- Email: [email protected]
Nb2O5 Schottky based ethanol vapour sensors: Effect of metallic catalysts
Accepted Manuscript Title: Nb2 O5 Schottky based ethanol vapour sensors: effect of metallic catalysts Author: Rosmalini Ab Kadir Rozina Abdul Rani Ahmad Sabirin Zoolfakar Jian Zhen Ou Mahnaz Shafiei Wojtek Wlodarski Kourosh Kalantar-zadeh PII: DOI: Reference:
S0925-4005(14)00494-8 http://dx.doi.org/doi:10.1016/j.snb.2014.04.083 SNB 16849
To appear in:
Sensors and Actuators B
Received date: Revised date: Accepted date:
27-12-2013 9-4-2014 24-4-2014
Please cite this article as: R.A. Kadir, R.A. Rani, A.S. Zoolfakar, J.Z. Ou, M. Shafiei, W. Wlodarski, K. Kalantar-zadeh, Nb2 O5 Schottky based ethanol vapour sensors: effect of metallic catalysts, Sensors and Actuators B: Chemical (2014), http://dx.doi.org/10.1016/j.snb.2014.04.083 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Nb2O5 Schottky based ethanol vapour sensors: effect
2
of metallic catalysts
3
Rosmalini Ab Kadir1,*, Rozina Abdul Rani1, Ahmad Sabirin Zoolfakar1, Jian Zhen Ou1,
4
Mahnaz Shafiei2, Wojtek Wlodarski1 and Kourosh Kalantar-zadeh1,*
cr
ip t
1
1
6
8
School of Electrical and Computer Engineering, RMIT University, Melbourne, VIC, Australia
2
Chemistry, Physics and Mechanical Engineering School, Queensland University of Technology,
an
7
us
5
Brisbane, QLD, Australia
10
M
9
Keywords: Niobium oxide, anodization, ethanol vapour, gas sensor, catalytic effect
14
te
13
Ac ce p
12
d
11
1
Page 1 of 37
ABSTRACT: The aim of this paper is to compare the performances of the highly porous Nb2O5
15
Schottky based sensors formed using different catalytic metals for ethanol vapour sensing. The
16
fabricated sensors consist of a fairly ordered nano-vein like porous Nb2O5 prepared via an
17
elevated temperature anodization method. Subsequently, Pt, Pd and Au were sputtered as both
18
Schottky contacts and catalysts for the comparative studies. These metals are chosen as they have
19
large work functions in comparison to the electron affinity of the anodized Nb2O5. It is
20
demonstrated that the device based on Pd/Nb2O5 Schottky contact has the highest sensitivity
21
amongst the developed sensors. The sensing behaviors were studied in terms of the Schottky
22
barrier height variations and properties of the metal catalysts.
an
us
cr
ip t
14
23 1. Introduction
25
Ethanol vapour sensors are in always in demand due to their significant applications in food,
26
biomedical, transportation and chemical industries as well as health and safety [1, 2]. Ethanol is
27
also commonly used as a fuel or a fuel additive. It is either utilized or appears as a by-product of
28
many chemical processes and is a common solvent in many industries. Ethanol is employed in
29
alcoholic beverages and is also a famous antiseptic. Obviously, sensing ethanol is of importance
30
for all the aforementioned applications [3-5].
Ac ce p
te
d
M
24
31
There are many types of ethanol vapour sensors based on different transducers such as
32
conductometric, surface acoustic wave, electrochemical, optical and Schottky diode based [6-11].
33
Amongst these transducers, Schottky diodes with nanostructured metal oxide sensing elements
34
offer simplicity of fabrication; reliable robustness and high sensitivity towards ethanol vapour
35
[12]. A Schottky diode based sensor generally consists of a metal-semiconductor contact. The
36
most common choices of metals are catalytic type noble metals including gold (Au), platinum
37
(Pt) and palladium (Pd) [13-17]. They catalyse the target gas molecules, while regain their
2
Page 2 of 37
metallic properties after the interaction. Additionally with their relatively large work functions,
39
these metals are likely to form Schottky contacts at their interface with most of the metal oxide
40
semiconductors [18]. Metal catalysts are employed in the chemical industry for numerous
41
chemical reactions. They operate by interacting with the target molecules, lowering the activation
42
barriers and hence increasing the rate of reactions. The metal catalysts also enhance the
43
sensitivity and reduce response and recovery time of the sensor. Au, Pt and Pd in the Schottky
44
diode based sensors enable catalytic decomposition of ethanol, which otherwise would not be
45
possible on the oxide surface. After the adsorption of ethanol, the local electronic properties of
46
the metal/metal oxide semiconductor changes that results in a change in the Schottky contact
47
behaviour and this alteration is used as the sensing parameter [19, 20].
