Subscriber access provided by University of South Dakota
Fossil Fuels
Nuclear-magnetic-resonance T1-T2 map division method for hydrogen-bearing components in continental shale Jinbu Li, Wenbiao Huang, Shuangfang Lu, Min Wang, Guohui Chen, Weichao Tian, and Zhiqiang Guo Energy Fuels, Just Accepted Manuscript • DOI: 10.1021/acs.energyfuels.8b01541 • Publication Date (Web): 16 Aug 2018 Downloaded from http://pubs.acs.org on August 19, 2018
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 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 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.
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 42 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
10
ABSTRACT:
11
There are many hydrogen-bearing components in shale, including kerogen, free oil,
12
adsorbed oil, free water, adsorbed water, and structural water. Measuring the content and
13
distribution of each component is important to understand the occurrence mechanism of shale oil.
14
The nuclear-magnetic-resonance (NMR) T1-T2 map can be used as a nondestructive technique to
15
distinguish hydrogen-bearing components in shale. In this paper, we examine the relaxation
16
characteristics of kerogen, shale, and clay minerals in continental shale under different oil or
17
water conditions using high-resolution low-field NMR instruments (frequency is 21.36 MHz,
18
echo time is 0.07 ms). The NMR T1-T2 map division method was established for each
19
hydrogen-bearing component. The relaxation characteristics of each component are as follows:
20
(1) Kerogen has the highest T1/T2 ratio, oil exhibits a higher T1/T2 than that of water, and the
21
mobility of water is greater than that of oil under saturated conditions. (2) The transverse
22
relaxation time of the free state is greater than the adsorbed state for oil and water. (3)
23
Intergranular pores of clay-rich continental shale shrink after saturation with water and result in
24
the main peak of the T2 value of free water at less than 1 ms, which differs from marine shale. (4)
25
Kerogen and structural water account for a large proportion of NMR signals in continental shale.
26
(5) The signals of some components in T1-T2 maps overlap because of the resolution limitation
27
of the NMR instrument. Organic matter abundance and oil saturation of shale, estimated by the
28
NMR T1-T2 map method, were in good agreement with the pyrolysis and distillation experiments,
29
which demonstrates the reliability of the NMR T1-T2 map division method for each
30
hydrogen-bearing component in continental 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
Keywords: NMR, T1-T2 map, shale, hydrogen-bearing component, kerogen
31 32
1.
INTRODUCTION
33
The low-field nuclear magnetic resonance (LFNMR), with magnetic strength less than 0.5 T
34
and frequencies of 2 MHz, 12 MHz, and 23 MHz, have been widely used to predict the
35
characteristics of reservoirs, including porosity,1,2 permeability,3,4 pore-size distribution,5,6 and
36
fluid saturation.7 At present, LFNMR with low echo time (TE) is often used to detect fluid in
37
nanopores (~2 nm)8–10 and to characterize the full pore-size distribution of shale reservoirs with
38
micro-nanopores.11–14 Components such as kerogen and clay structural water are also detected
39
under low-TE conditions (smaller than 0.1 ms).9,10,15 Because there are many types of
40
hydrogen-bearing components in shale (kerogen, free oil, adsorbed oil, free water, adsorbed
41
water, and structured water), and the resonance characteristics of each component are different, it
42
is necessary to establish a division method for the NMR signal distribution of each
43
hydrogen-bearing component in shale.
44
Previous studies mainly focused on the methods of signal-shielding16–18 and two-dimensional
45
NMR19–28 for the separation of NMR signals of hydrogen-bearing components in reservoirs. The
46
signal-shielding method involves immersion in MnCl2 or saturation with D2O, so that NMR can
47
only detect the signal of oil. The main problems of this method are the following: (1) the signal
48
of water may not be shielded in some small or dead holes where the MnCl2 or D2O cannot enter;
49
(2) Mn2+ can react with some clay minerals to destroy the pore structure,17,18 especially for
50
clay-rich continental shales; and (3) this method only separates the oil and water, but cannot
51
distinguish the kerogen signal. The two-dimensional NMR method is based on the transverse
52
relaxation (T2) and its link with other parameters, such as the diffusion coefficient (D),
ACS Paragon Plus Environment
Page 2 of 42
Page 3 of 42 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
53
longitudinal relaxation time (T1), and magnetic field gradient (G), to separate each
54
hydrogen-bearing component with different NMR characteristics. Previous studies have shown
55
the difficulty of measuring diffusion coefficients in nanoporous materials,28,29 and there are no
56
kerogen signals distinguished by the D-T2 method, therefore, the D-T2 method is not suitable for
57
identifying hydrogen-bearing components in shale oil reservoirs.
