Please cite this article in press as: T. Fu, Research on gas-sensing properties of lead sulfide-based sensor for detection of NO
2
and NH
3
at room temperature, Sens. Actuators B: Chem. (2009), doi:10.1016/j.snb.2009.03.075
ARTICLE IN PRESS
G Model
SNB-11469; No. of Pages 6
Sensors and Actuators B xxx (2009) xxx–xxx
Contents lists available at ScienceDirect
Sensors and Actuators B: Chemical
journal homepage: www.elsevier.com/locate/snb
Research on gas-sensing properties of lead sulfide-based sensor for detection
of NO
2
and NH
3
at room temperature
Tiexiang Fu
∗
School of Chemistry and Bioengineering, Changsha University of Science & Technology, Chiling Road, Changsha 410077, Hunan, PR China
article info
Article history:
Received 20 November 2008
Received in revised form 28 March 2009
Accepted 31 March 2009
Available online xxx
Keywords:
Nitrogen dioxide
Ammonia
Gas-sensing properties

2
) and ammonia (NH
3
) are foul-smelling
and harmful gases. Nitrogen dioxide is created by the high temper-
ature combustion of coal, chemical production, natural gas or oil
in power plants and also by the combustion of gasoline in inter-
nal combustion engines. One of the consequences of NO
2
being
released into the atmosphere is the formation of photochemical
smog. In addition, NO
2
is a cause of acid rain and is involved in the
depletion of ozone in the stratosphere. Ammonia is the most abun-
dant alkalinecomponent in the atmosphere, and it has been focuse d
air quality regulatory attention on the livestock and poultry indus-
tries [1,2]. Other typical sources of ammonia include fertilizers,
soils and production of chemicals, etc. The importance of the detec-
tion of these gases is evident. There are needs for nitrogen dioxide
and ammonia sensors are used in many situations including leak-
detection in air-conditioning systems and environmental sensing of
trace amounts ambient NH
3
or NO
2
in air, and the automatic control
of the chemical engineering production process involving ammonia
or nitrogen dioxide [3]. Generally, because these gases are toxic, it is
necessary to be able to sense low levels (∼ppm) of NO

3
-sensing properties of metal com-
plex [19–21]. A highly selective NH
3
sensor based on potassium
trisoxalateferrate(III) complex [22] andaNO
2
gas sensor based
on complex [Cr(bipyO
2
)Cl
2
]Cl [23] have also been reported by our
group.
Lead sulfide (PbS) is normally used as optical and semiconduc-
tor materials [24]. The practical application of PbS in a gas sensor
has only been studied by Y. Shimizu and his group. They carried
out systematically research determine the gas-sensing properties
of the Pb
1−x
Cd
x
S(x = 0.1, 0.2)-based and the metal-mono sulfide-
based (NiS, CdS, SnS and PbS) solid electrolyte sensor elements. The
sensor elements gave good SO
2
sensitivity at 300–400
◦
C [25,26].
But there is no report about the NO

Fig. 1. Schematic structure of the sensor.
was slowly dropped into this lead acetate solution (0.3 mol/L) and
then stirred at a temperature of 50
◦
C. The resulting precipitate was
washed with distilled water several times and filtered. Finally, PbS
particles were dried 2 h at 150
◦
C. Another kind of PbS particle was
prepared at reaction temperature of 80
◦
C similar to the method
and procedure described above.
H
2
S used for precipitating agent. 0.03 mol H
2
S was slowly injected
into 100 mL of lead acetate solution (0.3 mol/L) and vigorously
stirred at temperature 50
◦
C. The resulting precipitate was washed
with distilled water several times and filtered. Finally, PbS particles
were dried 2 h at 150
◦
C. Another kind of PbS particle was prepared
similar to the method and procedure described above, only with a
reaction temperature of 80
◦
C.

