Skip to main content
NCBI home page
The Innovation logoLink to The Innovation
. 2025 Jan 6;6(1):100740. doi: 10.1016/j.xinn.2024.100740

Identifying dehydration-induced shear velocity anomaly in the Earth’s core-mantle boundary

Songsong Han 1,2,3, Xiaoming Cui 1, Baolong Zhang 1, Xiaohong Li 1,2, Sidao Ni 1, Heping Sun 1,2, Mingqiang Hou 1,2,∗, Qingyang Hu 3,∗∗
PMCID: PMC11764039  PMID: 39872487

Abstract

The steep temperature gradient near the bottom of the mantle is known to generate a negative correlation between the shear wave velocity (VS) and the depth in most regions of the D″ layer, as detected by seismological observations. However, increasing VS with depth is observed at the D″ layer beneath Central America, where the Farallon slab sinks, and the origin of this anomaly has not been well constrained. Here, we calculate the thermoelastic constants and obtain the elastic wave velocities of hydrous phase H with various Al contents and cation configurations, which may act as a water carrier to the D″ layer. We find its VS to be substantially lower than the post-perovskite-type bridgmanite. The dehydration of Al-enriched phase H and the redistribution of Al from the hydrous component to dry silicates would gradually raise the VS below the top of the D″ layer. The presence of 3.5 wt % water is sufficient to compensate for the thermal effects to match the seismic anomaly at the bottom of the mantle beneath Central America. The positive slope of VS versus depth in the D″ layer may fingerprint deep water recycling.

Keywords: hydrous mineral, core-mantle boundary, seismic velocity anomaly, subduction plate, deep water

Graphical abstract

graphic file with name fx1.jpg

Open in a new tab

Public summary

  • •

    The chemistry of hydrous minerals is investigated at the conditions of the lowermost mantle.

  • •

    Seismic waves propagate faster in regions where mineral dehydration occurs.

  • •

    The incorporation of 3.5 wt % water explains the enigmatical increase of seismic velocity at the bottom of the Farallon slab.

  • •

    Water-mineral chemistry shapes the structure of Earth’s lowermost mantle.

Introduction

The D″ layer is the lowermost portion of Earth’s mantle and extends 100–400 km above the core-mantle boundary (CMB). It is characterized by the structural transition from bridgmanite (perovskite [Pv]) to post-perovskite (pPv) and the enrichment of Fe and Al.1,2 As the temperature increases from ∼2,500 to 4,000 K across the D″ layer,3,4 this layer exhibits large heterogeneous structures. Recent seismic tomography has revealed both vertical and lateral heterogeneities within the D″ layer.5,6,7 They are identified by strong thermochemical evolution8 and differentiation9 beneath Africa, the southern Pacific, and the Circum-Pacific, causing variations of velocities.10,11 Typically, the seismic velocity abruptly increases by 2%–3% because of the Pv-pPv phase transition12,13 and then slowly decreases at the depth approaching the CMB because of temperature-induced shear softening12,14,15 or higher Fe/Al concentrations.16,17,18,19 This trend has been confirmed through seismic observations in regions beneath Eurasia, Arctic, and the central Pacific.14,20,21,22

An exception to this conventional seismic velocity profile is the D″ layer beneath Central America, where the shear wave velocity (VS) gradually increases with depth toward the CMB.5,23 The enrichment of Fe/Al in pPv and the presence of Fe-rich ferropericlase (Fp) or CaSiO3-Pv reduce the VS of mineral aggregates,24,25,26,27 making them unlikely explanations for the observed VS increase. A new mechanism is therefore needed. Compared with other regions showing normal VS profiles,14,20,21,22 the Farallon slab has penetrated the lower mantle and reached the D″ layer beneath Central America.28,29 This cooler slab, with temperatures from ∼950– to 2,100 K between the mantle transition zone and the top of the D″ layer,30 hosts a variety of hydrous phases like Al-rich phase D, Al-bearing phase H, and δ-AlOOH30,31,32 and transports substantial water to the D″ layer.33 The chemistry of water may provide novel insights into the abnormal seismic properties observed in the region of our interests.34,35,36

Along the pressure-temperature (P-T) conditions of relatively cold slabs, water is primarily stored in dense hydrous magnesium silicates (DHMSs).37 Most DHMSs are stable only at depths shallower than 660 km.37,38,39,40,41,42,43 The hydrous phase H (MgSiO4H2) can be stabilized to the middle depth of the lower mantle and is predicted to decompose into Pv and H2O ice VII at ∼1,500 km based on first-principles calculations.44 Although the pure Mg-Si endmember cannot maintain long-term water reservoirs, phase H can form a δ-AlOOH-MgSiO4H2 solid solution through cation substitution, e.g., Mg2+ + Si4+ = 2Al3+. Nishi et al.45 found that phase H can accommodate up to ∼70% δ-AlOOH. Because δ-AlOOH is stable at conditions of 33–134 GPa and 1,350–2,300 K,31 the corresponding stability field of the solid solution can expand toward CMB conditions. Specifically, Ohira et al.30 reported that the phase H solid solution contains at least 56 mol % of the AlOOH component at 68 GPa and 2,010 K and 77 mol % at 128 GPa and 2,190 K, suggesting that phase H becomes increasingly Al enriched with depth. In their experiments, the fraction of the hydrated AlOOH-MgSiO4H2 of total silicates reaches 30 mol % at the P-T conditions of the D″ layer.30 A recent study by Kim et al. proposed a hydrogen-enriched E-layer on top of the core sourced from deeply subducted water.36 This model introduces core-sourced Fe and may lead to even more complex chemistry in the FeOOH-AlOOH-MgSiO4H4 solid solution.46,47

The presence of the aforementioned hydrous phases may substantially modify local seismic structures. For instance, the hydration of pyrolytic mantle with superhydrous phase B could explain the low compressional wave velocity (VP) and VS anomalies near cold subduction zones at the bottom of the mantle transition zone.48 Meanwhile, Al-bearing phase D is likely associated with negative VS anomalies in the uppermost lower mantle.49 Experimental results have shown that the VS of δ-AlOOH is ∼7% lower than that of Mg-Pv under pressures of the lowermost mantle, potentially contributing to the negative VS jump near 2,800 km depth near the northern margin of the large low-shear-velocity province beneath the Pacific.50

For phase H, Al-free phase H has lower velocities than Pv,51 and its dehydration will produce a VS impedance contrast of ∼17% at a depth of 1,300–1,700 km, generating seismic discontinuities in some subduction zones of the northwest Pacific.52 By incorporating up to 25 atom % Al, Liu et al.53 predicted that the Al-bearing phase H becomes denser, and its velocity is lower than the preliminary reference Earth model,54 making it a candidate to explain the low-velocity layer at ∼1,500 km depth. However, previous works on Al-bearing phase H often neglected the effects of temperature and relatively high Al content (>25 atom %); thus, they lack the essential information under geothermal temperatures. In this work, we employ first-principles calculations to model the thermoelasticity of Al-bearing phase H with higher Al content (50%, 75%, and pure Al endmember) and derive their velocities in deep subduction slab conditions. We attempt to evaluate the hypothesis that the dehydration transition from hydrous phases to pPv leads to an increased VS profile with depth, providing further constraints on the seismic features in the D″ layer and the ultra-deep water cycling.

