K–Ar ages of meteorites: Clues to parent-body thermal histories

Chemie der Erde 71 (2011) 207–226

Contents lists available at ScienceDirect

Chemie der Erde journal homepage: www.elsevier.de/chemer

Invited review

K–Ar ages of meteorites: Clues to parent-body thermal histories Donald D. Bogard ∗ Lunar and Planetary Institute, USRA, Houston, TX 77058, USA

a r t i c l e

i n f o

Article history: Received 21 December 2010 Accepted 7 March 2011 Keywords: Ar–Ar ages Meteorites Thermal histories Impacts Metamorphism

a b s t r a c t Whereas most radiometric chronometers give formation ages of individual meteorites >4.5 Ga ago, the K–Ar chronometer rarely gives times of meteorite formation. Instead, K–Ar ages obtained by the 39 Ar–40 Ar technique span the entire age of the solar system and typically measure the diverse thermal histories of meteorites or their parent objects, as produced by internal parent body metamorphism or impact heating. This paper briefly explains the Ar–Ar dating technique. It then reviews Ar–Ar ages of several different types of meteorites, representing at least 16 different parent bodies, and discusses the likely thermal histories these ages represent. Ar–Ar ages of ordinary (H, L, and LL) chondrites, R chondrites, and enstatite meteorites yield cooling times following internal parent body metamorphism extending over ∼200 Ma after parent body formation, consistent with parent bodies of ∼100 km diameter. For a suite of H-chondrites, Ar–Ar and U–Pb ages anti-correlate with the degree of metamorphism, consistent with increasing metamorphic temperatures and longer cooling times at greater depths within the parent body. In contrast, acapulcoites–lodranites, although metamorphosed to higher temperatures than chondrites, give Ar–Ar ages which cluster tightly at ∼4.51 Ga. Ar–Ar ages of silicate from IAB iron meteorites give a continual distribution across ∼4.53–4.32 Ga, whereas silicate from IIE iron meteorites give Ar–Ar ages of either ∼4.5 Ga or ∼3.7 Ga. Both of these parent bodies suffered early, intense collisional heating and mixing. Comparison of Ar–Ar and I–Xe ages for silicate from three other iron meteorites also suggests very early collisional heating and mixing. Most mesosiderites show Ar–Ar ages of ∼3.9 Ga, and their significantly sloped age spectra and Ar diffusion properties, as well as Ni diffusion profiles in metal, indicate very deep burial after collisional mixing and cooling at a very slow rate of ∼0.2 ◦ C/Ma. Ar–Ar ages of a large number of brecciated eucrites range over ∼3.4–4.1 Ga, similar to ages of many lunar highland rocks. These ages on both bodies were reset by large impact heating events, possibly initiated by movements of the giant planets. Many impact-heated chondrites show impact-reset Ar–Ar ages of either >3.5 Ga or <1.0 Ga, and generally only chondrites show these younger ages. The younger ages may represent orbital evolution times in the asteroid belt prior to ejection into Earth-crossing orbits. Among martian meteorites, Ar–Ar ages of nakhlites are similar to ages obtained from other radiometric chronometers, but apparent Ar–Ar ages of younger shergottites are almost always older than igneous crystallization ages, because of the presence of excess (parentless) 40 Ar. This excess 40 Ar derives from shock-implanted martian atmosphere or from radiogenic 40 Ar inherited from the melt. Differences between meteorite ages obtained from other chronometers (e.g., I–Xe and U–Pb) and the oldest measured Ar–Ar ages are consistent with previous suggestions that the 40 K decay parameters in common use are incorrect and that the K–Ar age of a 4500 Ma meteorite should be possibly increased, but by no more than ∼20 Ma. © 2011 Elsevier GmbH. All rights reserved.

Contents 1. 2. 3. 4.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ar–40 Ar method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ar age spectra, Ar diffusion, and K–Ar closure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ar–Ar ages by meteorite type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Ordinary chondrites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Enstatite meteorites and R chondrites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

39

∗ Tel.: +1 281 486 2195. E-mail addresses: [email protected], [email protected] 0009-2819/$ – see front matter © 2011 Elsevier GmbH. All rights reserved. doi:10.1016/j.chemer.2011.03.001

208 208 209 211 211 212

208

5. 6. 7.

D.D. Bogard / Chemie der Erde 71 (2011) 207–226

4.3. Acapulcoites and lodranites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4. IAB iron silicates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.5. IIE iron silicates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.6. Other iron silicates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.7. Mesosiderites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.8. HED meteorites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.9. Late collisional events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.10. Martian meteorites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Absolute K–Ar ages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Summary discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

39 Ar–40 Ar

212 213 214 215 216 217 218 220 221 222 223 223 223

method

1. Introduction

2.

Most ages of meteorites measured by various radiometric chronometers correspond to times of condensation of solid material in the early solar nebula, the formation time of the meteorite parent body, or formation of the meteorite itself, for example, by igneous processes. Chronometer systems used in such measurements include those based on long-lived parent nuclides (e.g., U–Th–Pb, Sm–Nd, and Rb–Sr) and short-lived, extinct nuclides (e.g., 53 Mn–53 Cr, 26 Al–26 Mg, 107 Pd–107 Ag, 182 Hf–182 W, 146 Sm–142 Nd, and 129 I–129 Xe). These chronologic data indicate that early condensates in our solar system formed ∼4.565 Ga ago (1 Ga = 109 years), and that most meteorite parent bodies formed a few million years afterward, or about 4.56 Ga ago. Reviews of such meteorite ages are given in Carlson and Lugmair (2000), Gilmour (2000), Krot et al. (2009), Wadhwa et al. (2009), Nyquist et al. (2009a), and Kleine et al. (2009). In contrast to most other isotopic chronometers, few K–Ar ages of meteorites represent their formation times as solid objects or formation times of the parent bodies. This situation exists because the K–Ar chronometer is easily reset by moderate heating and because most meteorites have experienced thermal events, produced either by internal metamorphism or impacts on the parent body. These thermal events span essentially the entire history of the solar system. The ease with which the thermal environment resets K–Ar ages permits this chronometer to be quite useful for determining the times of meteorite heating. In some cases, the K–Ar data obtained give additional information about these thermal events, such as heating temperature and post-heating cooling rate. These latter data derive from detailed characteristics of Ar diffusion in meteorite samples, and are acquired during the process of measuring the K–Ar age. So that the reader can better understand interpretations of K–Ar ages and other information derived from the Ar diffusion data, I first present some basics about measuring K–Ar ages and Ar diffusion. Following that, I summarize K–Ar ages for several classes of meteorites and review what these apparently tell us about the thermal histories of their parent bodies. In a paper with such a broad topic, it is not possible to reference all sources of data and discussion. Rather, I emphasize those papers that both present new data and review previously published age data or interpretations. Although sufficient data will be presented so as to establish obvious trends for a given meteorite class, some data on meteorites without a clear connection to a meteorite group will not be utilized. The major purpose of this paper is to summarize what K–Ar ages imply about thermal histories of some specific meteorite groups, and not to produce an exhaustive data complication. Hopefully, the many references given will serve to introduce the interested reader to the broader literature. Much of the data utilized were obtained in the author’s laboratory at the NASA Johnson Space Center (JSC) in Houston, TX.

For the past few decades, K–Ar ages of meteorites have not been determined by the older method of independent measurements of K and 40 Ar, but rather by utilizing the superior 39 Ar–40 Ar technique (cf. McDougall and Harrison, 1999, and references therein). The 39 Ar–40 Ar method (hereafter called Ar–Ar) involves fast neutron irradiation of the sample to convert a portion of the stable isotope 39 K to 39 Ar (half-life 269 years), which is then located in the same K lattice sites as is 40 Ar resulting from the natural decay of 40 K over time. The irradiated sample is heated at increasing stepwise temperatures in a high vacuum furnace, using an internal resistance coil, induction heating or a laser. Infrared (e.g., CO2 ) lasers are often used, or more rarely, a laser emitting at higher frequency, e.g., a UV laser. The 40 Ar/39 Ar ratio in the argon released in each step is measured on a mass spectrometer designed for this purpose. Because the 39 Ar serves as a proxy for 39 K, the measured 40 Ar/39 Ar ratio is a measure of the K–Ar age of the sample, without requiring a separate, absolute measurement of K abundance. With increasing extraction temperature, the Ar released may derive from different K-bearing phases and from different lattice sites, such as grain surfaces or grain interiors. An Ar–Ar age spectrum (Fig. 1) plots

4.7

0.20

4.6 0.15

0.10

K / Ca

4.5

4.4

39

Ar-40Ar AGE Ga

Age= 4.521 ±0.006 Gyr

GRA-95209 whole rock

0.05

4.3

4.2

0.00 0.0

0.1

0.2

0.3 39

0.4

0.5

0.6

0.7

0.8

0.9

1.0

Ar Cumulative Fraction

Fig. 1. Example of an Ar–Ar age spectrum, which plots age in Ga (rectangles, showing uncertainties) and the K/Ca ratio (stepped line) vs. the cumulative release of 39 Ar during stepwise temperature extractions of the GRA-95209 lodranite meteorite. Low temperature release of Ar shows effects of terrestrial weathering (loss of radiogenic 40 Ar and gain of atmospheric 40 Ar) and high temperature release shows effects of 39 Ar recoil in the reactor. An age of 4.521 ± 0.006 Ga is defined by the major K-bearing phase having a constant K/Ca ratio and releasing 6–84% of the total 39 Ar. Figure reproduced from McCoy et al. (2006).

D.D. Bogard / Chemie der Erde 71 (2011) 207–226

the calculated age against the fractional release of 39 Ar for each extraction as the temperature of the sample is increased. In many cases, the Ar–Ar technique permits one to identify terrestrial atmospheric Ar absorbed on grain surfaces and diffusive loss of some of the radiogenic 40 Ar that occurred over the age of the sample. Because Ar is a gas, it may be easily lost from some lattice sites, and determining this prior loss is important in Ar–Ar dating. Identifying such 40 Ar loss is not possible in classical K–Ar dating or in Ar–Ar dating if all Ar is extracted in one temperature step. In this paper, I will write Ar–Ar age when referring to data produced by this technique and K–Ar age when referring generally to the chronometer or age itself. Uncertainties of most Ar–Ar ages reported here are one sigma, and in some cases the level of the uncertainty is not given in the original work. Often Ar–Ar data (especially for terrestrial samples) are interpreted using an isochron plot, analogous to an isochron obtained from mineral separates in other chronometers. In an Ar–Ar isochron, 40 Ar is plotted against 39 Ar, and the isochron slope usually yields the age (McDougall and Harrison, 1999). Alternate isochrons plot 40 Ar/36 Ar ratios against 39 Ar/36 Ar ratios or 39 Ar/40 Ar against 36 Ar/40 Ar. For terrestrial samples the isochron intercept on the 40 Ar/36 Ar (or 36 Ar/40 Ar) axis gives the trapped component not produced by in situ decay of 40 K, usually atmospheric Ar, with or without the presence of some excess radiogenic 40 Ar inherited by the sample. However, in the case of meteorites, the presence of an argon component produced by cosmic rays in space often means that the isochron intercept does not identify a single trapped component. Often it is difficult to correct for this cosmogenic Ar. Ar–Ar dating of terrestrial samples is commonly performed on separates of K-bearing minerals, in order to simplify data interpretation. However, separates of K-bearing minerals are difficult to obtain for meteorites, because most meteoritic minerals are fine-grained and intergrown (e.g., Cressy, 1971). Because the 40 K half-life is relatively short (1.3 Ga) compared to typical meteorite formation ages (and compared to half-lives of radioactive parent nuclides in other chronometers), and because the 40 Ar/39 Ar ratio can be measured to high precision, the Ar–Ar chronometer has the potential of measuring an age very precisely. However, several complications in the Ar–Ar method produce some uncertainty in the accuracy of the derived age (cf. McDougall and Harrison, 1999). First, because the exact amount of conversion of 39 K to 39 Ar in the reactor is difficult to measure directly, the Ar–Ar age must be measured relative to the known age of a reference sample irradiated along with the unknown meteorite sample. (I–Xe ages are also measured relative to an age standard.) Several standard samples are used in Ar–Ar dating, and these have been calibrated against one another and compared to ages of the reference sample obtained by other radiometric methods. The uncertainty in age of the unknown sample introduced by the use of a standard sample is typically less than 0.5% (cf. Schwarz and Trieloff, 2007). Adopted ages for some standard samples have slightly changed over time, and the uncertainty in most reported Ar–Ar ages of meteorites do not include the uncertainty in the standard sample age. However, all measured Ar–Ar ages from all laboratories utilizing the same irradiation standard are affected to the same degree by any uncertainty in the absolute age of that standard. Most reported meteorites ages have been measured relative to two or three standard samples. A second complication in measuring Ar–Ar ages arises because the standard and unknown samples cannot be positioned in the same physical location during irradiation in the reactor, and the rate of conversion of 39 K to 39 Ar may slightly differ for the two. The effect of reactor position is monitored using multiple samples of the age standard or some other flux monitor, positioned throughout the irradiation package. The relative uncertainty this factor introduces into the age depends on several things, including the fast neutron

209

flux gradient across the sample package and relative positions of the various samples and flux monitors. Typically, the uncertainty this factor introduces into the age (as measured by the flux monitors and their relative positions) is included in the age uncertainty. It is typically no greater than a few tenths of a percent. A third uncertainty introduced into an Ar–Ar age determination arises from a correction that must be applied to measured 39 Ar (and to a lesser degree to 40 Ar) because of reactor production of 39 Ar from Ca (and 40 Ar from K). These correction factors are separately measured for each reactor used (e.g., Nyquist et al., 2006, appendix). The amount of the correction applied to 39 Ar increases with increasing Ca/K ratios, and for Ca/K of 200 (typical of eucrites) would be about 7%, but smaller for chondrites. The correction applied to 40 Ar is largest for high-K content and very young samples, and for most meteorites is minimal (<1%). The age uncertainty in making these reactor corrections is also typically included in the quoted age of the sample. Thus, when comparing Ar–Ar ages from the same laboratory or from different labs using the same age standard, the assigned age uncertainties for individual samples are generally sufficient. When comparing Ar–Ar ages reported by different labs using different Ar age standards, an additional uncertainty (but not necessarily an age difference) of some tenths of a percent may apply. There also exists some uncertainty in the half-life of the radioactive parent in all chronometers, and 39 K is no exception. Most laboratories use the K decay parameters recommended by Steiger and Jäger (1977). However, some evidence suggests that these K decay parameters are not correct (Renne et al., 1998, 2010), and that calculated Ar–Ar (and K–Ar ages) of meteorites are too young. There is not yet agreement as to how much determined Ar–Ar ages should be changed, but recent studies suggest that a 4.50 Ga age should be increased by perhaps 20 Ma (1 Ma = 1 × 106 years), as is discussed in Section 5. This uncertainty in absolute Ar–Ar age arising from uncertainty in the K decay parameters can affect interpretation of parent body histories for those meteorites that are relatively old, e.g., some chondrites and iron silicates. However, this age uncertainty becomes less important for those measured K–Ar ages that are significantly younger than the times of parent body formation. Note that all measured Ar–Ar ages from all labs are affected in the same manner by this uncertainty in the K decay parameters.

3. Ar age spectra, Ar diffusion, and K–Ar closure In most meteorites, not all K, and thus not all 40 Ar and 39 Ar, reside in a single lattice site with common Ar diffusion properties, and sometimes not in the same mineral phase. This fact often complicates Ar–Ar age interpretation. Different Ar diffusion sites often can be identified during stepwise temperature extraction by changes in the rate at which Ar degasses, changes in the K/Ca ratio of the phase degassing the Ar, and sometimes by changes in the Ar–Ar age. [Just as the K-bearing sites are identified by release of 39 Ar produced from K in the irradiation, so also are Ca-bearing sites identified by release of 37 Ar (half-life 35.1 days), which is produced from 40 Ca during neutron irradiation.] The Ar–Ar age spectrum (Fig. 1) for the GRA 95209 lodranite (McCoy et al., 2006) illustrates some of the points made above. The first ∼6% of the 39 Ar release gives evidence of terrestrial weathering of grain surfaces (loss of radiogenic 40 Ar, gain of atmospheric 40 Ar, and rapid changes in K/Ca), and the last ∼16% of the 39 Ar release gives evidence of reactor-induced 39 Ar recoil, from grain surfaces of a phase richer in K into grain surfaces of mafic phases such as pyroxene (see Bogard, 1995; McDougall and Harrison, 1999; Park et al., 2008a). Mafic phases typically show much lower K/Ca ratios and degas their Ar at higher temperatures. Such 39 Ar recoil effects may or may not be significant in a given Ar–Ar age spectrum. From 10% to 84% of the 39 Ar release, both the K/Ca ratio and the age of GRA 95209 give the same values within

210

D.D. Bogard / Chemie der Erde 71 (2011) 207–226

-3 -4

Log D / a2 sec-1

-5 -6 -7 Mt. Padbury Plag

-8

Savtschenskoje-LL Portales Valley #1

-9

Portales Valley #2

-10 0.5

0.6

0.7

0.8

0.9

1.0

1.1

1.2

1.3

1.4

1.5

1.6

1000 / T Fig. 2. Arrhenius plots of log D/a2 vs. reciprocal temperature in Kelvin for 39 Ar diffusion from four meteorite samples. D is the diffusivity and a is the diffusion distance. Each sample shows distinctly different Ar diffusion characteristics, with activation energies (in kcal/mole) of Mt. Padbury = 33, Savtschenskoje = 19, Portales Valley #1 = 69, and Portales Valley #2 = 49.

