Characterization of Gas Generated by Sequential Hydrous Pyrolysis of Potential Gas-Prone Source Rocks for Tight-Gas Reservoirs in the Rocky Mountain Area

CHARACTERIZATION OF GAS GENERATED BY SEQUENTIAL HYDROUS PYROLYSIS OF POTENTIAL GAS-PRONE SOURCE ROCKS FOR TIGHT-GAS RESERVOIRS IN THE ROCKY MOUNTAIN AREA

by Tingwei(Lucy) Ko

ABSTRACT

The source of unconventional gas in tight-gas reservoirs of the Rocky Mountain area is uncertain, but possible Cretaceous gas-prone source rocks include the Cameo coal zone, Mowry, Mancos, and Baxter/Hilliard Shales. Sequential hydrous pyrolysis experiments were conducted on immature samples of these source rocks to characterize their generated gases and evaluate their potential as sources for gas accumulations in the Greater Green River Basin and Piceance Basin. The experiments were conducted sequentially for 72 h at 300, 330, and 360°C, equivalent to measured vitrinite reflectance values of 1.0, 1.3, and 1.6 %Ro, respectively. After each 72 h experiment, the generated gas and expelled oil were removed from the reactor. Gases generated from each sequential experiment were analyzed for molecular composition and stable carbon and hydrogen isotopes.

All source rocks generated significant hydrocarbon and non-hydrocarbon gases (N2, H2S,

H2, and CO2). Cumulative yields of methane to butane increased with increasing thermal maturation. On a per gram of total-organic-carbon (TOC) basis, the methane yield from the Cameo coal at 360°C exceeded that at 300°C by almost eightfold. The cumulative methane yields on a TOC basis, from highest to lowest, are Cameo coal, Baxter/Hilliard, Mowry, and Mancos Shales. Cumulative gas wetness from both Mancos and Mowry Shales were high. With increasing thermal maturity, δ13C of methane from all four source rocks became lighter, consistent to the beginning trend of the convention model.

The Baxter/Hillard Shale generated the greatest amount of H2, H2S, and CO2, on a TOC 13 basis, whereas Cameo coal generated the least. The relatively positive δ C values for CO2 from the Mancos Shale and Baxter Shale suggests it was sourced from thermal decomposition of 13 carbonate minerals in the original rock samples. More negative δ C values for CO2 from the 13 Cameo coal indicate that gas is from an organic source, while intermediate δ C of CO2 from the Mowry Shale indicates contributions from both organic and inorganic sources or merely from the inorganic source. Gases generated from laboratory experiments are isotopically lighter than gas sampled from the Jonah and Piceance Basin fields. Two explanations are proposed: 1) Migration effects from the source to the reservoir and escaping gas from reservoirs may significantly alter final gas

iii compositions; 2) Gases from experiments only reach primary cracking stage of gas generation, whereas field gas may represent gas from secondary cracking of oils deeper in the basins. KEY WORDS: Carbon isotopes, natural gas, hydrous pyrolysis, Jonah Field, Piceance Basin

TABLE OF CONTENTS

ABSTRACT...... iii LIST OF FIGURES ...... viii LIST OF TABLES...... xv LIST OF ABBREVIATIONS...... xvii ACKNOWLEDGEMENTS...... xviii CHAPTER 1 INTRODUCTION ...... 1 1.1 Unconventional Tight Gas Resources...... 1 1.2 Research Objectives...... 2 CHAPTER 2 PREVIOUS RESEACH...... 5 2.1 The Process of Natural Gas Generation, from a Molecular Standpoint ...... 5 2.2 Sources of Natural Gas and Their Genetic Isotopic Signatures...... 12 2.2.1 Source of Natural Gas in the Piceance Basin and Jonah Field...... 13 2.2.2 Implications of Carbon and Hydrogen Isotopes in Gases ...... 15 2.2.3 Hydrogen Isotope Effects...... 22 CHAPTER 3 GEOLOGICAL BACKGROUND ...... 25 3.1 Piceance Basin, Colorado ...... 26 3.1.1 Source Rocks – Maturity and Potential...... 29 3.2 Greater Green River Basin, Wyoming...... 35 3.2.1 Source Rocks – Maturity and Potential...... 40 CHAPTER 4 METHODS & EXPERIMENTS ...... 47 4.1 Potential Source Rock Sample Collection ...... 47 4.2 Sources Rock Characterization...... 49 4.2.1 Sample Analysis in the SRA ...... 50 4.3 Hydrous Pyrolysis Experiments...... 55 4.3.1 Composite Sample Preparation ...... 56 4.3.2 Experimental Conditions...... 57 4.3.3 Hydrocarbon Collection ...... 57 4.4 Measured Vitrinite Reflectance ...... 58 4.4.1 Sample Preparation...... 58 4.4.2 Measured Vitrinite Reflectance...... 61 v

4.5 Gas-Chromatogram (GC) Analysis...... 62 4.6 Gas Composition and Isotope Analysis ...... 62 4.6.1 Gas Isotope Analysis ...... 63 CHAPTER 5 RESULTS...... 65 5.1 Core Description of the Mowry Shale Core ...... 65 5.2 Original Samples – Organic Geochemical Analysis...... 65 5.2.1 Organic Richness...... 65 5.2.2 Thermal Maturity...... 66 5.2.3 Organic Matter Type ...... 67 5.3 Mature Samples after HP – Results of SRA ...... 67 5.4 Hydrous Pyrolysis Experiments...... 67 5.4.1 Thermal maturity ...... 79 5.4.2 Gas Chromatograms of Expelled Oils...... 79 5.4.3 Gas Yields & Their Bulk Molecular Composition...... 86 5.4.3.1 The Bulk Yields...... 86 5.4.3.2 Hydrocarbon Gas Compositions and Gas Dryness ...... 92

5.4.3.3 CO2 and H2S...... 93 5.4.4 Non-hydrocarbon Gases ...... 93 5.4.5 Eh & pH of the Recovered Water...... 97 5.4.6 Isotopic Composition of Natural Gas...... 98 CHAPTER 6 DISCUSSION...... 104 6.1 Mass Balance ...... 104 6.2 Kerogen Cracking Products and Products Yield ...... 105 6.2.1 Hydrocarbon Gas Yields in Jonah Field & Southern Piceance Basin ...... 105 6.2.2 Bulk Compositions of Hydrocarbon Gases ...... 109 6.2.3 Controls on Gas Generation Products and Their Yields ...... 111 6.3 Non-hydrocarbon Gases...... 111 6.4 Eh & pH of the Recovered Water...... 123 6.5 Gas: Oil Ratio (GOR) ...... 124 6.6 The Relationship among Stable Isotopes, Kerogen Types, and Thermal Maturity...125 6.6.1 Empirical Diagrams ...... 139

6.7 Comparison to Other Published Hydrous Pyrolysis Gas Data...... 144 6.8 Comparison of Gas Isotopes Compositions in the Field and Laboratory Gases...... 147 CHAPTER 7 CONCLUSIONS ...... 149 CHAPTER 8 FUTURE WORK...... 152 REFERENCES ...... 153 CD-ROM...... 1 APPENDIX 4.1 Detailed Composite Description of Each Mancos Shale Core APPENDIX 4.2 Core Descriptions of the Mancos and Mowry Cores APPENDIX 4.3 Detailed Composite Description of Cameo Coal Samples APPENDIX 4.4 Mean Vitrinite Reflectance Measurement APPENDIX 4.5 Molecular Component Calculation for Each Hydrous Pyrolysis Experiment APPENDIX 5.1 SRA Pyrograms of Cameo Coal, Mowry, Mancos, and Baxter Shales APPENDIX 5.2 Vitrinite Reflectance Measurement of the Cameo Coal Chips

vii

LIST OF FIGURES

Figure 2.1 Maturation processes of sedimentary organic matters: Diagenesis (0-50°C), Catagenesis (50-200°C), and Metagenesis (>200°C) ...... 5

Figure 2.2 Van Krevelen diagram - Plot of H/C and O/C atomic ratio during thermal maturation ...... 7

Figure 2.3 The generation and yield of gas with increasing temperature, from sapropelic source and humic source...... 8

Figure 2.4 Schematic chemical structures of oil-generating sapropelic kerogen and generating humic kerogen...... 9

13 13 Figure 2.5 The δ C2-δ C3 versus ln(C2/C3) diagram shows genetic fractionation for thermogenic gases. Different trends of gases generated by primary cracking of kerogen (sub-vertical trend) from gases generated by secondary cracking of oils (sub-horizontal trend) ...... 17

Figure 2.6 Molecular proportions C2/C3 versus C1/C2, in logarithmic scales. (a) From pyrolysis experiments of Type II and Type III kerogen in a closed system; (b, c) same plot for natural gases of Angola and Kansas...... 18

13 13 Figure 2.7 The δ C1-δ C2 versus ln (C1/C2) diagram illustrates differentiation of maturity trends, mixing trends with biogenic methane and leakage trends..18

Figure 2.8 Distinction of thermogenic and bacterial origin of natural gas, combining the use of carbon isotopes and noble gases...... 19

13 Figure 2.9 The C2/C1 versus δ C1 diagram from model calculation of mixing and diffusion trends, with linear (a) and logarithmic (b) scales ...... 19

Figure 2.10 Diagram C2/C3 versus C2/iC4, distinguishing a maturity trend from a biodegradation trend...... 20

13 13 Figure 2.11 The δ C2- δ C3 versus C2/C3 diagram, ), with two examples of gas series from the Ceara Basin, Brazil and from Thailand...... 22

Figure 2.12 Gas generation temperatures predicted from the carbon isotope ratios of methane, ethane, and propane for gas derived from Type II kerogen...... 24

Figure 3.1 The paleogeography of the Piceance Basin in the Cretaceous...... 27

Figure 3.2 Generalized map of the Laramide tectonic elements in eastern Utah and western Colorado and Western Interior Seaway...... 28

viii

Figure 3.3 The stratigraphic column of the Piceance basin, from the Late Cretaceous through Tertiary ...... 29

Figure 3.4 Previous published and proposed nomenclature of the Cretaceous units in northwestern Colorado, based on biostratigraphy ...... 32

Figure 3.5 Contour map of thermal maturity values for the top of the Mancos Shale ...... 33

Figure 3.6 West-east cross section showing distribution of depositional facies and major coal zones in Mesaverde Group, southern Piceance Basin, Colorado.36

Figure 3.7 Vitrinite isoreflectance (Ro) trends at the base of the Cretaceous Mesaverde Group and the top of the Mancos Shale Group...... 36

Figure 3.8 Location of Jonah Field...... 38

Figure 3.9 Map of Jonah field showing the principal faults and the small anticlinal and synclinal folds ...... 38

Figure 3.10 Generalized stratigraphic column of the southwestern Wyoming ...... 39

Figure 3.11 Diagram relating the stratigraphic units in the Cretaceous, timing of thrust Movement, and thicknesses of stratigraphic in the foreland basin of Idaho, Wyoming and Utah...... 40

Figure 3.12 Contour map of the thermal maturity (vitrinite reflectance) at (a) the base of the Rock Springs Formation, (b) the top of the Mesaverde Group, and (c) the base of the Fort Union Formation...... 43

Figure 3.13 Map of the geographical extent of Hilliard-Baxter-Mancos shale in southwestern Wyoming and thermal maturities at the top and base of the Hilliard-Baxter-Mancos Total Petroleum System ...... 45

Figure 4.1 (a) The location of the two Mancos Shale cores in the DCA ; (b) The location the Mowry Shale core in the Rock Spring Uplift...... 52

Figure 4.2 Location Map of the Baxter Shale outcrop. Location: 1/4 NE, section 10, T18N, R104W, Sweetwater County, Wyoming. Three collection location of the outcrop Baxter Shale samples...... 53

Figure 4.3 McClane Canyon Coal Mine and the location map of the eight sample

locations along Coal Canyon...... 54

Figure 4.4 Typical Weatherford Laboratories Instruments Division’s SRA-TOCTM output...... 55

Figure 4.5 Hydrous pyrolysis equipment, setup, and temperature meter and controller. Electric heater housed in stainless steel and thermocouples covered near reactor housing with Marinite-I liners ...... 61

Figure 4.6 Illustration of gas collection system...... 61

Figure 4.7 Vitrinite (left) and inertinite (right) of the Cameo Coal chips under microscope ...... 62

Figure 5.1 Diagrams of TOC, S1, S2, and S3 versus depth (ft) in the two Mancos Shale cores ...... 73

Figure 5.2 Residual petroleum potential S2 versus TOC content for recognition of kerogen type for Mowry, Mancos, and Baxter Shales (A) and Cameo Coals (B) present in different scales...... 74

Figure 5.3 SRA hydrogen index versus Tmax temperature for kerogens from original source rocks...... 75

Figure 5.4 Psuedo-van Krevelen diagram for kerogens in the original source rocks...... 75

Figure 5.5 Diagrams of Tmax versus HI for the Cameo Coal, Mowry, Mancos, and Baxter Shales before and after sequential hydrous pyrolysis experiments...76

Figure 5.6 TOC versus HI diagrams for the Cameo Coal, Mowry, Mancos, and Baxter Shales. In general, both TOC and S2 decrease with increasing maturity. Arrows indicate maturation direction ...... 78

Figure 5.7 Gas chromatograms of generated oil/wax from Cameo Coal after sequential hydrous pyrolysis of 300/72, 330/72, 360/72 ...... 81

Figure 5.8 Gas chromatograms of generated oil/wax from Mowry Shale after sequential hydrous pyrolysis of 300/72, 330/72, 360/72 ...... 82

Figure 5.9 Gas chromatograms of generated oil/wax from Mancos Shale after sequential hydrous pyrolysis of 300/72, 330/72, 360/72 ...... 83

Figure 5.10 Gas chromatograms of generated oil/wax from Hilliard Shale after sequential hydrous pyrolysis of 300/72, 330/72, 360/72 ...... 84

Figure 5.11 Diagram of Pristane/ nC17 and Phytane/ nC18 for expelled oils generated

from sequential hydrous pyrolysis of Cameo Coal, Mowry, Mancos, and Baxter Shales. These two ratios can be used to infer oxidation/reduction and organic matter type in the source rocks...... 85

Figure 5.12 Cumulative yields of hydrocarbon gases of methane, ethane, propane, and butane from the four different source rocks- Cameo Coal, Mowry Shale, Mancos Shale, and Baxter Shale, on a per gram of TOC basis...... 90

Figure 5.13 Instantaneous yields of hydrocarbon gases of methane, ethane, propane, and butane from the four different source rocks- Cameo Coal, Mowry Shale, Mancos Shale, and Baxter Shale, on a per gram of TOC basis...... 91

Figure 5.14 Cumulative yields of hydrocarbon gases of methane, ethane, propane, and butane in Cameo Coal, Mowry, Mancos, and Baxter Shale, on a per gram of TOC basis. Gas: C1, C2 , C3 , C4 , and C5 ...... 94

Figure 5.15 Change of dryness of generated gases with increasing thermal maturity ...... 95

Figure 5.17 Cumulative yields of non-hydrocarbon gases (CO2, H2S, H2, and N2) in sequential hydrous pyrolysis experiments from Cameo Coal, Mowry Shale,Mancos Shale, and Baxter Shale, on a per gram of TOC basis...... 101

Figure 5.18 The cumulative δ13C values of the carbon dioxide generated from the 300 °C, 330 °C and 360 °C experiments from four source rocks- Cameo Coal, Mowry, Mancos, and Baxter Shale ...... 102

Figure 5.19 The cumulative δ13C values of methane, ethane, and propane generated from hydrous pyrolysis experiments versus the reciprocal of their carbon number ...... 102

Figure 6.1 Burial-history plot of SHB 13-27. Plot shows hydrocarbon generation windows for the Hilliard Shale through the Fort Union Formation...... 107

Figure 6.2 Timing of oil and gas generation from Type II and Type III source rocks in Jonah Field. Vertical dashed line at 5 Ma represents beginning of major uplift, erosion, and subsequent cooling...... 109

13 Figure 6.3 a) Cumulative stable carbon isotope of carbon dioxide (δ C- CO2) generated from sequential hydrous pyrolysis experiments in Cameo Coal, Mowry, Mancos, and Baxter Shales. (b) Instantaneous CO2 yields versus

its instantaneous carbon isotope change in the Mancos and Baxter Shales. (c) instantaneous CO2 yields versus its instantaneous carbon isotope change in the Cameo Coal and Mowry Shale ...... 116

Figure 6.5 X-Ray Diffraction (XRD) result of the Baxter Shale before (lower) and after (upper) sequential hydrous pyrolysis. The positive carbon isotope excursion near Cenomanian-Turonian (C/T) boundary was due to an increase in organic carbon burial rate during Cretaceous and drawdown in atmospheric pCO2 ...... 119

Figure 6.6 Carbon isotopic excursion in whole-rock carbonate (δcarb) and total organic carbon (δorg) at the Cretaceous Cenomanian/Turonian boundary...... 119

Figure 6.7 Eh-pH Diagram of the recovered water from sequential hydrous pyrolysis experiments on the Cameo Coal, Mowry, Mancos, and Baxter Shales...... 124

Figure 6.8 Cumulative gas: oil ratios (GOR) for the Cameo Coal, Mowry Shale, Mancos Shale, and Baxter Shale subjected to hydrous pyrolysis temperatures from 300 to 330 to 360°C for 72 h each...... 127

Figure 6.9 Cumulative gas: oil ratios (GOR) for the Cameo Coal, Mowry Shale, Mancos Shale, and Baxter Shale versus yields of expelled oil from sequential hydrous pyrolysis temperatures from 300 to 330 to 360°C for 72 h each ...... 127

Figure 6.10 Instantaneous yield of methane (a), ethane (b), and propane (c) versus instantaneous carbon isotope ratio change of methane, ethane, and propane from the Cameo Coal, Mowry, Mancos, and Baxter Shales. With increasing carbon number (from C1 to C3), the range of carbon isotopic values from a single source narrows ...... 131

Figure 6.11 Natural gas plots of gases from the three samples at selected maturity levels ...... 132

Figure 6.12 Natural gas plots of the Xujiahe Formation reservoired gases in the Sichuan Basin...... 133

Figure 6.13 Natural gas plots for gases generated from anhydrous pyrolysis of oil-prone source rocks...... 133

Figure 6.14 The δ13C isotopic values of methane, ethane, propane and n-butane for by using the molecular ratio C1/ (C2+C3) and carbon isotope ratios of

xii

13 methane δ C-CH4...... 134

Figure 6.15 Natural gas plots for hydrocarbons generated from various source rocks pyrolyzed under hydrous conditions at 330°C for 3 days...... 135

Figure 6.17 The distribution of stable carbon isotope composition of natural gases (C1 to C5) from oil-prone source rocks (open bars) and Jurassic coal measures (solid bars) in the Tarim Basin...... 138

Figure 6.18 The empirical Bernard diagram is used to classify natural gases in the field hydrothermal and anhydrous experiments of Smackover oil ...... 141

Figure 6.19 Diagram of the molecular proportion of C2/C3 versus C1/C2, in logarithmic scales. Data are from gases generated from the Cameo Coal, Mowry Shale, Mancos Shale, and Baxter Shale by experimental hydrous pyrolysis...... 142

13 13 13 Figure 6.20 Cross plot of δ C1 vs. δ C2–δ C3 for gases from kerogen and oil cracking...... 142

13 13 Figure 6.21 The δ C2- δ C3 versus C2/C3 diagram) of gases generated from primary cracking of the Cameo Coal (blue), Mowry (red), Mancos (green), and Baxter Shales (purple) in hydrous pyrolysis experiments ...... 143

Figure 6.22 The diagram C2/C3 versus C2/iC4, distinguishes a maturity trend from a biodegradation trend. Data are from gases generated from the Cameo Coal,Mowry Shale, Mancos Shale, and Baxter Shale in hydrous pyrolysis experiments ...... 144

Figure 6.23 All published gas isotope data on different source rocks by hydrous pyrolysis for Type II kerogen and Type III kerogen. Gas generated from secondary cracking of Smackover oil is also included in the plot for comparison, the linear trend in purple ...... 145

Figure 6.24 Gases generated from the Type III kerogen (mainly coals). Different under different heating programs and final temperatures in different 13 coals laboratories all show consistent values of δ C on C1 to C3 ...... 146

Figure 6.25 Gases generated from the Type II kerogen (mainly marine shales). Mancos,Mowry, and Polish Type II kerogen show consistent δ13C values on C1 to C3, but the Woodford, Baxter and Kimmeridge Shales 13 show distinct δ C values on C2 to C3...... 146

xiii

Figure 6.26 The cumulative δ13C of methane to propane in the Jonah Field. Field gas is the dashed line and lab gas is the solid line...... 148

Figure 6.27 The cumulative δ13C of methane to propane in the southern Piceance Basin. Field gas is the dashed line and lab gas is the solid line...... 148

xiv

LIST OF TABLES

Table 4.1 Detailed core information of the two Mancos Shale cores and one Mowry shale core in Denver, Colorado. The two Mancos cores from the Douglas Creek Arch area are 18.8 miles apart from each other ...... 51

Table 4.2 Calculation of volume of water to use for hydrous pyrolysis experiments ...... 60

Table 5.1 Results of SRA data for original unheated Cameo Coal outcrop samples. The calculated percentage before composite is based on the relative thickness of exposed outcrops ...... 68

Table 5.2 Results of SRA data for original unheated Mowry Shale core samples. Different facies were identified ...... 69

Table 5.3 Results of SRA data for original unheated Mancos Shale core samples. Well 6-15-1A-103 Federal has depth ranged from 3,392 to 3,404 ft. Well 15-29-4-103 Gentry has depth ranged from 3,358 to 3,569.5 ft...... 70

Table 5.4 Results of Leco TOC data for original unheated Baxter Shale outcrop samples. Data is provided by USGS...... 71

Table 5.5 Average geochemical parameters of the Upper Cretaceous Cameo Coal, Baxter, Mancos, and Mowry Shales from the Weatherford Source Rock Analyzer (SRA) ...... 71

Table 5.6 Measured vitrinite reflectance data of original source rock samples (Weatherford Laboratories, Texas)...... 72

Table 5.7 Comparisons of SRA data before and after compositing samples and sequential hydrous pyrolysis...... 77

Table 5.8 Measured vitrinite reflectance data of Cameo Coal chips representing equivalent sequential experimental conditions ...... 79

Table 5.9 Isoprenoid/ n-alkane ratios and odd-to-even predominance index (OEP) of experiment-expelled free oils (from Cameo Coal) and equipment rinses (from Mowry, Mancos, and Baxter Shales)...... 85

Table 5.10 The normalized mole percentage of generated gas from each sequential experiment on source rocks ...... 87

Table 5.11 Incremental yield of hydrocarbon gases (methane, ethane, propane, butane, and pentane) in sequential hydrous pyrolysis experiments...... 88

Table 5.12 Cumulative yield of hydrocarbon gases (methane, ethane, propane, butane, and pentane) in sequential hydrous pyrolysis experiments...... 89

Table 5.13 The roughly estimated CO2 and H2S gas dissolved in the recovered water after the sequential experiment ...... 99

Table 5.14 Instantaneous yield of non-hydrocarbon gases (CO2, H2S, H2, and N2) in sequential hydrous pyrolysis experiment...... 100

Table 5.15 Cumulative yield of non-hydrocarbon gases (CO2, H2S, H2, and N2) in sequential hydrous pyrolysis experiment...... 100

Table 5.16 Stable carbon isotopes of gases generated by sequential hydrous pyrolysis .103

Table 6.1 Recovery percentage of Cameo Coal, Mowry, Mancos, and Baxter Shales after sequential hydrous pyrolysis at 300, 330, and 360°C for 72 h...... 104

Table 6.3 Cumulative yields after sequential HP Experiments (unit: mole %)...... 112

Table 6.4 Measured Eh, pH, and temperature from the recovered water from sequential hydrous pyrolysis experiment on Cameo Coal, Mowry, Mancos, and Baxter Shales...... 124

xvi

LIST OF ABBREVIATIONS

CBM……Coal Bed Methane

EIA……..U.S. Energy Information Administration

HI……….Hydrogen Index

HP………Hydrous Pyrolysis

LK………Lowry Cretaceous

OEP……. Odd-to-Even Predominance index

OI……….Oxygen Index

RPSEA….Research Partnership to Secure Energy for America

SRA……..Source Rock Analyzer

TOC……..Total Organic Carbon

XRD…….X-ray Diffraction Analysis

xvii

ACKNOWLEDGEMENTS

This study would not have been possible without funding from RPSEA. I would like to first thank the institute for support of this project and my study at Colorado School of Mines.

To my advisor, Dr. Nicholas B. Harris, and my co-advisor, Dr. John B. Curtis for their guidance, patience, support, and time through this long process. Your insight into the subject and understanding of the project itself, have greatly improved my knowledge of petroleum geology and geochemistry. I would like to express my sincere gratitude to Dr. Michael D. Lewan for spending enormous amounts of his time, teaching me the hydrous pyrolysis experiment from the very beginning. Your involvement, guidance, patience, and knowledge are greatly appreciated. Thank you to my committee member Dr. John Humphrey for your instruction about stable isotopes and your reviews of my thesis.

For sample collection, I would like to thank Dr. Donna Anderson for help finding suitable Mancos Shale samples, and Dr. Paul Lillis and Dr. Mark Kirschbaum for the locations of several outcrops of the Cameo Coal. Thank you to all the USGS-CRC (Core Research Center) staff for cutting and sampling the cores for me. For sample analyses, I would like to thank Dr. Richard F. Wendlandt for doing the XRD analyses. Huge thanks to many people from the U.S. Geological Survey Energy Resource Lab: Augusta Warden, Mark Pawlewica, Robert Dias, Mark Drier, and Zachary Lowry, for performing the oil composition, gas composition and gas isotope analyses.

I also would like to thank all the faculty members at the CSM Geology & Geological Engineering Department for their instruction and giving me strong foundations in fundamental geoscience. Thank you to my CSM colleagues and friends for encouragement, discussions, and all kinds of help and backing. Special thanks to Sheven Poole, James Golab, and Jon Jay Charzynski for helping me correct my English grammar and writing.

Nothing can be completed without our very kind secretaries Debbie Cockburn and Marylin Schwinger. Thank you for so much assistance through these years. Finally, I would like to thank my friends and family in Taiwan and all my Taiwanese friends here in the U.S. Thank you for believing me and always being so supportive and encouraging.

xviii

CHAPTER 1

INTRODUCTION

1.1 Unconventional Tight Gas Resources The potential of unconventional gas resources - tight gas, coal bed methane (CBM), and shale gas - has raised predictions of higher U.S. gas production through at least 2020 (RPSEA, 2010). The tight-gas-sand reservoirs have accounted for the majority of the unconventional gas production in the United States for over 20 years (EIA Annual Energy Outlook, 2005). Even today, tight-gas reservoirs represent a growing and indispensible gas resource for indigenous U.S. gas supply. The western Rocky Mountain region has been the main production area for tight gas reservoirs over the past 15 years and makes up 20 % of the U.S. total natural gas reserves (Kuuskraa, 2009). The Rocky Mountain area now produces more than 50 % of U.S. total gas (EIA Annual Energy Outlook, 2009). The greatest potential for unconventional gases in the Rocky Mountain region lies in southwestern and south-central Wyoming, northwestern Colorado, and northeastern Utah (Potential Gas Agency, 2008). This large area consists mainly of three basins: the Greater Green River Basin in Wyoming, the Piceance Basin in Colorado, and the Uinta Basin in Utah. The technically recoverable resources in these three basins are about 26, 41, and 111 TCF, respectively (Potential Gas Agency, 2008). The major tight-gas fields in these three basins include Pinedale Anticline (ranked 3rd in proven gas reserves in the U.S., 2008, EIA), Jonah Field (ranked 8th) in the Greater Green River Basin; Mamm Creek (ranked 23rd) - Rulison (ranked 29th) – Parachute (ranked 22nd) – Grand Valley (ranked 11th), a cluster of fields in the Piceance Basin and the Greater Natural Buttes Field (ranked 5th) in the Uinta Basin (EIA, 2008 annual proven reserves summary). The sandstone reservoirs in these tight-gas fields typically have very low porosity and permeability, typically < 10% and < 0.1 mD, so understanding their unique petrophysical properties and applying hydraulic fracture stimulation are both essential to boost productivity (Shanley et al., 2004; Federal Energy Regulatory Commission, 2009). The regional tectonics, stratigraphy, sedimentation, eustacy, and paleogeography regarding the Rocky Mountain foreland basin and the Cretaceous Western Interior

Seaway in Colorado, Utah, and Wyoming have been well studied. However, details are sparse relating to the source rocks, with little quantitative information on their generation potential, bulk and isotopic compositional signatures of their generative gases. Recognizing differences between sources and reservoirs in isotope composition of gases can help understand migration processes. Evidence for vertical migration of gas within reservoirs includes carbon isotopic signatures of gases. Vertical migration of gas has been recognized in studies of variations in carbon isotopes in the northern, central, and southern part of the Piceance Basin (Johnson and Rice, 1993; Johnson and Roberts, 2003). Johnson and Rice (1993) discovered that gases in the Molina Member of the Wasatch Formation (Rulison and Grand Valley fields) are isotopically indistinguishable from gases in the underlying Mesaverde Formation. In the central part of the Piceance Basin (the Piceance Creek Dome field), gas from a sandstone reservoir in the lowermost part of the Eocene Green River Formation also appears to have identical isotopes to gases in the underlying Mesaverde Formation (Johnson and Roberts, 2003). However, developing useful gas migration models for the Rocky Mountain basins based on gas compositions cannot be done without knowing what formations are the sources of gas. Several candidates for gas-generating source rocks in the Rocky Mountain area have been identified by Law (1984) and Coskey (2004), but the exact source rocks have not been confidently identified. This study reports on laboratory experiments, including hydrous pyrolysis and Rock-Eval pyrolysis that were undertaken in order to characterize the bulk and isotopic compositions of gases generated from potential source rocks in two representative basins in the Rocky Mountain area - Jonah Field, Wyoming and the Piceance Basin, Colorado. These results will be valuable for identifying source rocks for gases in tight gas fields and to understand the possible mixing processes among different sources.

1.2 Research Objectives In order to understand volumes and the bulk and isotopic compositions of gas generated from source rocks in the Piceance Basin and Jonah Field, SRA analysis (Weatherford Source Rock AnalyzerTM) and hydrous pyrolysis experiments were carried out on four possible source formations. In the Rocky Mountain basins, formations

containing Type II marine, Type III terrestrial kerogen, or a mixture, are considered to be significant sources of gas by Coskey (2004) and Yurewicz et al. (2003). Those studies suggested primary source rocks in the Piceance Basin as the Cameo Coal, Mesaverde Group (Type III kerogen), and the Mancos Shale (Type II kerogen), and in the Jonah Field area, carbonaceous shales and coals of the Lance Formation and Mesaverde Group (Type III kerogen) and the Mowry and Hilliard Shale (Type II kerogen) (Coskey, 2004; Yurewicz et al., 2003). Experiments on candidate source rocks in the Piceance and Green River Basin were used to develop relationships between the gas volumes, the bulk and isotopic composition of gases from these source rocks and the thermal maturity, and to understand:

1. Whether there is a maturity effect on the isotopic compositions of gases. If so, what is the relationship between the maturity and isotopic composition of gas evolved from different kerogens?

Previous studies (Schoell, 1983; Chung et al., 1988; Schoell, 1988; Jenden et al., 1988; Jenden et al., 1993; Prinzhofer and Huc, 1995; Prinzhofer and Pernaton, 1997) all identified a strong relationship between the thermal maturity and the isotopic compositions of gases, especially with gases generated from kerogen early in primary cracking. However, recent studies of hydrous pyrolysis gases from Menilite Shale in Poland by Kotarba et al. (2009) indicates that thermal maturity has only subtle effects on the isotopic signatures of gases. Whether or not the level of thermal maturity affects the isotopic compositions of gases is the most critical question addressed in this study. A related question is whether, at a given maturity level, different types of kerogen produce gases of different isotopic composition? If both maturity and kerogen types affect the isotopic compositions of gases, which one has a greater impact on variations of stable isotopes in natural gases?

2. Gas generated from coals has CO2 and N2 as the main non-hydrocarbon

gases. Does the composition of non-hydrocarbon gases provide information about source rocks?

Sapropelic and humic organic matter generates different amounts and proportions of non-hydrocarbon gases. Comparison between laboratory-produced non-hydrocarbon gases and field-collected non-hydrocarbon gases may help indirectly interpret the source rocks. In addition, splits of the nonhydrocarbon gases generated from hydrous pyrolysis experiments were sent to Dr. Chris Ballentine for noble gas analysis. Gases of interests here include helium (He), neon (Ne), and argon (Ar). Because noble gases are inert and conservative, they can act as tracers for sources of gas components. However, the analyses and interpretations are not part of this thesis research. The next chapter will present the fundamental gas generation and isotope geochemistry concepts required for this study.

CHAPTER 2

PREVIOUS RESEARCH

2.1 The Process of Natural Gas Generation, from a Molecular Standpoint The evolution of sedimentary organic materials during burial and thermal maturation can be divided into diagenesis, catagenesis, and metagenesis (Figure 2.1) (Tissot and Welte, 1978). Distinctions between these stages are based both on maturation (temperature), time, and chemical processes.

Figure 2.1 Maturation processes of sedimentary organic matters: Diagenesis (0-50°C), Catagenesis (50-200°C), and Metagenesis (>200°C) (Modified from Tissot & Welte, 1978).

During diagenesis (Ro < 0.5 %), large geopolymers are altered and degraded from organisms by biological and low-temperature chemical degradation, condensation, and

5 polymerization (Tissot and Welte, 1984; Horsfield and Rullkötter, 1994). The main chemical reactions in this stage involve generation of biogenic gas, conversion of organic matter into kerogen, and the loss of water (Waples, 1984). These reactions are also accompanied by decarboxylation reactions in Type III terrestrial organic matters (Tissot and Welte, 1984; Hunt, 1996). During catagenesis (0.5% < Ro < 2.0%), the initial and principal stage of oil and gas formation, thermally-induced reactions transform kerogen into bitumen (Lewan, 1994), followed by bitumen decomposition into oil and thermogenic gas. Reactions are diverse and depend on the type of source material present. Lipid material (Type I and Type II kerogen) evolves relatively large amounts of low molecular weight aliphatic hydrocarbons (Tissot et al., 1978) in comparison to humic material, which tends to evolve smaller amounts of hydrocarbons and heteroatomic compounds (Durand and Espitalie, 1976; Larter and Douglas, 1980; Horsfield and Rullkötter, 1994). Dehydration and decarboxylation reactions continue to play an important role in kerogen and coal maturation (Senftle et al., 1986; Larter, 1989). Towards the end of catagenesis, kerogens become progressively hydrogen-deficient and depleted in volatile components (Tissot et al., 1974; Horsfield and Rullkötter, 1994). During metagenesis (2.0% < Ro < 4.0%), intensive thermal alteration of kerogen, bitumen, and petroleum occurs. Kerogen becomes thermally unstable, and only methane, hydrogen, and highly carbonized solid organic matter are stable (Horsfield and Rullkötter, 1994). In metagenesis, kerogen reorganizes into an almost pure carbon structure (Vandenbroucke, 2003). Instead of going through substantial condensation of aromatic rings, kerogen went through a parallelization of existing aromatic structures and transferred condensed rings into stacks (Oberlin et al., 1980). Further molecular structure arrangement of the residues causes slow loss of hydrogen, probably evolved as methane, leaving pyrobitumen as end products (Tissot and Welte, 1978; Durand, 1980). The van Krevelen-type diagram (van Krevelen, 1961, 1984; Stach et al., 1982) uses plot of H/C and O/C atomic ratios to describe the kerogen evolution in three maturation stages: diagenesis, catagenesis, and metagenesis (Figure 2.2) (Tissot et al., 1974; Tissot & Welte, 1984; Peters, 1986). During diagenesis, kerogen undergoes a significant decrease in oxygen and hydrogen and a correlative increase in carbon content, producing large quantities of CO2 and H2O, but very little or no thermogenic

hydrocarbons (Tissot & Welte, 1978; Vandenbroucke, 2003). In catagenesis, kerogen loses significant amount of hydrogen (decrease in H/C ratio) due to the generation and release of hydrocarbons. Aromatization of kerogen also increases during catagenesis (Tissot & Welte, 1984; Peters and Cassa, 1994). For all kerogens, the transition between catagenesis and metagenesis occurs when the atomic H/C ratio decreases to around 0.6 (Vandenbroucke, 2003). During metagenesis, only dry gas (methane) is generated from both remaining kerogens (primary cracking) and from previously generated liquid hydrocarbons (secondary cracking). An H/C ratio of 0.25 is typical of the kerogen at the end of metagenesis and the beginning of metamorphism (Hunt, 1996).

Figure 2.2 Van Krevelen diagram - plot of H/C and O/C atomic ratio during thermal maturation (Modified from Tissot & Welte, 1978, p.143; copyright Springer-Verlag)

Bacterial dry gas is generated at relatively low temperatures and shallow depths. After that, the thermal degradation of kerogen generates increasing amount of wet gas

and methane in oil window during catagenesis, and finally only dry gas at the end of catagenesis in most sedimentary basins (Tissot and Welte, 1978; Hunt, 1996) (Figure 2.3). The molecular structure change in kerogen with different thermal stresses, along with compositional change of generated gas, is described below. Kerogen structure determines its petroleum generation products because different bonds require different thermal energy to break the gas molecular precursors from the kerogen. Basic differences in the chemical structure of sapropelic (Type I and Type II) and humic (Type III and IV) kerogen are illustrated in Figure 2.4. Sapropelic kerogen is much richer in hydrogen than humic kerogen and is enriched in aliphatic structures and isolated ring structures (Behar and Vandenbroucke, 1987). Humic gas-generating kerogen contains only a few short chains with some single methyl groups and a large number of condensed rings with oxygenated functional groups (Behar and Vandenbroucke, 1987). In other words, the humic organic matter is characterized by an abundance of aromatic rings and heteroatoms.

Figure 2.3 The generation and yield of gas with increasing temperature, from sapropelic + source and humic source. The C2H6 represents hydrocarbon gases heavier than methane. The N2 is generated initially as NH3. The CO2 from organic source peaks at about 100°C. Humic organic matters generate more organic CO2 (Modified from Hunt, 1996).

