Marine and Petroleum Geology 132 (2021) 105233

Contents lists available at ScienceDirect

Marine and Petroleum Geology

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

Diagenetic evolution sequence and pore evolution model of

Mesoproterozoic Xiamaling organic-rich shale in Zhangjiakou, Hebei, based
on pyrolysis simulation experiments
Liangwei Xu a, b, c, Keji Yang b, c, *, Hao Wei d, Luofu Liu c, Xiao Li d, Lei Chen c, Tong Xu c,
Ximeng Wang c
a
The MOE Key Laboratory of Orogenic Belt and Crustal Evolution, School of Earth and Space Sciences, Peking University, Beijing, 100781, China
b
Hebei Key Laboratory of Strategic Critical Mineral Resources, Hebei GEO University, Shijiazhuang, 050031, China
c
State Key Laboratory of Petroleum Resources and Prospecting, China University of Petroleum, Changping, Beijing, 102249, China
d
Experimental Practice Teaching Centre of Hebei GEO University, Shijiazhuang, Hebei, 050031, China

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

Keywords: Organic-rich shale is an unconventional and complex petroliferous system that integrates a “source-reservoir-
Xiamaling shale cap”, and the coupled evolution of hydrocarbon generation, diagenesis and nanoscale pores is the key problem
Hydrous pyrolysis that affects shale gas accumulation. Current research methods for this issue mainly include direct observation
Low-pressure gas adsorption
and physical simulation. Direct observation ignores the heterogeneity and regional differences of the natural
Pore evolution model
Diagenetic evolution sequence
samples, and physical simulation mostly lacks an intuitive characterization and cannot clearly show the rela­
tionship between minerals and pore evolution characteristics in the same region. To avoid the effect of het­
erogeneity on the researched samples and clearly and intuitively reveal the coupled evolutionary relationship
among the hydrocarbon generation of thermal maturation, diagenesis and nanopore structures in shale, this
research included hydrous pyrolysis experiments on low mature marine shale samples to achieve various ma­
turities and diagenesis stages. The pyrolysis products at each stage were recovered and subjected to an ongoing
multidisciplinary analytical program. The results show that an increased temperature intensifies the evolution of
dissolution pores in unstable brittle minerals, promotes clay mineral conversion, and accelerates the develop­
ment of clay mineral pores and organic pores. The pore volume (PV) of the nanoscale pores reached its minimum
at 300 ◦ C, increased to its maximum at 500 ◦ C and decreased thereafter. The surface area (SA) of the nanoscale
pores reached their minimum at 300 ◦ C and then continuously increased, while those of macropores reached
their maximum at 500 ◦ C and then decreased. The diagenesis of the pyrolysis products were mainly dissolution,
clay mineral transformation, the thermal maturation of organic matter (OM), compaction and cement filling. The
diagenetic evolution was divided into four stages, and the diagenetic evolution sequence and pore evolution
model of the shale were established. Hydrocarbon generation, diagenesis and nanoscale pore evolution have
synergistic effects, which have great significance for shale reservoir evaluation and shale gas accumulation,
exploration and exploitation. The conceptual evolution model provides a quantitative prediction method for
hydrocarbon generation, diagenesis and nanoscale pore variations.

1. Introduction porosity and permeability, which greatly restricts the economic and
effective exploitation of shale gas. The pore evolution of shale reservoirs
As an unconventional petroliferous system that features “source- is a complex physical and chemical process controlled by the super­
reservoir-cap” combinations, shale has diverse origins and large pore position of diagenesis and hydrocarbon generation and expulsion (Jar­
size ranges, and the nanoscale pores of shale are more complex than vie et al., 2007; Cui et al., 2013; Lewan et al., 2014; Wu et al., 2015; Guo
those of conventional reservoirs (Loucks et al., 2012; Yu, 2013; Liu et al., and Mao, 2019). Different from conventional oil and gas reservoirs,
2019a; Zhao et al., 2019). Moreover, shale is characterized by low organic-rich shale reservoirs are not only affected by diagenesis but also

* Corresponding author. State Key Laboratory of Petroleum Resources and Prospecting, China University of Petroleum, Changping, Beijing, 102249, China.
E-mail address: [email protected] (K. Yang).

https://doi.org/10.1016/j.marpetgeo.2021.105233
Received 14 January 2021; Received in revised form 8 July 2021; Accepted 12 July 2021
Available online 21 July 2021
0264-8172/© 2021 Elsevier Ltd. All rights reserved.
L. Xu et al. Marine and Petroleum Geology 132 (2021) 105233

controlled by hydrocarbon generation. Compared with the effect of experimental results were reduced. Our research results provide a
diagenesis on framework mineral pores, the evolution of organic pores is reference and an approach for the prediction of favourable shale
mostly controlled by thermal maturation (Guo and Mao, 2019; Liu et al., reservoirs.
2019a; Zhang et al., 2019; Qiu et al., 2020). Organic-rich shale experi­
ences complex sedimentary burial, diagenetic evolution, hydrocarbon 2. Samples and methods
generation and structural transformation over its geological history.
These processes yield shale with diverse pore morphologies and complex 2.1. Samples
pore-structure parameters (Wang et al., 2020a; Zhao et al., 2019; Li
et al., 2021). As one of the main diagenetic processes of shale, the The original sample (XML-0) used for the hydrous pyrolysis experi­
thermal maturation of organic matter (OM) determines the ability of ments was collected from the Mesoproterozoic Xiamaling shale forma­
shale to generate and expel hydrocarbons, controls the evolution of the tion in Zhangjiakou, Hebei, China. Low mature black marine shale was
pore network in OM, and affects the variation in inorganic pores. In collected from fresh outcrops at depths of 3–5 m below the ground. The
addition to controlling the occurrence and migration of shale gas, sampling location of the Xiamaling shale has been reported in various
thermal maturation also has a significant effect on shale gas enrichment studies (Liu et al., 2011; Xu et al., 2018, 2021a). Prior to the hydrous
and accumulation (Jarvie et al., 2007; Mastalerz et al., 2013; Wei et al., pyrolysis experiments, any weathered surfaces were polished and
2020; Wang et al., 2020b). Accordingly, the differences in pore char­ treated with deionized water. Then, the original low mature Xiamaling
acteristics at different diagenetic stages are an important reason for the shale block sample was drilled into six columns with a length of 8–10 cm
differences in the current exploitation and development effects of ma­ and diameter of 2.5 cm, which were the only suitable dimensions for the
rine shale. Research on the coupling evolution of hydrocarbon genera­ sample cell in the autoclave used in the hydrous pyrolysis experiments.
tion, diagenesis and nanoscale pores has great significance for After the pyrolysis experiments were finished, the simulation products at
understanding the evolution of shale reservoirs and the mechanism of each stage were recovered and subjected to an ongoing multidisciplinary
shale gas accumulation and for shale gas exploration and exploitation. analytical program.
At present, the research methods used to determine the impact of
thermal evolution and diagenesis on pore structure mainly fall into two 2.2. Hydrous pyrolysis experiments
categories. The first category is direct observation, which uses high-
resolution equipment to analyse shale samples at different maturities Hydrous pyrolysis experiments were performed on an instrument
and stages of diagenesis and analyses the differences in diagenesis designed by the PetroChina Research Institute of Petroleum Exploration
characteristics and pore development to assess the relationship between and Development (Cui et al., 2013; Wu et al., 2015; Han et al., 2019).
these factors. However, this method ignores heterogeneity and the The apparatus is composed of 6 parts, i.e., a heating furnace system,
regional differences between samples and focuses mostly on organic pressure system, liquid supply system, production and collection system,
pores rather than framework mineral pores. Therefore, direct observa­ total control platform system and auxiliary system. The maximum
tion cannot reveal all the evolutionary characteristics of all pores in temperature of the device can reach 550 ◦ C, the maximum static rock
shale during diagenesis, and the research results of different scholars pressure is 350 MPa, and the maximum fluid pressure is 120 MPa. The
vary greatly (Curtis et al., 2012; Mathia et al., 2016; Liang et al., 2018; setting conditions can simulate shale diagenesis in the real burial pro­
Liu et al., 2019a; Wang et al., 2020b). The second category is physical cess to the maximum extent, simulate the diagenetic environment, and
simulation, in which samples with low maturity are selected, a tem­ complete the diagenetic evolution process of shale formation simulta­
perature sequence is set, hydrocarbon generation is induced by heating, neously. To simplify the evolution process of pore structures in shale
and the pore evolution at different stages is quantitatively analysed. This reservoirs under geological conditions, this experiment mainly consid­
method reduces the effects of sample heterogeneity and regional dif­ ered four factors: temperature, heating mode, fluid medium conditions
ferences on the experimental results to a certain extent. It has strong and compression process. To cover the entire evolution sequence from
comparability and can provide the overall characteristics of pore evo­ low mature to mature, highly mature and overmature, the system was
lution in the process of diagenesis. However, the disadvantages of heated from room temperature to 300 ◦ C, 350 ◦ C, 400 ◦ C, 450 ◦ C,
physical simulation are that it lacks visual characterization, it cannot 500 ◦ C, and 550 ◦ C at a heating rate of 10 ◦ C/h. The formation water for
readily show the relationship between minerals and pore evolution simulation was CaCl2, and the salinity of the formation water was 100
characteristics within the same region in a clear and visual way, and mL/L. The static rock pressure, temperature and time of the simulation
analysing the differences in the pore evolution between OM and inor­ are shown in Table 1.
ganic matter is difficult (Chen and Xiao, 2014; Sun et al., 2019; Xu et al.,
2019, 2021b; Han et al., 2019; Wang and Guo, 2019; Guo and Mao, 2.3. Organic geochemistry and mineralogy
2019; Wang et al., 2020b).
To avoid heterogeneity and regional differences in the researched The total organic carbon (TOC) content of the simulated samples was
samples to a maximum extent, the coupled evolutionary relationship measured using a LECO CS-230 analyser. Prior to measurement, the
among hydrocarbon generation, diagenesis, and nanopore structures in
shale must be clearly and intuitively revealed. Type II low mature Table 1
organic-rich shale from the Mesoproterozoic Xiamaling Formation in Parameter settings for the diagenetic physical simulation experiment.
Zhangjiakou, Hebei was selected, and hydrous pyrolysis experiments Sample ID XML- XML- XML- XML- XML- XML-
were conducted. Field emission scanning electron microscopy (FE-SEM) 1 2 3 4 5 6
observations were carried out on the pyrolysis products to carefully
Maximum 300 350 400 450 500 550
evaluate the diagenesis evolution and nanoscale pore structural varia­ temperature (◦ C)
tion characteristics with respect to temperature at the same location. Maximum pressure 42.5 45.4 51.5 52.5 62 72
Then, as part of an ongoing multidisciplinary analytical program, the (MPa)
characteristics of shale pore evolution at different diagenesis evolu­ Heating rate (◦ C/h) 10 10 10 10 10 10
Heating time (h) 28 30 32 34 38 46
tionary stages were quantitatively analysed, the influences of different Pressurization rate 1.52 1.5 1.53 1.43 1.54 1.43
minerals on pore evolution were clarified, the coupled relationship be­ (MPa/h)
tween diagenesis and the nanoscale pore evolution of shale was Keeping time (h) 48 48 48 48 48 48
revealed, and the influences of the sample heterogeneity caused by System state Closed Closed Closed Closed Closed Closed
Total time (h) 105 115 110 100 102 110
different tectonic regions and sedimentary environments on the

