1,2Jérôme Aléon, 3,4Johanna Marin-Carbonne, 3Kevin D. McKeegan, 5Ahmed El Goresy
Geochimica et Cosmochimica Acta (in Press) Link to Article [https://doi.org/10.1016/j.gca.2018.04.001]
1Centre de Science Nucléaire et de Science de la Matière, CNRS/IN2P3 – Université Paris-Sud UMR 8609, Bâtiment 104, 91405 Orsay Campus, France
2Institut de Minéralogie, de Physique des Matériaux et de Cosmochimie, UMR 7590, Sorbonne Université, Museum National d’Histoire Naturelle, CNRS, Univ. Pierre et Marie Curie, IRD, 61 rue Buffon, 75005 Paris, France
3Department of Earth, Planetary, and Space Sciences, University of California – Los Angeles, 595 Charles Young Drive East, Los Angeles, CA 90095-1567, USA
4Laboratoire Magma et Volcans, UMR 6524, Univ. Lyon, Univ. Jean Monnet Saint-Etienne, CNRS, Univ. Clermont Auvergne, IRD, 23 rue du Dr Paul Michelon, 42023 Saint-Etienne, France
5Bayerisches Geoinstitut, Universität Bayreuth, D-95440 Bayreuth, Germany
Oxygen, magnesium, and silicon isotopic compositions in the mineralogically complex, ultrarefractory (UR) calcium-aluminum-rich inclusion (CAI) E101.1 from the reduced CV3 chondrite Efremovka confirm that E101.1 is a compound CAI composed of several lithological units that were once individual CAIs, free-floating in the solar protoplanetary disk. Each precursor unit was found to have had its own thermal history prior to being captured and incorporated into the partially molten host CAI.
Four major lithological units can be distinguished on the basis of their isotopic compositions. (1) Al-diopside-rich sinuous fragments, hereafter sinuous pyroxene, are 16O-rich (Δ17O ≤ -20‰) and have light Mg and Si isotopic compositions with mass fractionation down to -3.5‰/amu for both isotopic systems. We attribute these peculiar isotopic compositions to kinetic effects during condensation out of thermal equilibrium. (2) Spinel clusters are 16O-rich (Δ17O ∼ -22‰) and have Mg isotope systematics consistent with extensive equilibration with the host melt. This includes (i) δ25Mg values varying between +2.6 ‰ and +6.5 ‰ close to the typical value of host melilite at ∼+5‰, and (ii) evidence for exchange of radiogenic 26Mg with adjacent melilite as indicated by Al/Mg systematics. The spinel clusters may represent fine-grained spinel-rich proto-CAIs captured, partially melted, and recrystallized in the host melt. Al/Mg systematics indicate that both the sinuous pyroxene fragments and spinel clusters probably had canonical or near-canonical 26Al contents before partial equilibration. (3) The main CAI host (Δ17O ≤ -2‰) had a complex thermal history partially obscured by subsequent capture and assimilation events. Its formation, referred to as the “cryptic” stage, could have resulted from the partial melting and crystallization of a 16O-rich precursor that underwent 16O-depletion and a massive evaporation event characteristic of F and FUN CAIs (Fractionated with Unknown Nuclear effects). Alternatively, a 16O-rich UR precursor may have coagulated with a 16O-poor FUN CAI having 48Ca anomalies, as indicated by perovskite, before subsequent extensive melting. The Al/Mg systematics (2.4 × 10-5 ≤ ≤ 5.4 × 10-5, where is a model initial 26Al/27Al ratio per analysis spot) are best understood if the FUN component was 26Al-poor, as are many FUN CAIs. (4) A complete Wark-Lovering rim (WLR) surrounds E101.1. Its Mg and Si isotopic compositions indicate that it formed by interaction of the evaporated interior CAI with an unfractionated 16O-rich condensate component. Heterogeneities in 26Al content in WLR spinels (3.7 × 10-5 ≤ ≤ 5.7 × 10-5) suggest that the previously reported age difference of as much as 300,000 years between interior CAIs and their WLRs may be an artifact resulting from Mg isotopic perturbations, possibly by solid state diffusion or mixing between the interior and condensate components.
The isotopic systematics of E101.1 imply that 16O-rich and 16O-poor reservoirs co-existed in the earliest solar protoplanetary disk and that igneous CAIs experienced a 16O-depletion in an early high temperature stage. The coagulation of various lithological units in E101.1 and their partial assimilation supports models of CAI growth by competing fragmentation and coagulation in a partially molten state. Our results suggest that chemical and isotopic heterogeneities of unclear origin in regular CAIs may result from such a complex aggregation history masked by subsequent melting and recrystallization.