Page 177 - High Temperature Solid Oxide Fuel Cells Fundamentals, Design and Applications
P. 177
154 High Temperature Solid Oxide Fuel Cells: Fundamentals, Design and Applications
Spacil's fabrication procedure was essentially modern. Nickel oxide was mixed
with stabilised cubic zirconia powder in an aqueous slurry of the type used for
slip casting. This slurry was applied to the electrolyte surface and fired at a
temperature up to 15 50°C. To provide a percolation path for electrons through
the anode, the nickel oxide was reduced to metal under hydrogen. The metal
being more dense than the oxide, the initial volume of the NiO component was
diminished by over 25% during this reduction step, and Spacil mentioned the
consequent enhanced porosity as advantageous for the anode. He also presented
alternative deposition procedures for the anode, such as plasma spraying. When
considering the performance of fuel cells today, it is interesting to note that
almost 40 years ago, he reported a power density of over 500 mW/cm2,
admittedly at the very high temperature of 1200°C.
The physical and chemical properties of both nickel and zirconia in the cermet
are critical to their compatibility and functionality. An intimate bonding at the
interface on the nanometre scale is necessary for the synergy of the materials and
for the establishment of the three-phase electrochemically active zone. Therefore
some level of physicochemical interaction or 'wetting' between metal and
ceramic is necessary, though the affinity of metallic nickel for zirconia is weak,
with a contact angle of 120" [7]. To promote bonding, Spacil suggested a lithium
carbonate surfactant flux: in current practice the powder specifications are
chosen to ensure a sufficient surface activity. Addition of metal dopants such as
titanium to the zirconia ceramic is another way to engineer suitable interfaces
[8]. It is useful, however, that this affinity of zirconia for nickel is limited, because
this inhibits interfacial reaction or elemental interdiffusion and allows the
two-phase nature of the cermet to be maintained under operating conditions. It
is known that in a particularly reducing environment, close to open-circuit
operation of a fuel cell with dry fuel, a nickel-zirconium intermetallic, Ni5Zr,
may form [9], but in normal operation the nickel is precipitated as the zirconium
component is reoxidised. This observation confirms the very low solubility of
nickel in stabilised zirconia, perhaps 2% at 1000°C [7], which has even made
possible a recent synthesis of nanodispersed cermet from homogeneous solutions
of nickel, zirconium and yttrium salts [lo].
The key to optimisation of durable efficient anodes in the decades since Spacil
lies in the improvement of materials specifications permitting a sensitive control
of cermet morphology. The original cermet used a high proportion of nickel, over
50% by volume, reduced from nickel oxide of grain size around 45 pm as sieved
through a 325 mesh screen, with non-connected inclusions of 10 pm zirconia
after sintering. The thermal expansion was therefore unduly high, since the
proportion of nickel was in excess of that required for electronic percolation
conductivity, and the lack of connectivity of the ceramic component permitted
long-term nickel aggregation while blocking oxygen ionic transport. With these
materials a temperature of 1550°C was required to sinter the anode to the
zirconia electrolyte substrate. Modern submicron ceramic powders sinter at
1400°C or lower, maintaining a higher specific surface area anode. Associated
with the reduced thermal expansivity of the cermet due to the increased ceramic
content, stresses during fabrication, reduction and operation are minimised,