2
L. Xu et al. Marine and Petroleum Geology 132 (2021) 105233

samples were placed in a crucible and treated with 5 % HCl to remove 3. Results
carbonates at 80 ◦ C. Pyrolysis data for the simulated samples were ob­
tained using a Rock-Eval 6 instrument. The reflectance of vitrinite-like 3.1. Geochemical characteristics
macerals of the initial shale was measured due to the lack of vitrinite
in the Xiamaling shale, and the macerals of the initial shales were Table 2 shows the geochemical characteristics of XML-0 and its
identified with a Leica MVP microscope photometer. According to the simulated samples. The TOC content of XML-0 is 7.18 %, S1+S2 is 35.46
given temperatures and heating rate of the pyrolysis experiments, the mg HC/g, and the hydrogen index (HI) is 488 HC mg/g TOC, which
EasyRo method was used to quantify the relationships between the shows that the original shale has a very high OM abundance and exhibits
maturity, temperature and time. The reaction degree (F) of the thermal good hydrocarbon generation potential. The EqRo of XML-0 is 0.58 %,
simulation experiment was used to establish the mathematical model of and the Tmax is 436 ◦ C, which indicates that XML-0 is low mature.
simulated vitrinite reflectance, and the formula is EASY%Ro = e− 1.6+3.7F Additionally, the EqRo of XML-6 is 2.57 %, and its Tmax is 612 ◦ C; this
(EASY%Ro is the simulated equivalent vitrinite reflectance in %) suggests that the sample had entered the overmature gas generation
(Lopatin, 1971; Waples, 1980; Sweeny and Burnham, 1990). stage at the end of the simulation. The TOC content and S1+S2 and HI
Mineralogy information was obtained using a Bruker D8 Advance X- values of the simulated samples gradually decreased, which implies that
ray diffraction (XRD) instrument, which was operated at 40 kV and 30 OM generated and expelled hydrocarbons in the simulation process.
mA with Cu Kα radiation. Stepwise scanning measurements were per­ The maceral identification results are shown in Fig. 1. Most areas of
formed at a rate of 4◦ /min between 3◦ and 85◦ (2θ). The relative mineral the microscopic fluorescent photographs show a brown colour, and the
percentages were obtained with a correction for Lorentz polarization main macerals in the photos are amorphous telalginite and lamalginite,
(Chalmers and Bustin, 2008). which are spherical algae that emit yellow-green fluorescence.
Furthermore, vitrinite-like macerals appear as grey bars under reflected
2.4. Low-pressure gas adsorption light, and OM mainly appears in flocculent, lamellate and globular
forms, which indicates that the kerogen type of XML-0 is type II and that
Low-pressure CO2 and N2 adsorption experiments were performed the OM could generate oil and gas during the simulation.
on a Micromeritics Tristar II 3020 SA analyser. Before analysis with CO2
and N2, the simulated samples were automatically out-gassed at 3.2. Mineral composition and content
approximately 110 ◦ C for 14 h under vacuum to remove adsorbed
moisture and volatile matter. The samples were measured at 273.15 K The mineral compositions of XML-0 and its simulated samples are
for the CO2 adsorption experiment, and N2 adsorption measurements listed in Table 3. XML-0 includes a large amount of quartz (34.9 %) and
were conducted at the temperature of liquid nitrogen (77.35 K at 101.3 clay minerals (31.5 %). The feldspar content is 12.3 %, and the calcite
kPa). Equilibrium times of 45 s and 30 s were set for CO2 and N2 content is 8.3 %. The content of pyrite and other minerals are 7.9 % and
adsorption, respectively. The relative pressure (P/P0) for the CO2 and N2 5.1 %, respectively. The clay minerals of XML-0 are dominated by mixed
adsorption experiments was in the range of 0.0001–0.03 and illite/smectite (I/S, 17.2 %) and illite (7.5 %), followed by smectite
0.00000001–0.995, respectively. The instrument software automati­ (16.2 %) and chlorite (2.7 %). Additionally, there is a small amount of
cally generated adsorption isotherms and calculated the SAs, pore vol­ kaolinite (6.7 %). The mineral compositions of the simulated samples
umes (PVs) and pore size distributions based on multiple adsorption mainly consist of clay minerals (35.8%–45.2 %) and quartz (28.3%–
theories (Mastalerz et al., 2012, 2013). 35.5 %), followed by feldspar at content that ranges from 7.7 % to 10.8
%. There is also a low content of calcite (4.4–6.2 %), pyrite (4.7–7.2 %)
2.5. Mercury intrusion porosimetry (MIP) and others (4.8–9.5 %). The dominant clay minerals are smectite
(11.5–15.6 %) and I/S (18.3–24.6 %), followed by illite (9.2–12.1 %)
The macropore structural characteristics were evaluated using an and kaolinite (4.4–7.5 %), with some chlorite (3.4–13.8 %).
automatic mercury porosimeter (AutoPore IV 9520, Micromeritics in­
strument). Prior to analysis, the samples were dried at 110 ◦ C for 24 h to 3.3. Pore structural parameter characterization
remove free water and then vacuumed. The accuracy of mercury in­
jection was 0.1 mL, the maximum working pressure of the mercury As shown in Table 4, the PVs of the micropores, mesopores and
porosimeter was 413 MPa, and the MIP pore size measurement range macropores of the 7 simulated samples are in the range of
was between 0.003 μm and 1000 μm. The porosity of the simulated 0.0029–0.0086 cm3/g (0.0059 cm3/g on average), 0.0083–0.0288 cm3/
samples was calculated based on the sample weight and the difference g (0.0184 cm3/g on average) and 0.0149–0.0285 cm3/g (0.0223 cm3/g
between the bulk and skeletal densities, which were obtained from the on average), respectively. The SAs of the micropores, mesopores and
mercury intrusion measurements. The experiment is described in detail macropores are in the range of 5.059–10.216 m2/g (with an average of
in various studies (Tian et al., 2013; Pan et al., 2015; Chen et al., 2017; 7.312 m2/g), 2.169–5.181 m2/g (with an average of 3.465 m2/g), and
Xu et al., 2021a). 0.815–4.258 m2/g (with an average of 2.213 m2/g), respectively. The
CO2 and N2 adsorption content are in the range of 0.74–2.38 cm3/g and
2.6. FE-SEM observation 2.62–23.68 cm3/g, respectively. The porosity of the simulated samples
ranges from 1.2593 % to 6.4568 %. The PV ratios of the micropores,
During the sample preparation process, Ar-ion polishing was used to mesopores and macropores are in the range of 10.54–14.78 % (12.51 %
reduce the surface roughness. The observed samples were polished using on average), 31.80–43.91 % (38.48 % on average), and 43.45–57.09 %
a Leica EM TIC 3X ion beam milling system, and a polished area of 5 × 5 (48.90 % on average), respectively. Accordingly, macropores contribute
mm was obtained. Then, the surfaces of the polished samples were the most to the PV, followed by mesopores, and micropores contribute
coated with Au to prevent electrostatic charging. To determine the the least. Macropores and mesopores are the major contributors to PV.
morphological and distribution characteristics and obtain quantitative The SA ratios of the micropores, mesopores and macropores are in the
data for the nanoscale pores, FE-SEM was applied by recording sec­ ranges of 52.54–62.90 % (57.43 % on average), 24.42–30.52 % (27.09
ondary electron imagery data at an operating current of 20 kV at 24 ◦ C % on average), and 10.13–22.64 % (15.49 % on average), respectively.
and a humidity level of 35 %. The generated SEM images provided the Accordingly, micropores contribute the most to the SA, followed by
locations of the pores and qualitative nanoscale pore structure infor­ mesopores, and macropores contribute the least. Therefore, micropores
mation for the samples. and mesopores are the main contributors to SA.

3
L. Xu et al. Marine and Petroleum Geology 132 (2021) 105233

Table 2
Geochemical parameter variations in XML-0 and the pyrolysis products.
Sample ID Simulation temperature (◦ C) TOC (%) EqRo (%) S1 (mg/g) S2 (mg/g) S1+S2 (mg/g) Tmax (◦ C) HI (mg/g TOC)

XML-0 Unheated 7.18 0.58 0.94 34.52 35.46 436 488

XML-1 300 7.12 0.72 2.85 41.26 44.11 446 456
XML-2 350 6.86 1.03 1.79 52.36 54.15 478 412
XML-3 400 5.87 1.74 2.43 21.59 24.02 521 359
XML-4 450 6.22 1.92 1.49 14.59 16.08 578 332
XML-5 500 5.68 2.23 1.32 12.51 13.83 598 259
XML-6 550 4.96 2.57 0.59 6.98 7.57 612 119

Fig. 1. Microscopic images of fluorescent and vitrinite-like macerals in XML-0. (a: amorphous, direct light; b: vitrinite-like macerals, reflected light; c: alginite,
direct light).

