Pore-types and pore-network evolution in Upper Devonian-Lower Mississippian Woodford and Mississippian Barnett mudstones: Insights from laboratory thermal maturation and organic petrology

International Journal of Coal Geology xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

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

Pore-types and pore-network evolution in Upper Devonian-Lower Mississippian Woodford and Mississippian Barnett mudstones: Insights from laboratory thermal maturation and organic petrology Lucy T. Koa,b,⁎, Stephen C. Ruppela, Robert G. Loucksa, Paul C. Hackleyc, Tongwei Zhanga, Deyong Shaod a Bureau of Economic Geology, John A. and Katherine G. Jackson School of Geosciences, The University of Texas at Austin, University Station, Box X, Austin, TX 78713, United States b Department of Geological Sciences, John A. and Katherine G. Jackson School of Geosciences, The University of Texas at Austin, Austin, TX 78712, United States c U.S. Geological Survey, 956 National Center, Reston, VA 20192, United States d The School of Geosciences, Lanzhou University, PR China

A R T I C L E I N F O

A B S T R A C T

Keywords: Barnett Woodford Macerals OM pore evolution AOM Bituminite Tasmanites

Pore-evolution models from immature organic-matter (OM) -rich Barnett (0.42%Ro) and Woodford (0.49%Ro) mudstones were compared with models previously developed from low-maturity OM-lean Boquillas (Eagle Fordequivalent) mudstones to investigate whether (1) different mineralogy (siliceous vs. calcareous) exerts different catalytic and sorption effects and influences OM-pore origin and evolution; and (2) different types of macerals show different OM pore evolution history. Laboratory gold-tube pyrolysis, scanning electron microscopy (SEM) and thin-section petrography, organic petrography, and geochemical characterization were used to investigate the role of bulk mineralogy, maceral type, and thermal maturation on OM-pore evolution. Results suggest that mineralogy has little impact on OM-pore development and evolution. Macerals, identified using both SEM (platy OM, particulate OM, organic–mineral admixtures, Tasmanites) and organic petrology (vitrinite, inertinite, amorphous organic matter [AOM]/bituminite, telalginite [Leiosphaeridia, Tasmanites]), do affect the origin and evolution of OM pores owing to differences in chemical compositions, generation kinetics, and activation-energy distributions between Tasmanites, matrix bituminite, and other types of macerals. Leiosphaeridia and Tasmanites in Woodford mudstone samples exhibit a delay in onset and a shorter period of petroleum generation and pore development compared to the matrix bituminite in the Barnett and Woodford mudstone samples. Pre-oil solid bitumen was observed to have migrated into initial primary mineral pore networks at the bitumen generation stage in both Barnett and Woodford samples. At higher levels of thermal maturation, the volume of primary mineral pores decreases and the pore volume composed of modified mineral pores and OM pores becomes greater. Pore evolution and pore-type heterogeneity in these mudstones is a function of the initial mineral pore network, types of kerogen and macerals, and generation kinetics of individual macerals upon thermal maturation.

1. Introduction It has been well documented that the formation of organic matter (OM) pores in mudstones is predominantly affected by thermal maturation processes (e.g., Bernard and Horsfield, 2014; Bernard et al., 2012a,b; Bohacs et al., 2013; Curtis et al., 2012; Ko et al., 2016; Loucks et al., 2009; Mastalerz et al., 2013; Milliken et al., 2013; Pommer and Milliken, 2015; Zargari et al., 2015). The change in size and shape of OM pores is related to continuous transformation of OM and can be associated with generated bitumen, pre-oil solid bitumen, oil, gas, ⁎

pyrobitumen, and char (e.g., Bernard et al., 2012a; Cardott et al., 2015; Curiale, 1986; Jacob, 1989; Ko et al., 2016; Landis and Castaño, 1995; Löhr et al., 2015; Loucks and Reed, 2014; Mastalerz and Glikson, 2000; Wood et al., 2015). The effect of mineralogy (clay mineralogy in particular) on the composition and yield of generated petroleum—with variations caused by differences in mineral adsorption and catalytic capacities—has been known for decades (e.g., Goldstein, 1983; Hetényi, 1995; Huizinga et al., 1987a,b; Johns, 1979; Pan et al., 2010; Tannenbaum et al., 1986; Tannenbaum and Kaplan, 1985; Wei et al., 2006a,b). Catalytic cracking

Corresponding author at: Bureau of Economic Geology, The University of Texas at Austin, University Station, Box X, Austin, TX 78713, United States. E-mail addresses: [email protected] (L.T. Ko), [email protected] (S.C. Ruppel), [email protected] (R.G. Loucks), [email protected] (P.C. Hackley).

http://dx.doi.org/10.1016/j.coal.2017.10.001 Received 11 May 2017; Received in revised form 2 October 2017; Accepted 3 October 2017 0166-5162/ © 2017 Elsevier B.V. All rights reserved.

Please cite this article as: Ko, L.T., International Journal of Coal Geology (2017), http://dx.doi.org/10.1016/j.coal.2017.10.001

International Journal of Coal Geology xxx (xxxx) xxx–xxx

L.T. Ko et al.

amounts of greenish gray mudstone, sandstone, dolostone, phosphate nodules, and pyrite (Cardott and Chaplin, 1993; Comer, 2007). In general, chert, composed of radiolarian remains and diagenetic silica, is more abundant in distal parts of the basin; mudstone, siltstone, and sandstone are more abundant in proximal areas. Siliceous mudstones and chert, composed primarily of microcrystalline, cryptocrystalline, and microfibrous quartz, are the two main lithofacies in the Woodford Formation at the study location. Woodford Formation rocks are, in general, stratigraphically age-equivalent to other OM-rich mudstone deposits, including the Chattanooga Shale in the Black Warrior Basin, the Antrim Shale in the Michigan Basin, the New Albany Shale in the Illinois Basin, the Marcellus Shale in the Appalachian Basin, and the Bakken Formation in the Williston Basin (Blakey, 2009; Meissner, 1978). The Woodford Shale, long known as the source rock for most of Oklahoma's petroleum reserves, has become an active target for shale-gas and shale-oil development in the Anadarko, Arkoma, and Marietta-Ardmore Basins (Cardott, 2011). In the area of investigation, Woodford mudstone is conformably overlain by the Mississippian Sycamore Formation, which is composed of silty limestone interbedded with dark shale and unconformably overlies the Silurian and Lower Devonian Hunton Group (Fishman et al., 2013; Ham, 1973).

of kerogen and/or bitumen by interaction of minerals with OM has been proposed by many previous studies (Espitalié et al., 1980, 1984; Hetényi, 1995; Horsfield and Douglas, 1980; Johns, 1979; Tannenbaum et al., 1986; Tannenbaum and Kaplan, 1985). Minerals in rocks can catalyze chemical reactions (i.e., CeC bond cleavage, defunctionalization, condensation, oxidation, and reduction) occurring during the maturation of kerogen. Much effort has been made to investigate the role of minerals in petroleum composition during the thermal maturation, specifically in regard to the presence or absence of minerals such as calcite, pyrite, quartz, calcium sulfate, illite, kaolinite, and montmorillonite (i.e. smectite). For example, montmorillonite shows a pronounced catalytic effect, significantly changing the chemical composition and yield of oil from kerogen, whereas other minerals such as calcite, pyrite, quartz, and calcium sulfate show weaker or no effect (e.g., Lao et al., 1989; Tannenbaum et al., 1986). These effects are related to montmorillonite having a high surface area and exposed cations on its surface (Johns, 1979); most importantly, montmorillonite contains sites where organic reactions can be catalyzed (Johns, 1979; Wei et al., 2006a,b). Both montmorillonite and illite can alter bitumen composition by adsorbing a considerable amount of asphaltene and polar compounds, whereas calcite and other non-clay minerals show little or no adsorption (Huizinga et al., 1987a,b; Tannenbaum and Kaplan, 1985). Although these studies have shown the effects of mineralogy on petroleum composition, the role of bulk mineralogy on pore development in unconventional reservoirs during maturation has received comparatively little investigation. The process by which pores evolve within different macerals during thermal maturation also remains unclear. Curtis et al. (2012), Loucks et al. (2012), Milliken et al. (2013), and Schieber (2010) have suggested that the type of kerogen (i.e., Type I, II, III, IV) might play a role in the formation of pores during thermal maturation. Variations in the relative proportion of macerals can affect geochemical properties of bulk OM and petroleum-generating potential (e.g., Horsfield et al., 1988; Schenk et al., 1990; Mastalerz et al., 2012). Only in the last several years, has the idea been introduced that OM pores may vary as a function of maceral type and the importance of integrating optical organic petrology with SEM petrography to distinguish different OM types (solid bitumen vs. different types of maceral) in which porosity is developed been recognized (Cardott et al., 2015; Cardott and Curtis, 2017; Fishman et al., 2012; Hackley and Cardott, 2016; Mastalerz et al., 2013). To investigate if bulk mineralogy, kerogen and maceral types, and thermal maturation change the shape and size of OM pores and the evolution of OM-pore and mineral-pore networks, we applied laboratory gold-tube pyrolysis, mineralogical and geochemical characterization, organic petrology, and SEM petrography methods. This study compares the pore-evolution model developed by Ko et al. (2016, 2017) for OM-lean Upper Cretaceous Eagle Ford mudstones to pore evolution in the Woodford and Barnett mudstones analyzed here. Specific research questions examined include the following: (1) Do differences in maceral type affect OM pore development and evolution? (2) Does bulk mineralogy, especially clay mineralogy, affect the timing of OM transformation and pore evolution?

2.2. Barnett mudstones

2. Geologic setting

Mississippian Barnett siliciclastic mudstones were deposited in a deeper-water foreland basin, the Fort Worth Basin (FWB), during the late Paleozoic Ouachita orogeny as the Gondwana plate approached the Laurussia craton (Loucks and Ruppel, 2007; Meckel et al., 1992). Barnett mudstones reach a maximum thickness of about 1000 ft. (304.8 m) in the northeast near the Muenster Arch and thin to the west and south (Montgomery et al., 2005; Pollastro et al., 2007). The Ouachita thrust–fold belt bounds the FWB to the east. The Muenster, Red River, Bend, and Lampasas Arches, as well as the Llano Uplift bound the FWB on the west and north (Montgomery et al., 2005; Pollastro et al., 2007). The Barnett Shale is composed of a mixture of siliceous mudstone, calcareous siliceous mudstone, argillaceous limestone, skeletal packstone, and phosphatic packstone to grainstone facies (Hickey and Henk, 2007; Loucks and Ruppel, 2007; Redmond, 2016). The Barnett Shale in the northern FWB has been studied extensively and developed as a general gas-shale reservoir model based on many pioneering studies (Bowker, 2007; Han et al., 2015; Jarvie et al., 2007; Loucks and Ruppel, 2007; Milliken et al., 2007; Milliken et al., 2012; Montgomery et al., 2005; Pollastro et al., 2007). Natural gas was first produced from Barnett strata starting in 1981 (Bowker, 2007). The Forestburg limestone subdivides the Barnett in the northern FWB into upper and lower units but is absent in the southern FWB (Loucks and Ruppel, 2007; Montgomery et al., 2005; Pollastro et al., 2007). The Barnett Formation unconformably overlies the Ordovician Ellenburger, Simpson, and Viola shelf carbonates and is directly overlain by the phosphatic glauconitic lime packstone unit and Pennsylvanian Marble Falls Group carbonates (Montgomery et al., 2005; Wood, 2013). In the southern FWB, the Barnett has been shown to unconformably overly the Mississippian Whites Crossing unit, the Devonian Houy Formation, or the Lower Ordovician Ellenburger Group (Redmond, 2016).

2.1. Woodford mudstones

3. Samples and methods

Upper Devonian–Lower Mississippian Woodford mudstones in southern Oklahoma were deposited in an epicontinental sea during global sea-level transgression, within a failed aulacogen structure (Blakey, 2009; Cardott and Chaplin, 1993; Ham, 1973; Lambert, 1993; Nicholas and Rozendal, 1975). From the Cambrian through Permian, the basin was primarily filled with carbonates and marine mudstones. The Woodford marine mudstone was deposited in a deeper-water euxinic environment and is siliceous, containing chert and subordinate

3.1. Core and outcrop samples The Devonian–Mississippian Woodford mudstone samples used in this study were collected by Dr. Geoffrey S. Ellis of the U.S. Geological Survey (Denver, Colo.) from a well-studied (e.g., Aufill, 2007; Ham, 1973; Kirkland et al., 1992; Krystyniak, 2005; Paxton and Cardott, 2008) road cut along Interstate 35 on the south flank of the Arbuckle Anticline in Carter County, Oklahoma, USA (Fig. 1). The outcrop has 2

International Journal of Coal Geology xxx (xxxx) xxx–xxx

L.T. Ko et al.

Fig. 1. Map of Texas and Oklahoma showing the distribution of the Barnett Formation, the general structural features, the location of the core (Lee C-5-1, Brown County, Tex.) in the Fort Worth Basin, and the outcrop location (‘X”) of the Woodford Mudstone used in this study. Modified after Fishman et al. (2013), Loucks and Ruppel (2007), and Redmond (2016).

We used a Leica DM4000 microscope with LED illumination and monochrome camera detection for reflectance analysis with computer program DISKUS-FOSSIL by Hilgers Technisches Buero and a Klein and Becker YAG calibration standard (0.908%Ro). Maceral identification was also performed using incident and transmitted-light petrography of thin sections.

been studied extensively and detailed stratigraphic and sedimentological analyses have been done. Chert and siliceous mudstone are the two major lithofacies in the Woodford Formation. We sampled both chert and mudstone lithofacies but for this study focused on the pore evolution of the siliceous mudstone sample. Mississippian Barnett mudstone subsurface core samples were collected from a shallow core, Houston Oil and Minerals, Lee, Sam, No. C5-1, located in Brown Country, Texas in the southwestern Fort Worth Basin (Fig. 1). This well was drilled by water-based mud. A sample (2.0 in. by 1.5 in.) was collected from the core at 1278 ft. (389.5 m). We are aware that weathering might have oxidized organic and inorganic components in the Woodford outcrop samples exposed at the surface and could affect experimental results. Therefore, geochemical analyses and petrographic examination were applied prior to heating experiments to examine the extent of weathering. Pyrite, total organic carbon (TOC) content, vitrinite reflectance, and Rock-Eval parameters, that are all very sensitive to weathering, do not show signs of alteration or anomaly (Littke et al., 1991; Lo and Cardott, 1995; Petsch et al., 2000, 2001). However, we did observe < 10% of pitted surface under SEM examination in both siliceous mudstone and chert facies of the Woodford Formation. The pitted surface is likely resulted from weathering of carbonate minerals such as calcite and dolomite. None of the siliceous minerals, pyrite, and OM were observed to have been altered under light and electron microscopy. Since the carbonate is a minor component in the Woodford mudstone, we think this slightly altered outcrop sample can be used to observe the evolution of pore in the OM through artificially designed heating experiments.

