Microscopic time‐dependent mechanical behavior of shale derived from nanoindentation

Understanding the microscopic time‐dependent mechanical behavior of shale is critical for assessing macroscopic creep and engineering applications. Grid nanoindentation experiments and nanoindentation creep tests were systematically conducted to investigate microscopic creep behaviors in shale. The indentation creep displacements and creep rates of the shale's soft, intermediate, and hard phases showed the same evolution patterns. The creep deformation was much higher in the soft phase than in the other two phases. However, the difference in the steady‐state creep rates between the three mechanical phases was negligible. A linear relationship was observed between the microscopic contact creep modulus and the microscopic Young's modulus, hardness, creep displacement, and creep rate. The primary mechanism of microscopic creep in shale revealed by the creep strain rate sensitivity parameter was the extension and closure of microcracks. The differences in the microscopic creep parameters derived from the experimental data using the deconvolution methods and representative point methods were evaluated, and the applicability of the two methods was described. The performances of commonly used creep models to predict the microscopic creep behaviors were evaluated. The Burgers model provided the best performance in predicting the steady‐state creep deformation and creep rate. The ability of the Mori–Tanaka and Voigt–Reuss–Hill models to derive macroscopic parameters from microscopic mechanical parameters was compared. Both methods provided macroscopic Young's modulus values close to the experimental values; however, neither could predict macroscopic creep parameters based on microscopic creep parameters. A linear relationship existed between the microscopic contact creep modulus and the microscopic Young's modulus, hardness, indentation creep deformation, and creep rate. The creep strain rate sensitivity parameter for the soft, intermediate, and hard phases of the shale was less than 0.33. The Burgers model can predict the microscopic steady‐state creep deformation and creep rate. The Mori–Tanaka method and Voigt–Reuss–Hill models could not predict macroscopic creep parameters based on microscopic creep parameters.