Table 3
Mineralogical composition of the original shale and physically simulated samples.
Sample ID Quartz (%) Feldspar (%) Calcite (%) Pyrite (%) Others (%) Clays (%) Smectite (%) I/S (%) Illite (%) Kaolinite (%) Chlorite (%)

XML-0 34.9 12.3 8.3 7.9 5.1 31.5 16.2 17.2 7.5 6.7 2.7
XML-1 35.2 10.8 6.2 7.2 4.8 35.8 15.6 18.3 9.2 7.5 3.4
XML-2 35.5 9.8 4.4 6.4 5.7 38.2 14.6 23.6 9.3 6.3 3.5
XML-3 34.2 8.3 5.2 6.3 6.6 39.4 13.5 24.6 9.3 6.1 5.5
XML-4 33.5 8.6 5.7 5.9 7.6 38.7 12.2 24.2 10.8 5.2 6.6
XML-5 30.2 7.8 5.1 5.4 8.9 42.6 11.7 23.6 11.5 4.7 9.5
XML-6 28.3 7.7 4.6 4.7 9.5 45.2 11.5 21.2 12.1 4.4 13.8

Table 4
Low-pressure gas adsorption and MIP measurement results.
Sample ID Original shale XML-1 XML-2 XML-3 XML-4 XML-5 XML-6 Average

CO2 adsorption Vmicro (cm3/g) 0.0052 0.0029 0.0046 0.0051 0.0064 0.0083 0.0086 0.0059
Smicro (m2/g) 5.718 5.059 5.569 7.112 7.556 9.956 10.216 7.312
Adsorption content (cm3/g) 0.95 0.74 0.86 1.51 2.04 2.38 2.27
N2 adsorption Vmeso (cm3/g) 0.0157 0.0083 0.0113 0.0187 0.0223 0.0288 0.0239 0.0184
Smeso (m2/g) 2.759 2.169 2.946 3.118 3.489 4.592 5.181 3.465
Adsorption content (cm3/g) 4.23 2.62 9.39 17.44 19.32 22.26 23.68
MIP Vmicro (cm3/g) 0.0186 0.0149 0.0178 0.0246 0.0263 0.0285 0.0257 0.0223
Smicro (m2/g) 1.063 0.815 1.137 1.287 2.884 4.258 4.049 2.213
Porosity (%) 1.5236 1.2593 1.6375 5.7698 6.0261 6.4568 5.5759
PV ratio (%) Micropore 13.17 11.11 13.65 10.54 11.64 12.65 14.78 12.51
Mesopore 39.75 31.80 33.53 38.64 40.55 43.90 41.07 38.56
Macropore 47.09 57.09 51.82 50.83 47.82 43.45 44.16 48.93
SA ratio (%) Micropore 59.94 62.90 57.70 61.75 54.25 52.94 52.54 57.43
Mesopore 28.92 26.97 30.52 27.07 25.05 24.42 26.64 27.09
Macropore 11.14 10.13 11.78 11.18 20.71 22.64 20.82 15.48

3.4. Pore-size distributions The micropore PV and SA distribution curves show three charac­
teristic peaks at 0.2–0.3 nm, 0.4–0.6 nm, and 0.8–0.86 nm (Fig. 3); they
Low-pressure CO2 isotherms for the 7 simulated samples all show suggest that pores with sizes of 0.2–0.3 nm, 0.4–0.6 nm, and 0.8–0.86
Type I adsorption isotherms (Fig. 2), which indicates that all the samples nm provide the majority of the PV and SA of micropores. With increasing
exhibit microporous features (Sing et al., 1985). Additionally, the simulation temperatures, the three-peak distribution of the PV variation
low-pressure CO2 isotherms show large differences, and XML-1 and curve with respect to pore size remained unchanged, but the PV and SA
XML-5 display the lowest and highest adsorption capacities at a P/P0 of increased significantly with pore sizes in the above three ranges due to
approximately 0.03, which identifies the lowest and highest micropo­ the influence of temperature and pressure. At 300–350 ◦ C (XML-0, XML-
rosities, respectively. 1), the PV and SA within the range of 0.4–0.6 nm varied little, and with

4
L. Xu et al. Marine and Petroleum Geology 132 (2021) 105233

represented by XML-2, XML-3, and XML-4 (Fig. 4b). The hysteresis loops
are similar in shape to those of type H4 and wider than those of XML-0
and XML-1, which indicates that V-shaped and wedge-shaped pores
have developed and that the pore connectivity has improved to a certain
extent compared with that of low-maturity samples. The third type is
represented by XML-5 and XML-6 (Fig. 4c). The hysteresis loops are
similar in shape to those of types H2 and H3, and this shows that the flat
pores open on all sides are developed and that the connectivity of the
pores is very good, which is beneficial for the adsorption, accumulation
and flow of shale gas.
The mesopore PV of the simulated samples is bimodal, with the main
peaks at approximately 2 nm and 40 nm. With increasing simulation
temperatures, the value of the micropore peak at approximately 2 nm
increased significantly, which was caused by the generation and
development of micropores in the simulated samples (Fig. 5a). The
Barrett-Joyner-Halenda (BJH) pore size distribution of the SA is unim­
Fig. 2. CO2 adsorption curves of the simulated samples at different simulation
odal. When the pore size is smaller, the contribution to the SA is greater,
temperatures.
and the micropores and small mesopores provide the majority of the SA
(Fig. 5b).
increasing temperatures, the PV and SA of the micropores in this range
The hysteresis loop shape of XML-1 is similar to that of XML-0. The
increased significantly, which implies that the micropore abundance in
hysteresis loop of the mercury intrusion curve is very narrow, and with
this aperture range increases with increasing simulated temperatures
only a small amount of mercury entering, there is no obvious inflection
(Chen and Xiao, 2014; Guo et al., 2017; Xu et al., 2021b). In the over­
point. There is a flat stage at pressures of 0.1–10 MPa, and the rate and
mature stage (XML-5, XML-6), the change rates of PV and SA no longer
quantity of mercury intrusion begin to increase after 10 MPa (Fig. 6a and
increased because the cracking gas of organic matter was basically
b). This result indicates that XML-0 and XML-1 are dominated by pores
depleted and could not generate micropores, and the micropores were
with sizes less than 10 nm and more than 10,000 nm, pores of other sizes
also damaged by compaction. Furthermore, some micropores were
are not developed, the pores are more dispersed, and the sorting features
gradually transformed into mesopores at this thermal maturity stage,
and connectivity are poor. Moreover, the pore development status of
which also caused no increase in the micropore volume and SA (Bernard
XML-1 is improved compared with that of XML-0 at each stage, and the
et al., 2012; Chen and Xiao, 2014; Wang and Guo, 2019; Xu et al.,
pore size and connectivity are better. The above results originated from
2021b).
two factors; one is that the organic acid generated in the process of
The N2 isotherms of the simulated samples can be divided into three
mature hydrocarbon generation diffuses in shale through effective
types as defined by IUPAC (Fig. 4) (Sing et al., 1985). The first type is
channels such as cracks and pores, which leads to a wider dissolution
represented by XML-0 and XML-1 (Fig. 4a). The hysteresis loops are
range of pores and the forming of a series of mesopores and macropores,
narrow and similar to those of type H4 to indicate that the amorphous
and the other factor is that clay minerals are constantly transformed,
pores, such as thin-necked and wide-bodied ink-bottle pores, are well
while smectite is transformed into illite, which results in an increase in
developed in the shale and that the micropores are highly developed.
the intergranular pores of clay minerals (Barth and Bjorlykke, 1993;
These types of pores are beneficial for the adsorption and accumulation
Vande Kamp, 2008; Bernard et al., 2012; Taylor and Macquaker, 2014).
of shale gas but are not conducive to gas flow. The second type is

Fig. 3. Pore volume (a) and surface area (b) distribution curves with respect to the pore diameter of the simulated shale samples.

Fig. 4. N2 adsorption curves of the simulated samples at different temperatures.

5
L. Xu et al. Marine and Petroleum Geology 132 (2021) 105233

Fig. 5. Pore volume (a) and surface area (b) distributions with the pore size derived from the N2 adsorption branch of the isotherms.

Fig. 6. Variation in the mercury intrusion curves and pore sizes of the simulated samples.

