Suggested Method of Specimen Preparation for Triaxial Tests on Partially Saturated Sand

Abstract

In geotechnical engineering, triaxial testing is widely adopted to evaluate the mechanical behavior of sand. Methods of specimen preparation for triaxial tests on dry and completely saturated sand are well established in the literature, whereas very little guidance exists on the preparation of partially saturated sand at relatively high degrees of saturation (typically Sr > 80 %). The purpose of this study is to elucidate the suitable method of specimen preparation for partially saturated sand using sodium percarbonate and investigate the effects of partial saturation on the undrained cyclic behavior of sand. Loosely and densely packed specimens are prepared in a dynamic triaxial apparatus through dry pluviation and tamping of sand. For tests on fully saturated sand, dry specimens are flushed with carbon dioxide and deaired water and saturated applying back pressure. For tests on partially saturated sand, sodium percarbonate is used in either fine powder or aqueous solution form to create oxygen bubbles in the voids, reducing the degree of saturation of specimens. The results suggest that not only the degree of saturation but also the way partially saturated specimens are prepared affects the liquefaction resistance. At the same level of saturation and principal stress difference, specimens prepared with the aqueous solution exhibit higher liquefaction resistance than those prepared with the powder. The solution of sodium percarbonate proves to be a more reliable and repeatable technique for preparing partially saturated triaxial specimens with relatively high degrees of saturation.