Early gangue rejection or metal preconcentration at coarse scale (millimetres) based on size has been identified as a feasible operating alternative whereby energy efficiency and unit metal productivity can be greatly increased. This is achieved by understanding and exploiting ore-specific preferential grade by size responses. Preferential grade by size refers to the propensity of some ores to naturally concentrate metal into specific size fractions during breakage. The magnitude of metal deportment is described through a Ranking Response parameter (RR). This parameter has been used to measure the extent of “liberation at coarse scale”. Mineral Liberation is defined as the measurable rock property that can link with a downstream separation technique which aims to concentrate valuable material to produce a saleable product. Liberation traditionally has been defined at grain scale whereby the efficiency of processes such as flotation is greatly dependent on particle properties at micro scale (microns). However, in size-based coarse separation the efficiency relies on having a processing stream with a strong grade variability across size fractions (i.e. high grade by size response) and therefore a high RR value.
This work aims to develop a model to predict preferential grade by size response, in terms of the RR of ores as a function of particle size distribution and size reduction process. To achieve these aims a novel methodology has been developed comprising a new preferential grade by size characterisation method coupled with Monte Carlo and comparative statistical methods (analysis of variance (ANOVA) and t-test). Six run of mine (ROM) bulk samples from 3 different geological style deposits (stock work vein hosted, Cu–Mo breccia porphyry and Cu–Mo volcanic porphyry) have been utilised in the analysis.
This methodology provides useful insights for the development of an optimum coarse separation circuit flowsheet design for preconcentration prior to energy intensive and inefficient grinding.
This paper may be located here.