an
us
cr
ip t
38
Metal oxides that have been widely investigated in Schottky diode based ethanol sensors
49
include SnO2, TiO2, ZnO, WO3, MoO3 and Ga2O3 [21-29]. Niobium oxide (Nb2O5) is an important
50
but yet much less studied metal oxide for Schottky diode based ethanol sensing. It is a n-type
51
semiconductor with a ~3.5 eV band gap that have shown to offer unique properties for
52
developing highly efficient solar-cells, electrochromic devices and sensors [30-35]. Devices
53
based on Nb2O5 have been reported for sensing of various gases and chemical components [36-
54
41]. Numerous methods have been established for fabricating Nb2O5 based gas sensing devices.
55
For example, Hyodo et al. and Rozina et al. have used Nb2O5 prepared via an anodization method
56
[36, 37], while Nb2O5 nanowires were synthesized by Wang et al. through thermal oxidation
57
process for hydrogen gas sensing [38]. Nb2O5 thin films doped with TiO2 have also been
58
developed via sputtering method for carbon monoxide gas sensing [40]. There are also
59
investigations regarding the effect of the calcination temperature on gas sensing properties of
60
Nb2O5 [39].
Ac ce p
te
d
M
48
3
Page 3 of 37
In this work, the Nb2O5 Schottky diode based sensors, integrated with different catalytic
62
metals, are developed and investigated. Three noble metals of Pt, Pd, and Au with different work
63
functions of 5.7, 5.6 and 5.4 eV respectively, are employed [42]. The effects of Pt, Pd and Au on
64
the ethanol sensing performances of the Schottky diode based sensors are fully studied and
65
discussed.
ip t
61
2. Experimental
us
67
cr
66
68 2.1. Fabrication
70
The Schottky diode based sensor devices were fabricated using niobium foil (99.9% purity,
71
Sigma Aldrich) of 0.25 mm thickness. The foil was cut into pieces of 1.0 cm × 1.0 cm, which
72
were cleaned with acetone, followed by isopropanol and rinsed with deionized water then dried
73
in a stream of high purity nitrogen gas. Nb2O5 film synthesized via an anodization process that
74
was conducted at 50 °C in the electrolyte of NH4F/ethylene glycol, containing a small amount of
75
water using a process that was based on our previous work [36]. Films were then annealed at 440
76
°C for 30 min at the ramp up/down of 2 °C/min for crystallization. The final stage of the
77
fabrication was the formation of the Schottky contacts. For this study, Pt, Pd and Au layers with
78
the thickness of ~20 nm each were deposited on the annealed Nb2O5 film using a GATAN
79
PECSTM sputter coating system. A shadow mask with a 2.5 mm diameter hole was utilized to
80
obtain a circular pad of catalytic contact. Sensor with Pt, Pd and Au contacts will be addressed as
81
Pt−Nb2O5, Pd−Nb2O5 and Au−Nb2O5, respectively, throughout this paper.
82
2.2. Structural characterizations.
Ac ce p
te
d
M
an
69
83
The morphologies of Nb2O5 films were studied using scanning electron microscopy (SEM -
84
FEI Nova NanoSEM), while the crystallographic properties of the films were studied via X-ray
4
Page 4 of 37
diffraction (XRD) analysis using a Bruker D8 DISCOVER microdiffractometer fitted with a
86
general area detector diffraction system (GADDS) that accommodates Cu-Kα radiation (l =
87
1.54178 Å). Analysis of the X-ray photoelectron spectrometry (XPS) was conducted using
88
Thermo Scientific K-alpha X-ray Photoelectron.
89
d
M
an
us
cr
ip t
85
2.3. Measurement setup
91
The gas sensing measurements devices were conducted in a LINKAM (Scientific Instruments)
92
customized gas testing chamber. A computerized mass flow control (MFC) multi-channel gas
93
calibration system was used, which allowed the sensors to be exposed to ethanol with different
94
concentrations by diluting it using 99.99 % dry synthetic air. This is carried out by adjusting the
95
flow rates of each gas channel, producing a specified concentration of the analyte vapour species
96
flowing at a total constant flow rate of 200 sccm to the LINKAM chamber. Electrical
97
connections to the Schottky gas sensors were achieved by physically connecting the probes to the
98
metal pads (top electrode) and Nb (bottom electrode) as schematically shown in Fig. 1. The
99
current−voltage (I−V) measurements were carried out using Keithley 2606 current source meter
100
as they were sequentially heated from room temperature to 270 °C and exposed to a mixture of
Ac ce p
te
90
5
Page 5 of 37
ethanol vapour and air. The sensors were mounted on heaters, which controlled their operational
102
temperatures. The measurements started with the Pd−Nb2O5 sensor, and it was observed that the
103
largest voltage shift occurred at 180 °C. Due to this observation and for comparison purposes, the
104
dynamic measurements of other sensors were also obtained at 180 °C. The dynamic responses
105
were measured as a change in the voltage magnitude utilizing an Agilent 34410A digital
106
multimeter to record the voltage alterations under the exposure to ethanol vapour with different
107
concentrations in air.
