WP4 Laboratory testing- progresshttps://www.beacon-h2020.eu/wp-content/themes/corpus/images/empty/thumbnail.jpg 150 150 Mary Westermark Mary Westermark https://secure.gravatar.com/avatar/a6e923d45e990c121a3301801ad91e24?s=96&d=mm&r=g
The experiments performed in the frame of Beacon are aimed at obtaining a more complete cognition of bentonite behaviour – by investigating different wetting dynamics or load scenarios, looking at different granulations and inhomogeneous systems or at different bentonite types, and by developing new methods of measurement. The work performed by various organisations has been structured into investigation of
- Hydro-mechanical behaviour of macroscopically homogeneous bentonite material
- Swelling into limited void
- Binary mixtures or artificial inhomogeneities
- Influence of the degree of saturation on the shearing behaviour at the bentonite – steel interface
The first bullet addresses the influence of initial granulometry of an unsaturated bentonite and of different hydro-mechanical paths on the final state after resaturation. Both types of experiments performed by EPFL on MX-80 samples showed a significant influence on final void ratio. The applied hydro-mechanical paths as well as the initial granulation influenced, however, mostly the inter-aggregate (macro) porosity. Additional constant load and constant volume tests performed by CU and CTU, respectively involved Cerny vrch samples in order to provide calibration data for this type of bentonite.
Regarding swelling into limited void, BGS and CIEMAT performed test series of bentonite swelling, using bentonite blocks in isochoric cells leaving a void at one end of the sample. BGS investigated the influence of the ratio between block and void volume at ambient and elevated temperature, concentrating on swelling pressure and pore pressure evolution. CIEMAT concentrated on the effect of hydration dynamics, performing experiments with liquid water resaturation and resaturation by a vapour phase at two differen suction states at ambient temperature. In all experiments, the bentonite was able to fill the complete void, no open space remained. In all experiments, however, an inhomogeneous water content (and thus dry density) distribution remained. A higher tempeature and/or a slower resaturation worked in favour of a more homogeneous water content. JYU and KIT have also been performing experiments on bentonite swelling into a limited void, concentrating on the development of new measurement techniques: Particle movement tracking (JYU) and small-scale spatially-resolved swelling pressure measurement.
CIEMAT and CEA performed hydration tests under constant volume conditions using initially heterogeneous samples. CEA performed small-scale dual density tests in diameter 57 mm with superimposed blocks of MX-80 bentonite compacted at different dry density or pellets/powder mixes. CIEMAT performed tests in 10×10 cm cylindrical samples half of which were composed of pellets and the other half of a compacted block of FEBEX bentonite. These tests show again that Final dry density gradients seem mostly driven by the speed of the hydration process and not directly by the initial heterogeneities. Dry density gradients tend to soften over time, although complete homogenization was not reached in any test. The local stresses in bentonite samples also exhibit large heterogeneities and do not converge at the same accessible time scales. Residual heterogeneities are not dependent on (or have a simple relation with) the initial sample heterogeneity. It seems that if the initial deformation is very large, it affects the macrostructure and becomes irreversible, conditioning the subsequent evolution of the system, as shown by porosimetry measurements. This would be supported by the fact that it is the macropore volume that gets mostly modified as a result of the initial hydration and swelling.
EPFL investigated the internal shearing of granular MX80 bentonite as well as the shearing at a steel-bentonite interface, for a fuller-type and a unifractional grain size distribution. For both tested granulations, the interface shearing strength of the material at hygroscopic state was lower than the internal shearing. Unifractional samples showed lower values of shear strength parameters in comparison to those prepared with a Fuller-type granulation. Samples characterized by a higher water content showed a higher peak of shearing strength when compared with samples of lower water content. The increase of strength appears to be governed by the increase of adhesion between steel-bentonite upon hydration.
All the experiments and results summarised here are documented in detail in the Deliverable D4.1/2 “Bentonite mechanical evolution – experimental work for the support of model development and validation” of the Beacon project, which is available at https://www.beacon-h2020.eu/deliverables/. Not all of the investigations are completed yet – future work includes the completion of ongoing experiments, and especially the development of new or improved test methods, like particle tracking, small-scale pressure measurement or X-CT in-situ observation is furthered.