Fluorine is a volatile component of most terrestrial magmas that affects: 1) physical properties, 2) trace element transport and ore deposition by fluids, and 3) the environmental impact of volcanic emissions. F in magmas is less subject than many other volatiles to near-surface perturbation of primary magmatic concentrations, and may therefore be one of the most useful proxies for the volatile content of subaerial lavas. Preliminary work by the investigators suggests that nominally anhydrous minerals (NAMs) may be the principal mineralogical host for terrestrial F, as has also been inferred for H. Furthermore, the original F contents of olivine megacrysts appear to be more faithfully recorded than the H contents. These data suggest a hypothesis that F is systematically partitioned between NAMs and may be used to better understand not only F geochemistry of the mantle, but also the behavior of H in natural mantle minerals. We have 4 main objectives in this work: 1) to determine the abundance of F in NAMs (including olivine, garnet, and pyroxene) from different parts of the upper mantle, 2) to show how major and minor elements (including H) are correlated with F in NAMs in order to infer F behavior during mantle processes (subduction-related melting and metasomatism) and crystal-chemical controls on F incorporation, 3) to measure the distribution of F between NAMs and quenched liquids from experiments to determine partition coefficients, and 4) to measure the compositional gradients of fluorine in new and existing experiments to quantify diffusion coefficients in NAMs. An important part of this proposal is using our understanding of the history of samples to investigate F distribution in a controlled petrologic context. Ideally, the results of this work will be to allow the use of trace fluorine contents in NAMs to help determine the processes responsible for forming the rocks that contain them.