sensing measurement, the sample gas concentrations were altered
by injecting sample gases into the airtight chamber using a sample
injector. A heating voltage (V
h
) was supplied to the coil for heating
the sensor and the working temperature of the sensor element was
Fig. 2. Electric circuit for gas-sensing measurement.
changed from room temperature to 85
◦
C by controlling the heater
voltage. Operating voltages (V
O
) of 2 V, 5 V, 10 V and 20 V dc were
applied across the circuit, and the voltage outputs (V
S
) across the
sampling resistor were recorded (range: from 0.1 to 2000 mV). The
electrical resistance of a sensor was measured in sample gases and
also in air. The transformational relation between the resistance (R)
of a sensor and the output voltage of the sampling resistor in circuit
is given by the following formula:
R = (V
O
− V
S
)R
S
/V
S
where R

3
)
concentration, was obtained by changing the NO
2
(or NH
3
)gas
concentration from 0.004 to 9.23% at room temperature.
The response or recovery time is the time for the voltage change
to reach 90% of the total change from V
(out)a
to V
(out)g
or vice versa.
The response and recovery time of sensors depended on the thick-
ness of the gas-sensing layer. The thinner the layer, the shorter the
response or recovery time was, in general. However, the responses
to NO
2
and NH
3
decreased when the thickness of the gas-sensing
layer was too thin. The response and recovery time of the sensor in
the present study were all measured at the optimizing thickness of
gas-sensing layer, about 5 ␮m.
Two sensors were heated for a week continuously, and then the
stability of the sensors was examined by comparing the data of
sensing properties at room temperature before and after the heat-
ing.
Please cite this article in press as: T. Fu, Research on gas-sensing properties of lead sulfide-based sensor for detection of NO

OH), acetone (CH
3
COCH
3
), benzene (C
6
H
6
), toluene
(C
7
H
8
), ethyl ether (C
4
H
10
O), tetrahydrofuran (C
4
H
8
O) and chloro-
form (CHCl
3
) were tested. The concentrations of these gases were
changed from 0 to 9.23% (the ratio of sample gas volume to air
volume). The sensor responses to sample gases were measured at
nine working temperatures (25, 30, 35, 40, 45, 55, 65, 75, 85
◦
C)

and 21.2 to
NH
3
(see Fig. 3(a)), respectively. The response magnitude of PbS
prepared with Na
2
Sat80
◦
C was about 105 to NO
2
and 296 to
NH
3
(see Fig. 3(b)), respectively. The experimental results of anti-
interference at room temperature and 5 V operating voltage showed
that, as long as the concentration of NO
2
and NH
3
is more than 0.6%,
other gases will not interfere with the determination of NH
3
and
NO
2
. The behavior of the sensors is also very similar at other oper-
ating voltages and other working temperatures. However, NH
3
will
not interfere with the determination of NO

2
Sat50
◦
C and H
2
Sat80
◦
C) were studied at four operating volt-
ages (2, 5, 10 and 20 V). The highest values of response to NO
2
at
room temperature are shown in Table 1. All sensors are sensitive
to NO
2
. The most responsive sensor to NO
2
was that based on PbS
prepared with Na
2
Sat50
◦
C. This sensor had the highest response
value at 5V operating voltages. The response and recovery time of
sensors at optimal thickness of gas-sensing layer (5 ␮m) are about
Table 1
Sensor’s highest value of response to NO
2
in a room temperature (each highest
response value appears at a different concentration).
Preparation of PbS with Operating voltage (V)

◦
C.
54 and 43 s, respectively. The average relative deviation was about
2.8%. The second most responsive sensor to NO
2
was the sensor
based on PbS prepared with H
2
Sat80
◦
C. The response and recovery
time of the sensor was about 135 and 98 s, respectively. The average
relative deviation was about 7.3%. The influences of precipitation
agents and temperature of preparation PbS on the sensor response
were not clearly explained. One of possible reasons is that sulfur
ion concentration, chemical reaction rate and temperature influ-
ence the surface area and the activation absorption center (position
and quantity) of precipitates.
The sensor response to NO
2
depends on the gas concentration.
The sensor response curve to NO
2
at the optimal operating volt-
age, based on PbS prepared with Na
2
Sat50
◦
C, is shown in Fig. 4.
Please cite this article in press as: T. Fu, Research on gas-sensing properties of lead sulfide-based sensor for detection of NO