3.3. Artificial gold-tube anhydrous pyrolysis and geochemical analysis Eight small-diameter rock cylinders (6 mm in diameter and 2–3 cm in length) were drilled perpendicular to bedding planes from each collected sample (Barnett siliceous mudstone, Woodford chert, and Woodford siliceous mudstone). In total, twenty-four samples were drilled, and one original sample from each lithofacies was used as a control group. Twenty-one cylindrical samples were pyrolyzed in gold tubes at the Guangzhou Institute of Geochemistry, Chinese Academy of Sciences. Samples were neither vacuum-dried nor oven-dried before being placed in gold tubes, so it is possible that some irreducible water may have been present. Anhydrous gold-tube pyrolysis experiments were conducted on 21 Barnett and Woodford cylindrical samples (7 for each). The gold tubes had an internal diameter of 6 mm and a wall thickness of 0.45 mm, and each was between 70 and 75 mm long, giving a total reactor volume of approximately 1.0 mL. Prior to loading the samples, the open-ended tubes were heated to 600 °C to remove any potential residual organic material contaminating the tubes. The tubes were then welded at one end using an Argon arc-welder. Each gold tube was placed in a separate stainless steel autoclave and inserted into a pyrolysis oven. The anhydrous pyrolysis experiments were conducted under isothermal conditions at temperatures of 130, 300, 310, 333, 367, 400, and 425 °C for 72-hr reaction time per experiment. A constant confining pressure was maintained at approximately 68.85 MPa (10,000 psi) by pumping water through the autoclave during pyrolysis to prevent rupturing of the gold tubes. When the 72-hr duration was reached, the autoclave was withdrawn from the oven and rapidly quenched to room temperature in cold water. After the autoclaves were depressurized, the gold tubes were taken out and placed in a vacuum line with a residual pressure of 0.1 Pa. The gold tube was pierced using a needle to allow the product gases (C1–C5, CO2, H2S, and H2) to be volatilized into the glass vacuum line. A dry ice and acetone (−80 °C) trap was used to collect C7–C14 light hydrocarbons.

3.2. Organic petrography Thin sections and polished pellets were made from Woodford outcrop and Barnett core samples for organic petrographic analysis, which was completed using a Zeiss Axio Imager A1m microscope equipped with white incident and transmitted light and epifluorescence illumination. Imaging was done with a Zeiss MRc digital camera at total magnifications of 200–500×, with the highest magnification using a 1.0 NA oil-immersion objective. An aliquot of 1 g or less crushed and sieved rock sample was mounted and polished to form pellets for analyses via ASTM D2797 (ASTM, 2016). Maceral identification and reflectance measurement (via ASTM D7708; ASTM, 2016) were performed on these polished pellets. 3

International Journal of Coal Geology xxx (xxxx) xxx–xxx

L.T. Ko et al.

Table 1 Bulk mineralogy (vol%) of the Barnett and Woodford mudrocks. The used conversion factor of TOC to OM is 1.24 (McIver, 1967; Tyson, 1995). Carbonates Sample #

Quartz

Plagioclase

Barnett siliceous mudstone Woodford siliceous mudstone Woodford chert

37.5

1.9

40.5

K-feldspar

6.7

Calcite

Dolomite

1.8

1.7

2.4

Clay minerals Ankerite

0.7

86.5

OM

Siderite

Pyrite

Illite + mica

I/S mixedlayer

2.5

5.0

22.5

9.0

1.3

11.7

11.7

1.00

4.5

Smectite

Kaolinite

Chlorite

5.0 4.1

13.0 21.0 8.0

electron (BSE), secondary electron (SE), and SE through-the-lens detector (TLD) images were collected at various instrument magnifications ranging from 124× to 80,000 × (horizontal field width = 2.41 mm–0.001 mm). Through-the-lens (TLD) detector was used for high-resolution imaging. Identification of mineral components was completed by thin-section petrography and SEM X-ray energydispersive spectroscopy (EDS) detector mapping. Two 30-mm2 Bruker XFlash silicon-drift energy-dispersive X-ray detectors were used for elemental identification, under an accelerating voltage of 15 keV, a spot size of 3.0–4.0, and a total count time of > 700 s. The best way to carry out the correlative microscopy study that combines organic petrology and SEM petrography is to examine the exact same area of interest. Because the amount of our samples is very limited, we could not prepare all samples according to methods used in Cardott and Curtis (2017), Fishman et al. (2012), and Hackley et al. (2017). Our approach to identify and examine macerals in mudstones by SEM is described below: (1) Macerals were recognized first in reflected white light, transmitted light, and epifluorescent light at magnifications of 200–500 ×. (2) To recognize macerals under SEM, low magnification images of 100–1500 × (horizontal field width [HFW]: 2.78 mm–199 μm) were required to identify maceral types in backscattered electron (BSE) mode. (3) Macerals were then examined at higher magnification (4,000–80,000 ×) in BSE mode to observe structure and porosity. This approach would eliminate the limitation of most high-resolution SEM images that observe areas relatively unrepresentative. All pore evolution were compared and observed at the same SEM scale. The changes in porosity, pore size, and shape were not quantified but qualitatively described based on visual estimation at the same scale.

The total number of moles of gas were calculated assuming ideal gas behavior. All generated petroleum (gas, liquid, and solid) was collected for compositional analyses. Identification and quantification of individual hydrocarbon (HC) and non-HC gas components were carried out using a two-channel Hewlett-Packard 6890 Series Gas Chromatograph (GC) that was custom-configured by Wasson ECE Instrumentation. The details of the GC operation conditions are described by Zhang et al. (2007). The yield of light hydrocarbons (C7–C14) was determined by GC. A known amount of C24D50 internal standard (ID) was added in a C7–C14 mixture with CH2Cl2 as the solvent, and the ratio of the ID content to its GC peak area was defined as a calibration factor. The yield of C7–C14 light oil was derived from the sum of the C7–C14 peak area multiplied by the calibration factor. Following GC analysis of C7–C14 hydrocarbons, the remaining solution was returned to the original vial, and the n-hexane solvent was allowed to evaporate under ambient conditions. The miniature core plug was then pulverized and placed into the vial. Approximately 4 mL CH2Cl2 was added into the vial and subjected to five 5-min ultrasonic treatments before settling for 72 h. After settling, both solid and liquid materials were carefully transferred to a filter funnel equipped with a 0.2 μm filter paper and washed with CH2Cl2. The filtered liquid was aspirated using a low vacuum, and the solid residue on the filter was reclaimed and then subjected to a second CH2Cl2 extraction. This extraction procedure was repeated multiple times until the solution was colorless. The generated extracted organic matter (EOM: C14 + compounds) was collected in 1.5 mL glass vials. After evaporating the solvent, the yield of EOM was determined by the weight difference of the glass vial before and after loading EOM. Saturate, aromatic, resins, and asphaltene (SARA) separation of the extract and quantification of extract fractions followed the procedures described by Bastow et al. (2007). A portion of the pyrolyzed cylindrical sample was pulverized. Part of the pulverized sample was analyzed (by GeoMark) for TOC using a LECO instrument and for Rock-Eval using Rock-Eval II instrument and conventional temperature programs (S1: 300 °C for 3 min; S2: 300 to 550 °C at 25 °C/min, hold at 550 °C for 1 min; S3: between 300 and 390 °C). The remaining sample was analyzed by nitrogen adsorption to investigate pore sizes ranging from 3 to 200 nm (not reported in this paper).

3.5. Mineralogical characterization KT GeoServices Inc., Colorado, provided X-ray diffraction (XRD) analysis for bulk mineralogy (Table 1). X-ray Diffraction (XRD) analyses were performed using a Siemens D500 automated powder diffractometer equipped with a copper X-ray source (40 kV, 30 mA) and a scintillation X-ray detector. The whole rock samples were analyzed over an angular range of 5° to 60° 2θ at a scan rate of one degree per minute. The glycol solvated oriented clay mounts were analyzed over an angular range of 2° to 36° 2θ at a scan rate of one degree per minute. Semiquantitative determinations of whole-rock mineral amounts were done using Jade Software (Materials Data, Inc.) with the Whole Pattern Fitting option. Detection limits for XRD are on the order of 1 to 5 wt%. The detection limits differ for each mineral species.

3.4. SEM petrography A flat surface was prepared from a portion of each post-pyrolysis Barnett and Woodford rock cylinder (without solvent extraction) for SEM analysis by Ar ion-beam milling using a Leica EM TIC020 Triple Ion Beam Miller. Because organic matter in the Woodford chert samples was solvent (CH2Cl2) extracted, these were not used for pore description. Each sample was milled for 10 h using an accelerating voltage of 8 keV and a current of 2.8 mA. An FEI Nova NanoSEM 430, fieldemission scanning electron microscope (FE-SEM) located at the Bureau of Economic Geology in Austin, Texas, was used to image pores and their association with OM and mineral grains under an accelerating voltage of 10–15 keV and a working distance of 4–5 mm. Backscattered

4. Working definitions used in this study Although the geochemical and petrographical definitions of kerogen, maceral, organic matter (OM), pre-oil solid bitumen, post-oil solid bitumen, migrabitumen, bitumen, pyrobitumen, asphalt, and char have been discussed and evolved in literature (e.g. Abraham, 1945; Curiale, 1986; Hwang et al., 1998; Jacob, 1989; Mastalerz and Glikson, 2000; Bohacs et al., 2013), the meanings of all these terms are not 4

International Journal of Coal Geology xxx (xxxx) xxx–xxx

L.T. Ko et al.

A

A-1

Stringy OM

Quartz

Phosphate peloid

OM-mineral admixture

Phosphate peloid

B Pyrite

B-1

Fig. 2. Thin-section and SEM photomicrographs of representative areas under plane polarized light (PPL) of Barnett and Woodford mudstones. Photomicrographs showing (A) Barnett Shale, laminated siliceous mudstone facies. Matrix is mainly composed of quartz, clay minerals, and organic matter (OM). (A-1) OM-rich, silty, argillaceous, siliceous mudstone. OM is largely amorphous. (B) Woodford Shale, weakly laminated chert facies. Radiolarian microfossils. Some radiolaria are filled and replaced by calcite or partly replaced by pyrite. (B-1) Radiolaria within mixed illite, OM, and microcrystalline quartz matrix. Some radiolaria are filled with OM. (C) Woodford Shale, laminated siliceous mudstone facies. Abundant compressed Tasmanites OM are the dominant form of OM in the Woodford siliceous mudstone. (C-1) Silty, Tasmanites-rich, argillaceous, siliceous mudstone. Some Tasmanites are filled with quartz. Abundant pyrite framboids are in the matrix.

Radiolarian

Radiolarian

Particulate OM

Radiolarian

agreed upon universally. Bitumen, solid bitumen, solid oil bitumen, migrabitumen, pre-oil solid bitumen, post-oil solid bitumen, and pyrobitumen are classical generic names for each bitumen and are defined based on solubility, fusibility, formation processes, thermal maturity, and geochemical properties. There is also a wide range of variation in bitumen reflectance and quantity and quality of extractable organic matter (EOM). Solid bitumen can occur in reservoirs as well as in the source rocks. However, there is little direct observation and discussion about solid bitumen in source rock reservoirs. Therefore, we provide here our working definitions for the terms we have used in this paper.

•

•

• Organic matter or organic material (OM): a general term referring to •

any liquid or solid materials enriched in organic carbon. Organic matter can have many forms, such as kerogen, bitumen, solid/solidified bitumen, residual/retained OM, pyrobitumen, and char. Kerogen: condensed and insoluble solid organic matter with no unique molecular structure; can be defined using bulk elemental composition and average molecular distributions (Walters, 2007). “Optically, kerogen can be described as mixtures of amorphous organic matter and macerals” (Walters, 2007). Kerogen will be converted to petroleum products when subjected to increased thermal maturation. Based on its origin, quality (hydrogen content), and depositional environments, kerogen is classified as Type I

• 5

(lacustrine), Type II (marine), Type I-S, II-S (sulfur-rich), Type III (terrigenous, gas-prone), and Type IV (terrigenous, inert carbon). Maceral: microscopically recognizable and morphologically distinct constituents of organic matter in mudstones and coals. The term maceral was first used by Stopes in 1935 (Tyson, 1995). Individual macerals have distinct physical and chemical properties. According to “ICCP System 1994,” macerals can be divided into vitrinite (International Committee for Coal Petrology (ICCP), 1998), inertinite (ICCP, 2001), huminite (Sykorova et al., 2005), and liptinite (Pickle et al., 2017). Bitumen: soluble in organic solvents (e.g. dichloromethane, toluene, carbon disulfide, and chloroform etc.), generally viscous, liquid organic matter, derived from (1) breakdown of less resistant biomacromolecules during diagenesis, and (2) thermal cracking of immature or low-maturity kerogen during catagenesis. Bitumen is a generic term and can have a wide range of viscosities (e.g., Bohacs et al., 2013; Loucks and Reed, 2014). Bitumen includes components such as hydrocarbon, resins, asphaltenes, and other non-hydrocarbons and is rich in polar components such as resins and asphaltenes (Tissot and Welte, 1984). Bitumen can be further cracked to oil and gas with increased thermal maturation. Solid/solidified bitumen: once mobile bitumen or subsurface petroleum fluids; solid or highly viscous at surface conditions due to

International Journal of Coal Geology xxx (xxxx) xxx–xxx

L.T. Ko et al.

•

•

expulsion of volatiles on the way up the wellbore and/or during handling and storage. Petrographers use solid bitumen to infer its void filling occurrence without reference to solubility. Very viscous, some of them can be dissolved in organic solvents; not exactly the same as “pre-oil solid bitumen” defined in Curiale (1986) but can be equivalent to pre-oil solid bitumen. Sometimes, solid bitumen is called residual or retained OM. The solid bitumen is not oil and gas indicators but can be indicative of the migration path. Pyrobitumen: secondary product from oil; exhibits a wide spectrum of composition ranging from polar compounds to nearly graphite (Curiale, 1986). May develop in reservoir rocks by thermal chemical alteration (TCA) or thermochemical sulfate reduction (TSR; Machel et al., 1995). TCA includes processes such as biodegradation, asphaltene precipitation, and thermal alteration (Kelemen et al., 2008). Pyrobitumen is also called “reservoir solid bitumen” and can result from reservoir alteration processes of once-liquid petroleum (Curiale, 1986). Pyrobitumen consists of insoluble, nonvolatile, solid hydrocarbon residues that “still retain some hydrocarbon generation capacity upon further heating” and can host pores (Bohacs et al., 2013; Loucks and Reed, 2014). This term can be equivalent to the term “post-oil bitumen” used by Curiale (1986). Bernard et al. (2012b) suggested that pyrobitumen hosts nanometersized spongy OM pores. Char: the “ultimate residue of HC generation with minimal hydrogen content and essentially no remaining potential for generating hydrocarbons,” derived from further heating of pyrobitumen and bitumen (Bohacs et al., 2013). Char can also host nanometersized spongy OM pores.

Bright-luminescing angular quartz (blue CL)

Bright-luminescing angular quartz (red CL)

Radiolarian Dull luminescence Fig. 3. SEM-CL image showing quartz in the Woodford chert sample. Interpreted radiolarian (yellow dashed line) composed of angular quartz silt particles cemented by dull luminescing authigenic quartz. Next to the radiolarian, the angular bright-luminescing angular quartz (blue to violet CL) is interpreted to represent quartz phenocrysts from volcanic or high-grade metamorphic rocks. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

5.1.2. Woodford chert and mudstone The Woodford chert and siliceous mudstone facies are both laminated (Fig. 2). The chert facies contains 93% quartz and < 5% clay minerals (illite), whereas the mudstone facies contains 50% quartz and 35% clay minerals based on XRD analysis (Table 1). Some of the chert was possibly recrystallized from dissolved radiolarian tests and sponge spicules that were originally composed of opal. Thin-section petrography shows a few intact recrystallized radiolarian microfossils and suggests that the diagenesis could have occurred very early. Recrystallized biogenic silica and subhedral to euhedral microscrystalline quartz compose the chert; however, the sizes of these crystallites are non-uniform in SEM. The silt-size quartz crystals in the Woodford samples were investigated using cathodoluminescence (CL) microscopy techniques (Milliken and Laubach, 2000; Milliken, 2013). Cathodoluminescence patterns show that the chert is dominated by authigenic quartz, indicated by dull luminescence, (Fig. 3). Cathodoluminescence analysis of the Woodford mudstone facies shows quartz in both detrital and authigenic forms (Fig. 4). Many brightly luminescent detrital grains have dull luminescent authigenic rims (Fig. 4c, d), implying that detrital quartz with authigenic microcrystalline quartz overgrowth is common. Most silt-sized grains are quartz; the rest are K-feldspar and dolomite. These silt grains are angular to subrounded and of detrital origin. Clay-sized minerals include illite, kaolinite, quartz, pyrite, and dolomite.

