You are on page 1of 8

Journal of Petroleum Science and Engineering 133 (2015) 810817

Contents lists available at ScienceDirect

Journal of Petroleum Science and Engineering


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

Experimental study on porosity and permeability of anthracite coal


under different stresses
Ya Meng a,b,n, Zhiping Li a,b, Fengpeng Lai a,b
a
Beijing Key Laboratory of Geology Evaluation and Development of Unconventional Natural Gas, China University of Geosciences (Beijing), Beijing 100083,
PR China
b
School of Energy Resources, China University of Geosciences (Beijing), Beijing 100083, PR China

art ic l e i nf o

a b s t r a c t

Article history:
Received 31 December 2014
Accepted 2 April 2015
Available online 20 April 2015

Coal porosity and permeability are key factors inuencing coal-bed methane well production. In order to
investigate the permeability behavior during anthracite coal seam methane production, the porosity and
permeability of anthracite coal sample from No. 3 coal seam in Southern Qinshui Basin of China in net
conning stress were measured in laboratory. The correlations between porosity, permeability and
effective stress were analyzed. Permeability damage rate, stress sensitivity coefcient and pore
compressibility factor were proposed to evaluate the effective stress-dependent sensitivity characteristics of anthracite coal. It turns out that, both porosity and permeability of coal sample decrease
exponentially with the increase of effective stress. If the effective stress is less than 5 MPa or 6 MPa,
stress sensitivity coefcient of coal reservoir changed greatly, and the stress sensitivity coefcient
decreases rapidly with effective stress increased. The permeability damage rate increases rapidly with
increasing effective stress, the stress sensitivity of coal reservoir enhanced; while in the effective stress is
greater than 5 MPa or 6 MPa, the stress sensitivity coefcient of the coal reservoir decreases as effective
stress increases slowly, and there is uctuation, the stress sensitivity of coal reservoir is reduced; while
permeability damage rate with the increase of effective stress increased more slowly. With the increase
of moisture content and temperature, the permeability damage rate of coal reservoir and stress sensitive
coefcient increase, and the stress sensitivity of coal reservoir enhanced.
& 2015 Elsevier B.V. All rights reserved.

Key words:
effective stress
anthracite coal
porosity
permeability
stress sensitivity

1. Introduction
Both porosity and permeability are key factors affecting the coalbed methane production. Porosity of coal usually refers to the sum of
matrix porosity and ssure porosity. The dual pore system of coal
regulates gas storage capacity, occurrence and transport of methane
through coal. Numerous researches on the relationship between
porosity and permeability of conventional oil and gas reservoirs have
been reported (Louis, 1969; Zimmerman and Bodvarsson, 1996;
Zimmerman, 2000). Pore compressibility factor and reservoir stress
sensitivity factor were dened to assess the stress-dependent permeability of conventional reservoirs (Lubinski, 1954; Biot, 1956; Louis,
1969; Wu et al., 1995; Jia et al., 1995; Jose, 1997; Min et al., 2004; Wang
et al., 2009). Jose (1997) pointed out that permeability loss of the tight
sandstone gas reservoir under conning pressure can reach as high as
90% of the initial permeability. As a typical unconventional gas
reservoir, coal is a type of low porosity and low permeability porous

n
Corresponding author at: School of Energy Resources, China University of
Geosciences (Beijing), No. 29 Xueyuan Road, Haidian District, Beijing 100083,
PR China. Tel.: 86 10 13521520515.
E-mail address: mengya629@163.com (Y. Meng).

http://dx.doi.org/10.1016/j.petrol.2015.04.012
0920-4105/& 2015 Elsevier B.V. All rights reserved.

media and is highly sensitive to effective stress. And coal has a high
afnity to gases, i.e. nitrogen, methane, and carbon dioxide. Therefore
permeability behavior of coal is more complex than that of conventional oil and gas reservoir. During gas production, coal permeability
decreases with the increase of effective stress at early water pumping
stage and then increases with the increasing shrinkage effect in later
gas extraction stage. A few permeability models were proposed and
widely used to describe the stress-dependent effect and matrix
shrinkage/expansion effect whereas these models did not take into
account the varying cleat compressibility (Shi and Durucan, 2004,
2010; Pan and Connell (2012); Palmer, 2009; Connell, 2009). Experimental studies showed that overall bituminous coal permeability
declines exponentially with the increase of effective stress (Enever
and Henning, 1997; McKee et al., 1988). Enever and Henning (1997)
found the exponential relationship between the permeability of coal
seam and the stress. Based on the study of coal seam permeability and
its relation to the buried depth in Piceance, San Juan and Black Warrior
basins in the United States, McKee et al. (1988) found that the
permeability reduces in the law of negative exponential function as
the buried depth of coal seam and the effective stress increase and the
aperture of coal seam cleat decreases. In order to investigate the
permeability behavior of high rank coal during early depletion of CBM