slight decrease with increasing NO
2
concentration. Other sensor’s
response curves also indicate similar responses to NO
2
gas, respec-
tively.
The operating voltage had an influence on sensor’s response
to NO
2
gas at room temperature. All the sensor responses to NO
2
increased and then decreased with an increase in operating voltage
at room temperature. The optimum operating voltage of sensor was
5V.
3.1.3. Sensor response to NH
3
Table 2 shows the highest values of sensor’s response to NH
3
at
room temperature. From Table 2, we can see that the most respon-
sive sensor to NH
3
was that based on PbS prepared with Na
2
Sat
80
◦
C. This sensor had the highest response value at 5 V operating
voltages. The response and recovery time was about 46 and 67 s,

2
Sat80
◦
C 12.9 296.0 69.4 31.9
H
2
Sat50
◦
C 12 225.1 43.6 17.5
H
2
Sat80
◦
C 11.2 27.2 18 15.2
Fig. 5. The correlation between NH
3
gas concentration and responses of the sensors
based on PbS was prepared with Na
2
Sat80
◦
C at the optimal operating voltage and
in a room temperature.
that NH
3
and NO
2
on the surface of PbS absorption ways and means
of electron transfer are different.
The variations in response of the sensor based on PbS prepared

than those at 20 V. The optimum operating voltage of the sensor
was also 5 V.
3.2. Relationship between gas response and temperature
The temperature dependences of the sensor response at 5 V
operating voltage are presented in Fig. 6. Fig. 6(a) shows that sen-
sor based on PbS prepared with Na
2
Sat50
◦
C in responses of NO
2
slightly decreased as the temperature increased from room tem-
perature to 47
◦
C. After that, the responses decreased sharply. When
the temperature exceeded 75
◦
C, the sensor did not respond to NO
2
.
Fig. 6(b) shows that responses to NH
3
of the sensor based on PbS
prepared with Na
2
Sat80
◦
C slightly decreased as the temperature
increased from room temperature to 42
◦

G Model
SNB-11469; No. of Pages 6
T. Fu / Sensors and Actuators B xxx (2009) xxx–xxx 5
Fig. 6. The responses as a function of temperature for two sensors based on PbS at
operating voltage 5 V. PbS was prepared under different conditions: (a) with Na
2
S
at 50
◦
C, sensor responses to NO
2
; (b) with Na
2
Sat80
◦
C, sensor responses to NH
3
.
at five concentrations. The measurement results showed popu-
larly consistent with what was done above. The average relative
deviations were in the range of about ±2.7 to NO
2
and ±3.8% to
NH
3
, respectively. The sensors’ electrical resistance in air changed
very poorly. The stability of the gas-sensing layers’ resistance in air
ensures a stable level for the gas sensors’ applications.
3.3. Discussion about the gas-sensing mechanism
It is known that each NO

concentration until reaching the absorp-
tion saturation. The comparison between the response curve to NH
3
and theoretical presumption shows a good agreement with each
other. The saturation response point corresponds to the saturation
point of gases adsorbed. However, the response curve to NO
2
is not
the same. The response increased significantly with an increase in
concentration until the NO
2
gas concentration exceeded a certain
value, as it can be seen in Fig. 4. The relation curves between the
response and gases concentration show a saturation point and an
inflexion.
The response saturation point corresponds to the saturation
point of gases adsorbed, but the inflexion phenomenon shows
that the sensing layer’s gas adsorption may be divided into two
stages. First, as a result of Van der Waals attraction, when the dis-
tance between NO
2
molecules and the surface of the gas-sensing
layer is shortened, the potential energy is decreased to the min-
imum of a nadir. At this stage, it is a physical adsorption and
few electrons transfer to the layers to make the electrical resis-
tance decrease. Second, when the distance is shortened further,
there is a deeper potential well which has a more steady adsorp-
tion, a weak coordination adsorption. The weak coordinate bonds
can be formed with transferring electrons partly from NO
2

of NO
2
and NH
3
adsorption quantities in the layers decreases. As
a consequence, a decrease in response to NO
2
or NH
3
of the gas-
sensing layer is observed.
It is understandable that temperature affected the sensor
response. Increasing temperature can speed up gas molecules’
motion to weaken the weak coordination adsorption. As a result,
the sensor responses decreased because of a decrease in the amount
of molecules adsorbed and electrons transferred from NO
2
or NH
3
to the gas-sensing layer.
4. Conclusions
In summary, The response and selectivity of four sensor types
based on PbS prepared under four conditions were tested at dif-
ferent temperatures and operating voltages. These sensors showed
high response to NO
2
and NH
3
at room temperature. No response
to H

