Field properties of cemented paste backfill at the Golden Giant mine

Cemented paste backfill (CPB) has been used for almost a decade in the mining industry and is gaining popularity world-wide. However, its design is largely based on material that is prepared, cured, and tested in the laboratory environment. Replicating the field mixing, placement and curing processes in a laboratory is difficult, and there are questions as to how representative the laboratory material is of the actual field material. Anecdotal evidence suggests that the CPB sometimes under performs, as evidenced by excessive sloughing of the exposed paste wall, and sometimes over performs, as suggested by the stable excavations in the pastefill. Only by understanding the field performance of cemented paste backfill can the design be optimised while ensuring safety. A field investigation of the Golden Giant mine's cemented paste backfill was undertaken to quantify the in situ properties and to provide the data needed for mix design optimisation. The investigation was two-pronged - in situ testing using a self-boring pressure meter, and testing of undisturbed samples of CPB. The investigation showed that the bulk properties of the in situ backfill are more variable than laboratory-prepared samples and, on average, tend to have a higher void ratio and lower degree of saturation. Field strengths derived from both the self-boring pressure meter (SBP) and triaxial testing are variable but consistently higher than the laboratory samples and this may be attributed to a higher cohesion developed in the field CPB. These results suggest that the current backfill design at the Golden Giant mine may be conservative. The self-boring pressure meter also provides an indication of the in situ stresses at the individual test locations. Together, the stress measurements provide an indication of the overall stress distribution in the test stopes. The results suggest that there is a complex interaction of self-weight, stress arching and post-placement mining-induced stress influencing the stress distribution in a backfilled stope. The measured stresses fall within the range predicted by simple, self-weight calculations and numerical modelling that considered stress arching and mining-induced stresses.