Untitled Document

Origin and composition of ore-forming fluids in the giant Golden Mile gold deposit, Kalgoorlie Western Australia

C.J.Heath, I.H. Campbell, M.Palin and W.J. Dunlap


Australia is the worldÕs third largest gold producer, and the Golden Mile AustraliaÕs most significant gold deposit. Despite producing more gold both historically and currently than any other deposit in the country, the Golden Mile remains geologically poorly understood. Located in the Archaean Yilgarn craton of Western Australia, the Golden Mile is fundamentally different in size, metal budget, structure and mineralogy from the many other lode gold deposits in this highly mineralised region. The aim of this study is to investigate the origin and composition of the ore bearing fluids responsible for producing this unique gold deposit.

Recent work at the Research School of Earth Sciences has focused on stable isotope and geochronology studies of the Golden Mile. Fifteen key samples spatially distributed throughout the deposit were selected for this purpose. Ten of the samples consist of vanadium sericite from tellurian rich ores, and 5 consist of sericite from sulphurous auriferous rich ores. Prior to mineral separation thin sections were cut and analysed by Electron Microprobe to determine major element geochemistry. Vanadium content of the V-sericites varied from 0.78 - 5.32 Wt %, and potassium from 6.75 - 8.41Wt %. Mineral separation was undertaken using a centrifuge heavy liquid technique. Average grainsize of the sericite separates obtained from the 15 samples was 26 mm ± 11 mm 1s, with purity at approximately 98% sericite. Quartz and pyrite were also separated from the whole rock samples at the same time as the sericite.

Previous geochronology at the Golden Mile remains ambiguous. Kent and McDougall (1995) dated two muscovite samples using the 40Ar/39Ar method, giving ages of 2628 ± 13 Ma, and 2630 ± 13 Ma for auriferous mineralisation. Kent and McDougall (1995) suggested that the Golden Mile samples should be considered minimum estimates due to evidence of hydrothermal activity post Au mineralisation, and that two samples cannot be considered truly representative of such a huge mineral deposit. Witt et al., (1996) proposed that the muscovite dated by Kent and McDougall (1995) may not represent the mineralisation event, and that the muscovite is potentially reset. For this reason the V-sericite samples analysed by this study are ideal candidates for 40Ar/39Ar geochronology, as they are directly related to the gold mineralisation event at the Golden Mile. Prior to 40Ar/39Ar analysis the samples were analysed using the K/Ar method to test for sample recoil during the 48 day irradiation period.

The fifteen sericite samples used for 40Ar/39Ar geochronology were also analysed for stable isotopes. Stable isotope analysis was employed to determine the source of the ore bearing fluid. Preliminary deuterium / hydrogen stable isotope analysis were undertaken on a subset of the 15 samples at the United States Geological Survey in Denver, Colorado. It is expected that the remaining D/H analysis will be completed in early 2002. Likewise, the 15 samples will be analysed for oxygen stable isotopes in early 2002. Quartz - sericite mineral pair d18O data will be used to provide temperature constraints on lode formation. When coupled with the D/H data, the oxygen data will also be used to help predict the source of the ore bearing fluid.

The final component of this study of the ore bearing fluid in the Golden Mile will consist of fluid inclusion analysis. Fluid inclusions plates have been constructed from the 15 key samples, as well as others. Proposed fluid inclusion studies include paragenetic, microthermometric, geobarometric and compositional analysis. As well as conventional analytical techniques, such as heating freezing, it is proposed that an infra-red and laser raman microprobe study be undertaken on the fluid inclusions.