Metal-silicate Differentiation Research Articles

Origin of fabrics and olivine chemical variations preserved in brachinite and brachinite‐like achondrite meteorites

AbstractBrachinites and brachinite‐like achondrites are olivine‐rich meteorites that represent materials after partial metal–silicate differentiation on multiple early Solar System bodies. Both meteorite types show macroscopic textures of olivine crystals, which make up >70 modal percent of their mineralogy. We investigated the orientations of olivine using electron backscatter diffraction (EBSD) and elemental compositions from paired brachinite‐like achondrites and one brachinite. The olivine orientations are characterized by a strong concentration of [010] axes with maxima perpendicular to the foliation/layering and a concentration of [001] axes distributed in a girdle or, in a few samples, as point maxima. Trace element abundances of the olivine in these meteorites determined using laser ablation inductively coupled plasma–mass spectrometry have uniformly low concentrations of sodium (<300 μg g−1), aluminum (<70 μg g−1), and titanium (<40 μg g−1) that are distinct from olivine in chondrites or within terrestrial lavas. Instead, brachinite and brachinite‐like olivine compositions broadly overlap those of olivine from melt‐depleted mantle lithologies on Earth. Evidence from olivine trace element geochemistry, in conjunction with mineral fabrics, supports that these meteorites formed as melt residues on their host planetary body(ies).

Ag isotopic and chalcophile element evolution of the terrestrial and martian mantles during accretion: New constraints from Bi and Ag metal-silicate partitioning

The Earth's timing of accretion and acquisition of moderately volatile compounds is uncertain. Hafnium-W and Mn-Cr isotopic data can bracket the timing of early planetary differentiation and core formation. The Ag-Pd system has also been utilized but its application has been limited by a lack of high pressure and temperature metal-silicate partitioning for Pd and Ag. Because Ag (and Bi) are volatile chalcophile siderophile elements, understanding their early distribution can constrain the origin of volatile elements in differentiated bodies and planets. Unfortunately, neither Ag or Bi have been studied across the wide range of pressure and temperature conditions that are relevant to accretion and core-mantle differentiation. Here, new high-pressure and temperature multi-anvil metal-silicate equilibrium experiments for Bi and Ag have been carried out at conditions relevant to planetary accretion and metal silicate differentiation that allow a more refined and complete understanding of element partitioning during core formation. The new metal-silicate partitioning data are combined with previously reported data, and utilized to predict the distributions of Bi, Pd, and Ag at conditions of accretion relevant for Earth and Mars. Application of the new partitioning results to Earth shows that D(Bi) and D(Ag) (D = metal/silicate concentration ratio) are lowered due to the effect of pressure and Si alloyed in the metallic liquid, resulting in higher predicted mantle Bi and Ag abundances than in the bulk silicate Earth (BSE), as well as high and variable Pd/Ag. The unradiogenic Ag isotopic composition of the BSE could have been generated by early accretion of volatile-poor (high Pd/Ag) pre-cursors, followed by later accretion of volatile–rich (low Pd/Ag) material, in agreement with earlier studies of Pd-Ag and Mn-Cr (Schönbächler et al., 2010). However, these main accretion phases would have to be followed by segregation of a sulfide liquid (at least 1.5% of magma ocean) at high pressures (>30 GPa), to explain the primitive upper mantle (PUM) Bi, Pd, and Ag, as well as Au, Pt, Cu and Ni concentrations as proposed previously. If the early accreted bulk Earth was volatile depleted with high Pd/Ag ratios, portions of the mantle may contain ancient domains that developed positive 107Ag isotopic anomalies (as also argued by noble gases, Nd, W, and Os isotopes). In comparison, Bi, Pd, and Ag concentrations in the martian mantle could have been set by simple metal-silicate equilibrium. Mars accreted and differentiated relatively rapidly, while also developing a deep magma ocean with a high Pd/Ag ratio that could have evolved positive 107Ag anomalies, in contrast to Earth. Measurements on shergottites may reveal these predicted Ag isotopic anomalies.