Please cite this article in press as: T. Fu, Research on gas-sensing properties of lead sulfide-based sensor for detection of NO
2
and NH
3
at room temperature, Sens. Actuators B: Chem. (2009), doi:10.1016/j.snb.2009.03.075
ARTICLE IN PRESS
G Model
SNB-11469; No. of Pages 6
6 T. Fu / Sensors and Actuators B xxx (2009) xxx–xxx
References
[1] G. Huyberechts, M. Van Muylder, M. Honor’e, J. Desmet, J. Roggen, Development
of a thin film ammonia sensor for livestock buildings, Sens. Actuators B 18–19
(1994) 296–299.
[2] H. Xin, Y. Liang, A. Tanaka, R.S. Gates, E.F. Wheeler, K.D. Casey, K.D. Heber,
J.Q. Ni, H. Li, Ammonia emissions from U.S. poultry houses. Part I. Mea-
surement system and techniques, in: Proceedings of the Third International
Conference on Air Pollution from Agricultural Operations, NC, 12–15 October,
2003.
[3] E.J. Connolly, B. Timmer, H.T.M. Pham, J. Groeneweg, P.M. Sarro, W. Olthuis, P.J.
French, A porous SiC ammonia sensor, Sens. Actuators B 109 (2005) 44–46.
[4] T.Maekawa, J. Tamaki, N. Miura, N. Yamazoe, Gold-loaded tungsten oxide sensor
for detection of ammonia in air, Chem. Lett. (1992) 639–642.
[5] E. Llobet, G. Molas, P. Molin‘as, J. Calderer, X. Vilanova, J. Brezmes, J.E. Sueiras,
X. Correig, Fabrication of highly selective tungsten oxide ammonia sensors, J.
Electrochem. Soc. 147 (2000) 776–779.
[6] C.N. Xu, N. Miura, Y. Ishida, K. Matsuda, N. Yamazoe, Selective detection of NH
3
over NO in combustion exhaust by using Au and MoO
3
doubly promoted WO

[12] S.T. Shishiyanu, T.S. Shishiyanu, O.I. Lupan, Novel NO
2
gas sensor based on
cuprous oxide thin films, Sens. Actuators B 113 (2006) 468–476.
[13] M. Penza, C. Martucci, G. Cassano, NO
x
gas sensing characteristics of WO
3
thin
films activated by noble metals (Pd, Pt, Au) layers, Sens. Actuators B 50 (1998)
52–59.
[14] T.S. Kim, T.B. Kim, K.S. Yoo, G.S. Sung, H.J. Jung, Sensing characteristics of dc
reactive sputtered WO
3
thin films as an NO
x
gas sensor, Sens. Actuators B 62
(2000) 102–108.
[15] C. Baratto, G. Faglia, E. Comini, G. Sberveglieri, A. Taroni, V. La Ferrara, L. Quer-
cia, G. Di Francia, A novel porous silicon sensor for detection of sub-ppm NO
2
concentrations, Sens. Actuators B 77 (2001) 62–66.
[16] J. Suhiro, G. Zhou, M. Hara, Fabrication of a carbon nanotubes-based gas
sensor using dielectrophoresis and its application for ammonia detection by
impedance spectroscopy, J. Phys. D: Appl. Phys. 36 (2003) 109–114.
[17] H. Chang, J.D. Lee, S.M. Lee, Y.H. Lee, Adsorption of NH
3
and NO
2
molecules on

sensor using
metal–sulfide electrode, J. Mater. Sci. 38 (2003) 4301–4305.
Biography
Tiexiang Fu is currently a professor at the School of Chemistry and Bioengineer-
ing, Changsha University of Science & Technology. His current research interests are
orientation complexes, gas sensors and electronic nose systems.