The Effect of Initial Static Shear Stress on Liquefaction Triggering of Coarse-Grained Materials

Soil liquefaction response is significantly affected by soil gradation (particle size, angularity, coefficient of uniformity) and density. However, the literature on the factors affecting liquefaction resistance with initial static shear stress (e.g., sloping ground) is more limited and primarily based on clean, poorly graded sands. As a result, the influence of particle size and gradation on the liquefaction potential of soils with initial shear stress is overlooked. In this study, 223 large-size cyclic simple shear tests were conducted on poorly and well-graded sands and gravels to evaluate the effects of soil gradation on the liquefaction resistance with the presence of initial static shear stress. Sandy and gravelly soils with coefficients of uniformity ranging from 1.6 to 42 were tested in a large-scale cyclic simple shear device under constant volume conditions, and the initial static shear stress correction factor $K_{\alpha }$ values were obtained. The results show that poorly graded sand specimens exhibit flow liquefaction, have a more significant vertical effective stress reduction as the initial static shear stress increased, but also exhibit beneficial effects of initial static shear stress even if loosely packed, mainly due to their more dilative nature. Well-graded sandy soils, on the other hand, did not have as an abrupt loss of stiffness compared to poorly graded sand specimens, but due to their higher coefficient of uniformity may be more contractive, causing more pronounced shear strain development at the last few cycles. Gravel content also affected the void ratio of sand, which influenced the onset of strain softening or hardening during cyclic loading. Dense specimens with initial static shear stress exhibit cyclic mobility, but this may not necessarily provide beneficial effects of the $K_{\alpha }$ correction factor, especially for higher coefficients of uniformity. The experimental results suggest that the widely used $K_{\alpha }$ correction factor approaches that were originally suggested based on poorly graded sand may be overoptimistic for both loose and dense soils when considering a broader spectrum of soils such as those encountered in engineering practice. It is proposed that the $K_{\alpha }$ correction factor should consider not only relative density and initial static shear stress but also particle size and gradation (i.e., determining the gravel content and the coefficient of uniformity), as well as angularity.