Search for: All records

Abstract Thermophysical properties of silicate liquids under extreme conditions are critical for understanding the accretion and evolution of super‐Earths and sub‐Neptunes. The thermal equation of state and viscosity of silicate liquids determine the adiabatic profiles and dynamics of magma oceans. However, these properties are challenging to constrain at elevated pressures in experiments. Here, we perform ab initio molecular dynamics simulations of MgSiO₃liquid across a wide range of pressures (0–1,200 GPa) and temperatures (2200–14000 K) and analyze its structure, the Grüneisen parameter, and viscosity. Our results reveal the clear temperature and pressure dependence of the Grüneisen parameter, which varies synchronously with the O‐O coordination number. The Grüneisen parameter shifts from positive to negative temperature dependence between ∼20 and 70 GPa, corresponding to a peak in the O‐O coordination number and SiO₅abundance. Initially, the Grüneisen parameter increases with pressure and then decreases, showing limited temperature dependence above ∼300 GPa, where its behavior resembles that of solids. Furthermore, we determine the adiabat and viscosity profiles of magma oceans in super‐Earths and sub‐Neptunes. The results suggest that the mantles of super‐Earths and sub‐Neptunes may solidify either from the bottom up or at pressures of ∼120–150 GPa, depending on the curvature of the mantle melting line. The low viscosity of magma oceans likely enhances convective currents and facilitate efficient differentiation. These thermophysical properties, now quantified up to terapascal pressures, enable updates to the mass‐radius relation of magma ocean exoplanets, showing notable differences compared to their solid counterparts. 

Abstract Earth's accretion was highly energetic and likely involved multiple global melting events. Following the Moon‐forming giant impact, extensive mantle melting and the separation of solids and melts under deep mantle pressures likely produced a basal magma ocean (BMO) beneath the solidified mantle. The presence and evolution of the BMO have been proposed to explain key geophysical and geochemical features of the lowermost mantle. Understanding the evolution of the BMO is crucial for testing these hypotheses, but its interaction with the core presents a significant challenge, as the mechanism of this exchange remains unclear. In this study, we develop a theoretical framework to assess the regime of BMO‐core exchange based on the compositions of the BMO and the core. We propose that during solidification, the BMO may evolve into a regime where the reaction at the BMO‐core interface drives compositional convection in liquids on both sides, if the core has a high enough Si content (–, under the assumption that the O content is –). In this scenario, the BMO‐core exchange would be much more efficient than previously estimated, buffering the tendency of FeO enrichment during crystallization and shortening the lifetime of the BMO. 

Radiogenic heat production is fundamental to the energy budget of planets. Roughly half of the heat that Earth loses through its surface today comes from the three long-lived, heat-producing elements (potassium, thorium, and uranium). These three elements have long been believed to be highly lithophile and thus concentrate in the mantle of rocky planets. However, our study shows that they all become siderophile under the pressure and temperature conditions relevant to the core formation of large rocky planets dubbed super-Earths. Mantle convection in super-Earths is then primarily driven by heating from the core rather than by a mix of internal heating and cooling from above as in Earth. Partitioning these sources of radiogenic heat into the core remarkably increases the core-mantle boundary (CMB) temperature and the total heat flow across the CMB in super-Earths. Consequently, super-Earths are likely to host long-lived volcanism and strong magnetic dynamos. 

Diffusional isotope fractionation has been widely used to explain lithium (Li) isotope variations in minerals and rocks. Isotopic mass dependence of Li diffusion can be empirically expressed as , where is the diffusivity of a Li isotope. The knowledge about temperature and compositional dependence of the factor which is essential for understanding diffusion profiles and mechanisms remains unclear. Based on the potential energy and interatomic forces generated by deep neural networks trained with ab initio data, we performed deep potential molecular dynamics (DPMD) simulations of several Li pseudo-isotopes (with mass = 2, 7, 21, 42 g/mol) in albite, hydrous albite, and model basalt melts to evaluate the factor. Our calculated diffusivities for 7Li in albite and model basalt melts at 1800 K compare well with experimental results. We found that in albite melt decreases from at 4000 K to at 1800 K. The presence of water appears to slightly weaken the temperature dependence of , with decreasing from to in hydrous albite melt. The calculated in model basalt melt takes much smaller values, decreasing from at 4000 K to at 1800 K. Our prediction of in albite and hydrous albite melts is in good agreement with experimental data. More importantly, our results suggest that Li isotope diffusion in silicate melts is strongly dependent on melt composition. The temperature and compositional effects on can be qualitatively explained in terms of ionic porosity and the coupled relationship between Li diffusion and the mobility of the silicate melt network. Two types of diffusion experiments are suggested to test our predicted temperature and compositional dependence of . This study shows that DPMD is a promising tool to simulate the diffusion of elements and isotopes in silicate melts. 

Diffusional isotope fractionation occurs in geochemical processes (such as magma mixing, bubble growth, and crystal growth), even at magmatic temperatures. Isotopic mass dependence of diffusion is commonly expressed as Di Dj ¼ mj mi
 b , where Di and Dj are diffusion coefficients of two isotopes whose masses are mi and mj. How the dimensionless empirical parameter b depends on temperature, pressure, and composition remains poorly constrained. Here, we conducted a series of first-principles molecular dynamics simulations to evaluate the b factor of Mg isotopes in MgSiO3 and Mg2SiO4 melts using pseudo-isotope method. In particular, we considered interactions between Mg isotopes by simultaneously putting pseudo-mass and normalmass Mg atoms in a simulation supercell. The calculated b for Mg isotopes decreases linearly with decreasing temperature at zero pressure, from 0:158  0:004 at 4000 K to 0:121  0:017 at 2200 K for MgSiO3 melt and from 0:150  0:004 at 4000 K to 0:101  0:012 at 2200 K for Mg2SiO4 melt. Moreover, our simulations of compressed Mg2SiO4 melt along the 3000 K isotherm show that the b value decreases linearly from 0:130  0:006 at 0 GPa to 0:060  0:011 at 17 GPa. Based on our diffusivity results, the empirically established positive correlation between b and solvent-normalized diffusivity (Di/DSi) seems to be applicable only at constant temperatures or in narrow temperature ranges. Analysis of atomistic mechanisms suggests that the calculated b values are inversely correlated with force constants of Mg at a given temperature or pressure. Good agreement between our first principles results with available experimental data suggests that interactions between isotopes of major elements must be considered in calculating b for major elements in silicate melts. Also, we discuss diffusion-controlled crystal growth by considering our calculated b values. 

Abstract Silicate melts have served as transport agents in the chemical and thermal evolution of Earth. Molecular dynamics simulations based on a deep neural network potential trained byab initiodata show that the viscosity of MgSiO₃melt decreases with increasing pressure at low pressures (up to ∼6 GPa) before it starts to increase with further compression. The melt electrical conductivity also behaves anomalously; first increasing and then decreasing with pressure. The melt accumulation implied by the viscosity turnover at ∼23 GPa along mantle liquidus offers an explanation for the low‐velocity zone at the 660‐km discontinuity. The increase in electrical conductivity up to ∼50 GPa may contribute to the steep rise of Earth's electrical conductivity profiles derived from magnetotelluric observations. Our results also suggest that small fraction of melts could give rise to detectable bulk conductivity in deeper parts of the mantle.