Pore structure characteristics of lower Silurian shales ...

International Journal of Coal Geology 156 (2016) 12–24

Contents lists available at ScienceDirect

International Journal of Coal Geology journal homepage: www.elsevier.com/locate/ijcoalgeo

Pore structure characteristics of lower Silurian shales in the southern Sichuan Basin, China: Insights to pore development and gas storage mechanism Feng Yang a,b,c,d,⁎, Zhengfu Ning b,c, Qing Wang b,c, Rui Zhang b,c, Bernhard M. Krooss d a

Key Laboratory of Tectonics and Petroleum Resources (China University of Geosciences), Ministry of Education, Wuhan 430074, PR China State Key Laboratory of Petroleum Resources and Prospecting, China University of Petroleum (Beijing), #18, Fuxue Rd, Changping, Beijing 102249, PR China c Ministry of Education Key Laboratory of Petroleum Engineering, China University of Petroleum (Beijing), #18, Fuxue Rd, Changping, Beijing 102249, PR China d Energy and Mineral Resources Group (EMR), Institute of Geology and Geochemistry of Petroleum and Coal, Lochnerstr. 4–20, RWTH Aachen University, 52056 Aachen, Germany b

a r t i c l e

i n f o

Article history: Received 16 September 2015 Received in revised form 29 December 2015 Accepted 30 December 2015 Available online 3 January 2016 Keywords: Shale Pore structure FE–SEM Porosity Gas storage Organic matter

a b s t r a c t Silurian shale in Sichuan Basin is currently the most important target zone for shale gas exploration and development in China. Pore structure characteristics of Lower Silurian Longmaxi shales from southern Sichuan Basin were investigated. The combination of field emission scanning electron microscope (FE–SEM) and argon ion beam milling was utilized to describe the nanometer-to micrometer-scale (N1.2 nm) pore systems. The shales were characterized by organic geochemical and mineralogical analyses. Total porosity, pore size distribution (PSD), specific surface area, and gas content were determined. Controls of organic matter richness, thermal maturity, and mineralogy on porosity were examined. The contribution of individual mineral components to total porosity was analyzed quantitatively. Total gas contents of the shales determined from canister desorption data were compared with theoretical (sorptive and volumetric) gas storage capacities. The total organic carbon (TOC) content of the shale samples ranges between 0.1 and 8.0 wt.% and helium porosity varies between 0.7 and 5.7%. Maturity in terms of equivalent vitrinite reflectance of bitumen (Reqv) ranges from 1.8 to 3.2%. TOC content is a strong control for the pore system of these shales, and shows a positive correlation with porosity. Porosity increases with increasing thermal maturity when Reqv is less than 2.5%, but decreases for higher thermal maturity samples. FE–SEM reveals four pore types related to the rock matrix that are classified as follows: organic matter (OM)–hosted pores, pores in clay minerals, pores of framework minerals, and intragranular pores in microfossils. Pores in clay minerals are always associated with the framework of clay flakes, and develop around rigid mineral grains because the pressure shadows of mineral grains prevent pores from collapsing. Pores of framework minerals are probably related to dissolution by acidic fluids, and the dissolution–related pores promote porosity of shales. A unimodal PSD exists in the micropore range of TOC– rich samples, while the PSD of carbonate–rich samples are bimodal. A PSD maximum in the micropore range is attributed by OM and another maximum in the range of mesopore–macropores is probably caused by the dissolution of carbonate minerals. Quantitative evaluation of the contribution of individual mineral components to porosity shows that the organic matter contributes approximately 62% to the total porosity. Framework minerals (quartz, feldspar, and carbonates, et al.) and clay minerals contribute 25% and 13%, respectively. The total gas content of these shales ranges from 0.4 to 6.2 m3/t, and the total gas contents of selected samples determined from canister desorption tests agree with the theoretically estimated original gas-in-place (OGIP). OM-hosted pores are the main space for gas storage, and accounted for about 78% (55% adsorbed gas plus 23% free gas) of the OGIP, while pores in the inorganic matter accommodate 22% free gas of the OGIP. © 2016 Elsevier B.V. All rights reserved.

1. Introduction Shale gas reservoirs are classified as non–buoyancy driven accumulations, continuous hydrocarbon plays that are composed of fine⁎ Corresponding author at: Key Laboratory of Tectonics and Petroleum Resources (China University of Geosciences), Ministry of Education, Wuhan 430074, PR China. E-mail address: [email protected] (F. Yang).

http://dx.doi.org/10.1016/j.coal.2015.12.015 0166-5162/© 2016 Elsevier B.V. All rights reserved.

grained sedimentary rocks that include shales, mudstones, limestones and siltstones. Natural gas is stored in shale reservoirs as: (1) free gas in pores and fractures, (2) adsorbed gas in organic matter and inorganic minerals, and (3) dissolved gas in “in-situ” liquid hydrocarbons and formation water (Curtis, 2002). Microstructures of shales exhibit a high degree of complexity and heterogeneity. Nanometer-to micrometer-scale pore systems have been found in the organic matter and matrix of inorganic grains of shale reservoirs, which have significant influence on gas

F. Yang et al. / International Journal of Coal Geology 156 (2016) 12–24

storage and fluid transportation. To elucidate the complex pore systems of shales, various methods have been developed for investigating the pore structure of shales. The fluid invasion methods, such as high-pressure mercury intrusion (MICP) methods, low-pressure gas adsorption, helium pycnometer, provide an estimation of petrophysical properties of shales, including porosity, specific surface area, and pore size distribution (Yang and Aplin, 2007; Bustin et al., 2008; Ross and Bustin, 2009; Chalmers et al., 2012a; Rexer et al., 2014; Yang et al., 2014; Ghanizadeh et al., 2014, 2015a,b). A combination of fluid invasion and radiation methods has also been used for the characterization of shales (Clarkson et al., 2012; Mastalerz et al., 2012; Ruppert et al., 2013). Direct imaging methods, including scanning electron microscopy (SEM) and transmission electron microscopy (TEM) imaging methods, scanning transmission X–ray microscopy, provide information on pore size, pore morphology, and connectivity of the pore networks (Potter et al., 2005; Slatt and O'Brien, 2011; Bernard et al., 2012; Dong et al., 2015). Mechanically polished methods used in thin sections produce surface topographic irregularities, which greatly exceed the size of many pores in shales. Recent development of argon ion beam milling, which provides cross-sections with exceptional high-quality, offer a new suitable alternative for high-resolution imaging. The combination of argon ion beam milling and SEM has provided a visual tool for observation of pores in shales (Loucks et al., 2009, 2012; Passey et al., 2010; Schieber, 2010; Curtis et al., 2012a; Klaver et al., 2012, 2015a,b; Milliken et al., 2013), and three-dimensional structural models of shales can be reconstructed from SEM images (Walls and Sinclair, 2011; Curtis et al., 2012b). Research during the last several years has identified a variety of pore types in unconventional reservoir rocks, such as pores associated with organic matter, interparticle mineral pores, intraparticle mineral pores, and microfractures (Loucks et al., 2009, 2012; Passey et al., 2010; Schieber, 2010; Slatt and O'Brien, 2011; Curtis et al., 2012a; Milliken et al., 2013). Porosity in shale reservoirs is a product of initial (depositional) porosity, compaction and chemical diagenesis (mineralogical transformation, cementation and dissolution). Differences in the origins and distribution of pore types have been shown to have different effects on gas storage and transportation (Passey et al., 2010; Milliken et al., 2013). Although considerable effort has been devoted to characterize the complex pore systems of shales, less attention has been paid to the controls on pore development. The relation of porosity to mineralogy and microstructures are still not well understood. Furthermore, gas in shale reservoirs is stored mainly as free gas and adsorbed gas. Evaluations about gas sorption capacity on shales have been carried out recently (Gasparik et al., 2012, 2014; Zhang et al., 2012; Rexer et al., 2013, 2014). However, sorption measurements on gas shales under “in-situ” conditions pose some technology challenges (Gasparik et al., 2014, 2015). The mechanism about gas adsorption and free gas storage needs further investigations. This research focuses on the pore structure characteristics of marine shales from Lower Silurian Longmaxi Formation in southern Sichuan Basin, China. The Longmaxi shale in Sichuan Basin is currently the most important target zone for shale gas exploration and development in China, and commercial gas flow rates were obtained in this area since 2009 (Dai et al., 2014; Wang et al., 2014a,b). However, unlike the marine gas shales in the United States, the marine shales in southern Sichuan Basin have very high maturity (equivalent vitrinite reflectance between 2.0% and 4.5%, Zou et al., 2010; Dai et al., 2014), and experienced several episodes of intensive tectonic activity after the hydrocarbon generation. Investigating the controls on the pore structure, porosity, mineralogy, and gas content of the highly over-mature shales provides insights into the generation, migration, and storage of natural gas in these shales. In this paper, we document and illustrate pore structure (size, distribution, arrangements, and origins of the pores) of gas shales from the southern Sichuan Basin by the combination of FE–SEM and argon ion beam milling. The controls of organic matter richness, thermal maturity, and mineralogy on porosity were examined. Besides,