their relative uncertainties and define the closure time of the Ar–Ar chronometer at 4.521 ± 0.006 Ga. In deciphering Ar–Ar ages, the JSC laboratory often utilized Ar diffusion characteristics, which may be calculated from the release of 39 Ar and 37 Ar as a function of extraction temperature, in order to compare the relative ease with which radiogenic 40 Ar can be lost from different samples (see Turner et al., 1978; Bogard and Hirsch, 1980). Knowing the Ar diffusion characteristics of a sample also may permit calculation of the temperature at which the K–Ar chronometer closed (see below). The Ar diffusion characteristics in a sample are expressed by two parameters, D/a2 , where D is diffusivity and ‘a’ is mean diffusion distance (typically determined by grain size), and by the activation energy, Q, typically given in kcal/mole or joules/mole. D/a2 is a measure of how fast Ar moves through the lattice, and Q is a measure of the energy resistance of the lattice to Ar diffusion. As expected, these two diffusion parameters tend to correlate. In the Ar–Ar age spectrum of the Monument Draw acapulcoite, McCoy et al. (1996) noted two distinct K-bearing phases, possessing different Ar diffusion properties and yielding different calculated K–Ar closure temperatures as a function of cooling rate. Using similarity in Ar–Ar ages of these two K-bearing phases, but differences in their Ar diffusion properties, McCoy et al. (1996) concluded that Monument Draw cooled from its peak metamorphic temperature of ∼1000 ◦ C to Ar–Ar closure at an average rate faster than 10 ◦ C/Ma. This study demonstrates how Ar–Ar data can be used to constrain meteorite thermal histories and cooling rates, and more examples will be given later. Fig. 2 gives four examples of 39 Ar diffusion characteristics of meteorites – plagioclase separated from the Mount Padbury mesosiderite, the Savtschenskoje LL chondrite, and two different samples (#1 and #2) from Portales Valley, a unique mixture of chondritic silicate and abundant iron metal (Bogard and Garrison, 1999, 2009a; Dixon et al., 2004). The Arrhenius plots shown, D/a2 against reciprocal temperature in Kelvin, are defined by 2–100%, 1–85%, 1–94%, and 0.5–92% of the total 39 Ar released, respectively. (The use of most of the total 39 Ar release from a single phase for calculating D/a2 data for these samples makes the calculations robust. 39 Ar released outside these ranges derived from other K-bearing phases, as mentioned above.) Each linear data trend defines the value of the diffusion parameter, D/a2 , at any given temperature, and the activation energy can be calculated from the Arrhenius

slope. These four samples show very different activation energies for Ar diffusion of 33, 19, 69, and 49 kcal/mole, respectively. Further, it is obvious from Fig. 2 that, at most temperatures over which 39 Ar degasses, the four samples show significantly different values of D/a2 . For example, at 723 ◦ C (1000/T = 1.0), which is near the mid point of 39 Ar and 40 Ar release for many meteorites, these four samples show a spread in D/a2 values of two orders of magnitude, from log −6.6 to log −4.5. Even two different silicate samples in Portales Valley give different values for both Ar diffusion parameters, and these differences indicate that the K-bearing lattice phases in the two samples are not identical. Turner et al. (1978) calculated Ar diffusion data for a suite of chondrites and reported a range of log D/a2 values at 1000 K of −4.2 to −8 and a range in activation energies, Q, of 20–70 kcal/mole. We have calculated similar ranges in these Ar diffusion parameters for many different meteorites measured in the JSC laboratory. The major conclusion to be drawn is that, because the parameters of Ar diffusion show significant variation among individual meteorites, then 40 Ar loss during thermal events can also be expected to show significant variation. The consequence of meteoritic K occurring in different phases, varying Ar diffusion properties, possible 40 Ar diffusion loss, and 39 Ar recoil effects, is that many Ar–Ar age spectra are complex and require interpretation. Complex Ar–Ar age spectra in meteorites have been discussed in detail by Turner (1969), Turner et al. (1978), Bogard et al. (1990, 2000, 2009), Bogard (1995), McCoy et al. (1996), Trieloff et al. (2003, appendix), Park et al. (2008a), and Bogard and Garrison (2009a). One of the more complex age spectra to interpret is where 40 Ar and 39 Ar degas from different K-bearing phases that partially overlap in the temperature of their release, and where the sample has experienced heating, which was sufficient to partially degas the lower temperature phase, but which produced a lesser degree of degassing of the higher temperature phase (Bogard, 1995). Thus, the reliability of a derived Ar–Ar age can depend not only on the factors discussed in the previous section, but also on how complex age spectra are interpreted. Typically an age is derived from only a part of the total age spectrum, and sometimes more than one thermal event may be recorded in the age spectrum. To deduce a reliable age from a sample, some workers require a minimum number of temperature steps or a minimum fraction of 39 Ar release all showing the same age within relative uncertainties. The JSC perspective is that to deduce a reliable age, one also should understand the whole age spectrum, including diffusion loss, 39 Ar recoil effects, multiple K phases, etc. Nevertheless, when several meteorites of a given type indicate similar Ar–Ar ages, the reliability of the derived age increases. In some chronometer systems, particularly K–Ar, the concept of closure temperature is important. Closure measures the temperature and time when the daughter nuclide from radioactive decay stops isotopic exchange among mineral phases, or in the case of K–Ar, the Ar stops diffusing out of the system (Dodson, 1973). Prior to closure, the chronometer does not record time. The closure temperature varies with the chronometer, the host mineral phase, the cooling rate, and the activation energy for diffusion. For some chronometers (e.g., Sm–Nd), closure occurs at high temperature, possibly approaching melting, and the closure time measures essentially the time when the solid material or the meteorite formed. However, most K–Ar systems have relatively low closure temperatures of a few hundred ◦ C in slow-cooling systems, and diffusive loss of 40 Ar can continue to occur during cooling of the meteorite within a parent body or following an impact that leaves the meteorite relatively hot (see discussion in Bogard, 1995). In diffusion of Ar (as with other species), diffusive loss depends on temperature, time, and diffusion distance, or grain size. Material heated to only moderate temperature for a longer time may lose more Ar than material heated to higher temperatures for a shorter time (e.g., Park et al., 2008a,b). In the following discussion of Ar–Ar

D.D. Bogard / Chemie der Erde 71 (2011) 207–226

ages of meteorites, I reference several specific examples where Ar diffusion and K–Ar closure were utilized to constrain the thermal histories.

211

4.55

4.50

4. Ar–Ar ages by meteorite type

Ordinary chondrites are divided into H, L, and LL types, partially based on the oxidation state of Fe (e.g., Brearley and Jones, 1998). Most chondrite groups are composed of mixtures of chondrules and matrix material, which experienced various processes prior to incorporation into parent bodies (see reviews by Keil et al., 1994; Scott, 2007). These materials accreted into objects about 80–95 km in diameter (Bennett and McSween, 1996), and these bodies experienced early heating from the presence of short-lived nuclides (e.g., 26 Al) and possibly from other mechanisms. As a consequence of this heating, chondrites experienced varying degrees of metamorphism. Increasing degrees of heating of ordinary chondrites are indicated by their petrologic grade, a number ranging from 3 to 6, where higher numbers represent greater heating and presumably greater depth in the parent body. The temperatures of these metamorphic events have been estimated to range from minimal heating up to ∼800 ◦ C or more for the highest metamorphic grades (Dodd, 1981). A comparison of oxygen isotopes in the three ordinary chondrite classes, H, L, and LL, and in E and R chondrites discussed in the next section, confirm that these classes of chondrites derived from multiple parent bodies (e.g., Brearley and Jones, 1998). Because of the complexity of these early events in the formation of chondrites and their parent bodies, chondrites do not show a single, unique age when examined by different chronometer systems. Their thermal histories occurred over an extended period of time. Components of chondrites (chondrules, CAIs, etc.) were formed by ∼4.565 Ga ago, and chondrite parent bodies had formed and were experiencing internal metamorphism by 4.560 Ga ago (Carlson and Lugmair, 2000; Amelin et al., 2002; Wadhwa et al., 2009; Nyquist et al., 2009a, Bouvier and Wadhwa, 2010). An analysis of the U–Pb chronometer in phosphate minerals from several equilibrated chondrites gave model ages ranging over 4.563–4.502 Ga, and these ages show an expected negative correlation with metamorphic grade (Göpel et al., 1994). The U–Pb system in phosphates has a relatively low closure temperature and is expected to be open at higher metamorphic temperatures. These data imply an ∼60 Ma period during which the U–Pb chronometer in different parts of the parent body cooled to below its closure temperature at different times. Brazzle et al. (1999) showed that for some of these same chondrites, 129 I–129 Xe ages obtained on phosphates and feldspar separates agreed with the Pb–Pb ages. However, some other I–Xe ages of chondrites give variations of tens of Ma and do not correlate well with metamorphic grade. It has been argued that iodine commonly occurs in soluble phases, and that some I–Xe ages reflect secondary aqueous processes on the parent bodies and not original formation times or even metamorphic cooling (Hohenberg and Pravdivtseva, 2008, and references therein). The Ar–Ar ages determined for several ordinary chondrites are shown in Fig. 3, and most ages are ∼4.43–4.53 Ga (Turner et al., 1978; Kaneoka, 1980; Pellas and Fiéni, 1988; Bogard et al., 2001; Trieloff et al., 2003; Dixon et al., 2004). Many other ordinary chondrites show younger ages down to ∼0.2 Ga, and most of these young ages represent impact heating and resetting (see Section 4.9). K–Ar ages of carbonaceous chondrites are uncommon and difficult to determine, although Jessberger et al. (1980) reported a clustering of Ar–Ar ages around 4.53 Ga for Allende. Most ordinary chondrite ages in Fig. 3 are unlikely to have been reset by impact, but probably reflect closure of the K–Ar chronometer during cooling after parent

Age, Ga

4.1. Ordinary chondrites 4.45

4.40

4.35

4.30 H

L

LL

EH

EL

EM

R

AL

Chondrite Type Fig. 3. Ar–Ar ages of ordinary, enstatite, and R-chondrites, as well as acapulcoites–lodranites, arranged by meteorite type: H = H chondrites, L = L chondrites, LL = LL chondrites, EH = EH enstatite chondrites, EL = EL enstatite chondrites, EM = enstatite melt rocks (diamond with cross) and aubrites (open diamond), R = R chondrites, and AL = acapulcoites and lodranites. Uncertainties reported for older chondrite ages are ∼30 Ma, and those for more recently reported ages are ∼5–20 Ma. Age uncertainties for acapulcoites–lodranites are given in Table 1. See text for data sources.

body metamorphism. The range in Ar–Ar ages of these chondrites, ∼100 Ma, is slightly greater than the ∼60 Ma age range shown by U–Pb in chondritic phosphates, and this difference is probably consistent with lower closure temperatures for K–Ar in feldspar compared to U–Pb in phosphates. Turner et al. (1978) estimated K–Ar closure temperatures of ∼100–400 ◦ C for several chondrites. These closure temperatures were calculated assuming cooling rates of 1–10 ◦ C/Ma, values that are consistent with some chondritic cooling rates determined from Ni diffusion in metal (Taylor et al., 1987; Bennett and McSween, 1993). Calculated closure temperatures increase modestly as cooling rates increase. For example, for Portales Valley sample #2 (Fig. 2), increasing the cooling rate from 1 ◦ C/Ma to 100 ◦ C/Ma increases the K–Ar closure temperature by about 65 ◦ C. Time periods up to ∼100 Ma for the K–Ar chronometer to close in strongly heated chondrites is consistent with the chondrite parent body thermal models of Bennett and McSween (1996). One earlier puzzle about Ar–Ar ages of chondrites was that younger ages did not correlate well with increasing metamorphic grade. Taylor et al. (1987) suggested that some parent asteroids underwent early collisional fragmentation and reassembly while they were still relatively hot, which altered the thermal profile as a function of depth. The L chondrite parent asteroid very likely experienced impact disruption 0.47 Ga ago (e.g., Korochantseva et al., 2007), and this event influenced the K–Ar ages of many L chondrites (see Section 4.9). Trieloff et al. (2003) showed that Ar–Ar ages (4.435–4.532 Ga, Fig. 3) for a suite of H chondrites of metamorphic grades 4–6, selected for low levels of shock-heating, do correlate with Pb–Pb ages measured in phosphates (4.504–4.563 Ga) and inversely correlate with metamorphic grade. These results support the interpretation that these H chondrite Ar–Ar ages reflect parent body cooling after metamorphism, and also support the “onion-layer” concept of chondrite parent bodies, whereby internal temperatures were greater in the deep interior and much lower near the surface.

212

D.D. Bogard / Chemie der Erde 71 (2011) 207–226

There exists petrologic evidence for impacts occurring on meteorite parent bodies very early in their history (e.g., Keil et al., 1994; Scott, 2002), and these impacts may have reset the Ar–Ar ages of some chondrites. Few old Ar–Ar ages (i.e., >4.3 Ga) exist for chondrites which were obviously strongly impact-heated, e.g. melted. Four examples are the PAT 91501 impact-melted L chondrite with an Ar–Ar age of 4.461 Ga (Benedix et al., 2008); a few LL chondrites whose Ar–Ar ages imply strong impact heating at ∼4.27 Ga (Dixon et al., 2004); the impact-melted Shaw L chondrite with an Ar–Ar age of 4.43 Ga (Turner et al., 1978); and MIL 05029, an L chondrite impact melt with an Ar–Ar age of 4517 ± 11 Ma (Weirich et al., 2011). MIL 05029 gives a relatively slow metal cooling rate of 14 ◦ C/Ma, implying formation in a very large impact event, which may have disrupted the L chondrite parent body and produced the lack of correlation between cooling rates and metamorphic grades in many L-chondrites (Weirich et al., 2011). Although impact resetting of Ar–Ar ages on ordinary chondrite parent bodies during the later stages of metamorphic cooling down to ∼4.3 Ga appears to have been rare, possibly all Ar–Ar ages of chondrites younger than ∼4.40 Ga resulted from impact heating. 4.2. Enstatite meteorites and R chondrites Far fewer chronological data exist for enstatite chondrites and aubrites. Aubrites are thought to have formed through igneous processes on enstatite-like parent bodies (Keil, 1989, 2010). The two recognized sub-classes of E chondrites, EL and EH, and aubrites are thought to have derived from separate parent bodies. In addition, the Shallowater aubrite is thought to have derived from a fourth parent body (Keil, 1989, 2010). Many E chondrites and aubrites display petrologic evidence of impact shock (e.g., Rubin et al., 1997). Bogard et al. (2010) reported new Ar–Ar ages for several enstatite meteorites and summarized older enstatite chronology from the literature. The Ar–Ar ages for several EH and EL chondrites are plotted in Fig. 3, along with ages for two enstatite impact melt rocks, for which EH vs. EL parentage is not known. The range of these enstatite chondrite ages, ∼4.43–4.54 Ga, is very similar to the range of Ar–Ar ages for ordinary chondrites. The parent body metamorphic history for enstatite chondrites, and particularly the maximum temperatures experienced, have few data constraints and are uncertain (Zhang et al., 1995; Zhang and Sears, 1996). If E chondrite parent bodies were metamorphosed to similar temperatures as were ordinary chondrite parent bodies, the Ar–Ar ages imply that these bodies were similar in size and experienced a similar cooling history over ∼0.1 Ga. However, some evidence discussed by Zhang and coworkers suggests that EL meteorites were not heated as high as were ordinary chondrites. Further, the two E chondrite melt rocks indicate that impact melting occurred during this period of metamorphic cooling. These two observations suggests that some Ar–Ar ages of EH and EL chondrites also may have been reset by impact heating (e.g., the EH age at 4.35 Ga), and that very early impacts (>4.3 Ga ago) may have been relatively common on these parent bodies. Some older Rb–Sr and U–Pb chronological data for enstatite chondrites and aubrites also are consistent with impact resetting shortly after parent body formation (see references and discussion in Bogard et al., 2010). Compared to ordinary chondrites, the thermal and chronological history of E chondrites appears more complex and uncertain. Among aubrites only Shallowater gives a well-defined Ar–Ar age of 4.535 ± 0.02 Ga. A few other aubrites give younger ages by the Ar–Ar, Rb–Sr, and Pb–Pb chronometers (e.g., Norton County Ar–Ar = 4.45 Ga; Fig. 2), and these younger ages are likely the result of impact heating (see discussion in Bogard et al., 2010). Shallowater, an unbrecciated aubrite, also has been precisely dated by the 129 I–129 Xe chronometer at an age of 4.5623 ± 0.0004 Ga (Gilmour et al., 2009; Hohenberg and Pravdivtseva, 2008, and ref-

erences therein). It has been suggested that Shallowater formed by impact melting, followed by a complex cooling history (Keil, 1989). Because the I–Xe chronometer in unaltered meteorite silicates has a higher closure temperature than does K–Ar, this might explain the small difference in these two ages (but also see Section 5 for an alternate explanation). It appears that all of the enstatite parent bodies experienced appreciable impacts very early in their history, probably more so than did parent bodies of ordinary chondrites. Because of this, the history of parent body metamorphism is difficult to discern from the limited enstatite chronology. The Rumuruti type, or R chondrites, is a relatively rare meteorite group, which is commonly brecciated (Rubin and Kallemeyn, 1989; Kallemeyn et al., 1996; Bischoff et al., 2011). Application of the Mn–Cr chronometer to a few samples implies a formation time of R chondrites consistent with that of other chondrites (Sigura and Miyazaki, 2006), but very little additional chronological data are available. On a three-oxygen isotope diagram, all chondrite groups plot above the terrestrial oxygen fractionation line, and R chondrites plot the furthest away. Because these differences in oxygen isotopes imply parent body formation in different parts of the solar system (Wiens et al., 1999), it may be informative to compare both the chronology and evidence for early impacts in R chondrites to data for other chondrite groups. Dixon et al. (2003) reported Ar–Ar studies of four R chondrites, but the age spectra are very complex and ages are not easy to derive. The preferred Ar–Ar ages are plotted in Fig. 3, and the meteorites with the oldest and youngest ages are breccias. These four Ar–Ar ages likely give little to no constraint on the cooling history of R chondrites, and all four ages may reflect impact heating. Nagao et al. (1999) reported classical K–Ar ages of ∼4.2 Ga for four R chondrites, but these should be considered minimum times for closure of this chronometer. The parent bodies of both E and R chondrites appear to have experienced a greater degree of early impacts in comparison to ordinary (H, L, and LL) chondrites, and this may be related to their formation in different parts of the solar system. 4.3. Acapulcoites and lodranites Acapulcoites and lodranites are thought to have derived from the same parent body, whose silicates were petrologically intermediate between ordinary and enstatite chondrites, and which was metamorphosed to higher temperatures than parent bodies of ordinary chondrites (Palme et al., 1981; McCoy et al., 1996, 1997a, 1997b; Mittlefehldt et al., 1996). Acapulcoites were heated to temperatures of ∼950–1000 ◦ C, approximately the cotectic melting point of iron–nickel metal and sulfides. Lodranites experienced greater metamorphism than acapulcoites, above 1000 ◦ C, and lost feldspathic components by partial melting and removal of basaltic liquids (McCoy et al., 1996, 1997a). Chondrules are rare in both meteorite types. The characteristics of acapulcoites and lodranites tend to overlap, and the classification of a given meteorite into one of the two classes may be problematic (Patzer et al., 2004). Among criteria that have been used to distinguish between them are these typical characteristics of lodranites – coarser grain textures, higher metal contents, and depletion of elements lost in partial melting (McCoy et al., 1997b). Using the U–Pb chronometer, Göpel et al. (1992) measured the age for the type meteorite, Acapulco, at 4557 ± 2 Ma, which indicates formation of the parent body at least this early. Three independent laboratories measured very similar Ar–Ar ages for Acapulco (Table 1), and the variation among these three ages (8 Ma) is smaller than the quoted uncertainties. The Ar–Ar ages of 7 different acapulcoites and lodranites (Table 1, plotting only the average age for Acapulco) are shown in Fig. 3. These ages cluster tightly at 4.49–4.52 Ga, in contrast to ages for the various chondrite classes. Interestingly, two meteorites with lodranite

D.D. Bogard / Chemie der Erde 71 (2011) 207–226 Table 1 Ar–Ar ages for acapulcoites–lodranites and silicates in some iron meteorites. Meteorite Acapulcoites–lodranites Acapulco

Monument Draw ALH 81261 EET 84302 ALH 81187 Gibson GRA 95209 IIE iron silicates Watson Kodaikanal Natschaëvo Colomera Weekeroo Station Miles Techado Tarahumara Other iron silicates Sombrerete Portales Valley #1 Portales Valley #2 NWA 176

Age (Ga)

±

Ref.

