⁵³Mn–⁵³Cr CHRONOMETRY OF CB CHONDRITE: EVIDENCE FOR UNIFORM DISTRIBUTION OF ⁵³Mn IN THE EARLY SOLAR SYSTEM

Precise age determination of planetary materials and their constituent phases can place important constraints on the processes responsible for the evolution of our solar system. While the use of long-lived radio nuclides can place the ages of planetary materials onto an absolute timescale, its temporal resolution is often limited due to the long half-lives of the parent nuclides. An exception to this is the U–Pb (or more specifically Pb–Pb) chronometer that can sometimes provide very precise absolute ages. However, application of this chronometer is limited to phases with very high uranium to lead ratio such as the calcium–aluminum-rich inclusions (CAIs), pyroxene in angritic meteorites, and some chondrules (Allègre et al. 1995; Amelin et al. 2002; Lugmair & Galer 1992; Amelin & Krot 2007; Amelin 2008; Connelly et al. 2008a, 2008b). In addition, complexity in the interpretation of the Pb isotopic data arises from the fact that it is often necessary to undertake intense chemical leaching (or washing) in order to remove the less-radiogenic components (e.g., Amelin et al. 2002; Amelin & Krot 2007), which hints toward the existence of secondary lead in the system, in addition to the radiogenic lead produced via decay of uranium.

The short-lived chronometers, on the other hand, can only be used to determine the "relative" ages, or require precise time anchors (i.e., sample with known absolute age) to place the chronological information on an absolute timescale. However, they are particularly sensitive to events that occurred in the early solar system. Within the short-lived chronometers, the ⁵³Mn–⁵³Cr system (half-life (t_1/2): 3.7 ± 0.4 Ma (Honda & Imamura 1971)) is particularly suited for the study of the first ∼20 million years of the solar system. However, the use of this system as a robust chronometer relies on the fundamental assumption that ⁵³Mn was uniformly distributed within the protoplanetary disk, which must be rigorously tested through cross-calibration with other chronometers. Here, we report the first high-precision ⁵³Mn–⁵³Cr age of chondrules and metal grain from CB chondrite Gujba. CB chondrites are ideal samples for this study because ages using other isotope systems, including both short- and long-lived chronometers, are available. However, the major challenge to the application of the Mn–Cr system arises from the geochemical nature of CB chondrites, where the chondrules are generally depleted in a moderately volatile element Mn, resulting in a low Mn/Cr ratio (Krot et al. 2001; Rubin et al. 2003). In order to overcome this problem, we have taken advantage of the recently developed analytical techniques for the high-precision Cr isotope analysis (Yamakawa et al. 2009). In addition, petrographical properties of these chondrules and metal grain were investigated using scanning electron microscopy and electron microprobe analysis to assist the interpretation of the isotopic data.

Chondrules and metal grain were extracted from CB chondrite Gujba. In all cases, samples were sliced into two pieces, one of which was used for Cr isotopic analysis, and the remaining portion was mounted onto the SpeciFix-40 epoxy resin for petrographical analysis. Backscattered electron (BSE) imaging was carried out using a Hitachi S-3100H scanning electron microscope equipped with a Horiba EMAX-7000 energy dispersive spectrometer. X-ray elemental maps were obtained using a JEOL JXA-8800 electron probe micro analyzer. Chondrule fragments for isotopic analysis were cleaned by ultrasonicating in distilled acetone, 0.1 mol l⁻¹ HCl and H₂O before pulverizing with SiN mortar. Chondrules were dissolved in HF+HNO₃ mixture using Teflon bombs at 190°C (samples labeled SC in Table 1) or in 7 ml Teflon vials (samples labeled MC in Table 1). Samples dissolved in 7 ml Teflon vials were leached in 6 mol l⁻¹ HCl for >48 hr under ultrasonic agitation (L = leachate) prior to total dissolution in HF+HNO₃ (R = residue). The metal grain (labeled M in Table 1) was dissolved in 6 mol l⁻¹ HCl and 1 mol l⁻¹ HNO₃. The chromium isotope ratios were measured using a Thermo-Finnigan TRITON-TI thermal ionization mass spectrometer. The chemical separation and mass spectrometry of Cr follows the method described in Yamakawa et al. (2009). The method for ⁵⁵Mn/⁵²Cr and Fe/Cr ratio measurements is shown in Makishima et al. (2010).

