Determining the Biogenicity of Microfossils in the Apex Chert, Western Australia, Using Transmission Electron Microscopy

Introduction: For over a decade, the oldest evidence for life on this planet has been microfossils in the 3.5 Ga Apex Chert in Western Australia [1]. Recently, the biogenicity of these carbon-rich structures has been called into question through reanalysis of the local geology and reinterpretation of the original thin sections. Although initially described as a stratiform, bedded chert of siliceous clasts, the unit is now thought to be a brecciated hydrothermal vein chert [2]. The high temperatures of a hydrothermal environment would probably have detrimental effects to early non-hyperthermophilic life, compared to that of a shallow sea. Conversely, a hydrothermal origin would suggest that if the microfossils were valid, they might have been hyperthermophilic. Apex Chert controversy. The Apex Chert microfossils were originally described as septate filaments composed of kerogen similar in morphology to Proterozoic and modern cyanobacteria [1, 3]. However new thin section analysis shows that these carbonaceous structures are not simple filaments [2]. Many of the original microfossils are branched and have variable thickness when the plane of focus is changed. Hydrothermal alteration of organic remains has also been suggested for the creation of these strange morphologies [4]. Another point of contention lies with the nature of the carbon material in these proposed microfossils. Kerogen is structurally amorphous, but transforms into wellordered graphite under high pressures and temperatures. Raman spectrometry of the carbonaceous material in the proposed microfossils has been interpreted both as partially graphitized kerogen and amorphous graphite [2, 5]. However, these results are inconclusive, since Raman spectrometry cannot adequately discriminate between kerogen and disordered graphite [6]. There are also opposing views for the origin of the carbon in the Apex Chert. The carbon would be biogenic if the proposed microfossils are indeed the remains of former living organisms. However, an inorganic FischerTropsch-type synthesis is also a possible explanation for the formation of large-aggregate carbonaceous particles and could also account for the depletion of C observed [2]. Methods: Previous studies of the Apex Chert have focused on petrographic analysis, stable isotope analysis, and now Raman spectrometry. However there are some benefits for using the more time-intensive methods of transmission electron microscopy (TEM). There are TEM techniques for imaging and determining both crystal structure and composition, all at the nanometer scale. High-resolution images can directly show local crystal structure, and may be used to differentiate between graphite and amorphous carbon, or kerogen. The two techniques for determining chemical composition are energy dispersive x-ray spectrometry (EDS) and electron energyloss spectrometry (EELS). As electrons pass through the sample, they can interact with it either elastically or inelastically. Inelastic collisions will cause the electrons in the beam to lose some of their energy, creating an energy spectrum within the transmitted beam. Dispersing the beam with a magnetic prism produces an electron energy-loss spectrum. Ionization edges appear on the spectrum when electrons lose energy to promote bonding electrons into empty orbitals, giving information on bonding within the material. Graphite and Kerogen in TEM. Both graphite and amorphous carbon affect the high-energy loss portion of the EELS spectrum at the carbon K-edge (Fig. 1). The differences between the two spectra are due to differences in carbon-carbon bonding. Due to more ordered crystal structure and regular bonding, the spectra of both ordered and disordered graphite show a sharp π* peak followed by a wide, flat-topped σ* peak. At the front edge of the σ* peak is another sharp peak called the “A” peak [7]. The spectrum for amorphous carbon, however, only has a small π* shoulder, a sloping σ* peak, and no “A” peak [7]. These spectra are different enough to be able to distinguish the two carbon phases.