Research on Quantitative Analysis for Nano-pore Structure

porosimetry (MIP) and gas adsorption (GA) methods show good agreement of ... and capillary pressure curve, which all reflects the pore structure of th...
0 downloads 0 Views 6MB Size
Article

Research on Quantitative Analysis for Nano-pore Structure Characteristics of Shale Based on NMR and NMR Cryoporometry Zhiqing Li, Zhiyu Qi, Xin Shen, Ruilin Hu, Runqiu Huang, and Qian Han Energy Fuels, Just Accepted Manuscript • Publication Date (Web): 22 May 2017 Downloaded from http://pubs.acs.org on May 26, 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 21

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

Research on Quantitative Analysis for Nano-pore Structure Characteristics of Shale Based on NMR and NMR Cryoporometry Zhiqing Lia,b,c*, Zhiyu Qia,b, Xin Shena,b, Ruilin Hua, Runqiu Huangc and Qian Hand a

Key Laboratory of Shale Gas and Geoengineering, Institute of Geology and Geophysics, Chinese Academy of Sciences,

Beijing 100029,China b

University of Chinese Academy of Sciences, Beijing 100049, China

c

State Key Laboratory of Geohazard Prevention and Geoenvironment Protection (Chengdu University of Technology),

Chengdu 610059, China d

Shanghai Niumai Electronic Technology Co., LTD, Shanghai 200333, China

*Correspongding author. Email: [email protected]

Abstract: Shale gas is a key natural gas resource of very great significance in China. Based on the differences in pore structure characteristics between marine and continental shale in China, the Weiyuan marine shale (sample 1#), Jiao Shiba marine shale (sample 2#) and Yaoqu continental shale (sample 5# and 6#) were selected to study the pore structure characteristics and controlling factors using cold field emission scanning microscopy (FE-SEM) and nuclear magnetic resonance (NMR). Nuclear magnetic resonance cryoporometry (NMRC) was employed to represent nano-scale pore structure. This method can be extended to microns measurement combining nuclear magnetic resonance relaxation analysis to detect in detail the pore structure of shales under the different aperture scales. The smaller the test temperature gradient is, the finer the result of pore distribution is. Test results show decreasing porosity from sample 5#, 2#, 6#, and 1# to 4#. NMRC, low field nuclear magnetic resonance (LFNMR), mercury intrusion porosimetry (MIP) and gas adsorption (GA) methods show good agreement of pore distribution in their respective scope of application. NMR method results in a much better estimate for the total pore volume than the more common MIP and GA. Hence; the pore structure of the reservoir shale can be evaluated more accurately by combining NMRC, LFNMR with GA and MIP. Thus the nano-pores of continental shale (5# Yaoqu shale) are clearly better developed and will more likely have a higher commercial exploitation value than marine shale. Key words: marine shale, continental shale, pore distribution, nano-pores, nuclear magnetic resonance cryoporometry(NMRC), mercury intrusion porosimetry (MIP).

1. Introduction The potential of global oil and gas resource is tremendous, and the proportion of conventional oil/gas to unconventional oil/gas resource is about 2:8 [2]. Pore characteristics of shale containing a large number of nano-pores are an important index used to measure and evaluate reservoir quality[2]. Gas is stored mainly in the holes of organic matter and clay minerals as adsorbed gas, and as free gas compressed in other pores[3]. The main research methods of shale microstructure includes amongst others, optical and electron microscope [4],

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

casting thin sections analysis [5], focused ion beam scanning electron microscopy (FIB-SEM) [6], mercury penetration, nitrogen adsorption method [7], fluid injection method, nuclear magnetic resonance imaging[8], field emission scanning electron microscopy (FE-SEM) combined with the argon ion polishing technology [9], Nano-CT reconstruction technology and nuclear magnetic resonance (NMR) [10]. T2 spectrum distribution of low-field NMR is considered a quick and effective means for quantitative microstructural characterization including total porosity, clay bound water, capillary and free porosity, pore distribution and derived permeability[11]. NMR is a non-destructive technique which has been widely applied to test the porosity of tight reservoirs and relies on relaxation time analysis of a liquid that fills the pores and the enhanced relaxation that occurs in a liquid at the solid/liquid interface. Mercury intrusion porosimetry method (MIP) however, mainly reflects the interconnected pores and throat in a certain aperture range while NMR reflects the entire pores and throat volume in a certain aperture range, including the unconnected pores [12]. It has close correlation between NMR T2 distribution and capillary pressure curve, which all reflects the pore structure of the rock [13]. The T2 distribution of free water can help construct the capillary pressure curve in the elimination of the effect of membrane bound water on the T2 distribution [14] and the fractal dimension calculated by T2 distribution can be used to represent the physical property of rock [15]. The combination of capillary pore distribution and T2 distribution, which reflects super capillary pore distribution, can get a complete pore distribution of the core including super capillary pore [16]. The unfrozen water content of frozen soil is tested by NMR to obtain T2 spectrum and the distribution of unfrozen water content under different temperatures[17]. Nuclear Magnetic Resonance Cryoporometry (NMRC) is a novel pore distribution measurement technique developed at the University of Kent and at Lab-Tools Ltd[18-19]. NMRC is used widely in measuring median pore size and pore distribution of porous material, such as bone microstructure[20], silica-gel, silica and activated carbons[21], biodegradable polymer microparticles[22] and drug release of polymeric nanoparticles[23]. NMRC is a very effective method for determining the PSDs of UF membranes with asymmetric pore structures [24]. The pore distribution functions obtained from the freezing and melting data based on NMRC were found to be similar in shape and width, which indicates that melting and freezing

