Reaction between silica or alkali-rich melts and mafic minerals - small scale kinetic fractionation with applications to global-scale mantle metasomatism (TP11)

点击翻译按钮获取中文摘要

Mantle metasomatism occurs by interaction of fluids (melts) with olivine-dominated mantle wallrock. The metasomatising agents are thought to be silicic or carbonatitic melts, with the first group prevailing. Non-basaltic silicic melts may be derived from partial melting of eclogite, i.e. deeply subducted oceanic crust. Metasomatised mantle rocks in ophiolites or xenoliths within volcanic systems contain information about the scales and mechanisms of chemical re-distribution within the Earth s mantle (like, f. ex., diffusion vs. reacting flow). Metasomatised mantle may melt at lower temperatures than pristine mantle rocks and thus have a strong imprint on the composition of directly observable volcanic rocks. In this proposal I ask for support of experiments that aim at the mechanisms of metasomatic exchange between melt and mafic minerals. The experiments will be focused on silicate melts that are out of equilibrium with mafic mantle minerals in two different ways: - SiO2-rich melt that reacts with olivine; - alkali-rich and Mg-poor melt that reacts with olivine and orthopyroxene. SiO2 metasomatism is a process that apparently occurs worldwide in various tectonic settings. Alkali-metasomatism with formation of Na(K)-amphibole, mica, diopside and other minerals at the expense of original mantle phases also occurs widespread as deduced from mantle xenoliths. The motivation for the suggested research comes from two different directions: - In experiments in the system olivine-quartz it was shown that the chemical composition of the evolving orthopyroxene rim is kinetically controlled and largely depends on grain boundary diffusion. Major and trace element concentratios in the orthopyroxene rims may deviate substantially from the values expected in bulk equilibrium between Opx and Ol. In these experiments the Opx/Qz interface is closed for element exchange other than the SiO2 component. This will be clearly different in the situation more relevant to mantle where the source of SiO2 is a silicic melt, but kinetic fractionation in such melt-olivine rim growth experiments is still expected; - Reaction rims between olivine and orthopyroxene and phonolite melt are recently investigated from the Heldburg phonolite, Thuringia. They consist of assemblies dominated by K-richterite-magnesian mica-diopsidic clinopyroxene that are very similar to several types of mantle metasomatites. The Heldburg phonolite plug is considered a natural experimental petrology lab . Controlled experimentation may answer open questions concerning these reaction rims, but first of all will give insight into the systematics of element exchange between alkaline melt and mafic mantle wallrock. In our experiments we will start out with chemically simple starting materials and single dopants and successively will introduce more chemical complexity. We will focus on the grain scale and grain boundary effects. This approach is complementary to earlier attempts of the experimental replication of mantle metasomatism in the lab using natural starting materials.