Keywords: rock salt, thermomechanical behaviour, multi-step creep, strain hardening, Lemaitre model, strain-hardening variable, hybrid model.

One of the main current issues of the nuclear industry is the problem of radioactive waste management. Among the solutions considered by the authorities, is the storage of these residues in an impermeable and tectonically stable deep geological formation. Most of the research work undertaken regarding this type of deep storage, concluded that four types of formations are likely to receive these wastes, namely: granite, schist, clay and rock salt.

In many European countries, the rock salt remains one of the potential rocks as a result of its low permeability, its dense and stable atomic structure, its abundance and its ability to present important creep deformations. These characteristics allow this material to ensure an efficient confinement for the waste disposals.

The EEC supported research program COSA I and II, which aimed at validation of the European computer codes for the modelling behaviour of the underground cavities built in rock salt [1], highlighted great divergence between the various research involved in the project about the characteristics of the rock salt constitutive law. Hereafter are some points that were subjected to animated debates:

• the existence of strain hardening,
• the intermediate principal stress influence nature,
• the existence and the form of a viscoplastic potential,
• the temperature influence nature.

The first point, that constitutes the main divergence, is the object of this paper. The various tests, conducted so far on the rock salt of Asse Mine (Germany), showed a coherent interpretation with the strain-hardening model [2]. On the other hand, several other authors suggested that no strain hardening characterized the in situ rock salt behaviour and that it is noted solely at the laboratory [3].

However, there is not a unanimous view regarding the presence or the absence of a stationary phase and the latter remains subjective to in situ data measures analysis.

Our work aims to provide a model reflecting the rock salt behaviour and thus providing a way to validate the laboratory tests as well as the in situ ones. In this paper, we will only focus on laboratory tests by means of multi-step creep tests with deviator and temperature changes The material chosen for this purpose is the rock salt extracted by Salin du Midi and Salines de l'Est in the Varangeville mine (Meurthe et Moselle, France).

The test results have revealed the strain-hardening character of rock salt superposed to the influence of temperature and stress. The interpretation on the test results with the model designed by Lemaitre [4], is not fully satisfactory for entire set of test results due to the fact that Lemaitre's model is used for the modelling of the viscoplastic behaviour of materials with only one strain hardening variable. Hence, a "hybrid" model introducing a double variable for strain hardening has been proposed. The validation of this model on the creep tests is very satisfactory and confirms the existence of two variables of strain hardening for rock salt behaviour

1

B. Come, "Le projet communautaire COSA : un exemple d'intercomparaison de codes de calcul géomécaniques pour le sel", Revue Française de Géotechnique, (10),1987.

2

K.H. Lux, S. Heusermann, "Creep tests on rock salt with changing load as a basis for the verification of theoretical material laws", Sixth International Symposium on Salt, (1), 17-35, 1983.

3

D.E. Munson, W.R. Wawersik, "Constitutive modeling of salt behavior - state of the technology", 7th ISMR Congress, Balkema, (3), 1797-1810,1993.

4

J. Lemaitre, "Détermination des lois de comportement des mat'eriaux 'elastoviscoplastiques", PhD Thesis, ONERA, Paris, 1970.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £140 +P&P)