The mercury injection curves of XML-2 and XML-3 exhibit inflection shrinkage pores are generated on the edges or interiors of OM. In the
points, and the rise rate is faster in the initial mercury injection stage, process of shedding functional groups, kerogen will form pores with
while the mercury inflow at the inflection point and the maximum rough walls, and these pores are located in OM and exhibit a high
inflow of XML-3 are higher than those of XML-2 (Fig. 6c and d). This abundance and different sizes (Ungerer, 1990; Dong et al., 2015).
result suggests that within the EqRo range of 1.03%–1.74 %, the simu­ The mercury injection curves of XML-4, XML-5 and XML-6 show
lated samples still mainly generate pores with sizes less than 10 nm and different amounts of mercury intake in each pressure range, and the
larger than 10,000 nm. The pores in XML-3 are more developed, and its mercury intrusion curve has an obvious inflection point. Thereafter, the
sorting features are less obvious than those of XML-2, which results in a mercury intake and injection rate significantly accelerate, and the
narrower hysteresis loop of XML-3 than XML-2 and leads to weakened maximum value of mercury is maintained. The mercury injection curve
permeability and connectivity. The sorting features and connectivity are has a wide hysteresis loop, is slightly convex and descends slightly
better. This is because in this hydrocarbon generation peak, kerogen (Fig. 6e and f). This result implies that the abundance of macropores
continues to condense and shrink in volume, and dissolution pores or decreases sharply in the overmature stage, whereas the pores in each

6
L. Xu et al. Marine and Petroleum Geology 132 (2021) 105233

aperture range increase rapidly, the pore size distribution is very OM (Chuhan et al., 2000; Huang et al., 2009a; Xu et al., 2019, 2021b). In
discrete, and the connectivity is good. The above result was obtained addition, the increase in illite may be related to kaolinite trans­
because in the dry gas-generating stage, both a large amount of CH4 and formation. Smectite conversion to illite consumes cations in feldspar and
organic nanopores are continuously generated. However, these nano­ overcomes the kinetic barrier of feldspar dissolution. Feldspar and
scale organic pores or clay mineral intergranular pores collapse under kaolinite will also transform to illite when there are enough cations in
high-temperature and high-pressure conditions, which results in a large the solution (Huang et al., 2009a; Peltonen et al., 2009). With increasing
reduction in mesopores and macropores and a sharp increase in the temperatures, the content of kaolinite and feldspar change with a similar
pores of each size. trend, which proves the conversion of kaolinite to illite. The chlorite
content tends to increase after decreasing with increasing temperature,
4. Discussion and the inflection point is between 400 ◦ C and 450 ◦ C.
With increasing simulation temperatures, EqRo increased from 0.58
4.1. Evolution of the mineralogy and geochemical parameters % to 2.57 %, and Tmax increased from 436 ◦ C to 612 ◦ C (Fig. 8), which
signals that the simulated sample passed the peak of hydrocarbon gen­
During the simulation process, the quartz content changed slightly, eration and entered the overmature gas generation stage. The TOC
while the clay mineral content and quartz content showed a trend of content decreased from 7.18 % to 4.96 %, the HI value decreased from
alternating changes, which was due to the conversion of smectite to illite 488 mg HC/g TOC to 119 mg HC/g TOC, and S1+S2 decreased from
and the release of siliceous material during the kaolinization of feldspar. 35.46 mg HC/g to 7.57 mg HC/g; these results imply that OM generated
Furthermore, heterogeneity may be another reason that causes quartz and expelled hydrocarbons in the simulation process. Additionally,
content variation (Huang et al., 2009b; Peltonen et al., 2009; Taylor and S1+S2/TOC decreased after increasing, and this shows that the OM in
Macquaker, 2014). The feldspar content gradually decreased with the sample generated a large amount of hydrocarbons, reached the peak
increasing temperature. Luan et al. (2016) observed thermal simulation stage of hydrocarbon generation at 350 ◦ C, and then began to expel
samples and found that the dissolution of feldspar significantly hydrocarbons. With increasing maturity, shale will expel large amounts
increased, while its stability significantly decreased as the temperature of organic acids while generating and expelling hydrocarbons, and these
increased. This conclusion is also verified by the current research. The acids play a significant promoting role in mineral conversion, pore
variation characteristics of calcite and feldspar are consistent, which is generation and evolution in simulated samples (Chen and Xiao, 2014;
related to the “relay dissolution” of calcite and feldspar. The calcite Wang et al., 2020b; Xu et al., 2021b).
content increased after decreasing and then decreased slightly, which is
also mentioned by Fan et al. (2011), i.e., recrystallization occurs due to 4.2. Pore morphology variation characteristics
retrograde dissolution at approximately 400–500 ◦ C. The pyrite content
is reduced because of oxidation or reaction with the organic acids 4.2.1. Morphology variation characteristics of pores in minerals
created during the generation of hydrocarbons from kerogen (Table 3,
Fig. 7).
4.2.1.1. Generation and evolution of dissolution pores in unstable brittle
The total content of the clay minerals constantly increased, while the
minerals. The formation and evolution of dissolution pores in unstable
variations in different types of clay minerals were relatively compli­
minerals such as feldspar and calcite were obvious in the simulated
cated. The smectite content decreased, the illite content increased, and
samples. With increasing thermal temperature, along with the genera­
the I/S content decreased after increasing. This is because the smectite in
tion of hydrocarbons, the OM in shale discharges large amounts of
I/S begins to convert to illite in large quantities at 400–450 ◦ C. The
organic acids, which causes the dissolution of feldspar and calcite, and
conversion from smectite to illite is a spontaneous reaction with low
then, it releases siliceous matter, fills the edges of the quartz of dissolved
energy consumption, which is catalysed by the thermal maturation of
particles, and results in the quartz-filling phenomenon. Furthermore, the

Fig. 7. Mineral composition and content variation of the simulated samples. (a) Quartz (%), (b) feldspar (%), (c) calcite (%), (d) pyrite (%), (e) smectite (%), (f) illite
and smectite mixed layer (I/S, %), (g) illite (%), (h) kaolinite (%), and (i) chlorite (%).

7
L. Xu et al. Marine and Petroleum Geology 132 (2021) 105233

Fig. 8. Variation in the geochemical indices for the samples with increasing simulation temperatures. a: TOC (%); b: EqRo (%); c: Tmax (◦ C); d: HI (mg HC/g); and e:
S1+S2/TOC (mg HC/g).

Fig. 9. Variation characteristics of quartz and unstable mineral pores in pyrolysis products. (a: XML-2, clastic quartz edge pore filled with secondary quartz; b: energy
spectrum of Fig. 9a; c: XML-3, intergranular pores filled with secondary quartz; d: XML-3, residual feldspar and intragranular pores; e: XML-5, residual feldspar and
intragranular pores; f; energy spectrum of Fig. 9e; g: XML-4, intragranular pores of calcite; h: energy spectrum of Fig. 9g; i: XML-5, intragranular pores of calcite).

8
L. Xu et al. Marine and Petroleum Geology 132 (2021) 105233

filled secondary quartz may also be generated by the conversion process before the samples were polished. The results show that with increasing
of other unstable minerals, such as the conversion of smectite and simulation temperatures, the lamellar I/S (Fig. 11a, b, c) becomes
kaolinite to illite (Barth and Bjorlykke, 1993; Berger et al., 1997; Chu­ elongated or filamentous (Fig. 11d, e, f) and then becomes fibrous illite
han et al., 2000; Huang et al., 2009b) (Fig. 9a, b, c). The morphology of (Fig. 11 g, h, i). Meanwhile, interlayer pores are gradually generated in
feldspar changes greatly and mainly undergoes edge dissolution and the clay minerals, the pore size gradually increases, and the pore con­
intragranular dissolution (Fig. 9d, e, f). The dissolution characteristics of nectivity improves. The key stage of clay mineral pore generation in the
calcite are similar to those of feldspar, while a “dissolution window” simulated samples corresponds to 350 ◦ C (EqRo = 1.03 %). When the
appears at 450 ◦ C with retrogressive dissolution (Fig. 9g, h, i) (Vande temperature exceeds 350 ◦ C, the clay mineral porosity variation is not
Kamp, 2008; Fan et al., 2011). obvious, which indicates that the generation and evolution of clay
Additionally, the dissolution pores in the original sample do not mineral pores mainly occurs from the immature stage to the oil gener­
develop in an obvious way (Fig. 10a). Furthermore, it can also be ation peak. This finding is consistent with the natural transformation of
observed under SEM that strawberry-like pyrite aggregates are formed clay minerals under actual geological conditions (Berger et al., 1997;
by the compact accumulation of polygonal crystals in granular form, Chuhan et al., 2000; Huang et al., 2009b).
such as micro- and nanoscale crystals. Single pyrite microcrystals are
scattered in the shale in the form of cubes and octahedra, which is 4.2.2. Morphological evolution of organic pores
consistent with the microscopic characteristics of pyrite described in the The pore development degree of XML-0 is relatively low, and few
literature (Wilkin et al., 1996; Loucks et al., 2012; Liu et al., 2019b) organic pores are generated. Small amounts of oil expulsion pores with
(Fig. 10b, c, f). When the thermal simulation temperature reached small sizes can be detected in some regions in the form of pits and el­
350–400 ◦ C (XML-2 and XML-3), a large number of dissolution pores lipses with isolated distributions (Fig. 12a), which is related to the
were generated in the feldspar (Fig. 9d, e, f), and various dissolution immaturity of the original sample. When the temperature rises to 350 ◦ C
pores could also be found in the pyrite particles (Fig. 10b, c, f). These (XML-2, EqRo = 1.03 %), hydrocarbon generation on the surface of the
pores were mainly generated by the dissolution of organic acids and the OM generates gas expulsion pores with widths of tens to hundreds of
oxidizing reaction. As the temperature reaches 400–500 ◦ C, dissolution nanometres that are spherical or elliptical in shape, are distributed in a
is enhanced, and the proportion of dissolution pores also increases. The relatively isolated manner, and are not easily distinguished from oil
size of the dissolution pores inside the particles increases to reach the expulsion pores (Fig. 12b). At 450 ◦ C (XML-4, EqRo = 1.92 %), a large
macropore scale, and some pores become interconnected and inter­ number of OM pores are generated, and the morphology changes
woven to form a pore network (Fig. 10d and e). There is no obvious significantly. The pore size is generally between ten and several hundred
increase in the dissolution pore size after the simulated temperature nanometres, and the shapes are mostly pits and honeycombs. As the
exceeds 550 ◦ C. In contrast, due to high confining pressure and me­ temperature continues to rise to the peak of gas generation, the OM
chanical compaction during the deposition period, some of the disso­ begins to pyrolyze or crack, and organic pores are further generated and
lution pores are damaged by compaction (Fig. 10f). form gas expulsion pore groups (Fig. 12c). At 550 ◦ C (XML-6, EqRo =
2.57 %), the number of organic pores in the simulated sample does not
4.2.1.2. Generation and evolution of clay mineral pores. To facilitate the increase further, and the size of the previously formed organic pores
identification of different clay minerals, the distribution and evolution does not increase. In contrast, the diameter of some organic pores de­
of clay minerals during thermal maturation were carefully observed creases substantially, and the morphology is destroyed (Fig. 12d). This

Fig. 10. Morphological evolutionary characteristics of the dissolution pores in pyrolysis products. (a: XML-0, dissolution pores are not obvious; b: XML-3, dissolution
pores of pyrite; c: XML-3, dissolution pores of pyrite; d: XML-4, dissolution pores are interconnected; e: XML-5, dissolution pores are interconnected; f: XML-6,
dissolution pores of pyrite are destroyed by compaction.)

