Page 49 - High Temperature Solid Oxide Fuel Cells Fundamentals, Design and Applications

30 High Temperature Solid Oxide Fuel Cells: Fundamentals, Design and Applications

measurements were to lead to useful results. Compounds of thorium and cerium
were effectively purified as ammonium double nitrates by crystallisation from
hot concentrated nitric acid. Pure lanthanum oxide was prepared by fractional
precipitation of hydroxides. Bearing in mind experience of oxide ceramics [42],
powders of mixed oxides were pressed to produce gastight discs, 25 mm in
diameter and 1-2 mm thick, which were sintered at temperatures up to 1920°C
in a stream of oxygen on a support stack of A1203/MgA1204/Th02 in alumina
tubes using a Tammann carbon tube furnace. Zr02 for this investigation was
available at that time only in the form of a mixture with Y2O3 as a residue from
the investigations of Peters because an embargo and the development of the
nuclear industry made it difficult to obtain.
There were additional reasons for concentrating in Rostock on solid electrolytes
based on Tho2. In 1948, Ryschkewitsch [42] pointed out that a large-scale
technical application of Tho2 was still lacking. During the 1950s, it seemed that
more zirconium than thorium was needed for the development of nuclear
energy. Furthermore the mixed oxides with Tho2 are crystallographically simpler
than those with Zr02. Some stocks of Tho2 existed for the fabrication of mantles
for gaslight.
In the investigations, carried out from 1955 to 1957, for cells with different
composition of the solid electrolyte, the electrode voltages were measured in the
temperature range between 300 and 1350°C and compared with
thermodynamically calculated values.
Schottky had shown that the efficiency of solid electrolyte fuel cells with
increasing load resistance decreases to zero if a noticeable part of the
conductivity of the electrolyte is of electronic nature [34]. Therefore the efforts
for purification and especially for separation of the polyvalent praseodymium
cations from the solid electrolyte material were made. In the case of Th-La mixed
oxides, with only 1 mol% La01.5 the ion transport number 1 was reached,
admittedly only with reducing gas on both electrodes (in CO,CO2 concentration
cells); in the oxygen/air cell even at 10 mol% LaOl.s this number was only near
0.8. A perfect disc of Ceo.9Lao.101.95 broke into pieces in CO,CO2/O2 between
700 and 840°C reaching the mean ionic transport number 0.8. For the
available Zr02 solid electrolyte (with 50 mol% Y01.5) in the oxygen/air cell,
the ion transport number was above 0.93.
On the basis of these results, the Boudouard equilibrium was investigated with
Tho.gLao.101.9s as solid electrolyte in the cell C0,C,Fe/Pe0,C0,C02, using only
the reactive carbon precipitated out of CO; iron in metallic or oxide form in the
electrodes supported the establishment of the electrode potential catalytically.
And with the Zr02 solid electrolyte in a C0,C02,Fe304/Pt,02 cell, the C02
dissociation equilibrium was investigated [43].
The good agreement between measured and thermodynamically calculated
data in these cases led to the most important by-product of SOFC development: if
solid electrolyte cells, charged with gases of known concentrations, deliver the
theoretically expected cell voltages, it also must be possible to calculate
unknown gas concentrations backwards from the cell voltages, measured
between the cell terminals in gas phases, which can be oxidising or reducing.