| Sumario: | The aim of the FEBEX (Full-scale Engineered Barriers Experiment) Project is to study the behavior of the near field components of a high-level radioactive waste (HLW) repository in crystalline rock, in accordance with a concept in which the canisters are surrounded by a clay barrier constructed of highly-compacted bentonite blocks. The FEBEX consists of a full scale "in situ" test (Grimsel, Switzerland) performed under natural conditions; a I/ mock-up" test (CIEMAT, Madrid), at almost full scale; and a series of laboratory tests designed to complement the information from the two largescale tests. During the pre-operational stage of the Pro¡ect, the planning, design and Installation of the two large-scale tests was accomplished, as well as the predictive modelling of their behavior and the basic characterisation of the bentonite. This report is one of those concerning the pre-operational stage of the Project, and refers to the selection and characterisation of the FEBEX bentonite, and to the fabrication of the blocks used in the "in situ" and ,, mock-up" tests. The clay utilized comes from the Cortijo de Archidona deposit (Almería), from where the S-2 clay used in previous ENRESA R&D Projects was also extracted. The results obtained in these Pro¡ects are summarised in Chapter 2, as some of the parameters considered in the predictive modelling were taken from the 5-2 bentonite database, due to their not having yet been determined for the FEBEX clay. For the FEBEX Project, about 300 tons of bentonite have been selected, homogenized and treated. At the factory, the clay was disaggregated and gently dried to a water content of around 14 %, all the material of particle size greater than 5 mm being re¡ected. The processed material was used for fabrication of the blocks for the two large-scale tests and for the laboratory tests. Samples of each processing lot were taken, and were mixed to obtain a single sample, whose reference is 70-IMA- 3-4-0, which is the one used by all the participant laboratories. Mineralogical and chemical characterisation was performed at CIEMAT and CSICZaidín, and the mechanical, hydraulic and thermal characterisation was carried out by CIEMAT and UPC-DIT; the results obtained at each laboratory were consistent with the rest of the laboratories. The FEBEX bentonite has a montmorillonite content greater than 90 %, with quartz, plagioclase, potassic feldspar, calcite and trydimite as accessory minerals. There may be illite/smectite interstratifications up to 15 %. The cation exchange capacity is higher than 100 meq/100g, and in the exchange complex 42 % Ca, 33 % Mg, 23 % Na and 2 % K are present. The identification properties are not exactly those expected for a material with this mineralogy. Both the liquid limit (100 %) and the external specific surface (32 m 2/g) are rather low, and the less than 2 .t.rn fraction obtained after intense dispersion amounts to only 65 %. This might mean that the particles are agglutinated in a certain way (possibly by a siliceous cement), although the consequences of this on the behavior of the clay have not been evaluated. Swelling pressure and saturated hydraulic conductivity were obtained for specimens compacted to dry densities of 1.60 y 1.70 g/cm 3 , along with the thermal conductivity for these densities and different degrees of saturation. The suction/water content relationship and a preliminary state surface for decreasing suctions and small stress changes were determined. The results obtained show slight variations with respect to the values obtained for the 3-2 clay, with a tendency of the FEBEX clay to be more impervious and to develop a higher swelling pressure. However, for the performance assessment of a repository (PA), these differences are considered to be in the expected range of variation of the bentonite properties. Two sets of blocks have been manufactured: one for the "mock-up" test, under the management and supervision of CIEMAT, and another for the "in situ" tests, under the management and supervision of AITEMIN. In both cases a crown design has been adopted, in order to minimize the void volume of the barrier. The blocks have been uniaxially compacted to pressures of 40 to 50 MPa, the resulting dry density being 1.77 g/cm 3 for the blocks for the "mock-up" test and 1.70 g/cm 3 for the blocks for the "in situ" test. Consequently, the average dry density of the installed barrier is 1.65 and 1.60 g/cm 3 , respectively.
|
|---|
