Mapping Shale Composition: A Breakthrough in Quantitative Mineral Analysis Using Mixed Phase X-Ray Spectra
Automated energy-dispersive X-ray spectroscopy in a scanning electron microscope (automated mineralogy) has been employed for decades to map the mineralogy of sediments and rocks at the micron scale. Current spectral analysis engines (SAE) automatically identify minerals from ED spectra, classify the dominant phase per acquisition point based on the X-ray spectra and the backscatter electron (BSE) intensity. The two existing approaches in automated mineralogical analysis—a) best spectral match classification, and; b) pre-defined compositional ranges of elemental matches with measured spectra—are both limited to report a single phase or a small number of pre-defined ‘boundary (mixed) phases’. Recent trials on mudstone samples have highlighted the limitations of these conventional EDS-based approaches for mineralogical identification in unconventional reservoir rocks. These techniques are limited because of the interaction of the e-beam with the sample, which occurs in a volume larger than the grain size in a typical shale rock. Therefore, the fine-grain size and mineralogic heterogeneity of mudstones commonly results in mixed X-ray spectra—complicating mineral identification. Decreasing the accelerating e-beam voltage reduces the interaction volume. However a minimum of 15 keV is required to excite characteristic energy lines of heavier elements such as Fe. In order to apply automated mineralogy to shale, a new approach to mixed spectra classification is required. Here we demonstrate the first results of a new SAE that can accurately quantify mineral and organic components within e-beam interaction volumes. Mineral reference spectra have been obtained from international mineral standards and purified extracted clays from shale rocks. X-ray spectra collected at each point are deconvolved using up to three reference spectra. The result is a mineral map that contains μm-cm scale textural information as well as, for the first time, spatially relevant data on the distribution of sub-micron mineralogy in mudstones. The calculated bulk composition results of 17 measurements on the compositionally diverse Argentinian Vaca Muerta Formation closely match independent XRF data. Spatial and modal mineralogical data can provide valuable contextual information for interpreting the geological/burial history, as well as can be utilized for the calculation of geomechanical properties, such as brittleness, which can inform conventional and unconventional exploration efforts.
AAPG Datapages/Search and Discovery Article #90189 © 2014 AAPG Annual Convention and Exhibition, Houston, Texas, USA, April 6–9, 2014