Immiscible two-liquid regions in the Fe-O-S system at high pressure: Implications for planetary cores

We have determined phase relations in the Fe-O and Fe-O-S systems in the range of 15-21 GPa and 1825-2300 °C. Below the liquidus temperatures, solid FeO and metallic liquids are observed in both the Fe-O and the Fe-O-S systems. An immiscible two-liquid region exists in the Fe-O binary system in the pressure range investigated, and the immiscibility gap between Fe-rich metallic liquid and FeO-rich ionic liquid does not greatly change with either pressure or temperature. On the other hand, an immiscible two-liquid region in the Fe-O-S ternary system narrows significantly with increasing pressure at constant temperature and vice versa, and it almost disappears at 21 GPa, and 2300 °C. Immiscible two-liquid regions are thus not expected to exist in the Fe-O-S system in the Earth's core, suggesting that both oxygen and sulfur can be incorporated into the core. Our results are consistent with a geochemical model for the core containing 5.8 wt.% oxygen and 1.9 wt.% sulfur as proposed by McDonough and Sun [McDonough, W.F., Sun, S.-S., 1995. The composition of the Earth. Chem. Geol. 120, 223-253].     

Making sound inferences from geomagnetic sounding

We examine the nonlinear inverse problem of electromagnetic induction to recover electrical conductivity. As this is an ill-posed problem based on inaccurate data, there is a critical need to find the reliable features of the models of electrical conductivity. We present a method for obtaining bounds on Earth’s average conductivity that all conductivity profiles must obey. Our method is based completely on optimization theory for an all-at-once approach to inverting frequency-domain electromagnetic data. The forward modeling equations are constraints in an optimization problem solving for the electric fields and the conductivity simultaneously. There is no regularization required to solve the problem. The computational framework easily allows additional inequality constraints to be imposed, allowing us to further narrow the bounds. We draw conclusions from a global geomagnetic depth sounding data set and compare with laboratory results, inferring temperature and water content through published Boltzmann-Arrhenius conductivity models. If the upper mantle is assumed to be volatile free we find it has an average temperature of 1409-1539 ° C. For the top 1000 km of the lower mantle, we find an average temperature of 1849-2008 ° C. These are in agreement with generally accepted mantle temperatures. Our conclusions about water content of the transition zone disagree with previous research. With our bounds on conductivity, we calculate a transition zone consisting entirely of Wadsleyite has <0.27  wt.% water and as we add in a fraction of Ringwoodite, the upper bound on water content decreases proportionally. This water content is less than the 0.4 wt.% water required for melt or pooling at the 410 km seismic discontinuity.     

Phase relations of iron and iron–nickel alloys up to 300 GPa: Implications for composition and structure of the Earth's inner core

We have investigated the phase relations of iron and iron–nickel alloys with 18 to 50 wt.% Ni up to over 300 GPa using a laser-heated diamond-anvil cell. The synchrotron X-ray diffraction measurements show the wide stability of hcp-iron up to 301 GPa and 2000 K and 319 GPa and 300 K without phase transition to dhcp, orthorhombic, or bcc phases. On the other hand, the incorporation of nickel has a remarkable effect on expanding the stability field of fcc phase. The geometry of the temperature–composition phase diagram of iron–nickel alloys suggests that the hcp–fcc–liquid triple point is located at 10 to 20 wt.% Ni at the pressure of the inner core boundary. The fcc phase could crystallize depending on the nickel and silicon contents in the Earth's core, both of which are fcc stabilizer.     

Another look at the core density deficit of Earth’s outer core

A constraint adopted in several geochemical studies of core composition is that the core density deficit (cdd) is 10%, with the implication that this number is based on robust geophysical evidence. The cdd is the perceived difference between the density of pure iron at core conditions and the seismically-determined density of the outer core. The importance of the cdd is that it limits the concentration of allowable light elements, such as sulfur and silicon, which, when mixed with Fe, or an Fe-Ni alloy, comprise the geochemical model of the inner core.We present evidence that the value of 10% for the cdd of the outer core is too high. Using a thermal-pressure equation-of-state, we find that for assumed melting temperatures of pure iron at the inner-outer core (ICB) pressure of 330 GPa ranging from 7500 to 4800 K, the cdd ranges from 2.9 to 7%, respectively. Reports that the cdd value of the outer core is less than 10% are found in a number of shock-wave studies, but the values reported here are apparently the lowest. Our cdd value for an assumed melting temperature of 6000 K for iron at 330 GPa is 5.4% and is compatible with proposed concentrations of Si and S impurities found from solubility studies at high P and T.     

