skip to main content

Title: The Speciation and Coordination of a Deep Earth Carbonate‐Silicate‐Metal Melt

Ab initio molecular dynamics calculations on a carbonate‐silicate‐metal melt were performed to study speciation and coordination changes as a function of pressure and temperature. We examine in detail the bond abundances of specific element pairs and the distribution of coordination environments over conditions spanning Earth’s present‐day mantle. Average coordination numbers increase continuously from 4 to 8 for Fe and Mg, from 4 to 6 for Si, and from 2 to 4 for C from 1 to 148 GPa (4,000 K). Speciation across all pressure and temperature conditions is complex due to the unusual bonding of carbon. With the increasing pressure, C‐C and C‐Fe bonding increase significantly, resulting in the formation of carbon polymers, C‐Fe clusters, and the loss of carbonate groups. The increased bonding of carbon with elements other than oxygen indicates that carbon begins to replace oxygen as an anion in the melt network. We evaluate our results in the context of diamond formation and of metal‐silicate partitioning behavior of carbon. Our work has implications for properties of carbon and metal‐bearing silicate melts, such as viscosity, electrical conductivity, and reactivity with surrounding phases.

more » « less
Author(s) / Creator(s):
 ;  ;  ;  
Publisher / Repository:
DOI PREFIX: 10.1029
Date Published:
Journal Name:
Journal of Geophysical Research: Solid Earth
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. null (Ed.)
    Volatiles including carbon and hydrogen are generally considered to be more soluble in silicate melts than in mantle rocks. How these melts contribute to the storage and distribution of key volatiles in Earth's interior today and during its early evolution, however, remains largely unknown. It is essential to improve our knowledge about volatiles-bearing silicate magmas over the entire mantle pressure regime. Here we investigate molten Mg FexSiO3 ( , 0.25) containing both carbon and hydrogen using first-principles molecular dynamics simulations. Our results show that the dissolution mechanism of the binary volatiles in melts varies considerably under different conditions of pressure and redox. When incorporated as CO2 and H2O components (corresponding to oxidizing conditions) almost all carbon and hydrogen form bonds with oxygen. Their speciation at low pressure consists of predominantly isolated molecular CO2, carbonates, and hydroxyls. More oxygenated species, including tetrahedrally coordinated carbons, hydrogen (O-H-O) bridges, various oxygen-joined complexes appear as melt is further compressed. When two volatiles are incorporated as hydrocarbons CH4 and C2H6 (corresponding to reducing conditions), hydroxyls are prevalent with notable presence of molecular hydrogen. Carbon-oxygen bonding is almost completely suppressed. Instead carbon is directly correlated with itself, hydrogen, and silicon. Both volatiles also show strong affinity to iron. Reduced volatile speciation thus involves polymerized complexes comprising of carbon, hydrogen, silicon, and iron, which can be mostly represented by two forms: C1−4H1−5Si0−5O0−2 (iron-free) and C5−8H1−8Si0−6Fe5−8O0−2. The calculated partial molar volumes of binary volatiles in their oxidized and reduced incorporation decrease rapidly initially with pressure and then gradually at higher pressures, thereby systematically lowering silicate melt density. Our assessment of the calculated opposite effects of the volatile components and iron on melt density indicates that melt-crystal density crossovers are possible in the present-day mantle and also could have occurred in early magma ocean environments. Melts at upper mantle and transition zone conditions likely dissolve carbon and hydrogen in a wide variety of oxidized and non-oxygenated forms. Deep-seated partial melts and magma ocean remnants at lower mantle conditions may exsolve carbon as complex reduced species possibly to the core during core-mantle differentiation while retaining a majority of hydrogen as hydroxyls-associated species. 
    more » « less
  2. The distribution of nitrogen in geologic systems is modulated by its partitioning between silicate (mineral and melt) and fluid phases. Under geologically applicable oxygen fugacity, pressure, and temperatures, nitrogen can be multiply-speciated, with N2 coexisting with reduced nitride (N􀀀 3) species. Non-polar, neutral species, including N2, tend to concentrate in fluids, while charged nitride species have a greater propensity to concentrate in silicate phases. The stoichiometry of converting N2 to single N atom nitride species implies that nitrogen speciation may depend on its concentration, and this leads to the hypothesis that the partitioning of nitrogen between silicate and fluid phases also depends on concentration, potentially biasing prior experimental work in doped systems and influencing the behavior of nitrogen in geologic systems. To test this hypothesis, we have completed a series high pressure (~1.75 GPa, 800 ◦C) experiments that react minerals, melts, and fluids with variable nitrogen concentrations (3.1–17.1 wt% N). Our results imply order-of-magnitude-scale increases in mineral/melt and melt/fluid partitioning as nitrogen concentrations decrease within natural ranges. For example, decreasing the N concentration from 2500 to 2 ppm increases predicted DN melt/fluid values by over an order of magnitude at constant PT conditions. This means that loss of nitrogen from a degassing magma or dehydrating slab is a self-limiting process that becomes increasingly inefficient as nitrogen concentration falls. Despite this, nitrogen remains highly concentrated in the atmosphere, which receives N from fluids exsolved from slabs and magmas. To maintain a nitrogen-rich atmosphere we therefore suggest that warm and oxidizing conditions have prevailed over subduction zones because warm slabs dehydrate under lower pressures where nitrogen is more easily partitioned into fluids, and oxidizing conditions also promote nitrogen partitioning into fluids. Concentration-dependent partitioning of nitrogen will also serve to moderate any initial variations of N/K in slab materials upon dehydration, and this may help to explain the relatively uniform N/K ratio of MORB mantle. We supplement our nitrogen concentration experiments with a temperature series (1.5–2 GPa, 750–950 ◦C). Our temperature series data reveal that at high temperature nitrogen favors melts over fluids, while temperature has no resolvable effect of biotite-fluid partitioning. 
    more » « less
  3. Abstract

