Pore characterization of isolated organic matter from high matured gas shale reservoir

Accepted Manuscript Pore characterization of isolated organic matter from high matured gas shale reservoir

Wenming Ji, Yan Song, Zhenhua Rui, Mianmo Meng, Hexin Huang PII: DOI: Reference:

S0166-5162(16)30481-5 doi: 10.1016/j.coal.2017.03.005 COGEL 2803

To appear in:

International Journal of Coal Geology

Received date: Revised date: Accepted date:

8 September 2016 12 March 2017 13 March 2017

Please cite this article as: Wenming Ji, Yan Song, Zhenhua Rui, Mianmo Meng, Hexin Huang , Pore characterization of isolated organic matter from high matured gas shale reservoir. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Cogel(2017), doi: 10.1016/j.coal.2017.03.005

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Pore characterization of isolated organic matter from high matured gas shale reservoir Wenming Jia,b, Yan Songa,c,*, Zhenhua Ruid,*, Mianmo Menga, Hexin Huange a

State Key Laboratory of Petroleum Resources and Prospecting, China University of Petroleum, Beijing 102249, China

b

Bureau of Economic Geology, Jackson School of Geosciences, The University of Texas at Austin, Austin, TX 78713, USA

c

Research Institute of Petroleum Exploration and Development, Beijing 100083, China

d

Independent Project Analysis, Inc., Ashburn, VA 20176, USA

e

Department of Geology, Northwest University, Xi’an 710069, China * Corresponding author: Yan Song: Tel. +86 10 89739068; E-mail address: [email protected]

PT

Zhenhua Rui: Tel. +1 5715357633; E-mail address: [email protected]

RI

Keywords

AC

CE

PT E

D

MA

NU

SC

Shale gas; Pore structure; Pore size distribution; Isolated organic matter; Lower Silurian shale

ACCEPTED MANUSCRIPT ABSTRACT Knowledge of pore structure of isolated organic matter (OM) provides guidance for better understanding the origin, distribution and characterization of pore system in gas shale reservoir, and can be used to better comprehend the storage and transport mechanism of natural gas in shale reservoir. Pore structure characteristics of isolated organic matters and its corresponding bulk shales were investigated for five high matured marine shale samples from the Lower Silurian Longmaxi formation in south China using X-ray diffraction (XRD), total organic carbon content (TOC) tests, field emission scanning

PT

electron microscope (FE-SEM) observation and ultra-low pressure nitrogen gas physisorption. The results indicate that OM hosted pores are abundant in bulk shale samples under FE-SEM observation. The BET Surface area and the BJH pore volume of isolated

RI

OM are approximately three times greater than that of bulk shale samples. Pore size distribution (PSD) of isolated OM from five

SC

shale samples show a consistent variation pattern. Additionally, with increasing TOC content, the PSD curves of isolated OM gradually grow down as result of the ductility of OM. Bimodal PSD in surface area and unimodal PSD in pore volume exist in

NU

both isolated OM and bulk shale samples, indicating surface area is mainly associated with micropores and fine mesopores (< 10nm) and larger pores are the dominate contributor to pore volume. Paralleled variation trends in PSD, especially in the pore size

MA

smaller than 4 nm, are observed between isolated OM and its corresponding bulk shale samples. However, the PSD divergences at the pore size larger than 10 nm between different samples are obvious. Therefore the pore network in gas shale reservoir is predominantly associated with the organic matter, especially small pores, and the mineral compositions are expected to be

AC

CE

PT E

D

responsible for lager pores.

ACCEPTED MANUSCRIPT 1. Introduction Growing attentions have been paid to the nanometer to micrometer scale pore network in shale rocks around the world during the last several years (Loucks et al., 2009, 2012; Mastalerz et al., 2012; Milliken et al., 2013; Clarkson et al., 2013; Labani et al., 2013; Tian et al., 2013; Wang et al., 2015; Signal, 2015; Fleury and Romero-Sarmiento, 2016). Natural gas produced in these pore networks, so called shale gas, lead America to energy independence and change the global energy market (McGlade et al., 2015). Unlike conventional oil and gas, shale gas system is a self-contained source-reservoir continuous petroleum system. Shale gas

PT

generated from biogenic and/or thermogenic process and stored in situ mainly as both adsorbed gas (i.e., on organic matter) and free gas (i.e., in pores and fractures) (Curtis, 2002; Jarvie et al., 2007). Pore system in shale rock is typically complicated and

RI

heterogeneous as a result of multiple pore type and origin, variable pore morphology and broad pore size range (Loucks et al.,

SC

2009, 2012; Clarkson et al., 2013). Based on the relationships between pores and grains, shale matrix-related pores can be classified as interparticle pore between grains and crystals, intraparticle pores within mineral particles, and organic matter pores

NU

(Loucks et al., 2012). In terms of the pore size spectrum, sub-micrometer pores in shale can be categorized into micropores with pore diameter less than 2 nm, mesopores with pore diameter between 2 nm and 50 nm, and macropores with pore diameter larger

MA

than 50 nm, following the International Union of Pure and Applied Chemistry (IUPAC) classification based on physical adsorption capability and capillary condensation theory (Rouquerol et al., 1994). Pore structure has a critical control on the shale gas storage and transport mechanism. Compared with large pores, more

D

surface areas and greater adsorption potential energies can be provided by small pores (Ross and Bustin, 2009). Therefore, the

PT E

adsorption behavior and the ratio of the adsorbed gas to free gas are variable in pores with different sizes. It is largest in micropores and decreases gradually from mesopores to macropores (Wang et al., 2015). Beliveau (1993) also figured out

CE

adsorption gas is more than free gas in pores with diameter less than approximate 20nm. Additionally, Rexer et al., (2014) suggested the vast majority of methane adsorption occurs in pores smaller than 6 nm based on the strong correlation between

