HETEROGENEOUS ISOTOPIC ANOMALIES OF SM AND GD IN THE NORTON COUNTY METEORITE: EVIDENCE FOR IRRADIATION FROM THE ACTIVE EARLY SUN

The Norton County aubrite is a fragmental impact breccia consisting of coarse and angular enstatite crystal fragments, embedded into a clastic matrix (Okada et al. 1988). In this study, whole rock, handpicked material of the light-colored phase, and chemical separates obtained by a sequential acid-leaching method were prepared from the same single fragment of the Norton County meteorite weighing about 5 g.

Three whole rock samples (hereafter referred to as NCB1, NCB2, and NCB3) and the handpicked sample (NCW) weighing about 0.2 g were decomposed individually by HF–HClO₄. The handpicked material of the light-colored phase was a concentrate of mainly enstatite crystals, as determined by electron probe micro-analysis.

Sequential acid-leaching treatments were carried out to obtain chemically different phases from the single samples. About 0.3 g of the powdered sample was leached using 5 mL of 0.1 M acetic-acid–ammonium acetate, 0.1 M HCl, 2 M HCl, and aqua regia, successively. Finally, the residue was decomposed by treatment with HF–HClO₄ with heating. These five leaching fractions (NCL) are designated as 1, 2, 3, 4, and 5, respectively. The leaching experiments were performed twice. The fractions obtained from the first and second experiments were identified by using subordination numbers 1 and 2, respectively, after the 1–5 notation.

The samples, namely, three whole rocks (NCB1, NCB2, and NCB3), one light-colored phase (NCW), and 10 leachates (NCL#1–1 to NCL#5–2) were then taken to dryness and redissolved in 1 mL of 2 M HCl. The solutions were divided into two portions: the main portion for Sm and Gd isotopic measurements and the rest for the determination of REE abundances.

Most of the sample solution was used for the Sm and Gd isotopic measurements. Sm and Gd were separated by a conventional method using cation exchange methods (Hidaka & Yoneda 2007). The solution was loaded onto a cation exchange resin column and washed with 4.7 mL of 2 M HCl before the REE were eluted with 3 mL of 6 M HCl. The solution was then evaporated to dryness and redissolved in a drop of 0.1 M HCl. The solution was loaded onto a second column packed with LN resin to separate Sm and Gd using 0.35 and 0.5 M HCl, respectively. A Micro-mass VG54–30 thermal ionization mass spectrometer equipped with seven Faraday cup collectors was used for the isotopic measurements of Sm and Gd. Data collection was performed using the static multi-mode. However, seven collectors are not sufficient to simultaneously measure all seven isotopes of Sm and Gd and associated isobaric interferences. Therefore, in this study, ¹⁴⁴Sm and ¹⁵²Gd were not monitored for Sm and Gd isotopic analyses, respectively.

For Sm isotopic measurements, isobaric interference from Nd (¹⁴⁴Nd, ¹⁴⁸Nd, and ¹⁵⁰Nd), and Gd (¹⁵²Gd and ¹⁵⁴Gd) are possible. The seven Faraday cup collectors were configured to monitor ¹⁴⁶Nd, ¹⁴⁷Sm, ¹⁴⁸Sm, ¹⁴⁹Sm, ¹⁵⁰Sm, ¹⁵²Sm, and ¹⁵⁴Sm during the Sm measurements. In the measurement of background after every 10 data collection during the Sm analytical scheme, the contribution from Gd interference was checked at mass 155. However, no Gd interference was detected in all samples. Correction for the instrumental mass fractionation was performed using an exponential law normalized to ¹⁴⁷Sm/¹⁵²Sm = 0.56081 (Russ et al. 1972).

