The problem of cavity expansion in unsaturated soils is investigated. A unified constitutive model for unsaturated soils is presented in a critical state framework using the concepts of effective stress and bounding surface plasticity theory. Consideration is given to the effects of suction and particle crushing in the definition of the critical state. A simple isotropic elastic rule is adopted. A loading surface and bounding surface of the same shape are defined using simple and versatile functions. A limiting isotropic compression line exists, towards which the stress trajectories of all isotropic compression load paths approach. A non-associated flow rule is assumed for all soil types. Isotropic hardening/softening occurs due to changes in plastic volumetric strains as well as suction for some unsaturated soils, enabling account of the phenomenon of volumetric collapse upon wetting. Results of isotropic compression tests, oedometric compression tests and drained and undrained triaxial compression tests performed on Kurnell (quartz) sand in saturated and unsaturated states and subjected to stresses sufficient to cause particle crushing are presented and used to calibrate the model. The model is also calibrated using results reported in the literature for triaxial tests performed on saturated and unsaturated speswhite kaolin and three load paths. For both soils the model leads to a much improved fit between simulation and experiment compared to that for models based on conventional plasticity theory. The model is implemented into a cavity expansion analysis using the similarity technique, extended for application to unsaturated soils. Cylindrical and spherical cavities are considered, as are drained and undrained conditions. Cavity expansion results for the bounding surface model and conventional plasticity models are compared for saturated conditions. Substantial differences highlight the importance of adopting a model that accurately describes stress-strain behaviour. Cavity expansion results for the bounding surface model and saturated and unsaturated conditions are also compared. Substantial differences, particularly in the limit pressure, highlight the major influence of suction and the importance of accounting for this when using cavity expansion theory to interpret results of the cone penetration and pressuremeter tests.
| Identifer | oai:union.ndltd.org:ADTP/188005 |
| Date | January 2004 |
| Creators | Russell, Adrian Robert, Civil & Environmental Engineering, Faculty of Engineering, UNSW |
| Publisher | Awarded by:University of New South Wales. School of Civil and Environmental Engineering |
| Source Sets | Australiasian Digital Theses Program |
| Language | English |
| Detected Language | English |
| Rights | Copyright Adrian Robert Russell, http://unsworks.unsw.edu.au/copyright |
Page generated in 0.0018 seconds