Search

Halogens show a range from moderate (F) to highly (Cl, Br, I) volatile and incompatible behavior, which makes them excellent tracers for volatile transport processes in the Earth’s mantle. Experimentally determined fluorine and chlorine partitioning data between mantle minerals and silicate melt enable us to estimate Mid Ocean Ridge Basalt (MORB) and Ocean Island Basalt (OIB) source region concentrations for these elements. This study investigates the effect of varying small amounts of water on the fluorine and chlorine partitioning behavior at 1280 °C and 0.3 GPa between olivine and silicate melt in the Fe-free CMAS+F–Cl–Br–I–H2O model system. Results show that, within the uncertainty of the analyses, water has no effect on the chlorine partitioning behavior for bulk water contents ranging from 0.03 (2) wt% H2O (DCl ol/melt = 1.6 ± 0.9 × 10−4) to 0.33 (6) wt% H2O (DCl ol/melt = 2.2 ± 1.1 × 10−4). Consequently, with the effect of pressure being negligible in the uppermost mantle (Joachim et al. Chem Geol 416:65–78, 2015), temperature is the only parameter that needs to be considered for the determination of chlorine partition coefficients between olivine and melt at least in the simplified iron-free CMAS+F–Cl–Br–I–H2O system. In contrast, the fluorine partition coefficient increases linearly in this range and may be described at 1280 °C and 0.3 GPa with (R 2 = 0.99): Dol/meltF=3.6±0.4×10−3×XH2O(wt%)+6±0.4×10−4. The observed fluorine partitioning behavior supports the theory suggested by Crépisson et al. (Earth Planet Sci Lett 390:287–295, 2014) that fluorine and water are incorporated as clumped OH/F defects in the olivine structure. Results of this study further suggest that fluorine concentration estimates in OIB source regions are at least 10% lower than previously expected (Joachim et al. Chem Geol 416:65–78, 2015), implying that consideration of the effect of water on the fluorine partitioning behavior between Earth’s mantle minerals and silicate melt is vital

In this experimental study we obtained new mineral/melt (DF = cmineral/cmelt) partitioning data for fluorine in a bimineralic hydrous eclogite under Earth's upper mantle conditions (4–6 GPa, 1460–1550 °C). Omphacitic clinopyroxene displays mineral/melt partition coefficients between DF = 0.056 ± 0.005 and DF = 0.074 ± 0.001. Garnet partition coefficients are consistently lower with an average partition coefficient of DF = 0.016 ± 0.003. We found that omphacitic clinopyroxene is the dominant nominally fluorine-free phase in subducted oceanic crust and hence omphacite is expected to be the major fluorine carrier during subduction of crust into the deeper mantle. Together with previously obtained partitioning data we propose that the oceanic crust can host more fluorine per mass unit than the underlying depleted oceanic mantle. If the majority of entrained fluorine is recycled into Earth's transition zone it is possible that the fluorine is either incorporated into high-pressure transition zone phases or released during high-pressure phase transformations and forming fluorine-rich small degree partial melts. Both scenarios are supported by elevated fluorine concentration in ocean island basalts, kimberlites, and lamproites. Combining the fluorine partitioning data with water partitioning data yields a plausible process to generate lamproitic magmas with a high F/H₂O ratio. The enrichment of fluorine relative to H₂O is triggered by multiple episodes of small degree melting that deplete the residual more in H₂O than in fluorine, caused by the approximately three times smaller mineral-melt partition coefficients of H₂O.

We report the results of a high-pressure high-temperature (HPHT) experimental investigation into the deformation of diamonds using the D-DIA apparatus. Electron backscatter diffraction (EBSD) data confirm that well-defined 300-700 nm wide {111} slip lamellae are in fact deformation micro-twins with a 60 degrees rotation around a < 111 > axis. Such twins formed at high confining pressures even without any apparatus-induced differential stress; mechanical anisotropy within the cell assembly was sufficient for their formation with very little subsequent lattice bending (<1 degrees per 100 mu m). When apparatus-induced differential stresses were applied to diamonds under HPHT conditions, deformation twin lamellae were generated, and continuous and discontinuous crystal lattice bending occurred (4-18 degrees per 100 mu m), including bending of the {111} twin lamellae. The (111) < 011 > slip system dominates as expected for the face-centred cubic (FCC) structure of diamond. Slip occurs on multiple {111} planes resulting in rotation around < 112 > axes. Deformation microstructure characteristics depend on the orientation of the principal stress axes and finite strain but are independent of confining pressure and nitrogen content. All of the uniaxially deformed samples took on a brown colour, irrespective of their initial nitrogen characteristics. This is in contrast to the two quasi-hydrostatic experiments, which retained their original colour (colourless for nitrogen free diamond, yellow for single substitutional nitrogen, Type lb diamond) despite the formation of (111) twin lamellae. Comparison of our experimental data with those from two natural brown diamonds from Finsch mine (South Africa) shows the same activation of the dominant slip system. However, no deformation twin lamellae are present in the natural samples. This difference may be due to the lower strain rates experienced by the natural samples investigated. Our study shows the applicabili, AuthorCount:9