Search

Seismic studies have found fine-scale anomalies at the core–mantle boundary (CMB), such as ultralow velocity zones (ULVZs)1,2and the core rigidity zone3,4. ULVZs have been attributed to mantle-related processes5–10, but little is known about a possible core origin. The precipitation of light elements in the outer core has been proposed to explain the core rigidity zone3, but it remains unclear what processes can lead to such precipitation. Despite its importance for the outer core11, the melting behaviour of Fe–Si–H at relevant pressure–temperature conditions is not well understood. Here we report observations of the crystallization of B2 FeSi from Fe–9wt%Si melted in the presence of hydrogen up to 125 GPa and 3,700 K by using laser-heated diamond anvil cells. Hydrogen dramatically increases the Si concentration in the B2 crystals to a molar ratio of Si:Fe ≈ 1, whereas it mostly remains in the coexisting Fe liquid. The high Si content in the B2 phase makes it stable in a solid form at the outermost core temperatures and less dense than the surrounding liquids. Consequently, the Si-rich crystallites could form, float and be sedimented to the underside of the CMB interface, and that well explains the core side rigidity anomalies3,4. If a small amount of the FeSi crystals can be incorporated into the mantle, they would form dense low-velocity structures above the CMB, which may account for some ULVZs10. The B2 FeSi precipitation promoted by H in the outermost core provides a single core-driven origin for two types of anomalies at the CMB. Such a scenario could also explain the core-like tungsten isotope signatures in ocean island basalts12, after the materials equilibrated with the precipitates are entrained to the uppermost mantle by the mantle plumes connected to ULVZs.

The Earth’s core–mantle boundary presents a dramatic change in materials, from silicate to metal. While little is known about chemical interactions between them, a thin layer with a lower velocity has been proposed at the topmost outer core (Eʹ layer) that is difficult to explain with a change in concentration of a single light element. Here we perform high-temperature and -pressure laser-heated diamond-anvil cell experiments and report the formation of SiO2and FeHxfrom a reaction between water from hydrous minerals and Fe–Si alloys at the pressure–temperature conditions relevant to the Earth’s core–mantle boundary. We suggest that, if water has been delivered to the core–mantle boundary by subduction, this reaction could enable exchange of hydrogen and silicon between the mantle and the core. The resulting H-rich, Si-deficient layer formed at the topmost core would have a lower density, stabilizing chemical stratification at the top of the core, and a lower velocity. We suggest that such chemical exchange between the core and mantle over gigayears of deep transport of water may have contributed to the formation of the putative Eʹ layer.

Light elements alloying with metallic Fe can change the properties and therefore play a key role in the structure and dynamics of planetary cores. Hydrogen and silicon are possible light elements in the rocky planets’ cores. However, hydrogen storage in Fe-Si alloy systems remains unclear at high pressures and high temperatures because of experimental difficulties. Taking advantage of pulsed laser heating combined with high-energy synchrotron X‑ray diffraction, we studied reactions between FeSi and H in laser-heated diamond-anvil cells (LHDACs) up to 61.9 GPa and 3500 K. We found that under H-saturated conditions the amount of H alloying with FeSi (0.3 and <0.1 wt% for the B20 and B2 structures, respectively) is much smaller than that in pure Fe metal (>1.8 wt%). Our experiments also suggest that H remains in the crystal structure of FeSi alloy when recovered to 1 bar. Further density functional theory (DFT) calculations indicate that the low-H solubility likely results from the highly distorted interstitial sites in the B20 and B2 structures, which are not favorable for H incorporation. The recovery of H in the B20 FeSi crystal structure at ambient conditions could open up possibilities to understand geochemical behaviors of H during core formation in future experiments. The low-H content in FeSi alloys suggests that if a planetary core is Si-rich, Si can limit the ingassing of H into the Fe-rich core.