Y. Meng et al. / Journal of Petroleum Science and Engineering 133 (2015) 810817

and its inuencing factors, the stress-dependent permeability of


anthracite coal was addressed based on laboratory experiments
(Meng and Li, 2013; Meng and Hou, 2013). Meng and Li (2013)
collected 12 high rank coal samples from the southeast margin of
China Ordos Basin and two low and medium rank coal samples from
the north margin of this basin and determined their air permeability
under a varying effective stress from 2.5 MPa to 20 MPa in laboratory,
the correlation models between permeability and stress of high rank
coal reservoir were established, and the controlling mechanism of
permeability variation was studied. Specically, variation in in-situ
water permeability is likely a function of the maceral composition,
mode of deformation, and degree of shearing of the coal seams
(Gentzis et al., 2007). However, CBM reservoir has low permeability
and strong gas adsorption capacity, which is quite different from the
conventional oil and gas reservoirs. CBM reservoir permeability is
controlled by coal reservoir stress regime, however researches on the
inuence of in-situ stress to the coal reservoir permeability were
insufcient due to the lack of coal reservoir stress and permeability
data. Thus the understanding of the permeability variation during the
exploration and development of CBM was still limited to some degree.
Previous experimental researches were mainly related to low-medium
rank coals, few researches on high rank coals has been reported. This
study was undertaken to address the stress-dependent porosity and
permeability of anthracite coal at a varying temperature and moisture
content. The porosity and permeability of anthracite coal sample from
No. 3 coal seam in Southern Qinshui Basin of China in net conning
stress were measured in laboratory. The correlations between porosity,
permeability and effective stress were analyzed. The results may be
applicable in developing strategies in exploration, well completion and
production of coal-bed methane.

2. Experimental method
2.1. Experiment samples
The Southern Qinshui Basin, located in Shanxi Province of the
Central China, is the most important production base for high
quality anthracite in China. The Southern Qinshui Basin measures
approximately 120 km from north to south and 80 km from east to
west, with an area of about 7000 km2. Coal seams, generated in
Carboniferous and Permian periods, contain abundant methane.
Permeability in the coal-bed reservoir is relatively high compared
to other coal-bed methane reservoirs in China. The exploration
and production tests in this eld have been conducted since 1990s.
The results show that the Qinshui Basin is a very promising coalbed methane reservoir with the most exploration wells, the best
development prospect, and a higher commercialized production in
the China's coal-bed methane reservoirs (Meng et al., 2011).
The coal samples were collected from the No. 3 coal seam in
Permian at Sihe coalmine in Southern Qinshui Basin. The burial depth
of the coal seam is from 350 m to 1200 m. According to approximate
analysis results on air dry basis, the moisture content is 1.36%, volatile
yield 8.12% and ash yield 22.63%. The maximum vitrinite reectance
Ro,Max is 3.12% and No. 3 coal seam is semi-bright coal in macrolithotype and is banded or homogeneous in coal texture. Cylindrical
coal samples were carefully drilled in the direction parallel to the
bedding plane. The diameter of coal samples is from 2.51 cm to
2.52 cm and the length from 4.75 cm to 5.11 cm. Dry coal samples
were prepared by drying the raw coal samples in the dryer for 48 h.
The basic data of the coal samples are shown in Table 1.
2.2. Experiment apparatus and procedure
We used an automatic porosity and permeability instrument
(AP 608) to test the porosity and permeability of coal under net

811

Table 1
Basic data of the coal samples.
ID

Diameter Length Moisture


content
(cm)
(cm)

Experimental
temperature
( 1C)

Description

1#
2#
3#
4#
5#
6#
7#
8#
9#
10#
11#

2.51
2.52
2.52
2.51
2.51
2.52
2.50
2.47
2.50
2.50
2.50

20
20
20
20
20
20
20
20
20
30
60

Fractured
Intact
Intact
Intact
Intact
Intact
Intact
Intact
Intact
Intact
Intact

4.75
5.09
5.11
4.99
5.08
4.96
2.72
3.44
2.60
5.00
5.00

Dry
Dry
Dry
Dry
Dry
Dry
30.42%
66.8%
100%
Dry
Dry

conning stress; gas source for conning pressure is supplied by


high-pressure air, the high-purity helium acts as the test gas
source. The experimental workow is shown in Fig. 1.
2.3. Experiment conditions
To understand the inuence of the coal reservoir stress on coal
porosity and permeability, we used the variation of net conning
pressure to simulate the variations of coal seam effective stress,
and measured the varying porosity and permeability with net
conning pressure, and then analyzed the relationships between
porosity, permeability and effective stress.
In this paper, the burial depth of coal seams in southern
Qinshui Basin ranges from 350 m to 1200 m, so the maximum
stress test was designed to 10 MPa, and the temperature test up to
60 1C. To avoid the inuence of slippage effect on permeability of
coal sample, in the course of the experiment, the displacement
pressure is xed. Conning pressure values were 3.5, 4.0, 5.0, 6.0,
7.0, 8.0, 9.0, and 10.0 MPa, and the temperature 20 1C. In order to
analyze the inuence of temperature on stress sensitivity, experimental temperature is set as 30 1C and 60 1C. To minimize gas
slippage effect, the same displacement pressure keeps constant
during the test process. Each stress point sustained long enough, at
least in balance for 30 min, and then gas permeability at this stress
point is measured.

