Quartz cement origins and impact on storage performance in Permian Upper Shihezi Formation tight sandstone reservoirs in the northern Ordos Basin, China

Quartz cement origins and impact on storage performance in Permian Upper Shihezi Formation tight sandstone reservoirs in the northern Ordos Basin, China

Accepted Manuscript Quartz cement origins and impact on storage performance in Permian Upper Shihezi Formation tight sandstone reservoirs in the north...

14MB Sizes 0 Downloads 2 Views

Accepted Manuscript Quartz cement origins and impact on storage performance in Permian Upper Shihezi Formation tight sandstone reservoirs in the northern Ordos Basin, China Dengke Liu, Wei Sun, Dazhong Ren, Changzheng Li PII:

S0920-4105(19)30300-6

DOI:

https://doi.org/10.1016/j.petrol.2019.03.061

Reference:

PETROL 5911

To appear in:

Journal of Petroleum Science and Engineering

Received Date: 5 September 2018 Revised Date:

23 February 2019

Accepted Date: 20 March 2019

Please cite this article as: Liu, D., Sun, W., Ren, D., Li, C., Quartz cement origins and impact on storage performance in Permian Upper Shihezi Formation tight sandstone reservoirs in the northern Ordos Basin, China, Journal of Petroleum Science and Engineering (2019), doi: https://doi.org/10.1016/ j.petrol.2019.03.061. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Quartz cement origins and impact on storage performance in

2

Permian Upper Shihezi Formation tight sandstone reservoirs in the

3

northern Ordos Basin, China1

4

Dengke Liu a, b, *, Wei Sun a, Dazhong Ren c, Changzheng Li d

5

Department of Geology, State Key Laboratory of Continental Dynamics, Northwest University, Xi’an 710069, China

School of Mining and Petroleum Engineering, Department of Civil and Environmental Engineering, University of Alberta, Edmonton, AB T6G 1H9, Canada

SC

b

c

Shaanxi Key Laboratory of Advanced Stimulation Technolgy for Oil & Gas Reservoirs, College of Petroleum

d

No.8 Oil Production Plant, Changqing Oilfield Company, PetroChina, Xi'an, 710021, China.

M AN U

Engineering, Xi’an Shiyou University, Xi’an 710065, China

TE D

Abstract: Authigenic quartz is an important cementing material that can degrade storage capacity, whereas the effect of quartz cement on microscale pores and throats in tight sandstone is controversial. Therefore, it is crucial to verify the sources of quartz cement as well as the controls on microscopic storage. Toward this end, a variety of tests were performed on Permian Upper Shihezi Formation sandstones, and they are the vital exploration and development interval for gas in the northern Ordos basin. We found that quartz cement is the most abundant interstitial mineral in the Upper Shihezi Formation tight sandstones, which forms at approximately 55 to 188 with a continuous process. All the samples can be divided into pore dominated and throat dominated types, and there is an increase in micropores and a decrease in macropores as the quartz cement content decreases. Chemical compaction and transformation of clay minerals were the main sources of silica. This work shows that although quartz cement would occupy the void space and lead to a loss of porosity, it could retard the compaction and preserve the pores, which is attributed to the limited compressibility of quartz cement supported rocks. Thus, the difference in the pore radius and throat radius is diminished, resulting in the decrease in pore-throat size heterogeneity in tight sandstones. Keywords: Quartz cement; Sources of silica; Storage performance; Upper Shihezi Formation; Ordos basin

EP

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

a

AC C

6 7 8 9 10 11 12

RI PT

1

1. Introduction Currently, research priorities in China for sandstone reservoirs, have evolved from predominantly conventional to unconventional sandstone reservoirs, with a decrease in the remaining reserves in the former (Zou et al., 2013; Jia et al., 2016; Rui et al., 2018). As the representative type of unconventional sandstone reservoir, tight sandstones gradually play a significant role in increasing reserves and stabilizing productivity due to their abundant potential 1 *

Corresponding author. 8510-111st nw, Edmonton, Canada. E-mail address: [email protected]

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

(Zou et al., 2012). Quartz cement is volumetrically important in the post deposition processes of tight sandstones, and it is one of the main cements that destroy the storage capacity in siliciclastic sandstones during deep burial. Therefore, the origin and main controls on the precipitation of quartz cement must be studied for storage performance assessment (Ehrenberg, 1990; Bjørlykke and Egeberg, 1993; Molenaar et al., 2007; Lander et al., 2008; Xi et al., 2015a, 2015b; Lai et al., 2018). Quartz cement originates from various silica sources. The external sources are unimportant since mass transport between sandstones and adjacent mudstones is limited by the low solubility of silica in formation water (Bjørlykke and Jahren, 2012). Hence, the internal sources, such as biogenic silica (mainly sponge spicules), chemical compaction and mineral redistribution, are the primary contributors to quartz cement precipitation (Bjørlykke and Egeberg, 1993; Worden and Morad, 2009; Hyodo et al., 2014). Although much research has been devoted to elucidating the sources of silica and origins of quartz cement (Kim and Lee, 2004; Worden and Morad, 2009; Xi et al., 2015b), the reactions and redistribution of quartz cement in tight sandstones and its impact on storage performance are still not completely understood (Peltonen et al., 2009; Xiao et al., 2018). In addition, the existing research has shown that quartz cement is one of the most important interstitial minerals in the Permian tight sandstones of the Upper Shihezi Formation in the northern Ordos Basin, China (Yang et al., 2008). Therefore, the authigenic quartz might be one of the most significant factors controlling storage performance in this study area. Quartz cement precipitated in pores can result in pore size distribution changes from primary pores blocked by the voids, and this can lead to a significant change in storage performance (Colón et al., 2004; Kim and Lee, 2004; Mousavi and Bryant, 2012). The aggregated authigenic quartz particles may also be porous and contain multiscaled pores, which could provide reservoir space. Therefore, the controversial questions about the origin of quartz cement and how quartz cement influences storage performance in tight sandstones need to be addressed. The following studies were carried out on thin sections by conducting scanning electron microscopy (SEM), cathode luminescence (CL), physical properties tests, X-ray diffraction (XRD), fluid inclusion tests, rate-controlled mercury porosimetry (RCP) and nuclear magnetic resonance (NMR) on the typical tight sandstone samples from the Permian Upper Shihezi Formation in the northern Ordos Basin, China. First, we used thin sections, SEM and CL images to identify the content and distribution of interstitial minerals, especially the quartz cement, among the space between the detrital grain particles, and then we used XRD to quantitatively evaluate the amounts of detrital grains and interstitial minerals. The pore morphology was also discerned by these optical observations. Second, we used the electronic probe tests to distinguish the quartz grains and the quartz overgrowth, compared their difference in chemical composition, and then conducted the fluid inclusion experiments to determine the formation temperatures of the quartz cement. RCP and NMR were also performed to ascertain the pore volumes and pore-throat size distribution. Finally, the sources of silica were discussed and the impacts of quartz cement on porosity and pore-throat size distribution were studied, and the implications of quartz cementation for tight sandstone storage were briefly analyzed.

AC C

38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81

2. Geological setting The Ordos Basin is a typical cratonic basin with lacustrine deposits (Yang et al., 2005; Luo et al., 2009). The basin located in northwest China and can be subdivided into six first class tectonic zones, namely, the Yimeng Uplift, Jinxi Fault-Fold Belt, Weibei Uplift, Tianhuan Depression,

ACCEPTED MANUSCRIPT

RI PT

Xiyuan Thrust Belt and Shanbei Slope Zones (Fig. 1 (a); Gao and Li, 2015; Huang et al., 2018). The research area, located in north Ordos Basin, is one of the most gas-rich areas in the Yimeng Uplift, Shanbei Slope and Tianhuan Depression Zone (Fig. 1 (a); Gao and Li, 2015; Huang et al., 2018). The Upper Shihezi Formation constitutes the middle strata of Permian rocks and consists mostly of grayish-green fine- to coarse-grained sandstones with intercalated limestones (Fig. 1 (b); Dai et al., 2005; Jia et al., 2017). The Upper Shihezi Formation is approximately 130-160 m, conformably overlies the Lower Shihezi Formation and is overlain conformably by the Shiqianfeng Formation (Fig. 1 (b); Dai et al., 2005; Jia et al., 2017). The depositional facies of the Upper Shihezi Formation in the Ordos Basin have been well studied by scholars, who believe that the sedimentary facies should be interpreted as fluvial (Yang et al., 2014). Previous research indicated that the geothermal gradient was approximately 30.3 /km when the mean lithostatic pressure coefficient was 0.86 (Yang et al., 2014). All samples were selected from the Upper Shihezi Formation.

