Silicate Earth Research Articles

Enrichment of heavy calcium isotopes in saprolite due to secondary mineral formation

Calcium stable isotopes were analyzed in well-characterized saprolite samples developed on a Mesozoic metadiabase dike near Cayce, South Carolina, to determine the degree and sense of isotopic fractionation during chemical weathering. The δ44/40Ca values (relative to NIST SRM 915a) of the saprolites, measured using double-spike TIMS, vary significantly (0.91‰ to 1.26‰), and are generally higher than the δ44/40Ca value of the unweathered metadiabase (0.98‰), which overlaps, within analytical uncertainties, with the δ44/40Ca estimate of the bulk silicate Earth. The δ44/40Ca values of the saprolites negatively correlate with bulk density, and Ti-normalized calcium concentrations, and positively correlate with chemical index of alteration (CIA) values. These trends reflect loss of light Ca isotopes to the hydrosphere during metadiabase weathering. Using insoluble incompatible elements ratios (e.g., Th/Nb) we estimate that the influence of aeolian dust is minimal (3% - 11%) and therefore unable to explain the observed variation in δ44/40Ca values of the saprolites. Selective weathering of rock-forming minerals like plagioclase and clinopyroxene cannot explain the high δ44/40Ca values (> 1 ‰) of the saprolites based on mass balance constraints. Samples with higher kaolinite/smectite ratio have higher δ44/40Ca values and lower concentrations of Ca, which suggests the loss of Ca from smectite-rich clays during progressive weathering. In this case, the lighter isotopes are preferentially lost, leaving the residual clays enriched in heavier isotopes of Ca. The δ44/40Ca values of saprolites correlate positively with δ26Mg and negatively with δ7Li values of the same samples, which further suggests that Ca stable isotopic variability of the Cayce metadiabase saprolites is controlled by formation of clay minerals and progressive loss of Ca during saprolitization. The enrichment of heavy Ca isotopes in saprolites with progressive loss of Ca can be modeled using Rayleigh distillation, with apparent fractionation factors between saprolites and fluid (α) of 1.00005 to 1.00015.

Chromium isotope variations in garnet-facies mantle rocks and their minerals: Implications for Cr isotope behavior in high-temperature processes

Previous work on Cr isotope composition of the lithospheric mantle focused on shallow spinel-facies peridotites and invoked the effects of partial melting and metasomatism. However, little is known on the Cr isotope composition and behavior in the deeper and more voluminous garnet-facies mantle, notably during generation and passage of magmas and fluids. Here, we present Cr isotope data for 26 garnet and ten spinel whole-rock peridotite xenoliths (mainly harzburgites) from the Udachnaya kimberlite on the Siberian craton and minerals from five of them, as well as for minerals of 14 eclogites and four garnet clinopyroxenites from the Dabie orogen in eastern China. Garnet (Gt) and clinopyroxene (Cpx) are major Cr hosts in the mantle. Our ionic modeling suggests the following order of 53Cr/52Cr values in coexisting minerals: garnet ≈ spinel > Cpx > olivine. Cr isotopes are equally partitioned between garnet and Cpx in four peridotites, while garnet is enriched in heavy Cr in one peridotite and all the eclogites (Δ53CrGt-Cpx 0.01‰ to 0.27‰, av. 0.10‰), but depleted in heavy Cr in the pyroxenites (Δ53CrGt-Cpx −0.03‰ to −0.16‰, av. −0.10‰). The Δ53CrGt-Cpx variations are attributed to effects of chemical composition, equilibration temperature and oxygen fugacity on equilibrium Cr isotope fractionation. Further, the anomalous Δ53CrGt-Cpx values in the Dabie pyroxenites and one peridotite may indicate disequilibrium inter-mineral isotope fractionation via kinetic diffusion during phase transitions as a result of pressure-temperature (P-T) changes and metasomatism.The whole-rock δ53Cr range in the garnet peridotites (−0.27 to 0.13‰; average −0.08 ± 0.18‰, 2 SD, n = 25, excluding an outlier with −0.86‰) overlaps bulk silicate Earth estimates whereas the spinel peridotites tend to have higher δ53Cr (−0.12 to 0.16‰; av. 0.02 ± 0.20‰, 2 SD, n = 10). While these rocks are residues of high degrees (30–40%) of melt extraction, their range contrasts with limited equilibrium Cr isotope fractionation we calculated for melting residues of both spinel and garnet peridotites (Δ53Crresidue-source < 0.03‰). The δ53Cr values in the spinel peridotites show correlations with modal abundances of Cpx, garnet and spinel, as well as major and trace element composition. We posit that the Cr isotope variations in these peridotites are somehow linked to solid–melt and inter-mineral diffusive exchanges during P-T changes, deformation and ingress of metasomatic media previously documented in the Udachnaya peridotites. The average δ53Cr value (−0.09 ± 0.16‰, 2 SD, n = 31) for garnet peridotites (excluding U9) in this and previous studies is similar to those previously obtained for (mainly off-craton) spinel peridotites and komatiites, and appears to be representative of the Cr isotope composition for bulk mantle. Numerical modeling for mantle magmas derived at different depths and by different melting degrees of garnet and spinel peridotites and eclogites yields a narrow δ53Cr range, possibly affected by oxygen fugacity.

