Solar cosmic ray records in lunar rock 64455

Available online at www.sciencedirect.com

Geochimica et Cosmochimica Acta 73 (2009) 2163–2176 www.elsevier.com/locate/gca

Solar cosmic ray records in lunar rock 64455 K. Nishiizumi a,*, J.R. Arnold b, C.P. Kohl b, M.W. Caffee c,1, J. Masarik d, R.C. Reedy e a

Space Sciences Laboratory, University of California, 7 Gauss Way, Berkeley, CA 94720-7450, USA b Department of Chemistry, University of California, San Diego, La Jolla, CA 92037-0524, USA c Center for Accelerator Mass Spectrometry, Lawrence Livermore National Laboratory, Livermore, CA 94551, USA d Nuclear Physics Department, Comenius University, SK-842 48, Bratislava, Slovakia e Institute of Meteoritics, MSC03-2050, University of New Mexico, Albuquerque, NM 87131-0001, USA Received 20 March 2008; accepted in revised form 15 December 2008; available online 10 January 2009

Abstract Cosmic ray produced 10Be (half-life = 1.36  106 yr), 26Al (7.05  105 yr), and 36Cl (3.01  105 yr) were measured in a depth profile of 19 carefully-ground samples from the glass-coated lunar surface rock 64455. The solar cosmic ray (SCR) produced 26Al and 36Cl in this rock are present in high concentrations, which in combination with the low observed erosion rate, 10 MeV) = 100 p/ cm2/4p. (B) The spectra for R0 = 90 MV, J(>10 MeV) = 100 p/cm2/4p in 64455 (depths in g/cm2, assuming a sphere with a radius of 7 g/cm2).

function of energy) for solar protons. The steep decrease in particle flux as a function of energy indicates far greater numbers of lower energy protons, resulting in the presence of SCR-produced nuclides in only the upper few g/cm2 of a surface (Reedy and Arnold, 1972). Solar-proton spectral parameters and their variations over time have been estimated using cosmogenic nuclides in lunar surface materials. Among the radionuclides mea53 10 sured are Mn (half-life = 3.7  106 yr), Be 6 (1.36  10 yr) (Nishiizumi et al., 2007), and 26Al (7.05  105 yr) (Norris et al., 1983; Nishiizumi, 2003). Cosmogenic radionuclide profiles in lunar rocks 10017 (SHRELLDALFF et al., 1970), 12002 (Finkel et al., 1971), 14321 (Wahlen et al., 1972), 14320 (Imamura et al., 1974), 68815 (Kohl et al., 1978; Nishiizumi et al., 1988), and 74275 (Nishiizumi et al., 1991) were used to determine the average SCR flux and energy spectrum over a time scale of months to 10 Myr. Measurements of SCR-produced 53 Mn in 10017 (Herr et al., 1971), 39Ar (269 yr) in 12053 (Begemann et al., 1972), 10Be, 26Al, 41Ca (1.0  105 yr), and 22Na (2.61 yr) in 74275 (Fruchter et al., 1982; Fink et al., 1998), and 14C (5,730 yr) in 68815 (Jull et al., 1998) have also been performed. SCR-produced noble gases were also measured in lunar rocks 61016 (Rao et al., 1993; Garrison et al., 1996) and 68815 (Reedy and Marti, 1991; Rao et al., 1994). Using comparisons of theoretical gardening models and observed SCR-produced activities in lunar cores Langevin et al. (1982) determined that the maximum possible R0 is 120 MV. The most detailed study performed to date utilized the highland breccia 68815 (2.0 Myr exposure age), which has a relatively high erosion rate (Kohl et al., 1978). Using 53 Mn and 26Al depth profiles in three lunar rocks (68815, 12002, and 14321), the average SCR spectrum and flux were and J(E best characterized by R0 = 100 MV >10 MeV) = 70 protons/cm2/s4p for the last 10 Myr (Kohl et al., 1978). These calculations assumed erosion rates of 0.5–2.2 mm/Myr for these rocks. However, this choice of R0 and J is not unique, it is possible to fit the data with R0 in the range 70–150 MV with appropriate adjustments

of the flux (J) and the erosion rate (Russ and Emerson, 1980). These calculations indicate that a high R0 corresponds to a low J while a low R0 corresponds to a higher J. The inability of these studies to uniquely resolve the solar-proton spectral index and the flux is common to most previous studies. This ambiguity can be addressed by measuring a medium- to high-energy product like 10Be, which is very sensitive to the SCR spectral parameter R0. Accordingly, R0 can be reasonably well constrained by the amount of 10Be in the surface layers of a lunar rock, i.e., the 10Be in excess of that produced by GCR. If there is no SCR-produced 10 Be detected, an upper limit can be set for the hardness of the SCR spectrum. In rock 68815, the fact that SCR-produced 10Be is undetectable indicates that R0 must have been lower than 85 MV for the last 2 Myr (Nishiizumi et al., 1988). Measurements of SCR-produced 21Ne, 22Ne, and 38 Ar profiles in 68815 also suggest that R0 is lower than 100 MV (Rao et al., 1994). Rao et al. (1994) also used a SCR flux (J) somewhat lower than that obtained from the 53 Mn and 26Al concentrations. The SCR-produced 21Ne, 22 Ne, and 38Ar profiles in 68815 can be brought into agreement with those obtained from radionuclide measurements by varying the erosion rate of the rock, as a higher SCR flux is needed if the erosion rate is higher. In this work, we extend our studies of detailed depth profiles of cosmogenic nuclides 10Be, 26Al, and 36Cl (3.01  105 yr) to the glass-coated rock 64455. Partial results were reported earlier (Arnold et al., 1993; Nishiizumi et al., 1995); we present here the final detailed depth profiles of 10Be, 26Al, and 36Cl using accelerator mass spectrometry (AMS) and compare them with new theoretical calculations of expected SCR production. 2. SAMPLE DESCRIPTION AND EXPERIMENTAL PROCEDURES The Apollo 16 basaltic impact melt breccia 64455 was collected from the northeast rim of a subdued crater on the northeast slope (10–15°) of Stone Mountain. It is a

Solar cosmic ray records in lunar rock 64455

small object, 5.6  4.0  2.5 cm weighing 56.7 g, and is coated with a thick (>2 mm) smooth dark glass layer (Ryder and Norman, 1980). In Figs. 2 and 3, the cubes labeled B1 and T1 designate bottom and top surfaces as described by Johnson Space Center/National Aeronautics and Space Administration (JSC/NASA). In fact, the side labeled B1 was the top of the rock on the moon as evidenced by the overlapping micro craters, which effectively destroy the original glass coated surface of the rock. The surface labeled T1 in Fig. 2, which is the bottom of the rock on the moon, shows very few such pits, and here the glass surface is preserved. Accordingly, for this work we reverse the JSC/ NASA designation. The glass coating is a splash coating of impact-melted material with a chemical composition similar to that of 64455, rather than a fusion crust (Morris et al., 1986; See et al., 1986). This observation is supported

2165

by the presence of a thin (1 mm) partially melted thermal aureole between the basalt core and the glass coating. Four distinct textural zones within the rock are: (1) a core of fractured highland basalt, (2) a rim of basalt with interstitial partial melt, (3) a thin crust of brown devitrified glass, and (4) an outer coating of fresh glass (Grieve and Plant, 1973). A 2.01 Myr 81Kr–Kr exposure age was obtained for the glass portion of 64455,17, indicating association with the 2.0 Myr South Ray cratering event (Arnold et al., 1993). The 38Ar age is 1.8 Myr, close to the 81Kr–Kr age, while the 21Ne age is 1.2 Myr, shorter than the 81Kr–Kr age (Bogard and Gibson, 1975; Bogard and Hirsch, 1975). Although its lunar surface orientation has not been definitely fixed photographically, the distribution of microcraters suggests a simple lunar surface irradiation (Schneider

Fig. 2. Sample rock 64455. The scale bars below the identification number represent centimeters. The cubes labeled B1 and T1 designate bottom and top surfaces as described by JSC/NASA. This work uses the reverse of this designation. Note, on the side labeled B1, which was in fact the top of the rock on the moon, the overlapping abundance of micro craters, which effectively destroys the original glass coated surface of the rock. The surface labeled T1 in the figure, which was in fact the bottom of the rock on the moon, shows very few such pits and here the glass surface is preserved. Pictures are NASA S-72-40134 and S-72-40135.

