Page 31 - High Temperature Solid Oxide Fuel Cells Fundamentals, Design and Applications

12 High Temperature Solid Oxide Fuel CeZls: Fundamentals, Design and Applications

and cracking in 100 mm x 100 mm plates. This has caused difficulties as larger
planar stacks have been constructed with thick interconnect plates.
Such large lanthanum chromite interconnect plates have generally been
made by powder processing methods. A fine powder of the desired composition
is prepared by mixing lanthanum, strontium and chromium nitrates, then
reacted with glycine at a high temperature [26]. The reaction mixture fluffs up
into a fine powder which can be readily compacted to form interconnect
plates, or extruded to make tube structures [27]. For example, calcia-doped
lanthanum chromite was co-extruded with YSZ to make an electrolyte tube
containing an interconnect strip along its length. This was co-fired to give a
dense composite.
A major difficulty with such interconnects is the difficulty of sintering to full
density. Lanthanum chromite powders do not sinter easily, especially in
oxidising atmospheres. Strontium-doped materials require a low partial pressure
of oxygen during sintering to become leak-tight, e.g. bar at a temperature
of 1720°C. Calcia-doped materials are better and can sinter in air at 1600°C. In
this case, especially with chromium deficiency, liquid phases appear during the
process and these help to pull the particles together. The downside is that these
liquids can soak away into surrounding porous materials, as Minh found when
co-firing his monolithic tape calendered composites [2 81.
To avoid these problems, Sulzer has used metal interconnects in their small-
scale residential SOFC heat and power unit [29]. The alloy is largely chromium,
with 5 wt% iron and 1 wt% yttria to give dispersion strengthening, made by
Plansee AG in Austria. This alloy has almost the same expansion coefficient as
YSZ and has the benefit of improved strength and toughness when compared
with lanthanum chromite. But it requires coating to prevent chromium
migration and is also expensive at the present time.
Another approach is to adopt a design similar to the lead acid battery and to
use wires brought out from the electrodes and connected externally. This is the
approach adopted by Adelan in their microtubular design. Clearly, the design of
the cells and how they fit into the overall stack is vitally important in deciding
such issues.

1.8 Cell and Stack Designs

A solid oxide fuel cell is a straightforward five-component entity as described in
Figure 1 .l. The main problem, which has been exercising engineers for the past
30 years, is that of designing cells which can be stacked to produce significant
power output. This power output is directly proportional to the cell area, so the
maximum area of YSZ membrane must be packed into the SOFC stack. This is
similar to a heat exchanger design exercise. Two plausible solutions are obvious:
a stack of flat plates or an array of parallel tubes. Typical heat exchanger
problems of joining, cracking and leakage are evident in the SOFC stacks because
of the complex materials and the high expansion coefficient. Of course the
difficulties are greater because of the temperature of operation. Additional