EPMA and LA-ICP-MS analyses of the minerals in the UG2&UG3 chromitites - Bushveld Complex
Description
The UG2 Reef constitutes the largest resource of PGE in the Eastern and Western Limbs of the Bushveld Complex, South Africa. The chromitites described here occur in the Upper Critical Zone (UCZ), a complexly layered sequence of norite-anorthosite with subordinate pyroxenite and harzburgite. In the Western Limb of the Bushveldd Complex (South Africa), the UG2 is overlain by two or three subordinate layers of chromitite, the “UG2 Leaders”, which may coalesce with the main seam to create a thick reef. At Winnaarshoek, in the Eastern Limb, the UG2 Leaders are poorly developed and are displaced by the UG3 and UG3a-b chromitites. Two sills are identified at Winnaarshoek, a pyroxenitic lower sill (UG2; UG2 Leaders) and a pyroxenite-harzburgite upper sill (UG3; UG3a-b). The UG2 and UG3 chromitites are underlain by thick layers of anorthosite, part of the primary stratigraphy, and overlain by pyroxenite related to the sills. Present data represents ICP-MC (Trace elements and REE) on bulk rock samples, as well as EPMA data (quantitative point analysis) on chromite, plagioclase, pyroxenes, and olivine from all these layers, from cores sampled across UG2-UG3-UG3a,b corresponding depths. The problem of the chromium budget, i.e., the anomalous thickness of chromitite contained in the pyroxenitic sills, is resolved by appealing to the entrainment of xenocrysts of Cr-spinel at depth. In some cases, the xenocrysts were sufficiently abundant as to form chromite slurries. The coalescence and bifurcation of individual chromitite layers suggests emplacement of multiple slurries. The composition of the Cr-spinel is tightly constrained and closely-spaced EPMA data have shown the absence of internal fractionation in the chromitites. Each layer of chromitite is characterised by the presence of loosely-packed grains of Cr-spinel (in a silicate matrix) interspersed with coarsely crystalline mosaics of Cr-spinel, the latter related to prolonged periods of annealing. Primary textures have been modified to such an extent that it is implausible to expect to find evidence of settling of crystals or basal growth. The mobile nature of the slurries was facilitated by the small size (<0.1 mm) of the rounded, dense grains of Cr-spinel. Some slurries sank to the base of sills and disaggregated the crystalline fabric of the footwall anorthosite, entraining the liberated, refractory crystals of plagioclase. The Cr-spinel and plagioclase in the UG2 and UG3, which in part could be described as chromite-anorthosites, are not related by cotectic crystallization. Each chromitite layer is characterized by a unique PGE signature, mineralization having been collected by the chromite slurries at depth and transported as microxenocrysts which piggybacked on the Cr-spinel.