9
L. Xu et al. Marine and Petroleum Geology 132 (2021) 105233

Fig. 11. Morphological evolution of clay intraparticle pores with increasing simulation temperatures. (a: XML-0, intraparticle pores of lamellar I/S; b: XML-1,
intraparticle pores of lamellar I/S; c: energy spectrum of Fig. 11b; d: XML-3, intraparticle pores of filamentous I/S; e: XML-3, intraparticle pores of filamentous
I/S; f: energy spectrum of Fig. 11e; g: XML-5, fibrous illite; h: energy spectrum of Fig. 11g; i: XML-6, fibrous illite.)

result occurs because with further increases in mechanical compaction 4.3. Porosity evolution characteristics
during the deposition period and experimental pressure, the shale is
subjected to stronger compaction, and the previously formed organic In this research, the porosity variation characteristics of the samples
pores collapse, which makes the preservation of such pores difficult. showed a five-stage evolutionary pattern with increasing temperatures,
However, the organic pores in rigid frames filled with brittle minerals i.e., slight decrease, slight increase, sharp increase, stable increase and
are well preserved (Figs. 10f and 12e), which suggests that compaction small decrease stages (Fig. 13).
does damage organic pores and that the rigid frame of brittle minerals In the slow decrease stage (from unheated to 300 ◦ C, EqRo < 0.72
plays a role in protecting organic pores during compaction. %), the primary pores are damaged by mechanical compaction. In
Additionally, substantial internal bubble-shaped pores (Fig. 12f and addition, the OM enters the hydrocarbon generation threshold and
g) and marginal shrinkage pores (Fig. 12h and i) are observed with generates a large amount of liquid hydrocarbons, which block the pore
respect to increasing maturity. Furthermore, microscopic observation throats and reduce the porosity of the sample. In the slow increase stage
shows that internal bubble-shaped pores are significantly more abun­ (300 ◦ C < T < 350 ◦ C, 0.72 % < EqRo < 1.03 %), the porosity increases
dant than marginal shrinkage pores at a higher simulated temperature from 1.26 % to 1.64 %. The simulated sample reaches the oil generation
(Fig. 12a, b, c). Dong et al. (2015) conducted simulation experiments on peak, and the OM begins to generate a large number of organic pores,
three different types of kerogen and found evolutionary characteristics which results in an increase in pore abundance. The cracking of OM into
of organic pores similar to those in this study. Ungerer (1990) reported liquid hydrocarbons or the filling of some original framework mineral
that the internal bubble-shaped pores and marginal shrinkage pores are pores by asphalt causes complex porosity changes, and no significant
closely related to the “dissociative hydrocarbon generation” and “par­ increasing trend is observed, which is consistent with previous research
allel removal of functional group hydrocarbon generation” evolutionary results (Chen and Xiao, 2014; Mathia et al., 2016; Xu et al., 2018). In the
pathways of OM and kerogen. sharp increase stage (350 ◦ C < T < 400 ◦ C, 1.03 % < EqRo < 1.74 %),
the porosity increases from 1.64 % to 5.77 % because a large amount of
OM is consumed, and this generates a large number of organic pores

10
L. Xu et al. Marine and Petroleum Geology 132 (2021) 105233

Fig. 12. SEM images of organic pores in pyrolysis products: (a) XML-0, oil expulsion pores; the internal pores are not developed; (b) XML-4, gas expulsion pores,
nanoscale pores are generated at the contact edge between OM and minerals, and the pore size further increases; (c) XML-5, gas expulsion pore groups, some of the
organic pores are interconnected to form mesopores and macropores; (d) XML-6, organic pores are affected by extrusion; (e) the brittle mineral skeleton protects the
organic pores; (f) (g): internal bubble-shaped pores; (h) (i): marginal shrinkage pores.

Fig. 13. Porosity variation characteristics of the pyrolysis products with increasing temperatures.

during the peak stage of gas generation. Furthermore, a large amount of another reason for the significant rise in porosity. In the stable increase
previously generated liquid hydrocarbon is cracked out, and the stage (400 ◦ C < T < 500 ◦ C, 1.74 % < EqRo < 2.23 %), the porosity
bitumen that filled the pores is also consumed by cracking, which is increases from 5.77 % to 6.46 %. At this high and overmature stage, a

11
L. Xu et al. Marine and Petroleum Geology 132 (2021) 105233

large number of organic pores are generated, and the liquid hydrocar­ strong compaction, which also causes the porosity to decrease.
bons and asphaltenes in the pores tend to be cracked out, but compac­
tion reduces the porosity in this stage; therefore, the porosity shows a 4.4. Diagenetic sequence and pore evolution model
steady increasing tendency (Dong et al., 2015; Xue et al., 2015; Luan
et al., 2016; Guo et al., 2020). In the small decrease stage (500 ◦ C < T < 4.4.1. Diagenetic types
550 ◦ C, 2.23 % < EqRo < 2.57 %), the porosity decreases from 6.46 % to By combining the thermal maturity, FE-SEM and XRD analysis re­
5.58 % because OM aromatization is intensified, some pores are sults, dissolution, clay mineral transformation, hydrocarbon generation,
destroyed and pore development slows during the overmature stage. compaction and cement filling were identified in the pyrolysis products.
Additionally, previously formed pores are largely destroyed due to

Fig. 14. Diagenesis characteristics in the simulation experiment. (a: XML-2, dissolution of calcite; b: XML-2, dissolution of calcite; c: energy spectrum of Fig. 14b; d:
XML-4, dissolution of feldspar; e: energy spectrum of Fig. 14d; f: XML-5, clay mineral transformation; g: XML-5, clay mineral transformation; h: energy spectrum of
Fig. 14g; i: XML-2, OM is compacted; j: XML-2, clay minerals are compacted; k: XML-4, recrystallization filling; l: XML-5, recrystallization filling).

12
L. Xu et al. Marine and Petroleum Geology 132 (2021) 105233

4.4.1.1. Dissolution characteristics. Dissolution occurs throughout the the pressure of the blue arrow, rigid mineral particles will be crushed to
entire diagenesis stage. It is aggravated with increasing temperature and form fractures, and the OM will also be deformed (Aplin et al., 2006;
is most obvious at 350 ◦ C (Figs. 1, 9 and 10, 14a, b, c). Hydrocarbon Wang et al., 2020c; Chen et al., 2021) (Figs. 12d and 14i, j).
generation can result in pore water becoming acidic in simulated sam­
ples (Barth and Bjorlykke, 1993; Slatt and O’Brien, 2011; Lu et al., 4.4.1.5. Cementation and filling. The authigenic minerals produced in
2015). Acidic water in pores reacts with unstable minerals such as diagenesis are filled into intergranular pores or dissolution pores, which
feldspar (R1), calcite (R2) and pyrite to form secondary pores such as results in the shrinkage of the pore space. It can also be seen that
intergranular, intragranular, intercrystalline, and intracrystalline dis­ authigenic minerals growing in pore throats block the seepage channels
solved pores, as well as dissolution cavities (mould pores) after complete (Fig. 14k and l). Authigenic minerals are formed in a variety of ways that
dissolution, which shows the characteristics of point dissolution, linear may include the recrystallization and cementation of the chemical
dissolution and surface dissolution (Fig. 14d and e) (Huang et al., 2009b; components released by mineral transformation. These authigenic
Guo and Mao, 2019; Xu et al., 2021b). minerals generally occupy the pore space (Vande Kamp, 2008; Peltonen
et al., 2009; Liu et al., 2019a; Zhang et al., 2019). Cementation and
2KAlSi3O8(k-feldspar) + 2H+ + H2O = Al2Si2O5(OH)4(kaolinite) +
filling are developed in the middle and late diagenesis stages, and they
4SiO2(siliceous) +2K+ – (R1)
are most developed in the middle diagenesis stage.
CaCO3 (calcite) + H+ = Ca2+ + HCO−3
————————————————— (R2) 4.4.2. Pore structure evolution characteristics
As the simulated temperature increases, the CO2 adsorption content
decreases at the initial stage and then decreases after increasing to the
maximum value at 500 ◦ C, while the N2 adsorption content decreases
4.4.1.2. Transformation of clay minerals. The clay minerals in the first and then decreases (Fig. 15). Both the CO2 and N2 adsorption
simulated samples are mainly composed of smectite, I/S and illite, fol­ content of the samples decrease to minimum values of 0.74 cm3/g and
lowed by kaolinite and chlorite. Interlayer water is lost from smectite in 2.62 cm3/g at 300 ◦ C, respectively, which indicates that the micropores
alkaline medium conditions and forms I/S, which further converts to and mesopores of XML-1 are not well developed. The CO2 adsorption
illite with a reduced formation ratio, which indicates a relatively high content of the samples rises to the maximum value at 500 ◦ C, and this
degree of illitization. With increasing temperatures, I/S converts from a shows that the micropores of XML-5 are the most developed. As the
flake shape to filamentous and flocculent yarn-like sheets (Figs. 11 and simulated temperature increases to 550 ◦ C, the CO2 adsorption amount
14f, g, h). Clay mineral conversion develops in the early and middle of the samples gradually decreases. This decrease occurs because the OM
diagenetic stages. Conversion from smectite to illite mainly appears has aromatized, cracking has ended, and the generated residual
during the middle stage of diagenesis. This conversion is a low-energy asphaltenes block the pores in the overmature stage. Additionally, the
consumption reaction, and kaolinite can be converted to illite with the compaction caused by the tremendous experimental pressure will also
participation of potash feldspar and K+ (Chuhan et al., 2000; Huang lead to a reduction in micropores. The increased N2 adsorption content is
et al., 2009a; Guo and Mao, 2019; Xu et al., 2019) according to the mainly due to the continuous transformation of micropores into meso­
following reaction: pores. Although the pore abundance is reduced by asphaltene plugging
and the compaction that results from the high experimental pressure,
Smectite + K+ + Al3+ → illite + Na+ + Ca2+ + Fe3+ + Mg2+ + SiO2 + H2O
pore abundance increases have a more obvious effect because of the
———————————————— (R3)
continuous transformation of micropores to mesopores.
KAlSi3O8 + 2K+ + Al2Si2O5(OH)4 (kaolinite) = 3KAl3Si3O10(OH)2 (illite) + The PV and SA variation characteristics of the micro-, meso- and
2SiO2 (siliceous) + 4 H2O + 2H+ macropores of the simulated samples are shown in Fig. 16. With
———————————————— (R4) increasing temperature, the PV of the micro-, meso- and macropores first
decreases to the minimum value at 300 ◦ C and then decreases after
increasing to the maximum value at 500 ◦ C (Fig. 16a, b, c). The SA of the
micro-, meso- and macropores first decreases to the minimum value as
4.4.1.3. Hydrocarbon generation. The thermal maturation of OM is one the temperature reaches 300 ◦ C and then keeps rising until the end of the
of the internal driving forces of shale diagenesis. First, with hydrocarbon simulation, except for the macropores, which reach their maximum
generation from OM and kerogen, organic acids can also be discharged, value at 500 ◦ C and then decrease again (Fig. 16a, b, c).
and decarboxylation can release CO2 to form carbonic acid, both of As the temperature increases to 300 ◦ C, the PV and SA of the micro-,
which could make the pore water a weakly acidic medium and cause meso- and macropores decline to their minimum values because me­
dissolution. K+, Al3+, and Mg2+ are released after the consumption of H+ chanical compaction leads to damage to the primary pores in the resi­
to convert the medium to alkaline conditions, which further promotes dues. In addition, the OM enters the threshold for hydrocarbon
the dissolution of quartz (Barth and Bjorlykke, 1993; Vande Kamp, generation, and a large amount of liquid hydrocarbons is generated,
2008; Lewan et al., 2014) (Fig. 9a, b, c, d, e, f), illitization of smectite which clogs the pore throats in the simulated samples and is another
(Fig. 11) and recrystallization of calcite (Fig. 14 k, l). In addition, with reason for this decline (Bernard et al., 2012; Liu et al., 2019a; Wang
hydrocarbon generation, substantial bubble-like pores inside and et al., 2020b; Chen et al., 2021). As the simulation temperature increases
shrinkage cracks beside the OM can be generated, and the generated from 300 ◦ C to 500 ◦ C, the PV and SA of the micropores, mesopores and
organic nanopores expand, connect and adjust continuously to form macropores increase, which results from the pore space in the samples
pores with larger sizes as the pyrolysis temperature increases (Hu, 2013; increasing due to the consumption of OM and the generation of various
Romero-Sarmiento et al., 2014; Tang et al., 2015) (Fig. 12). organic pores. Furthermore, large quantities of previously generated
liquid hydrocarbons are cracked, and a large amount of bitumen that
4.4.1.4. Compaction. Compaction mainly occurs in the early diagenetic fills the pores is also consumed by cracking to free the occupied pore
stage, which is the main reason for the large reduction in primary pores space, which increases the nanoscale pore space (Chen and Xiao, 2014;
and densification of the simulated samples. Under the compaction of the Xue et al., 2015; Xu et al., 2021b). As the temperature rises from 500 ◦ C
pressure shown by the blue arrow in Figs. 12d and 14i and j, the mineral to 550 ◦ C, the decrease in the PV and macropore SA is caused by the
lattice will collapse rapidly, rigid mineral particles will break, clay aromatization of OM and the residual asphaltene blocking the pores. The
minerals will undergo ductile deformation, bending and orientation, other cause is that the previously formed pores are largely destroyed
and the pores will be deformed. Under the effect of compression under

