Redox freezing and melting in the Earth's deep mantle resulting from carbon-iron redox coupling

Very low seismic velocity anomalies in the Earth's mantle may reflect small amounts of melt present in the peridotite matrix, and the onset of melting in the Earth's upper mantle is likely to be triggered by the presence of small amounts of carbonate. Such carbonates stem from subducted oceanic lithosphere in part buried to depths below the 660-kilometre discontinuity and remixed into the mantle. Here we demonstrate that carbonate-induced melting may occur in deeply subducted lithosphere at near-adiabatic temperatures in the Earth's transition zone and lower mantle. We show experimentally that these carbonatite melts are unstable when infiltrating ambient mantle and are reduced to immobile diamond when recycled at depths greater than ～250?kilometres, where mantle redox conditions are determined by the presence of an (Fe,Ni) metal phase. This 'redox freezing' process leads to diamond-enriched mantle domains in which the Fe(0), resulting from Fe(2+) disproportionation in perovskites and garnet, is consumed but the Fe(3+) preserved. When such carbon-enriched mantle heterogeneities become part of the upwelling mantle, diamond will inevitably react with the Fe(3+) leading to true carbonatite redox melting at ～660 and ～250 kilometres depth to form deep-seated melts in the Earth's mantle.     

The 'zero charge' partitioning behaviour of noble gases during mantle melting

Noble-gas geochemistry is an important tool for understanding planetary processes from accretion to mantle dynamics and atmospheric formation. Central to much of the modelling of such processes is the crystal-melt partitioning of noble gases during mantle melting, magma ascent and near-surface degassing. Geochemists have traditionally considered the 'inert' noble gases to be extremely incompatible elements, with almost 100 per cent extraction efficiency from the solid phase during melting processes. Previously published experimental data on partitioning between crystalline silicates and melts has, however, suggested that noble gases approach compatible behaviour, and a significant proportion should therefore remain in the mantle during melt extraction. Here we present experimental data to show that noble gases are more incompatible than previously demonstrated, but not necessarily to the extent assumed or required by geochemical models. Independent atomistic computer simulations indicate that noble gases can be considered as species of 'zero charge' incorporated at crystal lattice sites. Together with the lattice strain model, this provides a theoretical framework with which to model noble-gas geochemistry as a function of residual mantle mineralogy.     

Melting in the Earth's deep upper mantle caused by carbon dioxide

The onset of partial melting beneath mid-ocean ridges governs the cycling of highly incompatible elements from the mantle to the crust, the flux of key volatiles (such as CO2, He and Ar) and the rheological properties of the upper mantle. Geophysical observations indicate that melting beneath ridges begins at depths approaching 300 km, but the cause of this melting has remained unclear. Here we determine the solidus of carbonated peridotite from 3 to 10 GPa and demonstrate that melting beneath ridges may occur at depths up to 330 km, producing 0.03-0.3% carbonatite liquid. We argue that these melts promote recrystallization and realignment of the mineral matrix, which may explain the geophysical observations. Extraction of incipient carbonatite melts from deep within the oceanic mantle produces an abundant source of metasomatic fluids and a vast mantle residue depleted in highly incompatible elements and fractionated in key parent-daughter elements. We infer that carbon, helium, argon and highly incompatible heat-producing elements (such as uranium, thorium and potassium) are efficiently scavenged from depths of approximately 200-330 km in the upper mantle.     

Survival times of anomalous melt inclusions from element diffusion in olivine and chromite