¹⁵⁴Sm, ¹⁵⁶Dy, ¹⁵⁸Dy, and ¹⁶⁰Dy can interfere with Gd isotopes. The collectors were configured to monitor ¹⁵⁴Gd, ¹⁵⁵Gd, ¹⁵⁶Gd, ¹⁵⁷Gd, ¹⁵⁸Gd, ¹⁶⁰Gd, and ¹⁶¹Dy. Contribution from Sm interference was checked at mass 150 during the background measurement. Since Sm usually evaporates at lower temperature than Gd by using thermal ionization mass spectrometry, no Sm interference was detected in all of the Gd measurements. For normalization of Gd isotopic ratios, ¹⁵⁶Gd/¹⁶⁰Gd = 0.9361 has been generally used (Eugster et al. 1970). However, the ¹⁵⁶Gd/¹⁶⁰Gd isotopic ratio cannot be fixed to a constant value for the neutron-irradiated materials because of the possible isotopic shifts from ¹⁵⁵Gd to ¹⁵⁶Gd. In order to solve this issue, modified ¹⁵⁶Gd/¹⁶⁰Gd ratio (0.9361+α) is used for the normalization factor, assuming that a depletion of ¹⁵⁵Gd is equal to an enrichment of ¹⁵⁶Gd (Hidaka et al. 1995; Hidaka & Yoneda 2007). From the isotopic data for non-irradiated material, the sum of ¹⁵⁵Gd/¹⁶⁰Gd and ¹⁵⁶Gd/¹⁶⁰Gd isotopic ratios is fixed to 1.61290. The α values in individual samples are determined when the normalization by ¹⁵⁶Gd/¹⁶⁰Gd = 0.9361+α leads to ¹⁵⁵Gd/¹⁶⁰Gd = 0.67680–α.

Another minor aliquot of each sample solution was once evaporated to dryness and redissolved using 5 mL of 0.5 M HNO₃. A 0.05 g quantity of a 10 ppb indium solution was added to the individual sample solutions as an internal standard element to optimize the analytical conditions for REE measurements. ICP-MS (Agilent 7500cx) was used to determine the REE contents.

Among the samples of chemical separates, the isotopic data of fractions of NCL#3–NCL#5 from both two-times leaching experiments could not be obtained because of their low elemental abundances of Sm and Gd. All the other samples measured in this study show significant isotopic shifts of ¹⁵⁰Sm/¹⁴⁹Sm and ¹⁵⁸Gd/¹⁵⁷Gd. There is a significant difference in the Sm and Gd isotopic shifts, even in one of three whole rock samples, as shown in Table 1. The isotopic variation patterns of Sm and Gd illustrated in Figure 1 clearly show that the isotopic deviations come predominantly from neutron capture and are not from nucleosynthetic anomalies. Figure 2 shows correlation diagrams of isotopic shifts for (a) ¹⁴⁹Sm/¹⁵²Sm versus ¹⁵⁰Sm/¹⁵²Sm and (b) ¹⁵⁷Gd/¹⁶⁰Gd versus ¹⁵⁸Gd/¹⁶⁰Gd. These data give neutron fluences ranging from 1.89 × 10¹⁶ to 3.54 × 10¹⁶ n cm⁻². In particular, one of three sets of whole rock data (3.54 × 10¹⁶ n cm⁻²) is 2.3–2.9 times higher than data reported previously (Hidaka et al. 1999, 2006), although the other two sets of data are almost similar to each other and to previous data. The variation in neutron fluences can be caused by different depths in the case of a large-sized meteorite body. Actually, in the case of Norton County, the neutron fluences of the specimens taken from different depths at 10–37 cm vary significantly; they range from 1.22 × 10¹⁶ to 3.1 × 10¹⁶ n cm⁻² (Herzog et al. 2011). There is, however, no difference in depth among the samples used in this study. The three bulk samples were taken from a small fragment weighing about 5 g. Furthermore, the isotopic variations of Sm and Gd in the acid-leachates and the hand-picked sample are more heterogeneous and greater than those in the whole rock samples, corresponding to neutron fluences ranging from 1.82 × 10¹⁶ to 19.8 × 10¹⁶ n cm⁻².

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The chemical composition of the target material is yet another factor that alters the production rate of neutrons, and contributes to the variation in isotopic shifts of Sm and Gd. However, the neutrons produced in the target need to travel over long distances to reduce their energy to the thermal neutron level. Therefore, it is impossible to explain such an extremely large variation of the neutron production rate found in the three whole rock samples based on the difference of the chemical composition on the microscopic scale. The results are best explained by the migration of alien materials that had been highly irradiated into the Norton County matrix portion.