13

based on the gas content results determined from canister desorption tests and theoretical estimation, the gas storage behavior of these shales has been discussed. 2. Samples and experiments 2.1. Geological setting The Lower Silurian shale, deposited in a restricted marine basin environment, is widely developed in Sichuan Basin with a thickness of a few meters to 400 m (Zhou et al., 2014). The Silurian shales in this area have undergone a complicated tectonic evolution, which strongly affected the accumulation and preservation conditions of shale gas. The early Paleozoic strata experienced deep burial during the Early Mesozoic and an intense uplift from the Late Mesozoic to the Cenozoic (Zhou et al., 2014). At present, the burial depth of the Paleozoic shales is mostly within a range of 2000–7000 m in the Sichuan Basin. In some strongly uplifted areas, shales are buried at a shallow depth and even occur as outcrops (Wang et al., 2012). The lower section of the Lower Silurian Longmaxi Formation is mainly comprised of dark gray– black carbonaceous shale and arenaceous–silty mudstones interbedded with bioclastic limestone while the upper section mainly comprises of grayish–yellowish green shale and arenaceous mudstones. TOC of the Longmaxi Formation ranges from 0.4–18.4% with an average of 2.5% (Dai et al., 2014), and the high TOC sections mainly developed at the bottom of the Longmaxi Formation. The Longmaxi shale is highly over-mature with an equivalent vitrinite reflectance between 2.0% and 4.5% (Zou et al., 2010). 2.2. Samples Since the first cored well for shale gas in China was drilled in the Southern Sichuan Basin in 2008, more than 47 shale gas wells have been completed in Sichuan Basin, and commercial gas flows were obtained from many wells (Wang et al., 2012; Dai et al., 2014). Maximum gas yield of a single well in Longmaxi Formation in the Southern Sichuan Basin ranges from 3 × 103 m3/day to 430 × 103 m3/day (Dai et al., 2014). From the recently drilled wells N203 and W201 designed for the evaluation of the Silurian Longmaxi Formation in the Southern Sichuan Basin, 18 shale samples were collected for detailed analyses. In addition, 6 outcrops of the same formation were also sampled. Information on stratigraphic units, thickness distribution, geochemical characteristic, and well identification of the Silurian Longmaxi Formation source rocks in southern Sichuan Basin has been published previously by Wang et al. (2012) and Dai et al. (2014). 2.3. Experiments 2.3.1. Mineralogy XRD analysis was performed on powdered shale (~100 mesh) using a Rigaku D/max-2500PC diffractometer with 0.001° 2θ step size and 1 min step time at indoor temperature of 293.15 K and relative humidity (RH) of 70% (equipment maintenance). The relative mineral percentages were estimated semi-quantitatively using the area under the curve for the major peaks of each mineral. 2.3.2. Organic geochemistry The TOC content was determined on a LECO CS230 carbon/sulfur analyzer. Shale powders were treated using hydrochloric acid to remove the inorganic carbon, and then pyrolyzed up to 540 °C. Due to the absence of vitrinite in these Silurian shale samples, bitumen reflectance was measured using a microscope equipped with an oil-immersion objective lens and a photometer. The average bitumen reflectance was converted to equivalent vitrinite reflectance (Reqv), according to the equation of Feng and Chen (1988).

14


2.3.3. FIB–SEM imaging A FEI™ Quanta™ 200F scanning electron microscope was employed to observe the microstructure morphology. It could produce images of the pore structure with a high resolution of 1.2 nm (at 30 kV operating voltage and high vacuum mode) and a magnification of ×25 k–200 k. Shale samples were ground with fine grit sand paper. Then argon ion beam milling was utilized to produce a much flatter surface, and subsequently the surface was coated with gold to avoid charging. 2.3.4. Porosity, PSD and surface area Porosity of shale samples was calculated from bulk density and skeletal density. Samples were dried at 105 °C overnight, and then the skeletal densities (ρskeletal) were measured using a helium pycnometer based on Boyle's Law. The bulk density (ρbulk) of cylindrical plug samples was determined by caliper measurements. The bulk density of rock fragments was determined using the buoyancy method (Archimedes principle) after saturation with de-ionized water. Specific surface area and pore structure of the shales was investigated by low-pressure nitrogen adsorption using a QUADRASORB™ SI Four Station Surface Area and Pore Size Analyzer. About 200–500 mg powder samples (~100 mesh, 0.149 mm grain size was chosen for consistency with gas sorption tests) were outgassed at 383 K for 5 h under a vacuum of 10 μm Hg. Reagent grade nitrogen (99.999%) was used as adsorbent at 77 K, and adsorption–desorption isotherms were obtained from recording adsorption data at relative pressures (P/P0) from 0.01–1, and the equilibration time was set as 10 s.

bath at the reservoir temperature. The volumes of gas released were periodically recorded using a graduated cylinder. Gas content determination is terminated when the daily emission is less than 0.05 cm3/g for five consecutive days. Subsequently the “remainder gas” is determined by crushing the desorbed sample to a powder (~200 mesh) in a sealed mill. The total gas content, the sum of lost gas, measured gas, and residual gas, are reported normalized to sample mass. The methane sorption capacity of three selected shale samples was determined by a high-pressure manometric sorption apparatus. Drilled samples were carefully packed and rapidly sent to laboratory for sorption tests. Methane excess sorption isotherms were measured at in situ pressure and temperature conditions (the experimental pressures (up to 20 MPa) and temperatures (62–64 °C) were estimated according to the pressure and temperature gradients, Table 1). After the powdered samples (~100 mesh) were placed into the sample cell, the entire experimental setup was evacuated for 1 h using a vacuum pump, and then the apparatus was ready for sorption experiments. Similar measuring procedures were reported by Krooss et al. (2002) and Gasparik et al. (2014). The experimentally measured isotherms were fitted by the 3parameter excess sorption function based on a Langmuir-type absolute sorption isotherm (Gasparik et al., 2012, 2014; Yang et al., 2015). The adsorbed phase density required for this function was taken as the liquid density of methane at its boiling temperature and ambient pressure (421 kg/m3).

3. Results 2.3.5. Total gas content test and methane sorption analysis The total gas content of gas shales is subdivided into three components: lost gas, measured gas (desorbed gas), and remainder gas. The gas released from the shale core includes the adsorbed gas as well as free gas, thus the term “measured gas” is used instead of desorbed gas. Lost gas, the gas released from the time the target shale was drilled to the time that the sample was sealed in the desorption canister, can be estimated using the United States Bureau of Mines (USBM) method (Kissell et al., 1973). This procedure assumes that the amount of gas emitted is proportional to the square root of desorption time at reservoir temperature. The emitted gas is quantified after loading the core sample as quickly as possible in a sealed canister immersed in a water

3.1. Mineral compositions and organic geochemistry results The mineral compositions of the studied samples are presented in Fig. 1 and Table 1. Samples are either high in quartz and clay minerals or an intermediate mixture of quartz, clay and carbonate. Quartz content averages 34.8% (ranges from 14% to 70%), feldspar content averages 7% (ranges from 0% to 17.8%), carbonate minerals (calcite and dolomite) content averages 17.9% (ranging from 0% to 48.5%), and average clay mineral content is 37.9% (ranging from 17% to 59%). Clay minerals are dominated by illite (average 18%) with some samples also having illite/smectite mixed layer (average 10.7%) and chlorite (average 8.7%).