Potential of Earth’s core as a reservoir for noble gases: Case for helium and neon

doi: 10.7185/geochemlet.2028 lower mantle upper mantle OIB MORB outer core Low 3 He/ 22 Ne High 3 He/ 4 He Co re C on tr ib ut io n This study investigates metal-silicate partitioning of neon (D Ne) under the likely conditions of early Earth's core formation: up to 16 GPa, ∼ 3000 K and an oxygen fugacity near IW-2 (2 log units below the Iron-Wustite buffer). We find that the D Ne coefficients range between 10 −2 and 10 −1. These partition coefficients are only one of the controlling factors of noble gas distributions within the early Earth: because, even if D He and D Ne are low (∼10 −4), there may have been sufficient noble gases present in the mantle to supply a significant quantity of He and Ne to the core. Assuming gas-melt equilibrium of the molten proto-Earth with a nebular gas composition and con-comitant metal-silicate differentiation, the core would have inherited and maintained throughout Earth's history high 3 He/ 4 He ratios and low 3 He/ 22 Ne ratios (<0.6), making the core a potential source of primordial light noble gases in mantle plumes.

Reevaluation of metal interconnectivity in a partially molten silicate matrix using 3D microtomography

Understanding metal-silicate differentiation in small rocky bodies that accreted early in solar system history requires quantification of the effects of variable amounts of silicate melt and molten metal on the connectivity of metal-rich liquids. To shed light on this question, the equilibrium geometry and textural ripening of metal grains in the vicinity of the metal interconnection threshold have been determined experimentally. High pressure and temperature experiments were performed in the three-phase system forsterite + silicate melt + nickel at conditions of 1 GPa and up to 2080 K using piston-cylinder and Paris-Edinburgh presses. Sample textures were analyzed by 3D X-Ray microtomography either in-situ at the PSICHE beamline of the SOLEIL synchrotron, or on quenched samples using a laboratory Computed Tomography scan. Although dihedral angles point to textural equilibrium at the scale of individual grains at the end of each experiment, the attainment of textural equilibrium at sample scale is not straightforward. Depending on the relative proportions of the phases, different states of textural maturation are revealed. A particularly important issue is that time-resolved in-situ microtomography data show that cold (subsolidus) compression at the beginning of the experiment leads to soft metal grains being squeezed between silicates, leading to a forced interconnectivity of nickel. High temperature experiments with metal contents ≤20 vol% resulted in disruption of these forced networks, while the networks persisted in time for metal contents ≥25 vol%. This is taken to indicate that the stable interconnection threshold of pure nickel in a partially molten silicate matrix lies between 20 and 25 vol%. Therefore, we conclude that care must be taken when defining the interconnection threshold: not only should there be existence of a network of touching grains, but this network must persist in time in the absence of external forces and at pressure-temperature conditions that permit grain-boundary movement (i.e. excluding kinetically arrested systems). Growth of silicate grains is identified as the process driving textural maturation, and may explain the variability of interconnection thresholds reported in the literature. In addition, these considerations shed light on the diversity of textures observed in natural meteoritic samples (e.g. carbonaceous and ordinary chondrites and primitive achondrites), providing textural arguments to constrain the processes that affected these meteorites.

Two-stage formation of pallasites and the evolution of their parent bodies revealed by deformation experiments

Pallasites, stony-iron meteorites predominantly composed of olivine crystals and Fe-Ni metal, are samples of the interior of early solar system bodies and can thus provide valuable insights into the formation of terrestrial planets. However, pallasite origin is controversial, either sampling the core-mantle boundary or the shallower mantle of planetesimals that suffered an impact. We present high strain-rate deformation experiments with the model system olivine + FeS melt ± gold melt to investigate pallasite formation and the evolution of their parent bodies and compare the resulting microstructures to two samples of Seymchan pallasite. Our experiments reproduced the major textural features of pallasites including the different olivine shapes, olivine aggregates, and the distribution of the metal and sulfide phases. These results indicate that pallasites preserve evidence for a two-stage formation process including inefficient core-mantle differentiation and an impact causing disruption, metal melt injection, and fast cooling within months to years. Olivine aggregates, important constituents of angular pallasites, are reinterpreted as samples of a partially differentiated mantle containing primordial metallic melt not stemming from the impactor. The long-term retention of more than 10 vol% of metal melt in a silicate mantle sampled by olivine aggregates indicates high effective percolation thresholds and inefficient metal-silicate differentiation in planetesimals not experiencing a magma ocean stage.

