Quantitative Analysis of Nanopore Structural Characteristics of Lower

Nov 30, 2017 - gallium ion beam, the sample desk was rotated 52° (Figure 4A). Moreover, the ROI was sprayed with a thin ... Avizo 9.0 software was us...
0 downloads 13 Views 2MB Size
Subscriber access provided by READING UNIV

Article

Quantitative Analysis of Nano-pore Structural Characteristics of Lower Paleozoic Shale, Chongqing (Southwestern China) - Combining FIB-SEM and NMR Cryoporometry Shaoqing Tong, Yanhui Dong, Qian Zhang, Derek Elsworth, and Shimin Liu Energy Fuels, Just Accepted Manuscript • DOI: 10.1021/acs.energyfuels.7b02391 • Publication Date (Web): 30 Nov 2017 Downloaded from http://pubs.acs.org on December 3, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Energy & Fuels is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

1

Quantitative Analysis of Nano-pore Structural Characteristics of

2

Lower Paleozoic Shale, Chongqing (Southwestern China) -

3

Combining FIB-SEM and NMR Cryoporometry

4

Shaoqing Tong1,2, Yanhui Dong1,2,3, *, Qian Zhang1,2, Derek Elsworth3, Shimin Liu3

5 6 7 8 9 10

1

11

ABSTRACT

Key Laboratory of Shale Gas and Geoengineering, Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, China 2 University of Chinese Academy of Sciences, Beijing 100049, China 3 Department of Energy and Mineral Engineering, G3 Center and EMS Energy Institute, The Pennsylvania State University, University Park, PA-16802, USA * Corresponding author: Yanhui Dong, Email: [email protected]

12

Characterizing nano-pore structure is one of the most important factors in understanding gas

13

storage and transport in shale reservoirs – but remains a significant challenge. In this work, we

14

combine the benefits of Focused Ion Beam Scanning Electron Microscopy (FIB-SEM) and

15

Nuclear Magnetic Resonance Cryoporometry (NMR-C) to characterize the nano-pore structure of

16

lower Paleozoic shales from Chongqing, southwestern China. Mineral composition is qualified

17

through X-ray Energy Dispersive Spectroscopy (EDS) and Field-Emission Scanning Electron

18

Microscopy (FE-SEM). Voids in 2D micrographs are classified into types of meso- and/or micro-

19

fractures, inter particle pores (InterP pores), intra particle pores (IntraP pores), and pores in

20

organic matter (OM pores). An Otsu thresholding algorithm and an edge detection algorithm (in

21

Avizo) are used to segment OM, InterP and IntraP pores, and to establish 3D pore structure for

22

pore-network modeling (PNM). The pore size distribution (PSD) is measured via PNM and

23

compared to the PSD recovered from NMR-C. Results indicate that the small OM pores have

24

favorable potential for storing adsorbed gas because of their apparent interconnectivity and higher

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

25

porosity; conversely, InterP and IntraP pores are mostly isolated and with low porosity. The peaks

26

of the PSD, recovered from PNM, range from 10~20 nm and 60~120 nm and are in good

27

agreement with the peaks of the PSD recovered from NMR-C. This suggests that the NMR signals

28

in these discontinuous ranges are due to OM pores, InterP and IntraP pores. This indirectly

29

demonstrates that the probe liquid in the NMR-C experiments (i.e. water), may indeed enter the

30

nano-scale pores in the shale during centrifugal saturation. Comparison of the PSD from PNM and

31

NMR-C also shows that NMR-C has a higher sensitivity and accuracy in detecting nano-pores.

32

Combining FIB-SEM and NMR-C is a promising technique in detecting and characterizing the

33

nano-porosity of shales.

