Ceramics International 42 (2016) 9058–9064Contents lists available at ScienceDirectCeramics Internationaln Corr
E-m
http://d
0272-88journal homepage: www.elsevier.com/locate/ceramintInfluence of microwave-assisted hydrothermal treatment on the
properties of nickel oxide–zirconia based composites
L.B. Pinheiro a, F.N. Tabuti a, A.E. Martinelli b, F.C. Fonseca a,n
a Instituto de Pesquisas Energéticas e Nucleares, IPEN-CNEN/SP, São Paulo, SP 05508-000, Brazil
b Universidade Federal do Rio Grande do Norte, UFRN, Natal, RN 59072-970, Brazila r t i c l e i n f o
Article history:
Received 15 January 2016
Received in revised form
20 February 2016
Accepted 27 February 2016
Available online 28 February 2016
Keywords:
Ceramic composites
Coprecipitation
Microwave-assisted hydrothermal
Impedance spectroscopyesponding author.
ail address: fabiocf@usp.br (F.C. Fonseca).
x.doi.org/10.1016/j.ceramint.2016.02.165
42/& 2016 Elsevier Ltd and Techna Group S.ra b s t r a c t
The influence of a microwave-assisted hydrothermal treatment (MWH) on the properties of NiO–
ZrO2:8 mol%Y2O3–CeO2 composite was investigated. The composites were prepared by coprecipitation of
nickel and cerium salts in a suspension of ZrO2:8 mol%Y2O3 (YSZ). Simultaneous thermogravimetric and
differential thermal analysis revealed that the MWH promotes the crystallization of Ni(OH)2, as con-
firmed by X-ray diffraction. The sintering behavior of ceramic compacts was influenced by the MWH
treatment, which contributed to slightly higher final apparent densities. Impedance spectroscopy mea-
surements revealed that the electrical properties of sintered composites exhibit characteristics of mixed-
ionic–electronic conductors, a feature that was suppressed in the samples that underwent MHW
treatment. Such an effect evidences the influence of hydrothermal treatment on the final microstructure
of the composite.
& 2016 Elsevier Ltd and Techna Group S.r.l. All rights reserved.1. Introduction
Ceramic-supported Ni composites are studied for different ap-
plications in catalysis and electrochemistry. For example, Ni-
ZrO2:8 mol%Y2O3 (Ni–YSZ) cermet is the standard anode material
of solid oxide fuel cells (SOFC) having excellent properties, un-
heard by any other alternative material proposed for H2 fuelled
applications. Limitations due to low redox stability and sintering of
Ni particles at high operating temperatures are compensated by
both the high electronic conductivity and electrocatalytic activity
[1]. One of the main strategies for enhancing the properties of fuel
cell anode (or other catalytic systems [2]) is to tailor the micro-
structure by a strict control of the synthesis method. By controlling
parameters such as particle size and distribution of both the
ceramic and metallic phases can result in enhanced properties of
the cermet [3,4]. The nickel content strongly influences the elec-
trical conductivity of the cermet, which usually requires relative
volume fractions above the percolation threshold (
30 vol%). The
electrical conductivity is also dependent on the microstructure
and small particle size of Ni with high dispersion in the YSZ ske-
leton usually results in lower percolation threshold and better
redox resistance [3–6]. Thus, a wide range of synthesis techniques
has been reported for this cermet, such as, liquid mixture [6], gel-
precipitation [7] and combustion method [8]..l. All rights reserved.Aiming at applications requiring high-performance cermets
chemical synthesis routes are preferable over traditional solid-
state reaction to attain the desired Ni particle size. Particularly
the coprecipitation technique is efficient to synthesize particles
with reduced size and homogeneous size distribution [9]. Such
technique has been used for the synthesis of NiO–YSZ composites
[10–12]. Nevertheless, the main problem is to avoid the formation
of Ni complexes during the coprecipitation. However, by control-
ling the pH it is possible to obtain highly dispersed and homo-
geneous powders [11].
It is known that microwave energy can improve reaction ki-
netics, selectivity and yield of chemical reactions [13]. Therefore,
synthesis techniques based on chemical routes assisted by mi-
crowave treatments can benefit from such characteristics. Com-
bustion [14–16] and hydrothermal synthesis [17–19] are some
techniques usually modified for microwave-assisted heating. In
general, such studies showed a narrow particle size distribution
and better homogeneity in the final product in comparison with
the corresponding conventional synthesis methods. These im-
proved properties are attributed to unique thermal effects in mi-
crowave-heated materials, such as volumetric heating and inverse
temperature gradient [20].