108 109
3. Results and discussion
110 111
3.1
112
The SEM images of the cross-sectional and top views of nanoporous Nb2O5 after the annodization
113
and annealing processes are presented in Fig. 2. The thickness of the film in this figure is ~1 μm.
114
From Fig. 2(b) it is observed that Nb2O5 films are made of fairly organized pores, with inner
115
diameters ranging from 30 to 50 nm. Fig. 2(c) shows the higher magnification cross-sectional
116
images of the nanoporous Nb2O5 film. It is observed that the film is made of highly packed vein-
117
like nanostructured networks.
Ac ce p
te
d
M
Morphology and structural properties
an
us
cr
ip t
101
118
The XPS analysis was carried out in order to investigate the elemental composition of the
119
sensing films. As depicted in Fig 3(a-d), XPS survey spectra of the sensors show the presence of
120
Nb, O, Pt, Pd and Au elements with very little carbon as a contaminant. Inset of Fig. 3(a), shows
121
the high-resolution spectrum for the Nb region that exhibits the Nb3d3/2 peaks at 210.08 eV and
122
Nb3d5/2 at 207.28 eV. It is observed that the peaks obtained were in agreement with previous
123
reports [43], indicating that the film consists of stoichiometry Nb2O5. The inset of Fig. 3(b)
124
shows the high resolution XPS spectrum focusing on the Pt signature of Pt−Nb2O5 film. The
6
Page 6 of 37
peaks can be resolved into Pt4f peak with binding energies of 71.5 ± 0.2 eV for Pt4f7/2 and 74.4 ±
126
0.2 eV for Pt 4f5/2, respectively. These binding energies correspond to the Pt metallic state [44].
127
As can be seen in the inset of Fig. 3(c), two intense peaks centred at 341.3 ± 2 for Pd3d3/2 and
128
335.4 ± 2 for Pd3d5/2 are attributed to Pd the metallic state [45, 46]. Inset of Fig. 3(d) shows an
129
XPS spectrum of Au related area of the Au−Nb2O5 film. The Au 4f7/2 peak appears at a binding
130
energy of 83.98 eV ± 2 and the Au 4f5/2 peak appears at a binding energy of 87.71 eV± 2
131
indicating the metallic Au.
us
cr
ip t
125
Fig. 4 shows the XRD patterns of Pt−Nb2O5, Pd−Nb2O5 and Au−Nb2O5 films. All of them
133
display the Nb2O5 peaks which can be indexed to the orthorhombic phase of Nb2O5 (ICCD-27-
134
1003) depicted by the peaks of 22.6°, 28.3°, 36.6°, 42.4°, 46.2°, 49.7°, 55.1°, 58.3° and 63.1°
135
[31]. Fig. 4(a) shows the XRD pattern of Pt−Nb2O5 where the Pt peaks can be observed at 46.55°
136
and 67.85° [47]. Fig. 4(b) indicates the presence of Pd at peaks of 46.62° and 68.10°,
137
respectively, for the Pd−Nb2O5 sample [48]. Fig. 4(c) reveals the presence of three peaks located
138
at angles of 38.2°, 44.3° and 64.6°, which correspond to Au [49].
139
Ac ce p
te
d
M
an
132
140
3.2
Effect of metal catalyst on ethanol sensing
141
When the metal is brought into an intimate contact with the Nb2O5 semiconducting film, an
142
electron barrier is established at the interface. In equilibrium and in the absence of any gas
143
molecules, if the metal work function is larger than Nb2O5 electron affinity, a Schottky contact is
144
formed with a built in voltage that depends on the barrier height. For Pt, Pd and Au with
145
relatively large work functions, this barrier height is much more pronounced, which gives rise to
146
a Schottky behaviour for the contact. Nb2O5 conduction band edge is ~3.5 eV with reference to
147
vacuum. For the band gap of 3.5 eV, and when Nb2O5 is not doped, the Fermi level falls is at
7
Page 7 of 37
~5.25 eV. This means that Au makes a near ideal ohmic contact with intrinsic Nb2O5. In reality,
149
the anodized Nb2O5 is n-doped [36], which results in the electron affinity shifts closer to the
150
conduction band edge value. This means that it is likely that even the contact with Au is
151
Schottky. Fig. 5 demonstrates the I−V curves at room temperature for Pt−Nb2O5, Pd−Nb2O5 and
152
Au−Nb2O5 devices. The contact with Au is the closest to ohmic behaviour, while the Pt metal
153
establishes a Schottky contact with the largest barrier height.