3. Experiment results
3.1. Relationships between coal porosity, permeability and effective
stress
3.1.1. The relationship between permeability of coal and the effective
stress
The relationship between permeability of coal and the effective
stress is shown in Fig. 2. From Fig. 2, we can see that the relationship
between the gas permeability of coal and effective stress obeys the
negative exponential function as follows:
K i K 0 e  ap

where K i is the permeability under a specic effective stress,


10  3 m2; p is the variation value of effective stress from the initial
to a certain stress, MPa; K 0 is the permeability under initial effective
stress, 10  3 m2; and a is the regression coefcient, MPa  1.
Regression analysis results for six experimental samples are
shown in Table 2. K 0 ranges from 0.0267  10  3 m2 to
4.2562  10  3 m2, averagely 0.7819  10  3 m2; a ranges from
0.26 MPa  1 to 0.54 MPa  1, averagely 0.37 MPa  1.

812

Y. Meng et al. / Journal of Petroleum Science and Engineering 133 (2015) 810817

Fig. 1. Experimental workow (Meng and Li, 2013).

1.2

Porosity

y = 3.4933e

3.5
3.0

-0.0491x

R = 0.9711

0.8

2.5

0.003

y = 0.0267e

10

12

0.002

0.000

1.5
14

2.0
-0.0957x

y = 2.8397e

1.5

0.040

2.5

-0.3628x

2.0

Permeability/10-3 m2

3.0

Porosity

R = 0.9819

Porosity

3.0
0.006
2

R = 0.9144

y = 0.039e

10

12

0.000

1.5

14

y = 6.4715e

0.030

-0.0382x

5.0

R = 0.9496
y = 0.1121e

Porosity/%

Permeability/10-3 m2

5.5

0.040

0.020

-0.2615x

4.5

12

Permeability

-0.0548x

y = 4.8654e

0.030

R = 0.9871

4.0

0.020

3.5
3.0
-0.2919x

0.010

y = 0.067e
2

10

12

0.000

4.0

4.5

Porosity

2.5

R = 0.9877

1.0
14

5.0

R = 0.9811

10

0.040

Permeability/10-3 m2

Permeability
Porosity

0.050

Effective stress/MPa
6.0

1.5

0.060

0.000

2.0

-0.4306x

0.002

Effective stress/MPa

0.010

2.5

-0.1325x

y = 5.8803e

0.004

R = 0.9395

3.5

Permeability

R = 0.9797

1.0
14

12

4.0

0.008

-0.0483x

Porosity/%

Permeability/10-3 m2

Permeability
y = 3.4842e

10

0.010

3.5

0.060

Effective stress/MPa

0.080

0.000

2.5

R = 0.5481

y = 0.1901e

-0.5387x

R = 0.9868

Effective stress/MPa

0.020

3.0

Porosity

Porosity/%

0.0

0.004

0.001

2.0

R = 0.9902

3.5
Permeability

Porosity/%

y = 4.2562e

0.4

-0.3086x

Permeability/10-3 m2

Permeability

1.6

0.005

Porosity/%

4.0

Porosity/%

Permeability/10-3 m2

2.0

10

12

2.0

14

Effective stress/MPa

Effective stress/MPa

Fig. 2. Relationship between porosity and permeability of coal and the effective stress.

3.1.2. The relationship between porosity of coal and the effective


stress
The relationship between porosity of coal and effective stress is
shown in Fig. 2. The relationship between the porosity of coal and
effective stress obeys the negative exponential function as follows:

i 0 e  cp

where i is the porosity under a specic effective stress, %; p is


the variation value of effective stress from the initial to a certain
stress, MPa; 0 is the porosity under initial effective stress, %; and
c is the regression coefcient, MPa  1.
Regression analysis results for six experimental samples are
shown in Table 2. 0 ranges from 2.84% to 6.47%, averagely 4.51%; c
ranges from 0.038 MPa  1 to 0.133 MPa  1, averagely 0.07 MPa  1.

Y. Meng et al. / Journal of Petroleum Science and Engineering 133 (2015) 810817

813

Table 2
Relationship between porosity and permeability of coal and the effective stress.
Porosity

1#
2#
3#
4#
5#
6#
Average
Max. value
Minimum value

Permeability

Regression coefcient
c

Correlation coefcient
R2

Regression coefcient
a

K0

Correlation coefcient
R2

0.0491
0.0957
0.0483
0.1325
0.0382
0.0548
0.0698
0.1325
0.0382

3.4933
2.8397
3.4842
5.8803
6.4715
4.8654
4.5057
6.4715
2.8397

0.9711
0.5481
0.9819
0.9144
0.9496
0.9871
0.8920
0.9871
0.5481

0.3086
0.5387
0.3628
0.4306
0.2615
0.2919
0.3657
0.5387
0.2615

4.2562
0.0267
0.1901
0.0390
0.1121
0.0670
0.7819
4.2562
0.0267

0.9902
0.9868
0.9797
0.9395
0.9811
0.9877
0.9775
0.9902
0.9395

3.1.3. The relationship between coal porosity and permeability under


net conning stress
In order to more intuitively describe the inuence of coal
reservoir stress to the gas permeability, the dimensionless permeability K i =K 0 is dened as the ratio of gas permeability K i to initial
permeability of coal K 0 . The relationship between coal porosity
and the dimensionless permeability K i =K 0 under net conning
stress is shown in Fig. 3. The experiment results show an
exponential function relationship between the dimensionless
permeability K i =K 0 and porosity, that is, with the increase of
porosity, the dimensionless permeability K i =K 0 increases, which
obeys the law of exponential function. The regression analysis of
experimental results are dened as follows:
K i =K 0 mebi

r 2
8 2

0.8
1#
2#
3#
4#
5#
6#

0.6
0.4
0.2
0.0
0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

porosity/%
Fig. 3. Relationship between the porosity of coal and its permeability under the
effective stress.