Nickel isotope fractionation during metal-silicate differentiation of planetesimals: Experimental petrology and ab initio calculations

Metal-silicate fractionation of nickel isotopes has been experimentally quantified at 1623 K, with oxygen fugacities varying from 10−8.2 to 10−9.9 atm and for run durations from 0.5 to 1 h. Both kinetic and equilibrium fractionations have been studied. A wire loop set-up was used in which the metal reservoir is a pure nickel wire holding a silicate melt droplet of anorthite-diopside eutectic composition. During the course of the experiment, diffusion of nickel from the wire to the silicate occurred. The timescale to reach chemical equilibrium was fO2 dependent and decreased from 17 to 1 hour, as conditions became more reducing.The isotopic composition of each reservoir was determined by Multicollector-Inductively Coupled Plasma-Mass Spectrometry (MC-ICPMS) after Ni purification. The isotopic composition was found to be constant in the metallic wire, which therefore behaved as an infinite reservoir. On the contrary, strong kinetic fractionation was observed in the silicate melt (δNi down to −0.98‰.amu−1 relative to the standard). Isotopic equilibrium was typically reached after 24 hours. For equilibrated samples at 1623 K, no metal-silicate fractionation was observed within uncertainty (2SD), with ΔNiMetal-Silicate = 0.02 ± 0.04‰.amu−1.Theoretical calculations of metal-silicate isotope fractionation at equilibrium were also performed on different metal-silicate systems. These calculations confirm (1) the absence of fractionation at high temperature and (2) a weak temperature dependence for Ni isotopic fractionation for the metal-olivine and metal-pyroxene pairs with the metal being slightly lighter isotopically.Our experimental data were finally compared with natural samples. Some mesosiderites (stony-iron meteorites) show a ΔNiMetal-Silicate close to experimental values at equilibrium, whereas others exhibit positive metal-silicate fractionation that could reflect kinetic processes. Conversely, pallasites display a strong negative metal-silicate fractionation. This most likely results from kinetic processes with Ni diffusion from the silicate to the metal phase due to a change of Ni partition coefficient during cooling. In this respect we note that in these pallasites, iron isotopes show metal-silicate fractionation that is opposite direction to Ni, supporting the idea of kinetic isotope fractionation, associated with Fe-Ni interdiffusion.

Age and origin of IIE iron meteorites inferred from Hf-W chronology

Abstract Non-magmatic iron meteorites, including the IIE group, can provide important insights into the history of metal-silicate differentiation and collisions on planetesimals. To better constrain the evolution of metal segregation and impacts on the IIE parent body, W isotopic data are reported for 10 IIE iron meteorites and a metal vein from the Portales Valley H6 chondrite. In addition, Pt isotopic data were obtained to quantify cosmic ray-induced neutron capture effects on W isotopes. After correction for these effects, the IIE iron meteorites exhibit variable pre-exposure 182W compositions, translating into Hf-W model age clusters of ∼4–5 million years (Ma), ∼10 Ma, ∼15 Ma, and ∼27 Ma after CAI formation. These distinct 182W clusters likely represent samples from several discrete metallic melt pools on the IIE parent asteroid. The earliest metal segregation event at ∼4–5 Ma was likely facilitated by 26Al decay as an internal heat source. By contrast, the younger Hf-W model ages may not be chronologically meaningful, and probably reflect the effects of secondary mixing and re-equilibration of metal and silicates, likely facilitated by impacts on the IIE parent body. Thus, contrary to prior work, the Hf-W systematics of IIE iron meteorites do not require a protracted history of metal-silicate separation on the IIE parent body. Instead the results of this study are fully consistent with a single partial metal-silicate differentiation event driven by endogenic heating at ∼4–5 Ma, followed by one or multiple impact events causing mixing and re-equilibration of metal and silicates at a later stage. The exact timing of these impact event(s) remains poorly constrained, but they most likely occurred in the first few tens of Ma of Solar System history.

Corrigendum to “Early metal-silicate differentiation during planetesimal formation revealed by acapulcoite and lodranite meteorites” [Geochim. Cosmochim. Acta 216 (2017) 115–140

Corrigendum to “Early metal-silicate differentiation during planetesimal formation revealed by acapulcoite and lodranite meteorites” [Geochim. Cosmochim. Acta 216 (2017) 115–140

Genetics, crystallization sequence, and age of the South Byron Trio iron meteorites: New insights to carbonaceous chondrite (CC) type parent bodies.