cr
ip t
148
As a reference, the I−V curves of the Pd−Nb2O5 device was obtained at different operational
155
temperatures ranging from room temperature (23 °C) to 210 °C and then exposed to 1% ethanol
156
in ambient air for 10 mins. The current was set to 2 μA as at this value the voltage changes were
157
within the accepted range of the multimeter’s operation. Fig. 6 shows the I−V characteristics of
158
this device before and after the exposure to ethanol vapour. It is observed that the highest voltage
159
change occurred at the temperature of 180 °C. Based on this, as explained in the experimental
160
section and for direct comparison, the measurements of other sensors were also conducted at
161
180 °C.
te
d
M
an
us
154
Figs. 7(a) and (b) show the I−V characteristics of the sensors and dynamic response of the
163
sensors after the exposure to 1% ethanol concentration, respectively. From Figs. 7(a-i), it is
164
observed that the sensor with Pt contact gives the lowest forward current value at all voltages in
165
comparison to Pd and Au contacts (Figs. 7(a-ii,iii)). This is owing to the high Schottky barrier
166
height (SBH) for Pt−Nb2O5 device. The current is mainly governed by the SBH. The high work
167
function of the Pt contact metal leads to a high SBH and consequently a small number of free
168
charge carriers’ injection. Our results are in agreement with the work presented by Schmitz et al.
169
who observed that Pt on GaN resulted in the largest SBH as compared to Pd and Au [50].
Ac ce p
162
8
Page 8 of 37
The responses of the sensors biased at 2 μA, while being exposed to 1% ethanol vapour, are
171
shown in Fig. 7(b). It is observed that Pd−Nb2O5 showed the largest voltage shift followed by
172
Pt−Nb2O5 and Au−Nb2O5. The Pd−Nb2O5 sensor exhibited voltage shift of 2.64 V (at the baseline
173
of 14.0 V result in the response relative change of 19.3%), while Pt−Nb2O5 and Au−Nb2O5
174
sensors exhibited voltage shifts of 1.87 V (baseline of 19.9 V and the relative change of 10.1%)
175
and 1.07 V (baseline of 11.5 V and the relative change of 8.7%), respectively.
cr
ip t
170
Further measurements were conducted in the range of room temperature to 270 °C for
177
Pt−Nb2O5 and Au−Nb2O5 sensors in order to determine their optimum operating temperatures. It is
178
found that the optimum operating temperature of Pt−Nb2O5 was 200 °C, while it was 240 °C for
179
Au−Nb2O5. Fig. 8 shows the I−V characteristics and dynamic responses of the Pt–Nb2O5 and Au–
180
Nb2O5 sensors to 1 % of ethanol vapour at their respective optimum operating temperatures.
181
According to the presented results, the voltage shift of Pt–Nb2O5 and Au–Nb2O5 is smaller as
182
compared to Pd–Nb2O5 sensor. As such, it can be concluded that Pd–Nb2O5 sensor exhibits the
183
lowest operating temperature with the highest voltage shift. The voltage shift of each sensor is
184
attributed to the reduction of SBH, which can be calculated using the equations that are presented
185
as follows.
187
an
M
d
te
Ac ce p
186
us
176
According to thermionic emission theory for Schottky diodes, the dependence of forward current on the applied voltage is given by the expression [51]:
⎧ ⎛ qV ⎞ ⎫ I = I s ⎨exp ⎜ ⎟ − 1⎬ ⎝ kT ⎠ ⎭ ⎩
(1)
188
where q is the charge of one electron, V is the applied voltage, T is the absolute temperature in
189
Kelvin, k is the Boltzmann’s constant and Is is the saturation current as defined by:
9
Page 9 of 37
⎛ qΦB ⎞ I s = AA**T 2 exp⎜ − ⎟ ⎝ kT ⎠
(2)
in which A is the diode area, A** is the Richardson constant and ΦB is the zero bias Schottky
191
barrier height in eV. The SBH can be extracted from equation (2) as:
kT ⎛ AA**T 2 ⎞ ⎟ ln⎜ q ⎜⎝ I s ⎟⎠
(3)
cr
ΦB =
ip t
190
Here, Is can be obtained by the extrapolation of the current density at the zero voltage. Finally, the
193
change in Schottky barrier height can be determined from the following equation:
an
ΔΦB = (ΦBEth − ΦBAir )
us
192
(4)
where ΦBAir and ΦBEth are the SBHs in air and ethanol containing ambiences. Using equations (1)
195
to (4) the SBH differences of the sensors with Pt, Pd and Au metallic catalysts towards 1%
196
ethanol exposure were calculated and presented in Table 1. As can be seen, the Pd contact gives
197
the largest change in the barrier height, while Au based device resulted in the smallest change.