5. Results 5.1. Sample lithofacies, mineralogy, texture, and fabric 5.1.1. Barnett mudstone The Barnett sample is a siliceous mudstone, the most common lithofacies in the Barnett strata. These rocks are laminated and composed of 40% clay minerals (illite and mica, interlayered illite and smectite [I/ S], chlorite, and kaolinite) and 40% quartz (Table 1; Fig. 2). Thin-section petrography reveals that both detrital and biogenic quartz (sponge spicules) contribute to the total quartz content. Feldspar, carbonate, phosphate, and pyrite are minor components (Table 1). This facies is rich in OM, with an average total organic content (TOC) value of 4.5 wt % (range of 1.6 to 10.6 wt%) (Table 2; Redmond, 2016). Phosphate (apatite) grains form nodules, peloids, or are dispersed and admixed in the matrix with OM and other minerals (Fig. 2). Authigenic minerals include pyrite, dolomite, and phosphate. Agglutinated foraminifera were observed.

5.2. Macerals, TOC, and thermal maturity

Table 2 Rock-Eval, Leco TOC analyses, and solid bitumen reflectance measurement of immature Barnett and Woodford mudrocks. Immature or lowmaturity samples

Woodford chert

Barnett siliceous mudstone

Woodford siliceous mudstone

% carbonate Leco TOC (wt%) Rock-Eval S1 (mg/g TOC) Rock-Eval S2 (mg/g TOC) Rock-Eval Tmax (°C) Calculated Ro (%) Hydrogen Index (HI) Oxygen Index (OI) Solid bitumen reflectance (%)

3.33 6.22 0.89

11.43 9.70 1.14

4.25 16.7 1.93

38.04

53.72

89.74

432 0.60 612 6 0.40

421 0.42 554 15 0.43

425 0.49 537 4 0.38

5.2.1. Barnett mudstone The Barnett sample has a TOC content of 9.7 wt% (Table 2). Solid bitumen reflectance is 0.43%Ro (30 measurements, standard deviation = 0.06), and the Rock-Eval Tmax is approximately 420 °C (Table 2), indicating the Barnett sample is thermally immature. The Barnett siliceous mudstone sample contains scattered fluorescent telalginite particles (Fig. 5a, b). Telalginite is a minor part of the overall OM component which is dominated by faintly fluorescent amorphous organic matter (AOM), that is commonly oriented parallel to bedding (Fig. 5a, b). The AOM is sometimes identified as matrix bituminite (or amorphinite), a catch-all term for organic petrographers (e.g., Fishman et al., 2012), although it is generally understood that the terms AOM and matrix bituminite are used interchangeably to describe structureless kerogen (e.g., Hackley and SanFilipo, 2016; Hackley et al., 2016). Matrix bituminite (or AOM) in the Barnett sample ranges from 6

International Journal of Coal Geology xxx (xxxx) xxx–xxx

L.T. Ko et al.

A

B

C

D

Fig. 4. SEM-EDS and SEM-CL images showing varieties of quartz silt. (A, B) X-ray EDS maps of elements (Al, Si, Ca, K, Mg) displaying distribution of quartz (red), dolomite (purple), feldspar (yellow), and clay minerals (green) in the Woodford siliceous mudstone facies. Bedding is horizontal. (C, D) Quartz in multichannel electron cathodoluminescence (SE/CL) image. Brightly luminescent detrital grains (red and blue CL) have dull luminescent rims (white arrows), representing quartz of detrital and authigenic origins, respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Because reflectance and fluorescence are not observable with SEM petrography, maceral type identification is challenging. After comparing observations of OM size, abundance, morphology, and mineral associations in the Barnett mudstone from SEM and optical microscopy, we interpret pure OM (SEM organic matter category 1) to be telalginite macerals, OM–mineral admixtures and stringy/flaky OM (SEM categories 2 and 3, respectively) to be AOM or matrix bituminite, and particulate OM (SEM category 4) to be inertinite macerals. Although both OM–mineral admixtures and stringy/flaky OM represent AOM (or matrix bituminite), OM–mineral admixtures (SEM category 2) demonstrate a greater admixture of mineral matter than does the stringy/flaky OM (SEM category 3).

discrete lamellae with bright fluorescence and a relatively homogeneous gray reflecting surface to more-scattered accumulations with less fluorescence (or no fluorescence) and a greater admixture of mineral grains (Fig. 5). Additionally, scattered inertinite and vitrinite are present in the Barnett sample. The wispy network of gray-colored amorphous OM grades to adjacent gray reflecting solid bitumen (Fig. 5d). Our visual kerogen assessment is similar to the observation made by Jarvie et al. (2007) who also indicated that Barnett Shale has 95–100% AOM with occasional algal Tasmanites (telalginite) and minor terrigenous OM (vitrinite and inertinite). Using SEM petrography, four general forms of kerogen can be identified in the immature Barnett sample (Fig. 6): (1) Pure OM does not contain any minerals, and its occurrence is sparse. OM grains are in general oriented parallel to bedding and compacted. Pure OM particles range from 20 to 30 μm in length. (2) OM–mineral admixtures, also known as organo-mineral aggregates (Aplin and Macquaker, 2011), are the most abundant type of OM in the Barnett sample (Fig. 6). Admixtures—which consist of OM, minerals, and clay-mineral aggregates—are parallel to bedding. Admixture of minerals can be significant. (3) Stringy/flaky OM is the second most-abundant type of organic matter. OM particles are elongate and lamellar and have fairly well-defined boundaries. They are also parallel or subparallel to bedding but do not encase as much mineral and clay mineral as OM–mineral admixtures. Pyrite is the most common mineral associated with this type of OM. Stringy/flaky OM is roughly 50 to 200 μm long and 10 to 20 μm wide after compaction. (4) Particulate OM pieces with discrete shape and well-defined boundaries are randomly dispersed in the matrix. These seldom-compacted OM fragments are usually uncommon (< 5 vol% of the total OM) based on visual estimates, and their shape and size vary. They are interpreted to be part of plants, marine algae, marine phytoplankton, or the altered/oxidized remains of these organisms.

5.2.2. Woodford chert and siliceous mudstone The Woodford siliceous mudstone sample (TOC = 16.7 wt%) is richer in OM than the Woodford chert sample (TOC = 6.2 wt%) (Table 2). The Woodford chert contains solid bitumen with reflectance of 0.40%Ro (30 measurements, standard deviation = 0.07), and the Rock-Eval Tmax is approximately 432 °C (Table 2). The Woodford siliceous mudstone sample contains solid bitumen reflectance of 0.38%Ro (30 measurements, standard deviation = 0.07), and the Rock-Eval Tmax temperature is approximately 425 °C (Table 2). Both solid bitumen reflectance and Rock-Eval data show that the Woodford chert and siliceous mudstone facies are thermally immature. The dominant organic matter in both Woodford siliceous mudstone and chert is amorphous (AOM) and telalginite (in the form of Tasmanites microfossils) (Fig. 2). Tasmanites is a unicellular green algal cyst that generally is observed as elliptical discs compressed along the bedding plane. Some cysts are infilled by recrystallized quartz, calcite, or pyrite before compaction preserving the original spherical shape (Fishman et al., 2012; Hackley and Kus, 2015; Hart et al., 2013; Loucks and Ruppel, 2007; Schieber, 1996; Slatt and O'Brien, 2011). The 7

International Journal of Coal Geology xxx (xxxx) xxx–xxx

L.T. Ko et al.

B

A

Telalginite

AOM Phosphate peloids

Fig. 5. Optical photomicrographs under oil immersion showing maceral types identified from polished thin sections and pellets of Barnett siliceous mudstone. (A) Scattered fluorescent telalginite and faintly fluorescent AOM (a.k.a. matrix bituminite or amorphinite) were identified under epifluorescence. Matrix bituminite is the dominant organic component in Barnett mudstone. (B) Telalginite and AOM in transmitted light. (C) Matrix bituminite and inertinite identified in incident white light. (D) The wispy network of gray-colored AOM (matrix bituminite) grades to solid bitumen under incident white light.

Telalginite

AOM

D

C

Inertinite

Inertinite

Inertinite

Matrix bituminite Solid bitumen AOM

mudstones. (3) Particulate and discrete OM has well-defined boundaries, and discrete shapes, and resists compaction. This OM is randomly dispersed in the matrix and usually low in abundance (< 5 vol% of the OM) (Fig. 8a). Tasmanites OM is easily identified in both light and SEM microscopy. Based on size, abundance, morphology, and mineral associations, we interpret stringy, dispersed OM observed by SEM to be AOM (or matrix bituminite), and particulate and discrete OM to be inertinite or vitrinite.

Woodford siliceous mudstone sample contains abundant thick-walled Tasmanites and Leiosphaeridia (unknown spherical alga) that are flattened into the bedding planes (Fig. 7a). Maturity is low as suggested by the high-intensity green fluorescence (short wavelength) response (Bertrand et al., 1986; Pradier et al., 1991). Some of the Tasmanites have been partially bituminized as revealed by a decrease in fluorescence intensity, a shift to a more reddish color in fluorescence, and the development of a gray reflecting surface in reflected light (Fig. 7a, b). The matrix contains abundant dispersed fluorescent AOM (bituminite) (Fig. 7a, b). Some scattered inertinite is present and is very fine grained (Fig. 7b). The Woodford chert facies also contains abundant Tasmanites and Leiosphaeridia (Fig. 7c). Other petrographers also have identified telalginite in the Woodford as Foerstia (Cardott and Chaplin, 1993). Variations in telalginite morphology may represent different life-cycle phases of one species—i.e., active vs. resting (spore) phases—rather than different species (Telnova, 2012). Qualitatively, the OM is more oxidized in the chert than in the mudstone sample, as is evident through secondary mineralization (oxides) (Fig. 7d). Based on SEM petrography, the Woodford sample contains three forms of kerogen: (1) Telalginite (Tasmanites and/or Leiosphaeridia), the most abundant type of maceral, is characterized by its thin-to-thick wall, in some cases with discrete shape or encasing pyrite or quartz grains (Fig. 8a, b). The size of Tasmanites ranges from approximately 50 to 200 μm. (2) Stringy and dispersed OM has no well-defined boundaries and is generally admixed with minerals in the matrix (Fig. 8b). The abundance of minerals is less than in the OM–mineral admixtures and stringy/flaky OM categories observed by SEM in the Barnett

5.3. Geochemical properties and stages of petroleum generation 5.3.1. Barnett and Woodford mudstones Barnett and Woodford mudstones show distinct geochemical properties in terms of TOC; Rock-Eval S1, S2, and Tmax, hydrogen index (HI), and oxygen index (OI) values (Table 2). The Woodford siliceous mudstone sample contains the highest amount of TOC (16.7 wt%). As discussed above, both Barnett and Woodford mudstone samples are immature with Tmax values of approximately 430 °C (~ 0.6%Ro) (Table 2). Lewan (1983, 1987) also recognized the immature character of the Woodford Shale from the same outcrop location. The Barnett and Woodford organic-rich siliceous mudstone samples show higher RockEval S1 and S2 but lower Tmax and HI values when compared to lessorganic-rich Woodford chert (Table 2). During the gold-tube experiments, all samples showed decreases in the amount of TOC, S2, and HI values and increases in Tmax values (Fig. 9; Tables 3–5). Rock-Eval and TOC data from the Boquillas (Eagle Ford-equivalent) organic-lean mudstone (TOC = 1.9 wt%) are also plotted with the Barnett and 8

International Journal of Coal Geology xxx (xxxx) xxx–xxx

L.T. Ko et al.

(resins and asphaltenes) generation stage, (2) early oil generation stage, (3) mid-oil window and associated gas generation stage, (4) end of oil window generation stage, and (5) early oil cracking and associated wet gas generation stage. In the Barnett and Woodford siliceous mudstone samples, gradual increases in EOM from 130 °C to 333 °C experiments indicate that the Barnett and Woodford organic matter transformed from the kerogen to bitumen. Comparison of EOM and experimental temperature shows that above 333 °C, EOM (SARA components) gradually decreased with increasing temperature (Fig. 11). The amount of light oil (C7–C14) gradually increased to a maximum at 400 °C at the end of oil window, which corresponds to bitumen gradually being transformed into oil. From 400 °C to 425 °C, light oil yield began to decrease and gaseous hydrocarbon yield (C1–C3) continued to increase, which corresponds to the early oil cracking to wet-gas stage (Fig. 11). The saturate and aromatic fractions of the Barnett and Woodford samples did not exhibit similar trends to saturate and aromatic components of the Boquillas (Eagle Ford) sample (see Fig. 9 in Ko et al., 2016). Those of the Woodford chert showed more and higher rate of OM conversion per gram of TOC (Fig. 11).

Stringy/flaky OMr

Pure OMr

Phosphatic peloids

Pure OMr

5.4. Pre-heating pore networks 5.4.1. Barnett mudstone Pore classification and the definition of pore type as proposed by Loucks et al. (2012) and Ko et al. (2016, 2017) are used in this study. The interpreted pore network in the immature Barnett siliceous mudstone sample is associated with minerals especially clay minerals. The majority of these mineral pores are now preserved as elongated intraparticle pores between clay-mineral platelets, commonly a few microns in length (Fig. 12a). Interparticle pores, between minerals and kerogen particles, are less abundant and relatively small in size (Fig. 12b). Other less common intraparticle pores include phosphatic peloids (Fig. 12c), intra-pyrite-framboid OM pores (Fig. 12d), primary OM pores (Fig. 12e), and convoluted OM pores (Fig. 11f). Convoluted OM pores are related to the deformation of OM after compaction, lying between folded, twisted, or coiled kerogen that is clearly particulate (Ko et al., 2016).

OM mineral admixture

Discrete, particulate OMr

5.4.2. Woodford chert and siliceous mudstone The pre-heating pore network in the immature Woodford chert is predominantly composed of interparticle pores between quartz grains, between quartz and clay minerals (mainly illite), and intra-clay-platelet (intraparticle) pores. The pre-heating pore network in the immature Woodford siliceous mudstone sample is predominantly associated with intraparticle pores, especially from grain dissolution and clay-mineral platelets (Fig. 13a). Many intraparticle pores appear to be related to dolomite-rim dissolution (Fig. 13b). Other pore types include primary (inherited) OM pores in kerogen (Fig. 13c).

Fig. 6. SEM image of the Barnett siliceous mudstone sample. Ar-ion milled surface showing lamination, phosphatic peloids, and four distinct types of OM: (1) pure OM, (2) stringy/flaky OM, (3) OM–mineral admixture, and (4) discrete and particulate OM.

Woodford for comparison (Fig. 10). Stages and timing of OM conversion and petroleum generation in Barnett and Woodford mudstone samples were identified by the yield of generated gas (C1–C5), light hydrocarbons (C7–C14), hydrocarbons (C14 +), EOM, and SARA fractions with increasing experimental temperatures (Fig. 11; Tables 6–8). In general, the selected experimental temperatures simulate the generation of bitumen, early oil, middle of the oil window, and the very beginning of wet-gas generation (Fig. 11). However, between Barnett and Woodford mudstone samples, there are differences in the amount of generated oil, gas, and SARA fractions, as well as differences in the rate of conversion. The Barnett siliceous mudstone sample generated slightly more methane to pentane compared to the Woodford chert and siliceous mudstone samples from the middle oil window to early oil cracking to wet gas stage. The Woodford chert sample showed the greatest petroleum yield (EOM) compared to the other samples at the beginning of the oil window, especially the ARO and NSO (polar) compounds (Fig. 11). We suspect that there might have been some early migrated bitumen in the chert facies, which was not related to the thermal decomposition of original kerogen in the chert. Five oil and gas generation stages were defined based on yields of gas, light oil, EOM, and SARA fractions (Fig. 11): (1) bitumen or polar

5.5. Pore and pore-network evolution As with the previously studied Eagle Ford Group calcareous mudstone sample (Ko et al., 2016), Barnett and Woodford pores and pore networks also evolved with OM conversion during thermal maturation. Migration of pre-oil solid bitumen into the original mineral pore network at a very small scale was observed as a result of the kerogen being transformed into petroleum. The migration of pre-oil solid bitumen into the original mineral-pore network in the Barnett and Woodford samples implies that mineral pores were at least partly interconnected. 5.5.1. Barnett mudstone At the early bitumen generation stage (after 300 °C/72 h of heating), bubble-shaped OM-hosted pores were common in OM–mineral admixtures (AOM or matrix bituminite) in the matrix (Fig. 14a, b). Some shrinkage of kerogen was developed between interfaces of OM and minerals (Fig. 14a). We believe the shrinkage 9

International Journal of Coal Geology xxx (xxxx) xxx–xxx

L.T. Ko et al.

