Examining the reliability of current micro-and nano-indentation-based rock mechanical upscaling schemes: a comprehensive comparison with uniaxial/triaxial macroscopic mechanical testing

Shale and marlstone are two typical rocks for gas reservoirs and caprocks for the underground storage of carbon dioxide. Accurately estimating their mechanical properties is challenging due to these rocks' multiphase and multiscale structures. In this study, we conducted triaxial compressive, microindentation, and nanoindentation tests on samples collected in a borehole to investigate Young’s modulus at different scales. To upscale the mechanical properties, we utilized existing homogenization models, namely Mori–Tanaka (MT), Self-Consistent (SC), and Voigt-Reuss-Hill average (VRH), and compared the moduli upscaled from the nanoindentation technique and rock mineralogical method with those measured from microindentation and triaxial compressive tests. Results showed that the rock mineralogy-based MT and SC methods produced higher moduli than the nanoindentation-based MT method, and all of these were significantly larger than those obtained from the triaxial compressive tests. Additionally, the nanoindentation-based VRH method and the microindentation test yielded close moduli but still exhibited a certain gap compared to the macroscopic mechanical test results. This suggests that the current indentation-based mechanical upgrading method cannot accurately estimate the rock mechanics at the macroscale. Nevertheless, in the absence of a better solution, the above two methods can still be considered. The difference between the moduli measured by the compressive test and those upscaled through these homogenization models was attributed to the underestimation caused by the low elastic stiffness of pores, fractures, and grain contacts in the rocks. Therefore, more advanced methods are required to develop to make mechanical upscaling closer to the real situation of rocks.