Quantitative Structural Assessment of Heterogeneous Catalysts by Electron Tomography

Abstract

We present transmission electron microscope (TEM) tomography investigations of ruthenium-based fuel cell catalyst materials as employed in direct methanol fuel cells (DMFC). The digital three-dimensional representation of the samples not only enables detailed studies on number, size, and shape but also on the local orientation of the ruthenium particles to their support and their freely accessible surface area. The shape analysis shows the ruthenium particles deviate significantly from spherical symmetry which increases their surface to volume ratio. The morphological studies help to understand the structure formation mechanisms during the fabrication as well as the high effectiveness of these catalysts in the oxygen reduction reaction at the cathode side of fuel cells.

Three-dimensional analysis of Nafion layers in fuel cell electrodes

“…This makes it ideal to study materials with very different average atomic numbers. Thereby, electron tomography is quite widely used to study the nanoparticle distribution on their support in different catalyst materials 17 and particularly in fuel cell electrodes [18][19][20][21] .…”

Autonomous reconstruction and segmentation of tomographic data

“…Nevertheless, such data are used in extensive data analyses (see e.g. [24]). On the other hand, a missing wedge does not necessarily lead to artefacts in the reconstruction.…”

Removing the effects of the “dark matter” in tomography

“…Blurring and artefacts in tomograms make their processing for visualisation and accurate segmentation difficult, leading to a requirement for dedicated human intervention that is subjective and time-consuming [17,19,25,35,39]. Fig.…”

“…Nonetheless, much effort is being put into novel reconstruction algorithms which can notably improve the quality of tomograms, even with low sampling and limited angular range, using either the discretisation of intensities, the calculation of sinusoidal trajectories in sinograms, the incorporation of geometric prior knowledge, or compressive sensing [27][28][29][30][31]. Tomograms obtained after the reconstruction are 3D datasets containing a continuous range of intensity levels, and therefore in order to extract useful information, visualisation and segmentation are required in combination with the application of sophisticated algorithms, such as anisotropic non-linear diffusion, to reduce noise and enhance local structure without worsening the resolution or structural information, adaptive thresholding, or equalisation in real space [32][33][34][35]. In the present work, we focus on the effects of an experimental parameter: the minimum value of the intensity used to acquire a tomographic tilt series.…”