Iron snow, crystal floats, and inner-core growth: modes of core solidification and implications for dynamos in terrestrial planets and moons

Abstract

Recent planetary space missions, new experimental data, and advanced numerical techniques have helped to improve our understanding of the deep interiors of the terrestrial planets and moons. In the present review, we summarize recent insights into the state and composition of their iron (Fe)-rich cores, as well as recent findings about the magnetic field evolution of Mercury, the Moon, Mars, and Ganymede. Crystallizing processes in iron-rich cores that differ from the classical Earth case (i.e., Fe snow and iron sulfide (FeS) crystallization) have been identified and found to be important in the cores of terrestrial bodies. The Fe snow regime occurs at pressures lower than that in the Earth's core on the iron-rich side of the eutectic, where iron freezes first close to the core-mantle boundary rather than in the center. FeS crystallization, instead, occurs on the sulfur-rich side of the eutectic. Depending on the core temperature profile and the pressure range considered, FeS crystallizes either in the core center or close to the core-mantle boundary. The consequences of the various crystallizing mechanisms for core dynamics and magnetic field generation are discussed. For the Moon, revised paleomagnetic data obtained with advanced techniques suggest a peculiar history of its internal dynamo, with an early strong field persisting between 4.25 and 3.5 Ga, and subsequently a much weaker field. In addition, the long-lasting dynamo and the possible presence of an inner core, as inferred from a revised interpretation of Apollo seismic data, suggest core crystallization as a viable process of magnetic field generation for a substantial period during lunar evolution. The present-day magnetic fields of Mercury and Ganymede (if they occur on the iron-rich side of the Fe-FeS eutectic) and the related dynamo action are likely generated in the Fe snow regime and seem to be recent features. An earlier dynamo in Mercury would have been powered differently. For Mercury, MESSENGER data further suggest core formation under reducing conditions that may have resulted in an Fe-S-Si composition, further complicating the core crystallization process. Mars, with its early and strong paleo-field, likely has not yet started to freeze out an inner iron core.

Constraining the Date of the Martian Dynamo Shutdown by Means of Crater Magnetization Signatures

“…This, in turn, hints to the possibility of a dynamo magnetic field that has not been constant throughout its history. The iron snow regime has been shown to be a driving mechanism that can account for a weak magnetic field (see, e.g., Vilim et al, , for a study on Mercury and Breuer et al, , for a review on possible dynamo driving mechanisms for planetary bodies). In any case we note that since its visible magnetization is nonzero, its null space is also nonzero.…”

Thermal Conductivity of FeS and Its Implications for Mercury's Long‐Sustaining Magnetic Field

“…The temperature at Mercury's core mantle boundary has been investigated in numerous previous studies (Grott et al, ; Hauck et al, ; Tosi et al, ). The solidification of FeS below the CMB can be expected if the temperatures at the core mantle boundary decrease below 1,700 K (Breuer et al, ). We have obtained the thermal conductivity at 1,300 K, close to the eutectic temperature of Fe‐FeS system at 10 GPa (Fei et al, ; Morard et al, ).…”

“…It has been suggested that Mercury's dynamo is currently generated by chemical convection (Breuer et al, 2015;Cao et al, 2014;Chen et al, 2008;Dumberry & Rivoldini, 2015). The low magnetic field intensity of planet Mercury has been discussed using thermoelectric (Stevenson, 1987), thin shell (Stanley et al, 2005), thick shell (Heimpel et al, 2005), and feedback (Glassmeier et al, 2007) dynamo models.…”

Water and the Interior Structure of Terrestrial Planets and Icy Bodies

“…If the melting temperature has a lower gradient than the core temperature, iron will start to crystallize at the core-mantle boundary (CMB) and a growing snow zone will form during further cooling (McKinnon, 1996;Hauck et al, 2006). However, the unknown oxidation state of the interior during differentiation limits our ability to constrain whether the composition of a sulfur-bearing core is on the Fe-or FeSrich side of the eutectic composition (e.g., Scott et al, 2002;Breuer et al, 2015). Another important element could be hydrogen.…”