13
L. Xu et al. Marine and Petroleum Geology 132 (2021) 105233

Fig. 15. Variations in the maximum CO2 (a) and N2 (b) adsorption content during the simulation experiment.

Fig. 16. Pore volume and surface area variations of micropores (a, a’), mesopores (b, b’) and macropores (c, c’) during the thermal simulation experiment.

because of the strong experimental pressure. The decrease in the simulation products to the largest extent, the six simulated samples
micropore and mesopore SA is caused by the transformation of macro­ originated from one large shale sample, which is inevitably heteroge­
pores to mesopores and micropores and the increase in the number of neous in organic matter and mineral composition. Therefore, hetero­
micropores and mesopores (Chen and Xiao, 2014; Wang and Guo, 2019; geneity may be another reason for the PV and SA variation in the
Xu et al., 2019, 2021b). In addition, although the present hydrous py­ simulated samples.
rolysis experiments minimized the impact of heterogeneity on the

Fig. 17. Comprehensive diagram of the diagenetic evolution sequence and pore evolution model based on the hydrous pyrolysis experiment.

14
L. Xu et al. Marine and Petroleum Geology 132 (2021) 105233

4.4.3. Diagenetic and pore evolution model 5. Conclusion

According to the variations in maturity, pore structure, minerals and
hydrocarbon production, we established a comprehensive diagram of (1) The simulation temperature increase plays an important role in
the diagenetic evolution sequence and pore evolution model, and the promoting the evolution of framework mineral pores, intensifies
model can be divided into four stages (Fig. 17). the development of dissolution pores generated by unstable
In the early diagenetic stage (EqRo<0.6 %), the OM is in the brittle minerals, promotes the transformation of clay minerals,
immature stage, hydrocarbon generation is just beginning, organic pores and accelerates the formation and development of clay mineral
are few, the dissolution and conversion of clay minerals are weak, and pores, and two types of organic pores are generated and evolve,
the diagenesis intensity is low (Aplin et al., 2006; Milliken et al., 2012; namely, internal pores and edge shrinkage pores.
Taylor and Macquaker, 2014). The primary pores are largely destroyed, (2) The porosity evolves in five characteristic stages, and the PV and
and the clay minerals collapse and are compacted, with some primary SA of the micropores vary with the pore size and show three
pores (rigid intergranular pores of minerals) and clay mineral pores peaks. The PV and SA of the mesopores are bimodal and unim­
remaining (Fig. 12d and e, 14f, g). In this neutral or alkaline environ­ odal, respectively, and when the pore size is smaller, the contri­
ment, most of the smectite has been converted into illite or chlorite. bution of the pores to SA is greater. As the heating temperature
In the middle diagenetic stage (stage A) (0.6 %<EqRo<1.0 %), the increases, the abundance of macropores and macropores de­
diagenesis mainly involves dissolution, the transformation of clay min­ creases sharply, the pores in each size range increase rapidly, the
erals, and the thermal maturity of OM. With heating and clay mineral connectivity is gradually improved, and the pore size distribution
catalysis, kerogen generates a large amount of oil. With increasing is gradually dispersed.
temperatures, a large amount of I/S is transformed into illite (Aplin (3) Mesopores and macropores contribute the most to the PV, while
et al., 2006; Vande Kamp, 2008; Peltonen et al., 2009; Lewan et al., micropores and mesopores provide the majority of the SA. The
2014). Hydrocarbon generation leads to a change in the local medium PVs of the nanoscale pores first decrease to minimum values at
environment, and the dissolution of unstable minerals is substantially 300 ◦ C, then increase until reaching the maximum values at
enhanced, which promotes diagenesis. 500 ◦ C, and decrease thereafter. The SAs of the nanoscale pores
In the middle diagenetic stage (stage B) (1.0 %<EqRo<2.0 %), due to first decrease to the minimum values as the temperature reaches
increasing temperatures, residual kerogen and previously generated 300 ◦ C and then continuously increase until the end of the
liquid hydrocarbons are cracked to wet gas, I/S is generated in large simulation, except for the SA of the macropores, which reaches its
quantities and transformed into illite, and the clay mineral pores are maximum value at 500 ◦ C and then decreases.
mainly lamellar. The intensity of diagenesis reaches its peak at this (4) Hydrocarbon generation, pore evolution and diagenesis have
stage, and phenomena such as recrystallization occur (Fig. 14 h, i). The synergistic effects. Diagenesis mainly involves dissolution, the
residual primary pores are destroyed by compaction and filling, a large transformation of clay minerals, the thermal maturation of OM,
number of nanoscale organic pores are generated by the thermal compaction, cementation and filling. According to the classifi­
maturation of OM, and acidic fluid is expelled to promote the formation cation of the diagenesis stages, a comprehensive diagenetic evo­
of dissolution pores. Micro- and nanoscale pores are enlarged, adjusted lution sequence and pore evolution model were established.
and connected to form secondary pores with larger sizes (Bernard et al.,
2012; Chen and Xiao, 2014; Guo and Mao, 2019; Xu et al., 2021b). Declaration of competing interest
In the late diagenetic stage (EqRo>2.0 %), diagenesis mainly in­
volves the thermal maturity of OM and compaction. OM generates more The authors declare that they have no known competing financial
hydrocarbons (secondary hydrocarbon generation), residual kerogen is interests or personal relationships that could have appeared to influence
aromatized under high temperatures and generates CH4, and the nano­ the work reported in this paper.
scale pores are irregular in shape and continue to expand and connect
(Chen and Xiao, 2014; Guo and Mao, 2019; Wang et al., 2020b; Xu et al., Acknowledgements
2021b). OM forms internal pores and edge shrinkage pores (Ungerer,
1990; Dong et al., 2015). I/S is mainly dominated by illite in the form of This project was funded by the National Key Research and Devel­
filamentous and flocculent shapes, and a large number of round calcite opment Program of China (Grant No. 2019YFA0708501) and the Nat­
discs are generated (Figs. 11 and 14c, d). Dissolution is greatly weak­ ural Science Foundation of Hebei Province for Youth (Grant No.
ened with the decrease in fluid, and the dissolution pores are reduced by D2019403189). This research was also funded by the Key Research and
cemented fillings. At the end of hydrocarbon generation, under the Development Project of Hebei Province (Grant No. 19224205D) and the
conditions of high temperature and high pressure, compaction leads to Natural Science Foundation of Hebei Province (Grant No.
the closure of the pores in particles, and the organic pores begin to D2020403019). The authors would like to thank Dr. Ji Chen and Xu Bi
shrink because of the end of hydrocarbon generation through cracking. for helping with the ramped thermal simulation experiments and the
Our findings provide a conceptual model for the quantitative pre­ earlier version of this manuscript. Furthermore, the authors are indebted
diction of hydrocarbon generation, diagenesis and nanoscale pore var­ to the editor and anonymous reviewers for their insightful comments
iations according to thermal evolution, and we introduce a theoretical and suggestions, which significantly improved the manuscript.
model for the identification of prospective shale gas zones due to ther­
mal maturity mapping. Additionally, various factors, such as the types, References
content and occurrence of OM in shale and variations in the mineralogy
characteristics, geothermal fluids and depositional environment, may Aplin, A.C., Matenaar, I.F., McCarty, D.K., van der Pluijm, B.A., 2006. Influence of
mechanical compaction and clay mineral diagenesis on the microfabric and pore-
further evolve with increasing maturity. A pyrolysis thermal simulation
scale properties of deep-water Gulf of Mexico mudstones [J]. Clay Clay Miner. 54
can be conducted by choosing samples with different mineral charac­ (4), 500–514.
teristics and TOC content with various kerogen types from different Barth, T., Bjorlykke, K., 1993. Organic acids from source rock maturation: generation
deposition environments to improve the applicability and feasibility of potentials, transport mechanisms and relevance for mineral diagenesis [J]. Appl.
Geochem. 8 (4), 325–337.
the simulation results. Berger, G., Lacharpagne, J., Velde, B., Beaufort, D., Lanson, B., 1997. Kinetic constraints
on illitization reactions and the effects of organic diagenesis in sandstone/shale
sequences [J]. Appl. Geochem. 22 (1), 23–35.
Bernard, S., Wirth, R., Schreiber, A., Schulz, H.M., Horsfield, B., 2012. formation of
nanoporous pyrobitumen residues during maturation of the barnett shale (fort worth
basin) [J]. Int. J. Coal Geol. 103, 3–11.