The calculation of cosmic-ray-produced neutron equilibrium spectra using the neutron transport equation was developed by Lingenfelter et al. (1972), hereafter denoted as the LCH model. The combination of Sm and Gd isotopic data from one sample is very useful for the determination of the neutron energy spectrum. The latter provides information on the sizes of meteoroids and the depths of meteorite samples during irradiation. In particular, the difference in neutron capture resonance between ¹⁴⁹Sm (0.0973 eV) and ¹⁵⁷Gd (0.0314 eV) can be effectively used to estimate the energy level of thermal neutrons (Lingenfelter et al. 1972). Since the resonance in ¹⁴⁹Sm is at higher energy than that in ¹⁵⁷Gd, the ratio of the two isotopic shifts of ¹⁵⁰Sm/¹⁴⁹Sm and ¹⁵⁸Gd/¹⁵⁷Gd is dependent on the neutron energy spectral shape and can be calculated as follows:

$$\begin{equation*} \frac{{\varepsilon _{{\rm Sm}} }}{{\varepsilon _{{\rm Gd}} }} = \frac{{\frac{{\left({{{{}^{150}{\rm Sm}} \mathord{\left/ {\vphantom {{{}^{150}{\rm Sm}} {{}^{149}{\rm Sm}}}} \right. \kern-\nulldelimiterspace} {{}^{149}{\rm Sm}}}} \right)_{{\rm sample}} - \left({{{{}^{150}{\rm Sm}} \mathord{\left/ {\vphantom {{{}^{150}{\rm Sm}} {{}^{149}{\rm Sm}}}} \right. \kern-\nulldelimiterspace} {{}^{149}{\rm Sm}}}} \right)_{{\rm STD}} }}{{1\, + \,\left({{{{}^{150}{\rm Sm}} \mathord{\left/ {\vphantom {{{}^{150}{\rm Sm}} {{}^{149}{\rm Sm}}}} \right. \kern-\nulldelimiterspace} {{}^{149}{\rm Sm}}}} \right)_{{\rm sample}} }}}}{{\frac{{\left({{{{}^{158}{\rm Gd}} \mathord{\left/ {\vphantom {{{}^{158}{\rm Gd}} {{}^{157}{\rm Gd}}}} \right. \kern-\nulldelimiterspace} {{}^{157}{\rm Gd}}}} \right)_{{\rm sample}} - \left({{{{}^{158}{\rm Gd}} \mathord{\left/ {\vphantom {{{}^{158}{\rm Gd}} {{}^{157}{\rm Gd}}}} \right. \kern-\nulldelimiterspace} {{}^{157}{\rm Gd}}}} \right)_{{\rm STD}} }}{{1 \,+\, \left({{{{}^{158}{\rm Gd}} \mathord{\left/ {\vphantom {{{}^{158}{\rm Gd}} {{}^{157}{\rm Gd}}}} \right. \kern-\nulldelimiterspace} {{}^{157}{\rm Gd}}}} \right)_{{\rm sample}} }}}}. \end{equation*}$$

The value of ε_Sm/ε_Gd is therefore an indicator of the shape of neutron spectrum. The production rate of thermal neutrons also depends on the chemical composition and temperature of the target material (Lingenfelter et al. 1972). The chemical composition of the target is treated as the effective total macroscopic neutron capture cross section, Σ_eff (cm² g⁻¹), which is mainly estimated from the contents of Al, Ca, Fe, Mg, Si, and Ti as the major element component, and Sm, Eu, and Gd as the trace element component (Lingenfelter et al. 1972). In the LCH model, the value of ε_Sm/ε_Gd at different temperatures in the range from 0 to 400 K is predicted as a function of Σ_eff. The neutron energy spectrum proposed by the LCH model has been effectively used to discuss the exposure conditions of the planetary materials. Aubrites generally show higher neutron production rates than other meteorite species because they have lower contents of Fe and Ti as neutron absorbers of major element component in the rock sample. The macroscopic neutron capture cross section Σ_eff (cm² g⁻¹) is effectively used to characterize the chemical properties of the whole rock sample. Aubrites show lower Σ_eff values (0.0015–0.0023) than chondrites (0.0070–0.0096), lunar regolith (0.0042–0.0107), and so on (Hidaka et al. 1999). The estimated ε_Sm/ε_Gd, Σ_eff and the neutron fluence Ψ of individual fractions treated in this study are listed in Table 2.

