Fuel 211 (2018) 804–815

Contents lists available at ScienceDirect

Fuel
journal homepage: www.elsevier.com/locate/fuel

Material composition, pore structure and adsorption capacity of low-rank MARK

coals around the first coalification jump: A case of eastern Junggar Basin,
China
⁎
Shu Taoa,b, Shida Chena,b, , Dazhen Tanga,b, Xu Zhaoc,d, Hao Xua,b, Song Lia,b
a
School of Energy Resources, China University of Geosciences (Beijing), Beijing 100083, PR China
b
Coal Reservoir Laboratory of National Engineering Research Center of Coalbed Methane Development & Utilization, Beijing 100083, PR China
c
Exploration & Production Research Institute SINOPEC, Beijing 100083, China
d
Shandong Provincial Key Laboratory of Depositional Mineralization & Sedimentary Mineral, Shandong University of Science and Technology, Qingdao, Shandong
266590, China

A R T I C L E I N F O A B S T R A C T

Keywords: The first coalification jump (FCJ) has a significant impact on low-rank coal reservoir heterogeneity, and is of
Coalification jump great importance for coalbed methane (CBM) development. Here, a series of experiments were performed for 10
Material composition coal samples collected from eastern Junggar Basin, to compare the material composition, pore structure and
Pore structure adsorption capacity of lignite and candle coal. Contrast with the candle coal, the lignite has a higher inertinite
Adsorption capacity
content, larger pore volume, better connectivity, and greater specific surface area (SSA). During the process of
Low-rank coal
Eastern Junggar Basin
FCJ, the polycondensation of coal molecules and the compaction of coal matrix occur, leading to a rapid decline
of moisture, porosity and permeability, and the cell wall in the candle coal is badly crushed with clay minerals
filled from optical microscopy. In general, the larger total pore volume (1.7–300 nm, measured by N2 adsorp-
tion) contributes to the larger SSA. The SSA of candle coal mainly comes from the contribution of micropore
(< 10 nm), especially the 2–3 nm pores, while the micropore and transition pore (10–100 nm) contribute to most
of SSA of lignite. However, though the SSA of the candle coal is largely lower than that of the lignite, the CH4
adsorption capacity tends to decrease from the lignite to the candle coal due to material composition difference.
Low-field NMR was used to determine the pore and fracture system by analyzing the transverse relaxation time,
which showed that only two obvious peaks could be identified in lignite and three peaks at about 0.25 ms, 30 ms
and 200 ms are present in the candle coal. The fractal results indicate that the pore surface and complexity inside
the coal increase gradually from lignite to candle coal. These observations could deepen awareness and un-
derstanding of low-rank coal reservoir heterogeneity and the influence of FCJ on reservoir property.

1. Introduction
The development of coalification is not a straight line but a few
jumps because of the physicochemical changes of coal components [1].
This process of evolution is called the coalification jump. Being the key
geological event in the coal metamorphism of coal, the coalification
jump greatly affects the material composition and molecular structure
of coals. Moreover, it controls the physical properties of coal reservoirs,
such as the pore volume, surface area, CH4 adsorption ability, etc. [2,3].
The first coalification jump (FCJ) is characterized by the occurrence of
bituminization. As the coal rank increases, enriched-oxygen functional
groups gradually fall off. When the maximum vitrinite reflectance
(Ro,m) is less than 0.5%, the products are mainly CO2 and H2O. How-
ever, when Ro,m reaches 0.5–0.6%, as the aliphatic functional groups
and side chains begin to peel off the aromatic fused rings, the CH4-
dominated volatiles are formed [4,5]. Therefore, the FCJ is regarded as
the critical geological transition from lignite to bitumite, which usually
occurs at Ro,m of 0.5–0.6% [4–7]. The coal shape and its physical,
chemical, geochemical and petrological properties vary greatly before
and after the jump. The earliest exploration on the FCJ could be traced
back to late 20th century, which focused on the changes in material
composition and interior structure in coals caused by the jump [6–9].
Nevertheless, the heterogeneity features of the development of multi-
scale pore-fracture in low-rank coal reservoirs around the FCJ are rarely
studied in literature.
In the current study, ten coal samples collected from eastern
Junggar Basin with Ro in the range of 0.41–0.67% were analyzed using
a variety of techniques to reveal the characteristics of material

⁎
Corresponding author at: School of Energy Resources, China University of Geosciences, No. 29 Xueyuan Road, Haidian District, Beijing 100083, PR China.
E-mail address: 729210567@qq.com (S. Chen).

http://dx.doi.org/10.1016/j.fuel.2017.09.087
Received 22 July 2017; Received in revised form 9 September 2017; Accepted 21 September 2017
Available online 09 October 2017
0016-2361/ © 2017 Elsevier Ltd. All rights reserved.
S. Tao et al.
composition (including macerals, ash, moisture, volatile and major
elements) and pore-fracture properties (including pore size, shape,
structure, type, specific surface area (SSA) and fractal characteristics) of
low-rank coals around the FCJ. Furthermore, the effect of FCJ on the
adsorption capacity of coal was discussed. This study is important to
comprehensively and systematically reveal the material composition
and pore size distribution in low-rank coals from a microscopic per-
spective, and it also has practical significance for CBM exploration and
exploitation from low-rank coal reservoirs.
2. Geological setting and analytical procedures
2.1. Geological setting
The Junggar Basin is located in the northern Xinjiang Uygur
Autonomous Region, north-west of China, and it is the second largest
inland basin in China. The basin covers an area of 13 × 104 km2 with
370 km from north to south and 700 km from east to west. The study
area is situated in the east of Junggar Basin and is composed of fifteen
secondary structural units. The eastern Junggar region was strongly
transformed by Late Indosinian movement, which further enhanced the
chessboard-shaped structural framework with uplifts and depressions
alternation. The seismic reflection formation below Cretaceous of this
area has a chessboard-shaped structural framework with convex and
concave alternation and picturesque disorder (Fig. 1). The tectonic
uplift of Kelameili Mountain provides the source of the Jurassic for the
eastern Junggar region. Large-scale continental coal-forming lacustrine
in Yanshanian period made Badaowan formation (J1b) deposition have
filling up characteristics where only locally develop minable coal seam.
Fig. 1. Geographical position of the tectonic units, along with the stratigraphic column for the coal-bearing strata in eastern Junggar Basin.
805
Fuel 211 (2018) 804–815
During the deposition of the Xishan formation (J2x) where most of the
main coal seams occur, the basin was in a balanced subsidence stage.
The tectonic evolution and sedimentary characteristics control the coal
seam development characteristics. Therefore, Badaowan formation of
Lower Jurassic (J1b) and Xishanyao formation of middle Jurassic (J2x)
are two main coal-bearing strata, which are widely distributed in the
basin. Besides, most of the coal in the eastern Junggar Basin is typical
and favorable low-rank coal with Ro of 0.38–0.7%. In order to reveal
the differences in material composition, pore-fracture system and
characteristics of the adsorption capacity of low-rank coals around the
FCJ, ten fresh coal samples from Xishanyao Formation in different
mining areas (Fig. 1) were collected and analyzed.
2.2. Analytical procedures
2.2.1. Material composition
Firstly, coal samples were analyzed for proximate analysis under air
dried basis following the Chinese national standard GB/T 212-
2008[10]. According to ISO 7404.3-1994 [11] and ISO 7404.5-1994
[12], mean vitrinite reflectance (Ro) measurements and maceral ana-
lyses (500 points) were performed on the same polished section of the
coal samples using a Leitz MPV-3 photometer microscope. The X-ray
fluorescence spectroscopy (XRF) was used to determine the oxides of
major elements, including SiO2, Al2O3, CaO, K2O, Na2O, Fe2O3, MnO,
MgO, TiO2, and P2O5, following the method described in Ryu et al.
(2011) [13].
2.2.2. Pore structure
Based on the material composition analysis results, eight samples
S. Tao et al. Fuel 211 (2018) 804–815