AC

maximum methane sorption and CO2 sorption pore volume. The gas transport mechanism is variable in pores with different sizes. Gas flow in shales operate dominantly via diffusion and slippage flow in nanometer-size pores and Darcy continuum flow in larger pores depending on the Knudsen number (Javadpour et al., 2007). Therefore, characterizing the pore structure is very significant to better understand the shale gas storage and transport mechanism. A systematic multidisciplinary techniques, both qualitatively and quantitatively, could be utilized to elucidate the complex pore system in organic shales: scanning electron microscopy (SEM) images (Loucks et al., 2009, 2012; Milliken et al., 2013), atomic force microscopy (AFM) images (Javadpour, 2009), small or ultra-small angle neutron scattering (SANS/USANS) techniques ( Mastalerz et al., 2012; Clarkson et al., 2013), nuclear magnetic resonance (Signal, 2015; Fleury, 2016), low pressure N2 or CO2 physisorption (Clarkson et al., 2013; Labani et al., 2013; Tian et al., 2013; Wang et al., 2015), mercury injection capillary pressure (MICP) porosimetry (Clarkson et al., 2013; Ross and Bustin, 2009) and among other methods, with each

ACCEPTED MANUSCRIPT method characterizing a specific pore size range. Among these pores, organic matter hosted pore plays the most predominant role in organic rich shale pore system (Loucks et al., 2012; Milliken et al., 2013) and is major contributor to the shale gas adsorption capacity (Ross and Bustin, 2009; Zhang et al., 2012; Rexer et al., 2014; Ji et al., 2014, 2015). However, bulk characterization (e.g. gas physisorption, mercury intrusion) cannot distinguish OM-hosted pores from mineral-hosted pores (Löhr et al., 2015). Although SEM can be applied to identify the organic matter pore in shale, this is hampered by observation resolution (4-7nm) (Milliken et al., 2013) and the spatial limitation of the observed area and thus may not be able to achieve representative imaging

PT

for reservoir characterization (Mastalerz et al., 2012). Therefore, in this study a combination of isolated organic matter samples and its corresponding bulk shale samples were used for ultra-low N2 physisorption to elucidate the important role of organic

RI

matter in pore development of gas shale reservoir. Furthermore, Field emission scanning electron microscope (FE-SEM) equipped

SC

with an energy dispersive spectroscopy (EDS) system were also used to identify pore development in the shale samples. 2. Material and methods

NU

2.1 Samples

Great success in the Fuling shale gas field in Chongqing from the Upper Ordovician Wufeng Formation and the Lower

MA

Silurian Longmaxi shows an attractive exploration and development potential for shale gas in the Sichuan Basin, south China. A total of 5 core samples of the Lower Silurian Longmaxi shale were collected from Well YC-4 in the interval 655.8 ~ 739m (Table 1) in the southeastern edge around Sichuan Basin in Chongqing, China (Fig. 1). The shale samples are black in color and with

D

visible graptolite fossils. The Lower Silurian Longmaxi shale is characterized by huge thickness, high TOC content, favorable

PT E

types of organic matter, high maturities, abundant brittle minerals and strong gas generation intensity in the south Sichuan Basin (Dai et al., 2014). The Longmaxi Formation is a succession of organic-rich black mudstones deposited in a deep-water shelf

CE

environment, predominantly dysaerobic to anoxic continental margin basin, during the Early Silurian period. More detailed

Zou et al., 2015).

AC

information on the stratigraphy, geology, distribution and geochemistry of these shales can be obtained in Refs. (Dai et al., 2014;

SC

RI

PT

ACCEPTED MANUSCRIPT

Fig. 1. Map showing the location of the sample wells (B) on the edge of the southeastern Sichuan Basin in the Upper Yangtze Platform, south China (A).

2.2 Mineralogy and organic petrography

NU

The shale samples were crushed and sieved to less than 80 mesh particle size for ultra-low pressure nitrogen adsorption, X-ray diffraction (XRD) and total organic carbon (TOC) measurement. The TOC content was determined by a Leco CS230

MA

carbon/sulfur analyzer. 1 g of the shale samples, in a porous crucible, were treated using hydrochloric acid to remove carbonates. After 2 hours the samples were washed out using distilled water. After the water had drained from the crucible, the crucible and

D

sample were dried overnight at 70 °C. The total organic carbon content was then measured. Because no vitrinite occurs in rocks

PT E

earlier than Devonian due to the absent of higher land plants, the reflection of solid bitumen (R b) was measured. The reflection of solid bitumen (Rb) can be converted to equivalent vitrinite reflectance (Ro*) based on the following linear regression: Ro*= 0.618Rb+0.4 (Jacob et al., 1985).

CE

Bulk mineralogical compositions were determined from the X-ray diffraction patterns measured on a Bruker D8 DISCOVER diffractometer using Co Kα-radiation produced at 45 kV and 35 mA. Crushed samples were mixed with ethanol, hand ground and

AC

then smear mounted on glass slides for X-ray diffraction analysis. During the measurement, the sample was illuminated through a fixed divergence slit (1.8 mm, 1.45°), a graphite monochromator, and 58 mm, 0.3 mm spacing Soller slits. The diffracted beam was measured with a scintillation detector with counting time of 20 s for each step of 0.02° 2θ. Diffractograms were recorded from 2° to 76° 2θ. Quantitative phase analysis was performed by Rietveld refinement, with customized clay mineral structure models (Ufer et al., 2008). 2.3 FE-SEM observation Shale chip samples of about one centimeter square were polished to create a level surface using dry emery paper and then milled by a focused ion beam (FIB) system. In this FIB system, a focused 30 kV beam of argon ions mills the samples by sputtering away shale material via momentum transfer. Samples were coated with carbon to provide a conductive surface layer.

