Iv

Dr. David S. McKay Dr. Everett K. Gibson

NANOSIMS OPENS A NEW WINDOW FOR DECIPHERING ORGANIC MATTER IN TERRESTRIAL AND EXTRATERRESTRIAL SAMPLES*

DOROTHY Z. OEHLER1, FRANÇOIS ROBERT2, SMAIL MOSTEFAOUI2, ANDERS MEIBOM2, MADELEINE SELO2, DAVID S. MCKAY1 AND EVERETT K. GIBSON1

'Astromaterials Research and Exploration Science Directorate, NASA-Johnson Space Center, Houston, Texas 2Laboratoire d'Etude de la Matière Extraterrestre, Muséum National d'Histoire Naturelle, Paris, France

1. Introduction

Recognition of the earliest morphological or chemical evidence of terrestrial life has proved to be challenging, as organic matter in ancient rocks is commonly fragmentary and difficult to distinguish from abiotically-produced materials (Schopf, 1993; Van Zuilen et al., 2002; Altermann and Kazmierczak, 2003; Cady et al., 2003; Brasier et al., 2002, 2004, 2005; Hofmann, 2004; Skrzypczak et al., 2004, 2005). Yet, the ability to identify remnants of earliest life is critical to our understanding of the timing of life's origin on earth, the nature of earliest terrestrial life, and recognition of potential remnants of microbial life that might occur in extraterrestrial materials.

The search for earliest life on Earth now extends to early Archean organic remains; these tend to be very poorly preserved and considerably more difficult to interpret than the delicately permineralized microfossils known from many Proterozoic deposits. Thus, recent efforts have been directed toward finding biosignatures that can help distinguish fragmentary remnants of ancient microbes from either pseudofossils or abiotic organic materials that may have formed hydrothermally or in extraterrestrial processes (House et al., 2000; Boyce et al., 2001; Kudryavtsev et al., 2001; Schopf, 2002; Schopf et al., 2002, 2005a, b; Cady et al., 2003; Garcia-Ruiz et al., 2003; Hofmann, 2004; Brasier et al., 2005; Rushdi and Simoneit, 2005; Skrzypczak et al., 2005).

An exciting area of biosignature research involves the developing technology of NanoSIMS. NanoSIMS is secondary ion mass spectrometry (SIMS) for

* Adapted from Oehler, D.Z., Robert, F., Mostefaoui, S., Meibom, A., Selo, M. and McKay, D.S. (2006), Chemical Mapping of Proterozoic Organic Matter at Submicron Spatial Resolution, Astrobiology 6(6): 838-850. Permission to reprint portions of that article has been granted by Mary Ann Liebert, Inc., publisher of Astrobiology.

ultrafine feature, elemental and isotopic analysis. Its resolution approaches 0.05 |im for element mapping, which is 10-50 times finer than that attainable with conventional SIMS or electron microprobes. Consequently, NanoSIMS has the potential to reveal previously unknown, chemical and structural characteristics of organic matter preserved in geologic materials.

Robert et al. (2005) were the first to combine NanoSIMS element maps with optical microscopic imagery in an effort to develop a new method for assessing biogenicity. They showed that the ability to simultaneously map the distribution of "organic" elements [such as carbon (C), nitrogen (N), and sulfur (S)] and compare those element distributions with optically recognizable, cellularly preserved fossils could provide significant new insights into the origin of organic materials in ancient sediments.

This chapter details a recent NanoSIMS study which was designed to acquire new data relevant to establishing critical biosignatures (Oehler et al., 2006a-c). In this study, NanoSIMS was used to characterize element distributions of spheroidal and filamentous microfossils and associated organic laminae in chert from the ~0.85 billion year old (Ga) Bitter Springs Formation of Australia. Previous work established preservation of a diverse microbiota in the Bitter Springs Formation (Schopf, 1968; Schopf and Blacic, 1971), and there is no dispute within the scientific community regarding the biogenicity of any of the Bitter Springs structures evaluated in this new study. Thus, the NanoSIMS results described below provide new insight into - and can be used as a guide for assessing - the origin of less well understood organic materials that may occur in early Archean samples and in meteorites or other extraterrestrial samples.