    Super‐Earths ranging up to 10 Earth masses (ME) with Earth‐like density are common among the observed exoplanets thus far, but their measured masses and radii do not uniquely elucidate their internal structure. Exploring the phase transitions in the Mg‐silicates that define the mantle‐structure of super‐Earths is critical to characterizing their interiors, yet the relevant terapascal conditions are experimentally challenging for direct structural analysis. Here we investigated the crystal chemistry of Fe3O4as a low‐pressure analog to Mg2SiO4between 45–115 GPa and up to 3000 K using powder and single crystal X‐ray diffraction in the laser‐heated diamond anvil cell. Between 60–115 GPa and above 2000 K, Fe3O4adopts an 8‐fold coordinated Th3P4‐type structure (I‐43d,Z = 4) with disordered Fe2+and Fe3+into one metal site. This Fe‐oxide phase is isostructural with that predicted for Mg2SiO4above 500 GPa in super‐Earth mantles and suggests that Mg2SiO4can incorporate both ferric and ferrous iron at these conditions. The pressure‐volume behavior observed in this 8‐fold coordinated Fe3O4indicates a maximum 4% density increase across the 6‐ to 8‐fold coordination transition in the analog Mg‐silicate. Reassessment of the FeO—Fe3O4fugacity buffer considering the Fe3O4phase relationships identified in this study reveals that increasing pressure and temperature to 120 GPa and 3000 K in Earth and planetary mantles drives iron toward oxidation.

    more » « less
  4. Abstract

    The new metal‐enriched anomalous chondrite Sierra Gorda 013 (SG 013) contains two different lithologies. Lithology 1 (L1) is represented by anomalous CBa‐like chondrite material containing ~80 vol% of Fe,Ni‐metal particles and globules up to 6 mm in size; chondrules and clasts of types POP, BO, and SO (up to 5 mm in diameter); rare sulfides; and shock melted silicate–metal areas. It does not contain any fine‐grained matrix. Several chondrules contain chromite–pyroxene symplectites. Lithology 2 (L2) has a recrystallized texture with evenly distributed olivine, pyroxene and plagioclase. L2 does not have any chondrules or sulfides, and contains less Fe,Ni‐ metal (~25 vol%) than L1. Both lithologies contain reduced olivine (Fa2–4) and pyroxene (Fs3.5), similar to CBa chondrites. Similar to CBa, there is no Ni‐Co correlation in the SG 013 metal. Rare sulfides in L1 are enriched in V. Chromite was observed in both lithologies. Oxygen isotope compositions of both lithologies are different but in the range of CBa chondrites. Bulk major and trace element geochemistry of nonporphyritic chondrules and bulk siderophile compositions in metal globules of L1 indicate elemental fractionation during formation of metallic and silicate objects with records of the evaporation process: depletion in moderate and volatile elements with the exception of Cr. Bulk geochemistry of porphyritic chondrules of L1 and the silicate portion of L2 is similar and also indicates evaporation processes. The rare Earth element (REE) distribution of L1 chondrules records a very fractionated signature corresponding to possible differentiated precursor material, while the REE pattern of L2 is primitive chondritic. The formation of SG 013 could be explained by collisions of planetesimals producing an impact plume, the precursor material of which could be chondritic and possibly differentiated. Both lithologies were affected by secondary processes: L1 preserved the traces of shock events and partial melting resulting in formation of symplectites in chondrules, melt pockets, and metal–silicate melt between the metal globules; L2 was affected by shock thermal metamorphism (up to 900 °C) resulting in recrystallization.

