Liquefaction responses of fibre reinforced sand in shaking table tests with a laminated shear stack

Abstract

Mixing unidimensional tension-resistant elements into cohesionless soils, which is well known as fibre-reinforcement, is effective at strengthening the soils as well as increasing their liquefaction resistance in triaxial tests. In this study shaking table tests were performed on unreinforced and fibre reinforced sand models, contained in a laminated shear stack, to gather evidence as to whether or not the fibre reinforcement technology might reduce liquefaction susceptibility in a loading condition that has the semblance of an earthquake. The test results show that the excess pore pressures generated, and the amount by which the equivalent shear modulus is reduced, are significant in unreinforced and reinforced models when exposed to continuous sinoidal biaxial shaking simultaneously in horizontal and vertical directions. Fibre-reinforcement delays and reduces the build-up of excess pore pressures, slightly. Fibre-reinforcement also decreases slightly the attenuation of the acceleration amplitudes as well as the degradation of stiffness resulting from the generation of excess pore pressures. A constitutive model based on the rule of mixtures, which captures the load sharing between fibres, sand skeleton and pore water in triaxial loading conditions, and accounts for the anisotropy of the fibre orientation distribution, was modified and then used to study the shaking table deformation pattern. The stress the fibres impose on the sand skeleton in the vertical direction was determined and a new pore pressure ratio was used to assess whether or not liquefaction occurred. This enabled a mechanistic understanding of how the fibres alter the sand skeleton stresses and suppress liquefaction development. Liquefaction eventually occurred, or was close to occurring, in the reinforced model at different depths even though the fibres added stresses to the sand skeleton. The earth pressures prevailing in these 1 g shaking table tests were not sufficiently large for optimal interaction between fibres and sand skeleton. The fibres were not sufficiently tensioned as the confining (clamping) stresses provided by the neighbouring sand particles were not large enough. This means that the benefits of this reinforcement technology may be minor in practical situations where the fibre reinforced soil is at shallow depths.