Numerous observations have shown that shale formations may creep in and close the annulus between rock and casing in deep wells, thus effectively sealing off the well sections in a natural way. This movement of the rock is partly controlled by strain and partly by stress, and parts of the rock may be in a post-peak state while other parts are in a pre-peak state. The effect of pre-peak versus post-peak conditions on creep characteristics was studied in a set of stress relaxation tests on Pierre II shale. It was found that the stress relaxation process can be described with great precision both by a logarithmic function and a power-law function. However, only the power-law function fulfills the expected asymptotic behavior for infinite time. Temperature fluctuations can be accurately accounted for by an Arrhenius-type factor. The shear stress relaxation appears to be insensitive to pore pressure alterations. There is an indication that the relaxation rate is slightly higher when the rock is in a strain hardening state than in a strain softening state. No significant effect of exposure to NaOH was seen on the relaxation process.
Completion of oil and gas wells often implies that the annulus between rock and casing is left open in long sections. Such sections are potential pathways for leakage from the reservoir to the surface and needs to be sealed off at selected spots prior to plugging and abandonment. It is however a frequent and highly welcome observation that shale formations creep in, closes the annulus, and thus form a natural barrier that seals the well (Williams et al., 2009; Kristiansen et al., 2018). There are limited possibilities to monitor this process in the field as it goes on behind the casing, but the terminal state is nevertheless of great importance since it governs the long-term sealing efficiency of the barrier. It is therefore interesting to study long-term shale creep under controllable laboratory conditions.