In this project of the proposed research unit, we aim to experimentally investigate isotope fractionation in magmatic crystals, generated by chemical diffusion. In the first funding period, we aim to focus on olivine and Fe and Mg isotope fractionation, driven by Fe-Mg exchange diffusion, as well as diffusion-driven Li isotope fractionation. In a second funding period such investigations may be extended to other magmatic minerals, such as pyroxene and plagioclase, for which the diffusion rate will be investigated in other projects (# 1 and #2) of the research unit, during the first funding period.The use of isotopic zoning has recently been developed as an additional complementary tool in diffusion chronology, including Fe-Mg and Li isotope zoning in olivine. The advantage of isotopes is that at magmatic temperatures isotopes only significantly fractionate during diffusion, while equilibrium isotope fractionation is small. Isotopic zoning can thus be used to unequivocally identify diffusion-driven zoning and furthermore, in combination with chemical zoning to resolve complex magmatic scenarios in which magmatic crystals experienced several growth and diffusion stages. However, the extend of diffusion-driven isotopic zoning is yet barely known and isotopic profiles observed in natural olivines are only fitted to assumed models of isotopic fractionation during diffusion modelling. We propose the precise experimental determination of diffusion-driven isotope fractionation and its dependence on parameters such as temperature, oxygen fugacity, chemical composition and crystallographic orientation of the olivine. This will result in much better constraints on the initial boundary conditions that are assumed for the diffusion model, and consequently lead to the determination of more accurate constraints on the duration of magmatic processes. It will also help to improve estimates of diffusion-driven chemical fluxes (during partial melting or metasomatic events), based on mineral- or even bulk rock isotopic data.We will furthermore develop 3D models to predict the anisotropy of isotope fractionation profiles that should allow to identify sectioning effects in natural crystals zoning. This model and all findings on diffusion-driven isotope fractionation will be implemented into the user friendly software for diffusion modelling, which will be developed in project #6. In cooperation with project #8, we will already apply the experimentally calibrated isotope zoning, as determined in this project, for the investigation of natural olivine.