A variation in ε_Sm/ε_Gd as a function of Σ_eff is interpreted to be the result of differences in the neutron energy spectra (Lingenfelter et al. 1972). In general, aubrites show ε_Sm/ε_Gd values ranging from 0.37 to 0.66 (Hidaka et al. 1999, 2006). A difference in Σ_eff values might produce a slight variation in ε_Sm/ε_Gd values. Figure 3 shows a variation in ε_Sm/ε_Gd as a function of Σ_eff, which is interpreted to be the result of differences in the neutron energy spectra. A shaded zone reveals calculated neutron energy spectra at a temperature in the range from 0 to 400 K based on the LCH model. For comparison, the data of representative extraterrestrial materials such as the Apollo lunar regolith (A-15, 16, and 17), aubrites and chondrites are cited (Hidaka et al. 1999, 2000a, 2000b, 2006; Hidaka & Yoneda 2007), and also plotted in the same figure. The samples with high neutron fluences, NCW and NCL#2-2, show much higher ε_Sm/ε_Gd values (2.08–2.24) than others (0.49–0.58). The data suggest that the two samples were irradiated by higher energy neutrons relative to thermalized neutrons. Interestingly, similar results are reported for the acid leachates of the Kapoeta meteorite (howardite), which is known as a solar-gas-rich meteorite (Hidaka & Yoneda 2009). High ε_Sm/ε_Gd values of the Kapoeta leachates (ε_Sm/ε_Gd = 1.9) indicate the possibility of the interaction with solar cosmic rays (SCR) and activity from the early Sun.

Zoom In Zoom Out Reset image size

In general, irradiation by galactic cosmic rays (GCR) produces secondary neutrons that are slowed down to thermal energy in a target material. Therefore, the neutrons produced by GCR require a certain distance to travel in the target and lose their energies down to the thermal level. The meteorites with large meteoroidal bodies such as Norton County are able to produce thermal neutrons by GCR irradiation. However, the coexistence of a variable energy type of neutrons in the small fragments, as shown in this study, is difficult to explain when based only on simple GCR irradiation results. One probable interpretation is the existence of alien materials that experienced different irradiation histories from other materials in the Norton County meteorite matrix.

It is interesting to note that the high ε_Sm/ε_Gd samples of the Norton County fractions show high neutron fluences. Figure 4 is a diagram showing the relationship between ε_Sm/ε_Gd and neutron fluences for individual aubrite samples from both this study and previous studies (Hidaka et al. 1999, 2006). The data in the figure can be explained by the two-component mixing between high ε_Sm/ε_Gd with a high neutron fluence component and others. One of the whole rock samples, NCB2, and one of the leachates, NCL#1-2, also show slightly high ε_Sm/ε_Gd values (0.77 and 1.09, respectively), suggesting that these two samples might have been affected by highly irradiated materials. The previously reported ε_Sm/ε_Gd values of Norton County (0.50–0.51) are similar to those recorded for two of the three whole rock samples, NCB1 and NCB2, used in this study. As the LCH model predicts, the ε_Sm/ε_Gd value varies with the chemical composition Σ_eff, as shown in the figure. However, even considering the differences in Σ_eff values among individual samples, NCW, NCL#2-2 and NCL#1-2, it is difficult to obtain the high ε_Sm/ε_Gd values predicted by the LCH model.

Zoom In Zoom Out Reset image size

Very high ε_Sm/ε_Gd values of some components in Norton County cannot be explained, as far as we know, only from simple studies and from the Sm and Gd isotopic studies of meteorites and lunar samples, except in the case of Kapoeta. The acid leachates of Kapoeta also show high ε_Sm/ε_Gd values (=1.9), suggesting a different neutron irradiation environment for some components in Kapoeta (Hidaka & Yoneda 2009). As there is a significant amount of solar gas in Kapoeta, some components might have experienced SCR irradiation. Large enrichments of spallation-produced ²¹Ne and ³⁸Ar in the irradiated grains of Kapoeta are considered to be due to irradiation by energetic proton fluence from an active early (T Tauri) Sun (Caffee et al. 1987). The ε_Sm/ε_Gd values of the leachates from Kapoeta include large analytical uncertainties because of small isotopic shifts of Sm and Gd. On the other hand, the data from Norton County are much clearer, because the isotopic shifts of Sm and Gd are much larger than those of Kapoeta. This enables us to assess interaction with high-energy neutrons.