Melting of enstatite from 13 to 18 GPa under hydrous conditions

Experiments on MgSiO₃ enstatite were conducted in the pressure range from 13 to 18 GPa under hydrous conditions in order to clarify the effect of water on the melting phase relations of enstatite at pressures corresponding to the Earth’s mantle transition zone. In some previous experiments [Geol. Soc. Am. Bull. 79 (1968) 1685; Phys. Earth Planet. Inter. 85 (1994) 237], incongruent melting behavior to form Mg₂SiO₄ forsterite and SiO₂ enriched liquid up to 5 GPa was observed, and congruent melting behavior at pressures up to 12 GPa was observed. Under hydrous conditions, we found that the melting reaction changes from congruent to incongruent at around 13.5 GPa. Liquid formed above 13.5 GPa is enriched in MgO component relative to MgSiO₃ because it coexists with stishovite (SiO₂). Moreover, the solidus temperature decreases drastically at around 13.5 GPa, in unison with the change in the melting reaction. The solidus temperature is about 1400 °C at 13 GPa, but approximately 900 °C at 15 GPa. Our results show that the liquidus phase changes from clinoenstatite to stishovite with increasing pressure and water content above 13.5 GPa. MgSiO₃ enstatite is one of the major constituent minerals in the Earth’s mantle, and it is expected that MgO-enriched liquid will be generated in the transition zone if water is present.     

Static compression of hydrous silicate melt and the effect of water on planetary differentiation

High pressure experiments using the sink/float method have bracketed the density of hydrous iron-rich ultrabasic silicate melt from 1.35 to 10.0 GPa at temperatures from 1400 to 1860 °C. The silicate melt composition was a 50–50 mixture of natural komatiite and synthetic fayalite. Water was added in the form of brucite Mg(OH)₂ and was present in the experimental run products at 2 wt.% and 5 wt.% levels as confirmed by microprobe analyses of total oxygen. Buoyancy marker spheres were olivines and garnets of known composition and density. The density of the silicate melt with 5 wt.% water at 2 GPa and 1500 °C is 0.192 g cm^? 3 less than the anhydrous form of this melt at the same P and T. This density difference gives a partial molar volume of water in silicate melt of ～ 7 cm³ mol^? 1, which is similar to previous studies at high pressure. The komatiite–fayalite liquids with 0 and 2 wt.% H₂O, have extrapolated density crossovers with equilibrium liquidus olivine at 8 and 9 GPa respectively, but there is no crossover for the liquid with 5 wt.% H₂O. These results are consistent with the hypothesis that dense hydrous melts could be gravitationally stable atop the 410 km discontinuity in the Earth. The results also support the notion that equilibrium liquidus olivine could float in an FeO-rich hydrous martian magma ocean. Extrapolation of the data suggests that FeO-rich hydrous melt could be negatively buoyant in the Earth's D″-region or atop the core–mantle-boundary (CMB), although experiments at higher pressure are needed to confirm this prediction.     

Melting relations in the chloride–carbonate–silicate systems at high-pressure and the model for formation of alkalic diamond–forming liquids in the upper mantle

The experiments in the model system CaMgSi₂O₆–(Na₂CO₃, CaCO₃)–KCl are performed at 5 GPa and 1400–1600 °C in order to study the phase relations, including liquid immiscibility, in the chloride–carbonate–silicate systems with application to alkali and chlorine-rich liquids preserved in kimberlitic diamonds. Experiments in the boundary joins of the system demonstrated that both the carbonate–silicate and chloride–carbonate melts are homogeneous; while high-temperature (above 1800 °C) liquid immiscibility is assumed for the chloride–silicate join of the above system. Addition of silicate component into the chloride–carbonate melts and chloride component into the carbonate–silicate melts results in splitting of the homogeneous liquids into the immiscible chloride–carbonate brine and carbonate–silicate melt. Carbonate–silicate and chloride–carbonate branches of the miscibility gap converge within the carbonate-rich region of the system. Regular temperature evolution of the shape and size of the miscibility gap is deduced. With decreasing temperature, the convergence point moves toward more Si-rich compositions, expanding fields of homogeneous chloride–carbonate silica-saturated melts. This effect is governed by the precipitation of the silicate phases even from silica-bearing chloride–carbonate melts. In addition, experiments revealed regular evolution of both Cl-bearing carbonate–silicate melt and Si-bearing chloride–carbonate brine toward the low-temperature chlorine–bearing carbonatitic liquid with decreasing temperature. These trends are similar to the evolution of the melt and brine inclusions in some diamonds from Botswana, Brazil, Canada, and Yakutia, indicating their growth during cooling. The model for interaction of the chloride–carbonate brine with the mantle rocks is developed on the basis of the present experimental data. This model is applied to the chlorine-enriched kimberlites of the Udachnaya–East pipe.     

