration and development of
shale gas reservoirs, especially in the drilling and completion processes. Deep brittle shale samples from the
lower Silurian Longmaxi Formation located in Southwest China were characterized using X-ray diffraction
(XRD), swelling ratio tests, computerized tomography (CT), Field Emission Scanning Electron Microscope (FESEM) and uniaxial compressive strength (UCS) tests. Because of the physical characteristics of the samples, the
relationship between water activity and swelling ratio could be approximated by the FrenkelHalseyHill
(FHH) equation. The relationships between clay minerals, UCS and water activity were investigated. Based on
the physical description of the internal structure, the impact of water activity on micro-fractures was also
discussed. The results showed that multilayer adsorption was the main contributor to the swelling of the deep
brittle shale samples. The swelling ratio of brittle shale samples was described by the water activity of external
environment via a logarithmic expression of the FHH equation, with fractal dimensions ranging from 2.71 to
2.75. The mineralogical composition was a controlling factor on water activity, as shown by the positive
correlation between water activity and the total content of clay minerals (TCCM), especially the amount of
montmorillonite in illite/smectite mixed layer clay. When soaking in a solution with a water activity above the
critical water activity, the brittle samples developed micro-fractures because the clay swelled, acting as
cementing material in the weakened planes. The environmental water activity was used to evaluate mechanical
properties of the hydrated brittle shale. Brittle shale samples with larger water activity had a higher critical water
activity. Water activity analysis led to a better understanding of the characteristics of the brittle shale and the
performance of the downhole uid.
2015 Elsevier B.V. All rights reserved.

1. Introduction
Shale gas has recently received attention in the eld of energy production (Loucks and Ruppel, 2007; Gasparik et al., 2012). As an effective
way to cope with energy shortages and energy security, as previously
proven by the USA during the past few decades, shale gas exploration
is a viable option in China, especially in the southwest Sichuan Basin
(Chen et al., 2011; Wang et al., 2013). Compared with the Mesozoic
and Upper Paleozoic erathem as the target stratum for the USA's shale
gas industry, the Lower Paleozoic erathem for the Sichuan Basin is
deeper (Zhang et al., 2009). To reduce the cost of shale gas production,
China still requires fundamental research on the physicochemical properties of shale at increasing depths (Chen et al., 2011; Wang et al., 2013).
As a type of mud rock, shale is a ne-grained rock mainly composed
of clay mineral akes (Blatt and Tracy, 1996). Clay minerals form in the
presence of water (Ehlmann et al., 2011). The system, consisting of clay
minerals and water, has a signicant effect on the physicochemical
properties of the shale, which has been reported and observed by
Corresponding author. Tel.: +86 18810267543.
E-mail address: wenhang6024@126.com (H. Wen).

http://dx.doi.org/10.1016/j.clay.2015.02.015
0169-1317/ 2015 Elsevier B.V. All rights reserved.

using modern technologies. Water content plays an important role in

the mechanical properties of shale, especially the uniaxial compressive
strength (UCS) (Hsu and Nelson, 2002; Vales et al., 2004; Erguler and
Ulusay, 2009). It has also been shown to be related to the meso-scale
damage characteristics of shale hydration when studying the internal
morphology by computerized tomography (CT) (Shi et al., 2012; Ma
and Chen, 2014). Clay minerals can contain and adsorb water because
of their internal structure and type, which can be detected by X-ray diffraction (XRD) analysis (Ballard et al., 1994; Josh et al., 2012), and also
has been proved to be a controlling factor on bedding and deformability,
such as swelling (Wright, 2001; Fouche et al., 2004). However, compared with water content, water activity has a closer relationship with
the physicochemical properties of the material (Troller and Christian,
1978; Rockland and Nishi, 1980). Because water content cannot reect
the inuence of water on the material, water activity, as a description of
energy state, can precisely reveal the relationship between water and
the material. Water activity is a criterion used to measure the amount
of energy that would be needed for the water to be removed from the
material. The effect of environmental water activity on the mechanical
properties of a shale, such as the reduction of the compressive strength
caused by high water activity, has been observed by several researchers