where i is the porosity under net conning stress, %; K i is the


permeability under net conning stress,  10  3 m2; and m and b
are regression coefcients.
The experimental results for six samples under net conning
stress indicate that regression coefcient m ranges from 0.0002 to
0.1736, averagely 0.0299, b ranges from 0.6110 to 2.9236, averagely
1.7587, as shown in Table 3.
Besides porosity, the permeability of coal reservoir is closely
related to the pore structure. Based on capillary bundle model,
Kozeny (1927)and Carman (1937) gave a theoretical equation
governing the relationship between permeability (K) of the porous
medium and the porosity and average pore radius as follows:
K

1.0
Dimensionless permeability
(K i/K0)

ID

where r is the average pore radius of rock, m2 and is the


average pore tortuosity, dimensionless.
Tortuosity is the ratio of the actual length of pores, that is, uid
ow path, to exterior length of coal core. Eq. (4) shows that two
factors inuence the permeability of coal sample, including
porosity and average pore size. The coal is characterized by
complex pore distribution and large pore tortuosity which are
difcult to be measured and evaluated. The larger the pore
diameter within a certain range, the higher the porosity, the
higher permeability is; the smaller the pore diameter, the lower
the porosity of coal samples, the lower permeability is. Inuence
of pore diameter on the permeability is a power levels higher than
porosity. The coal reservoir with relatively large pore diameter but
low porosity may have relatively high permeability; however the
coal reservoir with small pore diameter but high porosity usually
has low permeability.
Coal reservoir is a dual pore system media, containing matrix
pore system and ssure pore system. The permeability of coal
reservoir mainly depends on coal ssures. Coal is usually fractured

Table 3
Relationship between coal porosity and permeability.
ID

Coefcient
m

Coefcient
b

Correlation coefcient
R2

1#
2#
3#
4#
5#
6#
Average
Max. value
Minimum value

0.0008
0.1736
0.0002
0.0027
0.0005
0.0018
0.0299
0.1736
0.0002

2.4090
0.6110
2.9236
1.7568
1.2988
1.5529
1.7587
2.9236
0.6110

0.9842
0.6620
0.9859
0.9875
0.9844
0.9737
0.9296
0.9875
0.6620

into many matrix blocks. Coal usually has densely developed


ssures which are very important for coal reservoir. These ssures
not only provide a reservoir space, but also connect the matrix
pore by interleave network system and greatly improve the
permeability of coal reservoirs. The theoretical relationship
between the permeability in the direction parallel to fracture
and fracture porosity, aperture is as follows:
Kf

f b2
12

where b is the average fracture aperture of coal reservoir, m; f is


the average fracture porosity of coal reservoir, decimal; and K f is
the fracture permeability of coal reservoir, m2.
As can be seen from Eq. (5), the larger the fracture aperture and
fracture porosity is, the higher the coal reservoir permeability is.
But fracture aperture has a power level higher inuence to the
permeability than fracture porosity.

Y. Meng et al. / Journal of Petroleum Science and Engineering 133 (2015) 810817

3.2. Relationship between permeability damage rate, stress


sensitivity coefcient, pore compressibility factor and effective stress

K 1  K 0min
Dk2
 100%
K1

where Dk2 is the permeability damage rate after the stress


increases to the highest stress point; K 1 is the permeability under
the rst stress point, 10  3 m2; and K min
'
is the permeability
under the highest stress point, 10  3 m2.
(2) Stress sensitivity coefcient is dened as follows:

K 

1 K
U
K 0 p

where K0 is the permeability under initial effective stress, 10  3 m2;


K is the variation of permeability, 10  3 m2; and p is the variation
of pressure, MPa.
The permeability damage rate has the same meaning as the
permeability stress sensitivity coefcient adopted in China's oil
and gas industry standards.
From Eq. (7), we can see that, the bigger the k is, the higher
the sensitivity of coal to effective stress will be. On the contrary,
the smaller the k is, the lesser the sensitivity of coal to effective
stress will be.
3.2.2. Pore compressibility factor
In order to quantitatively evaluate pore compression deformation of coal reservoir, we dened a pore compressibility factor as
follows:
 
1
8
Cp
0 p T
where Cp is the pore compressibility factor of coal reservoir,
MPa  1; 0 is the porosity under initial effective stress, %; and

=p T is the variation rate of coal porosity over effective stress
under isothermal conditions, MPa  1.
3.2.3. The relationship between permeability damage rate, stress
sensitivity coefcient and effective stress
Fig. 4 shows the curves of the permeability damage rate, stress
sensitivity coefcient and effective stress for six experimental coal
samples. As can be seen from Fig. 4, the permeability damage rate
and stress sensitivity coefcient increases with the variation of
effective stress, that is, with the increase of effective stress, the
coal reservoir stress sensitivity coefcient decreases, but there are
some uctuations; the permeability damage rate increases with
the increase of effective stress (Fig. 4).
Coal reservoir stress sensitivity coefcient varies signicantly
under effective stress of below 5 MPa or 6 MPa; with the increase
of effective stress, the coal reservoir stress sensitivity coefcient
declines sharply, permeability damage rate goes up rapidly. With
the increase of effective stress over 5 MPa or 6 MPa, the coal
reservoir stress sensitivity coefcient declines gently with some
uctuations, the permeability damage rate increases more slowly.
It shows that when the effective stress is below 5 MPa or 6 MPa
in No. 3 coal seam within Shanxi Formation Permian System in the