An improved model of metal/silicate differentiation during Earth's accretion

We improved the algorithm presented in Rubie et al. (2015) to model the chemical evolution of Earth driven by iron/silicate differentiation during the planet's accretion. The pressure at which the equilibration occurs during a giant impact is no longer a free parameter but is determined by the smooth particle hydrodynamic (SPH) simulations of Nakajima et al. (2021). Moreover, impacting planetesimals are now assumed to be too small to cause melting and differentiation and thus their materials are stored in the crystalline upper mantle of the growing planet until a hydrostatically relaxed global magma ocean forms in the aftermath of a giant impact, whose depth is also estimated from Nakajima et al. (2021). With these changes, not all dynamical simulations lead to a satisfactory reproduction of the chemical composition of the bulk silicate Earth (BSE). Thus, the latter becomes diagnostic of the success of dynamical models. In the successful cases also the BSE abundances of W and Mo can be reproduced, that were previously hard to fit (Jennings et al., 2021).

The deep Earth oxygen cycle: Mass balance considerations on the origin and evolution of mantle and surface oxidative reservoirs

The deep Earth oxygen cycle addresses the origin and long-term exchange of oxidized species between the mantle and exosphere and creates the geochemical context in which the interior and surface environment have evolved. The redox power of the bulk silicate Earth (BSE) can be measured in terms of its redox budget, RB, relative to a reference state of the predominant valences of redox-sensitive elements in the mantle. 82±4% (1.9±0.5 × 1023 moles) of the redox power resides in the mantle and 18±4% (4.2±0.5 × 1022 moles) in the exosphere. Vigorous outfluxes of oxidized species from the mantle (3.8±0.7 × 1013 mol/yr) can replenish the exosphere reservoir in 1.1±0.2 Ga, which requires efficient long-term recycling of redox power via subduction. Within uncertainties, recent (last 200 Ma) mantle RB outfluxes and subduction influxes (4.9±1.3 × 1013 mol/yr) are balanced, but outfluxes likely exceeded influxes earlier in Earth history. CO2 is 66±20% of the RB outfluxes but only 24±13% of the influxes. This is largely because of conjugate intervalence reactions; one in the shallow mantle creates carbonate at the expense of Fe2O3 and the other creates Fe2O3 from CO2 on the surface by a combination of organic carbon fixation and oxidative weathering.Scaling of mantle oxygen fugacity, fO2, to redox mass balance is approximately Δlog⁡(Fe3+Fe⊤)peridotite ≈14Δlog⁡(fO2)mantle . Consequently, inferences of secular evolution of mantle oxygen fugacity from the Archaean to the Proterozoic, amounting to about 1.5 log units in fO2, imply that the Archaean mantle had an Fe3+/FeT ratio <0.02, rather than the modern value of 0.04±0.01. Oceanic basalts derive from sources with greater redox budgets than mid-ocean ridges, and this is partly expressed as higher source Fe3+/FeT ratios, but more importantly, as greater source CO2 concentrations. In combination, these require that plumes in the deep upper mantle have Fe3+/FeT ratios significantly greater than the depleted mantle.The oxidative inventory of the BSE originated by a combination of H2O disproportionation in the atmosphere, leading to H2 escape, and FeO disproportionation in the deep mantle, with loss of Fe to the core. FeO disproportionation in a deep magma ocean inevitably produces a significant fraction of the BSE RB, but additional contributions are likely required. Further Fe loss could be from bridgemanite crystallization followed by Fe escape through a basal magma ocean. Gradual mixing of the resulting deep oxidized layer may account for secular oxidation of the mantle source regions of igneous rocks from 3 to 2 Ga. H2O disproportionation assisted in accumulation of the oxidized surface species, but was not a significant source of mantle oxidative power, as mechanisms of oxidative influx are quantitatively insufficient.