2. Materials and Methods

Analysis was performed on structures within a polished thin section, ~30 |im thick, of chert, collected by D.Z. Oehler from the Ellery Creek locality of the Bitter Springs Formation. For the current study, spheroidal (cf. Myxococcoides) and filamentous (cf. Eomycetopsis) microfossils as well as organic laminae were located within the section using optical microscopy (Fig. 1). Specimens were selected for NanoSIMS based on the quality of preservation and occurrence at the top surface of the thin section. The specimens were photographed using an Olympus BX60 Research and Polarizing Optical Microscope, outfitted with a Nikon DXM 1200F Digital Camera; 4x, 10x, and 40x dry objectives were used, in both transmitted and reflected light, and sketch maps were constructed for use with the photographs for locating the structures of interest in the NanoSIMS instrument. Photomicrographic focal series were taken (using the same optical microscope) from the top of the thin section to the base of the structures of interest in transmitted light, using a 100x oil immersion lens and Cargill type DF immersion oil (Formula Code 1261). Focal planes imaged spanned about 20 |im of the thin

Figure 1. Optical photomicrographs in transmitted light of organic spheroids, filaments, and laminae in a polished thin section of chert from the -0.85 Ga Bitter Springs Formation. (A-E) illustrate the spatial relationships among the different types of structures discussed in this paper. The small rectangles in (A) identify fields of view in (B) and (D), the small rectangle in (B) identifies the field of view in (C), and the small rectangle in (D) identifies the field of view in (E).

Figure 1. Optical photomicrographs in transmitted light of organic spheroids, filaments, and laminae in a polished thin section of chert from the -0.85 Ga Bitter Springs Formation. (A-E) illustrate the spatial relationships among the different types of structures discussed in this paper. The small rectangles in (A) identify fields of view in (B) and (D), the small rectangle in (B) identifies the field of view in (C), and the small rectangle in (D) identifies the field of view in (E).

section, and 8 to 12 images typically were taken, so that step size between individual images of a focal series was 1-3 |im. The thin section was subsequently cleaned to remove all oil and any contamination from fingerprints by ultrasonica-tion five times with reagent-grade ethanol for 2 min, each time. After the sample was cleaned with ethanol, it was dried in a 60°C oven for 1 h to drive off all solvents and finally coated with about 300 À of gold.

It was assumed that the ultrasonication procedure was adequate to remove any traces of immersion oil or fingerprints on the slide. This assumption seems reasonable in view of the facts that (1) the NanoSIMS carbon maps are mirrored by both N and S maps (which is suggestive of sedimentary organic matter rather than the immersion oil, which is composed of hydrocarbons and chlorinated hydrocarbons), (2) there is a one-to-one correspondence of the C, S, and N maps with organic, kerogenous structures seen optically, and (3) the initial sputtering by NanoSIMS removes the most surficial layer, so any surface contamination is removed before data are collected. In addition, none of the kerogenous structures imaged resides in any sort of a crack in the thin section, where traces of immersion oil conceivably could remain. Finally, it should be noted that the thin section studied was not embedded with epoxy during the thin sectioning process, as the sample was relatively dense, indurated and unfractured, so that none of the NanoSIMS data could be interpreted as originating from either epoxy or some combination of epoxy and immersion oil.

Chemical maps were produced with the Cameca NanoSIMS 50 of the Muséum National d'Histoire Naturelle in Paris, France. Using a focused primary beam of Cs+, secondary ions were sputtered from the sample surface. 12C-, 12C14N-, 32S-, 28Si-, and 16O- or 18O- were detected simultaneously (multicollection-mode) in electron-multipliers at a mass-resolving power of ~4,500 (MUM). At this mass-resolving power, the measured secondary ions were resolved from potential interference by other ions or molecules that fall close in mass to the ions of interest. Because nitrogen is detected as CN- in NanoSIMS instruments, it can only be detected in the presence of carbon. Images were obtained from a presputtered surface area by stepping the primary beam across the sample surface. Presputtering is done to remove the conductive coating and clean the surface of any contaminants before analysis. The primary beam was focused to a spot size of ~50-100 nm, and the step size was adjusted so that it was comparable to, but slightly smaller than, the size of the primary beam. An electron gun supplied electrons to the sputtered surface during analysis to compensate for positive charge deposition from the primary beam and to minimize specimen charging effects. Follow-up scanning electron microscopy was performed on the Jeol JSM-5910LV (at 15 kV, 10 mm W.D.) at Johnson Space Center, Houston, TX.

N/C atomic ratios were obtained from measured 12C14N- and 12C- yields by normalization to a kerogen standard that we prepared from a sample of the Eocene Green River Shale. This kerogen, which was extracted from the shale by standard HF-HCl techniques (Beaumont and Robert, 1999), comprised > 94% of the acid-insoluble residue; standard chemical techniques (Beaumont and Robert, 1999) were used to determine that the kerogen has an atomic N/C ratio of 0.025. The 12C14N-/12C- ratio of the standard was then measured in the NanoSIMS using operating conditions identical to those used for analyzing the Bitter Springs fossils (e.g., same presputtering, spot size, e-gun, etc.).

