CONSIDERATIONS IN DEVELOPING AN EMPIRICAL STRENGTH CRITERION FOR BIMROCKS

The strength of geological materials is one of the fundamental input parameters used in the design of civil and mining engineering works; including those projects to be constructed in complex geological mixtures or fragmented rocks such as melanges, fault rocks, coarse pyroclastic rocks, breccias and sheared serpentines. These often chaotic, mechanically and/or spatially heterogeneous rock masses are composed of relatively rock inclusions surrounded by weaker matrix, and maybe be considered bimrocks (block-in-matrix-rocks). It is almost always impossible to prepare standard core samples from bimrocks in order to perform laboratory studies. Therefore, for these rocks, determination of the mechanical parameters such as cohesion, friction angle and uniaxial compressive strength is extraordinarily challenging. Although there is sparse literature describing empirical and laboratory studies on bimrocks, there is no widely accepted empirical approach among the rock mechanics community due to the limitations of the existing empirical equations, which were developed largely for more tractable, relatively homogenous rock masses. In this study, an exhaustive database was developed by literature overviews and laboratory studies. Artificial bimrocks were also prepared in the laboratory for uniaxial and triaxial compression testing. Plaster of Paris, bentonite, cement and water were mixed in different ratios to fabricate matrix types with various strengths. In addition, real tuff and andesite blocks, fragmented to centimeter sizes to create blocks, were mixed with the matrix materials to create artificial bimrocks. Uniaxial and triaxial compression tests were conducted on specimens of pure matrix and artificial bimrock mixes having different block proportions. Finally, a series of statistical regression analyses were applied to the results of the laboratory strength tests to develop an empirical approach for estimating the overall strength of bimrock mass, by incorporating the Mohr-Coulomb strength model and the Hoek-Brown empirical criterion, both of which are widely used in rock engineering. The empirical approach based on the Hoek-Brown empirical equations was found to yield a slightly better predictive performance than an empirical approach based on the Mohr-Coulomb.