ACCEPTED MANUSCRIPT Each carbon-coated section was inspected using an FEI Helios NanoLab™ 650 FE-SEM. The FE-SEM images the shale sample surface with a resolution of 2.5 nm at 2 kV accelerating voltage and a working distance of 4 mm. EDS analyses of specific grains were also conducted for mineral identification. 2.4 Organic matter isolation: Organic matter samples were isolated using chemical methods. Shale samples (∼125 g each) were crushed to powder and then were Soxhlet extracted with chloroform for 72 h to remove soluble organic matter. The solid residue after solvent was

PT

continuously treated with HCl for about 12 h to remove carbonates. After washing with distilled water, the residue was treated with HF for 12 h to remove silicate minerals (Guthrie et al., 1994). The mixture was diluted with distilled water. The residue

RI

particle was treated acidified CrCl2 for 12 h to remove pyrite (Rexer et al., 2014). The residue was then washed again with

SC

distilled water for three times. The shale particles were separated from the solution by centrifuging (20 min, 3000 min−1). Samples were then freeze-dried (−5 °C) for 6 h and dried at ambient temperature. Details about the process and chemical reactions can be

NU

obtained from the Supporting Information in reference (Rexer et al., 2014). 2.5 Ultra-low pressure N2 adsorption

MA

Ultra-low pressure N2 adsorption analyses were performed at -196 °C and relative pressure ranged between 10 -7 and 0.995 using a Quantachrome® Autosorb-iQ2-MP apparatus at School of Petrochemical Engineering in Changzhou University. Because the N2 diffusion velocity is extremely low in microporous materials under ultra-low pressure, the equilibration time was set to 10

D

minutes for the threshold of relative pressure of 10 -6, decreased to 8 minutes for relative pressure between 10-6 and 10-5, and then

PT E

gradually decreased to 2 minutes for a relative pressure higher than 0.01. Detailed information can be obtained from Wang et al., (2015). Results include the pore surface area, pore volumes and average pore diameter. The pore surface area is calculated from

CE

the BET (Brunauer Emmett Teller) equation (Brunauer et al., 1938). The pore volume and pores size distribution of the shale samples were determined using the Barrett, Johner and Halenda (BJH) method for adsorption between relative pressure ranges of

AC

0.06 and 0.99 (Barrett et al., 1951). The average pore diameters were obtained using the ratio between the total amount of nitrogen adsorbed and the surface area assuming cylindrical pore geometry (Labani et al., 2013). 3. Results and discussions

3.1 Organic geochemistry and mineralogy The organic geochemistry and mineralogy composition of the studied shale samples are presented in Table 1. The total organic carbon (TOC) content ranges from 1.41% to 3.48%. The equivalent vitrinite reflectance values of theses samples are between 2.67% and 2.80% which indicates high maturity level for dry gas generation. The mineral compositions of the shale samples are consisting of a high proportion of clay (31% ~ 58%) and quartz (29% ~ 39%), moderate content of feldspar (6% ~ 21%) and carbonates (4% ~ 18%), and low abundance of pyrite (3% ~ 5%). For the entire shale samples the dominant clay minerals are illite (15% ~ 28%) and mixed layer illite/smectite (4% ~ 22%) with certain amount of chlorite (3% ~ 13%).

ACCEPTED MANUSCRIPT Table 1 Geochemical characteristic and mineralogical composition of the Longmaxi shale samples. Sample

Depth

TOC

Minerals (%)

Maturity Total clays

Quartz

Feldspar b

Carbonate c

Pyrite

22

58

29

6

4

3

27

13

52

32

8

5

3

10

25

16

51

34

7

4

4

4

28

4

36

36

12

12

4

3

15

13

31

39

7

18

5

ID

(m)

(%)

(RO*, %)

Chlorite

Illite

I/S

Sample 1

655.8

1.41

2.78

13

23

Sample 2

673.6

2.47

2.67

12

Sample 3

681.5

2.70

2.69

Sample 4

727.5

2.61

2.72

Sample 5

739.0

3.48

2.80

a

b

a

c

PT

Note: I/S = mixed layer Illite/smectite, Carbonates = calcite + dolomite, Feldspars = orthoclase + plagioclase, Ro* =equivalent vitrinite reflectance converted from the

RI

reflectance of bitumen (Rb) based on Ro*= 0.618Rb+0.4 (Jacob, 1985).

3.2 FE-SEM observation and pore types

SC

Field emission scanning electron microscope (FE-SEM) was utilized to examine pore type in the Longmaxi shale samples. Overall, FE-SEM images (Fig. 2) show that various types of pores are developed with pore size between several nanometers and

NU

several hundred nanometers in the Longmaxi shale samples. Among these pores, OM hosted pore are the most prevalent in the pore system with various pore shapes and pore sizes (e.g. Fig. 2A~C). The OM pores with irregular and elliptical shape are

MA

uniformly distributed independently within OM, but a few pores are extended to each other forming a larger pore and displaying complex shape in large OM particles (Fig. 2B). Interparticle pores are distributed between different mineral particles ranging

D

from soft clay minerals to hard rigid minerals mainly with slit shape (Fig. 2D~G). Sporadic intraparticle dissolution pores are also

PT E

identified within some minerals, e.g. quartz grain (Fig. 2D), albite grain (Fig. 2E), and calcite grain (Fig. 2F). Intercrystal pore within pyrite framboids (Fig. 2G) are often recognized, some of which are commonly filled by organic matter. Clay interlayer pores are also distinguished between clay platelets, for instance, chlorite (Fig. 2H) and illite (Fig. 2I). Interestingly, many pores

CE

between clay interlayers are often filled by organic matter with numerous OM pores, which forms a complex pore network (e.g. illite Fig.2A, B). From plenty of FE-SEM observations, we can conclude that the OM hosted pores are the most abundant in the

AC

high matured Longmaxi shale with certain amount of clay hosted pores and the other pore types are relatively seldom.

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

CE

Fig. 2. Field emission scanning electron microscope (FE-SEM) images of shale pore from the Longmaxi Formation in this study.

3.3 N2 adsorption and desorption isotherms

AC

The comparisons of ultra-low pressure nitrogen physisorption isotherms of 5 isolated OM and their corresponding bulk shale samples at liquid nitrogen temperature (-196°C) are illustrated in Fig. 3A~E. In order to clarify the detail of ultra-low pressure nitrogen physisorption isotherms of bulk shale samples, the amplification of nitrogen gas physisorption isotherms of bulk shales are displayed in Fig. 3F. As can be seen in Fig. 3, the adsorption amounts of N2 for isolated OM are far larger than bulk shale samples, indicating large pore volumes developed in organic matter. The N2 physisorption isotherms of both isolated OM and bulk shale samples show a significant hysteresis pattern (P/P0 > 0.5) with an absence of plateau at high pressure (P/P0 > 0.95). Such N2 isotherm shapes belong to Type IV isotherm (isotherm with hysteresis loop) according to the International Union of Pure and Applied Chemistry (IUPAC) classification (Sing et al., 1985), which indicate these materials contain both mesopores, which results in the hysteresis, and macropores, which lead to the absence of the plateau (Kuila et al., 2014). The hysteresis of isolated

ACCEPTED MANUSCRIPT OM is more prominent than that of bulk shale sample, indicating more mesopores exist in bulk shale sample than in isolated OM. Obviously, mineral compositions, especially clay minerals, are responsible for the more presences of macropores in bulk shale sample. Additionally, all the isotherms show adsorption at very low relative pressure (P/P0 < 0.01), indicating the existence of micropores. This observation suggests that organic matter may play a dominant role in the whole pore development of shale, however mineral compositions mainly contribute to the macropores development in gas shale reservoir.