ACS Paragon Plus Environment

Page 2 of 21

Page 3 of 21

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

processes occur in controlled pore glasses on the same length scale of the pore structure [25]. NMRC measurements have been performed on porous rocks for a number of national and international companies and universities, and it may be one of the most appropriate techniques for studying small-pore rock systems such as porous shales [26]. Fleury et al compared the pore distribution by NMR relaxation and NMRC in shales [27]. However not much has been reported on detailed characterization of shale based on NMRC. T2 spectrum of low field nuclear magnetic resonance (LFNMR) is employed in this work to understand and assess the pore distribution of marine and continental shale in China. NMRC is used to further detect shale pore distribution in the different pore size scale to study the characteristics of pore structure differences, combining FE-SEM with GA and MIP. 2. Materials and Methods 2.1. Fundamental principles of LFNMR and NMRC methods Low field nuclear magnetic resonance (LFNMR) involves the process where hydrogen nucleus of nonzero magnetic moment spins to generate Zeeman splitting in the external magnetic field and resonates to absorb radio frequency radiation in a certain frequency. Under the action of external radio-frequency pulse B1, the hydrogen atom generates magnetic resonance to reach a stable high-energy state, and then the external radio-frequency pulse B1 disappears. At the same time, the hydrogen atom recovers to the state of magnetic moment before magnetic resonance generation. The whole process is called relaxation process. The time required is called relaxation time, including longitudinal relaxation time T1 and transverse relaxation time T2. The distribution of T1 and T2 spectrum reflects the pore distribution. The longer T2 value represents H proton content in large pores, and the shorter T2 value represents H proton content in smaller pores. NMR Cryoporometry (NMRC) on the other hand is a new, simple and rapid method to determine pore distribution for studying freezing-melting of fluid in porous materials by NMR observation of the phase transition behaviour [18,28]. The method involves freezing a liquid in the pores and measuring the melting temperature by nuclear magnetic resonance. Since the melting point is depressed for crystals of small size, the melting point depression gives a measurement of pore size by analyzing the NMR signal as a function of temperature. Part of the pore water freezes at the temperature predicted by the Thompson (Kelvin) equation and the rest of the water does not freeze in the sense that it does not assume the structure of

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

Page 4 of 21

ice on cooling below this freezing point. This bound water exhibits a distribution of correlation times, and information about the width of this distribution is obtained[28]. The Gibbs and Thompson equation can be combined to relate to decrease in melting temperature to crystal dimensions. For a liquid confined within a pore in which a crystal is forming, assuming the contact angle between liquid, solid and pore wall is 180o, the temperature reduction ∆Tm of the melting point is given by the Gibbs-Thompson equation [29]; ΔTm=Tm-Tm(x)=4σTm/xρΔHf

(1)

Where, σ is the surface energy at the liquid-solid interface. Tm is the normal melting point of the bulk material. Tm(x) is the melting point of a crystal of linear dimension x. ∆Hf is the bulk enthalpy of fusion. ρ is the density of the solid. x is the pore diameter. For a particular liquid, therefore, we can write ΔTm=kGT/x

(2)

Where, kGT is a calibration constant to be determined empirically [18], which represents for characteristic of the liquid. If the pores are filled with liquid, the melting temperature of the liquid Tm(x) is related to the pore size distribution by dv(x)/dx=(dv(x)/dTm(x))﹒(dTm(x)/dx)=(dv/dTm(x))﹒(kGT/x2)

(3)

Where, the pore volume v(x) is a function of pore diameter x. The volume of pores with diameter between x and x+∆x is (dv / dx) ∆x. NMR cryoporometry is based on the theory of the melting point depressions (MPD) of liquid confined within a pore, which are dependent on the pore diameter. A porous material, saturated with liquid, which is then cooled, will produce a distribution of melting temperatures that depends on the pore distribution by raising the temperature after freezing all the liquid. When the temperature rises, the ice inside the small hole melts first, and then the ice inside the big hole melts. The signal intensity of NMR is a function of temperature (Figure 1). The signal intensity will be the total liquid volume in the sample at any given temperature and this must be differentiated to provide a pore distribution (equation 3). The water content in the sample increases continuously. The temperature scale (x-axis) can be mapped onto a pore size scale using equation (3). This is due to the differences of the potential energy of the water in the different size of pores, which leads to the other on time pore water freezing and thawing process [17]. If the mass and density of the absorbate are known then the signal

ACS Paragon Plus Environment

Page 5 of 21

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

intensity (y-axis) can be converted into a calibrated pore volume. The primary role of NMR is to detect the solid-liquid transition. There is a large difference between the broad signal of a solid and the narrowed signal of a liquid [30]. Under the different temperature conditions, the different T2 spectra and the corresponding pore distribution can be obtained. The smaller the temperature gradient, the smaller the pore size space, and the finer the features of the pore distribution. It offers a unique nondestructive method of obtaining full pore distributions in the range 2 to 100 nm at any point within a shale sample.

