Improving NO2 sensitivity by adding WO3 during processing of NiO sensing-electrode of mixed-potential-type zirconia-based sensor

Article abstract of DOI:10.1149/1.2747532

A planar N O2 sensor was fabricated by using an yttria-stabilized zirconia plate and NiO-based sensing electrode (SE). The improvement in sensing characteristic of mixed-potential-type planar N O2 sensor was examined by the addition of various metal oxides to NiO-SE. Among the metal-oxide additives tested, W O3 was found to give a significant enhancement in N O2 sensitivity. In addition, this enhancement was maximum when the initial W O3 content in the NiO-based paste before sintering was 10 wt %. The sensitivity (Δ electromotive force) to 50 ppm N O2 was as high as about 70 mV for the sensor using the 10 wt % W O3 -added NiO-SE at 800°C under the wet condition (containing 5 vol % H2 O). The sensitivity of this sensor was still 8 mV even down to 1 ppm N O2 at 800°C. The N O2 sensitivity was found to be invariant to different concentrations of H2 O and C O2 in the examined range of 5-15 and 5-20 vol %, respectively. It is speculated that the modification of morphology of the NiO-SE layer by the addition of W O3 is the main contributing factor for the improved N O2 sensitivity of the sensor.

Full text of DOI:10.1149/1.2747532

J246
Journal of The Electrochemical Society, 154 ͑8͒ J246-J252 ͑2007͒
0013-4651/2007/154͑8͒/J246/7/$20.00 © The Electrochemical Society
Improving NO2 Sensitivity by Adding WO3 during Processing
of NiO Sensing-Electrode of Mixed-Potential-Type
Zirconia-Based Sensor
,z
Norio Miura,a, Jian Wang,a Perumal Elumalai,a Taro Ueda,b Daisuke Terada,a
*
and Masaharu Haseic
aArt, Science and Technology Center for Cooperative Research, and bInterdisciplinary Graduate School of
Engineering Sciences, Kyushu University, Kasuga-shi, Fukuoka 816-8580, Japan
cRiken Corporation, Research and Development Division, Kumagaya-shi, Saitama 360-8522, Japan
A planar NO2 sensor was fabricated by using an yttria-stabilized zirconia plate and NiO-based sensing electrode ͑SE͒. The
improvement in sensing characteristic of mixed-potential-type planar NO2 sensor was examined by the addition of various metal
oxides to NiO-SE. Among the metal-oxide additives tested, WO3 was found to give a significant enhancement in NO2 sensitivity.
In addition, this enhancement was maximum when the initial WO3 content in the NiO-based paste before sintering was 10 wt %.
The sensitivity ͑⌬ electromotive force͒ to 50 ppm NO2 was as high as about 70 mV for the sensor using the 10 wt % WO3-added
NiO-SE at 800°C under the wet condition ͑containing 5 vol % H2O͒. The sensitivity of this sensor was still 8 mV even down to
1 ppm NO2 at 800°C. The NO2 sensitivity was found to be invariant to different concentrations of H2O and CO2 in the examined
range of 5–15 and 5–20 vol %, respectively. It is speculated that the modification of morphology of the NiO-SE layer by the
addition of WO3 is the main contributing factor for the improved NO2 sensitivity of the sensor.
© 2007 The Electrochemical Society. ͓DOI: 10.1149/1.2747532͔ All rights reserved.
Manuscript submitted March 28, 2007; revised manuscript received April 16, 2007. Available electronically June 25, 2007.
In order to reduce exhaust emissions and to improve fuel effi-
here the selection process of the oxide additive and the sensing
characteristics of mixed-potential-type zirconia-based NO2 sensor
using the WO3-added NiO-SE.
ciency, the automotive industries have recently introduced ultralean-
burn gasoline vehicles into the market.1,2 However, in the newly
introduced ultralean-burn gasoline vehicles, it is necessary to use
additionally a new NOx-storage catalyst. This is because NOx emit-
ted from these vehicles cannot be decomposed sufficiently into ni-
trogen under lean-burn ͑air-rich͒ conditions by a conventional three-
way catalyst. Moreover, because the NOx storage ability of the new
catalyst reaches saturation quickly, the catalyst has to be regenerated
by flowing fuel-rich gas periodically to decompose the absorbed
NOx. In this regard, in order to detect regenerating timing and to
monitor the catalyst performance, it is highly required to develop
high-performance NOx sensor. Such a high-performance solid-state
NOx sensor can also be used for an advanced on-board diagnosis
system which can inform a driver directly about the state of emis-
sion of his/her car, in the future.
So far, lots of potentiometric, amperometric, and impedancemet-
ric NOx sensors based on solid electrolyte have been examined and
reported.3 Among these NOx sensors, the mixed-potential-type NOx
sensors based on yttria-stabilized zirconia ͑YSZ͒ have been exam-
ined intensively, aiming at detection and monitoring of automotive
exhausts.4-41 The operating temperature of mixed-potential-type sen-
sors was usually lower than 700°C. This is because operation in
excess of 700°C usually results in poor NOx sensitivity.
