Application of Experimental Design to Study the Fully Softened Shear Strength

ABSTRACT The use of the fully softened shear strength has been recommended when conducting slope stability analysis of first-time slides in stiff fissured clays. Several studies in the literature address the influence of soil index properties and effective stress in the fully softened shear strength. However, limited studies have been published regarding the simultaneous characterization of the effect of soil properties, stresses, testing devices, and sample preparation in such shear strength. In this study, the effects of factors’ interaction on the secant fully softened shear strength friction angle are analyzed, and preliminary confidence intervals to quantify the difference between the friction angles measured using different testing devices are given. For this purpose, a two-phased statistical design of experiments was developed to conduct 59 laboratory tests using kaolinite, Pierre shale, and Charmouth mudstone in direct shear (DS), ring shear (RS), and direct simple shear (DSS) testing devices, and using low to highly disaggregated sample preparation methods. The index properties, as well as the fully softened friction angles, were measured. Both the magnitude of the fully softened friction angle and the curvature of the failure envelope were significantly dependent on the confining stress and testing device. The friction angle is significantly affected by the interaction between those two factors. Secant friction angle differences as large as 11° between the DS and RS testing devices were found.