H.St.C. O'Neill

The partitioning of Fe, Ni and Co between olivine, metal, and basaltic liquid: An experimental and thermodynamic investigation, with application to the composition of the lunar core

Abstract The partitioning of Fe, Ni and Co between Mg-rich olivine, lunar basaltic liquid and metal has been measured over a range of temperatures and pressures. The results of the olivine/metal partitioning may be compared to those predicted from thermodynamic calculation; in general, the agreement is good, although the calculated distribution coefficients are slightly greater then those experimentally determined. This discrepancy increases with increasing temperature and pressure, but it is not possible to ascertain unambiguously which thermodynamic data might be responsible. At lower temperatures, inversion of the equilibria to yield “cosmothermometers” for measuring the temperature of equilibration between olivine and metal gives results for the pallasite meteorites in excellent agreement with independent estimates. The results have been applied to Apollo 15 Green Glass, presumed to approximate a primitive melt from the lunar mantle, to deduce the composition of Fe-Ni-Co metal in equilibrium with the lunar mantle. This composition is approximately (by weight) 54–60% Fe, 38–45% Ni and 1% Co; consequently, if metal has separated from the lunar mantle under equilibrium conditions to form a small lunar core, this core will be nickel rich, with a Ni/Fe ratio of ~0.7 ±0.15.

Diamondiferous peridotite xenoliths from the Argyle (AK1) lamproite pipe, Western Australia

This paper describes a suite of peridotite xenoliths. some carrying diamonds at high grades, from the richly diamondiferous early Proterozoic (≈1180 Ma) Argyle (AK1) lamproite pipe, in northwestern Australia. The peridotites are mostly coarse garnet lherzolites but also include garnet harzburgite, chromite — garnet peridotite, a garnet wehrlite, and an altered spinel peridotite with extremely Cr-rich chromite. In all cases the garnet has been replaced by a kelyphite-like, symplectic intergrowth of Alrich pyroxenes, Al-spinel and secondary silicates. The peridotites have refractory compositions characterized by high Mg/(Mg+Fe) and depletion in lithophile elements (Al2O3 and CaO < 1%, Na2O≤0.03%) and high field strength cations such as Ti, Zr, Y, and Yb. Olivines have high Mg/(Mg+Fe) (Mg≠91–93) and, like olivine inclusions in diamonds from the Argyle pipe, contain detectable amounts of Cr2O3 (0.03%–0.07%) but have very low CaO contents (typically 0.04%–0.05%). Enstatites (Mg≠92–94) have comparatively high Cr2O3 (0.2%–0.45%) and Na2O (up to 0.18%) but very low Al2O3 contents (0.5%–0.7%). Diopsides (Mg≠92–94, Ca/(Ca+Mg+Fe)=0.37–0.43) are Cr-rich (0.7%–1.9% Cr2O3) and have low Al2O3 (0.7%–2.2%) and Na2O (0.5%–1.6%) contents. Many have high K2O contents, typically 0.1%–0.4% but up to 1.3% K2O in one xenolith. The chromite coexisting with former garnet is Mg-and Cr-rich [Mg/(Mg+Fe2+)=0.68–0.72, Cr/(Cr+Al)=0.72–0.79] whereas chromite in the spinel peridotite is even more Cr-rich (65% Cr2O3, Cr/(Cr+Al)=0.85, resembling inclusions in diamond. One highly serpentinized former garnet peridotite contains a Cr-rich (up to 13% Cr2O3) titanate resembling armalcolite but containing significant K2O (1%–2.5%), CaO (0.6%–2.2%), ZrO2 (0.1%–0.8%), SrO (0.1%–0.3%), and BaO (up to 0.58%): this appears to have formed as an overprint of the primary mineralogy. Temperatures and pressures estimated from coexisting pyroxenes and reconstructed garnet compositions indicate that the garnet lherzolites equilibrated at 1140°–1290° C and 5.0–5.9 GPa (160–190 km depth), within the stability field of diamond. Oxygen fugacties within the diamond forming environment are estimated from spinel-bearing assemblages to be reducing, with fO2 between MW and IW. The presence of significant K in the diopsides from the peridotite xenoliths and in diopsides from heavy mineral concentrate from the Argyle pipe implies metasomatic enrichment of the subcontinental lithosphere within the diamond stability field. The P-T conditions estimated for the Argyle peridotites demonstrate that diamondiferous lamproite magmas incorporate mantle xenoliths from similar depths to kimberlites in cratonic settings, and imply that Proterozoic cratonized orogenic belts can have lithospheric roots of comparable thickness to beneath Archaean cratons. These roots lie at the base of the lithosphere within the stability field of diamond. The xenoliths, the calcic nature of chrome pyropes from heavy mineral concentrate, and the diamond inclusion assemblage indicate that the lighosphere beneath the Western Australian lamproites is mostly depleted lherozolite rather than the harzburgite commonly found beneath Archaean cratons. Nevertheless, the dominance of eclogitic paragenesis inclusions in Argyle diamonds indicates a significant proportion of diamondiferous eclogite is also present. The form, mineral inclusion assemblage, and the C-isotopic composition of diamonds in the peridotite xenoliths suggest that disaggregated diamondiferous peridotites are the source of the planar octahedral diamonds which constitute a minor component of the Argyle production. These diamonds are believed to have formed from mantle carbon in reduced, refractory peridotite (Iherzolite-harzburgite) in contrast to the predominant strongly 13C-depleted eclogitic suite diamonds which contain a recycled crustal carbon component. The source region of the lamproites has undergone long-term (≥2 Ga) enrichment in incompatible elements.

