S.K. Tadokoro, E.N.S. Muccillo / Journal of Alloys and Compounds 344 (2002) 186–189
187
in an agate mortar. During the experiments, 26 variables
were controlled to ensure reproducibility. These variables
have been selected from previous works [10–13]. How-
ever, most of the combinations of these variables found in
the literature do result in powders containing a significant
amount of hard agglomerates. These hard agglomerates are
responsible for the poor densification of ceramic compacts.
This is the reason for the additional step (ultrasonic
dispersion, milling or hydrothermal treatment) used in
most works where solid solutions are prepared by the
coprecipitation technique. After many combinations of the
variables involved in the precipitation process, we have
found best results for: low-concentrations of the cation (0.1
g l21) and precipitant (1 g l21) solutions; precipitation pH
greater than 9; a suitable washing cycle with ammonium
solution–absolute ethanol–isopropyl alcohol of 4:2:2. The
azeotropic distillation was quite important to remove
residual water from the gel. The size of the prepared batch
is also important as many of these variables are best
controlled employing low-size batches.
Fig. 1. Pore size distribution curve obtained by the BJH method after
calcining the powder.
method assuming a cylindrical pore model is shown in Fig.
1. A monomodal distribution of pore sizes can be seen in
this curve. In general, zirconia-based solid solutions pres-
ent a bimodal distribution of pore sizes assigned to intra-
and interagglomerate pores. The latter are hardly removed
and usually cause an increase in time and/or temperature
required for complete densification of these ceramics. The
present result is attributed to the use of optimized parame-
ters during synthesis and to the complete elimination of
residual water from the precipitate by azeotropic distilla-
tion.
The linear shrinkage of a powder compact and its
derivative are shown in Fig. 2. Total shrinkage is 27% and
occurs in a temperature interval of 550–1255 8C. The
maximum shrinkage rate occurs at 1075 8C.
Fig. 3 is a representative micrograph of the fracture
surface of a sintered specimen (sintering temperature,
1170 8C). The main characteristic features are low-porosity
and grains with uniform shape and similar average size.
Sintering of the compacted powders at temperatures
between 1170 and 1200 8C resulted in high densification.
The specimen sintered at 1200 8C for 5 h, for example,
Cylindrical specimens were prepared by uniaxial and
isostatic pressing. The sintering was carried out in air for 5
h at selected temperatures.
The specific surface area of calcined powders was
obtained by the BET (Brunauer, Emmett, Teller) method
(ASAP 2010, Micromeritics). The BJH (Barret, Joyner,
Halenda) method was used for obtaining the pore size
distribution. The linear shrinkage (DIL 402 E/7, Netzsch)
of powder compacts was followed up to 1500 8C with
heating and cooling rates of 8 8C and 15 8C min21
,
respectively.
Fractured or polished and thermally etched surfaces of
sintered specimens were observed in a scanning electron
microscope (LEO 440I, Oxford). The grain size distribu-
tion was calculated over |1000 grains by a home made
program based on the Saltikov statistical analysis. Raman
spectroscopy (Renishaw Raman Microscope System 3000
coupled to an Olympus BH-2 microscope and to a CCD
detector cooled by Peltier) was used to characterize the
phase composition in sintered specimens. An Ar1 laser
(Omnichrome, model 170) with an exciting radiation of
514.5 nm was used.
Electrical resistivity measurements were carried out in a
HP4192A impedance analyzer between 5 Hz and 13 MHz
in the 280–540 8C temperature range. The results were
analyzed in the impedance mode, using a special software
[14]. Silver paste was used as electrode material.
3. Results and discussion
The value of specific surface area determined by the
BET method is 132.5 m2 g21 after calcining at 500 8C for
1 h. This calcination profile was selected from thermal
analysis results.
The pore size distribution determined by the BJH
Fig. 2. Curves of linear shrinkage and its derivative for powder compacts.