3D Imaging of PEMFC Electrode Microstructure by Electron Tomography and 3D FIB/SEM

Abstract

Scanning and transmission electron microscopies (SEM and TEM) are two imaging techniques widely used in the study of PEMFC electrode microstructure. They have played a significant role in the improvement of electrode components and the understanding of their main degradation mechanisms. However, the currently available description of the electrode microstructure does not allow yet to unambiguously identify the key microstructural factors that limit gas and proton transport within the electrode. These factors are crucial as they control the performance of the PEMFC, especially at high current density, and are therefore the subject of numerous studies, using either fine electrochemical measurements or modeling approaches.The objective of our work is to provide an improved description of the electrode microstructure by giving quantitative information on its porosity at different scales as well as on the ionomer distribution. For this purpose, we develop 3D characterization techniques applied to the PEMFC electrodes such as electron tomography and 3D FIB/SEM.The studied cathode electrode was prepared using TEC10E50E catalyst from Tanaka that is 50 wt % loading of Pt deposited on high-specific area carbon (HSAC), and Nafion D2020. The Pt loading of the electrode is 0.2 mgPtcm-2 and the ionomer/carbon ratio is 0.8.As a first step, electron tomography analyses were performed on the Pt/HSAC powder to determine the distribution of Pt nanoparticles (NPs) which can be located either on the carbon surface (outer NPs) or inside the carbon porosity (inner NPs)[1-3]. Our strategy is to record tilt image series by using two annular detectors: i) a high angle annular detector avoiding Pt NP diffraction contrast for the 3D image reconstruction of Pt NPs (without diffraction artifact) and ii) a lower angle annular detector that enhances carbon contrast for the 3D image reconstruction of the carbon phase. These two segmented images were then combined (Figure 1), allowing the calculation of the NP size histogram but also of the volume and the surface area of the two NP populations (inner and outer NPs).On the second step, the porosity of the electrode was analyzed by 3D FIB/SEM. A volume of around 10 X 10 x 5 μm3 was imaged with a cubic voxel size of 5 × 5 × 5 nm3. The solid phase (carbon, Pt NPs and ionomer) segmentation was based on a composite image created from two electron detectors (secondary and backscattered electrons) and using a machine learning software. Quantitative data such as the total porosity, the pore size distribution or the agglomerate size distribution of the solid phase were extracted from this volume. Figure 2a shows a 2D slice of this segmented volume where the solid phase is in black and the porosity in white.In order to analyze the electrode porosity with a higher spatial resolution, electron tomography experiment was also performed on a 150 nm thick slice of the electrode cut by ultramicrotomy after being embedded in an epoxy resin. The tilt image series were recorded with a pixel size of 0.5 nm and a field of view of 2 x 2 µm2. The 2D slice of the reconstructed volume (Figure 2b) shows that the largest solid agglomerates detected on the 3D FIB/SEM image can be identified as an agglomeration of numerous carbon particles. The challenge now is to quantify the porosity created between these carbon particles and to determine how many Pt NPs are located in these large agglomerates.Finally, the distribution of the ionomer was analyzed within the electrode similarly to [4]. For this analysis, a 150 nm slice of the electrode was cut by cryo-ultramicrotomy without embedding in epoxy resin as described in [5]. TEM images show that the ionomer is not homogeneously distributed on the surface of the entire carbon particles but is often observed in the concave surface created between the carbon particles (Figure 3). These observations suggest a ionomer distribution in agreement with the heterogeneous ionomer coating model proposed by Inoue et al. [6]. We also suggest that a part of the ionomer could be localized in the small pores located inside the large carbon agglomerates described in the previous paragraph. This could explain why the ionomer is so difficult to detect on TEM images. This work has been done in the frame of the FURTHER-FC project. This project has received funding from the Fuel Cells and Hydrogen 2 Joint Undertaking (now Clean Hydrogen Partnership) under Grant Agreement No 875025. This Joint Undertaking receives support from the European Union’s Horizon 2020 Research and Innovation program, Hydrogen Europe and Hydrogen Europe Research. Figure 1