4510 4514 4518 4517 4511 4519 4507 4493 4521

10 16 18 6 6 17 24 11 6

a b c a d d d e f

7

3676 (∼3400) 3745 4470 4490 4408 4489 4476

30 19 30 9 13 18

g g g g g g g h

4541 4477 4458 4524

12 11 16 13

i i i i

References: (a) McCoy et al., 1996; (b) Pellas et al., 1997; (c) Renne, 2000; (d) Mittlefehldt et al., 1996; (e) McCoy et al., 1997b; (f) McCoy et al., 2006, (g) Bogard et al., 2000, (h) Takeda et al., 2003, and (i) Bogard and Garrison, 2009a.

characteristics, Gibson, classified as a lodranite, and GRA 95209, classified as intermediate, show the lowest and highest Ar–Ar ages of the group. As lodranites were heated to somewhat higher temperatures, one might expect them to have taken longer to cool and thus to show younger ages than the acapulcoites. Also, if the acapulcoite–lodranite (A–L) parent body was heated to higher temperatures than parent bodies of chondrites, why are many Ar–Ar ages of chondrites significantly younger than A–L meteorites? For Acapulco, Pellas et al. (1997) plotted the Pb–Pb age, the Ar–Ar age, and three Pu fission track ages (based on lattice damage produced during fission of now extinct 244 Pu) against estimated closure temperatures for these chronometers (ranging over ∼700–300 K), and they derived a cooling history for the parent body. The track ages were measured in the minerals orthopyroxene, merrillite, and apatite, for which the temperatures at which tracks begin to form vary over ∼300–500 K. These authors concluded that the A–L parent body may have experienced rapid cooling (100 ± 40 C/Ma) down to about the Pb–Pb closure temperature of ∼720 K, followed by slow cooling at a rate of ∼2 ◦ C/Ma. The cooling rate of 2 ◦ C/Ma is similar to that for some H chondrites. Pellas et al. (1997) noted the absence of data which might reveal the cause of this two-stage cooling history. It is conceivable that the parent body underwent collisional disruption and reassembly very early in its history (e.g., Keil et al., 1994; Scott, 2002), which might explain the similarity in Ar–Ar ages for meteorites which apparently experienced different metamorphic temperatures. Interestingly, three acapulcoites–lodranites show quite variable amounts of excess 129 Xe resulting from the decay of extinct 129 I (half-life 17 Ma). GRA 95209, which gives the oldest Ar–Ar age in Table 1, contains significant excess 129 Xe, whereas the Gibson lodranite, giving the youngest Ar–Ar age, contains only a small excess of 129 Xe, and lodranite LEW 88280 contains no excess 129 Xe (Whitby et al., 2005). The difference in Ar–Ar ages between Gibson and GRA 94209 seems too small to explain these findings as due to slow cooling, and variations in excess 129 Xe among acapulcoites–lodranites may in part reflect variations in iodine content and early fluid movement, rather than cooling history (Whitby et al., 2005).

213

The cosmic ray exposure (CRE) ages of meteorites measure the time since they were impact-ejected as relatively small objects from depths of at least a few meters within their immediate parent objects. For ordinary chondrites, these ages range over two orders of magnitude, ∼1–100 Ma, and indicate ejection by many impact events. However, among CRE ages measured on ∼20 acapulcoites and lodranites, all but one fall in a narrow range of 4.6–6.8 Ma, suggesting that a very limited number of impacts initiated space exposure (Eugster and Lorenzetti, 2005). This finding may imply that most acapulcoites–lodranites were derived from a limited portion of the parent body and may account for the similarity in their K–Ar ages. Similar ages would be consistent with the two-stage cooling history suggested by Pellas et al. (1997) and with early impact disruption and cooling. Given the uncertainties in thermal history for this parent body, these data supply poor constraints on its initial size. 4.4. IAB iron silicates Several types of iron meteorites are thought to represent cores of differentiated asteroids from which the overlying silicate mantle has been removed by impacts. However, a few types of iron meteorites commonly contain inclusions of silicates and oxide minerals and did not likely form in asteroid cores. The IAB iron meteorites contain inclusions of a variety of compositions, including silicates having a chondrite-like composition (Mittlefehldt et al., 1998; Benedix et al., 2000). Silicates in a few IAB and IIE meteorites (discussed below) have an andesite-like, high-alkali composition, suggesting extreme partial melting products (Wasserburg et al., 1968; Takeda et al., 2000). Winonaites are composed of silicate resembling that in IABs, but without metal, and are thought to have originated on the same parent body as IAB meteorites (Benedix et al., 1998). The oxygen isotopic composition of IAB silicates is unlike that of chondrites (Clayton and Mayeda, 1996). Several models have been proposed to explain the formation of IAB meteorites, including parent body melting and fractional crystallization, incomplete differentiation, and impact-produced melting and differentiation (Kracher, 1985; Choi et al., 1995; Takeda et al., 2000; Wasson and Kallemeyn, 2002). Benedix et al. (2000) suggested a hybrid model in which a chondrite-like parent body experienced metamorphism up to 1200–1400 ◦ C, partial melting, and incomplete differentiation, which produced melts of different compositions. While still hot, a large impact caused IAB body breakup and reassembly, mixing metal and silicate. Like other meteorite parent bodies, the IAB parent formed relatively early. A high precision Hf–W age of 4561 ± 2 Ma for Caddo County measures the time of metal–silicate separation (Markowski et al., 2006). Analyses of the 187 Re–187 Os chronometer in metal from three IABs plot on a 4.61 Ga isochron defined by other types of iron meteorites (Shen et al., 1996). Niemeyer (1979a) reported 129 I–129 Xe ages of ∼4546–4574 Ma for silicate and troilite from five IABs, and Podosek (1970) reported an I–Xe age of 4558.6 Ma for silicate in Campo del Cielo (both normalized to a Bjurböle standard I–Xe age of 4562.3 Ma). On the other hand, slightly younger 147 Sm–143 Nd and 87 Rb–87 Sr ages of 4.50–4.53 Ga were reported for silicate in Caddo County (Stewart et al., 1996; Liu et al., 2002; Liu et al., 2003), and Niemeyer (1979b) reported Ar–Ar ages of ∼4.43–4.52 Ga for silicate from five IABs. Benedix et al. (1998) reported Ar–Ar ages of 4.53 and ∼4.51 Ga for two winonaite meteorites. Bogard et al. (2005) reported I–Xe ages of 4558 Ma for two IABs, an Ar–Ar age of ∼4.50 Ga for one IAB and ∼4.32 Ga for two other IAB silicates, and summarized older ages from the literature. Vogel and Renne (2008) reported Ar–Ar ages of plagioclase separates of four IAB silicates, including three previously measured. Vogel and Renne (2008) plotted all available Ar–Ar ages of IAB and winonaite meteorites in a descending sequence of ages, and

214

D.D. Bogard / Chemie der Erde 71 (2011) 207–226

Fig. 4. Ar–Ar ages of silicate from IAB iron meteorites arranged by descending age. Multiple samples were analyzed for some meteorites. Reprinted from Vogel and Renne (2008), copyright 2008, with permission from Elsevier.

their original graph is reproduced in Fig. 4. Note, however, these authors preferred to use different K decay parameters compared to those commonly used by other laboratories reporting meteorite ages, and the ages plotted in Fig. 4 are higher by 22 Ma compared to values that would be obtained with the decay parameters commonly used over the past >30 years (Steiger and Jäger, 1977). (This issue is discussed further in Section 5 on absolute K–Ar ages). Using the K decay parameters in common use, the Ar–Ar ages for IAB meteorites shown in Fig. 4 would range over ∼4.53–4.32 Ga, or a span of ∼200 Ma. Even different samples from the same meteorite show variations in Ar–Ar ages by some tens of Ma (Fig. 4). If the sequence of Ar–Ar ages shown in Fig. 4 represented a parent body with a simple history, the ages might be interpreted as representing K–Ar closure during prolonged metamorphic cooling, coupled with sample origins from different depths, as discussed above for H chondrites. However, several observations indicate that the IAB parent body experienced a more complex history involving impact heating and mixing (Benedix et al., 2000). For example, a hand-specimen-sized piece of Caddo County IAB, not a fragmental breccia, shows distinct areas of a gabbro-like material with transitional margins against both fine-grained ultramafic (winonaite-like) material and coarse metal veins (Takeda et al., 2000). This meteorite appears to have formed by impact mixing of materials which experienced different temperatures and degrees of differentiation. An analogous mixture of phases is present in the Rose City H chondrite, which experienced intense impact heating long after formation of its parent body (Bogard and Hirsch, 1980). The impact that affected Caddo County may have disrupted and reassembled the IAB parent body when it was partially melted (Benedix et al., 2000). The variations in Ar–Ar ages for IABs (Fig. 4) would thus be explained as measuring different closure times of individual samples in different thermal environments, and not as resetting by the impact event itself. When did the impact disruption of the IAB parent body occur? For consistency between the I–Xe and Ar–Ar ages, Bogard et al. (2005) suggested a relatively early impact mixing event >4.52 Ga ago, which rapidly cooled the silicates to below the closure temperature for the I–Xe chronometer, followed by relatively slow cooling through K–Ar closure. This slower cooling environment could have determined the different Ar–Ar ages among meteorites and would be consistent with cooling rates determined from Ni concentrations profiles in metal of seven IABs (Herpfer et al., 1994). However, Vogel and Renne (2008) found variations in Ar–Ar ages both among different inclusions of the same meteorite and among plagioclase

grains of different sizes within a given inclusion. They suggested that these age variations were produced through cooling of individual plagioclase grains in different parts of the parent body, which were then brought together during impact mixing. They suggested that to preserve different Ar–Ar ages among different grain sizes, mixing of dispersed silicates would have to occur at a very fine scale and that the mixed material would have to cool quickly to below K–Ar closure in order to preserve these age differences. One serious question about this suggested process is whether separate metamorphic cooling histories followed by impact assembly into individual silicate inclusions could have occurred on a scale as small as individual plagioclase grains ∼0.1 mm in size, and in so doing maintain the observed differences in Ar–Ar ages. Further, one could ask whether the original plagioclase grain sizes were preserved during their separation by crushing, sieving, magnetic and density separation, and hand picking. If not, measured grain sizes may not have been a controlled parameter in explaining Ar–Ar age variations among different grain sizes. Vogel and Renne (2008) suggested that the impact event would have to be shifted toward the younger Ar–Ar ages and that the impact occurred ∼4.47–4.49 Ga ago. In summary, diverse studies strongly indicate that the IAB parent asteroid experienced a complex thermal history involving internal metamorphism, partial melting, and relatively early impact disruption and reassembly. It remains unclear exactly when impact disruption of the IAB parent body occurred, but it must have been >4.47 Ga ago when that body was still relatively warm. To preserve old I–Xe ages would require moderate, early cooling. To explain differences in Ar–Ar ages among different inclusions of the same meteorite would require either differences in closure temperature during cooling after early impact mixing, or closure of these ages prior to the impact and rapid cooling after impact mixing so as not to reset the ages. Another question concerns the few younger Sm–Nd and Rb–Sr ages for Caddo County (∼4.50–4.53 Ga), as these chronometers presumably have higher closure temperatures compared to K–Ar. Thus, considerable uncertainty exists over the early and obviously complex history of the IAB parent asteroid. 4.5. IIE iron silicates The IIE iron meteorites (Wasson and Wang, 1986) contain abundant silicates of quite variable chemical composition and mineralogy and have an oxygen isotopic composition similar to that of H chondrites. Many silicate inclusions have igneous textures indicative of melting and crystallization, whereas others preserve unmelted chondritic textures (see Ruzicka et al., 1999, and references therein for descriptions of individual meteorites). The IIE irons have been divided into two groups, based on chemistry and mineralogy of their silicates. One group has near-chondritic silicate inclusions (e.g., Netschaëvo, Techado, and Watson), and the other group contains silicates that are igneous, fractionated, or highly evolved and alkali-rich (e.g., Weekeroo Station, Colomera, Kodaikanal, and Miles). For example, Colomera contains a sanidine crystal cms in length (Wasserburg et al., 1968) and Watson contains fragments of feldspar antiperthite with orthoclase lamellae containing 5.7% K, which exsolved from albitic feldspar (Olsen et al., 1994). The mechanism by which IIE silicates experienced melting and differentiation is poorly defined, and three models have been proposed: (1) near-surface mixing of metal and H chondrite melts; (2) mixing of metal and silicate melts at the core–mantle boundary of a chondritic parent body; and (3) planetary differentiation and impact melting (Ruzicka et al., 1999; Snyder et al., 2001, and references therein). Measured Ar–Ar ages of silicates from IIE iron meteorites are presented in Fig. 5 and Table 1. Early reported ages determined by various radiometric chronometers indicate that silicate

D.D. Bogard / Chemie der Erde 71 (2011) 207–226

by simple heating (Burnett and Wasserburg, 1967). Because the Hf–W data for Watson silicate do not show a tungsten isotopic anomaly, metal–silicate mixing occurred after the decay of 182 Hf (half-life 9 Ma), and the metal and silicate likely derived from separate bodies, possibly an iron and a H chondrite (Snyder et al., 2001). The collision of these two bodies likely produced the melting and mixing. The time of this event, 3676 Ma ago, is similar to the times of other major impact events recorded in eucrites and some chondrites, as is discussed in Sections 4.8 and 4.9. Why some IIE meteorites preserved their old ages, whereas ages for other IIE meteorites were reset 3676 Ma ago in a large impact event is unclear, but implies that the late impact did not heat all parts of the parent objects to the same degree. The very young 0.7 Ga Sm–Nd age for Weekeroo Station is not apparent in much older Ar–Ar, I–Xe, and Rb–Sr ages and does not likely signify a thermal event.

4.7 4.6 4.5 4.4 4.3

Age, Ga

215

4.2 4.1 4 3.9 3.8 3.7 3.6

4.6. Other iron silicates 0

1

2

3

4

5

6

7

8

9

10

11

12

Meteorite Fig. 5. Ages of IIE iron silicates (1–8) and three other metal rich meteorites (9–11) as measured by Ar–Ar (diamonds) and other radiometric chronometers (circles). Meteorites are: 1 = Watson, 2 = Kodaikanal, 3 = Natschaëvo, 4 = Colomera, 5 = Weekeroo Station, 6 = Miles, 7 = Techado, 8 = Tarahumara, 9 = Sombrerete, 10 = Portales Valley, 11 = NWA 176. Table 1 gives the Ar–Ar age uncertainties and data sources.

in Colomera, Weekeroo Station, and possibly Miles are relatively old, but that silicate in Kodaikanal and Netschaëvo have a much younger age of ∼3.7 Ga. Niemeyer (1980) reported an old 129 I–129 Xe age of 4.555 Ga for Weekeroo Station (but see Hohenberg and Pravdivtseva, 2008, for recalibration of monitor age), whereas Netschaëvo did not yield an I–Xe isochron. Kodaikanal gave a U–Pb age of 3676 ± 3 Ma (Göpel et al., 1985). Bogard et al. (2000) summarized these early literature ages and reported new Ar–Ar ages of silicate in four IIEs (Table 1). They found that Watson also had an age of 3676 ± 7 Ma, but that Colomera, Miles, and Techado had relatively old Ar–Ar ages of 4.41–4.49 Ga. These two age groups contain meteorites of both the fractionated and non-fractionated groups (above) as defined by chemical and mineral composition. Takeda et al. (2003) reported on the mineralogy of a large inclusion containing albitic plagioclase in the Tarahumara IIE meteorite and determined an Ar–Ar age of 4476 ± 18 Ma. Snyder et al. (2001) reported data for the Hf–W, Sm–Nd, and Rb–Sr radiometric systems for Weekeroo Station, Watson, and Miles, and they also reviewed earlier determined ages for IIEs and some other iron groups. The Sm–Nd and Rb–Sr data indicate a complex petrogenetic history likely involving impact melting and differentiation, and correlation of ages and specific events is difficult. Snyder et al. (2001) suggested typically old Rb–Sr ages of 4.3–4.5 Ga, but Sm–Nd model ages of 4.27 Ga for Miles, 3.04 Ga for Watson, and 0.70 Ga for Weekeroo Station. The relatively old (>4.5 Ga) radiometric ages for a few IIE meteorites indicate that the silicates differentiated and were mixed with metal during the time period that meteorite parent bodies were forming >4.5 Ga ago. The slightly younger Ar–Ar ages of 4.47–4.49 Ga for Colomera, Weekeroo Station, Tarahumara, and Techado probably represent K–Ar closure times during cooling on the parent body following this silicate–metal mixing. These ages are similar to those for some chondrites and may imply that the IIE parent body was similar in size. The close concordance of Ar–Ar, Pb–Pb, and Rb–Sr ages at 3.7 Ga for Watson, Kodaikanal, and Netschaëvo (particularly the precise 3676 ± 7 Ma Ar–Ar age for Watson and the 3676 ± 3 Ma U–Pb age for Kodaikanal) imply a second stage of melting, differentiation, and mixing for these three meteorites at this time. Conceivably, the young Ar–Ar ages could have been reset by strong heating alone. However, the young Rb–Sr ages found for different silicate inclusions of Kodaikanal cannot be explained

The chronologies determined for silicate in three other ironrich meteorites are informative, and help define thermal histories for these meteorites. Sombrerete is an unclassified iron containing evolved silicate inclusions similar to those in IIEs and likely experienced melting through internal metamorphism and impact mixing (Ruzicka et al., 2006). Northwest Africa 176, also unclassified, consists of polymineralic silicate inclusions of chondrite-like composition embedded in metal, which also likely experienced internal melting and/or impact melting and mixing (Liu et al., 2001). Portales Valley is H chondrite material mixed with abundant metal and likely also formed as a result of impact melting and metamorphism (Kring et al., 1999; Rubin et al., 2001; Ruzicka et al., 2005). Portales Valley is the only known chondrite to show a Widmanstäten structure in its metal, indicative of very slow cooling. These three meteorites probably derived from separate parent bodies, both from each other and from the other meteorite types discussed here. Bogard and Garrison (2009a) reported Ar–Ar and I–Xe ages of silicates in each of these meteorites (Table 1 and Fig. 5). The I–Xe ages for Portales Valley, Sombrerete, and NWA 176 are 4559.9 ± 0.5 Ma, 4561.9 ± 1.0 Ma, and 4544 ± 7 Ma, respectively, relative to an age of 4562.3 ± 0.4 Ma for the Shallowater age monitor. The Ar–Ar age for Sombrerete of 4541 ± 12 Ma is slightly younger than the I–Xe age, but the two ages overlap at the two sigma level. Although the ∼20 Ma age difference could measure a slow cooling history, this would not be consistent with the conclusion of Ruzicka et al. (2006) that Sombrerete cooled quickly after impact breakup and reassembly of a partially molten body. The younger Ar–Ar age may reflect a two-stage cooling history – early rapid cooling to close the I–Xe chronometer, followed by slow cooling after mixing of material having different temperatures during which the Ar–Ar chronometer closed. Bogard and Garrison (2009a) showed that I–Xe has a significantly higher closure temperature than K–Ar in some samples. Alternatively, the difference in Ar–Ar and I–Xe age may not be real, and corrections to K decay parameters in current use might bring the two ages into agreement (see Section 5). Two different samples of Portales Valley gave Ar–Ar ages which slightly differ, 4477 ± 11 and 4458 ± 16 Ma. Although these ages overlap within their uncertainties, Bogard and Garrison (2009a) suggested that measured differences in Ar diffusivities (Fig. 2) and slow cooling in the meteorite parent body also could explain the age difference. Using the Portales Valley Ar–Ar and I–Xe ages, along with the conclusion that its silicate was heated to 940–1150 ◦ C during impact of an already hot body (Ruzicka et al., 2006), Bogard and Garrison (2009a) presented a simple thermal model for the cooling history (Fig. 6). In this model, early cooling at a rate of ∼10 ◦ C/Ma (at 1000 K) from an initial temperature of ∼1350 K at a time 4560 Ma ago (the I–Xe age) is consistent with the two Ar–Ar ages, but not

216

D.D. Bogard / Chemie der Erde 71 (2011) 207–226

Bogard and Garrison, 2009a). Thus, in all three of these iron silicates, collisional heating and mixing occurred early, >4.53 Ga ago.