A

Fig. 7. Optical photomicrographs under oil immersion showing types of macerals identified from polished thin sections and pellets of Woodford siliceous mudstone and chert samples. (A) Woodford siliceous mudstone: thickwalled Tasmanites and Leiosphaeridia (unknown spherical alga) flattened in the bedding planes. The matrix contains abundant dispersed fluorescent AOM. (B) Woodford siliceous mudstone: AOM and inertinite identified in incident white light; same field as (A). (C) Woodford chert: abundant Tasmanites and Leiosphaeridia (unknown spherical alga). High-intensity green fluorescence response suggests low thermal maturity. (D) Woodford chert: mineralized Tasmanites, AOM, and inertinite identified in incident white light; same field as (C).

B Leiosphaeridia

AOM Fluorescent AOM

Inertinite

Telalginite (Tasmanites)

C

Leiosphaeridia

D

Inertinite Mineralized telalginite (Tasmanites)

Fluorescent AOM (amorphinite)

AOM

of the OM bubble pores increased in AOM, possibly related to the greater amount of petroleum generated (Figs. 11 and 14c). Most original interparticle and intraparticle mineral pores disappeared and became filled with petroleum (bitumen) at 310 °C/72 h of heating

pores—unlikely to have been developed in the subsurface—were formed when the sample was quickly quenched as the experimental temperature was reduced to room temperature. At the peak bitumen generation stage (after 310 °C/72 h of heating), the abundance and size

A

Particulate OM

B

Telalginite (Tasmanites)

Telalginite (Tasmanites)

Telalginite Stringy, dispersed OM in matrix

10

Fig. 8. SEM images of immature Woodford siliceous mudstone. Ar-ion milled surface showing lamination, phosphatic peloids, and three distinct types of kerogen (1) Tasmanites microfossil remains. (2) Stringy, dispersed OM. (3) Particulate and discrete OM. Q: quartz; Py: pyrite. (A) Compressed Tasmanites microfossil with quartz encased within and adjacent particulate, discrete OM. (B) Compressed Tasmanites microfossil and stringy OM dispersed in the matrix. Some Tasmanites are associated with pyrite framboids and some have quartz encased within. Matrix is composed of clay minerals, clay-sized quartz, and stringy, dispersed OM.

International Journal of Coal Geology xxx (xxxx) xxx–xxx

L.T. Ko et al.

A

C

B

Fig. 9. Plots of HI vs OI of the Barnett and Woodford samples during pyrolysis experiments. (A) Barnett siliceous mudstone samples (B) Woodford siliceous mudstone samples (C) Woodford chert samples.

Table 3 Geochemical data from Barnett siliceous mudstone facies samples from laboratory pyrolysis experiments. Sample ID

Heating temperature (°C)

Leco TOC (wt%)

Rock-Eval-2 S1 (mg HC/g)

Rock-Eval-2 S2 (mg HC/g)

Rock-Eval-2 Tmax (°C)

Hydrogen Index (mg/g)

Oxygen Index (mg/g)

BS-1 BS-10 BS-3 BS-6 BS-8 BS-12 BS-14 BS-2

Outcrop 130 300 310 333 367 400 425

9.70 10.60 10.80 10.90 9.09 7.23 7.48 7.76

1.14 0.48 6.94 10.39 11.69 0.09 0.20 4.63

53.72 56.93 47.18 42.55 20.37 2.54 1.21 1.49

421 420 428 431 437 456 549 340

554 537 437 390 224 35 16 19

15 13 9 9 11 4 4 5

Table 4 Geochemical data from Woodford siliceous mudstone facies samples from laboratory pyrolysis experiments. Sample ID

Heating temperature (°C)

Leco TOC (wt%)

Rock-Eval-2 S1 (mg HC/g)

Rock-Eval-2 S2 (mg HC/g)

Rock-Eval-2 Tmax (°C)

Hydrogen index (mg/g)

Oxygen index (mg/g)

WM-1 WM-10 WM-7 WM-2 WM-4 WM-12 WM-14 WM-6

Outcrop 130 300 310 333 367 400 425

16.70 15.40 13.60 13.20 12.30 10.30 10.60 9.99

1.93 2.04 9.27 10.89 20.52 25.50 12.49 8.48

89.74 90.75 69.16 60.11 38.15 12.84 5.80 2.32

425 420 431 435 428 392 391 387

537 589 509 455 310 125 55 23

4 3 3 5 3 3 3 3

Table 5 Geochemical data from Woodford chert facies samples from laboratory pyrolysis experiments. Sample ID

Experimental temperature stage (°C)

Leco TOC (wt%)

Rock-Eval-2 S1 (mg HC/g)

Rock-Eval-2 S2 (mg HC/g)

Rock-Eval-2 Tmax (°C)

Hydrogen index (mg/g)

Oxygen index (mg/g)

WF-1 WF-10 WF-4 WF-6 WF-7 WF-12 WF-13 WF-15

Outcrop 130 300 310 333 367 400 425

6.22 3.99 3.40 3.26 3.73 2.41 3.60 2.50

0.89 0.50 0.24 0.11 7.68 0.13 0.22 0.39

38.04 26.55 19.27 17.02 13.37 1.03 0.44 0.35

432 426 430 435 433 447 445 434

612 665 567 522 358 43 12 14

6 3 4 3 3 4 3 3

11

International Journal of Coal Geology xxx (xxxx) xxx–xxx

L.T. Ko et al.

Fig. 10. Plot of Rock-Eval S2 versus TOC of the Barnett, Boquillas, and Woodford mudrock samples showing original types of OM (Type II marine source) and decreases in S2 and TOC with increasing experimental temperature. Dash lines and arrows represent samples from immature to mature during gold tube pyrolysis (Ko et al., 2016).

immature

immature

immature

immature

mature

mature

mature

mature

C7-14 C1-3 EOM C4-5 (mg/g TOC) (mg/g TOC) (mg/g TOC) (mg/g TOC) 0.0 150.0 300.0 0.0 25.0 50.0 0.0 35.0 70.0 0.0 400.0 800.0

Petroleum Generation Stages

Sat, Arom (mg/g TOC) 0.0

120.0

240.0 0.0

Asph, NSO (mg/g TOC) 175.0

350.0

130

Immature

Temperature (oC)

290 310

Bitumen Generation

330

Early Oil

350

Oil Window

370 390

Early Oil Cracking to Wet Gas

410 430 450

Fig. 11. Plots of experimental temperatures (130, 300, 310, 333, 367, 400, 425 °C) and generated hydrocarbon components (gaseous and liquid petroleum). Five oil- and gas-generation stages are defined by trends from produced gas (C1–C3 and C4–C5), liquids (C7–C14, C14+), EOM (bitumen), and SARA fractions. Black (Barnett siliceous mudstone), blue (Woodford siliceous mudstone), and red (Woodford chert) indicate phase abundance at different temperatures. Sat: saturate (open squares); Arom: aromatic (solid squares); NSO: nitrogen, sulfur, oxygen, and heavy metals (solid squares); Asph: asphaltene (open squares). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

and sample preparation processes under vacuum conditions, either generated petroleum had migrated or the originally retained petroleum had escaped from the pore spaces. Some modified mineral pores are characterized by isopachous OM rims (Fig. 16c; Ko et al., 2016). Their sizes and shapes are determined by surrounding mineral framework grains. The modified mineral pores with isopachous OM rim were commonly observed during oil window. The particulate OM showed no evidence of pore development. At the middle oil window stage (after 367 °C/72 h of heating), the abundance of modified mineral pores significantly increased because of continued OM conversion (Fig. 17c). Spongy OM-hosted pores started to appear in both OM–mineral admixtures and pure OM patches (likely post-oil solid bitumen) (Fig. 17a, d). The particulate OM still showed no evidence of pore development (Fig. 17b). At the stage of early oil cracking and into the wet gas (after 400 and 425 °C/72 h of heating), the dominant pore types, modified mineral pores and OM spongy pores, were similar to those observed in the middle oil window although their abundances and volume are greater (Fig. 18a, b). Little change in

(Fig. 14c). Observed pre-oil solid bitumen migration into the phosphatic peloids implies that the latter have some interconnected pore network and act like a conduit for fluid movement (Fig. 15). No pores were found in the particulate OM under SEM (inertinite maceral). At the beginning of the oil window (after 333 °C/72 h of heating), the pure OM (telalginite) went through a thorough conversion (Fig. 16a). The continuous conversion of relatively viscous bitumen (rich in resin and asphaltene) to less viscous oil (rich in saturate and aromatic components) caused a significant volume loss, resulting in the formation of moldic spaces, fluid migration, and post-oil solid bitumen (or pyrobitumen). The predominant pore types and pore network were composed of modified mineral pores containing relic OM (Fig. 16). Modified mineral pores were resulted from generated, expelled, and migrated petroleum that pervades the connected primary mineral pore network (Ko et al., 2016). The relic OM (residual OM) can be solid bitumen as a result of compositional fractionation during thermal cracking and migration. The modified mineral pores should be filled with petroleum in the subsurface. However, during post experimental 12

International Journal of Coal Geology xxx (xxxx) xxx–xxx

L.T. Ko et al.

Table 6 Yields of generated gas and oil components from Barnett siliceous mudrocks after each experimental run. Heating Temperature (°C [°F])

C1–C3 (mg/g TOC)

C4–C5 (mg/g TOC)

C7–C14 (mg/g TOC)

EOM (mg/g TOC)

SAT (mg/g TOC)

ARO (mg/ g TOC)

NSO (mg/g TOC)

Asphaltene (mg/g TOC)

CO2 (mg/g TOC)

H2S (mg/g TOC)

130 300 310 333 367 400 425

0.01 6.69 9.84 24.24 91.17 175.79 244.07

0.00 1.22 1.44 6.34 24.83 39.20 39.16

0.84 14.01 22.23 34.60 50.60 52.17 40.65

34.67 217.03 314.37 341.83 197.85 62.28 40.10

3.80 11.01 22.38 30.60 20.61 28.84 6.07

9.27 49.75 73.75 95.20 70.28 28.03 22.03

12.01 62.08 82.29 92.96 39.46 16.77 11.63

3.84 78.92 114.29 106.56 61.32 13.97 3.64

0.43 94.22 66.49 68.48 146.91 186.37 227.25

0.00 2.17 0.45 3.02 16.09 29.40 51.99

(266) (572) (590) (631.4) (692.6) (752) (797)

Abbreviations: ARO = aromatic; C1–C3 = methane, ethane, and propane; C4–C5 = butane and pentane; C7–C14 = light oil; EOM = extractable organic matter; NSO = nitrogen, sulfur, oxygen, and heavy metals; SAT = saturate. Table 7 Yields of generated gas and oil components from Woodford siliceous mudrocks after each experimental run. Heating Temperature (°C [°F])

C1–C3 (mg/g TOC)

C4–C5 (mg/g TOC)

C7–C14 (mg/g TOC)

EOM (mg/g TOC)

SAT (mg/g TOC)

ARO (mg/g TOC)

NSO (mg/g TOC)

Asphaltene (mg/g TOC)

CO2 (mg/g TOC)

H2S (mg/g TOC)

130 300 310 333 367 400 425

0.01 5.47 10.81 33.55 84.06 156.35 201.20

0.00 0.43 1.59 9.45 20.32 33.37 30.92

0.21 11.16 18.83 26.12 37.72 38.75 15.78

14.95 176.47 288.89 408.09 252.82 93.12 54.93

N/A 15.14 14.18 24.79 24.66 5.57 6.91

N/A 51.29 75.32 122.60 94.89 35.41 28.04

N/A 40.77 66.14 75.32 41.84 11.35 12.63

N/A 63.41 123.79 307.97 83.68 39.36 5.47

0.17 25.05 32.24 30.32 69.13 67.01 90.96

0.00 0.00 0.00 0.01 0.02 0.06 0.09

(266) (572) (590) (631.4) (692.6) (752) (797)

Abbreviations: ARO = aromatic; C1–C3 = methane, ethane, and propane; C4–C5 = butane and pentane; C7–C14 = light oil; EOM = extractable organic matter; NSO = nitrogen, sulfur, oxygen, and heavy metals; SAT = saturate. Table 8 Yields of generated gas and oil components from Woodford chert after each experimental run. Heating temperature (°C [°F])

C1–C3 (mg/g TOC)

C4–C5 (mg/g TOC)

C7–C14 (mg/g TOC)

EOM (mg/g TOC)

SAT (mg/g TOC)

ARO (mg/g TOC)

NSO (mg/g TOC)

Asphaltene (mg/g TOC)

CO2 (mg/g TOC)

H2S (mg/g TOC)

130 300 310 333 367 400 425

0.01 1.25 1.75 36.81 66.30 111.28 195.06

0.00 0.07 0.07 12.73 16.34 32.03 31.97

1.56 5.20 8.07 39.06 45.88 62.81 45.74

20.51 124.59 240.89 709.92 219.95 66.20 42.21

N/A 9.37 17.00 50.50 21.91 4.58 7.02

N/A 35.30 67.16 210.02 70.61 33.49 21.39

N/A 31.67 53.48 123.99 30.98 13.74 9.93

N/A 40.89 94.00 271.67 51.12 14.10 5.07

1.40 6.50 6.00 29.60 44.20 53.60 69.80

0.00 0.00 0.00 0.00 0.00 0.10 0.20

(266) (572) (590) (631.4) (692.6) (752) (797)

Abbreviations: ARO = aromatic; C1–C3 = methane, ethane, and propane; C4–C5 = butane and pentane; C7–C14 = light oil; EOM = extractable organic matter; NSO = nitrogen, sulfur, oxygen, and heavy metals; SAT = saturate.

were not as abundant as those observed at this stage in the Barnett sample (Fig. 20a). Simultaneously, migration of petroleum (pre-oil solid bitumen) into the original intraparticle mineral pores associated with dissolved dolomite rims was observed at a very small scale. The sheltered pores between angular, stacking quartz silts and dissolution-rim pores in dolomite crystals were primary interparticle and intraparticle mineral pores (with no OM present) (Fig. 13a, b). After generation, expulsion, and migration of bitumen, many dissolution-rim pores were filled or partially filled with OM (likely pre-oil solid bitumen) which likely migrated from adjacent kerogen. Intriguingly, no pores were developed in Tasmanites OM (unicellular green algae) (Fig. 20b, d). Unchanged Tasmanites OM implies that even though, based on RockEval analysis, bulk OM in the Woodford is mainly marine Type II as in the Barnett, telalginite macerals (Tasmanites) in the two mudstones have different geochemical characteristics and transformation rates and ratios compared to the AOM (matrix bituminite) during thermal maturation. Importantly, the OM transformation ratio of the telalginite macerals can impact pore development and amount of petroleum generated significantly. At the end of bitumen and early oil generation (after 333 °C/72 h of heating), Tasmanites OM has high conversion to petroleum, leaving a large moldic space ranging from tens to hundreds

dominant pore types above 400 °C/72 h of heating is supported by geochemical analyses (Fig. 10), which showed no significant change in generated gas or liquid. Little or no pore development was observed in any particulate OM (inertinite maceral) under SEM throughout the experiment series. This implies that inertinite macerals (terrigenous kerogen) do not convert to petroleum, changing little or not at all during maturation into the wet gas window; therefore, they do not contribute much to total pore volume in the mudstone. Our experiments indicate that kerogen and maceral particles show variable pore development, resulting in inhomogeneity of pore types caused by differences in chemical composition from maceral type to maceral type (primary) which subsequently result in differences in specific gravity and generation kinetics (secondary) of maceral particles during thermal maturation. 5.5.2. Woodford mudstone As with the Barnett mudstone, at the early bitumen generation stage (after 300 °C/72 h of heating), shrinkage was observed especially between Tasmanites wall and surrounding mineral grains (Fig. 19). During the stage of main bitumen generation (310 °C/72 h), a few OM bubble pores were formed in stringy, dispersed OM; however, OM bubble pores 13

International Journal of Coal Geology xxx (xxxx) xxx–xxx

L.T. Ko et al.

