(38 Ar/36 Ar)Cosmo = 1.6± 0.1 Fig. 1. Cosmochron Diagram for Apollo 12

Lunar and Planetary Science XXXVII (2006) 1190.pdf

IMPLANTED AND COSMOGENIC 38Ar AND 36Ar IN LUNAR IMPACT SPHERULES. J. Levine1,2,R.A. Muller1,3, and P. R. Renne4,5, 1Department of Physics, University of California, Berkeley, California 94720, USA, 2Presently at Chicago Center for Cosmochemistry and Department of Geophysical Sciences, University of Chicago, Chicago, Illinois 60637, USA ([email protected]), 3Lawrence Berkeley National Laboratory, Berkeley, California 94720, USA, 4Berkeley Geochronology Center, 2755 Ridge Road, Berkeley, California 94709, USA, 5Department of Earth and Planetary Science, University of California, Berkeley, California 94720, USA.

Introduction: We recently published formation Identification of Implanted and Cosmogenic ages of 81 lunar impact spherules, determined by the Components: We measured abundances of 36-40Ar 40Ar/39Ar isochron technique [1]. These spherules by laser stepwise heating on individual spherules, from Apollo 12 soil sample 12023 may be compared degassing each spherule in 7-20 heating steps. All with those of Culler et al. [2], which were from 37Ar is created by neutron irradiation, chiefly by Apollo 14 sample 14163. On the basis of an 40Ca(n, )37Ar. After correction for reactor overabundance of impact spherules younger than 400 interferences, 38Ar and 36Ar measurements represent Ma, [2] argued for an increase of the meteoroid binary mixtures of implanted and cosmogenic bombardment flux in the inner solar system since that contributions. If we follow [5] and assume that time. The distribution of spherule ages we observed cosmogenic argon is created by spallation of calcium in [1] is consistent with this explanation, but (or other spallation targets co-distributed in the alternatives cannot be ruled out. In particular, we sample with Ca), then we may use a “cosmochron” suggested that the abundance of <300 Ma spherules diagram of 38Ar/36Ar vs. 37Ar/36Ar (so-called by in trench sample 12023 is the result of impact ejecta analogy with a 40Ar/36Ar vs. 39Ar/36Ar isochron) to in the lunar soil being stratified and overturned unravel the mixing of the two components (Fig. 1). relative to pre-impact stratigraphy. This is consistent The intercept of the trend line gives the 38Ar/36Ar with the models and laboratory experiments of ratio of the implanted component, and the slope S is Melosh [3, p. 79]. related to the cosmic ray exposure age by In addition to their geochronological implications, 38 36 ()Ar/ Ar P argon isotopes measured on lunar samples also S = 1 implanted CRE , 38 36 contain information about the abundance and ()Ar/ Ar 40F 40 cosmo distribution of calcium and potassium (which we 38 report in a companion abstract [4]), exposure to where P is the production rate of cosmogenic Ar 8 338 40 cosmic rays, and implantation of solar wind and solar (we adopt 1.4 10 cm Ar per gram of Ca per energetic particles. Here we report 38Ar/36Ar ratios of Myr from [4]), CRE is the effective duration of the implanted component in lunar spherules, as well cosmic ray exposure, F is the neutron fluence as cosmic ray exposure ages calculated from the experienced by the samples, and 40 is the fission- 40 37 38 37 spectrum averaged cross section for Ca(n, ) Ca. ratios of cosmogenic Ar to ArCa. Though we present only Apollo 12 spherule data here, we have The factor of 40 appearing in the denominator is the 40 38 36 also examined the Apollo 14 spherule data of [2], and atomic mass of Ca. While the Ar/ Ar ratio of the results are quite similar. implanted component can be determined from the cosmochron line, the isotopic composition of the Slope 0.0401 ± 0.0056 cosmogenic component must be assumed. We use 0.22 38 36 = ± CRE age 555 ± 78 Ma ( Ar/ Ar)cosmo 1.6 0.1 to encompass the value 0.215 Intercept 0.18917 ± 0.00083 of [6], but with a large uncertainty because of the MSWD = 0.653 (N = 7) 0.21 dependence of this ratio on specimen composition. Ar

36 0.205 On the basis of isotopic abundances, <10% of the 38 36

Ar/ Ar and Ar in the spherules is typically 38 0.2 cosmogenic; the rest is from the implanted 0.195 component. However, attempts to quantitatively 0.19 constrain the initial depth of the implanted argon in 0.185 the spherules have thus far been unsuccessful. Over -0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 the limited range of spherule sizes in this study 37Ar/36Ar (~200-500 μm), there is a weak tendency for larger Fig. 1. Cosmochron diagram for Apollo 12 spherule spherules to have contained more implanted argon H06. Ellipses contour 1σ. than smaller ones. Lunar and Planetary Science XXXVII (2006) 1190.pdf