1400 1300

1213-1423 K

Portales Valley

4.7. Mesosiderites

Cooling History 1200

Temperature, Kelvin

1100

11oC/Ma 8oC/Ma 10oC/Ma

1000

I-Xe Age

900 800 700

Ar-Ar 600

4.4oC/Ma 3.3oC/Ma 4.2oC/Ma

Ages

500 400 300 4.42

4.44

4.46

4.48

4.50

4.52

4.54

4.56

4.58

Time, Gyr B.P. Fig. 6. Possible thermal histories for Portales Valley defined by plotting Ar–Ar ages (dark circles) and I–Xe age (diamond) against calculated closure temperatures. The lines show three cooling models starting 4.56 Ga ago. One model assumes cooling continuously from an initial temperature of ∼1350 K, whereas two other models assume rapid initial cooling below the I–Xe closure temperature, followed by cooling at rates that satisfy the two measured Ar–Ar ages (filled circles) or Ar–Ar ages increased by 20 Ma to account for possible errors in 40 K decay parameters (open circles). Cooling rates at I–Xe and Ar–Ar closure are indicated for each cooling model. Figure reproduced from Bogard and Garrison (2009a).

with the I–Xe age. Instead, these authors assumed that the parent body impact and mixing occurred 4560 Ma ago and that the mixed components rapidly reached thermal equilibrium at a temperature slightly below the I–Xe closure temperature. In this case, slow cooling at a rate of ∼4 ◦ C/Ma (at 500 K) begins near the plotted I–Xe point and is consistent with both Ar–Ar ages as measured or even if these two ages are increased by ∼20 Ma to correct for K decay parameters (see Section 5). This second thermal model is consistent with conclusions of Ruzicka et al. (2006) for the formation history of Portales Valley and implies a parent body at least ∼30 km in diameter. This example again demonstrates how Ar–Ar ages, Ar diffusion data, and other radiometric ages can help define the thermal history of a meteorite. Using Ar–Ar and I–Xe ages, Bogard and Garrison (2009a) also presented a thermal cooling model for NWA 176, which was further constrained by a relatively fast cooling rate of ∼103 C◦ /Ma estimated from Ni diffusion profiles in the metal (Liu et al., 2001). Although the I–Xe age of 4544 ± 7 Ma was not as well defined, I–Xe closure in NWA 176 appears to have occurred ∼60 Ma after I–Xe closure in Portales Valley. Liu et al. (2001) estimated that the silicate was heated to 1100 ± 60 K. A thermal model that initiated cooling from 1100 K at a time 4544 Ma ago would be consistent with the I–Xe age and the Ni cooling rate, but not with the Ar–Ar age. Cooling initiated 4556 Ma ago, a more reasonable time for initial parent body formation, would be consistent with the I–Xe and Ar–Ar ages, but not with the metallographic cooling rate. However, if we assume the measured Ar–Ar age is ∼20 Ma too young due to errors in the K decay parameters (see Section 5), then all data become consistent with cooling that began as a result of parent body collisional heating and mixing ∼4554 Ma ago (see Fig. 11 in

Mesosiderites are a significant group of stony-iron meteorites composed of ∼17–90% metallic iron–nickel, with the remainder being fragments of igneous rock (basalt, gabbro, and dunite). Mesosiderite silicates are generally similar to basaltic howardites/eucrites (next section), but differ in compositional detail (Mittlefehldt et al., 1998). Mesosiderites appear to be mechanical mixtures of core and crustal materials from a differentiated parent object, and both impact and internal mechanisms have been suggested for their origins (Hewins, 1983; Mittlefehldt, 1990). Mesosiderite silicates have the same oxygen isotopic composition as eucrites (Clayton and Mayeda, 1996; Greenwood et al., 2006), and some workers have suggested a common parent body, whereas other workers advocate separate parent bodies. Based on mineral textures, it was concluded that mesosiderite silicates cooled relatively quickly at temperatures above ∼800 ◦ C (>10 ◦ C/103 years; Ganguly et al., 1994; Ruzicka et al., 1994). In contrast, Ni diffusion profiles in the metal suggest extremely slow cooling rates at much lower temperatures of ∼400 ◦ C (∼0.1 ◦ C/Ma, Powell, 1969; ∼0.02 ◦ C/Ma, Haack et al., 1996; ∼0.1–0.5 ◦ C/Ma, Yang et al., 1997). (See additional literature references to mesosiderite cooling history in Bogard and Garrison, 1998; Mittlefehldt et al., 1998; Scott et al., 2001). Scott et al. (2001) proposed that mesosiderites formed when a 200–400 km asteroid with a molten core was collisionally disrupted by another large object, causing molten metal globules to form around small, cool fragments of the basaltic crust. Following this disruption, the body reaccreted, leaving mesosiderites at depth, and producing the slow metal cooling rates. Obviously the thermal history of mesosiderites is an important parameter in understanding their formation. Mesosiderite silicates apparently formed relatively early, although precise chronological data are sparse. One mesosiderite gave ages of Pb–Pb = 4.555 ± 0.035 Ga, U–Pb = 4.560 ± 0.030 Ga, Rb–Sr = 4.542 ± 0.203 Ga, and Sm–Nd = 4.533 ± 0.094 Ga (Brouxel and Tatsumoto, 1991). Ireland and Wlotzka (1992) reported a 207 Pb/206 Pb age for zircon in a eucritic clast from one mesosiderite to be 4.563 ± 0.015 Ga. Stewart et al. (1994) reported Sm–Nd isochron ages for three clasts from two mesosiderites to be 4.52 ± 0.04, 4.48 ± 0.09, and 4.48 ± 0.19 Ga. Crozaz and Tasker (1981) determined 244 Pu fission track ages for six mesosiderites to be ∼3.9–4.2 Ga. Closure temperatures for track retention is generally much lower (∼100 ◦ C) than for these other chronometers. Bogard and Garrison (1998) and Bogard et al. (1990) found that Ar–Ar ages for 20 samples from 14 different mesosiderites were all significantly younger than 4.5 Ga. The Ar–Ar age spectra for these samples show varying degrees of upward slope, i.e., the ages steadily increase with increasing 39 Ar release. The age spectra for some samples slope very little, whereas others slope by a few hundreds of Ma between 10% and 90% 39 Ar release. The average age and the ages at 10% and 90% 39 Ar release for each sample are shown in Fig. 7. The average ages for 19 of the samples are generally similar to one another (3.8–4.1 Ga), and the average age of these 19 is 3.94 ± 0.10 Ga. Such young ages and sloped spectra could, in principle, have been produced either by varying amounts of diffusive loss of 40 Ar in one or more thermal events long after the parent body formed, or by very slow cooling, such that different K-bearing phases and grains sizes closed at different times. The release of argon as a function of temperature for these samples was used to calculate Ar diffusion properties and closure temperatures for the K–Ar chronometer (Bogard et al., 1990). Calculated activation energies for most samples range over 18–52 kcal/mole and closure temperatures over 50–340 ◦ C, for a cooling rate of 1 ◦ C/Ma.

D.D. Bogard / Chemie der Erde 71 (2011) 207–226

Fig. 5 for Portales Valley, except that mesosiderite cooling occurred much slower and at greater burial depth. This formation scenario for mesosiderites appears consistent with scenarios suggested by Haack et al. (1996) and Scott et al. (2001). After closure of Ni diffusion and all chronometers, the mesosiderite parent body must have been substantially disrupted by another collision in order to bring deep seated material nearer the surface. Finally, even later and smaller collisions initiated cosmic ray exposure, which for several mesosiderites range over ∼9–300 Ma (Terribilini et al., 2000), but did not reset the K–Ar chronometers.

Bondoc Vaca Muerta basalt V. M. gabbro V.M. Plag. (DM) V. M. Plag. (#5) Mt. Padbury Mt. Padbury Plag. Budulan Plag. Morristown Patwar gabbro Simondium QUE-86900 Esterville #1 Esterville #2 Emery Patwar basalt Hainholz Lowicz

4.8. HED meteorites

Pinnaroo Veramin 2.6

2.8

3.0

3.2

3.4 39

217

3.6

3.8

4.0

4.2

4.4

4.6

40

Ar- Ar Age, Ga

Fig. 7. Ar–Ar ages of mesosiderites. Most age spectra are significantly sloped. The diamonds indicate the average age for each sample and give an overall average age of 3.94 ± 0.10 Ga (omitting Pinnaroo). The ends of the horizontal lines give the ages at 10% and 90% of the 39 Ar release and indicate the degree of slope for each spectrum. Figure reproduced from Bogard and Garrison (1998).

Samples with strongly sloped age spectra tend to have higher activation energies for Ar diffusion and higher closure temperatures, as would be expected if differences in Ar diffusion characteristics among samples produced the sloped age spectra (Bogard and Garrison, 1998). One can independently use the K–Ar closure temperatures and sloped spectra ages to estimate cooling rates for the silicates through the temperatures of K–Ar closure. The cooling rate derived from the Ar data is ∼0.2 ◦ C/Ma (Bogard and Garrison, 1998), in excellent agreement with the cooling rate derived from Ni diffusion in metal at slightly higher temperatures (0.1–0.5 ◦ C/Ma; Yang et al., 1997). Such very low cooling rates and variable sloped age spectra are unique among meteorite types. As discussed above, the very nature of mesosiderites implies a large-scale impact event involving one or more differentiated parent bodies, which mixed near-surface silicate with deep-seated metal (Mittlefehldt et al., 1998; Scott et al., 2001). Bogard and Garrison (1998) interpreted the Ar–Ar data in the following scenario. In order to preserve the old silicate ages obtained by Pb–Pb, Sm–Nd, and Rb–Sr and the relatively fast cooling rates reported for silicate above ∼800 ◦ C, the impact-mixing event probably occurred relatively early (∼4.4–4.5 Ga ago) and quenched the silicate below the closure temperatures of these other chronometers. If the impact mixing event occurred later, it could not have heated the silicate above these closure temperatures. In order to explain the very slow cooling rates obtained from Ni diffusion in metal and from the Ar–Ar data, as well as the young Ar–Ar and 244 Pu fission track ages, the silicate–metal mixture was deeply buried during parent body reaccretion. The initial temperature of the buried mesosiderites did not possibly exceed ∼500 ◦ C, which is above the closure temperatures for Ni and Ar diffusion and for track retention, but below the closure temperatures of these other chronometers. The very slow cooling following this burial caused the K–Ar and track chronometers to remain open for several hundred Ma, and differences in Ar diffusion characteristics of samples produced the various sloped Ar–Ar age spectra. This cooling history proposed for mesosiderites is somewhat analogous to that proposed in

The HED (howardite–eucrite–diogenite) meteorites are igneous rocks which derive from a differentiated parent body thought to be the large (∼520 km diameter) asteroid 4-Vesta (Binzel and Xu, 1993; Mittlefehldt et al., 1998, and references therein). Eucrites, mostly basaltic, comprise the upper crust; diogenites, mainly composed of orthopyroxene, comprise the lower crust; and howardites are impact-produced, brecciated mixtures of these two types. The cumulate eucrites are grabbroic in texture and probably formed in basaltic intrusions at significant depth. Petrographic evidence, especially in pyroxene composition, show that most eucrites, both basaltic and cumulate, experienced varying degrees of internal metamorphism after formation (Takada and Graham, 1991; Yamaguchi et al., 1997), and many eucrites experienced impact brecciation. Analyses of short-lived chronometers in eucrites indicate that crust–mantle formation occurred ∼4.57 Ga ago (Lugmair and Shukolyukov, 1998; Srinivasan et al., 1999, 2007). Analyses of the long-lived chronometers of U–Pb, Sm–Nd, and Rb–Sr suggest a range of eucrite formation ages, from ∼4.57 Ga down to ∼4.4 Ga (Misawa et al., 2005a; Wadhwa et al., 2006). Extended periods of eucrite formation and of internal metamorphism both have been suggested to explain the observed range of ages. Several new Ar–Ar ages for eucrites and summaries of HED ages in the literature were presented by Bogard (1995) and Kunz et al. (1995). Bogard and Garrison (2003, 2009b) presented additional Ar–Ar ages. Some ages, especially earlier measurements, are poorly defined, possibly because many HED meteorites are breccias of clasts with different thermal histories. Ages of individual clasts from eucrites and howardites tend to be better defined than those of whole rock samples. A summary of 46 Ar–Ar ages for eucrites and howardites is presented in Fig. 8 as an age probability plot. To produce this plot, the age and 1-sigma uncertainty for each age determination is drawn as a Gaussian curve, where the peak of the curve gives the age and the spread of the curve measures the age uncertainty, with taller, narrower peaks representing smaller age uncertainties (Bogard and Garrison, 2003). The y-axis gives the relative probability for a given age. Two examples of individual age determinations, with 1-sigma age uncertainties of ±20 and ±50 Ma, are shown in Fig. 8 as the dotted curves. (To facilitate comparison with other plotted data, the probabilities of these two dotted curves have been increased by a factor of three.) The heavy, solid curve is the sum probability curve for these 46 analyses, except that the sum curve between 4.4 and 4.5 Ga has been divided by a factor of three to facilitate comparison. Without this normalization, this high-age curve would extend to a probability of 0.3 on the y-axis. The double age peak at 3.7–3.8 Ga is defined by eight individual overlapping age peaks with ages 3.68–3.81 Ga, and the narrow age peak at 4.48 Ga is defined by nine individual ages between 4.47 and 4.49 Ga and a tenth age at 4.51 Ga. The purpose in plotting Ar–Ar ages this way is to identify possible groupings of ages, and Fig. 8 shows that certain degassing ages are much more probable than others. For example, there are strong probabilities for K–Ar resetting events at 3.45–3.55, ∼3.8, 3.9–4.0, and ∼4.48 Ga (>90% probability of a degassing event between 4.47 and 4.49 Ga). However, there are low probabilities for

218

D.D. Bogard / Chemie der Erde 71 (2011) 207–226

Relative Probability for a Given Age

0.100 0.090 0.080

36 brecciated Eucrites

10 unbrecciated Eucrites (divided 3)

0.070 0.060 0.050 0.040 0.030 0.020 0.010 0.000 3300 3400 3500 3600 3700 3800 3900 4000 4100 4200 4300 4400 4500 4600

Ar-Ar AGE, Ma Fig. 8. Summed Gaussian age probability curve of Ar–Ar ages for 46 eucrites, including a few eucritic clasts in howardites. To produce this plot, the age and 1-sigma uncertainty for each age determination is drawn as a Gaussian curve, where the peak of the curve gives the age and the spread of the curve measures the age uncertainty, with taller, narrower peaks representing smaller age uncertainties. Gaussian age plots for two individual samples are shown as dotted curves. Meteorites defining the sharp summed age peak at 4.48 Ga are unbrecciated, and the height of this peak has been divided by three to facilitate comparison with the other data. Data from Bogard and Garrison (2003), with added subsequent data obtained at the NASA Johnson Space Center.

degassing events at ∼3.6 and 4.1–4.4 Ga, and no Ar–Ar ages occur below 3.3 Ga. The Ar–Ar ages plotted in Fig. 8 were all probably reset by impact heating and imply a whole new dimension to the thermal history of some meteorite parent bodies, as discussed below. Early analyses of radiometric ages of returned lunar rocks led to the suggestion that the lunar surface experienced a period of strong impact heating ∼3.9 Ga ago, long after its formation (Tera et al., 1974; Turner, 1977). The origin of these large impacting bodies and the time period over which they struck the Moon has remained controversial (e.g., Hartmann et al., 2000; Ryder, 2002). The range of Ar–Ar and some other radiometric ages for impact-heated lunar highland rocks is rather broad, ∼3.4–4.2 Ga (e.g., Turner, 1977; Bogard, 1995; Bogard and Garrison, 2003; Norman et al., 2010). Still, arguments have been made for both a relatively narrow time interval of <100 Ma for most of the impactor flux, and for a much longer time of several hundreds of Ma (Hartmann et al., 2000; Ryder, 2002). The issue has major implications far beyond the Moon. First, other bodies in the solar system (i.e., Earth, Mercury, Mars, large asteroids) were likely similarly bombarded, and this period of heavy bombardment played a major role in shaping their surfaces, just as it did for the Moon. Secondly, recent dynamical calculations suggest that orbits of the giant planets evolved early in solar system history, and that their migration, some toward and others away from the Sun, gravitationally disrupted both the inner portion of the Kuiper belt of objects and the asteroid belt, thereby producing the large objects that bombarded the Moon (Gomes et al., 2005; Bottke et al., 2007; Minton and Malhotra, 2009, and references therein). Clearly, defining the time period over which this early, intense bombardment occurred on the Moon and elsewhere is a first order planetary goal. Bogard (1995) argued that the Ar–Ar ages of many eucrites were reset by impact heating during this early, intense bombardment of the inner solar system and that the HED parent body, probably Vesta, could give an independent measure of the time period over which this bombardment occurred. The Ar–Ar ages plotted in Fig. 8 indicate that large impacts on Vesta occurred over a similar time period (3.5–4.1 Ga) as did impacts that reset lunar rock ages and formed lunar-sized basins. Because resetting of K–Ar ages by impact requires the sample to reside for a significant time in relative hot ejecta (or beneath the crater floor), only rather large impacts

are likely to reset K–Ar ages. For example, rocks ejected from a few specific lunar craters ≤1 km in diameter and returned by Apollo astronauts do not show K–Ar ages reset by these craters, and K–Ar ages of meteorites do not register the much later collisional events that initiated their cosmic ray exposure as small objects in space (Eugster, 2005). Thus, Bogard (1995) suggested that K–Ar ages reset by this early solar system bombardment would likely occur only for meteorites that originated from near-surface regions of relatively large parent bodies, as small parent bodies would be destroyed by large impacts. Among Ar–Ar ages of meteorites, eucrites and howardites commonly show early impact resetting and several ordinary chondrites also indicate such resetting (Bogard, 1995; Swindle et al., 2009; see Section 4.9). However, other meteorite groups show little evidence for impact resetting of Ar–Ar ages between 3.4 Ga and 4.1 Ga (IIE iron silicates are an exception). Of course, even Vesta could not have survived the very large impactors that formed the larger lunar basins, and the ages of Fig. 8 represent impacts significantly smaller than those that formed the large lunar basins. If the eucrite Ar–Ar ages of 3.4–4.1 Ga were produced by the same early, intense bombardment of the inner solar system that formed the lunar basins, what produced the narrow peak in Ar–Ar ages of eucrites at 4.48 Ga (Fig. 8)? It is significant that the 4.48 Ga eucrites are all unbrecciated, while the younger eucrites are all breccias. Somehow these eucrites with older ages escaped the later impact bombardment and brecciation events. Because eucrites in the 4.48 Ga group include samples that experienced different degrees of metamorphism and include a few cumulate eucrites, it is unlikely their Ar–Ar ages reflect cooling from this metamorphism, unless that cooling was sudden and produced by meteorite ejection from significant depths within the relatively warm parent body. Vesta is known to possess a few very large impact craters (up to ∼460 km diameter), and one of these events 4.48 Ga ago may have reset these Ar–Ar ages and a few Pb–Pb and Sm–Nd ages. Possibly, this early impact event ejected large fragments away from Vesta (Binzel and Xu, 1993; Sykes and Vilas, 2001), and these became the parent objects for those eucrites with older Ar–Ar ages (see discussion in Bogard and Garrison, 2003). 4.9. Late collisional events Many ordinary chondrites show younger K–Ar ages of <1 Ga, and K–Ar and U,Th–4 He ages of many L chondrites cluster around 0.5 Ga (Kirsten et al., 1963; Heymann, 1967; Wasson and Wang, 1991). A significant fraction of ordinary chondrites are breccias and/or exhibit shock effects produced by impact (Rubin et al., 1983; Stöffler et al., 1991). Some K–Ar ages were measured on chondrite impact melts. Thus, these K–Ar and U,Th–4 He ages have been attributed to resetting during impact heating on the chondrite parent bodies. The amount of 40 Ar degassing produced by this impact heating ranges from partial to essentially complete, and the reliability of derived impact ages varies considerably (e.g., see Korochantseva et al., 2007). Interestingly, these young K–Ar ages primarily occur among ordinary chondrites, and they are rare among other meteorite types. The review by Bogard (1995) summarized younger radiometric ages for chondrites available in the literature and noted that they mostly fall into two broad clusters – ∼3.4–4.0 Ga and <1.3 Ga. The older cluster of chondrite ages is attributed to the same early impactor population in the inner solar system that reset the Ar–Ar ages of shocked eucrites. This explanation is discussed in Section 4.8 and is not repeated here. However, the younger chondrite ages have a different cause. In essentially no case is the cosmic ray exposure age of a chondrite as old as its K–Ar age. This indicates that the impact events that reset these younger K–Ar ages did not launch the meteorites into space as small objects of less than a few meters in diameter. Rather, the impacts that reset K–Ar ages