199
d
te
It has been reported that at a temperature range of ~55 to ~200 °C, ethanol molecules start to dissociate into ethoxides on Pd, Pt and Au as [52-57]:
Ac ce p
198
M
194
C2H5OH (g) • CH3CH2O (a) + H (a)
(5)
200
in which (g) and (a) stand for the gaseous and adsorbed species, respectively. Ethoxides may be
201
further oxidized to acetaldehyde:
C2H5O (a) • CH3CHO (g) + H (a)
(6)
202
It is suggested that the H atoms are further broken down to H+ and an electron charge (e−) on
203
the catalytic metal layer. The accumulated absorbed H+ ions further spills through the catalytic
204
metal and intercalate with Nb2O5 to form HxNb2O5 [58]. As the system is maintained at an
10
Page 10 of 37
205
elevated temperature, it is also possible that HxNb2O5 breaks down into reduced niobium oxide
206
and produces H2O [36]. The process can be described as: 2HxNb2O5 • xH2O + 2Nb2O(5− x/2)
(7)
207
The released electrons and the formation of HxNb2O5 or Nb2O(5−
208
depletion region in the semiconductor and cause an effective reduction in the metal-
209
semiconductor barrier height, which allows more free charge carriers to flow. Exposing the
210
device back to air (oxygen environment with no ethanol molecules) allows the surface of the
211
reduced film to revert back to the original fully oxidized state as follows:
ip t
reduce the length of
an
Nb2O(5− x/2)+ (x/4)O2 • Nb2O5
us
cr
x/2)
(8)
This means that the original SBH values are regained and the sensors’ outputs return to their
213
original baselines.
M
212
The changes of SBH of Schottky-based gas sensors with a catalytic metal of larger work
215
function than the electron affinity of the semiconductor is schematically shown in Fig. 9. The
216
original SBH is determined by the work-function of catalytic metal and the electron affinity of
217
Nb2O5. When the metal is brought to an intimate contact with Nb2O5, electrons diffuse from the
218
metal to Nb2O5 in order to achieve a constant Fermi level throughout the system, forming a
219
thermal equilibrium. This results in the upward band bending within Nb2O5 at the interface with
220
metal. Consequently, further electrons in the metal see a barrier (SBH) against their migration
221
into the semiconductor. Upon the exposure to ethanol and its interaction with Nb2O5 near the
222
contact area, the SBH is reduced. When the sensor is exposed to air, oxygen molecules are
223
adsorbed at the metal-Nb2O5 interface, result in the SBH to increase.
Ac ce p
te
d
214
224
A plot of measured voltage shifts of the sensors equipped with different catalysts, under
225
different ethanol concentrations (ranging from 0.06 to 1 %) is shown in Fig. 10. At a low
226
concentration (0.06 and 0.12%), the Pt−Nb2O5 shows a higher voltage shift as compared to
11
Page 11 of 37
Pd−Nb2O5 and Au−Nb2O5. From the plots, it is clearly seen that the Pd−Nb2O5 device exhibited the
228
largest sensitivity towards ethanol at ethanol concentrations above 0.25%. It has been suggested
229
that the change in barrier height induced by hydrogen adsorption is proportional to the hydrogen
230
coverage of the surface [59]. It has also been reported that Pt is effective to increase the response
231
of the adsorption at low concentrations of hydrogen [60]. Ethanol and hydrogen breakdowns onto
232
a catalytic metal on a metal oxide semiconductor have a certain similarity as they both produce H
233
atoms in the process. Due to this, we concluded that they show a reasonably similar catalytic
234
behaviour towards ethanol. These explain why the barrier height changes of the Pd• Nb2O5 sensor
235
were higher than that of the Pt• Nb2O5 sensor at high concentrations of ethanol vapour.
an
us
cr
ip t
227
The response times were measured as time taken needs to reach 90% of its maximum response
237
and on the other hand the recovery time was defined as time that sensor needs to recover to its
238
90% of its baseline and are presented in Table 2. Generally, it is observed that Pd−Nb2O5 exhibit
239
faster response as compared to Pt−Nb2O5 and Au−Nb2O5. Response and recovery times for all
240
sensors display a strong dependence on the ethanol concentration. An increase in ethanol
241
concentration provides more ethanol atoms to be adsorbed on the oxide surface per unit time,
242
thereby favouring the fast electron kinetics. In addition, the increase of ethanol concentration can
243
increase the probability of collision between the atoms and then increase the reaction ratio [29].