Results

The crystal structure of Al-bearing phase H

Al cation substitution (25%, 50%, or 75% mol % Al in this study) can increase the thermodynamic stability of DHMSs.30,33,38,45,55,56,57,58 Panero and Caracas54 performed a static calculation (0 K) on a total of 630 structures of 9 Mg1-xAl2xSi2-xO6H2 compositions (e.g., hydrous alumina-silicates) up to 60 GPa. Their results highlight the role of configurational entropy in stabilizing hydrous phase H. Therefore, we should consider the effect of Al cation substitution on the elasticity of phase H.

In this work, the simulation is conducted within density functional theory (DFT) using the Vienna Ab initio Simulation Package (VASP; details in the supplemental materials and methods). We model the structures with 0% (MgSiO4H2), 25% (named Al0.25-phase H), 50% (named Al0.5-phase H), 75% (named Al0.75-phase H), and 100% (δ-AlOOH) Al. The energetics of these structures are important to obtain reasonable computational results. In phase H, the arrangements of Mg and Si atoms randomly occupy the cation sites.59 We acknowledge that ideal cation disordering would require an infinitely large supercell and is beyond the scope of the current work. In this work, we build 2 × 2 × 2 supercells on the basis of the monoclinic structure (ground-state model I by Tsuchiya and Mookherjee51), which consists of 64 atoms (16 cations, 32 oxygen atoms, and 16 H atoms, as shown in the center of Figure 1). Then, we substitute Mg and Si atoms with Al while keeping the highest achievable crystal symmetry. Depending on the cation configurations, we consider five possible structures for Al0.25-phase H, Al0.5-phase H, and Al0.75-phase H, respectively. Taking Al0.5-phase H for example, Al atoms may alternately take positions along the (001) (Figure 1A), (101) (Figure 1B), (111) (Figure 1C), (011) (Figure 1D), or (100) (Figure 1E) planes of the supercell, producing five configurations with different cation arrangements. Similar models were also made for Al0.25-phase H (Figure S1) and Al0.75-phase H (Figure S2). We conducted structural optimization for each type and calculated their phonon dispersion curves (Figure S3) to evaluate their relative stability. We found that (001) of Al0.5-phase H and (101) of Al0.75-phase H are mechanically stable and relatively more energetically favorable. The thermoelastic constants of all mechanically stable structures are calculated since they may appear at high temperatures. Variations of different configurations are used to constrain the uncertainty of the elasticity.

Figure 1.

Figure 1

Open in a new tab

Crystal structures of hydrous phase H with 0% and 50% Al incorporations

Supercells of phase H (2 × 2 × 2 of unit cell, 64 atoms) are viewed along the c axis, respectively. The structures omit atom spheres for clarity purposes. Gold, blue, and cyan polyhedra refer to Mg-O, Si-O, and Al-O octahedra, respectively. The crystal structure in the center is MgSiO4H2, (A–E) Al0.5-phase H with Al atoms aligned with respect to (A) (001), (B) (101), (C) (111), (D) (011), and (E) (100) planes of the supercell. Gray planes are reference planes of Al atoms.

Thermoelasticity of Al-bearing hydrous phase H

The temperature dependence of the nine main elastic constants of phase H with different Al contents is plotted in Figure 2 from 0 to 2,000 K at various pressures (40, 70, 100, and 130 GPa). We select (001) of Al0.5-phase H and (101) of Al0.75-phase H for a comparison. In general, all elastic constants soften with increasing temperature and stiffen with increasing pressure. However, such temperature effects are suppressed at higher pressures. C12, C13, and C23 are less sensitive to temperature compared with other elastic constants. From 0 to 2,000 K at 130 GPa, C12, C13, and C23 soften by 2.0%, 1.3%, and 1.2% for Al0.75-phase H, respectively. However, the corresponding reductions of diagonal elastic constants (namely C11, C22, C33, C44, C55, and C66) are much greater (see Table S1). The effects of temperature on Cij for AlOOH, MgSiO4H2, and Al0.5-phase H are similar to those of Al0.75-phase H (Table S1).

Figure 2.

Figure 2

Open in a new tab

Evolution of elastic constants of hydrous phase H

The elastic constants of δ-AlOOH (dash-dotted lines), Al0.75-phase H (dotted lines), Al0.5-phase H (dashed lines), and MgSiO4H2 (solid lines) at temperatures from 0 to 2,000 K. (A–I) The evolution of 9 independent elastic constants that describe the elasticity of phase H. Light blue, dark cyan, orange, and pink curves represent elastic constants at 40, 70, 100, and 130 GPa, respectively.

We also find that C44, C55, and C66 are more sensitive to the Al content. The incorporation of 75% Al into phase H stiffens shear elastic constants (e.g., C44, C55, and C66) by 4.9%, 8.8%, and 4.0%, respectively, at 100 GPa and 2,000 K. Such a compositional effect is depressed for other elastic constants, C11, C22, C33, C12, and C23. In addition, the values of C33 of Al0.5-phase H are apparently softer than MgSiO4H2 beyond ∼50 GPa (Figure S4). This is due to its slightly distorted crystal structure. Our simulation only considers one configuration for Al0.5-phase H and may underestimate the complexity of structures (Figure S4). The elastic constants caused by the distortion are summarized in Figure S5. The correlation between these elastic constants and P-T is complex (Figure S5), as they do not exhibit a monotonous variation with either pressure or temperature. However, their contributions to velocities are minor compared to the nine major elastic constants.

Elastic wave velocities under high P-T conditions

The thermoelastic properties are calculated within the quasi-harmonic approximation (QHA). Firstly, we establish the QHA P-T boundary (Figure S7) to determine the validity regions of the simulation. Since the P-T conditions of the subducting slab (Figure S8) lie well below the QHA boundary, our calculated velocities are safe to describe the seismic properties of the Al-bearing hydrous silicate. The velocities of AlOOH are compared with previous experiments50 at room temperature and static calculations51,60 (at 0 K) (Figure S6). Our results are consistent with the literature within an ∼5% difference, suggesting that our calculations are reliable. Figure S9 shows the interpolated elastic wave velocities of these phases across the entire P-T range of the deep lower mantle. The color gradients in Figure S9 also suggest that the temperature dependence of the VS decreases substantially with pressure, whereas the pressure dependence increases with temperature. In contrast, this behavior is much less pronounced for the VP.

