Precise Electrochemical Testing in Half-Cells: Controlling Catalyst Layer Formation and the Three-Phase Boundary in Gas Diffusion Electrodes

Abstract

Novel and tailored electrocatalysts need solid benchmarking for evaluation of e.g. the oxygen reduction reaction activity (ORR), and the performance in a model approach should strictly reflect their eventual performance in the fuel cell devices. It is strongly believed that the routinely used testing methods, like rotating disk electrode (RDE) method, provide erroneous results, as they do not mimic the original fuel cell conditions in terms of water management, presence of ionomer, relevant current densities, etc. Half-cell tests of gas diffusion electrodes (GDEs) have been shown to be a good compromise as these tests exhibit comparable oxygen mass transport to the electrode as a fuel cell device, in contrast to the limited mass transport in RDEs. Realistic operational conditions, such as potential range and higher current densities in the GDE set-ups, are more akin to membrane electrolyte assemblies (MEA) testing than RDEs without the hassle and cost of full-cell tests [1-4]. In this presentation, half-cell test results will be reported for gas diffusion electrodes prepared using various methods, namely: 1) Drop-casting; 2) Spray coating with sonication; 3) Air brushing with spray gun; and 4) Hand painting. Commercially available Pt/C HiSpec 4000 was used as electrocatalyst and results of parallel RDE and MEA tests with similar Pt loadings will be discussed. Detailed electrochemically active surface area (ECSA) analysis using Hupd and CO stripping studies of various loadings from 5μg cm-2 Pt loading to 50 μg cm-2 will be reported. Drop-casting method on GDE was found to be the most efficient process to study the ECSA using Hupd (38 m2 gPt -1) and CO-stripping (61 m2 gPt -1) which were achieved at a loading of 20 μgPt cm- 2. Spray-coating with ultrasonication with a diluted catalyst ink was the ideal procedure to fabricate a uniform thin layer of catalyst with multiple deposition cycles. The current densities achieved during ORR evaluation in half-cell set-up were almost 2 orders of magnitude higher than those achieved using the RDE method. Different strategies of implementing the Nafion ionomer were utilized and compared, like: 1) Nafion 117 membrane hot-pressed on the GDE; 2) Nafion solution sprayed on the surface of the GDE reaching a loading of 2 mg cm-2 and; 3) GDE without a membrane. This report will also aim to assess how the presence of phosphoric acid in the perchloric acid electrolyte and its purity affects the catalyst’s performance for the ORR reaction. Scanning electron microscopy (SEM) was used to study the catalyst layer formation on the gas diffusion layers (GDLs). The GDEs were characterized using X-Ray Absorption Spectroscopy (XAS) [5, 6], ex-situ and in-situ, allowing us to extract structural information through the extended X-ray absorption fine structure analysis (EXAFS) (catalyst particle size, composition, morphology, and oxidation state); furthermore, by using a subtractive-normalization technique in the near-edge region (Δμ XANES) it is possible to obtain adsorbate coverage (H, O, PO4 species adsorbed on Pt surface) and binding-site information for a detailed comparison between the above samples.