244
On the other hand, the recovery times for all sensors were longer due to the slow desorption
245
kinetics of the ethanol vapour from the interface. As can be seen, the device with the Pd catalytic
246
metal had the shortest response, while the response for the Au−Nb2O5 was the slowest.
Ac ce p
te
d
M
236
247
For the Pt and Au samples the recovery times at different concentrations are very close.
248
However, for the Pd sample the recovery time is much smaller at low concentrations than that of
249
the higher concentrations. As Pd is more active, it reacts three times faster than Pt with hydrogen
250
containing gas species [61]. Such faster interaction is likely to translate into shorter responses at
12
Page 12 of 37
251
lower concentrations of ethanol gas for the Pd-Nb2O5 based sensor. Obviously as the gas
252
molecules interaction with Pt and Au are not as fast, there is a minimal difference in the recovery
253
times between the low and high ethanol concentrations for Pt- and Au-Nb2O5 based sensors. As mentioned earlier, ethanol and hydrogen gas molecules breaking downs onto a catalytic
255
metal show certain similarities. It is then logical to compare our observations with previous
256
reports comparing the behaviors of Pt, Pd and Au on metal oxide semiconductors in response to
257
hydrogen. Armgath et al. have reported that the hydrogen solubility in Pd is about three orders
258
higher than the hydrogen solubility in Pt [61]. Due to this, it is suggested that similarly the
259
response of ethanol on the Pd device should be larger than that of the Pt device, which describes
260
why the Pd−Nb2O5 sensor response is almost twice as large as Pt−Nb2O5 device. It is believed that
261
the relatively small shift of the Au−Nb2O5 sensor as compared to the Pd−Nb2O5 and Pt−Nb2O5 is
262
mainly due to the mechanism of the hydrogen adsorption onto Au. It has been reported that the
263
H–Au interaction is weaker than the H–Pt interaction [62]. On Au, a higher hydrogen pressure is
264
required to reach the same fractional coverage as on Pt. Moreover on Au, the activation energy
265
for the dissociation of hydrogen is higher than on Pt. These results are also in agreement with
266
those reported by Yamamoto et al. They reported that their TiO2 Schottky diode based gas
267
sensors sensitivity to hydrogen decreased in the order of Pd > Pt > Au [63]. Additionally, Penza
268
et al. have investigated the effect of catalyst on WO3 sensors and concluded that devices with the
269
Pd catalyst provided the highest sensitivity towards hydrogen as compared to Pt and Au [64].
Ac ce p
te
d
M
an
us
cr
ip t
254
270
In order to investigate the selectivity of the sensors, further experiments were conducted. All
271
sensors were exposed to other gases including hydrogen and methane of the same concentration
272
(1% vol). Fig. 11 shows the dynamic responses, while Fig. 12 shows the voltage shifts of each
273
sensor under exposure of 1 % target gases at 180 °C. It is found that all the voltage shifts under
274
hydrogen gas exposure is slightly higher than that of to ethanol vapour. However, the exposures
13
Page 13 of 37
275
to methane gas resulted in a relatively smaller voltage shifts. Thus, the sensors exhibited a
276
stronger selectivity to ethanol and hydrogen and were less sensitive to methane gas. The Schottky
277
barrier height change of the three sensors under different target vapour/gases is shown in Table 1.
ip t
278 4. Conclusions
280
This work presented the investigations of the properties of Schottky diode based devices made of
281
metal catalysts and porous Nb2O5 towards ethanol sensing. The effects of Pd, Pt and Au metal
282
catalysts on the barrier height of the contacts and sensing properies of the devices were
283
comprehensively studied. Based on the presented results, it was concluded that sensor with the
284
Pd catalyst offered the best performance. They provided the highest sensitivity. Additionally,
285
short reponse and recovery were achieved with Pd catalyst as well. The better performance of the
286
Pd−Nb2O5 gas sensor was attributed to the properties of Pd catalyst, allowing more efficient
287
breakdown of ethanol molecules and better solubility of the H atoms, which resulted in the
288
highest Schottky barrier change. The presented investigation provide an in-depth vision
289
regarding the performance of various catalytic metals in forming Schottky contacts with Nb2O5
290
and their performance as ethanol vapour sensors.
us
an
M
d
te
Ac ce p
291
cr
279
292
Acknowledgement
293
The authors acknowledge the facilities, and the scientific and technical assistance, of the
294
Australian Microscopy & Microanalysis Research Facility at the RMIT Microscopy &
295
Microanalysis Facility (RMMF), at RMIT University.