D.D. Bogard / Chemie der Erde 71 (2011) 207–226

Fig. 9. Summed Gaussian age probability curve (heavy line) and individual sample curves (dashed lines) for Ar–Ar ages of many H chondrites. Essentially no ages occur between 1.3 and 3.5 Ga. Reprinted from Swindle et al. (2009), copyright 2009, with permission from the Meteoritical Society.

of chondrites presumably left the meteorites buried at least a few meters deep within the parent body regolith. Since the review of younger Ar–Ar ages of chondrites (Bogard, 1995), a number of additional impact-reset Ar–Ar ages of chondrites have been documented (e.g., Kring et al., 1996; Kunz et al., 1997; Grier et al., 2004; Swindle et al., 2006, 2009; Korochantseva et al., 2007; Swindle and Kring, 2008). A summary probability plot of Ar–Ar ages of shocked H chondrites is presented in Swindle et al. (2009) and is reproduced as Fig. 9. This plot was constructed in an analogous manner to the eucrite age probability plot in Fig. 8, and both individual age curves and a sum curve are shown. Note that many Ar–Ar ages of H chondrites plot at <1.5 Ga, and all but two of these plot at <1.0 Ga. There exists a concentration of reset ages over ∼300–600 Ma. No ages plot between 1.5 and 3.5 Ga. Seven H chondrites show Ar–Ar ages of ∼3.6–4.1 Ga and three more give ages during the parent body metamorphism period of >4.36 Ga. It has been proposed that Ar–Ar ages of ∼3.5–4.0 Ga among chondrites were reset by the same heavy bombardment that reset Ar–Ar ages of many eucrites (Bogard, 1995; Swindle and Kring, 2008; Swindle et al., 2009). Among LL chondrites, few Ar–Ar ages are <2 Ga, but there are a few ages of >3.8 Ga (Swindle and Kring, 2008; Dixon et al., 2004). Only a very few ordinary chondrites give Ar–Ar ages indicating shock-melting in the time period during which parent bodies were still cooling from metamorphism >4.35 Ga ago (Dixon et al., 2004; Benedix et al., 2008; Weirich et al., 2011). Many L chondrites show impact-reset Ar–Ar ages of <1.5 Ga. To a greater extent than H chondrites, Ar–Ar ages of many shocked L chondrites cluster between 0.45 Ga and 0.60 Ga (Swindle and Kring, 2008). Because so many L chondrites show K–Ar and U,Th–4 He ages of ∼0.5 Ga, it has been argued that a single impact near that time disrupted the whole L chondrite parent body (Heymann, 1967; Keil et al., 1994; Kunz et al., 1997). Strengthening this conclusion was the discovery of large quantities of L chondrite material in ∼0.48 Ga old Ordovician sediments in more than one terrestrial location (Schmitz et al., 2003; Heck et al., 2010, and references therein). This L chondrite breakup event was precisely Ar–Ar dated in the Ghubara L5 chondrite at 470 ± 6 Ma, consistent with the age of Ordovician sediments that contain the L chondrite material (Korochantseva et al., 2007). It is likely that those L chondrite Ar–Ar ages that are slightly older than 0.47 Ga reflect incomplete degassing by the event, but a few ages younger than 0.47 Ga may date later impacts. Ar–Ar ages of H chondrites show a wider range,

219

and it is not clear if any of these H chondrite events are associated with the 0.47 L chondrite event. Collectively, Ar–Ar ages of H, L, and LL chondrites suggest the occurrence of multiple large impact events. The concept of a breakup of the L chondrite parent body ∼0.47 Ga ago has produced a large amount of related literature, and it has even been suggested that infall of this material onto Earth may have contributed to the great diversification of living organisms that occurred at about this time (Schmitz et al., 2008). Why among chondrites are impact reset K–Ar ages of <1.0 Ga common, whereas reset K–Ar ages in the range of 1.5–3.5 Ga are practically non-existent? An impact on a parent object may eject meteorites into much more eccentric orbits or transfer fragments into proximity of an orbital resonance with Jupiter, from which they can be gravitationally perturbed into Earth-crossing orbits and then captured by Earth. The mean lifetime of asteroidal objects in orbital resonances has been calculated to be only a few million years, and the lifetime of objects in Earth-crossing orbits is similar (Gladman et al., 1997). Movement of an asteroidal object from within the belt into an orbital resonance is variable and depends on a number of factors. Breakup of the L chondrite parent body may have ejected large quantities of material directly into a gravitational resonance, and experimental data on the Ordovician L chondrite material indicates that these were captured by Earth relatively quickly (Heck et al., 2004). However, L chondrites involved in this 0.47 Ga event still fall on Earth, and the intermediate parent objects of these meteorites must have taken much longer to move into an orbital resonance. Most L chondrites have cosmic ray exposure ages between 4 and 50 Ma, much shorter than the 0.47 Ga breakup event, and there is no obvious difference in the distribution of CRE ages among L chondrites possessing low and high 40 Ar contents (Marti and Graf, 1992). H and LL chondrites generally show similar distributions of CRE ages (Marti and Graf, 1992). Most L chondrites must have resided in relatively large, intermediate parent objects after the 0.47 Ga event and during most of their migration time within the asteroidal belt. It would then require another, smaller collisional event to eject meteorites from these secondary objects, expose these L chondrites to cosmic rays, and possibly place them into orbital resonances or a path to Earth. The above perspective suggests some possible explanations for the bimodal distribution of impact-reset Ar–Ar ages among chondrites. First, large collisions in the asteroid belt may have been much more common in the time periods 3.4–4.1 Ga and possibly also <1.0 Ga, compared to the time period of 1.3–3.4 Ga ago. The discussion in Section 4.8 (which is not repeated here) does indicate a much higher asteroidal collision rate ∼3.5–4.0 Ga ago. The fact that chondrites (and eucrites) showing these older ages still fall on Earth indicates that their parent bodies or large intermediate-size fragments still reside in stable orbits and that secondary collisions initiate the journey of individual meteorites to Earth. But do abundant chondrite ages <1.0 Ga suggest an increased asteroid collision rate during this time, and if so why are these young ages not observed among other meteorite types? A higher collision rate for these young events seems unlikely, and an equilibrium in the sizefrequency of asteroids has probably existed for some time (Davis et al., 2002). However, we cannot exclude the possibility that a relatively few major collisions involving only chondrite parent bodies occurred <1 Ga ago. Another possible explanation for the Ar–Ar age distribution of chondrites across 0–3.4 Ga may be that original chondrite parent bodies and intermediate sized objects resulting from collisional events that reset K–Ar ages are positioned relatively close to orbital resonances, and that migration of fragments into orbital resonances and ejection toward Earth occurs on time scales of the order of 108 –109 years. This explanation might imply that large chondrite collisions which reset K–Ar ages also occurred in the asteroid belt earlier than 1 Ga ago, but that fragments resulting from those ear-

D.D. Bogard / Chemie der Erde 71 (2011) 207–226

lier impacts have already reached their final destination and are no longer falling on Earth. However, this explanation does not account for those chondrites with ages >3.4 Ga nor for the lack of young K–Ar ages among other meteorite types. Possibly, parent bodies and large fragments of other meteorite types are not located near orbital resonances, and only smaller collisions eject fragments into orbital resonances. A related factor in this explanation is whether a large impact on a parent body (which resets the K–Ar age) leaves the meteorite buried and protected on that same parent body, or ejects the meteorite as part of a large fragment toward an orbital resonance. In the former case, meteorites experiencing very early impacts might be preserved, whereas in the latter case, meteorites included within smaller fragments might have a limited lifetime in the asteroid belt. However, it seems ad hoc and strained to conclude that chondrites and eucrites with older ages were preserved on their original parent bodies, that chondrites with younger ages were ejected within fragments toward orbital resonances, and that most other meteorite types were located on parent bodies positioned such that they did not experience large, age-resetting collisions in the past Ga. In summary, Ar–Ar ages of chondrites and eucrites clearly measure the heating effects of large impacts in the asteroid belt over solar system history. These younger (<1 Ga) Ar–Ar ages of chondrites may provide one measure of the finite time required to move their impact fragments within the asteroid belt and into an orbital resonance, and thus account for the lack of reset Ar–Ar ages over 1.3–3.4 Ga. However, such explanations for chondrites do not readily account for the observation that only chondrites exhibit reset ages of <1 Ga. Possibly there is something different about the orbital locations of chondrite parent bodies, compared to parent bodies of other meteorite types, and that this difference accounts for differences in belt migration times. Conversely, the orbital locations of chondrite parent bodies may have allowed recent, large collisions, which did not occur in other meteorite types. The true explanation is not obvious. 4.10. Martian meteorites Currently about 50 martian meteorites have been identified, all igneous rocks. An overview of martian meteorites and a summary of the scientific findings for many individual meteorites are available in a compendium by Meyer (2009) and references therein. Mineralogically, martian meteorites are classified into several types – a single orthopyroxenite (ALH84001), several specimens of clinopyroxenites called nakhlites, two dunites, and many specimens classified as either basaltic, olivine–pyroxene–phyric (or lherzolitic), and olivine–phyric meteorites (Goodrich, 2002; Meyer, 2009). Ages of many martian meteorites have been determined by several radiometric techniques (Nyquist et al., 2001; Borg et al., 2003, 2005; Nyquist, 2006; Misawa et al., 2006a; Lapen et al., 2010; bibliography in Meyer, 2009). The only ancient meteorite is ALH84001, which shows shock effects and contains secondary martian carbonates. Sm–Nd studies of ALH84001 indicate it formed ∼4.5 Ga ago, but Rb–Sr and U–Pb studies of the carbonate and Ar–Ar measurements of the feldspar indicate that the shock event and carbonate formation occurred about 3.9–4.1 Ga ago (Turner et al., 1997; Borg et al., 1999a, 1999b, Cassata et al., 2010, and references therein). Fig. 10 plots the formation ages of several martian meteorites, as determined by various chronometers (not including K–Ar), against their ejection ages from Mars (Nyquist et al., 2001; Nyquist, 2006). The ejection age is the sum of the space (cosmic ray) exposure time and the terrestrial age. Data presented in this way tend to form clusters, and meteorites within a given cluster likely derived from a common location on Mars and were ejected by a single impact event, i.e., are source paired. (More recent data not plot-

Crystallization and Ejection Ages 5

Txtn (Ga)

220

1

ALH84001

Basaltic shergottites Lherzolitic shergottites Clinopyroxenites (Nakhlites) Dunite (Chassigny) Orthopyroxenite New Data

Nakhlites & Chassigny Y000593

DaG476 0.5

QUE94201

Y980459 EETA79001

Zagami Los Angeles Shergotty

0.1

0.5

Dhofar019

NWA 1460

SaU 005

Y793605 LEW88516 ALHA77005 5

1

10

20

30

Tej (Ma) Fig. 10. Plot of formation age (Txtn ) against Mars ejection age (Tej ) for several martian meteorites. Meteorites forming a group on this plot likely derived from a common site on Mars by a single impact. Figure furnished by L. Nyquist and modified from Nyquist (2006).

ted tend to be associated with one of these clusters.) There is still debate over how many impact events are required to eject martian meteorites, but most workers advocate 5–7 events. Interestingly, although nakhlites and dunites (Chassigny in Fig. 10) have different mineral proportions, they have very similar ages and form a tight cluster. Ar–Ar ages have been reported for several nakhlites and for Chassigny (Swindle and Olsen, 2004; Misawa et al., 2006b; Park et al., 2009; Cassata et al., 2010, and references therein). These Ar–Ar ages are all ∼1.3–1.4 Ga and are consistent with ages determined by other chronometers. These ages likely measure a common formation time for these meteorites. A second martian dunite, NWA 2737, is very strongly shocked, unlike Chassigny and the nakhlites. The Sm–Nd age of this meteorite is 1.42 ± 0.06 Ga, whereas its Rb–Sr data show considerable scatter (Misawa et al., 2005b). Its Ar–Ar age is only 160–190 Ma, and Bogard and Garrison (2008) suggested that this age was reset by an impact event prior to ejection of the meteorite from Mars. Argon diffusion characteristics obtained at low extraction temperatures for some martian meteorites have been used to infer maximum heating temperatures the meteorites could have experienced during their residence on Mars and still retain 40 Ar released at low temperatures (Shuster and Weiss, 2005, and references therein). Ar–Ar ages for most of the younger martian meteorites (formation ages of ≤0.6 Ga) cannot be determined, because of the presence of other 40 Ar components not produced by in situ decay of 40 K. This situation is the opposite of many other meteorite types, where the 40 Ar concentrations are lower than the formation ages would predict, because of Ar diffusion loss produced by metamorphic or impact heating. One reason for these older Ar–Ar ages in younger martian meteorites is that most contain trapped martian atmospheric gases, including 40 Ar shock-implanted from the martian atmosphere. This was the first direct evidence of a martian origin of these meteorites (Bogard and Johnson, 1983; Wiens et al., 1988; Treiman et al., 2000). These atmospheric gases are primarily, but not exclusively, contained in shock-melted phases, whose apparent K–Ar ages range as high as ∼6 Ga, and many analyses give apparent Ar–Ar ages older than 3 Ga. These old Ar–Ar ages are found in both stepwise temperature release of bulk samples (e.g., Bogard and Garrison, 1999) and in laser-probe analyses of spot-sized mineral phases (Walton et al., 2007). Isochron plots (40 Ar/36 Ar against 39 Ar/36 Ar; Section 2) of stepwise temperature data for a very few younger (<0.6 Ga age) martian meteorites yield correct Ar–Ar ages (i.e., Ar–Ar ages identical to ages determined by other isotopic techniques). Such isochron plots

D.D. Bogard / Chemie der Erde 71 (2011) 207–226

5000

and Park (2008) argued for an entirely different source for this excess 40 Ar, namely that it was inherited from the melt from which these meteorites crystallized ∼170 Ma ago. The implications of this source will not be discussed here, but it is pointed out that many terrestrial rocks contain similar amounts of 40 Ar inherited from the melt (Baxer, 2003; references in Bogard and Park, 2008). Thus, in both the presence of excess 40 Ar and in the source of this 40 Ar, many martian meteorites are quite different from other meteorite types.

Zagami 5000

Los Angeles NWA3171 Shergotty

4000

NWA2975, 1068

4000

Y-00097, 793605 EET79001

3000

Non-JSC

[ K ] ppm

Impact Glass 2000

3000

All 170 Myr Isochrons

1000

221

0

2000

0

10

20

30

40

50

60

1000

0 0

100

200

300 40

400 -7

500

600

3

Total Ar, 10 cm /g Fig. 11. Isochron plot of total 40 Ar against total potassium in whole rock, mineral, and melt glass samples of nine martian meteorites, all of which have measured formation ages of ∼170 Ma. Melt glass samples, containing martian atmospheric 40 Ar, plot far to the right. Samples plotting to the left are shown in the expanded figure, and the three dotted lines each represent a 170 Ma isochron. Most of these analyses are consistent with a 170 Ma Ar–Ar age and 1–2 × 10−6 cm3 /g of trapped 40 Ar, which was likely inherited from the melt. Figure constructed from two figures in Bogard et al. (2009).

give an Ar–Ar age of 353 ± 16 Ma for NWA 1460, consistent with its Sm–Nd and Rb–Sr age (Nyquist et al., 2009b), and an isochron for a plagioclase separate of Los Angeles gives an Ar–Ar age of ∼170 Ma, consistent with its Sm–Nd age (Bogard et al., 2009). However, most Ar–Ar analyses of younger martian meteorites do not yield meaningful Ar–Ar isochron ages (e.g., Bogard and Garrison, 1999; Walton et al., 2007; Bogard et al., 2009). Many of the martian meteorites have formation (crystallization) ages of ∼170 Ma, as determined by Sm–Nd, Rb–Sr, and/or Pb–Pb. Fig. 11 plots total 40 Ar against total K from Ar–Ar analyses of several martian meteorites that give these formation ages. This figure is a combination of two figures presented by Bogard et al. (2009). The plotted points identified as impact glass contain large amounts of trapped 40 Ar from the martian atmosphere. They scatter across the plot to the right, and isochrons drawn through any one of these analyses and the origin would imply variable (and false) K–Ar ages of several Ga. However, a much larger number of analyses of different phases (whole rock and mineral separates) of nine meteorites appear to be more consistent with a single isochron. These analyses, plotting nearer the y-axis, are shown in the expanded plot in the upper right of Fig. 11. Three isochrons, whose slope corresponds to a K–Ar age of 170 Ma but having different intercepts on the x-axis, are also shown as dashed lines in the expanded view. Obviously, none of the plotted points is consistent with a 170 Ma isochron that passes through the origin, and isochrons passing through the origin and any one of these points would imply false K–Ar ages ranging from a few hundred Ma to several Ga. Thus, all of these samples contain 40 Ar in excess of that expected from in situ decay. However, nearly all of the analyses are consistent with an isochron age of 170 Ma and relatively constant trapped 40 Ar concentrations of 1–2 × 10−6 cm3 /g. This is in contrast to analyses of impact glasses plotted in Fig. 11, where trapped 40 Ar ranges over ∼6–54 × 10−6 cm3 /g. It is unlikely that much of this excess 40 Ar is shock-implanted martian atmosphere (Bogard and Park, 2008; Bogard et al., 2009). These non-glass samples include different mineral phases and shock levels and have K concentrations varying by a factor of ∼80. It would be highly unlikely for a random process like shock implantation to emplace a nearly constant amount of martian atmospheric 40 Ar into such a diverse sample suite. Bogard