95 96 97 98 99 100 101 102 103 104 105 106

AC C

EP

TE D

M AN U

SC

82 83 84 85 86 87 88 89 90 91 92 93 94

Fig. 1. (a) Location map of the research area with inset map of the Ordos Basin region (upper right) (modified from Gao and Li, 2015; Huang et al., 2018) and (b) stratigraphic column from the Ordovician Majiagou Formation to Triassic Liujiagou Formation in the northern Ordos Basin (modified from Dai et al., 2005; Jia et al., 2017).

3. Experimental methods and data processing This study focuses on the Upper Shihezi Formation in the northern Ordos Basin where exploration and development well data are abundant for this study. All samples are from approximately 2.5-cm-diameter and 6.0-cm-length cylindrical cores drilled perpendicular to the bedding, which were dried at 110 for 24 h. More than 2000 helium porosity and nitrogen permeability data points under an ambient pressure of 20 MPa were collected. Three hundred thirty-two polished thin section samples, stained by Alizarin Red S and potassium ferricyanide,

ACCEPTED MANUSCRIPT were prepared for studying the petrology, diagenesis and pore systems in the research area by point counting at least 300 counts to check the validity of the sandstone composition data. An FEI Quanta 400 FEG scanning electron microscope (SEM) was used to determine the authigenic quartz, carbonate, clay minerals and visual pore characteristics for 233 samples. Cathode luminescence (CL) analyses for 46 samples were conducted with a Zeiss microscope equipped with a Gatan MonoCL3+ instrument. Twenty-nine samples were prepared for X-ray diffraction tests using an X’Pert PRO energy dispersive X-ray spectrometer. Elementary composition evaluations of minerals by electron probe analyses were done for 5 representative samples with 10 points using the JXA-8230 Electron Probe Microanalyzer, which enables detection of elements in the range of B5 to U92. In addition, seventy-eight samples were counted as doubly polished rock slices for homogenization temperatures (Th) of fluid inclusions within quartz overgrowth and the temperatures of phase transitions between -196 and 600 using a LINKAM THMS600 heating and cooling stage. The measured temperature standard deviation for Th is ±0.1 . Based on petrological and geochemical studies, in order to figure out the effects of quartz cement on storage performance, 12 samples with relatively rich quartz cement and trace amounts of interstitial minerals were selected and saturated with simulated formation water with CaCl2 salinity of 25,000 mg/L. The T2 spectrum derived from NMR was then measured in a Niumag NMR apparatus. The transverse relaxation time T2 satisfied the equation: T = · , where ρ, r and c

125 126 127 128 129 130 131 132 133

represent the surface relaxivity, pore radius and pore shape, respectively (Daigle and Johnson, 2016). ρ is a key calibrated parameter for T2 to convert r, which has to do with the relaxing strength of minerals, and c is equal to 2 because the pores in the samples were assumed to be cylindrical in the process of calibration (Saidian and Prasad, 2015; Li et al., 2015; Rosenbrand et al., 2015). After the NMR tests, twelve samples were redried, and then RCP tests were done on an ASPE 730 mercury porosimeter at a constant rate of 5×10-5 mL/min and maximum pressure of 6.2 MPa. The RCP tests were able to distinguish pores and throats by analyzing the increasing and decreasing trends of the intrusion pressure (Yuan and Swanson, 1989). The pore/throat radius can . be simplified as P = = (Washburn, 1921), where the contact angle between mercury and

134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150

the pore surface is assumed to be 140° and the mercury interfacial tension is set to 0.48 J/m2 in tight sandstones.

EP

TE D

M AN U

SC

RI PT

107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124

AC C

4. Results 4.1. Sandstone composition and texture The sandstones in the Upper Shihezi Formation are fine to coarse-grained and are generally texturally mature because of the relatively high proportion of well-sorted grains (Fig. 2 (a)). The roundness of the grains varies from subangular to subrounded (Fig. 2 (b)), and the grains contact mainly with line and point contacts (Fig. 2 (c)). Sutures can be found as well (Fig. 2 (d)). The sandstones in the Upper Shihezi Formation are mostly litharenite and sublitharenite (Fig. 2 (e); Folk, 1980). Point counting data reveal relative contents of quartz grains between 40.26% and 95.93% (av. 81.27%). Feldspars range from trace amounts to 4.66% (av. 1.75%), and rock fragments are between 4.07% and 58.44% (av. 16.98%). The major quartz grains are predominantly monocrystalline, and the most of the feldspar grains are orthoclase and plagioclase (Fig. 3 (a), Fig. 4 (g)). Dissolution is the major factor for the relatively low percentage of feldspar (Fig. 4(a, i)). The rock fragments consist of volcanic (av. 5.78%), metamorphic (av. 3.69%), and sedimentary rocks (av. 0.43%). The mineral component reveals that sediment provenance showed

ACCEPTED MANUSCRIPT little change during the Upper Shihezi depositional stage in the research area. Bulk sandstone XRD data reveal that the relative quartz content ranges from 57.89% to 96.58% (av. 81.66%), orthoclase ranges from 0.34% to 4.21% (av. 1.62%), plagioclase from 0.45% to 2.75% (av. 1.39%), calcite from 0.18% to 8.85% (av. 2.33%), and clay minerals were between 2.46% and 33.16% with an average of 12.99% (Fig. 3(a)). The contents of all the bulk minerals are variable and contain insignificant trends with the burial depth except calcite, which has a slightly negative correlation with burial depth (Fig. 3(a)).

SC

RI PT

151 152 153 154 155 156 157

158

Fig. 2. Petrological features in the Upper Shihezi Formation: (a) Sorting distribution; (b) roundness

M AN U

159 160 161 162 163

distribution; (c) thin section photomicrograph showing grain contact types; (d) thin section photomicrograph showing the suture; (e) ternary diagram showing the grain content of the Upper Shihezi Formation in the northern Ordos Basin (using Folk’s (1980) classification). PC-point contact; LC-line contact; Su-suture; N-the number of samples.

Proportion, %

(a)

(b)

Proportion, %

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

2500 2600

2700

2700

2800 2900 3000 3200 3300 3500 3700 3800 3900 4000 4200

3000 3100 3200 3300 3400 3500 3700 3800 3900 4000

Chlorite

Calcite

4100

2900

3600

3600

kaolinite

Plagioclase

EP

3400

2800

Depth, m

Depth, m

Orthoclase

3100

AC C

2400

2600

Illite/smectite mixed layer

Quartz

TE D

2500

Illite

Clay

2400

4100 4200

164 165 166

XRD: (a) bulk composition; (b) clay composition.

167 168 169 170 171 172 173 174

Kaolinite is the most abundant clay mineral and mainly occurs as booklet and vermicular aggregates (Fig. 3(b), 4(b, d)). The sandstone clay XRD data show that the relative amount of kaolinite ranges from 3.17% to 80.33% with an average of 46.96%, and it is slightly more abundant in the relatively deep sandstones (Fig. 3(b)). Illite occurs as honeycomb crystals and occasionally bridges the pores (Fig. 4(c, e)). The clay XRD data reveal a relative illite content between 6.79% and 53.46% with an average of 25.45% and an insignificant trend with burial depth (Fig. 3(b)). Mixed-layer illite/smectite interstratified with R=3 is a minor phase with a

4300

4300

Fig. 3. Vertical mineralogical composition distribution of Upper Shihezi Formation tight sandstones based on

ACCEPTED MANUSCRIPT relative amount between 0.18% and 21.18% and average of 6.78%, and it is variably present throughout the Upper Shihezi succession (Fig. 3(b), 4(d)). Rosette-shaped chlorite is also a vital clay mineral, and its relative content ranges from 5.21% to 35.36% with an average of 20.80% (Fig. 3(b), 4(e)). The chlorite clay XRD data suggest that the greatest abundance is over the 2400-3300 m depth range with slightly negative correlation to burial depth (Fig. 3(b)).

