Comment on “Ancient Asteroids Enriched in Refractory Inclusions” Dominik C. Hezel* and Sara S. Russell Sunshine et al. (Reports, 25 April 2008, p. 514) reported that certain asteroids contain 30 T 10 volume percent calcium- and aluminum-rich inclusions (CAIs). We contend that the amount of CAIs in CV chondrites is two to three times as low as the 10 volume percent assumed by the authors; thus, we question whether the CAI-rich bodies they studied are indeed older than known asteroids or formed before the injection of 26 Al into the solar nebula. S unshine et al.(1) reported the finding of certain asteroids that they claim to consist of 30 T 10 volume % of calcium- and aluminum-rich inclusions (CAIs). CAIs are the oldest objects found in chondrites, up to 1 to 2 million years (My) older than more common chondrules [e.g., ( 2–4)]. CAIs are a diverse group of objects and occur in variable abundances in the different chondrite groups. The formation conditions of the different types of CAIs are highly debated, and it is unclear how CAIs could have been stored for up to 2 My before being incorporated into asteroids together with chon- drules. The finding of Sunshine et al.(1) that some asteroids have high abundances of CAIs is of great importance, because these might indicate that CAIs formed in a separate region of the neb- ula. Some CAIs might then have escaped from this region, moved to other areas, and become trapped in asteroids that formed in regions that may have lacked any CAIs. This would strengthen the picture of a solar nebula with regions of different physical and chemical conditions, each favoring different components to form. How- ever, we contend that the modal abundances of 30 T 10 volume % CAIs proposed for certain asteroids by Sunshine et al.(1) are based on incorrect assumptions for chondrite CAI modal abundances and therefore may be too high. Sunshine et al.(1) combined spectral obser- vations of a set of asteroids with modeling of the observed spectra from laboratory spectra ob- tained from different chondritic materials. In par- ticular, they measured spectra from CAIs within the Allende CV3 meteorite, CAI-free Allende matrix, and different types of CAIs. They used the spectra to model the bulk Allende chondrite. Their modeling results indicate a CAI abundance of 10% in bulk Allende, which the authors note is in excellent agreement with the known abundance of CAIs in CV3 meteorites. This is because the authors assume 10 volume % of CAIs for CV chondrites [taken from table 3 in (5)]. That table contains a column designated “Refractory inclusion abundance 2 (vol%).” The superscripted “2” is described in the table caption as “Refractory inclu- sion abundance includes CAI+AOI.” The abbre- viation AOI stands for amoeboid olivine inclusions [or amoeboid olivine aggregates (AOAs)]. These are petrologically and chemically different inclu- sions than CAIs [e.g., (6)], and their reported modal abundances often exceed the reported CAI modal abundances in the same chondrite (7–10). As their name indicates, AOAs consist primarily of olivines, whereas CAIs have only minor amounts of olivine. Sunshine et al.(1) measured an AOA, which has a different spectra than CAIs. They did not include this spectra in their further discus- sions. To summarize, the numbers reported in table 3 in (5) and used by Sunshine et al. are not CAI modal abundances. The following CAI modal abundances have so far been reported for CV chondrites: 2.5 to 9.4% (7), 2.52% (11), and 0.65 to 1.89% (12). For CO chondrites, reported CAI abundances are 1.2 to 3.5% (8), 1.0 to 3.6% (9), 0.63 to 1.5% (13), and 0.85 to 1.38% (12). Despite these references, there is so far no good compilation of CAI modal abundances in the literature. There are certain problems associated with getting good numbers for CAI modal abun- dances that we summarize and discuss in (14). In (14), we combined theoretical considerations, modeling, and measurements to obtain CAI modal abundances in chondrites and found 3.0 +0.3 – 0.1 volume % for CAIs in CV and 1.0 +1.0 – 0.3 volume % for CAIs in CO