REDISTRIBUTION OF ALKALINE ELEMENTS IN ASSOCIATION WITH AQUEOUS ACTIVITY IN THE EARLY SOLAR SYSTEM

In our previous study, 35 chondrules with diameters from 100 to 600 μm were handpicked from the matrix portion of the Sayama meteorite. Although most of them were consumed in the previous isotopic study with complete decomposition and chemical treatments (Hidaka & Yoneda 2013), two remained, and were used in this study. The polished section was prepared from these two chondrules.

Prior to the identification of serpentine phases by Raman spectrometry, elemental mapping of Mg, Fe, Si, O, Al, Ca, and K of the samples was performed with a SEM-EDX (JEOL JSM-6390A). The beam size was 1 μm and the current was 1 nA at 15 kV of acceleration voltage. The data were treated with a standardless ZAF correction method for semi-quantitative electron probe micro-analyses of microscopical particles (e.g., Van Borm & Adams 1991).

Raman spectroscopy was completed on a Renishaw inVia Raman Reflex microscope equipped with a Leica DMLM microscope. The spectra were excited with a 532 nm LD laser and the Raman spectra were obtained on the thin section. The instrument had Streamline capabilities for rapid mapping 200 times faster than the traditional point-by-point mapping. It was used to collect spatially resolved chemical images of serpentinized phases transformed from olivine in chondrules on the thin section.

Determination of the elemental concentrations of Rb, Sr, Cs and Ba was performed with SHRIMP at Hiroshima University. The samples were sputtered with a 5 nA ${{{\rm{O}}}_{2}}^{-}$ primary ion beam. The mass resolution (M/ΔM at 1% of peak height) was set at 9000 to resolve the oxide ion species (MO⁺) from the mass region in this study. The masses of ⁸⁶Rb, ⁸⁷Rb+⁸⁷Sr, ⁸⁸Sr, ¹²⁰(²⁸Si₂¹⁶O₄), ¹³³Cs, ¹³⁵Ba, ¹³⁷Ba, ¹³⁸Ba, ¹³⁹La, and ¹⁴⁰Ce and backgrounds (at masses of 85.5 and 140.5) were monitored. Standard glass SRM 612 obtained commercially from NIST was used for the calibration of secondary ion ratios to the elemental concentrations. The count rate of each element, ${{C}_{{\rm{i}}}}^{+}$, was converted to the elemental concentration [C_i] by the following equation:

$$\begin{eqnarray*}&&\left[{C}_{{\rm{i}}}\right]={C}_{{\rm{i}}}^{+}\times {\alpha }_{{\rm{i}}}\times \displaystyle \frac{\left[{\mathrm{SiO}}_{2}\right]}{{\mathrm{Si}}_{2}{{\rm{O}}}_{4}^{+}}\end{eqnarray*}$$

where ${\alpha }_{{\rm{i}}}$ is a conversion factor for each element (i = Rb, Sr, Cs and Ba) obtained from the SHRIMP analysis of the reference material SRM612, and [SiO₂] is the concentration of SiO₂ at each analytical point measured by EPMA analysis.

The Raman spectra of serpentine phases were characterized by the peaks in the high wavenumber range at around 3550–3850 cm⁻¹ associated with OH stretching vibrations, and in lower wavenumber range at 230–240 cm⁻¹, 390 cm⁻¹, 680–690 cm⁻¹ and 1045–1105 cm⁻¹ associated with O–H–O vibrations, symmetric SiO₄, symmetric Si–O_b–Si stretching vibrations, and antisymmetric Si–O_nb stretching vibrations, respectively (Rinaudo & Gastaldi 2003; Petriglieri et al. 2015). Because serpentine is a phase composed of hydrous minerals, the spectra of O–H–O vibrations can be used as a sensitive and powerful probe to identify the serpentine minerals. In this study, two-dimensional (2D) Raman spectra mapping from the O–H–O vibrations was used for identification of the serpentine minerals in the chondrule grains. Figure 1(a) shows a typical Raman spectra of serpentine obtained from a spot analysis in the Sayama chondrule. Four peaks at wavenumber 230, 390, 690, and 3700 cm⁻¹ corresponding to the typical peaks to identify the serpentine were clearly observed in the spectrum. Figure 1(b) shows the 2D Raman map of a part of the chondrule of the Sayama meteorite based on the above four peaks. During the Raman spectroscopic observations, serpentine phases were identified in 10 μm-sized square region in chondrule-2, but not in chondrule-1.