34 35

KEYWORDS

36

Lower Paleozoic Shale, Nano-pore Structure, Pore Network Modeling, FIB-SEM, NMR-

37

Cryoporometry

38

39

1 Introduction

40

Shale gas is an important energy resource with rapid development in North America.1,2 It is

41

expected to play an increasingly important role in Asia and Europe in the near future.3 In recent

42

years, the lower Paleozoic shale gas reservoirs in the Sichuan Basin, southwestern China have

43

been explored and developed.4,5 The Weiyuan 201 Well (W201) was the first in China to provide

44

industrial-scale gas flow after hydraulic fracturing. The thermal maturity of the organic matter of

45

the ancient shales in the Sichuan Basin is higher than that of the Barnett, Marcellus and

ACS Paragon Plus Environment

Page 2 of 30

Page 3 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

46

Haynesville formations in North America - implying a lower degree of development of the pores

47

in the organic matter.6 Previous studies have indicated that shale pore structure has a strong

48

influence in controlling the shale gas content and storage and transport mechanisms.7,8,9 In

49

particular, pores in organic matter provide the main storage space for absorbed gas in shale.10,11,2,12

50

Hence, a correct understanding of the 2D and 3D pore structure is the foundation for shale gas

51

exploration, development, and research.9,13

52

As a typical low permeability reservoir, shale is a fine-grained, compact sedimentary rock

53

comprising pores and fractures at a variety of scales. The 2D and 3D pore morphology, size

54

distribution, and other features of shale have been investigated by many imaging methods. These

55

include field-emission scanning electron microscopy (FE-SEM), FE-SEM coupled with focused

56

gallium ion beam (FIB-SEM), or Nano X-ray computed tomography (Nano-XCT).2,5,14,15,16 Four

57

scales of void space can be identified in shale reservoirs. These include mesoscale fractures, meso-

58

and/or micro-fractures, inter particle and intra particle pores (InterP pores and IntraP pores), and

59

nano-scale pores in organic matter (OM pores).2,3,14,17,18 Imaging methods are able to distinguish

60

different types of pores in shale. Nano-XCT is a non-destructive method using transmission X-ray

61

attenuation imaging and is widely applied to detect pores in shales. Nano-XCT has been used to

62

study the development of nano-pores in the Chang 7 Member of organic-rich shale19 and to

63

characterize the connection of typical 3D pores in the Jiulaodong formation.2 However, the

64

minimum voxel size for Nano-XCT is approximately 32~50 nm, limiting the resolution of pores

65

that must transit of the order of three voxels.

66

FIB-SEM usually provides 3D micrographs at fine spatial resolutions of ~10 nm.20,21

67

Although milling and imaging layer-by-layer can partly damage the shale sample, nano-pores in

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

68

FIB-SEM micrographs are much more distinguishable than those recovered by Nano-XCT.

69

However, previous studies3,11,14 show that the majority of pores in organic matter range from 2 nm

70

to tens of nanometers. A few could be as large as few hundred nanometers. This means that

71

FIB-SEM could not cover the entire pore size distribution (PSD). In addition, FIB-SEM has a

72

small field of view.

73

Nuclear magnetic resonance cryoporometry (NMR-C) is a novel and emerging technique that

74

can probe PSD from 2 nanometers to several micrometers—generally more than three orders of

75

magnitude. It is based upon the pore-diameter-related melting and/or freezing points of pore

76

imbibed materials.11,12,22,23,24 For detection, NMR-C exploits the transverse relaxation time

77

difference between solid and liquid NMR signals.22 Firouzi, et al.22 first reported PSD

78

measurement using NMR-C method on shale, but they failed because the NMR instrumentation

79

used could not scan temperatures lower than 253 K to detect pores with size smaller than 6nm.

80

Using improved NMR instrumentation and saturation methods, later researches11,23,24 successfully

81

measured the PSD of shale using NMR-C. Fleury, et al.23 report compared the PSDs from NMR

82

relaxation and NMR-C in shales and found that the PSD obtained from cryoporometry is quasi

83

uniform from 2 nm up to 100 nm. Zhang, et al.24 utilized NMR-C to examine both bulk matrix and

84

pulverized shale samples with the results surprisingly indicating that the porosity of the bulk

85

sample was apparently smaller than that of the pulverized shale sample. Li, et al.25 used NMR-C,

86

NMR, mercury intrusion porosimetry, and gas adsorption methods to study pore structure

87

characteristics of shales in Southwestern China. They obtained good agreement in the PSDs

88

between the various methods. All previous studies focused on the applicability and accuracy of

89

NMR-C compared with other intrusive methods when probing pore structures in shale. However,

ACS Paragon Plus Environment

Page 4 of 30

Page 5 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

90

comparisons of NMR-C with imaging methods have been rarely reported.