Frozen pore

Figure 1. Principles of NMR cryoporometry

2.2. Advantages of nuclear magnetic resonance Mercury intrusion porosimetry (MIP) data reflects the minimum throat and the connectivity of the pore volume without the closed pores. When mercury enters the pores connecting the outside with fine throat, the volume of the small pores will be overestimated. The method tends to damage the pore matrix and probe the diameter of pore throats rather than the pores themselves. So it fails in non-destructive testing. In addition, when the sample has a complex pore structure, it is difficult and will take a long time to balance the pressure by using the gas adsorption (usually with nitrogen) method. Gas adsorption (GA) relies on the Kelvin relationship, concerning the change in the vapour pressure caused by this effect. Low field nuclear magnetic resonance (LFNMR) signals generated by the H atoms in the pores can be used to reflect the size and volume of all pore throats. And there is no material transfer in the test process. The method weakens the influence of the complexity of the pore structure to the test result. LFNMR can obtain the distribution of pores size from 2nm-100µm. Nuclear magnetic resonance cryoporometry (NMRC) is non-destructive, and can obtain the fine distribution of pores size from the 2nm-1µm range of different order of magnitude. The smaller the temperature gradient is, the more precise the pore distribution of different pore

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

Page 6 of 21

size is. NMRC returns an absolute signal that may be measured arbitrarily slowly, or in discrete steps, to obtain improved resolution or signal-to-noise (S/N) ratio [32]. The comparisons of the different methods are shown in Table 1. Table 1. Pore test methods comparison. Effective pore

Method

Sample weight

Sample size

Pore test range

MIP

1~2g

Less than Ø10.0mm×H15mm

3nm-400µm

100nm~100µm

1h

GA

3~8g

Less than Ø 6mm×H6mm

0.5~400nm

2~50nm

11h

LFNMR

1~2g

Less than Ø25mm×H30mm

2nm-100µm

2nm-10µm

10min

NMRC

1~2g

Less than Ø10mm×H25mm

2nm-1µm

2-100nm

3h

test range for shale

Test time

2.3. Sampling and test protocols Samples selected from key shale gas generating formations including the Weiyuan marine shale (sample 1#) in Sichuan, Jiaoshiba marine shale (sample 2#) in Chongqing, Yaoqu tuff (sample 4#) and Yaoqu continental shale (sample 5# and 6#) both in Shanxi (Figure 2) were used for this work. All the samples were parceled in vacuumed plastic films sealed with wax on the outer layer and subsequently used for thin section authentication analysis. A MicroMR12-025V low field nuclear magnetic resonance analyzer with resonance frequency of 11.854 MHz and equipped with a permanent magnet was used in the LFNMR experiment. Temperature was kept at less than 35℃ with a magnetic field intensity of 0.28T and a probe coil diameter of 25 mm. The shales were cut into less than Ø 25mm*H30mm sample size (Figure 3). The samples were vacuum pumped and saturated with kerosene (C15H32) for 8 hours because of the large number of clay minerals found in shale. The original samples and the kerosene saturated samples were tested separately to obtain the T2 spectrum, and subsequently demarcated with the standard sample to convert the nuclear magnetic signal intensity to parameters such as porosity, pore size distribution.

ACS Paragon Plus Environment

Page 7 of 21

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

Weiyuan Jiaoshiba Yaoqu

Figure 2. Location of sampling points.

Figure 3. LFNMR experiment samples

The NMRC experiment involved the use of a NMRC12-010V equipment with temperature control range of -35℃ to 0℃ and an echo time of 0.1ms. The equipment which is equipped with a permanent magnet has a probe coil diameter of 10mm and a sampling and principal frequency of 250 kHz and 11.05 MHz respectively. The width of the 90° and 180° pulse are respectively 2.8µs and 5.6µs while the refrigerating method is gas convection cooling. The shale samples were cut into sizes less than Ø10mm* H25mm (Figure 4) in addition to weight and density measurement. The samples were subsequently vacuum pumped for 8 hours and placed in distilled water (saturated) for 24 hours under pressure (20MPa). The porosity of the saturated samples was calculated by weighing samples using density balance and the saturated samples were wrapped with polytetrafluoroethylene (PTFE) membranes and inserted into the sample tube placed in the coil of NMR equipment. After the samples cooled to -35℃ until all the water was frozen in the cryostat for at least half an hour, the temperature rose gradually. T2 spectrum measurement (the amplitude of NMR signal from the liquid only) of the samples was undertaken to obtain the sample pore distribution with every 1℃ rise in the process from -35℃ to -10℃. The heat balance time is about 5min while the test time is about 3 to 5min. Furthermore, between -10℃ to 0℃, T2 spectrum measurement for every 0.5℃

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

Page 8 of 21

rise to obtain the sample pore distribution was conducted. With the final melting of all the ice in the pores, the T2 spectrum and the corresponding pore distributions were obtained under different temperature conditions.