Nitrogen (N) is extremely depleted in the bulk silicate Earth (BSE). However, whether the silicate magma ocean was as N‐poor as the present‐day BSE is unknown. We performed multi‐anvil experiments at 20 GPa and 1,673−2,073 K to determine the dihedral angle of Fe−Ni−N alloy melt in ringwoodite matrix to investigate whether percolation of Fe‐rich alloy melt in the solid mantle can explain N depletion in the BSE. The dihedral angles ranged from 112° to 137°, surpassing the wetting boundary. Our experiments suggest that N removal from the mantle by percolation of Fe‐rich alloy melt to the Earth's core is unlikely. Therefore, besides N loss to space during planetesimal and planetary differentiation, as well as its segregation into the Earth core, the stranded Fe‐rich metal in the deep mantle could be a hidden N reservoir, contributing to the anomalous depletion of N in the observable BSE. Understanding how and when the present‐day inventory of nitrogen (N) in the bulk silicate Earth (BSE) was established is important to gain insights into Earth's habitability. A key question remains as to why N is strongly depleted in the BSE than carbon and hydrogen. Efficient segregation of N into the metallic core in the final stage of Earth's formation is postulated to be one of the primary causes behind this depletion. However, it is not clear whether the silicate magma ocean (MO) was as depleted in N as the present‐day BSE due to the uncertainties in the degree of metal‐silicate equilibration during the final stages of Earth's formation. Post‐MO crystallization, Fe−Ni alloy precipitated in the reduced mantle owing to the disproportionation of ferrous iron. We used high‐pressure experiments to examine whether this Fe−Ni alloy melt can trap the excess N and percolate through the solid mantle to the core. Our experiments show that the percolation of Fe−Ni−N melt is unlikely owing to its dihedral angle in silicate phases being larger than the wetting boundary. Instead, the Fe−Ni−N alloy stranded in the mantle can be a hidden N reservoir and the present‐day BSE may not be as N‐depleted as predicted. Dihedral angles of Fe‐Ni‐N melt in ringwoodite matrix ranged from 112° to 137°, surpassing the wetting boundaryPercolation of Fe−Ni−N melt through the solid mantle cannot explain N depletion in the bulk silicate EarthFe−Ni−N alloy can be a hidden N reservoir in the mantle if excess N was present in the Earth's mantle post core‐mantle differentiation Dihedral angles of Fe‐Ni‐N melt in ringwoodite matrix ranged from 112° to 137°, surpassing the wetting boundary Percolation of Fe−Ni−N melt through the solid mantle cannot explain N depletion in the bulk silicate Earth Fe−Ni−N alloy can be a hidden N reservoir in the mantle if excess N was present in the Earth's mantle post core‐mantle differentiation

While the water storage capacities of the upper 700 km depths of the mantle have been constrained by high-pressure experiments and diamond inclusion studies, the storage capacity of the lower mantle remains controversial. A recent high-pressure experimental study on CaSiO3perovskite, which is the third most abundant mineral in the lower mantle, reported possible storage of H2O up to a few weight percent. However, the substitution mechanism for H in this phase remains unknown. We have conducted a series of density functional theory calculations under static-lattice conditions and high pressures to elucidate hydration mechanisms at the atomic scale. All of the possible dodecahedral (Ca2+→ 2H+) and octahedral (Si4+→ 4H+) substitution configurations for a tetragonal perovskite lattice have very small energy differences, suggesting the coexistence of multiples of H configurations in CaSiO3perovskite at mantle pressures and temperatures. The dodecahedral substitutions decrease the bulk modulus, resulting in a smaller unit-cell volume of hydrous CaSiO3perovskite under pressure, consistent with the experimental observations. Although the octahedral substitutions also decrease the bulk modulus, they increase the unit-cell volume at 1 bar. The H atoms substituted in the dodecahedral sites develop much less hydrogen bonding with O atoms, leading to a large distortion in the neighboring SiO6octahedra. Such distortion may be responsible for the non-cubic peak splittings observed in experiments on hydrous CaSiO3perovskite. Our calculated infrared spectra suggest that the observed broad OH modes in CaSiO3perovskite can result from the existence of multiples of H configurations in the phase. Combined with the recent experimental results, our study suggests that CaSiO3can be an important mineral phase to consider for the H2O storage in the lower mantle.

Accurate knowledge of the mineralogy is essential for understanding the lower mantle, which represents more than half of Earth’s volume. CaSiO3perovskite is believed to be the third-most-abundant mineral throughout the lower mantle, following bridgmanite and ferropericlase1–3. Here we experimentally show that the calcium solubility in bridgmanite increases steeply at about 2,300 kelvin and above 40 gigapascals to a level sufficient for a complete dissolution of all CaSiO3component in pyrolite into bridgmanite, resulting in the disappearance of CaSiO3perovskite at depths greater than about 1,800 kilometres along the geotherm4,5. Hence we propose a change from a two-perovskite domain (TPD; bridgmanite plus CaSiO3perovskite) at the shallower lower mantle to a single-perovskite domain (SPD; calcium-rich bridgmanite) at the deeper lower mantle. Iron seems to have a key role in increasing the calcium solubility in bridgmanite. The temperature-driven nature can cause large lateral variations in the depth of the TPD-to-SPD change in response to temperature variations (by more than 500 kilometres). Furthermore, the SPD should have been thicker in the past when the mantle was warmer. Our finding requires revision of the deep-mantle mineralogy models and will have an impact on our understanding of the composition, structure, dynamics and evolution of the region.