91

In this work we use both FIB-SEM and NMR-C to characterize nano-pore structure of shale.

92

3D pore structure modeling and pore network modeling (PNM) are used for the analysis of OM

93

pores as well as InterP and IntraP pores in shale from the W201 well. Additionally, the PSDs

94

obtained from PNM and NMR-C are quantitatively analyzed and contrasted to understand and

95

assess the corresponding types and structure of nano-porosity in shale.

96

2 Geological setting

97

The shale samples were taken from the Weiyuan area in western Chongqing, from the upper

98

Yangtze Platform. The Weiyuan area extends over about 2700 km2 in the southwestern part of the

99

Sichuan Basin (Figure 1). The region is bordered on the southeastern edge of the Leshan-Longnvsi

100

paleo-uplift and is characterized as a large-scale NEE-SEW oriented domal anticline.26 The

101

Weiyuan area has experienced three major orogenies, comprising the Caledonian, Indosinian, and

102

Himalayan. Emergent Proterozoic and Paleozoic strata in this area are mainly from the Sinian,

103

Cambrian, Ordovician, Silurian, and Permain periods (Figure 2).

104

The upper Sinian is subdivided into the Dengying and Doushantuo formations with a total

105

thickness of 659~670 m.26,27 The Dengying formation, comprises mainly microcrystalline dolomite

106

and is the main conventional gas reservoir in the Weiyuan area.27,28 A total of 766~1019 m of the

107

Cambrian unit consists of the Jiulaodong and Yuxiansi formations in the lower Cambrian and the

108

Xixiangchi formation in the upper Cambrian.29,30 The Jiulaodong formation contains two-thirds of

109

the sedimentary column of organic-rich black shales with a thickness of 230~400 m. It is uniformly

110

distributed in the Weiyuan area and is the dominant source rock and caprock of the Sinian gas

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

111

reservoir. It is also the target formation of the W201 well.29 The Yuxiansi and Xixiangchi formations

112

are primarily interbedded with sandstone and siltstone. The Ordovician strata are poorly developed

113

in the Weiyuan area due to the advent of the Caledonian orogeny. The lower Ordovician Luohanpo

114

and Dachengsi formations are dominated by dolomite with a total thickness of 131~280 m, while

115

the middle Ordovician Baota formation mainly consists of limestone with thick shales.

116

The lower Silurian Longmaxi formation is a set of organic-rich sedimentary black shales and

117

silty mudstones in southeastern Weiyuan, but is absent in the northwestern part of this region. The

118

Longmaxi formation is an important shale gas source in southwestern China. However, the 0~140 m

119

thinness of the Longmaxi formation in this region limits the reserves of shale gas. Due to uplift

120

during the Hercynian orogeny, the Devonian and Carboniferous strata are absent.26,28 The Permian

121

contains the lower Permian Maokou formation and the upper Permian Longtan formation. The

122

Maokou formation is characterized by limestone with a thickness of 204~459 m, while the Longtan

123

formation consists of limestone and shale with a thickness of 188~242 m.26

124 125

Figure 1. Schematic illustrating the primary structural outline and border of the Sichuan Basin, the

ACS Paragon Plus Environment

Page 6 of 30

Page 7 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

126 127

location of the Weiyuan gas field and the W201 well, relative to major cities.

128 129 130

Figure 2. Stratigraphic column of the Weiyuan area including chronology, lithology, and thickness. The grey denotes shale gas source formations in southwestern China.

131

3 Samples and Methods

132

3.1 Sample preparations

133

Shale samples from the marine sedimentary Jiulaodong formation were sourced from a depth

134

of 2749~2751 m the W201 well – a well with industrial-scale gas flow that was drilled by China

135

National Petroleum Corporation (CNPC) in 2011. 4 mm diameter cylindrical samples both

136

perpendicular and parallel to the beddings were recovered from core plugs and prepared for

137

FE-SEM, FIB-SEM and NMR-C experiments (Figure 3).