Table 1 Geochemical characteristics, mineralogical composition, and petrophysical parameters of the shale samples. Sample CN_001 CN_031 CN_033 CN_035 CN_036 CN_037 CN_0301 CN_0302 CQ_011 CQ_003 CQ_006 CQ_008 CQ_017 CQ_024 CW_11 CW_33 CW_51 CW_61 CW_76 CW_71 CW_86 CW_81 CW_95 CW_91 a

Stratigraphy

Deptha (m)

TOC (wt%)

Reqv (%)

Quartz + feldspar (wt%)

Carbonates (wt%)

Total clays (wt%)

Bulk density (g/cm3)

Skeletal density (g/cm3)

Porosity (%)

L. Silurian L. Silurian L. Silurian L. Silurian L. Silurian L. Silurian L. Silurian L. Silurian L. Silurian L. Silurian L. Silurian L. Silurian L. Silurian L. Silurian L. Silurian L. Silurian L. Silurian L. Silurian L. Silurian L. Silurian L. Silurian L. Silurian L. Silurian L. Silurian

2297.00 2117.64 2152.26 2261.20 2336.46 2386.23 2319.85 2377.84 outcrop outcrop outcrop outcrop outcrop outcrop 1379.50 1422.53 1453.72 1501.77 1508.55 1515.55 1521.55 1527.47 1533.30 1541.01

3.19 1.33 2.22 2.65 3.01 3.60 2.82 2.83 8.01 6.31 5.35 5.10 5.20 5.58 1.38 0.10 0.16 1.10 2.52 2.77 2.39 3.14 2.47 2.71

2.08 1.80 2.02 2.12 2.30 2.40 2.26 2.37 2.57 2.78 2.50 3.10 2.80 3.23 -

31.3 27.9 21.3 21.3 32.2 47.6 36.7 17.8 66.2 68.7 53.8 54.1 68.1 45.7 48.0 40.0 43.0 36.0 48.0 39.0 29.0 28.0 26.0 74.0

29.2 38.5 37.0 46.2 19.8 30.2 14.4 48.5 0.0 0.4 10.9 11.1 0.0 14.1 13.0 1.0 0.0 2.9 2.9 2.9 28.4 28.7 43.9 5.9

38.3 32.7 40.5 30.8 46.0 19.9 47.0 31.4 33.8 30.9 32.9 31.5 31.9 37.6 35.0 59.0 55.0 57.0 45.0 55.0 39.0 37.0 26.0 17.0

2.54 2.55 2.53 2.52 2.60 2.58 2.51 2.55 2.38 2.39 2.46 2.45 2.46 2.50 2.64 2.67 2.68 2.64 2.55 2.54 2.60 2.56 2.60 2.50

2.62 2.62 2.60 2.57 2.72 2.67 2.58 2.64 2.53 2.53 2.55 2.56 2.57 2.56 2.68 2.69 2.71 2.69 2.63 2.62 2.65 2.63 2.66 2.57

3.1 2.7 2.6 1.8 4.3 3.3 2.7 3.4 5.7 5.6 3.6 4.3 4.3 2.7 1.5 0.7 0.9 1.8 2.9 2.8 1.9 2.7 2.3 2.7

The pressure gradient of Silurian formation in the southern Sichuan Basin is 1.028 MPa/100 m, and temperature gradient is 3.06 °C/100 m.The average surface temperature is 16.6 °C.


Fig. 1. Mineralogical ternary plot of the shale samples.

No significant quantity of smectite was present in the studied samples. Pyrite is present in most of the samples and is up to 5%. The TOC contents of the shale samples range from 0.1 to 8.0 wt.% with an average value of 3.2 wt.% (Table 1). There is a moderate positive relationship between TOC content and quartz content of shale samples from N203 well (Fig. 2), indicating that quartz of these marine shale samples is at least partially of biogenic origin. Other studies on the Silurian shale in Eastern Sichuan Basin (Tian et al., 2013) and Devonian gas shales of Horn River Basin, Canada (Chalmers et al., 2012b) also suggest the presence of biogenic quartz. Equivalent vitrinite reflectance of the samples varies between 1.80% and 3.23%, confirming that these Silurian shales are over-mature as previously reported (Dai et al., 2014). 3.2. Porosity The average porosity of the Lower Silurian shale samples is 2.9% and ranges between 0.7% and 5.7%. A positive relationship between porosity and TOC content is observed (Fig. 3a). Samples with high TOC content have higher porosity. Quartz content also correlates positively with porosity (Fig. 3b). It has been considered that the positive relationship between quartz and porosity is due to the positive relationship between TOC and quartz content (Chalmers et al., 2012b). No relationships exist between porosity and total clay mineral contents of the shale samples. 3.3. PSD and specific surface area Pore size distributions (PSD) calculated from nitrogen adsorption isotherms by the Barrett–Joyner–Halenda (BJH) method are presented

Fig. 2. The relationship between TOC content and quartz content of shale samples from N203 well.

15

in Fig. 4. All the samples exhibit broad PSD, which are mainly unimodal for shale samples with high TOC content (N5%), the dominant pores being micropores (Fig. 4b). There is an exceptional moderate peak in mesopores of the PSD of TOC-rich sample CQ_017, which is probably due to the dissolution of feldspar. PSD of samples with moderate TOC content (b 3.5%), but high carbonate minerals content are different from the TOC-rich samples. PSD of samples with high carbonate minerals content are bimodal (Fig. 4a). One maximum in the micropore range is contributed by OM; and another maximum is in the range of mesopore–macropore range (20–70 nm) and probably results from the dissolution of carbonate minerals. The relative volumes of micro-, meso- and macropores in PSD of these samples are presented in Table 2. Micropore volumes of TOCrich samples are larger than those for carbonate-rich samples. TOC contents of the samples show a positive relationship with the contribution of micropore volumes, while an increase in carbonate mineral contents is associated with increasing contributions of meso- and macropore volumes (Fig. 5). A positive relationship exists between the Brunauer– Emmett–Teller (BET) specific surface area and TOC content (Fig. 6). Shale samples with higher TOC content have higher BET specific surface area and lower pore diameter. 3.4. Pore types from FE–SEM imaging FE–SEM imaging was performed on selected samples representing a typical range of TOC content and mineral compositions. FE–SEM images of the shale samples studied show various nanometer-to micrometersize pores. Based on the pore network and their relationships with minerals, pore types of these samples can be summarized as follows: OMhosted pores, pores in clay minerals, pores of framework minerals, and intragranular pores in microfossils. 3.4.1. OM-hosted pores OM-hosted pores are a significant component of pore systems in the organic-rich shales samples studied. Fig. 7a shows that a single OM grain with diameter of 4–5 μm could contain hundreds of pores with various shapes and sizes. The majority of the OM-hosted pores in the SEM images are mesopores, but macropores are also present. The shapes of these OM-hosted pores vary from ellipsoid to irregular polygons, but ellipsoid is the most common shape (Fig. 7a). The diameters of the OM-hosted pores in our FE–SEM images range from several to hundreds of nanometers, which have similar dimension to the OMhosted pores in Barnett shale (Loucks et al., 2009, 2012). It should be noted that most of the pores in FE–SEM images are meso–macropores, while low-pressure gas adsorption techniques can resolve much smaller pores (micro–mesopores) in shales (Bustin et al., 2008; Chalmers et al., 2012a,b). Organic matter with intraparticle porosity is found to be surrounded by quartz (Fig. 7b) or around pyrite crystals (Fig. 7c). Meso–macropores are preserved in the interstitial space of rigid grains (Fig. 7c). Macropores in the OM show interconnections with coarse mesopores (Fig. 7d). 3.4.2. Pores in clay minerals For the clay-rich samples, intragranular pores develop well in the clay grains, and intercrystalline pores also occur locally between clay minerals and quartz grains (Fig. 8). Intraparticle pores are located along cleavage planes of clay minerals, and show little preferential orientation (Fig. 8a). Many pores are linear in shape, which are defined by the lattice of randomly oriented clay mineral platelets. The lengths of these elongate intercrystalline pores are up to several micrometers, but the widths are narrow (dozens to hundreds of manometers). The clay flakes are deformed and split by compaction forces (Fig. 8b), and some cleavage intraparticle pores are enhanced by bending related to compaction (Fig. 8b). Intragranular pores in clay minerals are best developed in pressure shadows of the quartz and calcite grains, and the space between such rigid grains protects pores from collapse (Fig. 8c).

