PGE mineralization in porphyry Cu–Au deposits is characterized by substantially higher Pd/Pt (∼7–60) and Pd/Ir (∼100–10,500) than typical orthomagmatic sulfide deposits (e.g. Pd/Pt ∼0.6 and Pd/Ir ∼20 for the Bushveld). It has previously been suggested that the high mobility of Pd, relative to the other PGEs, may account for the preferential enrichment of Pd in porphyry Cu–Au deposits. However, the low compatibility of Pd in the volatile phase relative to the other PGEs, shown in this study, invalidates this explanation. We suggest that the PGE geochemistry of Pd-rich Cu–Au deposits is principally derived from the PGE characteristics of the magma from which the ore-forming fluid exsolved. Pd-rich porphyry Cu–Au deposits are associated with highly oxidized magmas. Prior to sulfide saturation Pd, Au and Cu behave as incompatible elements and concentrate in the melt with fractional crystallization, whereas Pt is depleted by early crystallization of a Pt-rich alloy and the other PGEs by the co-crystallization of the Pt-rich alloy and Cr spinel. As a consequence the Pd/Pt and Pd/Ir in the evolving melt and the magmatic volatile phases that exsolve from that melt, increase with increased fractionation. The high Pd content and high Pd/Pt (∼7–60) of Cu–Au porphyry ores therefore require the parent magma to have undergone extensive sulfide-undersaturated fractional crystallization prior to volatile exsolution.
Our study also showed that the altered dacites contain PGE abundances that are similar to those of fresh dacites although Pt and Rh are slightly enriched in the altered dacites, which indicates low mobility of PGEs during secondary hydrothermal alteration.