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

138 139 140 141 142 143

Figure 3. Sample preparation details. (A) 4 mm diameter cylindrical shale samples recovered by core drill. A third of the sample (orange box) is used for FE-SEM and FIB-SEM with the other two-thirds (blue box) used for NMR-C. (B) Argon ion polished samples for FE-SEM and FIB-SEM, sheathed in a silver conductive adhesive. (C) Shale sample in the NMR-C sample tube with ultrapure water.

144

3.2 Experimental facilities

145

3.2.1 FE-SEM and FIB-SEM

146

A FE-SEM equipped with an EDS and FIB-SEM were used to characterize the nano-pore

147

structure of the shale samples. The FE-SEM samples were prepared using a 4-mm diameter core

148

drill and argon ion polished (Figure 3 (A) and (B)). Both bedding-parallel and -perpendicular

149

surfaces of the shale samples were imaged with a Zeiss Merlin scanning microscope at 5 kV. A

150

voltage of 15 kV was subsequently used to qualitatively analyze minerals and organic matter via

151

EDS (BRUKER XFlash 6130).

152

The shale samples were placed into the chamber of a ZEISS Crossbeam 540 FIB-SEM, and

153

electron-beam-scanned at a vacuum of < 5×10-6 mbar. The region containing varied organic matter

154

and quartz, calcite, and pyrite was selected as the principal region of interest (ROI). To align the

155

shale sample surface perpendicular to the gallium ion beam, the sample desk was rotated 52 °

156

(Figure 4 (A)). Moreover, the ROI was sprayed with a thin layer of Pt to counter any damage from

157

the gallium ion beam. After deposition of Pt, the FIB-SEM automatically milled and imaged the

ACS Paragon Plus Environment

Page 8 of 30

Page 9 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

158

sample layer-by-layer (Figure 4 (B)). The layer thickness was 10 nm and the selected ROIs were

159

10.07×9.64×4.28 μm and 11.3010.268.18 μm respectively (Figure 4 (C)). Avizo 9.0 software

160

was used for analysis of the 2D and 3D micrograph data.

161 162 163 164 165

Figure 4. (A) Schematic diagram of the milling and imaging procedure in a FIB-SEM system. (B) Site on the shale sample milled in cross section by the gallium ion beam. (C) and (D) are the two ROIs containing organic matter, quartz, calcite and pyrite, with an extent of 10.07×9.64×4.28  and 11.3010.268.18 , respectively; voxel size is 10 nm3.

166

3.2.2 Nuclear magnetic resonance cryoporometry and nitrogen adsorption method

167

measurement

168

NMR-C method based on the differences in nuclear magnetic resonance relaxation between the

169

freezing or melting point of the fluids in the porous media. Since NMR distinguishes between the

170

solid and liquid states of the permeant (quantifying only the liquid mass) the NMR-C signal gives a

171

clear transition of permeant state with high signal-to-noise ratio. NMR-C relies on the

172

pore-size-dependent depression of the melting temperature of a pore-filling material given by the

173

Gibbs-Thomson equation,

174

∆      

175

Here,  and  are the bulk and the pore melting temperatures,  is the solid-liquid surface

  

∆ 



ACS Paragon Plus Environment

(1)

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

176

free energy, ∆

177

diameter. For arbitrarily shaped pores,  depends on the average curvature of the pore wall

178

rather than on a uniform size parameter such as . Then equation (1) may be rewritten as:

179

∆ 

180

Here, &' is the melting point depression constant. In a typical experimental protocol, the NMR

181

signal intensity I(T) from the molten fraction of the pore-filling material is measured as the initially

182

frozen sample is incrementally-warmed. To separate this signal from that of the frozen fraction, the

183

difference between the transverse relaxation times T2 in the respective states is exploited. The PSD

184

is then calculated by numerical differentiation of I(T). The PSD can be estimated if the NMR signal

185

intensity is properly calibrated in terms of the mass of the sample. Smaller temperature increments

186

may resolve smaller differential pore sizes and finer PSD features.11

#$% 

! is

Page 10 of 30

the latent heat of melting, " is the density of the solid, and is the pore



(2)

187

In these measurements, the NMR-C shale samples were first pretreated in a drying oven for 24

188

hours at 378.15 K. The dry 4-mm diameter cylindrical shale samples were then placed in 10 mm

189

diameter thin-walled high-resolution NMR sample tubes. The tubes are approximately 2 cm in

