Lifting the cloak of invisibility: Gold in pyrite from the Olympic Dam deposit, South Australia

Ehrig, Kathy1,2, Ciobanu, Cristiana3, Verdugo-Ihl, Max3, Dmitrijeva, Marija3, Cook, Nigel2, Slattery, Ashley4

1BHP Olympic Dam, Adelaide, Australia, 2School of Civil, Environmental and Mining Engineering, University of Adelaide, Adelaide, Australia, 3School of Chemical Engineering and Advanced Materials, University of Adelaide, Adelaide, Australia, 4Adelaide Microscopy, University of Adelaide, Adelaide, Australia

’Invisible gold’ in pyrite refers to gold either present within the sulfide lattice or as discrete nanoparticle (<100 nm-diameter) inclusions (NPs), making it undetectable by conventional optical and scanning electron microscopy. Investigation of “invisible gold” in chalcopyrite-pyrite ores from the Olympic Dam Cu-U-Au-Ag deposit (one of the world’s largest Au deposits) confirms the presence of Au in some arsenic-bearing pyrites at concentrations measurable by laser ablation inductively coupled-plasma mass spectrometry (LA-ICP-MS). Arsenic-bearing pyrite in the studied sample shows As-Co-Ni-oscillatory zoning patterns with variable complexity suggesting grain re-crystallization during replacement by chalcopyrite. LA-ICP-MS data obtained from 164 pyrite grains plot below the Au and As solubility limit empirically defined from studies of epithermal and Carlin-type deposits.

Several As-rich pyrite grains were analyzed using Scanning Transmission Electron Microscopy (STEM) and EDX-STEM analysis of foils obtained by Focused Ion Beam methods. Micron-scale, oscillatory zoning patterns observed on back-scattered electron (BSE) images and LA-ICP-MS element maps extend down to the nanoscale. Decoupling between trace elements is common, for example Ni depletion wherever As and Co are enhanced, with nucleation of discrete Co-As-bearing NPs (cobaltite/safflorite?)

Importantly, Au-bearing NPs are identified in all cases, in intimate association with other (sulpho)tellurides. In addition, abundant cassiterite and rare chalcopyrite NPs are also identified. Some of the largest Bi-Ag-telluride NPs contain electrum as tiny pore-attached NPs within the larger telluride. Nanometer-size electrum NPs were also identified in association with chalcopyrite. Silver-Au-telluride NPs form mono- or bi-component NPs. These NPs occur along trails displaying As-Co-enrichment, or formation of nm-wide lamellae of Bi-Pb-sulphotellurides marking pyrite twin boundaries. One wider lamella was identified from the layer stacking as a strongly disordered member of the aleksite series. Coarser tellurobismuthite (Bi2Te3), a few μm-wide, is associated with altaite (PbTe) at pyrite-chalcopyrite boundaries.

Pyrite displays kink- and screw-dislocations associated with trace element remobilization or NPs nucleation. These defects can be associated with either ‘marcasitization’ or loss of Fe (formation of pyrrhotite), within nanoscale domains affected by fluid percolation and pyrite recrystallisation. Twin planes in pyrite enriched in heavy elements (Bi-Pb-Te) represent zones of weakness and assist element exchange between host mineral and percolating fluids during coupled dissolution reprecipitation reactions (CDRR), analogous to those known for hematite from Olympic Dam.

Nanoscale textures in pyrite allow for interpretation of Au-NPs as Au released from solid solution in pyrite during CDRR associated with marginal chalcopyrite replacement. Nanoscale analysis lifts the cloak of invisibility for Au in pyrite at Olympic Dam. These results show that confirmation of whether gold occurs as NPs or in solid solution based solely on position above or below the solubility limit of Au in pyrite on a plot of Au vs. As is impossible without corroborative studies at the nanoscale.


Kathy completed a PhD from the University of California- Berkeley in 1991 and left San Francisco in 1992 to join the former WMC as a research geologist to work on the genesis of the Olympic Dam deposit. She currently leads the team who built the geometallurgical model of Olympic Dam.

