Trace element compositions of olivine from 75 mantle rocks of diverse origin, including xenoliths from kimberlites, basaltic lavas and orogenic peridotites, were determined by laserablation ICP-MS to study systematic variations between mantle lithologies, partitioning mechanisms in olivine and the potential of olivine for geothermobarometry and petrogenic mantle processes. Samples were selected to provide a wide range of forsterite contents (89.1-93.4), equilibration temperatures and pressures (750-1450°C; 15-80 kbar). Trace elements in olivine can be divided into three groups. Group I elements (Ni, Mn, Co, Cu, Zn, Li) show small concentration ranges and olivine is the major host mineral. These are mostly divalent elements and have ionic radii close to that of Mg. Group II elements (Cr, Al, V, Sc, Ca, Na) show large concentration ranges, which are mainly controlled by the equilibration temperature of the host rock. The elements are strongly concentrated in coexisting mantle minerals (garnet, clinopyroxene, spinel) and show a narrow range of bulk rock concentrations. They fit less comfortably in the olivine lattice than Group I elements because of their charge or size. Differences between garnet and spinel-facies rocks are apparent for Al, Ca and Sc. Group III elements (Ti, Zr, Nb, Y) show large ranges of concentration in olivine as well as in co-existing minerals, and are strongly dependent on bulk rock contents. Concentration differences between olivine from garnet and spinel-facies rocks are apparent for all these elements. They are strongly incompatible in olivine and other rock-forming mantle minerals because of their charge or size. Various mantle lithologies can be discriminated using olivine composition. Spinel, garnet and garnet-spinel peridotites can be distinguished in olivine Sc-Zr and MnO-Al 2 O 3 diagrams, whereas volcanic olivine is distinguished by high Ca and Al contents (picritic olivine) or high Nb contents (kimberlitic olivine). Since concentrations of Group III elements in olivine are diagnostic of whole-rock contents they can be used to trace the petrogenetic history of the  with P in kbar, Al Ol the concentration of Al in olivine in ppm, and Cr# Ol is Cr/(Cr+Al) in olivine. This thermometer predicts the temperature with a residual of 15°C based on calibration with two-pyroxene and Al-in-Opx geothermobarometers (Brey and Kohler, 1990). Although calibrated using lherzolites only, the thermometer performs well for clinopyroxenefree harzburgites and even spinel peridotites. An alternative thermometer is presented for the case where the presence of Cr 2+ is expected, e.g., for olivine inclusions in diamonds. The geochemical and thermobarometric information recorded by olivine can be a useful tool in studies of the petrogenesis of lithospheric mantle, olivine xenocrysts in mantle-derived magmas, the formation of diamonds and diamond exploration using detrital olivine.

. (2013) 'Nickel isotopic compositions of ferromanganese crusts and the constancy of deep ocean inputs and continental weathering eects over the Cenozoic.', Earth and planetary science letters., 375 . pp. 148-155. Further information on publisher's website: Use policyThe full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or charge, for personal research or study, educational, or not-for-prot purposes provided that:• a full bibliographic reference is made to the original source • a link is made to the metadata record in DRO • the full-text is not changed in any way The full-text must not be sold in any format or medium without the formal permission of the copyright holders.Please consult the full DRO policy for further details. AbstractThe global distribution of nickel (Ni) isotope values in ferromanganese crusts has been investigated by analysing 24 surface (zero-age) samples of crusts from various oceans by MC-ICPMS, using double-spike for mass bias correction. The isotopic compositions range between 0.9 and 2.5  (60Ni), with averages for the Atlantic samples permil ± error being slightly lighter than the ones at calculated from the Pacific samples (xy permil ± error), although the data show large overlapsre is considerable overlap in the Ni isotope compositions of waters from between all the measured oceans basins mea-sured. To complement the crustal surface study also a depth profile through one crust from the Pacific ocean was also analysed. This isotope profile suggest that large variations in nickel isotope composition (1.62.3 ) has occurred within this crust throughout the last 70 Myrs, although the mean value is similar to the average isotopic composition for ferromanganese crust surfaces from the Pacific Ocean.

We present a new technique for the accurate and precise determination of mass-dependent variations in nickel isotope compositions in geological materials. Our method involves an ion-exchange procedure comprising three columns and utilising the ability of Ni to form strong complexes with both ammonia and dimethylglyoxime. The separation procedure is independent of sample pH and works even for samples with large matrix to analyte ratios. Processed Ni solutions are free of matrix elements and direct isobars of Ni, and the yield is normally 85-95%. The purified Ni solutions were analysed using a Nu Plasma, multi-collector inductively coupled plasma mass spectrometer (MC-ICPMS), where instrumental mass fractionation-together with potential isotopic fractionation during chemical separation due to incomplete yield-was corrected for by a double-spike technique, where samples were spiked prior to column chemistry. Tests performed on both mixtures of synthetic and natural terrestrial standards demonstrates that the method is accurate. Replicate measurements of USGS reference materials (peridotite PCC-1, basalt BHVO-2, and shale SCo-1) yield a long-term external reproducibility (2 s.d.

This paper presents a detailed high-precision study of Ni isotope variations in mantle peridotites and their minerals, komatiites as well as chondritic and iron meteorites. Ultramac rocks display a relatively large range in δ 60 Ni (permil deviation in 60 Ni/ 58 Ni relative to the NIST SRM 986 Ni isotope standard) for this environment, from 0.15±0.07 to 0.36±0.08, with olivine-rich rocks such as dunite and olivine cumulates showing lighter isotope compositions than komatiite, lherzolite and pyroxenite samples. The data for the mineral separates shed light on the origin of these variations. Olivine and orthopyroxene display light δ 60 Ni whereas clinopyroxene and garnet are isotopically heavy. This indicates that peridotite whole-rock δ 60 Ni may be controlled by variations in modal mineralogy, with the prediction that mantle melts will display variable δ 60 Ni values due to variations in residual mantle and cumulate mineralogy. Based on fertile peridotite xenoliths and Phanerozoic komatiite samples it is concluded that the upper mantle has a relatively homogeneous Ni isotope composition, with the best estimate of δ 60 Ni mantle being 0.23±0.06 (2 s.d.). Given that >99% of the Ni in the silicate Earth is located in the mantle, this also denes the Ni isotope