Figure 4. NMRC experiment samples.

3. Results and Discussion 3.1. Thin section analyses and mineralogical composition Samples 2#, 5# and 6# have the characteristics of soiling hands. In addition, sample 1#, 2# and 4# contain mainly silt while up to 40% and 80% clay content were respectively observed in sample 5# and 6# with 6# containing many collophanes (Table 2). The X-Ray diffraction (XRD) mineral phases analyzed using the Japanese Neo-Confucianism X-Ray diffraction spectrometer (D/MAX-2400) shows that sample 1# possessed both the highest quartz content and brittleness coefficient (Table 3) while the quartz content and brittleness coefficient of 6# was observed to be the lowest although it had the highest pyrite content [2].

Table 2. Results of Thin section analyses No

1#

Location

Weiyuan

Name

Formation

Shale Marine

2#

Jiaoshiba

Shale

4#

Yaoqu

Tuff

Age

Structure type

features

Main

Dark grey

Silt 97%

Lower silurian

Aleuritic texture,

series,

massive structure

longmaxi

Aleuritic texture fine

Dark grey,

formation

laminated structure

soiling hands

Fine sand structure, massive structure

Component

Specimen

Dark grey

Silt 90%

Silt 90%

Secondary

Little clay and carbonaceous

Clay 5- 10%, collophane2%

Triassic 5#

Yaoqu

Shale

Continental

period, yanchang formation

6#

Yaoqu

Argillaceous silt structure, massive structure

Brown black, soiling hands

Clay

Argillaceous structure,

Brown black,

shale

tabular structure

soiling hands

ACS Paragon Plus Environment

Silt 60%

Clay 80%

Clay 40%, collophane < 1% Collophane 1520%, silt 2-5%

Page 9 of 21

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

Table 3. Results of Mineralogical composition (%). Clay

BI

minerals

brittleness

content

coefficient

Potassium No

Quartz

Albite

Calcite

Dolomite

Pyrite

Siderite

feldspar 1#

38.5

3.1

19.7

6.8

13.8

3.6

/

14.5

52.3

2#

36.3

0.9

8.6

6.4

5.5

5.5

/

36.8

42.7

4#

31.5

5.0

22.3

1.3

10.2

/

6.2

23.5

39.3

5#

28.6

0.8

15.4

/

/

/

/

55.2

34.1

6#

21.4

0.5

5.5

/

/

20.8

/

51.8

29.2

BI=quartz content/(quartz content + carbonatite content + clay content) 3.2. Pore structure characteristics and size distribution

3.2.1. T2 spectrum distribution The T2 spectra obtained from the sample tests are shown in Figure 5. NMR T2 spectrum of shale cores are categorized into four types: unimodal T2 spectrum (Type Ⅰ), bimodal T2 spectrum (Type Ⅱ),multimodal T2 spectrum (Type Ⅲ) and continuous single-peak T2 spectrum (Type Ⅳ). The different T2 spectrum types directly reflect various kinds of pore size distributions. NMR T2 distribution of sample 1#, Weiyuan shale (Figure 5a.) which is classified as Type Ⅲ has three discontinuous peaks. It reflects three kinds of pore distributions. The left peak of T2 spectral is much more significant than the other two, whose corresponding relaxation time is about 0.8ms. It belongs to micropore. The middle peak of T2 spectral has much lower amplitude with the relaxation time ranging from about 2ms to 9ms. This belongs to small hole. The right peak of T2 spectral which belongs to middle hole has the relaxation time ranging from about 30ms to 100ms. Based on FE-SEM, there are schistosity joints or bedding joints, various kinds of inter-connected pores or passing pores, primary intergranular pores and intergranular dissolution hole in sample 1# (Figure 6). Figure 5b shows that T2 distribution of sample 2# (Jiaoshiba shale) which is classified as Type Ⅱ, has two discontinuous peaks. It reflects two kinds of pore distributions. The left peak of T2 spectral represents micropore with corresponding relaxation time ranging from 0.02ms to 1ms. The pore size distribution for middle hole of right peak is 10-100ms. FE-SEM shows the pores are made up by schistosity joints or bedding joints, wedge-shaped pores, slit pores between particles (Figure 7a) and organic nanopores (Figure 7b).

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

Page 10 of 21

(b)

(a)

(c)

(d)

(e)

Figure 5. T2 spectral distribution of samples (a) sample 1#; (b) sample 2#; (c) sample 4#; (d) sample 5#; (e) sample 6#.