5. Absolute K–Ar ages Since publication by Steiger and Jäger (1977) of preferred decay parameters for 40 K and some other parent nuclides used in radionuclide chronometers, almost all K–Ar ages of meteorites have been reported using these values. However, a series of papers, Renne et al. (1998, 2010), Renne (2000), and Vogel and Renne (2008), proposed revised 40 K decay parameters. Derivation of revised 40 K decay parameters for geologic dating have been estimated in three ways: (1) by comparing the Ar–Ar age of a terrestrial sample with its precisely determined Pb–Pb age; (2) by neutron irradiating together different standard minerals, which are used in Ar–Ar dating and whose absolute ages have been separately measured; and (3) by assuming the oldest Ar–Ar ages among meteorites are contemporaneous with the formation time of their parent bodies. Each of these methods has inherent uncertainties, and suggested revisions to a meteorite age of 4500 Ma have ranged from +47 Ma to about +13 Ma. Comparison of a K–Ar age to an age determined by any other method assumes that the dated sample had a simple formation and cooling history and that the two chronometers closed at the same time. Unless the sample cooled relatively quickly, these assumptions may not be true. Meteorite ages presented in this paper address the third method. However, assuming that the oldest K–Ar ages of meteorites closed immediately after parent body formation allows no time for cooling down after parent body metamorphism or impact heating and gives only an upper limit to the amount by which an old meteorite age could be increased because of errors in the 40 K decay parameters. Previous estimates of amounts by which old meteorite ages could be increased before exceeding parent body formation times were ∼+47 Ma for the Acapulco primitive achondrite (Renne, 2000), ∼+30 Ma for the oldest Ar–Ar age among a suite of H chondrites (Trieloff et al., 2003); ∼+22 Ma for the oldest Ar–Ar age among a suite of silicate inclusions from IAB iron meteorites (Vogel and Renne, 2008); and ∼+20 Ma for silicate in three other metal-rich meteorites (Bogard and Garrison, 2009a). The oldest precise Ar–Ar ages reported for H chondrites are 4532 ± 16 Ma for Ste. Marguerite and 4533 ± 12 Ma for Monahans (2 sigma ages, Trieloff et al., 2003; Bogard et al., 2001). The Pb–Pb and I–Xe ages for Ste. Marguerite are both 4563 ± 1 Ma (Göpel et al., 1994; Gilmour et al., 2004), and the I–Xe age for Monahans halite is ∼4560 Ma (Busfield et al., 2004). Although the difference between Ar–Ar and I–Xe ages for these two chondrites is ∼30 Ma, both chondrites clearly experienced some internal metamorphism, and because I–Xe has a higher closure temperature than K–Ar (Bogard and Garrison, 2009a), their K–Ar ages very likely closed after their I–Xe ages. Several enstatite chondrites give Ar–Ar ages around 4.53–4.54 Ga, but with significant 2-sigma uncertainties (Bogard et al., 2010), and these do not permit tight constraints to be placed on age differences between Ar–Ar and older I–Xe ages (Kennedy et al., 1988). The Shallowater aubrite gave an Ar–Ar age of 4.535 ± 0.03 Ga (2-sigma, Bogard et al., 2010) and an I–Xe age of 4.562 Ga (Gilmour et al., 2009), an age difference of ∼30 Ma. Among acapulcoites–lodranites, the oldest Ar–Ar age of 4521 ± 12 Ma (2-sigma) for GRA 95209 is ∼36 Ma younger than the Pb–Pb age of 4557 ± 2 Ma for Acapulco phosphate (Göpel et al.,

222

D.D. Bogard / Chemie der Erde 71 (2011) 207–226

1994). Increasing the Ar–Ar ages of acapulcoites–lodranites to bring them into agreement with the formation time of the parent body would be inconsistent with the extended cooling history for Acapulco suggested by Pellas et al. (1997). The oldest reported Ar–Ar ages among silicate from IAB iron meteorites (calculated with decay parameters from Steiger and Jäger, 1977) are two samples from Caddo County with ages of 4530 ± 0.4 and 4540 ± 0.8 Ma and an age for the Pontlyfni winonaite of 4529 ± 26 Ma, although a few additional IAB analyses give ages of ∼4.53 Ga with large uncertainties (Bogard et al., 2005; Vogel and Renne, 2008, and references therein). The Caddo County and Campo del Cielo IABs give similar I–Xe ages of 4558–4559 Ma, and some other IAB silicates also give similarly old I–Xe ages (Bogard et al., 2005, and references therein). Thus, IABs might permit the K–Ar ages to be older by up to ∼30 Ma, although Vogel and Renne (2008) suggested that an ∼22 Ma increase in K–Ar ages of IABs was more appropriate. The oldest Ar–Ar ages for silicates from IIE irons (including Weekeroo Station) are ∼4.49 Ga with significant uncertainties, whereas the I–Xe age for Weekeroo Station is 4555 Ma (Brazzle et al., 1999). It seems unlikely that these IIE silicates could have a K–Ar age as old as the I–Xe age. The Ar–Ar ages of silicates from two unclassified iron meteorites, Sombrerete at 4541 ± 24 Ma and NWA 176 at 4524 ± 26 Ma (both 2-sigma), are each ∼20 Ma younger than the I–Xe ages of 4561.9 ± 1 Ma and 4544 ± 7 Ma for the same samples, respectively (Bogard and Garrison, 2009a). In addition, Ar–Ar ages for two samples of Portales Valley differ by ∼20 Ma, and the older age is ∼80 Ma younger than the I–Xe age of 4560 Ma. As discussed earlier, the thermal history for each of these samples would be brought into better agreement with the I–Xe ages if we assume that the K–Ar ages should be ∼20 Ma older than those measured. In summary, arguments have been made that 40 K decay parameters in current use are incorrect and produce Ar–Ar ages that are too young. Older Ar–Ar ages of several meteorites of different classifications, when compared to I–Xe and/or Pb–Pb ages for the same or similar meteorites, would permit, but not require these measured K–Ar ages to be increased by about 20 Ma. Possibly, all older (i.e., ∼4500 Ma) meteorite ages should be increased by some amount, not to exceed ∼20 Ma. However, the majority of meteorites of a given class give Ar–Ar ages younger than the oldest Ar–Ar ages of that meteorite class. Further, many meteorite K–Ar ages are younger than formation ages of that meteorite class, even after K–Ar ages are increased by 20 Ma. Thus, even if we increase the K–Ar age of all meteorites by an appropriate amount, strong evidence still exists for prolonged thermal histories involving parent body heating by internal metamorphism and/or collisional heating.

6. Summary discussion Among relatively unshocked H chondrites, the majority of Ar–Ar ages range from ∼4.53 to 4.43 Ga. A somewhat smaller number of Ar–Ar ages for L chondrites range from 4.52 to ≤4.45 Ga. Ar–Ar ages for LL chondrites range from 4.50 to 4.40 Ga. For at least the H chondrites, variations in ages obtained by Ar–Ar and some other chronometers are consistent with a layered parent body, where younger ages correlate with higher temperatures, greater degrees of metamorphism, greater subsurface depth, and longer cooling times. Ar–Ar ages of several enstatite meteorites, including EL and EH chondrites, two aubrites, and two enstatite impact melt rocks, range over ∼4.54–4.35 Ga and probably indicate that enstatite parent bodies were comparable in size to those of ordinary chondrites. Ar–Ar ages of R chondrites are difficult to measure, and none are older than ∼4.47 Ga. In contrast to chondrites, Ar–Ar ages of several acapulcoites and lodranites cluster

tightly at 4.52–4.49 Ga. Acapulcoites–lodranites are thought to derive from a single parent body and were internally metamorphosed to temperatures of >950 ◦ C. Possibly, the similarity in Ar–Ar ages among acapulcoites–lodranites represents a two-stage cooling history induced by a large impact event during metamorphic cooling. Ar–Ar ages of all of the above meteorite types probably represent slow cooling after internal metamorphism of the parent bodies and are consistent with parent bodies of ∼102 km diameter. Some researchers have suggested that the 40 K decay parameters in common use are incorrect and that K–Ar ages of old meteorites should be increased by different amounts ranging up to ∼30 Ma. From comparison of the oldest Ar–Ar ages to meteorite formation ages measured by other chronometers on different meteorite types, an increase of no more than ∼20 Ma for a sample that is 4500 Ma old appears permissible. With an adjustment of +20 Ma, the oldest ages for H chondrites and enstatite meteorites become ∼4.55 Ga and approach the ages measured by some other chronometers. The corrected oldest Ar–Ar ages for acapulcoites–lodranites become ∼4.54 Ga, still slightly less than the U–Pb age for Acapulco of 4557 ± 2 Ma. Thus, among chondrites and acapulcoites–lodranites, the K–Ar chronometer for a few meteorites may have closed relatively soon after the parent bodies formed, but K–Ar in many of these meteorites closed substantially later. K–Ar ages that are older by ∼20 Ma suggest faster cooling rates and somewhat smaller parent bodies than previously thought. Silicate inclusions in some iron meteorites represent mixing with metal after varying degrees of silicate differentiation, but the exact process of the silicate–metal mixing is still debated. Multiple analyses of silicate samples from several IAB meteorites give a continuous distribution of Ar–Ar ages over the range of 4.53–4.32 Ga (using old decay constants), a span of ∼200 Ma. Possibly, the range of IAB ages is the result of collisional breakup and reassembly of the parent body while it was still cooling from internal metamorphism, although details are unclear. Variable Ar–Ar ages within some IAB meteorites suggests it is unlikely that these IAB ages are solely the result of slow cooling after internal metamorphism. Ar–Ar ages of silicate inclusions from several IIE iron meteorites show a dichotomy. Based on chronometers other than K–Ar, five IIEs suggest formation ages of ∼4.55–4.28 Ga, although some of these age determinations are uncertain. Ar–Ar ages of these same meteorites range from 4.49 to 4.40 Ga and likely reflect cooling following the thermal process that produced the silicate differentiation and mixing with metal. If we increase these Ar–Ar ages of IABs and IIEs by +20 Ma, the oldest IAB age would require very early silicate–metal mixing ∼4.55 Ga ago, but the oldest IIE age of ∼4.51 Ga suggests impact mixing may have occurred substantially after parent body formation. In contrast to the other IIEs, silicates in three IIE meteorites give ages around 3.68 Ga, as determined by more than one radiometric chronometer, including Ar–Ar. These ages must represent another major impact heating event on the parent body ∼3.68 Ga ago (ignoring any correction for 40 K decay parameters), one that also produced differentiation of silicates. Ar–Ar ages of IAB iron silicates, IIE silicates, and unshocked eucrites all indicate major impact resetting about 4.49 Ga ago (old decay constants). Three other unclassified metal-rich meteorites give Ar–Ar ages of 4.46–4.54 Ga, and two of these give I–Xe ages of 4.560–4.562 Ga, whereas a third gives an I–Xe age of ∼4.544 Ga. If these Ar–Ar ages are increased by 20 Ma (to correct for 40 K decay parameters), the Ar–Ar and I–Xe ages for Sombrerete are in agreement and would indicate relatively fast cooling after silicate–metal mixing. The I–Xe and Ar–Ar ages (the latter corrected by +20 Ma) for Portales Valley and NWA 176 (and the metallographic cooling rate for NWA 176) are consistent with impact break-up and reassembly of both parent bodies relatively early, followed by rapid cooling of NWA 176 but slow cooling of Portales Valley. Thus, unlike the case with chondrites and acapulcoites–lodranites discussed above, there is

D.D. Bogard / Chemie der Erde 71 (2011) 207–226

strong indication in all of these iron meteorites of the role of early impact heating, and even parent body breakup and reassembly, in determining the K–Ar ages. Mesosiderites, likely impact-produced mixtures of metal and igneous silicates, give Ar–Ar ages around 4.0 Ga, which was long after the likely formation time of their parent body. Further, the Ar–Ar age spectra of many mesosiderites show an upward slope in age, consistent with very slow cooling. The degree of slope in these ages correlates with the ease of Ar diffusion among the samples. A cooling rate calculated from these Ar data suggest a very slow rate of ∼0.2 ◦ C/Ma, which is in good agreement with cooling rates determined for Ni diffusion in mesosiderite metal. Mesosiderites in particular present an enigma. Their very nature implies mixing of metal from an iron core with igneous silicates produced at or near a planetary surface, yet they give evidence of deep burial and very slow cooling. Only collisional breakup and reassembly of a large parent body seems a reasonable process to produce these characteristics, and a later collisional breakup may have been required to bring mesosiderites near the surface. Ar–Ar ages of HED meteorites are much more similar to ages of lunar highland rocks than to ages of most other meteorite types. HED ages form two groups. Those eucritic samples that are breccias or otherwise show shock effects (the majority of samples) give Ar–Ar ages in the range 3.4–4.1 Ga. These ages are similar to impact reset radiometric ages of many lunar highland rocks. It is widely believed that highland rock ages were reset by a period of late enhanced bombardment of the inner solar system ∼3.9 Ga ago, and the source of these impacting objects possibly derived from gravitational interactions of the giant gas planets with the asteroid belt and Kuiper belt bodies. Similar Ar–Ar ages among eucrites and lunar highland rocks indicate that this same bombardment also occurred on Vesta, the likely parent body of the HEDs. A smaller number of eucritic meteorites that are not brecciated have Ar–Ar ages which cluster tightly around 4.48 Ga. These older ages may reflect excavation from depth by a single large impact, but it is not clear how these meteorites escaped age resetting by later impacts. A few brecciated ordinary chondrites give impact reset Ar–Ar ages in a similar range, and so do three IIE meteorites. The rarity of impact reset Ar–Ar ages of 3.5–4.0 Ga among other meteorite types is likely due to their derivation from smaller parent bodies. Only relatively large parent bodies, like the Moon and Vesta, could sustain impacts sufficiently large to produce the heating required to reset K–Ar ages without destroying the body. A few chondrites give Ar–Ar ages of ∼3.5–4.0 Ga, and these were likely reset by the late enhanced bombardment that reset many Ar–Ar ages of eucrites. However, many strongly shocked ordinary chondrites commonly show Ar–Ar ages of less than 1 Ga. Essentially no Ar–Ar ages of chondrites or HED meteorites fall in the range of 1.3–3.5 Ga. Interestingly, Ar–Ar ages of <1 Ga are rare in meteorite types other than ordinary chondrites. These young chondrite ages possibly measure collisional events which produced large fragments that slowly moved into asteroidal orbits where they were gravitationally perturbed into Earth-crossing orbits in time scales on the order of 108 –109 years. However, this explanation does not easily account for the lack of young ages among other meteorite types. Because of unique orbital positions in the asteroid belt, possibly only chondrite parent bodies experienced recent intense collisions. The L chondrite parent body was likely collisional disrupted 0.47 Ga ago, and shortly thereafter delivered large quantities of fragments to Earth. The 0.47 Ga collisional event is commonly seen among L chondrites falling on Earth today. Among martian meteorites, only the nakhlites and a dunite give concordant ages of ∼1.3 Ga by multiple chronometers, including Ar–Ar. Most martian meteorites have formation ages of ≤0.6 Ga and are significantly shocked. Almost all of these meteorites show false Ar–Ar ages that are older. The excess 40 Ar in these samples

223

has two origins: martian atmospheric Ar implanted by shock and radiogenic 40 Ar inherited from the melt. In a few of these samples, isochron plots give the formation age. 7. Conclusion The main purpose of this paper is to utilize measured Ar–Ar ages to demonstrate that meteorites and their parent bodies have experienced extended and complex thermal histories involving both internal metamorphism and impact heating. K–Ar ages of meteorites demonstrate these complex thermal histories like no other chronometer. The formation ages of several meteorite parent bodies determined by various long-lived and short-lived chronometers are ∼4.56 Ga, and some igneous meteorites indicate somewhat later formation times. However, nearly all K–Ar ages of meteorites are younger than these ages, some much younger. Early in solar system history, residual heat from decay of short-lived radionuclides produced internal metamorphism, keeping the K–Ar chronometer open during slow cooling. The significant time required for some Ar–Ar ages to close is one measure of the significant size of some of these parent bodies (∼102 km diameter). These younger Ar–Ar ages result because of the relative ease with which 40 Ar diffuses and is lost from most silicate materials at relative modest temperatures. However, the important role of parent body collisions in heating these bodies and resetting K–Ar ages began early in solar system history, and in a few meteorite types geologically recent collisional heating is still observed. Some of these collisional events likely produced early disruption and reassembly of parent bodies, whereas others, such as the 0.47 Ga event in L chondrites, destroyed the immediate parent. K–Ar ages of meteorites thus document, not just the beginnings of our solar system, but its complex evolution throughout time. Acknowledgements I thank the Lunar and Planetary Institute and NASA for financial support during the preparation of this paper. I am grateful to Associate Editor Klaus Keil for soliciting this Invited Review and for helpful editorial comments, and to Nadia Vogel, Tim Swindle, and Larry Nyquist for permission to use their previously published figures. I appreciate valuable reviewer comments by T. Swindle and an anonymous reviewer. References Amelin, Y., Krot, A.N., Hutcheon, I.D., Ulyanov, A.A., 2002. Lead isotopic ages of chondrules and calcium–aluminum-rich inclusions. Science 297, 1678–1683. Baxer, E.F., 2003. Quantification of the factors controlling the presence of excess 40 Ar or 4 He. Earth Planet. Sci. Lett. 216, 619–634. Benedix, G.K., McCoy, T.J., Keil, K., Bogard, D.D., Garrison, D.H., 1998. A petrologic and isotopic study of winonaites: evidence for early partial melting, brecciation, and metamorphism. Geochim. Cosmochim. Acta 62, 2535–2553. Benedix, G.K., McCoy, T.J., Keil, K., Love, S.G., 2000. A petrologic study of the IAB iron meteorites: constraints on the formation of the IAB-winonaite parent body. Meteorit. Planet. Sci. 35, 1127–1141. Benedix, G.K., Ketcham, R.A., Wilson, L., McCoy, T.J., Bogard, D.D., Garrison, D.H., Herzog, G.F., Xue, S., Klein, J., Middleton, R., 2008. The formation and chronology of the PAT 91501 impact-melt L chondrite with vesicle–metal–sulfide assemblages. Geochim. Cosmochim. Acta 72, 2417–2428. Bennett, M.E., McSween, H.Y., 1993. Metallographic cooling rates of L-group ordinary chondrites. Lunar Planet. Sci. 24, 97–98 (abstract). Bennett, M.E., McSween, H.Y., 1996. Revised model calculations for the thermal histories of ordinary chondrite parent bodies. Meteorit. Planet. Sci. 31, 783–792. Binzel, R.P., Xu, S., 1993. Chips off asteroid 4 Vesta: evidence for the parent body of basaltic achondrite meteorites. Science 260, 186–191. Bischoff, A., Vogel, N., Roszjar, J., 2011. The Rumuruti chondrite group. Chemie der Erde-Geochemistry 71, in press. Bogard, D.D., 1995. Impact ages of meteorites: a synthesis. Meteoritics 30, 244–268. Bogard, D.D., Hirsch, W.C., 1980. 40 Ar–39 Ar dating, Ar diffusion properties, and cooling rate determinations of severely shocked chondrites. Geochim. Cosmochim. Acta 44, 1667–1682.