Fig. 12. SEM images showing types of pores (arrows) in the immature Barnett mudstone. (A) intra-clay-platelet pore; (B) interparticle (interP) pore between OM and mineral grains; (C) intra-apatite (intraP) pore; (D) intraframboid pore; (E) primary OM pore; (F) convoluted OM pore. Q: quartz; clay: clay minerals.

Mica

Kerogen Phosphate peloids

IntraP pores

modified mineral pores had increased slightly. A significant proportion of petroleum was still retained in the modified mineral pores (Fig. 22a, c). It is not until the cracking of oil to the wet-gas stage (after 425 °C/ 72 h of heating) that most fluids were entirely expelled, resulting in a significant increase in the abundance and size of modified mineral pores (Fig. 23). It is worth noting that OM spongy pores were uncommon and difficult to observe in the Woodford sample at the end of the oil window and at the cracking of early oil to the wet-gas stage.

of micrometers (Fig. 21). The dominant pore types in the Woodford siliceous mudstone are the moldic pores and modified mineral pores with relic OM (Fig. 21). However, some petroleum is still present in matrix pores (Fig. 21a). It could be that the petroleum that is still trapped within the mineral pores was too viscous to escape even under experimental vacuum conditions. Or it could be that the Woodford siliceous mudstone is too tight for petroleum to migrate. At the middle oil window stage (after 367 °C/72 h of heating), the abundance of 14

International Journal of Coal Geology xxx (xxxx) xxx–xxx

L.T. Ko et al.

Fig. 13. SEM images showing types of pores (arrows) in the immature Woodford mudstone. (A) intra-clay-mineral platelet pores and some are sheltered by quartz silts; (B) intraparticle (intraP) pores from dolomite dissolution; (C) primary OM pores. Q: quartz; F: feldspar; D: dolomite; Py: pyrite.

OM bubble pores OM bubble pores OM bubble pores Modified mineral pores Shrinkage pores

OM bubble pores Fig. 14. SEM images showing shrinkage pores and development of OM bubble pores in the Barnett siliceous mudstone sample at the early and peak bitumen generation stages. (A) Shrinkage pores, which are experimental artifacts, developed at contact between elongated kerogen and mineral grains. OM bubble pores developed within dispersed OM. (B) OM bubble pores developed within stringy, dispersed OM as a result of AOM conversion. (C) Development of OM bubble pores in the AOM or bituminite at the peak bitumen generation stage. Migration of petroleum (pre-oil solid bitumen) into the surrounding primary mineral pores forming modified mineral pores with relic OM. The Barnett mudrock has abundant quartz cement, forming a rigid framework. Feldspar is slightly dissolved. F: feldspar; Q: quartz; Ph: phosphate; Py: pyrite.

6. Discussion

compounds are expected during the oil window as a result of thermal cracking of bitumen (also see Fig. 9 of Ko et al., 2016). We suspect that the continuous decrease of SAT and ARO in Barnett and Woodford mudstones samples might be related to the anhydrous status of experimental conditions because the systems did not have enough hydrogen to facilitate petroleum (SAT and ARO) formation. Without the availability of hydrogen from water to form petroleum in Barnett and Woodford mudstones, the organic matter would likely go through dominantly a cross-linking reaction pathway forming insoluble solid pyrobitumen, instead of going through free radical mechanisms (Lewan, 1997). According to Huang (1996) and Behar et al. (2003), during anhydrous pyrolysis experiments in sealed gold tubes with external pressure

6.1. Generated petroleum components Examination of EOM as a function of experimental temperature shows that above 333 °C, EOM gradually decreased with higher temperature (Fig. 11), implying that the 333 °C/72 h run corresponds to the onset of oil generation. However, saturate (SAT) and aromatic (ARO) components continued to decrease with increasing temperature during maturation through the oil window, similar to the trends of resin and asphaltene components. Oil is rich in SAT and ARO compounds, whereas bitumen, solid bitumen, and pyrobitumen are rich in resins and asphaltene. Therefore, the increases in both SAT and ARO 15

International Journal of Coal Geology xxx (xxxx) xxx–xxx

L.T. Ko et al.

(confined environment), water could be generated during kerogen conversion and occurred in a liquid state in the gold tubes, making the system “hydrous.” We did not observe any evidence of water droplets on the wall of the gold tubes so we could not be certain if our simulation was hydrous. However, because samples were not oven-dry prior to pyrolysis, the presence of irreducible water cannot be ruled out. It is likely that our experiments were hydrous to some extent and then became anhydrous during the middle of the pyrolysis experiment.

Generated pre-oil solid bitumen forming migration network in the phosphatic peloid

6.2. Effect of maceral types on pore evolution Many petrographic and geochemical schemes such as morphology, texture, maceral and palynomorph composition, and bulk geochemical composition can be used to classify OM type (Walters, 2007). In this study, we attempted to integrate SEM petrography with organic petrography based on observation and description of texture, mineralmixing, and morphological characters made by organic petrologists. To further differentiate OM facies, “organic matter type, its source, and depositional environment” need to be taken into account (Tyson, 1995). The kerogen-type classification (Type I, II, IIS, III, or IV) based on a pseudo–van Krevelen diagram derived from hydrogen and oxygen indices is commonly used to display bulk geochemical results. Petrographic study identifies individual components of OM (macerals) and can provide confirmation of bulk geochemical data and the linkage between geochemical and sedimentological data. Using organic petrography can also distinguish the assessment of OM source, mixing, preservation, and maturation. This study demonstrates the lesscommon application of organic petrography to the study of maceral and

Fig. 15. SEM image showing original intra-particle pore networks in the phosphatic peloid in the Barnett sample as possible migration pathways (black lines) for generated petroleum (pre-oil solid bitumen) at the peak bitumen generation stage. Ph: phosphate.

Moldic pore

Modified mineral pores

Modified mineral pores

16

Fig. 16. SEM images showing modified mineral pores as dominant pore types in the Barnett sample at the beginning of the oil window. (A) Almost full conversion of telalginite forming a moldic pore. OM-mineral admixture develops few modified mineral pores (white arrows). (B) Macerals with different conversion capability. Particulate macerals (black arrows: possibly inertinite) have little to no conversion capability. OM-mineral admixture has converted to petroleum and formed modified mineral pores. (C) Modified mineral pores within dissolved feldspar and between clay-mineral platelets. (D) Stringy/flaky OM shows significant conversion (white arrows), forming pores within the original maceral. Q: quartz; F: feldspar; Ph: phosphate.

International Journal of Coal Geology xxx (xxxx) xxx–xxx

L.T. Ko et al.

Fig. 17. SEM images showing modified mineral pores (white arrows) and OM spongy pores (black arrows) as dominant pore types in the Barnett siliceous mudstone sample at the middle of the oil-window generation stage. (A) OM spongy pores in the OM–mineral admixtures. (B) Particulate OM with no visible pores. (C) Abundant modified mineral pores ranged from nanometers to micrometers in the matrix of the Barnett mudrock. (D) OM spongy pores (black arrow) in pure OM. Q: quartz; Clay: clay minerals.

OM spongy pores

Modified mineral pores

Fig. 18. SEM images showing modified mineral pores (black arrows) and OM spongy pores (white arrows) in the Barnett samples as dominant pore types in the late oil window and the early wet-gas window. The fact that dominant pore types do not change and are similar to pores in the middle oil window (Fig. 17) is confirmed by the geochemical analyses in Fig. 11, which show that during these two stages, no significant change in generated gas and liquid is observed. Clay: clay minerals.

Although they live and are commonly deposited in marine settings associated with Type II kerogen, these algae have relatively high hydrogen contents and aliphatic molecules, and are geochemically more similar to Type I kerogen (see Fig. 7 of Walters, 2007). Depending on the amount, type, and mixing of these algae in marine mudstones, it can be difficult to identify their occurrence by using only bulk geochemical analysis. Therefore, applying organic petrography to identify their

palynomorph assemblages for OM facies identification in mudstones and their associated pore evolution. Algae such as Gloeocapsomorpha (thought to be either a cyanobacterium or related to Botryococcus) (Derenne et al., 1992; Fowler et al., 2004), Botryococcus (green microalgae) (Derenne et al., 1992) and Tasmanites (Revill et al., 1994) have unusual lipid biochemistry and commonly produce organic-rich mudstones (Volkman et al., 2015). 17

International Journal of Coal Geology xxx (xxxx) xxx–xxx

L.T. Ko et al.

mudstone samples might correspond to one of the key differences, petroleum generation kinetics (Katz and Lin, 2014; Marzi and Rullkötter, 1992; Revill et al., 1994; Vigran et al., 2008) between Type I-like and Type II-contained source rocks. In the Barnett and Woodford marine mudstone cases, the differences are those of petroleum generation kinetics between lipid-rich algae (similar to Type I) and bacteria and marine algae (Type II). Activation energy distribution between algal-rich Type II kerogen and typical Type II kerogen is different, with algal-rich Type II kerogen having a much narrower distribution of activation energy than the typical Type II kerogen (Pepper and Corvi, 1995; Sundararaman et al., 1988; Tissot et al., 1987). The narrow activation energy distribution of algal-rich Type II kerogen means that the main stage of petroleum generation from algal-rich Type II kerogen is short (Katz and Lin, 2014). Another important aspect of activation energy distribution is that, in general, the mean activation energy of algal-rich Type II kerogen is higher than that of the typical Type II kerogen, requiring a higher level of thermal maturity to generate petroleum (Katz and Lin, 2014; Marzi and Rullkötter, 1992; Revill et al., 1994; Vigran et al., 2008). Revill et al. (1994) documented a very narrow activation energy distribution in the Latrobe Tasmanite kerogen in their Fig. 14 with the frequency factor of 8.9 × 1013 S− 1. Vigran et al. (2008) also reported the narrow activation energy distribution of the Tasmanites algae in Svalbard and the Barents Shelf source rocks in their Table 2 and suggest its kinetics is rather similar to that of the Eocene Green River Shale (data from Peters et al., 2006), the prototype of a Type I kerogen. Pore evolution in telalginite- and Tasmanites OM-rich Woodford samples corresponds to shorter and later OM conversion and petroleum generation of algal-rich Type II kerogen. Pores in telalginite and Tasmanites OM of the Woodford samples formed later in the oil window and developed much more quickly than in the AOM (matrix bituminite) of Barnett and Woodford mudstones. Micropores and pore canals (slit) had been previously observed on the surface of Tasmanites in the literature using transmitted electron microscopy (TEM) and confocal laser scanning microscopy (CLSM) (Hackley and Kus, 2015; Talyzina and

Fig. 19. SEM image of the Woodford siliceous mudstone sample at early bitumen generation stage. Shrinkage pores (white arrows) are artifacts between Tasmanites and minerals.

existence and occurrence is a necessary step in order to accurately predict the amount of petroleum generated and to select the appropriate pore-evolution model. Petrographic analysis, like any other method, has its limitations; for example, depending on maceral types, maceral identification can be difficult to conduct after the samples have matured through the peak oil window into the dry gas window as the maceral converts to petroleum and maceral interpretation relies on organic petrographers' experiences. The differences observed in OM pore evolution of macerals through laboratory pyrolysis simulation of the Barnett and Woodford siliceous

OM bubble pores Non-porous OM

Non-porous OM

Non-porous OM

OM shrinkage pores

18

Fig. 20. SEM images of the Woodford organic-rich siliceous mudstone sample at the bitumen generation stage. (A) Stringy, dispersed OM has gone through kerogen to bitumen conversion, forming bubble OM pores in the matrix. (B) Tasmanites are intact with no apparent conversion. No pores were observed. (C) Nonporous OM (unknown type). (D) Both particulate OM and broken pieces of Tasmanites show no apparent conversion. Some artifact shrinkage was observed. D: dolomite; F: feldspar; Py: pyrite; Q: quartz.

International Journal of Coal Geology xxx (xxxx) xxx–xxx

L.T. Ko et al.

better conversion potential than did the OM–mineral admixtures (AOM or matrix bituminite). Differences in OM conversion can be related to the different preservation state of AOM (or matrix bituminite) and the character of the mineral mixing from which it is liberated (Tyson, 1995). If the AOM (or matrix bituminite) has abundant mineral mixing or is degraded, it becomes increasingly heterogeneous and dull when viewed under fluorescence (Tyson, 1995). As a result, the conversion is not as good as that of the AOM with less mineral mixing and degradation. The conversion of inertinite, matrix bituminite, and Tasmanites in the Woodford were different from each other. As previously discussed, the transformation of Tasmanites OM to petroleum did not occur until the sample entered the early oil maturation window, where it occurred almost instantaneously, undergoing complete transformation to petroleum without the development of internal pores and leaving a moldic space. The particulate and less compacted OM (inertinite) shows no pore development as viewed in the SEM. Other techniques, such as the use of higher-resolution transmitted electron microscopy (TEM) or atomic force microscopy (AFM) instruments are required to further validate this observation.

Pores with trapped petroleum fluids

Modified mineral pores

6.3. Effect of mineralogy on pore development

Modified mineral pores

Although the Barnett and Woodford siliceous mudstones have different mineralogy than that of the Eagle Ford calcareous mudstones (Camp, 2014; Jennings and Antia, 2013; Ko et al., 2016; Pommer and Milliken, 2015; Schieber et al., 2016), we did not find that mineralogy has a significant effect on pore evolution. Even though the Barnett (25 to 40% clay minerals: illite, interlayered I/S, chlorite, and kaolinite) and Woodford (20 to 35% clay minerals: illite and kaolinite) have higher clay mineral content than that of the Eagle Ford limestone (< 5%: kaolinite and chlorite) and Eagle Ford marl (8 to 22% clay minerals: illite, interlayered I/S, chlorite, and kaolinite), the abundance and type of clay minerals do not seem to show any significant catalytic or sorption effect on generated petroleum and, thus, pore evolution. This might be related to the relatively low smectite content in these three major mudstone units. We hypothesize that there would be differences in petroleum fluid retention and pore development in mudstones with high smectite content (e.g., Hetényi, 1995; Lewan et al., 2014).

OM with little conversion

OM with significant conversion Fig. 21. SEM images of the Woodford organic-rich siliceous mudstone sample at the end of bitumen and early oil generation stage. (A) Modified mineral pores. Some pores occluded with retained oil. (B) Variations in OM pore development (OM conversion). Q: quartz.

6.4. Pore-connectivity implications Migration of petroleum (pre-oil solid bitumen) into the initial primary mineral pore network was observed in both Barnett and Woodford mudstone samples. When the Barnett samples reached bitumen generation, the majority of the primary intra-clay mineral platelets were filled with OM (Fig. 14c). The pre-oil solid bitumen network was also observed within some phosphatic peloids (Fig. 15). The initial mineral pore network of the Woodford mudstone is predominantly composed of intraparticle pores from dissolution of dolomite and feldspar and within clay- mineral platelets. As with the Barnett, these mineral pores later disappeared and were replaced by migrated petroleum (Fig. 22b), which implies that the pores (interparticle as well as intraparticle) are at least partially interconnected in these mudstones during the stages of thermal maturation and petroleum generation from bitumen to wet gas. Therefore, it is likely that the modified mineral pores form an effective pore network, especially when the amount of post-oil solid bitumen (or pyrobitumen) is minor.

