|
The
objective of the work presented in this dissertation has
been twofold: to develop a water-based system for tape
casting and to build laminar structures from the tapes. A
set of requirements on the binder was specified and latexes
were identified as suitable for water-based tape casting,
mainly owing to their low viscosity and high polymeric
content. A systematic approach was used to evaluate
different formulations throughout the processing chain to
find possibilities and limitations. Rheological measurements
were used to characterise stability and flow behaviour of
the suspensions and to study interactions between
constituents in the suspension. The effect of the dispersant
concentration was studied. It was found that, below optimum
dosage, i.e. monolayer coverage, bridging flocculation can
occur and the dispersant can adsorb onto latex particles.
Tape casting experiments were done to investigate possible
casting rates, resulting thickness and the thickness at
which spontaneous cracking occurs. The cast tapes were then
evaluated with regard to their quality, green density and
homogeneity. It has been shown in this work that very high
solids loading (> 55 vol%) can be reached in alumina
systems thanks to efficient dispersants and latex binders
with a high polymeric content. This in combination with an
efficient drying system of the tape caster enabled high
casting rates to be reached. The lowest viscous formulations
also gave tapes thicker than 0.5 mm without cracking.
Significant differences between binders were observed in
this work. For example, in a comparison between anionic and
nonionic latexes, the nonionic latex gave a more homogeneous
packing of the green tape and higher final density. In the
anionically stabilised latexes, phase separation between
surfactant and polymer occurred, creating pore channels in
the tape.
The
work on laminar structures focused mainly on ceramic
laminates with crack deflecting ability. These types of
structures have been shown to be more damage tolerant and to
have superior thermal shock resistance. Examples of
laminates of this type are SiC/graphite and dense/porous
laminates of alumina or SiC using fugitive particles. A
technique for fabricating the latter type of laminates was
developed in this work. The composite is a laminate of
alternating porous and dense layers of the same material. To
be able to co-sinter these types of layers, it is necessary
that the added fugitive particles do not affect the
stability of the suspension, i.e. the packing of the ceramic
matrix in the layers should be the same. Hereby the dense
and porous layers sinter with the same total shrinkage, and
the fugitive particles (mainly starch particles) leave voids
too large to sinter themselves and will shrink only by the
same amount as the surrounding matrix, leaving porosity. The
advantage of such a composite is that no chemical reaction
or thermal mismatch is present between the layers.
During
casting, when the suspension passes under the casting blade,
the shearing can cause particles and polymer to be oriented
in the direction of the flow. This can be used deliberately
to enhance mechanical, electrical or thermal properties.
However, it can also cause sintering anisotropy, making
close dimensional tolerances difficult to control. The
conditions that give rise to shrinkage anisotropy were
studied. Water-based systems were shown to be on a par with
or better than organic solvent-based systems. A desirable
particle orientation benefits from a high volume fraction,
elongated or plate-like particles and a high shear rate.
This was exploited in making a laminate structure with
alternating porous and dense layers from a coarse low
sinterability powder and a fine plate-like powder for the
respective layers.
|
