Closed system oxygen isotope redistribution in igneous CAIs upon spinel dissolution

1Jérôme Aléon
Earth and Planetary Sciences 482, 324-333 Link to Article [https://doi.org/10.1016/j.epsl.2017.11.027]
1Institut de Minéralogie, de Physique des Matériaux et de Cosmochimie UMR 7590, Sorbonne Universités, Museum National d’Histoire Naturelle, CNRS, UPMC, IRD, 61 rue Buffon, 75005 Paris, France
Copyright Elsevier

In several Calcium–Aluminum-rich Inclusions (CAIs) from the CV3 chondrites Allende and Efremovka, representative of the most common igneous CAI types (type A, type B and Fractionated with Unknown Nuclear isotopic anomalies, FUN), the relationship between 16O-excesses and TiO2 content in pyroxene indicates that the latter commonly begins to crystallize with a near-terrestrial 16O-poor composition and becomes 16O-enriched during crystallization, reaching a near-solar composition. Mass balance calculations were performed to investigate the contribution of spinel to this 16O-enrichment. It is found that a back-reaction of early-crystallized 16O-rich spinel with a silicate partial melt having undergone a 16O-depletion is consistent with the O isotopic evolution of CAI minerals during magmatic crystallization. Dissolution of spinel explains the O isotopic composition (16O-excess and extent of mass fractionation) of pyroxene as well as that of primary anorthite/dmisteinbergite and possibly that of the last melilite crystallizing immediately before pyroxene. It requires that igneous CAIs behaved as closed-systems relative to oxygen from nebular gas during a significant fraction of their cooling history, contrary to the common assumption that CAI partial melts constantly equilibrated with gas. The mineralogical control on O isotopes in igneous CAIs is thus simply explained by a single 16O-depletion during magmatic crystallization. This 16O-depletion occurred in an early stage of the thermal history, after the crystallization of spinel, i.e. in the temperature range for melilite crystallization/partial melting and did not require multiple, complex or late isotope exchange. More experimental work is however required to deduce the protoplanetary disk conditions associated with this 16O-depletion.