Experimental determination of activity-composition relations in Ni2SiO4− Mg2SiO4 and Co2SiO4-Mg2SiO4 olivine solid solutions at 1200 K and 0.1 MPa and 1573 K and 0.5 GPa

Abstract Activity-composition relations in the olivine solid solutions Ni 2 SiO 4 − -Mg 2 SiO 4 and Co 2 SiO 4 -Mg 2 SiO 4 have been determined at 1200 K and 0.1 MPa and at 1573 K and 0.5 GPa by equilibration with the corresponding oxide solutions. Both olivine solutions show small positive deviations from ideal (two site) mixing, which, within the limits of accuracy of the method, may be described by the simple regular solution model with parameters W Ni + Mg ol = 0.35 ± 1.0 kJ / g -atom and W Co - Mg ol = 1.37 ± 0.9 kJ / g -atom. The requirements of internal consistency between the two systems also show that the recent determination by Brousse et al . (1984) of the enthalpy of formation of Mg 2 SiO 4 is to be preferred over earlier work, and that their value is also probably more accurate than the uncertainty in their own measurements indicates; activities in the NiO-MgO system are close to ideal.

Thermodynamics of Co3O4: a possible electron spin unpairing transition in Co3+

AbstractThe free energy of the reaction:

Standard molar Gibbs free energies of formation of the tetragonal and hexagonal forms of germanium dioxide

Co_3 O_4 \rightleftarrows 3C_O O + \tfrac{1}{2}O_2

High pressure Raman spectroscopy of spinel-type ferrite ZnFe2O4

has been studied between 890 and 1,240 K using an e.m.f. technique. There is a phase transition in Co3O4 at 1,120±20 K which is accompanied by a large change in entropy (∼47 JK−1 mol−1 of Co3O4), and a rapid increase in unit cell volume and in electical conductivity. This is interpreted to be due to a partial change in electronic spin states in Co3 + from the spin-paired (low spin) configuration observed at room temperature to the spin-unpaired (high spin) state. The transition is probably not first order.

Experimental tests of low degree peridotite partial melt compositions: implications for the nature of anhydrous near-solidus peridotite melts at 1 GPa

Abstract The effect of melt composition on the solubility of Ni in haplobasaltic melts was investigated at 1 atm via additions of SiO2 (Quartz), Mg2SiO4 (Forsterite), and Na2SiO3 (Na-metasilicate) to the 1 atm diopside-anorthite eutectic melt composition. The experimental method applied was the mechanically assisted equilibration technique of Dingwell et al. (1994). Quenched glass samples were analyzed for major elements and Ni by the electron microprobe, and Ni was also determined in all samples by ICP-AES techniques. The Ni solubility and hence activity of NiO is not influenced by the addition of either SiO2 or Mg2SiO4 to the 1 atm anorthite-diopside eutectic composition. In contrast, addition of Na2SiO3 leads to a decrease in the Ni solubility up to 20 wt% of Na2SiO3 in the melt. Further addition results in a reversal of this trend, i.e., an increase of the Ni solubility up to about 45 wt% Na2SiO3. This observation is likely due to an exchange reaction of 2 Na for each Ca and/or Mg atom stabilizing aluminate tetrahedra in the melt structure. After all aluminate tetrahedra are stabilized by Na atoms, further addition of Na-metasilicate results in a network-modifying role of Na in the melt structure, creating new coordination possibilities for Ni. Activity coefficients for NiO in these melts have been calculated and compared with the literature. The present work leads to the conclusion that melt composition plays a minor role in determining the melt-olivine Ni partitioning. Other factors, amongst them temperature, are expected to play a much larger role.

Three water sites in upper mantle olivine and the role of titanium in the water weakening mechanism

The standard molar Gibbs free energies of formation of both the tetragonal and hexagonal forms of GeO2 have been determined in the temperature range 860 to 1070 K using solid-state galvanic cells. The results are: ΔfGmo(GeO2, hex.)/(J·mol−1)±120 = −547394+167.914(TK) and ΔfGmo(GeO2, tet.)/(J · mol−1)±110 = −568392+184.500(TK). These results predict the temperature of transformation from the tetragonal to the hexagonal form to be (1270±15) K in excellent agreement with preferred literature values.(1, 2)