190

length and filled with ultrapure water (Figure 3 (C)). The shale samples in the tubes were immersed

191

in water for ~10 h and then centrifuged at 5000 rpm for 3 h at room temperature to help saturate the

192

shales. The NMR-C measurements were conducted in a low field nuclear magnetic resonance

193

device manufactured by Niumag Corporation Ltd., China with a resonance frequency of 21 MHz

194

and a 10-mm probe. The temperature control system consisted of a compressor and a cooling bath to

195

provide a minimum temperature limit of 208.15 K. During the experiments, the temperature was

196

incremented at 0.1~2 K in each step with NMR signal measurements made over 10 mins to allow for

197

thermal equilibration at every temperature point.

ACS Paragon Plus Environment

Page 11 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

198

Shale samples were also characterized by the nitrogen adsorption method (NAM), and the

199

results were then compared to the NMR-C measurements to support the PSDs interpretation.

200

NAM measurements were conducted on a 3H-2000PS2 Instrument at 77.4 K with pulverized shale

201

samples.

202

3.3 Modeling methods

203

3.3.1 Pore structure modeling

204

Pore structure modeling is a direct approach to acquire the geometry of the pore space and was

205

employed here to characterize porosity and tortuosity.31 The first step is to segment the voids. This is

206

achieved simply by using the thresholds from the gray scale image histograms from each slice

207

followed by restacking the binarized three-dimensional images. Based on pore structure modeling,

208

the equivalent pore diameter can also be recovered and is widely used in quantitative analysis

209

defining the PSDs. The equivalent pore diameter can be given as:

210

()*  +

211

Here, ()* is the equivalent pore diameter and 4567) is the summation of voxels of connected pores.

212

Obviously, equation (3) describes a ball based equivalent pore diameter. However, one difficulty is

213

in accurately separating the connected pores at the correct throat links. Commonly, the connected

214

pores would not be divided in the pore structure modeling especially in the nano-pore structure of

215

shales. And then voxels of two or more connected pores are treated as one whole pore in calculating

216

the equivalent pore diameter with equation (3). This exaggerates the pore diameter. Therefore, pore

217

structure modeling is not a good idea to quantitatively analyze the PSD in this study.

3

,-./01 2



(3)

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

218

3.3.2 Pore network modeling (PNM)

219

An alternative technique to calculate equivalent pore diameters is to use pore network

220

modeling (PNM). This extracts a topologically representative network with idealized geometric

221

properties derived from the binarized three-dimensional images. We use the PNM method of Dong

222

and Blunt (2009),32 based on the maximal ball approach33 where spheres centered on each void

223

voxel are expanded to fill the pore. Both inflating and deflating search algorithms32 are used to

224

guarantee that the maximal ball is inscribed to the grain or image boundary. In principle, the larger

225

spheres represent pores defined as the larger voids in the rock, whereas the chains of smaller spheres,

226

connected by narrower pathways, define throats. Consequently, the PNM can distinguish pores and

227

throats as well as separate connected pores. Thus the pore diameter recovered from the PNM is not

228

overestimated, as it may be in pore structure modeling. The pore radius, throat radius, and pore

229

throat length are measured as shown in Figure 5 using Euclidean distances. Moreover, the PNM

230

provides other effective properties of pores and throats such as volume, shape factor, and

231

coordination numbers of pores. Further important output parameters contribute to the analysis of

232

nano-pore structure in shales.

233

In this study, 2D micrographs were imaged by FE-SEM for the analysis of nano-pore structure

234

in shales. 3D micrographs obtained from FIB-SEM were subsequently segmented into binary

235

images. The segmentation of OM pores in the 3D micrographs were based on an Otsu thresholding

236

algorithm34, while IntraP and InterP pores were segmented by an edge detection algorithm in Avizo

237

(due to depth of field issues with the larger size pores resolved by FIB-SEM35,36). Both pore

238

structure modeling and PNM were established for OM pores and IntraP/InterP pores in different part

239

of shales.

ACS Paragon Plus Environment

Page 12 of 30

Page 13 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

240 241 242 243

Figure 5. Illustration of the definitions of length and radius for pores and throats of PNM. 89 and 8: are the pore radii, while 8; is the radius of the connective throat. Terms