were selected to analyze the characteristics of pore-fracture by using
low temperature nitrogen adsorption (LTNA), mercury intrusion por-
osimetry (MIP), and nuclear magnetic resonance (NMR). MIP was
carried out according to the national standard, SY/T 5346-2005 [14],
by using Micromeritics Auto Pore IV 9500. Before the LTNA experi-
ments, all the samples were crushed and sieved to a size of
0.18–0.25 mm (60–80 mesh) (dried for 48 h), and then were tested
using Micromeritics ASAP2020 at a low temperature (77 K). Besides,
coal columns with the size of about 2.5 cm in diameter, 2–5 cm in
length were drilled along the direction parallel to face cleat, and then
saturated columns were prepared for the low field nuclear magnetic
resonance (NMR) measurements which were carried out using a
MiniMR60 instrument to obtain the comprehensive identification of
pore structure at different pore sizes, following the guidelines estab-
lished by Yao et al. (2010). [15] Coal permeability was tested with He
by using the automatic helium porosity/permeability analyzer (AP
608), following the Chinese petroleum industry standard SY/T 6385-
1999 [16]. The coal columns were heated for 24 h at 80 °C and then
cooled down to the ambient temperature (20 °C in constant).
2.2.3. Adsorption capacity
Finally, in order to reveal the influence of FCJ on the adsorption
capacity of coal, isothermal adsorption experiments were carried out.
All the coal samples were crushed and sieved to a size of 0.18–0.25 mm
(60–80 mesh). Then, 100–125 g samples were weighed for the
moisture-equilibrium treatment which was performed for at least four
days. After that, each sample was tested using a USA Terratek Isotherm
Measurement System (ISO-300), while the experimental temperature
and equilibrium pressure were maintained at 30 °C and 11 MPa, re-
spectively.
3. Results and discussion
3.1. Material composition
Coal maceral would undergo a qualitative change during the coa-
lification, which affects both the adsorption capacity and the hydro-
carbon generating potential of coal. [17–19]As shown in Table 1, the
huminite (lignite)/vitrinite (candle coal) content is the highest in the
eastern Junggar basin (45.34–81.94%, mean 63.24%), followed by in-
ertinite (8.79–38.58%, mean 23.08%) and exinite (1.06–5.37%, mean
3.02%). The inertinite content increases from lignite to candle coal,
while the exinite and minerals content shows a decreasing trend.
The low-rank coal samples have low ash yields (1.82–10.41%),
whereas they contain relatively high moisture content (1.91–13.57%)
and volatile yields (29.53–45.07%). With the decrease in moisture
content, the ash yield changes little from lignite to candle coal, but the
volatile yield increases obviously (Fig. 2a). The main reason for this
phenomenon is that the lignite contains a large amount of carboxyl and
hydroxyl functional groups, which render it a strong hydrophilic
ability. When the Ro value is close to 0.5%, aliphatic and cycloaliphatic
functional groups and side chains begin to peel off from the hydrogen-
rich vitrinite and the exinite, thus producing asphaltenes, which char-
acterize the onset of the bituminization of coal. The process is si-
multaneously by continuous precipitation in water and carbon dioxide,
and by the generation of a small amount of volatile matter (non-aro-
matic components).
A plot of the major elements, normalized to Chinese coal values of
Dai et al. (2012) [20] shows that the SiO2, K2O, TiO2 and P2O5 are
slightly depleted whereas Na2O, MgO and CaO show enrichment
(Fig. 2b).
In addition, the composition and distribution of macerals are found
to be different in lignite and candle coal (Fig. 3). In candle coal, the
microsporinite, cutinite, liptodetrinite and occasionally some resinite
are distributed in collinite (Fig. 3a). Furthermore, several pores are
filled with bituminite originated from the FCJ (Fig. 3b). The semi-
fusinite in candle coal is lens-shaped at macroscale. The cell wall is
badly crushed but granular in order, and the lumen is filled with clay
minerals (Fig. 3c). The maceral in lignite is dominated by homo-
collinite, where the microsporinite, liptodetrinite and a small amount of
resinite are distributed (Fig. 3d). The secondary fracture in lignite is
more developed than that in the candle coal, whereas the cell structure
is well preserved (Fig. 3e). Fig. 3f shows that the cellular structure is
clear and well-organized in lignite, and the boundaries between the
cells are easy to recognize.
3.2. Pore-fracture structure
The metamorphic process changes the physical and chemical states
of coal, and it alters the original pore-fracture characteristics [21]. In
order to reveal the difference in pore-fracture characteristics around the
FCJ, eight samples (Samples 1, 3, 5, 6, 7, 8, 9 and 10) were selected for
experiments including MIP, LTNA, and low field NMR. Compared with
the conventional natural gas, coalbed methane is mainly adsorbed on
inner surface of the coal matrix, and the larger SSA always means the
higher sorption capacity of coals. Usually, the micropore and transition
pore are regarded as absorption pore (< 100 nm) due to its huge SSA,
while the mesopore and macropore are considered as the major flow
paths for gas and water during the development of CBM, namely see-
page pore (> 100 nm). [21–23] Because the SSA of seepage pore is far
less than that of the adsorption pore, therefore, the adsorption in see-
page pore is usually ignorable. In addition, the MIP is easy to damage
the micropore structure due to its high injection pressure, and therefore
it is more applicable to analyze seepage pore [24]. Nitrogen adsorption
at 77 K has been generally accepted as the standard method for both
micropore and mesopore size analysis, but it is difficult to characterize
extremely narrow micropores (< 1 nm). [25,26] In this work, seepage
pore was characterized by MIP, while adsorption pore was determined

Table 1
Results of the proximate and maceral analyses.

Sample No. Formation Ro (%) Coal composition (%) Proximate analysis (%)

Vitrinite/Huminite Inertinite Exinite Mineral Mad Aad Vad

1 J2x 0.41 45.78 29.05 5.37 19.8 9.41 2.66 31.42

2 J2x 0.42 56.16 28.7 4.14 11 13.57 2.86 30.93
3 J2x 0.42 45.34 37.9 4.46 12.3 12.57 3.72 30.68
4 J2x 0.43 46.91 38.58 3.51 11 12.7 2.8 33.22
5 J2x 0.45 58.35 25.17 3.18 13.3 12.53 2.71 29.53
6 J2x 0.49 68.76 22.21 3.13 5.9 8.62 3.41 33.42
7 J2x 0.61 76.54 12.85 2.71 7.9 2.11 1.82 39.27
8 J2x 0.66 77.7 12.54 1.06 8.7 2.05 5.04 37.75
9 J2x 0.66 74.88 15.01 1.31 8.8 1.91 10.41 30.25
10 J2x 0.67 81.94 8.79 1.37 7.9 3.74 3.6 45.07

Notes: Mad = moisture content; Aad = ash yield; Vad = volatile yield; ad = air dried basis.