58
Recently, NMR T1-T2 maps provide better differentiation between the different
59
hydrogen-bearing phases. Many experiments were conducted on marine shales and sandstone
60
samples in this area.9,10,23–28 For example, Fleury performed NMR T1-T2 map studies on both the
61
kerogen at different maturities and on shale samples, and described the regions of water,
62
hydroxyls, and methane in the T1-T2 map for marine shales from Barnett and Fort Worth Basin.9
63
Washburn carried out NMR T1-T2 map experiments on four marine shale samples in the USA,
64
and used the T1-T2 map to identify the maturity of organic matter bearing in shale.10 Daigle
65
introduced a method based on the secular relaxation (T2sec) to distinguish the fluids through
66
plotting (T2sec versus T1/T2) for Bakken shales.19 Previous studies showed that a T1-T2 map
67
tested using a NMR instrument with high frequency (23 MHz) can differentiate each
68
hydrogen-bearing phase more easily than one with low frequency (2 MHz), that organic shale
69
cuttings that were oil saturated had higher T1/T2 ratios than those that were water saturated, and
70
the T2 of water in marine shales was greater than 1 ms.15,25,26 However, the above NMR T1-T2
71
map division methods established by different authors in previous studies are mainly aimed at
72
marine shales, and it is unclear whether their results are suitable for clay-rich continental shales.
73
In addition, the NMR signals of shale oil reservoirs were not systematically distinguished,
74
especially for the distinction between adsorbed oil and free oil, which are of great significance
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
75
Page 4 of 42
for evaluating the mobility of shale oil.
76
In our study, the NMR relaxation theory of hydrogen protons in different hydrogen-bearing
77
components in micro/nanoporous media was introduced. Second, continental shales from the
78
Shahejie Formation of Damintun Sag in the Bohai Bay Basin, China and minerals (i.e.,
79
montmorillonite) were selected and tested by high-resolution LFNMR T1-T2 map (frequency of
80
21.36 MHz and a TE of 0.07 ms, which can separate the hydrocarbon-bearing components very
81
well9,23,30,31) under the following conditions: original, extracted and dried states, saturated oil,
82
centrifugal oil, saturated water, centrifugal water, isolated kerogen, kerogen with adsorbed oil,
83
and clay minerals with different water contents. The NMR T1-T2 map division methods for
84
kerogen, adsorbed oil, free oil, structural water, adsorbed water, and free water in continental
85
shales were established. Finally, the T1-T2 map signal values of organic matter (e.g., kerogen,
86
adsorbed oil, and free oil) and water (structural and adsorbed water) were calculated and
87
compared with the results of pyrolysis, distillation, and X-ray diffraction (XRD) experiments to
88
verify the reliability of the NMR T1-T2 map division method of continental shale.
89
2.
THEORY
90
The two main relaxation mechanisms for NMR testing of porous media include longitudinal
91
relaxation (T1) and transverse relaxation (T2). For fluids with low viscosity (oil, water) in rock
92
pores, the relaxation time can be expressed as32,33
93
S 1 1 = + ρ1 T1 T1Bulk V
(1)
94
1 1 S Dγ 2 G 2TE 2 , = + ρ2 + T2 T2 Bulk V 12
(2)
95
where ρ1 and ρ2 are the longitudinal and transverse surface relaxivities, respectively, S is the pore
96
surface area, V is the pore volume, T1bulk and T2bulk are the longitudinal and transverse bulk
ACS Paragon Plus Environment
Page 5 of 42 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
97
relaxation times, respectively, D is the diffusion coefficient, γ is the gyromagnetic ratio, G is the
98
magnetic field gradient, and TE is the echo time. Pore fluid in a shale reservoir is mainly
99
dominated by surface relaxation, which is related to the pore size.34 With larger pore size,
100
samples have a longer fluid relaxation time.
101
However, the relaxation time of high-viscosity fluid (heavy oil) or solid materials (bitumen,
102
kerogen) is governed by intramolecular dipole-coupling interactions. For a spherical molecule,
103
the relaxation time can be expressed as35
104
1 8τ 2τ , = 2C + 2 2 T1 1 + 4ω 2τ 2 1 + ω τ
(3)
105
1 10τ 4τ = C 6τ + + , T2 1 + ω 2τ 2 1 + 4ω 2τ 2
(4)
106
where ω is the Larmor frequency, C is a constant, and τ is the correlation time, which can be
107
obtained by Debye's theory: τ =
108
4πµ a 3 3 kT
(5)
109
where µ is the viscosity of the fluid, a is the radius of the molecule, k is Boltzmann's constant,
110
and T is the absolute temperature.
111
The relationship whereby higher molecular size and viscosity of the fluid are evident with
112
the relevant time τ increases, T1 increases, and T2 decreases (Figure 1). Therefore, we can apply
113
this theory to laboratory analysis, whereby T1/T2 values are greater with higher fluid viscosity
114
and larger molecular size.24
115
3.
