Non-classical, non-linear elasticity in rocks: experiments in a triaxial cell with pore pressure control

In rocks and concrete, dynamic excitation leads to a fast softening of the material, followed by a slower recovery process where the material recovers part of its initial stiffness as a logarithmic function of time. This requires us to exit the convenient framework of time independent elastic properties, linear or not, and investigate non-classical, non-linear elastic behavior.

These phenomena can be observed during seismic events in affected infrastructure as well as in the subsurface. Since the transient material changes are not restricted to elastic parameters but also affect hydraulic and electric parameters as well as material strength, as documented for instance by long lasting changes in landslide rates, it is of major interest to characterize the softening and recovery phases. It may help us gain more insight in hazard prediction from both a geological and engineering perspective.

The underlying physics behind those non-classical, non-linear effects, sometimes referred to as “Nonlinear Mesoscopic Elasticity”, are not agreed upon. There is a lack of experiments that would allow us to discriminate between the existing models.: we aim to contribute to filling that knowledge gap.

Our experiments are made on a sample of Bentheim sandstone, initially dry and then fully saturated, in a triaxial cell. We subject the sample to loading and holding cycles in the microstrain range, while also varying confining pressure and pore pressure. Active acoustic measurements during those loading cycles with an array of 14 piezoelectric sensors allow us to monitor relative velocity changes during the experiment by using Coda Wave Interferometry (CWI).

We observe the dynamic softening as well as the recovery processes in the sample during repeated loading phases of different durations. We find that characteristics of the observed velocity changes vary depending on the observed sensor combination, indicating spatial variability of the response, as well as depending on the lapse time and frequency content of the acoustic measurements that we perform the CWI on.

These experiments serve to estimate the exact capabilities of our experimental setup in terms of signal quality, signal stability and lapse time dependent decorrelation of coda waves. We expect our results to inform a future series of similar but more refined experiments addressing the pore pressure dependence of the non-classical response of rocks.