A rigorous and systematic study on the effects of maximum particle size and sample scaling on the mechanical behavior of mine waste rock ( MWR) is presented. Materials tested were obtained from a similar location within the open pit of Barrick Gold Corporation's Goldstrike gold mine approximately 60 km northwest of Elko, Nevada. The collected field gradation was scaled using the parallel gradation technique in order to most accurately reproduce the fabric of the collected field material in the laboratory. The two MWR materials were shown to have drastically different mineralogy, which may be due to varying levels of in situ hydrogeologic weathering. Mechanical behavior of two MWR materials was systematically evaluated on 70-mm-diameter, 140-mm-tall specimens with a maximum particle size equal to 11.2 mm in drained isotropic and monotonic axisymmetric compression using a conventional triaxial device. Results from conventional triaxial tests in drained isotropic and monotonic axisymmetric compression were systematically compared to results presented by Fox (2011) for 150-mm-diamer, 300-mm-tall specimens with a maximum particle size equal to 25.0 mm for the same MWR materials also scaled using the parallel gradation technique and tested at the same levels of mean effective stress (p') in a large-scale triaxial device.

Intrinsic critical state parameters were defined for each material in conventional scale triaxial testing. Results suggest the parallel gradation technique provides a reasonable way to estimate values of intrinsic critical state parameters during isotropic compression in large-scale applications using conventional triaxial tests provided that the range of D _R,Initial is similar between scales. Values of the critical state friction angle (&phis;_c) were determined to be within 2° of the larger scale specimens. Dilatancy rates measured in conventional scale unweathered specimens was found to be, on average, approximately twice of that measured in the large-scale specimens of the same material. The weathered material indicated no significant changes in average dilatancy rates between scales. Bolton's (1986) fitting parameters Q and R were determined to decrease with decreasing particle size where Bolton's stress – dilatancy relationship was found to predict values of the peak friction angle (&phis;_p) more accurately using values of Q and R obtained in large-scale triaxial tests where conventional scale specimens yielded consistently unconservative predictions of &phis;_p suggesting that conventional triaxial tests should not solely be used to characterize the mechanical behavior of large-scale materials. Particle breakage measured after each triaxial test indicated a relatively constant shift from pre to post test particle-size distribution curves between conventional and large-scale specimens. Additional results and comparisons of values such as fractal dimension (D) and surface energy (Γ _se) are also discussed.