Application of Raman spectroscopy as thermal maturity probe in shale

Oct 4, 2018 - Raman spectroscopy was studied as a thermal maturity probe in a series ... modeling of signal and baseline functions to decrease subject...
0 downloads 0 Views 4MB Size
Subscriber access provided by University of Sunderland

Fossil Fuels

Application of Raman spectroscopy as thermal maturity probe in shale petroleum systems: insights from natural and artificial maturation series Paul C. Hackley, and N. Keno Lünsdorf Energy Fuels, Just Accepted Manuscript • DOI: 10.1021/acs.energyfuels.8b02171 • Publication Date (Web): 04 Oct 2018 Downloaded from http://pubs.acs.org on October 5, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Application of Raman spectroscopy as thermal maturity probe in shale petroleum systems: insights from natural and artificial maturation series Paul C. Hackley, U.S. Geological Survey, MS 956 National Center, Reston VA 20192, USA, [email protected] N. Keno Lünsdorf, Department of Sedimentology and Environmental Geology, GeorgAugust-Universität Göttingen, Göttingen, D-37077, Germany, [email protected] Abstract Raman spectroscopy was studied as a thermal maturity probe in a series of Upper Devonian Ohio Shale samples from the Appalachian Basin spanning from immature to dry gas conditions. Raman spectroscopy also was applied to samples spanning a similar thermal range created from 72-hour hydrous pyrolysis (HP) experiments of the Ohio Shale at temperatures from 300 to 360°C and isothermal HP experiments lasting up to 100 days of similar Devonian-Mississippian New Albany Shale. Raman spectra were treated by an automated evaluation software based on iterative and simultaneous modeling of signal and baseline functions to decrease subjectivity. Spectra show robust correlation to measured solid bitumen reflectance (BRo) values and were therefore used to construct logarithmic regression relationships for calculation of BRo equivalent values. Raman spectra show considerable differences between natural samples and HP residues with similar measured BRo values, indicating as-yet undetermined differences in carbon chemistry. We speculate this result may be due to differences in the sampling interactions of Raman vs. reflectance measurements, and the incomplete nature of maturation reactions in the time-limited hydrous pyrolysis residues. Samples used in this study are similar in organic assemblage (dominantly solid bitumen) to other commonly exploited North American shale petroleum systems, i.e., Bakken, Barnett, Duvernay, Fayetteville and Woodford shales. Therefore, results presented herein may be broadly applicable to other important shale plays. However, caution is suggested and Raman spectroscopy as a thermal probe may need individual calibration in each shale play due to differences in solid bitumen carbon chemistry. Key Words Raman spectroscopy; thermal maturity; solid bitumen reflectance; hydrous pyrolysis; New Albany Shale; Ohio Shale Introduction Production of hydrocarbons from shale petroleum systems in North America has increased dramatically in the last decade—the ‘shale revolution’—driven primarily by application of horizontal drilling and hydraulic fracturing technologies 1. Increased

ACS Paragon Plus Environment

1

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 35

hydrocarbon exploration in shale has caused need for fast and accurate determinations of thermal maturity and prediction of the locations of oil, wet gas and condensate, and dry gas production areas 2. Shale thermal maturity often is determined by reflectance measurements of organic matter, which in thermally mature shales is dominated by solid bitumen 3-6. Solid bitumen is the organic matter that occurs as a product of kerogen conversion to petroleum and from alteration of once-liquid oil 7-9, and it is identified through organic petrographic inspection by recognition of fracture-filling, groundmass (enveloping), and void-filling textures 10. Micro-Raman spectroscopy of organic matter, including solid bitumen, has been widely advanced as a thermal maturity probe by workers in petroleum systems 11, 12 and in low temperature metamorphic terranes 13-16. Typically the intensity and width of the first order Raman spectral parameters, consisting of D- (disordered or defect carbon) and G(graphitic carbon) bands 17 are measured and related to other thermal proxies such as organic reflectance or temperature derived from inorganic phase equilibria. In general, the Raman technique has proven most effective in overmature organic matter 18 due to fluorescence interference at lower thermal maturity 19. The issue of fluorescence interference has so far limited widespread application of Raman spectroscopy in the oil window relative to other widely used petroleum system thermal probes such as vitrinite reflectance measurement (VRo) and Tmax from programmed pyrolysis 20. Lack of a standardized Raman spectroscopic approach also has limited its application, with many combinations of instrument set-up (laser wavelength, laser power, spectral resolution, spatial resolution), data collection (number of points measured), and data reduction (baseline subtraction, peak-fitting) currently in use 21, 22. These drawbacks are due in part to relative novelty of Raman as a thermal proxy, with details for standardization and fluorescence suppression still to be worked out by future researchers. Despite the problems with Raman measurement of organic matter in petroleum systems, many works have provided VRo equivalent values from Raman spectroscopy, calculated from the regression analysis of measured Raman and reflectance parameters 11, 23-30. Studies of coal samples are most common 31, due in part to greater carbon abundance and larger organic particle size, and the higher reproducibility of VRo measurements in the humic coal systems. Raman studies of thermal maturity in dispersed organic matter are fewer, although more work has developed in recent years based on the data from coal systems. For example, Wilkins et al. 27 described the ‘Raman Maturity Method’ (RaMM), based on Raman spectral measurements of humic macerals calibrated to VRo values from a suite of 10 reference coals, followed by application of RaMM to Mesozoic marine shales of Australia 29 and a wide range of North American shales and mudrocks 28. This work suggested RaMM-calculated vitrinite reflectance generally corresponded to measured VRo values, particularly for samples with VRo >1.2%. In a work similar to ours, Zhou et al. 30 provided regression

