Page 54 - High Temperature Solid Oxide Fuel Cells Fundamentals, Design and Applications

History 35

the ‘excluded volume model’ [8 71. Doubts concerning the rapid intermixing
of the different cations of oxide electrolytes and oxide electrode layers were
resolved by in~~estigations with radionuclides [88-901, which confirmed the low
mobility of the cations in mixed oxides with fluorite structure. (In CaO-stabilised
zirconia the ratio of the self-diffusion coefficients of the anions to that of the
cations is larger than lo6 even at 1700°C).
At high oxygen pressures, oxide phases show defect electron (hole) conduction
(oxidation semiconduction) and at low oxygen pressures excess electron
conduction (reduction semiconduction). The transport number of excess
electrons in Zro.s$ao.lsOl.s3 as a function of the oxygen partial pressure could
be determined by measurements with a Ca,CaO/air cell [79]. The hole
conduction of zirconia-based solid electrolytes was noticed for the first time when
cells with Ni,NiO reference electrodes for gas potentiometry [44,91 J were
tested in air. The harmful oxygen permeability was measured potentiometrically
in 1965 1921.
Also in 1965, the fundamentals of gas potentiometry were presented,
including the range of free oxygen and of oxygen in equilibria, and the ‘neutral’
transition range [93]. Calculations and measurements in the case of
potentiometric titrations of different gases were in good agreement in all three
ranges [94]. (The sudden change of the cell voltage of a hydrogen/air cell zt the
equivalence point when oxygen was fed to the hydrogen had already been
shown graphically by Archer et al. [59].) Thus the investigations started for
SOFCs led to the development of oxygen sensors (lambda probes) now widely
used in automobiles. (A zirconia cell working potentiometrically was first
proposed by Loos in 1969 as a sensor for O2 and CO for the regulation of the air/
fuel ratio in cars [ 9 51 .)
Another less well-known by-product of SOFC development was the
electrochemical thermometry: i.e. the determination of elevated temperatures on
thethermodynamicscalewithC0,C02,H2,H20 11961 orOz concentrationcells [97].
The first investigations of polarisation phenomena in solid oxide fuel cells were
conducted by the research groups in Sverdlovsk [61,98], Frankfurt [63], Geneva
11691, GrenobIe [70] and Nagoya [71]. In the detailed investigations of fuel cells
with cerium-lanthanum mixed oxides by Takahashi et al. [71] the polarisations
observed were much smaller at the anode than at the cathode because by partial
reduction of the solid electrolyte, a mixed conductor (solid solution of Ce203 in
CeOz) was formed at the anode giving a depolarising interlayer. Detailed
investigations of the polarisation of solid electrolyte cells by determining the
complex admittance were first conducted by Bauerle in 19 69 [99].
The high conductivity of cerium-lanthanum mixed oxides and the favourable
polarisability of electrodes on such solid electrolytes was already stimulating
application ideas in the 1960s. But electronic conductivity of these electrolytes
above 600°C was seen as a weighty problem [71]. The influence of electronic
conductivity on the cell performance was investigated first by means of an
equivalent circuit [40,100]. The results, shown in Figure 2.8, led to the
conclusion that the ion transport number has to be greater than 0.9 if a solid
electrolyte was to be successful in a SOFC [loo].