To understand the effects of H2O on the mineral phases forming under the pressure-temperature conditions of the lower mantle, we have conducted laser-heated diamond-anvil cell experiments on hydrous ringwoodite (Mg2SiO4with 1.1 wt% H2O) at pressures between 29 and 59 GPa and temperatures between 1200 and 2400 K. Our results show that hydrous ringwoodite (hRw) converts to crystalline dense hydrous silica, stishovite (Stv) or CaCl2-type SiO2(mStv), containing 1 wt% H2O together with Brd and MgO at the pressure-temperature conditions expected for shallow lower-mantle depths between approximately 660 to 1600 km. Considering the lack of sign for melting in our experiments, our preferred interpretation of the observation is that Brd partially breaks down to dense hydrous silica and periclase (Pc), forming the phase assembly Brd + Pc + Stv. The results may provide an explanation for the enigmatic coexistence of Stv and Fp inclusions in lower-mantle diamonds.

Ultra‐low velocity zones (ULVZs) are anomalous structures, generally associated with decreased seismic velocity and sometimes an increase in density, that have been detected in some locations atop the Earth's core‐mantle boundary (CMB). A wide range of ULVZ characteristics have been reported by previous studies, leading to many questions regarding their origins. The lowermost mantle beneath Antarctica and surrounding areas is not located near currently active regions of mantle upwelling or downwelling, making it a unique environment in which to study the sources of ULVZs; however, seismic sampling of this portion of the CMB has been sparse. Here, we examine core‐reflected PcP waveforms recorded by seismic stations across Antarctica using a double‐array stacking technique to further elucidate ULVZ structure beneath the southern hemisphere. Our results show widespread, variable ULVZs, some of which can be robustly modeled with 1‐D synthetics; however, others are more complex, which may reflect 2‐D or 3‐D ULVZ structure and/or ULVZs with internal velocity variability. Our findings are consistent with the concept that ULVZs can be largely explained by variable accumulations of subducted oceanic crust along the CMB. Partial melting of subducted crust and other, hydrous subducted materials may also contribute to ULVZ variability. Earth's core‐mantle boundary (CMB), the interface between the solid silicate mantle and the molten iron‐rich outer core, is associated with a range of anomalous structures, including ultra‐low velocity zones (ULVZs). While generally associated with reduced seismic wave velocities and sometimes increased density, prior studies have reported a wide range of ULVZ characteristics, leading to many questions regarding their origins. The lowermost mantle beneath the southern hemisphere provides a unique environment to study ULVZs because it is located away from regions of large‐scale mantle upwelling and downwelling. Our study uses core reflected P‐waves (PcP) recorded by seismic stations in Antarctica to investigate this portion of the CMB for ULVZ presence. We find widespread evidence for variable ULVZ structure. Some of the imaged ULVZs can be modeled with a single layer, but others are more complex. We suggest that the ULVZs beneath the southern hemisphere are predominantly associated with subducted oceanic crust that has variable accumulations along the CMB. In some regions, hydrated subducted materials may also experience partial melting, which may contribute to the complicated ULVZ structures imaged in some locations. Core‐reflected P‐waves are used to investigate ultra‐low velocity zones in the lowermost mantle beneath the southern hemisphereResults show widespread evidence for variable ultra‐low velocity zones, some of which may indicate layered or gradational structureVariable accumulations of subducted oceanic crust with localized partial melting can explain these anomalous structures Core‐reflected P‐waves are used to investigate ultra‐low velocity zones in the lowermost mantle beneath the southern hemisphere Results show widespread evidence for variable ultra‐low velocity zones, some of which may indicate layered or gradational structure Variable accumulations of subducted oceanic crust with localized partial melting can explain these anomalous structures

We have synthesized pyrite-type PtO2(py-PtO2) at 50–60 GPa and successfully recovered it at 1 bar. The observed O–O stretching vibration in Raman spectra provides direct evidence for inter-oxygen bonding in the structure. We also identified the O–H vibrations in py-PtO2synthesized from the low-temperature areas, indicating hydrogenation, py-PtO2Hx(x≤ 1). Diffraction patterns are consistent with a range of degrees of hydrogenation controlled by temperature. We found that py-PtO2has a high bulk modulus, 314 ± 4 GPa. The chemical behaviors found in py-PtO2have implications for the hydrogen storage in materials with anion–anion bonding, and the geochemistry of oxygen, hydrogen, and transition metals in the deep planetary interiors.

To find the thermodynamically stable crystal structure of CaSiO3perovskite (CaPv) at high pressure and 300 K, we have conducted synchrotron X-ray diffraction (XRD) on thermally stress-annealed samples in a Ne pressure medium in the diamond-anvil cell at 28–62 GPa. Rietveld refinements of the diffraction patterns are significantly improved in fitting the positions and intensities of the split lines of CaPv if the starting model is a tetragonal perovskite-type structure with the SiO6octahedral rotation around the tetragonal c-axis. The result is in contrast with other previous experiments, but is consistent with first-principles calculations, reconciling the discrepancy between computations and experiments on the crystal structure of CaPv. We attribute the observed difference to the formation of the thermodynamically more stable phase under improved stress conditions in our experiments. Our fitting shows that the bulk modulus of CaPv is 223 ± 6 GPa when its pressure derivative fixed to 4, which is also consistent with first-principles calculations. The previous observations of the diffraction patterns of CaPv inconsistent with the first-principles studies could be due to the formation of a metastable crystal structure of CaPv under elevated deviatoric stresses.