16


Fig. 3. The relationship between (a) TOC content and porosity (b) Quartz content and porosity of shale samples from N203 well.

Macropores and mesopores are also located in the kerogen–clay aggregates, and these pores parallel with laminae of the clay minerals show strong post-compaction (many pores in kerogen–clay aggregates are closed) (Fig. 8d). 3.4.3. Pores of framework minerals Typical intragranular and intercrystalline pores in framework minerals such as carbonate rocks and pyrite are shown in Fig. 9. Dissolution-related pores are commonly observed when there are abundant carbonate grains scattered through the matrix of the shales. The shapes of pores in framework minerals may be controlled by their origin. Dissolution-related macropores are observed in calcite grains (Fig. 9a). Pyrite intercrystalline pores are located in pyrite framboids (Fig. 9b), though many of these pores are occluded with OM. Dissolution-related pores occur in feldspar grains and are along twin plans (Fig. 9c). Dissolutionrelated macropores, most likely formed by partial dissolution of carbonate are commonly isolated. Dissolution-related fractures are also found in carbonate minerals (Fig. 9d). The dissolution rarely removes the entire grains, and occurs along the grain boundaries.

concentrated in silicified Mallomonas (a subordinate class of Chrysophyta). 3.5. Measured total gas content Fig. 11 illustrates the evaluations of total gas content and lost gas of a shale sample using the USBM method. The total gas content of these shales studied ranged from 0.4 to 6.2 m3/t. The time used for the calculation of lost gas ranged from 4 to 6.5 h, and the lost gas makes up 20% to 54% of the total gas content. There is a positive relationship between the TOC content and measured total gas content of these shales (Fig. 12). Gas sorption capacities were obtained from the methane sorption isotherms measured at in situ pressure and temperature conditions (the maximum pressure up to 20 MPa, temperatures are between 62 and 64 °C) (Fig. 13). The Langmuir volumes for the selected samples varied between 1.02 and 2.27 std. m3/t rock. Shales with higher TOC content have higher sorption capacity. 4. Discussion 4.1. Effects of organic matter on porosity

3.4.4. Intragranular pores in microfossils The organisms in the petroliferous formations in the Silurian marine Longmaxi Formation consist mainly of algae, zooplankton, and fungi. During the process of deposition, these organisms were preserved and macropores can be found in the body cavities of the microfossils (Fig. 10). Fig. 10 show that connected intercrystalline macropores are

OM-hosted pores, resulting from volume loss associated with organic matter conversion during maturation, can contribute significantly to the total porosity of shales in the gas window. The positive relationship between the total porosity and TOC content is illustrated in Fig. 3a. This is also supported by the visible evidence that numerous pores are

Fig. 4. PSD of shale samples derived from the N2 adsorption branch of the N2 isotherms by the BJH method (a) carbonate-rich samples (b) TOC-rich samples.


17

Table 2 Pore structure parameters of selected shale samples. Sample

BET specific surface area (m2/g)

Pore volumea (cm3/g)

Average pore diameterb (nm)

Microporec (%)

Mesoporec (%)

Macroporec (%)

CN_001 CN_031 CN_033 CN_035 CN_036 CN_037 CN_0301 CN_0302 CQ_011 CQ_003 CQ_006 CQ_008 CQ_017 CQ_024

14.42 8.21 13.44 14.26 16.98 19.88 16.32 15.12 26.11 27.92 19.64 11.39 25.16 13.18

0.0298 0.0159 0.0321 0.0295 0.0411 0.0296 0.0307 0.0327 0.0297 0.0387 0.0201 0.0139 0.0420 0.0128

8.26 7.73 9.54 8.28 9.67 5.96 7.52 8.64 4.55 5.55 4.09 4.87 6.68 3.90

9.74 9.67 6.23 10.51 8.25 19.41 11.15 7.77 26.88 19.66 22.61 23.21 15.80 30.95

63.34 69.32 66.42 62.23 62.09 59.83 63.58 65.46 59.27 55.58 56.72 61.20 66.53 60.87

26.92 21.01 27.35 27.26 29.66 20.76 25.72 26.77 13.85 24.76 20.67 15.59 17.67 8.18

a b c

The pore volumes were obtained from the adsorption capacity of liquid nitrogen at the highest relative pressure point. The average pore diameters were calculated from total pore volume and BET specific surface area, according to Gurvich's rule (Gregg and Sing, 1982). Proportions of micro-, meso- and macropores were determined from N2 gas adsorption analyses.

concentrated in organic matter in FE–SEM images (Fig. 7). Our results are consistent with the observations in organic-rich shales from Marcellus Formation of North America (Milliken et al., 2013), where TOC content controls the OM-hosted porosity. The strong control of TOC content on the porosity of these marine shales is different from the low-TOC samples from the Lower Triassic Montney Formation in the Western Canadian Sedimentary Basin (Ghanizadeh et al., 2015b). Ghanizadeh et al. (2015b) have reported the petrophysical characteristics of selected Montney samples, and their low-TOC samples (TOC b 1%) have shown relatively high porosity, probablly related with inorganic matter. It is generally thought that porosity in the organic matter will increase with thermal maturity as hydrocarbons are generated (Jarvie et al., 2007; Loucks et al., 2009). In the over-mature shales studied here, porosity increases with thermal maturity at Reqv values below 2.5% (Fig. 14). After that, the porosity tends to decrease with enhanced thermal maturity. In order to analyze the effect of thermal maturity on porosity development, the porosity values of Silurian and Cambrian shales from the same area are also plotted in Fig. 14. These data suggest that there is an inflection in the relationship between porosity and thermal maturity. The FE–SEM study conducted by Wang et al. (2013) on Silurian and Cambrian shales in the Southern Sichuan Basin shows that when the thermal maturity is less than 2.5%, the OM-hosted pores develop well and their diameters are large in SEM images. However, when the thermal maturity is larger than 3.0%, the OM-hosted pores are less expressed and the diameters also become smaller. OM-hosted pores with diameters larger than 100 nm are not frequently observed in the SEM images of the Cambrian shales (Wang et al., 2013). The phenomenon of decreasing porosity at highly over-mature shales may be due to the carbonization of over-mature organic matter. During carbonization, OM-hosted pores are destroyed, merged, and collapsed, which

results in the decrease of both OM-hosted pores and porosity (Wang et al., 2014a; Yang et al., 2015). Furthermore, the preservation conditions are important to the porosity of shale. The shales with higher thermal maturity have been buried more deeply; thus mechanical compaction may have caused rearrangement of pores. The FE–SEM images show that OM-hosted pores surrounded by rigid framework minerals are obviously better preserved than the pores in kerogen–clay aggregates (Figs. 7 and 8d). 4.2. The uncertainties in bulk/skeleton density and porosity It is important to discuss the potential sources of uncertainty/error in the evaluation of petrophysical parameters of unconventional reservoirs, which commonly have low to extremely-low porosity and permeability values. The errors in bulk/skeleton density measurements may have a minimal effect on the porosity values of conventional reservoirs, but could significantly affect the porosity values estimated for unconventional reservoirs (Ghanizadeh et al., 2015c). The errors in bulk density measurements mainly originate from the precision of electronic balance and vernier caliper. The total error from the reading of vernier caliper is about ±42 μm for the experimental vernier caliper (measuring range: 0 – 150 mm; division value: 0.02 mm) (Dan, 2009), which will result in an error of 0.011 g/cm3 in the estimation of bulk density, and thus errors up to 0.41% in the estimated porosity values. The uncertainty from the precision of electronic balance (~ 0.1 mg) can be neglected in the porosity estimations. The possible sources of uncertainty on skeleton density were discussed by Ghanizadeh et al. (2015c). These uncertainties included sample cell volume, test duration, prepressurizing of the sample cell, and operator's bias on the choice of data. Ghanizadeh et al. (2015c) repeated helium pycnometer tests on

Fig. 5. The relationship between (a) TOC content and micropore volumes content (b) carbonate minerals content and meso–macropore volumes content for shale samples.