In the present study, NiO–YSZ was synthesized with small ad-
dition of CeO2. The addition of ceria aims at enhancing catalytic
activity of the composite [21,22]. Powders were synthesized by
hydroxide coprecipitation method followed by a microwave-as-
sisted hydrothermal (MWH) treatment aiming at the preparation
of homogeneous composites. The influence of the MWH treatment
L.B. Pinheiro et al. / Ceramics International 42 (2016) 9058–9064 9059was investigated and results compared with conventional heat-
treatment.Fig. 1. Thermogravimetric curves (a) and differential thermal analyses (b) of the as-
prepared, conventional heating (CH), and microwave-assisted hydrothermal
(MWH) treated NiO–YSZ–CeO2 powders.2. Experimental
2.1. Synthesis of NiO–YSZ–CeO2 powders
NiO–YSZ–CeO2 powders with nominal composition NiO:YSZ:
CeO2¼56:34:10 (wt%) were prepared by hydroxide coprecipita-
tion method. Such nominal composition was chosen to result in
samples with final NiO content close to the theoretical percolation
threshold of NiO (
30 vol%) in order to enhance possible effects
due to the different thermal treatments studied. Nevertheless, the
final composition is within the range used for the NiO–YSZ SOFC
composite anodes and other possible applications such as catalysts
for chemical vapor deposition of carbon nanotubes [23]. The
starting materials were Ni(CH3COO)2 4H2O (Aldrich), Ce
(NO3)3 6H2O (Aldrich), ZrO2:8 mol%Y2O3 (Tosoh), and NH4OH
(Vetec) as precipitation agent. The appropriate amounts of Ni
(CH3COO)2 4H2O and Ce(NO3)3 6H2O were dissolved in distilled
water (Ni2þ ion concentration: 0.1 mol L1) under continuous
magnetic stirring to obtain a homogeneous solution. The YSZ
powder was added to the solution and the resulting suspension
was heated at 95 °C. The NH4OH:H2O (1:1 volume) was added
dropwise to the suspension until pH¼9.5 was reached, thereby
forming the precipitate.
The precipitate mixed with YSZ powder was filtered and wa-
shed with distilled water and ethanol several times for complete
removal of ions NH4þ , NO−4, and CH3COO
2 . After washing, pre-
cipitates were dried at 100 °C for 12 h. The as-prepared powder
was heat treated by two different methods: i) microwave-assisted
hydrothermal (MWH) and ii) conventional heat treatment (CH).
The microwave-assisted hydrothermal treatment was performed
by placing the material in water suspension into teflon containers.
The sealed Teflon containers were inserted into an autoclave and
heated in microwave oven (800 W, 2.45 GHz) at 130 °C for 2 h
with 10 °C min1 heating rate. The average pressure achieved in
this heating condition was 1.4 atm. The conventional heat treat-
ment (CH) of powders was carried in an oven pre-heated at 130 °C
for 2 h. The as-prepared powder along with both MWH and CH
materials were calcined in air at 500 °C for 4 h to obtain NiO–YSZ–
CeO2 powders.
2.2. Characterization of NiO–YSZ–CeO2
The thermal decomposition of the as-prepared, MHW, CH, and
calcined powders were studied by simultaneous thermo-
gravimetry (TG) and differential thermal analysis (DTA) using a
Seteram Labsys. The maximum temperature was 1200 °C with
10 °C min1 heating rate in synthetic air flow (50 mL min1). The
chemical compositions of the calcined samples were determined
through energy dispersive X-ray spectroscopy (EDX). The linear
shrinkage of pellets (7 mm diameter) uniaxially pressed (4 MPa)
was measured using a Setaram Labsys dilatometer up to 1400 °C
with 10 °C min1 heating rate in air flow (50 mL min1). X-ray
powder diffraction (XRD) analyses were carried out for phase
identification in a Rigaku-Miniflex II operating at 30 kV, 15 mA,
0.05° 2θ° step and 2 s counting time, using Cukα1 radiation
(λ¼1.5406 Å). The crystallite sizes were calculated using Scherrer
formula [24,25].
The electrical properties of pellets sintered at 1200 °C for 3 h
were studied by impedance spectroscopy (IS) using a Solartron
1260 frequency analyzer in the 1 Hz–30 MHz frequency range
with 200 mV ac amplitude (zero bias dc). Such an amplitude was
found to be within the ohmic range for such measurements. The ISmeasurements were carried out in static ambient air in the 100–
700 °C temperature range, allowing the system stabilize at each
measuring temperature for at least 30 min. For IS measurements,
silver contacts were deposited onto the parallel surfaces of the
samples (typically, 
8 mm diameter and 1 mm thickness) and
cured at 600 °C for 1 h [26].