2166

K. Nishiizumi et al. / Geochimica et Cosmochimica Acta 73 (2009) 2163–2176

Fig. 3. The full slice in this figure is the original cut of rock 64455 by JSC/NASA. The slice with the ends removed is the sample we received (64455,82) and which we used in this work. The top of the picture is the top of the rock in its lunar orientation. The darker colored rim is caused by the glass coating on the rock; the lighter material is unaltered but shocked basalt. The top view of the sample 64455,82 marked with the cube B clearly shows micro craters covering the glass coating. The scale bars are millimeters.

and Ho¨rz, 1974). The bottom surface is covered by a smooth glassy layer without any evidence of micrometeorite damage (see Fig. 2). Blanford et al. (1974, 1975) measured cosmic ray tracks in 64455,14 and 64455,16, and they found high solar flare track densities in the surface subsample, 14, which is also heavily damaged by microcraters. They compared the track densities of 64455,14 with those of 68815 and concluded that the surface of rock 64455 was not eroded and that the top half of the rock had not been covered by soils (Blanford et al., 1974, 1975). A slice of sample 64455,82 (10  19  24 mm deep) was selected for our study based on high abundances of microcraters and cosmic ray tracks. In this slice, all exterior surfaces were completely covered by glass (see Fig. 3). This slice of sample 64455,82 was mounted on an X–Y–Z stage (Kohl et al., 1978). A surface area of about 10 mm  20 mm was ground using a dental drill. Depth measurements were done at each point of a 1 mm  1 mm grid; the depth resolution is 50 lm. The precision of the depth measurement is more than two times better than was possible for 68815 (Kohl et al., 1978). This improvement was made possible by two factors; a simpler grinding

orientation of 64455 relative to that of 68815 and the fact that we required less mass due to improved AMS detection sensitivity. The ground material was aspirated onto 0.45 lm Nucleopore filters with isopropanol, removed and weighed; the recoveries were typically >90 %, somewhat less for the smaller samples. The density of the glass, calculated from the grinding dimensions and the weight removed, is 2.74 g/cm3, in good agreement with the density of Ca feldspar, 2.76 g/cm3. We obtained 19 samples following this procedure, sampling locations ranging from the micrometeorite bombarded glass surface (top at the lunar surface) to the smooth glass surface (bottom at the lunar surface). The distance of each sample from the surface is shown in Table 1. Hereafter, ‘‘surface” refers to the micrometeorite bombarded surface as oriented to the zenith and ‘‘bottom” refers to the smooth glass surface oriented toward the lunar surface. All but one of the 19 samples were taken from thin layers ground from either the top or the bottom of the rock. One sample, ‘‘Mid” was taken from the middle of the rock with a coarse depth interval of 7.35–11.4 mm. Each sample was dissolved in an HF-HNO3 mixture along with Be (1.5–2 mg), Cl (5 mg), and Ni (0.6–1.7 mg)

Solar cosmic ray records in lunar rock 64455

2167

Table 1 Chemical composition of 64455.

T-1 T-2 T-3 T-4 T-5 T-6 T-8 T-10 Mid B-10 B-9 B-8 B-7 B-6 B-5 B-4 B-3 B-2 B-1

Depth (mm)

Wt (mg)

Mg (%)

Al (%)

K (ppm)

Ca (%)

Mn (ppm)

Fe (%)

0.00–0.22 0.22–0.40 0.40–0.59 0.59–0.91 0.91–1.38 1.38–1.88 3.48–4.00 7.18–7.74 4.35–11.4 17.6–18.2 18.2–19.1 19.1–19.5 19.5–20.2 20.2–20.7 20.7–21.1 21.1–21.4 21.4–21.7 21.7–21.9 21.9–22.0

45.93 49.63 64.88 65.45 76.18 103.59 90.28 91.19 72.46 36.71 56.96 34.47 132.22 117.80 102.35 116.92 100.99 60.00 32.82

4.82 4.81 4.90 5.02 5.12 5.24 6.00 5.82 5.37 5.40 5.42 5.13 5.15 5.00 4.98 4.85 4.76 4.63 4.72

13.4 13.3 13.5 12.4 12.6 12.6 11.9 11.7 12.1 11.8 11.7 11.7 12.8 13.0 12.9 12.5 13.4 13.5 13.4

880 850 920 860 900 1080 1850 1770 1690 1910 1850 1680 1200 1070 940 820 780 720 800

10.12 9.82 9.79 9.27 9.34 9.16 8.91 8.72 9.28 9.05 8.99 8.81 9.72 9.91 9.95 10.03 10.14 10.02 8.39

606 612 601 611 601 585 552 530 570 598 596 587 570 600 583 599 602 589 597

4.81 4.85 4.95 4.73 4.65 4.52 4.28 4.76 4.27 4.15 4.08 4.97 4.73 4.71 5.00 4.95 4.82 4.72 4.61

Average Average Average

All Interior Exterior

5.11 5.52 4.87

12.64 11.82 13.07

1190 1790 870

9.44 8.96 9.71

589 572 600

4.66 4.42 4.80

carriers. Beryllium, Al, and Cl were chemically separated from each sample, after taking chemical analysis aliquots, and purified for AMS measurements (e.g., Nishiizumi et al., 1984a,b). The 10Be, 26Al, and 36Cl concentrations were measured by AMS at the Lawrence Livermore National Laboratory (LLNL) (Davis et al., 1990). The concentrations of major target elements, except O and Si, were determined by atomic absorption spectrometry.

measurement errors and do not reflect the errors of chemical analysis (1.5–2%) or uncertainties in the absolute activities of the standards. The 10Be, 26Al, and 36Cl depth profiles in 64455,82 are shown in Fig. 4.

3. RESULTS

The 10Be activity level (about 6 dpm/kg) is consistent with a 2 Myr exposure history for the rock. The flat profile

The chemical compositions of the samples are shown in Table 1. The measured 10Be/Be ratios ranged from 2  1012 to 9  1012, the 26Al/Al ratios from 1  1011 to 7  1011, and the 36Cl/Cl ratios from 1  1012 to 5  1012. After correcting for the 10Be background due to 10B and the chemical blank (3  1014 for 10Be/Be, 3  1015 for 26Al/Al, and 1  1014 for 36Cl/Cl), the measured ratios were normalized to ICN 10Be, NBS 26Al, and NBS 36Cl AMS standards (Sharma et al., 1990; Nishiizumi, 2003; Nishiizumi et al., 2007). In our previous work on lunar surface rock 68815 (Nishiizumi et al., 1988), we adopted a 10Be half-life of 1.6  106 yr. Recent measurements indicate a half-life of 1.36  106 yr (Nishiizumi et al., 2007). However, since both measurements were normalized to the ICN 10Be standard, which was calibrated by radioactivity measurements, the observed 10Be activities in both 68815 and 64455 can be directly compared without any correction (Nishiizumi et al., 2007). The results are shown in Table 2 along with the depths (g/cm2) of the samples calculated using a density of 2.74 g/cm3. The depths of our samples were measured relative to a plane parallel to the top surface of the rock; the deepest samples are equivalent to depths of 6 g/cm2 (2.2 cm) on the moon. The errors quoted are ±1r AMS

4. DISCUSSION 4.1. Profiles

Table 2 10 Be, 26Al, and

T-1 T-2 T-3 T-4 T-5 T-6 T-8 T-10 Mid B-10 B-9 B-8 B-7 B-6 B-5 B-4 B-3 B-2 B-1

36

Cl concentrations in 64455.