Potassium isotope evidence for slab-derived fluids in the sub-arc mantle

Subduction recycles water, other volatiles and incompatible elements into Earth's mantle and plays a major role in exchanges between its surface and interior. Potassium (K), a highly fluid-mobile and incompatible alkali element, is strongly affected by these processes. The emerging K stable isotope system has been proposed as a new tracer of subduction events, e.g. to distinguish the inputs to arc magmas from sediments, altered basalts and other near-surface materials. Yet, it remains unknown if and how the K isotopic signatures of these components are affected by extraction of K-bearing media from dehydrating slabs and their transfer to the mantle wedge. Fragments (xenoliths) of sub-arc mantle carried to the surface by arc magmas may retain these slab-derived components. We report high-precision K isotope data on mantle harzburgite xenoliths from Avacha, an active arc volcano in Kamchatka (Russia) to further investigate the nature of slab-derived fluids and their impact on arc mantle. We find a narrow δ41K range for amphibole-rich cumulates from host andesite magmas (−0.53 ± 0.09‰ to −0.34 ± 0.09‰) that overlaps those of the bulk silicate Earth (BSE) and the upper continental crust (UCC). In contrast, the Avacha peridotites derived from the uppermost lithospheric mantle (refractory melting residues metasomatized by subduction zone fluids) show an extremely broad K isotopic range (δ41K ∼ 3‰) and are commonly enriched in heavy K isotopes (up to +1.16 ± 0.08‰) relative to the BSE and UCC; some of the peridotites show the highest δ41K values for any reported mantle rocks. The new data on the Avacha xenoliths, along with available K isotope data on arc lavas and eclogites, suggest that slab-derived fluids may develop heavy K isotopic signatures, distinct from those in the original subducting materials (e.g., marine sediments and altered oceanic crust). These results provide direct evidence that (a) subduction dehydration and fluid percolation in the mantle lithosphere can significantly fractionate K isotopes, and (b) slab-derived fluids infiltrated the sub-arc mantle wedge. Due to the large K concentration difference between the depleted mantle and slab-derived fluids, K isotopes may be a sensitive tracer of fluid activity in sub-arc mantle.

Earth's evolving geodynamic regime recorded by titanium isotopes.

Earth's mantle has a two-layered structure, with the upper and lower mantle domains separated by a seismic discontinuity at about 660 km (refs. 1,2). The extent of mass transfer between these mantle domains throughout Earth's history is, however, poorly understood. Continental crust extraction results in Ti-stable isotopic fractionation, producing isotopically light melting residues3-7. Mantle recycling of these components can impart Ti isotope variability that is trackable in deep time. We report ultrahigh-precision 49Ti/47Ti ratios for chondrites, ancient terrestrial mantle-derived lavas ranging from 3.8 to 2.0 billion years ago (Ga) and modern ocean island basalts (OIBs). Our new Ti bulk silicate Earth (BSE) estimate based on chondrites is 0.052 ± 0.006‰ heavier than the modern upper mantle sampled by normal mid-ocean ridge basalts (N-MORBs). The 49Ti/47Ti ratio of Earth's upper mantle was chondritic before 3.5 Ga and evolved to a N-MORB-like composition between approximately 3.5 and 2.7 Ga, establishing that more continental crust was extracted during this epoch. The +0.052 ± 0.006‰ offset between BSE and N-MORBs requires that <30% of Earth's mantle equilibrated with recycled crustal material, implying limited mass exchange between the upper and lower mantle and, therefore, preservation of a primordial lower-mantle reservoir for most of Earth's geologic history. Modern OIBs record variable 49Ti/47Ti ratios ranging from chondritic to N-MORBs compositions, indicating continuing disruption of Earth's primordial mantle. Thus, modern-style plate tectonics with high mass transfer between the upper and lower mantle only represents a recent feature of Earth's history.

Trace element redistributions during metamorphism of E-chondrites: Implications for reduced bodies and the Earth

We report on new trace element analyses of enstatite chondrites (ECs) to clarify their behavior during the metamorphism. During the transition from a type 3 to a type 5 or higher, silicates lose a large portion of their trace elements to sulfides. Our procedure allows us to obtain trace element abundances of the silicate fraction of an EC quite easily. The element patterns of these fractions (especially REE patterns) are quite different for EH and EL chondrites, and are furthermore dependent on the metamorphic grade. This procedure can be usefull to classify meteorites, in particular when the sulfides are altered. Applied to anomalous ECs, it allows direct recognition of the EH affinity of QUE 94204, and suggests that Zakłodzie, NWA 4301, and NWA 4799 derive from the same EH-like body of previously unsampled composition.We have used the concentrations obtained on the silicate fractions of the most metamorphosed chondrites to discuss the chemical characteristics of the primitive mantles of reduced bodies of EH or EL affinity (i.e., after core segregation). Our data indicate that these mantles are very depleted in refractory lithophile elements (RLEs), particularly in rare earth elements (REEs), and notably show significant positive anomalies in Sr, Zr, Hf, and Ti. These estimates imply that the cores contain most of the REEs, U and Th of these bodies. Interestingly, the inferred primitive mantles of these reduced bodies contrast with that of the Earth. If the Earth accreted essentially from ECs, one would expect similar signatures to be preserved, which is not the case. This mismatch can be explained either by a later homogenization of the bulk silicate Earth, or alternatively, that the materials that were accreted were isotopically similar to ECs, but mineralogically different (i.e., oldhamite-free).