300

150

Desorption branch of OM

Desorption branch of OM

Adsorption branch of shale

Adsorption branch of shale

Desorption branch of shale

Desorption branch of shale

PT

Adsorption branch of OM

Adsorption branch of OM Desorption branch of OM Adsorption branch of shale

SC

200

Adsorption branch of OM

(C)

Sample-3

Desorption branch of shale

NU

Volume adsorbed [cm3/g]

250

(B)

Sample-2

RI

(A)

Sample-1

100

MA

50

0 0.2

0.4

0.6

0.8

1 0

0.2

0.4

0.6

0.8

1

D

0

0

0.2

0.4

0.6

0.8

1

300

(D)

Sample-4 Adsorption branch of OM

200

Desorption branch of OM

(F)

Adsorption branch of shale Desorption branch of shale

50

Adsorption branch of shale

Desorption branch of shale

100

Sample-5 60

Adsorption branch of shale

150

(E)

Adsorption branch of OM

CE

Desorption branch of OM

70

Sample-5

40

Desorption branch of shale

AC

Volume adsorbed [cm3/g]

250

PT E

Relative pressure [P/P0]

30

20 50 10

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

Relative Pressure [P/P0] Fig. 3. Comparison of ultra-low pressure nitrogen gas physisorption isotherms of isolated organic matter (OM) and its corresponding bulk shale samples of the Longmaxi Formation at -196 °C. P0 is the saturated vapor pressure and P is the gas equilibrium pressure. The square-marked line indicates desorption branch while the circle-marked line indicates adsorption branch. Figure F show the amplification of nitrogen gas physisorption isotherms of bulk shale sample 5..

ACCEPTED MANUSCRIPT 3.4 Pore structure from N2 physisorption isotherms Pore structure basically refers to the specific surface area, total pore volume, pore size distribution, and average pore diameter. These parameters obtained from ultra-low pressure nitrogen physisorption experiment are presented in Table 2. The BET specific surface area of isolated organic matter ranges from 61.33 m2/g to 89.22 m2/g, with an average value of 78.86 m2/g, which is approximate three times greater than that of its corresponding bulk shale samples ranging from 24.34 m2/g to 29.93 m2/g, with an average value of 27.01 m2/g. The BJH pore volume of isolated organic matter varies from 0.3307 cm3/g to 0.4080 cm3/g,

PT

with a mean value of 0.3726 cm3/g, which is approximate four times greater than that of its corresponding bulk shale samples ranging from 0.0855 cm3/g to 0.1297 cm3/g, with a mean value of 0.1010 cm3/g. The average pore diameter of isolated organic

RI

matter ranges from 16.24 to 21.72 nm, with an average of 19.27 nm and its corresponding shale samples is between 14.18 to 18.11

SC

nm, with an average of 15.67 nm, Interestingly, the average pore diameters of isolated organic matter and its corresponding shale samples are almost equivalent, especially the sample 3.

Pore structure parameters of isolated organic matter and shale from nitrogen isotherms Isolated organic matter Sample BET surface area

BJH pore volume

Average pore diameter

(m2/g)

(cm3/g)

(nm)

Sample 1

89.22

0.3565

16.24

Sample 2

84.96

0.4080

19.01

Sample 3

84.63

0.3818

18.38

Sample 4

74.15

0.3861

Sample 5

61.33

0.3307

NU

Table2

Shale

BET surface area

BJH pore volume

Average pore diameter

(m2/g)

(cm3/g)

(nm)

24.34

0.0855

14.80

28.20

0.1032

15.44

29.93

0.1297

18.11

21.01

25.33

0.0949

15.82

21.72

27.26

0.0916

14.18

PT E

D

MA

ID

As can be seen in Table 2 and Fig. 4, there is a decrease trend in the BET surface area, whereas an increase trend in average

CE

pore diameter with increasing TOC content in isolated organic matter(Fig. 4A and C). This phenomenon may be due to the formation of many larger, complex pores in large organic matter particle, which derived from combination of smaller closely

AC

spaced pores during pore growth (Loucks et al., 2009).This can be well verified by FE-SEM observation (e.g. Fig. 2B).The BJH pore volume of isolated organic matter decreases with increasing TOC content except the sample 1 (see red dashed circle in Fig. 4B). This decreasing trend may be caused by the compaction of soft and ductile organic matter in shales with high TOC content during progressive burial of sediments. No significant linear variations in average pore diameter and BJH pore volume of bulk shale samples exists among different samples (Fig. 4E and F).The BET surface area in shales has a weak positive relationship with TOC content (Fig. 4D). Good positive linear trend in surface area and pore volume with TOC content in shale reservoir have been previously found by many researchers (e.g. Ross and Bustin, 2009; Tian et al., 2013; Ji et al., 2016). This indicates that organic matter pore is a significant contributor to the shale pore system.

ACCEPTED MANUSCRIPT 100

0.45

90

80

70

60

(C)

0.4

0.35

0.3 1

2 3 TOC content [%]

4

1

2 3 TOC content [%]

20 2 3 TOC content [%]

4

2 3 TOC content [%]

4

2

4

19 (E)

0.12

(F) 17

0.1

0.08

15

13

1

2

3 TOC content [%]

4

1

3 TOC content [%]

SC

1

15

PT

24

17

1

Average pore diameter [nm]

28

19

RI

(D)

21

4

0.14 BJH pore pore volume [cm3/g]

32 BET surface area [m2/g]

23 (B) Average pore diameter [nm]

BJH pore volume [cm3/g]

BET surface area [m2/g]

(A)

Fig. 4. Plots of BET surface area, BJH pore volume and average pore diameter versus TOC content of isolated organic matter (A, B and C) and bulk shale samples

NU

(D, E and F).

