Elementary Reaction Kinetics of the CO∕CO[sub 2]∕Ni∕YSZ Electrode

Abstract

Results of combined experimental and theoretical investigations of elementary chemical reaction processes of CO-CO 2 gas mixtures at nickel/yttria-stabilized zirconia ͑Ni/YSZ͒ solid oxide fuel cell ͑SOFC͒ model anode systems are presented. Temperature-programmed desorption and reaction measurements were performed in order to determine adsorption/desorption kinetic data as well as thermodynamic parameters for the CO/CO 2 /Ni and CO/CO 2 /yttria-stabilized zirconia ͑YSZ͒ heterogeneous reaction systems. From these data, an elementary kinetic reaction mechanism of the electrochemical CO oxidation at Ni/YSZ anodes was developed. Numerical simulations were performed for three different spillover mechanisms. Steady-state polarization curves and electrochemical impedance spectra were calculated, allowing for a direct comparison with experiments performed by Lauvstad et al. ͓J. Electrochem. Soc., 149, E506 ͑2002͔͒. Best agreement with the experimental data was obtained when assuming two consecutive charge-transfer steps from YSZ-O 2− via YSZ-O − to Ni-O, the second step being accompanied by oxygen spillover over the three-phase boundary.

SOFC modeling considering hydrogen and carbon monoxide as electrochemical reactants

“…Note that this electrochemical 13 approach also includes the gas-phase partial pressures, which are frequently neglected within the open literature and also in our previous models [3,37,44]. Here K eq the temperature dependent equilibrium constant for eqns (2) and (3), respectively and AV e is the electrochemically active surface area-to-volume ratio. For AV e an average value of 9·10 6 m 2 /m 3 is applied in this work, which can be compared to 70·10 6 m 2 /m 3 , the measured specific surface area for Ni/YSZ material developed for SOFC anodes [46].…”

“…There are a variety of designs developed for FCs, but they all operate with the similar basics and principles [1]. FCs are promising candidates for providing electrical power for future energy systems due to the high efficiency and low emissions of NO x compared to conventional power generation systems [2]. The solid oxide fuel cell (SOFC) operates at temperatures between 600-1000°C [3].…”

“…The generated oxygen ions are transported through the electrolyte, but the electrons are prevented to pass through it. The electrochemical reactions between hydrogen and oxygen ions, eqn (2), as well as between carbon monoxide and oxygen ions, eqn (3), take place in the anodic TPB [7][8]. …”

Experimental characterization and elementary reaction modeling of solid oxide electrolyte direct carbon fuel cell

“…Since experimental studies on SO-DCFC are rather complex, expensive, and time-consuming, comprehensive mathematical models are essential for the technology development. A validated mechanistic model would offer means to gain insight into the complex physical phenomena governing fthe uel cell performance that is not readily accessible experimentally, and it can also be useful tool for cell design and operating condition optimization.Modeling and experimental studies of SO-DCFCs have been reported recently by several researchers [6][7][8][9]. Numerous SOFC models considering the intricate interdependency among ionic and electronic conduction, gas transport phenomena, and electrochemical processes have been reported in the literatures for pure hydrogen, syngas or methane [10][11][12][13][14][15][16][17][18][19][20].…”

Bayesian inference of chemical kinetic models from proposed reactions

“…The standard approach to building models in such a case is to postulate reaction models and to compare them based on their ability to reproduce indirect system-level experimental data. Data-driven model comparison thus involves defining a metric of fit, e.g., the sum of squared residuals, and selecting a model that minimizes this metric, e.g., least-squares or regularized least-squares fitting (Hanna et al, 2014;Vogler et al, 2009;Yurkiv et al, 2011). The principal limitation of the least-squares fitting approach is that it only identifies a single "best" model and yields point estimates of the underlying parameter values, without providing a meaningful description of the uncertainty among competing models and in their parameters.…”