Transition Zone Pressures Research Articles

First-principles study of water incorporation in Fe-containing wadsleyite

Using first-principles density functional theory (DFT), we study the problem of water incorporation in iron bearing wadsleyite ((Mg,Fe)2SiO4) at transition zone pressures and temperatures under varying conditions of vacancy and ferric ion concentration. Our calculations suggest that about 30% of the Fe3+ may be found at Si tetrahedral site, which is in accordance with the experimental report of Bolfan-Casanova et al. [2012]. Considering first-principles optimized crystal structures of hydrous Fe bearing wadsleyite, we calculate their elastic properties, such as bulk and shear moduli, compressional and shear wave velocity, as a function of pressure and temperature. Our calculations find that pressure suppresses the P and S wave ‘velocity-reduction’ that is caused by hydration. However, the suppression is not significant. So it is very unlikely that the pressure suppression will severely affect the accuracy of seismic waves in detecting the water content of the mantle transition zone as recently proposed [Buchen et al., 2018]. Comparison of our theoretical results with the existing seismic data indicates that the water content of the transition zone decreases with increasing depth, with the region close to the 410 km discontinuity being saturated in water, corroborating the previously proposed theories on mantle convection [Bercovici and Karato, 2003].

Thermoelasticity of Water in Silicate Melts: Implications for Melt Buoyancy in Earth's Mantle

AbstractA comprehensive analysis of experimental data and theoretical simulations on the partial molar volume of water in silicate melt indicates that finite strain theory successfully describes the compression of the H2O component dissolved in silicate melt at high pressures and temperatures. However, because of the high compressibility of the water component, a fourth order equation of state fit is required to accurately simulate experimental results on water's volume in silicate melts at a deep upper mantle, transition zone, and lower mantle pressures. Data from previous shock compression experiments on hydrous minerals in which melting occurs along the Hugoniot are used to provide an experimental constraint on the partial molar volume of water in silicate melt at deep mantle temperatures and pressures. The equation of state of the water component indicates that, depending on elastic averaging technique, the amount of water that could be present in neutrally or negatively buoyant mafic/ultramafic melts above the 410 km seismic discontinuity is upper‐bounded at 5.6 wt%: smaller than previously inferred, and consistent with melt being confined to a narrow depth range above the 410 km discontinuity. If melt is predominantly distributed along grain boundaries in low aspect ratio films, extents of melting as low as 2% could produce observed seismic velocity reductions. The ability of the lowermost mantle to contain negatively buoyant hydrous liquids hinges on the trade‐off between iron content and hydration: at these depths, substantially higher degrees of hydration could be present within partial melts.

The elastic properties and anisotropic behavior of MgSiO3 akimotoite at transition zone pressures

Seismic wave velocities in the Earth's transition zone, between 410 and 660 km depth, are poorly matched by mineralogical models. Mainly due to the presence of majoritic garnet, wave velocities calculated for a peridotite composition lithology are slower than seismic reference models, particularly towards the base of the transition zone. One possible resolution for this discrepancy is if the MgSiO3 polymorph akimotoite replaces majoritic garnet in some regions of the transition zone. This is possible in peridotitic material if temperatures are slightly lower than a typical geotherm or in harzburgitic composition material. The presence of akimotoite might serve as an explanation not only for the discrepancy of seismic velocities at the base of the transition zone but also for observations of transition zone seismic anisotropy in regions of current subduction. In order to provide the data required to test this possibility, two high-quality single-crystals of MgSiO3 akimotoite were studied using combined Brillouin spectroscopy and X-ray diffraction up to 24.86(3) GPa, i.e. to the limit of the akimotoite stability field. The resulting equation of state yields an adiabatic bulk modulus and first pressure derivative of KS0 = 209(2) GPa and K′ = 4.4(1), respectively, and a shear modulus and first pressure derivative of G0 = 131(1) GPa and G′ = 1.7(1). This is in overall agreement with several experimental and computational studies, however, the resulting aggregate velocities are slower than previously reported, especially at high pressures. The elastic anisotropy of akimotoite is found to decrease only slightly with pressure; akimotoite hence remains the most elastically anisotropic mineral in the transition zone and may therefore play a major role in explaining seismic anisotropy observations in the proximity of subducting slabs.