3.5 BJH pore size distribution (PSD)

MA

The pore size distribution (PSD) can be displayed as cumulative, incremental, or differential distribution curves with respect to pore volume or surface area (Clarkson et al., 2013; Tian et al., 2013). The differential distribution curve can obtained from the

D

cumulative cure of pore volume (V) or surface area (S) versus pore diameter (D) by differentiation. Note that the dV/dlog(D) =

PT E

ln10 × dV/d(D) × D meaning that the derivative (y-axis value) is markedly amplified for larger pore sizes (Clarkson et al., 2013). Therefore, the parameter dV/dlog(D) can better show the characteristic of larger pore compared with the dV/d(D). The PSD curves derived from the desorption branch of the isotherm show a strong artificial pores peak at approximately 4 nm as the result

CE

of the tensile strength effect (TSE) phenomenon (Groen et al., 2003). Therefore, the 4 nm abnormal peaks in pore size

AC

distributions were ignored when we discuss the different PSD features derived from N2 desorption branch.

ACCEPTED MANUSCRIPT 0.06

60 (A)

(B) Sample 1 TOC=1.41% Sample 2 TOC=2.47%

0.04

Sample 3 TOC=2.70% Sample 4 TOC=2.61%

0.03

Sample 5 TOC=3.48%

0.02

50 dS/d(D) [m3/nm/g]

0.01

Sample 3 TOC=2.70%

30

Sample 4 TOC=2.61% Sample 5 TOC=3.48%

20

0 10 Pore diameter, D [nm]

100

1

0.5

10 Pore diameter, D [nm]

500 (C)

(D)

Sample 2

400 dS/dlog(D) [m2/g]

0.4

Sample 3 Sample 4

0.3

RI

Sample 1

Sample 5

300

200

NU

0.2

PT

1

100

Sample 1 TOC=1.41% Sample 2 TOC=2.47% Sample 3 TOC=2.70% Sample 4 TOC=2.61% Sample 5 TOC=3.48%

100

0.1

0

1

10 Pore diameter, D [nm]

100

0.5 (E) 0.4

D

Sample 1 TOC=1.41%

Sample 3 TOC=2.70% Sample 4 TOC=2.61%

0.2

Sample 5 TOC=3.48%

0 1

CE

0.1

PT E

Sample 2 TOC=2.47%

0.3

MA

0

10 Pore diameter [nm]

Cumulative surface area [m2/g]

dV/dlog(D) [cm3/g]

Sample 2 TOC=2.47%

40

10

0

Cumulative pore volume [cm3/nm/g]

Sample 1 TOC=1.41%

SC

dV/d(D) [cm3/nm/g]

0.05

1

10 Pore diameter [nm]

100

150 (F)

100

Sample 1 TOC=1.41%

50

Sample 2 TOC=2.47% Sample 3 TOC=2.70% Sample 4 TOC=2.61% Sample 5 TOC=3.48%

0 100

1

10 Pore diameter [nm]

100

AC

Fig. 5. Different presentation of BJH pore size distributions derived from N2 desorption branch of isotherms for kerogen samples in this study.

As illustrated in Figure 5, the different PSD curves of all isolated OM are well comparable displaying similar variation trend, illustrating that the organic matter pores in studied samples are analogously developed because of their relative homogeneities in both organic matter type and thermal maturity level. The plots of dV/d(D) of isolated OM show a distinct peak with pore size of 1~10 nm and a faint peak around 60 nm pore size (Fig. 5A). However, the plots of dV/dlog(D) show typical bimodal with pore peaks around 1 ~10 nm and 50 ~ 110 nm, respectively (Fig. 5C). These pore size distribution patterns suggest macropores contribute significantly to the total pore volume. This is because a large pore can provide pore space volume equal to many small pores (Tian et al., 2013). Unlike pore volume of isolated organic matter, both plots of dS/d(D) and dS/dlog(D) show unimodal with pore peak at 1 ~ 10nm. This phenomenon suggests the micropores and fine mesopores (< 10 nm) are the dominant

ACCEPTED MANUSCRIPT contributors to the surface area development in isolated organic matter.

0.01

12

(A)

0.008

Sample 1 TOC=1.41% Sample 2 TOC=2.47%

Sample 2 TOC=2.47% Sample 3 TOC=2.70%

0.006

Sample 4 TOC=2.61% Sample 5 TOC=3.48%

0.004

dS/d(D) [m2/nm/g]

dV/d(D) [cm3/nm/g]

(B)

Sample 1 TOC=1.41%

Sample 3 TOC=2.70%

8

Sample 4 TOC=2.61% Sample 5 TOC=3.48%

4

0

0 10 Pore diameter, D [nm]

0.1

100

1

10 Pore diameter, D [nm]

100 (D) 80

0.04 Sample 1 Sample 2

0.02

100

Sample 1 TOC=1.41% Sample 2 TOC=2.47% Sample 3 TOC=2.70% Sample 4 TOC=2.61% Sample 5 TOC=3.48%

40

20

MA

Sample 3

60

NU

0.06

dS/dlog(D) [m2/g]

0.08

SC

(C)

RI

1

dV/dlog(D) [cm3/g]

PT

0.002

Sample 4 Sample 5

1

10 Pore diameter, D [nm]

100

D

(E)

Sample 2 TOC=2.47%

0.1

Sample 3 TOC=2.70% Sample 4 TOC=2.61% Sample 5 TOC=3.48%

CE

0.05

PT E

Sample 1 TOC=1.41%

0 1

AC

Cumulative pore volume [cm3/g]

0.15

10 Pore diameter [nm]

0 1

10 Pore diameter, D [nm]

100

40 (F)

Cumulative surface area [m2/g]

0

30

20 Sample 1 TOC=1.41% Sample 2 TOC=2.47% Sample 3 TOC=2.70%

10

Sample 4 TOC=2.61% Sample 5 TOC=3.48%

0 100

1

10 Pore diameter [nm]

100

Fig. 6. Different presentation of BJH pore size distributions derived from N2 desorption branch of isotherms for shale samples in this study.