A multi-siderophile element connection between volcanic hotspots and Earth's core

The existence of resolvable 182W/184W deficits in modern ocean island basalts (OIB) relative to the bulk silicate Earth has raised questions about the relationship of these rocks to Earth's core. However, because the core is expected to host high abundances of highly siderophile elements (HSE: Os, Ir, Ru, Pt, Pd, Re), it would be expected that such heterogeneity is accompanied by correlating variability in HSE abundances among OIB, but this has not been observed. We report instead a relationship between the W isotopic compositions and Ru/Ir ratios of Hawai‘i and Iceland OIB, which represent two of Earth's primary mantle plumes. Previous studies have highlighted the unique behavior of Ru relative to Os and Ir during metal-silicate fractionation, particularly when sulfide phases are segregated with metal. Using the information from these studies, we construct models predicting the consequences for HSE fractionation of various scenarios in which 182W/184W deficits can be created. These models show that the observed trends are likely inconsistent with modern, active core-mantle interaction at the CMB, and instead the observed low-Ru/Ir, low-182W/184W OIB are best explained by metal-silicate interaction that happened at significantly lower pressures. Such conditions may reflect what is expected for metal-silicate equilibration during the process of core formation itself, meaning that the deep mantle sources of OIB, such as ultra-low velocity zones, may instead reflect preserved relics of core formation. An ancient origin for distinct domains now residing at the core-mantle boundary is consistent with geophysical and petrological observations, for example that the Mg/Fe ratio of ferropericlase in the D″ layer is in significant disequilibrium with the modern core. Additional work is required to constrain the behavior of HSE during silicate differentiation processes that may also generate low 182W/184W ratios. However, if modern OIB represent a direct link to the ancient processes of core formation, future geochemical studies may be able to unlock new information about the formation and evolution of the Earth using OIB.

Silicon and iron isotopes in components of enstatite chondrites: Implications for metal–silicate–sulfide fractionation in the solar nebula

AbstractSilicon and iron isotope compositions of different physically separated components of enstatite chondrites (EC) were determined in this study to understand the role of nebular and planetary scale events in fractionating Si and Fe isotopes of the terrestrial planet‐forming region. We found that the metal–sulfide nodules of EC are strongly enriched in light Si isotopes (δ30Si ≥ −5.61 ± 0.12‰, 2SD), whereas the δ30Si values of angular metal grains, magnetic, slightly magnetic, and non‐magnetic fractions become progressively heavier, correlating with their Mg# (Mg/(Mg+Fe)). White mineral phases, composed primarily of SiO2 polymorphs, display the heaviest δ30Si of up to +0.23 ± 0.10‰. The data indicate a key role of metal–silicate partitioning on the Si isotope composition of EC. The overall lighter δ30Si of bulk EC compared to other planetary materials can be explained by the enrichment of light Si isotopes in EC metals along with the loss of isotopically heavier forsterite‐rich silicates from the EC‐forming region. In contrast to the large Si isotope heterogeneity, the average Fe isotope composition (δ56Fe) of EC components was found to vary from −0.30 ± 0.08‰ to +0.20 ± 0.04‰. A positive correlation between δ56Fe and Ni/S in the components suggests that the metals are enriched in heavy Fe isotopes whereas sulfides are the principal hosts of light Fe isotopes in the non‐magnetic fractions of EC. Our combined Si and Fe isotope data in different EC components reflect an inverse correlation between δ30Si and δ56Fe, which illustrates that partitioning of Si and Fe among metal, silicate, and sulfidic phases has significantly fractionated Si and Fe isotopes under reduced conditions. Such isotope partitioning must have occurred before the diverse components were mixed to form the EC parent body. Evaluation of diffusion coefficients of Si and Fe in the metal and non‐metallic phases suggests that the Si isotope compositions of the silicate fractions of EC largely preserve information of their nebular processing. On the other hand, the Fe isotopes might have undergone partial or complete re‐equilibration during parent body metamorphism. The relatively uniform δ56Fe among different types of bulk chondrites and the Earth, despite Fe isotope differences among their components, demonstrates that the chondrite parent bodies were not formed by random mixing of chondritic components from different locations in the disk. Instead, the chondrite components mostly originated in the same nebular reservoir and Si and Fe isotopes were fractionated either due to gas–solid interactions and associated changes in physicochemical environment of the nebular reservoir and/or during parent body processing. The heavier Si isotope composition of the bulk silicate Earth may require accretion of chondritic and/or isotopically heavier EC silicates along with cumulation of refractory forsterite‐rich heavier silicates lost from the EC‐forming region to form the silicate reservoir of the Earth.

Spurious molybdenum isotope anomalies resulting from non-exponential mass fractionation

Mass-independent (nucleosynthetic) Mo isotope anomalies are uniquely useful for constraining genetic relationships among meteoritic and planetary materials and, by extension, the origin and nature of Earth's late-stage building blocks. The meaningful interpretation of such data, however, critically depends on the accurate correction of any natural and analytical mass-dependent isotope fractionation, which is commonly assumed to follow the ‘exponential law’. Here, using new high-precision Mo isotope data for a diverse set of terrestrial samples, we show that mass-dependent Mo isotope fractionation in nature typically does not adhere to this law, but is instead dominated by equilibrium and Rayleigh processes. We demonstrate that even moderate degrees of such non-exponential fractionation (i.e., mass-dependent isotope fractionation deviating from the exponential law) can result in significant spurious mass-independent Mo isotope anomalies that, when misinterpreted as nucleosynthetic anomalies, can lead to erroneous conclusions, particularly with respect to Earth's accretion history. Consequently, assessing the magnitude and origin of mass-dependent fractionation will be essential for future efforts to precisely determine the mass-independent Mo isotope composition of bulk silicate Earth and to identify potential nucleosynthetic isotope anomalies in terrestrial rocks.

