Three-Dimensional Stress-Strain Response and Stress-Dilatancy of Well-Graded Gravel

Abstract

The three-dimensional (3D) stress-strain response of uniform sands has been the focus of extensive laboratory investigations, resulting in well-established understanding of their 3D stress-strain-strength behavior. However, the applicability of stress-dilatancy theories and 3D stiffness, strength, and volumetric responses have not been sufficiently evaluated for well-graded fill soils commonly used in design. A series of drained true-triaxial tests at three levels of confining stress on specimens of well-graded Kanaskat gravel were conducted to address pertinent questions regarding this behavior. Constant mean effective stress paths during shearing corresponded to triaxial compression (TC), simple shear (SS), and triaxial extension (TE). Previously reported results from plane strain quasi-K consolidated (PSK) stress paths on the same material are incorporated and the general behavior suggest that 3D stress-dilatancy and frictional responses differ from those of uniform soils. The secant shear modulus at an initial stage of shearing was found to be stress path dependent with TC specimens being the highest. However, the stiffness at a more advanced stage of shearing was found to be relatively independent of the stress path and fitted power laws adequately capture pressure-dependent stiffness. Friction and dilation angles at failure are interpreted with respect to experimental boundary conditions and are significantly higher than those measured in the TC stress path. However, properly calibrated stress-dilatancy theories and 3D failure criterion are sufficiently capable of capturing this behavior. Fitting parameters are provided for a commonly used stress-dilatancy approximation and two 3D failure criteria for the soil investigated here. The data presented here should help those seeking to estimate the 3D response of a well-graded gravelly soil.