Effect of Water on Lattice Thermal Conductivity of Ringwoodite and Its Implications for the Thermal Evolution of Descending Slabs

AbstractThe presence of water in minerals generally alters their physical properties. Ringwoodite is the most abundant phase in the lowermost mantle transition zone and can host up to 1.5–2 wt% water. We studied high‐pressure lattice thermal conductivity of dry and hydrous ringwoodite by combining diamond‐anvil cell experiments with ultrafast optics. The incorporation of 1.73 wt% water substantially reduces the ringwoodite thermal conductivity by more than 40% at mantle transition zone pressures. We further parameterized the ringwoodite thermal conductivity as a function of pressure and water content to explore the large‐scale consequences of a reduced thermal conductivity on a slab's thermal evolution. Using a simple 1‐D heat diffusion model, we showed that the presence of hydrous ringwoodite in the slab significantly delays decomposition of dense hydrous magnesium silicates, enabling them to reach the lower mantle. Our results impact the potential route and balance of water cycle in the lower mantle.

Elastic properties of majoritic garnet inclusions in diamonds and the seismic signature of pyroxenites in the Earth's upper mantle

AbstractMajoritic garnet has been predicted to be a major component of peridotite and eclogite in Earth's deep upper mantle (>250 km) and transition zone. The investigation of mineral inclusions in diamond confirms this prediction, but there is reported evidence of other majorite-bearing lithologies, intermediate between peridotitic and eclogitic, present in the mantle transition zone. If these lithologies are derived from olivine-free pyroxenites, then at mantle transition zone pressures majorite may form monomineralic or almost monomineralic garnetite layers. Since majoritic garnet is presumably the seismically fastest major phase in the lowermost upper mantle, the existence of such majorite layers might produce a detectable seismic signature. However, a test of this hypothesis is hampered by the absence of sound wave velocity measurements of majoritic garnets with relevant chemical compositions, since previous measurements have been mostly limited to synthetic majorite samples with relatively simple compositions. In an attempt to evaluate the seismic signature of a pyroxenitic garnet layer, we measured the sound wave velocities of three natural majoritic garnet inclusions in diamond by Brillouin spectroscopy at ambient conditions. The chosen natural garnets derive from depths between 220 and 470 km and are plausible candidates to have formed at the interface between peridotite and carbonated eclogite. They contain elevated amounts (12–30%) of ferric iron, possibly produced during redox reactions that form diamond from carbonate. Based on our data, we model the velocity and seismic impedance contrasts between a possible pyroxenitic garnet layer and the surrounding peridotitic mantle. For a mineral assemblage that would be stable at a depth of 350 km, the median formation depth of our samples, we found velocities in pyroxenite at ambient conditions to be higher by 1.9(6)% for shear waves and 3.3(5)% for compressional waves compared to peridotite (numbers in parentheses refer to uncertainties in the last given digit), and by 1.3(13)% for shear waves and 2.4(10)% for compressional waves compared to eclogite. As a result of increased density in the pyroxenitic layer, expected seismic impedance contrasts across the interface between the monomineralic majorite layer and the adjacent rocks are about 5–6% at the majorite-eclogite-interface and 10–12% at the majoriteperidotite-boundary. Given a large enough thickness of the garnetite layer, velocity and impedance differences of this magnitude could become seismologically detectable.

High‐Pressure/Temperature Behavior of the Alkali/Calcium Carbonate Shortite (Na2Ca2(CO3)3): Implications for Carbon Sequestration in Earth's Transition Zone

AbstractThe behavior of shortite (Na2Ca2(CO3)3) has been probed using synchrotron‐based single crystal X‐ray diffraction and Raman spectroscopy at high pressures and following laser heating to illuminate carbon retention within the deep earth, and phase equilibria of alkali/calcium carbonate‐rich systems. Above 15 GPa, a transition to the shortite‐II structure occurs at 300 K. This phase is novel as it involves a large distortion of the carbonates, with an onset of 3 + 1 coordination and near‐dimerization of carbonate groups. Above 22 GPa, shortite‐II amorphizes. Samples laser heated at pressures between 12 and 30 GPa crystallize in a new structure, shortite‐III. Below 12 GPa, this phase appears to decompose into a mixture of shortite, nyerereite (Na2Ca(CO3)2), and aragonite (CaCO3) in accord with prior phase equilibria results. The high‐pressure behavior of nyerereite using Raman spectroscopy was also investigated to 25 GPa. The structural response of shortite to pressure is modulated by the sodium cations in the structure; hence, the behavior of alkali‐rich carbonates within kimberlitic systems at depth is likely dependent on the bonding and local geometry of alkali cations. Our results show that complex, dense high‐pressure structures are generated in the shortite system, and phase equilibria of the protoliths of carbonatites and kimberlites at deep upper mantle and transition zone pressures will involve intermediate alkali‐calcium carbonate phases, including the high‐pressure phases of shortite. Moreover, 3 + 1 coordination of carbon is observed at far lower pressures than other systems: this coordination could become important in complex carbonates and possibly liquids at substantially shallower depths than previously anticipated.