Depth (g/cm2)

10 Be (dpm/kg)

26 Al (dpm/kg)

36 Cl (dpm/kg)

0.00–0.06 0.06–0.11 0.11–0.16 0.16–0.25 0.25–0.38 0.38–0.51 0.95–1.10 1.97–2.12 1.19–3.11 4.81–5.00 5.00–5.23 5.23–5.34 5.34–5.53 5.53–5.66 5.66–5.78 5.78–5.86 5.86–5.94 5.94–6.00 6.00–6.03

5.94 ± 0.18 6.17 ± 0.22 7.04 ± 0.14 6.42 ± 0.15 6.30 ± 0.15 6.48 ± 0.18 5.88 ± 0.11 5.93 ± 0.13 5.91 ± 0.09 5.72 ± 0.29 6.10 ± 0.09 6.02 ± 0.10 6.70 ± 0.06 6.53 ± 0.05 6.51 ± 0.07 6.57 ± 0.06 6.49 ± 0.07 6.54 ± 0.21 6.61 ± 0.11

423.9 ± 8.4 379.4 ± 6.6 370.1 ± 6.2 342.3 ± 13.7 307.1 ± 8.9 275.4 ± 6.9 193.7 ± 2.9 133.2 ± 2.0 110.8 ± 3.1 74.7 ± 2.2 77.1 ± 2.4 68.6 ± 2.1 71.7 ± 2.1 70.5 ± 2.0 70.3 ± 1.7 69.5 ± 1.9 70.8 ± 1.7 70.2 ± 1.8 72.2 ± 1.5

17.71 ± 0.51 18.15 ± 0.36 19.50 ± 0.51 15.67 ± 0.27 15.61 ± 0.39 15.18 ± 0.21 13.66 ± 0.23 12.78 ± 0.20 13.20 ± 0.24 12.40 ± 0.19 12.20 ± 0.15 12.16 ± 0.19 12.71 ± 0.11 12.74 ± 0.11 13.09 ± 0.20 12.35 ± 0.20 11.90 ± 0.12 12.33 ± 0.11 12.02 ± 0.32

2168

K. Nishiizumi et al. / Geochimica et Cosmochimica Acta 73 (2009) 2163–2176

4.2. Erosion rates The SCR production rate in lunar surface rocks is defined by rigidity (R0), flux (J), erosion rate, and sample geometry. The absence of impact pits and the flat depth profiles of 26 Al and 36Cl on the bottom side of rock 64455 indicate that it was not tumbled within the last 2 Myr. Previous work indicates that there is no unique solution for R0 and J that fits SCR-produced 26Al, 53Mn, 21Ne, 22Ne, and 38Ar profiles if the erosion rate is unconstrained (Russ and Emerson, 1980; Rao et al., 1994). A significant indicator of the erosion rate is the surface of rock 64455, which is completely covered by 2 mm of smooth, glassy material. During grinding we observed that, although the top surface of the rock was damaged by micrometeorite impacts, there were visible, though small, areas indicating no erosion. These pristine areas on the upper surface are identical in appearance to the entire uneroded bottom surface. Additionally, except for a very few points, there are no individual craters deeper than 1 mm. Finally, cross-sectional photographs (e.g., Fig. 3) show that the thickness of the glassy layer on the top was equal (within 1 mm) to that of the bottom. Taken together, these observations imply that the erosion rate of 64455 is very likely less than 0.5 mm/Myr. 4.3. Theoretical calculations

Fig. 4. Measured 10Be, 26Al, and 36Cl activity depth profiles in 64455. The width of the error bars for depth indicates the depth interval sampled from the measurements during grinding. The error bars of activities indicate 1r errors of the AMS measurements.

and nearly undetectable SCR contribution to the 10Be profile are in good agreement with 10Be measurements in 68815 (Nishiizumi et al., 1988). Nearly all of the 10Be is produced by GCR, even in the uppermost surface. The near-zero contribution of SCR-produced 10Be constrains the spectral shape of the protons, expressed in rigidity, to low R0 values (Nishiizumi et al., 1988). On the other hand, 26Al and 36Cl profiles show clear SCR effects, in fact the measured activities for the surface sample (T-1) are the highest SCR-produced activities for these nuclides measured to date. After correction for the 2.0 Myr exposure age, these activities indicate that rock 64455 has a very low erosion rate, making it a nearly ideal lunar sample for the reconstruction of SCR activity.

4.3.1. The SCR model and its input Theoretical calculations for SCR-produced 10Be, 26Al, and 36Cl were made using the model of Reedy and Arnold (1972) and the new cross sections discussed below. This model calculates the fluxes as a function of energy from the slowing down and stopping of the incident particles, which is a well known process. The calculations were done for SCR spectral-shape parameters R0 = 125, 100, 90, 80, 70, 60, and 50 MV with an arbitrary flux above 10 MeV, J(>10 MeV), set at 100 protons/cm2/s4p. The activities expected in the layers were calculated for erosion rates of 0, 0.5 and 1 mm/Myr. The solar-proton flux incident on the rock was assumed to be omnidirectional from a half space (2p solid angle). The rock was assumed to be spherical with a radius of 7 g/cm2, and profiles were calculated in a series of spherical shells within this sphere. The shape of 64455 is not a perfect sphere, but a spherical approximation is adequate for the present calculation (Russ and Emerson, 1980). There are no good photographs or descriptions of the location and orientation of 64455 on the lunar surface. Heavy-ion tracks produced by cosmic ray nuclei in 64455 (Blanford et al., 1975) indicate that the rock’s soil line was tipped relative to the zenith by about 15°, which would imply that the solid angle exposed to space was a little less than 2p. Neither the slight dip in inclination nor deviations from a perfect sphere creates substantial uncertainties in our calculations. 4.3.2. Cross sections In this work, we use new excitation functions for protoninduced reactions. Earlier studies utilized cross section data that in many instances were only based on a few measurements and were supplemented with data extrapolated from

Solar cosmic ray records in lunar rock 64455

needed to compile and evaluate cross sections to many GeV. As part of this work we have compiled all the relevant cross sections available in the literature, Ph. D. theses, and nuclear databases. We also queried the international cross section compilation called Cross Section Information Storage and Retrieval System (CSISRS) at the National Nuclear Data Center in the Brookhaven National Laboratory. Michel et al. (1997) and Sisterson et al. (1997a) have filed their many measurements with CSISRS, occasionally with updates or corrections to their data. The new cross sections for 10Be production are similar to those used since 1990, however the post-1990 cross sections differ from those used in earlier work (e.g., Reedy and Ar-

analogous reactions (Reedy and Arnold, 1972). During the last two decades, many cross sections have been measured for the reactions of interest (e.g., Michel et al., 1997; Sisterson et al., 1997a,b); the present work uses these new measurements. Proton fluxes (J) derived from measured SCRproduced profiles in lunar samples are sensitive to the cross sections of the reactions that produce them in the target material (e.g., Sisterson et al., 1997c) so these new cross section data are a significant advance in our ability to model SCR production. Our SCR calculations require cross sections for producing 10Be, 26Al, and 36Cl by proton reactions from threshold energies to several hundred MeV. For GCR reactions, we

O(p,x)

10

26

Al(p,x) Al

Be

0

2

10

10

10

Cross section (mb)

Cross section (mb)

10

10

2169

Adopted (This work) Bo93 Di90b Mi95 Mi97 Sc96 Th86 Si97a Ki97 Am72 Li75 Ra77 We00 Yi69

-1

-2

-3

2

3

10

Adopted (This work) Mi97 Di90 Mi95 Ev71 Fu71 He75

0

10 1 10

4

10

1

10

10

2

10

10

3

Ki97 Sh93 Pa77 We90 Sc87 Si97b 4

10

E (MeV)

E (MeV)

Ca(p,x)

36

26

Cl

Si(p,x) Al

2

10

1

Cross section (mb)

Cross section (mb)

10

0

10

10

10

Adopted (This work) Im97 Mi97 Si97c Ba84 We90 Ch97

-1

1

10

-2 1

10

2

10

E (MeV)

10

3

10

4

0

10 1 10

Adopted (This work) Mi97 Fu71 Ki97 He75

OR73

Pa77

Mi89

Ra77b Ra79 Re73 De79

We90 Ar75 Si97a Bo93b

Mi95 Sc96 Th86

2

10

10

3

10

4

E (MeV)

Fig. 5. Compilation of experimentally determined cross sections for the indicated nuclear reactions. These are the principal reactions for the production of the nuclides 10Be, 26Al, and 36Cl in rock 64455 by solar-proton bombardment. The solid lines are the adopted excitation functions for each nuclear reaction. References are Am72 (Amin et al., 1972); Ar75 (Artun et al., 1975); Ba84 (Baros and Regnier, 1984); Bo93 (Bodemann et al., 1993a); Bo93b (Bodemann et al., 1993b); Ch97 (Chen et al., 1997); De79 (Dedieu, 1979); Di90 (Dittrich et al., 1990b); Di90b (Dittrich et al., 1990a); Ev71 (Evans, 1971); Fu71 (Furukawa et al., 1971); He75 (Reedy, 1987); Im97 (Imamura et al., 1997); Ki97 (Kim, 1997); Mi89 (Michel et al., 1989); Li75 (Lindstrom et al., 1975); Mi95 (Michel et al., 1995); Mi97 (Michel et al., 1997); Or73 (Reedy, 1987); Pa77 (Paillard, 1977); Ra77 (Raisbeck and Yiou, 1977); Ra77b (Raisbeck et al., 1977); Ra79 (Raisbeck et al., 1979); Re73 (Regnier et al., 1973); Sc87 (Schneider et al., 1987); Sc96 (Schiekel et al., 1996); Sh93 (Shibata et al., 1993); Si97a (Sisterson et al., 1997a); Si97b (Sisterson et al., 1997b); Si97c (Sisterson et al., 1997c); Th86 (Theis et al., 1986); We90 (Webber et al., 1990); Yi69 (Yiou et al., 1969).