296
14
Page 14 of 37
296 297
References
298 [1] G. Wen, Y. Zhang, S. Shuang, C. Dong, M.M.F. Choi, Application of a biosensor for
300
monitoring of ethanol, Biosens. Bioelectron., 23 (2007) 121-129.
301
[2] E. Comini, G. Faglia, G. Sberveglieri, Y.X. Li, W. Wlodarski, M.K. Ghantasala, Sensitivity
302
enhancement towards ethanol and methanol of TiO2 films doped with Pt and Nb, Sens. Actuators
303
B, 64 (2000) 169-174.
304
[3] C. Lamy, A. Lima, V. LeRhun, F. Delime, C. Coutanceau, J.M. Leger, Recent advances in
305
the development of direct alcohol fuel cells (DAFC), J. Power Sources, 105 (2002) 283-296.
306
[4] Y. Lin, S. Tanaka, Ethanol fermentation from biomass resources: current state and prospects,
307
Applied Microbiol. Biot., 69 (2006) 627-642.
308
[5] G. Kampf, F.A. Pitten, P. Heeg, B. Christiansen, Efficacy of two ethanol-based skin
309
antiseptics on the forehead at shorter application times, BMC Microbiol., 7 (2007) 85.
310
[6] M.Z. Ahmad, A. Wisitsoraat, A.S. Zoolfakar, R. Ab Kadir, W. Wlodarski, Investigation of
311
RF sputtered tungsten trioxide nanorod thin film gas sensors prepared with a glancing angle
312
deposition method toward reductive and oxidative analytes, Sens. Actuators B 183 (2013) 364-
313
371.
314
[7] A.S. Zoolfakar, M.Z. Ahmad, R.A. Rani, J.Z. Ou, S. Balendhran, S. Zhuiykov, K. Latham,
315
W. Wlodarski, K. Kalantar-Zadeh, Nanostructured copper oxides as ethanol vapour sensors,
316
Sens. Actuators B, 185 (2013) 620-627.
317
[8] S.J. Ippolito, A. Ponzoni, K. Kalantar-Zadeh, W. Wlodarski, E. Comini, G. Faglia, G.
318
Sberveglieri, Layered WO3/ZnO/36° LiTaO3 SAW gas sensor sensitive towards ethanol vapour
319
and humidity, Sens. Actuators B, 117 (2006) 442-450.
Ac ce p
te
d
M
an
us
cr
ip t
299
15
Page 15 of 37
[9] S.J. Ippolito, A. Ponzoni, K. Kalantar-zadeh, W. Wlodarski, E. Comini, G. Faglia, G.
321
Sberveglieri, Ethanol sensor based on layered WO3/ZnO/36° LiTaO3 SAW devices, in:
322
Transducers '05, Digest of Technical Papers, Vols 1 and 2, 2005, pp. 1915-1918.
323
[10] C. Elosua, I. Vidondo, F.J. Arregui, C. Bariain, A. Luquin, M. Laguna, I.R. Matias, Lossy
324
mode resonance optical fiber sensor to detect organic vapors, Sens. Actuators B, 187 (2013) 65-
325
71.
326
[11] Y. Zhao, Z.C. Feng, Y. Liang, SnO2 gas sensor films deposited by pulsed laser ablation,
327
Sens. Actuators B, 56 (1999) 224-227.
328
[12] K. Potje-Kamloth, Semiconductor junction gas sensors, Chem. Rev., 108 (2008) 367-399.
329
[13] L. Wang, H. Dou, Z. Lou, T. Zhang, Encapsuled nanoreactors (Au@SnO2): a new sensing
330
material for chemical sensors, Nanoscale, 5 (2013) 2686-2691.
331
[14] A. Lajn, H.v. Wenckstern, Z. Zhang, C. Czekalla, G. Biehne, J. Lenzner, H. Hochmuth, M.
332
Lorenz, M. Grundmann, S. Wickert, C. Vogt, R. Denecke, Properties of reactively sputtered Ag,
333
Au, Pd, and Pt Schottky contacts on n-type ZnO, J. Vac. Sci. Technol. B., 27 (2009) 1769-1773.
334
[15] J. Mizsei, P. Sipilä, V. Lantto, Structural studies of sputtered noble metal catalysts on oxide
335
surfaces, Sens. Actuators B, 47 (1998) 139-144.