The morphology of fractured surfaces of sintered samples was
investigated by field emission gun scanning electron microscopy
(FEG-SEM) using a FEI Inspect F50 microscope operating at 5 kV.3. Results and discussion
Thermogravimetric (TG) and differential thermal analysis (DTA)
curves for the as-prepared, CH, and MWH powders are shown in
Fig. 1. The as-prepared and CH samples exhibited three main mass
loss events up to 1200 °C taking place at similar temperatures
T
100, 280, and 390 °C. The MWH sample showed two mass loss
events at T
100 and 305 °C.
The initial mass loss at T
100 °C is attributed to water des-
orbed from the surface of the powders and corresponds to 
6%,
5%, and 3% for the as-prepared, CH, and MWH samples, respec-
tively. The total mass loss depends on the thermal history of the
powders and it was found to be 24%, 20%, and 15% for the as-
prepared, CH, and MWH samples, respectively. Nevertheless, the
main differences observed in thermal decomposition of the sam-
ples arise when the TG/DTA curves are analyzed in the tempera-
ture range of the main mass loss events (Fig. 1b). The mass loss is
9060 L.B. Pinheiro et al. / Ceramics International 42 (2016) 9058–9064
Fig. 2. X-ray diffraction patterns of synthesized (a) and calcined at 500 °C
(b) powders of NiO–YSZ–CeO2.
Table 2
Crystallite sizes and lattice parameters (a) obtained for calcinated powders, green
(ρ0), apparent densities (ρ), and fraction of the theoretical density (ρ/ρT).
as-prepared CH MWH
Crystallite sizes (nm)
NiO 8.7 7.7 9.4
YSZ 19.7 20 19.4associated with at least two convoluted exothermic peaks and one
endothermic, as shown in Fig. 1b. The convoluted exothermic peak
is attributed to the crystallization of ceria (
260 °C) [27] and
nickel hydroxide (
280 °C). Such attributions were confirmed by
performing separated thermal analyses of both NiO and CeO2 co-
precipitated precursors (not shown). The exothermic peaks are
more pronounced in both the as-prepared and CH samples,
whereas for the MWH sample the exothermic events were es-
sentially suppressed. The absence of exothermic peaks indicates
the presence of crystalline nickel hydroxide in the MWH powders.
The endothermic peak at T¼305 °C in DTA curve is possibly re-
lated to dehydroxylation of nickel hydroxide to nickel oxide [28]
(Ni(OH)2-NiOþH2O), which is likely to be the main event asso-
ciated with the second mass loss. Such a mass loss (
10%) was
similar for all samples, indicating that it is associated with nickel
phase. The as-prepared and CH samples showed a third mass loss
region (
390 °C) of 
2%, occurring after nickel hydroxide thermal
decomposition that is probably due to the removal of chemically
adsorbed water and nickel oxide hydration water.
The powders were calcined at 500 °C for 4 h and semi quanti-
tative chemical analyses were carried out by EDX. Table 1 sum-
marizes the phase compositions in oxide phases calculated from
EDX results. HfO2 is an impurity usually present in zirconia pow-
ders and was found as a minor phase. The phase compositions
were similar for the three samples, with average values being NiO:
YSZ:CeO2¼23:69:7 (wt%). Such composition corresponds to

27 vol% of NiO, which is close to the percolation threshold of the
semiconducting phase. However, such values are considerably
below the nominal ones due to losses of Ni during the coprecipi-
tation [9,11].
The influence of thermal treatments were on the NiO–YSZ–
CeO2 composites further analyzed by X-ray diffraction (XRD), as
shown in the Fig. 2.
The XRD data were indexed by comparison with standard files
hexagonal β-Ni(OH)2 (JCPDS 14-117), cubic YSZ (JCPDS 82-1246),
cubic CeO2 (JCPDS 34-394), and cubic NiO (JCPDS 47-1049). All the
observed peaks corresponded to the indexed phases and no
spurious phases were detected by the XRD. In Fig. 2a, the dif-
fraction peaks of β-Ni(OH)2 are clearly observed in the MWH
sample, while for both the as-prepared and CH powders the most
intense Ni(OH)2 peak (101) at 2θ
39° is not visible. Such a result
adds evidence to the relation between the suppression of the
exothermic peak occurring at 280 °C (Fig. 1b) and the presence of
crystalline nickel hydroxide in the MWH sample.