Figure 1 illustrates the different types of organic structures analyzed in the thin section and their spatial relationships to one another. The spheroids studied are fairly abundant and occur in clusters of a few to ~25 cells, most commonly between dark brown organic laminae; the cells are typically less than 10 |im in diameter and have distinct reticulate walls, 0.3-0.5 |im thick. The filaments consist of sinuous hollow tubes, are also abundant in the thin section, and occur intertwined in mat-like layers that grade into the dark organic laminae; the filaments are 3-5 |im in diameter, up to hundreds of microns long, and have somewhat diffuse granular walls, 0.4-0.7 |im thick. The organic laminae are planar features composed of morphologically indistinct organic material, as seen in optical microscopy. In thin section, they appear as strand-like fragments of organic matter that align to form parallel, wavy to crenulate surfaces. The material making up the laminae varies from morphologically diffuse and semitransparent to more distinct-bordered and dark brown in color. The laminae occur at intervals of a fraction of a millimeter to a few millimeters, and they have thicknesses from about 5 to 20 |im.

3. Results

NanoSIMS maps of C, N, S, Si, and O (measured as 12C-, 12C14N-, 32S-, 28Si-, and 16O- or 18O-) were acquired of the spheroidal and filamentous organic microfossils and the apparently amorphous organic laminae from a single thin section of the Bitter Springs Formation (Figs. 2-7). Results demonstrate an excellent correspondence between the optical images of the microfossils and the spatial (two-dimensional) distributions of C-, CN-, and S- (Figs. 2-4). Intense sputtering into one sample allowed penetration by the NanoSIMS to a focal plane, 2-3 | m

\ V

\)V/l2c14N

28Si

BOSSES . ■ ■■

Figure 2. Spheroidal organic microfossils in a polished thin section of chert from the Bitter Springs Formation. (A) Optical photomicrograph in transmitted light; the cells of this panel are part of the cluster illustrated in Fig. 1A and E, but at a different focal plane and rotated so that the optical and NanoSIMS images can be directly compared. (B-F) NanoSIMS element maps of the same area as in (A). Arrows show corresponding cells in the different panels. Scale in (A) applies to all. 12C, carbon; 12C14N, nitrogen measured as CN- ion; 32S, sulfur; 28Si, silicon; 18O, oxygen. Color scheme is red for carbon, blue for nitrogen, green for sulfur, pink for silicon, and lavender for oxygen. The more intense the color, the stronger the response, with white being the strongest response in all cases.

Figure 2. Spheroidal organic microfossils in a polished thin section of chert from the Bitter Springs Formation. (A) Optical photomicrograph in transmitted light; the cells of this panel are part of the cluster illustrated in Fig. 1A and E, but at a different focal plane and rotated so that the optical and NanoSIMS images can be directly compared. (B-F) NanoSIMS element maps of the same area as in (A). Arrows show corresponding cells in the different panels. Scale in (A) applies to all. 12C, carbon; 12C14N, nitrogen measured as CN- ion; 32S, sulfur; 28Si, silicon; 18O, oxygen. Color scheme is red for carbon, blue for nitrogen, green for sulfur, pink for silicon, and lavender for oxygen. The more intense the color, the stronger the response, with white being the strongest response in all cases.

Figure 3. Spheroidal organic microfossils in a polished thin section of chert from the Bitter Springs Formation. (A) Optical photomicrograph in transmitted light, focused at a plane slightly below that of Fig. 2A. Dotted oval indicates area sputtered in the NanoSIMS. (B) NanoSIMS map, illustrating carbon image after sputtering into the thin section to a focal plane similar to that illustrated in (A). Nitrogen and sulfur maps (not shown) were similar to the carbon map. Arrows show corresponding cells in (A) and (B) and in Fig. 2. 12C, carbon. Color scheme is red for carbon. The more intense the color, the stronger the response, with white being the strongest response in all cases.

Figure 3. Spheroidal organic microfossils in a polished thin section of chert from the Bitter Springs Formation. (A) Optical photomicrograph in transmitted light, focused at a plane slightly below that of Fig. 2A. Dotted oval indicates area sputtered in the NanoSIMS. (B) NanoSIMS map, illustrating carbon image after sputtering into the thin section to a focal plane similar to that illustrated in (A). Nitrogen and sulfur maps (not shown) were similar to the carbon map. Arrows show corresponding cells in (A) and (B) and in Fig. 2. 12C, carbon. Color scheme is red for carbon. The more intense the color, the stronger the response, with white being the strongest response in all cases.