The chemical composition of basaltic magma erupted at the Earth's surface is the end product of a complex series of processes, beginning with partial melting and melt extraction from a mantle source and ending with fractional crystallization and crustal assimilation at lower pressures. It has been proposed that studying inclusions of melt trapped in early crystallizing phenocrysts such as Mg-rich olivine and chromite may help petrologists to see beyond the later-stage processes and back to the origin of the partial melts in the mantle. Melt inclusion suites often span a much greater compositional range than associated erupted lavas, and a significant minority of inclusions carry distinct compositions that have been claimed to sample melts from earlier stages of melt production, preserving separate contributions from mantle heterogeneities. This hypothesis is underpinned by the assumption that melt inclusions, once trapped, remain chemically isolated from the external magma for all elements except those that are compatible in the host minerals. Here we show that the fluxes of rare-earth elements through olivine and chromite by lattice diffusion are sufficiently rapid at magmatic temperatures to re-equilibrate completely the rare-earth-element patterns of trapped melt inclusions in times that are short compared to those estimated for the production and ascent of mantle-derived magma or for magma residence in the crust. Phenocryst-hosted melt inclusions with anomalous trace-element signatures must therefore form shortly before magma eruption and cooling. We conclude that the assumption of chemical isolation of incompatible elements in olivine- and chromite-hosted melt inclusions is not valid, and we call for re-evaluation of the popular interpretation that anomalous melt inclusions represent preserved samples of unmodified mantle melts.     

Primary carbonatite melt from deeply subducted oceanic crust

Partial melting in the Earth's mantle plays an important part in generating the geochemical and isotopic diversity observed in volcanic rocks at the surface. Identifying the composition of these primary melts in the mantle is crucial for establishing links between mantle geochemical 'reservoirs' and fundamental geodynamic processes. Mineral inclusions in natural diamonds have provided a unique window into such deep mantle processes. Here we provide experimental and geochemical evidence that silicate mineral inclusions in diamonds from Juina, Brazil, crystallized from primary and evolved carbonatite melts in the mantle transition zone and deep upper mantle. The incompatible trace element abundances calculated for a melt coexisting with a calcium-titanium-silicate perovskite inclusion indicate deep melting of carbonated oceanic crust, probably at transition-zone depths. Further to perovskite, calcic-majorite garnet inclusions record crystallization in the deep upper mantle from an evolved melt that closely resembles estimates of primitive carbonatite on the basis of volcanic rocks. Small-degree melts of subducted crust can be viewed as agents of chemical mass-transfer in the upper mantle and transition zone, leaving a chemical imprint of ocean crust that can possibly endure for billions of years.     

Mineralogy of the mid-ocean-ridge basalt source from neodymium isotopic composition of abyssal peridotites

Inferring the melting process at mid-ocean ridges, and the physical conditions under which melting takes place, usually relies on the assumption of compositional similarity between all mid-ocean-ridge basalt sources. Models of mantle melting therefore tend to be restricted to those that consider the presence of only one lithology in the mantle, peridotite. Evidence from xenoliths and peridotite massifs show that after peridotite, pyroxenite and eclogite are the most abundant rock types in the mantle. But at mid-ocean ridges, where most of the melting takes place, and in ophiolites, pyroxenite is rarely found. Here we present neodymium isotopic compositions of abyssal peridotites to investigate whether peridotite can indeed be the sole source for mid-ocean-ridge basalts. By comparing the isotopic compositions of basalts and peridotites at two segments of the southwest Indian ridge, we show that a component other than peridotite is required to explain the low end of the (143)Nd/(144)Nd variations of the basalts. This component is likely to have a lower melting temperature than peridotite, such as pyroxenite or eclogite, which could explain why it is not observed at mid-ocean ridges.     

Recycling lower continental crust in the North China craton

Foundering of mafic lower continental crust into underlying convecting mantle has been proposed as one means to explain the unusually evolved chemical composition of Earth's continental crust, yet direct evidence of this process has been scarce. Here we report that Late Jurassic high-magnesium andesites, dacites and adakites (siliceous lavas with high strontium and low heavy-rare-earth element and yttrium contents) from the North China craton have chemical and petrographic features consistent with their origin as partial melts of eclogite that subsequently interacted with mantle peridotite. Similar features observed in adakites and some Archaean sodium-rich granitoids of the tonalite-trondhjemite-granodiorite series have been interpreted to result from interaction of slab melts with the mantle wedge. Unlike their arc-related counterparts, however, the Chinese magmas carry inherited Archaean zircons and have neodymium and strontium isotopic compositions overlapping those of eclogite xenoliths derived from the lower crust of the North China craton. Such features cannot be produced by crustal assimilation of slab melts, given the high Mg#, nickel and chromium contents of the lavas. We infer that the Chinese lavas derive from ancient mafic lower crust that foundered into the convecting mantle and subsequently melted and interacted with peridotite. We suggest that lower crustal foundering occurred within the North China craton during the Late Jurassic, and thus provides constraints on the timing of lithosphere removal beneath the North China craton.     