After calcining step at 500 °C the β-Ni(OH)2 was converted to
NiO and all samples exhibited similar XRD patterns (Fig. 2b). The
crystallite sizes of the calcined powders were calculated using the
Debye–Scherrer formula and the selected peaks corresponding to
the maximum relative intensity of each phase: the (311) reflection
at 2θ
59° for NiO, and the (111) reflection at 2θ
30° for YSZ.
Such analysis was not possible for CeO2 since it has a relative small
concentration and possibly very small crystallite size, as inferred
from the low intensity and broad diffraction peak at 2θ
28°
(Fig. 2b). The crystallite sizes, listed in Table 2, are 
9 nm and
20 nm for NiO and YSZ, respectively. Such values are in good
agreement with previously reported ones for similar calciningTable 1
Phase composition (wt%) calculated from EDX analysis.
As-prepared CH MWH
NiO 23.1 23.1 22.4
YSZ 68.9 68.9 69.6
CeO2 6.8 6.9 6.9
HfO2 1.2 1.1 1.1temperature [29]. It is interesting to note that the MWH samples
display the largest crystallite size of NiO, a feature possibly related
to the earlier crystallization of the precursor phase β-Ni(OH)2 as
compared to the other samples. The lattice parameters (a) were
calculated and shown in Table 2. Both YSZ and NiO were found to
have lattice parameters in good accordance with values of the
corresponding JCPDS files.a (Å)
NiO 4.182(1) 4.181(1) 4.194(1)
YSZ 5.142(1) 5.142(1) 5.143(1)
Density (g cm−3)
TMA ρ 5.77 5.67 5.84
(1400 °C) ρ0 2.83 – 2.71
IS ρ 5.04(6) 5.04(5) 5.19(9)
(1200 °C) ρ/ρT (%) 82.2(9) 81.8(8) 84.8(9)
L.B. Pinheiro et al. / Ceramics International 42 (2016) 9058–9064 9061
Fig. 3. Linear shrinkage of NiO–YSZ–CeO2 compacts.The sintering behavior of ceramic compacts was studied by
dilatometry analysis, as shown in Fig. 3. The total retraction ob-
served for compacts prepared with calcined powders were 19%,
22%, and 23% for the as-prepared, CH, and MWH, respectively
(Fig. 3a). Such total retractions are in agreement with the green
density (ρ0) values shown in Table 2. From room temperature up
to 
1100 °C samples display negligible retraction. Both the onset
of the shrinkage (
1050 °C) and the maximum densification rate
temperature (
1280 °C) are weakly dependent on the heat
treatment. However, at the maximum measuring temperature the
as-prepared sample has not attained the final sintering stages,
whereas both CH and MWH samples display a flattening of the
retraction curve suggesting the end of the sintering process at
T
1400 °C. The final apparent densities of samples were similar,
but the MWH showed slightly higher values than the other sam-
ples (Table 2).
The effect of the different thermal treatments on the properties
of the NiO–YSZ–CeO2 ceramic compacts sintered at 1200 °C was
studied. The apparent densities of the samples sintered at 1200 °C
(Table 2) are similar, but the MWH sample displayed slightly
higher values, in agreement with the dilatometry (Fig. 3). Phase
characterization of sintered samples by XRD is shown in Fig. 4. The
XRD data of sintered samples exhibit only the peaks correspond-
ing to both the NiO and YSZ. Differently from the calcined samples,
the high sintering temperature promoted the formation of theFig. 4. X-ray diffraction patterns of NiO–YSZ–CeO2 samples sintered at 1200 °C.solid solution YSZ:CeO2 [30]. The calculated lattice parameters (a)
for YSZ:CeO2 phase are aYSZ:Ce¼5.169(2) Å and 5.173(2) Å for as-
prepared and MWH samples, respectively. Cerium has larger ionic
radius than zirconium promoting an increase of the lattice para-
meter [31]. Thus, lattice parameter of sintered samples are higher
than the ones calculated for calcined samples (Table 2), because
the solid solution between YSZ and CeO2 at high temperature. By
using the Vegard's law [32] and the measured mass fraction of
CeO2 by EDX (corresponding to 
7 mol%) it is possible to calculate
the lattice parameter of the cubic solid solution aYSZ:Ce¼5.168 Å.
Such value is excellent agreement to the one calculated by the XRD
and indicates that ceria formed a complete solid solution with YSZ.
On the other hand, NiO exhibits lattice parameters (aNiO¼4.183 Å)
similar to those in Table 2.
