Horizontal and vertical variations in noble-metal distribution in the Great Dyke of Zimbabwe: A model for the origin of the PGE mineralization by fractional segregation of sulfide

Abstract

A study based largely on diamond drilling undertaken specifically for research purposes has shown that in the Darwendale Subchamber of the Great Dyke of Zimbabwe there are three principal sulfide zones that are located in Subunits 1a and 1b of Cyclic Unit 1. Sulfide Zone 1, the “Main Sulfide Zone” (MSZ) is subdivided into three subzones. The base of the MSZ occurs at the top of the main bronzitite layer of Cyclic Unit 1, extending into the overlying websterite. The noble metals are concentrated at the bases of each sulfide zone and subzone, and decrease in concentration rapidly upwards, whereas variations in the concentrations of base metals such as Cu correlate well with variations in the amounts of sulfide across each zone. When calculated in terms of their concentration in 100% sulfides, it is found that Pd decreases upwards most rapidly, followed by Pt and then Au, with the Cu concentration remaining essentially constant. The decrease of the noble metals is interpreted in terms of simple Rayleigh fractionation as the consequence of the removal of sulfide. The observed correlation between the tops and bottoms of sulfide zones with small variations in the Mg-number of bronzite is taken to indicate that, in Zones 2 and 3 sulfide segregation started as a result of crystallization increasing the S content of the resident magma in the chamber until it reached that of sulfide saturation, coupled with fresh inputs of sulfide-unsaturated, more primitive magma causing the resident magma to become unsaturated again. This recharge by primitive magma also raised the PGE content of the sulfides segregating from the trapped intercumulus liquid, indicating that the resident magma was becoming recharged in PGE in this way. Subunits (i) and (ii) of Sulfide Zone 1 are thought to contain too much sulfide to be explicable in this way, and it is suggested that mixing with a plagioclase-saturated magma, possibly generated around the flanks of the intrusion, was responsible for these subunits. The sulfides of each of the sulfide zones are more concentrated at the margin than at the axis of the Great Dyke, but have a lower tenor of noble metals. It is proposed that the axially-located sulfides were the consequence of magma-sulfide equilibration at a higher R factor (magma/sulfide ratio) than those at the margin. The magma-sulfide partition coefficients required to explain the percentage difference in the tenors of the different noble metals between axis and margin in this way are the same as those required to explain the vertical variations in noble-metal tenor in terms of the fractional segregation model.