Sr-Nd-Ca isotopic variations of Cenozoic calc-alkaline and alkaline volcanic rocks above a slab tear in Western Anatolia, Turkey

The tectonic mechanisms that may trigger a transition from calc-alkaline to alkaline volcanic activity above a subduction zone are enigmatic. We report major/trace elemental and Sr-Nd-Ca isotopic compositions of a suite of magmatic samples from the Miocene Gördes calc-alkaline dacites and the Quaternary Kula alkaline basalts from Western Anatolia, Turkey. Zircon sensitive high-resolution ion microprobe dating shows that the Gördes dacites formed at ca. 18 Ma. They are characterized by high-K calc-alkaline affinities, enrichment of light rare earth elements and large-ion lithophile elements (Rb, Th, U), positive Pb anomalies and negative Eu and high field strength element (HFSE) (Nb, Ta, and Ti) anomalies with high 87Sr/86Sri (0.70977−0.71010) and low εNd(t) (−7.3 to −7.0). They have small δ44/40Ca (0.59‰−0.69‰) values lower than that of mid-ocean ridge basalt (0.83% ± 0.11%). The Kula basalts exhibit alkaline affinities and are more enriched in incompatible elements (e.g., Rb, Ba, Th, Nb, Ta, Pb, and Sr) than those of typical oceanic island basalts. Their low 87Sr/86Sri (0.70307−0.70343) and high εNd(t) (+4.2 to +6.5) are consistent with an asthenospheric mantle source. Their δ44/40Ca values, ranging from 0.67‰ to 0.81‰, are higher than those of the dacites and lower than that of the bulk silicate earth (0.94‰ ± 0.10‰). Geochemical modeling suggests that variable degrees of partial melting of an asthenospheric mantle involving recycled oceanic crust may have resulted in low δ44/40Ca values of the Kula basalts. Combined with reported mantle tomographic images, we interpret that sub-slab asthenospheric mantle upwelling through a slab tear during subduction roll-back may have played a key role in the Sr-Nd-Ca isotopic variability from Miocene calc-alkaline dacites to Quaternary alkaline basalts in Western Anatolia.

Rubidium isotopic fractionation during magmatic processes and the composition of the bulk silicate Earth

Rubidium is a moderately volatile element with high incompatibility and fluid mobility. Its stable isotopes have great potential in tracing various geological processes. For example, lunar rocks are isotopically heavier than terrestrial ones, suggesting volatile loss by evaporation during or following the formation of the Moon. However, these studies rely on a poorly constrained estimate for the composition of the Earth’s mantle and a poor understanding of high-temperature processes which may act to fractionate stable Rb isotopes. It is therefore important to precisely characterize different rock types that sample the Earth’s mantle as well as to evaluate the importance of key isotopic fractionation processes. In order to address these issues, we established a high precision analytical method for Rb isotopic measurements using the Nu Sapphire CC-MC-ICP-MS (collision-cell multi-collector inductively coupled plasma mass spectrometer). In addition, we present a series of Rb isotopic data of volcanic rocks from Hekla volcano (Iceland) and MORB (mid-ocean ridge basalt) samples. We show that our method returns a high Rb sensitivity (∼500 V/μg·g−1 for 85Rb) and a long-term reproducibility of 0.03‰ on δ87Rb (the permil deviation of the 87Rb/85Rb ratio from the SRM 984 standard). This method uses a 2 ng/g Rb solution for analyses, allowing us to consume about 10 times less Rb to achieve similar or better precision than previous studies. Using this method, seven geostandards and one synthetic standard return Rb isotopic data consistent with previous work. Twenty-one Hekla volcanic rocks, spanning compositions from basalt to rhyolite, show limited Rb isotopic variation, with δ87Rb values varying from −0.17‰ to −0.07‰, demonstrating that magmatic evolution has an insignificant effect on Rb isotope ratios. A set of MORB samples (n = 15) from different mid-ocean ridges also span a limited Rb isotopic variation, displaying a range similar to the Hekla rock suite (−0.19 to −0.02‰). Combining our new data together with previously reported OIB data gives an average δ87Rb value of −0.12 ± 0.08‰ (2SD, n = 25), representing the current best estimate of the mantle’s isotopic composition. Considering the δ87Rb values of the upper continental crust (−0.14 ± 0.01‰, 2SE, n = 73), as inferred from recent measurements of granites, loess and sediments, and assuming this value represents the whole crust, the revised Rb isotopic composition of the bulk silicate Earth (and by extension the bulk Earth, assuming no Rb partitioned into the core) is −0.13 ± 0.06‰ (2SD).