ACS Paragon Plus Environment

2

Page 3 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

relationships specifically for solid bitumen reflectance, using results from polished samples of an artificially matured solid bitumen vein sample. Zhou et al. 30 concluded RBS and band intensity ratio (ID/IG) provided the best relation to solid bitumen reflectance, especially for moderate to high maturity solid bitumen. Sauerer et al. 25 documented a robust relationship (r2 = 0.95) between measured VRo and Raman band separation (RBS; the wavenumber difference in cm-1 between the first order Raman Dand G-bands) for shale and mudrock samples, but did not, however, provide an equation for the regression relationship. Their study used unprocessed rock chip samples with >4 wt.% total organic carbon (TOC) to illustrate a fast and efficient method for Raman thermal maturity determination. However, this method may not work as well for organic-lean samples or spectra may be influenced by presence of macromolecular abiotic reduced organic carbon intimately associated with mineral phases 32. For example, we have observed that first order Raman D- and G-bands of varying signal intensity above background can be observed from almost every acquisition of unprocessed rock chip shale samples. We presume high Raman signal intensity is from fortuitous analysis of shale organic matter whereas low signal intensity is from abiotic reduced organic carbon occurring in mineral association. In this contribution, we analyzed Raman spectral properties of natural and artificially matured shale sample sets spanning from immature to gas window thermal conditions. Oil-prone amorphous kerogen mixed with indigenous solid bitumen constitutes the bulk of the organic assemblage in these samples at immature conditions, whereas solid bitumen dominates the organic matter in thermally mature samples. The objectives of this work were to: 1. conduct Raman thermal analysis of well-documented sedimentary organic matter in the context of a commonly exploited shale petroleum system, and 2. to contrast the Raman spectral results from natural and artificially matured sample series. Methods Samples Samples included in this study are hydrous pyrolysis (HP) series using starting materials from the New Albany Shale and the Ohio Shale, lower Huron Member. These sample series were previously described and studied for several other works 6, 10, 33. The New Albany sample was collected from outcrop in Clarke County, Indiana, and the lower Huron from a road cut in Powell County, Kentucky (Table 1). We also used a natural maturation series comprised of immature to dry gas maturity samples from the lower Huron, collected from well core and cuttings across the northern Appalachian Basin (Table 1). Substituted into the lower Huron natural maturation series was one sample of the Cashaqua Shale Member of the Sonyea Formation, which has similar organic matter content (dominantly solid bitumen) to the lower Huron at its thermal maturity (boundary of wet gas/dry gas windows). Some individual samples from the

ACS Paragon Plus Environment

3

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 35

lower Huron natural maturation series also have been previously studied 33, 34. The naturally matured lower Huron samples were selected from a transect arranged roughly perpendicular to thermal maturity isolines as described from previous studies of thermal maturity in the Devonian of the northern Appalachian Basin 34-39. Samples used in this study are similar in organic composition to other commonly exploited North American shale petroleum systems of similar age, e.g., Bakken, Barnett, Duvernay, Fayetteville, Woodford, etc.4; therefore, Raman spectral results from this study should be broadly applicable to these similar shales in which the organic assemblage is dominated by solid bitumen. Hydrous Pyrolysis and Programmed Pyrolysis Hydrous pyrolysis (HP) followed the method of Lewan 40, 41 using crushed rock samples (2-4 g, 1-3 mm top size) loaded into stainless steel (SS-316) SwageLokTM mini-reactor vessels (25-35 mL internal volume) and covered with de-ionized water. A complete description of the HP method with SwageLokTM mini-reactors was recently provided by Hackley and Lewan 10. Experiments at temperatures of 300 to 360°C for 72 hours were applied to the lower Huron sample whereas the New Albany Shale sample was used for isothermal HP experiments at 320°C for periods of 24 to 2,400 hours. These different treatments were selected to evaluate how organic matter responded to thermal stress by varying temperature and time conditions, respectively. Bulk geochemical analyses were determined on the original samples and HP residues after crushing with mortar and pestle. Bulk analyses included total organic carbon (TOC) content by LECO and Rock-Eval II and Hawk programmed pyrolysis (analyzed at Weatherford and USGS in Denver) per the established methods 42. Sample 9781 Pittston 5320-5340' was analyzed multiple times via both Hawk and Rock-Eval II programmed pyrolysis technologies; however, reliable values for Tmax could not be obtained, even with solvent extracted aliquots, due to high maturity and lower organic matter abundance. Petrographic Preparation and Solid Bitumen Reflectance Samples were prepared for petrographic analyses according to ASTM D2797 43 wherein the rock particles were mounted in a thermoset plastic briquette, ground flat and then polished with successively finer abrasives until a 0.05-micrometer (µm) finishing stage. Solid bitumen reflectance analyses (BRo, %) followed ASTM D7708 44. In this technique, incident white light is reflected from solid bitumen positioned under the microscope crosshairs at 500x magnification, measured at a detector and compared to measured light reflected from a calibration standard. At least 20 measurements of solid bitumen reflectance were collected for each sample, with only 1 measurement of solid bitumen from each individual rock fragment. Solid bitumen reflectance was used instead of

ACS Paragon Plus Environment

4

Page 5 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

conventional vitrinite reflectance because of the relative scarcity and difficulties in identifying vitrinite in these shales 6. Sample briquettes were imaged under oil immersion on a Zeiss AxioImager microscope in white and blue incident light. A Leica DM4000 microscope equipped with LED illumination and monochrome camera detection was used for solid bitumen reflectance analysis with the computer program DISKUS-FOSSIL by Hilgers Technisches Buero. A YAG calibration standard (0.908% Ro) from Klein and Becker was used. Some older measurements 34 (Table 2) employed a J&M PMT system with a larger measuring aperture (~25 µm2) whereas more recent measurements with the DISKUS-FOSSIL system used a variable measuring aperture with the ability to adjust the measured area to