Entropy, dynamics, and freezing of CaSiO3 liquid

Wilson, Alfred J.; Stixrude, Lars

doi:10.1016/j.gca.2021.03.015

Cited by 6 publications

(21 citation statements)

References 73 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to compare our results with available experimental data (Richet et al, 1991), we computed the entropy of CaSiO 3 melt at 2000 K and ambient density using the 2PT approach. We found Ts ∼ 6.13 kJ/g, which is 0.28 kJ/g larger than the experimental value 5.85 kJ/g obtained at by Richet et al (1991), and lower than the 6.30 kJ/g obtained in Wilson and Stixrude (2021). The ∼3% difference between the Wilson and Stixrude (2021) and our 2PT data is also observed at 4000 and 6000 K, indicating our 2500 K data point is consistent with all other 2PT data obtained in this study at different temperatures.…”

Section: Melting Curves Entropy and Kinetics Of Meltingsupporting

confidence: 84%

“…The difference between our 2PT melting curve (especially above 6000 K) and that reported in Wilson and Stixrude (2021) likely comes from the remaining difference in the entropy of the melt. In particular, at 6000 K, the difference of ∼0.7 kJ/g in Ts (Figure S6 in Supporting Information ) will shift the melting pressure from 133 to 208 GPa, very close to the 2PT curve of Wilson and Stixrude (2021). Around 30% of this difference is due to a difference in the electronic entropy, which could be related to the use of different exchange‐correlation functionals (PBE vs. PBEsol).…”

Section: Melting Curves Entropy and Kinetics Of Meltingcontrasting

confidence: 78%

“…The associated dT/dp slopes decrease with increasing pressure. The 2PT melting point of 6025 K is about 1000 K lower than the other melting curves at 136 GPa, but only 440 K above the 2PT curve of Wilson and Stixrude (2021) and the Z‐method curve of Braithwaite and Stixrude (2019). The energies computed in the liquid from the TI compared to the 2PT method, indicate that the low 2PT melting curve can be explained by an overestimation of the melt entropy, which in turn underestimates the Gibbs free energy of the melt (Figures S5 and S6 in Supporting Information ).…”

Section: Melting Curves Entropy and Kinetics Of Meltingmentioning

confidence: 84%

“…The 120-and 720-atom cell solid-liquid coexistence curves are drawn with a smooth cubic spline-like function, with the 720-atom cell curve weighted by the point clusters at 18-21 GPa and 116-135 GPa. & Stixrude, 2019;Wilson & Stixrude, 2021). The classical MD study was carried out by heating homogenously until melting at a critical temperature, T h , which is much higher than the equilibrium melting temperature, T m .…”

Section: Figurementioning

confidence: 99%

See 3 more Smart Citations

Ab Initio Atomistic Simulations of Ca‐Perovskite Melting

Hernandez

Mohn

Guren

et al. 2022

Geophysical Research Letters

View full text Add to dashboard Cite

The lower mantle is dominated by the two silicate perovskites, bridgmanite and Ca-perovskite, constituting about 77 and 5 mol%, respectively, of peridotites and about 34 and 24 mol%, respectively, of recycled oceanic crust of basaltic composition (e.g., Stixrude & Lithgow-Bertelloni, 2012). The melting curves of the end member components MgSiO 3 , CaSiO 3 and MgO in the system CaO-MgO-SiO 2 system (CMS) may inform about the crystallization history of the magma ocean and the nature of the ultra-low velocity zones (e.g., Trønnes et al., 2021). Although the melting curves for MgO-periclase and MgSiO 3 -bridgmanite are reasonably well established through the lower mantle pressure range, the Ca-perovskite melting curve remains poorly constrained, in particular at the lowermost mantle pressures (Figure 1). The melting of Ca-perovskite at 14-15 and 16-60 GPa, respectively, has been investigated by multi-anvil (Gasparik et al., 1994) and laser-heated diamond anvil cell (LH-DAC) experiments (Shen & Lazor, 1995;Zerr et al., 1997). Multi-anvil experiments with resistive heating and W-Re thermocouples might provide more accurate and reproducible melting temperatures than the LH-DAC experiments, which are in poor agreement with each other with a difference in melting temperature of about 880 K at 45 GPa. Nomura et al. (2017) performed multi-anvil experiments along the MgSiO 3 -CaSiO 3 join at 24 GPa and developed a thermodynamic model, anchored to the Zerr et al. (1997) melting curve. Although the experimentally based melting curves converge in the 2500-2800 K range at 14-16 GPa, the large variation in their dT/dp slopes yields increasing discrepancies with increasing pressure. The melting curve of Gasparik et al. (1994) is steeper (212 K/ GPa in the 13-15 GPa range) than those determined by the LH-DAC experiments of Shen and Lazor (1995) and Zerr et al. (1997). The latter slopes, measured in the 17-20 GPa range are 32 and 61 K/GPa, respectively. The Ca-perovskite melting curve has also been investigated by classical molecular dynamics (MD) based on empirical atomic potentials (Liu et al., 2010) and density functional theory (DFT) MD calculations (BraithwaiteAbstract Melting curves of Ca-perovskite (pure CaSiO 3 ) were determined by ab initio density functional theory, using two solid-liquid coexistence methods and two free energy approaches, in the form of thermodynamic integration and two-phase thermodynamics. The melting curves based on the solid-liquid coexistence methods and thermodynamic integration rise steeply from 2000 K at 14 GPa to 7000 K at 136 GPa. The melting temperature at 136 GPa is 1400 K higher than previous ab initio predictions. The high thermal stability of Ca-perovskite is linked to its high-symmetry isometric structure and consistent with experiments, demonstrating that Ca-perovskite is the most refractory phase in basaltic compositions in the lower mantle pressure range. The steep dT/dp slope of the melting curve also shows that the Ca-perovskite liquidus field expands relative to those of bridgmanite...