The system iron-enstatite-water at high pressures and temperatures—formation of iron hydride and some geophysical implications

The system iron-enstatite-water was investigated at pressures around 5 GPa and at temperatures ranging from 1000 to 1200°C, using several different kinds of starting materials. Quenched samples showed the coexistence of iron, olivine and pyroxene. Synthesis of the Fe-containing olivine in the run products proves that a series of reactions, Fe + H₂O → FeH_x + FeO and FeO + MgSiO₃ → (Mg, Fe)₂SiO₄, have taken place. Spherical “balls of iron” were observed in the 1200°C run. This strongly indicates that the melting temperature of iron decreased by ～ 500 K by the possible dissolution of hydrogen. Following geophysical implications are derived from these experimental results. If water was retained in the hydrous minerals in the primordial material, the iron-water reaction is expected to occur throughout the core-formation process. The reaction product FeH_x will melt and then sink to form a proto-core and iron oxide will be dissolved in the Earth's mantle. The dissolution of hydrogen in the Earth's core is a natural consequence of the core-formation process.     

Variability of the topographic core-mantle torque calculated from core surface flow models

With the prospect of studying the relevance of the topographic core-mantle coupling to the variations of the Earth’s rotation and also its applicability to constraining the core surface flow, we investigate the variability of the topographic torque estimated by using core surface flow models accompanied by (a) uncertainty due to the non-uniqueness problem in the flow inversion, and (b) variance originating in that of geomagnetic secular variation models employed in the inversion. Various flow models and their variances are estimated by inverting prescribed geomagnetic models at the epoch 1980. The subsequent topographic torque is then calculated by using a core-mantle boundary topography model obtained by seismic tomography. The calculated axial and equatorial torques are found subject to the variability of order 10¹⁹ and 10²⁰  Nm, respectively, on which (b) is more effective than (a). The variability of the torque is attributed even to (a) and (b) of the large-scale flows (degrees 2 and 3). Yet, it still seems unlikely for the decadal polar motion with the observed amplitude to be excited exclusively by the equatorial topographic torque associated with any of reasonable core surface flow models. It is also confirmed that, with the topography model adopted here, the axial topographic torque on a rigid annulus in the core (coaxial with the Earth’s rotation axis) associated with any of reasonable flow models is larger by two orders of magnitude than the plausible inertial torque on such cylinders. This implies that any core surface flow model consistent with the topographic coupling does not exist, unless the topography model is appropriately modified. Nevertheless, the topographic coupling might provide not only a weak constraint for explaining the decadal LOD variations, but also the possibility to probe the core surface flow and the core dynamics.     

The Boussinesq and anelastic liquid approximations for convection in the Earth’s core

Convection in the Earth’s core is usually studied in the Boussinesq approximation in which the compressibility of the liquid is ignored. The density of the Earth’s core varies from ICB to CMB by approximately 20%. The question of whether we need to take this variation into account in core convection and dynamo models is examined. We show that it is in the thermodynamic equations that differences between compressible and Boussinesq models become most apparent. The heat flux conducted down the adiabat is much smaller near the inner core boundary than it is near the core-mantle boundary. In consequence, the heat flux carried by convection is much larger nearer the inner core boundary than it is near the core-mantle boundary. This effect will have an important influence on dynamo models. Boussinesq models also assume implicitly that the rate of working of the gravitational and buoyancy forces, as well as the Ohmic and viscous dissipation, are small compared to the heat flux through the core. These terms are not negligible in the Earth’s core heat budget, and neglecting them makes it difficult to get a thermodynamically consistent picture of core convection. We show that the usual anelastic equations simplify considerably if the anelastic liquid approximation, valid if αT?1, where α is the coefficient of expansion and T a typical core temperature, is used. The resulting set of equations are not significantly more difficult to solve numerically than the usual Boussinesq equations. The relationship of our anelastic liquid equations to the Boussinesq equations is also examined.     