    more » « less
  5. Abstract

    The mineral apatite, Ca10(PO4)6(F,OH,Cl)2, incorporates sulfur (S) during crystallization from S-bearing hydrothermal fluids and silicate melts. Our previous studies of natural and experimental apatite demonstrate that the oxidation state of S in apatite varies systematically as a function of oxygen fugacity (fO2). The S oxidation states –1 and –2 were quantitatively identified in apatite crystallized from reduced, S-bearing hydrothermal fluids and silicate melts by using sulfur K-edge X-ray absorption near-edge structure spectroscopy (S-XANES) where S 6+/ΣS in apatite increases from ~0 at FMQ-1 to ~1 at FMQ+2, where FMQ refers to the fayalite-magnetite-quartz fO2 buffer. In this study, we employ quantum-mechanical calculations to investigate the atomistic structure and energetics of S(-I) and S(-II) incorporated into apatite and elucidate incorporation mechanisms.

    One S(-I) species (disulfide, S22−) and two S(-II) species (bisulfide, HS−, and sulfide, S2−) are investigated as possible forms of reduced S species in apatite. In configuration models for the simulation, these reduced S species are positioned along the c-axis channel, originally occupied by the column anions F, Cl, and OH in the end-member apatites. In the lowest-energy configurations of S-incorporated apatite, disulfide prefers to be positioned halfway between the mirror planes at z = 1/4 and 3/4. In contrast, the energy-optimized bisulfide is located slightly away from the mirror planes by ~0.04 fractional units in the c direction. The energetic stability of these reduced S species as a function of position along the c-axis can be explained by the geometric and electrostatic constraints of the Ca and O planes that constitute the c-axis channel.

    The thermodynamics of incorporation of disulfide and bisulfide into apatite is evaluated by using solid-state reaction equations where the apatite host and a solid S-bearing source phase (pyrite and Na2S2(s) for disulfide; troilite and Na2S(s) for sulfide) are the reactants, and the S-incorporated apatite and an anion sink phase are the products. The Gibbs free energy (ΔG) is lower for incorporation with Na-bearing phases than with Fe-bearing phases, which is attributed to the higher energetic stability of the iron sulfide minerals as a source phase for S than the sodium sulfide phases. The thermodynamics of incorporation of reduced S is also evaluated by using reaction equations involving dissolved disulfide and sulfide species [HnS(aq)(2−n) and HnS(aq)(2−n); n = 0, 1, and 2] as a source phase. The ΔG of S-incorporation increases for fluorapatite and chlorapatite, and decreases for hydroxylapatite, as these species are protonated (i.e., as n changes from 0 to 2). These thermodynamic results demonstrate that the presence of reduced S in apatite is primarily controlled by the chemistry of magmatic and hydrothermal systems where apatite forms (e.g., an abundance of Fe; solution pH). Ultimately, our methodology developed for evaluating the thermodynamics of S incorporation in apatite as a function of temperature, pH, and composition is highly applicable to predicting the trace and volatile element incorporation in minerals in a variety of geological systems. In addition to solid-solid and solid-liquid equilibria treated here at different temperatures and pH, the methodology can be easily extended to different pressure conditions by just performing the quantum-mechanical calculations at elevated pressures.

    more » « less