Constellation, le dépôt institutionnel de l'Université du Québec à Chicoutimi

Résumé

Our project has investigated experimentally the role of Sdominated vapor in the formation of economically important platinum-group element (PGE) ores. The research has been conducted in two general directions: 1) experimental study of possibility of the PGE transport by a S vapor in the Fe-Ni-Cu sulfide system at magmatic temperatures; and 2) experimental study of the PGE enrichment mechanisms of a sulfide phase as a result of its desulfurization by the vapor phase and the formation of individual PGE alloys at low-S conditions. Open-system, with respect to the gas-phase, conditions were set up using the tube-in-tube technique in both series of experiments. In the vapor transport experiments, a S-rich donor, (Fe,Ni,Cu)1-xS of variable base-metal (BM) ratio, was doped with 2000 ppm of each PGE and Au. A S-poor pyrrhotite (Po) was used as the PGE receiver. The metal/S ratio of the system was varied by changing the donor/receiver ratio to assess whether the metals were transported as Sspecies or metals. In the system Fe-S-PGE the run products were receiver Po and donor PGE-bearing Po containing individual grains of Au. In the system Fe-Cu-Ni-S-PGE the run products were receiver Po and a donor association composed of monosulfide solid-solution [Mss, (Fe,Ni,Cu)1-xS] ± Cu-Au rich sulfide melt ± individual PGE phase (PGM). The S-removal experiments were carried out at 1000-980oC in the system Fe-Ni-Cu-S doped with 0.2 wt.% of Pt, Ir and 0.5 wt.% of Pd and with sulfur fugacity (fS2) buffered by S-poor pyrrhotite. The quantity of the Po buffer was 30-40 times larger than that of the S-rich sample composed of PGEbearing Mss ± Cu-rich sulfide melt. In the course of experiments, the S transfer from the PGE-bearing BM sulfide to the Po buffer has occurred via vapor phase thus simulating a natural desulfurization process. Final compositions of the run products from both series of experiments were determined by electron microprobe at McGill University (major PGE, BM and S), and by laser ablation ICP-MS at the University of Quebec in Chicoutimi (trace PGE with the detection limit for all PGE in the 10 to 30 ppb range). In the vapor transport experiments, the receiver Po contained significant quantities of transported Pt and Pd (PPGE), Os and Au, but little Ir, Ru and Rh (IPGE). In addition, a much higher quantity of Ni was transported trough the vapor-phase (700 ppm to 11300 ppm) compared to Cu (1000 ppm). There is a dependence of the amount of Ni and PPGE transported on the metal/S ratio of the system. This suggests that Ni and the PPGE were transported as base-metal sulfide species whereas Cu and the IPGE were transported as metals. The removal of sulfur from the PGE-bearing sulfide via a vapor phase in our desulfurization experiments resulted in: 1) exsolution of Pt and Ir from S-depleted base-metal sulfides in the form of PGE alloys leading to the enrichment of a sulfide residue in the heavy PGE; 2) partial or even complete melting of the sulfide leading to the formation of S-poor monosulfide solid solution [(Fe,Ni,Cu)1-xS], Fe-Pt-Ir alloy and highly mobile Cu-Ni-Pd rich sulfide liquid. This could separate palladium from platinum and iridium in natural ore systems resulting in a change from the initial Pd/Pt and Pd/Ir ratio. In most natural examples where fluid mobilization of PGE has been suggested, the remobilized material is enriched in Cu, Pd and to a lesser extent Pt. Our experiments suggest that transport by S-vapor is not the mechanism for this remobilization, as Pd was not fractionated from Pt, and Cu was not significantly transported by the vapor compared to Ni. A more promising candidate for the remobilization of the Cu and Pd is the Cu-rich sulfide melts that formed in the desulfurization experiments.