Nanoindentation of dry and fluid-saturated micro-porous rocks

In this report we explore the ability of nanoindentation technique to evaluate the pore-scale solid-fluid interactions in micro-porous rocks. We measure the creep deformation of a porous rock sample over a period of 3 minutes under a constant maximum force. The indentation tip is instrumented with a nano-DMA transducer which efficiently compensates for the thermal drifts. The candidate rock is a carbonate with micro-porous micritic cement. Secondary Electron (SE) images revealed a bimodal pore structure for this rock-type: regions (A) of micritic cement with micropores, and (B) with large grains and vuggy pores. The experiments were performed on dry rock samples as well as saturated with water (1 cp and buffered with 30 ppm calcite powder) and silicone oil (100 cp). Thus, the fluids presented a wide variation in viscosity and chemical reactivity. We then explored the size (maximum forces of 2, 4, and 8 mN) and loading rate (0.2-2 mN/sec) dependency of the observed creep behavior. The amount of total deformation within the 3 minutes of creep showed a uniform increase with a tendency to reach an equilibrium depth with creep rates (dh/h) below 5×10-3. The indentations in the water-saturated carbonate showed a 6-fold decrease in the Young's modulus (from 38 to 6 GPa) and 2-fold increase in creep magnitude (from 59 to 119 nm) compared with the dry indentations. We attribute these large differences to the possible chemical reaction of water and carbonate. This is further confirmed by comparing the hardness values, which showed that water softened the rock matrix by a factor of 4 (from 0.87 to 0.22 GPa). The carbonate sample saturated with oil, on the other hand, showed a higher modulus (47 GPa) and greater hardness (1.39 GPa), while the creep magnitude (31 nm) was half that observed in dry rock. We attribute this behavior to the viscous displacement of the pore fluid during consolidation of the poroelastic matrix. The loading rate-dependency and size (maximum load) sensitivity of the observed creep appear consistent with poroelasticity. We used Agbezuge and Deresiewicz's (1974) solution to derive poroelastic constants based on the recorded amount of creep. The analysis yields estimates of the diffusivity constant of the rock and the equilibrium creep depth. (We would like to acknowledge The U.S. Department of Energy (DOE) for their support)