Laboratory experiments on small-scale freezing sampling method for sand

The need to collect high-quality undisturbed samples to accurately determine in-situ liquefaction resistance of sandy soils in laboratory tests is widely recognized. Decades ago, the freezing sampling method was developed to minimize sample disturbance by pre-freezing the ground. However, due to the large scale and high cost of the work, this method is rarely used in common geotechnical surveys. To address this issue, a small-scale freezing sampling method has recently been developed. It reduces the sampling time and amount of required coolant, and can be applied to conventional borehole surveys. On the other hand, the quality of the samples collected by this new method has not been properly assessed. In this study, therefore, the appropriate conditions for inserting the freezing tube with the help of a cylindrical sand column were discussed, and the small-scale freezing sampling method was successfully implemented in a calibration chamber, whereby hollow cylindrical frozen samples with an outer diameter of 100 mm were collected within approximately 40 min and using only about 70 kg of liquid nitrogen. The focus of the quality assessments of the collected frozen samples was placed on the density and shear wave velocity measurements. The results showed that the changes in the small strain shear moduli were −0.5 % between the stages before and after the sampling, suggesting that the implementation of small-scale freezing sampling can be expected to improve the accuracy of in-situ liquefaction resistance estimates and, in turn, the accuracy of liquefaction assessments.