The temperature dependences of velocities are plotted in Figure 3 at three representative pressures of Earth’s lower mantle. It should be noted that the selected pressures in Figure 3A are relatively lower than those in Figures 3B–3D because the stability field of MgSiO4H2 is between ∼35 and 60 GPa.58 At 40 GPa, the VP and VS of MgSiO4H2 reduce by 6.3% and 8.9% from 0 to 2,000 K, while the corresponding reductions at 60 GPa are 5.1% and 7.3%, respectively. Therefore, the VS is more sensitive to temperature than the VP. Likewise, similar conclusions are drawn for AlOOH, Al0.75-phase H, and Al0.5-phase H (Figures 3B–3D; see Tables S2 and S3).

Figure 3.

Figure 3

Open in a new tab

The VP and VS of phase H with different Al contents

The elastic wave velocities are plotted over temperatures ranging from 0 to 2,000 K at various pressures for (A) Al-free phase H and (B) 90, (C) 110, and (D) 130 GPa. The velocities of Al-bearing structures represent the average velocities of different Al site substitutions. The uncertainties of the VP and VS for Al0.5-phase H are 1.8% and 2.2%, respectively, and those of Al0.75-phase H are 1.4% and 1.5%, respectively.

Previous experiments and calculations have confirmed that the thermal stability of hydrous phases is enhanced by increasing Al content.30,61 The effect of Al on the velocities of phase H is presented in Figure 4. As described above, the stability field of MgSiO4H2 is narrow; therefore, we select two P-T conditions, namely (40 GPa, 1,200 K) and (80 GPa, 1,800 K), sampled along the subducting slab geotherm.58 The incorporation of Al generally increases the elastic wave velocities of phase H. With the incorporation of 75% Al, the VS and VP propagate 6.2% and 3.7% faster, respectively, at 40 GPa and 1,200 K. Again, the VS is more sensitive to the incorporation of Al, likely due to the compositional effect on shear elastic constants C44, C55, and C66 (Figure 2). This compositional effect becomes even more pronounced at 80 GPa and 1,800 K (Figure 4). Therefore, we conclude that the Al content is the first-order factor in controlling the elasticity wave velocity, while the influence of cation configurations is the second order factor (Figure 4), albeit still important for accurately determining the overall velocities in a mineral assemblage.

Figure 4.

Figure 4

Open in a new tab

The Al content dependences of VP and VS for phase H

Error bars are uncertainties due to different cation substitutions.

Discussion

Implications for the lower mantle seismic structures

Our theoretical calculations provide constraints on the elastic wave velocities of hydrous phase H with various Al contents. These results allow us to explore seismic anomalies and detect deep water. Compared to the rest of the lower mantle, the D″ layer has the most contrasted seismic features62,63,64 due to the steep temperature gradient.3 In regions where subducting slabs have reached the CMB, the D″ layer shows a more complex correlation between seismic velocities and depth.5,65,66,67 The discontinuity near the top of the D″ layer is widely considered as the Pv-to-pPv polymorphic transition. Below the transition boundary, the VS decreases due to the spiking temperature15 or the incorporation of Fe/Al16,25 in the D″ layer, e.g., beneath the Indian subcontinent, Alaska, and the central Pacific.13,68,69,70

However, abnormally increased VS with depth has been detected by recent seismological observations in the D″ layer beneath Central America.5,23 As mentioned above, major components of the CMB are unlikely to account for such VS profiles because of their temperature-induced shear softening. The presence of partial melting at the CMB also does not seem to be a plausible explanation, as it typically results in sudden reductions of the VS by up to 30%–40%.71 Therefore, chemical heterogeneity, like deep water, should be considered.

Previous high-pressure experiments30,72 have refined the chemical stability fields of Al-enriched phase H and showed that it will dehydrate into Pv or pPv at lower-mantle conditions. By comparing the VS between Al-bearing phase H and pPv, we demonstrate that the VS of Al0.75-phase H, Al0.5-phase H, and Al0.25-phase H are slower than that of the pPv by 7.2(5)%, 7.3(2)%, and 11.5(7)%, respectively, along the slab geotherm at a depth range of 2,600–2,890 km (Figure S10). Consequently, the dehydration of Al-rich phase H results in a continuous increase of VS with depth in the D″ layer. During dehydration, the Al stored in Al-rich phase H preferentially enters surrounding silicates.73 The VS of pPv is negatively correlated with Al content (Figure S10).17,25 Therefore, the dehydration of Al-rich phase H, coupled with the redistribution of Al, can potentially compensate for the VS reduction due to the sharp temperature gradient.

Here, we explored different composition models by considering Al0.75-phase H and Al0.5-phase H. Our analysis suggests that at least 20% Al0.75-phase H must dehydrate to maintain a constant VS with depth. If the aggregate contains a higher amount of Al0.75-phase H, then the VS will increase with depth. To reproduce the seismic structure at the bottom of the lower mantle beneath Central America, we eventually optimized the model and established a mineral assemblage consisting of 26% Al0.75-phase H (equivalent to 3.5 wt % water), 55% pPv, and 19% Fp (Figure 5B). Upon dehydration, the excess Al in phase H diffuses into the surrounding pPv. Although Al incorporation decreases the VS of pPv, the effect is counteracted by the dehydration of Al0.75-phase H, as shown in Figure 5B. The final composition after dehydration will be 81% Al-pPv and 19% Fp (Figure S10). We also investigated the possibility of Al0.5-phase H, which has a slightly lower VS than Al0.75-phase H in the D″ layer. However, this comes at the expense of thermostability.30,61 Overall, our research demonstrates the feasibility of incorporating Al0.75-phase H into mineral assemblages to explain seismic anomalies. The increase in the VS with depth is a potential diagnostic feature for the existence of Al-enriched phase H. We also provide a dehydration profile of the VP as a reference (Figure S11).

Figure 5.