Experimental
Fabrication of sensor device
.— YSZ
plates
͑8 wt %
Y2O3-doped, 10 ϫ 10 mm, 0.2 mm thickness͒ were used for the
fabrication of the planar sensor. Commercial NiO powder ͑Kojundo
Chemical Lab. Co., Ltd., 99.97% purity͒ was thoroughly mixed with
an appropriate amount of each of various oxide additives in ethanol
by means of ultrasonication. After evaporating ethanol, the remain-
ing solid was mixed with ␣-terpineol ͑40 wt %͒ to obtain a paste.
The resulting paste was applied on the front side of the YSZ plate
attached with narrow Pt stripes ͑served as a current collector͒ by
means of the screen printing technique to make a sensing electrode.
A commercial Pt paste ͑Tanaka Kikinzoku Co., Ltd., TR 7907͒ was
printed on the back side of YSZ plate as a reference electrode ͑RE͒.
To make a good electrical contact with a measuring equipment, Pt
wires ͑0.1 mm in diameter͒ were fixed by using a Pt paste onto Pt
connecting spots of both SE and RE. The planar sensor thus ob-
tained was kept at 130°C for 2 h in atmospheric air and was subse-
quently sintered at 1400°C for 2 h in air. Figure 1 shows the sche-
Furthermore, because the temperature of exhausts from a gaso-
line engine occasionally rises to 900°C during accelerating or cruis-
ing at high speed, it is important that the NOx sensor should work
efficiently even at such a high temperature. We have first reported
that NiO could be used as a sensing electrode ͑SE͒ even at tempera-
tures of 800–900°C under wet conditions.36-39 However, the sensi-
tivity of the sensor needs to be further improved for actual applica-
tion.
Quite recently, we have found that the addition of Rh to NiO–SE
can bring about remarkable improvement in NO2 sensitivity.40 How-
ever, the use of Rh seems to not be preferable for actual application
because of its high cost. Thus, it is necessary to find a cheaper
alternative additive instead of Rh. Among the examined various ox-
ide additives not including precious metals, WO3 was found to give
the largest improvement effect in NO2 sensitivity. Thus, we report
Figure 1. Schematic view of front and back sides of the planar NO2 sensor
used.
*
Electrochemical Society Active Member.
z E-mail: miurano@astec.kyushu-u.ac.jp
Downloaded on 2015-06-17 to IP 130.194.20.173 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 154 ͑8͒ J246-J252 ͑2007͒
matic view of the planar sensor used in the present investigation.
J247
Characterization of sensing materials.— The crystal structures
of pure NiO and WO3-added NiO samples were examined with an
X-ray diffractometer ͑Rigaku, RINT 2100VLR/PC͒. The Cu K␣ ra-
diation ͑␭ = 1.5406 Å͒ and 0.5°/min angle step were used for all
measurements. The morphology of SE was observed by the use of a
scanning electron microscopy ͑SEM, Hitachi, S–3000N, operating at
20 kV͒ and the composition of SE was analyzed using an energy
dispersive X-ray analyzer ͑EDX, Horiba, EX-220SE͒.
Evaluation of sensing performances and electrochemical proper-
ties
.— The measurements of NOx-sensing characteristics were per-
formed using a conventional gas-flow apparatus equipped with a
furnace operating at 800°C. The sample gas containing various con-
centrations of NO2 ranging from 5 to 100 ppm was prepared by
diluting a parent gas ͑500 ppm NO2 in dry N2͒ with dry N2 and O2.
The ppm refers to the volume concentration of NO2 in the gaseous
mixture. The base gas was composed of 5 vol % O2 ͑+N2 balance͒.
Both the sample gas and the base gas were humidified with 5 vol %
͑50,000 ppm͒ H2O vapor by means of a homemade water-vapor
generator. Both the base gas and the sample gas were allowed to
flow over the sensor at a constant flow rate of 100 cm3/min. SE and
RE were exposed simultaneously to the sample gas or the base gas.
The difference in potential ͑emf͒ between SE and RE of the planar
sensor was measured at 800°C with a digital electrometer ͑Ad-
vantest, R8240͒ as a sensing signal. The potential of the SE was
always positive with respect to RE.
The current-voltage ͑polarization͒ curves were measured by
means of an automatic polarization system ͑Hokuto Denko, HZ-
3000͒ based on potential-sweep method at a scan rate of 2 mV/min
for a two-electrode configuration at 800°C in the base gas ͑5 vol %
O2 + N2 balance͒ and in the sample gas containing 50 ppm NO2
͑ϩ the base gas͒. The current axis of the anodic polarization curve
was subtracted from that of the cathodic polarization curve at each
potential so as to obtain the modified polarization curve in which the
current axis was expressed in an absolute scale. The complex-
impedance measurements of the sensor were performed by means of
a complex-impedance analyzer ͑Solarton, 1255 WB͒ in the fre-
quency range from 0.01 Hz to 1 MHz at 800°C.