Stability of gold nanoparticles in sulfur-bearing hydrothermal fluids: an experimental study

Liu, Weihua1, Chen, Miao1, Yang, Yi1, Mei, Yuan1, Etschmann, Barbara1, Brugger Joël1, Johannessen, Bernt 1

1CSIRO Mineral Resources, Clayton,, Australia, 2Monash University, Clayton,, Australia, 3Australian Synchrotron, Clayton,, Australia

Current theories of the gold deposit formation from hydrothermal fluids have been challenged by recent field and laboratory observations, suggesting that gold nanoparticles/colloids could be important in gold transport/deposition in ore fluids. In this study, the stability of gold nanoparticles (colloidal gold) in sulfur-bearing and citrate-bearing solutions was investigated at temperatures up to 225˚C using Synchrotron X-ray Near-edge Spectroscopy (XANES), and up to 350˚C with a visual check of colour change. The citrate-based colloidal gold solutions, with or without colloidal silica in the solution, are only stable up to 225˚C. In contrast, the gold colloids in Na2S solutions are not stable upon heating to 150˚C, but stable up to 300˚C when 0.5-1.5 wt% of colloidal silica is present in the solution. The colloidal gold particles started to aggregate and deposit from the solution with the aggregation and growth of silica particles at 350 ˚C. The concentrations of gold as colloids in the solutions are up to 0.5 mmol (~95 ppm), more than three orders of magnitude higher than gold solubility as aqueous complexes under the same condition calculated based on available thermodynamic data. These results provide the first evidence that high concentrations of gold nanoparticles are stable in sulfur-bearing fluids at elevated temperatures (~300˚C). This implies that the formation of gold nanoparticles is an effective way to concentrate gold in hydrothermal sulfur-bearing fluids to form high-grade gold ores.


Dr. Weihua Liu is a senior research scientist at the Commonwealth Scientific and Industrial Research Organisation (CSIRO) and was an Australian Research Council Future Fellow

Weihua’s research has been focused on the investigation of metal mobility and mineral/fluid reactions in hydrothermal systems, combining both experimental and theoretical techniques.

The unusual Imou porphyry Au-Cu deposit, Western Highlands PNG

Ireland, Timothy1; Federico Cernuschi2, Robert Sievwright1, Hannah Goswell1, Stefanova, Elitsa3, and John Dobe4.

1First Quantum Minerals Ltd, Canada, 2Eclectic Rock Ltd,  Punta del Este, Uruguay, 3Bulgarian Academy of Science – Geological Institute, Bulgaria, 4Footprint Resources Pty Ltd

Imou (4.944˚S; 142.806˚E) is a porphyry gold-copper deposit located western Papua New Guinea along structural strike between the deposits at Frieda River and Yandera. The area was highlighted during systematic geochemical exploration in the mid-1970s, when five porphyry centres were identified in a camp covering ~30 x 30 km. The Imou target was delineated in the 1990s as a stream sediment and soil Cu anomaly of >500 ppm coincident with Au anomalism >0.2 ppm and associated with a magnetic polyphase porphyritic intrusive complex. Two holes drilled in 1999 discovered the deposit, and a dozen recent holes permit this first attempt at geological description.

Magmatic rocks comprise equigranular and lesser porphyritic diorites that belong to the Nekiei intrusive suite, which was emplaced into a thrust duplex of obducted siltstone and oceanic crust. New zircon-U-Pb geochronology records at least 700 ky. but less than 2.1 my. of Late Miocene magmatic evolution. The intrusions are low to mid-K calc-alkaline series rocks composed primarily of calcic plagioclase and hornblende with rare biotite and quartz, and accessory titanomagnetite. Whole rock chemical proxies for porphyry fertility are comparable to causative suites in other deposits (Sr/Y up to 65 and V/Sc up to 16) but zircon in these rocks has relatively low concentrations of U, Th, LREE and Hf .The intrusions manifest in airborne magnetic surveys as discrete high amplitude magnetic ‘bullseye’ features, but no cogenetic large plutons are observed at surface, nor can be inferred from the magnetic response.