193 194 195 196

Fig. 4. Optical observation in the Upper Shihezi Formation tight sandstones: (a) Thin section photomicrograph

EP

181 182 183 184 185 186 187 188 189 190 191 192

showing feldspar dissolution and microcrystalline quartz; (b) SEM image showing kaolinite; (c) SEM image showing illite and chlorite; (d) SEM image showing illite/smectite mixed layer and kaolinite; (e) SEM image showing chlorite and illite; (f) thin section photomicrograph showing detrital quartz and quartz cement (red dashed

AC C

180

TE D

M AN U

SC

RI PT

175 176 177 178 179

lines represent the boundary); (g) SEM image showing quartz cement; (h) CL photomicrograph showing detrital quartz (light luminescent) and quartz cement (dark nonluminescent) with a suture; (i) thin section photomicrograph showing interparticle pores connected with dissolution pores; (j) thin section photomicrograph showing rock fragment dissolution and feldspar kaolinization; (k) thin section photomicrograph showing microcracks, interparticle pores and feldspar kaolinization; (l) thin section photomicrograph showing ferrocalcite. FD-feldspar dissolution; MQ-microcrystalline quartz; Ch-chlorite; Ka-kaolinite; Il-illite; I/S-illite/smectite mixed layer; QD-detrital quartz; QC-quartz cement. IP-interparticle pore; RFD-rock fragment dissolution; FK-feldspar kaolinization; MC-micro crack; Fc-ferrocalcite.

4.2. Quartz cementation The thin section studies show the presence of abundant quartz, and the quartz cement usually extends into pores occurring as elongate crystals that are easy to distinguish from the detrital

ACCEPTED MANUSCRIPT

M AN U

SC

RI PT

quartz due to their dust rims (Fig. 2(d), 4(f)). For the SEM observations, the authigenic quartz that partly or completely occludes the voids is approximately 40-110 µm in size with euhedral microcrystals (Fig. 4(g)). Furthermore, in the CL images, the detrital quartz grains are relatively light and luminescent and the quartz cement is dark and nonluminescent without zonation (Fig. 4(h)), which can be regarded as reliable evidence of a closed system (Molenaar et al., 2007). The point counting shows that the absolute quartz cement content ranges from a trace amount to over 15% with an average of 7.87%. This means that quartz cement is the most abundant interstitial mineral in the Upper Shihezi Formation sandstone. The electron probe analyses were performed on 5 samples with 10 analyzed points for the purpose of comparing the elementary differences between detrital quartz and quartz cement (Fig. 5, Table 1). The results show that Si4+ is the most abundant ion and that Al3+ is slightly enriched in detrital quartz, whereas other ions are in trace amounts (Fig. 5, Table 1). The precipitation temperature of quartz cement can be induced from fluid inclusions (Robinson and Gluyas, 1992). In this research, 78 quartz overgrowth fluid inclusion homogenization temperatures (Th) were measured (Table S1). The diameters were between approximately 1.3 and 11.3 µm and were irregular, triangular or elliptoid (Fig. 6 (a), Table S1). The majority of the fluid inclusions consist of liquid and gas phases, and over 85% of samples had gas liquid ratios less than 30% (Table S1). Figure 6 (b) illustrates the homogenization temperature distribution, which record the formation temperature of quartz cement, and the results reveal a Th of the fluid inclusions between 53.7 and 188 , predominantly centered in the range of 110 to 130 , which means that quartz cementation was a continuous process. As discussed in the preceding section, the onset of quartz cementation at approximately 55 and the slightly enriched Al3+ ions in detrital quartz correspond with observations and calculations made for other basins, for instance, the Songliao Basin in China (Xi et al., 2015b).

221

AC C

EP

TE D

197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220

222 223 224

Fig. 5. Micrographs and elemental distributions of the electron probe analysis: (a) micrograph of SEM showing the quartz cement; (b) elemental distribution of the quartz cement; (c) micrograph of SEM showing the detrital quartz;

ACCEPTED MANUSCRIPT 225

(b) elemental distribution of the detrital quartz. (red color cross and box represent the analyzed position).

226 30

Frequency, %

(a)

-15

-10

Y (um)

-5

0

FI

5

(b)

20

10

10

2 µm

15

0 50

20

227

10

70

90

110

130

150

170

190

Homogenization temperature,

20

Fig. 6. Microthermometric tests in Upper Shihezi Formation tight sandstones: (a) Micrograph of fluid inclusion

SC

under transmitted light at 25 ℃; (b) histogram of homogenization temperature for fluid inclusions in quartz cement. FI-fluid inclusion.

Table 1

The electron probe data of the Upper Shihezi Formation tight sandstones. -: no available data. Well

E31

Depth, m 3800

3954

L29

3671

L53

3943

L7

4155

Results, %

+

Na

Detrital quartz

-

Quartz cement

-

Detrital quartz

-

Quartz cement

-

Detrital quartz

-

2+

Mg -

Al

3+

0.11

Si

4+

99.2

K+

Ca2+

Ti4+

Mn2+

Fe2+

Total

-

-

-

-

-

99.3

-

-

99.4

-

-

-

-

-

99.4

-

0.08

98.5

-

-

-

-

-

98.6

-

-

98.2

-

-

-

-

-

98.2

-

0.07

98.8

-

-

-

-

-

98.8

Quartz cement

-

-

-

98.8

-

-

-

-

-

98.8

Detrital quartz

-

-

-

99.0

-

-

-

-

-

99.0

Quartz cement

-

-

-

98.9

-

-

-

-

-

98.9

Detrital quartz

-

-

0.16

98.8

-

-

-

-

-

98.9

Quartz cement

-

-

-

98.9

-

-

-

-

-

98.9

EP

HT1

Target minerals

4.3. Pore systems analysis 4.3.1. Pore types The pores in the Upper Shihezi Formation tight sandstones are composed of interparticle, dissolution, and intercrystalline pores related to interstitial minerals, and microcracks. The intergranular pores are the largest (up to 200 µm in diameter) and show triangular or polygonal shape (Fig. 4 (i, k)). Dissolution pores mainly occur within partly or completely dissolved unstable minerals, such as feldspar and rock fragments, the diameter of which typically is from 50 to 150 µm, and some dissolution pores increase the interparticle pore size by connection (Fig. 4 (a, i, j)). The intercrystalline types can be subdivided into three parts (Jullien et al., 2005; Xiao et al., 2018): within lattice interspace, pores within interstitial mineral particles and voids between interstitial mineral particles. The first two were too small to penetrate and make substantially less contribution to storage, so intercrystalline pores were defined as the pores within interstitial mineral particles in this research. The intercrystalline pores were extremely tiny in size (Fig. 4 (b-g)). The predominant pore microcrack types are micron-scale wide micro fractures associated

AC C

234 235 236 237 238 239 240 241 242 243 244 245 246 247 248

0 X (um)

M AN U

231 232 233

-10

TE D

228 229 230

-20

RI PT

-20

ACCEPTED MANUSCRIPT

(a)

S61

S139

S172

Z451

Frequency, %

Z452

20

Z46

SD32-46 SD50-471 S393

10

264

SD38-46

SD50-472 S396

0

40

Frequency, %

30

SC

RI PT

with brittle detrital quartz grains (Fig. 4 (k)). 4.3.2. Pore-throat size distribution by RCP The rate-controlled mercury porosimetry test is an effective way to appraise pore-throat size distribution, which could reveal distinct characteristics for the pores and throats with different radii by significant pressure fluctuations (Yuan and Swanson, 1989; Toledo, et al., 1994). The fundamental properties of the samples for tests are listed in Table S2 and the representative properties extracted by RCP on 12 samples with corresponding values are listed in Table S3. The throat radius (Rt) is between 0.43 and 1.64 µm (av. 0.85 µm), the pore radius (Rp) is between 138.81 and 168.74 µm (av. 153.63 µm), and the pore throat radius ratio is between 137.08 and 583.39 (av. 325.74) in the Upper Shihezi Formation tight sandstones (Table S3, Fig. 7). The curves from the studied samples reveal that the pore size distribution looks similar to the main pore radius peak center in the range from 110 to 180 µm (Fig. 7 (a)). The throat size and pore throat radius ratio distribution vary greatly. The throat size curves show approximately normal distribution with different kurtosis, whereas the pore throat radius ratio curves show unimodal (e.g., S172 and Z46) or bimodal distribution (e.g., S61 and SD32-46) (Fig. 7 (b, c)).

(b)

M AN U

249 250 251 252 253 254 255 256 257 258 259 260 261 262 263

30 20

S481

S482

S49

S61

S139

S172

Z451

Z452

Z46

SD32-46

SD38-46

SD50-471

SD50-472

S393

S396

10

0

0

100

200

300

400

Pore radius, µm 30

500

TE D Frequency, %

1

2

3

4

Throat radius, µm

(c)

20

10

EP

0

S61

S139

S172

Z451

Z452

Z46

SD32-46

SD38-46

SD50-471

SD50-472

S393

S396

0

268 269 270 271 272 273 274 275 276 277 278

0

500

1000

1500

Pore throat radius ratio

Fig. 7. Pore-throat size distribution from RCP of the Upper Shihezi Formation tight sandstones: (a) pore size

AC C

265 266 267

distribution; (b) throat size distribution; (c) pore throat radius ratio distribution.