Figure 5

Open in a new tab

The schematic diagram and VS profiles in the D″ layer

(A) A cartoon of the subducted Farallon slab29 (blue) and low-velocity anomalies28 (red). The little blue bits represent water released by the Al-enriched phase H. The Al-enriched phase H is carried into the D″ layer through Farallon subducting slabs. It would dehydrate due to the sharp temperature increment. The water released in slabs interacts with the hot thermal boundary layer above the CMB.28

(B) The VS of aggregate (26% Al0.75-phase H, 55% pPv, and 19% Fp) (dark blue line) was plotted along cold geotherm2 via fitting data. The aggregate composed of (Al, Fe)0.125-pPv25 and Fp74 is employed for comparison, and it exhibits a decreasing trend with depth in the D″ layer (gray line with triangles). The purple and dark green lines represent the aggregates containing Pv+Al0.75-phase H+Fp and Al-Pv+Fp, respectively. The changing colors indicate the Pv-pPv phase transition. The dashed dark blue lines are the increasing VS due to the dehydration. Green stars and gray lines represent the VS profiles beneath different regions.23,68,69,70

Implications for deep water

Aspects of deep water in Earth’s lower mantle are under extensive debate. While most hydrous minerals became unstable at the thermodynamic conditions of the lower mantle, nominal anhydrous minerals may store several oceans worth of water.37,75 For example, Fu et al.76 reported that bridgmanite can accommodate a relatively high amount of water, up to ∼1,020 ppm at 24 GPa and 1,800°C. Contrarily, Liu et al.77 suggested that bridgmanite is nearly dry, with water solubility below 50 ppm at the top of the lower mantle. Likewise, contradictory results also exist in Fp, with water solubility ranging from 20 to 1,900 ppm.78,79,80 The water storage content may be further enhanced by the presence of dense hydrous phases (e.g., hydrous magnesium silicates, FeO2Hx, ẟ-AlOOH, and hydrous stishovite),31,45,81,82 which are stable in the cold subducted slabs. They transport large quantities of water into the deep mantle83,84 and are detected through seismological and geodynamical observations.85,86 It is also reported that the accumulation of δ-AlOOH may contribute to large seismic structures like large low-shear-velocity provinces.50 In summary, a small amount of water might survive at the bottom of the Farallon slab, considering its cold nature.

The subduction of the paleo Farallon plate was initiated 200 mya, and its relics remain at the lowermost mantle.28,87,88,89 Over the geological timescale, water is carried into the lower mantle through hydrous phases. We hypothesize that the D″ layer beneath Central America is enriched in hydrous silicates within the Farallon subducting slab. The global model by van Keken et al.90 suggests that ∼3 × 1011 kg/year of water is transported to Earth’s lower mantle. Recently, Kim et al.36 proposed a water-rich layer at the outmost of the core and estimated that 1.9–2.7 × 1019 kg H2O has subducted to the CMB. The present water budget in the lower mantle is estimated to be 1.0–4.5 ocean water (with one ocean water = ∼1.4 × 1021 kg).37,75 In our model, the anomalous area (roughly 2.9 × 105 km2) with a thickness of 250 km (Figure 5B) underneath Central America therefore requires 1.5 × 1019 kg water at maximum (assuming 3.5 wt % water from this study). Notably, this requisite is orders of magnitude lower than the water budget of the lower mantle.

Previous studies have shown that the incorporation of a moderate quantity of Fe into phase H did not weaken the thermostability, and up to 30 mol % of (Fe, Al)-bearing phase H can coexist with pPv.91 It was also reported that 1 mol % Fe in δ-AlOOH increased its density by 0.02 g/cm3 and reduced VS by 0.03 km/s92,93 If the impact of the incorporation of Fe on elasticity is comparable with that of δ-AlOOH, then 8 mol % Fe would reduce the required amount of Al0.75-phase H from 26% to 18% (equivalent to 2.4 wt % water) to reproduce the seismic velocity profile at the D″ layer beneath Central America as shown in Figure S12.

Finally, water released from Al-enriched phase H influences the fine seismic structures and chemical composition of the lowermost mantle.8,94,95 Dehydrated fluids may react with Fp to form the pyrite-type (Mg, Fe)O2Hx,95,96,97 being detected as low-velocity zones (Figure 5A) located near the bottom of Farallon slab.28 The experimental results of (Al, Fe)OOH98 and (Mg, Fe)O2Hx95 imply that FeHx melts are formed through the exchange of core-mantle materials. Such core-mantle chemistry of hydrous phases may eventually modulate the seismic, electrical, and thermal properties of the core.99,100,101,102

Materials and methods

See the supplemental information for details.

Acknowledgments

This work was supported by the National Natural Science Foundation of China with grant nos. T2425016, 42394114, 42192535, and 42150101.

Author contributions

Conceptualization, M.H. and Q.H.; investigation, S.H., X.C., B.Z., X.L., S.N., H.S., M.H., and Q.H.; visualization, S.H. and Q.H.; formal analysis, S.H., M.H., and Q.H.; validation, S.H., M.H., and Q.H.; funding acquisition, H.S., M.H., and Q.H.; project administration, M.H. and Q.H.; writing – original draft, S.H.; writing – review & editing, M.H. and Q.H.

Declaration of interests

The authors declare no competing interests.

Published Online: January 6, 2025

Footnotes

Contributor Information

Mingqiang Hou, Email: houmq@apm.ac.cn.

Qingyang Hu, Email: qingyang.hu@hpstar.ac.cn.

Lead contact website

M.H.: https://apm.cas.cn/sourcedb/zw/rck/zgj_rck/202301/t20230105_6595422.html.

Q.H.: http://www.hpstar.ac.cn//channels/549.html.

Supplemental information

Document S1. Figures S1–S14, Tables S1–S3, and supplemental materials and methods
mmc1.pdf (2.8MB, pdf)
Document S2. Article plus supplemental information
mmc2.pdf (7.2MB, pdf)

