Quantitative Modeling of Infiltration in Layered Intrusions

Duke University, Durham NC

Investigators

Abstract

Boudreau EAR-0206905 Igneous intrusions such as the Bushveld Complex of South Africa and the Stillwater Complex of Montana host much of the world's reserves of economically valuable elements such as chromium and the platinum-group elements. Although subject to numerous studies, there is still surprisingly little agreement on how they crystallized and differentiated. A significant part of the proposed study involves extending a numerical model that links an equilibrium crystallization routine (based on the MELTS algorithm of Ghiorso and coworkers) with one-dimensional heat and mass transport equations that include compaction equations of McKenzie. Examples of areas the work would be extended include the addition of spinels to the program and investigation of the formation of economically important chromite-rich layers that occur in layered intrusions. On a parallel track, we will look at details of mineral-mineral trace element partitioning to test if concentrations of temperature-dependent trace elements are affected by compaction-driven recrystallization. Specifically, laser ablation-ICP-MS and electron microprobe mineral mapping and bulk mineral analysis will test if mineral-mineral partitioning is correlated textural and/or modal variations. Finally, we wish to test if one can infer concealed intrusion geometry from crystal size distribution (CSD) data. This will be tested by comparing existing data with new Lower and Critical zone samples to be collected where the Lower is both unusually thick over footwall embayments (i.e., the "Burgersfort Bugle") and where the Lower and Critical zone onlap the floor rocks in the eastern Bushveld. If successful, this will further demonstrate that texture is locally evolved and not necessarily a function of minerals transported from elsewhere.