The nucleosynthetic Mo, Ru, and W isotopic compositions of the South Byron Trio iron meteorite grouplet (SBT) are consistent with all three meteorites originating on a single parent body that formed in the carbonaceous chondrite (CC) isotopic domain within the Solar nebula. Consistent with a common origin, the highly siderophile element (HSE) concentrations of the SBT can be related to one another by moderate degrees of fractional crystallization of a parental melt with initially chondritic relative abundances of HSE, and with initial S and P contents of ~7 and ~1 wt. %, respectively. Tungsten-182 isotopic data for the SBT indicate the parent body underwent metal-silicate differentiation 2.1 ± 0.8 Myr after calcium aluminum rich inclusion formation, and thermal modeling suggests the parent body formed 1.1 ± 0.5 Myr after CAI formation. This accretion age is not resolved from the accretion ages of other CC and most noncarbonaceous (NC) type iron meteorite parent bodies. Comparison of the projected parental melt composition of the SBT to those projected for the IVA and IVB iron meteorite groups suggests that at least some portions of the CC nebular domain were more oxidized compared to the NC domain. In addition, comparison of the SBT parental melt S content to estimates for parent bodies of the IIAB, IIIAB, IVA, IID, and IVB "magmatic" iron meteorite groups suggests that CC type iron meteorite parental melts were characterized by a general depletion in S, in addition to depletions in some other moderately volatile elements. Based on chemical and O isotope similarities, prior studies have suggested the possibility of a common parent body for the SBT and the Milton pallasite. Molybdenum and Ru isotopic compositions of Milton also provide permissive evidence for this. The HSE concentrations in the Milton metal, however, cannot be related to the SBT by any known crystal-liquid fractionation or mixing path. Thus, Milton more likely formed on a different, chemically distinct, but genetically identical parent body present in the CC nebular domain.

Iron isotope constraints on planetesimal core formation in the early solar system

Abstract We determined the Fe isotope fractionation between the metal and silicate phases of two aubrite meteorites, Norton County and Mount Egerton. We find that the metallic phase is high in 57Fe/54Fe with respect to the silicate phase, with Δ57Femetal-silicate = 0.08‰ ± 0.04 for Mount Egerton and 0.09 ± 0.02‰ for Norton County. These data, combined with new measurements of 57Fe/54Fe of IIIAB iron meteorites, are used to constrain the origins of the high 57Fe/54Fe exhibited by all classes of iron meteorites. We find that if the parent bodies of the iron meteorites had chondritic bulk 57Fe/54Fe values, their cores must have been unusually small (≤8% by mass). Relaxing the constraint that the bodies were chondritic in their bulk iron isotope ratios allows for larger core mass fractions commensurate with usual expectations. In this case, the elevated 57Fe/54Fe values of iron meteorites are due in part to evaporation of melt during the accretion stages of the parent bodies and not solely the result of metal-silicate differentiation.

Early metal-silicate differentiation during planetesimal formation revealed by acapulcoite and lodranite meteorites

In order to establish the role and expression of silicate-metal fractionation in early planetesimal bodies, we have conducted a highly siderophile element (HSE: Os, Ir, Ru, Pt, Pd, Re) abundance and 187Re-187Os study of acapulcoite-lodranite meteorites. These data are reported with new petrography, mineral chemistry, bulk-rock major and trace element geochemistry, and oxygen isotopes for Acapulco, Allan Hills (ALHA) 81187, Meteorite Hills (MET) 01195, Northwest Africa (NWA) 2871, NWA 4833, NWA 4875, NWA 7474 and two examples of transitional acapulcoite-lodranites, Elephant Moraine (EET) 84302 and Graves Nunataks (GRA) 95209. These data support previous studies that indicate that these meteorites are linked to the same parent body and exhibit limited degrees (<2–7%) of silicate melt removal. New HSE and osmium isotope data demonstrate broadly chondritic relative and absolute abundances of these elements in acapulcoites, lower absolute abundances in lodranites and elevated (>2×CI chondrite) HSE abundances in transitional acapulcoite-lodranite meteorites (EET 84302, GRA 95209). All of the meteorites have chondritic Re/Os with measured 187Os/188Os ratios of 0.1271±0.0040 (2 St. Dev.). These geochemical characteristics imply that the precursor material of the acapulcoites and lodranites was broadly chondritic in composition, and were then heated and subject to melting of metal and sulfide in the Fe-Ni-S system. This resulted in metallic melt removal and accumulation to form lodranites and transitional acapulcoite-lodranites. There is considerable variation in the absolute abundances of the HSE, both among samples and between aliquots of the same sample, consistent with both inhomogeneous distribution of HSE-rich metal, and of heterogeneous melting and incomplete mixing of silicate material within the acapulcoite-lodranite parent body. Oxygen isotope data for acapulcoite-lodranites are also consistent with inhomogeneous melting and mixing of accreted components from different nebular sources, and do not form a well-defined mass-dependent fractionation line. Modeling of HSE inter-element fractionation suggests a continuum of melting in the Fe-Ni-S system and partitioning between solid metal and sulfur-bearing mineral melt, where lower S contents in the melt resulted in lower Pt/Os and Pd/Os ratios, as observed in lodranites. The transitional meteorites, EET 84302 and GRA 95209, exhibit the most elevated HSE abundances and do not follow modelled Pt/Os and Pd/Os solid metal-liquid metal partitioning trends. We interpret this to reflect metal melt pooling into domains that were sampled by these meteorites, suggesting that they may originate from deeper within the acapulcoite-lodranite parent body, perhaps close to a pooled metallic ‘core’ region. Petrographic examination of transitional samples reveals the most extensive melting, pooling and networking of metal among the acapulcoite-lodranite meteorites. Overall, our results show that solid metal-liquid metal partitioning in the Fe-Ni-S system in primitive achondrites follows a predictable sequence of limited partial melting and metal melt pooling that can lead to significant HSE inter-element fractionation effects in proto-planetary materials.