Hydration-reduced lattice thermal conductivity of olivine in Earth’s upper mantle

Earth's water cycle enables the incorporation of water (hydration) in mantle minerals that can influence the physical properties of the mantle. Lattice thermal conductivity of mantle minerals is critical for controlling the temperature profile and dynamics of the mantle and subducting slabs. However, the effect of hydration on lattice thermal conductivity remains poorly understood and has often been assumed to be negligible. Here we have precisely measured the lattice thermal conductivity of hydrous San Carlos olivine (Mg0.9Fe0.1)2SiO4 (Fo90) up to 15 gigapascals using an ultrafast optical pump-probe technique. The thermal conductivity of hydrous Fo90 with ∼7,000 wt ppm water is significantly suppressed at pressures above ∼5 gigapascals, and is approximately 2 times smaller than the nominally anhydrous Fo90 at mantle transition zone pressures, demonstrating the critical influence of hydration on the lattice thermal conductivity of olivine in this region. Modeling the thermal structure of a subducting slab with our results shows that the hydration-reduced thermal conductivity in hydrated oceanic crust further decreases the temperature at the cold, dry center of the subducting slab. Therefore, the olivine-wadsleyite transformation rate in the slab with hydrated oceanic crust is much slower than that with dry oceanic crust after the slab sinks into the transition zone, extending the metastable olivine to a greater depth. The hydration-reduced thermal conductivity could enable hydrous minerals to survive in deeper mantle and enhance water transportation to the transition zone.

Hydrogen mobility in transition zone silicates

We study the hydrogen mobility in ringwoodite and wadsleyite considering multiple charge-balanced defects, including Mg < = > 2H, Si < = > Mg + 2H, and the hydrogarnet defect, Si < = > 4H, using molecular dynamics simulations based on the density functional theory at transition zone pressures and temperatures between 1500 and 2500 K. We determine the diffusion coefficients and study in detail the mechanism of hydrogen mobility during lengthy simulations. Our results show that temperature, water concentration, and defect mechanism have a significant effect on mobility. We find that the fastest diffusion is for the Mg < = > 2H defect, while H is more mobile when incorporated as Si < = > Mg + 2H than as hydrogarnet defects. The computed diffusivities for ringwoodite are larger than for wadsleyite: at 2000 K, diffusivity is 1.13 × 10−09 m2/s for ringwoodite compared to 0.93 × 10−09 m2/s for wadsleyite. In general, the hydrogen atoms spend on the order of tens of picoseconds or more trapped in or around the vacancy sites with net migration between sites over timescales of tens of femtoseconds. At 2500 K, some of these hydrogen excursions take place over several angstroms, while at 2000 K, they do not always result in net diffusion. At 1500 K, most of the defects fail to make excursions from their defect sites resulting in diffusion.

Determination of calcium carbonate and sodium carbonate melting curves up to Earth's transition zone pressures with implications for the deep carbon cycle