Results: Implanted 38Ar/36Ar. Fig. 2 shows the spherules from the same scoop of soil (ours is from 38Ar/36Ar ratio of the implanted endmember for each ~20 cm depth in ejecta of Sharp Crater, the diameter of the 81 Apollo 12 spherules dated by [1]. Nearly of which is 14 m) have experienced vastly different all values are isotopically heavier than terrestrial exposure and transport histories in the lunar regolith. atmospheric air [7]. Though early estimates of the By contrast with the results of [5], we see no solar-wind 38Ar/36Ar ratio were indistinguishable particular clumping of exposure ages that might from air [e.g. 8], more recent measurements suggest reflect the Sharp impact event. Sharp Crater is much that the solar wind is offset toward lower 38Ar/ 36Ar smaller than that studied by [5], and may have ratios [9,10]. The offset toward higher 38Ar/36Ar excavated little fresh material from below the cosmic ratios which we observe is much too large to be ray penetration depth. Moreover, unlike some explained by production of 38Cl (which quickly mineral grains, our impact spherules were necessarily decays to 38Ar) during neutron irradiation. Either the created near the surface, meaning that each specimen solar wind 38Ar/36Ar ratio is greater than inferred by must have experienced cosmic ray irradiation prior to [10] or the implanted 38Ar and 36Ar in our spherules the Sharp event. is dominated by so-called solar energetic particle Acknowledgements: We are grateful to the Ann (SEP) implantation rather than the lower-energy solar and Gordon Getty Foundation and the Folger wind. Our measurements agree with the SEP isotopic Foundation for their support of our research. compositions of [6,9,10], both in the absolute values Jonathan Levine acknowledges the National Science and the spread in 38Ar/36Ar. Foundation for a Graduate Research Fellowship. The relative abundance of SEP argon compared Lunar samples were provided by the NASA Lunar with “normal” solar-wind argon in lunar samples is Sample Curator. orders of magnitude larger than the ratio of their References: [1] Levine, J. et al. (2005) GRL, 32, fluxes [9]. Wieler et al. [9] argue for very poor L15201, doi:10.1029/2005GL022874. [2] Culler, retention of the weakly implanted solar wind in lunar T.S. et al. (2000) Science, 287, 1785-1788. [3] plagioclase and ilmenite grains. The present Melosh, H.J. (1989) Impact Cratering: A Geologic observations suggest that glass too routinely loses Process. (New York: Oxford U. Press, 245 pp.). [4] most implanted argon from the low-energy solar- Levine, J. et al. (2006) LPSC, XXXVII (this volume). wind. [5] Turner, G. et al. (1971) Earth Planet. Sci. Lett. We hope that samples from the Genesis 12, 19-35. [6] Hashizume, K. et al. (2000) Earth spacecraft will convincingly identify the argon Planet. Sci. Lett. 202, 201-216. [7] Nier, A.O. (1950) isotopic composition of the solar wind and its relation Phys. Rev. 77, 789-793. [8] Eberhardt, P. et al. to both the terrestrial atmosphere and the trapped (1972) LSC, 3, 1821-1856. [9] Wieler, R. et al. component in lunar grains. (1986) Geochim. Cosmochim. Acta, 50, 1997-2017. Cosmic ray exposure ages. Cosmic ray exposure [10] Palma, R.L. et al. (2002) Geochim. Cosmochim. ages for the spherules dated in [1] are shown in Fig. Acta, 66, 2929-2958. 3. The most striking observation is that individual 1.4 2500

1.2 2000 0.215 1

0.8 1500 0.205 0.6

(nmol) per g Ca 1000

cosmo 0.4 0.195 Ar exposure age (Ma) implanted 37

Ar 500 38

Ar) 0.2 Ar/ 36 0.185 38

Ar/ 0 0 38

( 0 1000 2000 3000 4000 5000 0.175 40Ar/39Ar isochron age (Ma) Fig. 3. 38Ar/37Ar cosmic ray exposure age as a 0.165 40 39 0 1000 2000 3000 4000 5000 function of Ar/ Ar formation age for Apollo 12 40Ar/ 39Ar isochron age (Ma) spherules from [1]. Dashed line represents the physi- τ τ cal limit CRE = formation. The outlying datum comes Fig. 2. 38Ar/36Ar ratio of the implanted endmember in from a spherule that suffered considerable post- Apollo 12 spherules, as a function of formation age formation loss of Ar on the Moon; we suspect that we from [1]. Dashed line is atmospheric value from [7]. underestimated its isochron age in [1].