Fractured surfaces of the ceramics were observed in a FEG-SEM
microscope (Fig. 5). Secondary electron images in Fig. 5 display
phase contrast, being NiO particles darker than YSZ:CeO2 ones, as
confirmed by EDS analyses (not shown). Lowmagnification images
(Fig. 5a and c) reveal an irregular microstructure with a good
dispersion of NiO and a relatively high porosity, which was ex-
pected considering the calculated ρ/ρT
82% for the compacts
sintered at 1200 °C (Table 2). Both samples exhibit similar micro-
structure composed of denser regions, with larger concentration of
brighter YSZ:CeO2 grains, separated by more porous regions in
which NiO particles are predominant (Fig. 5a and c). Higher
magnification images (Fig. 5b and d) reveal that grain sizes are in
the submicrometric range. NiO particles have a faceted shape,
whereas YSZ:CeO2 exhibit a more rounded form. Nevertheless, it is
possible to compare in Fig. 5b and d that average grain size in the
as-prepared samples is considerably smaller than that of the MWH
sample, a feature more pronounced for the darker NiO grains. A
rough estimate of grain size during FEG-SEM analysis indicated
typical NiO grain sizes of 
450 nm for the as-prepared sample,
while MHW sample have grains with diameter 
750 nm. Such a
difference is likely to arise from the microwave heat treatment
during powder synthesis and it is in agreement with the formation
of crystalline Ni phases at early stages of the synthesis promoted
by the MHW, as shown in Figs. 1 and 2. Moreover, the larger grain
size of the MWH is in accordance with the higher crystallite size
calculated from XRD data (Table 2).
The observed microstructural features are reflected on the
electrical properties of the composites. The IS diagrams measured
between 300 °C and 700 °C are shown in Fig. 6. For To500 °C
(Fig. 6a and b), the IS diagrams exhibit a well-defined high fre-
quency semicircle (relaxation frequency f04100 kHz) and a low
frequency component in which at least two components could be
discerned: one in the frequency range 100 Hzo fo10 kHz and
another at lower frequencies (fo100 Hz). However, a direct at-
tribution to each IS component is not straightforward due to the
complex microstructure of composite samples that exhibit differ-
ent charge carriers and a rather complex distributions of both
grain size and pores. Thus, analyses of the IS data were focused on
the conductivity of the two components that could be more easily
separated. For To500 °C it was possible to separate the high and
low frequency components using a series-connected parallel
R-CPE equivalent circuit model [24,25,33]. At higher measuring
temperature (4500 °C), only the total electrical resistance was
determined at the intercept of the low frequency data with the
real axis. With increasing measuring temperature (Fig. 6c) the high
frequency semicircle rapidly decreases and the total resistance is
dominated by the low frequency components, as shown in the
impedance diagram measured at TZ500 °C (Fig. 6c and d). For
both the as-prepared and CH samples the magnitude of the elec-
trical resistance associated with each frequency component were
very similar in the whole temperature range. Interestingly, the
MWH sample exhibited a different behavior. For To500 °C the
9062 L.B. Pinheiro et al. / Ceramics International 42 (2016) 9058–9064
Fig. 5. FEG-SEM images of fractured surfaces of as-prepared (a, b) and MWH (c, d) samples.MWH showed the highest electrical resistance, being all the fre-
quency components significantly more resistive than the corre-
sponding ones of the other samples. Nevertheless, with increasing
measuring temperature, the MWH sample becomes less resistive
and at T
500 °C (Fig. 6c) all samples have comparable IS
diagrams.
Apparently, the low frequency components are responsible for
the different electrical behavior of such mixed ionic–electronic
composites [32,34]. Therefore, IS diagrams were analyzed to
compute the respective Arrhenius plots. Fig. 7 shows the Arrhenius
plots of the electrical resistivity of both high (Fig. 7a) and low
(Fig. 7b) frequency components at low To500 °C, and the total
resistance (Fig. 7c). Analyzing the Arrhenius plots revealed im-
portant features concerning the electrical properties of the
composites.
At low temperatures (To500 °C) the MHW sample has the
lowest electrical conductivity, being the low frequency component
the main resistive contribution to the total electrical resistivity. At
To500 °C all components exhibited thermally activated transport
and the calculated activation energies Ea are displayed on Table 3.