Petrology and thermal structure of the Hawaiian plume from Mauna Kea volcano

There is uncertainty about whether the abundant tholeiitic lavas on Hawaii are the product of melt from peridotite or pyroxenite/eclogite rocks. Using a parameterization of melting experiments on peridotite with glass analyses from the Hawaii Scientific Deep Project 2 on Mauna Kea volcano, I show here that a small population of the core samples had fractionated from a peridotite-source primary magma. Most lavas, however, differentiated from magmas that were too deficient in CaO and enriched in NiO (ref. 2) to have formed from a peridotite source. For these, experiments indicate that they were produced by the melting of garnet pyroxenite, a lithology that had formed in a second stage by reaction of peridotite with partial melts of subducted oceanic crust. Samples in the Hawaiian core are therefore consistent with previous suggestions that pyroxenite occurs in a host peridotite, and both contribute to melt production. Primary magma compositions vary down the drill core, and these reveal evidence for temperature variations within the underlying mantle plume. Mauna Kea magmatism is represented in other Hawaiian volcanoes, and provides a key for a general understanding of melt production in lithologically heterogeneous mantle.     

Trace element characteristics in the diopsides of peridotite xenoliths: a laser ablation-inductively coupled plasma-mass spectrometry study

Among the various xenoliths entrained by the Cenozoic Hannuoba basalts, peridotite is the most abundant one. The trace elements of the diopsides from the peridotite xenoliths were analysed by LA-ICP-MS. The overall depletion and some heterogeneity of the continental mantle beneath northern North China Craton were indicated by the characteristics of the trace elements. The ∑REE amount in diopside correlates with the Cr/(Cr+Al) ratio of diopside which is indicative of xenolith's partial melting degree. As the peridotite hosts and pyroxenite veins show similar REE distribution patterns, the composite xenoliths are probably formed by mantle deformation, rather than by the late metasomatism of mantle fluids/melts.     

Spin crossover and iron-rich silicate melt in the Earth's deep mantle

A melt has greater volume than a silicate solid of the same composition. But this difference diminishes at high pressure, and the possibility that a melt sufficiently enriched in the heavy element iron might then become more dense than solids at the pressures in the interior of the Earth (and other terrestrial bodies) has long been a source of considerable speculation. The occurrence of such dense silicate melts in the Earth's lowermost mantle would carry important consequences for its physical and chemical evolution and could provide a unifying model for explaining a variety of observed features in the core-mantle boundary region. Recent theoretical calculations combined with estimates of iron partitioning between (Mg,Fe)SiO(3) perovskite and melt at shallower mantle conditions suggest that melt is more dense than solids at pressures in the Earth's deepest mantle, consistent with analysis of shockwave experiments. Here we extend measurements of iron partitioning over the entire mantle pressure range, and find a precipitous change at pressures greater than ～76?GPa, resulting in strong iron enrichment in melts. Additional X-ray emission spectroscopy measurements on (Mg(0.95)Fe(0.05))SiO(3) glass indicate a spin collapse around 70?GPa, suggesting that the observed change in iron partitioning could be explained by a spin crossover of iron (from high-spin to low-spin) in silicate melt. These results imply that (Mg,Fe)SiO(3) liquid becomes more dense than coexisting solid at ～1,800?km depth in the lower mantle. Soon after the Earth's formation, the heat dissipated by accretion and internal differentiation could have produced a dense melt layer up to ～1,000?km in thickness underneath the solid mantle. We also infer that (Mg,Fe)SiO(3) perovskite is on the liquidus at deep mantle conditions, and predict that fractional crystallization of dense magma would have evolved towards an iron-rich and silicon-poor composition, consistent with seismic inferences of structures in the core-mantle boundary region.     