Zoom In Zoom Out Reset image size

Table 1 shows the elemental concentrations of Rb, Sr, Cs and Ba at 29 analytical points of the two chondrules. The abundances of the four elements vary widely in individual chondrule grains, and in particular those in serpentine phases (analytical spot numbers 2–1, 2–2, 2–3, and 2–4 in Table 1) are relatively higher than in other parts. This suggests selective adsorption of alkali elements into the serpentine phases in association with aqueous alteration.

Figure 2 shows the correlation between Rb and Sr (a), and between Cs and Ba (b) elemental abundances. Comparing the data obtained in this study with the CI reference data (Anders & Grevesse 1989), there is a significant chemical fractionation between Cs and Ba by the redistribution in the chondrules, but little fractionation between Rb and Sr. The Rb/Sr ratios collected from the chondrules range from 0.045 to 0.69, and most of the data points show a similar trend with the CI value (Rb/Sr = 0.295). On the other hand, the Cs/Ba ratios varied widely between 0.047 and 1.11, and most are higher than the CI value (Cs/Ba = 0.0799). The result shows the heterogeneous distributions of the alkaline elements in the chondrules. Interestingly, four analytical points from serpentine phases show higher contents of Rb, Sr, Cs, and Ba than other points. The higher Rb/Sr and Cs/Ba ratios at these four points suggest the selective adsorption of Rb and Cs in the serpentines in association with the early aqueous alteration on the parent body.

Zoom In Zoom Out Reset image size

Three Ba isotopes ¹³⁵Ba, ¹³⁷Ba and ¹³⁸Ba were selected for the in situ isotopic analysis in this study, because they have less isobaric interferences than the other Ba isotopes. ¹³⁵Ba and ¹³⁷Ba have no isobaric isotopes of other elements. Although ¹³⁸Ce and ¹³⁸La are isobaric interferences of ¹³⁸Ba, their isotopic abundances are very minor (0.25% and 0.09%, respectively). One of major concerns in this study is isotopic search for radiogenic ¹³⁵Ba decayed from ¹³⁵Cs. Although most of the Ba isotopic compositions of primitive materials in the solar system are generally affected by s- and r-process nucleosynthetic components that hide the contribution of the isotopic excess of ¹³⁵Ba formed by the decay of radioactive ¹³⁵Cs, those of the chemical separates from chondrules in the Sayama meteorite shows an excess of ¹³⁵Ba isotopic abundance up to (0.33 ± 0.06)%, which is not derived from the isotopic components from s- and r-process nucleosyntheses but possibly from the decay of ¹³⁵Cs (Hidaka & Yoneda 2013). ¹³⁷Ba/¹³⁸Ba = 0.15653 were used as a normalizing factor to make a correction of instrumental mass fractionation of ¹³⁵Ba/¹³⁸Ba ratios during the analyses. The data are shown in Table 2. The analytical results from SRM612 as a reference material provide ¹³⁵Ba/¹³⁸Ba = 0.09212 ± 0.00030 and ¹³⁷Ba/¹³⁸Ba = 0.15651 ± 0.00026 (1σ means of standard errors as analytical uncertainty). The isotopic ratios from most of the data points in the chondrules, shown in the analytical data points 1–1, 1–3, 1–4, 1–5, 1–9, 1–11, 1–12, and 2–7 in Table 2 as a typical example, are equivalent to those of SRM612 within the analytical uncertainties. Most of the other data points (1–7, 1–13, 1–14, 2–1, 2–2, 2–3, 2–4, 2–5, 2–6, and 2–8) show positive ¹³⁵Ba and ¹³⁷Ba isotopic anomalies, probably because of depletion of s-process isotopic components which are often observed in primitive materials in the solar system (Harper 1993; Hidaka et al. 2001, 2003; Andreasen & Sharma 2007; Carlson et al. 2007). On the other hand, one data point from 1 to 10 shows negative ¹³⁵Ba and ¹³⁷Ba isotopic anomalies caused by enrichment of s-process isotopes. We tried to find a spot showing isotopic excess of ¹³⁵Ba which suggests the existence of pure radiogenic ¹³⁵Ba decayed from ¹³⁵Cs, but with no success.

