Negligible Ca isotope fractionation in deep magma ocean suggests an unsolidified proto-Earth during Moon formation

1,2Wei Dai, 1Julien Siebert, 3Alexandre Corgne, 1Nicolas Wehr, 4Yunke Song, 1Frederic Moynier
Earth and Planetary Science Letters (in Press) Open Access Link to Article [DOI: 10.1016/j.epsl.2026.120342]
1Université Paris Cité, Institut de Physique du Globe de Paris, CNRS, 1 rue Jussieu, Paris, 75005, France
2Centre for Star and Planet Formation, Globe Institute, University of Copenhagen, DK-1350 Copenhagen K, Denmark
3Instituto Ciencias de la Tierra, Universidad Austral de Chile, Valdivia, Chile
4Bayerisches Geoinstitut, University of Bayreuth, Bayreuth, 95440, Germany
Copyright Elsevier

Considerable isotopic variation between Earth and major chondritic materials implies a possible fractionation in the early differentiate stage of the Earth. However, it is challenging to study isotopic fractionation at the high pressures relevant to the deep Earth due to the lack of natural samples. We present Ca isotopic fractionation (Δ44/40Capv-melt) between calcium perovskite (Ca-Pv, CaTiO3) and co-existing silicate glass (melt fraction) from experimental samples. The Δ44/40Capv-melt varies in experiments with different pressure, yields equilibrium fractionation factor [103lnα(1000K)] of 0.040.08+0.06 at 1 atm and 0.500.07+0.07 at 3 GPa, respectively. The variation on Ca isotope fractionation is likely caused by Ca-Pv phase structure transition from orthorhombic/tetragonal (coordination number (CN) = 8) to cubic (CN = 12) with the increase of pressure and temperature. Our model calculation shows that Ca perovskite is enriched in heavy Ca isotopes relative to silicate melts by 0.02 to 0.08 ‰ in the lower mantle. It suggests that the Ca isotope fractionation in the deep magma ocean is less than 0.05 ‰, regardless of the variation of isotopic fractionation factor caused by phase transition. We conclude that no considerable Ca isotope variation occurred in the deep mantle and the Ca isotope composition of upper mantle could be representative of the bulk silicate Earth. Considering the Nd isotopic differences between Earth and Moon, we suggest that the Moon formed primarily from a partly molten proto-Earth with advanced crystallization of perovskites in its lower mantle.

Discuss