Sound velocity measurements in dhcp-FeH up to 70 GPa with inelastic X-ray scattering: Implications for the composition of the Earth's core

We have determined the density evolution of the sound velocity of dhcp-FeH_x (x ≈ 1) up to 70 GPa at room temperature, by inelastic X-ray scattering and by X-ray diffraction. We find that the variation of V_P with density is different for the ferromagnetic and nonmagnetic dhcp-FeH_x, and that only nonmagnetic dhcp-FeH_x follows Birch's law. Combining our results with Birch's law for iron and assuming an ideal two-component mixing model, we obtain an upper bound of the hydrogen content in the Earth's inner core, 0.23(6) wt.% H, corresponding to FeH_0.13(3). The iron alloy with 0.23(6) wt.% H can satisfy the density, and compressional and shear sound velocities of the PREM inner core, assuming that there are no other light elements in the inner core.     

Heterogeneous accretion,composition and core–mantle differentiation of the Earth

A model of core formation is presented that involves the Earth accreting heterogeneously through a series of impacts with smaller differentiated bodies. Each collision results in the impactor's metallic core reacting with a magma ocean before merging with the Earth's proto-core. The bulk compositions of accreting planetesimals are represented by average solar system abundances of non-volatile elements (i.e. CI-chondritic), with 22% enhancement of refractory elements and oxygen contents that are defined mainly by the Fe metal/FeO silicate ratio. Based on an anhydrous bulk chemistry, the compositions of coexisting core-forming metallic liquid and peridotitic silicate liquid are calculated by mass balance using experimentally-determined metal/silicate partition coefficients for the elements Fe, Si, O, Ni, Co, W, Nb, V, Ta and Cr. Oxygen fugacity is fixed by the partitioning of Fe between metal and silicate and depends on temperature, pressure and the oxygen content of the starting composition. Model parameters are determined by fitting the calculated mantle composition to the primitive mantle composition using least squares minimization. Models that involve homogeneous accretion or single-stage core formation do not provide acceptable fits. In the most successful models, involving 24 impacting bodies, the initial 60–70% (by mass) of the Earth accretes from highly-reduced material with the final 30–40% of accreted mass being more oxidised, which is consistent with results of dynamical accretion simulations. In order to obtain satisfactory fits for Ni, Co and W, it is required that the larger (and later) impactor cores fail to equilibrate completely before merging with the Earth's proto-core, as proposed previously on the basis of Hf-W isotopic studies. Estimated equilibration conditions may be consistent with magma oceans extending to the core–mantle boundary, thus making core formation extremely efficient. The model enables the compositional evolution of the Earth's mantle and core to be predicted throughout the course of accretion. The results are consistent with the late accretion of the Earth's water inventory, possibly with a late veneer after core formation was complete. Finally, the core is predicted to contain ~ 5 wt.% Ni, ~ 8 wt.% Si, ~ 2 wt.% S and ~ 0.5 wt.% O.     

FeSi melting curve up to 70 GPa

The melting curve of iron monosilicide, FeSi, has been determined in a laser-heated diamond anvil cell from 6 up to 70 GPa by direct visual observation of the continuous laser-speckle motion in the liquid state. At 12 GPa and 1700 K, a discontinuous change in the slope of the melting curve indicates the first-order phase transition between the ?-FeSi (B20) and the CsCl-type FeSi structures (B2). During the phase transition the coordination number of both, Fe and Si atoms, increases from 7 to 8. Above this pressure, the melting curve rises steeply but shows significant flattening at higher pressures. A comparison with the melting curve of Fe shows that both curves cross at 32 ± 3 GPa, FeSi having higher melting temperatures (about 100 K) at high pressures.     

The Effects of the Concentration of Olivine Xenocrysts on the Viscosity of Kimberlite Melts: Experimental Evidence

