1 PACRIM 2015 Conference Paper Number: 166 High resolution characterisation of gold mineralisation at Plutonic Gold Mine, Western Australia: Evidence for the late- stage deposition of high-grade gold M.F. Gazley 1 , G. Duclaux 2 , L.A. Fisher 3 , R.M. Hough 4 , M.A. Pearce 5 1. GAusIMM, Research Scientist, CSIRO Minerals Resources Flagship, ARRC, PO Box 1130, Bentley, Perth 6102, Western Australia, Australia. Email: [email protected]2. Postdoctoral Fellow, Department of Earth Science, University of Bergen, Allegaten, 41, N-5020 Bergen, Norway. Email: [email protected]3. Group Leader, CSIRO Minerals Resources Flagship, ARRC, PO Box 1130, Bentley, Perth 6102, Western Australia, Australia. Email: [email protected]4. Research Director, CSIRO Minerals Resources Flagship, ARRC, PO Box 1130, Bentley, Perth 6102, Western Australia, Australia. Email: [email protected]5. Postdoctoral Fellow, CSIRO Minerals Resources Flagship, ARRC, PO Box 1130, Bentley, Perth 6102, Western Australia, Australia. Email: [email protected]
Transcript
1
PACRIM 2015 Conference
Paper Number: 166
High resolution characterisation of gold mineralisation at Plutonic Gold Mine, Western Australia: Evidence for the late-stage deposition of high-grade gold
M.F. Gazley1, G. Duclaux2, L.A. Fisher3, R.M. Hough4, M.A. Pearce5
1. GAusIMM, Research Scientist, CSIRO Minerals Resources Flagship, ARRC, PO Box 1130, Bentley, Perth 6102, Western Australia, Australia. Email: [email protected]
2. Postdoctoral Fellow, Department of Earth Science, University of Bergen, Allegaten, 41, N-5020 Bergen, Norway. Email: [email protected] 3. Group Leader, CSIRO Minerals Resources Flagship, ARRC, PO Box 1130, Bentley, Perth 6102, Western Australia, Australia. Email: [email protected]
4. Research Director, CSIRO Minerals Resources Flagship, ARRC, PO Box 1130, Bentley, Perth 6102, Western Australia, Australia. Email: [email protected]
5. Postdoctoral Fellow, CSIRO Minerals Resources Flagship, ARRC, PO Box 1130, Bentley, Perth 6102, Western Australia, Australia. Email: [email protected]
2
ABSTRACT Microcharacterisation such as that described in this paper provides understanding of a mineral system at multiple scales; understanding micro-scale variations is key to understanding the large-scale variations in a mineral system. More than 50 high-grade lode samples from across Plutonic Gold Mine (Plutonic), Marymia Inlier, Western Australia, were analysed by a variety of techniques, including micro X-ray fluorescence mapping (µXRF), micro X-ray computed tomography (µCT), scanning electron microscopy (SEM), field emission gun (FEG) SEM, electron back-scatter detector (EBSD), proton-induced X-ray emission microprobe (PIXE), and synchrotron XRF mapping. These data demonstrate that competency contrasts are key to governing the location of Au mineralisation at Plutonic. Micro-CT imaging of high-grade samples shows the Au filling in voids and pore-space in the rock, after having travelled along more permeable layers, already existing, or reaction-induced pores formed during hydrothermal circulation. Using 3-D visualisation the Au is seen to form a connected network which appears undisturbed at the scale of the samples and a high-grade Au deposition event was one of the last events to affect the rock. This interpretation is supported by SEM analyses on high-grade samples that reveal calc-silicate (epidote-clinozoisite) alteration that is not present in lower Au grade samples which are typically dominated by potassic alteration, and that sulphides are typically less abundant close to the Au. The replacement of biotite by chlorite is an alteration associated with the high-grade Au-mineralising fluid. Gold is coarsest where Au particles are located on the margins of minerals, typically sulphide and quartz grains, suggesting the importance of rheological contrast at the microscale. Furthermore, EBSD data reveals that Au is undeformed and over-prints peak metamorphic and retrograde minerals. Data confirm recent deposit-scale observations, suggesting that deposition of high-grade, visible Au was related to a hydrothermal fluid circulation event late in the geological history of Plutonic.
