Pergamon Geochimica et Cosmochimica Acta. Vol. 58, No. 13, pp. 2921-2926, 1994 Copyright 0 1994 Else&r Science Ltd Printed in the USA. All rights reserved 00 I6-7037/94 $6.00 + .OO 0016-7037(94)00089-l An ancient Sm-Nd age for a ferroan noritic anorthosite clast from lunar breccia 67016 CHANTAL ALIBERT,‘,’MARC D. NORMAN,‘,’and MALCOLM T. MCCULLOCH’ ‘Research School of Earth Sciences, Australian National University, Canberra A.C.T. 0200, Australia %entre de Recherches PCtrographiques et Gbchimiques (CRPG), BP 20, 54500 Vandoeuvre-l&-Nancy, France 3Planetary Geosciences, Dept. of Geology and Geophysics, School of Ocean and Earth Sciences and Technology, University of Hawaii, Honolulu HI 96822, USA zyxwvutsrqponmlkjihgfedcbaZYXWVUT (Received September 2 1, 1993; accepted in revised form February 10, 1994) Abstract-Strontium and neodymium systematics have been examined in a clast of ferroan noritic an- orthosite from Apollo 16 breccia 67016. Two splits (,328 and ,326) of the same clast give different Sm- Nd results. Split ,328 gives a well defined internal isochron age of 4.562 + 0.068 Ga and an initial ‘43Nd/ l”Nd ratio of0.50673 f 10 corresponding to tNd= 0. I f 0.2 (2u optimized error) relative to the Murchinson carbonaceous chondrite. The pyroxene separate from split ,326 lies on the same isochron. In contrast, the plagioclase and whole-rock from split ,326 fall below this line, indicating a small-scale disturbance of the Sm-Nd system. This may reflect either an isotopic exchange between the plagioclase and a low Sm/Nd mineral or a loss of radiogenic ‘43Nd from the plagioclase, possibly during the period of major impacts at -3.9 Ga. The preservation of an extremely old age for the noritic ferroan anorthosite 670 16,328 suggests a rapid cooling of this rock at an early stage in the evolution of the lunar magma ocean. This old age is also consistent with giant impact models for the formation of the Moon but implies a relatively early event (pre 4.50 Ga) and, therefore, rapid accretion and differentiation of the terrestrial planets. INTRODUCTION ALTHOUGH SOME CONSENSUS has been reached about the growth of planetesimals by runaway accretion, many aspects concerning the initial conditions, such as the role of nebular gases and size range of planetesimals, are still controversial (e.g., WETHERILL and STEWART, 1989; BARGEand PELLAT, 1991; KOLVOORD and GREENBERG, 1992; SAFRONOV, 1991). The possible range in these initial conditions corresponds to accretionary timescales which can vary by an order of mag- nitude, leading to very different implications for planetary differentiation. The relatively short formation interval of less than lo7 years which is expected for hierarchical accretion of massive parent bodies would result in rapid release of gravitational energy and global melting. Alternatively, for the Earth, a more extended formation interval (- 10’ years) would imply cooler conditions and, therefore, less extensive melting of the planet during its initial differentiation. The highlands crust of the Moon contains the oldest pre- served record of crustal formation in the Earth-Moon system. The age of the Moon’s primary crust can, therefore, provide an upper limit for the duration of planet formation in the inner Solar System. Remnants of the early lunar crust are represented by the ferroan anorthosites and the Mg-suite of dunites, troctolites, norites, and gabbronorites. Mg-suite rocks have provided Sm-Nd, U-Pb, and 40Ar-39Arplateau ages in the range 4.2-4.43 Ga (e.g., CARLSON and LUGMAIR, 1981; NYQUIS~ et al., 1981; COMPSTON et al., 1984; JESSBERGER et al., 1977; MAURER et al., 1978). More recently, CARLSON and LUCMAIR (1988) reported a precise Sm-Nd age of 4.44 k 0.02 Ga for ferroan anorthosite 60025, and SHIH et al. (1993) reported a Sm-Nd age of 4.46 + 0.07 Ga for a noritic clast from breccia 15445, suggesting a similar antiquity for both suites of rock. These relatively young ages are consistent with an origin of the Moon by a giant impact between a Mars-sized planetesimal and the proto-Earth, - 100 Ma after To, and are compatible with models of WETHERILL(1992) suggesting that accretion continues for -100 Ma before a giant impact becomes highly probable. In contrast, an older Rb-Sr age of 4.5 1 f 0.07 Ga has been obtained for the troc- tolite 76535 (PAPANASTASSIOU and WASSERBURG, 1976) and a U-Pb model age of 4.5 1 + 0.0 1 Ga for the ferroan anor- thosite 60025 (HANAN and TILTON, 1987). These latter ages are significantly older than the Sm-Nd ages obtained from the same rocks and have, therefore, been considered as un- reliable (LUGMAIR et al., 1976; CARLSON and LUGMAIR, 1988; PREMOand TATSUMO~O, 1992). For example, PREMO and TATSUMO~O (1992) call upon a late loss of Rb from olivine in the troctolitic cumulate 76535 and also show that using the Canyon Diablo Pb values as initial Pb gives a model age which is too old for this sample. This conclusion is also likely to be applicable to the ferroan anorthosite 60025. The ferroan noritic anorthosite clast analyzed here for strontium and neodymium isotopes is from feldspathic frag- mental breccia 67016, which was collected from the rim of North Ray Crater. Petrographic and geochemical accounts of breccia 670 16 have been given previously (NORMAN, 198 1; LINDSTROM and SALPAS,1983; NORMANet al., 1991; NOR- MAN and TAYLOR, 1992; NORMAN et al., 1993). The clast which we analyzed belongs to the mafic subgroup of the fer- roan anorthosite suite (JAMESet al., 1989). Clasts of similar composition have been described in other feldspathic frag- mental breccias from North Ray Crater (LINDSTROMand SALPAS, 1983; MCGEE, 1988). 40Ar-39Ar ages around 3.95 Ga have been reported for a dark clast of melt breccia and a plagioclase separate from breccia 67016 (TURNER and CA- DOGAN, 1975). 292 1