Structure and Melting of Fe, MgO, SiO 2 , and MgSiO 3 in Planets: Database, Inversion, and Phase Diagram

Junjie Dong (董俊杰), Gabriel-Darius Mardaru, Paul D. Asimow, Lars P. Stixrude, Rebecca A. Fischer · 2025

Read Structure and Melting of Fe, MgO, SiO 2 , and MgSiO 3 in Planets: Database, Inversion, and Phase Diagram by Junjie Dong (董俊杰), Gabriel-Darius Mardaru, Paul D. Asimow, Lars P. Stixrude, Rebecca A. Fischer (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-21734.txt.

Introduction

Knowing when rock melts deep inside a planet is key to modeling its interior, yet experts have long disagreed on the melting curves of even basic materials. Dong and co-authors build phase diagrams for four of them: iron, magnesium oxide, silica, and magnesium silicate.

The team combined an experimental database on phase equilibria with logistic regression and supervised learning, inverting the data into pressure-temperature diagrams that reach 5,000 gigapascals. The new diagrams settle long disputes about where each material melts and freezes. Their reach covers the interiors of giant planets and super-Earth exoplanets, where pressures far exceed anything in Earth's mantle. The authors offer the diagrams as a firmer base for models of rocky planet structure.

Catalog record

CREATORJunjie Dong (董俊杰), Gabriel-Darius Mardaru, Paul D. Asimow, Lars P. Stixrude, Rebecca A. Fischer
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE83 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2503-21734