Melting of carbonated eclogite or peridotite in the mantle influences the Earth's deep volatile cycles and bears on the long-term evolution of the atmosphere. Existing data on the melting curves of calcium carbonate (CaCO3) and sodium carbonate (Na2CO3) are limited to 7 GPa and therefore do not allow a full understanding of carbon storage and cycling in deep Earth. We determined the melting curves of CaCO3 and Na2CO3 to the pressures of Earth's transition zone using a multi-anvil apparatus. Melting was detected in situ by monitoring a steep and large increase in ionic conductivity, or inferred from sunken platinum markers in recovered samples. The melting point of CaCO3 rises from 1870 K at 3 GPa to ∼2000 K at 6 GPa and then stays within 50 K of 2000 K between 6 and 21 GPa. In contrast, the melting point of Na2CO3 increases continuously from ∼1123 K at 3 GPa to ∼1950 K at 17 GPa.A pre-melting peak in the alternating current through solid CaCO3 is attributed to the transition from aragonite to calcite V. Accordingly the calcite V-aragonite-liquid invariant point is placed at 13±1 GPa and 1970±40 K, with the Clapeyron slope of the calcite V to aragonite transition constrained at ∼70 K/GPa. The experiments on CaCO3 suggest a slight decrease in the melting temperature from 8 to 13 GPa, followed by a slight increase from 14 to 21 GPa. The negative melting slope is consistent with the prediction from our ab initio simulations that the liquid may be more compressible and become denser than calcite V at sufficiently high pressure. The positive melting slope at higher pressures is supported by the ab initio prediction that aragonite is denser than the liquid at pressures up to 30 GPa.At transition zone pressures the melting points of CaCO3 are comparable to that of Na2CO3 but nearly 400 K and 500 K lower than that of MgCO3. The fusible nature of compressed CaCO3 may be partially responsible for the majority of carbonatitic melts found on Earth's surface being highly calcic. It also provides a plausible explanation for low-degree melts of carbonated silicate rocks being particularly calcic at these depths. The melting curves of CaCO3 and Na2CO3 overlap with the estimated ocean-island geotherm at transition zone pressures, indicating that carbonatitic melt is readily generated from multi-component carbonate systems in the transition zone. The occurrence of such melt between the 410 and 660 km depths may facilitate the formation of ultradeep diamonds, produce low-velocity regions within the transition zone, and create a barrier to carbonate subduction into the lower mantle.

Global seismic data reveal little water in the mantle transition zone

Knowledge of the Earth's present water content is necessary to constrain the amount of water and other volatiles the Earth acquired during its formation and the amount that is cycled back into the interior from the surface. This study compares 410 and 660 km discontinuity depth with shear wave tomography within the mantle transition zone to identify regions with seismic signals consistent with water. The depth of the 410 and 660 km discontinuities is determined from a large updated dataset of SS-S410S and SS-S660S differential travel times, known as SS precursors. The discontinuity depths measured from binning and stacking the SS precursor data are then compared to the shear velocity model HMSL-S06 in the transition zone. Mapping all the possible combinations, very few locations match the predictions from mineral physics for the effects of water on discontinuity depth and shear velocity. The predictions, although not yet measured at actual transition zone temperatures and pressures, are a shallow 410 km discontinuity, a deep 660 km discontinuity, and a slow shear velocity. Only 8% of the bins with high-quality data are consistent with these predictions, and the calculated average water content within these bins is around 0.6 wt.%. A few isolated locations have patterns of velocity/topography that are consistent with water, while there are large regional-scale patterns consistent with cold/hot temperature anomalies. Combining this global analysis of long period seismic data and the current mineral physics predictions for water in transition zone minerals, I find that the mantle transition zone is generally dry, containing less than one Earth ocean of water. Although subduction zones could be locally hydrated, the combined discontinuity and velocity data show no evidence that wadsleyite or ringwoodite have been globally hydrated by subduction or initial Earth conditions.

Seismic signature of a hydrous mantle transition zone

Although water has a major influence on tectonic and other geodynamic processes, little is known about its quantity and distribution within the deep Earth. In the last few decades, laboratory experiments on nominally anhydrous minerals (NAMs) of the transition zone have shown that these minerals can contain significant amounts of water, up to 3.3wt%. In this study, we investigate if it is possible to use seismic observations to distinguish between a hydrous and anhydrous transition zone. We perform an extensive literature search of mineral experimental data, to generate a compilation of the water storage capacities, elastic parameters and phase boundary data for potentially hydrous minerals in the transition zone, and use thermodynamic modelling to compute synthetic seismic profiles of density, VP and VS at transition zone temperatures and pressures. We find that large uncertainties on the mineral phase equilibria (ca. 2GPa) and elastic properties produce a wide range of seismic profiles. In particular, there is a lack of data at temperatures corresponding to those along a 1300°C adiabat or hotter, which may be expected at transition zone pressures. Comparing our hydrous transition zone models with equivalent profiles at anhydrous conditions, we see that the depths of the 410 and 660 discontinuities cannot at present be used to map the water content of the transition zone due to these uncertainties. Further, while average velocities and densities inside the transition zone clearly decrease with increasing water content, there is a near-perfect trade-off with increases in temperature. It is therefore difficult to distinguish thermal from water effects, and the conventional view of a slow and thick transition zone for water and slow and thin transition zone for high temperature should be regarded with caution. A better diagnostic for water may be given by the average velocity gradients of the transition zone, which increase with increasing water content (but decrease for increasing temperature). However the significance of this effect depends on the degree of water saturation and partitioning between the NAMs. Since seismology is better able to constrain the thickness of the transition zone than velocity gradients, our study indicates that the most useful input from future mineral physics experiments would be to better constrain the phase relations between hydrous olivine and its high-pressure polymorphs, especially at high temperatures. Additionally, the uncertainties on the mineral seismic properties could be reduced significantly if the experimentally-observable correlations between bulk and shear moduli and their corresponding pressure derivatives would be published.