Figure 2. ͑Color online͒ Comparison of sensitivity ͑⌬emf͒ to 50 ppm NO2
at 800°C for the sensors using NiO-SE added with various metal-oxide ad-
ditives. ͑Base gas 5 vol % O2 + 5 vol % H2O + N2 balance.͒
WO3 is in the range of 5–15 wt %, while the sensitivity is degraded
when the content is more than 20 wt %. The maximum NO2 sensi-
tivity was observed at the initial content of 10 wt %. Thus, the de-
tailed sensing characteristics were examined for the sensor attached
with NiO-SE sintered after the addition of 10 wt % WO3. Hereafter,
we call such a NiO-based SE sintered after the WO3 addition
WO3-added NiO-SE.
Characterization of sensing-electrode materials.— In order to
examine the thermal stability of NiO and WO3-added NiO, X-ray
diffraction ͑XRD͒ patterns ͑not shown here͒ were recorded for each
of the oxide powders sintered at 1400°C for 2 h in air. It was ob-
served that, in all cases, NiO retains its crystallographic phase ͑face-
centered cubic͒ corresponding to JCPDS PDF ͑no. 44-1159͒. Due to
the sintering effect, all the peaks assigned to NiO were very narrow.
However, in the case of WO3-added NiO powders, there were no
Bragg reflections corresponding to WO3 and only the peaks for NiO
were observed. It was also confirmed from this XRD study that there
was no formation of solid solution between WO3 and NiO after
sintering at 1400°C. This indicates that, in the case of WO3-added
NiO-SE, most of the WO3 seemed to be evaporated from the NiO
matrix during the sintering process because of its ease evaporation at
Measurements of NO2 conversion to NO.— The catalytic activ-
ity of the gas-phase decomposition of NO2 ͑100 ppm NO2
+ 5 vol % O2 + N2 balance͒ to NO was evaluated for pure NiO,
10 wt % WO3-added NiO, and 50 wt % WO3-added NiO ͑ca. 0.2 g
each͒ sintered at 1400°C using a conventional flow cell as well as a
NOx analyzer ͑Yanaco, ECL-88A͒ in the temperature range of
200–900°C.
Results and Discussion
Selection of the best metal-oxide additive.— In order to select
the best oxide additive, each of various metal oxides, such as WO3,
ZnO, Ta2O5, SnO2, In2O3, Cr2O3, Fe2O3, CuO, Nb2O5, and CeO2,
was added ͑10 wt % each͒ to NiO-SE. Then, the sensitivity ͑⌬emf͒
to 50 ppm NO2 for the sensor attached with each of these SEs was
measured at 800°C in the presence of 5 vol % O2 and 5 vol %
water vapor. Here, the sensitivity is defined as the difference be-
tween the emf value in the sample gas and that in the base gas
͑⌬emf = emfsample − emfbase͒. It is seen from Fig. 2 that, among the
various metal-oxide additives examined, WO3, ZnO, Ta2O5, SnO2,
In2O3, and Cr2O3 can give some sort of improvement in NO2 sen-
sitivity, while the addition of other metal oxides, such as Fe2O3,
CuO, Nb2O5, and CeO2, shows decrement in NO2 sensitivity. Inter-
estingly, WO3 showed the highest improvement in NO2 sensitivity.
Figure 3 shows the variation of ⌬emf to 50 ppm NO2 at 800°C for
the sensor using NiO-SE sintered after the addition of various
amounts of WO3. The initial content of WO3 in the NiO-based
oxide paste before sintering was changed from 5 to 50 wt %. It is
seen that the NO2 sensitivity is improved when the initial content of
Figure 3. Variation of ⌬emf to 50 ppm NO2 at 800°C for the sensor using
NiO-SE added with different WO3 contents. ͑Base gas 5 vol % O2
+ 5 vol % H2O + N2 balance.͒
Downloaded on 2015-06-17 to IP 130.194.20.173 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

J248
Journal of The Electrochemical Society, 154 ͑8͒ J246-J252 ͑2007͒
NiO after sintering at 1400°C. There are no peaks corresponding to
W element in the NiO matrix in both cases. This implies that there is
no indication of the presence of WO3 in NiO matrix. This confirms
that almost all WO3 is evaporated from the NiO matrix, as specu-
lated above. Such a complete evaporation of WO3 lead to many
pores and bald surfaces in the case of 50 wt % WO3-added NiO-SE.