INTRODUCTION
Microcharacterisation techniques are becoming increasingly advanced, with speed of data collection and detection limits continually improving. These developments are constantly improving the geologist’s ability to resolve mineral system complexity. This paper documents high-resolution characterisation of Au mineralised samples from Plutonic Gold Mine (Plutonic), Western Australia. Maps of chemical variations, collected using micro X-ray fluorescence (µXRF), and 3-D phase distribution from high resolution micro X-ray computed tomography (µCT) are used to inform preparation of thin sections and polished blocks for more detailed characterisation. Analyses commonly encompass SEM imaging with chemical X-ray mapping and microprobe analyses, crystallographic measurements with electron back scatter diffraction (EBSD), and trace element mapping with proton-induced X-ray emission microprobe (PIXE). Finally, high-resolution, XRF mapping with the Maia detector at the Australian Synchrotron (SXRF) gives micron resolution of trace elements at ppm concentrations over cm areas to investigate mobility of path finder elements in samples of particular interest or importance. The data generated by this workflow allow us to answer questions that range from paragenesis of various minerals to determine conditions of Au mineralisation.
GEOLOGICAL SETTING
Plutonic Gold Mine (Plutonic) is located ~800 km northeast of Perth, Western Australia, at the southern end of the Archaean Plutonic Well Greenstone Belt in the Marymia Inlier, between the Yilgarn and Pilbara cratons. Plutonic is probably best considered to be an orogenic lode gold deposit with a long Au mineralising history. Plutonic Well Greenstone
3
Belt rocks were deposited between 2900 and 2720 Ma and metamorphosed at amphibolite facies conditions (~8 ± 2 kbar and ~600 ± 50°C) at 2660 – 2630 Ma (McMillan, 1996; Bagas, 1999; Vielreicher et al., 2002; Vickery, 2004; Gazley et al., 2011a). Lead isotope model ages for sulphide minerals at Plutonic suggest that some Au mineralisation occurred at ~2630 Ma, with a second event at ~2200 Ma (McMillan, 1996; Vielreicher et al., 2002; Gazley, 2011). These early events were followed by two subsequent Au-mineralising events, one at ~1830 Ma, during the Capricorn Orogeny (Vickery, 2004; Pirajno et al., 2004; Gazley, 2011), and a later one at ~1720 Ma, probably syn- or post- a hydrothermal event, which deposited base-metal associated Au (Vielreicher et al., 2002; Gazley, 2011). The later Au mineralising events are considered to have deposited most of the Au at Plutonic (Gazley, 2011; Duclaux et al., 2012, 2013a). Gold at Plutonic is primarily hosted in the Mine Mafic Package, a succession of mafic, subaqueous lava flow units interbedded with thin metasediments. Thin (typically < 1 m) lenses of metagraphitic shales, metashales and metacherts are interlayered with the mafic rocks on a cm- to dm-scale (Gazley et al., 2011b). The overall thickness of the package varies between 30 and 300 m. The entire sequence is cross-cut by at least two generations of late-stage dolerite dykes (Gazley et al., 2014a). The main style of Au mineralisation (Plutonic ‘brown’ lode; Vickery, 2004) typically occurs as thin (~1 – 3 m) lodes that comprise quartz-biotite-amphibole-epidote-arsenopyrite-pyrrhotite ± titanite ± carbonate ± chlorite ± chalcopyrite ± Au assemblages. Where these Au-bearing zones are well developed, they tend to be nearly parallel to both the stratigraphy, as marked by rare metasedimentary horizons (Awan, 2000), and the dominant foliation (Bagas, 1999). Often these metasedimentary units are mineralised. Other forms of Au mineralisation are less commonly observed at the Plutonic deposit: sub-economic to marginally economic Au grades in apparently unaltered metabasalt rock; lodes that mineralogically resemble Plutonic ‘brown’ lode but are hosted within shear zones that cross-cut the foliation; and late-stage quartz-carbonate-pyrrhotite ± chalcopyrite ± Au veins.
