Browsing by Author "Berry, AJ"

Now showing 1 - 3 of 3
  • No Thumbnail Available
    Item
    Oxidation state of uranium in mantle melts
    (Elsevier; Cambridge Publications, 2008-07) Berry, AJ; O'Neill, HSC; Foran, GJ
    The oxidation state of U exerts a first order control on its partitioning during partial mantle melting. An understanding of U partitioning, and its abundance relative to other members of the U decay series, is important for interpreting U series disequilibrium, with implications for the rate of melt transport at mid-ocean ridges. U LIII-edge X-ray absorption near edge structure (XANES) spectra were recorded for a synthetic mid-ocean ridge basalt (MORB), and a number of CMAS (CaO-MgOAl2O3-SiO2) compositions, containing 0.5 wt % U3O8, equilibrated at 1400 ˚C and controlled oxygen fugacities (fO2)at one-atmosphere. Spectra were obtained for both quenched glasses, and in situ at magmatic temperatures (Berry et al. 2003). In situ measurements allowed changes in the spectra in response to the imposed fO2 to be followed in real time. The spectra were recorded in fluorescence mode at the Australian National Beamline Facility (beamline 20B), Photon Factory, Japan. The spectra exhibit systematic variations in absorption edge energy and crest intensity with fO2. These changes occur between -4 and +6 log units of the nickel-nickel oxide (NNO) fO2 buffer. The fO2 range of the spectral variation indicates that U4+ must oxidise to U6+ through two stepwise one-electron reactions, involving a U5+ intermediate, rather than a direct two-electron process. The results also suggest that U5+ may be the dominant oxidation state at the fO2 conditions of MORB generation. © 2008 Elsevier Ltd
  • No Thumbnail Available
    Item
    Titanium substitution mechanisms in forsterite
    (Elsevier, 2007-07-30) Berry, AJ; Walker, AM; Hermann, J; O'Neill, HSC; Foran, GJ; Gale, JD
    Ti K-edge X-ray absorption near edge structure (XANES) and extended X-ray absorption fine structure (EXAFS) spectra, together with atomic scale calculations, indicate that Ti occupies the Si site in anhydrous forsterite. The energy and intensity of the 1s→3d transition in the XANES spectrum is diagnostic of Ti on a four-coordinate site. The EXAFS spectra determine a Ti–O bond length of 1.81±0.01 Å. This value is in excellent agreement with that of 1.8 Å determined computationally by periodic density functional theory and an embedded cluster approach for Ti on the Si site; the calculated bond length for Ti on a Mg site is 2.0 Å. Both computational methods further find that the direct substitution of Ti for Si is energetically favoured relative to substitution of Ti for Mg, charge balanced by either a Mg vacancy or Mg on a Si site. Together the results provide unambiguous evidence for Ti occupying the Si site in anhydrous forsterite. Ti-bearing olivine synthesised at upper mantle conditions in the presence of water, however, exhibits a pre-edge feature corresponding to six-fold coordination. This is consistent with the identification of a Ti-clinohumite-like point defect in samples of olivine from upper-mantle spinel peridotite. The change in Ti site thus provides a mechanism for the incorporation of water in olivine. The total Ti content of olivine will comprise varying contributions from Mg2[4]TiO4 and Ti-clinohumite-like Mg[6]TiO2(OH)2 substitutions. © 2007, Elsevier Ltd.
  • No Thumbnail Available
    Item
    XANES calibration for determining the oxidation state of iron in mantle garnet
    (Elsevier, 2010-11-01) Berry, AJ; Yaxley, GM; Woodland, AB; Foran, GJ
    Iron K-edge X-ray absorption near edge structure (XANES) spectra were recorded for synthetic garnets of the almandine–skiagite (Fe3Al2Si3O12–Fe3Fe2Si3O12), andradite–skiagite (Ca3Fe2Si3O12–Fe3Fe2Si3O12), Fe–knorringite–skiagite (Fe3Cr2Si3O12–Fe3Fe2Si3O12), and andradite–grossular (Ca3Fe2Si3O12–Ca3Al2Si3O12) solid solutions. Fe3+/ΣFe varied systematically in these samples from 0 to 1. The variation in the energy and intensity of spectral features as a function of Fe3+/ΣFe was investigated to identify correlations that could be used to determine Fe3+/ΣFe of unknowns. The pre-edge energy, which is commonly used for quantification, was found to be relatively insensitive for garnet, particularly at low values of Fe3+/ΣFe. The best correlation was for the absorption edge energy, which may provide an accurate and precise method for determining Fe3+/ΣFe of garnets in metapelitic rocks. The resulting calibration curve, however, significantly and non-systematically overestimates the Fe3+/ΣFe value of mantle garnets (from xenoliths and megacrysts) for which Fe3+/ΣFe had been determined previously by Mössbauer spectroscopy. This is probably a result of differences in composition between the synthetic and natural garnets. For the mantle garnets, Fe3+/ΣFe is strongly correlated with the intensity ratio of post-edge features at 7138.4 and 7161.7 eV. The deviation of the Mössbauer results from a linear fit is less than 0.01 (1 σ). This suggests a new method for determining Fe3+/ΣFe from XANES spectra that does not require precise energy calibration or spectral fitting. The accuracy and precision for mantle garnets (within the compositional range studied: 0.75 ≤ Mg/(Mg + Fe) ≤ 0.86, 3.7–6.3 wt.% CaO, and 0.3–7.4 wt.% Cr2O3) is comparable to those of Mössbauer spectrosopy, however, XANES spectra can be acquired in ~ 15 min, allowing large numbers of samples to be analysed or the distribution of Fe3+/ΣFe to be mapped with micron spatial resolution. © 2010, Elsevier Ltd.