Moczydlowska, 2000). In cross-section view under SEM, micropores in the Tasmanites cannot be resolved through the immature to bitumen maturation stage. However, at the beginning of oil window maturation, Tasmanites converts almost completely to petroleum, resulting in generation of pore space in the maceral. We do not think the immediate ‘loss and flow of hydrocarbons’ within the maceral promptly close up the space within Tasmanites because the surrounding siliceous silts are quite rigid and likely protect the majority of Tasmanites from collapsing. In the subsurface, we believe the Tasmanites would contain generated petroleum. Although Tasmanites OM in the Woodford has a Type I-like geochemical signature, this OM was deposited as marine facies. Abundant marine-derived bituminite or AOM is also observed in the Woodford mudstone. Sources of the matrix bituminite or AOM could be marine snow (organic–mineral aggregates), fecal pellets, and microbial mats (cyanobacteria and thiobacteria) (Tyson, 1995). Mastalerz et al. (2012) stated that “the relative proportions of individual macerals can significantly impact the geochemical characteristics of the organic matter.” We found that pore evolution and pore heterogeneity in mudstones are affected by the mixing of different maceral types, as well as their resulting geochemical properties. The stringy/flaky OM (AOM or matrix bituminite) in the Barnett sample had

6.5. Comparison of pore-evolution models Both the Barnett and Woodford siliceous mudstones in this study were deposited in a deep-marine environment (Cardott and Chaplin, 1993; Loucks and Ruppel, 2007). Based on organic petrography and Rock-Eval data, they both contain predominantly marine Type II 19

International Journal of Coal Geology xxx (xxxx) xxx–xxx

L.T. Ko et al.

Petroleum migration into the original primary mineral pores Pores with trapped fluids

Fig. 22. SEM images from the Woodford organic-rich siliceous mudstone sample in the oil-window generation stage. (A) Most pore spaces are still filled with generated petroleum fluids. Some fluids escaped during sample preparation processes, forming modified mineral pores with relic OM. (B) Evidence of petroleum (bitumen) migration into the original mineral pores (Fig. 11a shows the original mineral pores). (C) Wrinkled surface possibly indicate trapped petroleum fluids under SEM. (D) Modified intraparticle mineral pores in clay minerals. Q: quartz; Py: pyrite.

Modified mineral pores

Pores with trapped fluids Modified mineral pores

Modified mineral pores Tasmanites or Leiosphaeridia moldic pores

Fig. 23. SEM images showing modified mineral pores as the most abundant pore types in the Woodford organicrich siliceous mudstone sample at the end of the oil window and in the early wet-gas generation stage. (A) Tasmanites or Leiosphaeridia moldic pores. (B) Original kerogen that has gone through significant conversion and become modified mineral pores. (C) Modified mineral pores between pyrite framboids. (D) Modified mineral pores between mineral grains and clay minerals. Q: quartz; Py: pyrite.

Modified mineral pores

Modified mineral pores

window) (Figs. 14 and 16), to (4) predominant modified mineral pores (middle oil window) (Fig. 17), and finally to (5) modified mineral pores and OM spongy pores (end of oil window and early oil cracking to wet gas) (Fig. 18). The abundance of modified mineral pores and OM spongy pores increases from the middle of the oil window to early oil cracking to the wet-gas stages (Figs. 17 and 18). The pore-evolution history observed for the Barnett mudstone is similar to that reported by Ko et al. (2016) for Eagle Ford Group calcareous mudstones because the Barnett sample contains only few scattered telalginite particles.

kerogen with minor terrigenous Type III kerogen. Under similar temperature and pressure conditions (thermal maturation), the evolution of OM pores in the Barnett and Woodford mudstones should be similar; however, they are not. Figs. 24 and 25 present general pore- evolution models for the Barnett and Woodford mudstones, respectively. During OM conversion, the predominant pore types in the Barnett sample evolve from (1) primary mineral pores (immature) (Fig. 12), to (2) lesscommon OM bubble pores (early bitumen) (Fig. 14), to (3) co-dominant OM bubble pores and modified mineral pores with relic OM (early oil 20

International Journal of Coal Geology xxx (xxxx) xxx–xxx

L.T. Ko et al.

Original pore network (light blue) in the immature Barnett mudstone is predominantly composed of intraparticle pores between clay-mineral platelets and within phosphate peloids.

Bubble-shaped pores (white) started to develop in the stringy/flaky OM and in the matrix bituminite. Migrated preoil solid bitumen was observed within phosphate peloids.

Pores (white) continued to develop in the stringy/flaky OM and in the matrix bituminite. Stringy/flaky OM (upper) shows more OM pore development than OM-mineral admixture (lower).

Pores (white) continued to develop in the stringy/flaky OM and OM mineral admixture in the matrix, all are bituminite (or AOM). OM-hosted spongy pores were developed in the matrix bituminite.

The abundance of OM-hosted spongy pores increased in the matrix bituminite.

21

Fig. 24. A generalized model for the formation and evolution of OM and mineral pores observed in the Barnett siliceous mudstone facies sample. (I) Immature: The mudstone is significantly compacted and composed of much ductile OM and clay minerals. Pores are filled with connate water. The primary pore network consists of mostly intraparticle and a few interparticle pores. (II) Peak bitumen: Stringy/flaky OM and the OM in the matrix have gradually converted to bitumen, creating pore spaces (OM-hosted bubble-shaped pores). Bitumen migration (in phosphate peloid: black arrows) was observed. (III) Early oil: More conversion of kerogen to bitumen and some conversion of bitumen to oil has occurred, creating more pore spaces. Some pores are hosted by minerals (modified mineral pores with relic OM) and some are hosted by OM (OM-hosted bubble pores). (IV) Oil window: Most bitumen has converted to hydrocarbons. Some nanometer-sized OM-hosted spongy pores have begun to appear in the post-oil solid bitumen or pyrobitumen. Most pores are hosted by minerals (modified mineral pores with relic OM). (V) Early cracking of oil to wet gas: Similar to stage (IV) but with increased amount of OM-hosted spongy pores. Clay: clay minerals (green); F: feldspar (orange); P: modified mineral pores (white); Py: pyrite; Q: quartz (red); W: water (blue); distinct kerogen piece (black); amorphous kerogen (gray). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

International Journal of Coal Geology xxx (xxxx) xxx–xxx

L.T. Ko et al.

entire oil-window phase of maturation were less than those in the Barnett and Eagle Ford mudstone samples. Instead, a great amount of petroleum fluid was retained within the pore spaces of the Woodford siliceous mudstone sample and not expelled, even after post-experimental procedures and sample preparation processes. This retained petroleum fluid is reflected on the high S1 peaks on Rock-Eval pyrolysis pyrograms (please see supplement data) and relatively high S1 values in Table 4 throughout the oil window and early gas window (333 °C–425 °C). This might also suggest that the Woodford is ‘tighter’ (has lower permeability) than either the Barnett or the Eagle Ford mudstones. Therefore, it is more difficult for generated oil to expel and migrate. Third, fewer OM spongy pores were found in the Woodford sample, even during the cracking of oil to wet-gas generation stage, when compared with Barnett and Eagle Ford samples. This could be related to the fact that the Woodford samples had not generated significant amounts of pyrobitumen during gold-tube pyrolysis, as had the Barnett and Eagle Ford.

The Woodford siliceous mudstone sample has a different pore-evolution history because it is characterized by relatively abundant Tasmanites (telalginite) OM that did not undergo transformation to petroleum (bitumen) until the oil window (Fig. 25). When conversion of Tasmanites OM occurred, it happened quickly, expulsion was sudden, and transformation efficiency of Tasmanites OM was excellent (> 80%) (Figs. 21b, 23a, and 25). Our study is the first to demonstrate this observation. The Tasmanites has generation potential and OM conversion similar to that of Type I kerogen (Marzi and Rullkötter, 1992; Mukhopadhyay and Gormly, 1984; Revill et al., 1994; Tissot et al., 1974; Vigran et al., 2008). The rest of the macerals and AOM (matrix bituminite) in the Woodford samples behaved, relative to pore evolution, similarly to the Barnett and Eagle Ford mudstone samples. The onset of pore development began at the beginning of the bitumen generation window; pores evolve gradually with petroleum generation, and the OM transformation ratio is generally < 50%. Second, the occurrence and abundance of modified mineral pores throughout the

Original pore network (light blue) in the immature Woodford mudstone is predominantly composed of intraparticle pores between claymineral platelets, from dissolution of dolomite rims, and within pyrite framboids.

Bubble-shaped pores (white) started to develop in the matrix bituminite. Some of the original intraparticle pores are filled with migrated pre-oil solid bitumen.

Increasing numbers of OM bubble pores were developed in the matrix bituminite. An increasing amount of original intraparticle pores are filled with migrated pre-oil solid bitumen.

22

Fig. 25. A generalized model for the formation and evolution of OM and mineral pores observed in the Woodford siliceous mudstone. (I) Immature: The mudstone is significantly compacted because it contains much ductile material like OM and clay minerals. Pores are filled with connate water. The primary pore network consists of mostly intraparticle pores, with few interparticle pores. (II) Early bitumen: A small amount of OM in the matrix has gradually converted to bitumen, creating some pore spaces (OM-hosted bubble-shaped pores). (III) Peak bitumen: A small amount of OM in the matrix has gradually converted to bitumen, creating pore spaces (OM-hosted bubble pores and modified mineral pores). (IV) Early oil: Some Leiosphaeridia and Tasmanites (telalginite) have started to convert to petroleum. More pores have developed in the matrix, some filled with petroleum. (V) Oil window: Leiosphaeridia and Tasmanites (telalginite) almost fully converted to petroleum, creating abundant pore spaces. There is still much petroleum retained in the pores. (VI) Early cracking of oil to wet gas: Most petroleum in the matrix has been expelled, leaving abundant pores. Clay: clay minerals (green); D: dolomite (purple); P, P′, P″: pores (white); Q: quartz (red); W: water (blue); trapped fluid (HC: light gray); distinct kerogen piece (black); amorphous kerogen (gray). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

International Journal of Coal Geology xxx (xxxx) xxx–xxx

L.T. Ko et al.

Fig. 25. (continued)

Leiosphaeridia and Tasmanites started to transfer to petroleum, leaving large moldic pores (white). This implies that the transformation from kerogen to petroleum in these telalginite particles is exceptional (more than 80%). Pores were developed in bituminite, but some were still filled with petroleum (light gray).

Leiosphaeridia and Tasmanites continued to transfer to petroleum, leaving large moldic pores (white). At this stage, the majority of Leiosphaeridia and Tasmanites have been fully converted. Many pores in bituminite were filled with petroleum (light gray).

Pores in bituminite no longer contained petroleum. Some spongy-shaped pores were developed in the pyrobitumen in the matrix bituminite.

solid bitumen by identifying different bonding of carbon in organic compounds at microscopic scale (Bernard et al., 2012a,b; RomeroSarmiento et al., 2014).

The evolution of pores and their abundance corresponds to the amount of petroleum (bitumen, oil, and gas) generated (Fig. 11 and Table 6). For example, during early to middle oil window maturation when most of the kerogen and bitumen have converted to oil, volume loss from solid kerogen is the greatest, resulting in significant increases in pore space. At the same time, the rock contains the most accumulated oil, resulting in the greatest abundance of modified mineral pores. When the rock enters the middle to late oil window maturation, post-oil solid bitumen (or pyrobitumen) starts to form and occludes parts of the pore spaces in source rock reservoirs. Simultaneously, gas generation increases along with the formation of pyrobitumen, forming OM spongy pores in the post-oil solid bitumen (or pyrobitumen) (Fig. 11). Bernard et al. (2012b) suggested that the residual OM that hosts nanometersized OM pores (OM spongy pores in our definition) might be pyrobitumen (post-oil solid bitumen). Theoretically, pyrobitumen, or postoil solid bitumen, increases from oil window to gas window, which corresponds to our observation that the abundance of OM spongy pores increases from the middle of the oil window to the early oil cracking to wet gas stage in the Barnett. In the future, we would like to use scanning transmission X-ray microscopy (STXM) to differentiate kerogen vs.

7. Conclusions Laboratory gold-tube pyrolysis, SEM petrography, thin-section petrography, organic petrology, mineralogical identification, and geochemical characterization of Barnett and Woodford siliceous mudstones suggest that variations in mineralogy (siliceous vs. calcareous) do not have catalytic or sorption effects on pore evolution. Variations in the amount and types of clay minerals also did not show any significant catalytic or sorption effect on generated petroleum and, thus, pore development and evolution. This study reveals the under-appreciation of applying organic petrography to the study of maceral and palynomorph assemblages for OM facies identification in mudstones and their associated pore evolution. Maceral types, identified using both SEM (platy OM, particulate OM, organic-mineral admixtures, Tasmanites) and organic petrology (telalginite, vitrinite, inertinite, AOM,), do affect the evolution of OM 23

International Journal of Coal Geology xxx (xxxx) xxx–xxx

L.T. Ko et al.

manuscript. We also would like to thank the Bureau of Economic Geology Mudrock System Research Laboratory (MSRL) sponsors who supported this research: Anadarko, Apache, BHP Billiton, BP, Cenovus, Centrica, Chesapeake, Chevron, Cima, Cimarex, Concho, ConocoPhillips, Cypress, Devon, Encana, Eni, EOG, EXCO, ExxonMobil, FEI, Hess, Husky, IMP, Kerogen, Marathon, Murphy, Newfield, Oxy, Penn Virginia, Penn West, Pioneer, QEP, Samson, Shell, Statoil, Talisman, Texas American Resources, The Unconventionals, U.S. EnerCorp, Valence, and YPF. Publication was authorized by the Director, Bureau of Economic Geology, Jackson School of Geosciences, The University of Texas at Austin.

pores. This cause is related to differences in chemical compositions, generation kinetics, and activation energy distributions between Tasmanites, AOM, inertinite, and other types of macerals. In this study, we integrated SEM petrography with organic petrography based on observation and description of size, abundance, mineral mixing, and morphologic characteristics of organic components in the Woodford and Barnett mudstones. We interpreted SEM observations of pure OM to be telalginite macerals; OM-mineral admixtures and stringy/flaky OM to be AOM or matrix bituminite; and particulate OM to be inertinite or vitrinite macerals. Although both OM–mineral admixtures and stringy/ flaky OM represent AOM (or matrix bituminite), OM–mineral admixtures demonstrate a greater admixture of mineral matter than the stringy/flaky OM does. Applying organic petrography to identify the type and abundance of macerals especially algae such as Botryococcus and Tasmanites, is a necessary step in order to predict the amount of petroleum generated and associated pore-evolution models in organicrich mudstones (unconventional source rock reservoirs) accurately. Our results show that OM-rich Woodford rocks exhibit a markedly different pore evolution from Barnett and Eagle Ford mudstones because of the abundance of Tasmanites. Although the Barnett mudstone sample also contains telalginite maceral, its occurrence is rare. Therefore, the pore-evolution history observed for the Barnett sample is similar to that reported by Ko et al. (2016) for Eagle Ford Group calcareous mudstones. At higher levels of thermal maturation, the volume of primary mineral pores decreases and the pore volume composed of modified mineral pores and OM pores increases. The Woodford siliceous mudstone has a different pore-evolution model because it contains relatively abundant telalginite maceral (Tasmanites OM). During OM conversion, the predominant pore types change from (1) OM-hosted bubble-shaped pores (early bitumen), to (2) few OM-hosted bubble pores and modified mineral pores (peak bitumen), to (3) increased abundance of OM-hosted bubble pores and modified mineral pores (early oil; Tasmanites has started to convert to petroleum), to (4) predominant modified mineral pores (oil window; full conversion of Tasmanites OM to petroleum, creating abundant pore space but with petroleum retained in the pores as solid bitumen), and finally to (5) most abundant modified mineral pores (early cracking of oil to wet gas; most petroleum in the matrix has been expelled, leaving abundant pores). Unlike the Barnett siliceous and Eagle Ford calcareous facies, spongy OM pores were rare in the Woodford samples. Petroleum generation and pore development from algal-rich Type II kerogen (Tasmanites) might be similar to that of Type I kerogen which has a narrower activation energy distribution and a higher mean activation energy than does the typical Type II kerogen. These two differences result in algal-rich Type II kerogen having a later and shorter OM conversion and petroleum generation history in the Woodford siliceous mudstone, resulting in a different pore-development history and possibly a different expulsion model compared to Barnett and Eagle Ford mudstones. Because the Woodford has the highest TOC content among the samples or it is possibly tighter than the Barnett and Eagle Ford, generated petroleum is retained longer in the pore space of the rock.