Light stable Cr isotope compositions of mid-ocean ridge basalts: Implications for mantle source composition

Stable isotopes of chromium (Cr) and iron (Fe), both multivalent elements in silicate melts, are valuable proxies of redox conditions in magmatic systems. Partial melting and fractional crystallization have been identified as the dominant fractionation processes for Cr and Fe isotopes in high-temperature reservoirs that can lead to small but detectable isotope variations in igneous rocks. More recently, however, the source mineralogy has been suggested to play an important role, which can ultimately affect the magnitude of Cr and Fe isotopic fractionation during partial melting and thus the Cr and Fe isotopic composition of the originating melt.With the aim to investigate the role of magmatic processes and the source Cr and Fe isotope signatures on the isotope composition of these transition metals in normal mid-ocean ridge basalts (N-MORB), we analyzed 21 well-characterized MORB glasses from the southern East-Pacific Rise, Pacific-Antarctic Ridge and Mid-Atlantic Ridge. The Cr isotope compositions of N-MORB show a narrow range from −0.278 to −0.186 ‰ in δ53/52Cr (i.e., difference of a sample’s 53Cr/52Cr ratio relative to the international reference material NIST SRM979), with an average value of −0.237 ± 0.050 ‰ (2SD; n = 19). As such, the average N-MORB δ53/52Cr value is significantly lower than that of Bulk Silicate Earth of −0.12 ± 0.06 ‰ and the Cr isotope compositions reported for most ocean island basalts ranging from ∼ −0.23 to 0.00 ‰. Iron isotopic compositions of these MORB range from +0.032 to +0.137 ‰ in δ56/54Fe (i.e., difference of a sample’s 56Fe/54Fe ratio relative to the international reference material IRMM014) and are within the range of previously published MORB data. The lack of correlations between N-MORB Cr isotope signatures and indices of magmatic differentiation suggests a negligible control of this process on Cr stable isotopes. Partial melting modeling using pMELTS provides evidence that the significant offset towards lower δ53/52Cr values of N-MORB cannot be produced by decompression partial melting of a lherzolithic source such as the depleted upper mantle alone. Instead, we suggest that the lower δ53/52Cr values of MORB could be linked to intrinsic small-scale 53Cr-depleted pyroxene-rich domains in the upper mantle that influence the melt’s Cr isotopic budget during partial melting.

The nature of the Karoo mantle source region from the perspective of olivine in the Luenha picrites: an oxygen isotopic and chemical study

The Karoo large igneous province has been divided into rift zone and basin-related groups, with picrites from the Luenha river, Mozambique, representing an end-member of the latter. New O isotope, major and trace element data for olivine have been combined with MELTS crystallisation modelling to deconvolve compositional diversity associated with magma differentiation from source-derived heterogeneity. Three olivines populations have been discerned as follows: (1) the ‘main trend’, which records crystallisation from a variety of magma compositions; (2) the ‘low Cu trend’, which is inferred to represent xenocrysts or antecrysts; and (3) the ‘high CaO’ olivines, which record polybaric crystallisation of a primitive, little fractionated magma. The trace element variability in olivine phenocrysts relates partially to sampling of different parts of the same overall magma transport and storage systems, and partly to heterogeneity of parental magmas and their mantle sources. When the measured δ18Oolivine values have been converted into δ18Omelt values, the mean δ18Omelt values for the ‘main trend’ and ‘low Cu’ groups are indistinguishable from each other (5.7 ± 0.1‰, 2σ); however, the mean δ18Omelt value of 6.1 ± 0.1‰ for the ‘high CaO’ group is distinctly enriched. These data record source heterogeneity and suggest contributions from two mantle sources, one with elevated δ18O, and another with more ‘typical’ mantle δ18O. Combining these data with previously reported trace element and Nd and Sr isotope data support derivation from a mantle source similar to non-chondritic bulk silicate earth, but with minor contributions (1–2% for the enriched magmas) from a recycled sedimentary component. This points to the importance of a primitive mantle source for the basin-related successions in Karoo.

The indium isotopic composition of the bulk silicate Earth

Indium (In) behaves as a moderately volatile metal during nebular and planetary processes, and its volatility depends strongly on oxygen fugacity. The In isotopic composition of the bulk silicate Earth (BSE) could provide a critical constraint on the nature of Earth’s building blocks and mechanisms that lead to its volatile depletion. However, accurately and precisely determining the isotopic composition of In of the silicate Earth is challenging due to its low abundance in igneous rocks and the presence of significant isobaric interferences on its isotopes (e.g., 113Cd+ on 113In+ and 115Sn+ on 115In+). Here, we present a purification procedure for In from rock matrices and report the first dataset of In isotopic compositions of 30 terrestrial igneous rocks, one biotite geostandard, and one carbonaceous chondrite (Allende) measured on a Nu Sapphire collision-cell equipped multi-collector inductively-coupled-plasma mass-spectrometer (CC-MC-ICP-MS) with an external reproducibility of 0.11‰ (2SD). At this level of precision, we find no statistically significant difference in the In isotopic compositions of mid-ocean-ridge basalts (MORB), oceanic island basalts (OIB), and continental flood basalts (CFB). Furthermore, Canary Islands, Iceland and Afar lavas display no analytically resolvable In isotopic variations from basalts to rhyolites. Therefore, In isotope fractionation during igneous processes is smaller than our analytical uncertainty and the In isotopic compositions of basalts are likely to be representative samples of their mantle sources. The twenty-one terrestrial basalts from diverse geological settings have an average δ115In of 0.35 ± 0.07 ‰ (2SD). This value represents the current best estimate of the In isotopic composition of the mantle as well as of the bulk silicate Earth, assuming limited In isotope fractionation during mantle partial melting, and due to the small contribution of the continental crust to the In budget (<5%). This isotopic composition provides a baseline with which to compare with chondrites and differentiated planetary bodies in future studies.

