Metal-silicate Partitioning Experiments Research Articles

Metal–silicate partitioning experiments in the diamond anvil cell: A comment on potential analytical errors

We have performed simulations of the electron microbeam analysis of metal–silicate partitioning experiments in which a small (∼50μm) ball of metal is embedded in a silicate matrix. This geometry approximates that found in many high pressure diamond-cell experiments aimed at determining the partition coefficients (Di) of elements between metal and silicate. The simulations involved using the (publicly available) Monte Carlo simulation package PENEPMA which is based on the general-purpose electron–photon transport code PENELOPE. The code allows complex geometrical structures to be defined by quadric surfaces, which enclose homogeneous materials specified by the user. The package simulates both electron and photon transport, keeps track of the X-ray production mechanism and includes the continuum contribution.Our principal results came from simulating the effects of secondary fluorescence on the apparent metal–silicate partition coefficient of Ni for the case of an Ni-rich metal (26.8% Ni, 73.2% Fe) embedded in a silicate matrix. If, as is frequently the case, analysis of the silicate is performed 5–10μm from the silicate–metal interface then secondary fluorescence of Ni in the metal will increase the apparent NiO concentration in the silicate by ∼1000ppm. In the case considered, this is 25% of the Ni content of the silicate and the fluorescence effect decreases the apparent DNi from 110 to 84. For this type of experiment, therefore, the fluorescence effect reduces the apparent partition coefficient.We performed similar simulations of the analysis of the centre of a 10μm diameter metal ball embedded in the silicate matrix to determine apparent Si and O concentrations in the metal. In this case, because the electrons do not penetrate so far into the metal, the fluorescence effect is very small and can be neglected.Finally, we show that focused ion beam (FIB) sections, while providing the potential physically to separate metal from silicate present additional, but not insurmountable, problems of standardisation. The best results would be obtained with a section >3μm in thickness.We recommend that, in general, experimentalists consider simulating analysis of a typical experimental product before publication of results which may be affected by secondary fluorescence.

Oxygen and silicon contents of Earth's core from high pressure metal–silicate partitioning experiments

Oxygen and silicon partitioning between molten metal and silicate melts was measured in samples synthezised in piston-cylinder and multi-anvil presses between 2 and 21 GPa, 2273 and 2873 K, and at oxygen fugacities of 1.5–3.6 log units below the iron–wüstite buffer. Our partitioning data are used together with published data to parameterize the individual effects of pressure, temperature and composition on the partitioning of oxygen and silicon. Results show that the oxygen metal–silicate partition coefficient increases with increasing oxygen fugacity, temperature and pressure, whereas the silicon metal-silicate partition coefficient increases with decreasing oxygen fugacity, increasing temperature and pressure. Silicon and oxygen contents of Earth's core were derived for different core formation models. Considering single-stage core formation at 40 GPa, 3200 K, IW-2, the core would contain 1 to 3.5 wt.% silicon and 0.5 to 2.5 wt.% oxygen. In a continuous core-formation scenario, and depending on the oxidation path, Si core content varies from 1 to 11 wt.%, whereas oxygen content ranges from 0 to 2.5 wt.%. These models show that the oxygen content in the core cannot be significantly higher than 2.5 wt.%. In these compositional models, a range of combined silicon and oxygen concentrations in the core could satisfies the seismologically observed range of outer core density deficits.

Convergence of Ni and Co metal–silicate partition coefficients in the deep magma-ocean and coupled silicon–oxygen solubility in iron melts at high pressures

Models for a deep magma ocean have gained wide acceptance although with variations in the specific conditions at which core formation may have taken place. Preliminary high-pressure studies produced results consistent with metal–silicate equilibration at the base of a magma ocean that would have extended to as much as 60 GPa (corresponding to a depth of ~ 2000 km), > 2000 K and an oxygen fugacity two orders of magnitude below iron–wüstite (IW) buffer. However, up to now the magma models are based on extrapolations of low pressure (< 25 GPa) partition coefficient data that cannot be extrapolated to higher pressures. In this work, metal–silicate partitioning experiments were performed for pressures up to ~ 52 GPa and ~ 3500 K to investigate the behaviour of Ni and Co during terrestrial core formation using Laser-Heated Diamond-Anvil Cell (LHDAC) techniques. Our experimental results show that Ni and Co partitioning coefficients converge and remain similar above 30 GPa to the maximum pressure reached. In the range 30–52 GPa the data account for the relative depletions of Ni and Co ( e.g., the chondritic Ni/Co ratio) confirming evidence for a deep-magma ocean. The present results suggest a wide interval of pressure where the siderophile elements can match their mantle concentrations. We also show that both the solubilities of oxygen and silicon in molten Fe-rich alloy increase with increasing pressure. The experimental partition coefficient of Si ( D Si) together with D Ni and D Co all match the theoretical partition coefficients required for an equilibrium core–mantle differentiation at pressures above 30 GPa and for temperatures between 3000 and 3500 K.

