The possible load plate rigidity and scale effects on the frictional behavior of the interface between a quartz sand and high-density polyethylene (HDPE) geomembranes was investigated. The test materials included a quartz sand against smooth and textured HDPE geomembranes. The American Society for Testing and Materials D-5321 standard test method is adopted for measuring the shear strength of the interface between the test soil and geomembranes. The dimensions of the shear boxes are 100 x 100, 200 x 200, 300 x 300 and 400 x 400 mm(2). In addition to the use of conventional rigid load plates to transfer applied load to the test specimens, a silicon-membrane-type water bladder was also designed to attach at the bottom of the conventional rigid load plate to form a flexible load plate. Moreover, a measuring device was used to replace the lower shear box for measuring the variation of normal pressures at the shearing plane. The results of the study indicated that the use of conventional rigid load plate to perform direct shear test results in a concave pressure distribution at the shearing plane. A minimum of 5 cm thickness of soil in upper shear box is recommended in order to ensure the reproduction of normal pressures at the shearing plane when a rigid load plate is used for applying the normal load. The use of flexible load plate for large-scale direct shear test machine would provide an anticipated uniform pressure distribution at interface shearing plane. A thickness of 2-3 cm of sand layer in the upper shear box should be enough if loading is applied using a flexible load plate. The shear resistance associated with flexible and rigid load plate is quite similar to each other under same test conditions. However, the direct shear test performed using flexible load plate showed a better reproducibility of test data than that using rigid load plate. The size of shear box has an effect on the shear strength at the interface of sand/geomembrane systems. In general, the shear strength for larger shear box is greater than that associated with smaller shear box. However, the shear strength associated with 300 x 300 and 400 x 400 mm(2) shear boxes is almost identical to one another. Thus, it is recommended that minimum dimension of 300 x 300 mm(2) direct shear box should be used for determining the shear strength at the interface of a sand/geomembrane system. The shearing resistance associated with textured HDPE geomembrane is greater than that associated with a smooth-surfaced geomembrane. The shear strength associated with textured HDPE geomembranes increases with increasing shear strain. No peak values were observed. Stress-strain curves with peak values only occur for sand/smooth geomembrane system under high normal stress conditions. The friction angle for sand/textured geomembrane combinations is about 12degrees greater than that for sand/smooth geomembrane combinations. (C) 2002 Elsevier Science Ltd. All rights reserved.