The study of viscosity in sub-liquidus heterogeneous media, which includes kimberlite magma at the pressures and temperatures that prevail in the mantle, is an urgent task. We have conducted experiments in the serpentine–olivine, serpentine–CaCO3?olivine, and native kimberlite–olivine systems at a pressure of 4 GPa and temperatures of 1400?1600°С in a BARS high-pressure device using the technique of a falling Pt pellet. The samples were examined after experiments to find fine-grained chilled mass of crystals where the Pt pellet was observed at the time of chilling. The concentration of the solid phase was varied in the experiments between 10 and 50 wt %. We showed that when 50 wt % of olivine grains has been introduced, it was not possible to detect the motion of the Pt pellet, while when the concentration of olivine xenocrysts reached 10 wt %, the Pt pellet very rapidly descended to the bottom of the reaction volume. Viscosity was calculated using the Stokes method. We found that the viscosity of a homogeneous kimberlite melt at 4 GPa and 1600°С is below 2 Pa s, with the viscosity of a melt that contained up to 10 wt % of the solid phase being approximately constant. A kimberlite melt that contained 30 wt % of the solid phase had a viscosity on the order of 100 Pa s, while with 50 wt % of the solid phase the relative viscosity of an ultrabasic system increased to reach values over 1000 Pa s.     

Effect of hydrogen on the melting temperature of FeS at high pressure: Implications for the core of Ganymede

We have carried out in situ X-ray diffraction experiments on the FeS–H system up to 16.5 GPa and 1723 K using a Kawai-type multianvil high-pressure apparatus employing synchrotron X-ray radiation. Hydrogen was supplied to FeS from the thermal decomposition of LiAlH₄, and FeSH_x was formed at high pressures and temperatures. The melting temperature and phase relationships of FeSH_x were determined based on in situ powder X-ray diffraction data. The melting temperature of FeSH_x was reduced by 150–250 K comparing with that of pure FeS. The hydrogen concentration in FeSH_x was determined to be x = 0.2–0.4 just before melting occurred between 3.0 and 16.5 GPa. It is considered that sulfur is the major light element in the core of Ganymede, one of the Galilean satellites of Jupiter. Although the interior of Ganymede is differentiated today, the silicate rock and the iron alloy mixed with H₂O, and the iron alloy could react with H₂O (as ice or water) or the hydrous silicate before the differentiation occurred in an early period, resulting in a formation of iron hydride. Therefore, Ganymede's core may be composed of an Fe–S–H system. According to our results, hydrogen dissolved in Ganymede's core lowers the melting temperature of the core composition, and so today, the core could have solid FeSH_x inner core and liquid FeH_x–FeSH_x outer core and the present core temperature is considered to be relatively low.     

Compaction in a mantle with a very small melt concentration: Implications for the generation of carbonatitic and carbonate-bearing high alkaline mafic melt impregnations

Xenoliths entrained in alkaline basalts and kimberlites give strong evidence that mantle carbonatitic and carbonated high alkaline mafic silicate melts, which are initially produced at very low degrees of partial melting (?1%), percolate and accumulate to form impregnations with a melt concentration of up to 10%. At present no compaction model has explained such huge local amplification of melt concentration. Recently, Bercovici et al. [1] have shown that the commonly used equations of compaction are not sufficiently general to describe all melt percolation processes in the mantle. In particular, they show that, when the melt concentration in the mantle is very low, the pressure jump ΔP between the solid and liquid fractions of the mantle mush is very important and plays a driving role during compaction. 1-D compaction waves generated with two different systems of equations are computed. Three types of wave-trains are observed, i.e. (1) sinusoidal waves; (2) periodic waves with flat minima and very acute maxima (‘witch hat waves’); (3) periodic solitary waves with flat maxima and extremely narrow minima (‘bowler hat waves’). When the initial melt distribution in the mantle is quite homogeneous, the compaction waves have sinusoidal shapes and can locally amplify the melt concentration by a factor less than two. When there is a drastic obstruction at the top of the wetted domain, the pressure jump ΔP between solid and liquid controls the shape of the waves. If the computation assumes the equality of pressure between the two phases (ΔP=0), the compaction wave has a ‘bowler hat shape’, and locally amplifies the melt concentration by a factor less than 5. Alternatively, simulations taking into account the pressure jump between phases ΔP predict compaction waves with ‘witch hat shape’. These waves collect a large quantity of melt promoting the development of magmons with local melt concentration exceeding 100× the background melt concentration. It is inferred that in a mantle with very low concentrations of carbonatitic or high alkaline mafic silicate melt the magmons are about 1 km thick and reach, in less than 1 Ma, a melt concentration of about 10%. The magmons are likely generated below the lithosphere at some distance away from the center of hot spots. This can explain the development of mantle carbonatitic eruptions in the African rift and the carbonatite and high alkaline mafic silicate volcanic activity in oceanic islands.     

