a solution leads to a lower crystallographic formation
energy, and forms single phase materials having com-
plex compositions. Since the chemical reactants for
sol—gel processing can be purified conveniently by
distillation and crystallization, a coating of high purity
can be fabricated by sol—gel processing. Another bene-
fit is the ability to coat complex shapes, including
parts having blind holes and corners, long tapes and
wires, curved surfaces, and the inside surfaces of cylin-
ders. Also, coating adhesion is excellent [13—15].
The present work focuses on the alkali corrosion
The solution was kept boiling until a clear solution
was formed.
2.2.3. Alum ina solution
The precursor was ((CH ) CHO) Al. First aluminium
ꢀ ꢂ
ꢀ
isopropoxide was added to deionized water which was
°
heated to about 90 C while vigorously stirring. The
°
solution was kept at 90 C for 2 h, then 0.07 mol HCl
per mol alkoxide was added to peptize the sol par-
ticles. The sol was kept boiling in an open reactor until
a clear solution was formed.
and protective mechanism of Si N ceramics coated
ꢀ ꢁ
by oxide materials. The sol—gel coating techniques
were developed for application of the corrosion-resis-
tant coatings. The corrosion behaviour of coating
materials in alkali molten-salt, and alkali-containing
atmospheres was evaluated. The effect of the alkali
molten-salt, and alkali-containing atmospheres upon
coating characteristics such as microstructure, inter-
face adhesion, chemical composition of the surface,
weight changes, strength of coated and uncoated sam-
ples, was examined.
2.2.4. Mullite solution
The precursor were ((CH ) CHO) Al and Si(OC H ) .
ꢀ ꢂ
ꢀ
ꢂ ꢃ ꢁ
First aluminium isopropoxide was added to deionized
°
water which was heated to about 90 C while vigor-
ously stirring. The solution was kept at 90 C for 2 h,
°
then Si(OC H ) was added. Finally, 0.07 mol HCl per
ꢂ ꢃ ꢁ
mol alkoxide was added to peptize the sol particles.
The sol was kept boiling in an open reactor until
a clear solution was formed.
2. Experimental procedure
2.1. Materials
2.3. Coating techniques
A single batch of commercially available Si N with
CMZP, zirconia, alumina and mullite coatings on
silicon nitride substrate were prepared by sol—gel and
dip techniques. The procedures are described sche-
matically in Fig. 1. First, the ceramic samples were
immersed in the solutions for 5 min, and were taken
ꢀ ꢁ
6 wt % Y O (PY-6) as a sintering aid was used in this
ꢂ ꢀ
study. Samples used for the measurement of corrosion
rate and strength degradation were cut into rectangu-
lar coupons, 2 mm by 3 mm by 10 mm and 2 mm by
3 mm by 50 mm, respectively.
\ꢄ
out of the solutions at a rate of 4—12 cm min by the
In order to prepare crack-free coatings and enhance
dip machine. The coated samples were kept at room
°
the adhesion of the coatings to Si N substrate, the
carrier surface was treated to form an oxide layer.
First the silicon nitride samples were washed with
temperature for 4 days, and dried at 40—60 C for
ꢀ ꢁ
2 days to form clear gel coatings. After the drying
process, the samples coated by CMZP were fired at
°
°
1200 C for 24 h, while the samples coated by zirconia,
acetone, dried at 110 C for 2 h, then immersed in 20%
°
HF for 10 min. Finally, the samples were washed with
alumina or mullite were fired at 1500 C for 10 h. The
°
deionized water, and calcined at 1200 C for 6 h.
heating schedule consisted of a heating rate of
°
\ꢄ
°
0.5 C min from room temperature to 500 C; hold-
°
ing 2 h at 200, 300, 400 and 500 C, respectively; rapid
°
°
\ꢄ
2.2. Preparation of solutions for coatings
The basic principle of the sol—gel process is to form
a solution of the elements of the desired compound in
an organic solvent, polymerize the solution to form
a gel, then dry and fire this gel to displace the organic
components to form a final oxide.
heating from 500 to 1500 C at a rate of 5 C min
,
°
and 4 h hold at 1000 C and 10—24 h hold at
°
1200—1500 C.
2.4. Therm al shock test
Five coated Si N samples for each coating and each
ꢀ ꢁ
test, approximately 2 mm by 3 mm by 10 mm, were
2.2.1. CMZP solution
First the precursors, Ca(CH CO ) H O, Mg(C H O) ,
tested for thermal shock resistance. The samples were
°
heated to 300, 500 and 1000 C, respectively, and then
ꢀ
ꢂ ꢂ ꢂ
ꢂ ꢃ
ꢂ
°
Zr(C H O) , and (C H O) P(O), were mixed in
quenched in water at 25 C. The surface microstruc-
ꢂ ꢃ
ꢁ
ꢂ ꢃ
ꢀ
stoichiometric proportions in ethyl alcohol or de-
ionized water. Then the mixture was slowly stirred
while HCl was added dropwise until pH 2 was reach-
ed. A clear CMZP solution was formed.
ture of coatings and adhesion between the coatings
and Si N substrate were examined by scanning elec-
ꢀ ꢁ
tron microscopy (SEM).
2.5. Alkali corrosion test
The alkali corrosion resistance of both coated
2.2.2. Zirconia solution
The precursors, Zr(C H O) and YCl · 6H O, were
mixed in stoichiometric proportions in ethyl alcohol,
and uncoated Si N ceramics was examined by deter-
mining the weight loss and strength degradation of
ꢂ ꢃ
ꢁ
ꢂ
ꢀ ꢁ
°
then were kept at 90 C and stirred while 1 mol water,
samples exposed to Na CO molten-salt and sodium-
°
containing atmospheres at 1000 C for 50 h. Five
ꢂ
ꢀ
and 0.1 mol HCl per mol of Zr(C H O) was added.
ꢂ ꢃ
ꢁ
4456
Li