224

D.D. Bogard / Chemie der Erde 71 (2011) 207–226

Bogard, D.D., Johnson, P.J., 1983. Martian gases in an Antarctic meteorite. Science 221, 651–654. Bogard, D.D., Garrison, D.H., 1998. 39 Ar–40 Ar ages and thermal history of mesosiderites. Geochim. Cosmochim. Acta 62, 1459–1468. Bogard, D.D., Garrison, D.H., 1999. 39 Ar–40 Ar “ages” and trapped argon in martian shergottites, Chassigny, and ALH84001. Meteorit. Planet. Sci. 34, 451–473. Bogard, D.D., Garrison, D.H., 2003. 39 Ar–40 Ar ages of eucrites and the thermal history of asteroid 4-Vesta. Meteorit. Planet. Sci. 38, 669–710. Bogard, D.D., Park, J., 2008. 39 Ar–40 Ar dating of the Zagami martian shergottite and implications for the magma origin of excess 40 Ar. Meteorit. Planet. Sci. 43, 1–16. Bogard, D.D., Garrison, D.H., 2008. 39 Ar–40 Ar age and thermal history of martian dunite NWA 2737. Earth Planet. Sci. Lett. 273 (3–4), 386–392. Bogard, D.D., Garrison, D.H., 2009a. Ar–Ar and I–Xe ages and thermal histories of three unusual metal-rich meteorites. Geochim. Cosmochim. Acta 73, 6965–6983. Bogard, D.D., Garrison, D.H., 2009b. Ar–Ar impact heating ages of eucrites and timing of the LHB. Lunar Planet. Sci. 40 (#1131 CD-ROM, abstract). Bogard, D.D., Garrison, D.H., Jordan, J., Mittlefehldt, D., 1990. 39 Ar–40 Ar dating of mesosiderites: evidence for major parent body disruption <4 Ga ago. Geochim. Cosmochim. Acta 54, 2549–2564. Bogard, D.D., Garrison, D.H., McCoy, T.J., 2000. Chronology and petrology of silicates from IIE iron meteorites: evidence of a complex parent body evolution. Geochim. Cosmochim. Acta. 64, 2133–2154. Bogard, D.D., Garrison, D.H., Masarik, J., 2001. The Monahans chondrite and halite: 39 Ar–40 Ar age, solar gases, and parent body regolith neutron flux and thickness. Meteorit. Planet. Sci. 36, 107–122. Bogard, D.D., Garrison, D.H., Takeda, H., 2005. Ar–Ar and I–Xe ages and the thermal history of IAB meteorites. Meteorit. Planet. Sci. 40, 207–224. Bogard, D., Park, J., Garrison, D., 2009. 39 Ar–40 Ar “ages” and origin of excess 40 Ar in martian shergottites. Meteorit. Planet. Sci. 43, 905–924. Bogard, D.D., Dixon, E.T., Garrison, D.H., 2010. Ar–Ar ages and thermal histories of enstatite meteorites. Meteorit. Planet. Sci. 45 (Nr 5), 723–742. Borg, L.E., Nyquist, L.E., Shih, C.Y., Wiesmann, H., Reese, Y., Connelly, J.N., 1999a. Rb–Sr formation age of ALH84001 carbonates. In: Workshop on Martian Meteorites: Where Do We Stand, and Where Are We Going?,. Lunar and Planetary Institute (abstract #7030). Borg, L.E., Connelly, J.N., Nyquist, L.E., Shih, C.Y., 1999b. Pb–Pb age of the carbonates in the martian meteorite ALH84001. Lunar Planet. Sci. 30 (#1430 CD-ROM, abstract). Borg, L.E., Nyquist, L.E., Wiesmann, H., Shih, C.Y., Reese, Y., 2003. The age of Dar al Gani 476 and the differentiation history of the martian meteorites inferred from their radiogenic isotopic systematics. Geochim. Cosmochim. Acta 67, 3519–3536. Borg, L.E., Edmunson, J.E., Asmerom, Y., 2005. Constraints on the U–Pb isotopic systematics of Mars inferred from a combined U–Pb, Rb–Sr, and Sm–Nd isotopic study of the martian meteorite Zagami. Geochim. Cosmochim. Acta 69, 5819–5830. ´ D., Dones, L., 2007. Can planetesimals left over Bottke, W.F., Levison, H.F., Nesvorny, from terrestrial planet formation produce the lunar Late Heavy Bombardment? Icarus 190, 203–223. Bouvier, A., Wadhwa, M., 2010. The age of the solar system redefined by the oldest Pb–Pb age of a meteoritic inclusion. Nat. Geosci. 3, 637–641. Brazzle, R.H., Pravdivtseva, O.V., Meshik, A.P., Hohenberg, C.M., 1999. Verification and interpretation of the I–Xe chronometer. Geochim. Cosmochim. Acta 63, 739–760. Brearley, A.J., Jones, R.H., 1998. Chondritic meteorites. In: Papike, J. (Ed.), Planetary Materials. Reviews in Mineralogy, vol. 36. Mineralogical Society of America (Chapter 3). Brouxel, M., Tatsumoto, M., 1991. The Esterville mesosiderite: U–Pb, Rb–Sr, and Sm–Nd isotopic study of a polymict breccia. Geochim. Cosmochim. Acta 55, 1121–1133. Burnett, D.S., Wasserburg, G.J., 1967. Evidence for the formation of an iron meteorite at 3.8 × 109 years. Earth Planet. Sci. Lett. 2, 137–147. Busfield, A., Gilmour, J.D., Whitby, J.A., Turner, G., 2004. Iodine–xenon analysis of ordinary chondrite halide: implications for early solar system water. Geochim. Cosmochim. Acta 68, 195–202. Carlson, R.W., Lugmair, G.W., 2000. Timescales of planetesimal formation and differentiation based on extinct and extant radioisotopes. In: Canup, R., Righter, K. (Eds.), Origin of the Earth and Moon. Univ. Arizona Press, pp. 25–44. Cassata, W.S., Shuster, D.L., Renne, P.R., Weiss, B.P., 2010. Evidence for shock heating and constraints on martian surface temperatures revealed by 40 Ar/39 Ar thermochromometry of martian meteorites. Geochim. Cosmochim. Acta 74, 6900–6920. Choi, B.G., Ouyang, X., Wasson, J.T., 1995. Classification and origin of IAB and IIICD iron meteorites. Geochim. Cosmochim. Acta 59, 593–612. Clayton, R.N., Mayeda, T.K., 1996. Oxygen isotope studies of achondrites. Geochim. Cosmochim. Acta 60, 1999–2017. Cressy, P.J., 1971. The production rate of 26 Al from target elements in the Bruderheim chondrite. Geochim. Cosmochim. Acta 35, 1283–1296. Crozaz, G., Tasker, D.R., 1981. Thermal history of mesosiderites revisited. Geochim. Cosmochim. Acta 45, 2037–2046. Davis, D.R., Durda, D.D., Marzari, F., Campo-Bagatin, A., Gil-Hutton, R., 2002. Collisional evolution of small-body populations. In: Bottke, W.F., Cellino, A., Paolicchi, P., Binzel, R.P. (Eds.), Asteroids III. Univ. Arizona Press, pp. 545–558. Dixon, E.T., Bogard, D.D., Garrison, D.H., 2003. 39 Ar–40 Ar chronology of R chondrites. Meteorit. Planet. Sci. 38, 341–355.

Dixon, E.T., Bogard, D.D., Garrison, D.H., 2004. 39 Ar–40 Ar evidence for early impact events on the LL parent body. Geochim. Cosmochim. Acta 68, 3779–3790. Dodd, R.T., 1981. Meteorites: A Petrologic–Chemical Synthesis. Cambridge Univ. Press, 368 pp. Dodson, M.H., 1973. Closure temperatures in cooling geochronological and petrological systems. Contrib. Mineral. Petrol. 40, 259–274. Eugster, O., 2005. Cosmic-ray exposure ages of meteorites and lunar rocks and their significance. Chem Erde 63, 3–30. Eugster, O., Lorenzetti, S., 2005. Cosmic-ray exposure ages of four acapulcoites and two differentiated achondrites and evidence for a two-layer structure of the acapulcoite/lodranite parent asteroid. Geochim. Cosmochim. Acta 69, 2675–2685. Ganguly, J., Yang, H., Ghose, S., 1994. Thermal history of mesosiderites: quantitative constraints from compositional zoning and Fe–Mg ordering in orthopyroxene. Geochim. Cosmochim. Acta 58, 2711–2723. Gilmour, J., 2000. The extinct radionuclide timescale of the early solar system. Space Sci. Rev. 92, 123–132. Gilmour, J.D., Pravdivtseva, O.V., Busfield, A., Hohenberg, C.M., 2004. I–Xe and the chronology of the early solar system. In: Workshop on Chondrites and Protoplanetary Disk,. Lunar and Planetary Institute (abstract #9054). Gilmour, J.D., Crowther, S.A., Busfield, A., Holland, G., Whitby, J.A., 2009. An early I–Xe age for CB chondrite chondrule formation, and a re-evaluation of the closure age of Shallowater enstatite. Meteorit. Planet. Sci. 44, 573–580. Gladman, B.J., Migliorini, F., Morbidelli, A., Zappala, V., Michel, P., Cellino, A., Froeschle, C., Levison, H.F., Bailey, M., Duncan, M., 1997. Dynamical lifetimes of objects injected into asteroid belt resonances. Science 277, 197–201. Gomes, R., Levinson, H.F., Tsiganis, K., Morbidelli, A., 2005. Origin of the cataclysmic late heavy bombardment period of the terrestrial planets. Nature 435, 466–469. Goodrich, C.A., 2002. Olivine–phyric martian basalts: a new type of shergottite. Meteorit. Planet. Sci. 37, B31–B34. Göpel, C., Manhes, G., Allegre, C., 1985. Concordant 3,676 Myr U–Pb formation age for the Kodaikanal iron meteorite. Nature 317, 341–344. Göpel, C., Manhes, G., Allegre, C., 1992. U–Pb study of the Acapulco meteorite. Meteoritics 27, 226. Göpel, C., Manhès, G., Allègre, C.J., 1994. U–Pb systematic of phosphates from equilibrated ordinary chondrites. Earth Planet. Sci. Lett. 121, 153–171. Greenwood, R.C., Franchi, I.A., Jambon, A., 2006. New oxygen isotope evidence for the origin of mesosiderites and main group pallasites. Lunar Planet. Sci. 37, The Lunar and Planetary Institute (#1768 CD-ROM, abstract). Grier, J.A., Kring, D.A., Swindle, T.D., Rivkin, A.S., Cohen, B.A., Britt, D.T., 2004. Analyses of the chondritic meteorite Orvinio (H6): further insight into the origins and evolution of shocked H-chondrite material. Meteorit. Planet. Sci. 39, 1475–1493. Haack, H., Scott, E.R.D., Rasmussen, K.L., 1996. Thermal and shock history of mesosiderites and their large parent asteroid. Geochim. Cosmochim. Acta 60, 2609–2619. Hartmann, W.K., Ryder, G., Dones, L., Grinspoon, D., 2000. The time-dependant intense bombardment of the primordial Earth/Moon system. In: Canup, R.M., Righter, K. (Eds.), Origin of the Earth and Moon. Univ. Arizona Press, pp. 493–512. Heck, P.R., Schmitz, B., Baur, H., Halliday, A.N., Wieler, R., 2004. Fast delivery of meteorites to Earth after a major asteroid collision. Nature 430, 323–325. Heck, P.R., Ushikubo, T., Schmitz, B., Kita, N.T., Spicuzza, M.J., Valley, J.W., 2010. A single asteroidal source for extraterrestrial Ordovician chromite grains from Sweden and China: high-precision oxygen three-isotope SIMS analysis. Geochim. Cosmochim. Acta 74, 497–509. Herpfer, M.A., Larimer, J.W., Goldstein, J.I., 1994. A comparison of metallographic cooling rate methods used in iron meteorites. Geochim Cosmochim. Acta 58, 1353–1365. Hewins, R.A., 1983. Impact vs. internal origin for mesosiderites. In: Proc. 14th Lunar Planet. Sci. Conf.,. Lunar Planetary Institute, pp. B257–B266. Heymann, D., 1967. On the origin of hypersthene chondrites. Ages and shock effects of black chondrites. Icarus 6, 189–221. Hohenberg, C.M., Pravdivtseva, O.V., 2008. I–Xe dating: from adolescence to maturity. Chem. Erde—Geochem. 68, 339–351. Ireland, T., Wlotzka, F., 1992. The oldest zircons in the solar system. Earth Planet. Sci. Lett. 109, 1–10. Jessberger, E.K., Dominik, B., Staudacher, T., Herzog, F., 1980. 40 Ar–39 Ar ages of Allende. Icarus 42, 380–405. Kallemeyn, G.W., Rubin, A.E., Wasson, J.T., 1996. The compositional classification of chondrites: VIII. The R chondrite group. Geochim. Cosmochim. Acta 60, 2243–2256. Kaneoka, I., 1980. 40 Ar–39 Ar ages of L and LL chondrites from Alan Hills, Antarctica. Mem. Natl. Inst. Polar Res. (Special Issue 17), 177–188. Keil, K., 1989. Enstatite meteorites and their parent bodies. Meteoritics 24, 195–208. Keil, K., 2010. Enstatite achondrite meteorites (aubrites) and the histories of their asteroidal parent bodies. Chem. Erde—Geochem. 70, 295–317. Keil, K., Haack, H., Scott, E.R.D., 1994. Catastrophic fragmentation of asteroids: evidence from meteorites. Planet. Space Sci. 42, 1109–1122. Kennedy, B.M., Hudson, B., Hohenberg, C.M., Podosek, F.A., 1988. 129 I/127 I variations among enstatite chondrites. Geochim. Cosmochim. Acta 52, 101–111. Kirsten, T., Krankowsky, D., Zähringer, J., 1963. Edelgas und Kalium-Bestimmungen an einer grösseren Zahl von Steinmeteoriten. Geochim. Cosmochim. Acta 27, 13–42. Kleine, T., Touboul, M., Bourdon, B., Nimmo, F., Mezger, K., Palme, H., Jacobsen, S.B., Yin, Q.Z., Halliday, A.N., 2009. Hf–W chronology of the accretion and early evolution of asteroids and terrestrial planets. Geochim. Cosmochim. Acta 73, 5150–5188.

D.D. Bogard / Chemie der Erde 71 (2011) 207–226 Korochantseva, E.V., Trieloff, M., Lorenz, C.A., Buylin, A.I., Ivanova, M.A., Schwartz, W.H., Hopp, J., Jessberger, E.K., 2007. L-chondrite asteroid breakup tied to Ordovician meteorite shower by multiple isochron 40 Ar–39 Ar dating. Meteorit. Planet. Sci. 42, 113–130. Kracher, A., 1985. The evolution of the partially differentiated planetesimals: evidence from iron meteorite groups IAB and IIICD. J. Geophys. Res. 90, C689–C698. Kring, D.A., Swindle, T.D., Britt, D.T., Grier, J.A., 1996. Cat Mountain: a meteoritic sample of an impact-melted asteroid regolith. J. Geophys. Res. 101, 29,353–29,371. Kring, D.A., Hill, D.H., Gleason, J.D., Britt, D.T., Consolmagno, G.J., Farmer, M., Wilson, S., Haag, R., 1999. Portales Valley: a meteorite sample of the brecciated and metal-veined floor of an impact crater on an H-chondrite asteroid. Meteorit. Planet. Sci. 34, 663–669. Krot, A.N., Amelin, Y., Bland, P., Ciesla, F.J., Connelly, J., Davis, A.M., Huss, G.R., Hutcheon, I.D., Makide, K., Nagashima, K., Nyquist, L.E., Russell, S.S., Scott, E.R.D., Thrane, K., Yurimoto, H., Yin, Q.-Z., 2009. Origin and chronology of chondritic components: a review. Geochim. Cosmochim. Acta 73, 4963–4997. Kunz, J., Trieloff, M., Bobe, K.D., Metzler, K., Stöffler, D., Jessberger, E.K., 1995. The collisional history of the HED parent body inferred from 40 Ar–39 Ar ages of eucrites. Planet. Space Sci. 43, 527–543. Kunz, J., Falter, M., Jessberger, E.K., 1997. Shocked meteorites: argon-40–argon-39 evidence for multiple impacts. Meteorit. Planet. Sci. 32, 647–670. Lapen, T.J., Righter, M., Brandon, A.D., Debaille, V., Beard, B.L., Shafer, J.T., Peslie, A.H., 2010. A younger age for ALH84001 and its geochemical link to shergottite sources in Mars. Science 328, 347–351. Liu, M., Scott, E.R.D., Keil, K., Wasson, J.T., Clayton, R.N., Mayeda, T., Eugster, O., Crozaz, G., Floss, C., 2001. Northwest Africa 176: a unique iron meteorite with silicate inclusions related to Bocaiuva. Lunar Planet. Sci. 32 (#2152 CD-ROM, abstract). Liu, Y.Z., Nyquist, L.E., Wiesmann, H., Shih, C.Y., Takeda, H., 2002. Rb–Sr and Sm–Nd ages of plagioclase-diopside-rich material in Caddo County IAB iron meteorite. Lunar Planet. Sci. 33 (#1389 CD-ROM, abstract). Liu, Y.Z., Nyquist, L.E., Wiesmann, H., Shih, C.Y., Schwandt, C., Takeda, H., 2003. Internal Rb–Sr age and initial 87Sr/86Sr of a silicate inclusion from the Campo del Cielo iron meteorite. Lunar Planet. Sci. 33 (#1983 CD-ROM, abstract). Lugmair, G.W., Shukolyukov, A., 1998. Early solar system timescales according to 53 Mn–53 Cr systematics. Geochim. Cosmochim. Acta 62, 2863–2886. Markowski, A., Quitté, G., Halliday, A.N., Kleine, T., 2006. Tungsten isotopic composition of iron meteorites: chronological constraints vs. cosmogenic effects. Earth Planet. Sci. 242, 1–15. Marti, K., Graf, T., 1992. Cosmic-ray exposure history of ordinary chondrites. Annu. Rev. Earth Planet. Sci. 20, 221–243. McCoy, T.J., Keil, K., Clayton, R.N., Mayeda, T.K., Bogard, D.D., Garrison, D.H., Huss, G.R., Hutcheon, I.D., Wieler, R., 1996. A petrologic, chemical, and isotopic study of Monument Draw and comparison with other acapulcoites: evidence for formation by incipient melting. Geochim. Cosmochim. Acta 60, 2681–2708. McCoy, T.J., Keil, K., Muenow, D.W., Wilson, L., 1997a. Partial melting and melt migration in the acapulcoite–lodranite parent body. Geochim. Cosmochim. Acta 61, 639–650. McCoy, T.J., Keil, K., Clayton, R.N., Mayeda, T.K., Bogard, D.D., Garrison, D.H., Wieler, R., 1997b. A petrologic and isotopic study of lodranites: evidence for early formation as partial melt residues from heterogeneous precursors. Geochim. Cosmochim. Acta 61, 623–637. McCoy, T.J., Carlson, W.D., Nittler, L.R., Stroud, R.M., Bogard, D.D., Garrison, D.H., 2006. Graves Nunataks 95209: a snapshot of metal segregation and core formation. Geochim. Cosmochim Acta 70, 516–531. McDougall, I., Harrison, T.M., 1999. Geochronology and Thermochronology by the 40 Ar–39 Ar Method. Oxford University Press, 269 pp. Meyer, C., 2009. Mars Meteorite Compendium., http://curator.jsc.nasa.gov/antmet/mmc/index.cfm. Minton, D.A., Malhotra, R., 2009. A record of planet migration in the main asteroid belt. Nature 457, 1109–1111. Misawa, K., Yamaguchi, A., Kaiden, H., 2005a. U–Pb and 207 Pb-206 Pb ages of zircons from basaltic eucrites: implications for early basaltic volcanism on the eucrite parent body. Geochim. Cosmochim. Acta 69, 5847–5861. Misawa, K., Shih, C.Y., Reese, Y., Nyquist, L.E., Barrat, J.A., 2005b. Rb–Sr and Sm–Nd isotopic systematics of the NWA 2737 chassignite. Meteorit. Planet. Sci. 40 (Suppl.), A104 (abstract). Misawa, K., Yamada, K., Nakamura, N., Morikawa, N., Yamashita, K., Premo, W.R., 2006a. Sm–Nd isotopic systematics of lherzolitic shergottite Yamato-793605. Antarct. Meteorite Res. 19, 45–57. Misawa, K., Shih, C.Y., Reese, Y., Bogard, D.D., Nyquist, L.E., 2006b. Rb–Sr, Sm–Nd and Ar–Ar isotopic systematics of martian dunite Chassigny. Earth Planet. Sci. Lett. 246, 90–101. Mittlefehldt, D.W., 1990. Petrogenesis of mesosiderites: I. Origin of mafic lithologies and comparison with basaltic achondrites. Geochim. Cosmochim. Acta 54, 1165–1173. Mittlefehldt, D.W., Lindstrom, M.M., Bogard, D.D., Garrison, D.H., Field, S.W., 1996. Acapulco- and lodran-like achondrites: petrology, geochemistry, chronology, and origin. Geochim. Cosmochim. Acta 60, 867–882. Mittlefehldt, D.W., McCoy, T.J., Goodrich, C.A., Kracher, A., 1998. Acapulco- and lodran-like achondrites: petrology, geochemistry, chronology, and origin. Nonchondritic meteorites from asteroidal bodies. In: Papike, J.J. (Ed.), Planetary Materials. Reviews in Mineralogy, Vol. 36, pp. 1–195 (Chapter 4). Nagao, K., Okazaki, R., Sawada, S., Nakamura, N., 1999. Noble gases and K–Ar ages of five Rumuriti chondrites Y-75302, Y-791827, Y-793575, Y-82002, and Asuka881988. Antarct. Meteorite Res. 12, 81–93.

