Geomechanical characterisation of organic-rich calcareous shale using AFM and nanoindentation

Abstract

AbstractThe geomechanical integrity of shale overburden is a highly significant geological risk factor for a range of engineering and energy-related applications including CO$$_2$$ 2 storage and unconventional hydrocarbon production. This paper aims to provide a comprehensive set of high-quality nano- and micro-mechanical data on shale samples to better constrain the macroscopic mechanical properties that result from the microstructural constituents of shale. We present the first study of the mechanical responses of a calcareous shale over length scales of 10 nm to 100 $$\upmu$$ μ m, combining approaches involving atomic force microscopy (AFM), and both low-load and high-load nanoindentation. PeakForce quantitative nanomechanical mapping AFM (PF-QNM) and quantitative imaging (QI-AFM) give similar results for Young’s modulus up to 25 GPa, with both techniques generating values for organic matter of 5–10 GPa. Of the two AFM techniques, only PF-QNM generates robust results at higher moduli, giving similar results to low-load nanoindentation up to 60 GPa. Measured moduli for clay, calcite, and quartz-feldspar are $$22 \pm 2\,\hbox { GPa}$$ 22 ± 2 GPa , $$42 \pm 8\,\hbox { GPa}$$ 42 ± 8 GPa , and $$55 \pm 10\,\hbox { GPa}$$ 55 ± 10 GPa respectively. For calcite and quartz-feldspar, these values are significantly lower than measurements made on highly crystalline phases. High-load nanoindentation generates an unimodal mechanical response in the range of 40–50 GPa for both samples studied here, consistent with calcite being the dominant mineral phase. Voigt and Reuss bounds calculated from low-load nanoindentation results for individual phases generate the expected composite value measured by high-load nanoindentation at length scales of 100–600 $$\upmu$$ μ m. In contrast, moduli measured on more highly crystalline mineral phases using data from literature do not match the composite value. More emphasis should, therefore, be placed on the use of nano- and micro-scale data as the inputs to effective medium models and homogenisation schemes to predict the bulk shale mechanical response.