Strain based selection of friction angle for geotechnical design

Friction angle is the main shear strength parameter in case of cohesionless soils, which can be described as a function of critical state friction angle and dilatancy angle. When it comes to make a decision on selecting a design friction angle for geotechnical engineering applications, engineers either prefer to use peak friction angle accepting a smaller factor of safety or use the critical state friction angle as a design friction angle by neglecting the rewarding effects of dilatancy. This selection is a matter of safety concerns. However, in a project if it is predicted that the soil would not reach failure at any point within the geotechnical structure, then choosing peak friction angle as the design friction angle would be beneficial in terms of economy. Starting from this point, the goal in this study is to develop a method that would guide the engineer in the selection of an appropriate soil friction angle. For this purpose, it is attempted to quantify shear strain at failure as a function of soil properties. This way, it becomes possible to relate allowable structural deformations to magnitudes of possible shear strains, and then assess which friction angle is valid for the case at hand. Towards this goal, triaxial data of 8 different sands were investigated. These data was used to construct relationships between peak dilatancy angle (p) and shear strain (q) at which these peak dilatancy angles are measured. The reason for choosing this relationship is the inherent link between failure and peak dilatancy; peak dilatancy is always measured at the instance of maximum shear stress. Later, obtained p-q functions for all sands are plotted on the same graph. This graph revealed that p-q functions are dependent on the mean grain diameter (D50). As a result, a new method for assessing the shear strain limit for using peak friction angle as the design friction angle is proposed. This method uses D50 as input and calculates magnitude of shear strain at failure as a function of p. Combining this method with the equation proposed by Çinicioğlu et al. (2013) that calculates p as a function of in-situ stress and density of soil, it becomes possible to calculate shear strain at failure magnitude from the knowledge of in-situ stress, density and D50 values. As a result, comparing the shear strain at failure with the possible shear strains within the limits of structural deformations, designers can choose the appropriate friction angle for design.