Recent studies have shown that mineral end-member phases (δ-AlOOH phase, phase H, and stishovite) with rutile-type or modified rutile-type crystal structures and solid solutions between them in the MgO-Al2O3-SiO2system can store large amounts of water and can be stable at high pressures and high temperatures relevant to the Earth’s lower mantle. The Al-H charge-coupled substitution (Si4+→ Al3++ H+) has been proposed to explain the storage capacity found in some of these phases. However, the amount of H+found in some recent examples does not match the expected value if such substitution is dominant, and it is difficult to explain the larger water storage in stishovite with such a mechanism alone. An octahedral version of the hydrogarnet-like substitution (Si4+→ 4H+) has been proposed to explain the incorporation of protons in Al-free, water-rich stishovite. Yet, the high-pressure structural behavior of OH in this phase has not yet been measured. In this study, we report high-pressure Raman spectroscopy measurements on Al-free hydrous stishovite with 3.2 ± 0.5 wt% water up to 55 GPa. At ambient pressure, we find that the OH stretching modes in this phase have frequencies lying in between those in low-water aluminous stishovite and those in δ-AlOOH, suggesting a strength of the hydrogen bonding intermediate between these two cases. After decompression to 1 bar, we observe modes that are similar to the IR-active modes of anhydrous and hydrous stishovite, suggesting that the existence of Si defects in the crystal structure can activate the inactive modes. For both lattice and OH-stretching modes, our data show a series of changes at pressures between 24 and 28 GPa suggesting a phase transition (likely to CaCl2-type). While some of the lattice mode behaviors are similar to what was predicted for the AlOOH polymorphs, the OH mode of our hydrous stishovite shows a positive frequency shift with pressure, which is different from δ-AlOOH. All our spectral observations suggest that water-rich pure dense silica has a distinct proton incorporation mechanism from aluminous low-water stishovite and δ-AlOOH, supporting the proposed direct substitution.

Hexagonal close‐packed (hcp) structured Fe‐Ni alloy is believed to be the dominant phase in the Earth's inner core. This phase is expected to contain 4%–5% light elements, such as Si and H. While the effects of individual light element candidates on the equation of state (EoS) of the hcp Fe metal have been studied, their combined effects remain largely unexplored. In this study, we report the equations of state for two hcp‐structured Fe‐Si‐H alloys, namely Fe0.83Si0.17H0.07and Fe0.83Si0.17H0.46, using synchrotron X‐ray diffraction measurements up to 125 GPa at 300 K. These alloys were synthesized by cold compression of Fe‐9wt%Si in either pure H2or Ar‐H2mixture medium in diamond‐anvil cells. The volume increase caused by a H atom in hcp Fe‐Si‐H alloys is approximately eight times greater than that by a Si atom. We used the improved data set to develop a composition‐dependent EoS that covers a wide range of compositions. Our calculated density and bulk sound velocity of hcp Fe‐Si‐H alloys suggest a large trade‐off between Si and H contents in fitting the seismic properties of the inner core. Combining our new EoS with geophysical and geochemical constraints, we propose 1.6–3 wt% Si and 0.15–0.6 wt% H in the Earth's inner core. The Earth's inner core is believed to be made up of solid iron and nickel alloys, but there are also some lighter elements alloyed in. We studied how silicon and hydrogen affect the density and sound wave velocity of the iron alloys in the inner core. We combine our new data with existing datasets to create a model that shows that the inner core likely contains around 1.6%–3% silicon and 0.15%–0.6% hydrogen. These amounts can explain the density and sound velocity of the inner core as observed through seismic measurements. Equations of state are measured for Fe0.83Si0.17H0.07and Fe0.83Si0.17H0.46alloys in hexagonal close‐packed structuresInclusion of a H atom in hcp Fe‐Si‐H alloys results in a volume increase approximately eight times greater than that by a Si atomComparison of laboratory results with seismic data suggests that the Earth's inner core may contain 1.6–3 wt% Si and 0.15–0.6 wt% H Equations of state are measured for Fe0.83Si0.17H0.07and Fe0.83Si0.17H0.46alloys in hexagonal close‐packed structures Inclusion of a H atom in hcp Fe‐Si‐H alloys results in a volume increase approximately eight times greater than that by a Si atom Comparison of laboratory results with seismic data suggests that the Earth's inner core may contain 1.6–3 wt% Si and 0.15–0.6 wt% H