EDX analysis was also performed on 10 wt % WO3-added NiO and
50 wt % WO3-added NiO samples before sintering ͑the results were
not shown here͒. In both cases, the peaks corresponding to W ele-
ment were evidently present. The composition obtained from the
observed W peak intensity matched well with the nominal WO3
content before sintering. Based on the results obtained from XRD,
weight change measurement, SEM, and EDX analysis, it can be
concluded that the addition of WO3 to NiO can create many pores
on the NiO matrix and the amount of pores formed ͑after sintering͒
seems to depend strongly on the initial content of WO3 added before
sintering. Such a difference in porosity of NiO matrix leads to the
difference in NO2 sensitivity.
Figure 4. SEM images of the surface of ͑a͒ NiO, ͑b͒ 10 wt % WO3-added
NiO, ͑c͒ 50 wt % WO3-added NiO, and ͑d͒ 10 wt % Nb2O5-added NiO SEs.
Sensing performances of the NO2 sensor using WO3-added
NiO-SE
.— The response transients to various concentrations of NO2
were examined at 800°C for the planar sensor attached with pure
NiO-SE, 10 wt % WO3-added NiO-SE, and 50 wt % WO3-added
NiO-SE. The obtained results are shown in Fig. 6. It is seen that all
sensors respond well to NO2 and the steady-state emf values are
attained at each NO2 concentration rather quickly. The sensor at-
tached with 10 wt % WO3–NiO-SE shows the highest sensitivity
among the sensors tested in the examined range of 10–100 ppm.
This indicates that the NO2 sensitivity is significantly improved by
the addition of 10 wt % WO3 to NiO. However, the response and
recovery rates were found to be almost equal in all cases. The typi-
cal 70% response and recovery times for the sensor using 10 wt %
WO3–NiO and NiO-SEs are about 28 and 29 s, respectively. Figure
7 depicts the dependence of ⌬emf on NO2 concentration at 800°C
for the sensors attached with NiO-SE, 10 wt % WO3-added NiO-
SE, and 50 wt % WO3-added NiO-SE. In all cases, almost a linear
relationship between the ⌬emf and the logarithm of NO2 concentra-
tion is observed. The ⌬emf values for 10 wt % WO3–NiO-SE are
always much higher than those for NiO-SE and 50 wt %
WO3-added NiO-SE under the present condition. For instance, the
⌬emf value to 50 ppm NO2 at 800°C for the sensor using 10 wt %
WO3–NiO-SE and NiO-SE is about 70 and 42 mV, respectively.
The ⌬emf to NO2 in the range of 1–20 ppm was also measured at
800°C and the obtained results are shown in Fig. 8. It is surprising
that the sensor using 10 wt % WO3-added NiO-SE can detect even
1 ppm NO2 with acceptable sensitivity ͑⌬emf:8 mV͒. In this case,
⌬emf varies linearly with NO2 concentration in the range of
1–20 ppm. The reason for the occurrence of such a linear depen-
dance of NO2 sensitivity in the low concentration range is under
investigation.
It is well known that the amounts of water vapor and CO2 in real
exhausts from cars vary from 5 to 13 and 5 to 20 vol %, respec-
tively. Thus, it is important to examine the effects of water vapor
and CO2 on the NO2 sensitivity over the wide concentration range.
Figure 9 shows the dependence of emf to 50 ppm NO2 at 800°C on
the concentrations of water vapor and CO2 in the sample gas for the
sensor using 10 wt % WO3-added NiO-SE. It is seen that the emf
value of the sensor to 50 ppm NO2 is hardly affected by the con-
centrations of water vapor and CO2 in such a wide range. This
suggests that the present sensor can work well under actual exhaust
conditions.
We have previously reported that the addition of a small amount
of Rh to NiO also gives improvement in NO2 sensitivity.40 The
improvement degree of NO2 sensitivity of the present sensor using
10 wt % WO3-added NiO-SE is almost the same as that of the sen-
sor using Rh-added NiO-SE. Thus, the merit of the present sensor is
that the price of WO3 is much cheaper than that of Rh.
such a high temperature. It is noted that the evaporation temperature
of WO3 in air is reported to be 1400°C and the minimum detection
limit of minor constituent in XRD measurement is usually about
5%.
Because WO3 seemed to be evaporated from the NiO matrix,
there should be a weight change in the sample after sintering at
1400°C. Thus, to verify this, the weight of 10 wt % WO3-added
NiO was measured before and after sintering. As a result, it was
confirmed that there was a clear change in the weight of the sample
and the weight loss observed after sintering was exactly correspond-
ing to the weight of WO3 added.