116
3.1 Samples
SAMPLES AND EXPERIMENTS
117
Sixteen continental shale samples were selected from a key shale oil well designated S352
118
in the Shahejie Formation of Damintun Sag, Bohai Bay Basin, China (Figure 2), which had
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
119
formed in a deep and semi-deep lakes facies sedimentary environment. The current exploration
120
results show that more than 50 wells in this area have high oil and gas contents and some wells
121
have obtained industrial oil flow, which have shown good prospects for exploration and
122
development of shale oil.37 In the Shahejie Formation of well S352, the lithology of the upper
123
part (3180–3240 m) is black shale, the middle part (3240–3280 m) is argillaceous dolomite, and
124
the lower part is interbedded with black shale and argillaceous dolomite, the vertical distribution
125
of each sample is shown in Figure 2c.
126
Before NMR experiments, the geochemical characteristics and mineral compositions of
127
each sample were obtained from some basic experiments, such as Rock-Eval, distillation, and
128
XRD. After surface cleaning and being powdered to 100 mesh, the shale samples were placed
129
into a Rock Eval 6 analyzer (Vinci Technologies SA, France), and first heated at a temperature of
130
300°C for 3 min and then to 650°C at a heating rate of 50°C /min. The organic geochemistry
131
parameters, such as total organic carbon (TOC), volatile hydrocarbon content (S1), the cracking
132
hydrocarbons (S2), and the pyrolysis peak temperature (Tmax) were then obtained. In addition,
133
each sample was also crushed to 200 mesh and mixed with ethanol, ground with a mortar and
134
pestle, and smeared on a glass slide. The XRD experiments were performed using a X'Pert PRO
135
diffractometer (Malvern Panalytical Ltd., UK) with Cu Kα radiation (40 kV, 30 mA) to analyze
136
all the minerals. The geochemical and mineral compositions of each sample are shown in Table 1.
137
In terms of geochemical characteristics, the organic matter abundance of the selected samples is
138
high, TOC ranging from 0.24% to 10.1%, S1 ranging from 0.22 to 5.54 mg/g, and S2 ranging
139
from 0.7 to 40.34 mg/g. In addition, the oil content after heavy oil recovery ranges from 1.07 to
140
15.42 mg/g, according to the recovery method followed by Wang,38 and the oil saturation is
ACS Paragon Plus Environment
Page 6 of 42
Page 7 of 42 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
141
between 13% and 64.5%. The organic matter types are I and II1, and the maturity index (Tmax)
142
are greater than 435°C, indicating that shale of the Shahejie Formation has reached the mature
143
stage (except for Sample No.7). The clay content of selected samples is greater than 30% (except
144
for Sample No.7), with the highest value being almost 60%. The carbonate minerals are
145
dominated by dolomite, the dolomite content of Sample No.7 reached up to 80%.
146
Sample No. 2 was used as an example to study the characteristics of NMR T1-T2 mapping
147
of each hydrogen-bearing component in shale. In order to eliminate the effect of petrophysical
148
and organic abundance on the NMR tests, shale and kerogen were selected from the same rock
149
sample. The shale sample was divided into four parts: the first part was original state, the second
150
part was used for the preparation of kerogen and kerogen with adsorbed oil, the third part was
151
used for the preparation of saturated oil and centrifugal oil states, and the fourth part was used
152
for the preparation of saturated water and centrifugal water states.
153
In addition, montmorillonite with a purity of 99.99%, selected from San Diego County, Otay,
154
California, was used to study the NMR T1-T2 map of clay minerals. During the measurement
155
process, we controlled the water state by changing the heating temperature. Previous studies
156
have shown that the water type of the Shahejie Formation in Damintun Sag is NaHCO3 and its
157
total salinity is less than 9 g/L.39 In our experiments we used this salinity to simulate the water
158
characteristics of the geological conditions.
159
3.2 Experiments
160
(1) NMR experiment: The NMR experiment was carried out on a MicroMR23-060H-1
161
instrument (Shanghai Niumag, China) operated at 21.36 MHz and equipped with a 25.4 mm
162
probe. The measurement parameters of the NMR T1-T2 map were set as follows: waiting time,
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
163
1000 ms; number of echoes, 6000; echo times, 0.07 ms; number of scans, 64. The parameters
164
used in this study were chosen to detect the fluid in the nanopores and make it easier to
165
distinguish the hydrocarbon-bearing components, and were proved by other scholars researching
166
marine shales.9,30,31 After measurements, the Inversion Recovery Carr-Purcell-Meiboom-Gill
167
(IR-CPMG) sequence was detected.
168
(2) Other experiments:
169
a) Chloroform extraction experiment: The shale samples were extracted by chloroform for
170
240 h using the Soxhlet extraction method to remove the oil; the extraction temperature was set
171
at 85°C. After extraction, the shale samples were dried at 60°C for 24 h and stored in a
172
desiccator.