806
S. Tao et al. Fuel 211 (2018) 804–815

100
Sample 1 Sample 2 Sample 5
16 Mad Vad (a) 50
Sample7 Sample 9 Sample 10 (b)
14 The first coalification jump Na
45

Sample / Coal in China

10
12 y = 29.846x + 18.574
Mn
R² = 0.4558 Fe
40 Mg
10 Ca
Mad(%)

Vad(%)
Si Al P
8 35 1

6
30 K
4 Ti
y = -41.437x + 29.592 0.1
25
2 R² = 0.8904

0 20
0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.01
R (%)
Fig. 2. (a) Relationship between Mad, Vad and Ro around the first coalification jump; (b) The Chinese coal-normalized concentration of major elements.

Fig. 3. Characteristics of maceral in candle coal (a,b: reflection fluorescence, c: reflection single polarization) and lignite (d: reflection fluorescence, e,f: reflection single polarization by
optical microscopy. Notes: C represents the collinite; R represents the resinite; Cu represents the cutinite; Mis represents the microsporinite; LD represents the liptodetrinite; B represents
the bituminite; Cl represents the clay mineral; Sf represents the semifusinite; Hco represents the homocollinite; Fi represents the secondary fracture.

from N2 adsorption. The low-field NMR was used to evaluate both
seepage and adsorption pores, as well as their connectivity.
3.2.1. Seepage pore structure and seepage capacity
Pore structure of coal can be observed intuitively from mercury
injection and ejection curves [21]. As shown in Fig. 4, after injected a
high amount of mercury (0.1369–0.2535 ml/g), the mercury injection
curve (MIC) of lignite shows three different sections. When the pressure
is lower than 1 MPa, a steep slope on MIC is observed, which corre-
sponds to the pore diameter > 1000 nm (Fig. 5a). From 1 to 10 MPa,
the MIC is nearly a horizontal straight line, which indicates a good
development of macropores and mesopores. When the pressure is
higher than 10 MPa, the slope of MIC increases a little, indicating that
the adsorption pores are not developed. The MIC of candle coal shows
two distinct sections. When the pressure is between 2.41 MPa and
8.96 MPa, the amount of mercury injected barely increases, which re-
flects that the coal is extremely dense. The injection amount of mercury
increases linearly when the pressure is higher than 10 MPa, and the MIC
almost coincides with the ejection curve (Fig. 4). This kind of curve is
characterized by high micropore and transition pore volume, low see-
page pore volume, and poor pore connectivity. This pore size
distribution (PSD) is advantageous to CBM adsorption-desorption, but it
is unfavorable to CBM output (Fig. 5b).
As listed in Table 2, the total pore volume (TPV, equivalent to Vt in
Table 2) decreases greatly from lignite (0.1369–0.2535 cm3/g) to
candle coal (0.0385–0.05 cm3/g), with the increase of micropore and
transition pore proportion, the decrease of mesopore and macropore
volume. Meanwhile, the average pore diameter also shows a decreasing
trend. The lignite has a more uniform PSD and the better pore con-
nectivity. However, the candle coal has an equally polarized PSD and it
mainly develops adsorption pores and macropores. The difference of
pore volume percentage with different diameter between lignite coal
and candle coal suggests that the FCJ has a great impact on pore
structure of low-rank coals. During the compaction and coalification,
the number of macropore and fractures is reduced. Therefore, a sig-
nificant reduction in the macropore volume was observed with in-
creasing coal rank. Meanwhile, the coal molecules polycondensation
occurs during the FCJ process, which primarily affects the mesopore
structure, and resulting in the decrease of the mesopore volume with
increasing coal rank.
As we mentioned above, it is widely accepted that the micropore
and transition pore are adsorption pores due to their enormous SSA

807
S. Tao et al. Fuel 211 (2018) 804–815

0.25 Fig. 4. Mercury injection-extrusion curve of lignite and candle coal.

Ro=0.41% intrusion
Ro=0.41% Extrusion
Ro=0.42% intrusion
0.2 Ro=0.42% Extrusion
Lignite
Hg volume (ml·g -1)

Ro=0.45% intrusion
Ro=0.45% Extrusion
Ro=0.49% intrusion
0.15
Ro=0.49% Extrusion
Ro=0.61% intrusion
Ro=0.61% Extrusion
0.1 Ro=0.66% intrusion
Ro=0.66% Extrusion
Ro=0.66% intrusion
Ro=0.66% Extrusion Candle
0.05 coal
Ro=0.67% intrusion
Ro=0.67% Extrusion

0
0.001 0.01 0.1 1 10 100 1000
Pressure(MPa)

[15–28]. The mesopore and macropore, which are described as the porosity. During the FCJ, the amount of primary pore decreases
seepage pores, controlling the transportation and diffusion of CBM sharply, whereas that of the thermal pore amount increases gradually.
[29,30]. If value of 60% of the seepage pore proportion is taken as the Compared to the increase in the amount of micropore, the reduction in
dividing line of adsorption pore type (< 60%) and seepage pore type the amount of macropore dominates, leading to the decrease in coal
(≥60%), and considering the connectivity between the pores at dif- porosity. Fig. 7b shows that there is a positive linear correlation be-
ferent size level (closed type, proportion of mesopore lower than 20%; tween the permeability and seepage pore volume.
open type, proportion of mesopore > 20%), the pore can be divided
into four types, namely the open seepage pore, closed seepage pore,
open adsorption pore and closed adsorption pore. It can be seen from 3.2.2. Adsorption pore structure and adsorption capacity
Fig. 6 that lignite mainly develops the open seepage pore, whereas the 3.2.2.1. Adsorption pore structure. Although different coal samples have
candle coal mainly develops the closed adsorption pore type. different adsorption isotherms, the basic characteristics are similar. The
In general, the micropores with a large SSA corresponding to the adsorption volume of liquid nitrogen increases slowly low relative
small pores in coal matrix constitute the adsorption space. The transi- pressure (Fig. 8). When the relative pressure approaches 1 MPa, the
tion pore constitutes the capillary condensation and diffusion regions, adsorption volume increases rapidly. During the gas desorption process,
whereas the mesopore constitutes the slow seepage area for CBM. The the adsorbate gradually evaporates, and the difference in the adsorption
macropore, which corresponds to cleats or fractures, forms a strong loop of coal sample reflects the difference in pore shape. For most low-
laminar flow region. A relatively large macropore and mesopore vo- rank coal samples (except for Sample. 3 with Ro = 0.43%), the
lumes usually represent a strong seepage capacity and correspond to adsorption and desorption curves are separated with a small
high porosity and permeability. It can be seen that the permeability of hysteresis loop for P/P0 > 0.5 (pore diameter > 2.76 nm), suggesting
low-rank coal samples varies from 0.11 to 74.3 mD (Fig. 7), and a po- that the P/P0 values for the pore condensation and evaporation are
sitive correlation exists between the porosity and permeability (Fig. 7a). different, and some pores exist in open shapes [31]. In addition, there
It is important to note that the lignite in eastern Junggar Basin is soft might be some semi-open pores on the coal surface because the semi-
and friable, and it develops visible exogenous fractures. Therefore, the open pores do not contribute to the hysteresis loop. This kind of pore is
permeability of lignite is higher than that of the candle coal, which is beneficial to adsorption, desorption and diffusion of CBM.
related to the changes in macropore volume and the fracture density The results of N2 adsorption-desorption that calculated by BJH
during the process of coalification jump. The structure of lignite is loose model [32] and BET model [33] are presented in Table 3, respectively.
with good development of primary pores and fractures, which contain a It can be seen that the BET SSA of coal samples ranges from 0.19 to
large number of hydroxyl and carboxyl groups, and result in a high 5.6 m2/g, and BJH TPV ranges from 1.04 to 13.63 × 10−3 mL/g. The
average pore diameter of the coal sample varies between 2.45 nm and