Elasticity of superhydrous phase B, seismic anomalies in cold slabs and implications for deep water transport

Seismic anomalies in the vicinity of cold slabs, including low velocity zones and enhanced seismic shear wave splitting have been commonly attributed to the presence of hydrated slab material. The dense hydrous magnesium silicate (DHMS) superhydrous phase B (ShyB, Mg10Si3H4O18) is considered an important water carrier to transition zone depth due to its large pressure stability and abundance (up to 20vol.%) in hydrous peridotites. To interpret the observed seismic anomalies in terms of hydration and to assess the role of ShyB in deep water recycling, we have investigated the sound velocities and single-crystal elasticity of ShyB to pressures of 17.5GPa at ambient temperature by Brillouin scattering spectroscopy in diamond anvil cells. The Voigt–Reuss–Hill averages for the adiabatic bulk moduli KS, shear moduli μ and their pressure derivatives yield KS=150(2)GPa, μ=99(1)GPa, (∂KS/∂P)=4.7(2) and (∂μ/∂P)=1.44(5). The aggregate compressional and shear wave velocities of ShyB at transition zone pressures are comparable to those of hydrous iron-bearing ringwoodite but are significantly lower than anhydrous phases in peridotites. The calculated velocity contrast between dry and ShyB-bearing hydrous peridotite (containing 1.2wt.% of water) indicate that 17vol.% of ShyB decreases both compressional VP and shear VS wave velocities by at most 2.5% at transition zone depths. Our results, combined with data for the deformation mechanisms of ShyB, indicate that ShyB aggregates develop strong textures under down-dip compression regime and yield a maximum shear wave splitting of 1.1% in the plane perpendicular to the compression axis at 17.5GPa. Although ShyB in hydrous subducted peridotite would reduce its seismic velocities, the splitting geometry VSV>VSH generated by ShyB fabrics in peridotite is incompatible with the pattern observed in Tonga and Sangihe, and would require the contribution of other phases to be explained. Textured phase D is a plausible candidate to explain the shear wave splitting in the lower transition zone, while at intermediate transition zone depths, the coexistence of textured phase D and ShyB in hydrous deformed peridotite would generate the observed array polarization. However, the later scenario will require locally very high water concentrations at depth.

Melting of siderite to 20 GPa and thermodynamic properties of FeCO3-melt

The siderite (FeCO3) melting curve is determined through multi-anvil experiments at 6–20GPa, and 1300–1870°C. The experiments define a melting curve with a Clapeyron slope steepening from 45 to 18°C/GPa but without backbend at upper mantle conditions, i.e., siderite is denser than FeCO3-melt (FeCO3L). The melting curve fits Tm=1037(44)+70.0(88)*P−1.43(37)*P2 (valid from 5 to 20GPa) where pressure is in GPa and temperature in °C. Siderite melting is not stoichiometric, minor quench magnetite was always observed and is interpreted as the result of partial redox dissociation of FeCO3L leading to dissolved Fe3+ and CO2 in the carbonate melt. At pressures below ~6.8GPa, siderite does not melt but decomposes through an auto redox dissociation reaction to magnetite, a carbon polymorph and CO2. From the experimental determination of the pure siderite melting curve, we calculate thermodynamic properties of the FeCO3L end-member, which reproduce the siderite melting curve in P–T space better than 10°C. The metastable 1atm melting temperature is calculated to 1012°C. Siderite has the lowest melting temperature of the Ca–Mg–Fe carbonates, its melting curve may cross the mantle geotherm at transition zone pressures. The stability of siderite is not only dependent on pressure and temperature but also strongly on oxygen fugacity (fO2). Model calculations in P–T-fO2 space in a Fe–C–O2 system show that the siderite or FeCO3-melt maximum stability is always reached at conditions of the CCO buffer. Our experimental and thermodynamic data constitute a cornerstone to model carbonate melting in the deep Earth, necessary to understanding the deep carbon cycle.