(A)

Figure 2.4 Schematic chemical structures of oil-generating sapropelic kerogen and gas- generating humic kerogens (Hunt, 1996). (B)

For oil-prone source rocks, as thermal stress increases in the system, kerogen decomposes into bitumen first, involving cleavage of weak non-covalent bonds. Petrographic studies of rock from laboratory pyrolysis show a continuous organic network within the pore space established by expansion of bitumen into the pore system (Lewan, 1987). As thermal stress increases further, bitumen partially decomposes into oil. During this stage, kerogen content does not change significantly, but the bitumen content decreases. The overall maturation reaction of kerogen involves cleavage of covalent bonds (Lewan, 1985, 1989; Huizinga et al., 1988). When the carbon chains break off by β-scission to form liquid hydrocarbons, these highly aliphatic fragments are hydrophobic and will separate from the water-saturated bitumen phase as an immiscible oil phase (Lewan, 1994). The process in which kerogen decomposes into oil and gas is called primary cracking. During primary cracking, wet gas components, C2+ (C2 to C7), are formed in

concentrations of parts per billion (ppb) during catagenesis by low-temperature carbonium ion or free-radical reactions (Kissin, 1987). Carbonium-ion mechanisms

(those involving positively charged intermediaries) would be expected to produce dominantly branched products, whereas free-radical mechanisms (those involving neutrally charged intermediaries) would be expected to favor straight-chain products (Kissin, 1987; Seewald, 2003). The predominance of straight-chain alkanes in petroleum supports the free-radical mechanism, but the carbonium-ion mechanism may be important for the generation of branched low-molecular-weight compounds (Hunt, 1984; Seewald, 2003). Because its unique molecular structure, the major product of humic kerogen is

methane, and it can only form a small amount of C2+ gas. The cleavage of C-C bonds in sapropelic kerogens requires high energy, whereas the process of condensation of aromatic rings requires lower energy. If the mechanisms of gas formation from humic and sapropelic organic matters are correctly described, the isotopic dependence in methane with thermal maturity of humic organic matter would differ from that of sapropelic organic matter. Researchers have applied different pyrolysis methods to a variety of organic sources in order to develop models for the timing and quantity of natural gas generation in sedimentary basins. Laboratory pyrolysis experiments allows simulation of methane generating during primary cracking (Espitalie et al., 1987; Berner et al, 1995; Tang et al., 1996; Behar et al, 1997; Knauss et al., 1997; Cramer et al., 1998; Henry and Lewan, 1999) and secondary cracking (Pepper and Dodd, 1995; Behar et al, 1997; Horsfield et al., 1997; Schenk et al., 1997; Tsuzuki et al., 1999; Henry and Lewan, 1999; Waples, 2000; Hill et al., 2003; Tang and Schoell, 2005; Tang et al., 2005; Tian et al., 2008). In these studies, the calculated activation energy for the methane generation from Type II kerogens ranges from 126 to 262 KJ/mol (Lillack et al, 1991), suggesting the methane may be generated from primary cracking of kerogen over the range of 50°C to over 170°C, whereas activation energy distributions of Type III kerogen show a range from 167 KJ/mol to 352 KJ/mol, with peak generation potential for activation energy of about 245 KJ/mol (Cramer et al., 1998). The different distributions of activation energies of humic and sapropelic organic matter probably results from the different structures involved in methane formation (Galimov, 1988, 1989).

For primary cracking, a broad range of calculated activation energies (Ea) and frequency factors (A) in the Arrhenius equation is available in the literature. These calculations are done through mathematical optimization of data from experiments performed under different experimental conditions such as isothermal vs. non-isothermal, open-system vs. close-system, and hydrous vs. anhydrous pyrolysis. A temperature- dependent kinetic isotope fractionation concept was developed by Cramer et al. (1998) using activation energy in reactions involving 12C- and 13C-methane generation, which differed by ∆Ea = -102 J/mol. They applied temperature-dependent kinetic isotope fractionation from a set of parallel first-order reactions and found good match of carbon isotopic data between open-system pyrolysis experiment and model calculations. Tang et al. (2000) incorporated the relative cracking rates of isotopically substituted and un- substituted bonds to model the stable carbon isotope ratios in natural gases, based on both quantum chemistry calculation and isothermal closed-system data. With their model, the isotopic compositions of natural gases not only can be used to estimate the timing of gas generation but also to evaluate the maturity of source rocks in sedimentary basins. The instantaneous and cumulative gas accumulations can also be predicted from their model. Different pyrolysis techniques yield different kinetic parameters (Lewan and Ruble, 2002). Lewan and Ruble (2002) conducted a comparative study between isothermal hydrous pyrolysis and nonisothermal open-system pyrolysis on six well- characterized source rocks by using aliquots of the same source-rock samples. They concluded that isothermal hydrous pyrolysis gave better geologically controlled kinetic parameters and that the generation of expelled oil can be described by a single overall activation energy and frequency factor (Lewan and Ruble, 2002), in contrast to data from nonisothermal open-system pyrolysis, where data are best fit by a range of activation energies. More attention has been paid to hydrocarbon gases from secondary cracking. Data from field, laboratory, and theoretical calculations (Schenk, 1997; Domine, 1998; Waples, 2000) suggests that oil in reservoirs may be stable up to 200°C. Tian et al. (2008) investigated the yield of gas cracked from oil, the kinetic modeling of secondary cracking, and volume change of gas at various temperatures and pressures, based on new data on thermal stability of oil. (For secondary cracking, the assumption that cracking of

oil to gas follows first-order kinetics is frequently applied.) They proposed that a crude oil begins to crack at about 160°C and is completely cracked to gas at about 210°C, assuming a geological heating rate of 2°C/m.y (Tian et al., 2008). This new kinetic model will influence estimations of oil preservation in reservoirs. The application of carbon and hydrogen isotopic fractionation with respect to the thermal maturation of hydrocarbons is discussed in a later section.

2.2 Sources of Natural Gas and Their Genetic Isotopic Signatures Isotopes are atoms that have the same number of protons and electrons but different numbers of neutrons. Stable isotopes are nuclei that do not appear to decay to other isotopes on geologic timescales. The standard usage for stable hydrogen and carbon isotope is described here.

1 2 1 Hydrogen has two isotopes: 1 H and 1 H (deuterium, or D). The abundance of 1 H

2 (mass: 1.007825 amu) is 99.985% of the total hydrogen, whereas the abundance of 1 H (mass: 2.0140 amu) is 0.015 % (Lide and Frederikse, 1995). The masses of heaviest

2 isotopes of hydrogen differ substantially from the lightest isotopes: 1 H is 99.8 % heavier

1 than 1 H . The isotopic composition of hydrogen is expressed by the δD parameter, defined as: 2 D 2 D ( 1 ) spl − ( 1 ) std δD = [ H H ]×10 3 ‰ 2 D ( ) 1H std The standard for the isotopic composition of hydrogen is standard mean ocean water (SMOW). The difference in the masses of the isotopes affects the strength of covalent bonds that hydrogen forms with atoms of other elements. The theory of mass-dependent fractionation of the isotopes of an element has been presented by Urey (1947), Bigeleisen (1965), Bottinga and Javoy (1973), Richet et al. (1977), Melander and Sauders (1980), O’Neil (1986), and others. The two stable isotopes of carbon(C) and their abundances and masses are:

12 6 C : 98.90 %, 12.000000 amu

13 6 C : 1.10 %, 13.003355 amu The mass of 13C is 8.36 % greater than that of 12C, which causes the carbon isotopes to be fractionated by natural chemical and biological processes. The isotopic composition of carbon is expressed by the δ13C parameter, defined as: 13C 13C ( 12 ) spl − ( 12 ) std δ13C = [ C C ]×103 ‰ 13C ( ) 12C std where the standard is Pee Dee Belemnite (PDB).

2.2.1 Sources of Natural Gas in the Tight-Gas Reserovirs Natural gases in both Piceance Basin and Jonah Field of the Rocky Mountain area were generally thought to be of thermogenic origin, so only gas generated directly from kerogens or gas generated from oils are discussed below. Thermogenic methane starts to form in small amounts when the generation of shallow bacterial methane diminishes, so that the two processes overlap during burial. The idea that gas is generated in the early stage of the thermal cracking is supported by the evidence of the substantial quantities of methane dissolved in oil (Barker, 1990). Methane can be generated thermally from all types of kerogen throughout the oil window. The initial gas generated at a low thermal maturity (<0.5% Ro) is relatively dry

but the proportion of C2+ hydrocarbons increase with maturation during catagenesis (Hunt, 1996). With increasing thermal stress, long-chained organic compounds cleave off various short-chained compounds to form light hydrocarbons. Sapropelic Type I and Type II kerogen generates significant quantities of thermogenic hydrocarbon gases at temperatures over 70°C and the peak generation of methane occurs at about 150°C (Figure 2.3: Hunt, 1996). Small amounts of wet gas may also form from humic sources, depending on its hydrogen content. Dry thermogenic gas can form from all kerogen types in the high-temperature range from around 150°C to over 200°C (Figure 2.3). Oil cracking is a complicated process that involves many different parallel cracking reactions (Hill et al., 2003). Previous studies using artificial laboratory experiments on secondary cracking all showed that oil generation peaks at about 340°C

to 360°C under laboratory hydrous pyrolysis experiments for 72 hours (Ro = 1.3-1.5%) (Horsfield et al., 1992; Schenk et al., 1997; Wei et al., 2007) and is followed by intense oil cracking, based on the evidence of (1) the absence of generation of oil, (2) the reduced biomarker parameters, and (3) the formation of diamondoids (Dahl et al., 1999 and 2000; Peters et al., 2005; Wei et al., 2007). By extrapolating laboratory kinetic parameters to real geological conditions, the onset of gas generation by oil cracking is inferred to occur between 150°C to 225°C on geological time scales (Horsfield et al., 1992; Schenk et al., 1997; Tian et al., 2001; Waples, 2000; Wei et al., 2007). Different oil types do not affect the cracking rates or kinetics significantly (Waples, 2000); however, saturate and aromatic hydrocarbons requires different activation energies and frequency factors to simulate the cracking reaction (Song et al., 1994; Jackson et al., 1995; Behar and Vandenbroucke, 1996; Dominé et al., 1998; Behar et al., 1999, 2002; Domke et al., 2001; Dartiguelongue et al., 2006; Leininger et al., 2006; Behar et al., 2008). Dry bacterial gas, wet thermogenic gas, and dry thermogenic gas each have different carbon and hydrogen isotopic signatures. Oil and gas are strongly depleted in 13C relative to Pee Dee Belemnite (PDB) because 12C-12C bonds are weaker than the bonds formed by 13C; consequently, the maturation of organic matter at increasing temperature causes the residual kerogen to become enriched in 13C. There is an empirical relationship between δ13C value of methane and the thermal maturity of the source rocks – the δ13C of methane becomes heavier with increasing thermal maturity (Stahl, 1977, 1979; Schoell, 1983, 1988; Chung, 1988; Berner and Faber, 1988; Schoell, 1988; Jenden et al., 1988; Jenden et al., 1993). The δ13C value of bituminous coal varies only between narrow limits and has an average of about -25 ‰ (PDB) (Faure and Mensing, 2005). Peat evolves into lignite and bituminous coal through the progressive loss of hydrogen; the generated methane is enriched in 1H and 12C, relative to the residual organic matter. At the time of its deposition, coal is depleted in 13C because it is derived from vascular plants, which convert atmospheric CO2 into cellulose (Faure and Mensing, 2005). The plant assemblages during the Late Cretaceous were C3 plants (Latorre et al., 1997), which fractionate -27±6 ‰ with respect to atmospheric CO2 (Ehleringer and Cerling, 2002).

2.2.2 Implications of Carbon and Hydrogen Isotopes in Gases The chemical and isotopic compositions of natural gas are valuable for understanding generation and migration processes. Information on the genetic (thermal evolution of kerogen, including diagenesis and catagenesis) and post-genetic (secondary processes after oil or gas was generated) is commonly obtained from stable carbon and

hydrogen isotope ratios of C1- C5 hydrocarbons, especially when combined with bulk hydrocarbon compositional data (Prinzhofer and Huc, 1995). Genetic fractionation of gas is a function of the processes of primary cracking of kerogen and secondary cracking of oil (Prinzhofer and Huc, 1995). The genetic signature has been classified as biogenic, thermogenic, and coal-bed or shale gas, based on the processes involved in gas formation. Post-genetic fractionation involves one or more of the processes that occur after gas formation within the source rock, including migration from source to reservoir, leakage from traps, adsorption-desorption, biodegradation, and oxidation of the hydrocarbons (Prinzhofer and Huc, 1995). Observed gas compositions in nature are usually a mixture of products from different mechanisms and secondary processes. Empirical correlations between stable isotopic gas signatures and their source rocks are supported by significant amounts of field data (Stahl, 1977; Bernard, 1978; James, 1983; Schoell, 1983; Chung et al., 1988; Schoell, 1988; Jenden et al., 1988; Jenden et al., 1993; Whiticar, 1994; Prinzhofer and Huc, 1995; Prinzhofer and Pernaton, 1997). James (1983) was one of the first to demonstrate that carbon isotopes of natural gas components (C1 to C5) can be used to determine gas maturity– the carbon isotope of gas becomes heavier with increasing maturity. The carbon isotopes of natural gas can also be used to correlate gas to its source, reservoir gas with another reservoir gas, and recognize gas mixtures from multiple sources. His approach is useful for recognizing gas generated from primary cracking of kerogen and secondary cracking of oil, which is assigned as “genetic.” Schoell (1983) proposed that the carbon isotope of gas becomes

heavier with increasing thermal maturity. He further used the criteria such as the C2+ concentrations, hydrogen isotope variations in CH4, and carbon isotope variation in C2H6 to characterize secondary processes in natural gas. His assumption that during migration,

C2+ gases do not dissolve in pore waters can apply to gas migration through very tight

rocks such as shales. Because the C2+ gases have very low solubility, a fractionation in the bulk composition of gas would occur. Differences in the carbon isotopic compositions of gases may also reflect primary versus secondary cracking. Prinzhofer and Huc (1995) proposed an approach based on the methane-ethane-propane system to distinguish hydrocarbon gases generated by the primary cracking of kerogens from gases generated by the secondary cracking of oil, by 13 13 plotting δ Ci-δ Cj versus ln(Ci/Cj). In their anhydrous closed-system experiments (under

argon in gold tubes), the ethane/propane (C2/C3) ratio of the gas remained constant or slightly decreased during primary cracking, whereas this ratio increased drastically during secondary cracking of oil (Figure 2.5). In contrast, the methane/ethane (C1/C2) ratio increased progressively during primary cracking but remained almost constant 13 13 during secondary cracking (Figure 2.6). Additionally, by plotting δ C1-δ C2 versus

ln(C1/C2) (Figure 2.7), a thermogenic maturity trend can be determined: with increasing

maturity, the relative proportion of methane in (C1/C2) increases but the difference in 13 13 isotopic ratio decreases (δ C1-δ C2) and approaches zero. Figure 2.8 shows the diagram of (C2/C3 versus C2/iC4), developed to identify a maturity trend by Prinzhofer et al.

(2000). With increasing thermal maturity, the fraction of iC4 decreases more rapidly than that of C3. During biodegradation, fractions of C3 and nC4 are more easily altered than that of iC4 (Prinzhofer et al., 2000). The combination of chemical and isotopic composition of gas can be applied to identification of the secondary processes such as migration, mixing, and leakage. The 13 13 δ C1-δ C2 versus ln(C1/C2) diagram (Figure 2.7), can distinguish a thermogenic maturity trend from a mixing trend (with bacterial gas) (Prinzhofer and Huc, 1995). In the case of mixing thermogenic gas with bacterial gas, which normally contains methane with δ13C less than -55‰, the mixing trend will curve down and deviate from the maturation trend. In the same plot, diffusive leakage from a reservoir can be identified because the leaking gas will be enriched in methane and will be isotopically lighter than 13 the residual gas. Moreover, a cross-plot of C2/C1 versus δ C1 can also be used to identify the bacterial contribution, mixing effects or segregation effects during migration (Figure 2.9: Prinzhofer and Pernaton, 1997). The model calculations of mixing and diffusion 13 (Fick’s law) trends were plotted in this diagram of C2/C1 versus δ C1. Therefore, by

13 measuring or assuming C2/C1 and δ C1 values of the two mixing end members, a diffusive trend during migration or a mixing trend between thermogenic and bacterial gas can be identified. Characterizing the bulk and isotopic composition of natural gas can be applied to characterizing the degree of openness of the system. Prinzhofer et al. (2000) uses the 13 13 (δ C2-δ C3 versus C2/C3) to identify the degree of openness of the system and characterize the maturity of gas (Figure 2.10) (Lorant et al., 1998; Prinzhofer et al., 2000a, 2000b, 2003). Laboratory pyrolysis experiments in closed and open systems provide the possibility to simulate gas generation with two extreme boundary conditions that cannot be found in natural systems (perfectly closed and open systems never exist in natural systems). The degree of opening of the system is equivalent to the residence time of gas. Gases generated under closed system do not have the ability to react and transform themselves after being generated. This study applies a close-system approach in hydrous pyrolysis experiments to simulate the gas generation. Therefore, a closed system trend is expected.

Figure 2.6 Molecular proportions C2/C3 versus C1/C2, in logarithmic scales. (a) From pyrolysis experiments of Type II and Type III kerogen in a closed system, performed by Behar et al.(1991); (b, c) Same plot for natural gases of Angola and Kansas (Prinzhofer and Huc, 1995).

13 13 Figure 2.7 The δ C1-δ C2 versus ln (C1/C2) diagram illustrates differentiation of maturity trends, mixing trends with biogenic methane and leakage trends (Prinzhofer and Huc, 1995).

Figure 2.8 Diagram C2/C3 versus C2/iC4 (Prinzhofer et al., 2000a), distinguishing a maturity trend from a biodegradation trend. Some natural gas examples are shown exhibiting both processes. Field A, gases have a lower maturity than field B, both exhibiting a trend of biodegradation respectively (Prinzhofer and Battani, 2003).

13 Figure 2.9 The C2/C1 versus δ C1 diagram from model calculation of mixing and diffusion trends, with linear (a) and logarithmic (b) scales. The two mixing end members 13 have δ C of -80‰ and -35‰, and C2/C1 ratios of 0 and 0.3, respectively. The source gas has the same composition for the diffusion calculation (Prinzhofer and Pernaton, 1997).

13 13 Figure 2.10 The δ C2- δ C3 versus C2/C3 diagram (Lorant et al., 1998), with two examples of gas series from the Ceara Basin, Brazil (Prinzhofer et al., 2000b) and from Thailand. The first case presents a trend of maturity of an open system during the primary cracking of kerogen, the second case presents a wider range of maturity in a more closed system, ranging from the primary cracking to secondary cracking of oil and gas (Prinzhofer and Battani, 2003).

Theoretical and kinetics models have developed within the past twenty years to help explain the mechanism of natural gas formation, both chemically and isotopically (Gaveau et al., 1987; Galimov, 1988; Lorant et al., 1998; Clayton, 1991; Berner et al., 1995; Rooney et al., 1995; Tang et al., 2000). Theoretical considerations can quantify the gas isotopic fractionation and interpret the kinetics and mechanism of gas formation. 13 The first theoretical and kinetic model relating carbon isotopic values (δ C1, 13 13 δ C2, δ C3) to gas generation as a function of fractional conversion of kerogen or oil to gas from Type II kerogen, was developed by Rooney et al. (1995), based on the Rayleigh distillation theory. They tested their model on gases from the Delaware and Val Verde basins, West Texas and concluded that gas composition was relatively unaffected by migration or biogenic contamination. Coupling the Rooney et al. (1995) Rayleigh model to Burnham’s compositional model (1989), gas generation temperature can be derived from carbon isotopic fractionation for methane, ethane, and propane (Figure 2.11).

Different kinetic models for gases derived from Type III kerogen were suggested because the Type III kerogen generally shows larger fractionation of carbon isotopic composition 13 13 between ethane and propane (δ C1, δ C2) than does Type II kerogen. This might be explained by greater molecular heterogeneity in Type III kerogen (Rooney et al., 1995). Both Galimov (1988) and James (1990) also have had noted that kerogen composition is a source control on gas generation and isotopic composition. The most important assumption in Rooney’s (1995) Rayleigh model is that the relative rate of generation of methane and higher molecular weight hydrocarbon gases stays constant and gas is primarily generated from aliphatic hydrocarbon groups that are dominantly

homogeneous. However, later work showed that this is not true. That the ratio of C2+ hydrocarbons to methane is not constant during thermal maturation has been well established by several authors (Hill et al., 1996; Tang et al., 1996). Failure to identify the real kinetic isotopic fractionation factor (ε) during gas generation in the Rayleigh model makes Rooney’s method invalid. Newer stable carbon isotopic models were based on parallel first-order kinetic effects and calibration to laboratory pyrolysis (Tang and Jenden, 1995; Cramer, 1998; Lorant et al., 1998; Tang et al., 2000; Cramer et al., 2001; Tang et al., 2005). The kinetic isotope effect (KIE) was used to model isotopic fractionation for hydrocarbon generation from kerogen by Lorant et al. (1998). During primary cracking, they observed a decrease 13 of δ C in methane (C1) to butane (C4) in their isothermal anhydrous close-system 13 experiment. A divergence in δ C of C2-C4 was also observed at higher maturity, during

secondary cracking (Lorant et al., 1998). In addition, a successive decrease of C2-C5 at high pyrolysis temperature occurred during secondary cracking (Lorant et al., 1998). Their isotopic fractionation factor during secondary cracking was later improved by Tang et al. (2000) using a thermodynamic approach. Tang et al. (2000) applied quantum mechanical calculations to quantify isotope fractionation during natural gas generation. They successfully presented a kinetic model to fit chemical and isotopic composition of methane cracked from n-octadecane under isothermal closed-system conditions. The model not only provides quantitative 13 relationships among δ C1 (methane), total methane yield and methane generation rate,

but can be used to evaluate the source rock maturities at which specific accumulation were generated (Tang et al., 2000).

13 Figure 2.11 Gas generation temperatures predicted from the carbon isotope ratios (δ Cn) of methane, ethane, and propane for gas derived from Type II kerogen (data shown are from the Delaware and Val Verde basins). Temperatures obtained using Burnham's model for gas generation from Type II kerogen, assuming a heating rate of 1°C/Ma. (Rooney et al., 1995)

2.2.3 Hydrogen Isotope Effects in Kerogen during Thermal Transformation Deuterium (D) and hydrogen (H) have the largest relative mass difference of any stable isotope pairs and should therefore show the greatest fractionation in nature. However, few publications address the relationship among hydrogen isotope of gas, types of kerogen in source rocks, and maturity because the hydrogen isotope of gas might be affected by hydrogen isotope of surrounding water. Reddings et al. (1980) carried out one of the first investigations of the hydrogen isotopic composition of different types of kerogen: Type I kerogen from the Uinta Basin (Tissot et al., 1978), Type II kerogen from the Paris Basin and from two Posidonia shales in Northwestern Germany (Durand et al., 1972) and Type III kerogen from Douala Basin (Durand and Espitatle, 1976). They found that Type I kerogens maintained uniform hydrogen isotopic compositions throughout diagenesis and catagenesis and the deuterium isotopic composition increased slightly with respect to decreasing H/C ratios of 70 % (Figure 2.12). The deuterium isotopic composition in Type II and Type III kerogen increased by approximately 20 ‰ with decreasing H/C ratios and increasing thermal maturity (Figure 2.12). Hydrogen isotopic compositions did not show a linear change with atomic H/C ratios (Reddings et al., 1980). 22

More attention has been paid on hydrogen isotopic composition of kerogen and hydrocarbons in recent years (Burgoyne and Hayes, 1998; Hilkert et al., 1999; Li et al., 2001) because deuterium (D) and hydrogen (H) have the largest relative mass difference of any stable isotope pairs and should therefore show the greatest fractionation in nature. The factors that control δD values in hydrocarbons are the isotopic compositions of (1) source organic matter; (2) hydrogen exchange processes that involve D/H in water or clay minerals (Sessions et al., 2004; Alexander et al., 1982); (3) thermal maturation (Schimmelmann et al., 1999; Schimmelmann et al., 2001); (4) post-genetic processes, such as water washing, biodegradation, and migration (Lis et al., 2005). Schimmelmann et al. (1999) conducted experiments and used isotopic mass-balance calculation, showing that there is an increasing amount of hydrogen in organic kerogen derived from water and the derivation of hydrogen from water increases with increasing temperature and time. Schimmelmann et al. (2004) further suggested the δD values of oils might be affected more by their source rocks rather than the hydrogen exchange reaction between formation water and oils. Lis et al. (2006) investigated hydrogen isotopic signatures of non-

exchangeable hydrogen (δDn) and exchangeable hydrogen (δDex) in two matured Type II kerogens, from the New Albany Shale, Illinois Basin, and from the Exshaw Formation

(EF) in the Western Canada sedimentary basin. In both cases, δDn values increased with

thermal maturation up to Ro = 1.5 % and then leveled out, whereas δDex values first decreased to a minimum at Ro = 0.8-1.1 %, followed by a substantial increase at higher

thermal maturity. The increases in δDn values of Type II kerogens may be due to incorporation of water-derived deuterium from formation water during thermal maturation. The mechanisms for isotopic fractionation of D/H during thermal maturation of kerogens are still unknown.

12 21

22 13 23

31 32 14 33 24

25 34 15 35

Figure 2.12 Measured hydrogen isotopic fractionation during the evolution of different types of kerogens. Arrows show pathways of evolution. The deuterium isotopic composition show only slight increase with thermal maturity in Type I kerogen but has significant increase in Type II and Type III kerogens (Reddings et al., 1980).

CHAPTER 3

GEOLOGICAL BACKGROUND

This chapter describes the Cretaceous tectonic and sedimentary evolution of sedimentary basins in the Rocky Mountains, focusing specifically on critical source rock formations that may have been hydrocarbon source rocks for tight-gas fields. In the Early Cretaceous (ca. 130 Ma), an epicontinental seaway, known as the Western Interior Seaway, developed in North America (Figure 3.1) (Blakey, 2008), reaching its greatest extent during the Middle Cretaceous, when it stretched from central Utah to the western Appalachians and from Arctic Sea to the Gulf of Mexico (ca. 90Ma) (Figure 3.1) (Blakey, 2008). The Sevier orogeny defines the eastern margin of thin- skinned structural deformation in the North American Cordillera (Armstrong, 1968; Burchfield and Davis, 1975; Currie, 2002). The loading by the Sevier thrust belt produced a flexural foreland basin parallel to the thrust front, called the Cretaceous Western Interior Foreland Basin (Currie, 2002). Most of the Cretaceous strata in the Piceance Basin and Greater Green River Basin such as Mowry, Mancos, and Mesaverde Group, were deposited during the Sevier Orogeny, between the fore-bulge and back- bulge (Currie, 2002). Thick-skinned structural deformation of the Laramide orogeny east (in the foreland) of the Sevier thrust belt was developed during the Late Cretaceous (Johnson and Flores, 2003). The uplift of the Park and Sawatch Range and the deposition of the Lance and Wasatch Formations are considered the start of the Laramide Orogeny in the Piceance Basin and Greater Green River Basin (Johnson and Flores, 2003). Kauffman (1977, 1984, 1985) divided the Western Interior Seaway Basin into five zones trending approximately north-south, parallel to the trend of the basin. From the western Cordilleran thrust belt to the eastern stable craton platform, they are: (1) the foredeep basin, characterized by thick, coarse clastics associated with maximum subsidence and sedimentation rates (Jordan, 1981); (2) the forebulge zone characterized by discontinuous uplifts and arches that often formed erosional highs; (3) the axial basin, characterized by fine-grained mudstones, shales, and lime muds, an area of high subsidence and sedimentation rates in the deepest part of the basin; (4) the tectonic hinge zone between the foreland basin and stable platform, characterized by rapidly changing

facies, and complex regional disconformities with moderate subsidence rates and water depths; (5) the stable eastern platform zone, associated with low subsidence and sedimentation rates and containing alternating sequences of fine-grained clastic and carbonate deposited in shallow water. In the Late Cretaceous and early Tertiary, this foreland basin was tectonically divided into several small Laramide basins by uplifted blocks of crystalline rocks, causing the Western Interior Seaway to withdraw and retreat in pulses towards the northeast. During the Maastrichtian, these basins were gradually filled by sediments consisting mainly of terrestrial rocks from the western continent. These basins include two main tight-gas-sand basins that are addressed in this study: Greater Green River and the Piceance Basin. A brief summary of each basin with the emphasis on stratigraphy, source rocks, burial history, and petroleum system characteristics will be presented.

3.1 Piceance Basin, Colorado The Piceance basin is located in the northwestern Colorado. It is bounded in the north by the Uinta Mountains, to the east by the Grand Hogback Monocline along the western flank of the White River Uplift, to the southeast by the Elk Mountains, to the southwest by the Uncompahgre Uplift, and to the west by the Douglas Creek Arch. The basin is asymmetrical, with a gently dipping western flank bounded by the Uncompahgre Uplift and the Douglas Creek Arch, and a steeply overturned eastern flank along the White River Uplift (Johnson and Flores, 2003) (Figure 3.2). The Piceance Basin was part of the Western Interior Seaway during the Cretaceous. The Cretaceous Interior Seaway retreated eastward, resulting in the deposition of several transgressive-regressive depositional pulses. Cretaceous rocks, mostly shed from the Sevier Orogeny (Late Jurassic through Late Cretaceous), were deposited in the Western Interior Seaway, forming a large-scale stratigraphic succession containing several transgressive-regressive cycles. The stratigraphic column of the Late Cretaceous, in ascending order, contains the Dakota and Cedar Mountain Formations, Mowry Shales, Mancos Shale, Castle Gate Sandstone, and Iles Formation, lower Williams Fork Formation, and upper Williams Fork Formation of the Mesaverde Group (Figure 3.3). Above the Mesaverde Group, is the

Figure 3.1 (A) In the Early Cretaceous (ca. 130 Ma), the Western Interior Seaway of North America was an epicontinental seaway, bounded by a thrust belt in Nevada and western Utah, a transtensional basin in Arizona and southern California, and a classic continental (Andean-style) Cordilleran arc along the western coast of North America (Blakey, 2008). (B) During the Middle Cretaceous (ca. 90Ma), the (A) Western Interior seaway expanded to one of its greatest extents, stretching from central Utah to the western Appalachians and from Arctic Sea to the Gulf of Mexico (Blakey, 2008).

(B)

12a 12b Figure 3.2a and b. (a) Generalized map of the Laramide tectonic elements in eatern Utah and western Colorado (after Grose, 1972); (b) Generalized map of the Western Interior Seaway (modified after Patterson et al., 2003, Chapter 7, RMAG)

Figure 3.2c A cross section in Piceance Basin (A-A’). Piceance Basin is asymmetrical, with a gently dipping western flank bounded by the Uncompahgre Uplift and the Douglas Creek Arch, and a steeply overturned eastern flank along the White River Uplift (Zhang et al., 2008)

Figure 3.3 The stratigraphic column of the Piceance basin, from the Late Cretaceous through Tertiary. Abbreviations: o.m., orange marker; c.m., carbonate marker; l.p., Long Point Bed (Johnson and Roberts, 2003).

Tertiary Wasatch Formation, Green River Formation, and Uinta Formation (Hettinger and Kirschbaum, 2002, 2003). Tertiary outcrop strata are exposed at surface within the basin, with Mesaverde Group and Mancos Shale cropping out around the basin.

3.1.1 Source Rocks- Maturity and Potential Major hydrocarbon source intervals in the Piceance Basin include: the marine shales within the Mancos and Iles Formations; extensive coals within the Iles and Williams Fork Formations; and nonmarine shales within the Iles and Williams Fork Formations (Johnson and Roberts, 2003). The Mancos Shale is in excess of 5,000 ft. thick throughout most of the basin and act as an active source rock in both the Uinta and Piceance Basins (Kirschbaum, 2003). Coals within the Corcoran-Cozzette, Cameo- Fairfield, and lower Williams Fork sections are thought to have generated the largest volume of gas (Yurewicz et al., 2008), while the organic-rich shales within Iles and

Williams Fork Formation, with relatively low HIo (original hydrogen indices), are thought to have generated comparatively small amount of gas. The presence of the Upper Cretaceous Mowry Shale in the Piceance Basin is controversial. The Mowry Shale is observed to pinch out southward from Vernal, Utah, by Molenaar and Wilson (1990) and Molenaar and Cobban (1991). The Mowry Shale is present around Dinosaur National Monument and Rangely anticline, and is also present in the eastern and northern Piceance Basin (east of the Douglas Creek Arch: Anderson, 2010, personal communication). Farther south in the Uinta Basin and along the Douglas Creek Arch, the Mowry pinches out by depositional thinning and by grading into underlying Dakota Sandstone (Molenaar and Wilson, 1990) or into undifferentiated Dakota-Cedar Mountain deposits (Molenaar and Wilson, 1990). Fisher (2007) suggests that the Mowry Shale is not present or is remnant in the Douglas Creek Arch area. Overall, the Mowry Shale is likely not a good source rock in Piceance Basin because it is thin and of a proximal marine facies.

The Upper Cretaceous Mancos Shale The Upper Cretaceous Mancos Shale was deposited in offshore and open-marine environments of the Cretaceous Western Interior Seaway. The Mancos consists of dark- gray to black calcareous and bentonitic shales, with minor siltstone and sandstone (Kirschbaum, 2003). Early studies of the Mancos Shale include Cross and Purington (1899), Cobban and Reeside (1952), Weimer (1959, 1960), Kent (1965, 1968), Kauffman (1969, 1977, 1984, 1985), Molenaar and Wilson (1990), and Molenaar and Cobban (1991). Marine Mancos shales are present throughout nearly the entire Piceance Basin (Kirschbaum, 2003). The thickness of the Mancos Shale ranges from 3,450-4,150 ft. in the southern Piceance Basin (Fisher, 2007) to about 5,400 ft in the central part of the Uinta Basin (Hettinger and Kirschbaum, 2002). In the Douglas Creek Arch area (DCA), the Mancos Shale thins uniformly from north to south (Fisher, 2007). The Mancos Shale has a gradational upper contact with the overlying Mesaverde Group and a sharp basal contact with the underlying Dakota Sandstone. Detailed stratigraphic correlations within the Mancos Shale are still debated. Different terminologies were used in Uinta Basin, Piceance Basin, and DCA area.

Authors have subdivided the Mancos Group differently in their study areas based on biostratigraphy (ammonite and molluscan fossil), lithostratigraphy, geochemistry, geomechanical and petrophysical (well logs) data, or combinations of the above (Figure 3.4: O'Boyle, 1955; Berman et al., 1980; Spencer and Wilson, 1988; Molenaar and Cobban, 1991; Gardner, 1994; Anderson and Harris, 2006; Fisher, 2007; Kuzniak, 2009). A carbonate interval within Mancos Shale was recognized by some studies (Haskett, 1959; Longman et al., 1998; O'Boyle, 1955; Vincelette and Foster, 1992; Anderson and Harris, 2006; Fisher, 2007; Kuzniak, 2009). The lower part of the marine Mancos Shale includes the Tununk Shale, Ferron Sandstone, and Juana Lopez Members in the southern and eastern Uinta Basin, northeastern Utah, and northwestern Colorado (Molenaar and Cobban, 1991). The base of the lowest member, the Tununk Shale unconformably overlies the Dakota Sandstone. Where the Dakota is absent, the Tununk Shale sits unconformably on top of the Cedar Mountain Formation. The Mancos B, or Emery Sandstone, in the Piceance Basin consists of 500 to 1,000 ft of inter-bedded sandstone and laminated mudrock deposited during the Santonian and early Campanian (Kirschbaum, 2003). The lowest part of Mancos Shale is the most organic-rich section (Kirschbaum, 2003). Outcrop samples taken from the southwestern Piceance Basin have TOC values ranging from 0.18 to 3.36 %, with the highest value sampled from the lower 70 ft of the Mancos section near Delta, Colorado (Johnson and Rice, 1990; Kirschbaum, 2003). Geochemical analyses of lower Tununk Shale, Tununk Shale, and Niobrara Formation show the greatest potential for generating hydrocarbons (Kuzniak, 2009). A screening of geochemical analysis on cutting samples from the Hells Hole 9131 well in the DCA shows an average TOC value of 1.35 % and an average hydrogen index (HI) of 209 from the lower Mancos Shale (Fisher, 2007). Fisher (2007) proposed two source rock candidates (TOC> 1.5 %: HI>200) within lower Mancos Shale: the Niobrara-equivalent and the Tununk/ Juana Lopez member (Fisher, 2007).

Edwards et al., Stage 2009 Blackhawk Fm Campania n Star Point Ss

Blue Gate Shale (Upper) Santon ian Emery

Blue Gate Shale (Lower)

Ferron Ss Tununk Shale

Tununk Shale

Tununk

Dakota

Figure 3.4a Previous published and proposed nomenclature of the Cretaceous units in northwestern Colorado, mostly based on biostratigraphy (ammonite and molluscan fossils) (modified after Anderson and Harris, 2006)

Kuzniak (2009)

Figure 3.4 b Previous published and proposed nomenclature of the Cretaceous units in northwestern Colorado (modeified after Kuzniak, 2009)

An integrated study of the Lower Mancos Shale in the Uinta Basin, Utah by Anderson and Harris (2006) also showed the Mancos Shale has good potential as a source rock. The TOC contents in cutting samples from the Lower Mancos ranged from a minimum of 0.44 % to a maximum of 4.32 % (average TOC: 1.23 %). Vitrinite

reflectance (Ro) measurements show an increase in Ro from 0.68 % near the top of the lower Mancos to 0.89 % in the uppermost Dakota Sandstone (Anderson and Harris, 2006). The organic matter type identified from Rock-Eval pyrolysis indicated a range from dominantly Type II to mixture of Type II and Type III kerogens (Anderson and Harris, 2006). From modeling of U.S. DOE MWX (Multi-Well Experiment) wells, initial oil generation from the lower Mancos was about 76 Ma in the Piceance Basin. Peak gas generation from the lower Mancos was between 57 and 53 Ma (Nuccio and Roberts, 2003). Figure 3.5 shows the vitrinite reflectance contour map for the top of the Mancos Shale in both Uinta and Piceance Basins (Nuccio and Roberts, 2003). The data is from another integrated oil and gas assessment group by the U.S. Geological Survey.

Figure 3.5 Contour map of thermal maturity values for the top of the Mancos Shale (Nuccio and Roberts, 2003, Chapter 4, USGS DDS-069-B)

Studies have shown that oils are sourced from the Mancos Shale. Oil and gas are produced from the Mancos B on the Douglas Creek arch and in Rangely Field, Colorado

(Lillis et al., 2003). The oil is generally thought to be generated from Mancos Shales below Mancos B in both areas (Lillis et al., 2003). Chemical and isotopic compositions of gas collected from Morrison, Cedar Mountain, Dakota, Mancos B, Corcoran, and Cozzette Formations were interpreted to be have been sourced from the Mancos Shale (Johnson and Rice, 1990).

The Upper Cretaceous Cameo-Fairfield Coal Zone The thickest and most extensive Upper Cretaceous coals were deposited during Campanian to Maastrichtian in the Rocky Mountain area (Cross, 1988). Vertical aggradation of coastal-plain facies took place when the accommodation space was near maximum and when the rate of sea-level change was balanced by the rate of sedimentation (Cross, 1988). The lowest part of the Williams Fork Formation includes a well-developed coal sequence known as the Cameo-Wheeler Coal Zone, which conformably overlies and intertongues with the Rollins Sandstone Member of the Iles Formation (Figure 3.6). Both the Cameo-Wheeler Coals and the Rollins Sandstone are persistent units within the Mesaverde Group, extending westward from Grand Hogback to near the Colorado-Utah state line. The total thickness of Cameo-Wheeler Coal Zone ranges from 50 to 450 ft. and it contains about 87 net feet of coals. The Cameo Coal pinches out to the south beneath West Elk Mountain and to the west near the Colorado-Utah border (Hettinger and Kirschbaum, 2002). The South Canyon Coal Zone and Coal Ridge Coal Zone both overlie and intertongue with the Bowie Shale Member of the Williams Fork Formation (Figure 3.6). The thickness of the South Canyon coal zone is as much as 330 ft. thick and contains about 48 net ft. of coals (Hettinger et al., 2000). The Coal Ridge coal zone is as much as 200-500 ft thick and contains about 44 ft of net coal (Hettinger et al., 2000). Coal has a substantial capacity to adsorb gas in micropores and cleats. Thus, much of the gas generated during the early stages of thermogenic gas generation may have remained within the coal beds. Only after coals are saturated with gas does significant expulsion of gas occur. The ability of coal to store gas decreases with increasing rank (Juntgen and Karweil, 1996; Meissner, 1984), decreases with increasing temperature, and increases with increasing pressure (Mesissner, 1984; Wyman, 1984).

