Magnesium isotopic constraints on the origin of CBb chondrites

The magnesium isotopic composition of Calcium-, Aluminium-rich Inclusions (CAIs) and chondrules from the CBb chondrites HH237 and QUE94411 was measured using MC-ICPMS coupled with a laser ablation system.CAIs from CBb chondrites exhibit limited mass-dependent fractionation (δ25Mg′ (DSM3) < 1.3‰) and formed with undetectable 26Al (26Al/27Al < 4.6 × 10− 6). Petrographic observations suggest that CBb CAIs are igneous. The magnesium isotopic composition of CBb igneous CAIs contrast with that of CV3 igneous CAIs which are usually mass fractionated and formed with an elevated initial abundance of 26Al. We contend that the absence of 26Al in CAIs is due either to a late formation in the case of a stellar origin of 26Al, or to a lack of exposure to impulsive flares in the case of an irradiation origin of 26Al. In both cases, it implies that a protoplanetary disk was present ∼ 4563 Ma ago, when CBb chondrites agglomerated.Chondrules have δ25Mg′ (DSM3) varying from − 0.80 to 0.95‰. A rough negative correlation is observed between the δ25Mg′ of chondrules and their 24Mg/27Al ratio. This correlation is attributed to evaporation rather than mixing. Contrary to CAIs, chondrules from CBb chondrites have a magnesium isotopic composition similar to that of CV3 chondrules. This last result is surprising as CBb chondrules are significantly different from CV3 chondrules in mineralogy and chemistry.If chondrules from CBb chondrites formed in an impact-related vapour plume as proposed by Krot et al. [A.N. Krot, Y. Amelin, P. Cassen and A. Meibom, Young chondrules in CB chondrites formed by a giant impact in the early Solar System, Nature 436 (2005) 989–992], our data show that physical conditions in the vapour plume were similar to those of the solar accretion disk at the time and location of the formation of CV chondrules. We note that the oxygen isotopic composition of CAIs is incompatible with their remelting in the putative impact vapour plume. Alternatively, it is possible that CBb chondrules formed in a protoplanetary disk as the differences between these and “normal” CV3 chondrules can also be explained in term of spatial and temporal variations of the protoplanetary disk. We show that their young Pb–Pb age is not an argument in favour of an impact origin as protoplanetary disks can last as long as 10 Myr around protostars. If CBb chondrites formed in the solar accretion disk, we speculate they might be the last formed chondrite group. Such a hypothesis might shed light on the unique properties of CBb chondrites.