Correlating Cu-Fe sulfides and Au mineralization in the Ertsberg-Grasberg District of Papua, Indonesia using volumetric analysis and trace element geochemistry

The Ertsberg-Grasberg District in Papua, Indonesia, hosts two of the world’s largest intrusion-related Cu-Au deposits: the Ertsberg Intrusive System, a hybrid porphyry-skarn deposit, and the Grasberg Igneous Complex, a high-grade porphyry deposit. Cu mineralization within the Grasberg porphyry and Ertsberg skarn systems consists primarily of bornite and chalcopyrite, whereas native gold occurs as inclusions within, or along boundaries of these minerals. Experimental studies by other researchers have shown that at hydrothermal ore-forming temperatures (~300-700°C) and elevated sulfur activity, bornite and chalcopyrite can host 1000s ppm Au within the sulfide lattice or as nano-inclusions. Upon retrograde cooling of the hydrothermal system, the capacity of the Cu-Fe sulfides to host Au significantly decreases to ~10 ppm, suggesting that the Au becomes unstable within the Cu-Fe sulfide matrix and may passively migrate out of the sulfides and coalesce to form native gold grains. These data suggest that Cu-Fe sulfides could exert a strong control on the gold contents of porphyry deposits. However, the traditional model for native gold deposition in large hydrothermal systems relies primarily on fluid pulses, and does not consider gold contributions from the exsolution of previously precipitated gold within Cu-Fe sulfides.
To assess the role of Au-bearing Cu-Fe sulfides in the concentration of native gold in this setting, High Resolution X-ray Computed Tomography was used to measure the native gold grains’ shapes, textures, and occurrence modes, and map the extent of contiguous Cu-sulfides. HRXCT data were used to 3D modified Voronoi regions within the Cu-sulfides, as an estimate of diffusional domains that may have provided gold to the Au-grains during cooling. The modified Voronoi volumes are defined for each Au-grain as the set of points within the Cu-sulfide network that are closer to that Au-grain than any other, when measured along a path through the Cu-sulfide. HRXCT data for 11 Ertsberg-Grasberg District ore samples were processed with two different threshold criteria, which produced 16 linear correlation values between modified Voronoi volumes and gold grain volumes. Of these 16 correlation coefficients, none show statistically significant correlations. Because of the paucity of gold grains within our samples, calculated drainage regions commonly extended to the physical edge of the core samples, rendering their actual volumes ambiguous and the correlations associated with them un-interpretable. In cases where numerous gold grains were identified, correlation values were not statistically significant. In general, however, this analysis was impaired by the necessity of interpreting whether all grains in a sample were created by a single event and mechanism. Complementary Laser Ablation Inductively Coupled Plasma Mass Spectrometry (LA-ICP-MS) was used to assess and compare trace element variation within the Cu-Fe sulfides in the district and to constrain spatial variation of Au within Cu-Fe sulfides that contain native gold grains. LA-ICP-MS data show a strong positive correlation between Bi and Ag, and a moderate, positive correlation between Bi and Au, throughout the data. Au occurrences are scattered rather than uniform, where Au concentrations appear in patches throughout the sulfides. This may imply that the majority of non-visible gold exists in Au nanoparticles, rather than solid solution with the Cu-Fe sulfides. Spot analyses and 2D maps indicate the presence of an “enrichment halo” of elevated Au contents in Cu-Fe sulfides surrounding the gold grains. This “enrichment halo” can be interpreted as evidence in support of a hybrid Ostwald-type ripening process that coarsens gold inclusions in chalcopyrite at high-temperatures in porphyry-skarn systems.