Large layered mafic intrusions are important for a number of reasons. They are major sources for, among others, the platinum group elements (PGEs). Also, they represent the primary evidence for fractional crystallization, a process in which molten rocks solidify by the crystallization of a series of different minerals. As the minerals crystallize, they are removed from the melt and thereby no longer interact with it; this results in rock layers of different chemical or mineralogical composition.
Although fractional crystallization and crystal accumulation are the primary igneous processes occurring in these rocks, there is considerable uncertainty concerning details of how the partially molten rocks crystallized and, in particular, the extent to which postcumulus processes influenced lithology and geochemistry. For example, an active matter of debate is whether the stratiform PGE deposits, such as the Merensky Reef in the Bushveld complex, South Africa, owe their origin to such postcumulus processes. Our studies are directed toward understanding the postcumulus, magmatic environment. The approach in these studies is to develop petrologic models based on detailed examination of the PGE-bearing rocks. Most of our recent work concerns the Merensky reef, but current projects also include a geochemical and ion probe study of the PGE- bearing layer in the Great Dyke, Zimbabwe.
We also use experiments to understand the origin of PGE and other magmatic base metal sulfide deposits. In partitioning experiments, we produce coexisting sulfide and silicate melts which are then chemically analyzed to determine the relative abundance of metals in each of the melts. Further partitioning experiments on the PGEs and certain other elements are planned which will take advantage of new analytical techniques.