Petrophysical Properties of Opalinus Clay Drill Cores Determined from Med-XCT Images

The determination of petrophysical properties such as density, porosity and mineralogical composition of the rock are key objectives in cored sections of drilling campaigns. In view of the large amount of sample material that accumulates during a drilling campaign, a seamless determination of the properties along the cores is not feasible if only direct methods are used. Therefore, fast, non-destructive and affordable methods have been developed. Three-dimensional images of Opalinus Clay drill cores were acquired by using a medical X-ray computed tomographic scanner (med-XCT). The CT numbers of the images were density calibrated, which allowed to determine bulk density variations along drill cores. Then, a relationship between rock composition and bulk density was built in form of linear regression models to predict the porosity or the contents of major components from density calibrated image data. This relationship was established on the basis of rock samples, of which mineralogical compositions and porosities were measured in the laboratory. It turned out that the bulk density of Opalinus Clay is systematically related to porosity and the contents of clay minerals, quartz and calcite. With increasing density, porosity and clay minerals content decrease. This is because the pores and clay minerals together form the porous clay matrix and are thus structurally connected. The density of the porous clay matrix is comparatively low, and its content therefore controls the bulk density of Opalinus Clay. With a decrease in the content of the porous clay matrix, the calcite and quartz contents both increase, which is associated with an increase in bulk density. No systematic behavior was found for the accessories. Thus, their influence on bulk density is considered to be small. Med-XCT in combination with reference samples allows the determination of the rock composition and porosity along drill cores. In the case of Opalinus Clay, a larger number of reference samples (> ~50) are required to predict the properties with confidence.