173
b) Pressure saturation experiment: The chloroform-extracted shale samples were dried at
174
315°C for 24 h to remove the free and adsorbed water, and then the sample was placed into a
175
pressure saturator. Air was evacuated from the sample for 24 h, then saturated fluid was slowly
176
poured onto the shale samples, and the samples were placed in a high-pressure vessel. The
177
saturation time was 48 h, the pressure was 15 MPa; the saturation fluids were separately oil
178
(dodecane) and water (NaHCO3).
179
c) Centrifuge experiment: Samples were run for 6 h in a CSC-12 centrifuge rotor, which had
180
a radius of 13.5 cm and a speed of 10,000 rpm (converted into centrifugal force, 2.76 MPa), at a
181
temperature of 20°C.
182
(3) Preparation of kerogen and kerogen with adsorbed oil: The shale samples were
183
crushed to 100 mesh and soaked in distilled water for 4 h, followed by acid treatment (using 6
184
mol/L hydrochloric acid, 6 mol/L hydrochloric acid, and 40% hydrofluoric acid), alkaline
ACS Paragon Plus Environment
Page 8 of 42
Page 9 of 42 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
185
treatment (0.5 mol/L sodium hydroxide), pyrite treatment (6 mol/L hydrochloric acid and
186
arsenic-free Zn particles). Dichloromethane was added and stirred until the dichloromethane
187
volatilized, and then kerogen with adsorbed oil was obtained. Kerogen can be obtained by
188
chloroform extraction of the kerogen with adsorbed oil.
189
(4) Experimental test flow: The NMR instrument (TE of 0.07 ms) can detect the signal of
190
hydrogen-bearing components such as water, kerogen, oil, structural water, etc. In this study, the
191
targets of the NMR experiments were kerogen, kerogen with adsorbed oil, shale, extracted and
192
dried shale, saturated water, centrifugal water, saturated oil, centrifugal oil, and clay minerals
193
under different water conditions. The experimental testing process is illustrated in Figure 3.
194
3.3 Challenges and assumptions
195
(1) Challenges and solutions: There are three experimental challenges in performing
196
measurements: heterogeneity of samples, the T1-T2 map of kerogen with adsorbed oil, and testing
197
parameters. To overcome these challenges, shale and kerogen first must be selected from the
198
same sample to eliminate the effect of the petrophysics and organic abundance of samples on
199
NMR results. In addition, it is difficult to directly test the NMR signal of adsorbed oil since there
200
are also some NMR signals from shales and kerogen. In our study, the T1-T2 map differences
201
between the kerogen with adsorbed oil and kerogen were used to calculate the T1-T2 map of
202
adsorbed oil. Finally, the same NMR parameters must be used for testing kerogen, shales, and
203
minerals under different states to eliminate the effect of parameters on NMR signals; in
204
particular NMR instruments with a high frequency (greater than 20 MHz) and low TE (less than
205
0.07 ms) should be used to differentiate the components with short relaxation time, such as
206
kerogen and structural water.
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
207
(2) Assumptions: Fluid properties, mineral species, kerogen types and maturities, and shale
208
pore-structure characteristics will affect the NMR signal intensity and distribution. In our study,
209
the following assumptions were made in the NMR experiments: First, we used NaHCO3
210
solutions to represent water based on the type and salinity of the formation water in the research
211
area. As observed in previous studies, there is a small difference between dodecane and crude oil
212
in NMR signals,25 we used dodecane to represent oil. Second, as the main type of clay minerals
213
in the research area, montmorillonite was used to represent clays to study the water signals of the
214
NMR T1-T2 map, and the water state was controlled by changing the heating temperature. Third,
215
there is no obvious difference between the adsorption oil and swelling oil in the kerogen in NMR
216
spectroscopy, so we assumed that all of the oils were adsorbed oil and did not research swelling
217
oil in this study. Finally, in order to subtract the effect of kerogen and shale properties, kerogen
218
and shales were taken from the same sample in the research area, and we assumed that they can
219
represent the NMR characteristics of the continental shale in the oil window.
220
4.
221
4.1 Bulk fluid signal
RESULTS AND DISCUSSION
222
Various volumes, e.g.,0.2, 0.4, 0.6, 0.8, and 1.0 mL, of oil (dodecane) and water (NaHCO3
223
solution) were tested by NMR T1-T2 experimentation to develop a T1-T2 map signal distribution
224
of water and oil in bulk state and to determine the quantitative relationships between signal
225
intensity and its volume/mass. The results of these tests are shown in Figure 4 (0.4 mL oil and
226
water as a case study), which illustrate that the T1/T2 ratio of dodecane and water in the bulk state
227
is approximately equal to 1. The relaxation time of water is slightly longer, with a signal in the
228
range 1000