Effect of hydration on the single-crystal elasticity of Fe-bearing wadsleyite to 12 GPa

The single-crystal elastic properties of Fe-bearing wadsleyite with 1.93 wt% H 2 O (Mg 1.634 Fe 0.202 H 0.305 SiO 4 ) have been determined by Brillouin scattering. At ambient conditions, the aggregate bulk and shear moduli (K S0 , G 0 ) of this wadsleyite are 156.2(5) and 98.0(3) GPa, respectively. Compared to the corresponding anhydrous wadsleyite, 1.93 wt% H 2 O lowers K S0 and G 0 by 8.1% and 9.3%, respectively. High-pressure measurements up to 12 GPa show that the pressure derivative of the bulk modulus, K′ S0 = 4.8(1), is similar to that of the anhydrous Fe-wadsleyite with reported values of 4.6-4.74, but the addition of H 2 O increases the pressure derivative of the shear modulus, G 0 ′ from 1.5(1) to 1.9(1). This contrasts with the G 0 ′ of Fe-free wadsleyite, which is the same within uncertainty for the hydrous and anhydrous phases. As a result, both the compressional- and shear-wave velocities (v P , v S ) of hydrous Fe-bearing wadsleyite are about 200(±24) m/s slower than anhydrous Fe-bearing wadsleyite at transition zone pressures.

Thermodynamic properties of MgSiO3majorite and phase transitions near 660 km depth in MgSiO3and Mg2SiO4: A first principles study

[1] Thermodynamic properties of MgSiO3 tetragonal majorite have been calculated at high pressures and temperatures within the quasi-harmonic approximation based on density functional theory using the local density approximation (LDA) and the generalized gradient approximation (GGA). The LDA results compare exceptionally well with measured thermodynamic properties. A classical Monte Carlo simulation based on results from a cluster expansion method demonstrates that disorder between magnesium and silicon in the octahedral sites in MgSiO3 majorite does not occur below 3600 K at transition zone pressures. The ensuing calculations on phase boundaries of MgSiO3 between majorite, perovskite, and ilmenite show that a much better agreement with experiment can be obtained by using GGA rather than LDA, for LDA underestimates the transition pressures by as much as 11 GPa. The Clapeyron slopes predicted by GGA and LDA are close to each other: 0.9–1.7 MPa/K for majorite-perovskite transition, 6.9–7.9 MPa/K for majorite-ilmenite transition, and −7–−3 MPa/K for ilmenite-perovskite transition. The triple point predicted by GGA is located at 21.8 ± 1 GPa and 1840 ± 200 K which is ∼400 K lower in temperature than most experimental estimates. This result suggests that ilmenite is restricted to lower temperatures and that the majorite to ilmenite transition may occur in cold subducting slabs in the transition zone. Our calculation also reveals that wadsleyite decomposes to an assemblage of majorite plus periclase above 2280 K with a large negative Clapeyron slope (−22–−12 MPa/K) and that ringwoodite decomposes to ilmenite plus periclase below 1400 K (1.2 MPa/K). These two decomposition transitions may influence hot plumes and cold slabs near 660 km depth, respectively. Further calculations show that discontinuities in density, bulk modulus, and bulk sound velocity associated with the majorite to perovskite transition in MgSiO3 are much larger than those from the postspinel transition in Mg2SiO4 at conditions close to 660 km depth. This suggests that the large density discontinuity at 660 km depth as proposed by PREM (9.3%) might be accounted by a piclogite compositional model or marginally accounted by a pyrolite compositional model with, for example, 50 vol % ringwoodite, 45 vol % majorite, and 5 vol % other phases (such as calcium perovskite) at the bottom of the transition zone, provided that the density contrast between majorite and perovskite will not be greatly altered by the presence of other elements such as Fe, Al, Ca, and H. On the other hand, the smaller density discontinuity at 660 km depth as derived from impedance studies (4–6%) disfavors sharp contributions to seismic discontinuities from the majorite to perovskite transition.