18


Fig. 6. The relationship between TOC content and BET specific surface area and average pore diameter for shale samples.

three sample cells with different volume, and found that the uncertainty associated with the sample cell volume is less than 0.005 g/cm3. In theory, the smaller the void volume (difference between sample cell volume and sample volume) in a sample cell, the lesser the error is associated with the measured skeleton density. Thus, the sample cell was filled with as much sample as possible before performing the

helium pycnometry tests. For the test duration, our selected test results showed that as long as the pressure equilibrated was reached and the setup was leakage free, the extended test time would not change the skeleton density values too much (the variation is less than 0.009 g/cm3). Furthermore, for the same samples, the variation in skeleton densities between buoyancy method and helium pycnometer was in most cases less than 0.01 g/cm3, which will result in errors up to 0.38% in the estimated porosity values. Based on these calculations, conservative uncertainties are provided for calculated porosity values, and are incorporated in the figures. It should be noted that the porosity of shale samples were tested utilizing helium expansion, as most laboratories reported. However, the porosity measured using helium pycnometer is different from the porosity using “in–situ” gaseous hydrocarbons/Methane. Pore throats of many gas shales are so small that the shale may act as a molecular sieve. Thus, the helium (kinetic diameter: 0.26 nm) pycnometry measurement yields a higher porosity than other gases. However, because of methane sorption on the TOC-rich shales, thus the tested apparent helium skeleton density of TOC-rich samples is less than methane without correcting for sorption. In carbonate-rich samples or samples with low specific surface area, the apparent skeleton densities are similar since sorption effect is minimal. If sorption corrections are applied, the actual skeleton density measured using helium should exceed that of methane. 4.3. Contribution of minerals to porosity Quartz of the Lower Silurian shales is primarily biogenic in origin, as inferred from the positive relationship between the quartz content and

Fig. 7. OM-hosted pores of shale samples. (A) Ellipsoid or irregularly shaped pores in black organic matter. (B) Organic matter with intraparticle porosity surrounded by quartz. (C) OMhosted pores around pyrite crystals. (D) Macropores in the OM show interconnections with coarse mesopores.


19

Fig. 8. Pores in clay minerals. (A). Intraparticle pores in clay platelet. (B). Clay flakes are deformed and split, and intraparticle pores are enhanced by bending related to compaction (C). Intragranular pores and interparticle pores develop in pressure shadows of rigid grains. (D). Macropores and mesopores locate in the kerogen–clay aggregates, and these pores paralled with laminae of the clay minerals show strong post–compaction.

TOC (Fig. 2). Quartz-rich samples in these Lower Silurian shales appear to have higher porosity (Table 1, Fig. 3b). Higher porosities in the quartz-rich samples are probably related to the elevated TOC contents. Furthermore, primary porosity preservation is enhanced by quartz cements (Fig. 8c), which provides a rigid framework that prevents porosity from collapsing. Pores in clay minerals are associated with the framework of clay flakes (Fig. 8). These are well-developed around rigid quartz grains because the pressure shadows of these grains protect the pores from collapsing (Fig. 8c). Clay minerals are ductile and compact and deform easily, thus the presence of rigid grains is important for the preservation of porosity of ductile clay minerals. Many macropores and mesopores in the kerogen–clay aggregates are closed because there are no rigid skeletons to support the pore structure (Fig. 8d). The pore structure of edgeface arrangements in flocculated clay are commonly observed in Barnett and Woodford shales (Slatt and O'Brien, 2011). The explanation for all these openwork structures that survived burial and diagenesis for up to hundreds of millions of years could be the preservation in pressure shadows. Shales are commonly interbedded with silts and siltstones. With increasing silt content there would be an increase in pressure shadows and preserved porosity, resulting also in a decrease in sealing quality. While many pores of ductile clay minerals are lost by compaction and cementation, pores of framework minerals can develop in the subsurface under the influence of erosive fluids. Dissolution of carbonates and feldspars has been observed in many shales (Passey et al., 2010; Loucks et al., 2012) and this also can be seen on Fig. 9. Dissolution-related pores are primarily caused by the erosion of organic acidic fluids

associated with hydrocarbon generation (Surdam et al., 1989; Potter et al., 2005), which can be illustrated by the intense erosion of feldspar in the vicinity of organic matter (Fig. 9c). During the intermediate stage of diagenesis, organic acids are generated by transformation of organic matter, and secondary porosity can be improved by the corrosion with organic acids. Dissolution of carbonate minerals is an important mechanism for development of intraparticle porosity in shales. Diameters of dissolution-related pores are larger (mesopore–macropores) than the OM-hosted pores, and the PSD of carbonate-rich samples are bimodal, which is obviously different from the unimodal PSD of TOC-rich samples (Fig. 4). Pore size distributions (dV/dlog(W)) of low-TOC Montney samples (TOC b 1%) measured by Ghanizadeh et al. (2015b) have also shown maxima in the mesopore–macropore range. Porosity of these Montney samples is probably related with the dissolution of minerals since the Montney Formation has been reported to be rich in dolomite and feldspar (Chalmers and Bustin, 2012c). When carbonate content is low or the carbonate grains are scattered through the matrix, the dissolution-related pores are observed as isolated scallops. However, when abundant carbonate grains have been embedded into the layers of shale successions, the dissolution-related pores probably promote the porosity and permeability of shales. 4.4. Quantification of porosity as related to mineralogy Rock is a multi-mineral aggregate of different components, and the petrophysical property of a rock can be considered as the sum of the contributions from individual mineral and organic elements. This idea has been successfully used in the evaluations of several rock physical

20


Fig. 9. Pores of framework minerals. (A). Dissolution-related pores in calcite. (B). Intercrystalline pores in annular pyrite and pyrite crystals. (C). Dissolution-related pores in the rims of feldspar. (D). Dissolution-related fracture in carbonate minerals.

parameters (LeCompte et al., 2009; Wang et al., 2014b; Merkel et al., 2015; Yang et al., 2015). Shale is composed of organic matter, clay minerals and framework minerals (quartz, calcite, pyrite, etc.). The pore volume of shale matrix is made up of the intergranular volume (Vframework), the intragranular volume of individual mineral components (VClay) and the intraparticle volume of organic matter (VOM): V pore ¼ V OM þ V Framework þ V Clay

ð1Þ

Based on the evaluation and interpretation of FE–SEM images and petrophysical characterization, a combined physical model was used

to calculate the pore volume of shales. The expansion of Eq. (1) is V pore ¼ ρbulk V bulk wOM vOM þ wFramework vFramework þ wClay vClay ð2Þ Here Vpore (cm3) is the pore volume of the shale; Vbulk (cm3) is the bulk volume, and ρbulk (cm3/g) is the bulk density; wOM, wFramework, and wClay are the mass fractions of OM, framework minerals, and total clay minerals, respectively; vOM , vFramework , and vClay (cm3/g) is defined as the specific pore volume of OM, framework minerals, and total clay minerals, respectively. Considering that the experimental measured TOC (total organic carbon) is different from the TOM (total organic matter) of shales, a

Fig. 10. Intragranular pores in microfossils. (A). Intergranular pores in silicified mallomonas. (B). Intergranular pores in silicified mallomonas, and the boundary of pores show compaction.


Fig. 11. Evaluation about gas content of a shale sample using USBM direct method (Depth = 1501.44 m, formation pressure = 15.54 MPa, formation temperature = 62.6 °C). The total gas content in the USBM direct method is computed as the sum of the volume of lost gas, measured gas, and remainder gas. The lost gas volume was determined by extrapolating the first 4–6.5 h of measured gas readings to time zero.

proportionality coefficient of 0.85 was used to convert TOC to TOM content (TOC = 0.85 · TOM), and the mass fraction of all the minerals was corrected considering the TOM content. According to the definition of porosity (ϕ = Vpore/Vbulk), the porosity of shales can be expressed as: ϕ ¼ ρbulk

hw

TOC

0:85

vOM þ wFramework vFramework þ wClay vClay

i

ð3Þ

Combined with the bulk density, and mass fraction of individual (mineral and organic matter) components, multi–element nonlinear regression analysis was used to obtain the specific pore volumes of individual components. The regression coefficients are listed in Table 3 and the calculated vs. the measured porosity values are plotted in Fig. 15. The calculated porosities match the helium porosities reasonably, which indicates the feasibility of the developed method. The fitted vOM , vFramework , and vClay of the shale samples is 0.213 cm3/g, 0.0043 cm3/g, 0.0032 cm3/g, respectively, which indicates that the specific pore volume of organic matter is much larger than that of framework and clay minerals.

Fig. 12. Relationship between TOC content and the total gas content of shale samples determined by the USBM method.

