| Sumario: | The FEBEX in situ test simulated the engineered barrier of a nuclear waste repository and was in operation for 18 years under natural conditions. The barrier was composed of blocks of compacted FEBEX bentonite and a heater simulated the waste container. The gas permeability of bentonite samples taken at different positions around the heater during the final dismantling of the experiment was measured. These samples had initial dry density between 1.51 and 1.64 g/cm3 and water contents between 19.5 and 29.0%. Some of these samples were drilled between two blocks, therefore they had an interface along. The aim of the tests was to check: 1) the influence on gas permeability of the physical state of the samples (water content, dry density), 2) the effect of injection and confining pressures on gas permeability, 3) the change of gas transport properties with respect to the untretated, reference FEBEX bentonite and 4) the role of interfaces on gas transport. The gas permeability of the samples depended on water content and dry density, decreasing with the increase of both. These two variables changed across the barrier as a function of the distance to the heat and water sources, i.e. the heater and the granite. Thus, the gas permeability of the samples was also related to their position in the barrier, tending to be lower towards the granite, where the degree of saturation was higher. The gas permeability decreased with the accessible void ratio according to a power law, in the same way as expected for the FEBEX reference bentonite. Therefore, it seems that no changes on the gas transport properties of the bentonite took place during operation. Samples with an interface drilled in the internal ring of the barrier had higher permeability than samples of similar accessible void ratio with no interface, and it was necessary to apply higher confining pressures to reduce or suppress gas flow in them. In contrast, wetter samples drilled along interfaces of the intermediate and external rings of the barrier (which had very low accessible void ratio, due to saturation), had permeabilities closer to that corresponding to the same accessible void ratio in the reference bentonite.
|
|---|