The outcropping manifestation of the porphyry deposit at Imou is in many ways typical: within a geochemical footprint of ~2 km2 there are domains of A-family quartz veins that sometimes contain Cu-Fe-sulphides and/or magnetite, the distribution of which broadly coincides with the occurrence of hydrothermal magnetite, feldspar, anhydrite and muscovite. These quartz vein domains reach 50 vol% quartz in the vicinity of the magnetite-bearing assemblage. There are structurally controlled domains of late D-veins and phyllic alteration, followed by post-mineral porphyritic dykes. However, there are numerous differences with economic PCD described elsewhere. Cu-Au grade is not associated with the abundance of A-family quartz veins. Further, there is no petrographic nor bulk chemical evidence for large volumes of potassic alteration. Observed feldspars are albite and sporadic hydrothermal biotite represents only remobilisation of local potassium. There is little disseminated sulphide associated with this alteration, instead, grade development is associated with chalcopyrite-pyrite-(magnetite) fracture paints, and sulphide-anhydrite veins that cross-cut the quartz vein stages. These paragenetically late brittle veins are widely distributed at low frequency, but are associated with best Cu-Au grades. These late hydrothermal features are not systematically associated with a particular wallrock alteration assemblage, although chlorite-montmorillonite is the dominant alteration assemblage outboard of localised sericite- and albite-bearing assemblages.

We advance several working hypotheses for why Imou differs from other porphyries, and especially why metals did not precipitate efficiently with the early high temperature alteration and quartz veining. We speculate that many barren or subeconomic porphyries may have inferior Cu-Au grade development for similar reasons.


I’m a jack-of-all-trades explorer with experience in porphyry-epithermal systems, carlin-type deposits, sed-hosted Cu and sed-hosted Zn. I’m currently the Principal Exploration Geologist – and informal resident geochemist – at FQM. I did my postgrad research at CODES, and have spent much of my career working in E Europe and the central African copperbelt.  

Evolution of magmatic fertility for porphyry Au & Cu deposits through the prism of zircon chrono-chemistry, Balkan Peninsula, SE Europe

Ireland, Timothy1; Bilyarska Teodora2, Bilyarski, Stoimen1, Protic, Nenad1, and Stefanova, Elitsa3

1First Quantum Minerals Ltd,, Vancouver, Canada, 2St Kliment Ohridski University of Sofia, Sofia, Bulgaria, 3Bulgarian Academy of Science – Geological Institute, Sofia, Bulgaria

The accreted arc terranes of the Balkan western Tethyan host a complex metallogeny that includes almost all Cu and Au mineralisation styles typical of subduction-related arcs and post-collisional settings. Broadly, magmatism in the belt began as a series of submarine arcs constructed on thinned continental basement, and evolved to subaerial post-collisional magmatism during and after arc-continent collisions in the Late Cretaceous and Eocene. Each collisional event is associated with a phase of metallogenesis that changes in terms of metal budget and deposit style as collision occurs and then collapses, e.g. in the Timok and Panagjurishte districts, three phases of magmatism, each lasting 5-10 my. record a transition in space and time from sub-volcanic diorites associated with porphyry Cu-Au deposits, to more voluminous effusive andesites associated with epithermal deposit styles, and then monzonitic intrusions with a Au-dominant metal budget. The rocks from all of these three phases are the products of hydrous, intermediate magmas that yield equivalent ‘fertile-looking’ results in terms of whole rock magma chemistry proxies such as Sr/Y that explorers may use for area selection in arc environments.

A similar metallogenic progression occurs among less well-documented magmatism that occurred subsequent to the Alpine orogeny. In this case the main, early metallogenic phase is represented by porphyry and epithermal Cu-Au deposits associated with trachyandesite and monzonitic volcano-plutonic complexes. As post-collisional extension comes to dominate the regional tectonic environment the metallogeny becomes dominated by Pb-Zn in a series of ‘ore fields’ that we infer to be genetically related to syn-extensional granitoids, and thereafter by epithermal gold deposits in terrestrial grabens.

In this study we compiled all the available published whole rock chemistry and zircon chronology and chemistry for magmatic rock units associated with each of these metallogenic stages, and collected new complementary data from rock samples and stream sediments. The result is a coherent record of the temporal evolution of magmas associated with each of these metallogenic stages. We observe that the detrital zircon populations are closely related to the results from rock samples, however, we contend that the detrital record is a superior medium as it avoids sampling bias and undersampling. These detrital results were interpreted as a holisitic record of the magmatism in each district. There are patterns among this zircon chrono-chemical evolution that correspond to consistent aspects of regional metallogeny. Short periods of zircon crystallisation (i.e. <3 my) in which zircons span a wide range of compositions suggesting both primitive and enriched or evolved source contributions tend to characterise the metallogenic events in which Au is the primary economic commodity. In contrast, more protracted zircon crystallisation history and slower evolution to fertility implied by proxies such as Eu/Eu* characterises the major porphyry Cu camps. This approach may be applicable in other terranes wherein the magmatic-metallogenic history is not well constrained.