166

H. Wen et al. / Applied Clay Science 108 (2015) 165172

in laboratory experiments and eld cases (Chenevert, 1970; Chen et al.,

2003; Al-Bazali et al., 2008), but to date, no report on the water activity
of a shale and its impact on the shale physical and mechanical properties
has been published. This study focused on shale characteristics from a
water activity perspective.
The major goals of this paper are to investigate the water activity
characteristics of deep brittle shale from the Sichuan Basin in China.
An equation of a similar form to the FrenkelHalseyHill (FHH) equation, an isotherm equation containing the fractal dimension and describing the surface coverage as a function of the equilibrium pressure,
is applied to investigate the relationship between water activity of the
external environment and the swelling ratio of the brittle shale. The
relationship between the mineralogical composition and water activity
of the brittle shale will also be investigated. Furthermore, based on the
physical signicance of water activity, the effects of the water activity
in the external environment on the micro-fracture mechanisms and
mechanical properties of brittle shale will also be discussed.

cumulative thickness of 153 m. For the formation, the type of organic matter is III1, a favorable type for shale gas generation with an R0
(vitrinite reectance) value 2.0%, the formation was in the shale
gas generation stage. Test samples from the formation were cut
from preserved cores obtained from depths ranging from approximately 4350 m to 4370 m and 2377 m to 2415 m, respectively.
More detailed information on the stratigraphy, geology and petroleum potential of these cores can be obtained from Wang et al.
(2013) and the references therein.
The Lower Cambrian Niutitang Formation may also be a shale gas
reservoir (Fig. 2). The strata were below the Longmaxi Formation and
formed in a similar sedimentary environment (Nie et al., 2011). The
stratigraphy of the upper part of the formation consisted of gray silty
mudstone, siltstone and limestone, whereas the lower part of the formation consisted of black shale (Yang et al., 2014). Relevant detailed
geological information can be found in Nie et al. (2011) and Wang
et al. (2013). Because of the geological correlation between the
Longmaxi and the Niutitang Formations, the latter was used as a comparison to the former.

2. Experimental procedures
2.1. Materials

2.2. Equipment

Cores obtained from the Sichuan Basin located in Southwest China

were used in this study (Fig. 1). The cores consisted of black shale and
were obtained from the lower Silurian Longmaxi Formation (Fig. 2).
Geological evidence indicates that the Longmaxi Formation formed in
a shelf facies sedimentary environment (Wang et al., 2013). The
stratigraphy of the lower part of the Longmaxi Formation consists
of silty mudstone, silt and carbonaceous mudstone and carbon
mudstone. With TOC (total organic carbon) 0.57%, it has a

The samples were analyzed experimentally using XRD analysis,

water activity tests, swelling ratio tests, CT analysis, Field Emission
Scanning Electron Microscope (FE-SEM) analysis and UCS tests
(Chalmers et al., 2012).
XRD analysis was performed with a MiniFlex II Desktop X-ray
diffraction instrument from Rigaku Corporation in Japan. The sample
preparation, experiment operation and quantitative analysis were
performed according to the U.S. Geological Survey Open-File Report

Fig. 1. Isopach map of the Longmaxi Formation in the southern Sichuan Basin (Chen et al., 2011).

H. Wen et al. / Applied Clay Science 108 (2015) 165172

167

Fig. 2. Paleozoic stratigraphic column of the Sichuan Basin.

Modied from Chen et al., 2011.

01-041 (A Laboratory Manual for X-Ray Powder Diffraction). The test

samples were ground in air into dry and stable powder, with a particle
size smaller than 20 m, and then prepared into randomly oriented aggregate mounts and oriented aggregate mounts, which were used to
make oriented aggregate mounts for ethylene glycol and heat treatments. Cu K alpha weighted-mean radiation was used and the 2-theta
range of randomly oriented aggregate mounts and the oriented aggregate mounts were 345 and 2.515, respectively. The 2-theta
range for the ethylene glycol and heat treatments under 550 C were
2.530 and 2.515, respectively. The step size was 0.01 and scan
speed was 1/min. The DS (divergence slit) angle was 1.25, and the
width of RS (receiving slit) was 0.3 mm. The accuracy of randomly
oriented aggregate mounts was 0.1% and others were 1%.
Water activity was determined by an XLVF4-LabSwift-aw water activity meter made by Novasina in Switzerland. The sample preparation
and experiment operation were performed according to Novasina aw
Center Operation Manual (Novasina, 1988). The test samples, with
maximum volumes of 40 12 mm, were put into a sealed pot until
the environmental balance was reached at 24 C 1 C. The measuring
range was 0.031.00 with an accuracy of 0.01.
The swelling ratio was determined with a M4600 HTHP linear swelling device manufactured by Grace Instrument in the US. The test samples, weighing 10.5 g, were prepared by rst sieving to a maximum
particle size of 20 m and then pressing the material into plugs with diameter of 25.4 mm and height of N10.16 mm for 30 min at 13.8 MPa.
After soaking in 75 ml liquid samples under atmospheric conditions
(Chenevert, 1970), the M4600daqTM software collected axial swelling
ratio data until the program automatically stopped when the increasing
ratio of the data was less than 0.01%.