0.3
Ro=0.41% (a) 0.08
(b)
Ro=0.42%
Ro=0.45% Ro=0.61%
Hg volume (ml·g-1)

0.06
Ro=0.49% Ro=0.66%
Hg volume(ml·g-1)

0.2
Ro=0.66%
Ro=0.67%
0.04

0.1
0.02

0 0
1000000 10000 100 1 1000000 10000 100 1
Pore diameter (nm) Pore diameter (nm)

Fig. 5. Relationship between the total mercury injection and pore diameter.

808
S. Tao et al. Fuel 211 (2018) 804–815

Table 2
Pore distribution characteristics determined by MIP.

Sample No. Ro Vt (ml/g) MWE (%) DP (MPa) APD (nm) Proportion of pore distribution/%

(%) V1/Vt V2/Vt V3/Vt V4/Vt

1 0.41 0.2535 15.58 2.48 34.8 19.6 13.2 31.8 35.4

3 0.42 0.2231 20.17 2.42 27.5 21.83 14.27 30.8 33.1
5 0.45 0.1927 26.21 2.5 20.6 24.6 14.7 32.1 28.6
6 0.49 0.1369 34.26 1.4 15.9 25.9 20.8 28.6 24.6
7 0.61 0.075 57.60 2.97 10.2 40.67 24.67 21.6 13.07
8 0.66 0.0601 71.71 8.46 7.6 57.6 24.6 3.2 14.6
9 0.66 0.04 85.00 8.96 8 54.4 28.7 4.2 12.7
10 0.67 0.0385 94.55 2.41 7.2 59.28 28.77 3.38 8.57

Notes: Vt = Total mercury injection; MWE = Mercury withdrawal efficiency; DP = Displacement pressure; APD = Average pore diameter; V1 = Micropore volume (< 10 nm);
V2 = Transition pore volume (10–100 nm); V3 = Mesopore volume (100–1000 nm); V4 = Macropore volume (> 1000 nm).

0.016 67.2% 0.012 0.008 0.006

63.93% 53.25%
60.67%
Incremental intusion/ml·g-1

Incremental intusion/ml·g-1

Incremental intusion/ml·g-1
Incremental intusion/ml·g-1

0.012 0.006
0.008 0.004
39.33% 47.75%
Ro=0.45
0.008 32.8% Ro=0.41% 0.004 Ro=0.49%
36.07% Ro=0.42%

0.004 0.002
0.004 0.002

0 0 0 0
1 100 10000 1000000 1 100 10000 1000000 1 100 10000 1000000 1 100 10000 1000000

Pore diameter/nm Pore diameter/nm Pore diameter/nm

Pore diameter/nm

0.003
0.003 Ro=0.66 0.002 83.10% 0.002
82.20% Ro=0.66
65.30% Ro=0.61% 88.05% Ro=0.67
Incremental intusion/ml·g-1
Incremental intusion/ml·g-1

Incremental intusion/ml·g-1

Incremental intusion/ml·g-1
0.002 0.002 16.9%
34.70% 11.95%
17.8%
0.001 0.001

0.001 0.001

0 0 0 0
1 100 10000 1000000 1 100 10000 1000000 1 100 10000 1000000 1 100 10000 1000000
Pore diameter/nm Pore diameter/nm Pore diameter/nm Pore diameter/nm

Fig. 6. Pore size distribution of coal samples determined by MIP.

19.89 nm. From lignite to candle coal, the BJH TPV, BET SSA and the micropore has the least proportion (8.54–29.89%, mean: 17%).
average pore diameter reduce greatly. Fig. 9 indicates that the BJH TPV Therefore, the effect of FCJ on adsorption pores is relatively assuasive
is positively correlated with the BET SSA. However, the BET SSA and than that on seepage pores, which mainly results in a uniform bulk
the average pore diameter exhibit an opposite trend, suggesting that the compaction for micropore and transition pore, with an obvious decline
small pores are the main contribution to the BET SSA. Meanwhile, as of BJH TPV.
the coal rank rises, there is no obvious difference in pore volume per- However, the FCJ has an important impact on pore size distribution
centage with different diameters, as shown in Table 3. For both the (PSD) of the lignite and the candle coal. The PSDs (pore diameter
lignite and candle coal, transition pores are dominant (44–59.51%, ranges from 2 to 100 nm) of 8 coal samples are shown in Fig. 10. The
mean: 51.56%), followed by pores with diameter100–300 nm, while the PSD of lignite is bimodal, with the peak values at 8–10 nm and

80 (a) 80 (b)
Lignite Lignite
60 60
Permeability(mD)

Candle coal Candle coal

Permeability(mD)

y = 6.0937x - 29.402
40 R² = 0.9743
40
y = 3.5487x - 17.121
R² = 0.8712

20 20
y = 0.5744x - 0.5146
R² = 0.8285 y = 1.0081x - 0.4738
R² = 0.9815
0 0
0 10 20 30 0 5 10 15 20
Porosity(%) Seepage pore volume(cm3/g)
Fig. 7. (a) Relationship between porosity and permeability. (b) Relationship between seepage pore volume and permeability.

809
S. Tao et al. Fuel 211 (2018) 804–815

10 0.25 0.4 0.3

Ro=0.41% Ro=0.43% Ro=0.45% Ro=0.49%
8 0.2
0.3
Volume(mL/g)

0.2

Volume(mL/g)

Volume(mL/g)

Volume(mL/g)
6 0.15
Adsorption Adsorption Adsorption
0.2 Adsorption
4 Desorption Desorption Desorption
0.1 Desorption
0.1
0.1
2 0.05

0 0 0 0
0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1
Relative pressure Relative pressure Relative pressure Relative pressure

1.5 1
2 0.04 Ro=0.66%
Ro=0.61% Ro=0.66% Ro=0.67%
0.8
1.5 0.03
1

Volume(mL/g)

Volume(mL/g)
Volume(mL/g)
Volume(mL/g)

Adsorption 0.6
Adsorption Adsorption Adsorption
1 0.02
Desorption Desorption Desorption Desorption
0.4
0.5
0.5 0.01
0.2

0 0 0 0
0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1
Relative pressure Relative pressure Relative pressure Relative pressure

Fig. 8. Nitrogen adsorption-desorption curve of 8 selected coal samples.

Table 3
Results of the low temperature nitrogen adsorption.