show abstract

Section: Melting Curves Entropy and Kinetics Of Meltingsupporting

confidence: 84%

Section: Melting Curves Entropy and Kinetics Of Meltingcontrasting

confidence: 78%

Section: Melting Curves Entropy and Kinetics Of Meltingmentioning

confidence: 84%

Section: Figurementioning

confidence: 99%

See 2 more Smart Citations

Ab Initio Atomistic Simulations of Ca‐Perovskite Melting

Hernandez

Mohn

Guren

et al. 2022

Geophysical Research Letters

View full text Add to dashboard Cite

show abstract

“…Using the equations of state (EOSs) in this paper, the pressure at the boundary between a central metal core and the silicate mantle for the 4 M ⊕ planets we consider should be ∼450 GPa. The isoentropic adiabat of Wilson & Stixrude (2021) suggests that the temperature at this location could be as high as 14,000 K with an estimated maximum surface T of 6000 K. The conditions considered here are well above the liquidus for silicate and metal, so our planet beneath the atmosphere is considered to be entirely molten. Pressure effects on intramelt reactions are mitigated by the fact that as pressure increases beyond a few GPa, the thermochemistry of the melt species becomes less pressure-dependent.…”

Section: Chemical Thermodynamics At High P and Tmentioning

confidence: 96%

Chemical Equilibrium between Cores, Mantles, and Atmospheres of Super-Earths and Sub-Neptunes and Implications for Their Compositions, Interiors, and Evolution

Schlichting¹,

Young²

2022

Planet. Sci. J.

View full text Add to dashboard Cite

We investigate the equilibrium chemistry between molten metal and silicate and a hydrogen-rich envelope using 18 independent reactions among 25 phase components for sub-Neptune-like exoplanets. Both reactive and unreactive metal sequestered in an isolated core are modeled. The overarching effects of equilibration are oxidation of the envelope and reduction of the mantle and core. Hydrogen and oxygen typically comprise significant fractions of metal cores at chemical equilibrium, leading to density deficits that offer a possible alternative explanation for the low densities of the Trappist-1 planets. Reactions with the magma ocean produce significant amounts of SiO and H2O in the envelopes directly above the magma ocean. Molar concentrations in the envelopes of planets with reactive metal are H2 > SiO > CO ∼ Na ∼ Mg > H2O ≫ CO2 ∼ CH4 ≫ O2, while for the unreactive metal case, H2O becomes the second most abundant species, after H2, providing an arbiter for the two scenarios amenable to observation. The water abundances in the atmospheres exceed those in the mantles by at least an order of magnitude in both scenarios. The water concentrations in the silicate mantles are ∼0.01 and ∼0.1 wt% in the reactive and unreactive metal core cases, respectively, limiting the H2O that might be outgassed in a future super-Earth. Less dissolved water in the reactive core case is due to sequestration of H and O in the Fe-rich metal. The total hydrogen budget of most sub-Neptunes can, to first order, be estimated from their atmospheres alone, as the atmospheres typically contain more than 90% of all H.

show abstract