4.3.3. NMR-derived pore size distribution NMR-derived T2 spectra can characterize the full-range pore size distribution (Xiao et al., 2018). According to the equation mentioned above, the T2 relaxation time is proportional to the pore radius (Daigle and Johnson, 2016). Fig. 8 illustrates the T2 spectra of the samples in Upper Shihezi Formation tight sandstones. All samples show bimodal or quasi-bimodal patterns with great relaxation time span, meaning that pores and throats of different sizes were distributed. The T2 spectra have two forms: left-skewed distribution (micropores as the predominant pore space) and right-skewed distribution (macropores or microcracks as the predominant pore space). Their left peak center has a value of approximately 2.68 ms, and the right peak ranges from approximately 49.93 ms to 86.40 ms (Table S4).

S139, 3642.88m SD50-472, 3108.17m S393, 3666.42m Z452, 3137.23m S61, 3610.44m S172, 3690.87m SD38-46, 3026.23 m Z451, 3131.14m S396, 3836.13m SD50-471, 3106.3m Z46, 3004.8m SD32-46, 2989.28m boundary

100 90 80 70 60 50 40 30 20 10 0 0.1

1

10

100

RI PT

Frequency, %

ACCEPTED MANUSCRIPT

1000

10000

279 280

Fig. 8. Illustrations of NMR-derived T2 spectra in Upper Shihezi Formation sandstones.

281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313

5. Discussion 5.1. Silica sources Previous studies confirm that biogenic silica (especially sponge spicules), replacement of silicate minerals by carbonates, transformation of clay minerals, feldspar dissolution and chemical compaction are the main sources of silica (Boles and Franks, 1979; Burley and Kantorowicz, 1986; Dutton and Diggs, 1990; Bjørlykke and Egberg, 1993; Dutton, 1993; Rezaee and Tingate, 1997; Kim and Lee, 2004; Xi, et al., 2015). External sources can be excluded due to low SiO2 and Al3+ solubility (Bjørlykke and Jahren, 2012). Therefore, the preceding possible internal sources of silica will be discussed below. (1) Biogenic silica (Aase and Walderhaug, 2005; Weibel et al., 2010): Siliceous sponge spicule dissolution has been suggested as a source of silica in some places (Vagle et al., 1994). In the Upper Shihezi Formation tight sandstones, however, no distinct biogenic silica minerals are observed. Therefore, biogenic silica made no contribution to the quartz cementation in Upper Shihezi Formation tight sandstones. (2) Replacement of silicate mineral by carbonate (Burley and Kantorowicz, 1986; Kim and Lee, 2004): The thin sections show that ferrocalcite extensively corrodes detrital quartz grains and authigenic quartz overgrowths (Fig. 4 (l)), and these compelling characteristics provide evidence that the replacement of silicate minerals by carbonate is one of the sources of silica (Walker, 1960). However, the relatively low carbonate mineral point count (av. 0.5%) restricts the mass production of the silica; thus, this source was insignificant in Upper Shihezi Formation tight sandstones. (3) Feldspar dissolution or alteration (Hawkins, 1978; Rezaee and Tingate, 1997; Higgs et al., 2007): In Upper Shihezi Formation tight sandstones, feldspar dissolution is frequently observed where authigenic quartz develops, and it is always coupled with feldspar kaolinization (Fig. 4 (j, k)). Sporadically, even some microcrystalline quartz is adjacent to feldspar dissolution (Fig. 4 (a)), which means that the feldspar dissolution and alteration may be a possible source of silica because these reactions would generate aqueous silica that could be reprecipitated in favorable environmental conditions (Giles and Deboer, 1990; Bjørlykke and Jahren, 2012). There is an obvious feldspar increase with the kaolinite increase in Upper Shihezi Formation tight sandstones (Fig. 9 (a)), which also supports formation of kaolinite from feldspar dissolution. Nevertheless, the feldspar was a significantly small proportion of the detrital grains (Fig. 3 (a)); thus, this silica source was volumetrically unimportant in Upper Shihezi Formation tight sandstones. (4) Transformation of clay minerals (Boles and Franks, 1979; Metwally and Chesnokov,

AC C

EP

TE D

M AN U

SC

T2 relaxation time, ms

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

2012): The main types of clay mineral transformations in the Upper Shihezi Formation were smectite to illite and kaolinite to illite, which could be potential silica sources for quartz cementation (Thyberg et al., 2010; Bjørlykke, 2011; Bjørlykke, 2014). With increasing temperature and considerable content of potassium provided, the smectite could easily transform to illite through the mixed-layer I/S (Cama et al., 2000), and this process could release significant amounts of silica with the decrease in clay (Peltonen et al., 2009). Considerable work on the smectite to illite reaction has been previously undertaken by many scholars. It turns out that the transformation could release over 20 wt.% of silica as a proportion of the original smectite if this reaction is also associated with dissolution of K-feldspar, which could provide K+ (Saigal et al., 1988; Kim and Lee, 2004; Van de Kamp, 2008; Xi et al., 2015b; Niu et al., 2018). The K-feldspar dissolution is commonly observed in the research area by thin section (Fig. 2 (c), Fig. 4 (a, i)), which means that the source of potassium in the studied samples is not a trivial matter. In addition, the feldspar increase goes with the proportion of illite increase and is also evidence to support illite being generated from feldspar dissolution (Fig. 9 (b)). From the preceding petrologic analysis, volcanic rock fragments, which were one of the main sources of smectites (Mckinley et al., 2003; Shoval, 2004), were relatively abundant. Thus, the proportion of smectite at the eodiagenetic stage was probably rich, and the smectite to illite reaction might contributed a lot to the source of silica. Apart from the smectite to illite reaction, the kaolinite to illite reaction was also a significant process that released silica (Bjørlykke and Jahren, 2012). The obvious positive correlation between relative illite and kaolinite content reveals that the illitization of kaolinite might be an important stage in the formation mechanism of illite. Plentiful amounts of kaolinite would provide a wealth of sources for illite generation (Fig. 9 (c)). This reaction is likely to happen at relatively high temperature (over 140 ) (Xi et al., 2015b), and the homogenization temperature of the authigenic quartz can reach 188 (Fig. 6 (b)), which means that the illitization of kaolinite could release silica in deep burial accompanied by high temperature. The silica source provided by the kaolinite to illite reaction is one of the nonnegligible sources in the Upper Shihezi Formation tight sandstones due to the relatively high amounts of kaolinite (Fig. 3 (b)). However, the correlation between quartz cement and illite is a negative relationship (Fig. 9 (d)), suggesting that too much illite many retard the precipitation rate of silica and slow down the formation of quartz cement. The amounts of clay minerals were relatively low and most of the quartz cements were formed between 90-140 (Fig. 6 (b)), while the smectite to illite reaction occurs approximately between 70-90 (Peltonen et al., 2009; Xi et al., 2015b). Kaolinite would transform to illite when the temperature is over 140 , suggesting that clay mineral transformations is not the main source of quartz cements (Peltonen et al., 2009; Xi et al., 2015b). We still need to find the top silica source of quartz cement. (5) Chemical compaction (Dutton, 1993; Walderhaug, 1994; Gier et al., 2008): In the Upper Shihezi Formation, chemical compaction in the form of stylolites is common (Fig. 2 (d), Fig. 4 (f)). Chemical compaction is considered the most significant source of silica because the homogenization temperature of quartz cement ranges from approximately 50 to 190 in a continuous process and centers on a value of approximately 115 (Fig. 6 (b)), whereas the smectite to illite reaction only occurs at approximately 70-90 and the illitization of kaolinite usually occurs up to 140 (Peltonen et al., 2009; Xi et al., 2015b). In addition, the amounts of clay minerals were relatively low compared with quartz, demonstrating that chemical compaction may have made the greatest contribution to the sources of silica (Fig. 3 (a)). The content of

AC C

314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357

ACCEPTED MANUSCRIPT

Feldspar, %

Feldspar, %

8

4

8 6 4 2

0

0

0

5

370

10

15

Kaolinite, % 20

0

15 10 5 0

10

EP

5

5

10

15

20

Illite, %

y = 1.4023ln(x) + 3.9628 R² = 0.1619

(c)

0

20

20

TE D

Kaolinite, %

y = 2.1164x0.2613 R² = 0.2869

(b)

M AN U

6

2

371

10

y = 1.5789x0.3752 R² = 0.2608

(a)

SC

10

RI PT

dissolved detrital quartz induced by chemical compaction could be estimated by the proportion of ‘overlap quartz’ (Houseknecht, 1991). Fig. 9 (e) illustrates the point counting of overlap quartz to detrital quartz (OQ/DQ) versus quartz cement to detrital quartz (QC/DQ) by 20 grains/section for a subset of the thin section samples. A good relationship exists between the (OQ/DQ) and the (QC/DQ), indicating that chemical compaction was probably a main source of silica as well. In addition, the mean value for the (OQ/DQ) to (QC/DQ) ratio of the Upper Shihezi Formation tight sandstones is 0.91, which means that the silica budget was in near balance between the proportions of silica produced by chemical compaction. In addition, the clay minerals could promote the process of chemical compaction (Molenaar et al., 2007). Thus, chemical compaction might easily happen due to the widely developed clay minerals in the studied intervals. Therefore, chemical compaction was the top silica source of quartz cement in Upper Shihezi Formation tight sandstone.