21

Fig. 13. Methane excess sorption isotherms of selected shale samples measured at in situ pressure and temperature conditions (lines represent the fitted excess sorption isotherms).

The approach can be used to quantitatively evaluate the contribution of major mineral components to the total porosity of shale. The total porosity of these Lower Silurian shales ranges between 0.7% and 5.7%. The contribution to total porosity of individual minerals (the ratio of ρbulk wi vi to the total porosity) decrease in the order: organic matter (6.4%–85%, average 62%) N framework minerals (10.8%–46.9%, average 25%) N clay minerals (4.2%–48.7%, average 13%). This result indicates that the total porosity of these shales studied is mainly controlled by organic matter, which agrees with the positive relationship between TOC and porosity (Fig. 3a). The approach can be extended to other gas shale systems. It should be noted that this evaluation method should be performed on samples with similar lithotypes. If the differences in characteristic parameters of a series of samples are so substantial that a reasonable prediction cannot be achieved, samples can be divided to several groups to carry out the evaluation. 4.5. Porosity and original gas-in-place The original gas-in-place (OGIP) of shales can be rapidly estimated by the direct USBM method. Here the lost gas volume is estimated by extrapolating the measured amount of desorbed gas at early degassing time at reservoir temperature to the time when desorption began. The

Fig. 14. The relationship between Reqv and porosity of shales from Sichuan Basin.

22


Table 3 Specific pore volume of minerals derived from nonlinear regression to estimate the porosity. Coefficients

Specific pore volume (cm3/g)

Variable

vOM vFramework vClay

0.2126 0.0043 0.0032

Organic matter Framework minerals Clay minerals

“lost gas” estimate involves some uncertainty because it does not necessarily follow the Vemitted vs. √t relationship commonly used for extrapolation. This correlation, however, provides usually satisfactory results. The indirect method of laboratory high-pressure high-temperature gas sorption and porosity measurements can be used to calculate the OGIP when the in situ conditions are known. The total gas in place can be calculated as ntotal ¼ nfree þ nadsorbed þ ndissolved

ð4Þ

Here ntotal, nfree, and nadsorbed are the total gas, free gas, and adsorbed gas specific storage capacities, respectively; ndissolved is the gas storage capacity by dissolution in pore fluids such as bitumen and water, and is neglected here. The free gas and the excess amount of substance1 adsorbed gas capacities can be calculated as nfree ¼

ϕSg ρfree ðP; T Þ ρbulk

nadsorbed ¼ nL

mol kg

ρfree ðP; T Þ P 1− ρads P þ P L ðT Þ

ð5Þ mol kg

ð6Þ

Here nL(mol/kg rock) is the Langmuir (“monolayer”) sorption capacity; PL (MPa) is the Langmuir pressure; P (MPa) is the reservoir pressure; ρfree ðP; TÞ (mol/m3) is the molar density of the free gas at pressure P and temperature T, and ρads (mol/m3) is the density of the adsorbed gas phase; Sgis the gas saturation in the reservoir formation; ρbulk is the bulk density of the shale. The adsorbed gas density is used here as a fitting parameter that accounts for the fact that experimental excess sorption isotherms often exhibit maxima. Various attempts have been made to assess or estimate this value but it is not accessible to direct measurement. The excess sorption by definition and by experimental procedure is the amount of sorptive (gas) that can be stored in a given volume (Vvoid) in the presence of a given mass of adsorbent in addition to the amount of gas that could be accommodated in the same volume (in the absence of sorption effects) at the same pressure and temperature: nexcess ðP; T Þ ¼ ntotal ðP; T Þ−V void ρfree ðP; T Þ

ð7Þ

The excess amount (nexcess) is due to the combined effect of close packing (condensation) of the molecules at the surface of the sorbent, and consumption of a (small) portion of the original void volume by this condensed phase. As the density of the free gas increases at high pressures, the density difference between the sorbed gas and the free gas diminishes (and ultimately may vanish) so that the excess sorption decreases (and may become zero). Taking the free pore space of a given mass of porous rock ðvpore Þ as the void volume available for gas storage, the total gas content by combined volumetric (free gas) and adsorptive storage can be calculated as: ntotal ¼ ρfree ðP; T Þ vpore Sg þ nexcess ðP; T Þ

ð8Þ

1 Molar units (amount of substance) are used here for convenience; these can be readily converted to other unit systems (e.g. volumes at standard pressure and temperature conditions; 273.15 K and 101,325 Pa)

Fig. 15. Porosity values predicted using Eq. (3) vs. measured values of shale samples.

Here vpore (m3/kg rock) is the specific pore volume; nexcess (m3/kg rock) is excess sorption capacity; and the average gas saturation Sgof the Silurian shales from Sichuan Basin is taken as 0.6 (Wang et al., 2013). The calculated OGIP based on the experimental methane sorption isotherms and the original reservoir P/T conditions and the measured total gas content from USBM method of shale samples are shown in Fig. 16. The calculated OGIP agree well with the measured total gas contents (Fig. 16). It should be noted that the porosity used in the calculations were determined by helium pycnometry on unconfined (zero– stress) sample plugs. If the evolution of petrophysical properties with effective stress is considered, the calculated results can be used to simulate the gas production behavior. Detailed investigations of poro– elastic properties of shales will be discussed in future work. Understanding the gas storage behavior in the formation is critical to the accurate assessment of shale gas resources. The OGIP of these Lower Silurian shales from the southern Sichuan Basin ranges from 0.4 to 6.2 m3/t, which are comparable with those of typical North America shales (1.70–6.23 m3/t for Fayetteville shale and 1.70–2.83 m3/t for Marcellus shale) (Curtis, 2002; Sondergeld et al., 2010). These results indicate that the Lower Silurian shales in the Southern Sichuan Basin have good shale gas potential, though the formation has experienced strong tectonic uplifting (Zhou et al., 2014). The quantitative evaluation of the pore system indicates that the contribution of inorganic matter to the total porosity of these Lower Silurian shales in the southern Sichuan Basin is approximately 38%. However, that does not mean that the inorganic matter would also contribute so much (38%) to the OGIP because of moisture (the contribution to pore space is not equivalent to the contribution to gas storage capacity). Moisture is an important factor for gas capacity evaluation in shale gas reservoir systems because the amount and distribution of water can limit the volume of adsorbed and free gas. Organic matter in highly mature shale contributes marginally to the water uptake due to the low amount of hydrophilic oxygen functional groups (Gensterblum et al., 2013; Merkel et al., 2015), and this is also known as the characteristic of oil wetness of organic matter at the macro level (Wang and Reed, 2009). Merkel et al. (2015) have investigated the methane sorption of Haynesville (Ro = 2.5%) and Bossier shale (Ro = 2.2%) and concluded that the methane excess sorption capacity of clay minerals on the reservoir scale is negligible. The residual methane sorption capacity of moisture-equilibrated shales is found to be purely a function of organic matter in over-mature shales. Moisture mainly occupies the surface sites of hydrophilic clay minerals but does not cover the surface of organic matter in over-mature shales.


23

Fig. 16. Comparison between the calculated gas content (in std. m3/t) using Eq. (8) and measured total gas by USBM direct method. (a) CW_61, Depth = 1501.77 m, original formation pressure = 15.54 MPa, temperature = 62.6 °C; (b) CW_81, Depth = 1527.47 m, original formation pressure = 15.80 MPa, temperature = 63.39 °C.

The measured total gas content, calculated free gas and calculated total gas of shale cores are plotted vs. TOC content in Fig. 17. Because the methane sorption on clay minerals in the reservoir is negligible, organic matter accounts for the major portion of the sorptive gas storage capacity. The free gas content at TOC = 0% can be considered as free gas stored in the pore system of the inorganic matrix, and the gas stored in the organic matter is the total gas subtracted from that stored in the inorganic matrix. According to Fig. 17, the OGIP of the Silurian shales from W201 well contains an average of 54% (conservative uncertainties: ±14%) adsorbed gas and 46% (conservative uncertainties: ±14%) free gas under reservoir conditions, while the OGIP of N203 well contains 57% (conservative uncertainties: ±11%) adsorbed gas and 43% (conservative uncertainties: ±11%) free gas. Averagely, the OM-hosted pores approximately are accounted for 78% gas content (55% adsorbed gas plus 23% free gas) of the OGIP, while pores in the inorganic matter accommodates 22% free gas of the OGIP. Thus not only most of the porosity of the shales studied is provided by organic matter, but also the predominant gas content is stored in the OM–hosted pores of these Silurian shales.

