Quantifying excavation-induced rock mass damage in large open pits

Abstract

Open pit excavation induces rock mass damage as a result of stress release and blasting. The damage reduces the rock mass strength and stiffness. The strength reduction is typically quantified by the Hoek–Brown disturbance factor (D). Limitations of this approach are outlined in the paper. It is proposed that degradation of rock mass strength and stiffness results from damage followed by disturbance. Damage is caused by fracturing of intact rock (e.g. failure of rock bridges) that results from small strains (less than approximately 2 percent) and only increases porosity slightly. Disturbance is caused by rearrangement of intact rock blocks that results from large strains (perhaps 10 percent) and usually produces a large increase in porosity. The strength degradation produced by damage is quantified by the peak and post-peak strengths, and the strength degradation produced by disturbance is quantified by the residual strength. When using numerical models to study the progressive failure of rock masses, estimates of the damage characteristics are required. In many of these models, the Hoek–Brown failure criterion is treated as a yield criterion and the analysis is carried out using plasticity theory. The work described here is aimed at defining the appropriate evolution of rock mass strength as applied to slope stability studies. We do this by studying the stress-strain response of 3D bonded block models with low-porosity structures that range from blocky to disintegrated and a highporosity structure that mimics rockfill. The synthetic peak and post-peak strength envelopes are compared with Hoek–Brown strength envelopes with D factors of zero and one. The comparisons support the following conclusions. Peak and post-peak strength are similar for lower quality (Geological Strength Index less than approximately 40) rock masses. Hoek–Brown strengths for D = 1 are lower than post-peak strengths, with the difference being more pronounced for lower quality rock masses. For low quality rock masses, the Hoek–Brown strength for D = 1 may be lower than the residual strength. In the Hoek–Brown approach, as the D factor increases, the behaviour is cohesion weakening friction weakening; however, our work suggests that the behaviour for higher quality rock masses is cohesion weakening friction strengthening. This statement applies to the synthetic material, and laboratory evidence suggests that it also applies to real rock masses.