Modelling long-term behaviour of a natural gypsum rock

• Published in: Mechanics of materials 2005-12, Vol.37 (12), p.1223-1241
• Main Authors: Hoxha, Dashnor, Giraud, Albert, Homand, Françoise
• Format: Article
• Language: English
• Subjects: Condensed matter: structure, mechanical and thermal properties; Creep; Damage; Exact sciences and technology; Fundamental areas of phenomenology (including applications); Gypsum rock; Inelasticity (thermoplasticity, viscoplasticity...); Measurement and testing methods; Mechanical and acoustical properties of condensed matter; Mechanical properties of solids; Physics; Recovery; Relative humidity; Solid mechanics; Strain hardening; Structural and continuum mechanics; Viscoplasticity
• ISSN: 0167-6636; 1872-7743
• DOI: 10.1016/j.mechmat.2005.06.002

This paper deals with the modelling of the long-term behaviour of a natural gypsum rock, that manifests a volumetric strain variation during the creep, a significant influence of the relative humidity on the creep strain rate and an inverse creep during unloading. Based upon laboratory results a creep model describing primary and inverse creep is firstly constructed. The creep strains are seen as a sum of the recoverable and the viscoplastic strains. The firsts are results of two competitive creep mechanisms: the forward reversible creep strain, which is a function of the active stress and the backward creep itself a function of the cumulated deformation energy. According with laboratory tests, the viscoplastic strains have been considered to be mean-stress independent, which yields to an Odqvist-like creep model with some modifications in order to describe the primary creep and the volumetric strain evolution. The role of the relative humidity in the long-term behaviour of this rock is taken into account by the evolution of the creep activation energy. The model has a limited number of parameters, all accessible from standard laboratory tests. Its description was completed by the procedure of the parameters identification. Then, the predictions of the model have been compared with laboratory results and a good concordance has been found with laboratory data during the first stages of creep. Some discrepancies at the very late stages of the creep are explained by a damage of the rock during the creep. Accordingly, an extension of the model has been proposed that takes into account this damage. Consequently, the extended model fits better laboratory results and allows describing of the accelerated creep.