Liquefaction Susceptibility and Cyclic Response of Intact Nonplastic and Plastic Silts

This study presents the results of a laboratory test program that serves to improve the understanding of the liquefaction susceptibility and cyclic response of intact silts that span sand- and clay-like behaviors. Specimens were prepared from samples characterized with a plasticity index (PI) ranging from 0 to 39, fines content (FC) ranging from 29% to 100%, and overconsolidation ratio (OCR) ranging from 1.0 to 4.2, retrieved from five silt deposits in Western Oregon and Southwest Washington. The roles of PI, FC, and OCR on the 1D compression and monotonic and cyclic strength of nonplastic to plastic silts are identified. Hysteretic metrics proposed to quantify cyclic behavior provided an objective means to distinguish between qualitative judgments of sand-like, intermediate, and clay-like behavior. Prior soil index test-based liquefaction susceptibility criteria exhibited good to poor accuracy; modifications to existing criteria aligned with quantified hysteretic behavior, which together indicate that sand- and clay-like behavior is subject to the intensity and duration of cyclic loading. The variation of cyclic resistance ratio (CRR) and cyclic strength ratio, ${\tau }_{\mathrm{cyc}}/s_{u,\mathrm{DSS}}$, with the number of loading cycles, $N$, to reach single amplitude shear strain, $\gamma$, of 3% and 3.75% is presented. The ${\tau }_{\mathrm{cyc}}/s_{u,\mathrm{DSS}}$ for $N_{\gamma =3\%}=10$ and 30 appeared constant for $\mathrm{PI}\le 11$ and $\mathrm{PI}\ge 18$ and equal to 0.63 and 0.54, and 0.82 and 0.76, respectively, with an apparent linear trend for $11<\mathrm{PI}<18$. Despite higher void ratios, intact overconsolidated specimens exhibited greater CRR than their mechanically normally consolidated counterparts, highlighting the effects of OCR and natural soil fabric on cyclic resistance. Cyclic tests conducted on specimens consolidated using a quasi-stress history and normalized soil engineering properties (SHANSEP) method exhibited larger CRR than those tested using the recompression method, which is attributed to the smaller void ratios and potentially greater lateral stresses. The recompression technique is preferred for establishing the cyclic response to capture in-situ conditions when testing high-quality samples and where quantification of the preconsolidation stress is uncertain.