Permeability damage rate/%

3.2.1. Permeability damage rate and stress sensitivity coefcient


The stress sensitivity is higher for coal reservoir than for
sandstone. There are no industry standards about CBM reservoir
stress sensitivity test. Our experiment referenced the oil and gas
industry standards of China (SY/T5336, 5358, 6385) and adopted
increasing net conning pressure of coal samples to simulate the
effective stress variation of coal seam, and then determined the
permeability behavior of coal sample with change of net conning
pressure, and nally the reservoir stress sensitivity was analyzed.
(1) Permeability damage rate is calculated by Eq. (6).

0.30

100
1#PDR
2#PDR
3#PDR
4#PDR
5#PDR
6#PDR
1#SSC
2#SSC
3#SSC
4#SSC
5#SSC
6#SSC

80

60

40

0.25
0.20
0.15
0.10

20

0.05

stress sensitivity coefficient /MPa -1

814

0.00
0

10

12

Effective stress /MPa


PDR-Permeability damage rate ;SSC-stress sensitivity coefficient

Fig. 4. Relationship between permeability damage rates, stress sensitivity coefcient and the effective stress.

study area, coal reservoir permeability decreases rapidly with the


increase of effective stress, coal is highly sensitive to effective
stress; when the effective stress is over 5 MPa or 6 MPa, permeability decreases gently with the increase of effective stress, that is,
stress sensitivity is relatively weakened, and has some uctuation
due to the closure of micro-ssures.
Permeability damage rate under the effective stress of 10 MPa
ranges from 82.7% to 97.2%, average 89.3%; the maximum of the stress
sensitivity coefcient ranges from 0.018 MPa  1 to 0.250 MPa  1, the
minimum from 0.003 MPa  1 to 0.075 MPa  1, the average from
0.010 MPa  1 to 0.157 MPa  1, as shown in Table 4.
3.3. Relationship between the pore compressibility factor and the
effective stress
Fig. 5 shows the curves of the pore compressibility factor and
effective stress for six coal samples. As can be seen from Fig. 5,
pore compressibility factor varies with the increase of effective
stress. When effective stress is below 5 MPa or 6 MPa, coal
reservoir stress pore compressibility factor is relatively large and
decreases rapidly with the increase of the effective stress; when
the effective stress is over 5 MPa or 6 MPa, with the increase of
effective stress, coal reservoir pore compressibility factor has a
gently decreasing trend with some obvious uctuations.
The maximum of pore compressibility factor for six samples
ranges from 0.020 MPa  1 to 0.251 MPa  1, and the minimum from
0.003 MPa  1 to 0.070 MPa  1, the average from 0.011 MPa  1 to
0.115 MPa  1, as shown in Table 4. The pore compressibility factor
and stress sensitivity coefcient have almost the same variation
trend and law.
3.4. Moisture
CBM reservoir always has high water saturation in the development process, in order to investigate the inuence of water
saturation on permeability, we chose three coal samples of
different water saturations and carried out experimental analysis.
The water saturation of these three samples (7#, 8#, 9#) were
30.42%, 66.8% and 100% respectively.
The relationships between the dimensionless permeability and
effective stress for three samples are shown in Fig. 6. Under the
same water saturation, with the increase of effective stress, coal
permeability decreases in a negative exponential law. With the
increase of water saturation, stress sensitivity increases, permeability for coal sample of high water saturation drops more quickly
with the increase of the effective pressure, that is, the permeability
damage rate is bigger (Fig. 6). The water saturation of 7#, 8#, 9#
samples was 30.42%, 66.8% and 100%, respectively. Under the

Y. Meng et al. / Journal of Petroleum Science and Engineering 133 (2015) 810817

815

Table 4
Evaluation parameters of stress sensitivity for coal reservoir.
Permeability damage rate
(%)

1#
2#
3#
4#
5#
6#
Average
Max. value
Minimum value

Stress sensitivity coefcient


(MPa  1)

86.8
97.2
90.5
93.4
82.7
85.2
89.3
97.2
82.7

Average

Maximum

Minimum

Average

Maximum

Minimum

0.058
0.026
0.054
0.035
0.073
0.062
0.051
0.073
0.026

0.178
0.075
0.158
0.103
0.250
0.179
0.157
0.250
0.075

0.012
0.003
0.005
0.005
0.018
0.015
0.010
0.018
0.003

0.039
0.064
0.039
0.054
0.038
0.039
0.046
0.064
0.038

0.084
0.251
0.090
0.090
0.104
0.070
0.115
0.251
0.070

0.007
0.003
0.013
0.019
0.006
0.020
0.011
0.020
0.003

Pore compressibility factor/MPa-1

0.30
0.25
0.20

1#

2#

3#

4#

5#

6#

Table 5
Evaluation parameters of stress sensitivity for coal reservoir under different water
saturations.
ID

0.15
0.10

10

Permeability damage
rate

Stress sensitivity coefcient


(MPa  1)

(%)

(%)

Average Max.
value

Minimum
value

96.8
95.9
98.6

0.075
0.085
0.111

0.008
0.011
0.005

1
Sw=30.42%
Sw=66.80%

Permeability damage rate/%

100

Fig.5. Relationship between pore compressibility factor and effective stress.