15
L. Xu et al. Marine and Petroleum Geology 132 (2021) 105233

Chalmers, G.R.L., Bustin, R.M., 2008. Lower Cretaceous gas shales in northeastern British Lu, J., Ruppel, S.C., Rowe, H.D., 2015. Organic matter pores and oil generation in the
Columbia, Part I: geological controls on methane sorption capacity [J]. Bull. Can. Tuscaloosa marine shale [J]. AAPG (Am. Assoc. Pet. Geol.) Bull. 99 (2), 333–357.
Petrol. Geol. 56 (1), 1–21. Mastalerz, M., He, L., Melnichenko, Y.B., Rupp, J.A., 2012. Porosity of coal and shale:
Chen, J., Xiao, X.M., 2014. Evolution of nanoporosity in organic-rich shales during insights from gas adsorption and SANS/USANS techniques [J]. Energy Fuels 26 (8),
thermal maturation [J]. Fuel 129, 173–181. 5109–5120.
Chen, L., Jiang, Z.X., Liu, K.Y., Tan, J.Q., Gao, F.L., Wang, P.F., 2017. Pore structure Mastalerz, M., Schimmelmann, A., Drobniak, A., Chen, Y.Y., 2013. Porosity of Devonian
characterization for organic-rich Lower Silurian shale in the Upper Yangtze and Mississippian New Albany Shale across a maturation gradient: insights from
Platform, South China: a possible mechanism for pore development [J]. J. Nat. Gas organic petrology, gas adsorption, and mercury intrusion [J]. AAPG (Am. Assoc. Pet.
Sci. Eng. 46, 1–15. Geol.) Bull. 97 (10), 1621–1643.
Chen, Q., Yan, X.B., Liu, C.Y., Wei, X.L., Chen, Z., Qin, W.J., Hong, T.Y., 2021. Mathia, E.J., Bowen, L., Thomas, K.M., Aplin, A.C., 2016. Evolution of porosity and pore
Controlling effect of compaction upon organic matter pore development in shale: a types in organic-rich, calcareous, Lower Toarcian Posidonia Shale [J]. Mar. Petrol.
case study on the Lower Paleozoic in southeastern Sichuan Basin and its periphery Geol. 75, 117–139.
[J]. Oil Gas Geol. 42 (1), 76–85. Milliken, K.L., Esch, W.L., Reed, R.M., Zhang, T.W., 2012. Grain assemblages and strong
Chuhan, F.A., Bj Rlykke, K., Lowrey, C., 2000. The role of provenance in illitization of diagenetic overprinting in siliceous mudrocks, barnett shale (mississippian), fort
deeply buried reservoir sandstones from Haltenbanken and north Viking Graben, worth basin, Texas [J]. AAPG (Am. Assoc. Pet. Geol.) Bull. 96 (8), 1553–1578.
offshore Norway [J]. Mar. Petrol. Geol. 17 (6), 673–689. Pan, L., Xiao, X.M., Tian, H., Zhou, Q., Chen, J., Li, T.F., Wei, Q., 2015. A preliminary
Cui, J.W., Zhu, R.K., Cui, J.G., 2013. Relationship of porous evolution and residual study on the characterization and controlling factors of porosity and pore structure
hydrocarbon: evidence from modeling experiment with geological constrain [J]. of the Permian shales in Lower Yangtze region, Eastern China [J]. Int. J. Coal Geol.
Acta Geol. Sin. 87 (5), 730–736. 146, 68–78.
Curtis, M.E., Cardott, B.J., Sondergeld, C.H., Rai, C.S., 2012. Development of organic Peltonen, C., Marcussen, Ø., Bjørlykke, K., Jahren, J., 2009. Clay mineral diagenesis and
porosity in the Woodford Shale with increasing thermal maturity [J]. Int. J. Coal quartz cementation in mudstones: the effects of smectite to illite reaction on rock
Geol. 103, 26–31. properties [J]. Mar. Petrol. Geol. 26 (6), 887–898.
Dong, C.M., Ma, C.F., Luan, G.Q., Lin, C.Y., Zhang, X.G., Ren, L.H., 2015. Pyrolysis Qiu, Z., Liu, B., Dong, D.Z., Lu, B., Yawar, Z., Chen, Z.H., Schieber, J., 2020. Silica
simulation experiment and diagenesis evolution pattern of shale [J]. Acta diagenesis in the lower paleozoic wufeng and longmaxi formations in the sichuan
Sedimentol. Sin. 33 (5), 1053–1061. basin, south China: implications for reservoir properties and paleoproductivity [J].
Fan, M., He, Z.L., Li, Z.M., Yu, L.J., Zhang, W.T., Liu, W.X., Jiang, X.Q., 2011. Dissolution Mar. Petrol. Geol. 121, 104594.
window of carbonate rocks and its geological significance [J]. Oil Gas Geol. 32 (4), Romero-Sarmiento, M., Rouzaud, J.N., Bernard, S., Deldicque, D., Thomas, M., Littke, R.,
499–505. 2014. Evolution of Barnett Shale organic carbon structure and nanostructure with
Guo, H.J., Jia, W.L., He, R.L., Yu, C.L., Song, J.Z., Peng, P.A., 2020. Distinct evolution increasing maturation [J]. Org. Geochem. 71, 7–16.
trends of nanometer-scale pores displayed by the pyrolysis of organic matter-rich Sing, K., Everett, D., Haul, R., 1985. Reporting physisorption data for gas/sold systems
lacustrine shales: implications for the pore development mechanisms [J]. Mar. with special reference to the determination of surface area and porosity [J]. Pure
Petrol. Geol. 121, 104622. Appl. Chem. 54 (4), 603–619.
Guo, H.J., Jia, W.L., Peng, P.A., Zeng, J., He, R.L., 2017. Evolution of organic matter and Slatt, R.M., O’Brien, N.R., 2011. Pore types in the Barnett and Woodford gas shales:
nanometer-scale pores in an artificially matured shale undergoing two distinct types contribution to understanding gas storage and migration pathways in fine-grained
of pyrolysis: a study of the Yanchang shale with type II kerogen [J]. Org. Geochem. rocks [J]. AAPG (Am. Assoc. Pet. Geol.) Bull. 95 (12), 2017–2030.
105, 56–66. Sun, L.N., Tuo, J.C., Zhang, M.F., Wu, C.J., Chai, S.Q., 2019. Pore structures and fractal
Guo, S.B., Mao, W.J., 2019. Division of diagenesis and pore evolution of a permian characteristics of nano-pores in shale of Lucaogou formation from Junggar Basin
shanxi shale in the ordos basin, China [J]. J. Petrol. Sci. Eng. 182, 1–9. during water pressure-controlled artificial pyrolysis [J]. J. Anal. Appl. Pyrol. 140,
Han, H., Guo, C., Zhong, N.N., Pang, P., Chen, S.J., Lu, J.G., Gao, Y., 2019. Pore structure 404–412.
evolution of lacustrine shales containing Type I organic matter from the Upper Sweeny, J., Burnham, A.K., 1990. Evolution of a simple model of vitrinite reflactance
Cretaceous Qingshankou Formation, Songliao Basin, China: a study of artificial based on chemical kinetics [J]. AAPG (Am. Assoc. Pet. Geol.) Bull. 74 (10),
samples from hydrous pyrolysis experiments [J]. Mar. Petrol. Geol. 104, 375–388. 1559–1570.
Huang, K.K., Huang, S.J., Tong, H.P., Li, L.H., 2009a. Thermodynamic calculation of Tang, X., Zhang, J.C., Jin, Z.J., Xiong, J.Y., Lin, L.M., Yu, Y.X., Han, S.B., 2015.
feldspar dissolution and its significance on research of clastic reservoir. Geol. Bull. Experimental investigation of thermal maturation on shale reservoir properties from
China 28 (4), 474–482. hydrous pyrolysis of Chang 7 shale, Ordos Basin [J]. Mar. Petrol. Geol. 64, 165–172.
Huang, S.J., Huang, K.K., Feng, W.L., Tong, H.P., Liu, L.H., Zhang, X.H., 2009b. Mass Taylor, K.G., Macquaker, J.H.S., 2014. Diagenetic alterations in a silt- and clay-rich
exchanges among feldspar, kaolinite and illite and their influences on secondary mudstone succession: an example from the Upper Cretaceous Mancos Shale of Utah,
porosity formation in clastic diagenesis: a case study on the Upper Paleozoic, Ordos USA [J]. Clay Miner. 49 (2), 213–227.
Basin and Xujiahe formation, Western Sichuan depression [J]. Geochimica 38 (5), Tian, H., Pan, L., Xiao, X.M., Wilkins, R.W.T., 2013. A preliminary study on the pore
498–506. characterization of Lower Silurian black shales in the Chuandong Thrust Fold Belt,
Hu, H.Y., 2013. Porosity evolution of the organic-rich shale with thermal maturity southwestern China using low pressure N2 adsorption and FE-SEM methods [J]. Mar.
increasing. Acta Pet. Sin. 34 (5), 820–825. Petrol. Geol. 48, 8–19.
Jarvie, D.M., Hill, R.J., Ruble, T.E., Pollastro, R.M., 2007. Unconventional shale-gas Ungerer, P., 1990. State of the art of research in kinetic modelling of oil formation and
systems: the Mississippian Barnett Shale of north-central Texas as one model for expulsion [J]. Org. Geochem. 16 (1), 719–833.
thermogenic shale-gas assessment [J]. AAPG (Am. Assoc. Pet. Geol.) Bull. 91 (4), Vande Kamp, P., 2008. Smectite-illite-muscovite transformations, quartz dissolution, and
475–499. silica release in shales[J]. Clay and Clay minerals 56 (1), 66–81.
Lewan, M.D., Dolan, M.P., Curtis, J.B., 2014. Effects of smectite on the oil-expulsion Wang, F.T., Guo, S.B., 2019. Influential factors and model of shale pore evolution: a case
efficiency of the Kreyenhagen Shale, San Joaquin Basin, California, based on study of a continental shale from the Ordos Basin [J]. Mar. Petrol. Geol. 102,
hydrous-pyrolysis experiments [J]. AAPG (Am. Assoc. Pet. Geol.) Bull. 98 (6), 271–282.
1091–1109. Wang, P.F., Zhang, C., Li, X., Zhang, K., Yuan, Y., Zang, X.P., Cui, W.J., Liu, S.Y., Jiang, Z.
Liang, C., Cao, Y.C., Liu, K.Y., Jiang, Z.X., Wu, J., Hao, F., 2018. Diagenetic variation at X., 2020a. Organic matter pores structure and evolution in shales based on the he ion
the lamina scale in lacustrine organic-rich shales: implications for hydrocarbon microscopy (HIM): a case study from the Triassic Yanchang, Lower Silurian
migration and accumulation [J]. Geochem. Cosmochim. Acta 229, 112–128. Longmaxi and Lower Cambrian Niutitang shales in China [J]. J. Nat. Gas Sci. Eng.
Liu, H.M., Zhang, S., Song, G.Q., Wang, X.J., Teng, J.B., Wang, M., Bao, Y.S., Yao, S.P., 84, 103682.
Wang, W.Q., Zhang, S.P., Hu, Q.H., Fang, Z.W., 2019a. Effect of shale diagenesis on Wang, Y., Liu, L.F., Hu, Q.H., Hao, L.W., Wang, X.M., Sheng, Y., 2020b. Nanoscale pore
pores and storage capacity in the paleogene shahejie formation, dongying network evolution of Xiamaling marine shale during organic matter maturation by
depression, bohai bay basin, east China [J]. Mar. Petrol. Geol. 103, 738–752. hydrous pyrolysis [J]. Energy Fuels 34 (2), 1548–1563.
Liu, Y., Zhong, N.N., Tian, Y.J., Mu, G.Y., 2011. The oldest oil accumulation in China: Wang, Y.R., Nie, H.K., Hu, Z.Q., Liu, G.X., Xi, B.B., Liu, W.X., 2020c. Controlling effect of
meso-proterozoic Xiamaling formation bituminous sandstone reservoirs [J]. Petrol. pressure evolution on shale gas reservoir: a case study of the Wufeng-Longmaxi
Explor. Dev. 38 (4), 503–512, 2011. Formation in the Sichuan Basin[J]. Nat. Gas. Ind. 40 (10), 1–11.
Liu, Z.Y., Chen, D.X., Zhang, J.C., Lü, X.X., Wang, Z.Y., Liao, W.H., Shi, X.B., Tang, J., Waples, D.W., 1980. Time and temperature in petroleum formation: application of
Xie, G.J., 2019b. Pyrite morphology as an indicator of paleoredox conditions and Lopatin’s method to petroleum exploration. AAPG (Am. Assoc. Pet. Geol.) Bull. 64
shale gas content of the longmaxi and wufeng shales in the middle yangtze area, (6), 916–926.
south China [J]. Minerals 9 (7), 1–18. Wei, S.L., He, S., Pan, Z.J., Zhai, G.Y., Dong, T., Guo, X.W., Yang, R., Han, Y.J., Yang, W.,
Li, X.S., Zhu, H.J., Zhang, K.X., Li, Z., Yu, Y.X., Feng, X.Q., Wang, Z.X., 2021. Pore 2020. Characteristics and evolution of pyrobitumen-hosted pores of the overmature
characteristics and pore structure deformation evolution of ductile deformed shales Lower Cambrian Shuijingtuo Shale in the south of Huangling anticline, Yichang area,
in the Wufeng-Longmaxi Formation, southern China [J]. Mar. Petrol. Geol. 127, China: evidence from FE-SEM petrography [J]. Mar. Petrol. Geol. 116, 104303.
104992. Wilkin, R.T., Barnes, H.L., Brantley, S.L., 1996. The size distribution of framboidal pyrite
Lopatin, N.V., 1971. Temperature and geologic time as factors in coalification. Akad in modern sediments: an indicator of redox conditions [J]. Geochem. Cosmochim.
Nauk SSSR Izv Ser Geol 3, 95–106. Acta 60 (20), 3897–3912.
Loucks, R.G., Reed, R.M., Ruppel, S.C., Hammes, U., 2012. Spectrum of pore types and Wu, S.T., Zhu, R.K., Cui, J.G., Cui, J.W., Bai, B., Zhang, X.X., Jin, X., Zhu, D.S., You, J.C.,
networks in mudrocks and a descriptive classification for matrix-related mudrock Li, X.H., 2015. Characteristics of lacustrine shale porosity evolution, triassic chang 7
pores [J]. AAPG (Am. Assoc. Pet. Geol.) Bull. 96 (6), 1071–1098. member, ordos basin, NW China [J]. Petrol. Explor. Dev. 42 (2), 167–176.
Luan, G.Q., Dong, C.M., Ma, C.F., Lin, C.Y., Zhang, J.Y., X F, L., Lasbela, U., 2016. Xue, L.H., Yang, W., Zhong, J.A., Xu, Y., Chen, G.J., 2015. Porous evolution of the
Pyrolysis simulation experiment study on diagenesis and evolution of organic-rich organic-rich shale from simulation experiment with geological constrain, samples
shale [J]. Acta Sedimentol. Sin. 34, 1208–1216, 06. from Yanchang formation in Ordos Basin. Acta Geol. Sin. 89 (5), 970–978.

