Liquefaction Susceptibility Criteria for Silts and Clays

TLDR

New liquefaction susceptibility criteria for saturated silts and clays are presented, based on stress‑strain mechanics, to improve guidance for estimating potential strains and strength loss during seismic loading. The criteria predict clay‑like behavior for fine‑grained soils with PI ≥ 7 (or PI ≥ 5–6 for CL‑ML soils), providing a practical transition point consistent with available data. Test data show that silts and clays transition over a narrow PI range from sand‑like to clay‑like behavior, and the authors recommend reserving the term liquefaction for sand‑like soils while using cyclic softening failure for clay‑like soils, noting practical application issues.

Abstract

New liquefaction susceptibility criteria for saturated silts and clays are presented that are based on the mechanics of their stress-strain behavior and which provide improved guidance for selecting engineering procedures for estimating potential strains and strength loss during seismic loading. Monotonic and cyclic undrained loading test data for silts and clays show that they transition, over a fairly narrow range of plasticity indices (PI), from soils that behave more fundamentally like sands (sand-like behavior) to soils that behave more fundamentally like clays (clay-like behavior), with the distinction having a direct correspondence to the type of engineering procedures that are best suited to evaluating their seismic behavior. It is recommended that the term liquefaction be reserved for describing the development of significant strains or strength loss in fine-grained soils exhibiting sand-like behavior, whereas the term cyclic softening failure be used to describe similar phenomena in fine-grained soils exhibiting clay-like behavior. For practical purposes, clay-like behavior can be expected for fine-grained soils that have PI⩾7, although a slightly lower transition point for soils with a CL-ML classification (perhaps PI⩾5 or 6) would be equally consistent with the available data. Issues related to the practical application of these criteria are discussed.