Low osmium solubility in silicate at high pressures and temperatures

A new methodology to experimentally investigate metal/silicate partitioning for Os ( D Os metal/silicate) under high P– T conditions was devised that utilizes metal and silicate components with natural Os abundances and disparate 187Os/ 188Os. The isotopic contrast in starting components allows examination of isotopic exchange between metal and silicate. As is common for metal–silicate partitioning experiments involving highly siderophile elements (HSE), tiny nuggets of metal appeared in the silicate melts. Linear trends in the 187Os/ 188Os versus 1/ 188Os diagram for different silicate chunks from individual experimental charges permit projections of data to nugget-free silicate concentrations. The Os concentrations and isotope compositions in pure silicates reveal that the nuggets remove Os from the silicate melt while percolating through the melt, but contribute limited Os back to the silicate melt. Our data suggest that the rate of chemical equilibration regarding Os concentration between metal and silicate is much faster than that for isotopic exchange. Results for multiple experiments indicate extremely low Os solubility in silicate melt ( D Os metal/silicate > 2 × 10 5) at pressures as high as 2 GPa, and temperatures as high as 2400 °C. These results indicate that metal–silicate segregation during planetary differentiation at high temperatures and pressures may leave silicate that is extremely depleted in Os compared to the mantle abundances of Os observed, and that isotopic exchange of HSE between planetary cores and silicate mantles subsequent to primordial differentiation may be limited.

The effects of core formation on the Pb- and Tl- isotopic composition of the silicate Earth

We have performed metal-silicate partitioning experiments at 2 GPa and 1650–2180 °C to investigate the behaviour of Pb and Tl during terrestrial core formation. The aim was to test the hypothesis that metal core formation followed by late sulphide addition to the core resulted in the concentrations and isotopic compositions of Pb and Tl in the silicate Earth. We investigated D Pb met/sil and D TL met/sil as functions of the sulphur content of the metal and measured the equilibrium Tl isotope fractionation between the coexisting phases. Lead is moderately siderophile under the likely conditions (initially reducing [Wade, J., Wood, B.J., Core formation and the oxidation state of the Earth, Earth Planet. Sci. Lett. 236(2005) 78–95.]) of core segregation on Earth so that the μ( 238U/ 204Pb) of the bulk silicate Earth should have increased by a factor of 6.5 ( D Pb ∼ 13) as the core separated. In the case of Tl, core segregation should have reduced the Tl concentration of the BSE by about 50%. Neither the Pb nor Tl isotopic compositions of the bulk silicate Earth can, however, be completely explained by S-free iron core formation. Thallium isotopes were found not to be significantly fractionated by metal or sulphide separation from silicate. Addition of sulphur to the metal greatly increases metal-silicate partition coefficients for both Pb and Tl. D Pb met/sil increases by a factor of 15 and D TL met/sil by a factor of 45 as S increases from 0 to 35% in the metal phase. This means that extraction of sulphide from a molten mantle would result in D Pb sulph/sil of ∼ 40 and D TL sulph/sil of ∼ 60. We used the latter results to calculate the effects of late sulphide extraction on the Pb and Tl isotopic compositions of the silicate Earth. For a bulk Earth with μ of 0.7 addition of 1.6% sulphide to the core 100–140 Myr after the beginning of the solar system is sufficient to displace the Pb-isotopic composition of the silicate Earth into the region indicated by estimates in the literature. The Tl concentration and ɛ 205Tl of bulk silicate Earth are also consistent with this extent of sulphide addition. Raising μ of the Earth reduces the amount of sulphide addition to the core needed to satisfy the Pb and Tl isotopic constraints but for values of μ above 0.9 (< 0.65% sulphide) the sulphide has insufficient effect. Separation of sulphide from a mixture of silicate crystals and melt would, however, because of the incompatible nature of Pb and Tl amplify the sulphide effect. We show that, for initial μ of 0.7, as little as 0.1% sulphide extraction (at 150–200 Myr) from a mixture of 91% crystals and 9% melt would be sufficient to displace the isotopic compositions of Pb and Tl of the BSE to the required values. Late volatile loss of Pb and Tl is, therefore, not required by the partitioning and currently available isotopic data.