chondrites—far below the 10 volume % CAIs for CVand 13 volume % CAIs for CO chondrites indicated in (1). We acknowledge that CAI modal abundances might vary among meteorites of a single chondrite group and that CAI modal abundances in Allende could be higher than 3.0 +0.3 –0.1 volume %. However, if Al as a representative refractory element were concen- trated only in CAIs, Allende could not have more than 9.3 volume % CAIs (14). Because much of the Al in Allende is concentrated in chondrules and matrix, the value Sunshine et al. use for Allende is about two to three times too high. This means either their model used to calibrate their spectra needs a correction or something is missing in the interpretation of their spectra. One mis- conception could be that some CV chondrites do contain abundant spinel in their matrix (15, 16), opposite to what Sunshine et al. state, “In mete- orites, abundant aluminous spinels are only ob- served in CAIs ... ” In conclusion, we encourage Sunshine et al. (1) to check their calibration and calculations using a new assumption that Allende contains around 3 volume % CAIs. The asteroids they mea- sured may still have unusually high CAI modal abundances, but perhaps not as high as 30 vol- ume %; a range of 10 to 15 volume % would better agree with previous results (17). From the CAI modal abundances we obtained for carbonaceous chondrites, we calculate that CAIs contribute <10% to the Al budget of chondrites, with the exception of CV, where CAIs contribute ~25% to the Al budget. Even if the CAI modal abundance in all these chondrites were about three times higher, a large fraction of Al would still be con- tained in chondrules and matrix. CAIs are therefore not the dominant carrier of Al or possible 26 Al. As a result, the enhanced abundance of CAIs in the asteroids reported by Sunshine et al.( 1) do not necessarily require that these asteroids were melted, if all CAIs in them had canonical initial 26 Al of 5 × 10 -5 . This means that these asteroids need not necessarily be older than other asteroids from which we have samples or that their CAIs formed before a possible injection of 26 Al in the protoplanetary nebula. Whether this asteroid fam- ily provides the best candidates for sample-return missions may also need to be reconsidered. References 1. J. M. Sunshine, H. C. Connolly Jr., T. J. McCoy, S. J. Bus, L. M. La Croix, Science 320, 514 (2008). 2. G. J. MacPherson, A. M. Davis, E. K. Zinner, Meteoritics 30, 365 (1995). 3. S. S. Russell, G. Srinivasan, G. R. Huss, G. Wasserburg, G. J. MacPherson, Science 273, 757 (1996). 4. Y. Amelin, A. N. Krot, I. D. Hutcheon, A. A. Ulyanov, Science 297, 1678 (2002). 5. A. J. Brearley, R. H. Jones, in Reviews in Mineralogy, Vol. 36, Planetary Materials, J. J. Papike, Ed. (Mineralogical Society of America, Washington, DC, 1998). 6. A. N. Krot et al., Chem. Erde 64, 185 (2004). 7. H. Y. McSween Jr., Geochim. Cosmochim. Acta 41, 1777 (1977). 8. H. Y. McSween Jr., Geochim. Cosmochim. Acta 41, 477 (1977). 9. A. E. Rubin et al., Meteoritics 20, 175 (1985). 10. M. B. Norton, H. Y. McSween Jr., Lun. Planet. Sci. Conf. 38, 1807 (2007). 11. A. S. Kornacki, J. A. Wood, J. Geophys. Res. 89, (suppl.), B573 (1984). 12. C. May, S. S. Russell, M. M. Grady, Lun. Planet. Sci. Conf. 30, 1688 (1999). 13. S. S. Russell et al., Geochim. Cosmochim. Acta 62, 689 (1998). 14. D. C. Hezel, S. S. Russell, A. J. Ross, A. T. Kearsley, Meteorit. Planet. Sci.; http://arxiv.org/abs/0810.2174 (2008). 15. T. Murakami, Y. Ikeda, Meteoritics 29, 397 (1994). 16. D. C. Hezel, H. Palme, Earth Planet. Sci. Lett. 265, 716 (2008). 17. T. H. Burbine, M. J. Gaffey, J. F. Bell, Meteorit. Planet. Sci. 27, 424 (1992). 4 June 2008; accepted 6 October 2008 10.1126/science.1161386 TECHNICAL COMMENT Impacts and Astromaterials Research Centre (IARC), Depart- ment of Mineralogy, Natural History Museum, Cromwell Road, London SW7 5BD, UK. *To whom correspondence should be addressed. E-mail: d.hezel@nhm.ac.uk 14 NOVEMBER 2008 VOL 322 SCIENCE www.sciencemag.org 1050a on April 14, 2017 http://science.sciencemag.org/ Downloaded from