Sample No. Ro BJH TPV (10−3 ml/g) APD (nm) Pore volume percentage (%) BET SSA (m2/g)

(%) > 100 nm 10–100 nm < 10 nm

1 0.41 13.63 11.47 30.96 52.58 16.46 4.75

3 0.42 7.46 19.89 34.98 53.37 11.6 2.31
5 0.45 10.39 10.72 19.87 50.15 29.98 5.6
6 0.49 8.49 14.4 24.6 54.8 20.5 3.42
7 0.61 2.58 2.45 31.95 59.51 8.54 0.77
8 0.66 1.29 6.3 36.46 51.95 11.59 0.37
9 0.66 1.53 6.69 38.5 44.0 17.5 0.66
10 0.67 1.04 4.62 33.99 46.12 19.89 0.19

Notes: BJH TPV = Total pore volume; APD = Average pore diameter; BET SSA = Specific surface area.

6 (a) 25 (b)
BET specific surface area(m2/g)

BET specific surface area(m2/g)

5 Lignite Lignite y = -2.0428x + 8.304

20 R² = 0.5126
Candle coal Candle coal
4
15
3
y = 0.386x + 0.1628
R² = 0.5206 10
2
y = 0.34x - 0.0499
1
R² = 0.7515 5 y = -1.1797x + 25.908
R² = 0.5909
0 0
0 5 10 15 0 5 10 15
BJH total pore volume(10-3ml/g) Average pore diameter(nm)
Fig. 9. (a) Relationship between the BJH TPV and BET SSA. (b) Relationship between the average pore diameter and BET SSA.

40–60 nm, respectively. For candle coal, only one peak is observed, and important parameter to evaluate the gas content and CBM resources
the peak value is around 40–60 nm. The contribution of pores at dif- of coal reservoirs. [34–36] Fig. 12a indicates that the low-rank coal
ferent pore size to SSA is shown in Fig. 11, and the results show that the samples have a weak adsorption capacity, with the Langmuir volume
SSA distribution of lignite samples behaves as multimodal, and there (VL) ranging from 1.68 to 16.58 cm3/g and averaging 9.71 cm3/g. The
are three major peaks at 2–3 nm, 8–10 nm and 40–60 nm respectively. adsorption amount first increases quickly in low pressure region
These indicate that both micropore and transition pore are the main (< 4 MPa), and then it slowly increases as the pressure increases,
contribution to the total SSA of lignite coal (Fig. 11a). In contrast, the indicating that it is easy for methane to desorb with the decrease in
SSA distribution of candle coal samples behaves as unimodal with the pressure during the development process, and it means that the gas
peak appearing within 2–3 nm, which indicates that the BET SSA is production per unit pressure drop in early stages will be larger than that
mainly contributed by micropores (Fig. 11a). in later stages [37]. In addition, the VL value increases with the increase
in Ro (Fig. 12b). The average VL for candle coal is 15.66 cm3/g, which is
much higher than that of lignite (5.24 cm3/g).
3.2.2.2. Adsorption capacity. The coal adsorption capacity is an

810
S. Tao et al. Fuel 211 (2018) 804–815

0.0005
0.002
(a) (b)

Incremental Pore Volume(cm3·g-1)

Incremental Pore Volume(cm3·g-1 )

0.0004
0.0015 Ro=0.49% Ro=0.66%
Ro=0.45% 0.0003 Ro=0.67%
Ro=0.42% Ro=0.61%
0.001
Ro=0.41% 0.0002 Ro=0.66%

0.0005
0.0001

0 0
1 10 100 1 10 1 00
Pore diameter(nm) Pore diameter(nm)
Fig. 10. Pore size distribution of lignite (a) and candle coal (b).

0.25 Ro=0.49% 0.1

(a) (b) Ro=0.66%
Ro=0.45%
Incremental Pore Area (m2·g-1)

Incremental Pore Area (m2·g-1)

0.2 Ro=0.42% Ro=0.61%
0.08
Ro=0.41% Ro=0.67%
0.15 0.06 Ro=0.66%

0.1 0.04

0.05 0.02

0 0
1 10 100 1 10 1 00
Pore diameter(nm) Pore diameter(nm)
Fig. 11. BET SSA distribution at different pore size. (a) lignite (b) candle coal.

20
16 Ro=0 .67% (a) (b)
Ro=0 .66%

14 Ro=0 .61%
Adsorption capacity(cm3 ·g -1 )

Ro=0 .49%
Ro=0 .45%
12 Ro=0 .42% 15 y = 52.673x - 18.208
Ro=0 .41%
R² = 0.9502
10
VL(cm3·g -1)

8 10

4 5
2

0
0
0 2 4 6 8 10 12 0.4 0.5 0.6 0.7
Pressure(MPa) Ro(%)
Fig. 12. (a) Isothermal adsorption curves of coal samples. (b) Relationship between Ro and VL.

The impact of coal rank on the adsorption capacity is complicated only affected by the pore-fracture, but also by a comprehensive effect of
and it is influenced by many factors, such as the organic maceral, many other factors. Thus, the adsorption theory of middle-high rank
moisture, ash yield and pore-fracture structure. [38] Fig. 13 shows a coal is not completely applicable to low-rank coal, and the testing
similar changing trend for VL and vitrinite/huminite content but an methods for the low-rank coal need to be further explored.
opposite changing trend for inertinite and moisture content with the
increase in coal rank. In middle-high rank coals, because the methane
adsorbs on the surface of coal matrix, the adsorption capacity increases 3.2.3. Pore-fracture characterization by NMR and its fractal characteristic
with SSA. However, this phenomenon does not exist in the low-rank 3.2.3.1. Pore size distribution. T2 (Transverse relaxation time)
coal in the area currently studied (Fig. 14). This is because, firstly, there distribution characteristics of NMR can accurately reflect pore size
is not only the adsorbed gas, but also a large amount of free gas existing distribution. [15] Generally, the area and width of peak show the
in the pores with a diameter > 50 nm. Therefore, the VL value cannot proportion of a particular class of pores and the selectivity of pores,
completely represent the theoretical maximum gas content of low-rank respectively. [31] The number of peaks indicates the pore connectivity.
coal; secondly, methane mainly adsorbs in nanopores, although the SSA Fig. 15 shows that the T2 spectrum characteristics of lignite and candle
shows a decreasing trend from lignite to candle coal, the contribution of coal have significant differences. NMR T2 distribution of lignite shows a
micropore to the SSA gradually increases (Fig. 11b); thirdly, VL is not bimodal type, while the transverse relaxation times of two peaks are
located at 1 ms and 600 ms. The peaks of mesopore and macropore are

811
S. Tao et al. Fuel 211 (2018) 804–815

20 20 20
(a) (b) y = -0.519x + 20.719 (c)
15 15 R² = 0.7818
15 y = -1.1651x + 18.202

VL(cm3/g)
VL(cm3/g)

VL(cm3/g)
R² = 0.7718
10 10 10

5
5 5
y = 0.3649x - 13.978
R² = 0.8222
0
0 0 20 40 0
40 60 80 100 0 5 10 15
Vitrinite content (%) Inertinite content (%) Moisture content(%)

20 40 20
(f) Moisture content 15
20 100 (e)
(d) Langmuir volume

Moisture content(%)
Inertinite content (%)
Vitrinite content (%)
80 15 30 15
15 10

VL(cm3/g)
VL(cm3/g)

60
VL(cm3/g)

10 20 10
10
40 5
5 5 10 5
Langmuir volume 20 Inertinite content

Vitrinite content Langmuir volume

0 0 0 0 0 0
0.4 0.5 0.6 0.7 0.4 0.5 0.6 0.7 0.4 0.5 0.6 0.7
Ro(%) Ro(%) Ro(%)
Fig. 13. The relationship between vitrinite/huminite content (a and d), inertinite content (b and e), moisture content (c and f) and VL.