K. Nishiizumi et al. / Geochimica et Cosmochimica Acta 73 (2009) 2163–2176

4.3.3. Calculated solar-proton production profiles Model predictions for SCR-produced 10Be, 26Al, and 36 Cl are shown in Fig. 7 for a range of spectral shapes given an arbitrary omnidirectional (4p steradians) incident flux of 100 protons/cm2/s. The solid lines show the case of no erosion and the dashed lines show profiles for a 1 mm/Myr erosion rate. Profiles for a 1 mm/Myr erosion rate are shown only for the case of R0 = 100 MV. Our calculations indicate that a few elements dominate the production of these radionuclides in 64455. For 10Be, 94% is made from O, 3% from Al, and 1.5% each from Mg and Si. For 26 Al, 67% is made from Al and 31% from Si. For 36Cl, 95% is from Ca and 4% from K. The steep SCR production profile of 26Al, which has a low threshold energy for its production of about 15 MeV, is sensitive to any changes in the total flux or erosion rate and therefore constrains the erosion rate and J. A different rigidity changes the slope or shape of the SCR 26Al production profile: as is also the case for other low-energy products, such as 53Mn. For a given fluence, the production rates of SCR-produced 36Cl and 10Be are especially sensitive to the rigidity of the spectrum, but the shapes of the profiles do not change as much.

HET Code System (LCS) (Masarik and Reedy, 1994), again using updated excitation functions. The effective incident flux of GCR protons used for these calculations is known for several radionuclides measured down to about 400 g/

0

Cross section (mb)

10

10

Be

-1

10

10

C(p,x) Be 10

O(p,x) Be 10

Mg(p,x) Be

-2

10

10

Al(p,x) Be 10

Si(p,x) Be 10

Fe(p,x) Be -3

10

10

1

10

2

3

10

10

4

2

10

26

Al

1

10

Cross section (mb)

nold, 1972; Tuniz et al., 1984). The newest cross sections for 26 Al production differ slightly from those used by Reedy (1987) and are lower at energies above 400 MeV compared to Reedy and Arnold (1972). The newer cross sections for 36Cl were measured by Imamura et al. (1997), Michel et al. (1997), Sisterson et al. (1997c) and Fink et al. (2000). There were only a few earlier measurements at higher energies for 36Cl (e.g., Baros and Regnier, 1984), some of which are only in Ph. D. theses. The newest cross sections for making 36Cl from Ca for energies below about 200 MeV differ from those incorporated in previous compilations and used in earlier studies of lunar rocks. The compiled cross sections for each radionuclide were plotted as a function of energy for each major target element. In general, measurements from different research groups are in good agreement. A noteworthy exception are the cross sections for the production of 36Cl from Ca, where the measurements of Michel et al. (1997) near 64 and 69 MeV do not follow the trend measured by others. Accordingly, we disregarded these two values. Fig. 5 shows all the compiled cross section measurements and the smoothed curve that represents the best fit to the data for each major reaction: O for 10Be, Al and Si for 26Al, and Ca for 36Cl. For our calculations we use those curves. While it is hard to quantify the uncertainties in these cross sections, the agreement among most measurements suggests an uncertainty of 620%. The excitation functions for all target elements are shown in Fig. 6.

0

10

26

Mg(p,n) Al 26

Al(p,pn) Al

-1

10

26

Si(p,x) Al 26

Ca(p,x) Al

-2

10

26

Ti(p,x) Al 26

Fe(p,x) Al

-3

10

1

2

10

10

10

3

4

10

1

10

Cross section (mb)

2170

36

Cl

0

10

36

K(p,x) Cl 36

-1

Ca(p,x) Cl

10

36

Ti(p,x) Cl 36

Fe(p,x) Cl

4.3.4. Calculated GCR production profiles Because the observed profiles are a summation of SCR and GCR components, the GCR component must be subtracted from the measurements to compare experimental SCR profiles with model predictions. To calculate the amount of GCR-produced radionuclides we used the LA-

36

Ni(p,x) Cl -2

10

1

10

2

10

10

3

10

4

E (MeV) Fig. 6. Adopted excitation functions for the proton induced production of 10Be, 26Al, and 36Cl from each of the important target elements.

Solar cosmic ray records in lunar rock 64455

energy, the product of these fluxes and the cross sections for the appropriate nuclear reactions (Reedy et al., 1993; Reedy and Masarik, 1994b). For the cross sections, we relied on the values compiled by us, tested by earlier calculations (Reedy et al., 1993; Masarik and Reedy, 1994), and updated with new values from recent experiments (e.g., Michel et al., 1997). Neutrons dominate GCR production, and the cross sections used for GCR reactions differ from those of solar protons. These calculated GCR rates are estimated to have uncertainties of 20%. The calculated shapes of the GCR profiles for rate versus depth are much better determined than the absolute production rates.

2.5 R =100 MV 0 R =100 MV, ER=1 mm/Myr

10

Activity (dpm/kg)

Be

0

R =90 MV

2.0

0

R =80 MV 0 R =70 MV 0 R =60 MV

1.5

0

R =50 MV 0

1.0

0.5

0.0

800

4.4. Solar-proton fluxes and spectral shapes

R =100 MV

26

0

Al

700

R =100 MV, ER=1 mm/Myr 0

The measured activities as a function of depth are compared with the theoretical calculations to determine the flux of solar protons incident on lunar rock 64455. There are four parameters that vary in these comparisons; the erosion rate, the normalization used for the GCR production rates, the solar-protons’ spectral shape (R0), and intensity (J, expressed as the omnidirectional proton flux >10 MeV). The ranges of the erosion rates and GCR normalizations are constrained based on other observations, such as the nature of the surface indicating low erosion and previous results for GCR production in lunar samples, one example being the Apollo 15 deep drill core (Reedy and Masarik, 1994b).

Activity (dpm/kg)

R =90 MV 0

600

R =80 MV 0 R =70 MV 0 R =60 MV 0 R =50 MV

500

0

400 300 200 100 0

12 R =100 MV

36

Activity (dpm/kg)

0

Cl

10

R =100 MV, ER=1 mm/Myr 0 R =90 MV 0

R =80 MV 0 R =70 MV

8

0

R =60 MV 0 R =50 MV 0

6 4 2 0

0

1

2

2171

3

Depth (g/cm2)

4

5

6

Fig. 7. Theoretical SCR production profiles for 10Be, 26Al, and 36 Cl calculated using the average 64455 chemical composition and a 2.0 Myr exposure age. The depth profiles were calculated using the Reedy–Arnold model (Reedy and Arnold, 1972) and the new excitation functions (Fig. 6). The lines show SCR production rates (atom/min/kg) of each nuclide for seven sets of SCR parameters (R0 = 100, 90, 80, 70, 60, and 50 MV) with J = 100 protons/cm2/ s4p and no erosion. The SCR production rate for each nuclide with a 1 mm/Myr erosion rate for R0 = 100 MV is also shown.

cm2 depth in the Apollo 15 deep drill core (Masarik and Reedy, 1994). The target for the GCR calculations was assumed to be a sphere with the radius of the Moon. At the surface, the sphere was divided into spherical layers with thicknesses of 0.36 g/cm2 to account for variation of the particle fluxes with depth. The elemental composition and density were those of lunar rock 64455. The statistical errors of the calculated fluxes are 2–3%. Once the particle fluxes were determined with LCS, the production rates of cosmogenic nuclides were calculated by integrating, over

4.4.1. GCR production The activities produced by the GCR are important and must be carefully determined. While our theoretical GCR production rates tend to be in good agreement with measurements, such as in the Apollo 15 deep drill core (Reedy and Masarik, 1994b), some adjustments need to be made. GCR production rates were calculated for saturation, but in 64455 the activities for 26Al and 10Be are undersaturated due to the rock’s short exposure. For 64455’s nominal exposure age of 2 Myr, the corrections to the saturation production rates are 99% for 36Cl, 86% for 26Al, and 64% for 10Be. In many lunar rocks and cores used for SCR studies, there are samples below the depths reached by solar protons that can be used to experimentally normalize the GCR calculations. For lunar rock 64455, the deepest samples are close enough to the surface (6 g/cm2 or about 2 cm) that they contain some production by solar protons. Thus the SCR production in these deeper samples needs to be determined from the fits near the surface and then used to establish the correct GCR normalization. The better fits, presented below, give GCR normalizations of 0.80 for 36Cl, 0.83 for 26Al, and 0.60 for 10Be. These normalizations for 26Al and 10Be are in good agreement with the expected values, and both values would imply GCR exposure ages of 1.8 Myr. The disagreement with the expected value of 0.99 for 36Cl is a little more than expected, but within the range of our estimated GCR uncertainties. 4.4.2. Fitting the SCR production The values for the SCR flux (J) and spectral shape (R0) are best determined from the samples with significant SCR

