In-Situ Analysis of the in-Plane Current Distribution Difference between Electrolyte-Supported and Anode-Supported Planar Solid Oxide Fuel Cells by Segmented Electrodes

Problems of the SOFCs include current distribution that decays the total cell performance and efficiency, and causes electrode degradation chemically and thermo-mechanically. In the planar SOFCs, the fuel and oxidant distributions cause current and temperature distributions over the electrodes under the separator ribs and flow channels. Optimized design of the separator (interconnector) is hence required to improve the power generation characteristics and durability of practical fuel cell stacks. Although there have been a number of numerical analyses, very few experimental investigations validating in-situ current distributions to reveal the impact of the separator structure have been reported. We have therefore addressed measurements of in-plane spatial current variations of electrolyte-supported planar SOFCs having segmented cathodes under the rib and the flow channels so far 1,2 . In the present study, we extend the elucidation of the effect of the rib width on the current distribution to that of an anode-supported planar cell to develop generalized numerical model validated with the in-situ acquired distributions in the electrolyte-supported and anode-supported cells. We used the planar cell having three segmented cathodes assembled with segmented cathode separators for electrical insulation. The segmented cathodes were employed opposing to a rib and a set of parallel flow channels of the anode separator. The cell was composed of NiO-YSZ anode-support, YSZ electrolyte, and LSCF cathode (H.C. Starck). The electrode area for the channel was 1.3 cm 2 (2.5 x 0.5 cm) each while those for the ribs were 0.63 cm 2 (2.5 x 0.25 cm), 1.3 cm 2 (2.5 x 0.5 cm), 1.9 cm 2 (2.5 x 0.75 cm). The anode and cathode separator made of SUS430 had the flow channels with a width of 3 mm, a depth of 1 mm, and a length of 2.5 cm. Silver mesh was employed for the current collection of both sides. Current voltage (I-V) measurements were carried out under voltage control using three electric loads to reproduce the electrode equipotential of a single cell 3 . The anode and cathode were electrically connected with the four-terminal method. The anode NiO was reduced to Ni by feeding H 2 /N 2 mixture gas for 2 hours prior to measurements. During measurements, anode and cathode were fed upward with mixtures of H 2 /N 2 and dried air at constant flow rates, respectively. The cell was maintained at 800 °C by a tubular electric furnace at open circuit voltage (OCV). We prepared cathode sep