Understanding the mechanisms of performance degradation is of key importance for increasing the lifetime of solid oxide fuel cells (SOFC). Due to the complex multilayer structure and the partial thermomechanical incompatibility of the different layers, thermal stress in SOFC stacks can cause various problems such as crack formation, microstructural instability and delamination. In case of internal reforming of natural gas and other hydrocarbons a complex temperature distribution with large thermal gradients can be generated by the cooling effect of the fast endothermic steam reforming reaction [1]. Furthermore, hydrogen sulfide (H₂S) contaminations in the fuel gas will poison the catalytic active Ni within a Ni/8YSZ-anode and slow down electrochemical as well as catalytic reactions [2]. In order to advance the understanding and avoid degradation due to thermomechanical stress, a detailed knowledge of the temperature distribution and its interdependence with occurring loss mechanisms is necessary.

This work presents a non-isothermal 2D FEM gas channel model, capable of performance predictions for hydrocarbon-fueled single SOFC stack layers. Therefore, a previously developed isothermal model [3;4], incorporating relevant occurring loss mechanisms for hydro-carbon SOFC operation, was extended by the energy balance equations: joule heating due to ohmic losses as well as endothermic and exothermic catalytic and electro-catalytic reactions. By implementing experimentally determined heat conductivity parameters, heat transport is described realistically by heat conduction, convection and radiation within every cell layer. This enables the model to spatially predict the temperature distribution within Ni/YSZ-based SOFCs in hydro-carbon fuel operation in dependence of occurring loss mechanisms (Fig. 1). Furthermore, the deactivation of active catalyst surface area via Sulphur poisoning is regarded by implementing surface area dependent reforming kinetics. These were determined by measuring the gas conversion of fuels containing different amounts of H₂S in a specialized test rig with gas extraction and temperature tracking probes along the gas channel. The presented results will show, (i) how poisoning the Ni-catalyst will affect the reforming-activity and thus consequently the temperature distribution. And (ii), how these effects play out differently in anode-supported cells (ASC) compared to electrolyte-supported cells (ESC).

[1] L. Blum, L.G.J. de Haart, J. Malzbender, N. Menzler, J. Remmel, R. Steinberger-Wilckens, "Recent results in Jülich solid oxide fuel cell technology development", J. of Power Sources 241, p. 477-485 (2013).

[2] A. Kromp, S. Dierickx, A. Leonide, A. Weber and E. Ivers-Tiffée, "Electrochemical Analysis of Sulfur-Poisoning in Anode Supported SOFCs Fuelled with a Model Reformate", J. Electrochem. Soc. 159, p. B597-B601 (2012).

[3] H. Geisler, A. Kromp, A. Weber and E. Ivers-Tiffée, "Stationary FEM Model for Performance Evaluation of Planar Solid Oxide Fuel Cells Connected by Metal Interconnectors", J. Electrochem. Soc. 161, p. F778-F788 (2014).

[4] H. Geisler, S. Dierickx, A. Weber and E. Ivers-Tiffee, "A 2D Stationary FEM Model for Hydrocarbon Fuelled SOFC Stack Layers", ECS Trans. 68, pp. 2151-2158 (2015).