Loading...
Investigation of sulfur interactions on a conventional nickel-based solid oxide fuel cell anode during methane steam and dry reforming
Jablonski, Whitney S.
Jablonski, Whitney S.
Citations
Altmetric:
Advisor
Editor
Date
Date Issued
2015
Date Submitted
Collections
Research Projects
Organizational Units
Journal Issue
Embargo Expires
Abstract
Solid oxide fuel cells (SOFC) are an attractive energy source because they do not have undesirable emissions, are scalable, and are feedstock flexible, which means they can operate using a variety of fuel mixtures containing H₂ and hydrocarbons. In terms of fuel flexibility, most potential fuel sources contain sulfur species, which severely poison the nickel-based anode. The main objective of this thesis was to systematically evaluate sulfur interactions on a conventional, Ni/YSZ anode and compare sulfur poisoning during methane steam and dry reforming (SMR and DMR) to a conventional catalyst (Sud Chemie, Ni/K₂O-CaAl₂O₄). Reforming experiments (SMR and DMR) were carried out in a packed bed reactor (PBR), and it was demonstrated that Ni/YSZ is much more sensitive to sulfur poisoning than Ni/K₂O-CaAl₂O₄ as evidenced by the decline in activity to zero in under an hour for both SMR and DMR. Adsorption and desorption of H₂S and SO₂ on both catalysts was evaluated, and despite the low amount of accessible nickel on Ni/YSZ (14 times lower than Ni/K₂O-CaAl₂O₄), it adsorbs 20 times more H₂S and 50 times more SO₂ than Ni/K₂O-CaAl₂O₄. A one-dimensional, steady state PBR model (Detchemᴾᴮᴱᴰ) was used to evaluate SMR and DMR under poisoning conditions using the Deutschmann mechanism and a recently published sulfur sub-mechanism. To fit the observed deactivation in the presence of 1 ppm H₂S, the adsorption/desorption equilibrium constant was increased by a factor 16,000 for Ni/YSZ and 96 for Ni/K₂O-CaAl₂O₄. A tubular SAE reactor was designed and fabricated for evaluating DMR in a reactor that mimics an SOFC. Evidence of hydrogen diffusion through a supposedly impermeable layer indicated that the tubular SAE reactor has a major flaw in which gases diffuse to unintended parts of the tube. It was also found to be extremely susceptible to coking which leads to cell failure even in operating regions that mimic real biogas. These problems made it impossible to validate the tubular SAE model with experimental data. However, results from the tubular SAE model were compared with results from a previously developed tubular SAE model and showed good agreement for fuel channel exhaust gases.
Associated Publications
Rights
Copyright of the original work is retained by the author.