=================================================================================

Fine structures in an EEL spectrum fundamentally originate from the transition from a core level to unoccupied states under the dipole selection rule. Besides the ionization edges there is a fine structure superposed on the edge and extending up to ~ 50 eV from the EELS edge onset, reflecting the density of unoccupied states in the conduction band and known as the energy loss near edge structure (ELNES).

In principle, both the valence electron peaks and the ionization edges represent a fine structure that reflects the crystallographic or energy band structure of the specimen. The fine-structures of the ionization edges in the EEL spectra contain information that goes beyond the composition as the ELNES characteristics are determined mainly by the bonding states and thus, reflects the chemical neighborhood.

Although the fine structures of core loss spectra give information of the unoccupied density of states, the main application of EELS had still been the elemental quantification until the end of the 1980s mainly because the lack of modeling methods of the ELNES spectra. ELNES analysis has now been commonly applied since theoretical ELNES calculations become easier.

The fine structure of the edges can provide the following information:
        i) Chemical (bonding) effects:
           i.A) The bonding of ionized atoms.
           i.B) The density of states.
        ii) Structural (atomic arrangement) effects:
           ii.A) The coordination of the atom.
           ii.B) The distribution of other atoms around the ionized atom (radial distribution function, i.e. RDF).

In order to record the detailed fine structure of the edges, core-loss edges should be recorded with a small dispersion, e.g. 0.1 or 0.05 eV per channel.

An interpretation of near edge fine structures is not possible if the signal-to-noise ratio is too low.

Furthermore, in some laboratories, even though monochromators have been installed in the electron microscopes, the monochromators are turned off for routine work because:
          i) The loss of beam current with monochromators on,
          ii) High energy resolutions (e.g. < 0.5 eV) is not always necessary, for instance, in some cases of core-loss spectroscopy, the fine structures in a core-loss spectrum are dominated by lifetime broadening and solid-state effects [1] that normally does not need high resolutions.

However, a monochromator for the electron source or data deconvolution is still necessary in the frontiers of TEM-EELS if the energy spread of the available electron source in TEM is larger than the intrinsic fine structures of spectra.

In practice, several critical processing steps need to be considered and performed before the fine structure on a core-loss edge can be accurately analyzed. Since the ZLP is normally not in the same acquired spectrum, the energy calibration is less straightforward than with low-loss spectra. Ideally, the detector channel of the zero-loss peak should be calibrated to the voltage offset applied to the spectrometer drift tube. However, it is not unusual that the energy stability is poor or there are spectral shifts due to magnetic fields so that the energy calibration is not reliable.

In general, if the TEM specimen is too thick (t/λ > 0.4), a deconvolution process must be employed to remove the effect of plural scattering, since the increase of plural scattering intensity in the higher energy region of an ionization edge can cause some artifacts:
        i) Mask the fine structure;
        ii) Make the background signal on subsequent edges deviate significantly from the power law model.

[1] Mitterbauer, C., Kothleitner, G., Grogger, W., Zandbergen, H., Freitag, B., Tiemeijer, P., Hofer, F., 2003. Electron energy-loss near-edge structures of 3d transition metal oxides recorded at high-energy resolution. Ultramicroscopy 96, 469–480.