CT analysis was performed with a Micro XCT-400 micron X-ray

microscope/CT analyzer from Xradia Corporation in the US. The
test samples were cored by a Laboratory Drill Stand from ELE International in the UK into plugs with a diameter of 25.4 mm. The pixel
resolution was 988 1013, and the spatial resolution was
25.7 25.1 m2.
FE-SEM analysis was performed with a Quanta 200F eld emission
scanning electron microscope under circular scanning from FEI in
Holland. The test samples with maximum area of 54 mm were coated
with gold for 20 min using an SCD 500 sputter coater from Austria. The
samples were detected using the secondary electrons mode (SE) and
energy dispersive X-ray spectroscopy (EDX). The high voltage (HV) is
20.00 kV, the horizontal eld width (HFW) was 51.2 m, the magnication (mag) was 5000 and the working distance (WD) was approximately 10 mm. The resolution could reach 1.2 nm, and the maximum
magnication was 25200 K.
The UCS test results were collected using a GCTS-RTR-1500 rock mechanics testing machine. The test samples were 25.4 mm in diameter
and 40 mm in height. The UCS, static Young's modulus (E) and Poisson's
ratio (), dened as the tangent of the stressstrain curve at 50% peak
load, were measured with an axial strain ramp rate of 0.05% per minute.
3. Results
3.1. Compositional analysis
For the whole rock pattern, the samples were prepared in randomly oriented aggregate mounts, and the mineralogical composition was determined by the intensity area corresponding to the

168

H. Wen et al. / Applied Clay Science 108 (2015) 165172

presenting phases detected by the Bragg equation. To minimize the

preferred orientation and grain gradation effect, double-peak determination technology was used in mineral analysis; for instance,
quartz was determined by a 3.34 and 4.26 reections. For the
separated clay fraction pattern, samples were prepared in oriented
aggregate mounts; the mineralogical composition was determined
in the same manner as above. The mounts treated by ethylene glycol indicated that there was no a 17 reection on the X-ray
diffractogram. Thus, for the relative contents of clay minerals,
montmorillonite was absent and an illite/smectite (I/S) mixed
layer was present. All the X-ray diffractograms were performed
twice to ensure a consistent pattern. The XRD analysis results are
summarized in Table 1. Samples S-1, S-2 and S-3 were obtained
from a burial depth of 43504370 m and S-4, S-5 and S-6 from
2377 to 2415 m. The test samples contained abundant quartz and
clay minerals. The quartz content averaged 38.9% and ranged
from 16.3% to 65.6%, the carbonate mineral (calcite and dolomite)
content averaged 21.4% and was between 1.9% and 35.5%, and the
total content of clay minerals (TCCM) averaged 31.2%, with a
range between 21.1% and 40.1%. Compared with the Lower Cambrian Niutitang Formation in the Sichuan Basin, for which the average
values of quartz, carbonate and TCCM were 35.2%, 16.6%, and 26.9%,
respectively (Yang et al., 2014); the mineral composition of this
formation had a similar average value of quartz and carbonate;
and there was a higher average value of TCCM. However, for relative contents of clay minerals, the test samples contained an
abundant I/S mixed layer and illite, which was different from earlier descriptions (Chen et al., 2011; Wang et al., 2013). The I/S mixed
layer content averaged 63.7% and ranged from 44% to 81%, the illite
content averaged 15% and was between 4% and 39%, and the percentage of montmorillonite in the I/S mixed layer averaged 15.5%
with a range between 5% and 35%.
3.2. Water activity and swelling ratio results
Water activity and swelling ratio measurements were performed
three times, and the results were expressed as the mean value of
these measurements. The water activity of the brittle shale samples
varied between 0.73 and 0.81, showing an appreciable repeatability
with an SE value 5.77 10 3 (Table 2). Swelling did not take
place when the water activity of the shale was lower than the activity
of the external environment liquid (Chenevert, 1970). Therefore,
some saturated solutions with water activity above the critical
water activity were used in this study. The solvent used for these
samples was deionized water, and water activity ranged from 0.81
to 0.97 with excellent repeatability, that is, almost the same with
the standard value. The results of the swelling ratio analysis are
shown in Table 2. The swelling ratio data of the brittle shale samples
was obtained by recording measurements every 30 s and ranged
from 0.8% to 14.7%, exhibiting a good repeatability with an SE
value 1.15 10 3.

