Salt Creep: Transition Between the Low and High Stress Domains

Abstract

In 2014–2016, creep tests were performed in a dead-end drift of the Altaussee mine, where temperature and relative humidity experience very small fluctuations. These tests, which were several months long, proved that the creep rate of a natural salt sample is much faster in the 0.2–1 MPa deviatoric stress range than the creep rate extrapolated from standard laboratory creep tests performed in the 5–20 MPa range. In addition, the quasi-steady strain rate is a linear function of stress, and it is faster when grain size is smaller. These findings were consistent with microphysical models of pressure solution creep (rather than dislocation creep, which is the governing creep mechanism at high stresses). A gap in experimental data remained in the 1–5 MPa range, calling for a follow-up experimental program. In 2016–2019, three multi-stage creep tests were performed on salt samples from Hauterives (France), Avery Island (Louisiana, USA), and Gorleben (Germany), which had been tested in the 0.2–1 MPa range during the 2014–2016 campaign. Loads of 1.5, 3, and 4.5 MPa were applied successively on each sample for 8 months. Steady state was not reached at the end of each 8-month stage. However, tests results suggest that, in the 0.2–3 MPa range, the relationship between the strain rate and the applied stress is linear, a characteristic feature of pressure solution. For these three samples, the relationship between strain rate and deviatoric stress departs from linearity when the deviator is larger than approximately 3–4.5 MPa, pointing to a transition to dislocation creep at higher deviatoric levels.