Quartz cement, %

358 359 360 361 362 363 364 365 366 367 368 369

15

(d)

y = 5.5207x-0.344 R² = 0.1559

15 10 5 0

20

0

5

10

Illite, %

15

20

25

30

Illite, %

0.5

(e)

372 373 374 375 376 377 378

QC/DQ

AC C

0.4 0.3 0.2 0.1 0.0 0

0.1

0.2

0.3

0.4

0.5

OQ/DQ

Fig. 9. The fitting relationships in: (a) kaolinite and feldspar, (b) illite and feldspar, (c) illite and kaolinite from XRD results; (d) illite and quartz cement from thin sections results; (e) silica budget plot (modified from

Houseknecht, 1991) for a subset of the thin section samples. OQ-overlap quartz; DQ-detrital quartz; QC-quartz cement.

5.2. The effects of quartz cement on storage performance

ACCEPTED MANUSCRIPT

y = 15.106e-0.113x R² = 0.5139

15

total intrusion

10

pore body intrusion

2

391

4

6

Quartz cement, % 100

(c)

100

8

y = 22.393x0.4053 R² = 0.1618

2500

Depth, m

TE D

80 60 40 20 0 0

5

10

60

0.01 40

20

0

15

QC<3% 3%≤QC≤5% QC>5%

(d)

3000

3500

4000

20

0

5

10

15

20

Porosity, %

AC C

Porosity, %

80

IPD, %

Mercury intrusion saturation, %

EP

Proportion of QCRP, %

0.1

ICD, %

0

398 399 400 401 402 403 404

,%

IPD/ICD boundary

0

393 394 395 396 397

1

throat intrusion

5

392

10

(b)

Intrusion pressure, MPa

(a)

SC

Porosity, %

20

RI PT

The RCP tests can quantify the porosity and pore-throat size distribution, which are significant in tight sandstone storage appraisal. NMR can also reveal the pore-throat size distribution and quantify the proportion of different types of pores (Xiao et al., 2016). The storage performance is defined here as the porosity and pore-throat size distribution, which are functions of many controls in tight sandstones. 5.2.1. The effects of quartz cement on porosity Fig. 10 (a) illustrates that the core analysis porosity and the point counted quartz cement have a medium-well negative correlation, indicating that quartz cement can partly occlude the pores and cause loss of porosity. Thin section and SEM observation confirmed this phenomenon (Fig. 2 (d), 4 (f, g)). However, there still are a few samples that contain relatively large proportions of quartz cement with relatively high porosity (S61 and Z452, Table S2), which means that the pores among quartz cement aggregate may contribute to porosity to some extent.

M AN U

379 380 381 382 383 384 385 386 387 388 389 390

Fig. 10. Graph showing (a) the relationship between the point counted quartz cement and the core analysis porosity; (b) mercury intrusion curve measured by RCP grouped into two parts; (c) proportion of quartz cement related pores (QCRP) plotted against core analysis porosity; (d) variation of core analysis porosity against burial depth with different contents of quartz cement in Upper Shihezi Formation tight sandstones. IPD-interparticle pores dominant; ICD-intercrystalline pores dominant; QCRP-quartz cement related pores; QC-quartz cement.

The capillary pressure curves from these samples could be classified into two parts: interparticle pore dominated (IPD) and intercrystalline pore/throat dominated (ICD). The IPD part reveals that the total mercury injection curve trend is more in accordance with the pore mercury injection curve and shows an approximate plateau-like trend at the low-pressure stage, representing the pore-throat size distribution as “large pores combined with narrow throats,” whereas the pore mercury intrusion saturation of the ICD part remains about the same with the

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

intrusion pressure increase and shows exponential growth at the high-pressure stage, indicating that pore-throat heterogeneity is weak and that nearly all mercury is injected into the intercrystalline pores or throats of the samples in this stage (Xiao et al., 2018) (Fig. 10 (b)). Therefore, the proportion of the ICD part represents the porosity contributed by intercrystalline pores or throats. As the selected samples in this research contain relatively rich quartz cement with other interstitial mineral content in trace amounts, the proportion of the intercrystalline pores or throats could be defined as quartz cement related pores (QCRP). The QCRP, however, represent the pores among the aggregates of quartz cement and the narrowed throats resulting from being heavily filled by quartz cement (Fig. 4 (g)). The proportion of QCRP varies from 27.23% to 88.03% with an average of 60.19% and shows a weak positive correlation with porosity (Table S3, Fig. 10 (c)), indicating that QCRP indeed compensate for the loss of porosity to a certain degree, although they mainly destroy storage. In addition, the microcracks induced by compaction of detrital quartz can also increase the porosity (Fig. 4 (k)). To eliminate any consideration of compaction, Fig. 10 (d) illustrates the variation of core analysis porosity against burial depth and point counted quartz cement. In depths of approximately 3050 m (corresponding to similar compaction), the porosity decreased from 18.30% (SD32-46) to 10.41% (SD38-46) and then to 6.50% (SD50-472) with increasing quartz cement content (Table S2, Fig. 10 (d)). The depths of approximately 3600 m have a similar trend. That confirms that quartz cement filled the pore space and decreased the porosity directly. Furthermore, for the samples with similar amounts of quartz cement, the porosity decreases with increasing depth, and the reduction rate gradually increases as the samples shift from high to low quartz cement content, implying that quartz cement plays an important role in the partial prevention of porosity for tight sandstones (Fig. 10 (d)). 5.2.2. The effects of quartz cement on pore-throat size distribution According to the aforementioned results, the samples can be classified into two types based on the proportion of interparticle and intercrystalline pores: pore dominated samples and throat dominated samples. The pore dominated samples show that the trend of the total mercury injection curve is more in accordance with the pore mercury injection curve and that the pore volume contributed by interparticle pores is over 50% (Fig. 10 (b), 12 (a)), while the throat mercury injection curve is more in accordance with the total trend and the pore volume contributed by intercrystalline pores (throats) is more than 50% in the throat dominated samples (Fig. 10 (b), 12 (b)). The pore dominated samples tend to have commensurately greater amounts of quartz cement than the throat dominated ones, and the proportion of ICD increases with decreasing quartz cement (Table S2). As outlined above, the total mercury intrusion curve could be divided into a plateau-like trend and exponential growth. However, only pore dominated samples exert this trend, and the total mercury injection curves of the throat dominated samples are commonly in a straight line, showing no obvious distinction between the two groups (Fig. 11). This confirms that quartz cement increases the tight sandstone pore-throat size heterogeneity. The T2 spectra curves show significant differences in pore dominated samples and throat dominated samples. The curves of pore dominated samples show bimodal patterns with a right-skewed peak centered at approximately 70 ms (Fig. 9). The curves of the throat dominated samples, however, show quasi-bimodal to bimodal patterns with tails. All samples show a left-skewed distribution, and the peaks center at approximately 2.86 ms (Fig. 8). Thus, the pores were relatively larger in pore dominated samples than throat dominated according to the T2 values

AC C

405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448

ACCEPTED MANUSCRIPT that mentioned above to radius conversion equation (Daigle and Johnson, 2016). Furthermore,

450

according to Liu et al. (2009), the T2 relaxation time can be divided into three parts: <10 ms,

451

10~100 ms and >100 ms, which represent micropores, mesopores and macropores/microcracks,

452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471

respectively. Table S4 and Fig. 8 and the data in supplementary materials show that the proportions of mesopores and macropores are relatively high in the pore dominated samples, whereas micropores were the predominant pore type in throat dominated samples. Pore radius, throat radius and pore-throat radius ratios are the main parameters for the evaluation of pore-throat size distribution in RCP tests. There are various forms of throat radii, including the average throat radius (ra), the mainstream throat radius (rm, the contribution of calculated permeability reaches 95%) (Li et al., 2017) and the entry radius (re, corresponding to the entry pressure of the total mercury intrusion curve) to evaluate the relationship between quartz cement and pore-throat size distribution parameters. Fig. 12 (a) illustrates that quartz cement shows a better correlation with average throat radius than entry radius and mainstream radius, manifesting that the average radius is a more effective parameter for indicating the amount of quartz cement and that quartz cement indeed did not affect the permeability much. However, compared to the pore radius and the pore throat radius ratio, the throat radius plays little role in quartz cement: the pore radius increases with increasing quartz cement, and there is a medium-poor positive relationship between pore radius and point counted quartz cement (Fig. 12 (b)), meaning that the quartz cement may retard mechanical compaction and preserve the pore volume to a certain degree and the throat radius is less affected. In addition, the point counted quartz cement contains medium-poor negative correlation with the pore throat radius ratio (Fig. 12 (c)), indicating that its main role is to increase the tight sandstone heterogeneity, albeit slightly preserving the pore volume, consistent with the aforementioned results.