References

  • 1.Cobden L., Thomas C. The origin of D″ reflections: A systematic study of seismic array data sets. Geophys. J. Int. 2013;194(2):1091–1118. doi: 10.1093/gji/ggt152. [DOI] [Google Scholar]
  • 2.Catalli K., Shim S.H., Prakapenka V. Thickness and clapeyron slope of the post-perovskite boundary. Nature. 2009;462(7274):782–785. doi: 10.1038/nature08598. [DOI] [PubMed] [Google Scholar]
  • 3.Lay T., Hernlund J., Buffett B.A. Core–mantle boundary heat flow. Nat. Geosci. 2008;1(1):25–32. doi: 10.1038/ngeo.2007.44. [DOI] [Google Scholar]
  • 4.Frost D.A., Avery M.S., Buffett B.A., et al. Multidisciplinary constraints on the thermal-chemical boundary between Earth's core and mantle. G-cubed. 2022;23(3) doi: 10.1029/2021GC009764. [DOI] [Google Scholar]
  • 5.Li Y., Miller M.S. Seismic evidence for thermal and chemical heterogeneities in D″ region beneath Central America from grid search modeling. Geophys. Res. Lett. 2021;48(14) doi: 10.1029/2021gl092493. [DOI] [Google Scholar]
  • 6.Thomas C., Garnero E.J., Lay T. High-resolution imaging of lowermost mantle structure under the Cocos plate. J. Geophys. Res. 2004;109(B8):B08307. doi: 10.1029/2004jb003013. [DOI] [Google Scholar]
  • 7.Ma X., Thomas C. Small-scale scattering heterogeneities in the lowermost mantle from a global analysis of PKP precursors. JGR. Solid Earth. 2020;125(3) doi: 10.1029/2019JB018736. [DOI] [Google Scholar]
  • 8.Du Z., Deng J., Miyazaki Y., et al. Fate of hydrous Fe-rich silicate melt in Earth's deep mantle. Geophys. Res. Lett. 2019;46(16):9466–9473. [Google Scholar]
  • 9.Arveson S.M., Deng J., Karki B.B., Lee K.K.M. Evidence for Fe-Si-O liquid immiscibility at deep Earth pressures. Proc. Natl. Acad. Sci. USA. 2019;116(21):10238–10243. doi: 10.1073/pnas.1821712116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Ritsema J., Deuss A., van Heijst H.J., Woodhouse J.H. S40RTS: A degree-40 shear-velocity model for the mantle from new Rayleigh wave dispersion, teleseismic traveltime and normal-mode splitting function measurements. Geophys. J. Int. 2011;184(3):1223–1236. doi: 10.1111/j.1365-246X.2010.04884.x. [DOI] [Google Scholar]
  • 11.Koelemeijer P., Ritsema J., Deuss A., van Heijst H.J. SP12RTS: A degree-12 model of shear- and compressional-wave velocity for Earth's mantle. Geophys. J. Int. 2016;204(2):1024–1039. doi: 10.1093/gji/ggv481. [DOI] [Google Scholar]
  • 12.Hernlund J.W., Thomas C., Tackley P.J. A doubling of the post-perovskite phase boundary and structure of the Earth's lowermost mantle. Nature. 2005;434(7035):882–886. doi: 10.1038/nature03472. [DOI] [PubMed] [Google Scholar]
  • 13.Konishi K., Kawai K., Geller R.J., Fuji N. Waveform inversion of broad-band body wave data for the S-velocity structure in the lowermost mantle beneath the Indian subcontinent and Tibetan Plateau. Geophys. J. Int. 2012;191(1):305–316. doi: 10.1111/j.1365-246X.2012.05614.x. [DOI] [Google Scholar]
  • 14.Kawai K., Geller R.J., Fuji N. Waveform inversion for S-wave structure in the lowermost mantle beneath the Arctic: Implications for mineralogy and chemical composition. Geophys. Res. Lett. 2010;37(16) doi: 10.1029/2010gl043654. [DOI] [Google Scholar]
  • 15.Kawai K., Tsuchiya T. Temperature profile in the lowermost mantle from seismological and mineral physics joint modeling. Proc. Natl. Acad. Sci. USA. 2009;106(52):22119–22123. doi: 10.1073/pnas.0905920106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Li L., Brodholt J., Stackhouse S., et al. Elasticity of (Mg, Fe)(Si, Al)O3-perovskite at high pressure. Earth Planet Sci. Lett. 2005;240(2):529–536. doi: 10.1016/j.epsl.2005.09.030. [DOI] [Google Scholar]
  • 17.Tsuchiya T., Tsuchiya J. Effect of impurity on the elasticity of perovskite and postperovskite: Velocity contrast across the postperovskite transition in (Mg,Fe,Al)(Si,Al)O3. Geophys. Res. Lett. 2006;33(12):L12S04. doi: 10.1029/2006gl025706. [DOI] [Google Scholar]
  • 18.Shukla G., Wu Z., Hsu H., et al. Thermoelasticity of Fe2+-bearing bridgmanite. Geophys. Res. Lett. 2015;42(6):1741–1749. doi: 10.1002/2014gl062888. [DOI] [Google Scholar]
  • 19.Shukla G., Cococcioni M., Wentzcovitch R.M. Thermoelasticity of Fe3+- and Al-bearing bridgmanite: Effects of iron spin crossover. Geophys. Res. Lett. 2016;43(11):5661–5670. doi: 10.1002/2016gl069332. [DOI] [Google Scholar]
  • 20.Shen X., Zhou Y., Idehara K., et al. An examination of the presence and topography of the D″ discontinuity under the Russia–Kazakhstan border region using seismic waveform data from a deep earthquake in Spain. Earthq. Sci. 2014;27(2):209–215. doi: 10.1007/s11589-014-0071-2. [DOI] [Google Scholar]
  • 21.Avants M., Lay T., Russell S.A., Garnero E.J. Shear velocity variation within the D″ region beneath the central Pacific. J. Geophys. Res. 2006;111(B5) doi: 10.1029/2004jb003270. [DOI] [Google Scholar]
  • 22.He Y., Wen L., Zheng T. Seismic evidence for an 850 km thick low-velocity structure in the Earth's lowermost mantle beneath Kamchatka. Geophys. Res. Lett. 2014;41(20):7073–7079. doi: 10.1002/2014gl061249. [DOI] [Google Scholar]
  • 23.Whittaker S., Thorne M.S., Schmerr N.C., Miyagi L. Seismic array constraints on the D″ discontinuity beneath Central America. JGR. Solid Earth. 2016;121(1):152–169. doi: 10.1002/2015jb012392. [DOI] [Google Scholar]
  • 24.Thomson A.R., Crichton W.A., Brodholt J.P., et al. Seismic velocities of CaSiO3 perovskite can explain LLSVPs in Earth's lower mantle. Nature. 2019;572(7771):643–647. doi: 10.1038/s41586-019-1483-x. [DOI] [PubMed] [Google Scholar]
  • 25.Shukla G., Sarkar K., Wentzcovitch R.M. Thermoelasticity of iron- and aluminum-bearing MgSiO3 postperovskite. JGR. Solid Earth. 2019;124(3):2417–2427. doi: 10.1029/2018jb016379. [DOI] [Google Scholar]