Temperature dependence and mechanism of hydrogen incorporation in olivine at 12.5–14.0 GPa

The concentration and incorporation mechanisms of hydrogen into (Mg, Fe)2SiO4 olivine have been studied under water saturated conditions at 12.5–14.0 GPa and 1100–2000°C. The hydrogen content of olivine (Fo90–Fo95) coexisting with enstatite and hydrous melt increases from about 4600 ppmw at 1100°C to a maximum of 6250 ppmw at 1200°C. Above 1400°C, the hydrogen content of olivine decreases non‐linearly, reaching 160–240 ppmw in the 1800–2000°C range where there is a high melt fraction and no enstatite. Polarized infrared spectra of the recovered olivines exhibit OH absorption bands in both the 3450–3650 cm−1 range (group I) and 3160–3450 cm−1 range (group II). Results indicate that hydration of olivine is associated with both Mg and Si site vacancies. Microprobe analyses of the samples show a clear deficit of Si in olivine. The water storage capacity of olivine is probably no more than 1.5–2.0 times less than wadsleyite at 1400°C and transition zone pressures.

Constraints on the speciation of hydrogen in earth’s transition zone

The thermodynamic stability of hydrous B-type phases relative to hydrous β- and γ-phases is examined. The stability of phase B (Mg 12Si 4O 19(OH) 2) and superhydrous B (Mg 10Si 3O 14(OH) 4) relative to hydrous β-phase (with an end-member formula of Mg 7Si 4O 14(OH) 2) appears to be controlled by reactions that depend on whether coexisting MgSiO 3 resides in the garnet structure, or is present as a pyroxene. In majorite garnet-bearing systems, the volume of B-bearing assemblages is smaller than those containing hydrous β-phase; the corresponding volume change between B-assemblages relative to hydrous γ-phase-bearing assemblages is ambiguous. The increase in volume between superhydrous B-bearing assemblages and hydrous β-phase-bearing assemblages produces a free energy difference that is significantly greater than that expected from thermal contributions to the free energy. Phase B-bearing assemblages also have a smaller volume than hydrous β-phase assemblages, but the difference in volume is a factor of two smaller than that associated with superhydrous B: this may account for the greater stability range of superhydrous B compared to phase B. In contrast, pyroxene-bearing assemblages clearly favor the formation of hydrated β-phase. In accord with these expectations, heavily hydrated β- and γ-phases (those with greater than ∼1 wt.% H 2O) have been synthesized in the past primarily from melt-rich or pyroxene-bearing assemblages. To confirm these trends, we conducted synthesis experiments on samples at pressures between 15 and 25 GPa, utilizing the laser-heated diamond cell with chrysotile, San Carlos olivine, Bamble enstatite and a garnet lherzolite as starting compositions, with several weight percent water added to the anhydrous materials. At transition zone pressures, each of these materials yields a B-phase as the main crystalline hydrous phase. The combination of our calculations and experiments thus indicates that the B-phases are probably the main repository for hydrogen in the mantle between 400 and at least 520 km depth, particularly as their thermal stability is expected to be enhanced through the incorporation of levels of fluorine similar to those present in xenolithic phlogopites and amphiboles.

Single-crystal elasticity of ringwoodite to high pressures and high temperatures: implications for 520 km seismic discontinuity