A numerical model of the fractionation of olivine and molten sulfide from komatiite magma

A numerical model has been formulated that simulates the differentiation of mafic and ultramafic magmas by the fractionation of olivine and molten sulfide. The model is used to simulate the low-pressure differentiation of a komatiite magma series under both sulfide-undersaturated and sulfide-saturated conditions. Under sulfide-saturated conditions, the molecular ratio of olivine to sulfide removed from the silicate liquid is39 ± 2. Separation of this relatively small proportion of sulfide melt results in significantly different chemical trends in derivative liquids and fractionated material than are produced in the sulfide-undersaturated system, and this observation may be useful in mineral exploration. Comparison of the model results with published analyses of natural rocks indicate that the liquid equivalent members of the komatiite suite at Yakabindie, Western Australia, could be derivative liquids produced by fractional crystallization of olivine from a sulfide-undersaturated parental magma containing about 32 wt.% MgO. Derivation of a komatiitic pyroxenite with 20 wt.% MgO would require fractionation of 43.4 mol.% olivine whereas production of a komatiitic basalt with 12 wt.% MgO would involve removal of 58.5 mol.% olivine. Synvolcanic intrusive dunitic lenses at Yakabindie could have been produced by accumulation of material separated during about 3.8 mol.% fractionation of a similar parental magma, but the concentration of chalcophile elements in these bodies requires that the magma was sulfide-saturated.     

Self-diffusion in liquid Fe at high pressure

The self-diffusivity of liquid iron has been measured between 2 and 20 GPa and 1883 and 2393 K. There is a large pressure dependence across the pressure range studied and simple Arrhenius extrapolation to outer core conditions yields unrealistically low diffusivities. The large pressure and temperature range covered, however, allows different a priori diffusion models to be tested using the experimental results. The free volume model [J. Chem. Phys. 31 (1959) 1164] is found to be in excellent agreement with the experimental data and provides a basis for extrapolation of the results to outer core conditions. An extrapolation to core–mantle boundary (CMB) pressure at 4300 K yields a self-diffusivity for liquid iron of 2.5×10⁻⁵ cm² s⁻¹, which corresponds to a viscosity of 11 mPa s, in reasonable agreement with previous, theoretical, estimates.     

Experimental constraints on the chemical evolution of large icy satellites

We derive experimental constraints on the interior structure and mineralogy of large icy satellites by reacting material of chondritic chemistry with water at a pressure of 1.5 GPa, temperatures between 300°C and 800°C and a range of oxygen fugacities. Our results document the existence of three chemical processes that probably occur in large icy satellites as a result of high pressure hydrothermal processing: (1) the formation of low-density hydrated silicates, (2) the alloying of iron and sulfur to form FeS-dominated cores, and (3) the instability of organic material relative to carbonates. We construct new internal models of the thermal and structural state of Ganymede, and infer that the magnetic field of this body arises from convection within a mostly iron sulfide core. Simple thermochemical calculations are conducted to further explore the likely effects of composition and oxygen fugacity on the high pressure chemistry undergone by organic material within icy satellites. Both experimental and calculated results show that primordial organics are likely to have been largely oxidized to carbonates through hydrothermal processing early in Ganymede’s history, potentially sterilizing Ganymede’s H₂O layer.     

Geological settings and metallogenesis of high-grade iron deposits in China

The predominant types of high-grade iron deposits in China include skarn, sedimentary metamorphic(banded ironformation, BIF-type), continental/submarine volcanic-hosted and magmatic Fe-Ti-Voxide deposits. Based on a comprehensive review of current studies on these deposits, this paper suggests that the oxygen concentration in atmosphere played an important role for the formation of BIFs, whereas the tectonic setting and deep magmatic differentiation processes are more important for the other types. Notably, both high temperature and high pressure experiments and melt inclusion studies indicate that during the differentiation, high temperature magmas could develop iron-rich magma via liquid immiscibility but not pure oxide melt("iron ore magma"). Fe-P melt could be generated directly by liquid immiscibility under hydrous and oxidized condition. The formation of high-grade iron deposits is mostly associated with the processes related to multiple stages of superimposition, e.g., desiliconization and iron enrichment, removal of impurity, and remobilization and re-precipitation of iron. According to the temporal evolution, the high-grade iron deposit could be divided into multi-episode superimposition type(temporally discontinuous mineralization) and multi-stage superimposition type(temporally continuous mineralization). The former is represented by the sedimentary metamorphic iron deposit, and the latter includes those related to magmatic-hydrothermal fluids(e.g., skarn,volcanic-hosted and magmatic types).