Although high pressure enables alloying between hydrogen and iron, hydrogen‐to‐iron molar ratio (H/Fe) so far found in experiments is mostly limited to 1 in the close‐packed iron metal under high pressure. We report a H/(Fe + Ni) ratio of 1.8 ± 0.1 from (Fe,Ni)Hx(or x≥ 1.8) quenched from liquid, exceeding the amounts so far reported for densely packed Fe alloys. From the metastable behavior of the frozen (Fe,Ni)Hxliquid during decompression, we infer that the amount is a lower bound and therefore even a greater amount of H can be dissolved in the liquid part of Fe‐rich cores of planets. The significant H storage capacity of liquid Fe‐Ni alloy is important to consider for potential storage of H in the interiors of low‐density planets as well as rocky planets. Our new high‐pressure experiments show that much more hydrogen can dissolve in the iron‐nickel alloy liquid than what is known for solid iron‐nickel alloy at high pressures and high temperatures. These experimental observations open up the possibility for large internal hydrogen storage in the metallic cores for a range of planet types. Iron‐nickel alloy quenched from liquid at high pressures contains more than 3.1 wt% hydrogenLiquid iron‐nickel alloy can contain at least 80% more H than the solidHydrogen storage capacity of the liquid part of the core could be the largest among the internal layers of planets Iron‐nickel alloy quenched from liquid at high pressures contains more than 3.1 wt% hydrogen Liquid iron‐nickel alloy can contain at least 80% more H than the solid Hydrogen storage capacity of the liquid part of the core could be the largest among the internal layers of planets

Seismic images of Earth's interior reveal two massive anomalous zones at the base of the mantle, above the core, where seismic waves travel slowly. The mantle materials that surround these anomalous regions are thought to be composed of cooler rocks associated with downward advection of former oceanic tectonic plates. However, the origin and composition of the anomalous provinces is uncertain. These zones have long been depicted as warmer-than-average mantle materials related to convective upwelling. Yet, they may also be chemically distinct from the surrounding mantle, and potentially partly composed of subducted or primordial material, and have therefore been termed thermochemical piles. From seismic, geochemical and mineral physics data, the emerging view is that these thermochemical piles appear denser than the surrounding mantle materials, are dynamically stable and long-lived, and are shaped by larger-scale mantle flow. Whether remnants of a primordial layer or later accumulations of more-dense materials, the composition of the piles is modified over time by stirring and by chemical reactions with material from the surrounding mantle, underlying core and potentially from volatile elements transported into the deep Earth by subducted plates. Upwelling mantle plumes may originate from the thermochemical piles, so the unusual chemical composition of the piles could be the source of distinct trace-element signatures observed in hotspot lavas.

In situ multianvil press (MAP) studies have reported that the depth and the Clapeyron slope of the postspinel boundary are significantly less than those of the 660 km discontinuity inferred from seismic studies. These results have raised questions about whether the postspinel transition is associated with the discontinuity. We determined the postspinel transition in pyrolitic compositions in the laser‐heated diamond anvil cell (LHDAC) combined with in situ synchrotron X‐ray diffraction. The Clapeyron slope was determined to be −2.5 ± 0.4MPa/K and did not vary significantly with compositions and used pressure scales. Using Pt scales, our data indicate that the postspinel transition occurs in pyrolitic compositions at 23.6–24.5GPa (1850K). The transition pressure and slope are consistent with the depth and topography of the 660 km discontinuity. Our data reveal that inaccuracy in pressure scales alone cannot explain the discrepancy and technical differences between MAP and LHDAC contribute significantly to the discrepancy. The postspinel transition is consistent with the 660 discontinuityThe Clapeyron slope of the postspinel boundary is −2.5 MPa/KDiscrepancy in the previous studies is likely due to technical issues

The carbon and water cycles in the Earth's interior are linked to key planetary processes, such as mantle melting, degassing, chemical differentiation, and advection. However, the role of water in the carbon exchange between the mantle and core is not well known. Here, we show experimental results of a reaction between Fe3C and H2O at pressures and temperatures of the deep mantle and core‐mantle boundary (CMB). The reaction produces diamond, FeO, and FeHx, suggesting that water can liberate carbon from the core in the form of diamond (“core carbon extraction”) while the core gains hydrogen, if subducted water reaches to the CMB. Therefore, Earth's deep water and carbon cycles can be linked. The extracted core carbon can explain a significant amount of the present‐day mantle carbon. Also, if diamond can be collected by mantle flow in the region, it can result in unusually high seismic‐velocity structures. Carbon plays a vital role in geological processes occurring in the Earth's interior. While most carbon on Earth exists in its core, whether or not the core carbon can be added to the mantle is unclear due to the lack of knowledge of possible carbon transfer mechanism at the core‐mantle boundary (CMB). We conducted experiments by reproducing the extreme pressure and temperature conditions of the CMB. Our experiments show that water can react with the metallic iron core and liberate carbon as diamond, suggesting an important relationship between Earth's water and carbon cycles. In addition, our result predicts possible existence of diamond in some regions of the deepest mantle. Water reacts with iron‐carbon alloy to form diamond at the P–Tconditions expected for Earth's core‐mantle boundarySome of Earth's mantle carbon may come from the core through water‐induced reactions at the core‐mantle boundaryDiamonds formed at the core‐mantle boundary may result in high seismic velocity structures in the region Water reacts with iron‐carbon alloy to form diamond at the P–Tconditions expected for Earth's core‐mantle boundary Some of Earth's mantle carbon may come from the core through water‐induced reactions at the core‐mantle boundary Diamonds formed at the core‐mantle boundary may result in high seismic velocity structures in the region