ACS Paragon Plus Environment

Page 11 of 21

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

Intragranular dissolution pore

Primary intergranular pore

Figure 6. Electron micrograph of sample 1#

Microfracture

A

(a)

(b)

Figure 7. Electron micrograph of sample 2#:(a) Magnified photo with micro fracture; (b) Magnified photo of area A with organic holes.

Figure 5c. shows that T2 distribution of sample 4# (Yaoqu tuff) which is classified as Type I, has a continuous peak. It reflects a variety of pore distributions and shows certain connectivity. The relaxation time is from 0.02ms-100ms. The pore distribution of the left peak is from 0.02ms to 4ms,which represents for micropore. The pore distribution of the right peak is from 4ms to 90ms,which represents for middle-big hole (1-100µm) based on the pore naming criterion [2]. FE-SEM photo displays a large number of siderite aggregate intergranular pores and various sizes of mold inner holes (Figure 8). Figure 5d. shows that T2 distribution of sample 5# (Yaoqu shale) which is classified as Type Ⅱ, has two discontinuous peaks. It reflects two kinds of pore distribution including the left peak corresponding to micropore whose pore distribution is from 0.02ms to 2ms, and the right peak corresponding to middle hole and big hole whose pore distribution is from 6ms to 200ms. FE-SEM photo displays the intergranular holes of pyrite aggregate and developed intergranular seams (Figure 9).

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

Internal intergranular pore in siderite aggregate

Figure 8. Electron micrograph of sample 4#.

Figure 9. Electron micrograph of sample 5#.

Figure 5e. shows that T2 distribution of sample 6# (Yaoqu shale) which is classified as Type Ⅳ, has continuous single-peaks. It reflects three kinds of pore distribution. The left peak with pore distribution ranging from 0.02ms to 2ms and corresponding to micropore, the middle peak with pore distribution ranging from 2ms to 20ms and corresponding to small hole, while the relaxation time of right peak ranges from 20ms to 1000ms with wider distribution. It indicates that some relatively large pores, fractures, and small caves can be well developed. According to FE-SEM, the holes include the intergranular holes of pyrite aggregate and developed intergranular seams (Figure 10). So based on T2 spectral distribution of NMR and FE-FEM of shale, the relaxation time ranges include 0.02~1ms, 1~10ms, 10~100ms, 100~1000ms respectively corresponding to micropore (2~100 nm), small hole (0.1~1µm), middle hole (1~10µm) and big hole (10~100 µm).

ACS Paragon Plus Environment

Page 12 of 21

Page 13 of 21

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

Intergranular pore in pyrite aggregate

A

Intergranular seam

(b)

(a)

Figure 10. Electron micrograph of sample 6#:(a) Magnified photo with intergranular pore; (b) Magnified photo of area A with intergranular seam.

3.2.2. Pore size distribution The T2 transverse relaxation time of NMR fluid can be expressed as the following (Coates et al, 2009). 1 T2

S

 ρ2  

V pore

 Fs

ρ2 rc

(4)

Where, ρ is the transverse surface relaxation rate of rock (µm/ms), which is used to characterize the rock property. S/V is the specific surface area of pore (cm-1). F is the geometric shape factor (for spherical void F is equal to 3, for columnar throat roar F is equal to 2) while  is the capillary radius. Equation (5) can be obtained from equation (4). rc  c2 T2

(5)

Where, c is transition coefficient for capillary radius and transverse relaxation time (c =ρ ×F . The value of ρ is equal to 33 µm/s generally[31]. T2 relaxation time reflects the chemical environment of hydrogen proton in the sample, which is related to the binding force and its degree of freedom of hydrogen proton. The binding degree of hydrogen proton is related to the internal structure of sample. T2 distribution shows the pore distribution without considering the effect of body relaxation and diffusion. In porous media, the larger the aperture is, the longer the relaxation time of hydrogen proton is in the hole. The smaller the aperture is, the stronger the binding degree of hydrogen proton is in the hole and the shorter the relaxation time is. It means the position of the peak is related to the size of aperture. The higher the peak is, the larger the aperture is. The area (size) of the peak is related to the number of corresponding apertures. The greater the peak area, the more the pore amounts corresponding to the aperture. Using equation (5),

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

the relaxation time can be transformed into pore radius and the signal amplitude corresponds to the pore distribution (Figure 11).