Several attempts have been made to determine the bulk chemical compositions of aubrites. However, aubrites are coarse-grained rocks with minor and trace minerals very heterogeneously distributed throughout the rocks (Wolf et al. 1983). The REE contents of the aubrites provide important information to understand the formation and differentiation process of the meteorite parent body. In order to characterize the probable alien materials having high ε_Sm/ε_Gd and neutron fluence, REE analyses of the individual samples used for the isotopic studies were performed.

The occurrence of some of the exotic sulfides and metallic phases is explained by the formation of aubrites under highly reducing conditions. Among the sulfide minerals in the aubrites, oldhamite (CaS) is a particular material of interest. Oldhamite is known as a major REE carrier in the aubrites (Floss et al. 1990). The variety of REE patterns found in oldhamite from aubrites is considered not to have been produced by a single igneous process (Floss et al. 1990; Floss & Crozaz 1993). Some would have been difficult to produce by igneous processes, but could have been produced by nebular condensates. Furthermore, oldhamite is very refractory; it has a melting point of 2450 ºC and is calculated to be among the first condensates in a reduced solar gas (Larimer 1975; Larimer & Bartholomay 1979; Lodders & Fegley 1993). Since oldhamite is a water-soluble mineral, we planned to chemically separate the oldhamite by the first step of an acid-leaching procedure (see in the description of Section 2.1). The REE contents of individual fractions treated in this study are given in Table 3. The first and second leachates (NCL#1 and #2) of the sequential acid-leaching fractions show higher REE contents than the others (NCL#3, #4, and #5), suggesting the effective dissolution of the oldhamite in the first and second fractions.

The C1-chondrite normalized REE patterns of individual fractions are shown in Figure 5. Norton County shows two different REE patterns in the fractions from individual acid-leaching procedures over two trials. As shown in Figure 5(b), NCL#1-2 and NCL#2-2, from the first and second fractions in the second leaching run, have a relatively flat LREE, and small but clear depletions of Eu and Yb. Similarly, the REE pattern of the NCW phase in Figure 5(b) also has slightly negative anomalies of Eu and Yb. The presence of strong depletions in Eu and Yb in the REE pattern is a property of refractory materials, such as most of the hibonites produced in the early condensation situations (Ireland et al. 1992; Simon et al. 2002; Liu et al. 2009), because Eu and Yb show higher volatilities than the other REEs. Depletions in Eu and Yb imply that the inclusion experienced an intense heating event that allowed these elements to evaporate.

Zoom In Zoom Out Reset image size

On the other hand, the REE abundance patterns of the leachates NCL#1-1 and NCL#2-1 in Figure 5(a) show relatively flat and negative anomalies of Eu, but no negative anomalies of Yb. The results of the REE contents from two leaching experiments show the heterogeneous distribution of at least two types of oldhamite as an REE carrier. Oldhamite is considered to be analogous to hibonite and perovskite, the two most refractory Ca-bearing phases (Kornacki & Fegley 1986). Therefore, the leachates NCL#1-2 and NCL#2-2 might have predominantly included oldhamite of nebular origin, while the other leachates NCL#1-1 and NCL#2-1 might have predominantly included oldhamite of igneous origin.

Previous REE analyses of oldhamites in aubrites indicate that some grains formed by igneous processes on the parent body, and that others are relict grains that formed during nebular condensation (Floss et al. 1990; Floss & Crozaz 1993; Lodders et al. 1993; Wheelock et al. 1994). The results from our leaching experiments are consistent with the results of previous studies. Besides the concept of the nebula origin of oldhamite, the REE patterns of the leachates NCL #1-2 and NCL#2-2 can be simply explained from the mixing model between early condensates like Group-III hibonite and typical oldhamite without Eu and Yb depletions. A late igneous process might have produced the REE patterns of the mixture having Eu and Yb depletions. REE data also support the possible existence of early condensates from the solar nebula in the Norton County aubrite.