Spectral Electron Tomography as a Quantitative Technique to Investigate
Functional Nanomaterials.
Sara Bals
1
, Daniele Zanaga
1
, Bart Goris
1
, M. Pfannmöller
1
, S. Turner
1
and Bert Freitag
2
1.
EMAT, University of Antwerp, Antwerp, Belgium.
2.
FEI Company, P.O. Box 80066, KA 5600 Eindhoven, The Netherlands.
Electron tomography has become a versatile tool to investigate a broad variety of (hetero)nanostructures
in 3 dimensions (3D) [1]. Recent developments even pushed the resolution of the technique to the
atomic level [2-4]. In addition to 3D structural characterization, 3D spectral information containing
compositional, valency and electronic information can be obtained by 3D X-ray Energy Dispersive
Spectroscopy (XEDS) and electron energy loss spectroscopy (EELS).
In an increasing number of studies, XEDS is combined with tomography. These studies rely on newly
developed XEDS detectors [5] such as the Super-X detection system, which consists of 4 individual
detectors, symmetrically arranged around the TEM sample. Although qualitative results obtained by
XEDS tomography have already been reported, [6] it remains difficult to obtain quantitative information
by 3D XEDS and therefore further progress is required. By using the Super-X detector, one is able to
largely overcome problems that were previously related to extreme shadowing of XEDS detectors, but
some shadowing remains. [7] Since these shadowing effects will change for different tilt angles, the
projection principle for electron tomography is no longer fulfilled. In this presentation, we propose a
novel and straightforward approach to minimize the impact of shadowing effects during XEDS
tomography. The method relies on a synergistic combination of XEDS data, quantified using the ζ-factor
approach and 3D high angle annular dark field scanning transmission electron microscopy (HAADF-
STEM). HAADF-STEM yields a relatively high signal to noise ratio, and does not suffer from
shadowing effects other than those related to the “missing wedge”. XEDS, on the other hand, yields
chemical information, even when no prior knowledge of the sample is available or for samples that
contain elements with a small difference in atomic number Z. As a proof of principle, we will here apply
our methodology to different nanostructures containing a mix of Au and Ag.
In addition to 3D compostional data by XEDS, information on valency can be extracted from EELS
measurements. Recently, fine structure changes to the Ce M4,5 edge in the core-loss region of the EELS
spectrum were used to investigate valence changes at the surface of CeO2 nanoparticles. This is
illustrated in Figure 1, where the morphology of a truncated ceria nanoparticle is presented along with
spectra that are extracted from the 3D regions indicated in Figure 1a. From Figure 1b, it is obvious that
small differences in the spectra can be observed for voxels that are located in the inner part of the
reconstruction and voxels that are located at the surface. These differences directly correspond to a Ce
4+
-> Ce
3+
surface reduction. [8] These unique experiments revealed a clear facet-dependent reduction shell
at the surface of ceria nanoparticles, invisible to modern high resolution TEM structural imaging
techniques.
Finally, we will discuss the power of spectral tomography to reconstruct entire volume plasmon spectra
of materials in organic photovoltaic tandem cells. [9] As illustrated in Figure 2, such experiments
provide 3D structural correlations and compositional mapping for these materials, which is far from
straightforward using conventional electron tomography due to a lack of contrast. [10]
274
doi:10.1017/S1431927616002221
Microsc. Microanal. 22 (Suppl 3), 2016
© Microscopy Society of America 2016
https://doi.org/10.1017/S1431927616002221 Published online by Cambridge University Press