Figure 11. The pore size distribution of samples

As shown in Figure 11, the pore distribution of sample 1# is mainly concentrated in 2 nm ~ 70nm, 200nm ~ 1µm and 3µm ~ 10µm. The main pore range is from 2nm to 70nm by the numbers. The pore distribution of 2# is mainly concentrated in 2nm~100nm and 1µm~ 10µm. The main pore range is from 2nm to 100nm in terms of numbers. And the numbers of the pores in 2nm~ 100nm are more than the pores of 1# in 2nm ~ 70nm. The pore distribution of sample 4# is mainly concentrated in the 2nm~400nm and 400nm~10µm. The main pore range is from 2nm to 400nm by the numbers. And the numbers of pores size smaller than 100nm are the least in all the samples. The pore distribution of 5# is mainly concentrated in the 2nm~100nm and 500nm~30µm. The main pore range is from 2nm to 100nm in terms of numbers. But the numbers of pores from 2nm to 100nm are less than the pore number of 1# and 2#. The pore distribution of sample 6# is mainly concentrated in the 2nm~100nm, 200nm~1µm and 3µm~100nm while the main pore range is from 2nm to 100nm in terms of numbers. The number of pore from 2nm to 100nm is only less than those of sample 5#. In summary, in the range of 2nm~100nm, the number of pore decrease from sample 2#, 1#, 5# and 6# to 4#. The nanopore of samples 2# (Jiaoshiba shale) is the most developed, and it contains a large number organic pores (Figure 7b). Within 100nm~1µm pore size range, the pore number decreases from sample 1#, 4#, 6# and 5# to 2# while within 1µm~10µm pore size range, the pore number decreases from sample 4#, 2#, 5# and 6# to 1#.

ACS Paragon Plus Environment

Page 14 of 21

Page 15 of 21

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

3.2.3. NMRC test result analyses T2 spectrum distributions under different temperature are shown in Figure 12. It is obvious that the signal amplitude gradually increases with the temperature and the liquid content in the sample improves. So Gibbs-Thomson equation can be used to calculate the pore diameter according to the melting temperature as shown in equation (2). The liquid content of sample which represents the cumulative volume of pores can be obtained by the calibration curve between water content and signal intensity (Figure 13). So the variation trend of pore volume according the different aperture was acquired by differential distribution in equation (3) as shown in Figure 14.

Figure 12. T2 spectrum distribution of sample 2# under different temperature.

Figure 13. The calibration curve between water content and signal intensity

As shown in Figure 14, the porosity of sample 5# (Yaoqu shale) is the most developed, followed by sample 2# (Jaoshiba shale), sample 6# (Yaoqu shale) and sample 1# (Weiyuan

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

shale) while the pores of sample 4# (Yaoqu tuff) are the least. The pores of sample 2# are the most developed (ranging from 2nm to 10nm), followed by sample 5#, 1#, 6# and sample 4#. The organic holes of sample 2# provide more contribution for the pores distribution, and this distribution can be seen in Figure 6b.

Figure 14. Comparison of the pore volume distributions among different smples with pore diameter increasing.

And Within the scope of 10~50nm, the pores of sample 5# are the most developed, followed by sample 2#, 1#, 6# and sample 4#. The pores of sample 5# are the most developed in the range 50~100nm, followed by sample 6#, 1# , 2# and 4#. In the middle holes size range (between 100nm and 1µm), the pores of sample 6# are the most developed, followed by sample 1#, 5#, 4# and 2#. The intergranular holes of pyrite aggregate and developed intergranular seams can be seen in Figure 10 for sample 6#. So it is clear that the nanopores of continental shale in China, show better development that may make it have a higher commercial exploitation value than marine shale. 3.2.4. GA test result analyses N2 adsorption (GA) relies on measuring the amount of gas adsorbed by a porous solid as a function of pressure that is the sorption isotherm. Based on the capillary condensation principle, Kelvin equation is used to calculate diameter of micropore (2-100nm). Test results of N2 absorption are shown in Figure 15. Within the range of 2~10nm and 10~ 50nm, the pores of sample 2# (Jaoshiba shale) are the most developed, followed by sample 5#

ACS Paragon Plus Environment

Page 16 of 21

Page 17 of 21

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

(Yaoqu shale), sample 1# (Weiyuan shale), sample 6# (Yaoqu shale) and sample 4# (Yaoqu tuff) [2]. The distribution rules of pore ranging from 2~ 10nm are similar to the results by NMRC in Figure. 14. And in the range of 10 ~50nm, the pores of sample 5# are the most developed, followed by sample 2#, 1#, 6# and 4# (Figure. 14). The same order noticed in GA results (sample 2#, 5#, 1#, 6# and lastly 4#) from 10~50nm and 2~100nm is observed using LFNMR as shown in Figure 11. 3.2.5. MIP test result analyses Mercury intrusion porosimetry (MIP) can be used to measure the aperture size of sample and its distribution characteristics in the range of 100nm~100µm. The volumes of liquid mercury intruding into a porous sample were measured as a function of the applied external pressure [32]. The external pressure is inversely proportional to the net width of pores that the mercury can get into. The pressure readings can be converted into aperture using the Washburn equation.

(a)

(c)

ACS Paragon Plus Environment

(b)

(d)

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

Page 18 of 21

(e)

Figure 15. Nanometer pore distribution obtained by DFT model:(a)1# Weiyuan shale; (b) 2# Jiaoshiba shale; (c) 4# Yaoqu tuff; (d) 5# Yaoqu shale; (e) 6# Yaoqu shale.

