Page 174 - High Temperature Solid Oxide Fuel Cells Fundamentals, Design and Applications

Anodes 151

Y s7

Figure 6.1 Schematic of anode cermet structure, showing interpenetrating networks of pores and
conductors - nickrl for electrons, yttria-stabilised zirconia for oxide ions. The reactive sites are the contact
zonesofthe twoconductingphases whichareaccessible to fuel through theporosityr21.
discharge electrons to the conducting anode. This requires gas phase for the fuel
access, electrolyte phase for oxide ion entry and metal phase for electron output,
the so-called ‘three-phase boundary’ zone. For efficient operation this should not
be simply a linear structure at a two-dimensional interface of the solid materials,
but rather distributed to provide an active ‘volumetric’ reaction region in three
dimensions. Consequently, the fabrication of the anode is important in
determining this complicated three-phase structure.

6.3 Choice of Cermet Anode Components

Given these stringent requirements, only a few metallic or ceramic candidate
materials are available. After the ‘Nernst mass’, now known as yttria-stabilised
zirconia had been identified as the favoured high-temperature ceramic electrolyte,
Baur and Preis evaluated iron and graphite as anode materials [3]. Graphite, of
course, is susceptible to electrochemical oxidative corrosion, so that cell life with
a graphite anode is unduly short. Platinum also attracted attention due to its
high-temperature stability and catalytic properties, as did other transition metals
as presented in the historical review by Mobius [4]. Even platinum, however, was
unsuccessful as its bond to the electrolyte tends to fail in service, with the anode
layer spalling off probably due to electrochemical generation of water vapour at
the interface. The transition metals also have limitations. Iron is no longer
protected by the reducing activity of the fuel gas once the partial pressure of
oxidation products in the anode compartment exceeds a critical value, and it
then corrodes with formation of a red iron oxide. Cobalt is somewhat more stable,
but also more costly. Nickel has a significant thermal expansion mismatch to
stabilised zirconia, and at high temperatures the metal aggregates by grain
growth, finally obstructing the porosity of the anode and eliminating the
three-phase boundaries required for cell operation. As a consequence, all-metal
anodes have not found acceptance in SOFC technology. Pure ceramic oxide
anode technology is a very recent development and is discussed later.