First‐Principles Study of FeO₂H_x Solid and Melt System at High Pressures: Implications for Ultralow‐Velocity Zones

Abstract

Pyrite-type FeO 2 H x (P phase) has recently been suggested as a possible alternative to explain ultralow-velocity zones due to its low seismic velocity and high density. Here we report the results on the congruent melting temperature and melt properties of P phase at high pressures from first-principles molecular dynamics simulations. The results show that P phase would likely be melted near the core-mantle boundary. Liquid FeO 2 H x has smaller density and smaller bulk sound velocity compared to the isochemical P phase. As such, relatively small amounts of liquid FeO 2 H x could account for the observed seismic anomaly of ultralow-velocity zones. However, to maintain the liquid FeO 2 H x within the ultralow-velocity zones against compaction requires special physical conditions, such as relatively high viscosity of the solid matrix and/or vigorous convection of the overlying mantle.Plain Language Summary Ultralow-velocity zones (ULVZs) are 5-40-km-thick patches lying above Earth's core-mantle boundary. They are characterized with anomalously low seismic velocities compared with the ambient mantle and may contain important clues on the thermochemical evolution of the Earth. A recent experimental study argued that ULVZs may be caused by the accumulation of pyrite-type FeO 2 H x (P phase) at the bottom of the mantle. Here for the first time, we systematically study the thermoelastic properties of both FeO 2 H x solid and liquid phases. We find that P phase is likely melted near the core-mantle boundary and thus cannot be the source of ULVZs. Furthermore, in order for the molten product of P phase to cause ULVZs, the dense and nearly inviscid melts must be dynamically stable and confined within the ULVZs, which requires that the mantle is highly viscous and/or convects vigorously.

Chemical Reaction Between Metallic Iron and a Limited Water Supply Under Pressure: Implications for Water Behavior at the Core‐Mantle Boundary

“…The results of recent experimental studies suggest that subducted plates may deliver surface water to the deep lower mantle via CaCl 2 ‐type hydrous phases (Duan et al, 2018; Nishi et al, 2014, 2019; Ohira et al, 2014; Panero & Caracas, 2017; Sano et al, 2008; Tsuchiya, 2013; Walter et al, 2015) and pyrite‐type iron‐hydroxide FeO 2 H x (Liu et al, 2017; Nishi et al, 2017). These deep subducted hydrous phases may release water at the bottom of the mantle (Deng et al, 2019; Nishi et al, 2017; Piet et al, 2020) and/or form some reaction layers at the core‐mantle boundary via active interaction with iron from the core (Liu et al, 2017; Mao et al, 2017; Ohtani et al, 2018).…”

Thermoelastic Properties of B2‐Type FeSi Under Deep Earth Conditions: Implications for the Compositions of the Ultralow‐Velocity Zones and the Inner Core

“…The ULVZs exhibit 5%-20% reductions in compressional wave velocities (V P ) and 10%-30% reductions in shear wave velocities (V S ) compared to the surrounding mantle (McNamara et al, 2010;Thorne & Garnero, 2004), but a 5%-20% increase in density (Idehara, 2011;Reasoner & Revenaugh, 2000;Rost et al, 2005). Several mechanisms for the origin of ULVZs have been proposed, including: (a) partial melting due to high temperatures near the CMB (Andrault et al, 2012;Revenaugh & Meyer, 1997;Williams & Garnero, 1996), (b) isolated remnants of ancient subducted slabs (Dobson & Brodholt, 2005), (c) crystallization products from a basal magma ocean (Labrosse et al, 2007), and (d) enrichment of dense iron-rich phases (Deng et al, 2019;J. Liu et al, 2017;W.…”

“…The ULVZs exhibit 5%–20% reductions in compressional wave velocities (V P ) and 10%–30% reductions in shear wave velocities (V S ) compared to the surrounding mantle (McNamara et al., 2010; Thorne & Garnero, 2004), but a 5%–20% increase in density (Idehara, 2011; Reasoner & Revenaugh, 2000; Rost et al., 2005). Several mechanisms for the origin of ULVZs have been proposed, including: (a) partial melting due to high temperatures near the CMB (Andrault et al., 2012; Revenaugh & Meyer, 1997; Williams & Garnero, 1996), (b) isolated remnants of ancient subducted slabs (Dobson & Brodholt, 2005), (c) crystallization products from a basal magma ocean (Labrosse et al., 2007), and (d) enrichment of dense iron‐rich phases (Deng et al., 2019; J. Liu et al., 2017; W. L. Mao et al., 2006; Mergner et al., 2021; Wicks et al., 2010, 2017). Recent geodynamical modeling suggested that ULVZs located close to the edges of thermochemical piles in the lower mantle are most likely caused by compositionally distinct materials, whereas partial melting is possible only in the hottest regions of the piles (M. Li et al., 2017).…”

Nonideal Mixing Effects in Warm Dense Matter Studied with First-Principles Computer Simulations

“…The accuracy of this approach depends largely on the thermodynamic conditions at which it is applied [48]. A wide variety of systems has been explored under the assumption of ideal mixing [49][50][51][52][53][54][55][56]. The hypothesis works remarkably well for hydrocarbon mixtures in the warm dense matter regime [57], water-hydrogen mixtures [58], as well as in hydrogen-helium mixtures enriched in heavier elements [17,33,59,60].…”