No abstract

[1] We report analyses of hydrogen abundance in experimentally annealed and natural mantle minerals using FTIR and use these data to establish calibration lines for measurement of H 2 O concentrations in olivine, pyroxenes, garnet, amphibole and mica by secondary ion mass spectrometry (SIMS). We have reduced the detection limit for H 2 O analysis by SIMS to 2-4 ppm H 2 O (by weight) through careful attention to sample preparation and vacuum quality. The accuracy of the SIMS calibrations depends on the choice of FTIR extinction coefficients; however, all of the calibrations reported here are shown to be consistent with measurements on standards whose H 2 O abundance has been determined independently via manometry or nuclear reaction analysis. The resulting calibrations are accurate to 10-30% at the 95% confidence limit, with improvements possible through the use of higher-H 2 O standards. Using our SIMS calibration, we determined hydrogen concentrations in coexisting olivine, orthopyroxene, and glass from a single melting experiment at 2 GPa and 1380°C. Olivine/melt and orthopyroxene/melt partition coefficients are equal to 0.0020 ± 0.0002 and 0.0245 ± 0.0015, respectively, and the orthopyroxene/olivine coefficient is 12 ± 4 (2s uncertainties).

Basaltic lavas erupted at some oceanic intraplate hotspot volcanoes are thought to sample ancient subducted crustal materials. However, the residence time of these subducted materials in the mantle is uncertain and model-dependent, and compelling evidence for their return to the surface in regions of mantle upwelling beneath hotspots is lacking. Here we report anomalous sulphur isotope signatures indicating mass-independent fractionation (MIF) in olivine-hosted sulphides from 20-million-year-old ocean island basalts from Mangaia, Cook Islands (Polynesia), which have been suggested to sample recycled oceanic crust. Terrestrial MIF sulphur isotope signatures (in which the amount of fractionation does not scale in proportion with the difference in the masses of the isotopes) were generated exclusively through atmospheric photochemical reactions until about 2.45 billion years ago. Therefore, the discovery of MIF sulphur in these young plume lavas suggests that sulphur--probably derived from hydrothermally altered oceanic crust--was subducted into the mantle before 2.45 billion years ago and recycled into the mantle source of Mangaia lavas. These new data provide evidence for ancient materials, with negative Δ(33)S values, in the mantle source for Mangaia lavas. Our data also complement evidence for recycling of the sulphur content of ancient sedimentary materials to the subcontinental lithospheric mantle that has been identified in diamond-hosted sulphide inclusions. This Archaean age for recycled oceanic crust also provides key constraints on the length of time that subducted crustal material can survive in the mantle, and on the timescales of mantle convection from subduction to upwelling beneath hotspots.

[1] Abstract: We studied trace element geochemistry and petrology of the crust-mantle transition zone (MTZ) in the Samail massif of the Oman ophiolite to constrain the location where different primitive magmas mix beneath an oceanic spreading ridge. The MTZ is the deepest location where crystallization took place and thus is an ideal place to determine the compositional diversity of melts leaving the mantle, with various sources and degrees of depletion. We have reached three main conclusions: (1) More than 90% of our samples record equilibration with compositionally indistinguishable parental melts, similar to mid-ocean ridge basalts (MORB) and the melts that formed the crust in Oman. This suggests that mixing of diverse, polybaric partial melts of mantle peridotite occurred at or below the depth of the MTZ. The presence of distinct heterogeneity in less than 10% of our samples is similar to the nature and frequency of heterogeneity observed in melt inclusions in olivines from MORB. (2) Among the samples recording trace element equilibrium with MORB-like liquids are wehrlitic rocks, previously suggested to be cumulates from an unusual parental melt on the basis of petrological observation. (3) Systematics of Eu distribution among plagioclase and clinopyroxene in ''impregnated peridotites'' demonstrate that these minerals did not crystallize from ''trapped melt.'' As a consequence, it is not possible to use the modal proportion or texture of plagioclase + clinopyroxene impregnations to estimate the instantaneous melt porosity or the shape of melt pores at any time during the formation of these rocks.

No abstract

International audienc

Pb) mantle end-member, thought to result from recycled oceanic crust. Complete geochemical characterization of the HIMU mantle end-member has been inhibited due to a lack of deep submarine glass samples from HIMU localities. We homogenized olivine-hosted melt inclusions separated from Mangaia lavas and the resulting glassy inclusions made possible the first volatile abundances to be obtained from the HIMU mantle end-member. We also report major and trace element abundances and Pb-isotopic ratios on the inclusions, which have HIMU isotopic fingerprints. We evaluate the samples for processes that could modify the volatile and trace element abundances postmantle melting, including diffusive Fe and H 2 O loss, degassing, and assimilation. H 2 O/Ce ratios vary from 119 to 245 in the most pristine Mangaia inclusions; excluding an inclusion that shows evidence for assimilation, the primary magmatic H 2 O/Ce ratios vary up to 200, and are consistent with significant dehydration of oceanic crust during subduction and long-term storage in the mantle. CO 2 concentrations range up to 2346 ppm CO 2 in the inclusions. Relatively high CO 2 in the inclusions, combined with previous observations of carbonate blebs in other Mangaia melt inclusions, highlight the importance of CO 2 for the generation of the HIMU mantle. F/Nd ratios in the inclusions (30 6 9; 2r standard deviation) are higher than the canonical ratio observed in oceanic lavas, and Cl/K ratios (0.079 6 0.028) fall in the range of pristine mantle (0.02-0.08).

No abstract