Well-graded coarse-grained soils encompass a broad range of materials that are typically found in natural deposits and man-made fills. Current standard-of-practice estimates the behavior of these soils using sand-based engineering methods in liquefaction evaluations, assuming transferable strength, stiffness, stress-dilatancy, and liquefaction triggering behaviors. This generalization does not explicitly consider the effects of median grain size (D50) and gradation (Cu) on soil behavior, even though these parameters alter the soil fabric, index, and physical properties of the soil, and by extension their response to monotonic and cyclic loading. The aim of this work is to develop an experimental database: (1) to examine the role of gradation and grain size on the monotonic and cyclic response of coarse-grained soils, and (2) to evaluate the appropriateness of commonly used sand-based relationships to capture the mechanisms of well-graded soils.
A series of constant-stress (CS) and constant-volume (CV) monotonic direct simple shear (DSS) tests and cyclic CV-DSS tests were performed on four poorly- to well-graded quartzite soils but composed of materials from a single natural deposit to minimize the effect of other grain properties (e.g., mineralogy, deposition, particle shape). This experimental program involves a wide range of relative densities (20 – 75%), initial consolidation stresses (50 – 600 kPa), and static shear stresses (to target CSR levels of 0.07 to 0.18). Experimental results are used to identify and describe trends regarding soil strength, stiffness, dilative tendencies, and critical state under monotonic loading, and to develop liquefaction triggering resistance curves under cyclic shearing. Conclusions will provide an initial basis to assess the appropriateness and potential re-evaluation of using sand-based relationships to capture the static and dynamic mechanisms of well-graded, coarse-grained soils, with future work geared towards numerical simulations and advanced constitutive models to compare with the laboratory results from these direct simple shear tests.