20 proportion of seepage pore tends to decrease. Furthermore, the pore

connectivity is also reduced.
In addition, a comprehensive comparison of NMR pore size dis-
15 tribution can be used to further verify the characteristics of different
y = -2.5198x + 16.145 pore structures caused by the FCJ. According to the definitions of pore
R² = 0.7431
VL(cm3/g)

types (micropore and transition pore: < 2.5 ms; mesopore: 2.5–100 ms;
10 macropore: > 100 ms) [28], it can be seen that the NMR T2 distribution
is consistent with the mercury injection capillary pressure curve,
showing that lignite mainly develops seepage pores, whereas candle
coal mainly develops adsorption pores (Fig. 16). However, both MIP
5
and NMR have some inherent shortcomings. Firstly, the pore size dis-
tribution obtained from MIP can only reflect the range within the ca-
pillary diameter. Ultra-capillary pores, however, may not be reflected.
0 Moreover, the compression effect of coal may induce the illusion of
0 2 4 6 mercury intrusion under high pressure conditions. Therefore, a com-
BET specific surface area(m2/g) bination of MIP and NMR could characterize the pore structure of low-
rank coal reservoirs more accurately.
Fig. 14. Correlation between the BET SSA and VL.

3.2.3.2. Fractal characteristics. The NMR T2 distribution is related to

merged together, indicating a good interconnectivity. The candle coal the MIC, both of which reflect the pore structure of rocks to a certain
presents a trimodal type, whereas the peak values of T2 are located at degree. If the pore structure of coal reservoirs has a fractal nature, the
0.25 ms, 30 ms and 200 ms, respectively. Overall, the proportion of Lg (Sv) and Lg (T2) of NMR should have linear correlation as represented
adsorption pores tends to increase from lignite to candle coal, while the in Eq. (1) [39,40]

600
Ro=0.41% 400
R=0.61%
500 Ro=0.42% 350
Ro=0.66%
Ro=0.45%
300 Ro=0.66%
400 Ro=0.49%
Amplitude

250 Ro=0.67%
Amplitude

300 Bimodal 200

Triple peak

200 150

100
100
50

0 0
0.01 0.1 1 10 100 1000 10000 0.01 0.1 1 10 100 1000 10000
Transverse relaxation time /ms Transverse relaxation time /ms
Fig. 15. NMR T2 spectrum of coal samples in the eastern Junggar Basin.

812
S. Tao et al. Fuel 211 (2018) 804–815

3 4 Fig. 16. Pore size distribution according to NMR T2 spectrum.

36.62% Ro=0.42%
Ro=0.41% 3
2

Proportion/%
Proportion/%

32.25%
2
1 54.38%
1

0 0
0.01 1 100 10000 0.01 1 100 10000
Transverse relaxation time /ms Transverse relaxation time /ms
4
42.42% 4
Ro=0.45% 49.35%
3 Ro=0.49%
3
Proportion/%

Proportion/%

2 2
57.58%
50.65%
1 1

0 0
0.01 1 100 10000 0.01 1 100 10000
Transverse relaxation time /ms Transverse relaxation time /ms

5 6
67.44% 74.22%
Ro=0.61% Ro=0.66%
5
4
Proportion/%

4
Proportion/%

3
32.56% 3
2 25.78%
2
1 1

0 0
0.01 1 100 10000 0.01 1 100 10000
Transverse relaxation time /ms Transverse relaxation time /ms

7 8
82.93% 90.28%
6 7
Ro=0.66% 6 Ro=0.67%
5
Proportion/%

Proportion/%

5
4
4
3
3
2 17.07%
2
9.72%
1 1
0 0
0.01 1 100 10000 0.01 1 100 10000
Transverse relaxation time /ms Transverse relaxation time /ms

lg(Sv ) = (3−D)lg(T2) + (D−3)lgT2max
where, Sv is the percentage of cumulative pore volume in the TPV when
the transverse relaxation time is less than T2, %; D is the pore fractal
dimension; T2 is the transverse relaxation time, ms; T2max is the
transverse relaxation time corresponding to the maximum pore size,
ms.
During the calculation, the Sv–T2 double logarithm curve is divided
into two sections, which are used to calculate the fractal dimension DN1
of adsorption pores (T2 ≤ 2.5 ms), fractal dimension DN2 of seepage
pores (T2 > 2.5 ms) and total fractal dimension DNMR, respectively
(1) (Fig. 17a). In addition, for MIP, the fractal dimension DM1 of the ad-
sorption pores (Log P < −.5), fractal dimension DM2 of the seepage
pores (Log P > −.5) and the total fractal dimension DMIP are also
calculated based on the geometric theory of Washburn equation
(Fig. 17b) [41]. The results of the fractal dimension according to MIP
and NMR are shown in Table 4. The distribution of pore throat size is
different between NMR and MIP. The MIP reflects the minimum throat
and radius/volume of the pore connected to the throat, and can only
show the pore throat space in the part of mercury injection (Mercury
injection saturation < 100%). The pore throat reflected by NMR T2

813
S. Tao et al. Fuel 211 (2018) 804–815

1 0.5
NMR (a) MIP (b)
DN2=2.9158
y = 0.0842x - 0.1807
DN1=2.0015
-0.5
R² = 0.9127
y = 0.9985x - 0.2936
0 D MIP =2.2959
R² = 0.8877

log10dv/dp
-1.5
y = -0.7041x - 2.4477
lg W
R² = 0.9664
DNMR=2.6921 D M1 =2.0288
y = 0.3079x - 0.5169 -2.5 y = -0.9712x - 2.7061
-1 R² = 0.6562 R² = 0.9695

-3.5 D M2 =2.391
y = -0.609x - 2.5705
R² = 0.9476
-2 -4.5
-2 -1 0 1 2 3 -3 -1 1 3
lg T2 Log P

Fig. 17. Comparison of fractal dimensions of NMR, and MIP (Sample 7, Ro = 0.61%).

Table 4
Results of the fractal dimensions of NMR and MIP.