0.8

0.279
0.250
0.330

12

Effective stress/MPa

Dimensionless permeability
(Ki/K0)

Water
saturation

7# 30.42
8# 66.8
9# 100

0.05
0.00

Pore compressibility factor


(MPa  1)

0.40

80
60

7#PDR

8#PDR

9#PDR

7#SSC

8#SSC

9#SSC

0.30

0.20

40
0.10

20

Sw=100%

0.6

0.00

0.4

10

12

Effective stress/MPa
PDR-Permeability damage rate;SSC-stress sensitivity coefficient

0.2
0

Stress sensitivity coefficient/MPa-1

ID

Fig. 7. Relationship between permeability damage rate, stress sensitivity coefcient and the effective stress.

10

Effective stress/MPa
Fig. 6. Relationship between dimensionless permeability and effective stress under
different water saturations.

effective stress of 10 MPa, permeability damage rate was 96.8%,


95.9% and 98.6% respectively; the average of stress sensitivity
coefcient is 0.075 MPa  1, 0.085 MPa  1 and 0.111 MPa  1 respectively (Table 5).
Compared with dry sample, the stress sensitivity for three coal
samples of different water saturations are more obvious (Fig. 7 and
Table 5).
Under the effective stress for 10 MPa, permeability damage rate of
three samples with moisture ranges from 95.9% to 98.6%. The
maximum of stress sensitivity coefcient ranges from 0.25 MPa  1
to 0.33 MPa  1, the minimum ranges from 0.005 MPa  1 to
0.011 MPa  1, the average from 0.075 MPa  1 to 0.111 MPa  1. Comparing Tables 4 and 5, we can see that stress sensitivity coefcient
and permeability damage rate are higher for these three coal samples
of different water saturations than for dry ones.

The average of stress sensitivity coefcient for coal samples


with moisture is 1.77 times as that for dry ones. Water saturation
has a relatively high impact on coal reservoir stress sensitivity,
because water softens the coals and lowers its mechanical
strength. On the other hand, the gas/water pressure will offset
part of the total stress of the coal body and decrease the elastic
limit and shear strength, then the coal is easy to generate elastic
and plastic deformation and thereby the coal samples of different
water saturations are more sensitive to effective stress than dry
ones and the increase of effective stress will cause more signicant
permeability loss (Meng and Li, 2013).

3.5. Temperature
In order to investigate the inuence of temperature on coal
reservoir permeability, we chose two coal samples (10# and 11#)
and carried out stress sensitivity experiments at 30 1C and 60 1C
respectively.

816

Y. Meng et al. / Journal of Petroleum Science and Engineering 133 (2015) 810817

Dimensionless permeability
(Ki/K0)

Table 6
Evaluation parameters of stress sensitivity for coal reservoir under different
temperatures.

0.8

T=30

ID

T=60

0.6
0.4
0.2
0

Stress sensitive coefcient


(MPa  1)

(1C)

(%)

Average Max.
value

Minimum
value

81.3
96.0

0.067
0.070

0.015
0.010

10# 30
11# 60
0

10

Fig. 8. Relationship between dimensionless permeability and effective stress under


different temperatures.

80

0.30
10#PDR
11#PDR

0.20

10#SSC

40

11#SSC

0.10

20

0.00

0
0

10

Stress sensitivity coefficient/MPa -1

0.40

100

60

0.222
0.333

12

Effective stress/MPa

Permeability damage rate/%

Temperature Permeability damage


rate

12

Effective stress/MPa
PDR-Permeability damage rate;SSC-stress sensitivity coefficient
Fig. 9. Relationship between the permeability damage rate/stress sensitivity
coefcient and effective stress.

The relationship between dimensionless permeability and effective stress for the two coal samples under different temperatures are
shown in Fig. 8. At the same temperature, with the increase of
effective stress, permeability of coal sample reduces in the negative
exponential function law, and with the rise of temperature, stress
sensitivity increases. With the increase of the effective stress,
permeability drops more quickly for 11# coal sample of higher
temperature than 10# of lower. The higher the temperature is, more
obvious the stress sensitivity is, that is, the permeability damage rate
is bigger (Figs. 8 and 9). Under the effective stress of 10 MPa,
permeability damage rate is 81.3% for 10# and 96.0% for 11#. Average
stress sensitive coefcient is 0.067 MPa  1 for 10# and 0.070 MPa  1
for 11# (Table 6).
Coal reservoir permeability decreases with the increase of temperature, showing a temperature sensitivity. In general, the effect of
temperature rise on coal reservoir permeability can be divided into
two stages. At the low temperature stage, thermal expansion with
the rise of temperature causes a slight decline of permeability; at the
high temperature stage, coal cracking induced by thermal stress
above critical temperature will sharply increase coal reservoir
permeability. The experiments in this study aim to simulate the
permeability behavior of coal seam with a burial depth of less than
1000 m, so experimental temperatures are set as 30 1C and 60 1C
respectively, thermal cracking in this temperature range will not
occur. Therefore, in our experimental process, with the rise of
temperature, fracture aperture decreases due to thermal expansion
and thereby coal sample permeability decreases.