MICROCHARACTERISATION
The following workflow was carried out at the Australian Resources Research Centre, Perth, Australia. Samples of interest were also analysed by PIXE, using the CSIRO nuclear microprobe located at University of Melbourne, Australia and SXRF at the Australian Synchrotron (Melbourne, Australia). Samples up to ~15 x ~20 cm were mapped in a Bruker M4 Tornado using a Rh X-ray source (50 kV, 600 µA), under vacuum and a spot size of 25 µm. The only sample preparation that is required is that the surface of the sample is flat, i.e. cut. Maps are limited by data size, but areas ~10 x ~15 cm can be mapped with 75 µm step-sizes in around 5 hrs. Thin-sections can be mapped in around 2 hrs with 25 µm step-sizes. These data offer the opportunity for a first-pass semi-quantitative analysis of a sample. This information can be used to select representative thin sections from the sample. µXRF maps of a ~8 x ~15 cm slab of high-grade mineralisation from Plutonic shows the late, cross-cutting relationship between the vein of Au near the top of the sample and carbonate (indicated by Ca), sulphides (S) and scheelite (indicated by W) (Fig. 1). The 3-D gold distribution within a ~5 mm x ~5 mm block of high-grade mineralisation was examined using an XRADIA XRM500 3-D X-ray microscope with an accelerating voltage 160 kV and power of 10 W. Dedicated image processing workflows (e.g. Godel, 2013) were used to separate the gold from the gangue minerals (Fig. 2).
The back scatter electron (BSE) images presented in this paper were obtained using a Phillips XL40 W-filament scanning electron microscope (SEM). Mineral morphology, and textural relationships were imaged and their chemistry was measured using energy dispersive X-ray spectroscopy. Maps of chemical variation were created with sub-micron resolution to provide detailed areas of focus that complement the larger-scale XRF maps. The textural context of the gold that was imaged in 3-D (Fig. 2) is shown in Fig. 3. Electron backscatter diffraction measures the crystallographic orientation of a single point and can be used to create rasters of orientation data (Prior et al., 1999), to identify deformation (Prior et al. 2002) or crystallographic control during alteration reactions (Pearce et al., 2013). Various types of data can be extracted and plotted as a spatially referenced EBSD map including phase data (Fig. 4a) and crystallographic orientation (Fig. 4b-d). Samples were prepared by using a chemical-mechanical polish to remove damage imparted during conventional thin section making and then ion-beam milling because soft phases (such as gold) were targeted. The data from Plutonic show that in a rock made up of amphibole, quartz, albite and chlorite (Fig. 4a) the gold is present as both large twinned grains with minor variations in crystallographic orientation and a finer grained fraction (Fig. 4b). Large amphibole crystals (grains that are the same colour in Fig. 4c) are cross cut by quartz (Fig. 4d). The lack of deformation (variation in orientation within grains), tabular morphology of the growth twins and the cross-cutting relationship to the amphibole suggests that the gold is post-peak deformation and metamorphism.
DISCUSSION AND CONCLUSION
Microcharacterisation such as that described in this paper provides understanding of a mineral system at multiple scales (Duclaux et al., 2013b; Gazley et al., 2014b). Understanding the micro-scale variations in a mineral system is key to understanding the large-scale variations. The µXRF map (Fig. 1), suggests that competency contrasts govern the location of Au mineralisation. The more quartz-rich units in the µXRF map have preferentially fractured with small Au veins perpendicular to the fabric of the rock. This is consistent with the observation that the dominant control on Au mineralisation is the inter-flow sediments (typically quartz-rich) identified as key to controlling the location of Au mineralisation within mafic sequence (Gazley et al., 2011b). Evidence from the µXRF map suggests that Au is late-stage as it cross-cuts many of the features in the rock. This observation is further supported by the µCT image of a high-grade sample (Fig. 2), in which the Au fills in voids and available space in the rock. In 3-D it does not appear to be deformed, and was one of the last events to affect the rock. This interpretation is supported by available BSE and elemental maps from SEM work (e.g. Fig. 3); high-grade gold is often observed to be associated with chloritisation of biotite, suggesting a late-stage paragenesis. Further evidence for late-stage Au can be found in EBSD maps of Au grains, presented in Fig. 4. The gold grains cross cut the peak metamorphic mineralogy and they do not record any deformation textures that would be expected if the gold had been deformed post-emplacement. High resolution elemental map data collected by PIXE or SXRF can provide further quantitative data of micro-scale elemental distribution. Understanding how a deposit such as Plutonic formed requires evidence at a variety of scales from disparate datasets, and the integration of these data to form a genetic model that is robust at all scales.