Trace element signature of subduction-zone fluids, melts and supercritical liquids at 120-180 km depth

Fluids and melts liberated from subducting oceanic crust recycle lithophile elements back into the mantle wedge, facilitate melting and ultimately lead to prolific subduction-zone arc volcanism. The nature and composition of the mobile phases generated in the subducting slab at high pressures have, however, remained largely unknown. Here we report direct LA-ICPMS measurements of the composition of fluids and melts equilibrated with a basaltic eclogite at pressures equivalent to depths in the Earth of 120-180 km and temperatures of 700-1,200 degrees C. The resultant liquid/mineral partition coefficients constrain the recycling rates of key elements. The dichotomy of dehydration versus melting at 120 km depth is expressed through contrasting behaviour of many trace elements (U/Th, Sr, Ba, Be and the light rare-earth elements). At pressures equivalent to 180 km depth, however, a supercritical liquid with melt-like solubilities for the investigated trace elements is observed, even at low temperatures. This mobilizes most of the key trace elements (except the heavy rare-earth elements, Y and Sc) and thus limits fluid-phase transfer of geochemical signatures in subduction zones to pressures less than 6 GPa.     

Whole-mantle convection and the transition-zone water filter

Because of their distinct chemical signatures, ocean-island and mid-ocean-ridge basalts are traditionally inferred to arise from separate, isolated reservoirs in the Earth's mantle. Such mantle reservoir models, however, typically satisfy geochemical constraints, but not geophysical observations. Here we propose an alternative hypothesis that, rather than being divided into isolated reservoirs, the mantle is filtered at the 410-km-deep discontinuity. We propose that, as the ascending ambient mantle (forced up by the downward flux of subducting slabs) rises out of the high-water-solubility transition zone (between the 660 km and 410 km discontinuities) into the low-solubility upper mantle above 410 km, it undergoes dehydration-induced partial melting that filters out incompatible elements. The filtered, dry and depleted solid phase continues to rise to become the source material for mid-ocean-ridge basalts. The wet, enriched melt residue may be denser than the surrounding solid and accordingly trapped at the 410 km boundary until slab entrainment returns it to the deeper mantle. The filter could be suppressed for both mantle plumes (which therefore generate wetter and more enriched ocean-island basalts) as well as the hotter Archaean mantle (thereby allowing for early production of enriched continental crust). We propose that the transition-zone water-filter model can explain many geochemical observations while avoiding the major pitfalls of invoking isolated mantle reservoirs.     

An olivine-free mantle source of Hawaiian shield basalts

More than 50 per cent of the Earth's upper mantle consists of olivine and it is generally thought that mantle-derived melts are generated in equilibrium with this mineral. Here, however, we show that the unusually high nickel and silicon contents of most parental Hawaiian magmas are inconsistent with a deep olivine-bearing source, because this mineral together with pyroxene buffers both nickel and silicon at lower levels. This can be resolved if the olivine of the mantle peridotite is consumed by reaction with melts derived from recycled oceanic crust, to form a secondary pyroxenitic source. Our modelling shows that more than half of Hawaiian magmas formed during the past 1 Myr came from this source. In addition, we estimate that the proportion of recycled (oceanic) crust varies from 30 per cent near the plume centre to insignificant levels at the plume edge. These results are also consistent with volcano volumes, magma volume flux and seismological observations.     

A discontinuity in mantle composition beneath the southwest Indian ridge

The composition of mid-ocean-ridge basalt is known to correlate with attributes such as ridge topography and seismic velocity in the underlying mantle, and these correlations have been interpreted to reflect variations in the average extent and mean pressures of melting during mantle upwelling. In this respect, the eastern extremity of the southwest Indian ridge is of special interest, as its mean depth of 4.7 km (ref. 4), high upper-mantle seismic wave velocities and thin oceanic crust of 4-5 km (ref. 6) suggest the presence of unusually cold mantle beneath the region. Here we show that basaltic glasses dredged in this zone, when compared to other sections of the global mid-ocean-ridge system, have higher Na(8.0), Sr and Al2O3 compositions, very low CaO/Al2O3 ratios relative to TiO2 and depleted heavy rare-earth element distributions. This signature cannot simply be ascribed to low-degree melting of a typical mid-ocean-ridge source mantle, as different geochemical indicators of the extent of melting are mutually inconsistent. Instead, we propose that the mantle beneath approximately 1,000 km of the southwest Indian ridge axis has a complex history involving extensive earlier melting events and interaction with partial melts of a more fertile source.     