Because some of the Ba isotopic data set obtained from the Sayama chondrules are affected by an additional s-process isotopic component, the existence of radiogenic ¹³⁵Ba is hidden at the present state. However, because the isotopic ratio of the s-process component is given as ¹³⁵Ba/¹³⁷Ba = 2.145 by the stellar model (Arlandini et al. 1999), the radiogenic ¹³⁵Ba component can be estimated by subtracting the s-process component in the following equation.

$$\begin{eqnarray*}&&{\left(\displaystyle \frac{{}^{135}\mathrm{Ba}}{{}^{138}\mathrm{Ba}}\right)}_{\mathrm{meas}.}={\left(\displaystyle \frac{{}^{135}\mathrm{Ba}}{{}^{138}\mathrm{Ba}}\right)}_{\mathrm{initial}}+{\left(\displaystyle \frac{{}^{135}\mathrm{Ba}}{{}^{138}\mathrm{Ba}}\right)}_{s-\mathrm{process}}+{\left(\displaystyle \frac{{}^{135}\mathrm{Ba}}{{}^{138}\mathrm{Ba}}\right)}^{*}\end{eqnarray*}$$

where the subscripts of meas., initial, and s-process reveal measured values, initial values corresponding to the terrestrial standard value = 0.09212 obtained from the average of SRM612 analyses, and s-process isotopic component, respectively. The s-process isotopic component of (¹³⁵Ba/¹³⁸Ba)_s-process can be estimated from (¹³⁷Ba/¹³⁸Ba)_s-process × 2.145 based on the stellar model, and (¹³⁷Ba/¹³⁸Ba)_s-process is calculated from (¹³⁷Ba/¹³⁸Ba)_meas.-(¹³⁷Ba/¹³⁸Ba)_STD:

$$\begin{eqnarray*}{\left(\displaystyle \frac{{}^{135}\mathrm{Ba}}{{}^{138}\mathrm{Ba}}\right)}^{*} & = & {\left(\displaystyle \frac{{}^{135}\mathrm{Ba}}{{}^{138}\mathrm{Ba}}\right)}_{\mathrm{meas}.}-{\left(\displaystyle \frac{{}^{135}\mathrm{Ba}}{{}^{138}\mathrm{Ba}}\right)}_{\mathrm{initial}}\\ & & -{\left(\displaystyle \frac{{}^{137}\mathrm{Ba}}{{}^{138}\mathrm{Ba}}\right)}_{s-\mathrm{process}}\times 2.145.\end{eqnarray*}$$

Finally, (¹³⁵Ba/¹³⁸Ba)* is estimated from the measured Ba isotopic ratios of samples, (¹³⁵Ba/¹³⁸Ba)_meas. and (¹³⁷Ba/¹³⁸Ba)_meas. as follows:

$$\begin{eqnarray*}{\left(\displaystyle \frac{{}^{135}\mathrm{Ba}}{{}^{138}\mathrm{Ba}}\right)}^{*} & = & {\left(\displaystyle \frac{{}^{135}\mathrm{Ba}}{{}^{138}\mathrm{Ba}}\right)}_{\mathrm{meas}.}-0.09212\\ & & -\left\{{\left(\displaystyle \frac{{}^{137}\mathrm{Ba}}{{}^{138}\mathrm{Ba}}\right)}_{\mathrm{meas}.}-0.15651\right\}\times 2.145.\end{eqnarray*}$$