Lead dioxide is an important silica analog that has high-pressure behavior similar to what has been predicted for silica, only at lower pressures. We have measured the structural evolution and compressional behavior of different lead dioxide polymorphs up to 140 GPa in the laser-heated diamond-anvil cell using argon as a pressure medium. High-temperature heating prevents the formation of multi-phase mixtures found in a previous study conducted at room temperature using a silicone grease pressure medium. We find diffraction peaks consistent with a baddeleyite-type phase in our cold-compressed samples between 30 and 40 GPa, which was not observed in the previous measurements. Lead dioxide undergoes a phase transition to a cotunnite-type phase at 24 GPa. This phase remains stable to at least 140 GPa with a bulk modulus of 219(3) GPa for K′0= 4. Decompression measurements show a pure cotunnite-type phase until 10.5 GPa, where the sample converts to a mixture of baddeleyite-type, pyrite-type, and OI-type (Pbca) phases. Pure α-structured lead dioxide (scrutinyite) is found after pressure release at room pressure even though our starting material was in the β-structure (plattnerite). Pressure quenching to the α-structure appears to be a common feature of all group IVa oxides that are compressed to structures with greater density than the rutile-type structure.

Droplets of an iron-bearing calcia magnesia aluminosilicate (Fe-CMAS) melt were reacted under distinctly reducing conditions (fO₂= 2.4 × 10-13 and 6.4 × 10-15atm) at high temperature (~1400 °C) and ambient pressure. The low fO₂environment was maintained by a flowing gas mixture of CO and CO2, with a high content of CO. Molten metallic iron alloyed with silicon and carbon formed on the surface of the melt; no metal was observed in sample interiors. A color change from brown to pale blue confined to the outer layer of the melt indicated that essentially complete reduction of Fe3+to Fe2+had occurred in this region. Analysis of the reaction kinetics, particularly in comparison to melts of similar polymerization but free of CaO, reveals that the concentration of electron holes has decreased to such an extent that ionic transport in the melt is significantly slowed and the diffusion of CO as a neutral species becomes dominant and rate-limiting: molecular CO, initially incorporated into the melt as a physically dissolved species, subsequently reacts to form chemically dissolved (bonded into the melt structure) CO32-anions, consuming electron holes in the process. The chemical diffusion coefficient for CO in the reduced melt at 1400 °C is estimated as DCO≈ 4 × 10-4cm2/s, consistent with that of other, similarly sized molecular species (e.g., H2and H2O) for similarly polymerized melts, as reported by other investigators. Upon quenching, the droplet acts as a closed system. Internal redox couples see the reduction of the carbonate so as to form bubbles of CO, the composition of which are confirmed with Raman spectroscopy. The open-system reduction and closed-system quenching dynamics are analyzed following an Ellingham-diagram approach.

Significantly reduced detection of thermal radiation in gated spectroscopy allowed us to measure the Raman scattering of natural enstatite up to 1550 K at 1 bar. The intrinsic anharmonicity, ai = [(∂lnνi)/(∂T)]V, of the Raman-active modes in orthoenstatite (OEn) was obtained from temperature (T) shifts of vibrational frequencies measured in this study combined with previous high-pressure (P) Raman scattering data. Although the ai values of the lattice modes of OEn are similar to those for forsterite (Fo), the Si-O stretching modes have significantly lower intrinsic anharmonicity in OEn than in Fo, suggesting that the connectivity of the SiO4tetrahedra plays an important role in mode anharmonicity. At the phase transition at 1500 K, a doublet related to the stretching vibration of bridging O atoms in the SiO3chains becomes a singlet, and a doublet related to the stretching vibration of non-bridging O atoms remains as a doublet, consistent with the expected spectral change for a phase transition from OEn to protoenstatite. Two intense, low-frequency modes of OEn show a strong nonlinear decrease in frequencies with heating that cannot be explained solely by thermal expansion. This may indicate the reorganization of the structure around Mg atoms and unkinking of the SiO3chains at temperatures well below the phase transition