Such a difference diminishes with increasing measuring tem-
perature due to the higher Ea of the MWH sample. For TZ500 °C
all samples have comparable electrical conductivity. The high
frequency components (Fig. 7a) exhibited calculated Ea (Table 3)
that approach the reported values for oxygen ion conduction in
YSZ (
1 eV). However, the low frequency component (Fig. 7b) has
a marked lower Ea for both as-prepared and CH samples
(
0.75 eV), whereas the MHW sample has similar Ea
1 eV irre-
spectively the frequency range. The observed difference on the
activation energy values is a strong evidence that both as-prepared
and CH samples have a more mixed ionic–electronic contribution
on the low frequency component of the IS diagrams. Such a feature
is associated with the contribution of NiO to the electrical trans-
port that decreases the activation energy and increases the con-
ductivity at low temperature [33]. On the other hand, the MHWtreatment seems to suppress such a behavior and the low fre-
quency component exhibits considerably higher Ea, compatible
with YSZ-based electrolytes. Such an absence of mixed ionic–
electronic contribution indicates that electrical properties of the
MWH sample is significantly less affected by the electronic con-
ducting phase (NiO), probably due to the different grain size dis-
tribution evidenced on Fig. 5.
The Arrhenius plot of the total electrical resistivity (Fig. 7c)
reveals that the MHW sample is the only to exhibit a single Ar-
rhenius-like behavior in the entire temperature range, whereas
both the as-prepared and CH samples exhibited a deviation from
the Arrhenius behavior at T
450 °C. The MHW sample showed a
linear behavior in the entire temperature range investigated with
calculated Ea
1 eV, in perfect agreement with the expected be-
havior of YSZ-based oxygen conductors [1,24,25]. On the other
hand, both the as-prepared and the CH samples exhibited two
thermally activated processes. For To450 °C, as-prepared and CH
samples have Ea
0.73 eV, a value comparable to the ones pre-
viously reported for YSZ–NiO mixed ionic–electronic composites
[33]. Such relatively low Ea is a strong indication that NiO perco-
lated in the YSZ:CeO2 matrix at relatively low volume fraction
(
27 vol%) [5,33]. For TZ500 °C, all samples exhibited very si-
milar magnitude and temperature dependence of the total elec-
trical conductivity with Ea
1 eV. In this T range, the ionic con-
duction dominates the charge transport and the total electrical
conductivity of the MWH sample was slightly higher than the
others, possibly due to the higher apparent density.
Differences of the electrical properties revealed by IS data are
probably associated with microstructural features resulting from
the MWH treatment. The MHW sample exhibits an Arrhenius-like
behavior compatible with YSZ-based oxides, while the other
samples display a temperature dependence of the electrical con-
ductivity that indicates a mixed-conducting behavior. Similar be-
havior was previously observed as a function of NiO content [34],
i.e., samples with low relative volume fraction of NiO exhibit a
L.B. Pinheiro et al. / Ceramics International 42 (2016) 9058–9064 9063
Fig. 6. Impedance diagrams measured at 300 °C, 400 °C, 500 °C, and 700 °C for
sintered samples (1200 °C) using the as-prepared, conventional-heated (CH) and
microwave-assisted hydrothermal (MWH) powders. Numbers indicate the fre-
quency decimal logarithm.
Fig. 7. Arrhenius plots for the high (a), low (b) frequency components and total
(c) electrical conductivity. Straight lines correspond to the best linear fitting.linear Arrhenius and increasing the volume fraction of NiO results
in lower Ea and a deviation of the Arrhenius behavior. The present
results indicate that the chemical synthesis method resulted in a
good dispersion of phases that allowed the percolation of the NiO.
Taking into account that present samples have volume fraction
(
27 vol% NiO) close to the percolation threshold of the electronic
phase, it is reasonable to expect that minor microstructural
changes, such as different average grain size, can disrupt the
percolation path of the phases in the composite [5]. The IS and
FEG-MEV analyses showed that both as-prepared and CH samples
9064 L.B. Pinheiro et al. / Ceramics International 42 (2016) 9058–9064
Table 3
Activation energies (Ea) calculated from the Arrhenius plots.
