Accurate mineral quantification is critical for reservoir characterization, if reservoir properties are closely related to mineral assemblage and distribution of a reservoir. However, it is not easy to estimate the accurate mineralogy of natural rocks by a single analytical method. Although X-ray Diffraction (XRD) analysis is the most popular method for identifying mineralogy, it has limitations in quantifying it in principle. Furthermore, there is no single prevailing method for the quantification, especially in case of volcanic reservoirs.
This study established the mineral quantification workflow for volcanic reservoirs by integrated multiple analyses (XRD, X‐ray Fluorescence (XRF), Rock-Eval pyrolysis, thin section, and Scanning Electron Microscope (SEM) + Energy Dispersive X-ray Spectroscopy (EDS) with mineral mapping system). This method mainly involves the traditional stoichiometric concept; calculation of mineralogy from bulk rock chemical composition. In addition, our innovative workflow consists of the following two steps. First, the accuracy of the stoichiometric calculation, which has been difficult by now, is validated by comparing areal minerals from the SEM+EDS mineral mapping system to volumetric minerals calculated from the chemical composition of the same area. Second, we calculate the mineral composition of the rock samples of interest using the validated stoichiometric model. XRF provides the chemical composition, which is essential for stoichiometric calculation, while XRD constrains the mineral assemblage of the rock and Rock-Eval pyrolysis constrains the fraction of carbonate minerals to optimize the stoichiometric calculation.
We applied this method to the rhyolitic reservoir and the basaltic reservoir of domestic oil and gas fields in Tohoku, Japan. Both reservoir rocks are known as Green Tuff which erupted underwater around the Sea of Japan. Their reservoir qualities were improved by the hydrothermal alteration after the eruption. Simultaneously, the alteration resulted in complex mineral assemblages, and therefore, accurately understanding the mineralogy is essential for reservoir characterization but challenging. However, we successfully quantified the mineralogy for each reservoir rock by using this quantifying method. The mineralogical data contribute to estimating more reliable reservoir properties when used as calibration points for well log mineral analysis. Moreover, it also provides us with important geological information for the detailed evaluation of, for example, the potential acidizing stimulation of the field.
