However, this temperature could be lowered even further by reducing the thickness of the electrolyte layer to the 100 nm range. To prepare the latter, physical vapor deposition (PVD) methods, such as magnetron sputtering, leading to dense, pinhole-free thin films, are ideal. If, additionally, also the anode and cathode materials are kept thin, lower masses of the whole cells result, creating so-called micro-SOFCs. Thus, less energy would be required for operation, rendering Micro-SOFCs economically efficient as they could also exhibit increased lifetimes.
As preliminary tests of the feasibility of such thin film layers, model anodes and anode/electrolyte systems were prepared. As electrolyte material, YSZ was used, with the anode component consisting of a 1:1 nickel/copper alloy. The latter allows to potentially substitute the conventionally used nickel, which is an ideal catalyst for graphite growth.[4] The presence of copper has already been shown to inhibit carbon formation during operation.[5] The YSZ and NiCu thin films, as well as the NiCu/YSZ bilayer, were deposited via magnetron sputtering on heated silicon and sodium chloride single crystals (873 K) to allow for the creation of well-crystallized and -ordered specimens. The latter are ideally suited for various spectroscopic and microscopic experiments. The stability of the samples in a methane atmosphere at elevated temperatures was further investigated.[6]
X-ray photoelectron spectroscopy (XPS) indicates that minor carbon deposition occurs on YSZ (primarily consisting of sp3 species) upon treatment in methane at 1073 K. Both NiCu and NiCu/YSZ, in contrast, featured a very thick, strongly graphitic carbon overlayer. Elemental maps (STEM-EDX) reveal that, while the NiCu particles are rather homogeneous in the as-grown state, they form a core-shell structure upon reductive treatment in CH4- the particles consist of a nickel-rich core and a copper-rich outer layer. This shell also features variations in composition. While on those parts of the shell where the Ni/Cu ratio is about 3.5 to 5, carbon is deposited (also featuring carbon nanotubes), those with Ni/Cu ratios of about 0.5 do not exhibit any carbon overlayer at all. This suggests that the copper alloy indeed inhibits carbon growth. However, Ni/Cu ratios lower than 1:1 are required. According to DFT calculations, copper segregation is energetically favored and has been shown to occur in hydrogen.[5,8] Hence, it should be possible to generate the carbon-inhibiting core/shell structure starting from the 1:1 alloy by means of a pre-treatment in hydrogen at higher temperatures.
[1] Marinha, D.; Dessemond, L.; Djurado, E. Curr. Inorg. Chem. 2013, 3,2–22
[2] Atkinson, A.; Barnett, S.; Gorte, R. J.; Irvine, J. T. S.; McEvoy, A. J.; Mogensen, M.; Singhal, S. C.; Vohs, J. Nat. Mater. 2004, 3,17–27
[3] Verbraeken, M. Master’s Thesis, University of Twente, 2005
[4] Toebe, M. L.; Bitter, J. H.; van Dillen, A. J.; de Jong, K. P. Catal. Today 2002, 76,33–42
[5] Kim, H.; Lu, C.; Worrell, W. L.; Vohs, J. M.; Gorte, R. J. J. Electrochem. Soc. 2002, 149,A247–A250
[6] Götsch, T.; Schachinger, T.; Stöger-Pollach, M.; Kaindl, R.; Penner, S. Appl. Surf. Sci. 2017, 402,1–11
[7] Götsch, T.; Wallisch, W.; Stöger-Pollach, M.; Klötzer, B.; Penner, S. AIP Adv. 2016, 6 (2),025119
[8] Wolfbeisser, A.; Kovács, G.; Kozlov, S. M.; Föttinger, K.; Bernardi, J.; Klötzer, B.; Neymann, K. M.; Rupprechter, G. Catal. Today 2017,283, 134–143
Figure 1:STEM-EDX map of the NiCu/YSZ sample after methane treatment shows that the carbon deposition does not occur for those surfaces of the core/shell particles that have a Ni/Cu ratio of about 0.5 (marked by blue arrows).