Figure 4 shows the SEM images of ͑a͒ NiO, ͑b͒ 10 wt %
WO3-added NiO, and ͑c͒ 50 wt % WO3-added NiO samples on an
YSZ plate sintered at 1400°C for 2 h in air, along with ͑d͒ a
10 wt % Nb2O5-added NiO sample for comparison. It is seen that
the grain sizes of NiO range from 1 to 2 ␮m in pure NiO and
10 wt % WO3-added NiO. In 10 wt % WO3-added NiO matrix,
there appears to be increased porosity ͑larger pores͒ throughout the
matrix, compared with the pure NiO matrix. The increased porosity
may lead to lower gas-phase NO2 conversion to NO and then to
increased NO2 sensitivity. The formation of such pores could be due
to the evaporation of WO3 from the NiO matrix during sintering at
a high temperature of 1400°C. The surface of 50 wt % WO3-added
NiO consists of bald surfaces with less large NiO grains ͑size:
Ͼ2 ␮m͒ on YSZ, as seen in Fig. 4c. This implies that the 50 wt %
WO3 addition to NiO results in the formation of very porous NiO
matrix. In fact, the underlying YSZ can be clearly seen in this case.
Such large pores would decrease the number of reaction sites at the
interface between 50 wt % WO3-added NiO-SE and YSZ. Conse-
quently, the sensor using 50 wt % WO3-added NiO-SE is expected
to show diminished NO2 sensitivity. The addition of 10 wt %
Nb2O5 to NiO leads to a very dense matrix, as seen in Fig. 4d. Such
a dense matrix would retard the gas diffusion through itself, and
hence 10 wt % Nb2O5-added NiO-SE would show much lower NO2
sensitivity as given in Fig. 2. Thus, it can be said that WO3 can act
as a kind of pore-creating agent and Nb2O5 is working as an effec-
tive sintering agent. There is a possibility for the formation of solid
solution between Nb2O5 and NiO at such a high temperature of
sintering. However, the details in this regard are not considered in
this paper.
The evaporation of WO3 from the NiO matrix at 1400°C was
also verified on the basis of the results obtained from EDX analysis.
Figure 5 shows the SEM image of the cross-sectional view and EDX
spectrum of 10 wt % WO3-added NiO and 50 wt % WO3-added
Downloaded on 2015-06-17 to IP 130.194.20.173 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 154 ͑8͒ J246-J252 ͑2007͒
J249
Figure 5. SEM images of cross-sectional
view
and
EDX
spectrum
of
1400°C-sintered ͑a͒ 10 wt % WO3-added
NiO and ͑b͒ 50 wt % WO3-added NiO
SEs.
Explanation of WO3-addition effect.— In order to rationalize the
effect of WO3 addition on NO2 sensitivity of the sensor, the current-
voltage ͑polarization͒ curves of the sensor using each of the NiO-
based SEs were measured at 800°C in the base gas ͑5 vol % O2
+ N2 balance͒ and in the sample gas containing 50 ppm NO2 ͑ϩ the
base gas͒. The obtained results are shown in Fig. 10, in which the
current axis is expressed in an absolute scale. It is seen that the
polarization curve for the anodic reaction of O2 is almost same in
each sensor irrespective of the initial content of WO3. This implies
that the catalytic activity to the anodic reaction of O2 is not changed
even after 50 wt % addition of WO3. Such an unaltered rate of
oxygen reaction may be due to the presence of the minimally re-
Figure 7. ͑Color online͒ Variation of ⌬emf to different NO2 concentrations
in the range of 10-100 ppm at 800°C for the sensors attached with NiO-SE,
10 wt % WO3-added NiO, and 50 wt % WO3-added NiO-SE. ͑Base gas
5 vol % O2 + 5 vol % H2O + N2 balance.͒
Figure 6. ͑Color online͒ Response transients to NO2 for the sensors attached
with pure NiO-SE, 10 wt % WO3-added NiO, and 50 wt % WO3-added
NiO-SE at 800°C. ͑Base gas 5 vol % O2 + 5 vol % H2O + N2 balance.͒
Downloaded on 2015-06-17 to IP 130.194.20.173 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

J250
Journal of The Electrochemical Society, 154 ͑8͒ J246-J252 ͑2007͒
Figure 10. ͑Color online͒ Modified polarization curves obtained in
5 vol % O2 and in 50 ppm NO2 at 800°C for the planar sensors attached
with pure NiO-SE, 10 wt % WO3-added NiO-SE, and 50 wt % WO3-added
NiO-SE.