  • 26.Wu Z. Velocity structure and composition of the lower mantle with spin crossover in ferropericlase. JGR. Solid Earth. 2016;121(4):2304–2314. doi: 10.1002/2015jb012667. [DOI] [Google Scholar]
  • 27.Chen B., Jackson J.M., Sturhahn W., et al. Spin crossover equation of state and sound velocities of (Mg0.65Fe0.35)O ferropericlase to 140 GPa. J. Geophys. Res. 2012;117(B8) doi: 10.1029/2012jb009162. [DOI] [Google Scholar]
  • 28.Borgeaud A.F.E., Kawai K., Konishi K., Geller R.J. Imaging paleoslabs in the D″ layer beneath Central America and the Caribbean using seismic waveform inversion. Sci. Adv. 2017;3(11) doi: 10.1126/sciadv.1602700. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Fukao Y., Obayashi M. Subducted slabs stagnant above, penetrating through, and trapped below the 660 km discontinuity. JGR. Solid Earth. 2013;118(11):5920–5938. doi: 10.1002/2013jb010466. [DOI] [Google Scholar]
  • 30.Ohira I., Ohtani E., Sakai T., et al. Stability of a hydrous δ-phase, AlOOH–MgSiO2(OH)2, and a mechanism for water transport into the base of lower mantle. Earth Planet Sci. Lett. 2014;401:12–17. doi: 10.1016/j.epsl.2014.05.059. [DOI] [Google Scholar]
  • 31.Sano A., Ohtani E., Kondo T., et al. Aluminous hydrous mineral δ-AlOOH as a carrier of hydrogen into the core-mantle boundary. Geophys. Res. Lett. 2008;35(3) doi: 10.1029/2007gl031718. [DOI] [Google Scholar]
  • 32.Xu C., Inoue T. Melting of Al-rich phase D up to the uppermost lower mantle and transportation of H2O to the deep Earth. G-cubed. 2019;20(9):4382–4389. doi: 10.1029/2019gc008476. [DOI] [Google Scholar]
  • 33.Ohtani E., Yuan L., Ohira I., et al. Fate of water transported into the deep mantle by slab subduction. J. Asian Earth Sci. 2018;167:2–10. [Google Scholar]
  • 34.Pan S., Huang T., Vazan A., et al. Magnesium oxide-water compounds at megabar pressure and implications on planetary interiors. Nat. Commun. 2023;14(1):1165. doi: 10.1038/s41467-023-36802-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Gao H., Liu C., Shi J., et al. Superionic silica-water and silica-hydrogen compounds in the deep interiors of Uranus and Neptune. Phys. Rev. Lett. 2022;128(3) doi: 10.1103/PhysRevLett.128.035702. [DOI] [PubMed] [Google Scholar]
  • 36.Kim T., O’Rourke J.G., Lee J., et al. A hydrogen-enriched layer in the topmost outer core sourced from deeply subducted water. Nat. Geosci. 2023;16:1208–1214. doi: 10.1038/s41561-023-01324-x. [DOI] [Google Scholar]
  • 37.Ohtani E. Hydration and dehydration in Earth's interior. Annu. Rev. Earth Planet Sci. 2021;49(1):253–278. doi: 10.1146/annurev-earth-080320-062509. [DOI] [Google Scholar]
  • 38.Pamato M.G., Myhill R., Boffa Ballaran T., et al. Lower-mantle water reservoir implied by the extreme stability of a hydrous aluminosilicate. Nat. Geosci. 2014;8(1):75–79. doi: 10.1038/ngeo2306. [DOI] [Google Scholar]
  • 39.Kakizawa S., Inoue T., Nakano H., et al. Stability of Al-bearing superhydrous phase B at the mantle transition zone and the uppermost lower mantle. Am. Mineral. 2018;103(8):1221–1227. doi: 10.2138/am-2018-6499. [DOI] [Google Scholar]
  • 40.Ohtani E., Shibata T., Kubo T., Kato T. Stability of hydrous phases in the transition zone and the upper most part of the lower mantle. Geophys. Res. Lett. 1995;22(19):2553–2556. doi: 10.1029/95GL02338. [DOI] [Google Scholar]
  • 41.Frost D.J. The stability of hydrous mantle phases. Rev. Mineral. Geochem. 2006;62(1):243–271. doi: 10.2138/rmg.2006.62.11. [DOI] [Google Scholar]
  • 42.Liu X., Matsukage K.N., Nishihara Y., et al. Stability of the hydrous phases of Al-rich phase D and Al-rich phase H in deep subducted oceanic crust. Am. Mineral. 2019;104(1):64–72. doi: 10.2138/am-2019-6559. [DOI] [Google Scholar]
  • 43.Wang W., Tschauner O., Huang S., et al. Coupled deep-mantle carbon-water cycle: Evidence from lower-mantle diamonds. Innovation. 2021;2(2) doi: 10.1016/j.xinn.2021.100117. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Tsuchiya J., Umemoto K. First-principles determination of the dissociation phase boundary of phase H MgSiO4H2. Geophys. Res. Lett. 2019;46(13):7333–7336. doi: 10.1029/2019gl083472. [DOI] [Google Scholar]
  • 45.Nishi M., Irifune T., Tsuchiya J., et al. Stability of hydrous silicate at high pressures and water transport to the deep lower mantle. Nat. Geosci. 2014;7(3):224–227. doi: 10.1038/ngeo2074. [DOI] [Google Scholar]
  • 46.Thompson E.C., Campbell A.J., Tsuchiya J. Elasticity of ε-FeOOH: Seismic implications for Earth's lower mantle. JGR. Solid Earth. 2017;122(7):5038–5047. doi: 10.1002/2017jb014168. [DOI] [Google Scholar]
  • 47.Thompson E.C., Campbell A.J., Tsuchiya J. Elastic properties of the pyrite-type FeOOH-AlOOH system from first-principles calculations. G-cubed. 2021;22(5) doi: 10.1029/2021gc009703. [DOI] [Google Scholar]
  • 48.Xu C., Gréaux S., Inoue T., et al. Sound velocities of superhydrous phase B up to 21 GPa and 900 K. Geophys. Res. Lett. 2022;49 doi: 10.1029/2022gl098674. [DOI] [Google Scholar]
  • 49.Xu C., Gréaux S., Inoue T., et al. Sound velocities of Al-bearing phase D up to 22 GPa and 1300 K. Geophys. Res. Lett. 2020;47(18) doi: 10.1029/2020gl088877. [DOI] [Google Scholar]
  • 50.Mashino I., Murakami M., Ohtani E. Sound velocities of δ-AlOOH up to core-mantle boundary pressures with implications for the seismic anomalies in the deep mantle. JGR. Solid Earth. 2016;121(2):595–609. doi: 10.1002/2015jb012477. [DOI] [Google Scholar]
  • 51.Tsuchiya J., Mookherjee M. Crystal structure, equation of state, and elasticity of phase H (MgSiO4H2) at Earth's lower mantle pressures. Sci. Rep. 2015;5 doi: 10.1038/srep15534. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Song Z., Wu Z., Wang W., et al. Elasticity of phase H under the mantle temperatures and pressures: Implications for discontinuities and water transport in the mid-mantle. JGR. Solid Earth. 2022;127(11) doi: 10.1029/2022jb024893. e2022JB024893. [DOI] [Google Scholar]