To500 °C (Fig. 7a and b) Ea total (Fig. 7c)
High frequency Low frequency To500 °C T4500 °C
As-prepared 0.85 (5) 0.75(6) 0.73(3) 0.97(1)
CH 0.87(10) 0.76(8) 0.74(2) 0.98(9)
MWH 1.00(13) 1.00(6) 1.03(8) 1.03(9)have relative average grain size of both phases (YSZ:CeO2 and NiO)
that favor the percolation of the semiconducting phase, a feature
more clear at low temperature. On the other hand, the MWH
treatment resulted in the crystallization of Ni phases at early
stages of the synthesis, which promotes grain growth as compared
to the other samples. Such a grain growth in MWH samples, in-
ferred from both XRD and FEG-MEV analyses, decreases the dis-
persion of NiO in the YSZ:CeO2 matrix and hinders the percolation
path of the phases with marked consequences to the electrical
properties of the composite.4. Conclusion
The effects of a microwave hydrothermal (MWH) treatment
during synthesis of NiO–YSZ–CeO2 composites were investigated
and compared to a conventional heat treatment. Thermal analysis
and X-ray diffraction indicated that the microwave-assisted hy-
drothermal treatment favors the formation of crystalline β-nickel
hydroxide at low temperature. Such Ni-phase crystallizing at early
stages of the ceramic processing is likely to result in larger NiO
grain on sintered samples. Therefore, the electrical properties of
the composites showed that the microwave-treated sample has an
Arrhenius-like behavior characteristic of YSZ-ionic conductors,
whereas the remaining samples have a more clear contribution of
the semiconducting NiO. Such difference was attributed to the
disruption of the percolating path of NiO due to larger grain size
promoted by the microwave treatment. The presented results
evidence the importance of a detailed control of the processing of
mixed ionic–electronic composites.Acknowledgments
Thanks are due to the partial funding of, CNEN, FAPESP (2013/
26961-7, 2014/50279-4, and 2014/09087-4), and CNPq (554576/
2010-4). FCF, AEM, and LBP are thankful for CNPq Scholarships FCF
(300961/2013-8).References
[1] N.Q. Minh, Ceramic fuel-cells, J. Am. Ceram. Soc. 76 (1993) 563–588.
[2] P. Serp, M. Corrias, P. Kalck, Carbon nanotubes and nanofibers in catalysis,
Appl. Catal. A 253 (2003) 337–358.
[3] B.S. Prakash, S.S. Kumar, S.T. Aruna, Properties and development of Ni/YSZ as
an anode material in solid oxide fuel cell: a review, Renew. Sust. Energy Rev.
36 (2014) 149–179.
[4] M.S. Khan, S.-B. Lee, R.-H. Song, J.-W. Lee, T.-H. Lim, S.-J. Park, Fundamental
mechanisms involved in the degradation of nickel–yttria stabilized zirconia
(Ni–YSZ) anode during solid oxide fuel cells operation: a review, Ceram. Int. 42
(2016) 35–48.
[5] D.S. Mclachlan, M. Blaszkiewicz, R.E. Newnham, Electrical-resistivity ofcomposites, J. Am. Ceram. Soc. 73 (1990) 2187–2203.
[6] V Esposito, D.Z. De Florio, F.C. Fonseca, E.N.S. Muccillo, R. Muccillo, E. Traversa,
Electrical properties of YSZ/NiO composites prepared by a liquid mixture
technique, J. Eur. Ceram. Soc. 25 (2005) 2637–2641.
[7] M. Marinsek, K. Zupan, J. Macek, Preparation of Ni–YSZ composite materials
for solid oxide fuel cell anodes by gel-precipitation method, J. Power Sources
86 (2000) 383–389.
[8] A. Ringuedé, D. Bronine, J.R. Frade, Assessment of Ni/YSZ anodes prepared by
combustion synthesis, Solid State Ion. 146 (3–4) (2002) 219–224.
[9] C. Ding, K. Sato, J. Mizusaki, T. Hashida, A comparative study of NiO–
Ce0.9Gd0.1O1.95 nanocomposite powders synthesized by hydroxide and oxalate
coprecipitation methods, Ceram. Int. 38 (2012) 85–92.
[10] M.D. Liang, B. Yu, M.F. Wen, Y.C. Zhai, Preparation of nanosized NiO/YSZ
composite powders by situ-precipitation method, Rare Metal. Mater. Eng. 38
(2009) 443–446.
[11] W.K. Yoshito, V. Ussui, D.R.R. Lazar, J.O.A. Paschoal, Synthesis and character-
ization of NiO–8YSZ powders by coprecipitation route, Mater. Sci. Forum 498–
499 (2005) 612–617.
[12] C.M. Grgicak, R.G. Green, W.F. Du, J.B. Giorgi, Synthesis and characterization of
NiO–YSZ anode materials: precipitation, calcination, and the effects on sin-
tering, J. Am. Ceram. Soc. 88 (2005) 3081–3087.
[13] G.A. Tompsett, W.C. Conner, K.S. Yngvesson, Microwave synthesis of nano-
porous materials, ChemPhysChem 7 (2006) 296–319.
[14] H. Mohebbi, T. Ebadzadeth, F.A. Hesari, Synthesis of nano-crystalline NiO–YSZ
by microwave-assisted combustion synthesis, Powder Technol. 188 (2009)
183–186.
[15] K. Tahmasebi, M.H. Paydar, Microwave assisted solution combustion synthesis
of alumina-zirconia, ZTA, nanocomposite powder, J. Alloy. Compd. 509 (2011)
1192–1196.