The BSE images and the X-ray elemental maps of the chondrules are shown in Figures 1(a)–(e). The petrographical properties of the chondrules examined in this study are generally similar to those of chondrules studied by Rubin et al. (2003) and Krot et al. (2005). All five chondrules are skeletal-olivine chondrules, and ferromagnesian components are dominant. Olivine in these chondrules is forsteritic. Low-Ca pyroxene coexists with the skeletal-olivine in various proportions within the chondrules. The Al-rich mesostasis is distributed either as elongated laths along the ferromagnesian minerals (A-1-2-1-C1, A-1-3-3-C1, A-1-3-3-C2, and A-1-4-1-C3) or as tiny grains of <2–3 μm in size (A-1-3-2-C1). Grains of high-Ca pyroxene are distributed among the other phases in A-1-2-1-C1 (Figure 1(a)). The metal grain A-1-4-1-M1 (∼5 mm in diameter) is enriched in Fe (88.0 wt.% Fe). It contains 5.7 wt.% Ni, 5.1 wt.% Si, and 0.4 wt.% P. The distribution of Fe is generally homogeneous (Figure 1(f)). Sulfur-rich globules of up to several tens micron in size are heterogeneously distributed (Figure 1(g)). Compared to other chondrules, A-1-2-1-C1 is depleted in Mn and Cr, and shows a distinctively lower Mn/Cr ratio compared to other chondrules (Table 1). This may imply that the precursors of these chondrules were compositionally heterogeneous.

Zoom In Zoom Out Reset image size

The results for the Cr isotopic analysis are shown in Table 1 and Figure 2. Due to the relatively long exposure age of 26 ± 7 Ma of this meteorite (Rubin et al. 2003), correction for the spallation reaction was applied to the metal grain using the ⁵³Cr and ⁵⁴Cr production rates in Fe targets (Birck & Allègre 1985; Lugmair & Shukolyukov 1998). Our new data for ε⁵⁴Cr are in good agreement with the less precise data of this meteorite by Trinquier et al. (2007, 2008). It is characterized by a positive ε⁵⁴Cr value with a mean of +1.29 ± 0.02. Importantly, all chondrules and metal grain analyzed here have ε⁵⁴Cr values that overlap within the analytical uncertainty, indicating that chondrules and metal grain from Gujba evolved in an isotopically uniform reservoir. Homogenization of the Cr isotope may have taken place during chondrule/metal formation in an active region of the early solar nebula (e.g., Krot et al. 2001) or during an asteroid impact near or after the dissipation of dust in the protoplanetary disk (Rubin et al. 2003; Krot et al. 2005).

Zoom In Zoom Out Reset image size

The ε⁵³Cr values of Gujba, on the other hand, display a tight correlation with the ⁵⁵Mn/⁵²Cr ratio (Figure 2(a)). Both chondrules and the metal grain fall on a straight line, consistent with their formation from a common reservoir with an additional contribution from the decay of now extinct ⁵³Mn. The slope of the line calculated using the Ludwig Isoplot 3 program corresponds to an initial ⁵³Mn/⁵⁵Mn ratio of (3.18 ± 0.52) × 10⁻⁶. This value is clearly lower than that of chondrules from carbonaceous chondrites Allende with an initial ⁵³Mn/⁵⁵Mn ratio of (6.23 ± 0.82) × 10⁻⁶ (Yin et al. 2009b). It also appears to be lower than that of chondrules from primitive ordinary chondrite Chainpur ((5.1 ± 1.6) × 10⁻⁶; Yin et al. 2007), although strictly speaking, the relatively large uncertainty associated with the data for ordinary chondrite makes them marginally overlap within the uncertainty. Using angritic meteorites as time anchors (i.e., sample with a known absolute age and an initial ⁵³Mn/⁵⁵Mn ratio), the initial ⁵³Mn/⁵⁵Mn ratio of this sample can be converted to an absolute age. The two most commonly used time anchors are angrites D'Orbigny and LEW 86010, with an absolute age and an initial ⁵³Mn/⁵⁵Mn ratio of 4564.42 ± 0.12 Ma and (3.24 ± 0.04) × 10⁻⁶, and 4558.55 ± 0.15 Ma and (1.25 ± 0.07) × 10⁻⁶, respectively (Amelin 2008; Glavin et al. 2004; Lugmair & Shukolyukov 1998). Using these time anchors, the age of Gujba chondrules/metal grain translates to 4564.3 ± 0.9 Ma and 4563.5 ± 1.1 Ma, respectively. These ages are consistent with the generally young age of CB chondrite obtained using other chronometers (Table 2). An exception to this is the presence of small discrepancy between Pb–Pb age (4562.68 ± 0.49 Ma) and Mn–Cr age (4564.3 ± 0.9 Ma) of Gujba chondrules when D'Orbigny is used as a time anchor. This discrepancy cannot be eliminated by revision of ⁵³Mn half-life (cf. Yoneda et al. 2002), since the initial ⁵³Mn/⁵⁵Mn ratios of D'Orbigny and Gujba are very close (i.e., similar within the uncertainty). Thus, the above discrepancy implies that the initial ⁵³Mn/⁵⁵Mn and/or the absolute age of D'Orbigny need to be re-examined, and/or the Pb–Pb age of Gujba chondrules represents the age of an event subsequent to its formation that reset the Pb isotope systematics of the meteorite parent body. Among these possibilities, the ⁵³Mn/⁵⁵Mn ratio of D'Orbigny determined by Glavin et al. (2004) was confirmed by Yin et al. (2009a). On the other hand, a recent cross-calibration between I–Xe and Pb–Pb ages of various meteoritic samples, including Gujba chondrules, has demonstrated that chondrules from Gujba show discordant I–Xe and Pb–Pb ages, where the I–Xe system shows an age ∼1.5 Ma older than the Pb–Pb system (Gilmour et al. 2009). The I–Xe age obtained by these authors is in good agreement with the Mn–Cr age of this study. This, along with the observation that some chondrules in Gujba (chondrules 1 and 2 of Krot et al. 2005) clearly show disturbed Pb isotope signature represented by a younger (∼4552 Ma) age, may support the interpretation of high-precision Pb–Pb age of Gujba chondrules representing the "minimum age" of chondrule formation. A distinct possibility of age discrepancy may be the use of commonly assumed ²³⁸U/²³⁵U of 137.88 to calculate the Pb–Pb age of D'Oribigny. A recent study on the uranium isotopic variations in meteorites has shown that some components in meteorites such as CAIs display a range of ²³⁸U/²³⁵U ratios, probably resulting from the decay of extant ²⁴⁷Cm (Brennecka et al. 2010b). The ²³⁸U/²³⁵U ratio of pyroxene fraction of D'Oribigny has also been investigated, and a value of 137.822 ± 0.028 was obtained (Brennecka et al. 2010a). If we use this value instead of 137.88 to calculate the Pb–Pb age of D'Oribigny, its absolute age will be 4563.8 ± 0.4 Ma. This, in turn, will make the D'Orbigny anchored ages younger by 0.6 ± 0.4 Ma, changing the Mn–Cr age of Gujba chondrule/metal grain to 4563.7 ± 1.2 Ma. This age is within the uncertainty with the Pb–Pb age of Gujba chondrules (Krot et al. 2005). One concern with this treatment (other than the fact that only a single measurement for the ²³⁸U/²³⁵U ratio has so far been performed for D'Orbigny) is that if the uranium isotopic heterogeneity is a common feature in meteorites and their constituent phases, the ²³⁸U/²³⁵U ratio of Gujba chondrules may also be anomalous. If this is the case, further detailed comparison between Mn–Cr and Pb–Pb ages of Gujba chondrules will only be meaningful after their ²³⁸U/²³⁵U data are available.