  • 53.Liu G., Liu L., Yang L., et al. Crystal structure and elasticity of Al-bearing phase H under high pressure. AIP Adv. 2018;8(5) doi: 10.1063/1.5030375. [DOI] [Google Scholar]
  • 54.Dziewonski A.M., Anderson D.L. Preliminary reference Earth model. Phys. Earth Planet. In. 1981;25(4):297–356. doi: 10.1016/0031-9201(81)90046-7. [DOI] [Google Scholar]
  • 55.Panero W.R., Caracas R. Stability and solid solutions of hydrous alumino-silicates in the Earth’s mantle. Minerals. 2020;10(4):330. doi: 10.3390/min10040330. [DOI] [Google Scholar]
  • 56.Xu C., Inoue T., Kakizawa S., et al. Effect of Al on the stability of dense hydrous magnesium silicate phases to the uppermost lower mantle: Implications for water transportation into the deep mantle. Phys. Chem. Miner. 2021;48(9) doi: 10.1007/s00269-021-01156-4. [DOI] [Google Scholar]
  • 57.Nakatsuka A., Yoshiasa A., Ohkawa M., Ito E. Aluminous hydrous magnesium silicate as a lower-mantle hydrogen reservoir: a role as an agent for material transport. Sci. Rep. 2022;12(1):3594. doi: 10.1038/s41598-022-07007-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Ohtani E., Amaike Y., Kamada S., et al. Stability of hydrous phase H MgSiO4H2 under lower mantle conditions. Geophys. Res. Lett. 2014;41(23):8283–8287. doi: 10.1002/2014gl061690. [DOI] [Google Scholar]
  • 59.Bindi L., Nishi M., Tsuchiya J., Irifune T. Crystal chemistry of dense hydrous magnesium silicates: The structure of phase H, MgSiH2O4, synthesized at 45 GPa and 1000 °C. Am. Mineral. 2014;99(8-9):1802–1805. doi: 10.2138/am.2014.4994. [DOI] [Google Scholar]
  • 60.Tsuchiya J., Tsuchiya T. Elastic properties of δ-AlOOH under pressure: First principles investigation. Phys. Earth Planet. In. 2009;174(1-4):122–127. doi: 10.1016/j.pepi.2009.01.008. [DOI] [Google Scholar]
  • 61.Panero W.R., Caracas R. Stability of phase H in the MgSiO4H2–AlOOH–SiO2 system. Earth Planet Sci. Lett. 2017;463:171–177. doi: 10.1016/j.epsl.2017.01.033. [DOI] [Google Scholar]
  • 62.Garnero E.J., Jeanloz R. Earth's Enigmatic Interface. Science. 2000;289(5476):70–71. doi: 10.1126/science.289.5476.70. [DOI] [PubMed] [Google Scholar]
  • 63.Garnero E.J., Jeanloz R. Fuzzy patches on the Earth's core-mantle boundary? Geophys. Res. Lett. 2000;27(17):2777–2780. doi: 10.1029/2000GL008498. [DOI] [Google Scholar]
  • 64.Buffett B.A., Garnero E.J., Jeanloz R. Sediments at the Top of Earth's Core. Science. 2000;290(5495):1338–1342. doi: 10.1126/science.290.5495.1338. [DOI] [PubMed] [Google Scholar]
  • 65.Li Y., Miller M.S., Sun D. Seismic imaging the D″ region beneath the Central Atlantic. Phys. Earth Planet. In. 2019;292:76–86. doi: 10.1016/j.pepi.2019.05.005. [DOI] [Google Scholar]
  • 66.Sun D., Helmberger D., Miller M.S., Jackson J.M. Major disruption of D″ beneath Alaska. JGR. Solid Earth. 2016;121(5):3534–3556. doi: 10.1002/2015jb012534. [DOI] [Google Scholar]
  • 67.Lay T., Helmberger D.V. A lower mantle S-wave triplication and the shear velocity structure of D. Geophys. J. Int. 1983;75(3):799–837. doi: 10.1111/j.1365-246X.1983.tb05010.x. [DOI] [Google Scholar]
  • 68.Christopher J.Y., Thorne L. Evidence for a shear velocity discontinuity in the lower mantle beneath India and the Indian Ocean. Phys. Earth Planet. In. 1987;49(1):37–53. doi: 10.1016/0031-9201(87)90131-2. [DOI] [Google Scholar]
  • 69.Young C.J., Lay T. Multiple phase analysis of the shear velocity structure in the D″ region beneath Alaska. J. Geophys. Res. 1990;95(B11):17385–17402. doi: 10.1029/JB095iB11p17385. [DOI] [Google Scholar]
  • 70.Russell S.A., Reasoner C., Lay T., Revenaugh J. Coexisting shear- and compressional-wave seismic velocity discontinuities beneath the central Pacific. Geophys. Res. Lett. 2001;28(11):2281–2284. doi: 10.1029/2000GL012553. [DOI] [Google Scholar]
  • 71.Li Z., Leng K., Jenkins J., Cottaar S. Kilometer-scale structure on the core-mantle boundary near Hawaii. Nat. Commun. 2022;13(1):2787. doi: 10.1038/s41467-022-30502-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 72.Walter M.J., Thomson A.R., Wang W., et al. The stability of hydrous silicates in Earth's lower mantle: Experimental constraints from the systems MgO–SiO2–H2O and MgO–Al2O3–SiO2–H2O. Chem. Geol. 2015;418:16–29. doi: 10.1016/j.chemgeo.2015.05.001. [DOI] [Google Scholar]
  • 73.Ishii T., Ohtani E., Shatskiy A. Aluminum and hydrogen partitioning between bridgmanite and high-pressure hydrous phases: Implications for water storage in the lower mantle. Earth Planet Sci. Lett. 2022;583 doi: 10.1016/j.epsl.2022.117441. [DOI] [Google Scholar]
  • 74.Wu Z., Justo J.F., Wentzcovitch R.M. Elastic anomalies in a spin-crossover system: Ferropericlase at lower mantle conditions. Phys. Rev. Lett. 2013;110(22) doi: 10.1103/PhysRevLett.110.228501. [DOI] [PubMed] [Google Scholar]
  • 75.Peslier A.H., Schönbächler M., Busemann H., Karato S.-I. Water in the Earth’s Interior: Distribution and Origin. Space Sci. Rev. 2017;212(1-2):743–810. doi: 10.1007/s11214-017-0387-z. [DOI] [Google Scholar]
  • 76.Fu S., Yang J., Karato S.i., et al. Water Concentration in Single-Crystal (Al,Fe)-Bearing Bridgmanite Grown From the Hydrous Melt: Implications for Dehydration Melting at the Topmost Lower Mantle. Geophys. Res. Lett. 2019;46(17-18):10346–10357. doi: 10.1029/2019gl084630. [DOI] [Google Scholar]
  • 77.Liu Z., Fei H., Chen L., et al. Bridgmanite is nearly dry at the top of the lower mantle. Earth Planet Sci. Lett. 2021;570 doi: 10.1016/j.epsl.2021.117088. [DOI] [Google Scholar]