There are systematic decreasing trend in all PSD curves of isolated OM with increasing TOC contents (Fig. 5A~F). The organic matters are soft and ductile and are therefore prone to be compressed in shales with high TOC content during progressive burial of sediments, which results in the decreasing trend in PSD curves of isolated OM with increasing TOC contents. This can also be well proved by the decreasing BJH pore volume trend with TOC content in the isolated organic matter (Fig. 4B). However, exception is observed for the dV/dlog(D) of larger macropores (> 60 nm), which show that the curves of dV/dlog(D) of sample 4 and sample 5 turn higher than the curves of dV/dlog(D) of sample 1, sample 2 and sample 3 (Fig. 5C, E) at larger pore sizes (> 60

ACCEPTED MANUSCRIPT nm). In the large organic matter particle the small pores will connect with each other to form larger and more complex pores during pore growth in high thermal maturity stage as mentioned previously in session 3.2 and verified by FE-SEM observation (e.g. Fig. 2B), which leads to the abnormal increasing trend in the dV/dlog(D) of lager macropores (60 nm) in isolated organic matter with increasing TOC contents of shales. There may be more large organic matter particle in organic shale containing high TOC content. The different PSD curves of all bulk shale samples are illustrated in Figure 6. Overall, the different PSDs of bulk shale

PT

samples display similar peak patterns as isolated organic matter. There also exist two prominent peaks in the curves of dV/dlog(D) (Fig. 6C) and a weak second peak in the dV/dlog(D) of bulk shale samples (Fig. 6A) excluding the artificial 4 nm peak. Like the

RI

isolated organic matter, both the curves of dS/d(D) and dS/dlog(D) of bulk shale samples show unimodal with pore peak at 1 ~

SC

10nm (Fig. 6B and D). The difference is that all the PSD curves of bulk shale samples demonstrate no regularity and strong variation among different samples. This phenomenon may be due to the effect of mineral compositions within bulk shale samples

NU

on the pore development. Compared to the isolated organic matter, the dV/dlog(D) of bulk shale samples show more prominent

pore peak around 50 ~ 100 nm, indicating the mineral compositions mainly govern the macropore development in shale

MA

rocks. The discussions of Fig. 5 and Fig. 6 indicate strong heterogeneity of bulk shale samples and relative homogeneity of

AC

CE

PT E

D

isolated organic matter in PSD property.

ACCEPTED MANUSCRIPT 0.6

60

0.2

0

0 1

10 100 Pore diameter [nm]

1

0.02

0.4

0.2

0

0 1

10 Pore diameter [nm]

100

10 Pore diameter[nm]

1

0.2

0

0 1

10 100 Pore diameter [nm]

D

0.02

0.4

0.2

0

0 1

10 100 Pore diameter, D [nm]

1

PT E

OM Shale

dV/dlog(D) [cm3/g]

dV/d(D) [cm3/nm/g]

0.04

10 100 Pore diameter, D [nm]

0 1

10 100 Pore diameter [nm]

600 Sample 3 TOC=2.70% OM Shale

400

200

0 1

10 100 Pore diameter [nm]

600

Sample 4 TOC=2.61% OM Shale

Sample 4 TOC= 2.61% OM Shale

400

200

0

1

10 100 Pore diamter[nm]

1

60

Sample 5 TOC = 3.48% OM Shale

200

100

20

10 100 Pore diameter[nm]

0.6 Sample 5 TOC = 3.48%

40

10 Pore diamter[nm]

0

1

0.06

NU

0.4

dS/d(D) [m2/nm/g]

0.02

60 Sample 4 TOC=2.61% OM Shale

dS/d(D) [m2/nm/g]

OM

20

MA

Sample 4 TOC=2.61% 0.04

40

100

0.6 dV/ldlog(D) [cm3/g]

0.06

Sample 3 TOC=2.70% OM Shale

0 1

Sample 2 TOC=2.47% OM Shale

400

10 100 Pore diamter[nm]

60 Sample 3 TOC=2.70% OM Shale

dS/d(D) [m2/nm/g]

0.04

dV/dlog(D) [cm3/g]

dV/d(D) [cm3/nm/g]

10 100 Pore diameter[nm]

0.6 Sample 3 TOC=2.70% OM Shale

20

0 1

0.06

OM Shale

40

dS/dlog(D) [m2/g]

0.2

10 100 Pore diameter, D [nm]

600 Sample 2 TOC=2.47%

dS/dlog(D) [m2/g]

0.02

1

10 100 Pore diameter, D [nm]

60 Sample 2 TOC=2.47% OM Shale

0.4

0

dV/d(D) [cm3/nm/g]

1

dS/d(D) [m2/nm/g]

0.04

200

0

10 100 Pore diameter, D [nm]

0.6 Sample 2 TOC=2.47% OM Shale

dV/dlog(D) [cm3/g]

dV/d(D) [cm3/nm/g]

0.06

OM Shale

400

0 1

dS/dlog(D) [m2/g]

10 100 Pore diameter, D [nm]

20

PT

1

OM Shale

40

RI

0

Sample 1 TOC = 1.41% dS/log(D) [m2/g]

0.4

SC

0.02

OM Shale

Sample 1 TOC = 1.41% dS/d(D) [m2/nm/g]

dV/log(D) [cm3/g]

dV/d(D) [cm3/nm/g]

OM Shale

0.04

600

Sample 1 TOC = 1.41%

Sample 1 TOC = 1.41%

10 100 Pore diameter [nm]

600 Sample 5 TOC = 3.48%

Sample 5 TOC = 3.48% OM Shale

40

20

0

dS/dlog(D) [m2/g]

0.06

OM Shale

400

200

0 1

10 100 Pore diameter, D [nm]

1

10 Pore diameter, D [nm]

100

CE

Fig. 7. Comparison of Pore size distribution between isolated organic matter and its corresponding bulk shale samples in this study.