Nevertheless, the overall consistency in the ages of CB chondrite using different chronometers supports the uniform distribution of ⁵³Mn in the early solar nebula and argues against the model that proposes the presence of ⁵³Mn gradient within the early solar system (Lugmair & Shukolyukov 1998). The homogeneous distribution of ⁵³Mn allows the use of ⁵³Mn–⁵³Cr systematics as a powerful tool to investigate the high-resolution chronological information recorded in meteorites characterized by distinct ε⁵⁴Cr signatures (Trinquier et al. 2007; Yamakawa et al. 2010).

We have undertaken detailed Cr isotopic analyses of chondrules and metal grain of CB chondrite Gujba and confirmed the presence of now extinct ⁵³Mn in the early solar system where these materials formed. A precise ⁵³Mn/⁵⁵Mn ratio of (3.18 ± 0.52) × 10⁻⁶ was obtained from the slope of the linear correlation on the ε⁵³Cr versus ⁵⁵Mn/⁵²Cr evolution diagram. This value converts to absolute ages of 4563.7 ± 1.2 Ma and 4563.5 ± 1.1 Ma using angrites D'Orbigny and LEW 86010, respectively, as time anchors. These ages are in good agreement with the ages obtained for CB chondrites using other short- and long-lived chronometers, supporting the uniform distribution of ⁵³Mn in the early solar system. Nevertheless, the presence of small discrepancy between the D'Orbigny anchored Mn–Cr age (when ²³⁸U/²³⁵U = 137.88 is used to calculate the Pb–Pb age of D'Orbigny) and the Pb–Pb age of Gujba chondrules asks for further investigation of (1) the absolute age of D'Orbigny, (2) the Pb-Pb age of Gujba chondrules, and (3) the uranium isotopic composition of chondritic and achondritic meteorites and their constituent components in order to "fine-tune" the cross-calibration of the chronometers.

The authors thank Drs. Q.-Z. Yin (University of California Davis) and N. Tomioka (Okayama University) for discussion and comments, and the members of the Pheasant Memorial Laboratory for their assistance in the laboratory. This research was supported by Grants-in-aid from the Japan Society for the Promotion for Science (JSPS) to K.Y. and the program of the "Center of Excellence for the 21st Century in Japan" from the Ministry of Education, Culture, Sports, Science, and Technology (MEXT).