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Structure and Melting of Fe, MgO, SiO 2 , and MgSiO 3 in Planets: Database, Inversion, and Phase Diagram Junjie Dong (董俊杰), Gabriel-Darius Mardaru, Paul D. Asimow, Lars P. Stixrude, Rebecca A. Fischer Abstract We present globally inverted pressure–temperature ( P P – T T ) phase diagrams up to 5,000 GPa for four fundamental planetary materials, Fe, MgO, SiO 2 , and MgSiO 3 , derived from logistic regression and supervised learning, together with an experimental phase equilibria database. These new P P – T T phase diagrams provide a solution to long-standing disputes about their melting curves. Their implications extend to the melting and freezing of rocky materials in the interior of giant planets and super-Earth exoplanets, contributing to the refinement of their internal structure models. I Introduction Self-gravitation and internal heating of a planet induce high pressures and temperatures that alter the structure and physical properties of its interior. Knowledge of the phase equilibria of a planet’s constituent materials, including solid-solid phase transitions and melting, is essential, as these fundamental thermodynamic processes can regulate convective dynamics, thermal evolution, chemical differentiation, and magnetic field generation [cf. 181 , 117 , 182 , 166 , and references therein] . In light of the rapid discovery of exoplanets by ground- and space-based telescopes [cf. 182 , 166 , and references therein] , as well as the emerging geophysical constraints on giant planets in the outer Solar System from existing and planned missions [ 63 , 194 , 178 , 75 , e.g.] , there is a growing need to better characterize the material properties of giant planet and exoplanet interiors in order to better understand our observations. While extensive research has focused on the behavior of hydrogen [cf. 73 , and references within] and water [ 138 , 81 , 33 , 172 , 235 , e.g.,] at high pressure and temperature due to their stellar abundance and suitability for testing fundamental material theories, considerably less attention has been paid to rocky materials (MgO, SiO 2 , and MgSiO 3 ) and iron (Fe), except in the context of studying the Earth’s interior [ 126 , 208 ] . However, these materials also constitute the deep interiors of giant planets [ 117 , 133 ] and exoplanets [ 51 ] . The pressure–temperature ( P – T ) phase diagrams of these materials, especially their melting curves, remain poorly understood under the relevant conditions of giant planets and exoplanets. The centers of giant planets in our Solar System can reach pressures of 500–5,000 GPa [ 148 , 133 , 205 , 136 , 137 , 90 ] , while super-Earth exoplanets have central pressures of up to 5,000 GPa at 10 Earth masses [ 204 , 51 , 21 ] . In contrast to the low melting temperature and relatively uniform fluid behavior of hydrogen, rocky materials and iron metal are relatively refractory between 1,000–5,000 GPa [ 176 , 199 , 72 , 68 , 59 , 65 , 39 , 142 , 203 , 109 , 69 ] , and exhibits a wide range of exotic chemical behaviors at ultrahigh pressures potentially as the core of gaint planets or the mantle of super-Earth and sub-Neptune exoplanets, including the metallization of oxide solids [ 135 , e.g.,] , the diverse polymorphism of silicate solids [ 229 , e.g.,] , and the enhanced electrical conductivity of the silicate liquid [ 46 , 119 ] . These results highlight the profound implications of solid-solid phase transitions and melting in the interiors of giant planets and exoplanets. Previous efforts to directly determine the P – T phase boundaries of rocks and iron have typically focused on one phase boundary at a time, using a single set of phase equilibrium data consisting of 10–100 points near the phase boundary to bracket the transition [ 32 , e.g.] . Alternatively, equations of state for the materials are determined experimentally and then combined with extrapolated thermodynamic properties, such as enthalpy ( \Delta H ) and entropy ( \Delta S ), to construct a thermodynamic model of their phase equilibria [ 208 , 104 , e.g.] . The phase boundaries determined by the former approach often rely on visual inspection and freehand drawing based on a small data set and do not directly show interlaboratory discrepancies; the topology of the phase equilibria computed by the latter approach is very sensitive to the choice of thermodynamic parameters, which are often measured experimentally only at low temperatures and pressures and then extrapolated to the pressure and temperature ranges of planetary interiors [ 104 ] . Here, we present an alternative approach (Section II ) that combines all available static compression data from 1 bar to 100–300 GPa, supplemented by select dynamic compression and computational data obtained at much higher pressures up to 5,000 GPa (Section III ). We applied a global inversion algorithm based on logistic regression and supervised learning (Section II.2 ) to determine the phase transition boundaries (including melting curves) of MgO, SiO 2 , MgSiO 3 , and Fe. This global approach, based on an extensive database (Section II.1 ), allows us to conveniently infer the solid–solid phase transitions and melting curves of rocky materials and iron at pressure conditions relevant to giant planets and super-Earth exoplanets (1,000–5,000 GPa) without the need for complete information on their thermodynamic properties. We then performed thermodynamic modeling of the phase equilibria for MgO, SiO 2 , MgSiO 3 , and Fe between 1 bar to 200 GPa to validate our global inversion results. Finally, we discuss their implications for melting and freezing in planetary interiors (Section IV ). II Methods II.1 Phase Equilibria Database We have compiled a comprehensive phase equilibria database for Fe, MgO, SiO 2 , and MgSiO 3 , including a wide range of experimental and theoretical data produced over \sim 80 years, from the early 1940s to the present. The database includes \sim 3,300 entries for Fe, \sim 1,100 for MgO, \sim 1,600 for SiO 2 , and \sim 800 for MgSiO 3 . The data are categorized into three main types: 1. Experimental data from static compression techniques, such as multi-anvil press (MA) and diamond anvil cell (DAC) [ 84 , 132 ] ; 2. Experimental data from dynamic compression techniques, such as gas gun compression and laser and magnetic pulse compression [ 52 ] ; 3. Computationally-simulated data from first-principles quantum mechanical methods, such as density functional theory (DFT) calculations combined with molecular dynamics (MD) simulations as well as quantum Monte Carlo (QMC) simulations [ 236 ] . Most of the data entries are compiled from experimental observations of phase equilibria in static compression experiments, and among them we further distinguish two main types: 1. Direct observations: crossing of a phase transition or reaction boundary; 2. Indirect observations: identification of a stable phase. Direct observations involve the identification of a phase transition or reaction in a single experiment. We note that such observations can be influenced by kinetic barriers, making it difficult to unambiguously identify the equilibrium phase boundaries. For example, the sluggish kinetics of MgSiO 3 at low temperature often leads to conflicting observations of its transition or reaction boundary [ 112

Read the complete text