Table 2
Water activity (aw) and swelling ratio (sw) results for brittle shale samples (24 C 1 C).
Solute

H2O

K2SO4

KNO3

KCl

Sample

Water activity

1.00

0.97

0.94

0.84

0.81

S-1
S-2
S-3
S-4
S-5
S-6

0.76
0.73
0.74
0.81
0.80
0.79

8
6
5.8
14.7
14.5
13.4

5.4
3.9
4.5
10.6
10.9
10

4.2
3.5
3.1
7.5
7.3
7.2

2.7
1.7
2
5.3
4.8
5.1

1.1
0.9
0.8
2.8
2.4
2

KBr

3.3. Visualization by computerized tomography and eld emission scanning

electron microscope imaging
The macro- and meso-scale effects of the solution on the shale have
been previously studied by many researchers (Chenevert, 1970; Hsu
and Nelson, 2002; Chen et al., 2003; Erguler and Ulusay, 2009; Shi
et al., 2012; Ma and Chen, 2014). The factors controlling these effects,
that is, the mechanical properties and damage characteristics of shale
are controlled by the type and content of solute in the solution, have
been widely recognized. This test and the UCS tests in the next section
were only performed on sample S-5. The sample water activity ranged
from 0.79 to 0.81, which is essentially a constant value. The samples
for this test were labeled as S-5-1, S-5-2 and S-5-3 to allow microvisualization. The micro-fracture structures in the brittle shale samples
were visualized with CT and FE-SEM (Fig. 3). The CT test results were
from the plug cross-sections, while the FE-SEM test results showed
the morphology of the specically imaged area, especially possible failure zones. From top to bottom, the micro-structures of brittle shale samples immersed in different saturated solutions for 24 h are illustrated
(Fig. 3). The water activity of the solutions was 0.69 for KI, 0.81 for
KBr and 0.94 for KNO3 at 24 C 1 C. In Figs. 3a and b, the original
structures of the samples were still intact, while some darker spots appeared in Fig. 3a. However, long micro-fractures crossed this portion in
the CT test results (Fig. 3c), along with some defects that appeared in the
FE-SEM images. Based on the typical morphology of minerals shown in
the literature (Morin, 1993), the structures of illite and montmorillonite
occupied a larger area in the FE-SEM images. The clusters of tight units
with their relatively large, at-lying plates were evidence of illite, while
the tight plates with the clusters were the I/S mixed player. These results are consistent with the XRD analysis (Table 1). The EDX results
(Fig. 3) obtained from the same location in the FE-SEM test also conrmed that the locations were occupied by the chemical elements that
comprise illite and montmorillonite.

3.4. Uniaxial compressive strength results

The UCS results are shown in Fig. 4. Brittle shale samples used in this
test were also obtained from sample S-5. The UCS of this sample was

Table 1
XRD results for brittle shale samples.
Sample

S-1
S-2
S-3
S-4
S-5
S-6

Quartz (%)

35.1
44.8
65.6
44.5
16.3
26.9

Feldspar (%)

7.0
5.5
6.0
8.1
2.3
7.6

Calcite (%)

11.1
4.6
2.2
1.9
10.8
0

S*: Percentage of montmorillonite in I/S mixed layer

Dolomite (%)

19.8
14.0
5.1
0
24.7
34.2

Hematite (%)

0
0
0
0
7.8
0

Pyrite (%)

0
0
0
1.5
0
0

Siderite (%)

0.5
1.3
0
3.9
0
0

TCCM (%)

26.5
29.8
21.1
40.1
38.1
31.3

Results of clay minerals analysis (%)

Relative contents of clay
minerals

Mixed layer

I/S

Kaol

Chl

S*

81
58
69
56
44
74

6
8
19
14
39
4

7
19
6
16
4
3

6
15
6
14
13
19

10
6
5
35
24
13

H. Wen et al. / Applied Clay Science 108 (2015) 165172

S-5-1

(a) Aw=0.69

S-5-2

(b) Aw=0.81

S-5-3

(c) Aw=0.94

169

Fig. 3. Microstructures from samples with different environmental water activity. The rst column shows the CT test results from samples with a diameter of 25.4 mm, observed in full
scale.

100.57 MPa, E was 26.16 GPa and was 0.21. Before the test, the brittle
shale samples were prepared by soaking in the ve saturated solutions with different water activity (Table 2) for 24 h. For these
samples, the UCS was between 99.31 MPa and 55.15 MPa, whereas
E ranged from 27.4 GPa to 22.2 GPa, and varied between 0.19

and 0.41. Compared with the Lower Cambrian Niutitang Formation

from the Sichuan Basin, with a UCS of between 107 MPa and
164 MPa (Chen et al., 2014), the original samples in this test had
lower values due to the shallower depth from which the core was
obtained (Fig. 2).