TE D

M AN U

SC

RI PT

449

total intrusion throat intrusion pore body intrusion

0.1

EP

31.62%

ICD 80

70

60

50

40

30

20

0.001

1 total intrusion throat intrusion pore body intrusion

54.47% 0.1 IPD 0.01 ICD 0.001

0

100 90

Mercury intrusion saturation, %

80

70

60

50

40

30

20

10

0

Mercury intrusion saturation, %

Fig. 11. Mercury intrusion curves measured by RCP of the Upper Shihezi Formation tight sandstones: (a) typical pore dominated sample and (b) typical throat dominated sample. ICD-intercrystalline pore dominated; IPD-interparticle pore dominated.

4

y = 0.1019x + 1.4549 R² = 0.0411

3

Average radius Mainstream radius Entry radius

y = 0.0639x + 0.619 R² = 0.1854

y = 0.054x + 0.842 R² = 0.0515

2 1 0

200

y = 2.6169x + 144.23 R² = 0.3829

(b)

180 160 140 120 100

0

2

4

Quartz cement, %

6

8

800

Pore throat radius ratio

(a)

Pore radius, µm

Throat radius, µm

5

477 478 479

10

AC C

476

100 90

0.01

IPD

(b)

Intrusion pressure, MPa

Intrusion pressure, MPa

1

472 473 474 475

10

10

(a)

(c)

y = 448.73e-0.114x R² = 0.3394

600 400 200 0

0

2

4

Quartz cement, %

6

8

0

2

4

6

8

Quartz cement, %

Fig. 12. The relationships between the point counted quartz cement and (a) ra (average throat radius), rm (mainstream throat radius), and re (throat radius corresponding to entry pressure of total mercury intrusion curve);

ACCEPTED MANUSCRIPT 480

(b) pore radius; (c) pore throat radius ratio in Upper Shihezi Formation tight sandstones.

481

M AN U

SC

RI PT

Fig. 12 shows that quartz cement affects the pore-throat size distribution mainly through changing pore properties (corresponding to the maximal R-squared), whereas whether quartz cement directly leads to compaction is still debated. Fig. 13 (a) illustrates that in similar depths (approximately 3050 m and 3600 m), corresponding to similar mechanical compaction strength, the samples with relatively high amounts of quartz cement have higher pore radii, indicating that a higher quartz cement content can retard the mechanical compaction effect on the decrease in pore volumes to some extent. The maximum pressure in RCP tests is 6.2 MPa, corresponding to approximately 0.12 µm based on the Washburn equation (1921). Therefore, the NMR-derived pore size distribution should be used to characterize the full-range pore distribution. The NMR-derived pore size distribution shows the different morphologies, indicating that the proportion of intergranular pores and quartz cement-related pores is valid to characterize the pore size distribution in tight sandstones (Fig. 13 (b, c, d)). With increasing quartz cement, the pore size distribution varies from right-skewed quasi-bimodal or bimodal to left-skewed quasi-bimodal or bimodal patterns with separated T2 spectra distribution and finally to bimodal patterns with continuous T2 spectra distribution (Fig. 13 (b, c, d)). An obvious increase in the proportion of micropores corresponds to the short T2 population, and a decrease in the percentage of macropores corresponds to the long T2 population. This indicates that at the early diagenetic stage, quartz cement can retard mechanical compaction and compensate for the loss of pore volumes, while quartz cement formed in late diagenetic stages, albeit in small amounts, would fill the pores directly and decrease the radius of large pores and generate some minor pores within the aggregate of authigenic quartz. These are presumed to be incapable of restraining mechanical compaction that occurred mainly in the early rapid burial stage, leading to an amplitude and proportion increase in the short T2 population. 120

140

160

180

QC>5%

Frequency, %

Pore radius, µm

(a) 2500

TE D

482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504

EP

3%≤QC≤5% QC<3%

Frequency, %

Depth, m

AC C 509

QC>5%

1

10

100

1000

10000

T2 relaxation time, ms 100 90 80 70 60 50 40 30 20 10 0

Z452, 3137.23m S61, 3610.44m S172, 3690.87m SD38-46, 3026.23 m Z451, 3131.14m boundary

(c)

3%≤QC≤5% 0.1

1

10

100

1000

10000

T2 relaxation time, ms

Frequency, %

3500

100 90 80 70 60 50 40 30 20 10 0

S396, 3836.13m SD50-471, 3106.3m Z46, 3004.8m SD32-46, 2989.28m boundary

(d)

QC<3% 0.1

4000

S139, 3642.88m SD50-472, 3108.17m S393, 3666.42m boundary

(b)

0.1

3000

505 506 507 508

100 90 80 70 60 50 40 30 20 10 0

1

10

100

1000

10000

T2 relaxation time, ms

Fig. 13. The effects of quartz cement on the pore-throat size distribution in Upper Shihezi Formation tight sandstones: (a) the variation of pore radius against burial depth; (b) NMR-derived pore-throat size distribution with different point counted quartz cement. QC-quartz cement.

ACCEPTED MANUSCRIPT

0.1

Total intrusion

1

Throat intrusion Pore body intrusion

10

ICD/IPD boundary Pore threshold radius

15

100

525 526 527

100

80

60

40

20

0

Mercury intrusion saturation, %

10

5

Inflection point

(c)

1.5

rtt

y = 0.4274e0.1424x R² = 0.5549

Throat threshold radius Pore threshold radius

1

0.5 0

0 0

0.2

0.4

0.6

Throat radius, µm

0.8

1

0

2

4

6

8

Quartz cement, %

Fig. 14. Determination of (a) pore threshold radius (rpt) and (b) throat threshold radius (rtt); (c) point counted quartz cement versus pore/throat threshold radius.

TE D

528

EP

6. Conclusions The tight sandstones of the Upper Shihezi Formation are mostly litharenites and sublitharenites with abundant volcanic rock fragments. Quartz cement and kaolinite are the dominating interstitial minerals, whereas illite, chlorite and mixed layer I/S are minor. The Upper Shihezi Formation tight sandstones are characterized by poor physical properties, and the porosity decreases with burial depth. Interparticle pores, dissolution pores (feldspar and rock fragments), intercrystalline pores and microcracks were the major pore types. The RCP methods show that the pore radius, throat radius and pore throat radius ratio distribution curves were significantly different, and the quartz cementation alters the NMR-derived T2 spectra right-skewed peak to a left-skewed peak due to the increase in micropores. Quartz cement that formed chiefly coeval with or after chemical compaction was expected to be the most significant silica source, and the smectite to illite reaction as well as the kaolinite to illite reaction can also provide abundant silica sources in the Upper Shihezi Formation tight sandstones. K-feldspar dissolution was able to provide the potassium needed for quartz cementation. Silica sourced from replacement of silicate minerals by carbonate and feldspar dissolution was volumetrically insignificant. With decreasing quartz cement, the samples change from pore to throat dominated accompanied by micropore increases. Quartz cement can reduce the porosity, but the limited compressibility of grain-supported rocks due to the existence of early diagenetic rigid quartz was

AC C

529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548

2

(b)

Threshold radius, µm

rpt

M AN U

0.01

(a)

SC

RI PT

As stated above, the pore-throat size distribution in Upper Shihezi Formation tight sandstone is complex and anisotropic, making the use of the single and conventional RCP-derived parameters inappropriate. Hence, utilization of the threshold radius and the combination of different threshold radii may be regarded as an effective way to ascertain the effect of quartz cement on pore-throat size distribution. The pore threshold radius corresponding to little/no pore body intrusion (Fig. 14 (a)) and the maximum of the differential throat mercury saturation to pressure (inflection point) corresponding to the radius was defined as the throat threshold radius (Fig. 14 (b)). However, the point counted quartz cement has a poor correlation with the pore threshold radius and throat threshold radius (R-squared equal to 0.4404 and 0.2016, respectively). Thus, it is crucial to combine the two. The pore and throat threshold radii were used represent the pore and throat dominated samples, respectively. The point counted quartz cement has a medium-strong positive relationship with pore/throat threshold radius (R-squared equal to 0.5549), meaning that combinations of pore and throat threshold radii were the foremost resulting parameters to indicate the content of quartz cement compared to other single and conventional parameters (Fig. 14 (c)). Intrusion pressure, MPa dSHg/dP c, MPa-1

510 511 512 513 514 515 516 517 518 519 520 521 522 523 524

ACCEPTED MANUSCRIPT also able to compensate for the reduction of pore-throat radius. A single parameter can hardly represent the content of quartz cement, whereas the combination of the pore and throat threshold radius has a considerably strong contribution to assess it.