[16] B. Vaidhyanathan, D.K. Agrawal, R. Roy, Novel synthesis of nitride powders by
microwave-assisted combustion, J. Mater. Res. 15 (2000) 974–981.
[17] G.J. Choi, H.S. Kim, Y.S. Cho, BaTiO3 particles prepared by microwave-assisted
hydrothermal reaction using titanium acylate precursors, Mater. Lett. 41
(1999) 122–127.
[18] K.H. Kim, H.C. Park, S.D. Lee, W.J. Hwa, S.-S. Hong, G.-D. Lee, S.S. Park, Pre-
paration of submicron nickel powders by microwave-assisted hydrothermal
method, Mater. Chem. Phys. 92 (2005) 234–239.
[19] Y.B. Khollam, A.J. Deshpande, H.S. Patil, S.B. Deshpande, S.K. Date, Synthesis of
yttria stabilized cubic zirconia (YSZ) powders by microwave-hydrothermal
route, Mater. Chem. Phys. 71 (2001) 235–241.
[20] E.T. Thostenson, T.-W. Chow, Microwave processing: fundamentals and ap-
plications, Compos. Part A 30 (1999) 1055–1071.
[21] J. Feng, J. Qiao, W. Sun, P. Yang, H. Li, Z. Wang, K. Sun, Characteristic and
preparation of Ce0.5Zr0.5O2 as the anode support for solid oxide fuel cells by
phase inversion technology, Int. J. Hydrog. Energy 40 (2015) 12784–12789.
[22] M.G. Zimicz, D.G. Lamas, S.A. Larrondo, Ce0.9Zr0.1O2 nanocatalyst: influence of
synthesis conditions in the reducibility and catalytic activity, Catal. Commun.
15 (2011) 68–73.
[23] A.S. Ferlauto, D.Z. de Florio, F.C. Fonseca, V. Esposito, R. Muccillo, E. Traversa, L.
O. Ladeira, Chemical vapor deposition of multi-walled carbon nanotubes from
nickle/yttria-stabilized zirconia catalysts, Appl. Phys. A 84 (2006) 271–276.
[24] P. Scherrer, P. Göttinger Nachrichten, Math. Phys. 2 (1918) 98–100.
[25] F.C. Fonseca, R. Muccillo, Impedance spectroscopy of (yttria-stabilized zirco-
nia)-magnesia ceramic composites, Solid State Ion. 131 (2000) 301–309.
[26] F.C. Fonseca, R. Muccillo, Impedance spectroscopy analysis of percolation in
(yttria-stabilized zirconia)–yttria ceramic composites, Solid State Ion. 166
(2004) 157–165.
[27] B. Djuricic, S. Pickering, Nanostructured cerium oxide: preparation and
properties of weakly-agglomerated powders, J. Eur. Ceram. Soc. 19 (1999)
1925–1934.
[28] T. Subbaiah, R. Mohapatra, S. Mallick, K.G. Misra, P. Singh, R.P. Das, Char-
acterisation of nickel hydroxide precipitated from solutions containing Ni2þ
complexing agents, Hydrometallurgy 68 (2003) 151–157.
[29] K.W. Schlichting, N.P. Padture, P.G. Klemens, Thermal conductivity of dense
and porous yttria-stabilized zirconia, Journal. Mater. Sci. 36 (2001) 3003–3010.
[30] Y. Hinatsu, T. Muromura, Phase relations in the systems ZrO2–Y2O3–Nd2O3 and
ZrO2–Y2O3–CeO2, Mater. Res. Bull. 21 (1986) 1343–1349.
[31] R.D. Shannon, Revised effective ionic radii and systematic studies of intera-
tomic distances in halides and chalcogenides, Acta Cryst. A 32 (1976) 751–767.
[32] C. Wang, Y. Wang, W. Huang, B. Zou, Z.S. Khan, Y. Zhao, J. Yang, X. Cao, In-
fluence of CeO2 addition on crystal growth behavior of CeO2–Y2O3–ZrO2 solid
solution, Ceram. Int. 38 (2012) 2087–2094.
[33] F. Teocoli, D.W. Ni, S. Sanna, K. Thyden, F.C. Fonseca, V. Esposito, Fast mass
interdiffusion in ceria/alumina composite, J. Mater. Chem. A 3 (2015)
17135–17143.
[34] F.C. Fonseca, D.Z. De Florio, V. Esposito, E. Traversa, E.N.S. Muccillo, R. Muccillo,
Mixed ionic-electronic YSZ/Ni composite for SOFC anodes with high electrical
conductivity, J. Electrochem. Soc. 153 (2006) A345–A360.