3.6 Comparison of PSDs between isolated OM and bulk shales

AC

In order to further investigate the role of organic matter in shale pore system, we compare the pore size distributions derived from the isolated OM with its corresponding bulk shale sample in the same figure (Fig. 7). From these figures, paralleled variation trends in PSD curves are observed between isolated organic matter and its corresponding bulk shale samples, indicating the pore development are predominantly associated with the organic matter. Interestingly, with TOC content increasing, the divergence of PSDs between isolated organic matter and its corresponding bulk shale samples gradually diminish in the small pore size range of smaller than 4 nm, and turn convergent in the sample 5 (Fig.7). This indicates the small pores in gas shale reservoir, especially pores with size smaller than 4 nm, are mainly associated with organic matter. The PSDs variation trends of isolated organic matter and bulk shale samples at the pore size larger than 10 nm are significant different, indicating the mineral compositions, especially clay minerals, are expected to have a larger percentage of lager nanoscale pores (pore diameter > 10 nm). The larger pores related

ACCEPTED MANUSCRIPT with mineral matrix result in more complicated and heterogeneous pore size characterization of bulk shales. Since both organic matter and clay minerals contribute to the surface area and pore volume in gas shale reservoir, the pore structure results of the isolated OM from this study will be helpful to better understand the pore system in organic shale rock and to better model natural gas transport in shale reservoir for gas shale resource assessments and production planning. 4. Conclusion The pore structure and distribution of isolated organic matter (OM) and its corresponding bulk shale samples from the high

PT

maturity shales were investigated using the combination of field emission scanning electron microscope (FE-SEM) observation and ultra-low pressure nitrogen gas physisorption. These findings reveal important information about different role of organic

RI

matter and mineral composition in pore network of gas shale reservoir, which will in turn affect the gas storage and transport

SC

properties. Conclusions of this study are as follows:

(1) FE-SEM observations reveal that both OM hosted pores and mineral matrix pores are both developed in the Lower

NU

Silurian Longmaxi shales, and that the OM hosted pore are the most dominant with certain amount of clay associated pores in the pore system of the Longmaxi gas shale reservoir.

MA

(2) The BET specific surface area of isolated OM ranges from 61.33 m2/g to 89.22 m2/g, which is approximate three times greater than that of its corresponding shale samples ranging from 24.34 m2/g to 29.93 m2/g. The BJH pore volume of isolated organic matter varies from 0.3307 cm3/g to 0.4080 cm3/g, which is approximate four times greater than that of its corresponding

D

shale samples ranging from 0.0855 cm3/g to 0.1297 cm3/g.

PT E

(3) There exists a set of consistent shapes of PSD curves of isolated OM from different shale samples and a systematic decreasing trend in the PSDs with increasing TOC contents, especially in micropores and mesopores as a result of the ductile of

CE

OM. However, the presence of pores in minerals makes the PSD shapes of different bulk shale samples complicated and heterogeneous.

AC

(4) Bimodal pore size distribution in surface area and unimodal pore size distribution in pore volume exist in both isolated OM and bulk shale samples, indicating the surface area is mainly associated with micropores and fine mesopores (< 10nm) in both isolated OM and bulk shales and larger pores are the dominate contributor to pore volume. (5) Paralleled variation trends in PSD curves are observed between isolated OM and its corresponding bulk shale samples and with TOC content increasing, the PSD divergence gradually diminishes in the small pore size, indicating the pore network in gas shale reservoir is predominantly associated with the organic matter, especially pores with size smaller than 4 nm. The PSD variation trends of isolated OM and bulk shale samples at the pore size larger than 10 nm between different samples are significant different, indicating the mineral compositions are expected to have a larger percentage of lager pores. ACKNOWLEDGMENTS

ACCEPTED MANUSCRIPT The authors are indebted to Dr. C. Özgen Karacan and anonymous reviewers for their constructive comments and suggestions that have significantly improved the manuscript. This work was supported by the National Science and Technology Major Project (2011ZX05018-02) and National Natural Science Foundation of China (41472112). We also acknowledge the Chongqing Institute of Geology and Mineral Resources for providing the drill cores used in this study. Special acknowledgements are given to the Chinese Scholarship Council for sponsoring the first author to be a visiting scholar in the University of Texas at Austin.

PT

REFERENCES CITED Barrett EP, Joyner LG, Halenda PP. The determination of pore volume and area distributions in porous substances. I.

RI

Computations from nitrogen isotherms. J Am Chem Soc 1951; 73 (1): 373-380.

SC

Beliveau D. Honey, I shrunk the pores! J Can Petrol Technol 1993; 32(08): 15-17.

Brunauer S, Emmett PH, Teller EJ. Adsorption of gases in multimolecular layers. J Am Chem Soc 1938; 60: 309-319.

NU

Clarkson CR, Solano N, Bustin RM, Bustin AMM, Chalmers GRL, He L, et al. Pore structure characterization of North American shale gas reservoirs using USANS/SANS, gas adsorption, and mercury intrusion. Fuel 2013; 103: 606-6.

MA

Curtis JB. Fractured shale-gas systems. AAPG Bull 2002; 86(11): 1921-1938. Dai J, Zou C, Liao S, Dong D, Ni Y, Huang J, et al. Geochemistry of the extremely high thermal maturity Longmaxi shale gas, southern Sichuan basin. Org Geochem 2014; 74: 3-12.

effer L

re - am re J.

ore si e determination in modified micro-and mesoporous materials. Pitfalls and

PT E

Groen JC

D

Fleury M, Romero-Sarmiento M. Characterization of shales using T 1–T 2 NMR maps. J Petrol Sci Eng 2016; 137: 55-62.

limitations in gas adsorption data analysis. Micropor Mesopor Mat 2003; 60(1): 1-17.

CE

Guthrie JM, Pratt LM. Geochemical indicators of depositional environment and source-rock potential for the Upper Ordovician Maquoketa Group, Illinois Basin. AAPG Bull 1994; 78(5): 744-757.