Relative ion yields for SIMS analysis of trace elements in metallic Fe, Fe-Si alloy, and FeSi

The ion yields of a number of siderophile trace elements dissolved in Fe, Fe-Si alloy, and FeSi have been measured using secondary ion mass spectroscopy. Ion yields are a complex function of ionisation potential, energy distribution, oxide bond strength, and matrix material. The exact nature of secondary ion yield is poorly understood, yet there are regular trends throughout the ion yield pattern across the periodic table. Metal standards were constructed that consisted of a matrix metal (Fe, Fe 83Si 17, and FeSi) doped with 10 trace elements at a nominal concentration of 1 wt% each. The standards were quantitatively analysed by electron microprobe, and the exact concentrations were used to calibrate the count rates measured by the ion microprobe. The count rates for Si and Fe increased with increasing silicon concentration despite the corresponding decrease in Fe concentration. Similarly, the O count rate also increased with increasing Si concentration despite the fact that the only implanted oxygen was that of the primary beam, which remained constant throughout the analyses. The presence of Si in the metal matrix appears to enhance the overall ionisation of Fe and O. The ion yields, relative to Fe, of the trace elements themselves vary linearly with Si concentration. They also show an overall decrease in the range of ion yields in the Si-rich metals. Comparing the ion yields for the metals to SRM610 silicate glass shows a strong relationship with the first ionisation potential of these elements. In and Ga form the most extreme end with considerably higher ion yields in metal than in silicate glass. Zn, for some unknown reason, does not appear to conform to this trend in ionisation. The relative ion yields obtained in this study may be used in the future analysis of trace elements in Fe meteorites and metal-silicate partitioning experiments.

Siderophile geochemistry of Ga, Ge, and Sn: cationic oxidation states in silicate melts and the effect of composition in iron–nickel alloys

We report a series of metal–silicate partitioning experiments for Ga, Ge, and Sn to characterize the dependence of the partition coefficient, D, on oxygen fugacity, fO2. These were isothermal (1260°C) and isobaric (1 bar) experiments using a silicate composition that approximates a eucritic meteorite. It is well known that elements such as Ni, which exist in only one valence state under redox conditions of planetary interest, produce linear trends on log D vs log fO2 diagrams. For our experiments on Ga, Ge, and Sn, however, large deviations from linearity were evident and seemed to suggest unusual changes in the oxidation states for these elements in the silicate melt. But such an inference would have been mistaken because the metallic phase of these experiments, Ni–Fe alloys, was not of constant composition; instead, the Ni/Fe ratio was varied systematically to control oxygen fugacity. Although Ni and Fe form alloys that are not far from ideal, binary interactions between the trace elements and either major metallic component, Ni or Fe, are not necessarily similar.Using only information obtained from the literature on binary mixing properties among the metallic components, Fe, Ni, Ga, Sn, and Ge, a simple thermodynamic solution model was formulated to calculate the activity coefficients for the metallic components in our experimental system. It was found that, despite the slight deviations from ideality for Ni–Fe alloys, large differences exist between the way Ni interacts with trace elements and the way Fe does. Activity coefficients calculated from the thermodynamics of the metallic solution rationalized the experimentally derived log D vs log fO2 plots. When the independently derived activity coefficients for the trace elements in the alloys (γ) are used to plot log γ KD vs log fO2, the unusual oxygen fugacity dependencies can be fully reconciled with the expected valences of 3+ for Ga, 4+ for Ge, and predominantly 4+ at high oxygen fugacities changing to predominantly 2+ at low oxygen fugacities for Sn. If a planetary metal–silicate system evolves with increasing oxygen fugacity, the increasing Ni content of the metal will maintain the siderophile character of Ga, Ge, and Sn because these elements are more stable dissolved in Ni compared to Fe.