Re-Os isotopic evidence for long-lived heterogeneity and equilibration processes in the Earth's upper mantle

The geochemical composition of the Earth's upper mantle is thought to reflect 4.5 billion years of melt extraction, as well as the recycling of crustal materials. The fractionation of rhenium and osmium during partial melting in the upper mantle makes the Re-Os isotopic system well suited for tracing the extraction of melt and recycling of the resulting mid-ocean-ridge basalt. Here we report osmium isotope compositions of more than 700 osmium-rich platinum-group element alloys derived from the upper mantle. The osmium isotopic data form a wide, essentially gaussian distribution, demonstrating that, with respect to Re-Os isotope systematics, the upper mantle is extremely heterogeneous. As depleted and enriched domains can apparently remain unequilibrated on a timescale of billions of years, effective equilibration seems to require high degrees of partial melting, such as occur under mid-ocean ridges or in back-arc settings, where percolating melts enhance the mobility of both osmium and rhenium. We infer that the gaussian shape of the osmium isotope distribution is the signature of a random mixing process between depleted and enriched domains, resulting from a 'plum pudding' distribution in the upper mantle, rather than from individual melt depletion events.     

An interconnected network of core-forming melts produced by shear deformation

The formation mechanism of terrestrial planetary cores is still poorly understood, and has been the subject of numerous experimental studies. Several mechanisms have been proposed by which metal--mainly iron with some nickel--could have been extracted from a silicate mantle to form the core. Most recent models involve gravitational sinking of molten metal or metal sulphide through a partially or fully molten mantle that is often referred to as a 'magma ocean'. Alternative models invoke percolation of molten metal along an interconnected network (that is, porous flow) through a solid silicate matrix. But experimental studies performed at high pressures have shown that, under hydrostatic conditions, these melts do not form an interconnected network, leading to the widespread assumption that formation of metallic cores requires a magma ocean. In contrast, here we present experiments which demonstrate that shear deformation to large strains can interconnect a significant fraction of initially isolated pockets of metal and metal sulphide melts in a solid matrix of polycrystalline olivine. Therefore, in a dynamic (non-hydrostatic) environment, percolation remains a viable mechanism for the segregation and migration of core-forming melts in a solid silicate mantle.     

Water-rich basalts at mid-ocean-ridge cold spots

Although water is only present in trace amounts in the suboceanic upper mantle, it is thought to play a significant role in affecting mantle viscosity, melting and the generation of crust at mid-ocean ridges. The concentration of water in oceanic basalts has been observed to stay below 0.2 wt%, except for water-rich basalts sampled near hotspots and generated by 'wet' mantle plumes. Here, however, we report unusually high water content in basaltic glasses from a cold region of the mid-ocean-ridge system in the equatorial Atlantic Ocean. These basalts are sodium-rich, having been generated by low degrees of melting of the mantle, and contain unusually high ratios of light versus heavy rare-earth elements, implying the presence of garnet in the melting region. We infer that water-rich basalts from such regions of thermal minima derive from low degrees of 'wet' melting greater than 60 km deep in the mantle, with minor dilution by melts produced by shallower 'dry' melting--a view supported by numerical modelling. We therefore conclude that oceanic basalts are water-rich not only near hotspots, but also at 'cold spots'.     

