Page 57 - High Temperature Solid Oxide Fuel Cells Fundamentals, Design and Applications

The focal point of work on solid oxide fuel cells during this period was the
development of electrode materials. An early problem was poor adhesion of
the anode layers, which became obvious in 1963 [59]. Spacil as early as 1964
found the now well-known solution of using layers of nickel closely mixed with
solid electrolyte material [loll.
It was considerably more difficult to find a suitable cathode material. It turned
out to be a misconception to think that small concentrations of polyvalent
cations in mixed oxides with fluorite structure produce a high electronic
conductivity. Layers with a high concentration of praseodymium at the cathode
and high concentration of cerium at the anode had to be realised in order to
achieve anything near the desired conductivities [102]. Only mixed oxides with
uranium proved to be a good material for stable layers with eIectronic and ionic
conductivity, spreading the electrode reactions across the three phase
boundaries of electrode/electrolyte/gas. This idea was confirmed by the result
that thin mixed conducting interlayers between the pure electrolyte and the
metallic conductor considerably reduced polarisation phenomena [lo31 and
led to high current densities [104]. An optimised material with polyvalent
uranium ions in mixed oxides with fluorite structure suitable for sintering on
sohid oxide electrolytes without phase boundaries was developed by
Tannenberger in 1967 [lO5, 1061; it proved to be a favourable interlayer in
cathodes and anodes [ 10 7,lO 81.
Indium oxide with different additives was proposed as a cathode material in
1966 [lo91 and frequently used (e.g. [110, 107, 1081). However, electronically
conducting perovskites soon began to dominate the developments for both
cathode and interconnect. The use of Lal-xSrxCo03 for the air electrode of solid
oxide fuel celIs marked the beginning [lll], followed in 1967 by
recommendations of PrCoOs [112] and of mixtures of the oxides of Pr, Cr, Ni and
Co [ 11 31. Strontium-doped lanthanum chromite, even now the most important
ceramic interconnection material, was proposed by Meadowcroft in 1969 [114].
For cathodes, the situation in 1969 was summarised [ 11 51 as: 'It is apparent that
a fully satisfactory air electrode for high temperature zirconia electrolyte fuel
cells is still lacking.'
The SOFC stacks developed in 1963 were not safe enough for power
generation aboard spacecrafts. SOFCs for electrolysis of the atmosphere in
manned spacecrafts (recovery of oxygen from COz and H2O in stacks of bell-and-
spigot cells, carbon deposition, and hydrogen separation [116]) were also
investigated. For this application, electrolyte discs (6.3 cm diameter, 1.4-1.6
mrn thick) and for the first time mixed oxides of Zr and Sc [117] were used.
Flat-plate designs of SOFCs had been proposed earlier (Figure 2.9 [118,119]).
Terrestrial applications aimed at an economic SOFC system for the production
of electrical power from coal and air at an overall efficiency of 60% or greater.
With a conceptual 100 kW coal-burning fuel cell power system (Figure 2.10
[l20]), coal could be gasified using the heat and combustion products emerging
from the fuel cell stacks.
In 1968 General Electric took up the idea of electrochemical dissociation
of water vapour in solid oxide cells [121-1231. They hoped to produce cheap,