We evaluated the Kochi grid for FIB, TEM, and NanoSIMS, and the Okazaki cell for STXM-NEXAFS through the coordinated analysis of primitive extraterrestrial materials containing minerals and organics. The detailed discussion for minerals are located in Sect. 3.2 while organics are discussed in Sect. 3.3
In-situ analysis of mineral phases
The AMM, TT006b101, has a spherical shape of approximately 200 µm in size (approximately 13 µg in weight) (Fig. 4a) and is pressed onto a Gecko Tape. An ultra-thin section of the sample (10 × 8 × 0.1 µm3) was prepared and attached to the Kochi grid by the FIB (SMI-4050, Hitachi High-Tech Corp., Minato-ku, Japan) at the Kochi Institute of Core Sample Research, JAMSTEC (Fig. 4b).
We examined the detailed major elemental abundances, mineralogy, and microstructure to gain insight into its petrogenesis by TEM (JEM-ARM200F equipped with EDS, JEOL Ltd., Tokyo) followed by FIB (SMI-4050) to prepare an ultra-thin section. Based on the elemental and electron-diffraction analyses of the individual grains, the AMM was confirmed to consist of olivine [(Mg,Fe)2SiO4], magnetite (Fe3O4), and interstitial Ca-Mg-Fe-Al-bearing amorphous silicate (Fig. 4c), where olivine and magnetite occur as euhedral to subhedral grains of several micrometers in size. The petrography suggests that the precursor material of the AMM was extensively heated to be completely melted and then was partially crystallized by rapid cooling. When the hydrated carbonaceous chondrites containing abundant phyllosilicates experienced extensive heating, a mineral assemblage of olivine, magnetite, and SiO2-rich amorphous material was formed (Toppani et al. 2001). Note that the phyllosilicates in the precursor chondritic material would have also been affected by heating and dehydration processes during its atmospheric entry.
Next, we applied rastered ion imaging by the Japan Agency for Marine-Earth Science Technology (JAMSTEC) NanoSIMS 50L ion microprobe (Ametek CAMECA, Inc., Gennevilliers Cedex, France) to acquire an isotope map of oxygen (18O/16O ratio) as well as elemental maps of Si and Mg as 24Mg16O, Al as 27Al16O, Ca as 40Ca16O, and Fe as 56Fe16O for the sample (Fig. 4d). The detailed measured conditions and calculation of δ18OSMOW were published in a previous work (Ito and Messenger 2008). The elemental ratio maps (Fig. 4d) show the constituent mineralogical features of olivine, magnetite, and a Ca-Mg-Fe-Al-bearing amorphous silicate in the section analyzed by the TEM-EDS elemental and crystallographic analyses (Fig. 4c). The obtained δ18OSMOW isotopic composition of the sample’s mineral phases shows a homogeneous distribution of 12.7 ± 2.2 per mil (Fig. 4d). We did not find a clear difference in the δ18O of each phase within the analytical uncertainties. This δ18O value is broadly consistent with previous O isotopic compositions for various AMMs, which suggests that heavy O isotopic enrichment was caused by atmospheric entry heating or thermal metamorphism in the parent body (Matrajt et al. 2006; Engrand and Dobrica 2012).
In-situ analysis of organic matter
A systematic investigation of the nanoglobules in Yamato, (Y)-791198, which is composed of unheated CM2.4 chondrites (Nakamura 2005; Rubin et al. 2007), was carried out utilizing FIB, STXM-NEXAFS, NanoSIMS, and TEM analyses. The universal sample holders of the Kochi grid for FIB, TEM, NanoSIMS, and the Okazaki cell for STXM-NEXAFS were used.
We prepared an ultra-thin section (30 × 30 × 0.1 µm3) of the Y-791198 matrix using the FIB at the Kochi Institute for Core Sample Research, JAMSTEC. The NEXAFS spectra of C K-edge of the section and the nanometer-scale carbon distribution were measured by an STXM at the UVSOR BL4U (Fig. 5a). C and N as 12C14N elemental images of the same section were obtained by the JAMSTEC NanoSIMS (Figs. 5b-c) as well as the H, C, and N isotope maps (details of the analytical conditions are located in Ito et al. 2014). Subsequently, a TEM-EDS analysis was performed to obtain carbon X-ray maps and ultra-high magnification images of each carbon enriched region (Fig. 5d).
Four nanoglobule candidates, G1 – G4, were found in the ultra-thin section by combining STXM C K-edge spectral image (Fig. 5a) and NanoSIMS 12C and 12C14N elemental images (Figs. 5b-c). The STXM C K-edge spectral image was generated by an accumulation of all the spectral peak intensities in each pixel after a baseline collection at 280 eV. The C-rich regions defined by the STXM C K-edge spectral image are broadly consistent with those of the NanoSIMS 12C and 12C14N elemental images (Figs. 5a-c). Table 1 summarizes the results obtained by the TEM-EDS (size) and NanoSIMS (H, C, and N isotopic compositions) analyses.