Kimberlite ascent by assimilation-fuelled buoyancy

Kimberlite magmas have the deepest origin of all terrestrial magmas and are exclusively associated with cratons. During ascent, they travel through about 150 kilometres of cratonic mantle lithosphere and entrain seemingly prohibitive loads (more than 25 per cent by volume) of mantle-derived xenoliths and xenocrysts (including diamond). Kimberlite magmas also reputedly have higher ascent rates than other xenolith-bearing magmas. Exsolution of dissolved volatiles (carbon dioxide and water) is thought to be essential to provide sufficient buoyancy for the rapid ascent of these dense, crystal-rich magmas. The cause and nature of such exsolution, however, remains elusive and is rarely specified. Here we use a series of high-temperature experiments to demonstrate a mechanism for the spontaneous, efficient and continuous production of this volatile phase. This mechanism requires parental melts of kimberlite to originate as carbonatite-like melts. In transit through the mantle lithosphere, these silica-undersaturated melts assimilate mantle minerals, especially orthopyroxene, driving the melt to more silicic compositions, and causing a marked drop in carbon dioxide solubility. The solubility drop manifests itself immediately in a continuous and vigorous exsolution of a fluid phase, thereby reducing magma density, increasing buoyancy, and driving the rapid and accelerating ascent of the increasingly kimberlitic magma. Our model provides an explanation for continuous ascent of magmas laden with high volumes of dense mantle cargo, an explanation for the chemical diversity of kimberlite, and a connection between kimberlites and cratons.     

Partial melting of the dry mafic continental crust: Implications for petrogenesis of C-type adakites

C-type adakites have been commonly considered as a result of partial melting of the mafic lower continental crust (LCC) at high pressure, as supported by high P-T experiments on hydrous basalts. However, because the mafic eclogitic LCC is generally dry, experiments on water-bearing materials cannot be used to constrain the melting processes of the dry mafic LCC. Due to the lack of systematic melting experimental studies on dry mafic rocks at crustal pressures, MELTs software was applied to simulating melting of the dry mafic LCC at 1–2 GPa. Comparison of model results with experimental data indicates that, when melting degree is greater than 20%, melts from the dry mafic LCC at 1–3 GPa cannot produce the C-type adakitic melt with high SiO₂ content (~70%). Although the limited experimental results about dry mafic rock melting at 1–2 GPa in the literature suggest that low degree melting (<10%) cannot produce silicic melt either, MELTs software simulation shows that, at pressure >1.8 GPa, low-degree melting can produce dacitic melt with high K₂O/Na₂O (~1) if SiO₂ content of the melt is controlled by residual garnet. Furthermore, the simulation also suggests that, if pressure is <1.8 GPa, abundant plagioclase (plg) in the residual phase may decrease SiO₂ content in the melt to below 62%, much lower than that of the C-type adakites observed in eastern China. Given the high P-T conditions required to produce melts with high SiO₂ and extremely low HREE contents, such melts could easily be contaminated by other crustal-derived melts, implying that the C-type adakites from eclogite melting could be less commonly observed in the outcrops than previously believed. Besides the interpretation that garnet fractionates Sr, Y, and REE, high Sr/Y and La/Yb could be also produced by multiple ways such as inheriting the source features and fractional crystallizing clinopyroxene (cpx). Therefore, it may be problematic using high Sr/Y and La/Yb as criteria to identify adakites. Instead, REE patterns with strong depletion of HREE relative to MREE (e.g. high Gd/Yb) could be a better parameter to identify the role of garnet and thus adakites. Finally, geochemical models based on MELTs simulation indicate that Eu anomaly cannot be simply used to constrain the role of plg in magmatism because Eu anomaly in the melt is a function of source characteristics, oxygen fugacity (fO₂) of magmatic systems, and plg/mafic minerals mode ratio.     

The Earth's mantle

Seismological images of the Earth's mantle reveal three distinct changes in velocity structure, at depths of 410, 660 and 2,700 km. The first two are best explained by mineral phase transformations, whereas the third-the D" layer-probably reflects a change in chemical composition and thermal structure. Tomographic images of cold slabs in the lower mantle, the displacements of the 410-km and 660-km discontinuities around subduction zones, and the occurrence of small-scale heterogeneities in the lower mantle all indicate that subducted material penetrates the deep mantle, implying whole-mantle convection. In contrast, geochemical analyses of the basaltic products of mantle melting are frequently used to infer that mantle convection is layered, with the deeper mantle largely isolated from the upper mantle. We show that geochemical, seismological and heat-flow data are all consistent with whole-mantle convection provided that the observed heterogeneities are remnants of recycled oceanic and continental crust that make up about 16 and 0.3 per cent, respectively, of mantle volume.