Coal beds generally start to expel significant amounts of gas at medium-volatile bituminous ranks and higher (Ro≥ 1.1%) (Rice, 1993). During the Late Cretaceous, the Cameo Coal was not buried deeply and the coalification temperature was low (Johnson and Nuccio, 1986; Law et al., 1989). As Tertiary sediments accumulated, the Cameo Coal underwent continuous burial and reached a maximum burial depth at the end of Laramide orogeny (~36Ma). The thickness of the Tertiary formations is more than 12,000 ft. along the deepest part of the foreland basin and thins to 1,200 ft in the DCA area (Johnson and Nuccio, 1986). At the Multi- well Experiment (MWX) site in the Piceance Basin, the timing of maximum burial of Mesaverde coals was between 35 and 9 Ma (Barker, 1989). The maximum burial of the lowest coal of the Mesaverde was 12,600 ft. at 35 Ma (Johnson and Nuccio, 1986). Peak gas generation in the Piceance basin was between 36 and 20 Ma (Nuccio and Roberts, 2003), based on modeling studies. Thermal maturities based on vitrinite reflectance in the coal interval of the Mesaverde

Group range from an Ro of 0.60 % or less in outcrops around the margins of the Piceance

Basin to Ro values exceeding 1.35 % in deeper parts of the basin (Figure 3.7) (Johnson

and Roberts, 2003). Maximum Ro values are about 2.1 % in the deepest trough of the Piceance Basin.

3.2 Greater Green River Basin, Wyoming Jonah Field in the northwestern part of the Greater Green River Basin, in Sublette County, southwestern Wyoming, is the focus of this study on gas compositions in the Greater Green River Basin (Figure 3.8). The Pinedale Field, another large tight-gas-sand field, is located just northeast of the Jonah Field (Figure 3.8). These fields are located in the Hoback Basin, a smaller basin located in the northern LaBarge Platform and Sandy Bend Arch, and the west of the Pinedale Anticline, and considered to be part of the Greater Green River Basin. The Hoback Basin is a northwest-trending trough southwest of the Wind River Range. Jonah Field is a wrench-fault-bounded structural trap. The south-bounding fault and west-bounding fault form a wedge-like up-dip closure in the southwest of the field and are lateral seals of Jonah Field (Warner, 1998).

Figure 3.6 West-east cross section showing distribution of depositional facies and major coal zones in Mesaverde Group, southern Piceance Basin, Colorado. Abbreviation: Ss., Sandstone; ss., sandstone; Mbr., member; cz., coal zone; Sh., Shale; Gp., Group; Fm., Formation; pt, part. (Modified after Hettinger et al., 2000; Johnson and Roberts, 2003, Chapter 7, USGS DDS-69-B)

Figure 3.7 Vitrinite reflectance (Ro) trends at the base of the Cretaceous Mesaverde Formation Group and the top of the Mancos Shale Group (Nuccio and Roberts, 2003, Chapter 4, USGS DDS-069-B)

The downdip limit of Jonah Field is the Sand Draw Syncline that separates Jonah field from the Pinedale Anticline to the northeast. Several northeast-trending arcuate shear faults subdivide the field into several pressure compartments within the Jonah Field structural wedge. Six important fault zones and six main folds are identified and modeled in his study, as shown in Figure 3.9. Each of the compartments shows different productivity characteristics, which suggests they are isolated from each other (Warner, 2000). The stratigraphic units in the Jonah Field, from the deepest well penetrated horizons up to the surface are, in ascending order, are the Upper Cretaceous Mowry Shale, Frontier Formation, Hilliard Shale, Rock Spring Formation (Mesaverde), Ericson Sandstone member (Mesaverde), Almond Sandstone member (upper Mesaverde equivalent), Lance Formation, Unnamed Tertiary Formation, Tertiary Fort Union Formation, Wasatch Formation and Green River Formation (Law et al., 1986; Warner, 1997; Johnson et al, 2004: Figure 3.10). Both the Jonah Field and Pinedale Anticline are located west of the maximum transgression during the Maastrichtian. Several transgressions and regressions of the Western Interior Seaway occurred in the Greater Green River Basin during the Cretaceous, in response to episodic thrust belt deformation and eustatic change (Tyler et al., 1995). The thrusting culminated with the Laramide Orogeny during the Eocene. Loading of the thrust sheets causes more rapid subsidence in the western portion of the Green River Basin (Horne, 2009). At this time, Jonah Field was located northwest of the Lewis shoreline. Simultaneously, the Wind River Uplift and Granite Mountains were thrusted to the southwest and south. Sediment was rapidly shed from the rising uplifts and generally filled the basins above relative sea level. The depositional environments were alluvial-plain and upper coastal plains, where fluvial channels, flood-basin splays, flood-plain mudstones, and paleosols accumulated. The timing of the tectonic movements during the Cretaceous affected the distribution of source and reservoir facies in different stratigraphic intervals; therefore, affecting the subsequent migration of hydrocarbons (Figure 3.11) (Horne, 2009). One example is the difference in burial history between two adjacent fields - Jonah Field (Stud

Horse Butte 13-271) and the Pinedale Area (Wagon Wheel 1).

Figure 3.8 Location of Jonah Field. The Pinedale Field is located northwest of the Jonah Field (Modified from Hanson et al., 2004).

Figure 3.9 Map of Jonah field showing the principal faults and the small anticlinal and synclinal folds: WF = west fault; SBWF = Stud Horse Butte west fault; SBF = Stud Horse Butte fault; SBEF = Stud Horse Butte east fault; AF = Antelope fault; SJF = south Jonah fault; SDS = Sand Draw syncline; SBS = Stud Horse Butte syncline; SDN = Sand Draw nose; CN = Cabrito nose; SBA = Stud Horse Butte anticline; YPN = Yellow Point nose (Hanson et al., 2004).

1‐3: Mesaverde‐Lance‐Fort Union Total Petroleum System (TPS) 4: Lewis TPS 5. Mesaverde TPS 6. Niobrara TPS 7. Hilliard‐Baxter‐Mancos TPS 8. Mowry TPS

Figure 3.10 Generalized stratigraphic column of the southwestern Wyoming (USGS Southwestern Wyoming Province Assessment Team, 2005).

Gases from the Lower Mesaverde (Rock Springs Formation) in the Pinedale Anticline are modeled to have undergone expulsion and migration at 54 Ma when the Lance Formation in the Jonah area had only reached a burial depth of about 8,000 ft (middle Lance) and would not have been fully compacted (Coskey, 2004). Gas generation at the Pinedale Anticline began about 4 million years earlier than Jonah Field (SHB 13-27). Modeling indicates that higher lithostatic pressure and hydrocarbon concentration in the Pinedale Area may have helped hydrocarbons migrate updip to the lower-pressured Jonah area (Coskey, 2004).

Figure 3.11 Diagram relating the stratigraphic units in the Cretaceous, timing of thrust movements, and thicknesses of stratigraphic in the foreland basin of Idaho, Wyoming, and Utah (Horne, 2009).

3.2.1 Source Rocks – Maturity and Potential Potential source rocks in the Jonah Field contain Type II, Type III, or mixed Type II/III kerogens. Major hydrocarbon source intervals in the Jonah Field include: the Mowry Shale, the Baxter (Hilliard) Shale, coals and carbonaceous shales in the Lower Mesaverde Formation, shales in the Upper Mesaverde Formation (possibly equivalent to the Lewis Shale), shales in the Lance Formation, and coals in the Fort Union Formation. Coskey (2004) examined lithology, thickness, vitrinite reflectance, TOC, and other geochemical characteristics of these source rocks and suggested both Lance and Fort Union Formations are unlikely to be good source rocks. The thickness of the Mowry Shale is about 350 ft and the Hilliard Shale is more than 3,500 ft in the Jonah Field. The Mowry Shale has relatively low TOC which might be due to dilution from western- derived clastic sediments and to elevated maturity (Coskey, 2004). The base of the Hilliard Shale is at about 21,000 ft, and some faults penetrate the Hilliard Shale that may act as conduits allowing vertical migration. The Lower Mesaverde Formation (Rock Springs and Ericson Formations) contains abundant humic coals, weak fluorescence and

oil/condensate stains (Coskey, 2004) and has TOC of 1 % to more than 8 % (Law, 1984). The thermal maturity of the Upper Mesaverde Formation (mostly siltstone and mudstone)

is about 0.73 to 1.00 % Ro (Coskey, 2004). Candidates for hydrocarbon source rocks in Jonah Field have been studied and modeled in several papers (e.g. Law, 1984; Spencer, 1987; Dickinson, 1989; Hanson et al., 2004; Coskey, 2004). Average gas composition in

Jonah Field is approximately 91 % C1, 8.5 % C2+, and 0.5 % CO2 and the methane isotopic compositions are indicative of thermogenic gas (δ13C ratio = -36 ‰: Hanson et al., 2004).

The Upper Cretaceous Coals in Mesaverde & Lance Formations Coals and organic-rich carbonaceous shales are presumed to be the primary source of gases in the Mesaverde and Lance Formations (Law, 1984). Coal is present mainly in the Rock Springs Formation, and the Almond Formation equivalent of the Mesaverde Group, with only minor thickness of coals in the Ericson and Lance Formations. Tyler et al. (1995) measured from 0 to over 75 ft of coals in the Rock Springs Formation with the thickest accumulation northwest of the Rock Springs uplift. Coal-bearing strata are more than 13,000 ft deep in the Pinedale Area (Tyler et al., 1995). Maps of vitrinite reflectance (Ro) measured from coal chips and cuttings were constructed from three stratigraphic horizons: the base of the Rock Springs Formation, the top of the Mesaverde Group, and the base of the Fort Union Formation, as shown in Figures 3.12 (Finn et al., 2005). Vitrinite reflectance at these 3 horizons in Jonah Field ranges from 0.8 to 2.0% Ro (Finn et al., 2005). Coskey (2004) applied 1D burial history modeling and petroleum system analysis to the gas accumulation at Jonah Field by assuming the Lance interval contained Type III kerogen with average initial TOC values ranging from 1.00 to 1.25 %. Coskey (2004) suggested that, although several coal intervals within Lance and underlying Mesaverde were capable of generating significant amounts of gas, there was insufficient coal-bearing section to have generated all the gas trapped in the Jonah Field. He concluded that gas must have been generated and migrated from either deeper sources such as Hilliard Shale or Mowry Shale or migrated laterally from other unknown source areas (Coskey, 2004).

The Upper Cretaceous Baxter/Hilliard Shale The Hilliard Shale consists of three units: a lower black shale, a middle sandstone, and a thick upper unit of interbedded sandstone, shale, and coal (Frerichs and Steidtmann, 1971; Adams, 1972). Figure 3.13 shows the geographical extent of Hilliard-Baxter- Mancos shale in southwestern Wyoming (Finn and Johnson, 2005). The thickness of this stratigraphic interval ranges from 3,500 to 6,000 ft. (Finn and Johnson, 2005). The lower shale unit varies considerably across the basin. The middle sandstone is found to be continuous throughout the basin. The upper shale and sandstone unit is composed of a sequence of thick dark gray to black shales interbedded with thin, discontinuous sandstone (Frerichs and Steidtmann, 1971). Numerous thick coal beds are also present. The depositional environment was interpreted to be paludal or mixed marine-nonmarine environment and corresponds to a typical regressive sequence common during the Cretaceous in the Rock Mountain Area (Finn and Johnson, 2005). The Hilliard Shale is more than 3,700 ft. thick in the Jonah Field area with present day TOCs ranging from 0.25 to 1.39 % (Law, 1984). Thermal maturities at the base of the Hilliard-Baxter-Mancos section are shown in

Figure 3.13 (Finn and Johnson, 2005). Thermal maturities increase from 0.6 % Ro at the base and around the margin of the basin to greater than 1.3 % Ro in the deepest part of the basin. Vitrinite reflectance is more than 2.0 % Ro in the deep trough of the Hoback Basin. The Upper Cretaceous Mowry Shale The Mowry Shale was deposited in a rapidly subsidence foreland basin with development of the Meade-Crawford Thrusts (Burtner et al., 1994). It is an organic-rich, hard, dark-gray, siliceous mudstone with abundant fish scales and bentonitic layers (Figure 3.14) (Burtner and Warner, 1984; Johnson, 2003). The top of the Mowry Shale is typically placed by the top of the Clay Spur Bentonite Bed (Nixon, 1973; Burtner and Warner, 1984). The Mowry Shale overlies the Dakota or Muddy J Sandstone in most of the Rocky Mountain basins. The Mowry is recognized as a high resistivity zone on well logs, whereas low resistivity signals (high gamma-ray values) can be related to bentonite in the subsurface of the southwestern Wyoming (Nixon, 1973; Kirshbaum and Roberts, 2005). The Mowry thickens to nearly 300 ft. to the northeast of the Piceance Basin

a. The base of the Rock b. The top of the Mesaverde c. The base of the Paleocene Springs Formation Group Fort Union Formation Figure 3.12 Contour map of the thermal maturity (vitrinite reflectance) at (a) the base of the Rock Springs Formation, (b) the top of the Mesaverde Group, and (c) the base of the Fort Union Formation (Modified after Finn et al., 2005). (Burtner and Warner, 1984; Molenaar and Wilson, 1990). The organic matter content of the Mowry Shale was first investigated by Schrayer and Zarella (1963, 1966, and 1968). An increase in organic matter content from the northwestern to southeastern Mowry Sea is consistent with thicker deposition and better preservation of fine-grained shales near the axial part of the seaway (Nixon, 1973). However, Surdam et al. (2007) suggested that conversely, the Mowry Shale thins from about 600 feet in the northwest to about 250 feet in the southeast of the Greater Green River Basin. Byers and Larson (1979) defined three different depositional facies within the Mowry Shale: anoxic, laminated mudstone; oxic, bioturbated mudstone; and oxic bioturbated sandstone. They found that the Mowry Shale becomes more oxic and coarser from southwest to northeast. The Mowry Shale has the highest TOC among the Cretaceous shales (Burtner and

Warner, 1984). TOC, hydrogen index (HI), and maximum temperature (Tmax) were

measured in 19 samples of the Mowry Shale and ranges from 1.6 to 2.4 %, 32 to 276, and 424 to 450°C, respectively. A mixture of Type II and Type III kerogen was found in the Mowry Shales (Burtner and Warner, 1984). Mowry Shales are mature throughout most of

the Greater Green River Basin. Vitrinite reflectance values (Ro) less than 0.6% are found near basin margin and the Rock Spring Uplift (Kirshbaum and Roberts, 2005). Based on the thermal maturation modeling, Surdam et al. (2007) also suggested 80 mg of gas was generated for each gram of TOC in the Mowry Shale - 18 mg of gas were expelled, making it a potential shale-gas prospect in Wyoming. Initial petroleum generation of the Mowry Shale may occur during the early Late Cretaceous (100-80 Ma) because of the sufficient TOC and level of maturity in the Mowry Shale, prior to the Hogsback and Prospect thrusts (Kirshbaum and Roberts, 2005). A second period of petroleum generation may have taken place after the burial of foredeep deposits beneath the Absaroka Thrust (Burtner et al., 1994). One-dimensional thermal maturity and burial history modeling was performed on several wells in the Greater Green River Basin east of Wyoming thrust belt (Roberts et al., 2005; Kirshbaum and Roberts, 2005). Gas generation from secondary cracking of oil in deeper parts of the basin started at about 56 Ma and ended at about 41 Ma (Kirshbaum and Roberts, 2005). Gas-prone, Type III source rocks show the beginning of gas generation starts about 78 Ma (Roberts et al., 2005).

Figure 3.14 Stratigraphic correlations of the Mowry Shale in southwestern Wyoming and Wyoming Thrust Belt. Generalized depositional environments, significant fossils, significant oil and gas producing units, and potential source rocks are shown. Subsurface equivalent names are shown for the Moxa arch area; B1–B5 are benches within the second Frontier. Abbreviations: Sh, shale; Mbr, member; Al, Albian; Ce, Cenomanian; Tu, Turonian; Co, Coniacian. (Kirschbaum and Roberts, 2005).

CHAPTER 4

METHODS & EXPERIMENTS

This chapter describes the SRA analysis (open system, anhydrous) and hydrous pyrolysis (close system, hydrous) experiments on different source rocks to evaluate source rock potential and simulate hydrocarbon generation, respectively. Oil and gas generated from hydrous pyrolysis experiments were then analyzed by gas chromatography (GC) and Isotope Ratio Mass Spectrometer (IRMS). A series of compounds in generated oils, the bulk composition of generated natural gas, and the carbon isotopic composition of methane in natural gas provide information on compositions as they evolve with increasing thermal stress.

4.1 Potential Source Rock Sample Collection and Composite Immature potential source rocks collected directly from outcrops and cores were used in hydrous pyrolysis experiments. The immature source rocks are defined as rocks before or at the beginning of the oil generation. Initial hydrocarbon generation for oil-

prone source rocks generally commences at a vitrinite reflectance (Ro) of 0.5-0.6 % and is

complete at a vitrinite reflectance of 1.3-1.5 % Ro (Hunt, 1996). However, based on natural datasets and several hydrous pyrolysis experiments on immature humic coals of

Carboniferous through Tertiary age, a range of 0.85-1.8 % Ro would better define the “oil window” for humic coals (Type III kerogens) (Petersen, 2002). Rock samples from cores and crushed chips of shales are appropriate for use in hydrous pyrolysis experiments because they allow the embedded organic matter to mature in contact with natural mineral fabrics and interstitial water (Lewan, 1993). Cores are ideal for the experiment but can be difficult to acquire. Crushed rocks from unweathered outcrops ranging in size from 0.5 to 2.0 cm, can be a practical alternative

(Lewan, 1993). Fresh, immature (<0.6% Ro) source rock samples are hard to obtain because many outcrop samples are deeply weathered and have the potential to produce inaccurate geochemical information. Wells drilled deep into the source rock interval and cored deep enough to be able to get a 400 gram sample of immature rocks are very rare. Two Mancos Shale cores from the Douglas Creek Arch Area and one Mowry Shale core from the Rock Spring Uplift in the Greater Green River Basin were the only

47 available cores qualified for sampling. These core samples were collected from the Core Research Center of the U.S. Geological Survey (USGS-CRC) in Denver, Colorado. Table 4.1 and Figure 4.1 provide information on these three cores. The two Mancos B wells, which are about 19 miles apart (Figure 4.1a), have been cored deeply enough to reach approximately 13 ft. below productive reservoir rocks (the Mancos B Formation) into a condensed sections called the Upper Blue Gate Shale (Fisher, 2007). Well 15-29-4-103 GENTRY contains the Upper Blue Gate Shale from 3,392 ft to 3,404 ft (13 ft thick). Well 6-15-1S-103 FEDERAL contains the Upper Blue Gate Shale from 3,558 ft to 3,569.5 ft (12.5 ft thick). A thin slice of slabbed core samples was cut and sampled throughout both condensed sections and samples from the two wells were later composited in the 1:1 ratio. However, due to the limited availability of Mancos core material, the proportion of shales within each foot of the core was not the same. Appendix 4.1 provides a detailed composite description of each Mancos Shale core. Both condensed sections visually show homogeneous lithology but sedimentary structures within each condensed section are different. Well 15-29-4-103 GENTRY (Library Number D339) contains thinly laminated shale with natural fractures parallel to bedding; shale in Well 6-15-1S-103 FEDERAL (Library Number D351) is non-laminated and contains small spotted disseminated organisms (possible foraminifera because the core effervesced actively with HCl) (Appendix 4.2). The Mowry Shale core from the Green River Basin was from a well located at the Rock Springs Uplift drilled in the 1940’s (Figure 4.1b) and was obtained from the USGS- CRC facility. Although more than half of the core was missing (Mowry Shales ranged from 2,430 ft to 2,471 ft), it still provided enough shale material for hydrous pyrolysis experiments. This Mowry Shale core was immature (Figure 4.1). Five facies were identified and detailed core descriptions and pictures of the core are available in Appendix 4.2. Baxter Shale samples from outcrop were collected by Michael Lewan in 2003 from a fresh railroad cut in southwestern Wyoming (1/4 NE, section 10, T18N, R104W, Sweetwater County, Wyoming) (Figure 4.2). The outcrops are located on five terraces and are about four meters high (Lewan, 2010, personal communication). The sampled terrace consisted of dark-grey sandy mudstone and has 0.5 to 2.0 cm thick bedding

planes. The rock samples are calcareous and contain minor glauconite lenses. Three Baxter Shale samples from three different stratigraphic layers were collected from the outcrop and were later composited with a 1:1:1 ratio (Figure 4.2). Pyrite is visible throughout the three thin sections of the Baxter Shale samples, suggesting no or very little weathering occurred. Outcrop samples of Cameo Coal were obtained from several coal seams in Coal Canyon and McClane Canyon Coal Mine in western Colorado (Figure 4.3). Figure 4.3 provides the locations of the collected coal samples from Coal Canyon. Stop 1 is close to a small abandoned strip mine near the bend of the Coal Canyon. A previous coal-lease operator stripped off the carbonaceous mudrocks, thin coals, and sandstones immediately overlying the Cameo coal, creating a 45-ft highwall (Cumella and Cole, 2003). Along the Coal Canyon outcrop, several good exposures of coal seams occurred in both the western and eastern sides of Coal Canyon Road. A total of seven coal seam samples were collected in Coal Canyon. Two fresh coal samples were collected from the upper and lower coal seam in the McClane Canyon Coal Mine area (19 miles north of Loma in Highway 139, CO). The organic geochemistry of these Cameo Coal samples shows similar characteristics. A total of nine Cameo Coal samples were composited, according to their relative thicknesses in outcrop (Appendix 4.3). Mean vitrinite reflectance of these immature source rocks were analyzed by the Weatherford Laboratories in the Woodlands, Texas. At least 5 grams of each source rock was sent to the Weatherford Laboratories for vitrinite reflectance measurement. The average value of the vitrinite reflectance for immature Cameo Coal, Mowry, and Mancos Shale is 0.62%, 0.68%, and 0.67%, respectively (Appendix 4.4).

4.2 Source Rock Characterization: Source Rock Analyzer (SRA) The Weatherford Laboratories Instruments Division’s Source Rock AnalyzerTM (SRA) applies a Rock-Eval pyrolysis approach to rapidly characterize petroleum- generative potentials of source rocks. Pyrolysis is defined as the heating of organic matter in the absence of oxygen to yield hydrocarbons. Rock-Eval pyrolysis of whole rocks is a rapid open-system heating process in an inert atmospheric. It provides information on the

type of organic matter, evaluates organic richness, and quantifies the organic matter (Espitalié et al., 1977). The Weatherford SRA quantitatively determines: (1) The mass of free hydrocarbons (S1); (2) The amount of hydrocarbons generated from thermal cracking of

nonvolatile organic matter (S2); (3) The amount of carbon dioxide (CO2) produced during pyrolysis of kerogen up to a temperature of 400°C (S3). The S1 and S2 values are measured by using a Flame Ionization Detector (FID) while S3 is measured by using an infrared radiation (IR) detector. The SRA also determines the total organic carbon (TOC) of the sample and the temperature (Tmax) during the pyrolysis at which the maximum release of hydrocarbons occurs from cracking of kerogen.

4.2.1 Sample Analysis in the SRA Core samples from the Mancos and Mowry Shale and outcrop samples from the Cameo Coal and Baxter Shale were crushed, pulverized, and sieved through a 40-mesh sieve (about 0.422 mm). Samples of pulverized Mancos, Mowry, and Baxter Shale analyzed by the SRA were between 93 to 103 mg while Cameo Coal samples were only 5 mg because coals generally have very high TOC (> 60%). High TOC source rocks like coals need relatively small amount of samples in order to measure accurately TOC content under SRA detection limits. Accurately weighed, pulverized samples were put into an SRA crucible and placed in the SRA-Agilent autosampler. The sample within crucible is then raised and put into the oven at 325 °C. After a sample was put into the oven at 325 °C, it was rapidly heated to 340 °C. Rock samples were held isothermally at 340 °C for 3 minutes, allowing free hydrocarbons (S1) to be volatilized and detected by the FID. The CO2 liberated at the same time was detected by the IR cell (S3). After the isothermal heating, the oven temperature was ramped up to 640 °C at a programmed constant heating rate of 25°C/min, during which hydrocarbons were generated from the thermal degradation of

the kerogen (S2). The S2 is reported as milligram (mg) of S2 per gram of rocks and roughly indicates the generative potential of source rocks. The pyrolysis was carried under an inert helium atmosphere.

Table 4.1 Detailed core information of the two Mancos Shale cores and one Mowry shale core in Denver, Colorado. The two Mancos cores from the Douglas Creek Arch area are 18.88 miles apart from each other. Formation Operator Well Name Type Location Latit Longi Sample Lithol Field County Sta ude tude Interval(ft) ogy te Aspen MOUNTAIN 1 CR Full NWNW 41.557 109.058 2430 to shale Greater Green Sweet WY (Mowry) FUEL SUPPLY HETZLER (chips) sec 8 T18N 2495 River Basin water R103W Mancos COSEKA 6-15-1S-103 Slabbed NENESW Sec 15 39.960 108.945 3392 to shale Douglas Creek Rio CO RESOURCES FEDERAL T1S R103W 3403.5 Arch Area Blanco Mancos COSEKA 15-29-4-103 Slabbed SWNWNE sec 29 39.686 108.979 3358 to shale Douglas Creek Rio CO RESOURCES GENTRY T4S R103W 3569.5 Arch Area Blanco

51 6-15-1S-103 FEDERAL

15-29-4-103 GENTRY

4.1 a

Figure 4.1 (a) The location of the two Mancos Shale cores in the DCA (Modified after Kirschbaum and Roberts, 2005).

Jonah Field

1 CR Hetzler 52

4.1 b

Figure 4.1 (b) The location of the Mowry Shale core in the Rock Spring Uplift from Well 1 CR Hetzler. The well location is superimposesd on the maturity map (Modified after Kirschbaum and Roberts, 2005).

030514-01

1.9 m

53 030514-01: Sample of dark-grey sandy 030514-02 mudstone from a 25 cm interval. Sample is slaby to blocky

1.05 m 030514-02: Sample of dark-grey sandy mudstone from a 23 cm interval. Sample is 030514-03 slaby to blocky tracks 030514-03: Sample of dark-grey sandy mudstone from a 23 cm interval. Sample is slaby to blocky

Figure 4.2 Location Map of the Baxter Shale outcrop. Location: 1/4 NE, section 10, T18N, R104W, Sweetwater County, Wyoming. Three collection location of the outcrop Baxter Shale samples (Lewan, 2003, personal communication).

Figure 4.3 McClane Canyon Coal Mine and the location map of the eight sample locations along Coal Canyon.

After the heating run was complete, the oven was cooled to 580°C to ensure that no pyrolysis components were released during oxidation. Pure air was then flushed into the carrier gas line for 5 minutes to displace helium from the whole system. The remaining kerogen was oxidized while the oven was held at 580°C. During this time,

carbon monoxide (CO) and carbon dioxide (CO2) were released and measured by two

different IR cells to determine residual hydrocarbons (S4). A typical pyrogram is shown in Figure 4.4. A blank and a standard sample were run before each sequence to control the quality. A standard check was run after a whole sequence to again assure the detectable quality. If a sequence had more than 10 samples, a checked standard was run after every 7 samples. Data on the standard were provided in the result chapter.

TM Figure 4.4 Typical Weatherford Laboratories Instruments Division’s SRA-TOC output.

4.3 Hydrous Pyrolysis Experiments The most appropriate experimental technique to provide constraints on the composition of gas in the source rock is hydrous pyrolysis (Lewan, 1993 and 1994), which is expected to simulate the observed isotopic composition of natural system. A hydrous pyrolysis experiment involves heating a source rock or a kerogen sample in the presence of liquid water at subcritical temperatures in a closed reactor (Lewan et al., 1979; Lewan, 1985). Gas generated during the laboratory hydrous pyrolysis experiments

(Henry and Lewan, 1999; Lewan and Henry, 1999) can be collected and analyzed for comparison to gases present in the reservoirs.

4.3.1 Composite Sample Preparation Sample of the four source rocks, Cameo Coal, Mancos, Mowry, and Baxter Shales, were composited and crushed to gravel size (2.4-9.5 mm). The composite samples were sieved to three sizes: < 2.4 mm, 2.4 -9.5 mm, and > 9.5 mm. The resulting samples in the 2.4 to 9.5-mm fraction were collected for sequential hydrous pyrolysis experiments. The amounts of rock and water used in experiments were measured to ensure all the rock samples submerged in liquid water throughout the experiments, as well as to ensure the expanded volume of water at experimental temperature did not exceed the volume of the reactor. Average bulk density of each source rock sample was measured and calculated (Table 4.2). The specific volumes of water in the liquid and vapor phase at experimental conditions are determined from ASME Steam Tables (Table 4.2) (ASME, 1979). Using the equation presented by Lewan (1993), the experimental temperature was designed under 374 °C to avoid supercritical fluid phase. In each experiment, different amounts of crushed source rock and distilled water were loaded into the reactor, due to the limited availability of some source rocks. The weight of each source rock sample was measured, and the rock samples were placed into the reactor. A metal rod was used to ensure rock samples were leveled out and well compacted at the bottom of the reactor. A tight-fitting carbonized Cr-Ni screen was placed directly on top of the rock chips before the distilled water was added. For the Cameo Coal HP experiment, three extra equal-sized coal chips cut perpendicular to bedding from the same piece of coal were placed on top of the screen for determination of vitrinite reflectance after each sequential experiment. The reactor then was closed and evacuated for 5 minutes before distilled water was injected into the reactor. The loading procedure minimized the presence of pre-existing gas or water in the pores of rocks and maximized contact between water and the pore surfaces (Kotarba et al., 2009). After loading rock samples and distilled water into the reactors, the reactors were purged with

1,000 psia of helium to check for leaks and then vented until 25 psia of helium remained inside the reactor.

4.3.2 Experimental Conditions Hydrous Pyrolysis (HP) experiments (Figure 4.5) were conducted in 4 autoclave reactors made of Hastelloy C 276 (Parr Instrument Company, Illinois, US). The reactor was heated by an electrical heater (Parr Instrument Company, No. 4962, Illinois, US) which was controlled by a time-proportioning controller (Parr Instrument Company, No. 4821) with a Type J (iron-constantan) thermocouple in a stainless steel sheath (Lewan, 1983). Temperature was monitored by an additional Type J thermocouple (Omega Engineering, Inc., AHG-J-24) connected to a Fluke Monitoring System connected to a computer that connects temperatures every 30 seconds (Omega Engineering, Inc., Model 199). Gas was collected in two 30-cc stainless steel cylinders connected to the reactor, with one thermocouple and two manometers to monitor temperature and pressures during gas collection (Figure 4.6). Sequential HP experiments were employed in this study because of limitations in sample quantity. Experiments on four source rocks were carried out at 300 °C/ 72 hrs, 330 °C/ 72 hrs, and 360 °C/ 72 hrs. After 300 °C, at 72 hrs, gas was collected from the two stainless steel cylinders. The reactor was then opened to collect any oil that may have formed and suspended on the water surface. After collecting gas and oil, the reactor was closed and then run at 330 °C/ 72 hrs. The gas and oil collection was repeated and the final 360 °C/ 72 hrs run started. Final gas and oil collection was repeated after the final sequential experiment was done. For the Cameo Coal experiments, a piece of coal on top of the screen was removed after each sequential experiment.

4.3.3 Hydrocarbon Collection The reactor was air cooled to ambient temperature before collecting hydrocarbon samples. The reactor (reactor and gage block) was weighed to determine if any loss of mass occurred during experiment- all the experiments showed less than 1 gram of weight loss, indicating no leakage during any experiment. The observed small weight loss (less than 1 gram) is a result of the burning off the anti-seize paste on the bolt threads. With

57 the two evacuated gas cylinders closed, the gas control valve on the reactor gage block was then slowly opened until pressure and temperature of generated gas reach equilibrium in the whole system. The two stainless-steel 30-cc cylinders were opened to collect gas and closed when pressure drop stabilized. The reactor was disconnected from gas collection system and moved under a fume hood to vent off the remaining gas. Any liquid or water dropped from collection stem of the gage block when venting was collected and weighed. The expelled oil phase floating above water phase (if present) was first collected with a Pasteur pipette and transferred to a glass vial. This is designated as the free oil portion of the expelled oil. One of the three Cameo Coal chips was collected in each sequence in order to help determine the thermal maturity related to the experimental conditions later. The remaining dispersed expelled oil adhering to the reactor wall, pipette, reactor head, gage block, and Ni-Cr screen were collected with a benzene rinse. This benzene rinse was filtered through a 0.45µm Teflon hydrophobic filtered paper with vacuum air in order to get the equipment rinse and recovered water. This benzene rinse was designated as the equipment rinse portion of the expelled oil. The recovered equipment rinse was dried under hood and weighed until less than 0.003 g difference of weight appeared on the scale after 30 minutes. The recovered water was filtered through a 0.45µm hydrophilic filtered, collected and measured with Eh and pH meter.

4.4 Measured Vitrinite Reflectance Vitrinite Reflectance is the most commonly used maturation indicator. Coal is comprised of a number of distinct types of organic matter, called macerals (Crelling and Dutcher, 1980). Four Cameo Coal chips- CHIP#0 (original chip), CHIP#1 (300°C/72 hr), CHIP#2 (330°C/72 hr), and CHIP#3 (360°C/72 hr) –collected from the experiments were measured for their thermal maturity by vitrinite reflectance (Figure 4.7).

4.4.1 Sample Preparation Coal chips taken out from reactors of hydrous pyrolysis experiments were air-dried until less than 0.003 gram of moisture loss within a half hour. A small piece was taken from the original coal sample and put directly into a premade epoxy plug that has been

predrilled to a depth of about 2 to 3 mm (with a 5 mm milling cutter, a flat bottomed drill) (Pawlewicz, personal communication). The coal samples were then mixed with a two-part epoxy resin and stirred to ensure a good coal/epoxy contact. These were then set aside to allow the epoxy to cure. The preparations were then polished to a high drill) (Pawlewicz, personal communication). The coal samples were then mixed with a two- part epoxy resin and stirred to ensure a good coal/epoxy contact. These were then set aside to allow the epoxy to cure. The preparations were then polished to a high degree using a series of two grit paper sizes (first 60 grit and then 600) and two polishing compound (0.3 micron and then a 0.05 micron size) steps. This polishing imparts a near scratch-free surface on the coal chips which ensures more accurate reflectance measurements.

4.4.2 Measured Vitrinite Reflectance Reflectance was measured using a Zeiss Axiophot® reflected-light microscope which is fitted with a 50 X oil immersion lense and uses incident light (100 W tungsten lamp) filtered with a 546 nm (green) filter by Mark J. Pawlewicz at the U.S. Geological Survey Energy Resource Lab in the Denver Federal Center. The reflectance measurements were

made under oil immersion (Cargille PCB free Type A immersion oil, ne = 1.518 @ 23°C). The purpose of using oil immersion is to concentrate light from above and block surrounding light. The vitrinite particles were measured in a random orientation and a total of 25 measurements were acquired to get a mean random vitrinite reflectance (Rv, in %). The system was calibrated by synthetic glass and sapphire.

4.5 Gas Chromatography (GC) Analysis Whole oil analysis was performed using a Hewlett Packard 6890 Gas Chromatography (HP 6890 GC) equipped with a 60 m × 0.32 mm ZB-1 fused silica capillary column. Analyses were performed by Zachary K. Lowry with the U. S. Geological Survey at the Denver Federal Center. The GC was programmed from 40° to 320°C at 4.5°C/min and held for 18 minutes at final temperature (320°C). The Flame Ionization Detector (FID) output was digitized for calculating relative amounts of hydrocarbon components.

Table 4.2 Calculation of volume of water to use for hydrous pyrolysis experiments.

Mowry Shale Mancos Shale Baxter Shale Cameo Coal Temp of Run(°C) 300 330 360 300 330 360 300 330 360 300 330 360 From ASME Steam Tables Gamma l 1.4036 1.562 1.894 1.4036 1.562 1.894 1.4036 1.562 1.894 1.4036 1.562 1.894 Gamma v 21.643 12.967 6.943 21.643 12.967 6.943 21.643 12.967 6.943 21.643 12.967 6.943 Input Variables Density of Rock (g/cc) 1.844 1.844 1.844 1.889 1.889 1.889 1.911 1.911 1.911 1.165 1.165 1.165 Weight of Rock Samples (g) 300 300 300 325 325 325 400 400 400 200 200 200 Volume Reactor (cc) 1025 1025 1025 1025 1025 1025 1025 1025 1025 1025 1025 1025 Calculated Variables Volume Rock Samples (cc) 162.7 162.7 162.7 172.0 172.0 172.0 209.3 209.3 209.3 171.7 171.7 171.7 Volume remaining in reactor (cc) 862.3 862.3 862.3 853.0 853.0 853.0 815.7 815.7 815.7 853.3 853.3 853.3 Difference in gamma l and v 20.239 20.239 20.239 20.239 20.239 20.239 20.239 20.239 20.239 20.239 20.239 20.239

Water Mass used(g) 60 400 400 400 390 390 390 375 375 375 390 390 390 Liquid volume @ run Temp (ml) 540.6 592.3 718.3 526.2 575.8 695.8 506.3 554.3 670.7 526.2 575.7 695.6

The percent of reactor volume filled 69% 73.7% 86% 68% 73.0% 85% 70% 74.5% 86% 68% 72.9% 85%

Figure 4.6 Illustration of gas collection system.

Figure 4.7 Vitrinite (left) and inertinite (right) in the Cameo Coal chips under microscope.

4.6 Gas Composition and Isotope Analysis Chemical quantification was acquired by an HP 6890 series Gas Chromatography (GC) that has been custom configured by Wasson ECE Instrumentation for analysis of light natural gases. Analyses were performed by Augusta Warden with the U. S. Geological Survey at the Denver Federal Center. Gas samples were collected in 30-cc stainless steel (316) cylinders and then injected into the light-natural gas GC. The GC contains eight columns and three detectors (one FID and two TCDs) so that the hydrocarbon and nonhydrocarbon components were analyzed on a single injection. An autosampler was used to introduce samples to the Gas Chromatograph. Helium is the carrier gas for the FID. One thermal conductivity detector (TCD) analyzes the following components: carbon dioxide, ethane, ethylene, acetylene, hydrogen sulfide, argon/oxygen, nitrogen, methane and carbon monoxide and uses helium as a carrier gas. A second TCD analyzes for helium and hydrogen using nitrogen as the carrier gas. Standard gases of known chemical compositions are used to calibrate the GC. For the U.S.G.S. analyses, a specially-blended refinery hydrocarbon-gas standard supplied by Air Liquide was used to calibrate the instrument. All molecular gas data were reported in mole volume percent and calibrated to the standard in mole percent. The total molar gas yield from gas collection of each sequential hydrous pyrolysis experiment was calculated by using the Ideal Gas Law PV = nRT and molecular weight

of each component (Appendix 4.5). For each experiment, a total mass (in grams) for the measured gas components was determined. The pressure (P) of the gas-collection system was monitored and measured by two manometers. Room temperature (T) at the time of analysis was measured by a thermocouple placed inside the thermal well of the reactor. The reactor was weighed just before being connected to the gas collection system. The reactor was weighed again after venting off the remaining gas in the reactor. The difference before and after venting the reactor approximately represents the mass of gas generated from the reaction during heating. The mass of gas is mainly from gas accumulated in the headspace of the reactor, but a small amount of dissolved gas may also be included. Also included in the mass of the headspace gas is the weight of helium gas (≈ 25 psia) which was placed into the system prior to the experiment. The volume taken into account for the calculation with the idea gas law includes the volume (V) of the headspace of the reactor above the oil/water level, the volume of the gauge block, and the volume of the gas collection system. The volume of the gas collection system was pre-measured and formulated by Michael Lewan at the U.S. Geological Survey, Denver, CO as the equation of:

Head space volume: Vh = 30.4h-0.716 (where h is the height above oil/water surface); Gauge block volume: 8.2 cm3; Gas collection system volume: 21.1cm3

4.6.1 Gas Isotope Analysis Carbon isotopic analyses were also run on collected gases. The stable isotopes of carbon (12C and 13C) are determined by an isotope ratio mass-spectrometer. The following equation is used to calculate the ratio difference (δ) in per mil units, relative to a standard: 13 δ C (‰) = [(Rsample/ Rstandard)-1]*1000 R = 13C/12C Standard = Vienna-Pee Dee belemnite (VPDB)

A MicroMass Optima mass spectrometer equipped with a HP 6890 gas chromatograph and Carlo Erba elemental analyzer were used to determine carbon isotopic ratios. All analyses were run by Robert Dias and Mark Drier, at the U.S. Geological Survey Energy Resource Lab in the Denver Federal Center.