K. Nishiizumi et al. / Geochimica et Cosmochimica Acta 73 (2009) 2163–2176 8.0 10

7.0

Be

5.0 4.0

10

Obs. Be Calc. SCR+GCR SCR GCR

3.0

10

Be (dpm/kg)

6.0

2.0 1.0 0.0 26

Al 26

Obs. Al Calc. SCR+GCR SCR GCR

300

200

26

Al (dpm/kg)

400

100

0 25 36

20

Cl (dpm/kg)

production, which are those within about 2 g/cm2 of the surface. Eight samples in this depth range allow the fit to be fairly well constrained. The best fits produce the least variation for J and R0 among these top samples while simultaneously not producing monotonic production rate variations as a function of the sample’s depth. As noted in previous studies, variations in R0 of 10 MV do not greatly change the quality of the fits, although these variations do cause changes in the values for J(>10 MeV), with lower values of J(>10 MeV) corresponding to higher values of R0. While a range of values for R0 and J(>10 MeV) produces good fits, in most cases the particle fluxes above 30 and 60 MeV for these fits are less variable. Protons with these energies produce nuclides below the topmost surface. These higher energies are also closer to the reaction threshold energies for most SCR-produced radionuclides. The effective reaction threshold energies for the three radionuclides in this study are about 15 MeV for 26Al, 40 MeV for 10 Be, and 20 MeV for 36Cl (Fig. 6). Thus, protons with energies of about 10 MeV do not produce any of these three radionuclides. The measured profiles and the calculated production rates as a function of depth for SCR, GCR, and total production are shown in Fig. 8 for the best fits. Because the SCR production of 10Be was low, it is not possible to get a good fit to the measurements. For example, even a hard spectrum (R0 = 125 MV) could be made to fit the data if the GCR normalization was lowered to 0.50 instead of 0.60. To constrain the 10Be profile it would be necessary to have deeper samples than possible with 64455. Thus, the profile for 10Be was not fit independently, and its SCR values in Fig. 8 were calculated with the SCR parameters from the 26Al fit.

Cl 36

Obs. Cl Calc. SCR+GCR SCR GCR

15

10

36

2172

5

4.4.3. Solar-proton fluxes during the last 2 Myr determined from 26Al The best fits for 26Al have R0 values of 80–90 MV and erosion rates of about 0.5 mm/Myr; increasing the erosion rate results in a slightly higher flux. None of the fits were good for all surface samples, and the measured activities in the top three samples (0–0.16 g/cm2) were lower than the calculated values, even for an erosion rate of 1 mm/ Myr. Increasing the hardness (R0) reduced the disagreement near the surface but increased the disagreement for deeper samples. A solar-proton spectrum with a value for R0 of 85 MV but fewer protons below 100 MeV did not work any better than a pure R0 = 90 MV spectrum. Adopting R0 = 90 MV and an erosion rate of 0.5 mm/ Myr as the best fit, the omnidirectional (4p) flux above 10 MeV was 73 protons/cm2/s. These results are slightly different from our preliminary result using older cross sections (Nishiizumi et al., 1995) in which we calculated R0 = 75 MV and J(>10) = 100 protons/cm2/s. The present results are compared with those from other studies for the 1–5 Myr time period in Table 3. As noted in Reedy (1998), the fluxes above 30 and 60 MeV show smaller variations than the fluxes above 10 MeV. The averages of the previous work are R0  90 MV and J(>10 MeV) = 80 protons/cm2/s for the 1–5 Myr time period, similar to the re-

0

0

1

2

3

4

5

6

Depth (g/cm2) Fig. 8. Calculated best-fit production rates compared to data for 10 Be, 26Al, and 36Cl in 64455. The dotted line is the GCR contribution to the production for each nuclide, the dashed line is the SCR production and the solid line represents the combined total production. SCR parameters are R0 = 90 MV and J = 73 protons/cm2/s4p for both 10Be and 26Al and R0 = 70 MV and J = 196 protons/cm2/s4p for 36Cl. An erosion rate of 0.5 mm/ Myr was used for both 10Be and 26Al. For 36Cl the best fit is with an erosion rate 0.5 mm/Myr for half the surface and no erosion for the remainder.

sult presented here. The average proton fluxes >30 and >60 MeV during the last 1 Myr and longer periods also agree fairly well with previous work. 4.4.4. Solar-proton fluxes during the last 0.5 Myr determined from 36Cl The better fits for 36Cl had R0 values of 70 MV; values of R0 of about 80 MV also produce good fits, although the scatter in the measured activities in the top four samples makes it hard to fit all data. This variability in the top four

Solar cosmic ray records in lunar rock 64455

2173

Table 3 Average omnidirectional integral fluxes (in protons/cm2/s) of solar protons above three energies for the last five solar cycles (top) and for various time periods in the past (bottom) and approximate exponential-rigidity spectral shapes (R0). The lunar samples used and the adopted erosion rates are shown for results using long-lived radionuclides. For 1–5 Myr, the fluxes >30 and >60 MeV show less scatter than the fluxes >10 MeV and the values of R0. The average fluxes >30 MeV and >60 MeV are higher for the more recent time periods. Period

Data

Sample

1996–2006 1986–1996 1976–1986 1965–1975 1954–1964 1954–2006 10 kyr 0.2 Myr 0.3 Myr 0.5 Myr 1 Myr 1 Myr 1 Myr 1 Myr 1 Myr 2 Myr 5 Myr a

Erosion, mm/Myr

12002

14

C 41 Ca 81 Kr

68815 74275 68815

36

64455 68815 68815 Several 74275 64455 68815 68815

Cl Al —a —a —a —a 21 Ne 53 Mn 26

Using both 1.36 Myr

0.25 2.2 3.0 Several 0.5

10

Be and 0.705 Myr

26

Reference

R0 (MV)

>10 MeV

>30 MeV

>60 MeV

Reedy (2006, updated) Reedy (1998) Goswami et al. (1988) Reedy (1977) Reedy (1977) and Sisterson et al. (1996) Average of above Jull et al. (1998) Fink et al. (1998) Reedy and Marti (1991) and Reedy et al. (1999) This work Kohl et al. (1978) Nishiizumi et al. (1988) Michel et al. (1996) Fink et al. (1998) This work Rao et al. (1994) Kohl et al. (1978)

64 65 40 90 100

278 152 63 92 227

61 31 5 30 82

10 — 1 8 35

80 113 80 80

150 103 198 160

42 42 56 48

12 17 16 15

70 100 70 125 100 90 85 100

196 70 150 55 89 73 68 70

46 25 35 24 32 24 21 25

11 9 8 11 12 8 6 9

Al.

samples is most likely the result of erosion. The fits indicate that this sample had an erosion rate of between 0 and 0.5 mm/Myr. We adopt a best-fit having R0 = 70 MV, and an erosion rate of 0 for 50% of the surface and 0.5 mm/Myr for the other half. The actual erosion history may be more complicated than this simple case. As noted above, some regions appear to show no erosion while others clearly show the removal of surface material. This fit yields an omnidirectional (4p) flux above 10 MeV of 196 protons/cm2/s, a softer spectrum than those previously proposed for 15008 and 74275 (Reedy and Nishiizumi, 1998), but in good agreement with preliminary fits to 36Cl in 64455 and other samples (Reedy, 1998) and with 81Kr in 68815 (Reedy et al., 1999). Table 3 shows a comparison of our 36Cl measurements with previous results for exposure during the last 10 kyr to 0.5 Myr. Again, there is some scatter in the values for R0 and fluxes >10 MeV, but the fluxes >30 and >60 MeV show less variation. These fluxes for 10 kyr to 0.5 Myr for >30 and >60 MeV are significantly higher than those for the time period 1–5 Myr, but are similar to the average fluxes measured since 1954, in agreement with results in (Reedy, 1998). 5. CONCLUSIONS Based on this study we conclude that (1) Rock 64455 provides the best depth profiles for 10Be, 26 Al, and 36Cl measured to date. The erosion rate for this rock was likely less than 0.5 mm/Myr, which reduces the possible range for this important parameter. Erosion rates in other lunar rocks are higher, possibly more than 2 mm/Myr (e.g., Kohl et al., 1978).

