Application and Quality Control of Core Data for the Development and Validation of Elemental Spectroscopy Log Interpretation

Core or cuttings samples are often analyzed for chemical and mineral composition to provide ground truth for developing petrophysical or geological applications or for validating log interpretations of elemental concentrations, mineralogy, and matrix properties. Unfortunately, some core data are inaccurate, but they are rarely subjected to quality-control measures and can therefore lead to erroneous conclusions regarding the validity of the log data. For core data, elemental concentrations are generally measured by X-ray fluorescence (XRF) or inductively coupled plasma (ICP) techniques. The best way to validate results from these techniques is to test certified reference materials that are composed of sedimentary minerals similar in composition to samples of interest. Core mineralogy is most commonly analyzed by X-ray diffraction (XRD) or Fourier-transform infrared spectroscopy (FTIR). Laboratory results can be evaluated by analyzing known mixtures of certified minerals. Once it has been established that sources of accurate core elemental concentrations and mineralogy are available, it is advisable to implement routine quality-control monitoring. An example of a quality-control measure is a technique that requires independent analyses for elemental and mineral concentrations. The technique assumes that the minerals have fixed elemental compositions. Measured mineralogy is used to compute elemental concentrations of the major elements, including Si, Al, K, Fe, S, Ca, Mg, and Na. Derived elemental concentrations are compared with the measured elemental concentrations. Deviations between the derived and measured concentrations are used to evaluate the quality of the input data. Examples of both good and poor inputs for elemental and mineral data are shown. Once the quality of the data is proved to be good, it is possible to use the data to validate the accuracy of interpretations developed for elemental spectroscopy logs, such as the closure model to convert concentrations to yields and models to interpret mineralogy.