CHAPTER 5

RESULTS

This section discusses several subjects related to source rock properties, their petroleum generation capacities, stable carbon isotopes of natural gas generated from specific source rocks, and their relationship with kerogen types and thermal maturities.

5.1 Core Description of the Mowry Shale Core Five facies were identified in the incomplete Mowry Shale core (only 30 ft of core is preserved in a 57 ft section). Facies I consists of parallel-laminated shale with reworked shell debris and calcite cement. Facies II is slightly laminated and highly fractured shale with calcite cements. Facies III is thinly laminated bentonite. Facies IV consists of shale with densely distributed shells (both in-situ and reworked) with abundant fish scales. Facies V is shale with foraminifera beds. Facies VI is parallel- laminated shale with in-situ shell beds. A detailed core description is attached in Appendix 4.2.

5.2 Original Samples – Organic Geochemical Analyses The four representative source rocks in this study- Cameo Coal, Mowry, Mancos, and Baxter Shales- all exhibit distinctive geochemical characteristics. Screening pyrolysis analyses of the original (prior to the hydrous pyrolysis experiments) Cameo Coal, Mowry and Mancos Shale samples were done using the Weatherford Source Rock AnalyzerTM (SRA). Leco TOC data for the original Baxter Shale sample was provided by Michael Lewan of the U.S. Geological Survey, Denver, CO. Tables 5.1 to 5.4 list the results of organic geochemical analyses of the four source rock samples.

5.2.1 Organic Richness Table 5.5 gives the average geochemical parameters of the four source rocks - Cameo Coal, Mowry, Mancos, and Baxter Shales – prior to their being subjected to hydrous pyrolysis experiments. The nine outcrop Cameo Coal samples give the following geochemical attributes: TOC values range from 40 to 80 wt%, hydrogen index (HI)

ranges from 79 to 131 mg HC/g TOC, production index (PI) ranges from 0.02 to 0.05 mg HC/g TOC. The Mowry Shale core (20 samples) shows TOC ranging from 1.87 to 2.93 wt%, HI from 82 to 226, PI from 0.07 to 0.28. The two Mancos Shale cores (total 26 samples) give very different values: one has TOC ranging from 1.33 to 1.73 wt%, HI from 81 to 133, PI from 0.05 to 0.1 while the other Mancos core has TOC ranging from 1.37 to 3.09 wt%, HI from 97 to 258, PI from 0.06 to 0.12 (Figure 5.1). These two Mancos cores were blended together in a 1:1 ratio for the sequential hydrous pyrolysis experiment. (According to the relative thickness of the condensed sections from gamma ray logs, they should be mixed in the 2:3 ratio but due to the limited amount of available samples, a 1:1 ratio was applied when mixing). The three Baxter Shale outcrop samples give TOC values ranging from 0.53 to 0.72 wt%, HI from 72 to 84, PI from 0.06 to 0.10. The SRA pyrograms of Cameo Coal, Mowry Shale, and Mancos Shale are attached in Appendix 5.1. Based on the above results of the screening analyses, the four source rock samples are appropriately immature for the study using sequential hydrous pyrolysis.

5.2.2 Thermal Maturity

The nine outcrop Cameo Coal samples give Tmax values range from 426 to 438

°C. The Mowry Shale core (20 samples) shows Tmax from 433 to 446 °C. The two

Mancos Shale cores (total 26 samples) give Tmax from 433 to 438 °C and Tmax from 434 to 439 °C. The three outcrop Baxter Shale samples give Tmax from 436 to 439 °C (Figure 5.3). Kerogen in these source rocks is all thermally immature based on geochemical data, organic petrography, and vitrinite reflectance data (Table 5.5). The Cameo Coal

shows a vitrinite reflectance of 0.61-0.64 % Ro, Mowry Shale shows a vitrinite

reflectance of 0.66 to 0.70 % Ro, and the Mancos Shale shows a vitrinite reflectance of

0.63 to 0.71 Ro. The calculated vitrinite reflectance from average Tmax for the Cameo

Coal averages an Ro = 0.66%, Mancos Shale averages an Ro = 0.69%, Mowry Shale averages an Ro = 0.73%, and the Baxter Shale averages an Ro = 0.66 % (calculated Ro =

(Tmax*0.018)-7.16; Jarvie et al., 2001).

5.2.3 Organic Matter Type

Figure 5.2 shows plots of residual petroleum potential: S2 versus TOC content of the four source rocks indicate present-day source rock consisted of either Type III or mixed Type II/III kerogen. Figure 5.3 is a plot of hydrogen index versus Tmax temperature and Figure 5.4 shows a pseudo-van Krevelen diagram with HI and OI on the x and y- axes. Both diagrams indicate that the source rock samples can be genetically classified as having Type-III or mixed Type-II/III kerogens. The Cameo Coal samples are composed mainly of Type III kerogens. The Mowry, Mancos, and Baxter Shales are all composed of mixed Type II/III kerogens. It is worth noting the difference between the two Mancos cores, one plotting with a higher Type II content than the other.

5.3 Mature Samples after Hydrous Pyrolysis- Results of SRA As source rocks generated and expelled hydrocarbons, the amount of organic

matter in the source rock decreased. TOC and hydrogen index (S2/TOC) decreased because of the conversion of reactive kerogen. Table 5.7 shows the comparisons of SRA data of the original core/outcrop samples, the composite samples with different ranges of particle sizes (2.4 to 9.5mm and less than 2.4 mm), and the recovered source rocks after sequential hydrous pyrolysis. Figures 5.5 and 5.6 show the geochemical evolution of the organic matter in the four source rocks. With increasing thermal maturity, HI, OI, TOC,

and S2 decreased and Tmax increased in Mowry, Mancos, and Baxter Shales. The only one exception is the Cameo Coal. The TOC content of the Cameo Coal increased after the sequential hydrous pyrolysis (Figure 5.6). Coal generally usually contains about 60-80% carbon (mostly organic). During the maturation process, both carbon components and non-carbon components in the coal decreased. Therefore, the relative proportion of carbon components in coal may gradually increase with increasing thermal maturity.

5.4 Results of Sequential Hydrous Pyrolysis The oils generated during hydrous pyrolysis from each of the four source rocks were analyzed for their bulk compositions, and the generated gases were analyzed for their bulk and stable carbon and hydrogen isotopic compositions. The recovered water after hydrous pyrolysis experiments was analyzed for its Eh and pH values and the recovered

Table 5.1 Results of SRA data for original unheated Cameo Coal outcrop samples. The calculated percentage before composite is based on the relative thickness of exposed outcrops.

Location Sample ID S1 S2 Tmax (°C) S3 TOC (wt%) HI OI PI S1/TOC Percentage coal canyon 00203Ko 1.82 76.1 433.1 4.63 59.60 128 8 0.02 0.03 0.12 coal canyon 003Ko 2.63 60.18 425.5 23.25 60.19 100 39 0.04 0.05 0.05 McClane(upper) 00501Ko 2.02 94.00 437.3 11.15 77.81 121 14 0.02 0.04 0.16 McClane(lower) 00502Ko 1.48 72.2 438.1 14.10 72.87 99 19 0.02 0.02 0.14 coal canyon 006Ko 2.61 81.82 430.0 17.08 64.13 128 27 0.03 0.04 0.06 coal canyon 007Ko 1.36 35.00 433.0 25.54 47.24 74 54 0.04 0.03 0.05 coal canyon 009Ko 1.58 42.43 434.6 17.34 48.94 87 35 0.04 0.03 0.10 coal canyon 01001Ko 2.07 67.83 433.6 15.51 51.65 131 30 0.03 0.04 0.15 coal canyon 012Ko 1.88 33.65 435.1 16.85 42.43 79 40 0.05 0.04 0.17

Table 5.2 Results of SRA data for original unheated Mowry Shale core samples. Different facies were identified.

Sample Depth (ft) FACIES S1 S2 Tmax (°C) S3 TOC (wt%) HI OI PI S1/TOC 2430-2435 N/A 0.44 4.63 440.7 0.23 2.33 199 10 0.09 0.19 2430-2435 N/A 0.48 4.73 437.1 0.31 2.43 194 13 0.09 0.2 2430-2435 N/A 0.48 4.95 438.1 0.27 2.49 199 11 0.09 0.19 2430-2435 II 0.44 4.35 438.0 0.23 2.41 180 10 0.09 0.18 2430-2447 II 0.28 2.81 435.3 0.29 2.29 123 13 0.09 0.12 2430-2447 I 0.52 6.31 438.4 0.24 2.69 234 9 0.08 0.19 2430-2447 I 0.53 6.15 437.5 0.26 2.67 230 10 0.08 0.2 2430-2447(duplicate) I 0.54 6.15 437.9 0.28 2.69 229 10 0.08 0.2 2447-2454 II 0.16 1.71 435.8 0.31 2.09 82 15 0.09 0.08 2447-2454 (duplicate) II 0.3 2.22 438.0 0.28 2.11 105 13 0.12 0.14 2447-2454 II 0.32 2.18 434.5 0.24 2.10 104 11 0.13 0.15 2447-2454 N/A 0.35 3.79 440.0 0.26 2.17 175 12 0.08 0.16 2447-2454 N/A 0.45 4.12 438.7 0.31 2.42 170 13 0.10 0.19 2454-2459 IV 0.6 6.63 438.4 0.29 2.93 226 10 0.08 0.2 2454-2459 IV 0.52 6.46 437.5 0.36 2.9 222 12 0.07 0.18 69 2454-2459 I 0.41 3.84 439.7 0.27 2.31 166 12 0.10 0.18 2454-2459 N/A 0.33 3.68 439.6 0.32 2.4 153 13 0.08 0.14 2454-2459 N/A 0.35 3.94 439.6 0.32 2.64 149 12 0.08 0.13 2459-2463 N/A 0.34 3.59 437.1 0.34 2.63 137 13 0.09 0.13 2459-2463 V 0.33 3.91 435.3 0.34 2.69 145 13 0.08 0.12 2459-2463 I 0.56 5.77 442.7 0.27 2.86 202 10 0.09 0.2 2459-2463 I 0.55 5.71 440.3 0.25 2.84 201 9 0.09 0.19 2463-2471 IV 0.24 3.18 435.5 0.38 2.28 140 17 0.07 0.11 2463-2471 II 0.35 3.29 433.2 0.38 2.72 121 14 0.10 0.13 2463-2471 I 0.35 4.00 438.0 0.25 2.56 156 10 0.08 0.14 2463-2471 I 0.35 3.83 434.8 0.23 2.52 152 9 0.08 0.14 2471-2495 N/A 0.6 1.81 445.1 0.78 1.88 96 41 0.25 0.32 2471-2495 (duplicate) N/A 0.68 1.74 443.6 0.43 1.87 93 23 0.28 0.37 2471-2495 II 0.31 1.59 444.4 0.23 1.88 85 12 0.16 0.16 2471-2495 II 0.3 1.64 445.9 0.25 1.91 86 13 0.16 0.16 2471-2495 N/A 0.66 2.41 443.1 0.4 2.11 114 19 0.21 0.31 2471-2495 (duplicate) N/A 0.71 2.58 445.7 0.4 2.13 121 19 0.22 0.33

Sample Depth (ft) S1 S2 Tmax (°C) S3 TOC (wt%) HI OI PI S1/TOC 3392 0.14 2.18 434.1 0.15 1.64 133 9 0.06 0.09 3393 0.12 1.81 435.7 0.06 1.5 121 4 0.06 0.08 3394 0.15 1.6 434.6 0.17 1.51 105 11 0.08 0.10 3395 0.12 1.25 435.4 0.11 1.37 91 8 0.09 0.09 3396 0.16 1.45 436.2 0.16 1.48 98 11 0.1 0.11 3397 0.14 1.7 435.7 0.13 1.59 107 8 0.08 0.09 3398 0.1 1.12 433.2 0.15 1.33 84 11 0.09 0.08 3399 0.14 1.33 437.6 0.12 1.45 92 8 0.09 0.09 3400 0.13 2.06 436.1 0.16 1.73 119 9 0.06 0.07 3401 0.1 1.95 437.9 0.1 1.65 119 6 0.05 0.06 3402 0.11 1.56 433.5 0.15 1.5 104 10 0.07 0.07 3403 0.12 2.04 434.4 0.12 1.7 120 7 0.06 0.07 3404 0.13 1.4 435.1 0.19 1.42 99 14 0.08 0.09 70 3404(duplicate) 0.12 1.41 433.8 0.2 1.43 99 14 0.08 0.09

Sample Depth (ft) S1 S2 Tmax (°C) S3 TOC (wt%) HI OI PI S1/TOC 3558 0.5 7.41 438.8 0.25 3.09 240 8 0.06 0.16 3558.5 0.44 6.39 435.5 0.23 2.71 236 9 0.06 0.16 3559 0.43 5.62 435.6 0.19 2.63 213 7 0.07 0.16 3560 0.4 4.52 437.2 0.33 2.39 190 14 0.08 0.17 3561 0.46 4.69 437.7 0.21 2.34 201 9 0.09 0.2 3561.8 0.39 4.65 436.5 0.21 2.24 207 10 0.08 0.17 3562 0.47 5.1 438.8 0.23 2.45 209 10 0.08 0.19 3563 0.31 2.91 435.6 0.29 1.79 163 16 0.1 0.186 3564 0.39 4.4 437.9 0.18 2.41 183 8 0.08 0.16 3565 0.42 5.34 437.6 0.21 2.34 228 9 0.07 0.18 3565.8 0.39 5.13 438.0 0.19 2.26 227 8 0.07 0.17 3566 0.16 1.23 434.7 0.64 1.38 89 47 0.12 0.12 3566(duplicate) 0.16 1.2 438.2 0.63 1.37 87 46 0.12 0.11 3567 0.57 5.35 436.0 0.28 2.4 222 12 0.1 0.24 3568 0.57 5.87 436.8 0.21 2.7 217 8 0.09 0.21 3569 0.36 4.56 434.1 0.34 2.5 183 14 0.07 0.14 3569.5 0.61 7.04 437.7 0.23 2.73 258 8 0.08 0.22 70

Table 5.4 Results of Leco TOC data for original unheated Baxter Shale outcrop samples. Data is provided by USGS.

Sample ID S1 S2 Tmax (°C) S3 TOC (wt%) HI OI PI S1/TOC 030514-1 0.05 0.51 435.00 0.05 0.61 84.00 8.00 0.09 8.00 030514-2 0.04 0.38 435.00 0.08 0.53 72.00 15.00 0.10 8.00 030514-3 0.04 0.60 434.00 0.08 0.72 84.00 11.00 0.06 6.00

Table 5.5 Average geochemical parameters of the Upper Cretaceous Cameo Coal, Baxter, Mancos, and Mowry Shales from the Weatherford Source Rock Analyzer (SRA). UK: Upper Cretaceous.

COMPOSITE UK Cameo Coal UK Baxter Shale UK Mancos Shale UK Mowry Shale SAMPLES

Avg S1 (mg/g) 1.86 0.04 0.26 0.47 71

Avg S2 (mg/g) 64.76 0.5 3.09 3.97

Avg S3 (mg/g) 15.25 0.07 0.19 0.32

Avg Tmax (°C) 434.21 434.67 436.15 438.87 Avg TOC (%) 59.45 0.62 1.90 2.46 Avg HI 107.52 80.02 148.69 158.39 Avg OI 27.45 11.32 10.79 13.27 Avg PI 0.03 0.08 0.07 0.11

Avg S1/TOC 0.03 7.33 0.13 0.19 Calculated Avg 0.66 0.66 0.69 0.73 Ro (%)* *The calculated vitrinite reflectance (Ro: %) is based on the formula: Ro(calculated) = (Tmax*0.018)-7.16 (Jarvie et al., 2001)

Table 5.6 Measured vitrinite reflectance data of original source rock samples (Weatherford Laboratories, Texas).

WFT Source WELL Depth Vitrinite Reflectance UV Fluorescence ID Rocks NAME (ft) R.o.Ave. No. Conf. Qual. Form Color

BH-47028-NH000502 Mowry Shale R567 2430-2445 0.70 13 F G BH-47028-NH000503 Mowry Shale R567 2471-2495 0.66 11 F F BH-47028-NH000504 Mancos Shale D339 3403.6 0.63 5 P P BH-47028-NH000505 Mancos Shale D351 569.5 0.71 4 P F Algae YO-O BH-45573-NH000318 Cameo Coal outcrops Coal Mine 0.61 52 H F BH-45573-NH000319 Cameo Coal outcrops Coal Mine 0.64 49 H F Algae YO BH-45573-NH000320 Cameo Coal outcrops Coal Mine 0.61 59 H F Algae YO

Figure 5.1 Diagrams of TOC, S1, S2, and S3 versus depth (ft) in the two Mancos Shale cores.

18 Mancos D339 TYPE I

) Mancos D351 k

c 16 oil-prone

o Mowry Shale r

usually lacustrine g Baxter Shale

HC/ 14 g TYPE II m ,

2 oil-prone S

( 12 usually marine L Mixed TYPE II -III A I

T oil-gas-prone

N E

T 10

N P 8 RBO CA Organic Lean

DRO 6 TYPE III gas-prone HY G N I 4 N

I A M RE 2 TYPE IV inert 0 0.00.51.01.52.02.53.03.54.04.55.0 (A) TOTAL ORGANIC CARBON (TOC, wt.%)

300 TYPE I oil-prone TYPE II )

k usually lacustrine oil-prone c o 250

r usually marine g / HC

g m , 2 200 Cameo Coal (S Mixed TYPE II-III L Other Shales

A I oil-gas-prone NT

E T O 150 N P TYPE III

CARBO gas-prone O 100

HY G N I

N Organic

I A 50 Lean M

E R TYPE IV inert 0 0 1020304050607080 (B) TOTAL ORGANIC CARBON (TOC, wt.%)

Figure 5.2 Residual petroleum potential S2 versus TOC content for recognition of kerogen type for (A) Mowry, Mancos, and Baxter Shales and (B) Cameo Coals. Note the different scales on the two plots. Genetic boundaries are drawn after Langford and Blanc- Valleron (1990).

Mancos D339 Mancos D351 Cameo Coal Mowry Shale Baxter Shale 75

Figure 5.3 (Above) SRA hydrogen index versus Tmax temperature for kerogens from original source rocks.

Figure 5.4 (Right) Psuedo-van Krevelen diagram for kerogens in the original source rocks.

(A) Mancos Shale (B) Cameo Coal

After Sequential HP Original Composite (L)

Original Composite

After HP

Original Composite

After HP

Figure 5.5 Psuedo-van Krevelen Diagrams of HI versus OI for the Cameo Coal, Mowry, Mancos, and Baxter Shales before and after sequential hydrous pyrolysis experiments.

Table 5.7 Comparisons of SRA data before and after compositing samples and sequential hydrous pyrolysis. UK: Upper Cretaceous.

SAMPLE Sample size (mm) Sample weight (mg) S1 (mg/g) S2 (mg/g) Tmax(°C) S3 (mg/g) TOC (wt.%) HI OI PI S1/TOC

UK Cameo Original Composite(s) < 2.4 5 3.03 61.07 430.3 13.13 50.27 122 34 0.05 0.06 Original Composite(L) 2.4‐9.5 5.7 3.06 62.08 431.4 12.04 48.12 129 32 0.05 0.06 After Sequential 2.4‐9.5 7.6 16.41 32.39 509.3 0.01 55.34 59 0 0.34 0.30 After Sequential 2.4‐9.5 6 13.65 28.61 505.3 0.01 56.1 51 0 0.32 0.24 UK Mowry Original Composite(s) < 2.4 100.8 0.71 3.56 439.3 0.18 2.42 148 7 0.17 0.29 Original Composite(s) < 2.4 98.9 1.39 7.25 435.3 0.17 2.84 256 6 0.16 0.49 Original Composite(L) 2.4‐9.5 100.1 0.57 3.06 440.7 0.25 2.42 126 10 0.16 0.24 77 Original Composite(L) 2.4‐9.5 98.7 0.65 3.56 441.4 0.19 2.43 147 8 0.15 0.27 Original Composite(L) 2.4‐9.5 97.7 1.28 6.9 436.4 0.20 2.87 241 7 0.16 0.45 After Sequential 2.4‐9.5 97.5 0.11 0.39 458.3 0.04 2.02 19 2 0.22 0.05 After Sequential 2.4‐9.5 99 0.12 0.39 459.1 0.03 1.97 20 1 0.23 0.06 After Sequential 2.4‐9.5 101.6 0.17 0.65 456.1 0.04 1.99 33 2 0.21 0.09 UK Mancos Original Composite(s) < 2.4 97.4 0.37 2.59 438.1 0.16 1.88 139 9 0.13 0.20 Original Composite(L) 2.4‐9.5 100 0.38 2.69 438.5 0.16 1.81 148 9 0.12 0.21 After Sequential 2.4‐9.5 99.5 0.05 0.18 462.5 0.04 1.41 13 3 0.22 0.04 After Sequential 2.4‐9.5 99 0.08 0.31 459.0 0.04 1.44 21 3 0.21 0.06 UK Baxter Original Composite(s) < 2.4 100.5 0.03 0.34 439.1 0.06 0.82 42 7 0.07 0.04 Original Composite(s) < 2.4 96.3 0.04 0.4 432.9 0.06 0.85 47 7 0.08 0.05 Original Composite(L) 2.4‐9.5 100.7 0.02 0.3 440.1 0.03 0.58 45 5 0.06 0.03 Original Composite(L) 2.4‐9.5 100.9 0.04 0.59 434.9 0.06 0.76 78 7 0.06 0.05 After Sequential 2.4‐9.5 97.9 0.01 0.05 587.6 0.01 0.51 8 1 0.21 0.02 After Sequential 2.4‐9.5 98.7 0.02 0.06 551.4 0.01 0.51 9 1 0.24 0.04

(core) (core)

(outcrop)

(core) 78

Figure 5.6 TOC versus HI diagrams for the Cameo Coal, Mowry, Mancos, and Baxter Shales. In general, both TOC and S2 decrease with increasing maturity. Arrows indicate maturation direction.

rocks were analyzed for their organic geochemical properties after the experiments.

5.4.1 Thermal Maturity - Vitrinite Reflectance The vitrinite reflectance data includes analyses of Cameo Coal chips from each Cameo Coal sequential experiment. One of the three Cameo Coal chips was taken out from each sequential experiment during the collection phase and dried under the hood until less than 0.3 mg of change in mass was observed within 30 minutes. The measured vitrinite reflectance values from the Cameo Coal chips following the sequential experiments of 72 hours at 300°C, 330°C, and 360°C average 1.02%, 1.26%, and 1.63% respectively (Table 5.8; Raw data are available in Appendix 5.2).

Table 5.8 Measured vitrinite reflectance data of Cameo Coal chips representing equivalent sequential experimental conditions. Experimental N/A 300/72hr 330/72hr 360/72hr 360/72hr Conditions (Noble gas) Cameo Coal Chip # Original 1 2 3 4

Mean Ro (%) 0.58 1.02 1.26 1.63 1.64 Std. Deviation 0.02 0.04 0.03 0.03 0.02 Minimum 0.54 0.94 1.20 1.54 1.58 Maximum 0.61 1.08 1.31 1.67 1.70

5.4.2 Gas Chromatograms of Expelled Oil Twelve expelled oils generated from three sequential hydrous pyrolysis experiments for each of four samples were collected and analyzed by gas chromatography (GC). Figures 5.7 to 5.10 shows gas chromatograms of the generated oil from each source rock in the order of sequential experimental conditions (300°C/72h, 330°C/72h, 360°C/72h). The non-biomarker maturity parameters, isoprenoid/ n-alkane ratios and odd-to- even predominance index (OEP: Scalan and Smith, 1970) listed in Table 5.9 can be used

to characterize thermal maturity and redox conditions. Pristane (Pr: iC19) and phytane

(Ph: iC20) are common components in petroleum because they derive from either the phytol side chain of chlorophyll a in phototrophic organisms, or from the bacteriochlorophyll a and b in purple sulfur bacteria (e.g. Brooks et al., 1969; Powell and McKirdy, 1973; Peters et al., 2004). Reducing conditions promote cleavage of phytol to phytane (iC20) while oxic conditions favor conversion of phytol to pristane (iC19). 79

Therefore, the pristine/phytane (Pr/Ph) ratio can be used as a redox conditions indicator. The expelled oil from the Cameo Coal has a high Pr/Ph ratio (>3.0), suggesting input from terrestrial organic matter under oxydizing conditions. The expelled oils from the Mowry, Mancos, and Baxter Shales have Pr/Ph ratios of less than 2.0, indicating predominance of marine-sourced organic matter. The Pr/Ph ratio becomes significantly lower with increasing temperature in the Cameo Coal sequential experiment, while the Mancos and Baxter Shales show a slight decrease. The inverse trend is observed in the Mowry Shale, with increasing Pr/Ph ratios at higher temperatures.

The Pr/nC17 and Ph/nC18 ratios decrease with increasing thermal maturity in all experiments for all four source rocks (Table 5.9). This observation is consistent with previous studies, and the reduction of these two ratios can be attributed to more n-alkanes being generated from the cracking of kerogen (Tissot et al., 1971; Peters et al., 2004), and the fact that branched alkanes are less stable than n-alkanes (Fabuss et al., 1964, Hill et

al., 2003). Figure 5.11 shows the variation of Pr/nC17 and Ph/nC18 ratios for expelled oil

collected after each sequential experiment. The Pr/nC17 and Ph/nC18 plot can be used to infer oxidation/reduction and organic matter type in the source-rock depositional environment (Peters et al., 1999). Increasing thermal maturation displaces points toward the lower left of the diagram in general. The relative abundance of odd versus even n-alkanes can be used to estimate the thermal maturity of oil. Carbon preference index (CPI) or odd-to-even predominance (OEP) values significantly below or above 1.0 indicates low thermal maturity. Values of 1.0 or close to 1.0 suggest that an oil is thermally mature. In this study, Scalan and Smith’s (1970) odd-to-even predominance index (OEP) was applied to the maturation of oil. With increasing thermal maturity in the Cameo Coal, Mancos and Baxter Shales, OEP in oils decreases slightly, approaching to 1.0. The OEP of oils generated from the Mowry samples increased very slightly with increasing thermal maturity, trending away from 1.0 (Table 5.9).

nC24 nC26 iC18

nC22 nC29 nC20 nC28 HP-3479 nC14 nC13 nC17 nC18 iC16 nC16 Cameo Coal 300°C/ 72hr Pr/Ph = 6.78 Solvent nC12 nC30 CS2 iC14 Pr/nC17 = 1.47 iC13 Ph/nC18 = 0.22 OEP = 1.32 nC11 Ph Free Oil nC33 nC10

nC23 nC25 nC21 nC19 nC17 nC18 nC15 nC27 nC13 HP-3484 Cameo Coal

nC29 330°C/ 72hr nC11 Pr/Ph = 5.72 iC18 Pr Pr/nC17 = 0.44 iC16 Ph/nC18 = 0.08 nC31 OEP = 1.16 Free Oil nC 8 nC33 Ph nC35

nC11 nC12

nC13

nC15 HP-3487

nC9 nC17 Cameo Coal nC18 360°C/ 72hr

nC21 Pr/Ph = 2.42 nC nC 8 23 Pr/nC17 = 0.06

nC25 Ph/nC18 = 0.03 OEP = 1.10 Free Oil nC29 nC35

Figure 5.7 Gas chromatograms of generated oil from Cameo Coal after sequential hydrous pyrolysis of 300/72, 330/72, 360/72.

nC17

nC18 nC20

nC16

nC22 nC24 HP-3480 Mowry Shale 300°C/ 72hr nC 26 Pr/Ph = 1.49 nC28 Pr/nC17 = 0.34 Pr nC30 Ph/nC18 = 0.25 iC18 nC Ph 14 nC32 OEP = 1.00

nC34 iC16 Equipment Rinse

nC13 iC15

nC17 nC19

nC21

nC23 HP-3483 Mowry Shale nC15 nC25 330°C/ 72hr Pr/Ph = 1.56 nC27 nC29 Pr/nC17 = 0.30 Pr Ph/nC = 0.21 iC18 18 nC31 nC14 Ph OEP = 1.10 nC33

iC16 Equipment Rinse nC35

nC37 iC15 nC13

nC17 nC15

nC18 HP-3488 Mowry Shale nC21

nC14 360°C/ 72hr Pr/Ph = 1.66 nC23 Pr/nC17 = 0.31 nC25 Ph/nC18 = 0.21 Pr iC18 nC27 OEP = 1.14 nC13

iC16 nC29 Ph Equipment Rinse

iC15 nC31

nC12 nC33 iC13 nC35 nC11 nC37

Figure 5.8 Gas chromatograms of generated oil/equipment rinse from Mowry Shale after sequential hydrous pyrolysis of 300/72, 330/72, 360/72.

nC15

nC17 HP-3481

nC14 Mancos Shale Pr nC18 300°C/ 72hr Pr/Ph = 2.03 nC13 Pr/nC17 = 0.77 iC18 nC20 nC22 Ph/nC18 = 0.52 iC16 nC Ph 24 OEP = 1.24

nC12 iC15 nC26 Equipment Rinse

iC14 nC28

nC30

nC11 iC13 nC32

nC34

nC17

nC19 nC18 HP-3485 nC 21 Mancos Shale nC15 nC23 330°C/ 72hr nC25 Pr/Ph = 1.76

nC27 Pr/nC17 = 0.30 nC29 Ph/nC18 = 0.19

iC18 Pr OEP = 1.14 nC31 Ph Equipment Rinse iC16 nC33 nC13 nC35

iC15 nC37

nC39

nC17

nC19 HP-3489 Mancos Shale nC21

nC15 360°C/ 72hr Pr/Ph = 1.59 nC 23 Pr/nC17 = 0.12 Ph/nC18 = 0.08 nC25 OEP = 1.07 nC27 Equipment Rinse

nC13 nC29

nC31

nC33 nC35

Figure 5.9 Gas chromatograms of generated oil/equipment rinse from Mancos Shale after sequential hydrous pyrolysis of 300/72, 330/72, 360/72.

nC17 HP-3482 nC18 nC20 Baxter Shale

iC18 nC22 300°C/ 72hr Pr/Ph = 1.77

nC24 Ph Pr/nC17 = 1.13 nC26 nC28 Ph/nC18 = 0.75 iC16

nC15 nC30 OEP = 1.45 Equipment Rinse nC16 β-carotane nC32 iC15

nC34 iC13 iC14

nC19 nC21

nC18 nC23 HP-3486 Baxter Shale nC17 nC25 330°C/ 72hr nC27 Pr/Ph = 1.44 nC29 Pr/nC17 = 0.50

iC18 Pr Ph/nC18 = 0.30

Ph OEP = 1.21 nC31 Equipment Rinse nC15 nC33

iC16

nC35 nC37

nC23 nC21

nC25 HP-3490

nC27 Baxter Shale nC19 360°C/ 72hr nC18 Pr/Ph = 1.43 nC nC17 29 Pr/nC17 = 0.24 Ph/nC18 = 0.14 iC18 OEP = 1.18 nC31 Equipment Rinse Pr nC16 nC 15 Ph

nC33 nC14

nC35

nC37

Figure 5.10 Gas chromatograms of generated oil/equipment rinse from Baxter Shale after sequential hydrous pyrolysis of 300/72, 330/72, 360/72.

CPI(OEP) Calculation reference: Scalan and Smith, 1970

Figure 5.11Diagram of Pristane/ nC17 and Phytane/ nC18 for expelled oils generated from sequential hydrous pyrolysis of Cameo Coal, Mowry, Mancos, and Baxter Shales. These two ratios can be used to infer oxidation/reduction and organic matter type in the source rocks. With increasing thermal maturity, data points move towards lower left corner of the diagram. Modified after Peters et al., 1999.

Table 5.9 Isoprenoid/ n-alkane ratios and odd-to-even predominance index (OEP) of experiment-expelled free oils (from Cameo Coal) and equipment rinses (from Mowry, Mancos, and Baxter Shales).

Cameo Pr/Ph Pr/nC17 Ph/nC18 OEP 300/72 6.78 1.47 0.22 1.32 330/72 5.72 0.44 0.08 1.16 360/72 2.42 0.06 0.03 1.1

Mowry Pr/Ph Pr/nC17 Ph/nC18 OEP 300/72 1.49 0.34 0.25 1 330/72 1.56 0.3 0.21 1.1 360/72 1.66 0.31 0.21 1.14

Mancos Pr/Ph Pr/nC17 Ph/nC18 OEP 300/72 2.03 0.77 0.52 1.24 330/72 1.76 0.3 0.19 1.14 360/72 1.59 0.12 0.08 1.07

Baxter Pr/Ph Pr/nC17 Ph/nC18 OEP 300/72 1.77 1.13 0.75 1.45 330/72 1.44 0.5 0.3 1.21 360/72 1.43 0.24 0.14 1.18

5.4.3 Gas Yields and Their Bulk Molecular Composition The total molar gas yield from gas collection of each sequential hydrous pyrolysis experiment was calculated by using the Ideal Gas Law PV = nRT. Normalized gas compositions in mole percentage from each sequential hydrous pyrolysis experiment are shown in Table 5.10.

5.4.3.1 The Bulk Yields Gas generated at 300°C, 330°C, and 360°C for 72 hours had different instantaneous and cumulative yields from different source rocks, as shown in Table 5.11 and 5.12. Instantaneous yield indicates the gas yield from each sequential experiment while cumulative yield includes incremental gas yield from previous runs and gas yields from latest sequential experiment. Cumulative yields of hydrocarbons such as methane, ethane, propane, butane, and pentane, increased with increasing thermal temperature. On a per gram of total organic carbon (TOC) basis, the Cameo Coal had the highest methane and ethane yield throughout the experiments (Figure 5.12). After the sequential experiment at 360°C, cumulative yields of methane were Cameo Coal > Baxter > Mowry > Mancos Shale, whereas cumulative yields of ethane were Cameo Coal > Mowry Shale > Mancos ≈ Baxter Shale. The cumulative yield of methane from the Cameo Coal 360°C-run exceeded its 300°C-run yield by 8X. The Mowry Shale samples generated most cumulative propane and butane when temperature reached 360°C and the Baxter Shale generated the least. Surprisingly, the cumulative yields of propane and butane exceeded ethane in the Mowry and Mancos Shale. Although the investigation of the cause for different yields is beyond the scope of this study, the different kinetics of gas generation of different compounds within these source rocks, especially in the Mowry and Mancos Shale, might contribute to the variation of these gas yields. However, the instantaneous yields from each sequential experiment did not necessarily show an increase of hydrocarbon gases with increasing thermal stress. Instantaneous methane generation decreased after the 330°C run in the Mancos Shale (Figure 5.13). Instantaneous ethane generation in the Mancos Shale decreased first after the 300°C run but increased after the 330°C experiment (Figure 5.13).

Table 5.10 The normalized mole percentage of generated gas from each sequential experiment on source rocks.

Experimental Condition Rocks He H2 CO2 H2S O2 & Ar N2 CH4 C2H6 C3H8 iC4 nC4 iC5 nC5 nC6 nC7 Others Total 300/72 Cameo 14.0 0.9 81.3 0.3 0.0 0.2 2.0 0.5 0.4 0.1 0.1 0.1 0.0 0.0 0.0 0.1 100 330/72 Cameo 24.9 2.4 54.2 1.1 0.0 0.1 9.7 3.5 2.4 0.4 0.6 0.2 0.2 0.1 0.0 0.3 100 360/72 Cameo 21.5 4.0 33.1 1.0 0.0 0.0 25.1 8.1 4.4 0.5 1.2 0.2 0.4 0.2 0.1 0.3 100 300/72 Mowry 44.2 1.7 52.5 1.0 0.0 0.2 0.0 0.1 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.1 100 330/72 Mowry 56.1 11.5 24.7 5.3 0.0 0.1 0.6 0.5 0.2 0.0 0.1 0.0 0.1 0.0 0.0 0.7 100 360/72 Mowry 60.8 8.0 13.3 9.3 0.0 0.0 3.5 1.4 0.9 0.2 0.4 0.1 0.2 0.1 0.0 1.8 100 300/72 Mancos 25.8 1.3 71.9 0.7 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 100 330/72 Mancos 39.4 3.6 52.7 2.8 0.0 0.0 0.8 0.2 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.3 100 360/72 Mancos 45.7 3.8 42.6 5.3 0.0 0.0 0.8 0.6 0.3 0.1 0.2 0.0 0.1 0.0 0.0 0.6 100 87 300/72 Baxter 17.6 0.4 81.7 0.4 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 100 330/72 Baxter 31.3 1.2 65.2 1.7 0.0 0.1 0.3 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 100 360/72 Baxter 35.3 1.4 58.5 3.8 0.0 0.0 0.6 0.1 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.2 100

Table 5.11 Instantaneous yield of hydrocarbon gases (methane, ethane, propane, butane, and pentane) in sequential hydrous pyrolysis experiments.

