Page 288 - High Temperature Solid Oxide Fuel Cells Fundamentals, Design and Applications

Testing of Electrodes, CelIs and Short Stacks 265

Note that there is a potential loss across the electrolyte due to the electrolyte
resistance. The potential steps at the interfaces are now smaller compared to the
OCV condition due to the losses originating from the polarisation resistance of
the electrode processes. Figure 10.2~ illustrates the case of an anode-supported
cell. The potential is given for both the position of the ‘reference’ electrode where
no current flows across the electrolyte (i.e. the electrical potential is constant
across the electrolyte), and for a position far away from the electrode edges as in
Figure 10.2b.
As the potential along the anode, which is a very good electronic conductor, is
the same everywhere (the anode constitutes an isopotential plane), the potential
of the anode at the ‘reference’ electrode must be equal to the potential in the
middle of the current-bearing part of the anode. Thus the two potentia1 curves in
Figure 10.2~ must start at the same point. Therefore, as seen in Figure 10.2c, the
potential difference between the ‘reference’ electrode and the upper electrode in
Figure 10.1 is simply the total polarisation of the full cell. Thus it is clear that the
‘reference’ electrode measures only the emf of the cell with the actual gas
compositions at the ‘reference’ and inside the support at the lateral position
opposite to the reference while the current is flowing. Thus, no information on
what happens on any of the working electrodes can be derived from such
measurements. If the concentrations of the reactants and products at the lateral
position of the reference electrode were the same as in the active electrode/
electrolyte interfaces, then it would be possible to deduce the total concentration
overpotential. This is, however, in general not the case, and this means that
the voltage difference actually measured between the working and the
‘reference’ electrodes cannot be assigned any clear meaning. In the present
context, it is also helpful to remember that a single-electrode potential cannot be
measured directly: it is only possible to measure a potential difference between
two electrodes.
To avoid such problems, it is necessary to test the electrodes using a suitable
three-electrode set-up or a symmetrical two-electrode cell [40], even though
both have their shortcomings. Some examples of useful set-ups for studying
electrode performance are briefly presented here. The set-ups are based on zirconia
pellets with an eIectrode arrangement suitable for three-electrode studies.
Specific material choices and mounting details are given in the caption to Figure
10.3. Such pellet-like geometries, where the reference electrode can be suitably
placed (in a bore as in the figure or as a ring around the pellet) are suitable for
fundamental studies of electrode kinetics. The pellet-like test cell geometries
depicted in Figure 10.3 suffer from two disadvantages: it is difficult to ensure that
the fabrication process for the electrodes used is identical to the one used for the
actual cells, and the ohmic resistance between the working and the reference
electrodes is quite substantial which may result in a ‘signal to noise’ problem
when very good electrodes are studied. An improved three-electrode geometry
using a ring-shaped working electrode is currently being investigated [3 91.
To measure a particular electrode performance in detail, a symmetrical cell
with identical electrodes on each side can be used as shown in Figure 10.4. This
has a platinum mesh to make good contact with the electrodes and two platinum