A nonmagnetic differentiated early planetary body

Paleomagnetic studies of meteorites have shown that the solar nebula was likely magnetized and that many early planetary bodies generated dynamo magnetic fields in their advecting metallic cores. The surface fields on these bodies were recorded by a diversity of chondrites and achondrites, ranging in intensity from several μT to several hundred μT. In fact, an achondrite parent body without evidence for paleomagnetic fields has yet to be confidently identified, hinting that early solar system field generation and the dynamo process in particular may have been common. Here we present paleomagnetic measurements of the ungrouped achondrite NWA 7325 indicating that it last cooled in a near-zero field (<∼1.7 μT), estimated to have occurred at 4563.09±0.26 million years ago (Ma) from Al–Mg chronometry. Because NWA 7325 is highly depleted in siderophile elements, its parent body nevertheless underwent large-scale metal-silicate differentiation and likely formed a metallic core. This makes NWA 7325 the first recognized example of an essentially unmagnetized igneous rock from a differentiated early solar system body. These results indicate that all magnetic fields, including those from any core dynamo on the NWA 7325 parent body, the solar nebula, young Sun, and solar wind, were <1.7 μT at the location of NWA 7325 at 4563 Ma. This supports a recent conclusion that the solar nebula had dissipated by ∼4 million years after solar system formation. NWA 7325 also serves as an experimental control that gives greater confidence in the positive identification of remanent magnetization in other achondrites.

Silicon stable isotope fractionation between metal and silicate at high-pressure, high-temperature conditions as a tracer of planetary core formation

AbstractThe largest differentiation event in Earth and other terrestrial planets was the high-pressure, high-temperature process of metal core segregation from a silicate mantle. The abundant element silicon (Si) can be partially sequestered into the metallic core during metal–silicate differentiation, depending on pressure, temperature and planetary oxidation state. Knowledge of the Si content of a planet's core can constrain the conditions of core formation, but in the absence of direct samples from planetary cores, quantifying core Si content is challenging. One relatively new tool to study core formation in terrestrial planets is based on combining measurements of the Si stable isotopic composition of planetary crust and mantle samples with measurements of the Si stable isotope fractionation between metal and silicate at high-temperature and high-pressure conditions. In this study we present the results of a small set of high-pressure, high-temperature (HPT) experiments and combine these with a review of literature data to investigate how the Si isotope fractionation behaviour between metal and silicate varies as a function specifically of experimental run time and temperature. We show that although there is no debate about the sign of fractionation, absolute values for Si isotope fractionation between metal and silicate are difficult to constrain because the experimental database remains incomplete, and because Si isotopic measurements of metals in particular suffer from the absence of a true inter-laboratory comparison. We conclude that in order to derive accurate quantitative estimates of the Si content of the core of the Earth or other planets a wide range of additional experiments will be required.

