Prediction of clay-rich rock failure coupling local and global non-destructive measurement techniques.

Understanding the damage processes in clay-bearing rocks is a decisive factor in geological engineering. But, more generally, they may also contribute to localized deformation and thus the rupture of fault gauges in seismic zones. Due to their physical and mechanical properties, such as low permeability, fracture healing potential, and effective radioactive elements adsorption capacity, several European countries plan to use these impermeable rocks to confine their nuclear waste in deep geological repositories. However, structural damage could lead to uncontrolled radionuclide dispersion by advective transport, thus the early detection of the damaging nucleation and evolution is decisive in geological engineering.

This study aims to explore the interplay between P-wave ultrasonic velocity and the micro-deformation mechanism identified by digital image correlation (DIC) in order to predict the failure of the sample.

To this end, uniaxial compression tests are performed on small-scale Tournemire shale samples (8 mm in width and twice as long) using a home-designed miniature loading frame under controlled relative humidity of RH = 75% and RH = 20%. These tests involve two simultaneous measurements: 1) the axial and lateral P-wave travel times, recorded by an active emission system, and 2) the full displacement field on the sample surface, based on Digital Image Correlation (DIC) applied to high-resolution optical or Environmental Scanning Electron Microscopy (ESEM) images. The latter allows for the calculation of 2D full strain fields in order to characterize the deformation at different scales while performing simultaneous acoustic measurements. The processed images are representative of two distinct scales: one at the microscale using the ESEM with a resolution of 24 nm/pixel and the other at the mesoscale using an optical camera with a resolution of 0.55 μm/pixel.

The results allow us to discuss the global evolution of the acoustic wave velocities during the uniaxial loading process with respect to the identified active micro-damage mechanisms.