EVIDENCE for LATE FORMATION and YOUNG METAMORPHISM in the ACHONDRITE NAKHLA

We have determined an age T = 1.87 AE for Nakhla and an initial s7Sr/S%r, I -0.7028. A rather thorough Sr equilibration took place at this time between all the phases of this meteorite, but there is clear evidence of somewhat incomplete isotopic equilibrium during the 1.87 AE event, which is not consistent with the derivation of Nakhla from a magma at this time. Nakhla has a young model age of $.6-2.9 AE which requires that gross differentiation processes for its parent body occur between 8.6 and 1.87 AE, significantly later than 4.6 AE. The most reasonable model is that Nakhla formed as an igneous rock at '-' $ AE and was metamorphosed at 1.87 AE. Several similarities exist between the Nakhla parent planet and the Earth and indicate the existence of other objects with close terrestrial chemical affinities. We report here an age of 1.87 AE for the achondrite Nakhla as determined by the Rb-Sr method. This investigation is part of a systematic search in meteorites for evidence of long term planetary evolution and for indications of the intense bombardment history at ~ 4.0 AE postulated for the moon by Tera et a/.[19,78, 1974a, b]. The Rb-Sr internal isochron age of $.8 AE on Kodaikanal [Burnett and Wasserburg, 1967] clearly demonstrated the existence of differentiation processes for the parent planets of meteorites on a long time scale. More recently, Papanastassiou et al. [1974] have shown that the achondrite Kapoeta contains basaltic clasts with Rb-Sr internal isochron ages of 8.6AE and 8.9AE. Our interest in Nakhla was stimulated by its peculiar rare earth element pattern [Schmitt and Smith, 1968] and the low 4øAr-a9Ar gas retention age of 1.8 AE for Lafayette and Nakhla [Podosek and Huneke, 1978; Podosek, 1978]. A sample of Nakhla (A) was provided to us by R. Schmitt. A preliminary analysis of a total rock (Table 1) yielded a model age of TBABI = $.6 AE and indicated that this meteorite was considerably ounger than reported earlier by Pinson et al. [1965], In order to verify that this discrepancy was not due to heterogeneity of the meteorite, the sample from the Harvard collection (• 489b) which was used by Pinson et al. [1965] was also analyzed (sample B). From our results, it appears that the earlier analyses are in error. Study of the meteorite fragments with the binocular microscope showed that a brown stain permeated many of the pale green pyroxene grains along cracks and cleavages. Clusters of white, fine grained, feathery plagioclase crystals, intergrown with an SiO2 polymorph, occasionally surrounded the pyroxene grains as rims and were present.in the junctions between pyroxene grains. Inspection of thin sections howed the brown stain to permeate many of the pyroxene and olivine grains on a fine scale. This brown stain proved to be a "glass" with extremely high K (up to 9%). Material interior to the fragments provided to us was chipped for analysis. Mineral separates were prepared using heavy liquids and a Frantz magnetic separator. Electron microprobe analyses were done on the mineral separates by A. Chodos. The compositions of the major phases are shown in Table 2 and are in good agreement with those obtained by Prior[1912]. While the pyroxenes were generally quite uniform, a pyroxene was also found which was Copyright ̧1974 by the American Geophysical Union richer in Fe and poorer in Mg, (19.4% FeO and 9.7% MgO). No Ni was detected in the olivine. The thin films of brown glass were difficult to analyze and gave poor sums but were typically K rich (2-9% K20). They appeared to be of variable composition and relatively rich in FeO. A typical analysis for one of these brownies is listed in Table 2. We consider this type of material to be the devitrified glass of a late stage mesostasis. The high K glass was concentrated in the lower density fractions; however, the bulk of the 2.6 and 2.5 g/cm '3 mineral separates consisted of plagioclase with a tendency to be somewhat more sodic than the typical plagioclases. The glass was intimately intergrown in these plagioclase crystals in small brown blebs and coatings. Considerable care was necessary to remove alkali rich interstitial glass from the pyroxene separates. A high purity pyroxene separate was obtained by successive crushing and density separations followed by magnetic separation and hand picking. The analytical data for Rb and Sr and the K concentrations are given in Table 1. The blank levels were 0.01 ng Rb, 0.1 ng Sr and 10 ng K, and have a negligible ffect on the data. Most samples were first dissolved; a less than one percent aliquot of the total solution was spiked and the K, Rb and Sr concentrations were determined without chemical separation. The remaining solution was then spiked optimally. The glass separate from fragment A was spiked and then split into two aliquots which were processed independently; the results are in good agreement. The first total sample of fragment A was not spiked totally. The analyses of two independently spiked aliquots of the dissolved sample consisting of 8 and 50 percent of the total solution are in good agreement. The other total samples were totally spiked. Table 1. Rb, Sr, K Restfits Sample Wgt. K c Rb d 88Sr d mg 10-Smole/g 87Rb/86Sr 87Sr/86Sre TBAB I