The stable Cr isotopic compositions of chondrites and silicate planetary reservoirs

The depletion of chromium in Earth’s mantle (∼2700ppm) in comparison to chondrites (∼4400ppm) indicates significant incorporation of chromium into the core during our planet’s metal–silicate differentiation, assuming that there was no significant escape of the moderately volatile element chromium during the accretionary phase of Earth. Stable Cr isotope compositions – expressed as the ‰-difference in 53Cr/52Cr from the terrestrial reference material SRM979 (δ53/52CrSRM979 values) – of planetary silicate reservoirs might thus yield information about the conditions of planetary metal segregation processes when compared to chondrites. The stable Cr isotopic compositions of 7 carbonaceous chondrites, 11 ordinary chondrites, 5 HED achondrites and 2 martian meteorites determined by a double spike MC-ICP-MS method are within uncertainties indistinguishable from each other and from the previously determined δ53/52CrSRM979 value of −0.124±0.101‰ for the igneous silicate Earth. Extensive quality tests support the accuracy of the stable Cr isotope determinations of various meteorites and terrestrial silicates reported here. The uniformity in stable Cr isotope compositions of samples from planetary silicate mantles and undifferentiated meteorites indicates that metal–silicate differentiation of Earth, Mars and the HED parent body did not cause measurable stable Cr isotope fractionation between these two reservoirs. Our results also imply that the accretionary disc, at least in the inner solar system, was homogeneous in its stable Cr isotopic composition and that potential volatility loss of chromium during accretion of the terrestrial planets was not accompanied by measurable stable isotopic fractionation. Small but reproducible variations in δ53/52CrSRM979 values of terrestrial magmatic rocks point to natural stable Cr isotope variations within Earth’s silicate reservoirs. Further and more detailed studies are required to investigate whether silicate differentiation processes, such as partial mantle melting and crystal fractionation, can cause stable Cr isotopic fractionation on Earth and other planetary bodies.

Highly Siderophile Element and187Os Signatures in Non-cratonic Basalt-hosted Peridotite Xenoliths: Unravelling the Origin and Evolution of the Post-Archean Lithospheric Mantle

The highly siderophile elements (HSE) consist of the Platinum Group Elements (PGE: Ru, Rh, Pd, Os, Ir, Pt) along with rhenium and gold. These transition elements show relative chemical inertness and high market values, which respectively earned them the additional names of noble metals and precious metals. The HSE show a very pronounced affinity for iron metal, which translates into metal/silicate partition coefficients similar to or higher than 10,000 over large ranges of both pressure and temperature (e.g., O’Neill et al. 1995; Borisov and Palme 2000; Ertel et al. 1999, 2001, 2006, 2008; Fortenfant et al. 2003, 2006; Brenan et al. 2005; Cottrell and Walker 2006; Brenan and McDonough 2009; Laurenz et al. 2010; Mann et al. 2012; see Brenan et al. 2016, this volume for detailed review). Consequently, the HSE are thought to have been efficiently sequestered within the metallic core of our planet during the metal–silicate differentiation of Earth, leaving the silicate counterpart almost HSE-barren. Investigations of mantle peridotites since the 1970s revealed ng.g−1 level abundances as well as close-to-chondritic proportions of the HSE (Chou 1978; Jagoutz et al. 1979; Mitchell and Keays 1981; McDonough and Sun 1995; Becker et al. 2006; Fischer-Godde et al. 2011). Such abundances and inter-HSE fractionations are not predicted for the silicate Earth left after separation of the metallic core for low- or high-pressure core–mantle differentiation (see Brenan et al. 2016, this volume). The close agreement between the osmium isotopic compositions of fertile mantle peridotites and those of chondritic meteorites (Walker et al. 2002a), which requires nearly identical Re/Os ratios in these two reservoirs, provides particularly convincing evidence that the mantle’s HSE content cannot simply represent the residue left after core formation. …

Core formation and core composition from coupled geochemical and geophysical constraints

The formation of Earth's core left behind geophysical and geochemical signatures in both the core and mantle that remain to this day. Seismology requires that the core be lighter than pure iron and therefore must contain light elements, and the geochemistry of mantle-derived rocks reveals extensive siderophile element depletion and fractionation. Both features are inherited from metal-silicate differentiation in primitive Earth and depend upon the nature of physiochemical conditions that prevailed during core formation. To date, core formation models have only attempted to address the evolution of core and mantle compositional signatures separately, rather than seeking a joint solution. Here we combine experimental petrology, geochemistry, mineral physics and seismology to constrain a range of core formation conditions that satisfy both constraints. We find that core formation occurred in a hot (liquidus) yet moderately deep magma ocean not exceeding 1,800 km depth, under redox conditions more oxidized than present-day Earth. This new scenario, at odds with the current belief that core formation occurred under reducing conditions, proposes that Earth's magma ocean started oxidized and has become reduced through time, by oxygen incorporation into the core. This core formation model produces a core that contains 2.7-5% oxygen along with 2-3.6% silicon, with densities and velocities in accord with radial seismic models, and leaves behind a silicate mantle that matches the observed mantle abundances of nickel, cobalt, chromium, and vanadium.

