Search

Compacted bentonite is planned to be used as buffer and backfill materials for the containment of radioactive waste in underground repositories. The performance of these barriers depends on the swelling capacity of bentonite upon hydration. Prolonged interaction between bentonite and alkaline fluids from neighbouring concrete structures can impair the swelling capacity due to profound changes in the chemical composition of bentonite. The coupled hydro-chemo-mechanical behaviour of bentonite under such conditions is at present not well understood. This paper presents for the first time a combined experimental and modelling study that addresses this coupled behaviour with the aim of understanding the key mechanisms leading to swelling pressure loss. Two experiments are presented in which compacted Wyoming bentonite was saturated with either clay or cementitious water, leading to different initial swelling capacities. The samples were subsequently subject to a flow of a KOH-rich cementitious water leading to a slow but sustained decrease in swelling pressure in both tests. The main novelty is the application of a recently developed hydro-chemo-mechanical model for bentonite for interpretation of the experiments. The model accounts for the impact of montmorillonite dissolution, cation exchange reactions, and changes in salinity on the swelling capacity of bentonite. The model results show a relatively good agreement with experimental measurements and suggest that the decrease in swelling capacity of bentonite is driven primarily by an increase in potassium fraction in the interlayer water and by montmorillonite dissolution. • Compacted Wyoming bentonite is saturated and percolated with cementitious (KOH rich) water. • Sustained decrease in the swelling pressure upon cementitious water injection. • Hydro-chemo-mechanical (HCM) model for swelling clays based on the Barcelona Basic Model (BBM). • Swelling pressure loss is dominated by K+ uptake in the interlayer. [ABSTRACT FROM AUTHOR]

This paper aims to theoretically investigate and characterize the failure mechanism of hydrated bentonite plugs, when used as a sealing material in wellbores. Using the proposed theoretical models, it was determined that when hydrated bentonite plugs failed due to friction at the walls, created by internal swelling pressure or through internal shear failure, the resulting failure pressure was linear with the plug height (H), coefficient of friction or internal shear strength, and inversely proportional to plug diameter (D). On the other hand, if the friction at the walls was created by plug weight then the failure pressure was a function of H 2 /D. However, in this case, the frictional strength increased parabolically with height, and, at reasonable plug heights, the frictional strength became greater than shear strength, and resulted in failure due to internal shear failure, which also made the failure pressure a function of H/D. This suggests that plug strength is usually a linear function of H/D. Moreover, the coefficient of friction and internal shear strength are dependent on final moisture content of the hydrated plug, the salinity of the hydration water, pressure, temperature and hydration time. The strength of a bentonite plug is inversely proportional the total volume by which it expands during hydration and takes in excess of 185 days to achieve this state. Plug strength decreased in saline water, although even at elevated concentrations sufficient strength remained to produce and adequate seal.From very limited data, shown here, Queensland bentonite, from the 5D zone of Amcol's Gurulmundi mine, appeared to have similar plugging strength as Wyoming bentonite. • Developed novel models to predict the failure pressure of bentonite plugs. • Provided insights into the failure mechanisms of bentonite plugs. • Bentonite plug strength is a linear function of plug height to diameter ratio. • Friction coefficient is dependent on final moisture content of the hydrated plug. • Bentonite plug strength increases with hydration time and decreases in saline water. [ABSTRACT FROM AUTHOR]

Abstract: The NAGRA concept for a high level waste repository includes bentonite pellets for backfilling around the spent nuclear fuel canisters. Pellets have therefore been produced from granulated Wyoming bentonite (Am. Colloid., brand name MX-80), which have been tested as part of NAGRA’s research program with respect to its sealing properties. Two types of pellet material were used, one containing only bentonite and the other containing a mixture of 70% bentonite and 30% sand. As reference, standard granulated MX-80 bentonite material was also used. The pellets were industrially crushed in order to give better grain size distribution with respect to field compaction. In all tests a 0.2M NaCl test solution was used, simulating the groundwater composition of Opalinus Clay host rock. Swelling pressure and hydraulic conductivity were measured for all three materials at five different samples densities, ranging from 1500kg/m3 to 2050kg/m3 after full water saturation. The results unambiguously show that there is no significant difference at any density between the MX-80 pellet material and the original granulated MX-80 material with respect to swelling pressure or hydraulic conductivity. Thus, the basic sealing properties in the pellet materials are not affected by the drying or compaction treatment. [Copyright &y& Elsevier]