Using the CaIrO3-type structure model (space group Cmcm), lattice parameters and atomic positions of the MgGeO3post-perovskite (pPv) phase were determined based on Rietveld refinements at 78-109 GPa and first-principles calculations based on density functional theory. The reproducibility of structural parameters obtained for different samples, consistency with theoretical calculations, and good agreement with expected bond lengths based on structurally similar materials all provide evidence for both validity of CaIrO3-type structure model for the pPv phase in MgGeO3exceeding 1 Mbar and reliability of structural parameters obtained by Rietveld refinements approaching 1 Mbar. The MgGeO3pPv phase exhibits strong anisotropy in axial compressibility, with the b-axis being most compressible. The polyhedral bulk modulus for the GeO6octahedron is 1.9× larger than that for the MgO8hendecahedron. Examination of neighboring O-O distances shows that the O-O distance aligned along the a direction is one of the longest and that aligned along c is one of the shortest, and these may be related to the lower compressibility along c compared with a. Comparison of structural features of MgGeO3pPv with those for MgSiO3, NaMgF3, and CaIrO3pPv show that MgSiO3pPv has more similarity with NaMgF3and MgGeO3pPv than with CaIrO3pPv in such parameters as degree of octahedral distortion, implying that both NaMgF3and MgGeO3pPv are better analogs to MgSiO3pPv than CaIrO3pPv.

In-situ Raman measurements using a gated spectroscopy system revealed irreversible changes at 800-1000 °C in a natural red spinel (with 2 cation mol% Cr and 1 cation mol% Zn) and at 1100-1200 °C in a natural clear spinel (without Cr or Zn). Our observations of rapid broadening of a mode at 409 cm−1and the appearance of two weak modes at 210 and 520 cm−1at the transition temperature confirm the association of these features with cation disordering proposed by previous quench studies. Furthermore, we found that the frequencies of modes at 313 and 666 cm−1change at the transition temperature. The discontinuous frequency decrease of the mode at 313 cm−1and the increase in the frequency of the mode at 666 cm−1can be explained by the entrance of heavier Al atoms into the tetrahedral sites and the entrance of lighter Mg atoms into the octahedral sites, respectively. Our study demonstrates that in-situ Raman spectroscopy is a powerful tool for studying cation disordering in spinel-structured minerals at high temperature.

Raman spectra of Fe2O3were measured to 62 GPa in a diamond anvil cell. All group theoretically predicted Raman-active phonon modes were detected to 54 GPa. In addition, some high-pressure spectra show an IR-active Eumode (~660 cm-1), possibly induced by surface defects or stress. This mode is related by a factor of two to a mode at 1320 cm-1. The assignment of the 1320 cm-1mode has been controversial (two-magnon scattering or two-phonon scattering), and our observation supports the phonon assignment. All Raman-active phonons show nonlinear pressure-induced shifts. The mode Grüneisen parameters and their logarithmic volume dependences for two low-frequency phonons, A1gand Eg, become negative and infinite, respectively, near 50 GPa as a result of the instability of the corundum structure at this pressure. Using the observed Raman-active phonons together with acoustic phonons previously measured by ultrasonics, and Kieffer’s model, we calculate the phonon contribution to the thermodynamic parameters of hematite. Comparison with experimentally measured values allows an estimation of an upper bound to the magnon contribution to the heat capacity at ambient pressure. This increases continuously above the Morin temperature and reaches a maximum at the Néel temperature (~37%). The Raman spectra change dramatically at pressures greater than 54 GPa as indicated by the appearance of new peaks and a significant increase in background. Although direct structural analysis is not possible due to the low signal-to-background ratio and the lack of polarization information, we were able to examine the consistency of our Raman observation with the corundum-to-perovskite phase transformation using the results for an analog system: MgSiO3ilmenite (ordered corundum type) and perovskite. This analysis shows that observed new features in Fe2O3Raman spectra may not be consistent with the GdFeO3perovskite structure.

Equation of state fits to experimental P,V,T data were examined by the inversion of synthetic data sets using the thermoelastic parameters of MgSiO3perovskite. Our results show that by extending the pressure and temperature range to 130 GPa and 2500 K, the volume dependence of the Grüneisen parameter, q (∂lng/∂lnV), could be resolved to ~10% under the best circumstances. However, simulations also showed strong correlation between the bulk modulus, KT0, and its pressure derivative, K'T0, and q within the currently accepted uncertainty of elastic parameters for MgSiO3perovskite. We considered the effect of random error based on the reported uncertainty for different measurement techniques. Even though the laser heated diamond-anvil cell (LHDAC) technique has significantly larger temperature uncertainty, the ability to extend the pressure and temperature ranges allows for improved resolution of higher order thermodynamic parameters. However, systematic error from temperature inhomogeneity in the LHDAC sample could result in overestimation of q. We also performed Birch-Murhanghan-Debye (BMD) equation of state (EOS) fits for currently available data sets. Consistent with the simulation results, combining recent LHDAC (Fiquet et al. 1998) and resistance heated diamond-anvil cell (RHDAC) (Saxena et al. 1999) with lower P-T measurements (Ross and Hazen 1989; Wang et al. 1994; Utsumi et al. 1995; Funamori et al. 1996) we obtained q = 2.0(3) and γ0= 1.42(4). The difference between q = 2.0(3) and the normally assumed value of q =1 strongly affects calculated values for higher order thermoelastic parameters [e.g., a, (∂KT/∂T)P] as well as first order parameters, such as density and bulk modulus at lower mantle conditions. However, possible systematic error sources need to be further investigated and measurements at higher P-T conditions promise to yield better constraints on the thermoelastic parameters of MgSiO3perovskite

The stability of synthetic cation-exchanged zeolites having the chabazite framework (CHA) and their cation-water interaction were studied using high-temperature reaction calorimetry. Four cations (K, Na, Li, and Ca) were exchanged into CHA.