No DN1 R2 DN2 R2 DNMR R2 DM1 R2 DM2 R2 DMIP R2

1 1.3159 0.9242 2.8994 0.7956 2.6244 0.6332 2.3721 0.9362 2.0810 0.891 2.0530 0.9591
2 1.6128 0.934 2.9311 0.8063 2.7086 0.6085 2.3059 0.9518 2.2733 0.9552 2.0667 0.9742
3 1.6900 0.9291 2.9297 0.8677 2.7290 0.6096 2.1847 0.9703 2.1892 0.8541 2.0156 0.9815
4 1.6398 0.9299 2.9312 0.8402 2.7207 0.6034 2.3119 0.9208 2.2798 0.8988 2.1926 0.9795
5 2.0015 0.8877 2.9158 0.9827 2.6921 0.6562 2.0288 0.9695 2.3910 0.9476 2.2959 0.9664
6 2.1174 0.8482 2.9259 0.9783 2.7577 0.5868 1.9113 0.9233 2.2204 0.8021 2.2879 0.9615
7 2.0701 0.8627 2.9525 0.9956 2.7868 0.5572 1.7359 0.9634 2.3310 0.9146 2.2320 0.8397
8 2.0171 0.8744 2.9737 0.9457 2.8310 0.5158 1.8900 0.945 2.3314 0.8181 2.3244 0.9102

14 14 100 (b) 100

(a) Contributionof micropore for BET
Propotion of micropore volume/%

90 Propotion of micropore volume 90

BET specific surface area/m2·g-1

Contributionof micropore for BET/%

12 BET specific surface area 12
Average pore diameter/nm

Average pore diameter 80 80

10 10 70 70
y = -7.7388x + 22.466
R² = 0.6936 60 60
8 8
50 y = 38.298x + 3.6451 50
R² = 0.9272
6 6
40 40

4 4 30 30
y = -5.6074x + 12.604
R² = 0.6503 20 20
y = 54.994x - 68.383
2 2
10 R² = 0.8838 10
0 0 0 0
1.0 1.5 2.0 2.5 1.0 1.5 2.0 2.5
Fractal dimension, DN1 Fractal dimension, DN1

Fig. 18. (a) Relationship between DN1 and BET SSA and average pore diameter. (b) Relationship between DN1 and proportion of micropore volume and contribution of micropore to BET
SSA.

spectrum distribution is the entire pore throat (including dead pores), low-rank coal, with the increase of micropores, both pore volumes and
whose radius is greater than that of the MIP. Due to the difference in the average pore diameter decrease, while DN1 decreases. The lignite pri-
pore radius range of NMR and MIP (RNMR > RMIPA), the total fractal marily consists of the seepage pore, the adsorption pore, and a small
dimension exhibits a relationship of DNMR > DMIP, with the regression proportion of micropore (Table 3; < 24.8%). Therefore, the surface is
coefficients (R2) of NMR lower than that of the MIP. Moreover, the relatively smooth and the coal exhibits a strong homogeneity, resulting
adsorption pore fractal dimension DN1 (1.32–2.11) shows an opposite in a low DN1 value (1.32–1.69). The candle coal, after the FCJ, has
variation trend with DM1 (1.74–2.37), while the seepage pore fractal experienced mechanical compaction and dehydration, and pore struc-
dimension DN2 (2.89–2.97) is relatively close to DM2 (2.08–2.39). In ture is rearranged. From lignite to candle coal, the pore volume de-
addition, the R2 of DN2 (0.796–0.996) is consistent with the R2 of DM2 creases sharply, and the volume of Macropores and mesopores de-
(0.85–0.96). In other words, the distribution of seepage pore with NMR creases faster than that of the micropors. Therefore, the percentage of
is similar to that with MIP, which indicates that the adsorption pore micropore increases continuously (> 41.5%). Meanwhile, with the in-
fractal dimension DN1 is of great importance for the analysis of ad- crease of surface roughness, the coal heterogeneity enhances and DN1
sorption space of coal. increases.
The fractal dimension DN1 is closely related to the development of To sum up, for low-rank coals reservoirs, the FCJ has a significant
micropores and SSA, and it can be used to characterize the complexity/ impact on reservoir heterogeneity, including material composition,
irregularity of the pore-fracture surface and the heterogeneity of coal pore structure, seepage capacity and adsorption capacity, which must
[40]. Fig. 18 shows that the DN1 is negatively correlated with the BET be taken into consideration in the future CBM development strategy. In
SSA and the average pore diameter, while it has a positive correlation general, the lignite has a higher pore volume, better pore connectivity,
with the proportion of micropore volume and the rate of contribution of larger porosity and greater permeability, which is beneficial to the
micropore to SSA, with the regression coefficient as high as 0.9272. For production of CBM. For the candle coal, during the polycondensation of