Timing of metal–silicate differentiation in the Eagle Station pallasite parent body

The time of the metal–silicate differentiation of the Eagle Station pallasite (ESP) parent body was investigated using the 26Al–26Mg short-lived chronometer (half-life of 0.72Myr). The Mg isotope ratios were measured in ESP olivines by both MC–SIMS and HR-MC–ICPMS, allowing us to check the consistency between the results given by two different analytical protocols and data reduction processes. Results show that the two datasets are consistent, with a (δ26Mg*)av. value of –0.003 (±0.005)‰ (2 s.e., n=89). Such a value, associated with data from the 182Hf–182W short-lived systematics (half-life of 8.9Myr), indicates an ESP parent body metal–silicate differentiation occurring most likely at least at ∼2Ma, but possibly 4Ma, after CAI formation. From the 27Al/24Mg ratios measured in ESP olivines using MC–SIMS, the duration of the olivine crystallization process was inferred to have lasted over ∼275kyr if the core has differentiated as early as 2Ma after CAIs, while in the case of a core differentiation occurring 4Ma after CAIs, the silicate–silicate differentiation should have lasted for another 4Myr.

Evidence for Mo isotope fractionation in the solar nebula and during planetary differentiation

Mass-dependent Mo isotope fractionation has been investigated for a wide range of meteorites including chondrites (enstatite, ordinary and carbonaceous chondrites), iron meteorites, and achondrites (eucrites, angrites and martian meteorites), as well as for lunar and terrestrial samples. Magmatic iron meteorites together with enstatite, ordinary and most carbonaceous chondrites define a common δ98/95Mo value of −0.16±0.02‰ (relative to the NIST SRM 3134 Mo standard), which is interpreted to reflect the Mo isotope composition of bulk planetary bodies in the inner solar system. Heavy Mo isotope compositions for IAB iron meteorites most likely reflect impact-induced evaporative losses of Mo from these meteorites. Carbonaceous chondrites define an inverse correlation between δ98/95Mo and metal content, and a positive correlation between δ98/95Mo and matrix abundance. These correlations are mainly defined by CM and CK chondrites, and may reflect the heterogeneous distribution of an isotopically light metal and/or an isotopically heavy matrix component in the formation region of carbonaceous chondrites. Alternatively, the elevated δ98/95Mo of the CM and CK chondrites could result from the loss of volatile, isotopically light Mo oxides, that formed under oxidized conditions typical for the formation of these chondrites.The Mo isotope compositions of samples derived from the silicate portion of differentiated planetary bodies are heavy compared to the mean composition of chondrites and iron meteorites. This difference is qualitatively consistent with experimental evidence for Mo isotope fractionation between metal and silicate. The common δ98/95Mo values of −0.05±0.03‰ of lunar samples derived from different geochemical reservoirs indicate the absence of significant Mo isotope fractionation by silicate differentiation or impact metamorphism/volatilization on the Moon. The most straightforward interpretation of the Mo isotope composition of the lunar mantle corresponds to the formation of a lunar core at a metal–silicate equilibration temperature of 1800±200°C. The investigated martian meteorites, angrites and eucrites exhibit more variable Mo isotope compositions, which for several samples extend to values above the maximum δ98/95Mo=+0.14‰ that can be associated with core formation. For these samples post-core formation processes such as partial melting, metamorphism and in the case of meteorite finds terrestrial weathering must have resulted in Mo isotope fractionation. Estimates of the metal–silicate equilibration temperatures for Mars (2490±770°C) and the angrite parent body (1790±230°C) are thus more uncertain than that derived for the Moon. Although the Mo isotope composition of the bulk silicate Earth has not been determined as part of this study, a value of −0.16‰<δ98/95Mo<0 can be predicted based on the chondrite and iron meteorite data and by assuming a reasonable temperature range for core formation in the Earth. This estimate is in agreement with four analyzed basalt standards (−0.10±0.10). Improved application of mass-dependent Mo isotope fractionation to investigate core formation most of all requires an improved understanding of potential Mo isotope fractionation during processes not related to metal–silicate differentiation.

