This grant provides funding to investigate whether a unique relationship exists between small strain shear modulus and effective stresses at failure during triaxial compression for dilatant soils. This hypothesis was developed based on preliminary tests involving isotropically consolidated drained triaxial compression tests on an artificially cemented silty sand with shear wave velocity measurements made throughout shearing. There were two important findings from these tests: 1.) the ratio of small strain shear modulus to principal effective stress at failure was constant for the given soil, independent of the degree of cementation, density, and consolidation stress; and 2.) there was a clear trend of the small strain shear modulus increasing during shear to a maximum value and then decreasing despite continued increases in the deviator stress up to failure. To test this hypothesis, Ko-consolidated drained triaxial compression tests will be performed on three different soils in a dilative state: coarse, angular sand, non-plastic silt, and high plasticity clay. Shear wave velocity measurements will be made throughout the consolidation and shear phases of each test.
If these findings are representative of dilatant or structured soils in general, then there are several important implications for engineering practice. A significant application of this concept would be the estimation of effective stress strength parameters directly from in situ shear wave velocity measurements in soils that have been traditionally problematic to sample and test in the laboratory, such as dense sands and silts or sensitive clays. Establishment of the proposed relationship would potentially change how geotechnical site investigations and laboratory testing program are designed. Another application would be that in situ shear wave velocity measurements, for example from cross-hole or seismic cone penetration tests, could be used as an early warning system for the onset of failure in cemented, structured, or sensitive soils during staged construction of embankments and natural slopes which fail progressively.
The main objective of this project was to study physical properties of dense soils using both destructive and non-destructive methods. In particular, we focused on linking the small strain stiffness of these soils with their strength at large strains. The small strain shear modulus was obtained by measuring the shear wave velocity of soil samples in the laboratory non-destructively, and from that we calculated the small strain shear modulus using elastic theory. The small strain shear modulus is an important soil parameter for estimating the response of soils to earthquakes, and it is influenced by the density, stress state, fabric and other factors of the soil. The strength of soil can be measured by many means in the laboratory, all of which are destructive to the samples. For this study we performed drained triaxial tests, in which an all-around pressure is applied to cylindrical samples of soil to simulate stresses in the ground, and then the samples are compressed vertical until the soil fails. We hypothesize that there is a unique relationship between the small strain shear modulus (G0) and the effective stress at failure (i.e. strength, σ'1f) of dense or dilative soils. The relationship between these two parameters (one non-destructive, the other destructive) is believed to be constant (G0/σ'1f= M) as long as the specimens exhibit dilative (or dense) behavior. To test this hypothesis, samples of different types of soils (sand, silt, and clay) were prepared at different densities and stress states, and both the small strain shear modulus and strength was measured. The results strongly supported the proposed hypothesis. As long as the samples were dense (or in the case of the clay samples, over-consolidated), there was a clear relationship between the small strain shear modulus and the effective stresses at failure (i.e. strength) for dilative soils. This finding may have important implications for understanding the behavior of soils that are difficult to sample in the field. Applications may include in situ monitoring of sensitive soils during construction and the estimation of the strength of dense sands and gravels in situ.