Test results of MIP are shown in Figure 16 [2]. Within the range of 100nm~1µm, the pores of sample 1# (Weiyuan shale) are the most developed, followed by sample 4# (Yaoqu tuff), then sample 6# (Yaoqu shale), sample 2# (Jiaoshiba shale) and sample 5# (Yaoqu shale). In the range of 100nm ~1µm, the pores of sample 4# are the most developed, followed by sample 1#, sample 6#, sample 5# and sample 2# by LFNMR (Figure 11). Within the range of 1~10µm, the pores of sample 4# are the most developed, followed by sample 6#, sample 2#, sample 1# and sample 5#. In the same range from 1µm to 10µm, the pores of sample 4# are the most developed, followed by sample 6#, sample 2#, sample 1# and sample 5# by LFNMR (Figure 11).

Figure 16. Pore distribution characteristics of samples by mercury injection test.

ACS Paragon Plus Environment

Page 19 of 21

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

4. Conclusions N2 adsorption and NMRC show good co-linearity of pore size from 2~10nm and 10~50nm while N2 adsorption and LFNMR show good agreement in range of 10~50nm and 2~100nm. Furthermore, MIP and LFNMR also show good consistency in the middle hole (1~10µm) and the general trend in the small hole from 100nm to1µm. NMR method results in a much better estimate for the total pore volume than the more common MIP and GA. Especially, NMRC method offers a unique nondestructive method of obtaining full pore distributions in the range 2 to 100 nm at any point within a shale sample. The smaller the temperature gradient is, the more precise the different pore distribution is. And the method is relatively rapid, reproducible, and accurate (Mitchell et al,2008), and the results are consistent with those of the conventional GA method. NMRC, LFNMR and MIP, GA show good agreement of pore distribution in their respective scope of application. It is clear that the nanopores of continental shale (5# Yaoqu shale) are better developed, and may have a higher commercial exploitation value than marine shale. Continental shale gas exploration and development may lead to a new round of shale gas revolution. Hence, the pore structure of the reservoir shale can be evaluated more accurately by combining NMRC, LFNMR with GA and MIP.

Acknowledgements This research is financially supported by the Strategic Priority Research Program of the Chinese Academy of Sciences, Grant No. XDB10030100, XDB10030102; Youth Innovation Promotion Association, CAS (No. 2017092); National Natural Science Foundation of China, Grant No. 41672316, 41330643; Opening fund of State Key Laboratory of Geohazard Prevention and Geoenvironment Protection (Chengdu University of Technology), Grant No. SKLGP2016K011; China Postdoctoral Science Foundation, Grant No. 2016M590871 and CAS Key Technology Talent Program.

References [1] Zou C.N., Yang Z., Zhang G.S., 2014.Conventional and unconventional petroleum “orderly accumulation”: concept and practical significance. Petroleum Exploration and Development41,14-27. [2] Li Z.Q., Oyediran I.A., Huang R.Q., Hu F., Du T.T., Hu R.L., Li X., 2016. Study on pore structure characteristics of marine and continental shale in China. Journal of Natural Gas Science and Engineering, 33, 143–152. [3] Pan Z. J., Connell L. D., 2015. Reservoir simulation of free and adsorbed gas production from shale.

