| Summary: | The experimental work performed within the FEBEX under natural conditions and at full scale, a “mockup” test performed at almost full scale and a large number of laboratory tests. Most of the laboratory tests have been carried out prior to or in parallel with the two large-scale tests; in other words, on samples of the same barrier material as in the in situ and “mock-up” tests but not on the barriers themselves. However, part of the laboratory tests (those included in this document) have been performed on samples of bentonite obtained during the partial dismantling of the in situ test, performed in 2002 after five years of heating and hydration. Two types of tests have been performed: 1) characterization of the bentonite, with a view to gaining insight into the state of the barrier and for comparison with the results of THM and THG model predictions, and 2) tests to determine the changes in THM and THG properties occurring during operation as a result of the combined action of temperature, water, joints and solutes. Presented below is a summary of the conclusions drawn from the results of the laboratory tests included in this document. The distribution of the humidity and dry density of the bentonite in vertical sections shows axial symmetry. There are no major variations between the average humidity of sections located around the heater and those others that are outside its area of influence. Neither are there differences between sections located in areas of different hydraulic conductivity of the granite. This homogeneity, along with the radial distribution of humidity in vertical sections, is a result of the control exercised by the bentonite over hydration kinetics, due to the major difference between the permeabilities of the bentonite and the granite. The average degree of saturation of all the bentonite extracted is 85 percent. During operation the bentonite underwent a generalized increase in volume that has led to a reduction from the dry density of the compacted blocks (1.70 g/cm3) to the average dry density of the barrier calculated from the measurements performed on the samples extracted (1.58 g/cm3). This is due to the expansion of the bentonite to fill the construction voids and to the slight decompression experienced by the barrier during dismantling and sampling. The mineralogical and geochemical characterization performed would appear to indicate that there have not been major modifications in the bentonite during operation. The cation exchange capacity has increased with respect to the initial value, fundamentally due to the generalised increase in exchangeable potassium and calcium. An increase in exchangeable sodium has been observed towards the granite in both the sections subjected to a thermal gradient and in the isotherm sections. The hydration of the bentonite in the blocks in contact with the granite causes the dissolution of the most soluble minerals (sulphates, carbonates and chlorides), which are transported towards the inner part of the barrier. As of the date of dismantling, this has generated different saline fronts depending on the mobility of the dissolved ions: the behaviour of Na, Ca and Mg is similar to that of the chloride, while the mobility of the sulphate is considerably lower. The movement of solutes is quicker in areas affected by temperature. The chemical and mineralogical composition of the bentonite in contact with the concrete plug is similar to that of the untreated bentonite. Only an increase in aluminium content and the occasional appearance of larger quantities of calcite and gypsum are recorded. An increase has been observed also in exchangeable Ca, Na and K, along with an increase in the salinity of the interstitial water. Various thermal, hydraulic and mechanical properties of the bentonite extracted from the barrier have been determined, in order to check for possible changes in THM behaviour caused during the experiment. The water retention capacity of the samples obtained from the dismounted barrier is similar to that of the samples of bentonite not subjected to the experiment. The hydraulic conductivity of the samples from dismantling depends fundamentally on dry density (as occurred with the bentonite not used in the barrier), as a result of which it is related to their position in the barrier. The swelling capacity of the bentonite has not changed irreversibly after five years of operation under repository conditions. The pre-consolidation pressure of the samples from the dismounted barrier has decreased from the initial value of 40 MPa to values of less than 10 MPa (lower in the outer ring of the barrier), this being due to microstructural changes associated with the increase in volume experienced during hydration.The tests performed on remoulded samples of bentonite from dismantling, with joints parallel and perpendicular to flow, show that once saturated the medium becomes homogeneous and the joints seal completely, as a result of which the hydro-mechanical properties of the material depend only on its dry density.
|
|---|
