GSA Annual Meeting in Indianapolis, Indiana, USA - 2018

Paper No. 243-4
Presentation Time: 9:00 AM-6:30 PM

AN EXPERIMENTAL ASSESSMENT OF PLATINUM, PALLADIUM AND NICKEL PARTITIONING BETWEEN COEXISTING SULFIDE AND BASALTIC MELTS


REO, George A., FRANK, Mark R. and EHLICH, Joshua J., Department of Geology and Environmental Geosciences, Northern Illinois University, Davis Hall, Room 312, DeKalb, IL 60115

This study examined the equilibrium concentrations of platinum, palladium and nickel in coexisting experimentally synthesized sulfide and basaltic melts to evaluate the formation of platinum group element (PGE)-rich sulfide ore bodies in layered mafic intrusions (LMI). Metallic Pt, Pd, and Ni were equilibrated with a silicate melt, a sulfide melt, and pyrrhotite at 1100 °C in fused quartz tubes to mimic the formation of PGE-rich LMI. Experiments were conducted in a vertical tube furnace equipped for gas-mixing with O2 fugacity buffered to a value equivalent to QFM (quartz-fayalite-magnetite), and S2 fugacity (fS2) measured by analyzing the composition of quenched pyrrhotite. Whereas O2 was kept constant, fS2 ranged from 10-10 to 10-2 bars, allowing for an analysis of its potential impacts on PGEs in LMIs. Major elemental concentrations in the quenched sulfide and silicate melts were determined by using a CAMECA SXFiveFE EPMA. Trace metal concentrations in the quenched silicates were ascertained by using an Agilent 7500ce ICPMS coupled with a Geolas laser ablation system. Nernst style partition coefficients between sulfide and silicate melts for a given element were determined (Dxsul/sil). Dxsul/sil was defined as the concentration of an element in the quenched sulfide melt divided by the concentration in the quenched silicate melt. DPtsul/sil was found to be 1.4 x 105 to 1.5 x 107, decreasing with increasing fS2. Palladium behaved similarly, as DPdsul/sil ranged from 1.8 x 105 to 5.3 x 105 and decreased with increasing fS2. DNisul/sil ranged from 3.1 x 102 to 4.5 x 103 and showed no trend over the range of fS2 studied. Cu/Ni and Pt/Pd in the sulfide melts was 0.9 to 3.3 and 0.1 to 0.5, respectively. These ratios generally decreased with increasing fS2. Metal saturated low fS2 sulfide-silicate melt systems produced sulfide melts with lower Fe concentrations and higher Pt and Pd concentrations, resulting in higher DPGEsul/sil values. The results of this study suggest that the scavenging of Pt and Pd from the silicate melt is favored in orthomagmatic sulfide mineral assemblages with low fS2. Further, the Fe content of the sulfide melt may be a useful proxy for determining PGE enrichment in PGE-rich LMI.