Fig. 4. Mechanical properties under different environmental water activity.

Fig. 5. Plots of ln(sw 100) vs. ln[ln(1/aw)] from the swelling and water activity tests.

170

H. Wen et al. / Applied Clay Science 108 (2015) 165172

4. Discussion
4.1. Effect of water activity on the swelling ratio of brittle shale
Based on the data obtained from the water activity (aw) and swelling
ratio (sw) tests (Table 2), the relationship between aw of the saturated
solutions and sw of the brittle shale samples is illustrated (Fig. 5).
The linear relationship can be expressed as follows:
ln sw
100 Const B  ln ln 1=aw 

aw p=p0

which is similar to the logarithmic form of the simple adsorption isotherm, the FHH equation (Avnir and Jaroniec, 1989; Jaroniec, 1995;
Sahouli et al., 1997), and where sw 100 corresponds to n, which is
the amount of water adsorbed at the relative pressure p/p0 and absolute
temperature T. Const is a characteristic constant, B corresponds to D-3 in
which D is the fractal dimension, p is the equilibrium pressure and p0 is
the saturated vapor pressure (Jaroniec, 1995). This similarity may be explained as follows: as nely divided material, the brittle shale samples in
the swelling ratio test had a very low amount of water retained by capillary condensation. The swelling predominantly occurred though multilayer adsorption (Prost et al., 1998). The amount of water in the
multilayer adsorption process, which is a direct function of water activity, can be calculated by multiplying the total surface area by the thickness of the lm of water adsorbed. It was inferred that the brittle shale
samples with a higher content of montmorillonite would have a higher
swelling ratio because of the special internal structure, as conrmed by
the data in Tables 1 and 2. The equations for regression lines in Table 3
show reasonable ts, suggesting that the above inference is likely. From
the equations, the following observations can be obtained: the brittle
shale samples were fractal; the fractal dimensions were between 2.71
and 2.75, similar to those for the Lower Cambrian Niutitang Formation
in the Sichuan Basin (Yang et al., 2014); and the brittle shale samples
with a higher fractal dimension had higher water activity (Fig. 6),
while the higher water activity samples had a higher swelling ratio in
the same water activity environment (Table 2). This can be explained
by the fact that a higher fractal dimension indicates that the brittle
shale samples have a more irregular surface, which provides a greater
surface area on which adsorption can occur (Yang et al., 2014). These
analyses further indicate that the relationship between the water activity and swelling ratio is reliable.
4.2. Relationships between water activity and the composition of brittle
shale
The TCCM in brittle shale differs from that of conventional reservoir
rock, which generally has a high value (Loucks and Ruppel, 2007). The
water activity is positively correlated with TCCM (Fig. 7a). Brittle shale
samples with higher TCCM have a greater water activity. However,
water activity does not have an obvious correlation with the amount
of I/S mixed layer clay (Fig. 7b). The relationship between the water activity and montmorillonite in the TCCM (in I/S mixed layer) of the brittle
shale samples is illustrated in Fig. 7c. A good correlation exists between

Table 3
Fractal dimensions derived from FHH model.
Sample

Regression line equation

Fitting coefcient

Fractal dimension

S-1
S-2
S-3
S-4
S-5
S-6

Y = 4.9107 0.2879X
Y = 4.6114 0.2902X
Y = 4.627 0.2874X
Y = 5.717 0.254X
Y = 5.5939 0.2752X
Y = 5.5468 0.2712X

0.6758
0.7031
0.6402
0.7398
0.7116
0.6414

2.7121
2.7098
2.7126
2.746
2.72428
2.7288

Fig. 6. Relationship between the water activity and fractal dimension of brittle shale
samples.

them; brittle shale samples with a higher water activity have a higher
content of montmorillonite in TCCM. This may be due to montmorillonite being an expansive clay mineral, which usually has a larger total surface. When it is buried in the formation, it will contain more water
molecules, thus having a higher water activity, whereas illite will not.
This idea is supported by the relationships between the water activity
and mineralogical composition of the brittle shale (Tables 1 and 2)
and also agrees well with previous research (Zhang et al., 2008a,b). Because of factors such as temperature and time, montmorillonite will
generally transform into I/S mixed layer clays or illite with increasing
depth (Moore and Reynolds, 1989). With increasing depth, water activity decreases (Zhang et al., 2008b). This is supported by Table 2. The
water activity of samples S-1, S-2 and S-3, obtained from deep strata,
is lower than that of samples S-4, S-5 and S-6 from shallow strata.
Mineralogical composition is the primary cause of physical and chemical properties of brittle shale, and the water activity can be used as a
parameter for evaluating the mineralogy of brittle shale, especially the
clay mineral content.

