Numerical model of planar anode supported solid oxide fuel cell fed with fuel containing H2S operated in direct internal reforming mode (DIR-SOFC)

Abstract Experimental analysis of a planar 100 mm × 100 mm SOFC cell was conducted during operation at 1173 K in direct internal reforming (DIR) mode. In the first phase the rate of direct internal reforming was varied from 0 to 100% what corresponds to complete external reforming and complete DIR, respectively. In the second phase 1.2 ppm(v) of H 2 S was introduced to the feeding gas and the variation of the rate of direct internal reforming was repeated. Following the experimental analysis the numerical model was proposed to determine the correlation between the presence of the poisoning agent and the electrochemical performance. The effect on the resistance of the cell was studied. The lumped volume model was applied to predict the cell voltage. With the use of the experimental data it was possible to determine the relative change of the model parameters which describe the ionic and electronic conductivity of the SOFC. Model was adopted for predictive modeling of the solid oxide fuel cell, operated in DIR-SOFC mode with and without the presence of hydrogen sulfide. Additionally, literature data measured for a cell operated in complete internal reforming mode with variation of the sulfur content in the feeding gas were analyzed to define the effect of H 2 S content on the performance drop. Relative change of the resistance of a cell was correlated with the rate of internal reforming and the content of sulfur. Results of the analysis show that the degradation of the performance of SOFC due to sulfur poisoning during operation in DIR mode can be modelled with high fidelity. Change of the ionic and electronic resistance of a cell accounted for the maximum of 34 and 53%, respectively when the rate of DIR was altered between 0 and 100%. The contribution of the sulfur poisoning accounts for 69 and 79% when the H 2 S content varies in the range of 0.001–5 ppm(v). With average relative prediction error below 3%, the proposed approach finds good application in simulating the performance of a cell exposed to different gas mixtures with different levels of sulfur in the fuel stream.