(2) The SCR-produced 26Al depth profile yields an exponential-rigidity spectral shape (R0) of 90 MV and an omnidirectional flux (J) above 10 MeV of 73 protons/cm2/s for an erosion rate of 0.5 mm/Myr. All possible fits are consistent with the very low amount of SCR-produced 10Be. (3) The SCR-produced 36Cl depth profile is best fit using our newly-evaluated 36Cl cross sections with an R0 of 70 MV, erosion rates of 0 mm/Myr for half the exposed surface and 0.5 mm/Myr for the remainder, and a J(>10 MeV) of 196 proton/cm2/s. (4) Our results confirm the observation of Reedy (1998) that the fluxes above 30 and 60 MeV have less variation than R0 and fluxes above 10 MeV and that average solar-proton fluxes over the last 1–5 Myr are less than those for 0.01–0.5 Myr and for 1954–2006. (5) The results in Table 3 for several radionuclides plus modern solar proton measurements have similar integral fluxes (especially for >30 and >60 MeV) from modern times to 0.5 Myr ago. There have been many samples and nuclides used for the 1–5 Myr time period, and all give similar results for >30 and >60 MeV integral fluxes over this time period. (6) Our results and the average fluxes for comparable time periods (Table 3) suggest a change in the fluxes above 30 and 60 MeV (the energies that are best determined from such SCR profiles) at 0.5 Myr ago. The fluxes above 30 MeV for the last 0.5 Myr and last 2 Myr are about 45 and 25 protons/cm2/s, respectively. As about a quarter of the 26 Al in 64455 was made in the most recent 0.5 Myr with an average flux of about 45 protons/ cm2/s, the flux from 0.5–2 Myr must have been 18 protons/cm2/s, or 2.5 lower than during the

2174

K. Nishiizumi et al. / Geochimica et Cosmochimica Acta 73 (2009) 2163–2176

most recent 0.5 Myr. Such a large change in the average flux of solar protons is not expected from records of GCR-produced radionuclides in the Moon (e.g., Reedy and Masarik, 1994a) or from terrestrial records. The records for 1–2 Myr are based on many groups and samples and are fairly consistent. Additional work for the average fluxes over last 0.5 Myr, which are not as well determined, is needed to confirm any change in SCR fluxes during this period. (7) We now have reliable experimental SCR profiles and model calculational methods with good cross sections at all energies for all target nuclides for 10Be, 26Al, and 36Cl.

ACKNOWLEDGMENTS We wish to thank F. Ho¨rz for suggesting 64455 for our study and for valuable discussions. Technical support was furnished by D. Matriano for part of the sample preparation and M. Keogh for setup of the grinding apparatus. We also thank I. Leya, M.N. Rao, and an anonymous reviewer for their valuable comments. This work was supported by NASA grants NAG 9-33, NAGW 3514, NAG512846, NNG06GF22G, and NNG06GF66G, Slovak grant agency grant APVV-0569-07, and was performed under the auspices of the U.S. D.O.E. by LLNL under contract W-7405-ENG-48.

REFERENCES Amin B. S., Biswas S., Lal D. and Somayajulu B. L. K. (1972) Radiochemical measurements of 10Be and 7Be formation cross sections in oxygen by 135 and 550 MeV protons. Nucl. Phys. A 195, 311–320. Arnold J. R., Kohl C. P., Nishiizumi K., Caffee M. W., Finkel R. C. and Southon J. R. (1993) Measurements of cosmogenic nuclides in lunar rock 64455. Lunar Planet. Sci. XXIV. Lunar Planet. Inst., Houston. pp. 39–40 (abstr.). Artun O., Cassagnou Y., Legrain R., Lisbona N., Roussel L., Alard J. P., Baldit A., Costilhes P., Fargeix J., Roche G. and Tamain J. C. (1975) Multinucleon removal induced by highenergy protons. Phys. Rev. Lett. 35, 773–775. Baros F. and Regnier S. (1984) Measurement of cross sections for 22 Na, 20–22Ne and 36–42Ar in the spallation of Mg, Al, Si, Ca, and Fe. Production ratios of some cosmogenic nuclides in meteorites. J. Phys. 45, 855–861. Begemann F., Born W., Palme H., Vilcsek E. and Wa¨nke H. (1972) Cosmic-ray produced radioisotopes in Apollo 12 and Apollo 14 samples. Proc. Lunar Sci. Conf. 3, 1693–1702. Blanford G. E., Fruland R. M., McKay D. S. and Morrison D. A. (1974) Lunar surface phenomena: solar flare track gradients, microcraters, and accretionary particles. Proc. Lunar Sci. Conf. 5, 2501–2526. Blanford G. E., Fruland R. M. and Morrison D. A. (1975) Longterm differential energy spectrum for solar-flare iron-group particles. Proc. Lunar Sci. Conf. 6, 3557–3576. Bodemann R., Lange H.-J., Leya I., Michel R., Schiekel T., Ro¨sel R., Herpers U., Hofmann H. J., Dittrich B., Suter M., Wo¨lfli W., Holmqvist B., Conde´ H. and Malmborg P. (1993a) Production of residual nuclei by proton-induced reactions on C, N, O, Mg, Al and Si. Nucl. Instrum. Methods Phys. Res. B82, 9–31. Bodemann R., Lange H.-J., Leya I., Michel R., Schiekel T., Ro¨sel R., Herpers U., Hofmann H. J., Dittrich B., Suter M., Wo¨lfli

W., Holmqvist B., Conde´ H. and Malmborg P. (1993b) Production of residual nuclei by proton-induced reactions on C, N, O, Mg, Al and Si. Fed. Rep. Germany report to the I.N.D.C No. 037/LN, 49. Bogard D. D. and Gibson E. K. (1975) Volatile gases in breccia 68115. Lunar Sci. VI, 63–65 (abstr.). Bogard D. D. and Hirsch W. C. (1975) Noble gas studies on grain size separates of Apollo 15 and 16 deep drill cores. Proc. Lunar Sci. Conf. 6, 2057–2083. Chen C.-X., Albergo S., Caccia Z., Costa S., Crawford H. J., Cronqvist M., Engelage J., Greiner L., Guzik T. G., Insolia A., Knott C. N., Lindstrom P. J., McMahon M., Mitchell J. W., Potenza R., Russo G. V., Soutoul A., Testard O., Tull C. E., Tuve´ C., Waddington C. J., Webber W. R. and Wefel J. P. (1997) Systematics of isotopic production cross sections from interactions of relativistic 40Ca in hydrogen. Phys. Rev. C 56, 1536–1543. Davis J. C., Proctor I. D., Southon J. R., Caffee M. W., Heikkinen D. W., Roberts M. L., Moore T. L., Turteltaub K. W., Nelson D. E., Loyd D. H. and Vogel J. S. (1990) LLNL/UC AMS facility and research program. Nucl. Instrum. Methods Phys. Res. B52, 269–272. Dedieu J. (1979) Mesures de sections efficaces de production d’isotopes de longue pe´riode et leur inte´reˆt astrophysique. Ph. D. thesis, University of Bordeaux, No. 1525. Dittrich B., Herpers U., Hofmann H. J., Wo¨lfli W., Bodemann R., Lu¨pke M., Michel R., Dragovitsch P. and Filges D. (1990a) AMS measurements of thin-target cross sections for the production of 10Be and 26Al by high energy protons. Nucl. Instrum. Methods Phys. Res. B52, 588–594. Dittrich B., Herpers U., Lu¨pke M., Michel R., Hofmann H. J. and Wo¨lfli W. (1990b) Determination of cross sections for the production of Be-7, Be-10 and Na-22 by high-energy protons. Radiochim. Acta 50, 11–18. Evans, J. C. (1971) A study of solar flare history by measurement of 22 Na and 26Al production profiles in lunar rocks. Ph. D. thesis, University of California, San Diego. Fink D., Klein J., Middleton R., Vogt S., Herzog G. F. and Reedy R. C. (1998) 41Ca, 26Al, and 10Be in lunar basalt 74275 and 10Be in the double drive tube 74002/74001. Geochim. Cosmochim. Acta 62, 2389–2402. Fink D., Vogt S. and Hotchkis M. (2000) Cross-sections for 36Cl from Ti at Ep = 35–150 MeV: applications to in-situ exposure dating. Nucl. Instrum. Methods Phys. Res. B172, 861–866. Finkel R. C., Arnold J. R., Imamura M., Reedy R. C., Fruchter J. S., Loosli H. H., Evans J. C., Delany A. C. and Shedlovsky J. P. (1971) Depth variation of cosmogenic nuclides in a lunar surface rock and lunar soil. Proc. Lunar Sci. Conf. 2, 1773– 1789. Fruchter J. S., Evans J. C., Reeves J. H. and Perkins R. W. (1982) Measurement of 26Al in Apollo 15 core 15008 and 22Na in Apollo 17 rock 74275. Lunar Planet. Sci. XIII. Lunar Planet. Inst., Houston. pp. 243–244 (abstr.). Furukawa M., Shizuri K., Komura K., Sakamoto K. and Tanaka S. (1971) Production of 26Al and 22Na from proton bombardment of Si, Al, and Mg. Nucl. Phys. A 174, 539–544. Garrison D. H., Bogard D. D., Rao M. N. and Reedy R. C. (1996) Determination of 2 Ma average flux and rigidity for energetic solar protons from lunar rock 61016. Lunar Planet. Sci. XXVII. Lunar Planet. Inst., Houston. pp. 395–396 (abstr.). Goswami J. N., McGuire R. E., Reedy R. C., Lal D. and Jha R. (1988) Solar flare protons and alpha particles during the last three solar cycles. J. Geophys. Res. 93, 7195–7205. Grieve R. A. F. and Plant A. G. (1973) Partial melting on the lunar surface, as observed in glass coated Apollo 16 samples. Proc. Lunar Sci. Conf. 4, 667–679.

