EXPERIMENTAL INVESTIGATION OF TIME-DEPENDENT EFFECT ON SHEAR STRENGTH PARAMETERS OF SAND-GEOTEXTILE INTERFACE

Various methods have been suggested for keeping geosynthetics in place during shearing including clamping [41] and gluing [42]. Clamping was avoided as it may increase the likelihood of a progressive failure, and thus reduce the measured peak interface shear strength [4, 41]. In the present study, two procedures were evaluated for tests. In the first approach, the sand was poured in a half shear test apparatus box, and the geotextile layer was located on it and the rest of the half box was then filled with the same sand. In the second approach, the geotextile layer was carefully adhered to a piece of rigid block with a thickness such that half of the shear test box was occupied. The other box half was filled with the sand, and the test was performed. The test results indicated that the results of the two configurations of tests had a very small difference. Therefore, for simplicity and in order to minimize the potential for any movement of the geotextile during shearing, the geotextile specimen was glued to a rigid block. The authors believe that the glue does not influence the compressibility and internal shear deformation of the geotextile since the glue does not penetrate into the geotextile, and it only covers one side of the surface of the geotextile. The viscosity of the glue was such that it did not penetrate into the voids of GT. The rigid block with glued geotextile was kept under compressive stress for one hour to ensure a proper bonding. The application of compressive stress for one hour also helped to reduce elongation of the geotextile during shearing and encouraged a sliding type of failure. The variation of ft/ft=0 or δt/δt=0 with time for only dry sand and dry sand-geotextile interface is presented in Fig. 5, respectively. Here ftand ft=0 denote the sand friction angle after elapsed time of t and that at no elapsed time, respectively. Characters δt and δt=0 represent respectively, the sand-geotextile interface angle at an elapsed time of t and that at no elapsed time. As observed, the increase in values of ft/ft=0 or δt/δt=0 is significant for the first two hours. Beyond this, the variation of ft/ft=0 or δt/δt=0 is very small. Also, the variation of δt/δt=0 with time for sand-geotextile interfaces is greater than that for the sand alone. As a whole, the values of ft/ft=0 or δt/δt=0 increase up to 2.5%, 3.3% and 3.9% at 720 minutes after the sample is poured in the mold for dry sand, dry sand-GTN.20 geotextile, and dry sand-GTN.50 geotextile