Figure 8. ͑Color online͒ Variation of ⌬emf to different NO2 concentrations
in the range of 1-20 ppm at 800°C for the sensors attached with NiO-SE and
10 wt % WO3-added NiO-SE. ͑Base gas 5 vol % O2 + 5 vol % H2O + N2
balance.͒
observed values in each case confirms that the sensing mechanism
of the present sensor is also based on the mixed-potential model, as
reported before.4-7,10,19-21,38-41
quired reaction sites in all cases irrespective of different morphol-
ogy. Thus, the sensor exhibited almost equal response/recovery
rates. The polarization curve for the cathodic reaction of NO2 shifts
upward when the SE of the sensor is changed from pure NiO to
10 wt % WO3-added NiO, while the polarization curve shifts down-
ward when the SE is changed to 50 wt % WO3-added NiO. This
suggests that the catalytic activity to the cathodic reaction of NO2 is
strongly dependent on the initial content of WO3 added to NiO and
is highest when the WO3 content is 10 wt %. We have previously
reported that, even though the rate of anodic reaction of oxygen is
unaltered, the sensitivity can be changed due to the change in the
rate of cathodic reaction of NO2 for a mixed-potential-type sensor.40
As a result, the intersection ͑estimated emf value͒ between the an-
odic polarization curve and the cathodic polarization curve increases
largely from 41 to 69 mV when the SE of the sensors is changed
from pure NiO to 10 wt % WO3-added NiO and decreases to
19 mV when the SE is changed to 50 wt % WO3-added NiO. Each
intersection value is in good agreement with the actual ⌬emf value
of each NO2 sensor. Such a close coincidence of the estimated and
The increase or decrease in the catalytic activity to the cathodic
reaction of NO2 caused by the addition of WO3 to NiO-SE was also
substantiated by the results of complex-impedance measurements.
Figure 11 gives Nyquist plots measured in the base gas ͑5 vol %
O2 + N2 balance͒ and in the sample gas containing 400 ppm NO2
͑ϩ the base gas͒ at 800°C under wet conditions for the sensor at-
tached with pure NiO-SE, 10 wt % WO3-added NiO-SE, and
50 wt % WO3-added NiO-SE. It is seen that, in all cases, the
complex-impedance plots are in the form of a compressed semi-
circle in the examined frequency range. The Nyquist plots in the
base gas obtained for the sensors attached with both 10 wt %
WO3-added NiO-SE and 50 wt % WO3-added NiO-SE are almost
Figure 9. ͑Color online͒ Dependence of ⌬emf to 50 ppm NO2 on different
concentrations of H2O and CO2 for the sensor using 10 wt % WO3-added
NiO-SE at 800°C.
Figure 11. ͑Color online͒ Complex impedance plots obtained in ͑a͒ 5 vol %
O2 and ͑b͒ 100 ppm NO2 at 800°C for the planar sensors attached with pure
NiO-SE, 10 wt % WO3-added NiO, and 50 wt % WO3-added NiO-SE.
Downloaded on 2015-06-17 to IP 130.194.20.173 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 154 ͑8͒ J246-J252 ͑2007͒
J251
It should be noted here that the initial content of WO3 was varied
from 5 to 50 wt %, and only 10 wt % WO3 addition gives a large
improvement in NO2 sensitivity. If a trace amount of WO3 would
present on the NiO surface and affect the NO2 sensitivity drastically,
the same influence ͑improvement͒ should be observed in all the
WO3-added samples. But it was not so, as seen from Fig. 3. For
example, 20 and 50 wt % WO3-added samples gave even a dimin-
ished sensitivity. Thus, only in the case of 10 wt % WO3 addition,
an appropriate morphology of NiO matrix can be obtained, which
leads to the large improvement in sensitivity.
Based on the above-mentioned results obtained from the mea-
surements of the polarization curves, the complex-impedance plots,
and the gas-phase NO2 decomposition catalysis, the improved NO2
sensitivity by the addition of 10 wt % WO3 to NiO can be attribut-
able to increased porosity, which in turn leads to the improved ki-
netic effect on the cathodic reaction of NO2 and the lower gas-phase
catalytic activity.
Conclusions
Figure 12. ͑Color online͒ Temperature dependence of NO2 conversion to
NO for the gas-phase reaction on pure NiO, 10 wt % WO3-added NiO, and
50 wt % WO3-added NiO samples sintered at 1400°C.
The mixed-potential-type planar sensors based on a YSZ plate
and NiO-based SEs were fabricated and examined for detection of
NO2 at high temperatures. Among the several examined metal-oxide
additives ͑WO3, ZnO, Ta2O5, SnO2, In2O3, Cr2O3, Fe2O3, CuO,
Nb2O5, and CeO2͒ to NiO-SE, WO3 was found to give rather large
improvement in NO2 sensitivity. The maximum enhancement was
observed when the initial WO3 content in NiO before sintering at
1400°C was 10 wt %. The sensitivity ͑⌬emf͒ to 50 ppm NO2 was
as high as about 70 mV for the sensor using the 10 wt %
WO3-added NiO-SE at 800°C in the presence of 5 vol % H2O and
5 vol % O2. The coexistence of H2O and CO2 gave no influence on
the NO2 sensitivity. Based on the results of measurements for the
polarization curves, the complex-impedance plots, and the gas-phase
catalysis, the improvement effect in NO2 sensitivity by the addition
of 10 wt % WO3 to NiO was speculated to be attributable to the
modification of morphology, which lead to the improved kinetic
effect on the cathodic reaction of NO2 and the lower gas-phase NO2
conversion rate.
the same as that of the sensor using pure NiO-SE. This suggests that
the rate of the anodic reaction of O2 occurring at the interface of
WO3-added NiO-SE/YSZ is the same as that of pure NiO-SE/YSZ.