Solar cosmic ray records in lunar rock 64455 Herr W., Herpers U. and Woelfle R. (1971) Spallogenic 53Mn (T2  106 y) in lunar surface material by neutron activation. Proc. Lunar Sci. Conf. 2, 1797–1802. Imamura M., Nishiizumi K., Honda M., Finkel R. C., Arnold J. R. and Kohl C. P. (1974) Depth profiles of 53Mn in lunar rocks and soils. Proc. Lunar Sci. Conf. 5, 2093–2103. Imamura M., Nishiizumi K., Caffee M. W. and Shibata S. (1997) Measurements of proton-induced production cross sections for 36 Cl from Ca and K. Nucl. Instrum. Methods Phys. Res. B123, 330–333. Jull A. J. T., Cloudt S., Donahue D. J., Sisterson J. M., Reedy R. C. and Masarik J. (1998) 14C depth profiles in Apollo 15 and 17 cores and lunar rock 68815. Geochim. Cosmochim. Acta 62, 3025–3036. Kim K. J. (1997) Cross sections of 10Be and 26Al by 30 to 500 MeV protons on thin targets of C, Mg, Al, Si, and SiO2. M. S. thesis, San Jose State University. Kohl C. P., Murrell M. T., Russ, III, G. P. and Arnold J. R. (1978) Evidence for the constancy of the solar cosmic ray flux over the past ten million years: 53Mn and 26Al measurements. Proc. Lunar Planet. Sci. Conf. 9, 2299–2310. Langevin Y., Arnold J. R. and Nishiizumi K. (1982) Transport processes on the lunar surface: comparison of model calculations with radionuclides data. J. Geophys. Res. 87, 6681–6691. Lindstrom P. J., Greiner D. E., Heckmann H. H., Cork B. and Bieser F. S. (1975) Isotope production cross sections from the Fragmentation of 16O and 12C at relativistic energies. Lawrence Berkeley Laboratory, LBL-3650, 10. Masarik J. and Reedy R. C. (1994) Effects of bulk composition on nuclide production processes in meteorites. Geochim. Cosmochim. Acta 58, 5307–5317. Michel R., Dittrich B., Herpers U., Peiffer F., Schiffmann T., Cloth P., Dragovitsch P. and Filges D. (1989) Proton-induced spallation at 600 MeV. Analyst 114, 287–293. Michel R., Gloris M., Lange H.-J., Leya I., Lu¨pke M., Herpers U., Dittrich-Hannen B., Ro¨sel R., Schiekel T., Filges D., Dragovitsch P., Suter M., Hofmann H.-J., Wo¨lfli W., Kubik P. W., Baur H. and Wieler R. (1995) Nuclide production by protoninduced reactions on elements (6 6 Z 6 29) in the energy range from 800 to 2600 MeV. Nucl. Instrum. Methods Phys. Res. B103, 183–222. Michel R., Leya I. and Borges L. (1996) Production of cosmogenic nuclides in meteoroids: accelerator experiments and model calculations to decipher the cosmic ray record in extraterrestrial matter. Nucl. Instrum. Methods Phys. Res. B113, 434–444. Michel R., Bodemann R., Busemann H., Daunke R., Gloris M., Lange H.-J., Klug B., Krins A., Leya I., Lu¨pke M., Neumann S., Reinhardt H., Schnatz-Bu¨ttgen M., Herpers U., Schiekel T., Sudbrock F., Holmqvist B., Conde´ H., Malmborg P., Suter M., Dittrich-Hannen B., Kubik P.-W., Synal H.-A. and Filges D. (1997) Cross sections for the productions of residual nuclides by low- and medium-energy protons from the target elements C, N, O, Mg, Al, Si, Ca, Ti, V, Mn, Fe, Co, Ni, Cu, Sr, Y, Zr, Nb, Ba, and Au. Nucl. Instrum. Methods Phys. Res. B129, 153–193. Morris R. V., See T. H. and Ho¨rz F. (1986) Composition of the Cayley formation at Apollo 16 as inferred from impact melt splashes. Proc. Lunar Planet. Sci. Conf. 17, J. Geophys. Res. 91, E21–E42. Nishiizumi K., Elmore D., Ma X. Z. and Arnold J. R. (1984a) 10Be and 36Cl depth profiles in an Apollo 15 drill core. Earth Planet. Sci. Lett. 70, 157–163. Nishiizumi K., Klein J., Middleton R. and Arnold J. R. (1984b) 26 Al depth profile in Apollo 15 drill core. Earth Planet. Sci. Lett. 70, 164–168. Nishiizumi K., Imamura M., Kohl C. P., Nagai H., Kobayashi K., Yoshida K., Yamashita H., Reedy R.

2175

C., Honda M. and Arnold J. R. (1988) 10Be profiles in lunar surface rock 68815. Proc. Lunar Planet. Sci. Conf. 18, 79–85. Nishiizumi K., Arnold J. R., Sharma P., Kubik P. W. and Reedy R. C. (1991) Cosmogenic 36Cl production on the lunar surface. Lunar Planet. Sci. XXII. Lunar Planet. Inst., Houston. pp. 979– 980 (abstr.). Nishiizumi K., Kohl C. P., Arnold J. R., Finkel R. C., Caffee M. W., Masarik J. and Reedy R. C. (1995) Final results of cosmogenic nuclides in lunar rock 64455. Lunar Planet. Sci. XXVI. Lunar Planet. Inst., Houston. pp. 1055–1056 (abstr.). Nishiizumi K. (2003) Preparation of 26Al AMS standards. Nucl. Instrum. Methods Phys. Res. B223–B224, 388–392. Nishiizumi K., Imamura M., Caffee M. W., Southon J. R., Finkel R. C. and McAninch J. (2007) Absolute calibration of 10Be AMS standards. Nucl. Instrum. Methods Phys. Res. B258, 403–413. Norris T. L., Gancarz A. J., Rokop D. J. and Thomas K. W. (1983) Half-life of 26Al. Proc. Lunar Planet. Sci. Conf. 14, J. Geophys. Res. 88, B331–B333. Paillard P. (1977) Contribution a` l’e´tude expe´rimentale de la production de 26Al par spallation des noyaux Fe, Si, Al. Ph. D. thesis, University of Bordeaux. Raisbeck G. M., Menninga C., Brodzinski R. and Wogman N. (1977) Cross sections for the production of 26Al from targets of Si, Al, and Fe irradiated by protons of 600 MeV. In Proc. 15th Int. Cosmic Ray Conf. 2, (Plovdiv). pp. 116–119. Raisbeck G. M. and Yiou F. (1977) The Be-10 problem revisited. In Proc. 15th Int. Cosmic Ray Conf. 2, (Plovdiv). pp. 203–207. Raisbeck G. M., Lestringuez J., Salau¨n S., Vlieks A., Bourle`s D. and Yiou F. (1979) Some nuclear cross section measurements relevant to cosmic ray propagation. In Proc. 16th Int. Cosmic Ray Conf 2, (Kyoto) (OG 9.3-8). pp. 207–211. Rao M. N., Garrison D. H., Bogard D. D. and Reedy R. C. (1993) Solar-flare implanted 4He/3He and solar proton produced Ne and Ar concentration profiles preserved in lunar rock 61016. J. Geophys. Res. 98, 7827–7835. Rao M. N., Garrison D. H., Bogard D. D. and Reedy R. C. (1994) Determination of the flux and energy distribution of energetic solar protons in the past 2 Myr using lunar rock 68815. Geochim. Cosmochim. Acta 58, 4231–4245. Reedy R. C. and Arnold J. R. (1972) Interaction of solar and galactic cosmic-ray particles with the moon. J. Geophys. Res. 77, 537–555. Reedy R. C. (1977) Solar proton fluxes since 1956. Proc. Lunar Sci. Conf. 8, 825–839. Reedy R. C., Arnold J. R. and Lal D. (1983) Cosmic-ray record in solar system matter. Annu. Rev. Nucl. Part Sci. 33, 505–537. Reedy R. C. (1987) Predicting the production rates of cosmogenic nuclides in extraterrestrial matter. Nucl. Instrum. Methods Phys. Res. B29, 251–261. Reedy R. C. and Marti K. (1991) Solar-cosmic-ray fluxes during the last ten million years. In The Sun in Time (eds. C. Sonett, M. Giampapa and M. Matthews). University of Arizona Press, Tucson, pp. 260–287. Reedy R. C., Masarik J., Nishiizumi K., Arnold J. R., Finkel R. C., Caffee M. W., Southon J., Jull A. J. T. and Donahue D. J. (1993) Cosmogenic-radionuclide profiles in Knyahinya. Lunar Planet. Sci. XXIV. Lunar Planet. Inst., Houston. pp. 1195–1196 (abstr.). Reedy R. C. and Masarik J. (1994a) Numerical simulation of cosmogenic nuclide production in lunar rock. Meteoritics 29, 521 (abstr.). Reedy R. C. and Masarik J. (1994b) Cosmogenic-nuclide depth profiles in the lunar surface. Lunar Planet. Sci. XXV. Lunar Planet. Inst., Houston. pp. 1119–1120 (abstr.).