Microcline + albite and albite + quartz standard mixtures were quantitatively analyzed using Rietveld refinement to investigate the reliability of the method. The method was also applied to the determination of weight percentages of component minerals in a perthite sample. The analysis resulted in a standard deviation of approximately I wt% for the microcline + albite mixtures and 4 wt% for the albite + quartz mixtures, indicating that the accuracy of quantitative Rietveld analysis is far better than that of conventional XRD methods in which only a few select peaks are used. The inclusion of the Dollase-March preferred-orientation correction function slightly improved the accuracy of the results. That function also improved the rate of convergence and the accuracy of unit-cell parameters. However, the Rietveld-Toraya correction function did not work well with the quantitative analyses. The result of the quantitative analysis of the albite + quartz mixtures was poorer than that of the microcline + albite mixtures, and the weight fraction of albite tended to be overestimated. The analytical result of the microcline + albite mixtures implies that the Rietveld method is more accurate when component minerals have similar preferred-orientation characteristics. However, if component minerals have different crystal habits, as represented by the albite + quartz mixtures, the deviation is much greater and systematic errors may occur. Quantitative phase analysis of a perthite specimen using the Rietveld method resulted in very accurate values for the mineral proportion and unit-cell parameters of the constituent alkali feldspars, demonstrating the power of the Rietveld method to analyze multiphase samples.

One of the central challenges in accurately estimating the mantle melting temperature is the sensitivity of the probe for detecting a small amount of melt at the solidus. To address this, we used a multichannel collimator to enhance the diffuse X‐ray scattering from a small amount of melt and probed an eutectic pyrolitic composition to increase the amount of melt at the solidus. Our in situ detection of diffuse scattering from the pyrolitic melt determined an anhydrous melting temperature of 3,302 ± 100 K at 119 ± 6 GPa and 3,430 ± 130 K at the core‐mantle boundary (CMB) conditions, as the upper bound temperature. Our CMB temperature is approximately 700 K lower than the previous estimates, implying much faster secular cooling and higher concentrations of S, C, O, and/or H in the region, and nonlinear, advocating the basal magma ocean hypothesis. The heat stored in the Earth's deep interior has been the primary fuel for a range of global processes from mantle convection to surface tectonics, but quantitative estimation of the heat remains uncertain. The melting temperature of mantle materials is one of the key parameters to understanding the thermal evolution and present‐day state of the Earth's interior, but it has been poorly constrained, with recent measurement discrepancies as large as 600 K. Here, we report melting temperatures of mantle compositions measured over a wide range of pressures expected for the lower mantle. We investigated a eutectic mantle composition using multichannel collimator filtered X‐ray diffraction in combination with the laser‐heated diamond‐anvil cell. Fitting our melting data over the range of 46–145 GPa led to a solidus temperature of 3,430 ± 130 K at the core‐mantle boundary. This temperature is approximately 700 K lower than the previous estimates, implying much faster secular cooling at the lower mantle than previously believed. Furthermore, our solidus curve constrained for a wide pressure range is strongly nonlinear and thus supports the basal magma ocean hypothesis. Combined usage of multichannel collimator and laser‐heated diamond‐anvil cell improved detection of silicate meltThis led to the lowest solidus temperature of anhydrous pyrolite, 3,430 ± 130 K, at the core‐mantle boundary conditionsWe advocate Fe‐enriched provinces at the core‐mantle boundary to be originated from the magma ocean

Recent first‐principles calculations have proposed that bulk sound speed (VΦ) decreases at the perovskite (Pv) → postperovskite (PPv) transition. Yet some measurements suggest an increase in VΦat the PPv transition in (Mg,Fe)SiO3and NaMgF3. Our new measurements on NaMgF3up to 50 GPa under quasi‐hydrostatic stress conditions reveal that the VΦof PPv remains lower than that of Pv near the PPv transition, consistent with a first‐principles study on NaMgF3. Combined with similar observations in (Mg,Fe)SiO3, MgGeO3, and CaIrO3, our result on NaMgF3suggests that the VΦdecrease is a robust trend in the PPv transition and is controlled more by crystal structure than by chemical composition. Our finding also strengthens the proposal that the elastic properties of PPv can explain the small P‐wave increase across the D″ discontinuity and the anticorrelation between the VΦand S‐wave velocity anomalies in the D″ region.