16
L. Xu et al. Marine and Petroleum Geology 132 (2021) 105233

Xu, L.W., Liu, L.F., Jiang, Z.X., Chen, L., Wang, Y., 2018. Methane adsorption in the low- artificial thermal maturation: a pyrolysis study of the Mesoproterozoic Xiamaling
middle-matured neoproterozoic xiamaling marine shale in Zhangjiakou, Hebei [J]. marine shale with type II kerogen from Zhangjiakou, Hebei, China [J]. Energy
Aust. J. Earth Sci. 65 (5), 691–710. Explor. Exploit. 37 (1), 493–518.
Xu, L.W., Yang, K.J., Wei, H., Liu, L.F., Li, X., Chen, L., Xu, T., Wang, X.M., 2021a. Full- Yu, B.S., 2013. Classification and characterization of gas shale pore system [J]. Earth Sci.
scale pore structure characteristics and the main controlling factors of Front. 20 (4), 211–220.
Mesoproterozoic Xiamaling shale in Zhangjiakou, Hebei, China [J]. Nanomaterials Zhang, S., Liu, H.M., Wang, M., Liu, X.J., Liu, H.L., Bao, Y.S., Wang, W.Q., Li, R.Z.,
11 (2), 1–34. Luo, X., Fang, Z.W., 2019. Shale pore characteristics of Shahejie Formation:
Xu, L.W., Yang, K.J., Zhang, L.F., Liu, L.F., Jiang, Z.X., Li, X., 2021b. Organic-induced implication for pore evolution of shale oil reservoirs in Dongying sag, north China
nanoscale pore structure and adsorption capacity variability during artificial thermal [J]. Petroleum Research 4 (2), 113–124.
maturation: pyrolysis study of the Mesoproterozoic Xiamaling marine shale from Zhao, J.H., Jin, Z.J., Hu, Q.H., Liu, K.Y., Liu, G.X., Gao, B., Liu, Z.B., Zhang, Y.Y.,
Zhangjiakou, Hebei, China [J]. J. Petrol. Sci. Eng. 202, 108502. Wang, R.Y., 2019. Geological controls on the accumulation of shale gas: a case study
Xu, L.W., Wang, Y., Liu, L.F., Chen, L., Chen, J., 2019. Evolution characteristics and of the early Cambrian shale in the Upper Yangtze area [J]. Mar. Petrol. Geol. 107,
model of nanopore structure and adsorption capacity in organic-rich shale during 423–437.

17