The single-crystal elastic moduli of spinel-structured γ-(Mg 0.91Fe 0.09) 2SiO 4 (ringwoodite) were measured to 16 GPa at room temperature, and to 923 K at ambient pressure by Brillouin spectroscopy. A third-order finite-strain equation of state yields pressure derivatives of 4.1 (2) and 1.3 (1) for the adiabatic bulk modulus ( K S ) and shear modulus ( μ), respectively. The pressure derivatives of aggregate elastic moduli for all three polymorphs (α, β, and γ) of (Mg,Fe) 2SiO 4 in the Mg-rich end are equal within the mutual experimental uncertainties, with ( ∂K S / ∂P) T =4.1–4.3 and ( ∂μ/ ∂P) T =1.3–1.4. A linear decrease of the elastic moduli and sound velocities with temperature adequately describes the data. The temperature derivatives of K S and μ are −0.021 and −0.016 GPa/K, respectively. An extrapolation of our data to transition zone pressures and temperatures indicates that the shear and compressional impedance contrasts associated with β-(Mg,Fe) 2SiO 4→γ-(Mg,Fe) 2SiO 4 transition are sufficient to produce a visible discontinuity at 520 km depth even with only a moderate (30–50%) amount of olivine. Moreover, the partitioning of Fe between β-(Mg,Fe) 2SiO 4 and γ-(Mg,Fe) 2SiO 4 and majoritic garnet will produce a relatively sharp seismic discontinuity (in the order of 10–15 km) even when the latent heat of reaction is taken into account. The visibility of the discontinuity in regional and global studies could be inhibited by large topography that would accompany any local variations in temperature or Fe content. The high V P contrasts observed in some localities are difficult to explain by the β→γ transition and are suggestive of a chemical boundary. This is consistent with chemical heterogeneity in the transition zone, probably related to subduction. One of the possible explanations for high velocity contrasts near 520 km is a “buried Moho”—a boundary between garnetite (former oceanic crust) and γ-(Mg,Fe) 2SiO 4-enriched (former peridotite) layers.

Diamonds from the asthenosphere and the transition zone

Over the last two decades inclusions in diamonds have been recognised from a number of mines which represent chemically pristine samples of the sublithospheric mantle. X-ray diffraction studies show that isochemical phase transitions occurring in the plastic deformation field of diamond have completely eliminated crystal structures which are stable at transition zone and lower mantle pressures only. In the case of lower mantle inclusions ultra-deep parageneses may nevertheless be readily recognised by their unique chemical composition. For the chemically much less distinct asthenospheric and transition zone parageneses, majorite garnet is the only reliable indicator. From the majorite garnet record, the deeper upper mantle and transition zone appear to be dominated by basaltic material. Apart from elevated Si contents, the major element composition of majorite garnets is not significantly different from lithospheric eclogitic garnets, although there is a bias to higher Na and Ti and lower Fe contents. Coexisting clinopyroxene completely falls within the compositional ranges of the normal eclogitic suite. Trace element studies indicate that the majoritic source rocks have REEN patterns which range from MORB type to strongly LREE enriched. As petrological data are not consistent with a basaltic asthenosphere and transition zone, deep diamonds may represent either an exceptional sampling bias or, more likely, preferential diamond formation in eclogitic environments. This may be due to the pyrolitic parts of the transition zone being too reducing for diamond formation. It is speculated that sublithospheric eclogitic diamonds originated in subducted oceanic crust, which is relatively oxidised compared to pyrolite mantle due to sea water alteration. In such a scenario diamonds may originate either from redox reactions involving reduced mantle fluids and slab material or from reduction of carbonates which occurs invariably with increasing depth as pressure promotes the formation of ferrous iron rich minerals and therefore decreasing f O2. The latter process may explain the isotopically heavy composition (δ13C up to +0.9%) of some asthenospheric and transition zone diamonds. Isotopically light diamonds (δ13C down to −24.4 %) are probably related to chemical fractionation processes since subducted organic matter which is frequently invoked to cause such signatures should convert to diamond at much shallower depth.

Melting relations of hydrous pyrolite in CaO‐MgO‐Al2O3‐SiO2‐H2O System at the transition zone pressures

Phase relations and melt compositions in CaO‐MgO‐Al2O3‐SiO2‐pyrolite under hydrous (+2% of H2O) and anhydrous conditions have been determined at 13–20 GPa and 1600–2220°C. Liquidus and solidus temperatures for the hydrous system are about 50–100°C and 180–240°C lower than those for the dry system, respectively. Majorite is a liquidus phase of the hydrous pyrolite from 13 to 20 GPa. Olivine is a liquidus phase at 13 GPa and both periclase and majorite are the liquidus phases at 20 GPa in the dry pyrolite. We observed expansion of the stability field of anhydrous phase B in hydrous experiments. Compositions of partial melts at 13–20 GPa are generally similar in dry and hydrous systems, but hydrous melts contain more SiO2 at 13–17 GPa. The melts formed by low degree of melting have Al2O3‐depleted and CaO‐rich compositions. Trends of hydrous melt compositions are generally consistent with those of aluminum‐depleted komatiite magmas.