The Influence of Confining Stress on Liquefaction Resistance of Uniform Ottawa C109 Sand and Uniform Pea Gravel

It is shown by several studies that soils become less resistant against liquefaction as effective overburden stress increases at the same cyclic stress ratio (CSR) level. Therefore, the correction factor Kσ is used to refer to the normalized resistance to liquefaction at an initial effective overburden stress of 1 atm, which is commonly represented in the empirical field-performance database. In other words, the correction factor Kσ is utilized to adapt empirical liquefaction resistance charts for confining stresses smaller or especially greater than 1 atm. The influence of high confining pressure on liquefaction resistance has received limited attention compared to other factors, even though its impact can be significant and critical in certain cases. This gap is even more pronounced for gravelly soils, which have fewer documented liquefaction case histories compared to sands and silty sands, leaving the effect of high confining pressure on their liquefaction resistance less explored in the literature. This study presents results from Large Scale Cyclic Simple Shear tests (12-in. diameter) conducted on two subrounded materials as uniform sand (Ottawa C109 Sand, OS) and uniform gravel (Pea Gravel, PG). The tests were performed on reconstituted specimens at two after consolidation density levels: loose (DR ≈ 45%) prepared using air pluviation, and dense (DR ≈ 85%) prepared using dry tamping, under various shear loading and confining stress conditions as 1, 2.5, 4, and 5.5 atm. Several studies have shown that many gravels experience a greater reduction in cyclic strength under high confining pressures compared to sands, suggesting lower Kσ values for gravels. On the other hand, this study shows that the uniform gravel experiences cyclic strength increase with increasing confining pressure as opposed to uniform sand. Hence, one should always be careful while applying the Kσ correction to liquefaction resistance of gravelly soils, and specific empirical and/or semiempirical Kσ estimation plots and equations should be developed for gravels as the response might be different than similar sands.