RI PT

Acknowledgements This paper is co-funded by the National Natural Science Foundation of China (No. 51874242) and the National Natural Science Foundation for Young Scientists of China (No. 41702146). The authors would like to thank Dr. Zhehui Jin from University of Alberta for polishing the languages. We truly appreciate engineer Junxiang Nan for the data processing. References

Aase, N.E. and Walderhaug, O., 2005. The effect of hydrocarbons on quartz cementation: diagenesis in the Upper

SC

Jurassic sandstones of the Miller Field, North Sea, revisited. Petroleum Geoscience, 11(3), 215-223.

Bjørlykke, K. and Egeberg, P.K., 1993. Quartz cementation in sedimentary basins. AAPG bulletin, 77(9), 1538-1548.

Bjørlykke, K., 2011. Open-system chemical behaviour of Wilcox Group mudstones. How is large scale mass

Geology, 7(28), 1381-1382.

M AN U

transfer at great burial depth in sedimentary basins possible? A discussion. Marine and Petroleum

Bjørlykke, K. and Jahren, J., 2012. Open or closed geochemical systems during diagenesis in sedimentary basins: Constraints on mass transfer during diagenesis and the prediction of porosity in sandstone and carbonate reservoirs. AAPG bulletin, 96(12), 2193-2214.

Bjørlykke, K., 2014. Relationships between depositional environments, burial history and rock properties. Some principal aspects of diagenetic process in sedimentary basins. Sedimentary Geology, 301, 1-14.

TE D

Boles, J.R. and Franks, S.G., 1979. Clay diagenesis in Wilcox sandstones of southwest Texas: implications of smectite diagenesis on sandstone cementation. Journal of Sedimentary Research, 49(1), 55-70. Burley, S.D. and Kantorowicz, J.D., 1986. Thin section and SEM textural criteria for the recognition of cement dissolution porosity in sandstones. Sedimentology, 33(4), 587-604. Cama, J., Ganor, J., Ayora, C. and Lasaga, C.A., 2000. Smectite dissolution kinetics at 80 C and pH

EP

8.8. Geochimica et Cosmochimica Acta, 64(15), 2701-2717. Colón, C.F.J., Oelkers, E.H. and Schott, J., 2004. Experimental investigation of the effect of dissolution on sandstone permeability, porosity, and reactive surface area1. Geochimica et Cosmochimica Acta, 68(4), 805-817.

AC C

549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592

Dai, J., Li, J., Luo, X., Zhang, W., Hu, G., Ma, C., Guo, J. and Ge, S., 2005. Stable carbon isotope compositions and source rock geochemistry of the giant gas accumulations in the Ordos Basin, China. Organic Geochemistry, 36(12), 1617-1635.

Daigle, H. and Johnson, A., 2016. Combining mercury intrusion and nuclear magnetic resonance measurements using percolation theory. Transport in Porous Media, 111(3), 669-679.

Dutton, S.P. and Diggs, T.N., 1990. History of quartz cementation in the Lower Cretaceous Travis Peak Formation, East Texas. Journal of Sedimentary Research, 60(2), 191-202. Dutton, S.P., 1993. Influence of provenance and burial history on diagenesis of Lower Cretaceous Frontier Formation sandstones, Green River Basin, Wyoming. Journal of Sedimentary Research, 63(4), 665-667. Ehrenberg, S.N., 1990. Relationship between diagenesis and reservoir quality in sandstones of the Garn formation, Haltenbanken, mid-Norwegian Continental shelf (1). AAPG bulletin, 74(10), 1538-1558. Folk, R.L., 1980. Petrology of sedimentary rocks. Hemphill Publishing Company. Austin, Texas, 21, 714-727.

ACCEPTED MANUSCRIPT Gao, H. and Li, H., 2015. Determination of movable fluid percentage and movable fluid porosity in ultra-low permeability sandstone using nuclear magnetic resonance (NMR) technique. Journal of Petroleum Science and Engineering, 133, 258-267. Gier, S., Worden, R.H., Johns, W.D. and Kurzweil, H., 2008. Diagenesis and reservoir quality of Miocene sandstones in the Vienna Basin, Austria. Marine and Petroleum Geology, 25(8), 681-695. Giles, M.R. and De Boer, R.B., 1990. Origin and significance of redistributional secondary porosity. Marine and Petroleum Geology, 7(4), 378-397.

RI PT

Hawkins, P.J., 1978. Relationship between diagenesis, porosity reduction, and oil emplacement in late Carboniferous sandstone reservoirs, Bothamsall Oilfield, E Midlands. Journal of the Geological Society, 135(1), 7-24.

Higgs, K.E., Zwingmann, H., Reyes, A.G. and Funnell, R.H., 2007. Diagenesis, porosity evolution, and petroleum emplacement in tight gas reservoirs, Taranaki Basin, New Zealand. Journal of Sedimentary Research, 77(12),

SC

1003-1025.

Houseknecht, D.W., 1991. Use of cathodoluminescence petrography for understanding compaction, quartz cementation, and porosity in sandstones.

Huang, H., Chen, L., Sun, W., Xiong, F., Ji, W., Jia, J., Tang, X., Zhang, S., Gao, J. and Luo, B., 2018. Pore-throat

China. Fractals, 26(02), 1840005.

M AN U

structure and fractal characteristics of Shihezi Formation tight gas sandstone in the Ordos Basin,

Hyodo, A., Kozdon, R., Pollington, A.D. and Valley, J.W., 2014. Evolution of quartz cementation and burial history of the Eau Claire Formation based on in situ oxygen isotope analysis of quartz overgrowths. Chemical Geology, 384, 168-180.

Jia, C., Jia, A., Zhao, X., Guo, J. and Tang, H., 2017. Architecture and quantitative assessment of channeled clastic deposits, Shihezi sandstone (Lower Permian), Ordos Basin, China. Journal of Natural Gas Geoscience, 2(1),

TE D

11-20.

Jia, C., Zheng, M. and Zhang, Y., 2016. Some key issues on the unconventional petroleum systems. Petroleum Research, 1(2), 113-122.

Jullien, M., Raynal, J., Kohler, E. and Bildstein, O., 2005. Physicochemical reactivity in clay-rich materials: tools for safety assessment. Oil & gas science and technology, 60(1), 107-120.

EP

Kim, Y. and Lee, Y.I., 2004. Origin of quartz cement in the Lower Ordovician Dongjeom formation, Korea. Journal of Asian Earth Sciences, 24(3), 327-335. Lai, J., Wang, G., Cai, C., Fan, Z., Wang, S., Chen, J. and Luo, G., 2018. Diagenesis and reservoir quality in tight gas sandstones: The fourth member of the Upper Triassic Xujiahe Formation, Central Sichuan Basin,

AC C

593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636

Southwest China. Geological Journal, 53(2), 629-646.

Lander, R.H., Larese, R.E. and Bonnell, L.M., 2008. Toward more accurate quartz cement models: The importance of euhedral versus noneuhedral growth rates. AAPG bulletin, 92(11), 1537-1563.