(3)

(4)

(5)

(6) 5. Conclusions Multi–methods (FE–SEM, helium pycnometry, and low–pressure gas adsorption) have been adopted to characterize the pore characteristics of Silurian shales from Sichuan Basin, China. New insights on pore system development and gas storage have been drawn: (1) Combination of FE–SEM and argon ion beam milling show that four pore types can be identified in the matrix of the shales studied: OM–hosted pores, pores in clay minerals, pores in framework minerals, and intragranular pores in microfossils. (2) TOC content exerts a strong control on the porosity and its distribution in these shales. The TOC content shows a positive correlation with porosity. Porosity increases with increasing thermal

maturity at Reqv values below 2.5%, but decreases for higher thermal maturity samples. The porosity of organic matter and clay minerals tends to be preserved by the presence of rigid grains that provide a framework preventing these pores from collapsing. The porosity of framework minerals is mainly related to carbonate dissolution. Pore size distributions of TOC-rich samples are unimodal, indicating the contribution of organic matter. PSD of carbonate-rich samples are bimodal. One maximum is in the micropore range, contributed by OM and another maximum is in the range of mesopore–macropores, resulting from the dissolution of carbonate minerals. A procedure has been demonstrated to quantitatively evaluate the contribution of individual mineral components to total porosity. Quantification of porosity as related to mineralogy shows that organic matter contributes approximately 62% to the total porosity. Framework minerals (quartz, feldspar, and carbonates, et al.) and clay minerals contribute 25% and 13%, respectively. The total gas content of these Lower Silurian shales determined from canister desorption tests ranges from 0.4 to 6.2 m3/t, and the total gas contents of selected samples agree with the theoretically estimated OGIP. The OM-hosted pores accounted for about 78% gas content (55% adsorbed gas plus 23% free gas) of the OGIP, while pores in the inorganic matter accommodate 22% of the free gas of the OGIP.

Acknowledgments The authors would like to acknowledge the financial support of the National Natural Science Foundation of China (Grant No.51274214).

Fig. 17. Adsorbed and total gas contents (OGIP) with respect to TOC content at original reservoir conditions of Silurian shales from Southern Sichuan Basin. (a) W201 well; (b) N203 well.

24


References Bernard, S., Horsfield, B., Schulz, H.–.M., Wirth, R., Schreiber, A., Sherwood, N., 2012. Geochemical evolution of organic–rich shales with increasing maturity: a STXM and TEM study of the Posidonia Shale (Lower Toarcian, northern Germany). Mar. Pet. Geol. 31, 70–89. Bustin, R.M., Bustin, A.M.M., Cui, X., Ross, D.J.K., Pathi, V.S.M., 2008. Impact of shale properties on pore structure and storage characteristics. SPE Paper 119892 presented at the SPE shale gas production conference in Fort Worth, Texas, November 16–18, 2008. Chalmers, G.R.L., Bustin, R.M., 2012c. Geological evaluation of Halfway–Doig–Montney hybrid gas shale–tight gas reservoir, northeastern British Columbia. Mar. Pet. Geol. 38, 53–72. Chalmers, G.R.L., Bustin, R.M., Power, I.M., 2012a. Characterization of gas shale pore systems by porosimetry, pycnometry, surface area, and field emission scanning electron microscopy/transmission electron microscopy image analyses: examples from the Barnett, Woodford, Haynesville, Marcellus, and Doig units. AAPG Bull. 96, 1099–1119. Chalmers, G.R.L., Ross, D.J.K., Bustin, R.M., 2012b. Geological controls on matrix permeability of Devonian gas shales in the Horn River and Liard basins, northeastern British Columbia, Canada. Int. J. Coal Geol. 103, 120–131. Clarkson, C.R., Freeman, M., He, L., Agamalian, M., Melnichenko, Y.B., Mastalerz, M., et al., 2012. Characterization of tight gas reservoir pore structure using USANS/SANS and gas adsorption analysis. Fuel 95, 371–385. Curtis, J.B., 2002. Fractured shale-gas systems. AAPG Bull. 86, 1921–1938. Curtis, M.E., Cardott, B.J., Sondergeld, C.H., Rai, C.S., 2012a. Development of organic porosity in the Woodford Shale with increasing thermal maturity. Int. J. Coal Geol. 103, 26–31. Curtis, M.E., Sondergeld, C.H., Ambrose, R.J., Rai, C.S., 2012b. Microstructural investigation of gas shales in two and three dimensions using nanometer-scale resolution imaging. AAPG Bull. 96, 665–677. Dai, J.X., Zou, C.N., Liao, S.M., Dong, D.Z., Ni, Y.Y., Huang, J.L., Wu, W., Gong, D.Y., Huang, S.P., Hu, G.Y., 2014. Geochemistry of the extremely high thermal maturity Longmaxi shale gas, southern Sichuan Basin. Org. Geochem. 74, 3–12. Dan, Y.X., 2009. The orgin analysis of measuring error for vernier calipers. Metrol. Meas. Technol. 36, 12–15. Dong, T., Harris, N.B., Ayranci, K., Twemlow, C.E., Nassichuk, B.R., 2015. Porosity characteristics of the Devonian Horn River shale, Canada: insights from lithofacies classification and shale composition. Int. J. Coal Geol. 141–142, 74–90. Feng, G.X., Chen, S.J., 1988. Relationship between the reflectance of bitumen and vitrinite in rock. Nat. Gas Ind. 8, 20–24. Gasparik, M., Ghanizadeh, A., Bertier, P., Gensterblum, Y., Bouw, S., Krooss, B.M., 2012. High-pressure methane sorption isotherms of black shales from the Netherlands. Energy Fuel 26 (8), 4995–5004. Gasparik, M., Bertier, P., Gensterbluma, Y., Ghanizadeh, A., Krooss, B.M., Littke, R., 2014. Geological controls on the methane storage capacity in organic-rich shales. Int. J. Coal Geol. 123, 34–51. Gasparik, M., Gensterbluma, Y., Ghanizadeh, A., Weniger, P., Krooss, B.M., 2015. Highpressure/high-temperature methane-sorption measurements on carbonaceous shales by the manometric method: experimental and data-evaluation considerations for improved accuracy. SPE J. 20, 1–20. Gensterblum, Y., Merkel, A., Busch, A., Krooss, B.M., 2013. High-pressure CH4 and CO2 sorption isotherms as a function of coal maturity and the influence of moisture. Int. J. Coal Geol. 118, 45–57. Ghanizadeh, A., Amann–Hildenbrand, A., Gasparik, M., Gensterblum, Y., Krooss, B.M., Littke, R., 2014. Experimental study of fluid transport processes in the matrix system of the European organic-rich shales: II. Posidonia Shale (Lower Toarcian, northern Germany). Int. J. Coal Geol. 123, 20–33. Ghanizadeh, A., Bhowmik, S., Haeri–Ardakani, O., Sanei, H., Clarkson, C.R., 2015a. A comparison of shale permeability coefficients derived using multiple non-steady-state measurement techniques: examples from the Duvernay Formation, Alberta (Canada). Fuel 140, 371–387. Ghanizadeh, A., Clarkson, C.R., Aquino, S., Ardakani, O.,.H., Sanei, H., 2015b. Petrophysical and geomechanical characteristics of Canadian tight oil and liquid-rich gas reservoirs: I. pore network and permeability characterization. Fuel 153, 664–681. Ghanizadeh, A., Clarkson, C.R., Aquino, S., Vahedian, A., Ardakani, O.,.H., Sanei, H., 2015c. Impact of solvent–extraction on fluid storage and transport properties of Montney Formation. SPE Paper 175954 Presented at the SPE/CSUR Unconventional Resources Conference in Calgary, Alberta, Canada, October 20–22, 2015. Gregg, S.J., Sing, K.S.W., 1982. Adsorption, Surface Area and Porosity. second ed. Academic Press, New York. Jarvie, D.M., Hill, R.J., Ruble, T.E., Pollastro, R.M., 2007. Unconventional shale–gas systems: the Mississippian Barnett shale of north–central Texas as one model for thermogenic shale–gas assessment. AAPG Bull. 91, 475–499. Kissell, F.N., McCulloch, C.M., Elder, C.H., 1973. The directmethod of determiningmethane content of coal beds for ventilation design. US Bureau of Mines, Report of Investigations. 7767, pp. 1–17. Klaver, J., Desbois, G., Urai, J.L., Littke, R., 2012. BIB–SEM study of the pore space morphology in earlymature Posidonia Shale fromthe Hils area, Germany. Int. J. Coal Geol. 103, 12–25. Klaver, J., Desbois, G., Littke, R., Urai, J.L., 2015a. BIB–SEM characterization of pore space morphology and distribution in postmature to overmature samples from the Haynesville and Bossier shales. Mar. Pet. Geol. 59, 451–466. Klaver, J., Hemes, S., Houben, M., Desbois, G., Radi, Z., Urai, J.L., 2015b. The connectivity of pore space in mudstones: insights from high–pressure Wood's metal injection, BIB– SEM imaging, and mercury intrusion porosimetry. Geofluids 1–15.

