L. Jia et al. / Journal of Alloys and Compounds 414 (2006) 152–157
153
coefficient of anode can be managed to match with those
of other SOFC components as YSZ is mixed with Ni in an
arbitrary ratio [3,7]. The performances of anode could be
affected greatly by the quality of raw materials of NiO and
YSZ. The results of Jiang and Badwal indicated that the Y-
TZP (Ni/3 mol%Y2O3–ZrO2) in the anode cermet played an
important role in modifying the electrode microstructure and
kinetics of the fuel oxidation reaction [8]. So, the quality
of YSZ powders would have great influences on the perfor-
mance of the Ni–YSZ cermet.
The quality of YSZ powders, including the particle size
distribution and the specific surface area, could be influ-
enced greatly by pre-calcining it at different temperature.
Accordingly, the performances of the Ni–YSZ cermet could
be affected by mixing pre-calcining YSZ powders in NiO. In
this paper, the effects of the YSZ pre-calcining temperature
on Ni–YSZ cermet will be discussed, including the porosity
of open holes, sintering shrinkage characteristic, resistivity
after reducing in H2 and polarization, etc.
tivity under reducing atmosphere was measured with the dc
four-probe technique using a resistance testing system based
on Keithley 2400 SourceMeter and an AI 708 digital con-
troller. The reduction gas was hydrogen provided by a GCH
2500 high purity hydrogen generator.
Another part of the powders was pressed into a column
which diameter and length were all about 6 mm. These sam-
ples were used to investigate the shrinkage characteristics uti-
lizing a Netzsch DIL 402C/3/G dilatometer (Netzsch GmbH,
Germany) with air purging. In this measurement, the tem-
perature was raised from room temperature to 500 ◦C at a
rate of 5 K/min, and from 500 ◦C to 900 ◦C at 10 K/min, then
holding at 900 ◦C for 2 h, finally raised the temperature to
1300 ◦C at 10 K/min and holding at 1300 ◦C for 2 h.
Finally, some adhesives were added in the third part to
form slurry. The slurry was coated onto the 8YSZ electrolyte
green tapes of pellets with a diameter of 13 mm. The sam-
ples were baked at 350 ◦C to burn out the adhesives and then
co-sintered at 1300 ◦C for 10 h. In the sintering process, a
ceramic tile as small weight was placed on top of the sample
to avoid the pellet’s distortion [9]. The anodes on YSZ elec-
trolyte were reduced in H2 at 700 ◦C. Then the silver paste
(DAD-87) was painted on to the opposite side of the anode
The cathode was symmetrical with respect to the anode and
the reference electrode was painted beside the cathode. These
samples were then sealed to one end of ceramic tubes by
applying the technique of reference [10]. Three-electrode
method was used to measure the impedance spectroscopy and
polarization of anode in reducing atmosphere (H2) employed
SI 1260 Impedance/Gain-phase Analyzer and SI 1287 Elec-
trochemical Interface (Solartron Instruments, UK).
2. Experimental
The pre-calcined YSZ powders were prepared by fir-
ing commercial YSZ powders (China Building Material
Academy, Beijing, China)at900 ◦C, 1100 ◦Cand1200 ◦Cfor
2 h in air, respectively. The particle size distribution was mea-
sured with a Mastersizer 2000 laser size analyzer (Malvern
Instrument Ltd., UK) combined with a Hydro2000MU acces-
sory. Water was used as a dispersant, the pump speed was
2000 rpm and the ultrasonic set was 20.
NiO nano-powders were synthesized by a precipitation
method with Ni(NO3)2·6H2O and ammonia and finally fired
at 400 ◦C for 1 h, the average grain size of NiO was 8–14 nm.
Then the raw powders of NiO, YSZ (commercial powders
and the pre-calcined powders at different temperatures) and
amylum were mixed in agate mortar for 1 h and dried. The
contents of the materials prepared are summarized in Table 1.
Then the mixed powders were divided into three parts.
One part was pressed into pellets with a diameter of about
13 mm and a thickness of about 0.5 mm. The green tapes were
then sintered at 1400 ◦C for 2 h. These samples were used to
measure the porosity of open holes and the resistivity after
reducing in hydrogen. The method of measuring the coeffi-
cient of water absorption was used to estimate the porosity of
open holes of different samples. This method was achieved
by measuring the samples weight before and after absorbing
water using FA2004 balance (0.1 mg resolution). The resis-
3.1. Particle size distribution
Fig. 1 shows the particle size distribution curves of raw
YSZ powders and the pre-calcined ones at different temper-
atures. The curves of raw powders, 900 ◦C and 1100 ◦C pre-
calcined powders, show three peaks, they are about 0.6 m,
3 m and 20 m, respectively. During the measurement pro-
cess, the particle ratio with small and middle size is related
with the ultrasonic dispersion. The 3 m peak should be
depressed along with the ultrasonic dispersion time increas-
ing, at the same time, the 0.6 m peak rise up. It means that
the middle particles are some soft congeries that may be dis-
persed by ultrasonic. The 0.6 m peak becomes lower with
the pre-calcining temperature increased and it disappeared
for the powders that pre-calcining at 1200 ◦C. According to
the particle size distribution data, specific surface area val-
ues calculated by Malvern’s software are listed in Table 2. It
is obvious that the specific surface area decreased with the
pre-calcining temperature increasing. It should be attributed
to the aggregating and sintering of small particles.
Table 1
Prepared Ni–YSZ cermet
Abbreviation
YSZ raw powder (50 wt%)
NiO (wt%)
NZM
Commercial YSZ
50
50
50
50
NZM900
NZM1100
NZM1200
YSZ powders pre-calcined at 900 ◦C
YSZ powders pre-calcined at 1100 ◦C
YSZ powders pre-calcined at 1200 ◦C