The representative atomic number contrast images (high-angle annular dark field in scanning-TEM mode (HAADF-STEM)) and carbon X-ray images of nanoglobules G1 and G3 are shown in Fig. 5. The size of the grains ranges from 650 nm to 1000 nm in diameter (Table 1). The grains are mainly composed of carbon, and their size and shape are similar to those of previous reports (e.g., Nakamura-Messenger et al. 2006; De Gregorio et al. 2013).
The STXM C K-edge spectra of G1 to G4 show peak intensities at 285 eV (aromatic or olefinic carbon), 286.5 eV (oxygen substituted double-bonded carbon, e.g., enolic carbon), 288.4 eV (carbonyl carbon in amide moieties), and 290.2 eV (carbonate CO3) (Fig. 5f). These peaks exist at a slightly lower energy, approximately 0.3 eV, in comparison with the peaks found by Vinogradoff et al. (2018) due to the surrounding organics that have a wide variety of molecular configurations (e.g., De Gregorio et al. 2013).
We found two types of nanoglobules as a result of the C K-edge spectra that suggest G1 is a ketone-rich nanoglobule, and G2 to G4 are aromatic nanoglobules even though they showed no clear size and morphology difference from each other. Similar features of nanoglobules were reported in De Gregorio et al. (2013). As Flynn et al. (2010) indicated that organic matters in-situ in the Murchison and Orgueil chondrites differ in their C K-edge spectra from samples of IOM acid extraction from those same chondrites, the similarity found in this study could be fortuitous. However, it may simply be a result of variation in the carbon functional group in each nanoglobule.
The C (δ13CPDB) and N (δ15NAir) isotopic compositions in the nanoglobules (G1 to G4) in Y-791198 showed a large variation (Table 1) that is consistent with nanoglobules in the Bells (CM2) and Murchison (CM2) chondrites (De Gregorio et al. 2013). No isotopic “hot-spots” with highly enriched 15N (δ15NAir) were observed in the globules. G1 shows as negative δ15NAir of approximately − 300 per mil and low degree and negative δ15N were observed in the organic matter of the QUE99117 (CR3) and MET 00426 (CR3) chondrites (Floss et al. 2014). These N isotopic characteristics are expected to occur in ion–molecule reactions (Wirström et al. 2012). The H isotopic compositions of G1, G2, and G4 showed high D-enrichment, implying they have interstellar origins, while G3 had only a moderate D-enrichment (δDSMOW = approximately + 200 per mil). The H isotopic variation found in the nanoglobules was not caused by analytical artifact as we carefully chose an analytical sequence (NanoSIMS isotope map followed by TEM observation) to avoid H isotopic fractionation by electron beam irradiation.
We confirmed that the developed coordinated analytical system of the TEM, NanoSIMS, and STXM-NEXAFS provided the same quality analytical dataset as that of previous works by each instrument.
Coordinated Analysis For The Ryugu Asteroidal Sample
As shown in Fig. 6, we established a coordinated analytical sequence for the future analysis of the Ryugu asteroidal samples from non-destructive analyses at synchrotron radiation facilities, such as 3D-CT (computed tomography), XRD (X-ray diffraction), and STXM-NEXAFS, to destructive analyses, such as TEM, SIMS, and laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS). This coordinated sequence has the potential to obtain the complex characteristics inside a sample. Regions-of-interest (ROIs) inside the sample can be found through synchrotron-based 3D-CT and XRD analyses. Prior to conducting a series of microanalysis, we used an FIB to extract the ROIs based on a 3D characterization of the sample (Uesugi et al. 2014b).
The coordinate analysis proposed in this study, which involves utilizing instruments, devices, and methods, has been further developed by related works (Kodama et al. 2020; Shirai et al. 2020; Uesugi et al. 2020). Kodama et al. (2020) described the development of a surface treatment technique by FIB for obtaining high-quality electron back-scattered diffraction (EBSD) patterns from minerals in AMMs. Uesugi et al. (2020) developed a vertically aligned carbon nanotubes (VACNT) holder for synchrotron-based CT and XRD analyses. Shirai et al. (2020) stated that the elemental abundances of VACNT, polyimide film, and synthetic quartz glass will be used for the analysis of the Ryugu samples to evaluate possible contaminations during the sample handling process as they concluded that these materials showed low levels of contaminants and are therefore adequate for use as sample holders for the Ryugu samples.
An in-house non-air exposing sample loading system that utilized the Okazaki cell for sample transfer between an STXM and a glove box under N2 or Ar conditions is used for analyzing anaerobic materials at the UVSOR BL4U. A non-air exposing system that includes a glove box is also available for synchrotron-based CT and XRD at SPring-8 (Fig. 7). Note that an in-house non-air exposing sample holder for NanoSIMS is currently under development, and will be ready for the analysis of the Hayabusa2 returned samples. Commercial non-air exposing sample holder systems are available for TEM and FIB. We have not yet installed these systems, though they will be installed before the analysis. We plan to use the developed holders (the Kochi grid and clamp), FFTC, Okazaki cell, and coordinate analysis under non-air exposing systems between institutes for extraterrestrial samples (i.e., the asteroid Ryugu) (Fig. 7).