This research consists of full-scale shake table tests to investigate liquefaction of sandy soils. Consideration of the potential and consequences of liquefaction is critical to the performance of any structure built in locations of high seismicity underlain by saturated granular materials as it is the leading cause of damage associated with ground failure. In certain cases the financial losses associated with liquefaction can significantly impact the financial future of an entire region. Most liquefaction triggering studies are performed in the field where liquefaction has been previously observed, or in tabletop laboratory testing. The study detailed herein is a controlled laboratory test performed at full scale to allow for the measurement of field-scale index testing before and after cyclic loading. Testing was performed at the Parson’s geotechnical and Earthquake Laboratory at Cal Poly San Luis Obispo on the 1-dimensional shake table with a mounted flexible walled testing apparatus. The testing apparatus, originally constructed for soil-structure interaction experiments utilizing soft clay was retrofitted for the purpose of studying liquefaction. This research works towards comparing large-scale simple-shear liquefaction testing to small-scale simple-shear liquefaction testing of a #2/16 Monterey sand specimen. The bucket top was modified in order to apply a vertical load to the soil skeleton to replicate overburden soil conditions. Access ports were fitted into the bucket top for instrument cable access and to allow cone penetration testing before and after cyclic loading. A shear-wave generator was created to propagate shear waves into the sample for embedded accelerometers to measure small strain stiffness of the sample. Pore-pressure transducers were embedded in the soil sample to capture excess pore water pressure produced during liquefaction. Displacement transducers were attached to the bucket in order to measure shear strains during cyclic testing and to measure post-liquefaction volumetric deformations. The results of this investigation provide an empirical basis to the behavior of excess pore water production, void re-distribution, shear wave velocity, shear strain and cone penetrometer tip resistance of #2/16 Monterey sand before, during, and after liquefaction in a controlled laboratory environment at full-scale.

Although clays are generally considered stable materials under seismic conditions, recent failures initiated in clay layers after earthquakes have emphasized the need to study the cyclic and post-cyclic behavior of these materials. Moreover, if strength loss as a result of cyclic loading were to occur in the material comprising the dam and/or dam foundation, the consequences of failure could be substantial. The objective of this study is to evaluate the effect of plasticity characteristics, mineralogical composition, and accumulated energy on the cyclic behavior, post-cyclic shear strength and the degradation in shear strength due to cyclic loading in normally consolidated clays. Seventeen soil samples prepared in the laboratory from kaolinite, montmorillonite, and quartz were tested using static and cyclic simple shear apparatuses. In addition, the results of cyclic simple shear tests on twelve natural samples were provided by Fugro Consultants, Inc. in Houston, TX. Using the results, cyclic strength curves were developed to represent 2.5%, 5% and 10% double amplitude shear strains. These curves were used to examine the influences of mineralogical composition, plasticity characteristics and shear strain on the cyclic resistance of soil samples. A power function was used to represent the cyclic strength curves. The samples were found to become increasingly resistant to cyclic loading as the plasticity index increased. Moreover, the soils with montmorillonite as the clay mineral were noted to have consistently higher cyclic resistances than the soils with kaolinite as the clay mineral. By examining the power functions, it was found that the cyclic strength curve approaches linearity as the plasticity index increases in soils having kaolinite as the clay mineral. However, the opposite trend is observed in soils having montmorillonite as the clay mineral. The study shows that the post-cyclic shear strength increases with increasing plasticity index. Moreover, the post-cyclic shear strengths of soils with montmorillonite as the clay mineral were significantly higher than the post-cyclic shear strengths of soils with kaolinite as the clay mineral. The degradation in shear strength due to cyclic loading appeared unaffected by mineralogy, but a greater reduction in strength was noted with decreasing plasticity index. The post-cyclic shear strength was also found to reduce as the number of cycles required to cause 10% double amplitude shear strain increased. The energy approach considering the accumulated energy per unit volume in the soil mass as a result of cyclic loading was also utilized in this study. The results from the energy approach were independent of the cyclic wave form, but were still dependent on the amplitude of the cyclic load used during the testing. An increase in the amplitude of the cyclic loading function results in a decrease in the accumulated energy per unit volume. Furthermore, an increase in the liquid limit and/or plasticity index of the soils containing kaolinite as the clay mineral shows an increase in the accumulated energy, whereas an increase in plasticity of the soils containing montmorillonite as the clay mineral results in a decrease in the amount of accumulated energy. In both types of materials, the amount of accumulated energy per unit volume is found to increase with increasing double amplitude shear strain. Relationship between the ratio of post-cyclic undrained shear strength to the baseline undrained shear strength and the accumulated energy is also determined. / Ph. D.