The Nyquist plot in the sample gas shrinks when the SE changes
from pure NiO to 10 wt % WO3-added NiO and expands when the
SE changes to 50 wt % WO3-added NiO. This suggests that the rate
of cathodic reaction of NO2 at the interface of 10 wt % WO3-added
NiO-SE/YSZ is higher than that at the interface of pure NiO-SE/
YSZ. Thus, such an increased rate of cathodic reaction of NO2 at the
interface between 10 wt % WO3-added NiO-SE and YSZ is one of
the contributing factors for enhancing the NO2 sensitivity of the
present sensor. At the interface of 50 wt % WO3-added NiO-SE/
YSZ, the rate of cathodic reaction of NO2 seems to be much lower
than that for pure NiO. Such a lower reaction rate might be because
the number of reaction sites at the interface ͑50 wt % WO3-added
NiO-SE/YSZ͒ is significantly less than those at the interface of pure
NiO-SE/YSZ. It is evident from the SEM images ͑Fig. 4 and 5͒ that
the surface of 50 wt % WO3-added NiO-SE is composed of many
bald surfaces throughout the SE matrix. At these bald surfaces, the
electrochemical reaction of NO2 would hardly proceed. In this case,
there is also a possibility that the Pt-current collectors may contact
with YSZ to some extent and influence the emf response of the
sensor. Thus, as a result, the sensor using 50 wt % WO3-added
NiO-SE exhibited the diminished NO2 sensitivity compared with the
sensors using both pure NiO-SE and 10 wt % WO3-added NiO-SE.
We have already reported that the catalytic activity to the
gas-phase NO2 conversion to NO also plays an important role in
deciding the NO2 sensitivity of the mixed-potential-type
sensors. 31,32,36-41 Thus, it is essential to examine the gas-phase NO2
conversion to NO on pure NiO, 10 wt % WO3-added NiO, and
50 wt % WO3-added NiO samples sintered at 1400°C. Figure 12
shows the temperature dependence of NO2 conversion to NO for all
the samples. It is quite interesting that the 10 wt % WO3-added NiO
exhibits lower NO2 conversion than pure NiO at each temperature
examined. Such a lower gas-phase catalytic activity of SE layer also
plays an important role for the increment of NO2 sensitivity by the
10 wt % WO3 addition, because the undecomposed NO2 can reach
the interface of SE/YSZ in this case. The lower gas-phase catalysis
is mainly due to increased porous NiO matrix ͑less fewer reaction
sites͒ due to evaporation of WO3, as seen from SEM images ͑Fig.
4͒. The gas-phase catalytic activity of 50 wt % WO3-added NiO
was much lower than those of pure NiO and 10 wt % WO3-added
NiO, probably due to bigger pores and more bald surface. However,
the contribution of such a lower catalytic activity to the lower NO2
sensitivity is not cleared yet in this case.
Acknowledgments
This work was supported in part through the Effective Promotion
of Joint Research with Industry, Academia, and Government, Spe-
cial Coordination Fund for Promoting Science and Technology and
the Grant-in-Aid for Scientific Research on Priority Area, Nanoion-
ics ͑439͒, by MEXT.
Kyushu University assisted in meeting the publication costs of this
article.
References
1. F. Menil, V. Coillard, and C. Lucat, Sens. Actuators B, 67, 1 ͑2000͒.
2. N. Docquier and S. Candel, Prog. Energy Combust. Sci., 28, 107 ͑2002͒.
3. J. W. Fergus, Sens. Actuators B, 121, 652 ͑2006͒.
4. N. Miura, G. Lu, N. Yamazoe, H. Kurosawa, and M. Hasei, J. Electrochem. Soc.,
143, L33 ͑1996͒.
5. N. Miura, H. Kurosawa, M. Hasei, G. Lu, and N. Yamazoe, Solid State Ionics,
86–88, 1069 ͑1996͒.
6. N. Miura, G. Lu, and N. Yamazoe, Sens. Actuators B, 52, 169 ͑1998͒.
7. G. Lu, N. Miura, and N. Yamazoe, Ionics, 4, 16 ͑1998͒.
8. T. Hibino, Y. Kuwahara, T. Otsuka, N. Ishida, and T. Oshima, Solid State Ionics,
107, 213 ͑1998͒.
9. T. Hibino, Y. Kuwahara, T. Otsuka, N. Ishida, and T. Oshima, Solid State Ionics,
107, 217 ͑1998͒.
10. N. Miura, G. Lu, and N. Yamazoe, Solid State Ionics, 136–137, 533 ͑2000͒.
11. W. Geopel, G. Reinhardt, and M. Rosch, Solid State Ionics, 136–137, 19 ͑2000͒.
12. E. D. Bartolomeo, E. Traversa, M. Baroncini, V. Kotzeva, and R. V. Kumar, J. Eur.
Ceram. Soc., 20, 2691 ͑2000͒.