AC

Jacob H. Disperse solid bitumens as an indicator for migration and maturity in prospecting for oil and gas. Erdol Kohle Erdgas Petrochem 1985; 38(8): 365-365. Jarvie DM, Hill RJ, Ruble TE, Pollastro RM. Unconventional shale-gas systems: The Mississippian Barnett Shale of north-central Texas as one model for thermogenic shale-gas assessment. AAPG Bull 2007; 91(4): 475-499. Javadpour F, Fisher D, Unsworth M. Nanoscale gas flow in shale gas sediments. J Can Petrol Technol 2007; 46(10): 55-61. Javadpour F. Nanopores and apparent permeability of gas flow in mudrocks (shales and siltstone). J Can Petrol Technol 2009; 48(08): 16-21. Ji W, Song Y, Jiang Z, Chen L, Li Z, Yang X et al. Estimation of marine shale methane adsorption capacity based on experimental investigations of Lower Silurian Longmaxi formation in the Upper Yangtze Platform, south China. Mar Pet Geol 2015; 68: 94-106.

ACCEPTED MANUSCRIPT Ji W, Song Y, Jiang Z, Wang X, Bai Y, Xing J. Geological controls and estimation algorithms of lacustrine shale gas adsorption capacity: A case study of the Triassic strata in the southeastern Ordos basin, China. Int J Coal Geol 2014; 134: 61-73. Ji W, Song Y, Jiang Z, Meng M, Liu Q et al. Fractal characteristics of nano-pores in the Lower Silurian Longmaxi shales from the Upper Yangtze Platform, south China. Mar Pet Geol 2016; 78: 88-98. Kuila U, McCarty DK, Derkowski A, Fischer TB, Topór T, Prasad M. Nano-scale texture and porosity of organic matter and clay minerals in organic-rich mudrocks. Fuel 2014; 135: 359-373.

PT

Labani MM, Rezaee R, Saeedi A, Al Hinai A. Evaluation of pore size spectrum of gas shale reservoirs using low pressure nitrogen adsorption, gas expansion and mer cury porosimetry: A case study from the Perth and Canning Basins, Western Australia. J

RI

Petrol Sci Eng 2013; 112: 7-16.

SC

Löhr SC, Baruch ET, Hall PA, Kennedy MJ. Is organic pore development in gas shales influenced by the primary porosity and structure of thermally immature organic matter? Org Geochem 2015, 87: 119-132.

NU

Loucks RG, Reed RM, Ruppel SC, Hammes U. Spectrum of pore types and networks in mudrocks and a descriptive classification for matrixerelated mudrock pores. AAPG Bull 2012; 96: 1071-1098.

MA

Loucks RG, Reed RM, Ruppel SC, Jarvie DM. Morphology, genesis, and distribution of nanometer-scale pores in siliceous mudstones of the Mississippian Barnett Shale. J Sed Res 2009; 79: 848-861. Mastalerz M, He L, Melnichenko YB, Rupp JA. Porosity of coal and shale: insights from gas adsorption and SANS/USANS

D

techniques. Energy Fuels 2012; 26(8): 5109-5120.

PT E

McGlade C, Speirs J, Sorrell S. Unconventional gas–a review of regional and global resource estimates. Energy 2013; 55: 571-584.

CE

Milliken KL, Rudnicki M, Awwiller DN, Zhang T. Organic matter–hosted pore system, Marcellus Formation (Devonian), Pennsylvania. AAPG Bull 2013; 97(2): 177-200.

AC

Rexer TF, Mathia EJ, Aplin AC, Thomas KM. High-pressure methane adsorption and characterization of pores in Posidonia shales and isolated kerogens. Energy Fuels 2014; 28(5): 2886-2901. Ross DJ, Bustin RM. The importance of shale composition and pore structure upon gas storage potential of shale gas reservoirs. Mar Pet Geol 2009; 26: 916-927. Rouquerol J, Avnir D, Fairbridge CW, Everett DH, Haynes JH, Pernicone N, Ramsay JDF, Sing KSW, Unger KK. Recommendations for the characterization of porous solids (Technical Report). Pure Appl Chem 1994; 66 (8): 1739-1758. Sigal R F. Pore-size distributions for organic-shale-reservoir rocks from nuclear-magnetic-resonance spectra combined with adsorption measurements. SPE J 2015; 1-7.

ACCEPTED MANUSCRIPT Sing KS, Everett DH, Haul RAW, Moscou L, Pierotti RA, Rouquerol J, Siemieniewsha T. Reporting physisorption data for gas/solid systems with special reference to the determination of surface area and porosity. Pure Appl Chem 1985; 57: 603-619. Tian H, Pan L, Xiao XM, Wilkins RWT, Meng ZP, Huang BJ. A preliminary study on the pore characterization of Lower Silurian black shales in the Chuandong Thrust Fold Belt, southwestern China using low pressure N 2 adsorption and FE-SEM methods. Mar Pet Geol 2013; 48: 8-19.

PT

Ufer K, Stanjek H, Roth G, Dohrmann R, Kleeberg R, Kaufhold S. Quantitative phase analysis of bentonites by the Rietveld method. Clay Clay Miner 2008; 56: 272–282.

RI

Wang G, Ju Y. Organic shale micropore and mesopore structure characterization by ultra-low pressure N2 physisorption:

SC

Experimental procedure and interpretation model. J Nat Gas Sci Eng 2015; 27: 452-465.

Zhang T, Ellis GS, Ruppel SC, Milliken K, Yang R. Effect of organic-matter type and thermal maturity on methane adsorption in

NU

shale-gas systems. Org Geochem 2012; 47: 120-131.

Zou C, Yang Z, Dai J, Dong D, Zhang B, Wang Y, et al. The characteristics and significance of conventional and unconventional

AC

CE

PT E

D

MA

Sinian–Silurian gas systems in the Sichuan Basin, central China. Mar Petrol Geol 2015; 64: 386-402.

ACCEPTED MANUSCRIPT

MA

NU

SC

RI

PT

Graphical abstract

D

Highlights

OM hosted pores are the most abundant in high maturity shale under FE-SEM observation.



Surface area and pore volume of isolated OM is about 3 times greater than bulk shale.



The agreement among PSD curves of different isolated OM is remarkably good.



Minor pores (< 10nm) contribute to surface area and larger pores govern pore volume.



The pore system in shale is mainly associated with OM and minerals contribute more to large pores.

AC

CE

PT E