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

Journal of Natural Gas Science and Engineering, 22, 359-370. [4] Ji L. M., Qiu J. L., Xia Y. Q., 2012. Micro-pore characteristics and methane adsorption properties of common clay minerals by electron microscope scanning. Acta Petrolei Sinica, 33,249 - 256. [5] Chen J., Zhou G.Y., Zhao X.L., He C., 2005. Overview of study methods of reservoir rock pore structure. Special Oil and Gas Reservoirs, 12,11-14. [6] Hinai A.A., Rezaee R., Esteban L., Labani M., 2014.Comparisons of pore size distribution: A case from the Western Australian gas shale formations. Journal of Unconventional Oil and Gas Resources, 8, 1-13. [7] Xie X. Y., Tang H.M., Wang C.J., 2006. Contrast of nitrogen adsorption method and mercury porosimetry method in analysis of shale’s pore size distribution.Natural Gas Industry, 26, 100- 102. [8] Wang S., 2009. Analysis of rock pore structural characteristic by nuclear magnetic resonance. Xinjiang Petroleum Geology, 30,768 -771. [9] Li J.J., Yin J.X., Zhang Y.N., Lu S.F., Wang W.M., Li J.B., Chen F.W., Meng Y.L. ,2015. A comparison of experimental methods for describing shale pore features - A case study in the Bohai Bay Basin of eastern China. International Journal of Coal Geology, 152, 39-49. [10] Meng M.M., Ge H.K., Ji W.M., Shen Y.H., Su S., 2015. Monitor the process of shale spontaneous imbibition in co-current and counter-current displacing gas by using low field nuclear magnetic resonance method. J. of Nat. Gas Sci. Eng., 27, 336-345. [11] Rezaee R., Saeedi A., Clennell B., 2012. Tight gas sands permeability estimation from mercury injection capillary pressure and nuclear magnetic resonance data. J. Petroleum Sci. Eng., 88, 92-99. [12] Zhang, L.H., Guo J.J., Tang H.M., Liu J., Li Q. R., He J.Y., 2015. Pore structure characteristics of Longmaxi shale in the southern Sichuan Basin. Natural Gas Industry, 35, 22-29. [13] Li T.J., Li, Z.F.; Zhao, Y.C.; Guo,Z.H.; Wang, C.J. Study on pore structure consistency based on Nuclear magnetic resonance (NMR) and mercury intrusion method. Natural Gas Industry 2006, 26, 57-59. [14] He Y.D., Mao Z.Q., Xiao L.Z., Ren X.J., 2005. An improved method of using NMR T2 distribution to evaluate pore size distribution. Chinese Journal of Geophysics, 48, 373-378. [15] Zhang C.M., Chen Z.B., Zhang Z.S., Li S., Ling H.S., Sun T.D., 2007. Research on pore fractal structure of reservoir rock based on T2 distribution of Nuclear magnetic resonance. Journal of Oil and Gas Technology, 29, 80-86. [16] Yun H.Y., Zhao W.J., Liu B.K., Zhou C.C., Zhou F.M., 2002. Researching rock pore structure with T2 distribution. Well logging technology, 26, 18- 21. [17] Tan L., Wei C.F., Tian H.H., Zhou J.Z., Wei H.Z., 2015. Experimental study of unfrozen water content of frozen soils by low-field nuclear magnetic resonance. Rock and Soil Mechanics, 36,1566-1572. [18] Strange J. H., Rahman M. R., 1993. Characterization of porous solids by NMR. Physical Review Letters, 71, 3589- 3591. [19] Webber J. B. W., Anderson R., Strange J. H., Tohidi B., 2007. Clathrate formation and dissociation in vapor/water/ice/hydrate systems in SBA-15, sol–gel and CPG porous media, as probed by NMR

ACS Paragon Plus Environment

Page 20 of 21

Page 21 of 21

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

relaxation, novel protocol NMR cryoporometry, neutron scattering and ab initio quantum–mechanical molecular dynamics simulation. Magnetic Resonance Imaging, 25 , 533- 536. [20] Fantazzini P., Viola R. Alnaimi S. M. Strange J. H., 2001.Combined MR- Relaxation and MRCryoporometry in the study of bone microstructure. Magnetic Resonance Imaging, 19, 481–484. [21] Dore J.C., Webber J.B.W., Strange J.H., 2004. Characterisation of porous solids using small- angle scattering and NMR cryoporometry. Colloids and Surfaces A: Physicochem. Eng. Aspects, 241,191– 200. [22] Petrov O. V., Fur´o I., Schuleit M., Domanig R., Plunkett M., Daicic J., 2006. Pore size distributions of biodegradable polymer microparticles in aqueous environments measured by NMR cryoporometry. International Journal of Pharmaceutics, 309,157–162. [23] Gopinathan N., Yang B., Lowe J. P., Edler K. J., Rigby S.P., 2014. NMR cryoporometry characterisation studies of the relation between drug release profile and pore structural evolution of polymeric nanoparticles. International Journal of Pharmaceutics, 469,146–158. [24] Jeon J.D., Kim S.J., Kwak S.Y., 2008. 1H nuclear magnetic resonance (NMR) cryoporometry as a tool to determine the pore size distribution of ultrafiltration membranes. Journal of Membrane Science,309, 233– 238. [25] Petrov O. V., Furó I., 2010. A joint use of melting and freezing data in NMR cryoporometry. Microporous and Mesoporous Materials, 136,83– 91. [26] Webber J. B. W., Corbett P., Semple K.T., Ogbonnaya U., Teel W. S., Masiello C. A., Fisher Q. J., Valenza J. J., Song Y. Q., Hu Q. H., 2013. An NMR study of porous rock and biochar containing organic material. Microporous and Mesoporous Materials, 178, 94-98. [27] Fleury M., Fabre R., Webber J.B.W, 2015. Comparison of pore size distribution by NMR relaxation and NMR cryoporometry in shales. SCA, 25-36. [28] Overloop K., Vangerven L., 1993. Freezing Phenomena in Adsorbed Water as Studied by NMR. Journal of Magnetic Resonance, Series A, 101, 179-187. [29] Jackson C. L., McKenna G. B., 1990. The melting behavior of organic materials confined in porous solids. J. Chem. Phys, 93, 9002- 9011. [30] Mitchell J., Webber J. B. W., Strange J.H., 2008. Nuclear magnetic resonance cryoporometry. Physics Reports, 461, 1-36. [31] Coates G., Xiao L.Z., Pramm E.M., 2007. NMR logging principles and application.Translated by Meng, F.Y. , Beijing, Petroleum Industry Press, 36-39. [32] Petrov O. V., Furó I., 2009. NMR cryoporometry: Principles, applications and potential. Progress in Nuclear Magnetic Resonance Spectroscopy, 54, 97-122.

ACS Paragon Plus Environment