Soil liquefaction and post-liquefaction behavior of ground and its affecting factors

Abstract

Soil liquefaction refers to a phenomenon wherein a saturated soil loses strength in response to a dynamic load, usually earthquake shaking causing to increase in pore water pressure. Increase in pore water pressures is the result of rapid loading situation during seismic events where there is not enough time for dissipation of excess pore water pressures through natural drainage. This phenomenon and subsequent ground settlements which causes large deformation of soil layers, may seriously damage the structures. Since Niigata and Alaska earthquakes in 1964, more researches focused on the basic mechanism and various aspects of liquefaction and associated problems. The present study attempts to investigate the soil liquefaction and subsequent ground deformations that may seriously damage the structures. The primary factors affecting the soil liquefaction will be discussed in order to quantify their effect on the liquefaction and post-liquefaction behaviour of the ground. There are various factors affecting the soil liquefaction. Among them, the packing density and void ratio can be mentioned, that are key mechanical characteristics of soil in geotechnical engineering. Previous experimental studies on sand-silt mixtures have shown that silty fines significantly affect the soil fabric and resulting mechanical behavior. Some bilinear models have been proposed to capture minimum and maximum void ratios. In Chapter 2, previously-reported data from 37 soil mixtures with 338 individual specimens for which the minimum void ratios are available were studied. The results helped form the basis to develop a nonlinear model capable of estimating the shape of the minimum void ratio curve of the sand-silt mixtures. The equations of minimum void ratios were derived for soil mixtures in coarse and fine grain dominant regions. The results show that the predicted values of minimum void ratios by quadratic polynomial equations agree well with the measured data; the root mean square error (RMSE) is 0.03. The proposed model only requires one data from the experimental test to predict the minimum void ratio for any soil mixture with various fine contents. There are some mathematical models that predict the limit void ratios, however, the results obtained through these models have shown some discrepancies with the experimentally measured limit void ratios; in particular, the model predictions were inaccurate around the threshold fines content in case the limit void ratio is minimum. Previous studies revealed that in a sand-silt mixture, the parameters affecting the maximum void ratio are almost same as those of the minimum void ratio. To overcome this issue, Chapter 3 extends the mathematical model for estimating minimum void ratios of soil mixtures to predict the maximum void ratios of soil mixtures as a function of fines content. The extended model is verified by data from 37 soil mixtures with various soil samples. The predictions fitted well with the experimental results, whether there were small or big reductions in the maximum void ratios and with various fines content for both coarse- and fine-dominant regions with the root mean square error equal to 0.043. Moreover, 44 tests were conducted on 22 different mixed graded soil samples from the Nakdong River soil in Busan to obtain the minimum and maximum void ratios of soil mixtures at different fines content. The predictions of the test results for the minimum and maximum void ratios via the model confirmed the applicability of the proposed nonlinear model. In Chapter 4, a SPT-N based investigation is carried out to assess the susceptibility of liquefaction in Eco-Delta city, located in the southwestern part of Busan city in South Korea. Data from 229 sites are analyzed for the earthquake of 7.5 magnitude with a peak horizontal acceleration of 0.2g to carry out the liquefaction potential index (LPI) through two deterministic methods which have different factors of safety (FS). The liquefaction probability is investigated by the deterministic and reliability methods and the liquefaction hazard maps are generated. To observe the effect of fines content and plasticity index on the liquefaction susceptibility, three different cases are considered. It is found that among the four approaches used, Overseas Coastal Area Development Institute of Japan (OCDI) method showed more sensitivity to changes of fines content and plasticity index. The Eco-Delta city is found to be highly vulnerable to liquefaction having 91% of sites with LPI values greater than 15. A solidification (compaction) process always occurs after liquefaction due to the dissipation of excess pore water pressure and causes permanent settlement in the ground. In Chapter 5, data from Nakdong River soil in Busan, South Korea is used to simulate 53-m-deep free-field ground with a 10-m-deep layer of liquefiable sandy soil. A sensitivity analysis was performed by subjecting three different soil densities to four different sinusoidal motions with different amplitudes and frequencies to investigate the differences in the outcome. The excess pore water pressures at various depths and settlements were estimated in 12 analysis cases. Compared to the initial settlement due to liquefaction, significantly greater settlement due to the dissipation of the excess pore water pressure and solidification was found (78% of the final settlement in the weakest motion case), which should be considered in liquefaction analysis. By regression analysis of the results, an exponential function of normalized depth is proposed to accurately estimate the final settlement after dissipation of excess pore water pressure based on the motion properties at any particular depth.