Tungsten in the mantle constrained by continental lithospheric peridotites: Less incompatible and more abundant

Isotope dilution W whole-rock data, coupled with data for other trace elements in 42 peridotite xenoliths from Tariat (Mongolia), Yangyuan and Hannuoba (China), and Letlhakane (Botswana) are used to constrain the W abundance and behavior in the continental lithospheric mantle. Tungsten concentrations are >11 ng g−1, even in the most depleted peridotites, and higher than the W abundance estimate of the bulk silicate Earth (BSE) in all but one sample. Combined W-Th-U systematics reveal fractionation of W from Th and U with increasing degree of melt depletion, while incompatible element-enriched samples display no such fractionation. Integrated melt reactions and degrees of melt depletion are estimated based on either residual Th or residual Yb abundances for each sample, relative to a BSE precursor. This allows modelling of different partial melting scenarios that consider maximized trace element retention (Th-based model; where measured Th contents are decoupled from Al2O3 contents) or correction for metasomatic trace element contributions (Yb-based model; where measured Yb contents are well-coupled with Al2O3 contents). Modelled residual W concentrations reveal W excesses of 11–19 ng g−1, even when correcting for metasomatic W contributions. Combined with available in situ trace element mineral data for the rock-forming silicates and spinel, which all display W abundances much higher than expected, these findings suggest that either (i) W is less incompatible than previously thought, or that (ii) the peridotites experienced W metasomatism after the last melt depletion event with subsequent re-equilibration. The latter is unlikely, as similar geochemical patterns are observed for oceanic lithospheric mantle based on published data, requiring such W metasomatism to be a global phenomenon. We thus conclude that W is not as incompatible as Th and U during upper mantle melting and potential W-Th-U fractionation mechanisms are examined to explain near-constant W/Th and W/U in mantle melts. We propose revised W concentrations for the BSE of 27 ± 16 ng g−1 to 31 ± 14 ng g−1 and for the depleted mantle of 22 ± 13 ng g−1.

Gallium isotopic constraints for the origin of the Earth-Moon system

Comparatively heavy isotopic compositions of moderately volatile elements (MVE) in lunar rocks have been advocated to reflect the loss of light isotopes during devolatilization processes from the Moon. In this study we present new gallium (Ga) isotope data for lunar highland rocks, with a focus on the Ferroan Anorthosite Suite (FAS). These are commonly thought to be direct crystallization products from the late lunar magma ocean (LMO) and should contain the majority of the Ga inventory of the Moon. As such, FAS rocks are crucial for identifying the processes that drove Ga isotope fractionation as well as for inferring the Ga isotopic composition of the bulk Moon.Our data reveal that FAS samples have a range in δ71Ga from −0.27 to 0.22‰ and are generally isotopically light in Ga compared to other lunar lithologies, but straddle values typical of terrestrial rocks. Although Ga is defined as an MVE, these Ga isotope variations do not correspond with concentrations of more volatile elements, indicating that Ga isotope variations in the FAS are not primarily controlled by devolatilization processes. Instead, the Ga isotopic compositions of bulk FAS rocks broadly correlate with the composition of plagioclase, with the calcium content of plagioclase decreasing as Ga becomes isotopically heavier. This suggests that fractionation of Ga isotopes in FAS rocks was caused by the preferential incorporation of isotopically light Ga into plagioclase during the later solidification stages of the LMO. The progressive crystallization and extraction of plagioclase forces the residual melt towards increasingly heavier Ga isotope ratios, corroborating similar conclusions derived from correlations between δ71Ga and Eu* in the mare basalt suite.Using Ga isotope partitioning calculations, we demonstrate that an isotope fractionation coefficient between plagioclase and coexisting melt of −0.3 to -0.4‰ could explain the observed range of δ71Ga values in FAS, mare basalt suite rocks, and KREEP. These calculations allow for a first order estimate of the Ga isotopic composition of the bulk silicate Moon prior to plagioclase fractionation and suggest it was close to the composition of the bulk silicate Earth. This would imply that the Moon did not lose a substantial fraction of its Ga inventory during accretion, consistent with new constraints from Rb-Sr isotope systematics that indicate the Moon's volatile deficit was primarily inherited from Theia. In conjunction with the overlap in non-mass-dependent isotope ratios, these collective observations could be reconciled if Theia and the proto-Earth formed in similar regions of the inner Solar System that were already volatile-depleted.