I’m a jack-of-all-trades explorer with experience in porphyry-epithermal systems, carlin-type deposits, sed-hosted Cu and sed-hosted Zn. I’m currently the Principal Exploration Geologist – and informal resident geochemist – at FQM. I did my postgrad research at CODES, and have spent much of my career working in E Europe and the central African copperbelt.  

Controls on gold endowments of porphyry deposits

 Massimo, Chiaradia1

1Department of Earth Sciences, University of Geneva, Geneva, Switzerland

Porphyry deposits are natural suppliers of ~75% copper and ~20% gold to our society. Nonetheless, gold endowments of porphyry deposits are characterized by a wide range going from a few tons to >2500 tons of gold. Here, I propose a model to explain the reasons of the large variations in metal endowments of porphyry Cu-Au deposits.

Porphyry Cu-Au deposits define two distinct trends in Au versus Cu tonnage plots: Cu-rich (Au/Cu ~4*10-6) or Au-rich (Au/Cu ~80*10-6). Cu-rich porphyry deposits are related to Andean-type subduction and typical calc-alkaline magmatism in thick continental arcs. In contrast, Au-rich porphyry deposits are associated with high-K calc-alkaline to alkaline magmatism in late to post-subduction or post-collision and extensional settings, and also with calc-alkaline magmatism. The largest Au-rich porphyry deposits are associated with high-K calc-alkaline to alkaline magmatism. Geochronological data at individual porphyry deposits suggest that gold endowments for both trends grow larger the longer the mineralization process is. However, Au is precipitated at much higher rates in Au-rich (~4500 tons Au/Ma) than in Cu-rich porphyry deposits (~100 tons Au/Ma).

Monte Carlo modelling of petrologic processes suggests that the different rates of gold precipitation in Cu-rich and Au-rich porphyry deposits most likely result from a 5-12 times better efficiency of gold precipitation in Au-rich than in Cu-rich deposits. The reason of the different efficiencies of gold precipitation is the different depths of formation of Cu-rich and Au-rich porphyry deposits which favour (deep level) or not (shallow level) a decoupling of Au and Cu precipitated from the magmatic-hydrothermal fluids. Interestingly, Au-rich porphyry deposits formed at shallower levels are also associated with magmatic rocks that have evolved at average shallower levels than Cu-rich deposits, as suggested by systematically lower Sr/Y values of the former (Au-rich systems) with respect to the latter (Cu-rich systems). Monte Carlo modelling shows that the higher gold endowments of Au-rich porphyry deposits associated with alkaline magmas require higher gold contents in the parental magmas such as those that are typical of alkaline magmas but not of calc-alkaline ones. This suggests an additional petrogenetic control in the formation of the Au-richest porphyry deposits associated with variably alkaline magmas.

Whereas depth of porphyry formation and chemistry of magmas (alkaline versus typical calc-alkaline) seem to control the Au-rich versus Cu-rich nature of porphyry Cu-Au deposits, the correlation of the Cu and Au endowments with ore deposition duration suggests that the final Cu and Au endowments of these deposits are determined by the cumulative number of mineralizing steps that are ultimately controlled by magma volume and ore process duration. The difference is that variably alkaline systems and shallow crustal calc-alkaline systems are inherently associated with magmas, whose fluids are tectonically (i.e., shallow emplacement) and chemically (alkaline magmas) optimized for high gold precipitation efficiency. In contrast, typical calc-alkaline (high Sr/Y) magmas form in a geodynamic context that favours enormous magma accumulations, which are necessary to produce behemothian Cu-rich deposits, but are emplaced at depths at which the exsolved fluids are less efficient for gold precipitation.

Biography to come.

About the GSA

The Geological Society of Australia was established as a non-profit organisation in 1952 to promote, advance and support Earth sciences in Australia.

As a broadly based professional society that aims to represent all Earth Science disciplines, the GSA attracts a wide diversity of members working in a similarly broad range of industries.

Conference Managers

Please contact the team at Conference Design with any questions regarding the conference.
© 2020 Conference Design Pty Ltd