2176

K. Nishiizumi et al. / Geochimica et Cosmochimica Acta 73 (2009) 2163–2176

Reedy R. C. (1998) Studies of modern and ancient solar energetic particles. Proc. Indian Acad. Sci. (Earth and Planetary Sciences) 107, 433–440. Reedy R. C. and Nishiizumi K. (1998) Factors affecting the interpretation of solar-proton-produced nuclides and some chlorine-36 results. Lunar Planet. Sci. XXIX. Lunar Planet. Inst., Houston. #1698 (abstr.). Reedy R. C., Marti K., Lavielle B. and Gilabert E. (1999) Solarproton fluxes from krypton-81 revisited. Meteor. Planet. Sci. 34, A96 (abstr.). Reedy R. C. (2006) Solar-proton event-integrated fluencies during the current solar cycle. Lunar Planet. Sci. XXXVII. Lunar Planet. Inst., Houston. #1419 (abstr.). Regnier S., Lagarde M., Simonoff G. N. and Yokoyama Y. (1973) Production de 26Al dans Fe et Si par protons de 0.6 et 24 GeV. Earth Planet. Sci. Lett. 18, 9–12. Russ, III, G. P. and Emerson M. T. (1980) 53Mn and 26Al evidence for solar cosmic ray constancy-an improved model for interpretation. In Proc. Conf. Ancient Sun (eds. R. O. Pepin, J. A. Eddy and R. B. Merrill). Pergamon Press, pp. 387–399. Ryder G. and Norman M. D. (1980) Catalog of Apollo 16 rocks. Publication 52, JSC/NASA, Houston 1144. Schiekel T., Sudbrock F., Herpers U., Gloris M., Lange H.-J., Leya I., Michel R., Dittrich-Hannen B., Synal H.-A., Suter M., Kubik P. W., Blann M. and Filges D. (1996) Nuclide production by proton-induced reactions on elements (6 6 Z 6 29) in the energy range from 200 MeV to 400 MeV. Nucl. Instrum. Methods Phys. Res. B114, 91–119. Schneider E. and Ho¨rz F. (1974) Microcrater populations on Apollo 17 rocks. Icarus 22, 459–473. Schneider R. J., Sisterson J. M., Koehler A. M., Klein J. and Middleton R. (1987) Measurement of cross sections for aluminum-26 and sodium-24 induced in aluminum. Nucl. Instrum. Methods Phys. Res. B29, 271–274. See T. H., Ho¨rz F. and Morris R. V. (1986) Apollo 16 impact-melt splashes; petrography and major-element composition. Proc. Lunar Planet. Sci. Conf. 17, J. Geophys. Res. 91, E3–E20. Sharma P., Kubik P. W., Fehn U., Gove H. E., Nishiizumi K. and Elmore D. (1990) Development of 36Cl standards for AMS. Nucl. Instrum. Methods Phys. Res. B52, 410–415. Shibata S., Imamura M., Nagai H., Kobayashi K., Sakamoto K., Furukawa M. and Fujiwara I. (1993) Measurements of 10Be and 26Al production cross sections with 12 GeV protons by accelerator mass spectrometry. Phys. Rev. C 48, 2617–2624. SHRELLDALFF, Shedlovsky J. P., Honda M., Reedy R. C., Evans J. C., Lal D., Lindstrom R. M., Delany A. C., Arnold J. R., Loosli H. J., Fruchter J. S. and Finkel R. C. (1970) Pattern of bombardment produced radionuclides in rock

10017 and in lunar soil. In Proc. Apollo 11 Lunar Sci. Conf. pp. 1503–1532. Sisterson J. M., Schneider R. J. I., Jull A. J. T., Donahue D. J., Cloudt S., Kim K., Beverding A., Englert P. A. J., Castaneda C., Vincent J. and Reedy R. C. (1996) Revised solar cosmic ray fluxes estimated using measured depth profiles of 14C in lunar rocks; the importance of good 14C cross section determinations. Lunar Planet. Sci. XXVII. Lunar Planet. Inst., Houston. pp. 1209–1210 (abstr.). Sisterson J. M., Kim K., Beverding A., Englert P. A. J., Caffee M., Jull A. J. T., Donahue D. J. and McHargue L. (1997a) Measurement of proton production cross sections of 10Be and 26 Al from elements found in lunar rocks. Nucl. Instrum. Methods Phys. Res. B123, 324–329. Sisterson J. M., Kim K., Beverding A., Englert P. A. J., Caffee M. W., Vincent J., Castaneda C. and Reedy R. C. (1997b) Measuring excitation functions needed to interpret cosmogenic nuclide production in lunar rocks. In Application of Accelerators in Research Industry (eds. J. L. Duggan and I. L. Morgan). AIP Press, New York, pp. 811–814. Sisterson J. M., Nishiizumi K., Caffee M. W., Imamura M. and Reedy R. C. (1997c) Revised 36Cl production rates in lunar rock 64455 using new cross section measurements. Lunar Planet. Sci. XXVIII. Lunar Planet. Inst., Houston. pp. 1329–1330 (abstr.). Theis S., Englert P., Michel R., Aylmer D., Herzog G. F., Kruse T. H., Moniot R. K., Tuniz C., Jermakian A., Klein J. and Middleton R. (1986) 10Be and 26Al production from Mg, Al, Si, and O by 600-MeV protons in granodiorite spheres. Lunar Planet. Sci. XVII. Lunar Planet. Inst., Houston. pp. 887–888 (abstr.). Tuniz C., Smith C. M., Moniot R. K., Kruse T. H., Savin W., Pal D. K., Herzog G. F. and Reedy R. C. (1984) Beryllium-10 contents of core samples from the St. Severin meteorite. Geochim. Cosmochim. Acta 48, 1867–1872. Wahlen M., Honda M., Imamura M., Fruchter J. S., Finkel R. C., Kohl C. P., Arnold J. R. and Reedy R. C. (1972) Cosmogenic nuclides in football-sized rocks. Proc. Lunar Sci. Conf. 3, 1719– 1732. Webber W. R., Kish J. C. and Schrier D. A. (1990) Individual isotopic fragmentation cross sections of relativistic nuclei in hydrogen, helium, and carbon targets. Phys. Rev. C 41, 547– 565. Yiou F., Seide C. and Bernas R. (1969) Formation cross sections of lithium, beryllium, and boron isotopes produced by the spallation of oxygen by high-energy protons. J. Geophys. Res. 74, 2447–2448. Associate editor: Rainer Wieler