Quantitative Mineral Mapping – A Building Block for Geometallurgy at Olympic Dam

Two irrefutable facts in geology and metallurgy are: ore deposits are comprised of variable mixtures of minerals which are systematically distributed throughout the deposit, and mineralogy exerts the primary control on metallurgical performance. If the distribution of process critical minerals can be quantified and mapped, then a logical assumption is that metallurgical performance can be predicted across the deposit.Olympic Dam is a +9 billion tonne Fe-oxide Cu-U-Au-Ag deposit which is hosted entirely within a magmatic-hydrothermal breccia complex consisting of over seventy ore and gangue minerals. Each has a unique set of metallurgical performance properties. Fortunately, 15 of the 70 minerals accounts for >98 per cent of the total rock mass.The mineral mapping journey commenced in the early drilling days shortly after the discovery of Olympic Dam in 1975. Geologists quickly recognised that the copper-iron sulphides had a predictable zoned distribution. However, geological models reflected the presence or relative abundances, not quantitative concentrations, of the minerals. This type of mineral mapping remained relatively unchanged for ~20 years and was of limited use outside of geology.Several things changed during the mid-1990s: assaying costs decreased significantly, instruments capable of automatically measuring mineralogy became readily available, the recognition of two quantifiable relationships at OD (ie mineralogy is a function of chemical composition (ie assays), and metallurgical performance is a function of assays, mineralogy, texture and process conditions. The linkage of metallurgical performance to mineralogy and mineralogy back to assays allows for every assayed sample in the resource model to be assigned mineral concentrations and metallurgical performance predictors. At Olympic Dam, block models (ie 3D maps) are routinely produced for 27 elements, density, 15 process critical minerals and greater than 50 metallurgical performance variables. These geometallurgy-enabled resource models are used for mine planning, process plant design and process optimisation studies. *This is an abstract only. No paper was prepared for this abstract.*