4. Discussion
Coal is a dual porosity medium, which contains matrix pore
system and fracture system. The fracture system contributes in a

major way to the permeability of the coal seam. So the permeability depends strongly on the aperture of the fractures in coal.
Increasing normal stress will narrow the fractures and then lower
the permeability of the coal.
Experimental results show that coal reservoir permeability
varies exponentially with the increase of effective stress, that is
K i K 0 e  ap . In the drainage process of CBM wells, the water
pumping lowers the coal reservoir pressure, which leads to the
increase of effective stress and induces the decline of permeability.
This process is just the dynamic changing of the coal reservoir
permeability. Therefore, the relationship between the coal reservoir permeability and effective stress (k  e ) reects the relationship between the coal reservoir permeability and producing
pressure drop in the drainage process (k  P).
The No. 3 coal seam in Shanxi Formation Permian System in the
southern Qinshui basin is up to 1200 m deep, its reservoir pressure
is below 10 MPa and the reservoir pressure gradient averages
0.71 MPa/100 m. According to the CBM production experience
from US, reservoir pressure can drop to abandonment pressure
of about 0.7 MPa. That is, coal reservoir pressure varies from
10 MPa to 0.7 MPa, which corresponds to the effective stress
variation in the actual CBM well drainage.
From permeabilitystress curves (Figs. 2, 6 and 8), with the
increase of effective stress, coal reservoir permeability reduces,
which obeys the law of negative exponential function.
When the effective stress is below 5 MPa or 6 MPa, the corresponding drop of gas well uid level is from 0 m to 500 m or 600 m,
with the increase of effective stress, the coal reservoir permeability
drops rapidly, showing a high stress sensitivity. When the effective
stress is over 5 MPa or 6 MPa, the corresponding drop of gas well
uid level is above 500 m or 600 m, with the increase of effective
stress, the coal reservoir permeability decreases much more slowly,
showing a weak stress sensitivity.
Coal seam is in a certain in situ stress environment and the
stress of coal seam varies spatially. Under low in-situ stress regime
in shallow coal seam, with the increase of effective stress, plastic
deformation of coal reservoir due to compaction under stress
results in a signicant reduction in permeability; under high insitu stress regime in the deep, coal reservoir stress sensitivity will
abate due to very small initial aperture and minor compaction.
In the early drainage of a CBM well, coal seam is usually water
saturated, water pressure drops due to pumping and correspondingly
the effective stress increases and coal permeability declines quickly,
thereby the drainage in this stage should be sustained, stable and
slow. When the water pressure drops to the critical desorption
pressure (Pc), water saturation of coal reservoir decreases as the
increasing gas desorption, and gas saturation increases, the stress
sensitivity of coal seam diminishes, gas well uid level in this stage
should keep stable. In the middle and later stage of drainage,
continuous gas desorption will lead to signicant coal matrix
shrinkage and an increasingly rise in permeability and thereby gas
production goes up. In short, the change of permeability by above
two aspects will impact on the gas production. On the one hand, the
permeability variation induced by in-situ stress is a static effect; on

Y. Meng et al. / Journal of Petroleum Science and Engineering 133 (2015) 810817

817

are compressed and closed, signicant elasticplastic deformation occurs in coal, so the coal reservoir permeability
decreases obviously.
(5) The variation of permeability induced by in-situ stress is
greater than that by dynamic changes of the reservoir pressure
in the gas production process.

Acknowledgment

Fig. 10. Relationship between permeability and effective stress, reservoir pressure
(Luo, 2006).

This study was nancially supported by the National Science


and Technology Major Project in 12th Five-Year Plan Period under
Grant no. 2011ZX05038-2-1, and Shanxi Provincial Basic Research
ProgramCoal Bed Methane Joint Research Foundation under
Grant no. 2013012002. The authors also thank the reviewers and
editors for their constructive comments and suggestions on
improving this manuscript.
References

the other hand, the permeability variation induced by reservoir


pressure change during production process is a dynamic effect.
Fig. 10 shows the relationship between the permeability and effective
stress, and reservoir pressure. The X-axis denotes permeability and
the Y-axis effective stress. The heavy solid line represents the
correlation curve of permeability and effective stress, the dotted line
variation path of reservoir pressure during gas production.
In Fig. 10, K0 is the ground permeability at effective stress of
eff 0 , Ke is the reservoir permeability at reservoir pressure of Pe,
and Kwf is the permeability at reservoir pressure of Pwf, where Pe is
the initial reservoir pressure and Pwf borehole pressure.
As can be seen from Fig. 10, the variation range of permeability
is much wider for K0-Ke than for Ke-Kwf, which reects that the
impact of in-situ stress on permeability is greater than that of
dynamic changes of the reservoir pressure in the gas production
process (Luo, 2006).
5. Conclusions
(1) With the increase of effective stress, coal reservoir porosity
and permeability decreases in the law of negative exponent
function. Under the compressive stress, the increase of stress
results in normal compaction deformation, the porosity of coal
pore and fracture decreases and permeability drops sharply.
(2) When the effective stress is below 5 MPa or 6 MPa, stress
sensitivity coefcient and pore compressibility factor decrease
sharply with the increase of the effective stress, the permeability
damage rate increases markedly, that is, the coal has a high
sensitivity to effective stress; when the effective stress is over
5 MPa or 6 MPa, with the increase of effective stress, stress
sensitivity coefcient and pore compressibility factor uctuate
and decrease gently, the permeability damage rate varies slowly.
(3) The water saturation and temperature of coal reservoir have a
marked impact on production of CBM wells. Increasing the
moisture content and temperature will enhance the stress
sensitivity of coal reservoir. The permeability of water-bearing
coal sample drops quickly with the increase of the effective
pressure, stress sensitivity is more obvious, that is, the permeability damage degree caused by stress is relatively big. With
the rise of the temperature, the permeability damage rate and
the stress sensitive coefcient go up.
(4) In the process of CBM production, with the extraction of water
and gas from coal reservoir, the coal reservoir pressure
decreases gradually, resulting in the increase of effective stress
in coal reservoir, and the pores and fractures of coal reservoir