225

Niemeyer, S., 1979a. I–Xe dating of silicate and troilite from IAB iron meteorites. Geochim. Cosmochim. Acta 43, 843–860. Niemeyer, S., 1979b. 40 Ar–39 Ar dating of inclusions from IAB iron meteorites. Geochim. Cosmochim. Acta 43, 1829–1840. Niemeyer, S., 1980. I–Xe and 40 Ar–39 Ar dating of silicate from Weekeroo Station and Netaschaëvo IIE iron meteorites. Geochim. Cosmochim. Acta 44, 33–44. Norman, M.D., Duncan, R.A., Huard, J.J., 2010. Imbrium provenance for the Apollo 16 Descartes terrain: argon ages and geochemistry of lunar breccias 67016 and 67455. Geochim. Cosmochim. Acta 74, 763–783. Nyquist, L.E., 2006. Martian meteorite ages and implications for martian cratering history. In: Workshop on Surface Ages and Histories: Issues in Planetary Chronology,. Lunar and Planetary Institute (abstract #6010). Nyquist, L.E., Bogard, D.D., Shih, C.Y., Greshake, A., Stöffler, D., Eugster, O., 2001. Ages and geologic histories of martian meteorites. Space Sci. Rev. 96, 105–164. Nyquist, L., Yamaguchi, A., Bogard, D., Shih, C.-Y., Karouji, Y., Ebihara, M., Reese, Y., Garrison, D., Takeda, H., 2006. Feldspathic clasts in Yamato 86032: Remnants of the lunar crust with implications for its formation and history. Geochim. Cosmochim. Acta 70, 5990–6015. Nyquist, L.E., Kleine, T., Shih, C.-Y., Reese, Y.D., 2009a. The distribution of short-lived radioisotopes in the early solar system and the chronology of asteroid accretion, differentiation, and secondary mineralogy. Geochim. Cosmochim. Acta 73, 5115–5136. Nyquist, L.E., Bogard, D.D., Shih, C.Y., Park, J., Reese, Y.D., Irving, A.J., 2009b. Concordant Rb–Sr, Sm–Nd, and Ar–Ar ages for Northwest Africa 1460: a 345 Ma old basaltic shergottite related to “lherzolitic” shergottites. Geochim. Cosmochim. Acta 73, 4288–4309. Olsen, E., Davis, A., Clarke, R.S., Schultz, L., Weber, H.W., Clayton, R., Mayeda, T., Jarosewich, E., Sylvester, P., Grossman, L., Wang, M., Lipschutz, M.E., Steele, I.M., Schwade, J., 1994. Watson: a new link in the IIE iron chain. Meteoritics 29, 200–213. Palme, H., Schultz, L., Spettel, B., Weber, H.W., Wänke, H., Christophe, Michel-Levy, M., Lorin, J.C., 1981. The Acapulco meteorite: chemistry, mineralogy and irradiation effects. Geochim. Cosmochim. Acta 45, 727–752. Park, J., Garrison, D.H., Bogard, D.D., 2008a. 39 Ar–40 Ar ages of martian nakhlites. Geochim. Cosmochim. Acta 73, 2177–2189. Park, J., Bogard, D.D., Mikouchi, T., McKay, G.A., 2008b. The Dhofar-378 martian shergottite: evidence of early shock melting. J. Geophys. Res. Planets 113, E08007, doi:10.1029/2007JE003035. Park, J., Garrison, D.H., Bogard, D.D., 2009. 39 Ar–40 Ar ages of martian nakhlites. Geochim. Cosmochim. Acta 73, 2177–2189. Patzer, A., Hill, D.H., Boynton, W.V., 2004. Evolution and classification of acapulcoites and lodranites from a chemical point of view. Meteorit. Planet. Sci. 39, 61–85. Pellas, P., Fiéni, C., 1988. Thermal histories of ordinary chondrite parent asteroids. Lunar Planet. Sci. 19, 915–916 (abstract). Pellas, P., Fiéni, K., Trieloff, M., Jessberger, E.K., 1997. The cooling history of the Acapulco meteorite as recorded by the 244 Pu and 40 Ar–39 Ar chronometers. Geochim. Cosmochim. Acta 61, 3477–3501. Podosek, F.A., 1970. Dating of meteorites by the high-temperature release of iodinecorrelated 129 Xe. Geochim. Cosmochim. Acta 34, 341–366. Powell, B.N., 1969. Petrology and chemistry of mesosiderites. I: Textures and composition of nickel–iron. Geochim. Cosmochim. Acta 33, 789–810. Renne, P.R., 2000. 40 Ar/39 Ar age of plagioclase from Acapulco meteorite and the problem of systematic errors in cosmochronology. Earth Planet Sci. Lett. 175, 13–26. Renne, P.R., Karner, D.B., Ludwig, K.R., 1998. Absolute ages aren’t exactly. Science 282, 1840–1841. Renne, P.R., Mundil, R., Balco, G., Min, K., Ludwig, K.R., 2010. Joint determination of 40 K decay constants and 40 Ar*/40 K for the Fish Canyon sanidine standard, and improved accuracy for 40 Ar/39 Ar geochronology. Geochim. Cosmochim. Acta 74, 5349–5367. Ryder, G., 2002. Mass flux in the ancient Earth–Moon system and benign implications for the origin of life on Earth. J. Geophys. Res. 107, 1–14. Rubin, A.E., Kallemeyn, G.W., 1989. Carlisle Lakes and Allan Hills 85151: members of a new chondrites grouplet. Geochim. Cosmochim. Acta 53, 3035–3044. Rubin, A.E., Rehfeldt, A., Peterson, E., Keil, K., 1983. Fragmental breccias and the collisional evolution of ordinary chondrite parent bodies. Meteoritics 18, 179–196. Rubin, A.E., Ulff-Moller, F., Wasson, J.T., Carlson, W.D., 2001. The Portales Valley meteorite breccia: evidence for impact melting and metamorphism of an ordinary chondrite. Geochim. Cosmochim. Acta 65, 323–342. Rubin, A.E., Scott, E.R.D., Keil, K., 1997. Shock metamorphism in enstatite chondrites. Geochim. Cosmochim. Acta 61, 847–858. Ruzicka, A., Boynton, W.V., Ganguly, J., 1994. Olivine coronas, metamorphism, and the thermal history of Morristown and Emery mesosiderites. Geochim. Cosmochim. Acta 58, 2725–2741. Ruzicka, A., Fowler, G.W., Gregory, A., Snyder, G.A., Prinz, M., Papike, J.J., Taylor, L.A., 1999. Petrogenesis of silicate inclusions in the Weekeroo Station IIE iron meteorite: differentiation, remelting, and dynamic mixing. Geochim. Cosmochim. Acta 63, 2123–2143. Ruzicka, A., Kilgore, M., Mittlefehldt, D.W., Fries, M.D., 2005. Portales Valley: petrology of a metallic-melt meteorite breccia. Meteorit. Planet. Sci. 40, 261–295. Ruzicka, A., Hutson, M., Floss, C., 2006. Petrology of silicate inclusions in the Sombrerete ungrouped iron meteorite: implications for the origins of IE-type silicate-bearing irons. Meteorit. Planet. Sci. 41, 1797–1831. Schmitz, B., Häggström, T., Tassinari, M., 2003. Sediment dispersed extraterrestrial chromite traces a major asteroid disruption event. Science 300, 961–964.

226

D.D. Bogard / Chemie der Erde 71 (2011) 207–226

Schmitz, B., Harper, D.A., Peucker-Ehrenbrink, B., Stouge, S., Alwmark, C., Cronholm, A., Bergström, S.M., Tassinari, M., Xiaofeng, W., 2008. Asteroid breakup linked to the great Ordovician biodiversification event. Nat. Geosci. 1, 49–53. Schwarz, W.H., Trieloff, M., 2007. Intercalibration of 40 Ar–39 Ar standards NL-25, HB3gr hornblende, GA1550, SB-3, HD-B1 biotote and BMus/2 muscovite. Chem. Geol. 242, 218–231. Scott, E.R.D., Haack, H., Love, S.G., 2001. Formation of mesosiderites by fragmentation and reaccretion of a large differentiated asteroid. Meteorit. Planet. Sci. 36, 869–881. Scott, E.R.D., 2002. Meteorite evidence for the accretion and collisional evolution of asteroids. In: Bottke, W.F., et al. (Eds.), Asteroids III. Univ. of Arizona Press, Tucson, pp. 697–709. Scott, E.D.R., 2007. Chondrites and the protoplanetary disk. Annu. Rev. Earth Planet. Sci. 35, 577–620. Shen, J.J., Papanastassiou, D.A., Wasserburg, G.J., 1996. Precise R-Os determinations and systematics of iron meteorites. Geochim. Cosmochim. Acta 60, 2887–2900. Shuster, D.L., Weiss, B.P., 2005. Martian surface paleotemperatures from thermochronology of meteorites. Science 309, 594–597. Sigura, N., Miyazaki, A., 2006. Mn–Cr ages of Fe-rich olivine in two Rumuruti (R) chondrites. Earth Planets Space 58, 689–694. Snyder, G.A., Lee, D.-C., Ruzicka, A.M., Prinz, M., Taylor, L.A., Halliday, A.N., 2001. Hf–W, Sm–Nd, and Rb–Sr isotopic evidence of late impact fractionation and mixing of silicates on iron meteorite parent bodies. Earth Planet. Sci. Lett. 186, 311–324. Srinivasan, G., Goswmi, J.N., Bhandari, N., 1999. 26Al in eucrite Piplia Kalan: plausible heat source and formation chronology. Science 284, 1348–1350. Srinivasan, G., Whitehouse, M.J., Weber, I., Yamaguchi, A., 2007. The crystallization age of eucritic zircon. Science 317, 345–347. Steiger, R.H., Jäger, E., 1977. Subcommission on geochronology: convention on the use of decay constants in geo- and cosmochronology. Earth Planet. Sci. Lett. 36, 359–362. Stewart, B., Papanastassiou, D.A., Wasserburg, G.J., 1994. Sm–Nd chronology and petrogenesis of mesosiderites. Geochim. Cosmochim. Acta 58, 3487–3509. Stewart, B., Papanastassiou, D.A., Wasserburg, G.J., 1996. Sm–Nd systematics of a silicate inclusion in the Caddo County IAB iron meteorite. Earth Planet. Sci. Lett. 143, 1–12. Stöffler, D., Keil, K., Scott, E.D.R., 1991. Shock metamorphism of ordinary chondrites. Geochim. Cosmochim. Acta 55, 3845–3867. Swindle, T.D., Olsen, E.K., 2004. 40 Ar–39 Ar studies of whole rock nakhlites: evidence for the timing of formation and aqueous alteration on Mars. Meteorit. Planet. Sci. 39, 755–766. Swindle, T.D., Kring, D.A., 2008. Chronological evidence for the late heavy bombardment in ordinary chondrite meteorites. In: Workshop on Early Solar System Impact Bombardment,. Lunar & Planetary Institute (abstract #3004). Swindle, T.D., Kring, D.A., Olson, E.K., Isachsen, C.E., 2006. Ar–Ar dating of shockmelted ordinary chondrites: chronology of asteroidal impacts. Lunar Planet. Sci. 37 (#1454 CD-ROM, abstract). Swindle, T.D., Isachsen, C.E., Weirich, J.R., Kring, D.A., 2009. 40 Ar–39 Ar ages of Hchondrite impact melt breccias. Meteorit. Planet. Sci. 44, 747–762. Sykes, M.V., Vilas, F., 2001. Closing in on HED meteorite sources. Earth Planets Space 53, 1077–1083. Takada, H., Graham, A.L., 1991. Degree of equilibration of eucritic pyroxenes and thermal metamorphism of the earliest planetary crust. Meteoritics 26, 129–134. Takeda, H., Bogard, D.D., Mittlefehldt, D.W., Garrison, D.H., 2000. Mineralogy, petrology, chemistry, and 39 Ar–40 Ar ages and exposure ages of the Caddo County IAB iron: evidence for early partial melt segregation of a gabbro area rich in plagioclase-diopside. Geochim. Cosmochim. Acta 64, 1311–1327. Takeda, H., Bogard, D.D., Otsuki, M., Ishii, T., 2003. Mineralogy and Ar–Ar age of the Tarahumara IIE iron, with reference to the origin of alkali-rich materials. In: International Symposium, Evolution of Solar System Materials: A New Perspective from Antarctic Meteorites,. National Institute of Polar Research, pp. 134–135 (abstract). Taylor, G.J., Maggiore, P., Scott, E.R.D., Rubin, A.E., Keil, K., 1987. Original structures, and fragmentation and reassembly histories of asteroids: evidence from meteorites. Icarus 69 (Issue 1), 1–13.

Tera, F., Papanastassiou, D., Wasserburg, G.J., 1974. Isotopic evidence for a terminal lunar cataclysm. Earth Planet. Sci. Lett. 22, 1–21. Terribilini, D., Eugster, O., Mittlefehldt, D.W., Diamond, L.W., Vogt, S., Wang, D., 2000. Mineralogical and chemical composition and cosmic-ray exposure history of two mesosiderites and two irons. Meteorit. Planet. Sci. 35, 617–628. Treiman, A.H., Gleason, J.D., Bogard, D.D., 2000. The SNC meteorites are from Mars. Planet. Space Sci. 48, 1213–1230. Trieloff, M., Jessberger, E.K., Herrwerth, I., Hopp, J., Fléni, C., Ghélls, M., Bourot-Denise, M., Pellas, P., 2003. Structure and thermal history of the H-chondrite parent asteroid revealed by thermochronometry. Nature 422, 502–506. Turner, G., 1969. Thermal histories of meteorites by the 39 Ar–40 Ar method. In: Millman, P. (Ed.), Meteorite Research. D. Reidel, Dordrecht, Netherlands, pp. 407–417. Turner, G., 1977. Potassium–argon chronology of the Moon. Phys. Chem. Earth 10, 145–195. Turner, G., Enright, M.C., Cadogan, P.H., 1978. The early history of the chondrite parent bodies inferred from 40 Ar–39 Ar ages. In: Ninth Lunar Planet. Sci., pp. 989–1025, Lunar Science Institute. Turner, G., Knott, S.F., Ash, R.D., Gilmour, J.D., 1997. Ar–Ar chronology of the martian meteorite ALH84001: evidence for the timing of the early bombardment of Mars. Geochim. Cosmochim. Acta 61, 3835–3850. Vogel, N., Renne, P.R., 2008. 40 Ar–39 Ar dating of plagioclase grain size separates from silicate inclusions in IAB iron meteorites and implications for the thermochronological evolution of the IAB parent body. Geochim. Cosmochim. Acta 72, 1231–1255. Wadhwa, M., Amelin, Y., Bogdanovski, O., Shukolyukov, A., Lugmair, G., Janney, P., 2009. Ancient relative and absolute ages for a basaltic meteorite: implications for timescales of planetesimal accretion and differentiation. Geochim. Cosmochim. Acta 73, 5189–5201. Wadhwa, M., Srinivasan, G., Carlson, R.W., 2006. Timescales of planetesimal differentiation in the early solar system. In: Lauretta, D.S., McSween, H.Y. (Eds.), Meteorites and the Early Solar System II. Univ. Arizona Press, pp. 715–732. Walton, E.L., Kelly, S.P., Spray, J.G., 2007. Shock implantation of martian atmospheric argon in four basaltic shergottites: a laser probe 40 Ar/39 Ar investigation. Geochim. Cosmochim. Acta 71, 497–520. Wasserburg, G.J., Sanz, H.G., Bence, A.E., 1968. Potassium-feldspar phenocrysts in the surface of Colomera, an iron meteorite. Science 161, 684–687. Wasson, J.T., Wang, J., 1986. A nonmagmatic origin of group IIE iron meteorites. Geochim. Cosmochim. Acta 50, 725–732. Wasson, J.T., Wang, S., 1991. The histories of ordinary chondrite parent bodies: U,Th–He age distributions. Meteoritics 26, 161–167. Wasson, J.T., Kallemeyn, G.W., 2002. The IAB iron meteorite complex: a group, five subgroups, numerous grouplets, closely related, mainly formed by crystal segregation in rapidly cooling melts. Geochim. Cosmochim. Acta 66, 2445–2473. Weirich, J.R., Wittman, A., Isachsen, C.E., Rumble, D.R., Swindle, T.D., Kring, D.A., 2011. The Ar–Ar age and petrology of Miller Range 05029: evidence for a large impact in the very early solar system. Meteoritics Planet. Sci. 45, 1868–1888. Whitby, J.A., Crowther, S., Busfield, A., Gilmour, J.D., 2005. Inhomogeneity on the lodranite parent body inferred from I–Xe systematics. Lunar Planet. Sci. 36 (#1658 CD-ROM, abstract). Wiens, R.C., Becker, R., Pepin, R.O., 1988. The case for a martian origin of the shergottites: trapped and indigenous gas components in EETA79001 glass. Earth Planet. Sci. Lett. 77, 149–158. Wiens, R.C., Huss, G.R., Burnett, D.S., 1999. The solar oxygen isotopic composition: predictions and implications for solar nebula processes. Meteorit. Planet. Sci. 34, 99–107. Yamaguchi, A., Taylor, G.J., Keil, K., 1997. Metamorphic history of the eucritic crust of 4 Vesta. J. Geophys. Res. 102, 13381–13386. Yang, C.W., Williams, D.B., Goldstein, J.I., 1997. A new empirical cooling rate indicator for meteorites based on the size of the cloudy zone of the metallic phases. Meteorit. Planet. Sci. 32, 423–429. Zhang, Y., Sears, D.W.G., 1996. The thermometry of enstatite chondrites: a brief review and update. Meteorit. Planet. Sci. 31, 647–656. Zhang, Y., Benoit, P.H., Sears, D.W.G., 1995. The classification and complex thermal history of the enstatite chondrites. J. Geophys. Res. 100 (No. E5), 9417–9438.