Figure 3 shows a correlation diagram between the Cs/Ba elemental abundance ratio and the calculated (¹³⁵Ba/¹³⁸Ba)* values. As the results of calculation, the contribution of the radiogenic component is not clearly observed from the analytical points of chondrule-1. On the other hand, the excess of the calculated (¹³⁵Ba/¹³⁸Ba)* are found in five of eight data points from chondrule-2. The data points shown in Figure 3 are scattered and do not construct a clear isochron line. In order to subtract s-process components from the total ¹³⁵Ba isotopic abundance, the s-process isotopic ratio of ¹³⁵Ba/¹³⁷Ba = 2.145 is assumed on the basis of the stellar model (Arlandini et al. 1999).

Zoom In Zoom Out Reset image size

In this study, the correction of instrumental mass fractionation is not made to determine the ¹³⁵Ba/¹³⁸Ba and ¹³⁷Ba/¹³⁸Ba ratios, because the Ba isotopic analyses of the reference material SRM612 did not show significant fractionation. A comparison of the isotopic results of reference materials from SHRIMP analyses with previous TIMS analyses (Andreasen & Sharma 2006; Carlson et al. 2007; Bermingham et al. 2014) shows that the data are in good agreement within the analytical uncertainties. The isotopic deviations associated with the instrumental mass fractionation are considered to be not larger than the analytical uncertainties. In this study, the isotopic ratios from some data points are equivalent to those of the standard material SRM612 within the analytical uncertainties.

Finding a reasonable correction method is also required for the estimation of radiogenic ¹³⁵Ba from the Cs–Ba isotopic analysis, because ¹³⁵Ba isotope in the early solar materials is often suffered by significant amount of additional input of s-process component. Although subtractions of other isotopic ratios of ¹³⁵Ba/¹³⁷Ba based on the stellar model from the s-process component have been tried in this study, none of them provides a better linear correlation between the calculated radiogenic ¹³⁵Ba/¹³⁸Ba ratios and the Cs/Ba elemental abundance ratios. The unclear correlation between Cs/Ba and ¹³⁵Ba isotopic excess is considered to have resulted from significant disturbance of the Cs–Ba system by aqueous alteration. Our previous study on Cs–Ba analysis of bulk chondrules from the Sayama meteorite also provides isotopic evidence for the reconstruction of the ¹³⁵Cs–¹³⁵Ba isochron after resetting by intense aqueous alteration. The Cs–Ba systematics of the bulk chondrules from the Sayama chondrite shows a modified isochron with a slope of ¹³⁵Cs/¹³³Cs = (6.8 ± 1.9) × 10⁻⁴ and an apparently high initial ${\varepsilon }^{135}$ Ba (=+12.9) value. In this study, assuming that the five data sets that were selected based on having a positive (¹³⁵Ba/¹³⁸Ba)* ratio, the data points 2–1, 2–2, 2–4, 2–6, and 2–8 in Table 2, show the radiogenic ¹³⁵Ba component, they provide a slope of (3.9 ± 9.1) × 10⁻⁴ corresponding to ¹³⁵Cs/¹³³Cs = (2.7 ± 6.3) × 10⁻⁴. However, the result shows little linearity because of large uncertainty.

Previous studies on the estimation of the initial ¹³⁵Cs/¹³³Cs ratio in the solar system provide variable ratios between 1.6 ×10⁻⁴ and 4.8 ×10⁻⁴ from several approaches, and the initial abundance of ¹³⁵Cs in the solar system is still unclear (McCulloch & Wasserburg 1978; Hidaka et al. 2001; Hidaka & Yoneda 2011, 2013). Furthermore our previous study on the bulk analyses of the Sayama chondrules suggests the possibly large redistribution of Cs in the chondrules during the aqueous alteration in the early solar system.