Experimental evidence for the absence of iron isotope fractionation between metal and silicate liquids at 1 GPa and 1250–1300 °C and its cosmochemical consequences

Iron isotope fractionation during metal–silicate differentiation has been proposed as a cause for differences in iron isotope compositions of chondrites, iron meteorites and the bulk silicate Earth. Stable isotope fractionation, however, rapidly decreases with increasing temperature. We have thus performed liquid metal–liquid silicate equilibration experiments at 1250–1300°C and 1GPa to address whether Fe isotope fractionation is resolvable at the lowest possible temperatures for magmatic metal–silicate differentiation. A centrifuging piston cylinder apparatus enabled quantitative metal–silicate segregation. Elemental tin or sulphur was used in the synthetic metal-oxide mixtures to lower the melting temperature of the metal. The analyses demonstrate that eight of the 10 experimental systems equilibrated in a closed isotopic system, as was assessed by varying run durations and starting Fe isotope compositions. Statistically significant iron isotope fractionation between quenched metals and silicates was absent in nine of the 10 experiments and all 10 experiments yield an average metal–silicate fractionation factor of 0.01±0.04‰, independent of whether graphite or silica glass capsules were used. This implies that Fe isotopes do not fractionate during low pressure metal–silicate segregation under magmatic conditions. In large bodies such as the Earth, fractionation between metal and high pressure (>20GPa) silicate phases may still be a possible process for equilibrium fractionation during metal–silicate differentiation. However, the 0.07±0.02‰ heavier composition of bulk magmatic iron meteorites relative to the average of bulk ordinary/carbonaceous chondrites cannot result from equilibrium Fe isotope fractionation during core segregation. The up to 0.5‰ lighter sulphide than metal fraction in iron meteorites and in one ordinary chondrite can only be explained by fractionation during subsolidus processes.

Isotopic constraints on the age of the Earth’s core: Mutual consistency of the Hf-W and U-Pb systems

The estimation of the time of Earth’s core formation on the basis of isotopic systems with short-lived and long-lived parent nuclides gives significantly different results. Isotopic data for the 182Hf-182W system with a 182Hf half-life of approximately 9 Myr can be interpreted in such a way that the core was formed 34 Myr after the origin of the solar system assuming complete core-mantle equilibrium. Similar estimates on the basis of the U-Pb isotopic system suggest a significantly longer mean time of core formation of approximately 120 Myr. If the Earth’s core were formed instantaneously, both isotopic systems would have shown identical values corresponding to the true age. The discrepancy between the U-Pb and Hf-W systems can be resolved assuming prolonged differentiation of prototerrestrial material into silicate and metallic phases, which occurred not simultaneously and uniformly in different parts of the mantle. This resulted in the isotopic heterogeneity of the mantle, and its subsequent isotopic homogenization occurred slowly. Under such conditions, the mean isotopic compositions of W and Pb in the mantle do not correspond to the mean time of the separation of silicate and metallic phases. This is related to the fact that the exponential function of radioactive decay is strongly nonlinear at high values of the argument, and its mean value does not correspond to the mean value of the function. There are compelling reasons to believe that the early mantle was heterogeneous with respect to W isotopic composition and was subsequently homogenized by convective mixing. This follows from the fact that the lifetime of isotopic heterogeneities in the mantle is close to 1.8 Gyr for various long-lived isotopic systems. There is also no equilibrium between the mantle and the core with respect to the contents of siderophile elements. Because of this, the mean isotopic ratios of W and Pb cannot be used for the direct computation of the time of metal-silicate differentiation in the Earth. Such estimation requires more sophisticated models accounting for the duration of the differentiation process using several isotope pairs. Given the prolonged core formation, which has probably continued up to now, the question about its age becomes ambiguous, and only the most probable growth rate of the core can be estimated. The combined use of the U-Pb and Hf-W systems constrains the time of formation of 90% of the core mass between 0.12 and 2.7 billion years. These model estimates could have been realistic under the condition of complete disequilibrium between the silicate and metallic phases, which is as improbable as the suggestion of complete equilibrium between them on the whole Earth scale.