4.3. Effect of water activity on the micro-fracture mechanism of brittle shale

Many studies have shown that the micro-fracture mechanism of
brittle shale is dependent on the characteristics of the brittle shale and
the environment, including their chemical and physical properties
(Matin and Mazumdar, 2009; Long et al., 2011; Yao et al., 2013).
Water activity is a critical control on the micro fracture mechanism of
brittle shale, as evidenced by a series of results reecting a relationship
between the water activity of the solutions and the extent of the microfracture of the brittle shale samples (Fig. 3). The decrease in the swelling
ratio of the brittle shale samples with the water activity of the solutions
(Table 2) also indicates that the closer to the critical activity the water
activity of the solution is, the lower the deformability of the brittle
shale samples. There was no microscopic structure change in the brittle
shale sample treated by the solution with a water activity close to the
critical value, while there were micro-fractures and even micro-holes
in samples treated with solution above the critical value (Fig. 3). The
mechanism for the solution water activity as an important property
with respect to the structure development of brittle shale can be
described at the microscopic level as follows. The solution with high
water activity provided brittle shale a higher equilibrium pressure
environment that would make the surface of clay adsorb more water
molecules; the water molecules were bound in the interlayer of the
multilayer sequence, which made the clay undergo crystalline swelling
(Jo et al., 2004). When the brittle shale with weakened planes cemented
by clay (Fig. 3) was soaked in the solution, micro-fractures and even
micro-holes would appear as a result of the deformability of the
cementing material (Lal, 1999).

H. Wen et al. / Applied Clay Science 108 (2015) 165172

171

et al., 2014). This is also well illustrated by the CT and FE-SEM images
(Fig. 3).
5. Conclusion

(a) Relationship between water activity and TCCM

(b) Relationship between water activity and I/S mixed

layer in TCCM

In this paper, an equation similar to the FHH equation has been applied to investigate the relationship between the water activity of the
external environment and the swelling ratio of brittle shale. The relationship between the mineralogical composition and water activity of
the brittle shale has been investigated. Furthermore, based on the physical signicance of water activity, the effects of the external environment water activity on the micro-fracture mechanism and mechanical
properties of brittle shale have also been discussed. The following
conclusions have been reached:
(1) The swelling ratio of brittle shale samples can be described by the
water activity of the external environment in loglog space using
FHH theory. It is also affected by mineralogical composition.
Based on the theory, the fractal dimensions of the brittle shale
samples range between 2.71 and 2.75 normally, which also
indicates that the relationship between the water activity and
swelling ratio is reliable.
(2) TCCM is a controlling factor on the water activity, as shown by
the positive correlation between them. A similar relationship
also exists between montmorillonite in the I/S mixed layer and
water activity. No obvious relationship exists between water
activity and I/S mixed layer clay.
(3) After contact with a high water activity solution, the appearance
of micro-fractures and micro-holes in the brittle shale sample is
due to the swelling of the clay during which the cementing
material occurs in the weakest areas, and the higher surface
area will absorb more water molecules in a higher equilibrium
pressure environment. The solution with a higher water activity
above the critical water activity results in a larger extent of
micro-fracturing in the brittle shale.
(4) A negative correlation between water activity of the solutions
and both UCS and E is observed; a positive relationship between
water activity and is also observed. This may be related to the
observation that the brittle shale samples soaked with a higher
water activity solution will produce more micro-fractures.

Acknowledgments

(c) Relationship between water activity and montmorillonite

(S) in TCCM (in I/Smixed layer)
Fig. 7. Relationship between TCCM, I/S mixed layer, S, and water activity of brittle shale
samples.

