PGE Tenor and Metal Ratios within and below the Merensky Reef, Bushveld Complex: Implications for its Genesis

Abstract

The Merensky Reef is the consequence of several major influxes of magma into the Bushveld magma chamber that gave rise to at least two cyclic units, including the Pre-Merensky and Merensky cyclic units. Analysis of the S, Ni, Cu, platinum group element (PGE) and Au contents of 1140 samples from 24 profiles through the rocks forming these units and two profiles extending downward from the Reef to close to the UG-2 horizon of the Western Bushveld Complex provide important petrogenetic constraints on the origin of the Reef. Particular attention is paid to variations in the calculated PGE tenors and Cu/Pd ratios of sulfides, both vertically and laterally. All of the profiles show a decrease in Pt tenor and concomitant increase in Cu/Pd ratio upward across the pyroxenite of the Merensky unit and laterally away from certain centres that are interpreted as feeder zones. The data also document that more mafic horizons within the Merensky to UG-2 interval of the southwestern Bushveld are characterized by much lower Cu/Pd (<100–400) than the ambient ratio (∼4000). This is interpreted as demonstrating that these horizons are the consequence of incursions of magma whose composition has been modified in a staging chamber by reaction with sulfide resident there. The favoured interpretation for the Merensky Reef is that the Bushveld chamber(s) was occupied by a resident magma crystallizing orthopyroxene and plagioclase. Pre-Merensky and, subsequently, Merensky magmas entered the chamber(s) intermittently from localized feeder structures, mixing or mingling with the resident magma and giving rise to the more mafic cumulates. At specific horizons, including the Merensky Reef, the mixing or mingling caused sulfide immiscibility. In these cases, as the new magma pulses flowed laterally away from the entry zones they deposited sulfide, becoming depleted in PGE. On coming to rest, the magma continued to segregate sulfide and crystallize pyroxene, so that the sulfides became progressively depleted in PGE and acquired a higher Cu/Pd upward through the pyroxenite. This model requires that the Merensky magmas could dissolve up to two orders of magnitude more PGE than current measurements on dry diopside–anorthite melts have indicated. It is suggested that very PGE-rich magmas can develop as a result of sulfide-unsaturated magma reacting with early formed sulfide in a staging chamber, dissolving FeS and enriching the remaining sulfide in PGE. Later magmas equilibrating with this enriched sulfide will themselves become highly enriched in PGE.