Krooss, B.M., Van Bergen, F., Gensterblum, Y., Siemons, N., Pagnier, H.J., David, P., 2002. High-pressure methane and carbon dioxide adsorption on dry and moisture-equilibrated Pennsylvanian coals. Int. J. Coal Geol. 51 (2), 69–92. LeCompte, B., Franquet, J.A., Jacobi, D., 2009. Evaluation of Haynesville shale vertical well completions with a mineralogy based approach to reservoir geomechanics. SPE Paper 124227 Presented at the SPE Annual Technical Conference and Exhibition in Louisiana, USA, 4–7 October, 2009. Loucks, R.G., Reed, R.M., Ruppel, S.C., Jarvie, D.M., 2009. Morphology, genesis, and distribution of nanometer-scale pores in siliceous mudstones of the Mississippian Barnett Shale. J. Sediment. Res. 79, 848–861. Loucks, R.G., Reed, R.M., Ruppel, S.C., Hammes, U., 2012. Spectrum of pore types and networks in mudrocks and a descriptive classification for matrix–related mudrock pores. AAPG Bull. 96, 1071–1098. Mastalerz, M., He, L., Melnichenko, Y.B., Rupp, J.A., 2012. Porosity of coal and shale: insights from gas adsorption and SANS/USANS techniques. Energy Fuel 26, 5109–5120. Merkel, A., Fink, R., Littke, R., 2015. The role of pre-adsorbed water on methane sorption capacity of Bossier and Haynesville shales. Int. J. Coal Geol. 147–148, 1–8. Milliken, K.L., Rudnicki, M., Awwiller, D.N., Zhang, T.W., 2013. Organic matter-hosted pore system, Marcellus Formation (Devonian), Pennsylvania. AAPG Bull. 97, 177–200. Passey, Q.R., Bohacs, K.M., Esch, W.L., Klimentidis, R., Sinha, S., 2010. From oil-prone source rock to gas-producing shale reservoir–geologic and petrophysical characterization of unconventional shale–gas reservoirs. SPE Paper 131350 Presented at the CPS/SPE International Oil & Gas Conference and Exhibition in Beijing, China, 8–10 June, 2010. Potter, P.E., Maynard, J.B., Depetris, P.J., 2005. Mud & Mudstones. Spring Verlag, Berlin. Rexer, T.F.T., Benham, M.J., Aplin, A.C., Thomas, K.M., 2013. Methane adsorption on shale under simulated geological temperature and pressure conditions. Energy Fuel 27, 3099–3109. Rexer, T.F.T., Mathia, E.J., Aplin, A.C., Thomas, K.M., 2014. High-pressure methane adsorption and characterization of pores in Posidonia shales and isolated kerogens. Energy Fuel 28, 2886–2901. Ross, D.J.K., Bustin, R.M., 2009. The importance of shale composition and pore structure upon gas storage potential of shale gas reservoirs. Mar. Pet. Geol. 26, 916–927. Ruppert, L.F., Sakurovs, R., Blach, T.P., He, L., Melnichenko, Y.B., Mildner, D.F.R., Alcantar– Lopez, L.,., 2013. A USANS/SANS study of the accessibility of pores in the Barnett Shale to methane and water. Energy Fuel 27, 772–779. Schieber, J., 2010. Common themes in the formation and preservation of intrinsic porosity in shales and mudstones: illustrated with examples from across the Phanerozoic. SPE Paper 132379 Presented at the SPE Unconventional Gas Conference in Pittsburgh, Pennsylvania, 23–25 February, 2010. Slatt, R.G., O'Brien, N.R., 2011. Pore types in the Barnett and Woodford gas shales: contribution to understanding gas storage and migration pathways in fine-grained rocks. AAPG Bull. 95, 2017–2030. Sondergeld, C.H., Ambrose, R.J., Rai, C.S., Moncriff, J., 2010. Micro-structural studies of gas shales. SPE Paper 131771 Presented at the SPE Unconventional Gas Conference, in Pittsburgh, Pennsylvania, 23–25 February, 2010. Surdam, R.C., Crossey, L.J., Hagen, E.S., Heasler, H.P., 1989. Organic–inorganic interactions and sandstone diagenesis. AAPG Bull. 73, 1–23. Tian, H., Pan, L., Xiao, X.M., Wilkins, R.W.T., Meng, Z.P., Huang, B.J., 2013. A preliminary study on the pore characterization of Lower Silurian black shales in the Chuandong Thrust Fold Belt, southwestern China using low pressure N2 adsorption and FE– SEM methods. Mar. Pet. Geol. 48, 8–19. Walls, J., Sinclair, S., 2011. Digital rock physics provide critical insights to characterize Eagle Ford. The American Oil & Gas Reporter, pp. 1–4. Wang, F.P.,., Reed, R.M.,., 2009. Pore networks and fluid flow in gas shales. SPE Paper 124235 Presented at the SPE Annual Conference and Exhibition in Louisiana, USA, 4–7 October, 2009. Wang, Y.M., Dong, D., Li, J., Wang, S., Li, X., Wang, L., Cheng, K., Huang, J., 2012. Reservoir characteristics of shale gas in Longmaxi Formation of the Lower Silurian, southern Sichuan. Acta Pet. Sin. 33, 551–565. Wang, F.Y., Guan, J., Feng, W.P., Bao, L.Y., 2013. Evolution of overmature marine shale porosity and implication to the free gas volume. Pet. Explor. Dev. 40, 819–824. Wang, Y.M., Dong, D.Z., Cheng, X.Z., Huang, J.L., Wang, S.F., Wang, S.Q., 2014a. Electric property evidences of the carbonification of organic matters in marine shales and its geologic significance. Nat. Gas Ind. 34, 1–7. Wang, Y.M., Dong, D.Z., Yang, H., He, L., Wang, S.Q., Huang, J.L., Pu, B.L., Wang, S.F., 2014b. Quantitative characterization of reservoir space in the Lower Silurian Longmaxi Shale, southern Sichuan, China. Sci. China Earth Sci. 57, 313–322. Yang, Y.L., Aplin, A.C., 2007. Permeability and petrophysical properties of 30 natural mudstones. J. Geophys. Res. 112 (B03206). Yang, F., Ning, Z.F., Liu, H.Q., 2014. Fractal characteristics of shales from a shale gas reservoir in the Sichuan Basin, China. Fuel 115, 378–384. Yang, F., Ning, Z.F., Zhang, R., Zhao, H.W., Krooss, B.M., 2015. Investigations on the methane sorption capacity of marine shales from Sichuan Basin, China. Int. J. Coal Geol. 146, 104–117. Zhang, T.W., Ellis, G.S., Ruppel, S.C., Milliken, K., Yang, R.S., 2012. Effect of organic–matter type and thermal maturity on methane adsorption in shale–gas systems. Org. Geochem. 47, 120–131. Zhou, Q., Xiao, X.M., Tian, H., Pan, L., 2014. Modeling free gas content of the Lower Paleozoic shales in the Weiyuan area of the Sichuan Basin, China. Mar. Pet. Geol. 56, 87–96. Zou, C.N., Dong, D.Z.,., Wang, S.J., Li, J.Z., Li, X.J., Wang, Y.M., Li, D.H., Cheng, K.M., 2010. Geological characteristics, formation mechanism and resource potential of shale gas in China. Pet. Explor. Dev. 37, 641–653.