The authors are grateful for nancial support to the National Natural
Science Foundation of China (NSFC) (No. 51234006), NSFC for Creative
Research Groups (No. 51221003), NSFC for Distinguished Young (No.
51325402) and NSFC for Major projects (No. 51490650).
References

4.4. Effect of water activity on the mechanical properties of brittle shale

The mechanical properties of the brittle samples, e.g., UCS, E and ,
correlate with water activity (Fig. 4). Based on the relationship between
water activity and water content, the observed trend is in agreement
with the results from previous work (Meier et al., 2013). Three major
factors affect mechanical properties in brittle shale: UCS, E and . The
higher the saturated solution water activity is, the more signicantly
the UCS and E decrease and increases. The largest decline and the
maximum growth rate can be up to 45.17% and 96.82%, respectively.
The results indicate that a higher water activity environment will
make a brittle shale sample weaker, as it becomes more susceptible to
the generation of micro-fractures, similar to sandstone (Rathnaweera

Al-Bazali, T., Zhang, J.G., Chenevert, M.E., Sharma, M.M., 2008. Factors controlling the
compressive strength and acoustic properties of shales when interacting with
water-based uids. Int. J. Rock Mech. Min. 45, 729738.
Avnir, D., Jaroniec, M., 1989. An isotherm equation for adsorption on fractal surfaces of
heterogeneous porous materials. Langmuir 5, 14311433.
Ballard, T.J., Beare, S.P., Lawless, T.A., 1994. Fundamentals of shale stabilization: water
transport through shales. SPE Fonnation Eval. 6, 129134.
Blatt, H., Tracy, R.J., 1996. Petrology: Igneous, Sedimentary and Metamorphic. W. H. Freeman, New York.
Chalmers, G.R., Bustin, R.M., Power, I.M., 2012. Characterization of gas shale pore systems
by porosimetry, pycnometry, surface area, and eld emission scanning electron/
transmission electron microscopy image analyses: examples from the Barnett,
Woodford, Haynesville, Marcellus, and Doigunits. AAPG Bull. 96, 10991119.
Chen, G.Z., Chenevert, M.E., Sharma, M.M., Yu, M.J., 2003. A study of wellbore stability in
shales including poroelastic, chemical, and thermal effects. J. Pet. Sci. Eng. 38,
167176.
Chen, S.B., Zhu, Y.M., Wang, H.Y., Liu, H.L., Wei, W., Fang, J.H., 2011. Shale gas reservoir
characterisation: a typical case in the southern Sichuan Basin of China. Energy 36,
66096616.

172

H. Wen et al. / Applied Clay Science 108 (2015) 165172

Chen, T.Y., Feng, X.T., Zhang, X.W., Cao, W.D., Fu, C.J., 2014. Experimental study on mechanical and anisotropic properties of black shale. Chi. J. Rock Mech. Eng. 33,
17721779.
Chenevert, M.E., 1970. Shale control with balanced-activity oil-continuous muds. J. Pet.
Tech. 22, 13091316.
Ehlmann, B.L., Mustard, J.F., Murchie, S.L., Bibring, J., Meunier, A., Fraeman, A.A., Langevin,
Y., 2011. Subsurface water and clay mineral formation during the early history of
Mars. Nature 479, 5360.
Erguler, Z.A., Ulusay, R., 2009. Water-induced variations in mechanical properties of claybearing rocks. Int. J. Rock Mech. Min. 46, 355370.
Fouche, O., Wright, H., Le Cleac'h, J.M., Pellenard, P., 2004. Fabric control on strain and rupture of heterogeneous shale samples by using a non-conventional mechanical test.
Appl. Clay Sci. 26, 367387.
Gasparik, M., Ghanizadeh, A., Bertier, P., Gensterblum, Y., Bouw, S., Krooss, B.M., 2012.
High pressure methane sorption isotherms of black shales from the Netherlands.
Energ Fuel 26, 49955004.
Hsu, S.C., Nelson, P.P., 2002. Characterization of Eagle Ford Shale. Eng. Geol. 67, 169183.
Jaroniec, M., 1995. Evaluation of the fractal dimension from a single adsorption isotherm.
Langmuir 11, 23162317.
Jo, H.Y., Benson, C.H., Edil, T.B., 2004. Hydraulic conductivity and cation exchange in nonprehydrated and prehydrated bentonite permeated with weak inorganic salt solutions. Clays Clay Miner. 52, 661679.
Josh, M., Esteban, L., Delle, P.C., Sarout, J., Dewhurst, D.N., Clennell, M.B., 2012. Laboratory
characterisation of shale properties. J. Pet. Sci. Eng. 8889, 107124.
Lal, M., 1999. Shale stability: drilling uid interaction and shale strength. Latin American
and Caribbean Petroleum Engineering Conference. Venezuela, Caracas, Apr 2123.
SPE 54356.
Long, P.Y., Zhang, J.C., Tang, X., Nie, H.K., Liu, Z.J., Han, S.B., Zhu, L.L., 2011. Feature of
muddy shale ssure and its effect for shale gas exploration and development. Nat.
Gas Geosci. 22, 525532.
Loucks, R.G., Ruppel, S.C., 2007. Mississippian Barnett Shale: lithofacies and depositional
setting of a deep-water shale-gas succession in the Fort Worth Basin, Texas. AAPG
Bull. 91, 579601.
Ma, T.S., Chen, P., 2014. Study of meso-damage characteristics of shale hydration based on
CT scanning technology. Pet. Explor. Dev. 41, 249256.
Matin, A., Mazumdar, S., 2009. Deformation mechanisms in the frontal Lesser Himalayan
Duplex in Sikkim Himalaya. India J. Earth Syst. Sci. 118, 379390.
Meier, T., Rybacki, E., Reinicke, A., Dresen, G., 2013. Inuence of borehole diameter on the
formation of borehole breakouts in black shale. Int. J. Rock Mech. Min. 62, 7485.
Moore, D.M., Reynolds, R.C. Jr, 1989. X-Ray Diffraction and the Identication and Analysis
of Clay Minerals. Oxford University Press, New York, pp. 146157.

