Effect of saturation procedures on direct simple shear testing of silt tailings

The direct simple shear (DSS) test carried out under constant volume (CV) conditions forms one of the primary laboratory techniques to characterise soils and tailings. The use of CV conditions to simulate undrained shearing is supported by historical evidence on the testing of a saturated clay and sands, with this evidence being incorporated into current guidelines and state of practice procedures. However, some recent comparisons of the results of undrained hollow cylinder simple shear (HCSS) and CV DSS tests on predominately silt gold tailings adopting state of practice test procedures (i.e. inundation of the sample after loose moist tamping) showed much less post-peak strength loss in the gold tailings than the undrained HCSS tests. The current study investigated this discrepancy further by carrying out DSS tests under high back pressures, undrained simple shear tests with flexible membrane and constant cell pressure and DSS tests after flushing with carbon dioxide and with use of a small back pressure. In all cases, the undrained tests or DSS tests with greater effort put towards saturation exhibited greater post-peak strength loss more consistent with the HCSS and the critical state line. The importance of these results on the estimation of tailings brittleness in engineering practice was outlined.