Source Rocks Cameo Coal Mowry Shale Mancos Shale Baxter Shale TOC (wt %) 59.3 2.45 1.9 0.62 Temperature (°C) 300 330 360 300 330 360 300 330 360 300 330 360 Gases(mg/g TOC) Methane 0.63 1.75 5.19 0.00 0.80 4.50 0.46 1.83 1.62 0.00 2.35 3.79 Ethane 0.34 1.17 2.50 0.00 0.59 3.22 0.39 0.20 1.53 0.00 0.78 1.26 Propane 0.29 1.20 3.12 0.30 1.14 3.41 0.07 0.89 2.27 0.00 0.67 1.23 n‐Butane 0.12 0.40 0.92 0.32 0.64 1.89 0.00 0.60 1.27 0.00 0.52 0.71 88 iso‐Butane 0.11 0.26 0.39 0.06 0.20 0.78 0.00 0.17 0.45 0.00 0.26 0.24 Total Butane 0.24 0.66 1.31 0.39 0.83 2.67 0.00 0.78 1.71 0.00 0.77 0.95 Iso‐Pentane 0.07 0.14 0.21 0.16 0.18 0.57 0.00 0.21 0.37 0.00 0.00 0.29 n‐Pentane 0.06 0.16 0.34 0.24 0.36 0.97 0.00 0.43 0.74 0.00 0.32 0.29 Total Pentane 0.13 0.30 0.56 0.40 0.55 1.54 0.00 0.64 1.11 0.00 0.32 0.59 Gas Ratios

C1/C2 1.82 1.49 2.07 N/A 1.34 1.40 1.18 9.34 1.06 N/A 3.00 3.01

C2/C3 1.20 0.98 0.80 0.00 0.52 0.94 5.87 0.22 0.67 N/A 1.17 1.03

C3/C4 1.21 1.83 2.38 0.78 1.37 1.28 N/A 1.15 1.33 N/A 0.86 1.29

iC4/nC4 0.91 0.64 0.42 0.20 0.31 0.41 N/A 0.29 0.35 N/A 0.50 0.33

Table 5.12 Cumulative yield of hydrocarbon gases (methane, ethane, propane, butane, and pentane) in sequential hydrous pyrolysis experiments. Source Rocks Cameo Coal Mowry Shale Mancos Shale Baxter Shale TOC (wt %) 59.3 2.45 1.9 0.62 Temperature (°C) 300 330 360 300 330 360 300 330 360 300 330 360 Gases(mg/gTOC) Methane 0.63 2.38 7.56 0.00 0.80 5.30 0.46 2.30 3.92 0.00 2.35 6.14 Ethane 0.34 1.52 4.02 0.00 0.59 3.81 0.39 0.59 2.11 0.00 0.78 2.04 Propane 0.29 1.49 4.61 0.30 1.44 4.85 0.07 0.96 3.23 0.00 0.67 1.89 n‐Butane 0.12 0.52 1.45 0.32 0.96 2.85 0.00 0.60 1.87 0.00 0.52 1.23 iso‐Butane 0.11 0.37 0.76 0.06 0.26 1.04 0.00 0.17 0.62 0.00 0.26 0.49 Total Butane 0.24 0.89 2.21 0.39 1.22 3.89 0.00 0.78 2.49 0.00 0.77 1.72

89 Iso‐Pentane 0.07 0.21 0.42 0.16 0.34 0.92 0.00 0.21 0.58 0.00 0.00 0.29 n‐Pentane 0.06 0.22 0.56 0.24 0.61 1.58 0.00 0.43 1.17 0.00 0.32 0.61 Total Pentane 0.13 0.43 0.99 0.40 0.95 2.49 0.00 0.64 1.75 0.00 0.32 0.91

Figure 5.12 Cumulative yields of hydrocarbon gases of methane, ethane, propane, and butane from the four different source rocks- Cameo Coal (blue), Mowry Shale (pink), Mancos Shale (orange), and the Baxter Shale (green), on a per gram of TOC basis.

Figure 5.13 Instantaneous yields of hydrocarbon gases of methane, ethane, propane, and butane from the four different source rocks- Cameo Coal (blue), Mowry Shale (pink), Mancos Shale (orange), and the Baxter Shale (green), on a per gram of TOC basis.

5.3.4.2 Hydrocarbon Gas Compositions and Gas Dryness

Hydrocarbon gases (C1 to C5) generated from the Cameo Coal, Mowry, Mancos, and Baxter Shales under sequential hydrous pyrolysis experiments varied in their bulk

compositions. Figure 5.14 shows the relative amount of cumulative gas (C1 to C5) generated from each source rock. Methane was the dominant component in the Cameo Coal, Mancos Shale, and Baxter Shale during the experiments, but comprised less than 30% of the hydrocarbon gas generated from the Mowry Shale. Gases generated in the experiments varied in molecular composition as a function of kerogen type and experimental conditions. The dryness of the generated gases was calculated using (C1)/ (C1-C4). Cumulative dryness increased from zero to 0.38 to 0.51 and from zero to 0.73 to 0.73 in the Mowry and Baxter Shales, in the experiments at 300°C, 330°C, and 360°C, respectively. For both Mowry and Baxter Shales, gas dryness

increased significantly from an equivalent Ro of 1.0% to 1.3%. Cumulative gas dryness decreased from 0.72 to 0.71 to 0.55 in the Mancos Shale at equivalent vitrinite reflectance of 1.0 % to 1.3% to 1.6% (Figure 5.15). Gas dryness was high throughout the sequence of experiments in the Cameo Coal, remaining at approximately 0.6. The Baxter Shale had highest gas dryness (0.73). Both Mowry and Mancos Shales had low cumulative gas dryness (about 0.55). The relationship between dryness (or wetness) and thermal maturity is well established for Type I and Type II kerogens, but not for Type III kerogen (Whiticar, 1994).

The relative molecular compositions of C1/C2, C2/C3, C3/C4, and iC4/nC4 also varied with experimental conditions and different source rocks (Table 5.10). From

experimental temperature of 300 to 330 to 360°C, the instantaneous C1/C2 ratio of gases generated from Cameo coal changed from 1.82 to 1.49 to 2.07, and that from Mancos Shale changed dramatically from 1.18 to 9.34 to 1.06. No significant changes with increasing temperature occurred in C1/C2 in the Mowry and Baxter Shale experiments.

The C2/C3 ratio of gases generated from Cameo Coal decreased from 1.20 to 0.98 to 0.80

but increased from the Mowry Shale from 0 to 0.52 to 0.94. The C2/C3 ratio of gases generated from the Mancos Shale varied from 5.87 to 0.22 to 0.67. Ranges for experimentally generated gas ratios correspond to the typical ranges for natural gases

(Nikonov, 1972: C2/C3 = 1.5-5.0 and C3/C4 = 1-3; Oudin, 1993: iC4/nC4 < 1 for thermogenic gases generated during catagenesis).

5.4.3.3 CO2 and H2S The normalized total gas yield was then compared to the mass difference from before and after venting the reactor for each experiment. It is very likely that the gas dissolved in water came out of solution during the abrupt pressure drop when the reactor was vented. The amount of this dissolved gas was estimated by the difference between the calculated mass (using PV = nRT) and the measured loss of mass (before and after venting the reactor). The gas dissolved in the water would most likely be carbon dioxide

(CO2) and some hydrogen sulfide (H2S) since these compounds have relatively high solubilities in water. Figure 5.16 shows the comparison of the calculated mass of gas and the measured weight change of the reactor before and after venting. The Baxter Shale data deviated the most from 1:1 line, suggesting that it had the highest level of gas dissolved in water. This observation suggests that the Baxter Shale generated the most

CO2, and the Cameo Coal generated the second most.

5.4.4 Non-hydrocarbon Gases

Significant amounts of non-hydrocarbon gases such as nitrogen (N2), carbon

dioxide (CO2), hydrogen (H2), and hydrogen sulfide (H2S) were also generated from the four source rocks during hydrous pyrolysis experiments. Table 5.13 shows the instantaneous yields of these gases accumulated in the headspace of the reactors. These non-hydrocarbon gases, especially CO2 (solubility in water ≈1.7g/L @20°C, 1atm) and

H2S (solubility in water ≈4g/L@20°C, 1atm), have relatively high solubility in water; therefore the total amount of non-hydrocarbon gases generated from each rock includes both the aqueous phase of gas dissolved in water and the free gas in the headspace of the reactor. In general, the samples that contain mixed Type II/III organic matter (Baxter, Mancos, and Mowry Shale) generate more non-hydrocarbon gases than the Type III organic matter sample (Cameo Coal). Carbon dioxide was by far the largest contributor to the non-hydrocarbon gases. On a per gram of total organic carbon (TOC) organic matter

Cameo Coal Mowry Shale 8.0 6.0

7.0 5.0 ) ) C C O 6.0 T O g T / g 4.0 / 5.0 (mg

Methane (mg

l d e l

4.0 Ethane e 3.0 Yi

Yi e

v Propane e v ti i a 3.0 t l Butane a l u 2.0 Methane u

m Pentane

m 2.0 Ethane Cu Cu 1.0 Propane 1.0 Butane Pentane 0.0 0.0 300 330 360 300 330 360 Temperature (C) Temperature (C)

94 Mancos Shale Baxter Shale 4.0 6.5 6.0 3.5 5.5 ) Methane C 5.0 O

) 3.0 T Ethane C g

/ 4.5 O

T Propane g 2.5

/ 4.0 (mg Butane d l 3.5 e (mg Pentane

2.0 Yi d

l 3.0 e e v i Yi

t 2.5 e 1.5 Methane a l v u ti 2.0 a

Ethane m l

u 1.0 Propane Cu 1.5 m

Cu Butane 1.0 0.5 Pentane 0.5 0.0 0.0 300 330 360 300 330 360 Temperature (C) Temperature (C) Figure 5.14 Cumulative yields of hydrocarbon gases of methane, ethane, propane, and butane in Cameo Coal, Mowry, Mancos, and Baxter Shale, on a per gram of TOC basis. Gas: C1 (blue), C2 (red), C3 (green), C4 (orange), and C5 (light blue).

Figure 5.15 Change of dryness of generated gases with increasing thermal maturity. Gases generated from the Cameo Coal (blue) show dryness slightly changed from 0.65 to 0.60 to 0.63. Gases generated from the Mowry Shale (pink) show dryness increases from 0.00 to 0.38 to 0.51. Gases generated from the Mancos Shale (orange) show dryness decreases from 0.72 to 0.71 to 0.55. Gases generated from the Baxter Shale (green) show dryness increases from 0.00 to 0.73 and stay constant at 0.73.

hydrocarbon gases. On a per gram of total organic carbon (TOC) basis, the Baxter Shale

generated the most CO2, H2S, and H2 throughout the sequential experiments (Figure

5.17), while the Mancos and Mowry Shales generated intermediate amounts, and the Cameo Coal generated the least amount. Carbon dioxide occurred in the analyzed natural gases in concentrations from 13.29 to 81.65 mol. % (Table 5.10). The high content of

CO2 from the Baxter Shale (5048 mg/ g TOC) might be indicative of the incorporation and decomposition reactions of inorganic sources, such as carbonate materials, with increasing thermal stress, a possibility discussed more fully in Chapter 6. The instantaneous yield of CO2 was highest after the first isothermal heating of 300°C/ 72h and decreased considerably for sequential heating iterations after 330°C/ 72h and 360°C/ 72h (Table 5.11). The δ13C values of carbon dioxide generated from the experiments varied for 13 different source rocks (Figure 5.18). The δ C values of CO2 ranges from -26.28 ‰ to -

24.95 ‰ in the Cameo Coal, from -7.25 ‰ to -5.97 ‰ in the Mowry Shale, and from - 1.65 ‰ to 1.62 ‰ in the Mancos and Baxter Shales (Figure 5.18).

Comparison of mass of gas 14 y = 1.5781x - 2.4245 R² = 0.9982 12 Cameo Coal

) y = 1.9045x - 2.995

10g R² = 0.9998 ( Mowry g

in Shale 8 nt e Mancos V Shale

ter 6

af Baxter y = 1.3987x - 1.1709 4 R² = 0.9915 Shale s Loss

Ga 2

of y = 1.0263x - 0.3217

t R² = 0.9344 0 igh

e 02468101214

W Calculated Total Gas (g)

Figure 5.16 The diagram shows the amount of dissolved gas estimated by the difference between the calculated mass (using PV = nRT) on the x-axis and the measured loss of mass (before and after venting the reactor) on the y-axis from the Cameo Coal(blue), Mowry Shale (pink), Mancos Shale (orange), and the Baxter Shale (green). The Baxter Shale (in green) shows the largest difference between calculated headspace gas and measured gas loss before and after venting the reactor, suggesting possibly the highest level of high-solubility gas dissolved in water. The gas is most likely to be carbon dioxide or hydrogen sulfide.

Hydrogen was generated in the second largest quantities, after CO2 of the non- hydrocarbon gases, occurring in concentrations from 0.37 to 11.47 mol. % (Table 5.10).

Hydrogen (H2) can be derived from both water and organic matter. Hydrogen can also participate in the formation of H2S, making it complicated to interpret its behavior and quantify its actual amount. On a per gram of TOC basis, the Baxter Shale generated the greatest amount of hydrogen, and the Cameo Coal generated the least amount. With increasing thermal stress, the hydrogen yield increased only slightly in the Cameo Coal

(Type III) but increased dramatically in the other shales, which contain mixed Type II/III organic matter (Figure 5.17).

The generation of hydrogen sulfide (H2S) took place primarily in the 300°C and 330°C run and very little was generated in the 360°C run. On a per gram TOC basis, the

Baxter Shale generated the most H2S (g), Mowry Shale the second, Mancos Shale the

third, and the Cameo Coal generating the least H2S (Figure 5.17). The amount of

hydrogen sulfide from each sequential experiment was generally less than 5 mol % (Table

5.10). An exception was the Mowry Shale, which generated 9.28 mol % of H2S during the 360°C/72h run. Nitrogen was also produced during the experiments in concentrations from 0 to 0.23 mol. %. Figure 5.17 shows that on a per-gram TOC basis, the Mowry Shale generated more than 3.5 mg of cumulative nitrogen at experimental temperature of 360°C. The Baxter and Mancos Shale generated about an equal amount of nitrogen of 2.7 mg at experimental temperature of 360°C. In order to quantify the amount of non-hydrocarbon gases generated in each experiment (especially CO2 and H2S), the quantity of each gas plus their related aqueous species must be determined. Solubility calculations for CO2 and H2S required using pH values of the recovered water and followed the method published in Lewan, 1997 (Table 7), in which only one species of gas assumed to exist in the water phase at one time. Because the recovered water was not collected until the end of the three sequential experiments, only the gas generated from the last isothermal heating at 360°C/72h could be calculated. Significant amounts of both CO2 and H2S gas were dissolved in the recovered water (Table 5.13). Approximately, the Baxter Shale had the most dissolved

CO2 in the water phase. In all source rocks, significant proportions of hydrogen sulfide gas were dissolved in the water.

5.4.5 Eh and pH Value of the Recovered Water The Eh, pH value, and the measured temperature of the recovered water were measured after the collection phase was finished. The negative Eh value of the recovered water indicates strong reducing capability of the source rock. The reducing capacity of a source rock is related to its amount of hydrogen generated. Among the Type II/III marine

shales, Mowry has the highest TOC content, Mancos has the second highest, and the Baxter has the least. This trend is reflected in the Eh value where water from the Mowry Shale experiment has the most negative Eh value (Eh: -355.80). Although the Type III terrestrial organic matter (Cameo Coal) has TOC of around 60 to 80 wt%, much higher than the mixed Type II/III organic matter (marine shales), the Type III organic matter shows weaker reducing capability to the water in the reactor, resulting in a less negative Eh of -234.10.

Source Rocks Cameo Mowry Mancos Baxter pH 4.79 6.35 6.41 6.45 Eh ‐234.10 ‐355.80 ‐352.50 ‐345.00 Temperature (°C) 19.70 19.70 19.70 19.60

5.4.6 Isotopic Compositions of Natural Gas The cumulative δ13C values of generated hydrocarbon gases from the four source rocks are given in Table 5.16. No or very little methane, ethane, and propane were generated from the Mowry, Mancos, and Baxter Shale at the first 300°C run, so the carbon isotopic data from these species are not available. The cumulative carbon isotope 13 (δ C) of iso-butane (iC4), n-butane (nC4), iso-pentane (iC5), and n-pentane (nC5) of the

Cameo Coal were measurable but the data should be used with caution as the iC4, nC4,

iC5, nC5 all represent very small peaks outside the normal analytical window. Figure 5.19 illustrates the cumulative δ13C values of the generated hydrocarbon gases from the Cameo Coal, Mowry, Mancos, and Baxter Shales with their reciprocal carbon numbers on the x-axis. Carbon isotope ratios (δ13C) of methane generated from these four source rocks all become lighter with increasing experimental thermal temperatures. This result is opposite to most of the previously published data in the field which showed the carbon isotope ratio of gas become progressively enriched in 13C, with increasing thermal stress (Faber, 1987; James, 1983; Whiticar, 1994).

Table 5.13 The estimated CO2 and H2S gas dissolved in the recovered water after the sequential experiment.

Experimental Conditions 360°C/72h Source Rocks Cameo Mowry Mancos Baxter

CO2(g) 2.274 0.341 1.485 2.605

CO2(aq) 3.842 1.057 4.949 9.253

Total CO2 (gram) 6.116 1.398 6.434 11.858

H2S(g) 0.054 0.184 0.142 0.130

H2S(aq) 0.271 1.072 0.845 0.800

Total H2S (gram) 0.326 1.256 0.987 0.930 pH 4.799 6.342 6.409 6.451

Table 5.14 Instantaneous yield of non-hydrocarbon gases (CO2, H2S, H2, and N2) in sequential hydrous pyrolysis experiments.

Source Rocks Cameo Coal Mowry Shale Mancos Shale Baxter Shale TOC (wt. %) 59.3 2.45 1.9 0.62 Temperature (°C) 300 330 360 300 330 360 300 330 360 300 330 360 Gases(mg/gTOC)

Carbon Dioxide (CO2) 69.89 26.93 18.76 257.11 91.75 46.37 702.49 343.98 240.39 2727.41 1271.01 1049.13

Hydrogen (H2) 0.18 0.43 0.45 3.68 15.33 25.07 5.37 14.16 22.97 9.05 25.98 52.37

Hydrogen Sulfide (H2S) 0.10 0.02 0.000 0.73 0.31 0.02 0.25 0.17 0.00 1.06 0.87 0.00

Nitrogen (N2) 0.04 0.05 0.10 0.39 1.95 1.27 0.58 1.08 0.97 0.57 1.09 1.18

Table 5.15 Cumulative yields of non-hydrocarbon gases (CO2, H2S, H2, and N2) in sequential hydrous pyrolysis experiments.

Source Rocks Cameo Coal Mowry Shale Mancos Shale Baxter Shale

100 TOC (wt. %) 59.3 2.45 1.9 0.62 Temperature (°C) 300 330 360 300 330 360 300 330 360 300 330 360 Gases(mg/g TOC)

Carbon Dioxide (CO2) 69.89 96.82 115.58 257.11 348.86 395.23 702.49 1046.47 1286.86 2727.41 3998.42 5047.55

Hydrogen (H2) 0.18 0.61 1.06 3.68 19.01 44.08 5.37 19.53 42.49 9.05 35.03 87.40

Hydrogen Sulfide (H2S) 0.10 0.12 0.12 0.73 1.04 1.06 0.25 0.41 0.41 1.06 1.93 1.93

Nitrogen (N2) 0.04 0.09 0.19 0.39 2.34 3.61 0.58 1.66 2.64 0.57 1.66 2.84

100

101

Figure 5.17 Cumulative yields of non-hydrocarbon gases (CO2, H2S, H2, and N2) in sequential hydrous pyrolysis experiments from Cameo Coal (blue), Mowry Shale (pink), Mancos shale (orange), and Baxter Shale (green).

101

Figure 5.18 The cumulative δ13C values of the carbon dioxide generated from the 300 °C, 330 °C and 360 °C experiments from four source rocks- Cameo Coal (blue), Mowry Shale (red), Mancos Shale (green), and Baxter Shale (purple).

C3H8

0.33 0.5 1.0

Figure 5.19 The cumulative δ13C of methane, ethane, and propane generated from hydrous pyrolysis experiments versus the reciprocal of their carbon number. Cameo Coal (green), Mowry Shale (orange), Mancos Shale (purple), and Baxter Shale (blue).

102

Table 5.16 Stable carbon isotopes of gases generated by sequential hydrous pyrolysis.

13 13 13 ) Experimental #, Rocks δC1 δ C2 δ C3 C1 C2 C3 C1/(C2+C3) C2/C1 iC4 C2/C3 C2/iC4 ln(C1/C2 ln(C2/C3) ln(C2/C3) Conditions(°C/h) (‰) (‰) (‰) HP‐3479 300/72 Cameo ‐34.7 ‐29.9* ‐29.5 2.00 0.49 0.40 2.25 0.25 0.10 1.23 4.90 1.41 0.20 0.20 HP‐3484 330/72 Cameo ‐37.3 ‐30.4 ‐29.4 9.67 3.53 2.36 1.64 0.37 0.39 1.50 9.05 1.01 0.40 0.40 HP‐3487 360/72 Cameo ‐38.7 ‐29.4 ‐28.6 25.09 8.05 4.40 2.02 0.32 0.52 1.83 15.48 1.14 0.60 0.60 HP‐3480 300/72 Mowry N/A N/A N/A 0.00 0.09 0.00 0.00 N/A 0.01 N/A 9.00 N/A N/A N/A HP‐3483 330/72 Mowry ‐36.8 ‐33.1 ‐32.6* 0.59 0.45 0.16 0.97 0.76 0.04 2.81 11.25 0.27 1.03 1.03 HP‐3488 360/72 Mowry ‐39.4 ‐32.8 ‐32.8 3.54 1.43 0.92 1.51 0.40 0.17 1.55 8.41 0.91 0.44 0.44 HP‐3481 300/72 Mancos N/A N/A N/A 0.13 0.01 0.04 2.60 0.08 0.00 0.25 N/A 2.56 ‐1.39 ‐4.17 HP‐3485 330/72 Mancos ‐36.9 N/A ‐31.8* 0.77 0.20 0.03 3.35 0.26 0.02 6.67 10.00 1.35 1.90 ‐4.71 HP‐3489 360/72 Mancos ‐37.5 ‐32.3 ‐31.7 0.79 0.59 0.27 0.92 0.75 0.06 2.19 9.83 0.29 0.78 ‐1.22 0.00 0.00 0.00 0.00 0.00 0.00 N/A N/A N/A N/A N/A 103 HP‐3482 300/72 Baxter N/A N/A N/A HP‐3486 330/72 Baxter ‐36.4 ‐43.4* ‐33.9* 0.33 0.05 0.04 3.67 0.15 0.01 1.25 5.00 1.89 0.22 ‐4.52 HP‐3490 360/72 Baxter ‐37.2 ‐38.9* ‐33.8* 0.58 0.10 0.07 3.41 0.17 0.01 1.43 10.00 1.76 0.36 ‐3.89

Experimental Conditions Rocks iC4(‰) nC4(‰) iC5(‰) nC5(‰) HP-3479 300/72 Cameo -28.42* -28.70* -27.38* -27.71* HP-3484 330/72 Cameo -28.97* -28.58* -28.02* -28.23* HP-3487 360/72 Cameo N/A N/A N/A N/A HP-3491 360/72 Cameo -28.63* -28.57* -28.18* -28.10* *Data in grey boxes are measurable and repeatable but high errored because they are barely above the detection limits.

103

104

CHAPTER 6

DISCUSSION

6.1 Mass Balance Mass balance calculations are presented here to show the mass change due to the artificial thermal maturation of the source rocks. The total mass of expelled hydrocarbons plus the recovered rocks should be equal or very close to the mass of the original rocks placed in the reactor before heating. Table 6.1 shows the recovery percentage for each sample before and after the sequential hydrous pyrolysis experiments. The Cameo Coal has about 85 % recovery, the Mowry and Mancos Shales have near 98 % recovery, and the Baxter Shale has 100 % recovery. These high recoveries indicate that sample handling and quality control were good throughout the experimental procedures and no leakage occurred. The lower recovery rate in the Cameo Coal might be related to its high moisture content or its components and structures. Coal has more heteroatoms such as N, O and S, making its chemical structure relatively easy to break down during thermal maturation. The breakdown of the coal structure results in the release of water, gas, or other small molecules originally trapped within the coal. The tendency of the trapped water and low-molecular volatile material to escape is high during either the collection phase or when the recovered rocks were dried under a fume hood for several days. Therefore, we can expect a relatively low recovery of coal samples when compared to shale samples.

Table 6.1 Recovery percentage of Cameo Coal, Mowry, Mancos, and Baxter Shales after sequential hydrous pyrolysis at 300, 330, and 360°C for 72 h. Sequential Total Weight Loss Recovered Total Original Recovery HP Expelled after venting Rocks (g) Recovered Rock (g) (%) Experiments Oil (g) the reactor(g) (g) Cameo Coal 4.929 20.4 155.5 175.934 212.05 85.28 Mowry Shale 0.677 3.1 291.1 294.2 300 98.29 Mancos Shale 0.503 8.8 310 318.8 325.2 98.19 Baxter Shale 0.049 16.2 384.9 401.1 400.5 100.16

104

6.2 Products Yield The amount of oil and gas generated under different experimental conditions (300°C/72h, 330°C/72h, 360°C/72h) has been shown to vary with source rock and kerogen type (Lewan, 1983). Lewan (1983) suggested that in a non-sequential experiment at 330°C for 72 h, Type II kerogen is transformed to bitumen with only small amounts of bitumen transformed to oil, while in a non-sequential experiment at 360°C for 72 h with the same kerogen, nearly complete transformation of bitumen to oil occurred, with only small amount of oil cracking to gas. Experiments conducted here were sequential hydrous pyrolysis experiments at 300, 330, and 360°C for 72 h each. These conditions were equivalent to measured thermal maturities of 1.0% Ro, 1.3% Ro and 1.6% Ro. Based on previous studies (Schoell, 1983), I conclude that gas in the experiments was mainly generated from the primary cracking of organic matter, not from the secondary cracking of oil. In this section, the normalized gas yield (on a TOC basis) in the experiments and compared to the published estimated field gas production or reserves at the Jonah Field.

6.2.1 Hydrocarbon Gas Yields in the Jonah Field and southern Piceance Basin For Type II or mixed Type II/III kerogen-bearing marine source rocks, the thermal maturity of 1.6 % Ro represents a maturation stage just above the oil generation window, into the wet gas window (Dow, 1977). This suggests that conditions during experiments only simulate gas generated from the primary cracking of source rocks. For Type III kerogen-bearing terrestrial source rocks, this level of thermal maturity also represents completion of the oil generation stage but marks only the beginning of the dry gas window (Dow, 1977; Behar et al., 1992, 1997). Dry gas may be generated from humic coals from a vitrinite reflectance of about 1.5 % up to 3.0 % Ro (Peterson, 2006), which suggests that large quantities of gas generated in the dry gas window was not simulated in this study. Literature exists on the study and modeling of the gas generation capabilities of possible source rocks in both Jonah Field and the Piceance Basin. Coskey (2004) applied 1D burial history modeling and petroleum system analysis in the SHB 13-27 well in Jonah Field, showing that the Mesaverde and lower Lance Formations have only reached

105

the middle mature oil window (0.7 to 1.0 % Ro) (Figure 6.1). However, the Hilliard (Baxter equivalent) and Mowry Shales both entered the main gas generation window (1.3 to 2.6 % Ro) as early as 47 Ma (Figure 6.1). His study suggested that the present-day gas production might be sourced from a mixture of primary cracking from coals and carbonaceous shales in the Mesaverde and Lance Formations, and from the secondary cracking of oil generated from the Hilliard and Mowry Shale. In addition, a larger fetch area might also be available based on regional geology and basin paleogeometry (Coskey, 2004). The regional pressure and structural gradients at the time of expulsion may favor the direct gas migration from Pinedale Anticline toward the Jonah Field area (Coskey, 2004). The laboratory-generated primary-cracking gas from each source rock can provide a quantification of gas yields. From the cumulative methane yield of source rocks in this study, the estimation of total cumulative methane yield from the Mesaverde coal,

Baxter and Mowry Shales in Jonah Field, is about 19 mgCH4/ gTOC, assuming that the Mesaverde coal and Cameo Coal have similar generative capacity for methane gas1. If the assumption is correct, the cumulative ethane yield is expected to be about 9.87 mgC2H6/ gTOC, propane about 11.35 mgC3H8/ gTOC, and butane about 7.82 mgC4H10/ gTOC in the Mesaverde coal, Baxter and Mowry Shales (Table 5.12). Using the generalized stratigraphic column and TOC values from Coskey (2004) to calculate the hydrocarbon gas generation (C1 to C4) from primary cracking in the Lance and Mesaverde carbonaceous shales, Mesaverde coal, Baxter and Mowry Shales, the estimated gas generation is only 18.3 tcf2 (based on a productive source rock area of close to 17,000 acre, the size of Jonah Field) (Table 6.2). A current gas in-place estimate for Jonah Field is about 8.3 tcf (DuBois et al., 2004). If we assumed the saturation threshold of 20%, the estimated gas generation is only 3.65 tcf. This large difference may suggest that gas might have migrated from a larger fetch area (in other words, there was lateral as well as vertical migration) or generated from the secondary cracking of oil from

1 Cumulative methane generation at the final 360°C/72 h for Mowry Shale is 5.30 mg/ gTOC, Baxter Shale is 6.14 mg/ g TOC, and Cameo Coal (Mesaverde Coal) is 7.56 mg/ gTOC. Total methane generation from these three source rocks is 5.30+6.14+7.56 = 19.0 mg/ gTOC. 2 A simple layer model was assumed to calculate the source rock generation capability, assuming a fetch area only directly under the Jonah Field. Gas generation from one source rock (tcf) = [thickness of the source rock*productive area*density*TOC (%)*CnH2n+2 (mg/ gTOC)]/ estimated rock density (g/cc) The estimated rock density (g/cc) = rock mass (g)/volume of rock (cc) 106

the deeper sources or DuBois et al.’s (2004) gas in-place calculations (Sg*Φ*h*A) might be incorrect due to the uncertainties of Rw in the Archie’s equation

Figure 6.1 Burial-history plot of SHB 13-27. Plot shows hydrocarbon generation windows for the Hilliard Shale through the Fort Union Formation. Modified after Coskey (2004).

Evidence for secondary cracking of oil generated in the Hilliard/Baxter Shale and Mowry Shale in Jonah Field is also provided in Figure 6.2 which depicts the burial history model of the Wagon Wheel well in the Pinedale Anticline, northeast of Jonah Field. Modeling by Roberts et al. (2005) suggests that both Mowry and Hilliard/Baxter Shales were buried deeply enough in that location to undergo secondary cracking of oil from 80 Ma to 50 Ma (Roberts et al., 2005). Although the Mowry Shale underneath Jonah Field area does not have particularly rich present-day TOC and may not have good source rock quality, the Mowry Shale in the south of the Jonah Field shows very good source rock character (John Curtis, personal communication) and on the west of the Jonah Field, there is Mowry-derived oil production (Curtis, personal communication). It is possible that the early generated oil and gas migrated from west to east and south to north into the Jonah Field. The west to

107

east migration might correspond to the tectonic movements of the Wyoming Thrust Belt such as Absaroka, Charlston, Darby, and Hogsback thrust belts. It was not possible to make a realistic calculation on the gas generation potential in the Piceance Basin following the approach used above, because the maturation history, source rock quality, and the fetch area in the Piceance Basin, either present-day or in the past for the gas accumulation, are not clearly understood.

Table 6.2 Estimated hydrocarbon gas generation (methane to butane) from primary cracking of kerogen in carbonaceous shales of the Lance Formation and Upper Mesaverde Group, and coals in the Lower Mesaverde Group, Hilliard Shale, and Mowry Shale, based on the experimental data. Carbonaceous shales in Lance Formation and Upper Mesaverde Group are assumed to be similar to the Mancos Shale. The total gas generation from primary cracking only is about 18.3 tcf. Volumetric gas generation (C1 to C4) was estimated by: (Productive area in Jonah Field source rock thickness/ density of the source rock)*TOC (wt%)*cumulative gas yield (mg gas/gTOC)

108

6.2.2 Bulk Compositions of Hydrocarbon Gases The bulk compositions of gases generated in the laboratory are distinctly different from gases collected at Jonah Field and in the Piceance Basin fields. Gases generated from the hydrous pyrolysis experiments in the lab are wetter, with a much higher fraction of non-hydrocarbon gases. Experimental results show that gas dryness for gases generated in experiments equivalent to a thermal maturity of 1.6% Ro is about 51 % for the Mowry Shale, 55 % for the Mancos Shale, and 63 % for the Cameo Coal (Figure 5.13). The proportion of non-hydrocarbon gases generated in the laboratory is about 60%

109

in the Cameo Coal, 43 % in the Mowry Shale, 62% in the Mancos Shale, and about 71 % in the Baxter Shale (Table 6.3). However, in both the Jonah Field and Piceance Basin, produced gases and mud gas are relatively dry. In Jonah Field, methane constitutes more than 85 mol. % of the produced gas (Robinson and Shanley, 2004). Gas component analyses from the Lance

Formation in Jonah 3-7 Well (production interval: 10,122 to 10,380 ft) gave CH4 at

85.630 mol %; C2H6 at 7.703 mol %; C3H8 at 3.458 mol %; i-C4H10 at 0.791 mol %; n-

C4H10 at 0.681 mol %; i-C5H12 at 0.193 mo l%; n-C5H12 at 0.134 mol %; CO2 is 0.806 mol %; and N2 is 0.604 mol % (Robinson and Shanley, 2004). In the Piceance Basin, gas

dryness (C1/C1-5) was reported from several reservoir rocks by Tyler et al. (1995). Gas dryness varies from 0.99 to 0.78 (average 0.94) from 20 Cameo coalbed gas samples; gas is generally wetter from the Williams Fork Formation (excluding Cameo Coal) than the Cameo coalbed gases, with dryness values ranging from 0.79 to 1.00 (average: 0.90); gas is generally dry to very dry from the Iles Formation (some samples are very wet and might be associated with oil). In general, field gas in the Piceance Basin contains about

70 to 90% methane, 0.1 to 18 % C2+ gas, and 0 to 25 % CO2, with a significant amount of nitrogen (Reservoir Evaluation Report, 2005). Several possible reasons are proposed to explain the difference in bulk compositions between experimental and field gases: 1) gas generated from the laboratory

only represents thermal maturity up to 1.6 % Ro, whereas gas generated and mixed in the reservoir may represent gas expelled from higher maturity source rocks or secondary cracking of oil; 2) in the field, significant amounts of non-hydrocarbon gases are dissolved in the formation water or oil, as they migrate from source to reservoir (Price and Schoell, 1995); 3) after hydrocarbon expulsion, secondary processes such as migration and/or biodegradation may leave the heavier components behind along the migration conduit, transporting only the lighter components (methane) into the reservoir. Therefore, less dryness is common to all laboratory pyrolysis methods (Mango et al., 1994; Price and Schoell, 1995). It is impossible to directly quantify the amount of gas generated in natural systems because either the light volatile compounds have been expelled or lost or the heavy compounds dissolved, precipitated or absorbed once the petroleum is expelled. The only way to quantify the whole process of gas generation is to

110

perform artificial maturation experiments in the laboratory – a situation in which nearly 100% primary migration (expulsion) can be simulated. The bulk composition of gases generated in the laboratory should therefore represent the nearly ideal compositions of gas generated by primary cracking in the natural system.

6.2.3 Controls on Gas Generation Products and Their Yields Thermal degradation of organic matter leads to oil and gas generation (Peters, 1986). The prediction of both the quantity and quality of gas composition remains problematic. The chemical structure of kerogen with the thermal maturity overprint determines both the generated gas products and their yields during primary cracking. Humic kerogen contained in the coal has only a few short side chains with a single methyl group, with a large number of condensed rings, while sapropelic kerogen as contained in the marine shale has many long chains and individual ring structures that can break off to form methane. Therefore, more methane would be generated during primary cracking from humic Type III kerogen than from sapropelic Type II kerogen until the source rock becomes hydrogen-limited. Because the Cameo Coal contains mainly Type III kerogen and the Mowry, Mancos, and Baxter Shales contain mixed Type II/III kerogen, we expect to see more methane generated from those marine shales. However, it was not possible to estimate gas generation from secondary cracking in this study because the experimental temperature and time were not sufficient enough to crack the oil to gas. Both anhydrous and hydrous pyrolysis models predict that amounts of gas generated from the cracking of reservoir oil are greater than the gas yields obtained from kerogen (Henry and Lewan, 1999). Therefore, we expect more gas generated from oil cracking.

6.3 Non-hydrocarbon Gases

The non-hydrocarbon gases H2, CO2, N2, and H2S are generally the dominant gaseous products from hydrous pyrolysis experiments (Kotarba et al., 2009). The large yields of these non-hydrocarbon gases from hydrous pyrolysis, compared to the fraction of these gases in natural systems, may be explained by the usually high reactivity and solubility of these gases, as gas might be lost during the process of dissolution in water

111

Table 6.3 Cumulative yields after sequential HP Experiments (unit: mole %)

Proportion of Rocks He H2 CO2 H2S O2 & Ar N2 CH4 C2H6 C3H8 iC4 nC4 iC5 nC5 nC6 nC7 Others Total non‐hydrocarbon gases Cameo 60.3 7.3 168.7 2.4 0.0 0.3 36.8 12.1 7.2 1.0 2.0 0.5 0.6 0.3 0.1 0.7 300.0 0.60 Mowry 161.1 21.2 90.5 15.6 0.0 0.4 4.1 2.0 1.1 0.2 0.6 0.2 0.3 0.1 0.1 2.7 300.0 0.43 Mancos 110.9 8.7 167.2 8.8 0.0 0.1 1.7 0.8 0.3 0.1 0.2 0.1 0.1 0.1 0.0 0.9 300.0 0.62 Baxter 84.2 3.0 205.3 5.8 0.0 0.1 0.9 0.2 0.1 0.0 0.1 0.0 0.0 0.0 0.0 0.2 300.0 0.71 112

112

and migration in a natural petroleum system (Hunt, 1996; Kotarba et al., 2009). Hydrous pyrolysis experiments commonly produce large quantities of carbon

dioxide (Lewan, 1997). Carbon dioxide (CO2) is the main component of the non- hydrocarbon gases typically generated in the hydrous pyrolysis experiments. Carbon dioxide is highly soluble in the aqueous fluid. At the standard temperature and pressure environment, the solubility of CO2 in water follows a volume ratio of 1:1 (CO2 to water). At reservoir pressures, the volume ratio increases; for example at 7,000 psi (equivalent to

a depth of about 15, 000 ft), the volume ratio becomes 30:1 (CO2 to water, Hunt, 1996). It

is difficult to predict the amount of CO2 actually generated by source rocks in these tight- gas basins, and the amount of CO2 dissolved in the water in the reactor, by knowing only the pH value of water, because hydrogen is the strongest acid that determines pH value, and there are more than one species of non-hydrocarbon gas. Other organic acids can also contribute to the pH value. By applying Henry’s Law constants and assuming that only one species of gas exists in the water phase and saturates the water, an estimated total

CO2 yield can be estimated by several steps: 1) during the gas collection phase, the gaseous to aqueous phase ratio for CO2 can be calculated by the Ideal Gas Law and

Henry’s Law (Table 5.13); 2) if we can also obtain the cumulative gaseous CO2 at the

end of the sequential experiment, we can use the gaseous to aqueous CO2 phase ratio to

roughly estimate the cumulative CO2 dissolved in the water phase; 3) The total CO2 yield would then be the sum of the cumulative gaseous CO2 and aqueous CO2. 3 An estimated gaseous to aqueous phase ratio for CO2 is about 1: 1.7 for the Cameo gas, 1:3 for the Mowry gas, 3: 10 for the Mancos gas, and about 1:3.6 for the Baxter gas (Table 5.13). On a per gram of TOC basis, the cumulative carbon dioxide yield in the Cameo Coal headspace gas is 115.6 mg/gTOC, the Mowry Shale is 395.23 mg/gTOC, the Mancos Shale is 1286.86 mg/gTOC, and the Baxter Shale is 5047.55 mg/gTOC. By using the gaseous to aqueous phase ratio for CO2, the estimated total

generation of CO2 is 0.3 g, 1.6 g, 5.6 g, and 23.0 gCO2/ gTOC for the Cameo Coal, Mowry, Mancos, and Baxter Shale, respectively.

