The formation and evolution of the Moon’s crust inferred from the Sm-Nd isotopic systematics of highlands rocks

Lars E. Borg, William S. Cassata, Josh Wimpenny, Amy M. Gaffney, Charles K. Shearer

Geochimica et Cosmochimica Acta
In Press, Journal Pre-proof, Available online 19 September 2020

LINK

“Ages determined for magnesian and ferroan anorthosite crustal rock suites overlap, suggesting they formed contemporaneously about 4.3 to 4.5 Ga. A notable exception is the Sm-Nd age previously determined on Mg-suite gabbronorite 67667 which is at least 100 Ma younger than the youngest ferroan anorthosite. New chronologic measurements of 67667 presented here yield concordant Sm-Nd and Rb-Sr mineral isochron ages of 4349 ± 31 Ma and 4368 ± 67 Ma, suggesting the samples is older than previous estimates. Furthermore, a whole rock Sm-Nd isochron of Mg-suite rocks from the Apollo 14, 15, 16, and 17 landing sites yields an age of 4348 ± 25 Ma, indicating that Mg-suite magmatism was widespread and roughly contemporaneous on the lunar nearside. Analysis of Sm-Nd internal isochron ages confirms that Mg-suite magmatism was restricted to a period between about 4.33 and 4.35 Ga at the Apollo 14, 15, 16, and 17 landing sites and was synchronous with magmatism at the Apollo 16 site associated with the ferroan anorthosite suite between 4.35 and 4.37 Ga. Magnesian- and ferroan anorthosite suite rocks with ages younger than ∼4.33 Ga appear to have experienced slow cooling in the deep lunar interior, so that the ages record when the samples cooled below the closure temperature of the Sm-Nd isotopic system and not the time they crystallized.

The ages determined for Mg-suite and ferroan anorthosite suite rocks are concordant with the age determined for the formation of urKREEP of 4350 ± 34 Ma using the Sm-Nd isotopic systematics of 67667 and measurements completed on norite 78238, troctolite 76535, KREEP basalt 15386, and gabbronorite NWA 773. Crystallization ages of Mg-suite and FAS are also concordant with the average of 146Sm-142Nd ages previously determined for the formation of the mare basalt source region of 4333 ± 30 Ma. The similarity of ages for Mg-suite magmatism, ferroan anorthosite suite magmatism, urKREEP formation, and formation of the mare basalt source regions implies the processes that produced these rocks were petrogenetically linked. It also implies that both early-stage and late-stage lunar magma ocean cumulates formed over a relatively short duration of <40 Ma. Late and somewhat rapid solidification of a lunar magma ocean can account for the concordance of ferroan anorthosite suite rocks, urKREEP, and the mare basalt source regions. However, the major and trace element compositions of Mg-suite magmas preclude them from being a primary differentiation product of the lunar magma ocean. Instead, the Mg-suite could be produced as a result of mixing of magma ocean solidification products during density driven overturn occurring immediately after, or perhaps during, solidification of the lunar magma ocean. This scenario not only accounts for the chronology of the various rock suites, but is consistent with the petrogenesis of the Mg-suite that involves the interaction between pre-existing Mg-rich, plagioclase-rich, and urKREEP-rich cumulates of the magma ocean.”