Page 50 - High Temperature Solid Oxide Fuel Cells Fundamentals, Design and Applications

History 31

Accordingly the potentiometric determination of gas concentrations with solid
electrolyte cells and the first designs of probes for the in situ analysis of hot gases
(reference electrode e.g. Ni,NiO or a gas with known oxygen partial pressure)
were patented in 19 5 8 [44]. The first calculations of oxygen partial pressures in
purified nitrogen, using the measured cell voltages of solid oxide cells, were
performed in 19 5 5.
In 195 7 Kiukkola and Wagner reported thermodynamic investigations on
metal/metal oxide systems, for the first time using CaO-stabilised ZrOz (especially
2ro.s jCao.1 jO1.85) as solid electrolyte [45]. But they could not realise the
intended measurements by using a gas reference electrode because their solid
electrolytes (sintered at 1400-1450°C) were porous [46]. The investigations of
Peters and Mann on metal/ metal oxide systems with gastight Tho.9Lao.101.95
solid electrolyte were performed using reference electrodes with CQ,CQ2
mixtures [47].
The electronic part of conduction of Thoz electrolytes could be observed
increasing with the oxygen partial pressure (oxidation semiconduction) even
with pure white mixed oxides (purified from polyvalent cations). During the
establishment of the electrode potentials there were signs of solubility of oxygen
in the lattice. These facts led to the conclusion that the electronic conduction
arose in the anion sublattice and that generally, in mixed oxides with oxide ion
vacancies, holes can exist in the form of monovalent negative oxide ions.
For understanding the oxide ion conduction in mixed oxides, there was the
problem, already seen by Wagner [22], that the radius of the oxide ions is larger
than that of all cations in the crystals. Along with the concentration, it is always
the mobility of the charge carriers which determines the conductivity of
homogeneous bodies. By space-geometrical considerations it could be shown
[40,48] that the fluorite lattice in particular offers better possibilities for the
motion of the larger anions than it does for the smaller cations (Figure 2.6).
Furthermore, from geometrical calculations it was clear that with decreasing
radius of the cations down to a lower limit the possibilities for the motion of
cations decrease and those of the anions increase.
The fact that the cations are firmly held in their places in the oxide ion
conductors has much importance for the long-term stability of fuel cells. The
comprehension of the low cation mobility supported the suggestion of
incorporating polyvalent cations in mixed oxides with fluorite structure for
obtaining electronic conducting layers and producing stable electrodes at the
oxide ion conductors by sintering [49] (aiming at a continuous row of mixed
crystals with the electrolyte material, contrary to the recommendation of
Schottky [34]). The layers of mixed conductors should ensure ideal conditions
for the conduction of oxide ions and electrons and also for the transfer reactions
in the electrodes. After these ideas had been presented in the Class of Chemistry
of the Academy of Sciences in Berlin in 1958 [50], there were substantial
doubts in the discussion; a statistical mixing of all the different cations in a
homogeneous solid phase at high temperatures was considered to be very
probable. In this and other cases, important questions remained. For example,
the cause of the relatively stable cell voltages, which were repeatedly observed