3 Table 5.13 shows the estimated CO2 gas dissolved in the water after the sequential HP experiments. The gaseous to aqueous phase ratio for CO2 from the Cameo Coal gas is 2.274:3.842 ≈ 1:1.7. 113

There are three possible indigenous sources of CO2. The first one is the thermal degradation of organic matter occurring during diagenesis and catagenesis, mostly by decomposition of carbonyl (C=O), methoxyl (-OCH3), and phenolic hydroxyl (-OH) groups (Hunt, 1996). In general, Type III kerogen-bearing source rocks tend to generate

more organic-sourced CO2 than Type II kerogen-bearing marine shale. The second source involves reactions between kaolinite and carbonates. This reaction produces chlorite and carbon dioxide, at temperatures above 100°C and reaching equilibrium at 160°C (Hutcheon and Abercrombie, 1989). The third source is high temperature thermal decomposition of carbonates. Magnesium carbonate decomposes to magnesium oxide and carbon dioxide at a temperature between 250 to 800°C (however, this high temperature is unlikely to occur in the reservoir). These three different sources give different δ13C values. Thermal destruction of carbonates gives δ13C ranging from -10 to +2 ‰ (Katz, 2002) while the thermal degradation of organic matter gives δ13C of -12 to - 29 ‰. The Mowry, Mancos and Baxter Shales all generate significant amounts (more than 1g/ gTOC) of carbon dioxide, implying that organic matter is not the only source for

CO2. In fact, not all TOC is converted into the carbon component of hydrocarbons. Hunt (1996) applied a material balance approach, suggesting the theoretical maximum convertible TOC in different types of kerogen. For Type II kerogen, 42% of the total carbon can convert to generate oil, while 6.2 % can convert to generate gas. For Type III kerogen, only 18 % of the organic carbon can convert to oil, and 7.2 % can convert to gas (Orr, 1992). If this theoretical consideration is correct that starting with one gram of carbon in an initial Type II kerogen, 0.42 grams are expected to convert o oil and 0.06 grams are converted to gas. In order to generate one gram of CO2, equivalent 0.27 grams of carbon is required. This leaves 0.25 grams of carbon in the residual unit of kerogen.

Therefore, in order to generate significant amount of CO2, inorganic source is necessary.

The mass balance calculation cannot be performed here to estimate the amount of CO2

gas being generated because there are several organic sources of CO2 gas: gas generated directly from kerogen, from bitumen, or from oil. Without knowing the transformation ratio of kerogen to gas, kerogen to bitumen to gas, or kerogen to bitumen to oil to gas, the mass balance calculation cannot be accurately performed.

114

The carbon isotopic composition of carbon dioxide can help with understanding

the source of CO2. Figure 6.3 shows the cumulative stable carbon isotope composition of 13 carbon dioxide generated from the Cameo Coal experiments (δ C-CO2 = -25 to -26‰), indicating that carbon dioxide is mainly from the thermal transformation of organic matter because organic matter tends to have a composition close to -30 ‰. The relatively 13 13 positive δ C of CO2 generated in experiments on the Mancos and Baxter Shales (δ C = 0.59 to 1.18‰) suggests an inorganic source, most likely from thermal decomposition of carbonate materials, because marine carbonates have a composition close to 0 ‰ (Faure 13 13 and Mensing, 2005). Intermediate δ C of CO2 from the Mowry Shale experiments (δ C = -6.35 to -5.97‰) might indicate contributions from both organic and inorganic sources, with a higher proportion of CO2 from thermal decomposition of carbonate minerals. It is also possible that the CO2 is decomposed from thermal decomposition of diagenetic carbonates, an inorganic source. In our experiments, the rate of instantaneous carbon dioxide generation declines with increasing maturity in all four source rocks, consistent with the idea that the generation of CO2 from kerogen takes place at a relatively early stage of maturation. The mineralogy of the Mancos Shale was studied by Fisher (2007) applying petrophysical log techniques (porosity and sonic tool data) to identify the lithology and matrix mineralogy of intervals within the lower Mancos Shale. His study showed that more than 80% of the minerals in the lower Mancos Shale are either dolomite or calcite. The Juana-Lopez and Tununk Shales and Dakota Silt-equivalent have the highest dolomite content (> 80%), and the Frontier-equivalent member having the highest calcite content (> 60%). In this study, we used samples of the upper Blue Gate Shale member in the Lower Mancos Shale, which Fisher (2007) calculated to have more than 70 % dolomite, less than 30% calcite, and less than 10 to 40% quartz. However, there are no XRD analyses to calibrate his well log interpretation. These data can only imply the existence of variable calcite/dolomite in the Lower Mancos Shale (Figure 6.4). Qualitative XRD analyses were completed on two Baxter Shale samples: one before the hydrous pyrolysis experiment, representing the original source rock; and the other after the hydrous pyrolysis experiment, as shown in Figure 6.5. The result shows that the Baxter Shale has both calcite and dolomite before and after the experiments.

115

Figure 6.3 (a) Cumulative stable carbon isotope composition of carbon dioxide (δ13C- CO2) generated from sequential hydrous pyrolysis experiments in Cameo Coal (blue), Mowry (red), Mancos (green), and Baxter Shales (purple). (b) Instantaneous CO2 yields versus its instantaneous carbon isotope change in the Mancos and Baxter Shales. (c) Instantaneous CO2 yields versus its instantaneous carbon isotope change in Cameo Coal and Mowry Shale.

The original sample is composed of a higher proportion of dolomite than calcite; however, after the sequential 300, 330, 360°C heating for 72 h each, the relative

116

proportion of calcite exceeds the dolomite, along with the formation of a small amount of smectite. The δ13C-carbonate in Cretaceous formations from the U.S. Western Interior Seaway ranged from 0 to +3 ‰ (Hayes et al., 1989; Pratt et al., 1993), as shown in Figure 13 6.6, similar to the δ C of CO2 value measured in experiment with the Mancos and Baxter Shale. There are no published data regarding carbonate components in the very siliceous Mowry Shale. Further investigation on mineralogic compositions of the Mowry Shale would be necessary, in order to quantify the inorganic-sourced CO2 in the Mowry Shale.

Analyses of CO2 generated from coals in the Piceance Basin and Jonah Field are available in the literature, but no CO2 data from the marine shales in these basins are

published. CO2 from gases produced from the Mesaverde coal beds in the Greater Green River Basin ranges from less than 1 to more than 25 % mole fraction (Tyler et al., 1995).

CO2 content from gases produced from the Cameo Coal ranges from 1.3 to 14.3 % and

averages 5.3 % (Tyler et al., 1995). The Cameo Coal gases have the highest CO2 content near the Divide Creek Anticline and the south part of the Piceance Basin (Tyler et al.,

1995). CO2 content from gases produced from the Williams Fork Formation ranges from less than 1 % to more than 20 % and averages 5 %. Carbon dioxide content from gases produced out of Iles Formation is usually less than 2 %. The difference in the CO2 content of gases produced from the Williams Fork and Iles formations might indicate different gas sources in these two formations (Tyler et al., 1995). Coal has very high storage capacity (adsorption) for gases. Usually more than 95% of gas resides in the coal matrix as an adsorbed layer on the internal coal surface (Cui et al., 2004). Carbon dioxide is preferentially adsorbed over methane by coal because CO2 has better affinity for the solid surface than CH4 (Rogers, 2003). A typical methane isotherm shows that at an initial reservoir pressure of 1,620 psia, the methane

storage capacity of coal is about 450 scf/ ton (TICORA, 2009). The CO2 adsorption saturation capacity of the coal is more than that of methane and nitrogen (Cui et al., 2004). In this study, even though our coal sample has been crushed into gravel size, most of the macropores or cleat networks might be destroyed, the micropore structures can still

be intact. A great deal of CO2, CH4, and N2 might still exist inside the pore structure

system in the Cameo Coal. Except coals, shales can also have good CO2 adsorption

117

Figure 6.4 Summary of mineral percentages taken from lower Mancos Shale interval matrix for Hells Hole 9131 and Lower Horse Draw 2186. Nine sub-groups are identified in the lower Mancos Shale- A: Lower Mancos Shale all intervals B: Upper Blue Gate Shale C: Niobrara-equivalent D: Lower Blue Gate Shale E: Lower Blue Gate Silt F: Frontier-equivalent G: Juana Lopez H: Tununk I: Dakota Silt-equivalent. (Fisher, 2007)

118

Dolomite Calcite Recovered Baxter Shale samples after HP

smectite

Original composite Baxter Shale samples

Figure 6.5 X-Ray Diffraction (XRD) result of the Baxter Shale before (lower) and after (upper) sequential hydrous pyrolysis.

Figure 6.6 Carbon isotopic excursion in whole-rock carbonate (δcarb) and total organic carbon (δorg) at the Cretaceous Cenomanian/Turonian boundary (Hayes et al., 1989). The total amount of time represented is on the order of 1 million years (Kump and Arthur, 1999). The positive carbon isotope excursion near Cenomanian-Turonian (C/T) boundary was due to an increase in organic carbon burial rate during Cretaceous and drawdown in atmospheric pCO2 (Arthur et al., 1988). Prior to the excursion, ΔB was approximately - 28‰. During the event, δorg increased by 4 ‰ and δcarb increased by only 2 ‰. ΔB is the carbon isotopic composition of sediment carbonate and organic carbon.

119

capacity but it depends mostly on the TOC content (Nuttall et al., 2005). At reservoir pressure of 7, 000 psi, equivalent to a depth of about 15,000 ft, the

volume ratio of CO2 dissolved in a 40° API gravity oil is about 170:1 (CO2 to oil) (Hunt, 1995), more than five times greater of its solubility in water. This high solubility under reservoir conditions causes a significant amount of CO2 to be dissolved in either

formation water or oil. The high solubility of CO2 might lead to CO2 being lost during the process of migration. Therefore, a huge difference of cumulative CO2 yield exists between laboratory-simulated and reservoir-produced CO2.

Hydrogen (H2) is generated in the second largest quantities of non-hydrocarbon gases. The cumulative hydrogen yield at an equivalent vitrinite reflectance of 1.6% Ro is

87 mgH2/ gTOC in the Baxter Shale, 44 and 42.5 mgH2/ gTOC in the Mowry and

Mancos Shale, and only 1.06 mgH2/ gTOC in the Cameo Coal (Table 5.15). Hydrogen

(H2) can be derived from both water and organic matter. Type II kerogen from marine shale has a higher atomic H/C ratio than Type III kerogen from terrestrial coals (Tissot et al., 1974; Lewan et al., 1985; Baskin, 1997), so more hydrogen from marine shales is available for conversion to hydrocarbons. In the experiments, instantaneous hydrogen generation increases with increasing thermal maturity (Table 5.14). With increasing thermal stress, bonds which hold hydrogen in the kerogen are easy to break. Hydrogen is a common constituent of many well gases and subsurface waters, but it is rarely reported in the literature because it is not included in conventional gas analyses (Hunt, 1996). Zinger (1962) found up to 43 % hydrogen in the gas dissolved in oil-field waters of Paleozoic rocks of the lower Volga region of Russia. In western Siberia, Nechayeva (1968) reported that he found hydrogen in about 15% of all gas samples analyzed and about 60% of the samples with hydrogen concentration under 1%. The high concentration of hydrogen from organic-rich Jurassic sediments suggests that it comes from the thermal decomposition of organic matter. More recent information on hydrogen content in oil or gas fields was not found. Since hydrogen is the most effective reducing agent among the natural gases, the amount of generated hydrogen determines the redox condition of the reservoir. It may not be sensible to compare the hydrogen content in naturally occurring hydrocarbon gases and gases generated in the laboratory because in natural geological traps, hydrogen is so light, mobile, and reactive that it cannot be retained for a long time

120

(Hunt, 1996), while in closed-system laboratory experiments, hydrogen is retained in the reactor, collected, and measured.

Nitrogen (N2) is the third most abundant non-hydrocarbon gas generated in the experiments. Under laboratory conditions, nitrogen can be derived from the thermal evolution of kerogen (organic) or inorganic nitrogen in sedimentary rocks (Kotarba, 1988; Krooss et al., 1995). Ammonia dissolved in pore water can be oxidized to nitrogen through contact with heavy metal oxides or ferric oxide or meteoric waters containing oxygen. Nitrogen can also be derived from the atmosphere. Certain micro-organisms can take up atmospheric nitrogen and transfer it to organic nitrogen compounds (biological fixation) (Krooss et al., 1995). In gas fields, nitrogen can be derived from multiple sources, including volcanic, radiogenic, primordial, and atmospheric sources (Krooss et al., 1995). The cumulative nitrogen yield from hydrous pyrolysis experiments ranged

between 0.97 and 1.27 mgN2/ gTOC from Mowry, Mancos, and Baxter Shales at an

equivalent vitrinite reflectance of 1.6% Ro. The Cameo Coal only generated 0.10 mgN2/ gTOC. The different cumulative yield of N2 between the marine source rocks and the Cameo Coal may have been due to the fact that sapropelic organic matter is richer in nitrogen components (Krooss et al., 1995); consequently, more nitrogen can be produced from the sapropelic organic matter. This corresponds to our experimental results (Figure 5.17). The Mowry, Mancos, and Baxter Shales which contain more sapropelic kerogen, generate more nitrogen than the Cameo Coal which contains mainly humic kerogen. The reaction kinetics is another controlling factor in forming nitrogen from kerogen. Instantaneous nitrogen generation increases with increasing maturity (Table 5.14) because the higher thermal energy in the system results in faster and more bond-breaking reactions. Nitrogen is produced from the Piceance Basin and Jonah Field in the Greater

Green River Basin. N2 content in Mesaverde coalbed gases in the Greater Green River Basin ranges from less than 1 to 20 % and averages about 4 % (Tyler et al., 1995). Nitrogen content in the Williams Fork Formation in the Piceance Basin ranges from 0.5 to 37.5 % and averages 8.8 % (Tyler et al., 1995). However, only a small amount of nitrogen (Table 5.10) was generated during sequential hydrous pyrolysis experiments. This might be related to the fact that nitrogen is usually formed at higher temperatures

121

than methane (Krooss et al., 1995), and the experiments carried out in this study may not have been able to reach the main nitrogen generation stage.

H2S has very high solubility in water, more than twice that of CO2 in water (Suleimenov and Krupp, 1994; Weiss, 1974). Under any laboratory experimental conditions, hydrogen sulfide is most likely generated from organic matter during catagenesis, through thermal decomposition of reduced organic sulfur compounds in kerogen (Orr, 1986). The generation of H2S increases with increasing temperature and

pressure and the formation of H2S is related to the dispersed organic matter (Orr, 1986).

In both the Piceance Basin and Jonah Field in the Greater Green River Basin, no H2S production was reported (Johnson and Roberts, 2003); however, H2S is produced in the Uinta Basin, Utah (Harris, personal communication).

Because H2S is even more soluble in water than CO2, the same approach was applied to calculating the amount of actual H2S dissolved in the water as was used for

calculating the dissolved CO2. By applying Henry’s Law constants and assuming that only one species of gas existed in the water phase, an estimated gaseous to aqueous phase

ratio for H2S is higher than the ratio for CO2. The gaseous to aqueous phase ratio for H2S is fairly consistent, about 1: 5 or 1:6 for gases generated from all source rocks (Table 5.13). On a per gram of TOC basis, the cumulative hydrogen sulfide yield at equivalent vitrinite reflectance of 1.6% Ro shows about 2 mgH2S/ gTOC in the Baxter Shale, 1.06 and 0.41 mgH2S/ gTOC in the Mowry and Mancos Shales, and only 0.12 mgH2S/ gTOC in the Cameo Coal. By using the gaseous to aqueous phase ratio for CO2, the estimated

total generation of H2S is 0.7 mg, 7.2 mg, 2.8 mg, and 13.8 mg H2S/ gTOC.

At a temperature of 104°C (220°F) and a pressure of 2,000 psi, H2S is forty times

more soluble in water than at STP (standard temperature and pressure). Any H2S formed by the kerogen is disseminated by fluid migration and precipitated as pyrite if it is on contact with iron. In rocks with low iron content, such as in some carbonate reservoirs,

H2S from kerogen may enter the reservoirs (Orr, 1977). Pyrite was found in the floodplain and U facies of the Lance Formation associated with paleosol development in the Pinedale Anticline (Govert, 2009; Chapin, 2009). Pyrite was also identified in thin sections from samples of the Niobrara and Juana-Lopez outcrops in the western Piceance Basin (Kuzniak, 2009).

122

6.4 Eh/pH of Recovered Water Redox potential (Eh) and pH were measured on recovered waters of sequential hydrous pyrolysis experiments on the Cameo Coal, Mowry Shale, Mancos Shale, and Baxter Shale. Redox potential is a measure of the ability of an environment to supply electrons to an oxidizing agent or to take up electrons from a reducing agent. It is analogous to pH of an environment, which is a measure of its ability to supply protons (hydrogen ions) to a base or to take up protons from an acid (Krauskopf and Bird, 1995). The strongest reducing agent is hydrogen. When non-hydrocarbon gases such as

hydrogen sulfide (H2S) and carbon dioxide (CO2) are released from the source rocks in the experiments and dissolve in water, they react to form hydrosulfide ion (HS−), sulfide 2− – -2 ion (S ), bicarbonate ion (HCO3 ), carbonate ion (CO3 ), and carbonic acid (H2CO3), making the water weakly acidic. These ions are only subtle contributors to the acidity of water, when compared to hydrogen gas. Table 6.4 shows the measured Eh, pH, and temperature of the recovered water from experiments on Cameo Coal, Mowry, Mancos, and Baxter Shales. The Nernst Equation was applied to calculate the partial pressure of hydrogen gas dissolved in the water. In general, marine source rocks (Mowry, Mancos, and Baxter Shales) give higher

PH2 than the terrestrial source rocks (Cameo Coal). Among the marine shales, the Mowry Shale has the highest partial pressure of hydrogen gas (0.216 bar), the Mancos shale has the second highest (0.127 bar), and the Baxter Shale has the least (0.059 bar). Partial pressure of hydrogen from Cameo Coal experiment only gives 0.022 bar. The calculated partial pressure of hydrogen results correspond with the calculated amount of hydrogen gas in the headspace of the reactor from each experiment. Generated hydrogen gas in the headspace is 0.025 g, 0.026 g, 0.017 g, and 0.007 g for the Cameo Coal, Mowry, Mancos, and Baxter Shale, respectively. In Figure 6.7 – a plot of the four source rocks on an Eh- pH diagram is shown, indicating reducing environmental conditions of the four source rocks during their maturation processes. The recovered water from the Cameo Coal shows the lowest pH value, indicating the highest hydrogen generation.

123

Table 6.4 Measured Eh, pH, and temperature from the recovered water from sequential hydrous pyrolysis experiment on Cameo Coal, Mowry, Mancos, and Baxter Shale.

Source Rocks Cameo Mowry Mancos Baxter pH 4.79 6.35 6.41 6.45 Eh (mV) ‐234.10 ‐355.80 ‐352.50 ‐345.00 Temperature (°C) 19.70 19.70 19.70 19.60

*logPH2 ‐1.662 ‐0.665 ‐0.897 ‐1.231

PH2 (bar) 0.022 0.216 0.127 0.059 *Partial pressure of hydrogen (g) was calculated by the Nernst Equation: Log PH2 = -33.8295*Eh-(2.0003*pH)

Figure 6.7 Eh-pH Diagram of the recovered water from sequential hydrous pyrolysis experiments on the Cameo Coal (blue), Mowry (red), Mancos (green), and Baxter Shales (purple).

6.5 Gas: Oil Ratio (GOR) The gas: oil ratios (GOR) for the generated hydrocarbon gases and expelled oils from the Cameo Coal, Mowry, Mancos, and Baxter Shales are given in Table 6.5. The gas and oil volume calculation from sequential hydrous pyrolysis data applied the method in Lewan and Henry (1999). The hydrocarbon gases consist of methane, ethane, propane, n-butane, i-butane, n-pentane, and i-pentane. Figure 6.8 shows the increase in cumulative GOR with increasing experimental temperature, assuming an average API gravity of 30° in expelled oils. All four source rocks show a curvilinear increase in cumulative GOR with increasing experimental temperature, with the Baxter Shale having the highest GOR and GOR increase. Overall GOR at 1.6% Ro ranges from 768 to 1878 scf/bbl. Although

124

Type III kerogen-bearing source rocks (coals) are typically described as a gas-prone source rocks, the Cameo sample generated less hydrocarbon gas per gram of TOC than the oil-prone source rocks (marine shales). Figure 6.9 plots the cumulative GORs with their expelled oil for the four source rocks. The trend for Mowry and Mancos Shale with Type II and mixed Type II/III kerogen is similar, while the Baxter Shale has the highest increase in GOR with very little oil generation. This observation is consistent with results from previous hydrous pyrolysis studies, which have also shown that Type I and Type II kerogens generate significantly more gas than Type III kerogen (Lewan, 1993; Hunt, 1996, Lewan and Henry, 1999); results from open-system pyrolysis studies (Behar et al., 1997) are also consistent with these findings. However, metagenesis, a stage with significant amount of gas generation, does not necessarily follow the same GOR trend (Lewan and Henry, 1999). In the southern Piceance Basin, the GOR in the Grand Valley and Parachute fields ranged from 1 to 10 bbl/MMcf, and ranged from 1 to 50 bbl/MMcf in the Mamm Creek Field (Nelson and Santus, 2010). The GOR averaged about 10 bbl/MMcf in the Rulison Field (Nelson and Santus, 2010). In Jonah Field, the averaged GOR was 86.2 mmcf gas/bbl, reported by Robinson and Shanley (2004)

6.6 The Relationship among Stable Isotopes, Kerogen Types, and Thermal Maturity Figure 5.19 shows the δ13C values of the generated hydrocarbon gases with reciprocal carbon number. Figure 6.10 shows the instantaneous yield versus instantaneous change of δ13C in methane, ethane, and propane. The carbon isotope values of gaseous hydrocarbons become heavier with increasing carbon number in experiments on the Cameo Coal, Mowry and Mancos Shale (Figure 5.19), consistent with experimental observations by Chung et al. (1988) and several other authors (eg. Burruss et al., 2003; Kotarba et al., 2009). Kinetic isotope effects (KIE) state that the kinetics of the cracking of alkyl side chains from high molecular weight organic matter favors the breaking of 12C-12C bonds, resulting in 12C enrichment of the product gas molecules (Chung et al., 1988). Chung et al. (1988) proposed a relatively simple mechanism of gas generation from kerogen based on several assumptions: 1) the 13C is homogeneously distributed in all carbon atoms; 2) all gas hydrocarbons are formed by the

125

same reaction mechanism; 3) gas is thermally cracked from isotopically homogeneous parent molecules; 4) gas hydrocarbons are not formed by condensation reactions. Therefore, during thermal maturation, methane is most depleted in 13C and progressively higher carbon number alkanes and kerogen are increasingly less depleted in 13C. In contrast, the carbon isotopes of ethane in the Baxter Shale have lighter δ13C values than methane, creating a concave trend in Chung’s natural gas plot of δ13C versus reciprocal carbon number (Figure 5.19). In fact, some of the Baxter mud gases have ethane with δ13C more negative than methane (Lewan, personal communication). However, it is still worth noting (Table 5.16, Chapter 5) that while measurement of δ13C of ethane and propane generated from the Baxter Shale in these experiments is repeatable, there are potential measurement uncertainties due to the small volume (low peak height) of the generated gas. Therefore, the representation of the Baxter Shale gas pattern is arguable.

Table 6.5 Cumulative gas: oil ratio (GOR) of the generated gas and expelled oil in each sequential experiment. A sensitivity test was done on the API gravity of oil, assuming API gravity values of 10°, 20°, 30°, 40°, and 50° for the expelled oil. API (10) API (20) API (30) API (40) API (50) Rocks Experimental GOR(scf/bbl) GOR(scf/bbl) GOR(scf/bbl) GOR(scf/bbl) GOR(scf/bbl) Condition Cameo 300/72 84.8 79.2 74.3 69.9 66.1 Cameo 330/72 324.4 303.0 284.3 267.7 252.9 Cameo 360/72 877.0 819.1 768.4 723.6 683.7 Mowry 300/72 24.7 23.0 21.6 20.4 19.2 Mowry 330/72 244.5 228.3 214.2 201.7 190.6 Mowry 360/72 1004.7 938.4 880.3 829.0 783.2 Mancos 300/72 22.7 21.2 19.9 18.8 17.7 Mancos 330/72 175.8 164.2 154.0 145.0 137.0 Mancos 360/72 1028.7 960.8 901.3 848.8 802.0 Baxter 300/72 0.0 0.0 0.0 0.0 0.0 Baxter 330/72 594.9 555.6 521.2 490.9 463.8 Baxter 360/72 2143.7 2002.2 1878.3 1768.8 1671.2

126

Figure 6.8 Cumulative gas: oil ratios (GOR) for the Cameo Coal (blue), Mowry Shale (red), Mancos Shale (green), and Baxter Shale (purple) subjected to hydrous pyrolysis temperatures from 300 to 330 to 360°C for 72 h each.

360/72

360/72 360/72

330/72

330/72 330/72 300/72 300/72 300/72

Figure 6.9 Cumulative gas: oil ratios (GOR) for the Cameo Coal (blue), Mowry Shale (red), Mancos Shale (green), and Baxter Shale (purple) versus yields of expelled oil from sequential hydrous pyrolysis temperatures from 300 to 330 to 360°C for 72 h each.

127

Gas generated from the Cameo Coal, Mowry and Mancos Shale in the experiments all show “dogleg” trends (Figure 5.19), similar to that reported by Kotarba et al. (2009), for gases generated from different kerogen types in Polish source rocks at sequential hydrous pyrolysis runs of 330 and 355°C for 72 h, and to that reported by Burruss et al. (2003) and Lillis (1999), for gases generated from Hue Shale and Shublik Shale (NPRA source rocks) at sequential hydrous pyrolysis runs of 300, 320, 340 and 360°C for 72 h. Tian et al. (2010) applied closed-system anhydrous pyrolysis using sealed gold tubes to investigate chemical and isotopic compositions of generated gases from low- maturity kerogen, from high-maturity kerogen and from crude oil from marine source rock samples in the Tarim Basin, China. Their experimental (anhydrous) results are very similar to the hydrous pyrolysis results presented here. For both the low-maturity kerogen and high-maturity kerogen, the natural gas compositions form a subtle dogleg trend on a cross-plot of δ13C and reciprocal of carbon number (Figure 6.11). Gases generated from the low-maturity kerogen show less linear trend than those generated from the high- maturity kerogen. Gases generated from the oil cracking have compositions plotting closest to a linear relationship between δ13C and reciprocal of carbon number. Several recent publications reporting closed-system anhydrous pyrolysis results also described dogleg trends in cross-plots of δ13C and reciprocal of carbon number, from both laboratory-derived and field gases (Dai et al., 2005; Dai et al., 2009; and Zou et al., 2007). Dai et al. (2009) analyzed field gas samples of Type II and II kerogens from Xujiahe Formation in the Sichuan Basin, SW China. Figure 6.12 shows the dogleg trend for produced gases from coal source rocks which have not undergone significant secondary alteration. Zou et al. (2007) applied confined system pyrolysis with a heating rate of 20°C/ h (Behar et al., 1997) to study gas generated from Type III kerogen isolated from a late Paleozoic coal from Ordos Basin, northern China. They found out that gases derived from cracking of coal samples all show dogleg trends. Zou et al. (2007) also compared their carbon isotope data of gases to other published gas isotope data from the closed-system pyrolysis: from Jurassic coal (Liu et al., 2002, 2003), from a calcareous shale (Type I kerogen, Huang et al., 1999), and from a Type II kerogen (Lorant et al., 1998) (Figure 6.13). Zou et al. (2007) therefore developed an improved natural gas plot

128

(Chung et al., 1988) that differentiates coal-derived gas from oil-associated gas (Figure 6.13). They also found that for oil-associated gases, the concave trend has higher thermal maturity than the linear trend (Figure 6.13), indicating the δ13C of ethane and propane becomes heavier with increasing thermal temperature. However, for coal-derived gases, the δ13C of ethane and propane does not show any consistent change with increasing thermal temperature. All these recently published studies provide different ways to explain the classic natural gas plot and suggest that it is necessary to re-interpret the plot. Willette (2010, unpublished PhD thesis) applied hydrous pyrolysis to investigate the influence of water, water salinity, grain mineralogy, and oil compositions on the gas 13 isotopes during secondary cracking of oil to gas. She suggested that δ C trend of C1 to

C3 from both the hydrous and anhydrous laboratory experiment is very close to linear and similar to values determined from an associated gas (unknown GOM oil) in a natural system (Figure 6.14). After my revisiting the sub-linear trend in the natural gas plot reported by Chung et al. (Figure 6.15, 1988) for gases generated from different kerogen types by hydrous pyrolysis at 330°C for 72 h (Figure 6.15), I found out that in fact, two out of four trends in Chung’s natural gas plot also show dogleg trends during primary cracking (330°C for 72 h). Other recently published papers using the Chung’s cross-plot of δ13C versus 1/carbon number, from either laboratory pyrolyzed gases or field collected gases showed a variety of shapes – trends which are dogleg, linear, concave or convex (Dai et al., 2005; Dai et al., 2009; Burruss and Laughrey, 2010). This might indicate that not only the thermal maturity and types of kerogen have strong influence on the carbon isotope composition of gases but other factors such as the homogeneity of kerogen’s structure, dominant component of kerogen, or the distribution of activation energies within parental organic matter, can determine the shape of Chung’s natural gas plot. In addition, Chung et al. (1988) made several unrealistic assumptions in their model. For example, it is nearly impossible for all gas hydrocarbons to be formed by the same reaction mechanism nor in reality, it is reasonable to assume that gas is cracked from isotopically homogeneous parent molecules because different types of kerogen are composed of different molecular structures. The assumption that all gas hydrocarbons can be formed by the same reaction mechanism is disproved by several publications. Four

129

reaction mechanisms (kinetic schemes) were proposed by Behar et al. (1992) for both primary and secondary cracking, based on anhydrous pyrolysis experiments,: 1) depolymerization reactions of kerogen which involves the most labile compounds

containing C-O and C-S bonds; 2) C-C bond cracking of C6+ saturated chains; 3)

demethylation reaction of aromatic structures such as C9-C13 compounds; 4) C-C bond

cracking of C3-C5 compounds. Lewan (1997) proposed two reaction mechanisms that involve kerogen transformation to bitumen: cross-linking (between radical sites and adjacent carbons to form a more cyclic and insoluble molecule) and thermal-cracking thermal cracking of the β-positions of the free-radical sites to form low-molecular-weight hydrocarbons) reaction pathways for thermal maturation of bitumen in a closed-system under hydrous pyrolysis conditions. Seewald (2003) concluded that the free- radical mechanism (also called homolytic cleavage mechanism: Kissin, 1987) might be more critical than the carbonium-ion mechanism (also called heterolytic cleavage mechanism) for petroleum generation. Chung et al. (1988) simplified the reaction mechanism to only thermal-cracking reaction from isotopically homogeneous parent molecules. Although he later suggested that methane might be generated from different or additional functional groups, and ethane and propane might be generated from structurally similar functional groups, his natural gas plot trend, based on his assumptions, can only explain one of the several scenarios of reaction mechanisms. A new model is therefore proposed in this study for explaining different trends in natural gas plot for Type II and Type III kerogen without any alteration of gas compositions by secondary processes such as gas migration and leakage. For Type II kerogen, the dogleg trend represents gas mainly generated from primary cracking, while the linear trend shows more gas generated from secondary cracking (Figure 6.16). This new model is supported by gases generated from both closed-system hydrous and anhydrous pyrolysis experiments (Chung et al., 1988; Lorant et al., 1998; Huang et al., 1999; Burruss et al., 2003, Kotarba et al., 2009; Tian et al., 2010; this study). The explanation for this phenomenon is that during secondary cracking, heavy hydrocarbons

are broken into smaller molecules, increasing dryness (C1/C1-5) and decreasing the proportion of heavy hydrocarbon gases. The carbon isotope composition of the residual 13 12 12 C2+ gases becomes enriched in C due to the preferential cracking of the C- C bonds

130

(a) With increasing amount of methane generated from source rocks, the 13 δ C1 values become lighter. The methane generation decreases in the Mancos Shale.

(b) With increasing amount of ethane generated from source rocks, the 13 δ C2 values stay almost constant throughout the experiments in the

Cameo Coal and Mowry Shale. (c) With increasing amount of propane generated from source rocks, the 13 δ C3 values stay constant throughout the experiments.

Figure 6.10 Instantaneous yield of methane (a), ethane (b), and propane (c) versus instantaneous carbon isotope ratio change of methane, ethane, and propane from Cameo Coal (blue), Mowry (red), Mancos (green), and Baxter (purple) Shales. With increasing carbon number (from C1 to C3), the range of carbon isotopic values narrows.

131

Figure 6.11 Natural gas plots of gases from the three samples at selected maturity levels. (a) LN oil; (b) KP kerogen (high-maturity kerogen); (c) TZ kerogen (low- maturity kerogen). The Ro was calculated using the kinetic parameters of Hill et al. (2007). Modified after Tian et al. (2010).

132

Figure 6.12 Natural gas plot of the Xujiahe Formation reservoired gases in the Sichuan Basin. Seventy-two out of the 76 analyzed gas samples display a normal carbon isotopic 13 13 13 13 trend for C1 to C4 gases (i.e., δ C1 < δ C2 < δ C3 < δ C4), whereas the remaining gases 13 13 13 13 show a partial reversal trend (δ C2 > δ C3 or δ C3 > δ C4 ). The partial reversal is most likely from the mixing of coaly sourced gases that were generated at different thermal maturity levels. This suggestion is supported by the presence of two phases of hydrocarbon fluid inclusions in the reservoirs. Modified after Dai et al. (2009).

Figure 6.13 Natural gas plot for gases generated from anhydrous pyrolysis of oil-prone source rocks (a) and coals (b). The curves are concave for overmature oil-associated gas and convex for overmature coal-derived gases. (1) Type I kerogen (solid triangle, Huang et al., 1999); (2) Type II kerogen (solid square, Lorant et al., 1998); (3) Type III kerogen (solid circle, Zou et al., 2007); (4) coal (open circle, Liu et al., 2003); (5) xylite heated in open-system anhydrous pyrolysis (open square, Berner et al., 1995). Solid line: isotope curve of gas generation; dashed line: isotope curve shift due to secondary cracking.

133

Figure 6.14 The δ13C isotopic values of methane, ethane, propane and n-butane for hydrothermal and anhydrous experiments of Smackover oil. Isotopic data from pyrolysis experiments of other oil families included for comparison. Natural gas sample of GOM oil (open circle) is included for comparison (Willette, 2010).

13 in C2+ alkanes (Zou et al., 2007). This causes the δ C of C2 and C3 to increase in the 13 dogleg trend and the dogleg trend becomes more linear while the δ C of C1 shifts to slightly more negative (depleted in 13C). Further increase in thermal stress might tilt the 13 slope of δ C of C2 and C3 further upward, leaving a downward “concave” curve in the natural gas plot, as shown in Figure 6.13 (dashed line) (Zou et al., 2007). Tian et al. (2010) also showed the results that support this new model. They found out that methane from oil cracking is more enriched in 12C than methane from kerogen cracking while ethane and propane become extremely enriched in δ13C with respect to their starting samples at the thermal levels at which they began to crack.

However, for Type III kerogen, this explanation that the residual C2+ gases 13 12 12 becomes enriched in C due to the preferential cracking of the C- C bonds in C2+ alkanes might be different, because less C2+ gas is generated from humic source rocks

134

(Figure 6.17). Therefore, with increasing thermal maturity, the trend is expected to shift 13 13 upwards (enriched in C) or the slope of δ C for C2 and C3 being slightly tilted up at the end. The change in carbon isotope gas compositions depends mainly on the molecular structure of the Type III kerogen. Since the Type III kerogen is usually very heterogeneous, more variation in the natural gas plot is expected.

Figure 6.15 Natural gas plots of gas hydrocarbons generated from various source rocks pyrolyzed under hydrous conditions at 330°C for 3 days. The subbituminous coal and Green River Shale trends show dogleg shape. Modified after Chung et al. (1988). Previous interpretations of the non-linear plot point to microbial alteration or oxidation of the gas. However, microbial and oxidation effects are unlikely to occur under laboratory conditions. Very few quantitative calculations are available to calculate 13 how much microbial gases are necessary for decreasing δ C of C1 to make the observed dogleg trend. 13 It is possible that even for the same kerogen type, the δ Cn versus 1/n trend in the natural gas plot is not only affected by the thermal maturity but also by the kerogen

135

structure and composition. The isotopic signatures of C2 and C3 generated from Type III (Cameo Coal) and the mixed Type II/III kerogen (Mowry, Mancos, and Baxter Shale) are very distinctive (Figure 5.19). However, δ13C of methane shows overlap between the mixed Type II/III and Type III kerogens. This observation is similar to Zou et al. (2007) who compared the δ13C of methane from both marine source rocks and coals in the Tarim Basin in NW China and found that the methane carbon isotope composition has a large 13 overlap between coal-derived and oil- associated gas (δ C1 = -52.0 ‰ to -22.0 ‰, PDB standard) (Figure 6.17). The overlap value of δ13C of methane might imply that other factors such as the dominant reaction mechanisms may be more important than the indigenous kerogen type for methane generation. It has been widely accepted that with increasing thermal maturity, gas becomes more enriched in 13C (James, 1983; Schoell, 1983; Chung et al., 1988; Berner et al., 1995; Andresen et al., 1995; Lorant et al., 1998; Huang et al., 1999, Cramer et al., 2001; Gaschnitz et al., 2001; Liu et al., 2003; Zou et al., 2007). High-maturity gas is usually dominated by isotopically heavier methane, but results from these experiments show the opposite trend- gas becomes more depleted in 13C with increasing experimental temperatures (Figure 5.19). Through detailed examination of previously published data, it was realized that, in some reports, δ13C of methane first becomes more depleted in 13C, then enriched again with increasing maturity (Table 4, Andresen et al., 1995; Table 2, Lorant et al., 1998; Figure 6, Tang et al., 2000; Table 3, Zou et al., 2007; Figure 5, 6, and7, Tian et al, 2010). Among these laboratory pyrolysis results, only Andresen et al. (1995) applied hydrous condition in the laboratory. The highest temperature that Andresen et al. (1995) reached in hydrous pyrolysis experiments was 365°C (for 10 days), indicating it is very likely that the three source rocks that they used, the Miocene brown coal, the Ness Formation coal, and the Kimmeridge Clay Formation, underwent only primary cracking, just as in the experiments described here. Lorant et al. (1998) suggested that the earliest methane was generated from isotopically heavier methyl group precursors, bound with relatively labile C-O or C-S bonds (Chung and Sackett, 1980; Smith et al., 1985) at conditions of isothermal heating at 300°C/72h. Later, methane was generated from isotopically lighter methyl group

136

Figure 6.16 A new model is proposed in this study for explaining different trends in the natural gas plot for Type II and Type III kerogen without any alteration by secondary processes such as gas migration and leakage. For Type II kerogen, the dogleg trend represents gas mainly generated from primary cracking, while the linear trend shows more gas generated from secondary cracking.