Biot, M.A., 1956. Theory of deformation of a porous viscoelastic anisotropic solid.


J. Appl. Phys. 27, 457467.
Carman, P.C., 1937. Fluid ow through a granular bed. Trans. Inst. Chem. Eng. 15,
150167.
Connell, L.D., 2009. Coupled ow and geomechanical processes during gas
production from coal seams. Int. J. Coal Geol. 77, 222233.
Enever, J.R.E., Henning, A., 1997. The relationship between permeability and
effective stress for Australian coal and its implications with respect to coalbed
methane exploration and reservoir modeling. In: Proceedings of the 1997
International Coalbed Methane Symposium, pp. 1322.
Gentzis, T., Deisman, N., Chalaturnyk, R.J., 2007. Geomechanical properties and
permeability of coals from the Foothills and Mountain regions of western
Canada. Int. J. Coal Geol. 69, 153164.
Jia, W.R., Li, F.K., Xiao, J.X., 1995. A study on some issues of development disposition
of a low permeability oileld. Pet. Explor. Dev. 22 (4), 4751.
Jose, G., 1997. Numerical simulation of coupled uid-ow/geomechnical behavior
of tight gas reservoirs with stress sensitive permeability. SPE39055, 115.
Kozeny, J., 1927. Uber Kapillare Leitung der Wasser in Boden. Sitzungsber. Akad.
Wiss. Wien 136, 271306.
Louis, C., 1969. A study of groundwater ow in jointed rock and its inuence of the
stability of rock masses, Rock Mechanics Research Report 10. Imperial College,
London.
Lubinski, A., 1954. Theory of elasticity for porous bodies displaying a strong pore
structure. In: Proceedings of the 2nd U.S. National Congress of Applied
Mechanics, pp. 247256.
Luo, R.L., 2006. A Study of Deformation and Percolation Mechanisms of Deep Gas
Reservoir and its Application. China University of Petroleum, Beijing.
McKee, C.R., Bumb, A.C., Koenig, R.A., 1988. Stress-dependent permeability and
porosity of coal and other geologic formations. SPE Form. Eval. 3, 8191.
Meng, Z.P., Zhang, J.C., Wang, R., 2011. In-situ stress, pore pressure, and stressdependent permeability in the Southern Qinshui Basin. Int. J. Rock Mech. Min.
Sci. 48, 122131.
Meng, Z.P., Li, G.Q., 2013. Experimental research on the permeability of high-rank
coal under a varying stress and its inuencing factors. Eng. Geol. 162, 108117.
Meng, Z.P., Hou, Q.L., 2013. Coupling model of stress-dependent permeability in
high-rank coal reservoir and its control mechanism. Chin. J. Geophys. 56 (2),
667675.
Min, K.B., Rutqvist, J., Tsang, C.F., 2004. Stress-dependent permeability of fractured
rock masses: a numerical study. Int. J. Rock Mech. Min. Sci. 41 (7), 11911210.
Palmer, I., 2009. Permeability changes in coal: analytical modeling. Int. J. Coal Geol.
77, 119126.
Pan, Z.J., Connell, L.D., 2012. Modeling permeability for coal reservoirs: a review of
analytical models and testing data. Int. J. Coal Geol. 92, 144.
Shi, J.Q., Durucan, S., 2004. Drawdown induced changes in permeability of
coalbeds: a new interpretation of the reservoir response to primary recovery.
Transp. Porous Media 56, 116.
Shi, J.Q., Durucan, S., 2010. Exponential growth in San Juan Basin Fruitland coalbed
permeability with reservoir drawdown: model match and new insights. SPE
Reserv. Eval. Eng. 6, 914925 SPE-123206-PA.
Wang, L.Q., Liu, H.Q., Zhen, S.G., et al., 2009. Quantitative research on stress
sensitivity of low permeability reservoir. Acta Pet. Sin. 30 (1), 96103.
Wu, X.Y., Chen, Z.A., Sun, D.M., et al., 1995. An experimental study of changes of
porosity of sandstones with pressure. Acta Geophys. Sin. 38 (Suppl. 1),
S275S280.
Zimmerman, R.W., 2000. Coupling in poroelasticity and thermoelasticity. Int. J.
Rock Mech. Min. Sci. 37, 7987.
Zimmerman, R.W., Bodvarsson, G.S., 1996. Hydraulic conductivity of rock fractures.
Transp. Porous Media 23, 130.