Carbon budget of Earth's deep mantle constrained by petrogenesis of silica-poor ocean island basalts

The convecting mantle is the Earth's largest carbon reservoir that controls the carbon outflux from the interior to surface, but its carbon content remains largely uncertain due to limited constraints on carbon inventory of the deep mantle. This knowledge gap can be filled by a global assessment of CO2 in ocean island basalts (OIBs), which are associated with upwelling mantle plumes from the deep mantle. However, this approach is hindered because of limited samples of undegassed, primary melts from intraplate magmatic settings and the debate on the origin of volatile-rich alkaline OIBs from plume versus lithospheric mantle. Here we examine the origin of silica-poor, alkaline OIBs and constrain CO2 in the primary melts of global OIBs by applying the newly developed liquid-based thermobarometer and primary melt correction scheme for silica-poor mantle-derived magmas. The averaged primary melt compositions of alkaline OIBs from individual island groups are more CO2-rich (∼3–11 wt% CO2) than those of subalkaline OIBs (∼0–7 wt% CO2), but both series from the same island groups yield final equilibration with the mantle at ∼3–5 GPa and mantle potential temperatures of ∼1430–1530°C. Our findings indicate that alkaline and subalkaline OIBs are generated from carbon-rich and carbon-poor domains, respectively, in the mantle plume sources, rather than deriving from two different sources in the plume and lithosphere mantle or variable extents of melting of a source with identical carbon content. Despite the source heterogeneity between alkaline and subalkaline OIBs, CO2 in the averaged primary melts of individual island groups displays strong positive correlations with Nb and Ba, which define robust CO2/Nb (1850 ± 196) and CO2/Ba ratios (226 ± 22) for the less-degassed deep, OIB source mantle. Combining these ratios with primitive mantle Nb and Ba contents, we estimate that the deep mantle on average has ∼330–400 ppm carbon. With an averaged CO2 content of ∼4 wt% in their primary melts, global OIBs are likely produced from the deep mantle source by ∼3% melting. The mean CO2 content in the primary melts of global OIBs, together with the previous estimate of plume magma production rate, yields a flux of ∼5.5 × 1012 mol/yr CO2 or ∼66 Mt/yr carbon through ocean island volcanism. If the OIB source mantle carbon budget is mostly primordial, our estimation provides a useful reference point for undifferentiated bulk silicate Earth to further examine carbon distribution and/or loss through Earth's history.

VapoRock: Thermodynamics of Vaporized Silicate Melts for Modeling Volcanic Outgassing and Magma Ocean Atmospheres

Silicate vapors play a key role in planetary evolution, especially dominating early stages of rocky planet formation through outgassed magma ocean atmospheres. Our open-source thermodynamic modeling software “VapoRock” combines the MELTS liquid model with gas-species properties from multiple thermochemistry tables. VapoRock calculates the partial pressures of 34 gaseous species in equilibrium with magmatic liquid in the system Si–Mg–Fe–Al–Ca–Na–K–Ti–Cr–O at desired temperatures and oxygen fugacities (fO2, or partial pressure of O2). Comparison with experiments shows that pressures and melt-oxide activities (which vary over many orders of magnitude) are reproduced to within a factor of ∼3, consistent with measurement uncertainties. We also benchmark the model against a wide selection of igneous rock compositions including bulk silicate Earth, predicting elemental vapor abundances that are comparable to (Na, Ca, and Al) or more realistic than (K, Si, Mg, Fe, and Ti) those of the closed-source MAGMA code (with maximum deviations by factors of 10–300 for K and Si). Vapor abundances depend critically on the activities of liquid components. The MELTS model underpinning VapoRock was calibrated and extensively tested on natural igneous liquids. In contrast, MAGMA’s liquid model assumes ideal mixtures of a limited set of chemically simplified pseudospecies, which only roughly approximates the nonideal compositional interactions typical of many-component natural silicate melts. Finally, we explore how relative abundances of SiO and SiO2 provide a spectroscopically measurable proxy for oxygen fugacity in devolatilized exoplanetary atmospheres, potentially constraining fO2 in outgassed exoplanetary mantles.

Eoarchean-Paleoarchean crustal material in the southern North China Craton and possible mantle reservoir of early Earth

The depleted mantle and the continental crust are largely geochemically and isotopically complementary. However, the question of when the depleted mantle reservoirs developed on Earth remains a topic of considerable debate. In this study, we report the existence of a ca. 3.8 Ga detrital zircon from the quartzite of the Paleoproterozoic Songshan Group in the southern North China Craton. In situ zircon hafnium isotopic characteristics of the 3.8–3.2 Ga detrital zircons indicate the presence of source rocks as old as ca. 4.5 Ga in the southern North China Craton. Together with the global zircon U-Pb-Hf isotope dataset from the North China Craton, Jack Hills, Acasta as well as available μ142Nd values of ancient rocks from Archean craton worldwide, the new results indicate that the silicate Earth has differentiated at 4.4–4.5 Ga almost immediately after accretion, developing continental crust and a complementary depleted mantle reservoir at that same time.