137

precursors, or mixes of it, making the δ 13C value of methane generated from 330°C/72h and 360°C/72h lighter or more depleted than the earliest generated methane. Both Y. Tang (unpublished data) and C. Bomen & F. Lorant (unpublished data) have observed this overall isotopic fractionation trend, showing that isotope ratios become lighter in the early stage of gas generation and would become heavier in the late stage of gas 13 generation. They interpreted that the decreasing δ C-CH4 values can be explained by mixing with methane generated from isotopically lighter, more tightly-bound precursors (Tang and Jenden, 1998; Tang et al., 2000). Cramer et al. (2004) using four independent reactions to explain the carbon isotope and dynamics of thermal methane generation from a Carboniferous coal. His model result showed that the lighter isotope ratio at the beginning of the reaction is the result of mixing two parallel reaction complexes, which are demethylation and release of groups which cross link ring structures or secondary cracking of long chain hydrocarbons within the molecular network of the coal (Fig.6, Cramer, 2004). However, this explanation contradicts established kinetic isotope effects, specially that the 12C-12C bond is easier to break than the 13C-13C bond. Therefore, it is probably fair to state that it is not presently possible to arrive at general principles for controls on the isotopic composition of gases. The heating rate is another critical factor. The Arrhenius equation best describes the temperature effects on the rate constant:

k = Af exp (-Ea/RT)

138

where both Af and Ea are constants, Af is the frequency factor and Ea is the activation energy. Tang et al. (2000) applied theoretical mathematical modeling to calculate stable carbon isotope compositions in natural gases. They found that the fractionation factor between two different species such as two different gases is strongly dependent on the temperature. Laboratory experimental temperatures (360°C) are usually higher than geological heating temperatures (200°C), so the different reaction rates of generation 13 12 between CH4 and CH4 in nature, is more significant than that in the laboratory experiments. Therefore, we expect to see more significant isotopic fractionation in gases from the field than in laboratory experiments, based on the theoretical Arrhenius equation.

6.6.1 Empirical Diagrams Empirical diagrams using carbon isotope compositions have been developed by several authors to classify and correlate origins of natural gas. The Bernard diagram (Figure 6.18) modified after Bernard (1978), and Faber and Stahl (1984), compares the 13 molecular ratio C1/ (C2+C3) of natural gas and its δ C1. The diagram is mainly used to distinguish thermogenic gas from bacterial gas, or to identify the source of thermogenic gas and its maturation path. Secondary factors such as mixing can also be recognized in the diagram. Gases generated from the Cameo Coal, Mowry, Mancos, and Baxter Shales by sequential hydrous pyrolysis experiments, all plot within the thermogenic gas area; source rock (or kerogen) types are on this plot. It is likely that the distinctions between Type II and Type III kerogen can only be made at the maturities in the secondary cracking range. The red arrow in Figure 6.18 indicates the direction of increasing experimental heating temperatures, but noneof the experimental data reported here follows the direction of increasing maturity in the Bernard diagram.

Figure 6.19 shows the plot of ln(C1/C2) versus ln(C2/C3) that Prinzhofer and Huc (1995) proposed for differentiating hydrocarbon gases generated by primary cracking of kerogens versus from gases produced by secondary cracking of oil. For the Mowry Shale, the ethane/propane (C2/C3) ratio of the generated gas slightly decreases with increasing maturity, while the methane/ethane (C1/C2) ratio of the gas increases. This molecular

139

variation corresponds to the direction of the primary cracking identified by Prinzhofer and Huc (1995). However, for the Cameo Coal, Mancos and Baxter Shales, the

ethane/propane (C2/C3) ratios of the generated gases increase while the methane/ethane

(C1/C2) ratios of the gases decrease from the 300°C/72h to the 330°C/72h experiments.

For the Cameo Coal, both the (C1/C2) and (C2/C3) ratios slightly increase from 330°C/72h to 360°C/72h conditions. In other words, only gases generated from the Mowry Shale follow the primary cracking path identified by Prinzhofer and Huc (1995). 13 13 13 Tian et al. (2010) proposed a cross-plot of δ C1 vs. δ C2–δ C3 which may also be effective in distinguishing gas from oil cracking and from kerogen cracking (Figure 6.20: Tian, 2006; Guo et al., 2009). Generally, carbon isotopic values of methane from oil cracking are more negative than those from kerogen cracking at similar maturities. Although gases from the cracking of both kerogen and oil are characterized by a positive 13 13 13 relationship between δ C1 and the δ C2–δ C3, they have different trends. Gases generated from Cameo Coal, Mowry, Mancos, and Baxter Shales seem to be in good agreement with fields identified on this plot. Another approach to distinguishing primary cracking from secondary cracking is 13 13 13 13 by plotting δ Ci-δ Cj versus ln(Ci/Cj). The δ C2- δ C3 versus C2/C3 diagram (Lorant et al., 1998) use both the chemical and isotopic signatures of ethane and propane to characterize the maturity of the gas, as shown in Figure 6.21. An important parameter to take into account is the degree of openness of the system. Gases generated from primary cracking of kerogen from experimental samples plotted in the primary cracking field on cross-plot, except for gases generated from the Baxter Shale. However, there is potential 13 13 inaccuracy in determination of δ C2 of gas in the Cameo Coal, the δ C3 of gases in the 13 13 Mancos and Mowry Shales, and especially the δ C2 and δ C3 of gases in the Baxter Shale, due to very little amount of ethane and propane generated from the Baxter Shale. Therefore, the Baxter Shale data points in this plot might not be accurate. Prinzhofer et al. (2000) characterized the maturity of gas by using the chemical signatures of the C2-C4 fraction of gases. This approach is based on the principle that iC4 decreases more rapidly than C3 with increasing thermal maturity. In contrast, nC4 is more easily altered than iC4 during biodegradation. Gases generated from the four source rocks are plotted on the diagram, and they all follow predicted trends of increasing maturity

140

Figure 6.18 The empirical Bernard diagram is used to classify natural gases in the field 13 by using the molecular ratio C1/ (C2+C3) and carbon isotope ratios of methane δ C-CH4. Modified after Bernard (1978) and Faber and Stahl (1984). Gases generated from the Cameo Coal (blue), Mowry Shale (red), Mancos Shale (green), and Baxter Shale (purple) in the laboratory are plotted on the Bernard Diagram. It shows that the laboratory- generated gas does not follow the maturity trend as field gases do, but they are appropriately plotted in the thermogenic gas area. Red arrows indicate increasing laboratory temperatures (thermal maturity).

141

330

360 330 360 360 360 330 300 330

300

Figure 6.19 Diagram of the molecular proportion of C2/C3 versus C1/C2, in logarithmic scales. Data are from gases generated from the Cameo Coal (blue), Mowry Shale (red), Mancos Shale (green), and Baxter Shale (purple) by experimental hydrous pyrolysis. Arrows point towards the path of increasing maturation. Modified after Prinzhofer and Huc (1995).

13 13 13 Figure 6.20 Cross plot of δ C1 vs. δ C2–δ C3 for gases from kerogen and oil cracking. Modified after Tian et al. (2010).

142

(Figure 6.22). Biodegradation was not expected to occur in our laboratory experiments. Compositions of gases generated in these laboratory experiments are not generally consistent with empirical trends developed on the basis of field gases in these widely

used diagrams, with the exception of the C2/C3 versus C2/iC4 diagram (Prinzhofer et al., 13 13 2000) and the δ C2- δ C3 versus C2/C3 diagram (Lorant et al., 1998). The discrepancy between these experimental results and trends deduced from field gases may arise because of secondary processes such as dissolution of gases in the fluid and migration. Laboratory can only simulate gases generated and expelled within the source rocks. Any processes happen after expulsion can alter the gas composition significantly. Therefore, none of the empirical diagrams are recommended for use to interpret laboratory generated gases.

13 13 Figure 6.21 The δ C2- δ C3 versus C2/C3 diagram (Lorant et al., 1998) of gases generated from primary cracking of the Cameo Coal (blue), Mowry (red), Mancos (green), and Baxter Shales (purple) in hydrous pyrolysis experiments.

143

Figure 6.22 The diagram C2/C3 versus C2/iC4, distinguishes a maturity trend from a biodegradation trend. Data are from gases generated from the Cameo Coal (blue), Mowry Shale (red), Mancos Shale (green), and Baxter Shale (purple) in hydrous pyrolysis experiments. Modified after Prinzhofer et al. (2000).

6.7 Comparison to Other Published Hydrous Pyrolysis Gas Data Isotope data on gases generated by hydrous pyrolysis from different source rocks with Type II and Type III kerogen have been published by Chung et al. (1988), Andresen et al. (1995), and Kotarba et al. (2009). The temperature of these pyrolysis experiments are all below 365°C and the longest duration for experiment is 240 hours, indicating all experiments generated gases by primary cracking. Figure 6.23 plots all published data, together with data from gas generated from secondary cracking of Smackover oil by hydrous pyrolysis experiment (Willette, 2010). The gas generated from secondary cracking of Smackover oil gives a linear trend, whereas gases generated from other primary cracking pyrolysis experiments give dogleg or convex trends. Figure 6.24 compares all gases generated from the Type III kerogen (mainly coals). Different coals under different heating programs, ending at different final 13 temperatures in different laboratories all show consistent values of δ C of C2 and C3. Figure 6.25 compares all gases generated from the mixed Type II/III kerogen (mainly marine shales). The Mancos, Mowry, and Polish Type II kerogen give similar δ13C values

144

13 of C2 and C3, but the Woodford, Baxter and Kimmeridge Shales show distinct δ C values of C2 and C3. One possibility is that both the Woodford and Kimmeridge Shales are much more lipid-rich and TOC-rich (Mnich, 2009; Schwarzkopf, 1992) than most of the other samples. Alternatively, the difference in δ13C may be due to the different chemical structures in Type II kerogen, indirectly related to their slightly different sources and depositional environments.

Figure 6.23 All published gas isotope data on different source rocks by hydrous pyrolysis for Type II kerogen and Type III kerogen (Chung et al., 1988; Andresen et al., 1995; Kotarba et al., 2009; Willette, 2010). Gas generated from secondary cracking of Smackover oil is also included in the plot for comparison, the linear trend in purple.

145

Figure 6.24 gases generated from the Type III kerogen (mainly coals). Different coals under different heating programs and final temperatures in different laboratories all show 13 consistent values of δ C of C1 to C3.

Figure 6.25 Gases generated from the Type II kerogen (mainly marine shales). Mancos, 13 Mowry, and Polish Type II kerogen show consistent δ C values of C1 to C3, but the 13 Woodford, Baxter and Kimmeridge Shales show distinct δ C values of C2 to C3.

146

6.8 Comparison of Gas Isotope Compositions in Field and Laboratory Gases Carbon isotope data of the mud gas samples collected from Jonah Field and the Piceance Basin were plotted in Figure 6.26 and 6.27 respectively, to compare with the gas isotope data in the potential source rocks in Jonah Field and the Piceance Basin (Harris and Philp, unpublished data). The median and mode carbon isotope ratios from field gas samples are used to represent the overall field data. The carbon isotope ratios of ethane and propane from field gases are heavier than those from gases generated from experiments on potential source rocks, while methane compositions of field gases are similar to experimental gases (Jonah Field) or are lighter than the experimental gases (Piceance Basin). Two possible reasons are proposed: 1) Possibility of migration effects, both from source into the reservoir and escaping gas from the reservoir that may significantly alter final gas compositions; 2) Gas isotopes from the lab only reach early- stage of petroleum generation: primary cracking, whereas field gas represents gas from both primary cracking of kerogen and secondary cracking of oils. Oil that was generated from the lab and that collected from the field have different characteristics: waxy oil from the laboratory does not exist in the field (Meade and Brown, personal communication). In contrast, both the Jonah Field and Piceance Basin fields produce condensate (Meade and Brown, personal communication), which represents the relatively mature stage of petroleum generation.

147

Figure 6.26 The cumulative δ13C of methane to propane in the Jonah Field. Field gas is the dashed line and lab gas is the solid line.

Piceance 1 Piceance 2 Piceance 3 Piceance 4 Piceance 5 Piceance 6 Piceance 7 Piceance 8 Piceance 9

Figure 6.27 The cumulative δ13C of methane to propane in southern Piceance Basin. Field gas is the dashed line and lab gas is the solid line. 148

CHAPTER 7

CONCLUSIONS

Hydrous pyrolysis has been recognized as one of the better laboratory methods for simulating thermogenic gas generation from marine source rocks and coals (Qin et al., 1994; Teerman and Hwang, 1991, Andresen et al., 1995; Kotarba and Lewan, 2004). The experiments on one coal source rock (Cameo Coal) and three marine source rocks (Mancos Shale, Mowry Shale, and Baxter Shale) were conducted as part of an investigation of source rocks for natural gas in the Rocky Mountain area. Experimental conditions simulated thermal

maturities of 1.0, 1.3, and 1.6 %Ro. Gases under these conditions were generated by primary cracking and because of the experimental conditions, gas compositions were not affected by secondary processes such as migration, dissolution in formation waters or leaking from reservoirs.

Significant findings are:

• On a per gram TOC basis, hydrocarbon gas yields for the four source rocks were relatively similar, varying by less than 100% for methane and by 250% for propane and butane. Because the Cameo Coal is much richer in organic carbon than the other source rocks, however, gas yields on a per-volume of source rock basis are highest from the Cameo Coal.

• During primary cracking, the bulk composition of generated hydrocarbon gases varied by formation and in some cases, thermal maturity. Gases generated from the Cameo Coal and Mancos Shale were relatively dry and changed little at low thermal maturity but gas in the Mancos Shale becomes wetter as thermal maturity increased. Gases generated from the Baxter and Mowry were wet at low thermal maturity but became dryer as thermal maturity increased. Kerogen type has been considered to be the main controlling factor on the bulk composition of gas.

However, the different yields and proportion of methane and C2+ gases generated

149

from the mixed Type II/III kerogen (Mowry, Mancos, and Baxter Shales) during primary cracking suggests that the kerogen structure is also very critical in determining the bulk composition of gases.

• Non-hydrocarbon gases such as CO2, N2, H2S, H2 were also generated in the

sequential hydrous pyrolysis experiment, with CO2 being generated in the most 13 significant amount. The δ C of CO2 suggests that carbon dioxide generated from the Mancos and Baxter Shales had an inorganic source - from the thermal decomposition of carbonate, although the specific reaction is unknown. Carbon dioxide generated from the Cameo Coal is entirely sourced from organic matter. Carbon dioxide generated from the Mowry Shale is possibly derived entirely from authigenic carbonate or from a mixture derived from both organic and inorganic sources, with large proportion of gas sourced from carbonate materials.

• There was no consistent trend in the carbon isotopic composition of gases generated from the four source rocks as a function of thermal maturity. Carbon isotopic compositions of methane became isotopically more negative from

equivalent thermal maturity of 1.0 % to 1.6% Ro, contrary to expected behavior in which gases become more positive with increasing thermal maturity. The carbon isotopic compositions of ethane and propane gases were virtually unchanged or became slightly more positive with increasing maturity.

• The carbon isotopic compositions of gases generated from individual source rock became more positive with increasing carbon number, consistent with numerous previous studies. Previous studies suggested that where the carbon isotopic composition varies linearly as a function of the reciprocal carbon number (Chung plot), gas had been generated from a single source rocks. However, these experiments clearly show non-linear trends in gases generated from single source rocks. An improved model is proposed in this study for the mixed Type II/III and Type III kerogen in which the dogleg trend on Chung’s plots represents gas mainly generated from primary cracking. Linear trends on these plots in fact

150

reflect gas generated from secondary cracking. This can be explained by the fact that during secondary cracking, ethane and propane both crack to form methane. Therefore, ethane and propane are expected to become enriched in 13C, whereas methane is expected to be depleted in 13C.

• Gases generated from Type III and mixed Type II/III kerogen give overlapping methane carbon isotope signatures. However, the carbon isotopic composition of ethane and propane can successfully differentiate different types of kerogen. 13 Ethane and propane generated from Type III kerogen show heavier δ C2 and 13 δ C3 than those generated from the mixed Type II/III kerogen.

• Gas isotope data from laboratory hydrous pyrolysis experiment differ considerably from gas isotopic compositions in field samples- the gas isotopes of the field gases are heavier than the ones from laboratory gases. Two possible explanations are suggested: 1) Possibility of migration effects, both from source into the reservoir and escaping gas from the reservoir that may significantly alter final gas compositions; 2) Gas isotopes from the lab only reach the early-stage of petroleum generation: primary cracking, whereas field gas may represent gas from both primary cracking of kerogen and secondary cracking of oils.

151

152

CHAPTER 8

FUTURE WORK

Conclusions and limitations identified by this study indicate several areas of future work that would be valuable advancing our understanding of tight gas sand reservoirs and in applying gas isotopes in these fields. The scope of this study was essentially restricted to the experimental temperature using source rock samples, so only gases generated from primary cracking of the potential source rocks were simulated. It would be useful to obtain gas isotope data from secondary cracking of oils generated from marine shales and gas isotope data from the entire dry gas window from coals. Characterizing the isotope composition of gas formed from both primary and secondary cracking and correlating the data to thermal maturation will be valuable when compared to the field gas. The relationship among carbon isotope, kerogen type in source rocks, and thermal maturity has to be redefined separately for primary, secondary cracking, and the combination of both of them. Representative immature core samples will be ideal for the study. The Baxter Shale used in this study is essentially a siltstone and, as a result, the generation potential may have been underestimated. The Mancos Shale core only represents the very upper small portion of the lower Mancos Shale Group. Obtaining and carrying out experiments on samples more representative of these formations as a whole would be very beneficial. Elemental analysis and mineralogy identification would be useful before the hydrous pyrolysis experiment. Sample preparation such as acid washing might be necessary if the carbonate portion of the source rock exceeds 50% of its original components. The limitation in the accuracy of some carbon isotope ratios hindered an accurate determination of the controls on thermal maturity and kerogen type. In addition,

developing a model for the inorganic source of CO2 in the Mowry Shale will require further investigation into its mineral composition, both before and after the hydrous pyrolysis experiments. Additional analysis of carbon isotopes of solid kerogen can provide useful information on the extent of fractionation from kerogen to gas. The extent of fractionation can provide constrains in the future theoretical calculation and modeling.

152

153

REFERENCES

Adams, P. R., 1972, Foraminiferal biostratigraphy and paleoecology of the Hilliard Formation type area, Lincoln County, Wyoming [Master Thesis]: Laramide, University of Wyoming, 56 p.

Alexander, R., R. I. Kagi, and A. V. Larcher, 1982, Clay catalysis of aromatic hydrogen- exchange reactions, Geochimica et Cosmochimica Acta, v. 46, p. 219–222.

Andresen, B., T. Throndsen, A. Raheim, J. Bolstad, 1995, A comparison of pyrolysis products with models for natural gas generation, Chemical Geology, v. 126, p. 261–280.

Anderson, D. S., 2010, Colorado School of Mines, personal communication.

Anderson, D. S., and N. B. Harris, 2006, Integrated sequence stratigraphic and geochemical resource characterization of the Lower Mancos Shale, Uinta Basin, Utah, Utah Geological Survey Open File Report 483, 68 p.

Armstrong, R. L., 1968, Sevier orogenic belt in Nevada and Utah, GSA Bulletin, vol. 79, p. 429– 458.

Arthur, M. A., W. E. Dean, L. M. Pratt, 1988, Geochemical and climatic effects of increased marine organic carbon burial at the Cenomanian/ Turonian boundary, Nature, v. 335, p. 714-717.

ASME (American Society of Mechanical Engineers), 1979 and 1993.

Barker, C. E., 1989, Fluid inclusion evidence for paleotemperatures within the Mesaverde Group, Multiwell Experiment site, Piceance Basin, Colorado, in B. E. Law and C. W. Spencer, eds., Geology of tight gas reservoirs in the Pinedale anticline area, Wyoming, and at the Multiwell Experiment site, Colorado, U.S. Geological Survey Bulletin, v. 1886, p. M1 M11.

153

Barker, C., 1990, Calculated volume and pressure changes during the thermal cracking of oil to gas in reservoirs, AAPG Bulletin, v. 74, p. 1254-1261.

Baskin, D. K., 1997, Atomic H/C ratio of kerogen as an estimate of thermal maturity and organic matter conversion, AAPG Bulletin, v. 81, p. 1437-1450.

Behar, F. and M. Vandenbroucke, 1987, Chemical modeling of kerogens, Organic Geochemistry, v. 11, no. 1, p. 15-24.

Behar, F., S. Kressmann, J. L. Rudkiewicz, and M. Vandenbroucke, 1992, Experimental simulation in a Confined system and kinetic modeling of kerogen and oil cracking, Organic Geochemistry, v. 19, p. 173-189.

Behar, F., M. Vandenbroucke, Y. Tang, F. Marquis, and J. Espitalié, 1997, Thermal cracking of kerogen in open and closed systems: Determination of kinetic parameters and stoichiometric coefficients for oil and gas generation, Organic Geochemistry, v. 26, p. 321– 339.

Behar, F., H. Budzinski, M. Vandenbroucke, and Y. Tang, 1999, Methane generation from oil cracking: kinetics of 9-methylphenanthrene cracking and comparison with other pure compounds and oil fractions, Energy & Fuels, v. 13, p. 471-481.

Behar, F., F. Lorant, and M. D. Lewan, 2008, Role of NSO compounds during primary cracking of a Type II kerogen and a Type III lignite, Organic Geochemistry, v. 39, p. 1-22.

Berman, A. E., J. Poleschook, D., and T. E. Dimelow, 1980, Jurassic and Cretaceous systems of Colorado, in H. Kent, and K. Porter, eds., Colorado geology: Denver, Colorado, Rocky Mountain Association of Geologists, p. 111-128.

Bernard, B. B., 1978, Light hydrocarbons in marine sediments: Ph.D. dissertation, Texas A&M University, College Station, Texas, 144 p.

Berner, U. and E. Faber, 1988, Maturity related mixing model for methane, ethane and propane, based on carbon isotopes, Organic Geochemistry, v. 13, p. 67-72.

154

Berner, U., E. Faber, G. Scheeder, and D. Panten, 1995, Primary cracking of algal and land-plant kerogens: kinetic models of isotope variations in methane, ethane and propane, Chemical Geology, v. 126, p.233–245.

Berner, U. and E. Faber, 1996, Empirical carbon isotope/maturity relationships for gases from algal kerogens and terrigenous organic matter, based on dry, open-system pyrolysis, Organic Geochemistry, v. 24, p. 947-955.

Bigeleisen, J., 1965, Chemistry of isotopes, Science, v. 147, p. 463-471.

Blakey, R. C., 2008, Pennsylvanian-Jurassic sedimentary basins of the Colorado Plateau and southern Rocky Mountains in, Miall, A. D. ed., Sedimentary Basins of United States and Canada: Elsevier, Amsterdam, p. 245-296.

Bottinga, Y., and M. Javoy, 1973, Comments on oxygen isotope geothermometry, EPSL, v. 20, p. 250- 265.

Bowker, K. A., and J. W. Robinson, 1996, Jonah field: A basin-centered sweetspot in the northern Green River basin, Wyoming, in S. Longacre, B. Katz, R. Slatt, and M. Bowman, eds., Compartmentalized reservoirs: Their detection, characterization and management: 1997 AAPG Rocky Mountain Section Meeting Official Program Book and Extended Abstracts Volume, p.25–30.

Brooks, J. D., K. Gould, J. W. Smith, 1969, Isoprenois hydrocarbons in coal and petroleum, Nature, v. 222, p.257-259.

Burchfield, B. C., and G. A. Davis, 1975, Nature and controls on Cordilleran orogenesis, western U.S.: extensions of an earlier synthesis, American Journal of Science, v. 275A, p. 363– 396.

Burgoyne, T. W., and J. M. Hayes, 1998, Quantitative production of H2 by pyrolysis of gas chromatographic effluents, Analytical Chemistry, v. 70, p. 5136-5141.

Burnham, A.K., 1989, A simple kinetic model of petroleum formation and cracking, Lawrence

155

Livermore Natl. Lab., Livermore, Calif., Intern. Pubi. UCID-21665, 11 pp. Chung, H.M. and Sackett, W.M., 1979. Use of stable carbon isotope compositions of pyrolytically derived methane as maturity indices for carbonaceous materials, Geochimica et Cosmochimica Acta, vol. 43, p. 1979-1988.

Burruss, R. C., P. G. Lillis, and T. S. Collett, 2003, Geochemistry of natural gas, North Slope, Alaska, Implication of gas resources, NPRA, U.S. Geological Survey Open-File Report 03- 041.

Burruss, R. C., and C. D. Laughrey, 2010, Carbon and hydrogen isotopic reversals in deep basin gas: Evidence for lmits to the stability of hydrocarbons, Organic Geochemistry (In Press.)

Burtner, R.L., A. Nigrini, and R. A. Donelick, 1994, Thermochronology of Lower Cretaceous source rocks in the Idaho-Wyoming thrust belt, AAPG Bulletin, v. 78, p. 1613–1636.

Burtner, R.L., and Warner, M.A., 1984, Hydrocarbon generation in Lower Cretaceous Mowry and Skull Creek Shales of the Northern Rocky Mountain area, in Woodward, Jane, Meissner, F.F., and Clayton, J.L., eds., Hydrocarbon source rocks of the Greater Rocky Mountain Region: Rocky Mountain Association of Geologists, p. 449–467. Chung, H. M., J. R. Gormly, and R. M. Squires, 1988, Origin of gaseous hydrocarbons in subsurface environments: Theoretical considerations of carbon isotope distribution, Chemical Geology, v. 71, p. 97–103.

Byers, C.W., and Larson, D.W., 1979, Paleoenvironments of Mowry Shale (Lower Cretaceous), western and central Wyoming, AAPG Bulletin, v. 63, p. 354–375.

Chapin, M., A. Govert, G. Ugueto, 2009, Examining detailed facies and rock property variation in Upper Cretaceous, Tight Gas Reservoirs, Pinedale Field, Wyoming (abs.): AAPG National Convention, p. 1-6.

Chung, H.M., J. R. Gormly, and R. M. Squires, 1988, Origin of gaseous hydrocarbons in Subsurface environments: theoretical considerations of carbon isotope distribution, Chemical Geology, v. 71, p. 97–103.

Chung, H. M. and H. M. Sackett, 1979, Use of stable isotope compositions of pyrolytically

156

derived methane as maturity indies for carbonaceous materials, Geochimica et Cosmochimica Acta, v. 43, p. 1979-1988.

Clayton, C. J., 1991, Carbon isotope fractionation during natural-gas generation from kerogen, Marine Petroleum Geology, v. 8, p. 232–240.

Cobban, W.A., and J. B. Reeside, Jr., 1952a, Frontier Formation, Wyoming and adjacent areas, AAPG Bulletin, v. 36, no. 10, p. 1913-1961.

Cobban, W.A., and J. B. Reeside, Jr., 1952b, Correlation of the Cretaceous formations of the western interior of the United States, GSA Bulletin, v. 63, p. 1011-1043.

Coskey, R.J., 2004, Burial-history modeling for the Jonah Field area: an unusual and possibly unique gas accumulation in the Green River Basin, Wyoming, in J.W. Robinson and K.W. Shanley, eds., Jonah Field: Case Study of a Giant Tight-Gas Fluvial Reservoir: AAPG Studies in Geology 52, chapter 7, p. 93-125.

Cramer, B., B. M. Krooss, and R. Littke, 1998, Modeling isotope fractionation during primary cracking of natural gas: A reaction kinetic approach, Chemical Geology, v. 149, p. 235– 250.

Cramer, B., E. Faber, P. Gerling, and B. M. Krooss, 2001, Reaction kinetics of stable carbon isotopes in natural gas- insights from dry, open system pyrolysis experiments, Energy & Fuel, v.15, p. 517-532.

Cramer, B., 2004, Methane generation from coal during open system pyrolysis investigated by isotope specific, Gaussian distributed reaction kinetics, Organic Geochemistry, v. 35, p.379-392.

Cross, C.W., and C.W. Purington, 1899, Description of the Telluride quadrangle, Colorado: U.S. Geological Survey Geologic Atlas Folio 57, 19 p.

Cross, T. A., 1988, Controls on coal distribution in transgressive-regressive cycles, Upper Cretaceous, Western Interior, USA, in Wilgus, C. K., C. A. Ross, and H. Posamenties eds., Sea-level changes – An integrated approach, SEPM Special Publication No. 42, 400p.

157

Cui, X., R. M. Bustin, G. Dipple, 2004, Selective transport of CO2, CH4, and N2 in coals: insights from modeling of experimental gas adsorption data, Fuel, v.83, p. 293-303.

Cumella, S.P., and R. D. Cole, 2003, Stratigraphic Architecture and reservoir characteristics of the Mesaverde Group, southern Piceance Basin, Colorado, RMAG and AAPG Field Trip Guidebook, 47 p.

Currie, B. S., 2002, Structural Configuration of the Early Cretaceous Cordilleran Foreland-Basin system and Sevier Thrust Belt, Utah and Colorado, The Journal of Geology, v. 110, p. 697 - 718.

Curtis, J. B., 2008, Potential Supply of Natural Gas in the United States, Potential Gas Agency, Colorado School of Mines.

Curtis, J. B., 2010, personal communication.

Cushman, Jr., R.A., 1994, Palynostratigraphy of the upper Cretaceous Mancos Shale in western Colorado: Ph.D. dissertation, Colorado School of Mines, Golden, Colorado, T-3466, 686 p.

Dahl, J. E., J. M. Moldowan, K. Peters, G. Claypool, M. Rooney, G. Michael, M. Mellos and M. Kohnen, 1999, Diamondoid hydrocarbons as indicators of oil cracking, Nature, v. 399, p. 54–56.

Dahl, J. E., J. M. Moldowan, and P. A. Lipton, 2000, Diamondoids as source indicators, Molecular Organic Geochemistry Industrial Affiliates, Stanford University, Workshop, Internal Report.

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

Dai, J., C. Zou, J. Li, Y. Ni, G. Hu, X. Zhang, Q. Liu, C. Yang, A. Hu, 2009, Carbon isotopes of Middle–Lower Jurassic coal-derived alkane gases from the major basins of northwestern China, International Journal of Coal Geology, vol. 80, p.124-134.

158

Dartiguelongue, C., F. Behar, H. Budzinski, G. Scacchi, and P.M. Marquaire, Thermal stability of dibenzothiophene in closed system pyrolysis: Experimental study and kinetic modeling, Organic Geochemistry, v. 37, p. 98-116.

Dickinson, W. W., 1989, Analysis of vitrinite maturation and Tertiary burial history, northern Green River Basin, Wyoming, U.S. Geological Survey Bulletin, n. 1886-F, p. F1– F13.

Dominé, F., D. Dessort, and O. Brevart, 1998, Towards a new method of geochemical kinetic modeling: implications for the stability of crude oils, Organic Geochemistry, v. 28, p. 597– 612.

Domke, S. B., R. F. Pogue, F. J. R. Van Neer, and C. M. Smith, 2001, Investigation of the kinetics of ethylbenzene pyrolysis using a temperature-scanning reactor, Industrial & Engineering Chemistry Research, v. 40, p. 5878-5884.

Dow, W.C., 1977, Kerogen studies and geological interpretations, Journal of Geochemical Exploration, v. 7, p. 79–99.

DuBois et al., 2004, Geology of the Jonah Field, Sublette County Wyoming, in J.W. Robinson and K.W. Shanley, eds., Jonah Field: Case Study of a Giant Tight-Gas Fluvial Reservoir: AAPG Studies in Geology 52, chapter 5.

Durand, B., J. J. Fripiat, and R. Pelet, 1972, Alkylammonium decomposition on montmorillonite surfaces in an inert atmosphere, Clays and Clay Minerals, v. 20, p. 19-33.

Durand, B. and J. Espitalié, 1976, Geochemical studies on the organic matter from the Douala Basin (Cameroon)--II. Evolution of kerogen, Geochimica et Cosmochimica Acta, v. 40, p. 801-808.

Durand, B., 1980, (ed.) Kerogen, Insoluble Organic Matter from Sedimentary Rocks, Éditions Technip, Paris.

Ehleringer, J. R., T. E. Cerling, and M. D. Dearing, 2002, Atmospheric CO2 as a global change driver influencing plant-animal interactions, Integrative and Comparative Biology, v. 42, p. 424-430.

159

EIA (Energy Information Administration) Annual Energy Outlook, 2005 and 2009.

Ellis, M.S., V.L Freeman, and J.R. Donnell, 1988, Cross sections showing coal beds and coal zones in the Mesaverde Formation in the Carbondale 30' x 60' quadrangle, west-central Colorado: U.S. Geological Survey Coal Investigations Map C–97–B, scale 1:100,000.

Espitalié, J., J. L. Laporte, M. Madec, F. Marquis, P. Leplat, J. Paulet, and A. Boutefeu, 1977, Méthode rapide de caractérisation des roches-mères, de leur potentiel pétrolier et de leur degré Devolution, Rev. Inst. Fr. Pet., vol. 32, p.23-42.

Espitalié, J., P. Ungerer, I. Irwin, and F. Marquis, 1987, Primary cracking of kerogens:

Experimental and modelling C1, C2-C5, C6-C15, and C15 classes of hydrocarbons formed, in L. Mattavelli and L. Novelli, eds., Advances in Organic Geochemistry, Pergamon, Oxford (1988), pp. 893–899.

Faber, E. and W. Stahl, 1984, Geochemical surface exploration for hydrocarbon in North Sea, AAPG Bulletin, v. 68, p.363-386.

Fabuss, B.M., J. O. Smith and Satterfield, 1964, Thermal cracking of pure saturated hydrocarbons, in J. McKetla, Jr., ed., Advances in Petroleum Chemistry and Refining Vol. 5, Wiley, New York (1964), pp. 156–201 chapter 4 .

Faure and Mensing, 2005, Isotopes: Principles and Applications, 3rd Edition, Wiley.

Federal Energy Regulatory Commission (FERC), 2009, website accessed http://www.ferc.gov/industries/gas/gen-info.asp

Finn, T. M., R. C. Johnson, and S. B. Roberts, 2005, The Mesaverde-Lance-Fort Union composite total petroleum system, southwestern Wyoming Province, Chapter 10 of Petroleum Systems and Geological Assessment of Oil and Gas in the Southwestern Wyoming Province, Wyoming, Colorado, and Utah, U.S. Geological Survey Digital Data Series DDS-69-D.

Fisher, R., 2007, Stratigraphic, Geochemical, and Geomechanical Analysis of the Lower Mancos

160

Shale, Douglas Creek Arch, Northwest Colorado, U.S.A.: M.S. Thesis, Colorado School of Mines, Golden, 179 p.

Frerichs, W. E. and J. R. Steidtmann, 1971, Sea level variations during deposition of the Hilliard Formation (Upper Cretaceous), western Wyoming, Wyoming Geological Association 23rd Annual Field Conference Guidebook, p. 157-168.

Galimov, E.M., Shabaeva, I. Yu., 1985, Carbon isotope composition of CH4 and CO2 in sediments of the Middle American Trench, in von Huene, R., Aubouin, J., eds., Initial Reports of the Deep Sea Drilling Project, 84, U.S. Government Printing Office, Washington, pp. 693–694.

Galimov, E. M., 1988, Sources and mechanisms of formation of gaseous hydrocarbons in sedimentary rocks, Chemical Geology, v. 71, p. 77–95.

Galimov, E. M., 1989, Sources and formation mechanisms of hydrocarbon gases in sediments, Geochemistry International, v. 26, p.1–15.

Gardner, M.H., and T.A. Cross, 1994, Middle Cretaceous paleogeography of Utah, in M.V. Caputo, J.A. Peterson, and K.J. Franzcyk eds., Mesozoic systems of the Rocky Mountain region, USA: Denver, Rocky Mountain section of the Society for Sedimentary Geology, p. 471 - 501.

Gaveau, B., R. Letolle, and M. Monthioux, 1987, Evaluation of kinetic parameters from 13C isotopic effect during coal pyrolysis, Fuel, v. 66, p. 228–231.

Govert, 2009, Petrogenetic controls on the porosity of tight-gas sandstones in the Lance Formation, Pinedale Anticline, Sublette Country, Wyoming [Master Thesis]: Colorado School of Mines, Golden, Colorado.

Grose, L. T., 1972, Tectonics: Geologic Atlas of the Rocky Mountain Region: RMAG, p. 35-44.

Hanson et al., 2004, Structural Geology, Seismic Imaging, and Genesis of the Giant Jonah Gas Field, Wyoming, U.S.A., in J.W. Robinson and K.W. Shanley, eds., Jonah Field: Case Study of a Giant Tight-Gas Fluvial Reservoir: AAPG Studies in Geology 52, chapter 6.

161

Harris, N.B., 2010, Colorado School of Mines, personal communication

Harris, N. B. and R. P Philp, unpublished data.

Haskett, G. I., 1959, Niobrara Formation of Northwest Colorado, in J. D. Haun, and R. J. Weimer, eds., Symposium on Cretaceous Rocks of Colorado and Adjacent Areas: Denver, Rocky Mountain Association of Geologists, p. 46-49.

Hayes, J.M., B. N. Popp, R. Takigiku, and M. W. Johnson, 1989, An isotopic study of biogeochemical relationships between carbonates and organic carbon in the Greenhorn Formation, Geochimica et Cosmochimica Acta, v. 53, p. 2961–2972.

Henry, A. A. and M. D. Lewan, 1999, Comparison of kinetic model predictions of deep-gas generation: USGS Open-File Report No. 99-326, 34p; also in USGS Bulletin 2173.

Hettinger, R.D., L.N.R Roberts, and T. A. Gognat, 2000, Investigations of the distribution and resources of coal in the southern part of the Piceance Basin, Colorado, Chapter O of Kirschbaum, M.A., Roberts, L.N.R., and Biewick, L.R.H., eds., Geologic assessment of coal in the Colorado Plateau—Arizona, Colorado, New Mexico, and Utah: U.S. Geological Survey Professional Paper 1625–B, 61 p, [available only on CD-ROM].

Hettinger, R.D. and M.A. Kirschbaum, 2002, Stratigraphy of the Upper Cretaceous Mancos Shale (Upper Part) and Mesaverde Group in the Southern Part of the Uinta and Piceance Basins, Utah and Colorado, U.S. Geologic Survey, Geologic Investigations Series I-2764.

Hettinger, R.D. and M.A. Kirschbaum, 2003, Stratigraphy of the Upper Cretaceous Mancos Shale (upper part) and Mesaverde Group in the southern part of the Uinta and Piceance basins, Utah and Colorado, in Petroleum systems and geologic assessment of oil and gas in the Uinta-Piceance Province, Utah and Colorado: U.S. Geological Survey Digital Data Series DDS-69-B, chapter 12, 21 p.

Hilkert, A. W., C. B. Douthitt, H. J. Schlüter, and W. A. Brand, Isotope ratio monitoring gas chromatography/Mass spectrometry of D/H by high temperature conversion isotope ratio mass spectrometry, Rapid Communications in Mass Spectrometry, v. 13, p. 1226-1230.

162

Hill, R. J., Y. Tang, I. R. Kaplan, and P. D. Jenden, 1996, The influence of pressure on the thermal cracking of oil, Energy & Fuels, v. 10, p. 973-882.

Hill, R. J., Y. Tang, and I. R. Kaplan, 2003, Insights into oil cracking based on laboratory experiments, Organic Geochemistry, v. 34, p. 1651– 1672.

Horne, J. C., 2009, The influence of depositional environments on coal stratigraphy, a field seminar, The Rocky Mountain Association of Geologist, 101 p.