Figure 4. Filamentous microfossils in a polished thin section of chert from the Bitter Springs Formation. (A-C) Optical photomicrographs in transmitted light; (B) and (C) are at lower magnifications to illustrate the tube-like morphology and intertwined habit of these fossils. These filaments are part of the mass of filaments illustrated in Fig. 1A-C but at a different focal plane and rotated slightly so that the optical and NanoSIMS images can be compared. (C) is about 20 |m below the focal plane of (A) (but in the exact same locality of the thin section), and a focal series of 11 photomicrographs, each taken successively 1-3 |im lower in the section, demonstrates that these filaments form an entangled mass throughout the entire 20 |im that was imaged. (D-F) NanoSIMS element maps. Arrows show corresponding cells in the different panels. 12C, carbon; 12C14N, nitrogen measured as CN- ion; 32S, sulfur. Color scheme is red for carbon, blue for nitrogen, and green for sulfur. The more intense the color, the stronger the response, with white being the strongest response in all cases.

Figure 4. Filamentous microfossils in a polished thin section of chert from the Bitter Springs Formation. (A-C) Optical photomicrographs in transmitted light; (B) and (C) are at lower magnifications to illustrate the tube-like morphology and intertwined habit of these fossils. These filaments are part of the mass of filaments illustrated in Fig. 1A-C but at a different focal plane and rotated slightly so that the optical and NanoSIMS images can be compared. (C) is about 20 |m below the focal plane of (A) (but in the exact same locality of the thin section), and a focal series of 11 photomicrographs, each taken successively 1-3 |im lower in the section, demonstrates that these filaments form an entangled mass throughout the entire 20 |im that was imaged. (D-F) NanoSIMS element maps. Arrows show corresponding cells in the different panels. 12C, carbon; 12C14N, nitrogen measured as CN- ion; 32S, sulfur. Color scheme is red for carbon, blue for nitrogen, and green for sulfur. The more intense the color, the stronger the response, with white being the strongest response in all cases.

Figure 5. NanoSIMS images of a wall contact between two spheroidal microfossils in chert from the Bitter Springs Formation. (A and B) Relatively low magnification element maps. (C-F) Highresolution element maps. White rectangle in (A) shows area of high-resolution images in (C-F). Arrows in (E and F) tie locations of the silicon globules in (F) with corresponding locations on the carbon map in (E). In (F), the diffuse area of Si response in the central portions of the spheroids is likely due to silica in the host chert. Dotted white ovals in (E and F) are reference areas to tie the two images for comparison. 12C, carbon; 12C14N, nitrogen measured as CN- ion; 32S, sulfur; 28Si, silicon. Color scheme is red for carbon, blue for nitrogen, green for sulfur, and pink for silicon. The more intense the color, the stronger the response, with white being the strongest response in all cases.

Figure 5. NanoSIMS images of a wall contact between two spheroidal microfossils in chert from the Bitter Springs Formation. (A and B) Relatively low magnification element maps. (C-F) Highresolution element maps. White rectangle in (A) shows area of high-resolution images in (C-F). Arrows in (E and F) tie locations of the silicon globules in (F) with corresponding locations on the carbon map in (E). In (F), the diffuse area of Si response in the central portions of the spheroids is likely due to silica in the host chert. Dotted white ovals in (E and F) are reference areas to tie the two images for comparison. 12C, carbon; 12C14N, nitrogen measured as CN- ion; 32S, sulfur; 28Si, silicon. Color scheme is red for carbon, blue for nitrogen, green for sulfur, and pink for silicon. The more intense the color, the stronger the response, with white being the strongest response in all cases.

deeper in the section. A comparison of the optical image at a similar focal plane with the C-, CN-, and S- maps at that lower plane in the NanoSIMS also demonstrates a three-dimensional correspondence between results of NanoSIMS and optical microscopy (Fig. 3). Importantly, the host chert matrix is essentially lacking in significant C-, CN-, and S-, and these ions are present only in structures identified as microfossils using optical microscopy (Fig. 2).

