Variability in results from mineralogical and organic geochemical interlaboratory testing of U. S. Geological Survey shale reference materials

Abstract

The expansion of unconventional petroleum resource exploration and production in the United States has led to an increase in source rock characterization efforts, particularly related to bulk organic and mineralogical properties. To support the analytical and research needs of industry and academia, as well as internal work, the U.S. Geological Survey (USGS) has collected and prepared shale geochemical reference materials (GRMs) from several major shale petroleum systems in the U.S. The sources of these materials are the Late Cretaceous Boquillas (lower Eagle Ford-equivalent) Formation (roadcut near Del Rio, TX), Late Cretaceous Mancos Shale (outcrop near Delta, CO), Devonian–Mississippian Woodford Shale (outcrop near Ardmore, OK), Late Cretaceous Niobrara Formation (quarry near Lyons, CO), Middle Devonian Marcellus Shale (creek bed in LeRoy, NY), and Eocene Mahogany zone oil shale of the Green River Formation (oil shale mine near Rifle, CO). Of particular interest in the development of these GRMs has been the examination of variability between laboratories and specific methods or instruments in commonly made measurements, including major- and trace-element concentrations, X-ray diffraction (XRD) mineralogy, total organic carbon (TOC) content, and programmed pyrolysis (PP) parameters. For the component concentrations and parameters we measured, the techniques and instrument types included: (1) elemental analysis by X-ray fluorescence, inductively coupled plasma mass spectrometry, and instrumental neutron activation analysis; (2) XRD mineralogy with various preparatory methods (spray drying or micronizing with or without internal standard); (3) TOC by combustion with infrared detection after carbonate removal or the PP approach; (4) PP by Rock-Eval 2 or more recently developed instruments (Rock-Eval 6, Source Rock Analyzer or SRA, and Hydrocarbon Analyzer With Kinetics or HAWK). Overall, the results showed that the selected shales cover a wide range of source rock organic and mineralogical properties. Major- and trace-element chemistry results showed low heterogeneity consistent with other USGS GRMs. Comparison of TOC results showed coefficients of variation (COV) of around 5% and the most consistent organic geochemical results between different laboratories and methods. Arguably the most relevant PP measurement, S2 or kerogen hydrocarbon-generating potential (mg-HC/g-rock), showed a somewhat wider range of variability than TOC (COV ~10%), but was consistent between the three modern instruments and the industry-standard Rock-Eval 2. Major phase mineralogy (mineral concentrations ≥10 wt. %, organic-free basis) were comparable between laboratories, but variability in minor phase identification and quantification was observed. Utilization of these shale GRMs as quality control samples and testing materials is expected to help support analytical and experimental efforts in the continued development of unconventional petroleum resources.