References Abraham, H., 1945. Asphalts and Allied Substances. D Van Nostrand Company, New York (1491 pp). American Society of Testing and Materials (ASTM) D2797, 2016. Standard Practice for Preparing Coal Samples for Microscopical Analysis by Reflected Light: Annual Book of ASTM Standards: Petroleum Products, Lubricants, and Fossil Fuels. Gaseous Fuels: Coal and Coke. sec. 5, v. 5.06. ASTM International, West Conshohocken, PA, pp. 542–546. http://www.astm.org/Standards/D2797.htm. American Society of Testing and Materials (ASTM) D7708, 2016. Standard Test Method for Microscopical Determination of the Reflectance of Vitrinite Dispersed in Sedimentary Rocks: Annual Book of ASTM Standards: Petroleum Products, Lubricants, and Fossil Fuels. Gaseous Fuels: Coal and Coke. sec. 5, v. 5.06. ASTM International, West Conshohocken, PA, pp. 923–932. http://www.astm.org/ Standards/D7708.htm. Aplin, A.C., Macquaker, J.H.S., 2011. Mudstone diversity: Origin and implications for source, seal, and reservoir properties in petroleum system. AAPG Bull. 95, 2031–2059. Aufill, M., 2007. High Resolution Magnetic Susceptibility of the Oklahoma Woodford Shale and Relationships to Variations in Outcrop Spectral Gamma-Ray Response. Master's thesis. Oklahoma State University, Stillwater, OK (165 p). Bastow, T.P., van Aarssen, B.G.K., Lang, D., 2007. Rapid small-scale separation of saturate, aromatic, and polar components in petroleum. Org. Geochem. 38, 1235–1250. Behar, F., Lewan, M.D., Lorant, F., Vandenbroucke, M., 2003. Comparison of artificial maturation of lignite in hydrous and nonhydrous conditions. Org. Geochem. 34, 575–600. Bernard, S., Horsfield, B., 2014. Thermal maturation of gas shale systems. Annu. Rev. Earth Planet. Sci. 42, 635–651. http://dx.doi.org/10.1146/annurev-earth-060313054850. Bernard, S., Horsfield, B., Schulz, H.-M., Wirth, R., Schreiber, A., Sherwood, N., 2012a. 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. http://dx.doi.org/10.1016/j.marpetgeo.2011.05.010. Bernard, S., Wirth, R., Schreiber, A., Schulz, H.-M., Horsfield, B., 2012b. Formation of nanoporous pyrobitumen residues during maturation of the Barnett Shale (Fort Worth Basin). Int. J. Coal Geol. 103, 3–11. http://dx.doi.org/10.1016/j.coal.2012. 04.010. Bertrand, P., Pittion, J.-L., Bernaud, C., 1986. Fluorescence of sedimentary organic matter in relation to its chemical composition. Org. Geochem. 10, 641–647. http://dx.doi. org/10.1016/0146-6380(86)90061-6. Blakey, R., 2009. Paleogeography and geologic evolution of North America, Late Devonian (360 Ma). Map: http://jan.ucc.nau.edu/~rcb7/namD360.jpg (accessed 09.18.2016). Bohacs, K.M., Passey, Q.R., Rudnicki, M., Esch, W.L., Lazar, O.R., 2013. The spectrum of fine-grained reservoirs from ‘shale gas’ to ‘shale oil’/tight liquids: Essential attributes, key controls, practical characterization. In: International Petroleum Technology Conference, Beijing, China, 26–28 March, (IPTC-16676-MS, 16 pp.). Bowker, K.A., 2007. Barnett Shale gas production, Fort Worth Basin: issues and discussion. AAPG Bull. 91, 523–533. http://dx.doi.org/10.1306/06190606018. Camp, W.K., 2014. Diagenesis of Organic-rich Shale: Views from Foraminifera Penetralia, Eagle Ford Formation, Maverick Basin, Texas (abs.). (AAPG Search and Discovery Article # 51054). Cardott, B.J., 2011. Four distinct Woodford Shale plays in Oklahoma—gas, condensate, oil, and biogenic methane. In: AAPG Hedberg Conference, . http://www. searchanddiscovery.com/abstracts/pdf/2011/hedberg-texas/abstracts/ndx_cardott. pdf (accessed 01.23.2017). Cardott, B.J., Chaplin, J.R., 1993. Guidebook for Selected Stops in the Western Arbuckle Mountains, Southern Oklahoma, Oklahoma Geological Survey Special Publication 93–3. (61 p). Cardott, B.J., Curtis, M.E., 2017. Identification and nanoporosity of macerals in coal by scanning electron microscopy. Int. J. Coal Geol in press. https://doi.org/10.1016/j. coal.2017.07.003. Cardott, B.J., Landis, C.R., Curtis, M.E., 2015. Post-oil solid bitumen network in the Woodford Shale, USA — a potential primary migration pathway. Int. J. Coal Geol. 139, 106–113. Comer, J.B., 2007. Lithologic characteristics and gas production potential of Woodford Shale in the southern midcontinent. In: GSA Annual Meeting, Denver, Colo., 28–31 October 2007. Curiale, J.A., 1986. Origin of solid bitumens, with emphasis on biological marker results.

Acknowledgements The Woodford outcrop samples for this study were kindly provided by Dr. Geoffrey S. Ellis of the Central Energy Resources Team at the U.S. Geological Survey, Denver, Colorado. The authors would like to thank Lanzhou University, China, for carrying out the pyrolysis experiments with the support of the Major State Basic Research Development Program of China (Grant number: 2012CB214701). The authors also thank Patrick Smith for providing CL training and imaging. Dr. Mike Lewan, Dr. Kitty Milliken, Dr. Terri Olson, Dr. Robert Reed, and Dr. Joan Spaw provided insightful discussions, suggestions, and inspiration during the manuscript preparation. Editor Shifeng Dai, reviewers Joe Curiale, and two anonymous reviewers provided constructive, detailed, and insightful comments which helped improve the quality of the 24

International Journal of Coal Geology xxx (xxxx) xxx–xxx

L.T. Ko et al.

pore types. In: Camp, W.K., Diaz, E., Wawak, B. (Eds.), Electron Microscopy of Shale Hydrocarbon Reservoirs, pp. 101–113 AAPG Memoir 102. Johns, W.D., 1979. Clay mineral catalysis and petroleum generation. Annu. Rev. Earth Planet. Sci. 7, 183–198. Katz, B., Lin, F., 2014. Lacustrine basin unconventional resource plays: key differences. Mar. Pet. Geol. 56, 255–265. http://dx.doi.org/10.1016/j.marpetgeo.2014.02.013. Kelemen, S.R., Walters, C.C., Kwiatek, P.J., Afeworki, M., Sansone, M., Freund, H., Pottorf, R.J., Machel, H.G., Zhang, T., Ellis, G.S., Tang, Y., Peters, K.E., 2008. Distinguishing solid bitumens formed by thermochemical sulfate reduction and thermal chemical alteration. Org. Geochem. 39, 1137–1143. Kirkland, D.W., Denison, R.E., Summers, D.M., Gormly, J.R., 1992. Geology and organic geochemistry of the Woodford Shale in the Criner Hills and western Arbuckle Mountains. In: Johnson, K.S., Cardott, B.J. (Eds.), Source Rocks in the Southern Midcontinent, 1990 Symposium, pp. 38–69 Oklahoma Geological Survey Circular 93. Ko, L.T., Loucks, R.G., Zhang, T., Ruppel, S.C., Shao, D., 2016. Pore and pore network evolution of Upper Cretaceous Boquillas (Eagle Ford-equivalent) mudrocks: results from gold-tube pyrolysis experiments. AAPG Bull. 100, 1693–1722. http://dx.doi. org/10.1306/04151615092. Ko, L.T., Loucks, R.G., Ruppel, S.C., Zhang, T., Peng, S., 2017. Origin and characterization of Eagle Ford pore networks in the south Texas Upper Cretaceous shelf. AAPG Bull. 101, 387–418. http://dx.doi.org/10.1306/08051616035. Krystyniak, A.M., 2005. Outcrop-Based Gamma-Ray Characterization of the Woodford Shale of South-central Oklahoma. Master's thesis. Oklahoma State University, Stillwater, Oklahoma (145 p). Lambert, M.W., 1993. Internal stratigraphy and organic facies of the DevonianMississippian Chattanooga (Woodford) Shale in Oklahoma and Kansas: source rocks in a sequence stratigraphic framework: chapter 11. AAPG Stud. Geol. 37, 163–176. Landis, C.R., Castaño, J.R., 1995. Maturation and bulk chemical properties of a suite of solid hydrocarbons. Org. Geochem. 22, 137–149. Lao, Y., Korth, J., Ellis, J., Crisp, P.T., 1989. Heterogeneous reactions of 1-pristene catalyzed by clays under simulated geological conditions. Org. Geochem. 14, 375–379. Lewan, M.D., 1983. Effects of thermal maturation on stable organic carbon isotopes as determined by hydrous pyrolysis of Woodford Shale. Geochim. Cosmochim. 47, 1471–1479. http://dx.doi.org/10.1016/0016-7037(83)90306-X. Lewan, M.D., 1987. Petrographic study of primary petroleum migration in the Woodford Shale and related rock units. In: Doligez, B. (Ed.), Migration of Hydrocarbons in Sedimentary Basins. Editions Technip, Paris, pp. 113–130. Lewan, M.D., 1997. Experiments on the role of water in petroleum formation. Geochim. Cosmochim. 67, 3691–3723. Lewan, M.D., Dolan, M.P., Curtis, J.B., 2014. Effects of smectite on the oil-expulsion efficiency of the Kreyenhagen Shale, San Joaquin Basin, California based on hydrouspyrolysis experiments. AAPG Bull. 98, 1091–1109. http://dx.doi.org/10.1306/ 10091313059. Littke, R., Klussmann, U., Krooss, B., Leythaeuser, D., 1991. Quantification of loss of calcite, pyrite, and organic matter due to weathering of Toarcian black shales and effects on kerogen and bitumen characteristics. Geochim. Cosmochim. 55, 3369–3378. Lo, H.B., Cardott, B.J., 1995. Detection of natural weathering of Upper McAlester coal and Woodford Shale, Oklahoma, U.S.A. Org. Geochem. 22, 73–83. Löhr, S.C., Baruch, E.T., Hall, P.A., Kennedy, M.J., 2015. Is organic pore development in gas shales influenced by the primary porosity and structure of thermally immature organic matter? Org. Geochem. 87, 119–132. http://dx.doi.org/10.1016/j. orggeochem.2015.07.010. Loucks, R.G., Reed, R.M., 2014. Scanning-electron-microscope petrographic evidence for distinguishing organic-matter pores associated with depositional organic matter versus migrated organic matter in mudrocks. GCAGS J. 3, 51–60. Loucks, R.G., Ruppel, S.C., 2007. Mississippian Barnett Shale: lithofacies and depositional setting of a deep-water shale-gas succession in the Fort Worth Basin, Texas. AAPG Bull. 91, 579–601. http://dx.doi.org/10.1306/11020606059. 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. http://dx.doi.org/10.2110/jsr. 2009.09. 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. http://dx.doi.org/10.1306/08171111061. Machel, H.G., Krouse, H.R., Sassen, R., 1995. Products and distinguishing criteria of bacterial and thermochemical sulfate reduction. Appl. Geochem. 10, 373–389. Marzi, R., Rullkötter, J., 1992. Qualitative and quantitative evolution and kinetics of biological marker transformations–laboratory experiments and application to the Michigan Basin. In: Moldowan, J.M., Albrecht, P., Philp, R.P. (Eds.), Biological Markers in Sediments and Petroleum. Prentice-Hall, Englewood Cliffs, NJ, pp. 18–41. Mastalerz, M., Glikson, M., 2000. In-situ analysis of solid bitumen in coal: examples from the Bowen Basin and the Illinois Basin. Int. J. Coal Geol. 42, 207–220. http://dx.doi. org/10.1016/S0166-5162(99)00040-3. Mastalerz, M., Schimmelmann, A., Lis, G.P., Drobniak, A., Stankiewicz, A., 2012. Influence of maceral composition on geochemical characteristics of immature shale kerogen: insight from density fraction analysis. Int. J. Coal Geol. 103, 60–69. http:// dx.doi.org/10.1016/j.coal.2012.07.011. Mastalerz, M., Schimmelmann, A., Drobniak, A., Chen, Y., 2013. Porosity of Devonian and Mississippian New Albany Shale across a maturation gradient: insights from organic petrology, gas adsorption, and mercury intrusion. AAPG Bull. 97, 1621–1643. http:// dx.doi.org/10.1306/04011312194. McIver, R.D., 1967. Composition of kerogen— clue to its role in the origin of petroleum. In: Proceedings of the 7th World Petroleum Congress, Mexico City, Mexico, pp. 25–36.