13. T. Ono, M. Hasei, A. Kunimoto, T. Yamamoto, and A. Noda, JSAE Rev., 22, 49
͑2001͒.
14. S. Zhuiykov, M. Muta, T. Nakano, M. Hasei, N. Yamazoe, and N. Miura, Electro-
chem. Solid-State Lett., 4, H19 ͑2001͒.
15. J. W. Yoon, M. L. Grilli, E. D. Bartolomeo, R. Polini, and E. Traversa, Sens.
Actuators B, 76, 483 ͑2002͒.
16. N. F. Szabo, H. Du, S. A. Akbar, A. Soliman, and P. K. Dutta, Sens. Actuators B,
82, 142 ͑2002͒.
Downloaded on 2015-06-17 to IP 130.194.20.173 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

J252
Journal of The Electrochemical Society, 154 ͑8͒ J246-J252 ͑2007͒
17. N. F. Szabo, H. B. Du, S. A. Akbar, A. Soliman, and P. K. Dutta, Sens. Actuators
B, 82, 142 ͑2002͒.
18. E. L. Brosha, R. Mukundan, D. R. Brown, and F. H. Garzon, Sens. Actuators B, 87,
47 ͑2002͒.
19. N. Miura, S. Zhuiykov, T. Ono, M. Hasei, and N. Yamazoe, Sens. Actuators B, 83,
222 ͑2002͒.
20. S. Zhuiykov, T. Ono, N. Yamazoe, and N. Miura, Solid State Ionics, 152–153, 810
͑2002͒.
21. T. Ono, M. Hasei, A. Kunimoto, and N. Miura, Electrochemistry (Tokyo, Jpn.), 71,
405 ͑2003͒.
22. R. Mukundan, E. L. Brosha, and F. H. Garson, J. Electrochem. Soc., 150, H279
͑2003͒.
23. D. C. Skelton, R. G. Tobin, D. K. Lambert, C. L. Di Maggio, and G. B. Fisher,
Sens. Actuators B, 96, 46 ͑2003͒.
24. L. P. Martin, A. Q. Pham, and R. S. Glass, Sens. Actuators B, 96, 53 ͑2003͒.
25. A. Dutta, N. Kaabbuathong, M. L. Grilli, E. Di Bartolomeo, and E. Traversa, J.
Electrochem. Soc., 150, H33 ͑2003͒.
30. D. Y. Wang and E. Detwiler, Sens. Actuators B, 99, 571 ͑2004͒.
31. T. Ono, M. Hasei, A. Kunimoto, and N. Miura, Solid State Ionics, 175, 503 ͑2004͒.
32. N. Miura, K. Akisada, J. Wang, S. Zhuiykov, and T. Ono, Ionics, 10, 1 ͑2004͒.
33. R. Ramamoorthy, S. A. Akbar, and P. K. Dutta, Sens. Actuators B, 113, 162
͑2006͒.
34. E. L. Brosha, R. Mukundan, R. Lujan, and F. H. Garzon, Sens. Actuators B, 119,
398 ͑2006͒.
35. S. W. Song, L. P. Martin, R. S. Glass, E. P. Murray, J. H. Visser, R. E. Soltis, R. F.
Novak, and D. J. Kubinski, J. Electrochem. Soc., 153, H171 ͑2006͒.
36. P. Elumalai, J. Wang, S. Zhuiykov, D. Terada, M. Hasei, and N. Miura, J. Electro-
chem. Soc., 152, H95 ͑2005͒.
37. N. Miura, J. Wang, M. Nakatou, P. Elumalai, and M. Hasei, Electrochem. Solid-
State Lett., 8, H9 ͑2005͒.
38. P. Elumalai and N. Miura, Solid State Ionics, 176, 2517 ͑2005͒.
39. N. Miura, J. Wang, M. Nakatou, P. Elumalai, S. Zhuiykov, and M. Hasei, Sens.
Actuators B, 114, 903 ͑2006͒.
40. J. Wang, P. Elumalai, D. Terada, M. Hasei, and N. Miura, Solid State Ionics, 177,
2305 ͑2006͒.
41. P. Elumalai, V. V. Plashnitsa, T. Ueda, M. Hasei, and N. Miura, Ionics, 12, 331
͑2006͒.
26. S. Kading, S. Jakobs, and U. Guth, Ionics, 9, 151 ͑2003͒.
27. W.-F. Zhang, P. Schmidt-Zhang, and U. Guth, Solid State Ionics, 169, 121 ͑2004͒.
28. U. Guth and J. Zosel, Ionics, 10, 366 ͑2004͒.
29. N. F. Szabo and P. K. Dutta, Solid State Ionics, 171, 183 ͑2004͒.
Downloaded on 2015-06-17 to IP 130.194.20.173 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).