Page 276 - High Temperature Solid Oxide Fuel Cells Fundamentals, Design and Applications

Electrode Polarisations 253

Although measurements of separate cathode and anode overpotentials
contain errors, it is possible to obtain the total cathode + anode overpotential
with reasonable accuracy, by subtracting the ohmic contribution from the total
voltage, at a given current density.
In a typical AC impedance measurement, the losses occurring in cell operation
are represented by the ohmic resistance Ro and the polarisation resistances
Zcath + Zanod = Zpo~. The terminals of the measurement device can be connected
either to the working electrodes or to a working electrode and a reference
electrode on either side, in order to measure Zpol, Zcath and .&nod, respectively. Due
to ohmic losses in the cell a part of the potential difference across the cell is
included in each of the measured I-V-characteristics or impedance curves. The
distribution of the electrolyte resistance to the measured electrode impedances
significantly depends on the arrangement of the electrodes. In the case of an ideal
electrode alignment (Le. electrode misalignment < < electroIyte thickness) this
electrode arrangement can provide useful, although not completely free of
errors, information about the polarisation processes of the individual electrodes
[46]. Otherwise significant errors may occur due to the inhomogeneous current
density distribution 120,471.
In addition to providing information on polarisations, impedance
spectroscopy is also useful in simulating reaction mechanisms; this is illustrated
here for the cathodic reaction. In the reaction model presented here, on137 the
following steps are considered: (1) Dissociative adsorption of oxygen at the
cathode surface: (2) Surface diffusion of adsorbed oxygen along the cathode
surface through the pores: and (3) Reduction of adsorbed oxygen at the TPB and
subsequent vacancy exchange and oxygen incorporation into the electrolyte.
The electrolyte surface is assumed to remain inactive because of its low
electronic conductivity. For further simplification, it is assumed that the oxygen
surface diffusion proceeds sufficiently fast and therefore can be neglected.
The dissociative adsorption of oxygen is assumed to proceed via the reaction

where kads and hes are the rate constants for adsorption and desorption of
molecular oxygen, ‘s’ is a vacant active surface site for oxygen and Oan is an
oxygen atom adsorbed on an active site. The adsorbed oxygen then diffuses along
the pore walls of the cathodes and enters the TPB-region where it reacts
according to the following equation

and transfers into the electrolyte bulk. Here, k,d and k,, are the rate constants for
the oxygen exchange reaction, in forward and reverse directions, respectively.
The law of mass action applied to reactions (2 7) and (28) yields