Org. Geochem. 10, 559–580. Curtis, M.E., Cardott, B.J., Sondergeld, C.H., Rai, C.S., 2012. Development of organic porosity in the Woodford Shale with increasing thermal maturity. Int. J. Coal Geol. 103, 26–31. http://dx.doi.org/10.1016/j.coal.2012.08.004. Derenne, S., Metzger, P., Largeau, C., van Bergen, P.F., Gatellier, J.P., Sinninghe Damste, J.S., de Leeuw, J.W., Berkaloff, C., 1992. Similar morphological and chemical variations of Gloeocapsomorpha prisca in Ordovician sediments and cultured Botryococcus braunii as a response to changes in salinity. Org. Geochem. 19, 299–313. Espitalié, J., Madec, M., Tissot, B., 1980. Role of mineral matrix in kerogen pyrolysis: influence on petroleum generation and migration. AAPG Bull. 64, 59–66. Espitalié, J., Senga Makadi, K., Trichet, J., 1984. Role of the mineral matrix during kerogen pyrolysis. Org. Geochem. 6, 365–382. http://dx.doi.org/10.1016/01466380(84)90059-7. Fishman, N.S., Hackley, P.C., Lowers, H.A., Hill, R.J., Egenhoff, S.O., Eberl, D.D., Blum, A.E., 2012. The nature of porosity in organic-rich mudstones of the Upper Jurassic Kimmeridge Clay Formation, North Sea, offshore United Kingdom. Int. J. Coal Geol. 103, 32–50. http://dx.doi.org/10.1016/j.coal.2012.07.012. Fishman, N.S., Ellis, G.S., Paxton, S.T., Boehlke, A.R., Egenhoff, S.O., 2013. Gas storage in the Upper Devonian–Lower Mississippian Woodford Shale, Arbuckle Mountains, Oklahoma: How much of a role do chert beds play? In: Chatellier, J., Jarvie, D.M. (Eds.), Critical Assessment of Shale Resource Plays, pp. 81–107. AAPG Memoir 103. 10:1306/13401726H53493. Fowler, M.G., Stasiuk, L.D., Hearn, M., Obermajer, M., 2004. Evidence for Gloeocapsomorpha prisca in Late Devonian source rocks from southern Alberta, Canada. Org. Geochem. 35, 425–441. Goldstein, T.P., 1983. Geocatalytic reactions in formation and maturation of petroleum. AAPG Bull. 67, 152–159. Hackley, P.C., Cardott, B.J., 2016. Application of organic petrography in north American shale petroleum systems: a review. Int. J. Coal Geol. 163, 8–51. http://dx.doi.org/10. 1016/j.coal.2016.06.010. Hackley, P.C., Kus, J., 2015. Thermal maturity of Tasmanites microfossils from confocal laser scanning fluorescence microscopy. Fuel 142, 343–350. Hackley, P.C., SanFilipo, J.R., 2016. Organic petrology and geochemistry of Eocene Suzak bituminous marl, north-central Afghanistan: depositional environment and source rock potential. Mar. Pet. Geol. 73, 572–589. http://dx.doi.org/10.1016/j.marpetgeo. 2016.02.029. Hackley, P.C., Fishman, N., Wu, T., Baugher, G., 2016. Organic petrology of mudrocks from the lacustrine Lucaogou Formation, Santanghu Basin, northwest China: application to lake basin evolution. Int. J. Coal Geol. 168, 20–34. Hackley, P.C., Valentine, B.J., Voortman, L.M., van Oosten Slingeland, D.S.B., Hatcherian, J., 2017. Utilization of integrated correlative light and electron microscopy (iCLEM) for imaging sedimentary organic matter. J. Microsc. 267, 371–383. Ham, W.E., 1973. Regional geology of the Arbuckle Mountains, Oklahoma: Guidebook for field trip no. 5. November 10-11, with contributions by Amsden, T. William. American Association of Petroleum Geologists, Geological Society of America, and Society of Economic Paleontologists and Mineralogists (compiled by T. L. Rowland, 56p). Han, Y., Mahlstedt, N., Horsfield, B., 2015. The Barnett Shale: compositional fractionation associated with intraformational petroleum migration, retention, and expulsion. AAPG Bull. 99, 2173–2202. Hart, B.S., Macquaker, J.H.S., Taylor, K.G., 2013. Mudstone (“shale”) depositional and diagenetic processes: Implications for seismic analyses of source-rock reservoirs. Interpretation 1, B7–B26. Hetényi, M., 1995. Simulated thermal maturation of type I and III kerogens in the presence, and absence, of calcite and montmorillonite. Org. Geochem. 23, 121–127. http://dx.doi.org/10.1016/0146-6380(94)00120-P. Hickey, J.J., Henk, B., 2007. Lithofacies summary of the Mississippian Barnett Shale, Mitchell 2 T.P. Sims well, Wise County, Texas. AAPG Bull. 91, 437–443. http://dx. doi.org/10.1306/12040606053. Horsfield, B., Douglas, A.G., 1980. The influence of minerals on the pyrolysis of kerogen. Geochim. Cosmochim. 44, 1119–1131. http://dx.doi.org/10.1016/0016-7037(80) 90066-6. Horsfield, B., Yordy, K.L., Crelling, J.C., 1988. Determining the petroleum-generation potential of coal using organic geochemistry and organic petrology. Org. Geochem. 13, 121–129. Huang, W.L., 1996. Experimental study of vitrinite maturation: effects of temperature, time, pressure, water, and hydrogen index. Org. Geochem. 24, 233–241. Huizinga, B.J., Tannenbaum, E., Kaplan, I.R., 1987a. The role of minerals in the thermal alteration of organic matter: III. Generation of bitumen in laboratory experiments. Org. Geochem. 11, 591–604. Huizinga, B.J., Tannenbaum, E., Kaplan, I.R., 1987b. The role of minerals in the thermal alteration of organic matter: IV. Generation of n-alkanes, acyclic isoprenoids, and alkenes in laboratory experiments. Geochim. Cosmochim. 51, 1083–1097. Hwang, R.J., Teerman, S.C., Carlson, R.M., 1998. Geochemical comparison of reservoir solid bitumens with diverse origins. Org. Geochem. 29, 505–517. International Committee for Coal Petrology (ICCP), 1998. The new vitrinite classification (ICCP System 1994). Fuel 77, 349–358. International Committee for Coal Petrology (ICCP), 2001. The new inertinite classification (ICCP System 1994). Fuel 80, 459–471. Jacob, H., 1989. Classification, structure, genesis and practical importance of natural solid bitumen (“migrabitumen”). Int. J. Coal Geol. 11, 65–79. 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. Jennings, D.S., Antia, J., 2013. Petrographic characterization of the Eagle Ford Shale, South Texas: Mineralogy, common constituents, and distribution of nanometer-scale

25

International Journal of Coal Geology xxx (xxxx) xxx–xxx

L.T. Ko et al.

Schenk, H.J., White, E.G., Schwochau, K., 1990. Structural modifications of vitrinite and alglnite concentrates during pyrolytic maturation at different heating rates. A combines infrared, 13C NMR and microscopical study. Org. Geochem. 16, 943–950. Schieber, J., 1996. Early diagenetic silica deposition in algal cysts and spores: a source of sand in black shales? AAPG Bull. 66, 175–183. Schieber, J., 2010. Common themes in the formation and preservation of intrinsic porosity in shales and mudstones—illustrated with examples across the Phanerozoic. In: Soc. Petr. Eng. (SPE) Unconventional Gas Conference, Pittsburgh, Penn., USA, 23–25 February, (Paper # SPE 132370). Schieber, J., Lazar, R., Bohacs, K., Klimentidis, B., Dumitrescu, M., Ottmann, J., 2016. An SEM study of porosity in the Eagle Ford Shale of Texas—pore types and porosity distribution in a depositional and sequence stratigraphic context. In: Breyer, J.A. (Ed.), The Eagle Ford Shale: A Renaissance in U.S. Oil Production. AAPG Memoir 110, pp. 167–186. Slatt, R.M., 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. Sundararaman, P., Teerman, S.C., Mann, R.G., Mertani, B., 1988. Activation energy distribution: a key parameter in basin modeling and a geochemical technique for studying maturation and organic facies. In: Proceedings 17th Annual Indonesian Petrology Association. Vol. 1. pp. 169–185. Sykorova, I., Pickel, W., Christanis, K., Wolf, M., Talor, G.H., Flores, D., 2005. Classification of huminite – ICCP System 1994. Int. J. Coal Geol. 62, 85–106. Talyzina, N.M., Moczydlowska, M., 2000. Morphological and ultrastructural studies of some acritarchs from the Lower Cambrian Lükati Formation, Estonia. Rev. Palaeobot. Palynol. 112, 1–21. Tannenbaum, E., Kaplan, I.R., 1985. Role of minerals in the thermal alteration organic matter—I: generation of gases and condensates under dry condition. Geochim. Cosmochim. 49, 2589–2604. http://dx.doi.org/10.1016/0016-7037(85)90128-0. Tannenbaum, E., Huizinga, B.J., Kaplan, I.R., 1986. Role of minerals in thermal alteration of organic matter—II: a material balance. AAPG Bull. 70, 1156–1165. Telnova, O.P., 2012. Morphology and ultrastructure of Devonian prasinophycean algae (Chlorophyta). Paleontol. J. 46, 534–548. Tissot, B.P., Durand, B., Espitalie, J., Combaz, A., 1974. Influence of nature and diagenesis of organic matter in formation of petroleum. AAPG Bull. 58, 499–506. Tissot, B.P., Pelet, R., Ungerer, P., 1987. Thermal history of sedimentary basins, maturation indices, and kinetics of oil and gas generation. AAPG Bull. 71, 1445–1466. Tissot, B.P., Welte, D.H., 1984. Petroleum Formation and Occurrence, 2nd ed. SpringerVerlag, Berlin (699 pp). Tyson, R.V., 1995. Sedimentary Organic Matter: Organic Facies and Palynofacies. Springer Netherlands (615 pp). Vigran, J.O., Mork, A., Forsberg, A.W., Weiss, H.M., Weitschat, W., 2008. Tasmanites algae—contributors to the Middle Triassic hydrocarbon source rocks of Svalbard and the Barents Shelf. Polar Res. 27, 360–371. Volkman, J.K., Zhang, Z., Xie, X., Qin, J., Borjigin, T., 2015. Biomarker evidence for Botryococcus and a methane cycle in the Eocene Huadian oil shale, NE China. Org. Geochem. 78, 121–134. Walters, C.C., 2007. The origin of petroleum. In: Hsu, C.S., Robinson, P.R. (Eds.), Practical Advances in Petroleum Processing, Volume 1, Chapter 2, pp. 79–101. Wei, Z., Moldowan, J.M., Dahl, J., Goldstein, T.P., Jarvie, D.M., 2006a. The catalytic effects of minerals on the formation of diamondoids from kerogen macromolecules. Org. Geochem. 37, 1421–1436. http://dx.doi.org/10.1016/j.orggeochem.2006.07. 006. Wei, Z., Moldowan, J.M., Paytan, A., 2006b. Diamondoids and molecular biomarkers generated from modern sediments in the absence and presence of minerals during hydrous pyrolysis. Org. Geochem. 37, 891–911. http://dx.doi.org/10.1016/j. orggeochem.2006.04.008. Wood, S.G., 2013. Lithofacies, Depositional Environments, and Sequence Stratigraphy of the Pennsylvanian (Morrowan-Atokan) Marble Falls Formation, Central Texas. Master's thesis. University of Texas at Austin, Austin, TX (276 p). Wood, J.M., Sanei, H., Curtis, M.E., Clarkson, C.R., 2015. Solid bitumen as a determinant of reservoir quality in an unconventional tight gas siltstone play. Int. J. Coal Geol. 150–151, 287–295. Zargari, S., Canter, K.L., Prasad, M., 2015. Porosity evolution in oil-prone source rocks. Fuel 153, 110–117. Zhang, T., Ellis, G.S., Wang, K., Walters, C.C., Kelemen, S.R., Gillaizeau, B., Tang, Y., 2007. Effect of hydrocarbon type on thermochemical sulfate reduction. Org. Geochem. 38, 897–910.

Meckel, L.D., Smith, D.G., Wells, L.A., 1992. Ouachita foredeep basins: regional paleogeography and habit of hydrocarbons. In: Macqueen, R.W., Leckie, D.A. (Eds.), Foreland Basins and Fold Belts: Chapter 15, AAPG Memoir 55, Tulsa, Oklahoma, pp. 427–444. http://dx.doi.org/10.1306/M55563C17. Meissner, F.E., 1978. Petroleum geology of the Bakken Formation, Williston basin. North Dakota and Montana. In: The Economic Geology of the Williston Basin: Montana Geological Society 24th Annual Conference; 1978 Williston Basin Symposium, pp. 207–227. Milliken, K.L., 2013. SEM-based cathodoluminescence imaging for discriminating quartz types in mudrocks. In: Unconventional Resources Technology Conference, Denver, CO, 12–14 August 2013, (URTeC #1582467). Milliken, K.L., Laubach, S.E., 2000. Brittle deformation in sandstone diagenesis as revealed by scanned cathodoluminescence imaging with application to characterization of fractured reservoirs. In: Pagel, M., Barbin, V., Blanc, P., Ohnenstetter, D. (Eds.), Cathodoluminescence in Geosciences: Chapter 9, pp. 225–243. Milliken, K.L., Choh, S.-J., Papazis, P., Schieber, J., 2007. “Cherty” stringers in the Barnett Shale are agglutinated foraminifera. Sediment. Geol. 198, 221–232. Milliken, K.L., Esch, W.L., Reed, R.M., Zhang, T., 2012. Grain assemblages and strong diagenetic overprinting in siliceous mudrocks, Barnett Shale (Mississippian), Fort Worth Basin, Texas. AAPG Bull. 96, 1553–1578. Milliken, K.L., Rudnicki, M., Awwiller, D.N., Zhang, T., 2013. Organic matter-hosted pore system, Marcellus Formation (Devonian), Pennsylvania. AAPG Bull. 97, 177–200. Montgomery, S.L., Jarvie, D.M., Bowker, K.A., Pollastro, R.M., 2005. Mississippian Barnett Shale, Fort Worth Basin, northcentral Texas: gas-shale play with multi-trillion cubic foot potential. AAPG Bull. 89, 155–175. http://dx.doi.org/10.1306/ 09170404042. Mukhopadhyay, P.K., Gormly, J.R., 1984. Hydrocarbon potential of two types of resinite. Org. Geochem. 6, 439–454. Nicholas, R.L., Rozendal, R.A., 1975. Subsurface positive elements within Ouachita foldbelt in Texas and their relation to Paleozoic cratonic margin. AAPG Bull. 59, 193–216. Pan, C., Geng, A., Zhong, N., Liu, J., 2010. Kerogen pyrolysis in the presence and absence of water and minerals: steranes and triterpenoids. Fuel 89, 336–345. http://dx.doi. org/10.1016/j.fuel.2009.06.032. Paxton, S.T., Cardott, B.J., 2008. Oklahoma gas shales, field trip guidebook. In: Oklahoma Geological Survey Open File Report 2–2008, (87 p). Pepper, A.S., Corvi, P.J., 1995. Simple kinetic models of petroleum formation. Part 1: oil and gas generation from kerogen. Mar. Pet. Geol. 12, 291–319. Peters, K.E., Walters, C.C., Mankiewicz, P.J., 2006. Evaluation of kinetic uncertainty in numerical models of petroleum generation. AAPG Bull. 90, 387–403. Petsch, S.T., Berner, R.A., Eglinton, T.I., 2000. A field study of the chemical weathering of ancient sedimentary organic matter. Org. Geochem. 31, 475–487. Petsch, S.T., Smernik, R.J., Eglinton, T.I., Oades, J.M., 2001. A solid state 13C-NMR study of kerogen degradation during black shale weathering. Geochim. Cosmochim. 65, 1867–1882. Pickle, W., Kus, J., Flores, D., Kalaitzidis, S., Christanis, K., Cardott, B.J., Misz-Kennan, M., Rodrigues, S., Hentschel, A., Hamor-Vido, M., Crosdale, P., Wagner, N., ICCP, 2017. Classification of liptinite – ICCP system 1994. Int. J. Coal Geol. 169, 40–61. Pollastro, R.M., Jarvie, D.M., Hill, R.J., Adams, C.W., 2007. Geologic framework of the Mississippian Barnett Shale, Barnett–Paleozoic total petroleum system, Bend arch–Forth Worth Basin, Texas. AAPG Bull. 91, 405–436. http://dx.doi.org/10.1306/ 10300606008. Pommer, M.E., Milliken, K.L., 2015. Pore types and pore-size distributions across thermal maturity, Eagle Ford Formation, southern Texas. AAPG Bull. 99, 1713–1744. http:// dx.doi.org/10.1306/03051514151. Pradier, B., Bertrand, B., Martinez, L., Laggoun-Defarge, F., 1991. Fluorescence of organic matter and thermal maturity assessment. Org. Geochem. 17, 511–524. http://dx.doi. org/10.1016/0146-6380(91)90115-Z. Redmond, L., 2016. Lithofacies, Depositional Environments, and Depositional Model of the Mississippian Barnett Formation in the Southern Fort Worth Basin. Master's thesis. University of Texas at Austin, Austin, TX (162 p). Revill, A.T., Volkman, J.K., O'Leary, T., Summons, R.E., Boreham, C.J., Banks, M.R., Denwer, K., 1994. Hydrocarbon biomarkers, thermal maturity, and depositional setting of tasmanite oil shales from Tasmania, Australia. Geochim. Cosmochim. 58, 3803–3822. Romero-Sarmiento, M.-F., Rouzaud, J.-N., Bernard, S., Deldicque, D., Thomas, M., Littke, R., 2014. Evolution of Barnett Shale organic carbon structure and nanostructure with increasing maturation. Org. Geochem. 71, 7–16.

26