Ultra-high-resolution images (obtained with the smallest possible primary beam spot, around 50 nm in diameter, which was rastered over a small area, typically 10 x 10 |im2, in order to collect sufficient secondary ions from each pixel) show that the C-, CN-, and S- distributions are identical to one another for both the spheroidal and the filamentous microfossils (Figs. 5 and 6). The spheroidal microfossils are defined by wall-like structures that consist of distinct globules enriched in C-, CN-, and S- (Fig. 5). In contrast, the filamentous microfossils appear to consist of more diffuse, irregular, and "less packaged" material enriched in C-, CN-, and S- (Fig. 6). These observations are likely to reflect differences in the structures of the biological precursors of the two types of microfossils: the spheroidal microfossils comprising

Figure 6. Filamentous microfossils in a polished thin section of chert from the Bitter Springs Formation. (A) Optical photomicrograph in transmitted light. (B-F) High-resolution NanoSIMS element maps. Black rectangle in (A) shows area of detail in (B-F). Scale in (B) applies to (B-F). 12C, carbon; 12C14N, nitrogen measured as CN- ion; 32S, sulfur; 28Si, silicon; 16O, oxygen. Color scheme is red for carbon, blue for nitrogen, green for sulfur, pink for silicon, and lavender for oxygen. The more intense the color, the stronger the response, with white being the strongest response in all cases.

Figure 6. Filamentous microfossils in a polished thin section of chert from the Bitter Springs Formation. (A) Optical photomicrograph in transmitted light. (B-F) High-resolution NanoSIMS element maps. Black rectangle in (A) shows area of detail in (B-F). Scale in (B) applies to (B-F). 12C, carbon; 12C14N, nitrogen measured as CN- ion; 32S, sulfur; 28Si, silicon; 16O, oxygen. Color scheme is red for carbon, blue for nitrogen, green for sulfur, pink for silicon, and lavender for oxygen. The more intense the color, the stronger the response, with white being the strongest response in all cases.

remnants of actual cell walls and the filamentous forms probably representing remnants of extracellular mucilaginous sheaths common to filamentous cyanobacteria.

The Si- and O- maps also reflect the morphology of the microfossils (Fig. 2), even though some Si- and O- yields are detected from the host chert (SiO2) as well (Fig. 5F). However, in the higher-resolution images of Figs. 5 and 6, the Si- and O- yields display differences in detail from the distributions of C-, CN-, and S-. In the spheroids, the Si- distribution shows a more open texture than is apparent in the C- map (cf Fig. 5A and B), and in the highest resolution (Fig. 5E and F), the globules of Si- alternate spatially with globules of C-. In the filaments, Si- and O- distributions appear to be thicker and more continuous than the simultaneously collected C-, CN-, or S- ions (cf. Fig. 6B-D with Fig. 6E and F).

The NanoSIMS elemental maps of the organic laminae exhibit relationships among C-, CN-, S-, Si-, and O- similar to those observed in the spheroids and filaments, and the images show densely packed structures reminiscent of the filamentous microfossils and collapsed spheroids (Fig. 7).

CN-/C- ratios of the spheroids, filaments, and laminae were measured in multiple localities on the NanoSIMS maps. Results show major differences in both absolute values and ranges (Table 1).

Figure 7. Organic lamina in a polished thin section of chert from the Bitter Springs Formation. (A) Optical photomicrograph in transmitted light; the location of this lamina and its spatial relationship to other structures discussed is illustrated in Fig. 1A and D. (B-F) NanoSIMS element maps of the area in (A). Arrows show reference points for comparison. The white ovals show the same region in (A-F). Scale in (A) applies to all. Sputtered areas are particularly evident in (E and F) as the large circular regions that extend well beyond the lamina. 12C, carbon; 12C14N, nitrogen measured as CN-ion; 32S, sulfur; 28Si, silicon; 18O, oxygen. Color scheme is red for carbon, blue for nitrogen, green for sulfur, pink for silicon, and lavender for oxygen. The more intense the color, the stronger the response, with white being the strongest response in all cases.

Figure 7. Organic lamina in a polished thin section of chert from the Bitter Springs Formation. (A) Optical photomicrograph in transmitted light; the location of this lamina and its spatial relationship to other structures discussed is illustrated in Fig. 1A and D. (B-F) NanoSIMS element maps of the area in (A). Arrows show reference points for comparison. The white ovals show the same region in (A-F). Scale in (A) applies to all. Sputtered areas are particularly evident in (E and F) as the large circular regions that extend well beyond the lamina. 12C, carbon; 12C14N, nitrogen measured as CN-ion; 32S, sulfur; 28Si, silicon; 18O, oxygen. Color scheme is red for carbon, blue for nitrogen, green for sulfur, pink for silicon, and lavender for oxygen. The more intense the color, the stronger the response, with white being the strongest response in all cases.

Table 1. Nitrogen to Carbon Ratios.

Sample

Measured CN-/C-

Was this article helpful?

0 0

Post a comment