Induced partial saturation method for soil liquefaction mitigation in large scale shake testing
MetadataShow full item record
Due to the hazards of soil liquefaction and limitations of traditional liquefaction mitigation approaches at sites with existing superstructures, this research focuses on investigating a new liquefaction mitigation method--Induced Partial Saturation (IPS) which reduces the degree of soil saturation by introducing small amounts of gas (oxygen bubbles) into liquefaction-susceptible sand making the sand partially saturated, an idea initially developed by researchers at Northeastern University. A large-scale 1-g geotechnical laminar box (6m height, 2.75m wide, 5m long) which can simulate the response of fully saturated sand and partially saturated sand up to 20ft (6m) depth subjected to base shaking has been developed by the Structural Engineering and Earthquake Simulation laboratory (SEESL), University at Buffalo. The laminar box system consists of 39 laminates vertically stacked together, shaking base, computer-controlled high-speed actuators, strong floor, advanced instrumentations, and a laboratory hydraulic filling for placing sand. One large-scale fully saturated sand experiment (LG-1) and two large-scale partially saturated sand experiments (IPS-1 and IPS-2) were performed with two different treatment methods, involving nearly 16ft (5m) deep sand deposit, and the results are presented henceforth. One of the latter two experiments involved partially saturating the top 8 feet of sand (IPS-1), and the other involved partially saturating the entire deposit of sand in the laminar box (IPS-2). Shake tests were conducted on the laminar box holding the loose sand deposit. Accelerometers, temposonics, potentiometers, piezometers, conductivity probes and cameras were used to observe and record soil response during and after the conclusion of shaking motion. Cone Penetration Tests (CPT) were performed to assess the geotechnical engineering properties of soil. A comparison between fully saturated sand responses, partially IPS treated induced partial saturation sand response and fully IPS treated induced partial saturation sand response reveal the influence of IPS on soil behavior due to strong shaking. Liquefaction, significant sand boils, and ground settlement occurred when fully saturated sand was subjected to a shaking base motion (LG1). Soil did not liquefy in the treated zone (top 8ft of deposit) in IPS1 shake tests although limited vertical settlement occurred. Fully saturated sand (LG-1) liquefied at all depths in all the LG-1 tests except at depths close to 12'-16' for the later tests where loose sand densified due to repeated shaking. Fully treated sand in IPS-2 did not liquefy at any depth under similar base shaking motion and pore pressure transducers revealed a delay in excess pore water pressure development with respect to time, when compared to excess pore water pressure generation in fully saturated sand (LG1). The partially treated sand in IPS-1 did not liquefy in the treated zone (8ft from the free surface of sand) and exhibited similar excess pore water pressure response. Further analysis of the work are presented in three other separate PhD dissertations at Northeastern University and another MS thesis report at University at Buffalo. The experiment results presented demonstrate that the IPS method can be further improved as an effective method to reduce the soil liquefaction potential and reducing liquefaction-induced ground settlement especially useful at sites with pre-existent superstructures already constructed.