Horsfield, B., H. J. Schenk, N. Mills, and D. H. Welte, 1992, An investigation of the in-reservoir conversion of oil to gas: compositional and kinetic findings from closed-system programmed-temperature pyrolysis, Organic Geochemistry, v. 19, p. 191-204

Horsfield, B. and J. Rullkötter, 1994, Diagenesis, catagenesis, and metagenesis of organic matter, The petroleum system- from source to trap, in L.B. Magoon, and W.G. Dow, eds., 1994, AAPG Memoir 60, p. 189-199.

Horsfield, B., 1997, The bulk composition of first-formed petroleum in source rocks, Springer Verlag, Heidelberg, Federal Republic of Germany, 335-402 pp.

Huang, D., Liu, B., Wang, T., Xu, Y., Chen, S., Zhao, M., 1999, Genetic type and maturity of Lower Paleozoic marine hydrocarbon gases in the eastern Tarim Basin, Chemical Geology, v. 162, p 65-67.

Huizinga, B. J., Z.A. Aizenshtat, and K. E. Peters, 1988, Programmed pyrolysis-gas chromatography of artificially matured Green River kerogen, Energy & Fuels, v. 2, p. 74- 81.

Hunt, J. M., 1984, Generation and migration of light hydrocarbons, Science, v. 226. no. 4680, p. 1265 – 1270.

Hunt, J. M., 1996, Petroleum geochemistry and geology: New York, W. H. Freeman and Co., Second Edition, 743 p.

163

Hutcheon, I., and H. J. Abercrombie, 1989, The role of silicate hydrolysis in the origin of CO2 in sedimentary basins, Proceedings of the 6th International Symposium on Water-Rock Interaction, A. A. Balkema, Rotterdam, The Netherlands, p 321-324.

James, A. T., 1983, Correlation of natural-gas by use of carbon isotopic distribution between hydrocarbon components, AAPG Bulletin, v. 67, p.1176–1191.

James, A. T., 1990, Correlation of reservoired gases using the carbon isotopic compositions of wet gas components, AAPG Bulletin, v. 74, p. 1441-1458.

Jenden, P. D., K. D. Newell, I. R. Kaplan, and W. L. Watney, 1988, Composition and stable- isotope geochemistry of natural gases from Kansas, Midcontinent, USA, Chemical Geology, v. 71, p. 117–147.

Jenden, P. D., D. J. Drazan, and I. R. Kaplan, 1993, Mixing of thermogenic natural gases in northern Appalachian basin, AAPG Bulletin, v. 77, p. 980–998.

Johnson, R. C., and V. F. Nuccio, 1986, Structural and thermal history of the Piceance Creek Basin, western Colorado, in relation to hydrocarbon occurrence in the Mesaverde Group, in C. W. Spencer and R. F. Mast, eds., Geology of tight reservoirs: AAPG Studies in Geology 24, p. 165–205.

Johnson, R.C., 1989, Geologic history and hydrocarbon potential of Late Cretaceous-Age low permeability reservoirs, Piceance Basin, western Colorado: U.S. Geological Survey Bulletin 1787-E.

Johnson, R. C. and D. D. Rice, 1990, Variations in composition and origins of gases from coal bed and conventional reservoirs, Wind River Basin, Wyoming, in Keefer, W.R., Metzger, W.J., and Godwin, L.H., eds, Oil and gas and other resources of the Wind River Basin, Wyoming: Wyoming Geological Association Special Symposium Volume, p. 319–335.

Johnson, R.C., and R.M. Flores, 2003, History of the Piceance Basin from Latest Cretaceous through Early Eocene and the characterization of Lower Tertiary sandstone reservoirs in

164

Peterson, K.M., Olson, T.M., and Anderson, D.S., eds., Piceance Basin 2003 Guidebook: Rocky Mountain Association of Geologists, Denver, CO, p. 21-62.

Johnson, R.C., and S. B. Roberts, 2003, The Mesaverde total petroleum system, Uinta-Piceance Province, Utah and Colorado: in Petroleum Systems and Geologic Assessment of Oil and Gas in the Uinta-Piceance Province, Utah and Colorado, Chapter 7, USGS Digital Data Series DDS-69-B, 2007: U.S. Geological Survey, Denver, CO, p. 1-41.

Johnson et al., 2004, Regional Stratigraphic Setting of the Maastrichtian Rocks in the Central Rocky Mountain Region, in J.W. Robinson and K.W. Shanley, eds., Jonah Field: Case Study of a Giant Tight-Gas Fluvial Reservoir: AAPG Studies in Geology 52, chapter 4.

Jordan, T., 1981, Thrust loads and foreland basin evolution, Cretaceous, Western United States, AAPG Bulletin, v. 65., no. 12, p. 2506-2520.

Jüntgen, H. E., and J. Karweil, 1966, Gasbildung und Gasspeicherung in Steinkohlenflozen: Parts I and II: Erdol Kohle, Erdgas, Petrochem, v. 19. p. 251–258, 339–344.

Katz, B. J., 2002, Gas Geochemistry – a key to understanding formation and alteration processes, Indonesian Petroleum Association, 28th annual convention proceedings, vol. 1, p. 789-802.

Kauffman, E.G., 1977, Geological and biological review: Western Interior Basin: in Kauffman, E. G., ed., Cretaceous facies, faunas, and paleoenvironments across the Western Interior Basin: The Mountain Geologist, v. 14, no. 3-4, p. 75-99.

Kauffman, E.G., 1984, Paleobiogeography and evolutionary response dynamic in the Cretaceous Western Interior seaway of North America, in Westermann , G.E.G., ed., Jurassic- Cretaceous biochronology and paleogeography of North America: Geological Association of Canada, Special Paper 27, p. 274-306.

Kauffman, E.G., 1985, Cretaceous evolution of the western interior basin of the United States, in L. M. Pratt, E.G. Kaufman, F.B. Zelt eds., Fine-grained deposits and biofacies of the Cretaceous Western interior seaway: evidence of cyclic sedimentary processes, Society of Economic Paleontologists and Mineralogists (SEPM) field trip guidebook No. 4,

165

p. IV – XIII.

Kent, H.C., 1965, Biostratigraphy of the lower Mancos Shale (Cretaceous) in northwestern Colorado: Boulder, Colorado, University of Colorado Press, 162 p.

Kent, H.C., 1968, Biostratigraphy of the Niobrara-equivalent part of the Mancos Shale (Cretaceous) in northwestern Colorado, AAPG Bulletin, v. 52, p. 2098-2115.

Kirschbaum, M. A. and L. N. R., Roberts, 2005, Geologic Assessment of Undiscovered Oil and Gas Resources of the Mancos/Mowry Total Petroleum System, Uinta-Piceance Province, Utah and Colorado, U.S. Geological Survey Digital Data Series DDS-69-B, Chapter 6.

Kissin, Y. V., 1987, Catagenesis and composition of petroleum: origin of n-alkanes and isoalkanes in petroleum crudes, Geochimica et Cosmochimica Acta, v. 51, p. 2445-2457.

Knauss, K. G., S. A. Copenhaver, R. L. Braun, and A. K. Burnham, 1997, Hydrous pyrolysis of New Albany and Phosphoria Shales: production kinetics of carboxylic acids and light hydrocarbons and interactions between the inorganic and organic chemical systems, Organic Geochemistry, v. 27, p. 477-496.

Kotarba, M. J., 1988. Geochemical criteria for origin of natural gases accumulated in the Upper Carboniferous coal-seam-bearing formations in Walbrzych Coal Basin, Zesz. Nauk. Akad. Gorn.-Hutn. (Acad. Min. Metall. Bull.) No. 1199 Geologia (Geology), v. 42, pp. 1–119 (in Polish with English abstract).

Kotarba, M. J. and M. D. lewan, 2004, Characterizing thermogenic coalbed gas from Polish coals of different ranks by hydrous pyrolysis, Organic Geochemistry, v. 35, p. 615-646.

Kotarba, M. J., and K. Nagao, 2008, Composition and origin of natural gases accumulated in the Polish and Ukrainian parts of the Carpathian region: gaseous hydrocarbons, noble gases, carbon dioxide and nitrogen, Chemical Geology, v. 255, p. 426–438.

166

Kotarba, M. J., J. B. Curtis, and M. D. Lewan, 2009, Comparison of natural gases accumulated in Oligocene strata with hydrous pyrolysis gases from Menilite Shales of the Polish Outer Carpathians, Organic Geochemistry, v. 40, p. 769-783.

Krauskopf, K. B., and D. K. Bird, 1975, Introduction to Geochemistry – 2nd Edition, McGraw- Hill Science/ Engineering/ Math, 640 p.

Krooss, B. M., R. Littke, B. Müller, J. Frielingsdorf, K. Schwochau, E. F. Idiz, 1995, Generation of nitrogen and methane from sedimentary organic matter: implications on the dynamics of natural gas accumulations, Chemical Geology, v. 126, p. 291–318.

Kuuskraa, V. A., 2009, Paradigm Shift in Domestic Natural Gas Resources, Supplies and Costs, the EFI Gas-to-Market & Energy Conversion Forum, Washington, D.C., powerpoint presentation.

Kuzniak, K., 2009, New stratigraphic interpretations, geochemistry, and petrophysics of the Lower Mancos Group, Douglas Creek Arch, northwestern Colorado, U.S.A.: M.S. Thesis, Colorado School of Mines, Golden, 161 p.

Langford, F. F., M. M. Blanc-Valleron, 1990, Interpreting Rock-Eval pyrolysis data using graphs of pyrolizable hydrocarbons vs. total organic carbon, AAPG Bulletin, v. 74, p. 799-804.

Larter, S. R., and A. G. Douglas, 1980, A pyrolysis-gas chromatographic method for kerogen typing, in A. G. Douglas and J. R. Maxwell, eds., Advances in Organic Geochemistry 1979: Oxford, Pergamon Press, p. 579-584.

Larter, S. R., 1989, Chemical models of vitrinite reflectance evolution, Geologische Rundschau, v.78, p. 349-359.

Latorre, C., J. Quade, and W. C. McIntosh, 1997, The expansion of C4 grasses and global change in the late Miocene: Stable isotope evidence from the Americas, Earth and Planetary Science Letters, v. 146, no. 1-2, p. 83-96.

167

Law, B.E., 1984, Relationships of source-rock, thermal maturity, and overpressuring to gas generation and occurrence in low-permeability Upper Cretaceous and Lower Tertiary rocks, Greater Green River Basin, Wyoming, Colorado and Utah, in J. Woodward, F. F. Meissner, and J. L. Clayton, eds., Hydrocarbon source rocks of the greater Rocky Mountain Region: Rocky Mountain Association of Geologists Guidebook, p. 469–490.

Law, B.E., R. M. Pollastro, and C. W. Keighin, 1986, Geological Characterization of Low- Permeability Gas Reservoirs in Selected Wells, Greater Green River basin, Wyoming, Colorado, and Utah, AAPG Special Publication: Geology of Tight Gas Reservoirs, vol. 24, p.253-269

Law, B. E., and R. C. Johnson, 1989, Structural and stratigraphic framework of the Pinedale anticline, Wyoming, and the multiwell experiment site, Colorado, in B. E. Law and C. W. Spencer, eds., Geology of tight gas reservoirs in the Pinedale anticline area, Wyoming, and at the multiwell experiment site, Colorado: U. S. Geological Survey Bulletin, v. 1886, p. B1– B11.

Leininger, J. P., F. Lorant, C. Minot, and F. Behar, 2006, Mechanisms of 1-methylnaphthalene pyrolysis in a batch reactor, Energy & Fuel, v. 20, p. 2518-2530.

Lewan, M. D., J. C. Winters, and J. H. McDonald, 1979, Generation of oil-like pyrolysates from organic-rich shales, Science, v. 203, p. 896-899.

Lewan, M. D., 1985, Evaluation of petroleum generation by hydrous pyrolysis experimentation: Philosophical Transactions of the Royal Society of London, series A, v. 315, no. 1531, p. 123–134.

Lewan, M. D., 1987, Petrographic study of primary petroleum migration in the Woodford Shale and related rock units, in B. Doligez, ed., Migration of hydrocarbons in sedimentary basins: Paris, Editions Technip, p. 113-130.

Lewan, M. D., 1993, Laboratory simulation of petroleum formationhydrous pyrolysis, in M. Engel and S. Macko, eds., Organic geochemistry—principles and applications: New York, Plenum Press, p. 419–442.

168

Lewan, M. D., 1994, Assessing natural oil expulsion from source rocks by laboratory pyrolysis, in L. Magoon and W. Dow, eds., The petroleum system—from source to trap: AAPG Memoir 60, p. 52–70.

Lewan, M. D., 1997, Experiments on the role of water in petroleum formation, Geochimica et Cosmochimica Acta, v. 61, no. 17, pp. 3691-3723. Lewan, M. D. and A. A. Henry, 1999, Gas: oil ratios for source rocks containing type-I, -II, -IIS, and –III kerogen as determined by hydrous pyrolysis: USGS Open-File Report No. 99-327, 34p; also in USGS Bulletin 2173.

Lewan, M. D., and T. E. Ruble, 2002, Comparison of petroleum generation kinetics by isothermal hydrous and non-isothermal open-system pyrolysis, Organic Geochemistry, v. 33, p. 1457– 1475.

Lewan, M. D., 2010, personal communication.

Li, M., Y. Huang, M. Obermajer, C. Jiang, L. R. Snowdon, and M. G. Fowler, 2001, Hydrogen isotopic compositions of individual alkanes as a new approach to petroleum correlation: case studies from the Western Canada Sedimentary Basin, Organic Geochemistry, v. 32, p. 1387-1399.

Lide, D. R. and H. P. R. Frederikse, 1995, CRC Handbook of Chemistry and Physics, CRC Press, New York.

Lillis, P. G., M. D. Lewan, A. Warden, S. M. Monk, and J. D. King, 1999, Identification and characterization of oil types and their source rocks, Chapter OA, in The Oil and Gas Resource Potential of the 1002 Area, Arctic National Wildlife Refuge, Alaska, by ANWR Assessment Team, U.S. Geological Survey Open-File Report 98-34.

Lillack, H., W. Esser, and K. Schwochau, 1991, Evolution of hydrogen and methane by nonisothermal pyrolysis of petroleum source rocks, in Organic Geochemistry: Advances and applications in energy and the natural environment, Manchester, Manchester University Press, pp. 306-309.

169

Lis, G. P., A. Schimmelmann, and M. Mastalerz, 2005, FTIR absorption indices for thermal

maturity in comparison with vitrinite reflectance Ro in type-II kerogens from Devonian black shales, Organic Geochemistry, v. 36, p. 1533-1552.

Lis, G. P., A. Schimmelmann, and M. Mastalerz, 2006, D/H ratios and hydrogen exchangeability of type-II kerogens with increasing thermal maturity, Organic Geochemistry, v. 37, p. 342- 353.

Liu, Q., W. Liu, S. Qin, W. Wang, B. Gao, 2002, Gas generating characteristics of coal rock and its maceral in thermal simulation, Petroleum Geology and Experiment, v. 24, p. 147–151 (in Chinese, English abstract).

Liu, W.H., Y. Song, Q. Y. Liu, S. F. Qin, X. F. Wang, 2003, Evolution of carbon isotopic composition in pyrolytic gases generated from coal and its main macerals, Acta Sedimentologica Sinica, v. 21, p.183–190 (in Chinese, English abstract).

Longman, M. W., B. A. Luneau, and S. M. Landon, 1998, Nature and distribution of Niobrara lithologies in the Creataceous Western Interior Seaway of the Rocky Mountain region, The Mountain Geologist, v. 35, p. 137-170.

Lorant, F., A. Prinzhofer, F. Behar, and A. Y. Huc, 1998, Carbon isotopic and molecular constraints on the formation and the expulsion of thermogenic hydrocarbon gases, Chemical Geology, v. 147, p. 249 –264.

Makismov, S. P., 1982, New data on gas prospects of Lower Permian terrigenous deposits of Poland, International Geology Review, v. 24, p. 1064-1070.

Melander, L., and W. H. Saunders, 1980, Reactions Rates of Isotopic Molecules, Wiley, New York.

Meissner, F. F., 1984, Cretaceous and lower Tertiary coals as sources for gas accumulations in the Rocky Mountain area, in J. Woodward, F. F. Meissner, and J. L. Clayton, eds., Hydrocarbon source rocks of the greater Rocky Mountain region: Rocky Mountain Association of Geologists guidebook, p. 401–431.

170

Mnich, C. A., 2009 Gochemical signatures of stratigraphic sequences and sea-level change in the Woodford Shale, Permian Basin : M.S. Thesis, Colorado School of Mines, Golden, 88p.

Molenaar, C.M., and W.A. Cobban, 1991, Middle Cretaceous stratigraphy on the south and east sides of the Uinta Basin, northeastern Utah and northwestern Colorado, United States Geological Survey Bulletin 1787-P, 34 p.

Molenaar, C.M., and B.W. Wilson, 1990, The Frontier Formation and associated rocks of northeastern Utah and northwestern Colorado, U.S. Geological Survey Bulletin 1787-M, 21 p.

Montgomery, S. L., and J. W. Robinson, 1997, Jonah field, Sublette County, Wyoming: Gas production from overpressured Upper Cretaceous Lance sandstones of the Green River basin, AAPG Bulletin, v. 81, p. 1049– 1062.

Nechayeva, O. L., 1968, Hydrogen in gases dissolved in water of the western Siberian plain, Dokl.Akad. Nauk SSSR, v. 179 (4), P. 961-962.

Nelson, P. H., and S. L. Santus,2010, Gas, oil, and water production from Grand Valley, Parachute,Rulison, and Mamm Creek Fields in the Piceance Basin, Colorado, U.S. Geological Survey Open-File Report 2010-1110, 34 p.

Nixon, R.P., 1973, Oil source beds in Cretaceous Mowry Shale of northwestern interior United States, AAPG Bulletin, v. 57, p.136–161.

Nuccio, V.F., and L. N. R. Roberts, 2003, Thermal maturity and oil and gas generation history of petroleum systems in the Uinta-Piceance Province, Utah and Colorado, Chapter 4, in Petroleum systems and geologic assessment of oil and gas the Uinta-Piceance Province, Utah and Colorado, USGS Uinta-Piceance Assessment Team, compilers: U.S. Geological Survey Digital Data Series DDS-69-B.

Nuttall, B., C. F. Eble, J. A. Drahovzal, R. M. Bustin, 2005, Analysis of Devonian black shales in Kentucky for potential carbon dioxide sequestration and enhanced natural gas production, Final Report, 120 p.

171

Oberlin, A., J. L. Boulmier, and M. Villey, 1980, Electron microscopic study of kerogen microtexture: selected criteria for determining the evolution stage of kerogen, in B. Durrand, ed., Kerogen, Technip, Paris, pp. 191–241.

O'Boyle, C. C., 1955, The Cretaceaous rocks of northwest Colorado, in H. Ritzma, and S. Oriel, eds., Geology of Northwest Colorado Annual Field Conference, Rocky Mountain Association of Geologists, p. 32-35.

O’Neil, J. R., 1986, Theoretical and experimental aspects of isotope fractionation, in J.W. Valley, H.P. Taylor Jr. and J.R. O'Neil, eds., Stable Isotopes in High Temperature Geological Processes: Reviews in Mineralogy and Geochemistry, v. 16, pp. 1–40.

Orr, W. L., 1992, Rate and mechanism of non-microbial sulfate reduction: Thermochemical sulfate reduction. Memo, after oral presentation, at 95th Annual Geological Society of America Meeting 1982.

Orr, W. L., 1977, Geological and geochemical controls on the distribution of hydrogen sulfide in natural Gas, in R. Campos and J. Goni, eds., Advances in organic Geochemistry, 1975, Enadisma, Madrid: Empressa nacional adaro de investigaciones mineras, pp. 571-597.

Orr, W. L., 1986, Kergoen/asphaltene/sulfur relationships in sulfur-rich Monterey oils, Organic Geochemistry, v. 10, p.499-516.

Pawlewicz, M. J., 2010, personal communication.

Patterson, P. E., K. Kronmueller, and T. D. Davies, 2003, Sequence Stratigraphy of the Mesaverde Group and Ohio Creek Conglomerate, Northern Piceance Basin, Colorado, in K. M. Peterson, T. M. Olson, and D. S. Anderson, eds., Piceance Basin 2003 Guidebook, pp. 115-129.

Pepper, A. S., and P. J. Corvi, 1995, Simple kinetic models of petroleum formation: Part I: oil and gas generation from kerogen, Marine and Petroleum Geology, v. 12, p. 291-319.

172

Pepper, A. S., and T. A. Dodd, 1995, Simple kinetic models of petroleum formation: Part II: oil- gas cracking, Marine and Petroleum Geology, v. 12, p. 321-340.

Peters, K. E., 1986, Guidelines for evaluating petroleum source rock using programmed pyrolysis, AAPG Bulletin, v. 70, p. 318-329.

Peters, K. E, and M. R. Cassa, 1994, Applied source rock geochemistry: Chapter 5: Part II Essential Elements, AAPG Memoir 60: The Petroleum System- from Source to Trap, p.93- 120.

Peters, K. E, M. J. Clutson, and G. Robertson, 1999, Mixed marine and lacustrine input to an oil-cemented sandstone breccia from Brora, Scotland, Organic Geochemistry, v. 30, p. 237-248.

Peters, K.E. Peters, C.C. Walters and J.M. Moldowan, 2004, The Biomarker Guide, Volumes 1 & 2, Biomarkers and Isotopes in Petroleum Exploration and Earth History, The University Press, Cambridge, 1155 pp.

Petersen, H. I., 2002, A re-consideration of the “oil window” for humic coal and kerogen Type III source rocks, Journal of Petroleum Geology, v. 25, p. 407-432.

Petersen, H. I., 2006, The petroleum generation potential and effective oil window of humic coals related to coal composition and age, International Journal of Coal Geology, v. 67, p. 221- 248.

Potential Gas Agency, 2008.

Powell, T. G., D. M. McKirdy 1973, Relationship between ratio of pristane to phytane, crude oil composition and geological environment in Australia, Nature, v. 243, p. 37-39.

Pranter, M. J., A. I. Ellison, R. D. Cole, and P. E. Patterson, 2007, Analysis and modeling of intermediate-scale reservoir heterogeneity based on a fluvial point-bar outcrop analog, Williams Fork Formation, Piceance Basin, Colorado, AAPG Bulletin, v. 91, p. 1025-1051.

173

Pratt, L. M., M. A. Arthur, W. E. Dean, P. A. Scholle, 1993, Paleo-oceanographic cycles and events during the Late Cretaceous in the Western Interior Seaway of North America, in Caldwell, W.G.E., Kauffman, E. G., eds., Evolution of the Western Interior Basin, Geological Association of Canada Special Paper 39, pp. 333–353.

Prinzhofer, A. and A. Huc, 1995, Genetic and post-genetic molecular and isotopic fractionation in natural gas, Chemical Geology, v. 126, p. 281–290.

Prinzhofer, A. and E. Pernaton, 1997, Isotopically light methane in natural gas: bacterial imprint or diffusive fractionation, Chemical Geology, v. 142, p. 193–200.

Prinzhofer, A., M. R. Mello, L. C. da Silva Freitas, and T. Takaki, 2000a, New geochemical characterization of natural gas and its use in oil and gas evaluation, in M. R. Mello and B. J. Katz, eds., Petroleum systems of South Atlantic margins: AAPG Memoir 73, p. 107–119.

Prinzhofer, A., M. R. Mello, and T. Takaki, 2000b, Geochemical characterization of natural gas: a physical multivariable approach and its applications in maturity and migration estimates, AAPG Bulletin, v. 84, p. 1152-1172.

Prinzhofer, A. and A. Battani, 2003, Gas Isotopes Tracing: an Important Tool for Hydrocarbons Exploration, Oil & Gas Science and Technology – Rev. IFP, v. 58, no. 2, p. 299-311.

Qin, K., Yang, Q., Gua, S., Lu, Q., Shu, W., 1994, Chemical structure and hydrocarbon formation of the Huanxian brown coal, China, Organic Geochemistry, v. 21, p. 333–341.

Reddings, C. E., Schoell M., Monin J. C., and Durand B., 1980, Hydrogen and carbon isotopic composition of coals and kerogens, in A. G. Douglas and J. R. Maxwell eds., Advances in Organic Geochemistry, 1979, Pergamon Press, pp. 711–723.

Reservoir Evaluation Report, 2005. C&C Resevoirs, the analog company.

Rice, D. D., 1993, Composition and origins of coalbed gas, in B. E. Law and D. D. Rice, eds., Hydrocarbons from coal, AAPG Studies in Geology, v. 38, p. 159–184.

174

Richet, P., Y. Bottinga, and M. Javoy, 1977, A review of H, C, N, S, and Cl stable isotope fractionation among gaseous molecules, Annual Review of EPSL, v. 5, p. 65-110.

Roberts, L. N. R., and M.A. Kirschbaum, 1995, Paleogeography of the late Cretaceous of the western interior of middle North America-Coal distribution and sediment accumulation: USGS Professional Paper 1561, 115 p.

Roberts, L. N.R., M. D. Lewan, and T. M. Finn, 2005, Burial history, thermal maturity, and oil and gas generation history of petroleum systems in the southwestern Wyoming province, Wyoming, Colorado, and Utah, Chapter 3 of Petroleum Systems and Geological Assessment of oil and gas in The southwestern Wyoming Province, Wyoming, Colorado, and Utah, U.S. Geological Survey Digital Data Series DDS-69-D.

Robinson, J. W. and K. W. Shanley, 2004, Jonah field: T28-29N, R107-109W, Sublette County, Wyoming, in J. W. Robinson and K. W. Shanley, eds., Jonah field: Case study of a tight- gas fluvial reservoir: AAPG Studies in Geology 52, p. 159–182.

Rogers, R. E., Coalbed methane: principles and practice, Pearson Education, Inc., 345 p.

Rooney, M. A., G. E. Claypool, and H. M. Chung, 1995, Modeling thermogenic gas generation using carbon isotope ratios of natural gas hydrocarbons, Chemical Geology, v. 126, p. 219– 232.

“RPSEA” website accessed September 13th, 2010, http://www.rpsea.org/

Ruble, T. E., M. D. Lewan, and R. P. Philp, 2001, New insights on the Green River petroleum system in the Uinta basin from hydrous pyrolysis experiments, AAPG Bulletin, v. 85, no. 8, p. 1333-1371.

Rubey, W. W., 1928, Origin of the siliceous Mowry Shale of the Black Hills region, USGS Professional Paper 154, p. 153-170.

Seewald, J. S., 2003, Organic–inorganic interactions in petroleum-producing sedimentary basins, Nature, v. 426, p.327-333.

175

Scalan, E. S. and J. E. Smith, 1970, An improved measure of the odd-even predominance in the normal alkanes of sediment extracts and petroleum, Geochimica et Cosmochimica Acta, v. 34, p. 611-620.

Schenk, H. J., R. Di Primio, and B. Horsfield, 1997, The conversion of oil into gas in petroleum reservoirs. Part I: Comparative kinetic investigation of gas generation from crude oils of lacustrine, marine and fluviodeltaic origin by programmed temperature closed-system pyrolysis, Organic Geochemistry, v. 26, p. 467– 481.

Schimmelmann, A., M. D. Lewan, and R. P. Wintsch, 1999, D/H isotope ratios of kerogen, bitumen, oil, and water in hydrous pyrolysis of source rocks containing kerogen types I, II, IIS, and III, Geochimica et Cosmochimica Acta, v. 63, p. 3751-3766.

Schimmelmann, A., J. P. Boudou, M. D. Lewan, and R. P. Wintsch 2001, Experimental controls on D/H and 13C/12C ratios of kerogen, bitumen and oil during hydrous pyrolysis, Organic Geochemistry, v. 32, p. 1009-1018.

Schimmelmann, A., A. L. Sessions, C. J. Boreham, D. S. Edwards, G. A. Logan, and R. E. Summons, 2004, D/H ratios in terrestrially-derived petroleum systems, Organic Geochemistry, v. 35, p. 1169-1195.

Schoell, M., 1983, Genetic characterization of natural gases, AAPG Bulletin, v. 67, p. 2225– 2238.

Schoell, M., 1988, Multiple origins of methane in the Earth, Chemical Geology, v. 71, p. 1–10.

Schrayer, G.J., and W.M. Zarrella, 1963, Organic geochemistry of shales – I, Distribution of organic matter in the siliceous Mowry Shale of Wyoming, Geochimica et Cosmochimica Acta, v. 27, p.1033-1046.

Schrayer, G.J., and W.M. Zarrella, 1966, Organic geochemistry of shales – II, Distribution of extractable organic matter in the siliceous Mowry Shale of Wyoming, Geochimica et Cosmochimica Acta, v. 30, p. 415-434.

176

Schrayer, G.J., and W.M. Zarrella, 1968, Organic carbon in the Mowry Formation and its relation to the occurrence of petroleum in Lower Cretaceous reservoir rocks; in G.R. Wulf, ed., Black Hills area South Dakota, Montana, Wyoming, 20th Field Conference Guidebook Laramie, Wyoming Geological Association, p.35-39.

Schwarzkopf, T. A., 1992, Source rock potential (TOC + hydrogen index) evaluation by integrating well log and geochemical data, Organic Geochemistry, v. 19, p. 545-555.

Senftle, J. T., S. R. Larter, B. W. Bromely and J. H. Brown, 1986, Quantitative chemical characterization of vitrinite concentrates using pyrolysis-gas chromatography: Rank variation of pyrolysis products, Organic Geochemistry, v. 9, p. 345–350.

Sessions, A. L., S. P. Sylva, R. E. Summons, and J. M. Hayes, 2004, Isotopic exchange of carbon-bound hydrogen over geologic timescales, Geochimica et Cosmochimica Acta, v. 68, no. 7, p. 1545–1559.

Shanley, K. W., 2004, Fluvial reservoir description for a giant, low permeability gas field: Jonah field, Green River basin, Wyoming, U.S.A., in J. W. Robinson and K. W. Shanley, eds., Jonah field: Case study of a tight-gas fluvial reservoir: AAPG Studies in Geology 52, p. 159–182.

Smith, J. W., D. Rigby, K. W. Gould, and G. Hart G., 1985, An isotopic study of hydrocarbon generation processes, Organic Geochemistry, v. 8, p. 341-347.

Snowdon, L. R., 2001, Natural gas composition in a geological environment and the implications for the processes of generation and preservation, Organic Geochemistry, v. 32, p. 913-931.

Spencer, C. W., 1987, Hydrocarbon generation as a mechanism for overpressuring in Rocky Mountain region, AAPG Bulletin, v. 71, p. 368-388.

Stach, E., M. Th. Mackowsky, M. Teichmüller, G.H. Taylor, D. Chandra, and R. Teichmüller, 1982, Stachs Textbook on Coal Petrology (3rd edn.), Gebrüder Borntraeger, Berlin.

177

Stahl, W. J., 1977, Carbon and nitrogen isotopes in hydrocarbon research and exploration, Chemical Geology, v. 20, p. 121–149.

Stahl, W. J., 1979, Carbon isotopes in petroleum geochemistry, in E. Jager and J. C. Hunziker, eds., Lectures in isotope geology, Springer-Verlag, p. 274-282.

Suleimenov, O. M., and R.E. Krupp, Solubility of hydrogen sulfide in pure water and in NaCl solutions, from 20 to 320°C and st saturation pressures, Geochimica et Cosmochimica Acta, v. 58, p. 2433-2444.

Surdam, R., Z. Jiao, R. Bentley, R. DeBruin, 2007, Shale Gas Potential of the Mowry Shale in Wyoming Laramide Basins, GSA Annual Meeting, Denver, Abstract.

Tang, Y., P. D. Jenden, A. Nigrini, and S. C. Teerman, 1996, Modeling early methane generation in coal, Energy & Fuels, v. 10, p. 659–671.

Tang, Y., J. K. Perry, P. D. Jenden, and M. Schoell, 2000, Mathematical modeling of stable carbon isotope ratios in natural gases, Geochimica et Cosmochimica Acta, v. 64, p. 2673– 2687.

Tang Y. and M. Schoell, 2005, A new genetic scheme for natural gas formation and isotopic evidence for oil cracking, Geochimica et Cosmochimica Acta, v. 69, no. 10, p. 501-550.

Tang, Y., Y. Huang, G. F. Ellis, Y. Wang, P. G. Kralert, B. Gillaizeau, Q. Ma, and R. Hwang, 2005, A kinetic model for thermally induced carbon and hydrogen fractionation of individual n-alkanes in crude oil, Geochimica et Cosmochimica Acta. v. 69, p. 4505-4520.

Teerman, S. C. and R. J. Huang, 1991, Evaluation of the liquid hydrocarbon potential of coal by artificial maturation techniques, Organic Geochemistry, v. 17, p. 749-764.

Tian H., X. Xiao, R. W. T. Wilkins, and Y. Tang, 2008, New insights into the volume and pressure changes during the thermal cracking of oil to gas in reservoirs: Implications for the in-situ accumulation of gas cracked from oils, AAPG Bulletin, v. 92, no. 2, p. 181-200.

178

Tian H., X. Xiao, R. W. T. Wilkins, H. Gan, L. Guo, L. Yang, 2010, Genetic origins of marine gases in the Tazhong area of the Tarim basin, NW China: Implications from the pyrolysis of marine kerogens and crude oil, International Journal of Coal Geology, v. 82, p. 17-26.

TICORA, 2009.

Tissot, B., B. Y. Califet-Debyser, G. Deroo, and J. L. Oudin, 1971, Origin and evolution of hydrocarbons in early Toarcian shales, Paris basin, France, AAPG Bulletin, v. 55, p. 2177-2193.

Tissot, B., B. Durand, J. Espitalie, A. Combaz, 1974, Influence of nature and diagenesis of organic matter in formation of petroleum, AAPG Bulletin, v. 58, p.499-506.

Tissot, B., and D. H. Welte, 1978, Petroleum formation and occurrence: a new approach to oil and gas exploration, Springer-Verlag, New York, 521 p.

Tissot, B., and D. H. Welte, 1984, Petroleum occurrence and formation, – Approach to oil and gas exploration, 2nd Edition: Springer-Verlag, Berlin-Heidleberg-New York, 699 p.

Tsuzuki, N., N. Takeda, M. Suzuki, and K. Yokoi, 1999, The kinetic modeling of oil cracking by hydrothermal pyrolysis experiments, International Journal of Coal Geology, v. 39, p. 227– 250.

Tyler, R., and R. G. McMurry, 1995, Genetic stratigraphy, coal occurrence, and regional cross section of the Williams Fork Formation, Mesaverde Group, Piceance Basin, northwestern Colorado: Colorado Geological Survey Open File Report 95-2, 42 p.

Uinta-Piceance Assessment Team, 2003, Petroleum systems and geologic assessment of oil and gas in the Uinta-Piceance Province, Utah and Colorado: United States Geological Survey Data Digital Series, DDS-69-B, version 1. ch. 1-28.

Urey, H. C., 1947, The thermodynamic properties of isotopic substances, Journal of the Chemical Society, p. 562-581.

179

Vandenbroucke, M., 2003, Kerogen: from types to models of chemical structure, Oil & Gas Science and Technology, v. 58, no. 2, p. 243-269. van Krevelen D. W., 1961, Coal: typology, chemistry, physics, constitution, Elsevier, Amsterdam, 2nd print 1981. van Krevelen, D. W., 1984, Organic geochemistry – old and new, Organic Geochemistry, v. 6, p. 1-10.

Vincelette, R. R., and N. H. Foster, 1992, Fractured Niobrara of Northwestern Colorado, in J. W. Schmoker, E. B. Coalson, and C. A. Brown, eds., Geological studies relevant to horizontal drilling: examples from Western North America Rocky Mountain Association of Geologists, p. 227-241.

Waples, D. W., 1984, Thermal models for oil generation, in J. Brooks and D. Welte, eds., Advances in petroleum geochemistry, v. 1: London, Academic Press, p. 7-67.

Waples, D. W., 2000, The kinetics of in-reservoir oil destruction and gas formation: Constraints from experimental and empirical data, and from thermodynamics, Organic Geochemistry, v. 31, p. 553– 575.

Warner, E. M., 1998, Structural geology and pressure compartmentalization of Jonah field, Sublette County, Wyoming, in R. M. Slatt, ed., Compartmentalized reservoirs in Rocky Mountain basins: Rocky Mountain Association of Geologists, p. 29–46.

Warner, E. M., 2000, Structural geology and pressure compartmentalization of Jonah field on 3-D Seismic Data and Subsurface Geology, Sublette County, Wyoming, The Mountain Geologists, v. 37, p. 15–30.

Wei, A., J. M. Moldowan, S. Zhang, R. Hill, D. M. Jarvie, H. Wang, F. Song, and F. Fago, 2007, Diamondoid hydrocarbons as a molecular proxy for thermal maturity and oil cracking: Geochemical models from hydrous pyrolysis, Organic Geochemistry, v. 38, p. 227-249.

Weimer, R.J., 1959, Upper Cretaceous stratigraphy, Colorado, in Rocky Mountain Association of Geologists symposium on Cretaceous rocks of Colorado and adjacent areas, p. 9-16.

180

Weimer, R.J., 1960, Upper Cretaceous stratigraphy, Rocky Mountain area, AAPG Bulletin, v. 44, no. 1, p. 1-20.

Weiss, R. F., 1974, Carbon dioxide in water and seawater: the solubility of a non-ideal gas, Marine Chemistry, v. 2, p. 203-215.

Willette, D. C., 2010, Geologic factors responsible for the generation of natural gas and pyrobitumen through the pyrolysis of natural oils [Ph.D. Dissertation]: Colorado School of Mines, Golden, Colorado.

Whiticar, M. J., 1994, Correlation of natural gases with their sources, in L. B. Magoon and W. G. Dow, eds., The petroleum system—from source to trap, AAPG Memoir 50, p. 261–283.

Wyman, R. E., 1984, Gas resources in Elmworth coal seams, in J. A. Master, ed., Elmworth— Case study of a deep basin gas field, AAPG Memoir 38, p. 173–188.

Yurewicz, D. A., K. M. Bohacs, J. D. Yeakel, and K. Kronmueller, 2003, Source Rock Analysis and Hydrocarbon Generation, Mesaverde Group and Mancos Shale, Northern Piceance Basin, Colorado, Rocky Mountain Association of Geologist, Chapter 8, p.130-153.

Yurewicz, D. A., K. M. Bohacs, J. Kendall, R. E. Klimentidis, K. Kronmueller, M. E. Meurer, T. C. Ryan, and J. D. Yeakel, 2008, Controls on gas and water distribution, Mesaverde basin- centered gas play, Piceance Basin, Colorado, in S. P. Cumella, K. W. Shanley, and W. K. Camp, eds., Understanding, exploring, and developing tight-gas sands— 2005 Vail Hedberg Conference: AAPG Hedberg Series, no. 3, p. 105–136.

Zinger, A. S., 1962, Molecular hydrogen in gas dissolved in waters of oil-gas fields, lower Volga region, Geokhim, v. 10, p. 890-898.

Zhang, E., R. J. Hill, B. J. Katz, and Y. Tang, 2008, Modeling of gas generation from the Cameo coal zone in the Piceance Basin, Colorado, AAPG Bulletin, v. 92, p. 1077-1106.

Zou, Y., Y. Cai, C. Zhang, X. Zhang, and P. Peng, 2007, Variations of natural gas carbon isotope- type curves and their interpretation – A case study, Organic Geochemistry, v. 38, p. 1398- 1415.

181

Zumberge, J. E., K. A. Ferworn, J. B. Curtis, 2009, Gas character anomalies found in highly productive shale gas wells, GoldSchmidt Conference Abstracts 2009, A1539, Davos, Switzerland.

182