814
S. Tao et al.
coal molecules and the compaction of coal matrix in the FCJ process, its
porosity and permeability decline rapidly, while its adsorption capacity
tends to increase. These phenomena indicate that reasonable fracturing
operation needs to be taken to improve the connectivity and the see-
page capacity during the development process of CBM within the
candle coal. The observations in this study could deepen awareness and
understanding of low-rank coal reservoir heterogeneity and the influ-
ence of FCJ on reservoir property, which have important directive to
production practices of CBM in low-rank coal seams.
4. Conclusions
(1) The FCJ has a significant impact on low-rank coal reservoir het-
erogeneity. Contrast with the candle coal, the lignite has a higher
inertinite content, larger pore volume, better connectivity, and
greater SSA. During the process of FCJ, the polycondensation of
coal molecules and the compaction of coal matrix occur, leading to
a rapid decline of moisture, seepage pore volume, porosity and
permeability, and the cell wall in the candle coal is badly crushed
with clay minerals filled from optical microscopy.
(2) According to the N2 adsorption test results, the PSD of lignite ap-
pears to be bimodal, with two peak values present at 8–10 nm and
40–60 nm, whereas the PSD of candle coal shows only one peak at
40–60 nm. The larger pore volume (1.7–300 nm, measured by N2
adsorption) contributes to the larger SSA. The SSA of candle coal
mainly comes from the contribution of micropore (< 10 nm),
especially the 2–3 nm pores, while the micropore and transition
pore (10–100 nm) contribute to most of SSA of lignite.
(3) With increasing coal rank, the SSA decrease rapidly. However,
different from middle-high rank coals, the CH4 adsorption capacity
tends to decrease from lignite to candle coal due to material com-
position difference.
(4) The NMR T2 distribution of lignite shows a bimodal large pore type,
whereas the candle coal shows a trimodal small pore type, in-
dicating that the lignite mainly develops seepage pore and the
candle coal mainly develops adsorption pore. From lignite to candle
coal, the pore connectivity gradually. The NMR fractal results in-
dicate that the pore surface and complexity inside the coal increase
gradually from lignite to candle coal.
Acknowledgements
This work was supported by the National Natural Science
Foundation of China (41502157, 41530314, 41772132), the Key
Project of the National Science & Technology (2016ZX05043-001), and
the Fundamental Research Funds for the Central Universities
(53200859306). The authors are grateful to anonymous reviewers and
the editor for their careful reviews and detailed comments, which
helped to substantially improve the manuscript.
References
[1] Li W, Zhu YM, Wang G, Jiang B. Characterization of coalification jumps during high
rank coal chemical structure evolution. Fuel 2016;185:298–304.
[2] Kędzior S. Methane contents and coal-rank variability in the upper Silesian coal
basin. Poland Int J Coal Geol 2014;139:152–64.
[3] Guo H, Cheng Y, Wang L, Lu S, Jin K. Experimental study on the effect of moisture
on low-rank coal adsorption characteristics. J Nat Gas Sci Eng 2015;24:245–51.
[4] Hoffmann K, Huculakmączka M. The utilization possibility of waste lignite as a raw
material in the process of obtaining humic acids preparations. Pol J Chem Technol
2013;14:1–6.
[5] Shibaoka M. Micrinite and exudatinite in some Australian coals, and their relation
to the generation of petroleum. Fuel 1978;57:73–8.
[6] Bustin RM, Guo Y. Abrupt changes (jumps) in reflectance values and chemical
compositions of artificial charcoals and inertinite in coals. Int J Coal Geol
1999;38:237–60.
[7] Kopp OC, Lii MEB, Clark CE. Volatiles lost during coalification. Int J Coal Geol
2000;44:69–84.
[8] Canónico U, Tocco R, Ruggiero A, Suárez H. Organic geochemistry and petrology of
815
Fuel 211 (2018) 804–815
coals and carbonaceous shales from western venezuela. Int J Coal Geol
2004;57:151–65.
[9] Riediger CL. Solid bitumen reflectance and Rock-Eval Tmax as maturation indices:
an example from the “Nordegg Member”, Western Canada Sedimentary Basin. Int J
Coal Geol 1993;22:295–315.
[10] GB/T 212–2008, 2008. Chinese National Standard. Proximate analysis of coal (in
Chinese).
[11] ISO 7404.3-1994, 1994. Methods for the petrographic analysis of bituminous coal
and anthracite-part 3: method of determining maceral group composition.
[12] ISO 7404.5-1994, 1994. Method for the petrographic analysis of bituminous coal
and anthracite-part 5: method of determining microscopically the reflectance of
vitrinite. MOD.
[13] Ryu JS, Jacobson AD, Holmden C, Lundstrom C, Zhang ZF. The major ion, δ44/
40Ca, δ44/42Ca, and δ26/24Mg geochemistry of granite weathering at pH=1 and
T=25°C: power-law processes and the relative reactivity of minerals. Geochim
Cosmochim Acta 2011;75:6004–26.
[14] SY/T 5346–2005, 2005. Chinese petroleum and natural gas industry standard. Rock
capillary pressure measurement (in Chinese).
[15] Yao Y, Liu D, Che Y, Tang D, Tang S, Huang W. Petrophysical characterization of
coals by low-field nuclear magnetic resonance (NMR). Fuel 2010;89:1371–80.
[16] SY/T 6385–1999, 1999, Chinese petroleum and natural gas industry standard. The
porosity and permeability measurement of core in net confining stress, [In Chinese].
[17] Levine JR. Coalification: the evolution of coal as source rock and reservoir rock for
oil and gas. Law BE, Rice DD, editors. Hydrocarbons from coal, 38. AAPG Studies in
Geology; 1993. p. 39–77.
[18] Pan J, Hou Q, Ju Y, Bai H, Zhao Y. Coalbed methane sorption related to coal de-
formation structures at different temperatures and pressures. Fuel 2012;102:760–5.
[19] Scott AR, Kaiser Jr WR. Thermogenic and secondary biogenic gases, san juan basin,
colorado and new mexico – implications for coalbed gas producibility.
1994;78:1186–209.
[20] Dai S, Ren D, Chou C-L, Finkelman RB, Seredin VV, Zhou Y. Geochemistry of trace
elements in Chinese coals: a review of abundances, genetic types, impacts on human
health, and industrial utilization. Int J Coal Geol 2012;94:3–21.
[21] Xu H, Tang DZ, Liu DM, Tang SH, Yang F, Chen XZ, et al. Study on coalbed methane
accumulation characteristics and favorable areas in the Binchang area, south-
western Ordos Basin. China Int J Coal Geol 2012;95:1–11.
[22] Chen S, Tao S, Tang D, Xu H, Li S, Zhao J, et al. Pore structure characterization of
different rank coals using N2 and CO2 adsorption and its effect on ch4 adsorption
capacity: a case in Panguan syncline, Western Guizhou, china. Energy Fuels
2017;31:6034–44.
[23] Yao Y, Liu D, Tang D, Tang S, Huang W. Fractal characterization of adsorption-
pores of coals from north China: an investigation on CH4 adsorption capacity of
coals. Int J Coal Geol 2008;73:27–42.
[24] Laxminarayana C, Crosdale P. Role of coal type and rank on methane sorption
characteristics of Bowen Basin, Australia coals. J Int J Coal Geol 1999;40:309–25.
[25] Lee GJ, Pyun SI, Rhee CK. Characterisation of geometric and structural properties of
pore surfaces of reactivated microporous carbons based upon image analysis and
gas adsorption. Microporous Mesoporous Mater 2006;93:217–25.
[26] Thommes M, Cychosz KA. Physical adsorption characterization of nanoporous
materials: progress and challenges. Adsorption 2014;20:233–50.
[27] Li J, Liu D, Yao Y, Cai Y, Xu L, Huang S. Control of CO2 permeability change in
different rank coals during pressure depletion: an experimental study. Energy Fuels
2014;28:987–96.
[28] Yao Y, Liu D, Tang D, Tang S, Che Y, Huang W. Preliminary evaluation of the
coalbed methane production potential and its geological controls in the Weibei
Coalfield, Southeastern Ordos Basin. China Int J Coal Geol 2009;78:1–15.
[29] Shi JQ, Durucan S. Drawdown induced changes in permeability of coalbeds: a new
interpretation of the reservoir response to primary recovery. Transp Porous Media
2004;56:1–16.
[30] Shi JQ, Durucan S. CO2 storage in deep unminable coal seams. Oil Gas Sci Technol
2005;60:547–58.
[31] Sing KSW, Everett DH, Haul RAW, Moscou L, Pierotti RA, Reuquerol J, et al.
Reporting physisorption data for gas/solid systems with special reference to the
determination of surface area and porosity (Recommendations 1984). Pure Appl
Chem 1985;57:603–19.
[32] Barrett EP, Joyner LG, Halenda PP. The determination of pore volume and area
distributions in porous substances. I. computations from nitrogen isotherms. J Am
Chem Soc 1951;73:373–80.
[33] Brunauer S, Emmett PH, Teller E. Adsorption of gases in multi-molecular layers. J
Am Chem Soc 1938;60:309–19.
[34] Busch A, Gensterblum Y, Krooss BM, Littke R. Methane and carbon dioxide ad-
sorption-diffusion experiments on coal: upscaling and modeling. Int J Coal Geol
2004;60:151–68.
[35] Krooss BM, Bergen FV, Gensterblum Y, Siemons N, Pagnier HJM, David P. High-
pressure methane and carbon dioxide adsorption on dry and moisture-equilibrated
pennsylvanian coals. Int J Coal Geol 2002;51:69–92.
[36] Ross HE, Hagin P, Zoback MD. CO2 storage and enhanced coalbed methane re-
covery: reservoir characterization and fluid flow simulations of the Big George coal,
Powder River Basin, Wyoming, USA. Int J Greenh Gas Con 2009;3:774–86.
[37] Meng YJ, Tang DZ, Qu YJ, Xu Hao, Li Y. Division of the stages of coalbed methane
desorption based on the Langmuir adsorption isotherm. 2015;8:57–65.
[38] Shah A, Chiu E, Ames D, Harrigan S, Mckenzie D. Adsorption characteristics of coal
in constant-pressure tests. Can Geotech J 2009;46:1165–76.
[39] Zhang Z, Weller A. Fractal dimension of pore-space geometry of an eocene sand-
stone formation. Geophysics 2014;79:77–87.
[40] Zhou S, Liu D, Cai Y, Yao Y. Fractal characterization of pore–fracture in low-rank
coals using a low-field NMR relaxation method. Fuel 2016;181:218–26.
[41] Leon CALY. New perspectives in mercury porosimetry. Adv Colloid Interface Sci
1988;77:341–72.