Li, J., Yin, J., Zhang, Y., Lu, S., Wang, W., Li, J., Chen, F. and Meng, Y., 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. Li, P., Zheng, M., Bi, H., Wu, S. and Wang, X., 2017. Pore throat structure and fractal characteristics of tight oil sandstone: A case study in the Ordos Basin, China. Journal of Petroleum Science and Engineering, 149, 665-674. Liu, W., Xiao, Z., Yang, S., 2009. Comparative studies on methods of evaluation of reservoir pore structure by using NMR (nuclear magnetic resonance) well logging data. Oil Geophysical Prospecting, 44(6), 773-778. Luo, J.L., Morad, S., Salem, A., Ketzer, J.M., Lei, X.L., Guo, D.Y. and Hlal, O., 2009. Impact of diagenesis on

ACCEPTED MANUSCRIPT reservoir quality evolution in fluvial and lacustrine deltaic sandstones: evidence from Jurassic and Triassic sandstones from the ordos basin, china. Journal of Petroleum Geology, 32(1), 79-102. McKinley, J.M., Worden, R.H., Ruffell, A.H. and Morad, S., 2003. Smectite in sandstones: a review of the controls on occurrence and behaviour during diagenesis. In Clay Mineral Cements in Sandstones (Vol. 34, 109-128). International Association of Sedimentologists Special Publication. Metwally, Y.M. and Chesnokov, E.M., 2012. Clay mineral transformation as a major source for authigenic quartz in thermo-mature gas shale. Applied Clay Science, 55, 138-150.

RI PT

Molenaar, N., Cyziene, J. and Sliaupa, S., 2007. Quartz cementation mechanisms and porosity variation in Baltic Cambrian sandstones. Sedimentary Geology, 195(3-4), 135-159.

Mousavi, M.A. and Bryant, S.L., 2012. Connectivity of pore space as a control on two-phase flow properties of tight-gas sandstones. Transport in porous media, 94(2), 537-554.

Niu, X., Yan, D., Zhuang, X., Liu, Z., Li, B., Wei, X., Xu, H. and Li, D., 2018. Origin of quartz in the lower

SC

Cambrian Niutitang Formation in south Hubei Province, upper Yangtze platform. Marine and Petroleum Geology, 96, 271-287.

Peltonen, C., Marcussen, Ø., Bjørlykke, K. and Jahren, J., 2009. Clay mineral diagenesis and quartz cementation in mudstones: The effects of smectite to illite reaction on rock properties. Marine and Petroleum Geology, 26(6),

M AN U

887-898.

Rezaee, M.R. and Tingate, P.R., 1997. Origin of quartz cement in the Tirrawarra sandstone, southern Cooper Basin, South Australia. Journal of Sedimentary Research, 67(1), 168-177.

Robinson, A. and Gluyas, J., 1992. Duration of quartz cementation in sandstones, North Sea and Haltenbanken basins. Marine and Petroleum Geology, 9(3), 324-327.

Rosenbrand, E., Fabricius, I.L., Fisher, Q. and Grattoni, C., 2015. Permeability in Rotliegend gas sandstones to gas and brine as predicted from NMR, mercury injection and image analysis. Marine and Petroleum Geology, 64,

TE D

189-202.

Rui, Z., Cui, K., Wang, X., Lu, J., Chen, G., Ling, K. and Patil, S., 2018. A quantitative framework for evaluating unconventional well development. Journal of Petroleum Science and Engineering, 166, 900-905. Saidian, M. and Prasad, M., 2015. Effect of mineralogy on nuclear magnetic resonance surface relaxivity: A case study of Middle Bakken and Three Forks formations. Fuel, 161, 197-206.

EP

Saigal, G.C., Morad, S., Bjorlykke, K., Egeberg, P.K. and Aagaard, P., 1988. Diagenetic albitization of detrital K-feldspar in Jurassic, Lower Cretaceous, and Tertiary clastic reservoir rocks from offshore Norway; I, Textures and origin. Journal of Sedimentary Research, 58(6), 1003-1013. Shoval, S., 2004. Deposition of volcanogenic smectite along the southeastern Neo-Tethys margin during the

AC C

637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680

oceanic convergence stage. Applied Clay Science, 24(3-4), 299-311.

Thyberg, B., Jahren, J., Winje, T., Bjørlykke, K., Faleide, J.I. and Marcussen, Ø., 2010. Quartz cementation in Late Cretaceous mudstones, northern North Sea: changes in rock properties due to dissolution of smectite and precipitation of micro-quartz crystals. Marine and Petroleum Geology, 27(8), 1752-1764.

Toledo, P.G., Scriven, L.E. and Davis, H.T., 1994. Pore-space statistics and capillary pressure curves from volume-controlled porosimetry. SPE Formation Evaluation, 9(01), 46-54. Vagle, G.B., Hurst, A. and Dypvik, H., 1994. Origin of quartz cements in some sandstones from the Jurassic of the Inner Moray Firth (UK). Sedimentology, 41(2), 363-377. Van de Kamp, P.C., 2008. Smectite-illite-muscovite transformations, quartz dissolution, and silica release in shales. Clays and Clay Minerals, 56(1), 66-81. Walderhaug, O., 1994. Precipitation rates for quartz cement in sandstones determined by fluid-inclusion microthermometry and temperature-history modeling. Journal of Sedimentary Research, 64(2a), 324-333.

ACCEPTED MANUSCRIPT Walker, T.R., 1960. Carbonate replacement of detrital crystalline silicate minerals as a source of authigenic silica in sedimentary rocks. Geological Society of America Bulletin, 71(2), 145-152. Washburn, E.W., 1921. The dynamics of capillary flow. Physical review, 17(3), p.273. Weibel, R., Friis, H., Kazerouni, A.M., Svendsen, J.B., Stokkendal, J. and Poulsen, M.L.K., 2010. Development of early diagenetic silica and quartz morphologies—Examples from the Siri Canyon, Danish North Sea. Sedimentary Geology, 228(3-4), 151-170. Worden, R. and Morad, S. eds., 2009. Quartz cementation in sandstones (Vol. 14). John Wiley & Sons.

RI PT

Xi, K., Cao, Y., Jahren, J., Zhu, R., Bjørlykke, K., Haile, B.G., Zheng, L. and Hellevang, H., 2015a. Diagenesis and reservoir quality of the Lower Cretaceous Quantou Formation tight sandstones in the southern Songliao Basin, China. Sedimentary Geology, 330, 90-107.

Xi, K., Cao, Y., Jahren, J., Zhu, R., Bjørlykke, K., Zhang, X., Cai, L. and Hellevang, H., 2015b. Quartz cement and its origin in tight sandstone reservoirs of the Cretaceous Quantou formation in the southern Songliao basin,

SC

China. Marine and Petroleum Geology, 66, 748-763.

Xiao, D., Jiang, S., Thul, D., Lu, S., Zhang, L. and Li, B., 2018. Impacts of clay on pore structure, storage and percolation of tight sandstones from the Songliao Basin, China: Implications for genetic classification of tight sandstone reservoirs. Fuel, 211, 390-404.

M AN U

Xiao, D., Lu, S., Lu Z., Huang W. and Gu, M., 2016. Combining nuclear magnetic resonance and rate-controlled porosimetry to probe the pore-throat structure of tight sandstones. Petroleum Exploration and Development, 43(6), 1049-1059.

Yang, H., Fu, J., Wei, X. and Liu, X., 2008. Sulige field in the Ordos Basin: Geological setting, field discovery and tight gas reservoirs. Marine and Petroleum Geology, 25(4-5), 387-400.

Yang, R., Fan, A., Van Loon, A.J., Han, Z. and Wang, X., 2014. Depositional and diagenetic controls on sandstone reservoirs with low porosity and low permeability in the eastern Sulige gas field, China. Acta Geologica

TE D

Sinica (English Edition), 88(5), 1513-1534.

Yang, Y., Li, W. and Ma, L., 2005. Tectonic and stratigraphic controls of hydrocarbon systems in the Ordos basin: A multicycle cratonic basin in central China. AAPG bulletin, 89(2), 255-269. Yuan, H.H. and Swanson, B.F., 1989. Resolving pore-space characteristics by rate-controlled porosimetry. SPE Formation Evaluation, 4(1), 17-24.

EP

Zou, C., Zhu, R., Liu, K., Su, L., Bai, B., Zhang, X., Yuan, X. and Wang, J., 2012. Tight gas sandstone reservoirs in China: characteristics and recognition criteria. Journal of Petroleum Science and Engineering, 88, 82-91. Zou, C., Zhang G., Yang, Z., Tao S., Hou, L., Zhu, R., Yuan, X., Ran, Q, Li, D., Wang, Z., 2013. Concepts, characteristics, potential and technology of unconventional hydrocarbons: On unconventional petroleum

AC C

681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713

geology. Petroleum Exploration and Development, 40(4), 413-428.

ACCEPTED MANUSCRIPT

Highlights Chemical compaction was expected to be the most significant silica source. Quartz cement can reduce the porosity but retard the mechanical compaction. The combination of pore and throat threshold radius can better appraise quartz

AC C

EP

TE D

M AN U

SC

RI PT

cement.