Morin, J., 1993. Soil crusting and sealing. FAO Information Division, Editorial Group, Soil
Tillage in Africa: Needs and Challenges. Food and Agriculture Organization of the
United Nations, Rome (Chapter 5).
Nie, H.K., Zhang, J.C., Li, Y.X., 2011. Accumulation condition of the Lower Cambrian shale
gas in the Sichuan Basin and its periphery. Acta Pet. Sin. 32, 959966.
Novasina, A.G., 1988. Novasina aw center Operation Manual Navasina A.G., Pffkon,
Switzerland.
Prost, P., Koutit, K., Benchara, A., Huard, E., 1998. State and location of water adsorbed on
clay minerals: consequences of the hydration and swellingshrinkage phenomena.
Clays Clay Miner. 46, 117131.
Rathnaweera, T.D., Ranjith, P.G., Perera, M.S.A., 2014. Salinity-dependent strength and
stressstrain characteristics of reservoir rocks in deep saline aquifers: an experimental study. Fuel 122, 111.
Rockland, L.B., Nishi, S.K., 1980. Inuence of water activity on food product quality and
stability. Food Technol. (USA) 34, 4259.
Sahouli, B., Blacher, S., Brouers, F., 1997. Applicability of the fractal FHH equation. Langmuir 13, 43914394.
Shi, B.Z., Xia, B.R., Lin, Y.X., Xu, J., 2012. CT imaging and mechanism analysis of crack development by hydration in hare-brittle shale formations. Acta Pet. Sin. 33, 137142.
Troller, J.A., Christian, J.H.B., 1978. Water ActivityBasic Concepts. Water Activity and
Food. Academic Press, New York, pp. 112.
Vales, F., Minh, D.N., Gharbi, H., Rejeb, A., 2004. Experimental study of the inuence of the
degree of saturation on physical and mechanical properties in Tournemire shale
(France). Appl. Clay Sci. 26, 197207.
Wang, S.B., Song, Z.G., Cao, T.T., Song, X., 2013. The methane sorption capacity of Paleozoic
shales from the Sichuan Basin. China Mar. Pet. Geol. 44, 112129.
Wright, H., 2001. Role de la mine'ralogie, de la texture et de la structure dans la
de'formation et la rupture des argilites de l'Est. ENPC, Paris, p. 185.
Yang, F., Ning, Z.F., Liu, H.Q., 2014. Fractal characteristics of shales from a shale gas reservoir in the Sichuan Basin. China Fuel 115, 378384.
Yao, J., Sun, H., Huang, Z.Q., Zhang, L., Zeng, Q.D., Sui, H.G., Fan, D.Y., 2013. Key mechanical
problems in the development of shale gas reservoirs. Sci. Sin.Phys. Mech. Astron. 43,
15271547.
Zhang, J.G., Al-Bazali, T.M., Chenevert, M.E., Sharma, M.M., 2008a. Factors controlling the
membrane efciency of shales when interacting with water-based and oil-based
muds. SPE Drill. Complet. 6, 150158.
Zhang, J.G., Rojas, J.C., Clark, D.E., 2008b. Stressed-shale drilling strategy-water-activity
design improves drilling performance. SPE Drill. Complet. 12, 385393.
Zhang, J.C., Jiang, S.L., Tang, X., Zhang, P.X., Tang, Y., Jin, T.Y., 2009. Accumulation types and
resources characteristics of shale gas in China. Nat. Gas Ind. 29, 109114.