<p> Compositional analysis of reservoir rock is a vital aspect of oil exploration and production activities. In a broad sense, knowing the mineral composition of a reservoir can help with characterization and interpretation of depositional environments. On a smaller scale, identifying mineralogy helps calibrate well logs, identify formations, design drilling and completion programs, and screen for intervals with potential problem minerals, such as swelling clays. The petroleum industry utilizes two main methods to find compositional mineralogy, x-ray diffraction (XRD) and thin section analysis. Both methods are time consuming, expensive, and destructive. An alternative method for compositional analysis that includes quantitative mineralogy is a valuable prospect, especially if it had the potential to characterize the total organic content (TOC). </p><p> The remote sensing community has been using infrared spectroscopy to analyze mineralogy for years. Within the last ten years, the advancement of infrared spectrometers and processing programs have allowed infrared spectra to be taken and analyzed faster and easier than before. The objective of this study is to apply techniques used in remote sensing for quantitatively finding mineralogy to the petroleum industry. While developing a new methodology to compositionally analyze reservoir rock, a database of infrared spectra of relevant minerals has been compiled. This database was used to unmix spectra using a constrained linear least-squares algorithm that is used in the remote sensing community. A core has been scanned using a hand-held infrared spectrometer. Results of the best method show RMS error from mineral abundance to be under five percent.</p>

<p> <b>Petrophysical evaluation of lithology and mineral distribution with an emphasis on feldspars and clays, middle and upper Williams Fork Formations, Piceance Basin, Colorado.</b> Understanding accessory mineralogy occurrence and distribution is critical to evaluating the reservoir quality and economic success of tight&ndash;gas reservoirs, since the occurrence of iron&ndash;rich chlorites can decrease resistivity measurements and the occurrence of potassium feldspar increases gamma&ndash;ray measurements, resulting in inaccurate water saturation and net&ndash;to&ndash;gross calculations, respectively. This study was undertaken to understand the occurrence and distribution of chlorite and potassium feldspar in the middle and upper Williams Fork Formations of the Piceance Basin at Grand Valley Field. </p><p> Eight lithofacies are identified in core based on grain&ndash;size, internal geometry, and sedimentary structures. Four architectural elements (channel fill, crevasse splay, floodplain, and coal) were determined from lithofacies relationships, and then associated with well&ndash;log responses. Logs and models were used to determine the occurrence and distribution of lithology, architectural elements, chlorite and potassium feldspar, as well as the relationships between minerals and lithology and architectural elements. Net&ndash;to&ndash;gross ratios vary stratigraphically, from 8% to 88%, with a higher average in the middle Williams Fork Formation (58.3%) than in the upper Williams Fork Formation (48.5%). Volumetric proportions vary stratigraphically for both channel fills (18&ndash; 75%) and crevasse splays (1&ndash;7%). </p><p> The average volume percent of chlorite and potassium feldspars are both &lt;1%, with P ₅₀ values of 1.3% and 7%, respectively. Chlorite is pervasive at the base of the middle Williams Fork Formation: almost 90% of the sandstones in sand&ndash;rich intervals contain chlorite. The distribution of chlorite did not vary between reservoir architectural elements, with 70% of both crevasse splays and channel fills containing chlorite. The results of this study show that, for the middle and upper Williams Fork Formations at Grand Valley Field, 1) there are eight lithofacies and four architectural&ndash;element types identified from core; 2) the occurrence and distribution of accessory minerals (&lt;10%) of chlorite and potassium feldspar can be accurately estimated from limited core and well&ndash;log data; 3) chlorite occurrence does not vary significantly between reservoir architectural elements; 4) the abundance of chlorite near completion intervals and the occurrence of potassium feldspar in calculated mudstone lithologies indicate a need to re&ndash;evaluate the utilization of saturation models and lithology calculations in reservoir&ndash;quality evaluations.</p>