  • 78.Bolfan-Casanova N., Keppler H., Rubie D.C. Water partitioning between nominally anhydrous minerals in the MgO–SiO2–H2O system up to 24 GPa: implications for the distribution of water in the Earth’s mantle. Earth Planet Sci. Lett. 2000;182(3):209–221. doi: 10.1016/S0012-821X(00)00244-2. [DOI] [Google Scholar]
  • 79.Murakami M., Hirose K., Yurimoto H., et al. Water in Earth's lower mantle. Science. 2002;295(5561):1885–1887. doi: 10.1126/science.1065998. [DOI] [PubMed] [Google Scholar]
  • 80.Litasov K., Ohtani E., Langenhorst F., et al. Water solubility in Mg-perovskites and water storage capacity in the lower mantle. Earth Planet Sci. Lett. 2003;211(1-2):189–203. doi: 10.1016/s0012-821x(03)00200-0. [DOI] [Google Scholar]
  • 81.Nishi M., Kuwayama Y., Tsuchiya J., Tsuchiya T. The pyrite-type high-pressure form of FeOOH. Nature. 2017;547(7662):205–208. doi: 10.1038/nature22823. [DOI] [PubMed] [Google Scholar]
  • 82.Lin Y., Hu Q., Walter M.J., et al. Hydrous SiO2 in subducted oceanic crust and H2O transport to the core-mantle boundary. Earth Planet Sci. Lett. 2022;594 doi: 10.1016/j.epsl.2022.117708. [DOI] [Google Scholar]
  • 83.Martin H. Effect of steeper Archean geothermal gradient on geochemistry of subduction-zone magmas. Geol. 1986;14(9):753–756. doi: 10.1130/0091-7613(1986)14<753:EOSAGG>2.0.CO;2. [DOI] [Google Scholar]
  • 84.van Hunen J., Moyen J.-F. Archean Subduction: Fact or Fiction? Annu. Rev. Earth Planet Sci. 2012;40(1):195–219. doi: 10.1146/annurev-earth-042711-105255. [DOI] [Google Scholar]
  • 85.Niu Y. Origin of the LLSVPs at the base of the mantle is a consequence of plate tectonics – A petrological and geochemical perspective. Geosci. Front. 2018;9(5):1265–1278. doi: 10.1016/j.gsf.2018.03.005. [DOI] [Google Scholar]
  • 86.Sun D., Helmberger D., Lai V.H., et al. Slab Control on the Northeastern Edge of the Mid-Pacific LLSVP Near Hawaii. Geophys. Res. Lett. 2019;46(6):3142–3152. doi: 10.1029/2018gl081130. [DOI] [Google Scholar]
  • 87.van der Meer D.G., Spakman W., van Hinsbergen D.J.J., et al. Towards absolute plate motions constrained by lower-mantle slab remnants. Nat. Geosci. 2009;3(1):36–40. doi: 10.1038/ngeo708. [DOI] [Google Scholar]
  • 88.Goes S., Agrusta R., van Hunen J., Garel F. Subduction-transition zone interaction: A review. Geosphere. 2017;13(3):644–664. doi: 10.1130/ges01476.1. [DOI] [Google Scholar]
  • 89.Boschman L.M., van Hinsbergen D.J.J., Kimbrough D.L., et al. The dynamic history of 220 million years of subduction below Mexico: A correlation between slab geometry and overriding plate deformation based on geology, paleomagnetism, and seismic tomography. G-cubed. 2018;19(12):4649–4672. doi: 10.1029/2018gc007739. [DOI] [Google Scholar]
  • 90.van Keken P.E., Hacker B.R., Syracuse E.M., Abers G.A. Subduction factory: 4. Depth-dependent flux of H2O from subducting slabs worldwide. J. Geophys. Res. 2011;116(B1) doi: 10.1029/2010jb007922. [DOI] [Google Scholar]
  • 91.Yuan H., Zhang L., Ohtani E., et al. Stability of Fe-bearing hydrous phases and element partitioning in the system MgO–Al2O3–Fe2O3–SiO2–H2O in Earth's lowermost mantle. Earth Planet Sci. Lett. 2019;524 doi: 10.1016/j.epsl.2019.115714. [DOI] [Google Scholar]
  • 92.Su X., Zhao C., Lv C., et al. The effect of iron on the sound velocities of δ-AlOOH up to 135 GPa. Geosci. Front. 2021;12(2):937–946. doi: 10.1016/j.gsf.2020.08.012. [DOI] [Google Scholar]
  • 93.Ohira I., Jackson J.M., Solomatova N.V., et al. Compressional behavior and spin state of δ-(Al,Fe)OOH at high pressures. Am. Mineral. 2019;104(9):1273–1284. doi: 10.2138/am-2019-6913. [DOI] [Google Scholar]
  • 94.Mao H.-K., Hu Q., Yang L., et al. When water meets iron at Earth's core–mantle boundary. Natl. Sci. Rev. 2017;4(6):870–878. doi: 10.1093/nsr/nwx109. [DOI] [Google Scholar]
  • 95.Hu Q., Liu J., Chen J., et al. Mineralogy of the deep lower mantle in the presence of H2O. Natl. Sci. Rev. 2021;8(4) doi: 10.1093/nsr/nwaa098. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 96.Hou M., He Y., Jang B.G., et al. Superionic iron oxide–hydroxide in Earth's deep mantle. Nat. Geosci. 2021;14(3):174–178. doi: 10.1038/s41561-021-00696-2. [DOI] [Google Scholar]
  • 97.Tang R., Liu J., Kim D.Y., et al. Chemistry and P-V-T equation of state of FeO2H at the base of Earth’s lower mantle and their geophysical implications. Sci. Bull. 2021;66(19):1954–1958. doi: 10.1016/j.scib.2021.05.010. [DOI] [PubMed] [Google Scholar]
  • 98.Zhang L., Yuan H., Meng Y., Mao H.K. Discovery of a hexagonal ultradense hydrous phase in (Fe,Al)OOH. Proc. Natl. Acad. Sci. USA. 2018;115(12):2908–2911. doi: 10.1073/pnas.1720510115. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 99.Li Y., Vočadlo L., Sun T., Brodholt J.P. The Earth's core as a reservoir of water. Nat. Geosci. 2020;13(6):453–458. doi: 10.1038/s41561-020-0578-1. [DOI] [Google Scholar]
  • 100.Tagawa S., Gomi H., Hirose K., Ohishi Y. High-temperature equation of state of FeH: Implications for hydrogen in Earth's inner core. Geophys. Res. Lett. 2022;49(5) doi: 10.1029/2021gl096260. [DOI] [Google Scholar]
  • 101.Ko B., Chariton S., Prakapenka V., et al. Water-induced diamond formation at Earth's core-mantle boundary. Geophys. Res. Lett. 2022;49(16) doi: 10.1029/2022gl098271. [DOI] [Google Scholar]
  • 102.Liu J., Sun Y., Lv C., et al. Iron-rich Fe-O compounds at Earth's core pressures. Innovation. 2023;4(1) doi: 10.1016/j.xinn.2022.100354. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Document S1. Figures S1–S14, Tables S1–S3, and supplemental materials and methods
mmc1.pdf (2.8MB, pdf)
Document S2. Article plus supplemental information
mmc2.pdf (7.2MB, pdf)

Articles from The Innovation are provided here courtesy of Elsevier

RESOURCES