336
[16] I.T. Weber, A. Valentini, L.F.D. Probst, E. Longo, E.R. Leite, Influence of noble metals on
337
the structural and catalytic properties of Ce-doped SnO2 systems, Sens. Actuators B 97 (2004)
338
31-38.
339
[17] L. Wang, Z. Lou, R. Wang, T. Fei, T. Zhang, Ring-like PdO-NiO with lamellar structure for
340
gas sensor application, Journal of Materials Chemistry, 22 (2012) 12453-12456.
341
[18] Y. Hu, J. Zhou, P.-H. Yeh, Z. Li, T.-Y. Wei, Z.L. Wang, Supersensitive, fast-response
342
nanowire sensors by using Schottky contacts, Adv. Mater., 22 (2010) 3327-3332.
Ac ce p
te
d
M
an
us
cr
ip t
320
16
Page 16 of 37
[19] M. Calatayud, A. Markovits, M. Menetrey, B. Mguig, C. Minot, Adsorption on perfect and
344
reduced surfaces of metal oxides, Catal. Today, 85 (2003) 125-143.
345
[20] K. Hiehata, A. Sasahara, H. Onishi, Local work function analysis of Pt/TiO2 photocatalyst
346
by a Kelvin probe force microscope, Nanotechnology, 18 (2007).
347
[21] K.D. Schierbaum, U.K. Kirner, J.F. Geiger, W. Gopel, Schottky-barrier and conductivity gas
348
sensors based upon Pd/SnO2 and Pt/TiO2, Sens. Actuators B 4(1991) 87-94.
349
[22] T.-Y. Wei, P.-H. Yeh, S.-Y. Lu, Z. Lin-Wang, Gigantic enhancement in sensitivity using
350
Schottky contacted nanowire nanosensor, J. Am. Chem. Soc., 131 (2009) 17690-17695.
351
[23] S. Kandasamy, A. Trinchi, W. Wlodarski, E. Comini, G. Sberveglieri, Hydrogen and
352
hydrocarbon gas sensing performance of Pt/WO3/SiC MROSiC devices, Sens. Actuators B 111
353
(2005) 111-116.
354
[24] M. Shafiei, A.Z. Sadek, J. Yu, K. Latham, M. Breedon, D. McCulloch, K. Kalantar-zadeh,
355
W. Wlodarski, A hydrogen gas sensor based on Pt/nanostructured WO3/SiC Schottky diode,
356
Sens. Lett., 9 (2011) 11-15.
357
[25] M. Shafiei, J. Yu, M. Breedon, N. Motta, Q. Wu, Z. Hu, L. Qian, K. Kalantar-zadeh, W.
358
Wlodarski, Hydrogen gas sensors based on thermally evaporated nanostructured MoO3 Schottky
359
diode: A comparative study, in: Proceedings of IEEE Sensors Conference, Limerick, 2011, pp.
360
8-11.
361
[26] M. Shafiei, J. Yu, M. Breedon, A. Moafi, K. Kalantar-zadeh, W. Wlodarski, R.B. Kaner, K.
362
Galatsis, Pt/MoO3 nano-flower/SiC Schottky diode based hydrogen gas sensor, in: 9th IEEE
363
Conf. Sens. Kona, 2010, pp. 354-357.
Ac ce p
te
d
M
an
us
cr
ip t
343
17
Page 17 of 37
[27] S.J. Ippolito, S. Kandasamy, K. Kalantar-zadeh, W. Wlodarski, Hydrogen sensing
365
characteristics of WO3 thin film conductometric sensors activated by Pt and Au catalysts, Sens.
366
Actuators B 108 (2005) 154-158.
367
[28] J. Yu, Enhancement of electric field properties of Pt/nanoplatelet MoO3/SiC Schottky diode,
368
J. Phys. D. Appl. Phys., 43 (2010) 025103.
369
[29] J. Yu, S.J. Ippolito, M. Shafiei, D. Dhawan, W. Wlodarski, K. Kalantar-zadeh, Reverse
370
biased Pt/nanostructured MoO3/SiC Schottky diode based hydrogen gas sensors, Appl. Phys.
371
Lett., 94 (2009) 013504.
372
[30] R.A. Rani, A.S. Zoolfakar, J.Z. Ou, R. Ab. Kadir, H. Nili, K. Latham, S. Sriram, M.
373
Bhaskaran, S. Zhuiykov, R.B. Kaner, K. Kalantar-zadeh, Reduced impurity-driven defect states
374
in anodized nanoporous Nb2O5: the possibility of improving performance of photoanodes, Chem.
375
Commun., 49 (2013) 6349-6351.
376
[31] J.Z. Ou, R.A. Rani, M.-H. Ham, M.R. Field, Y. Zhang, H. Zheng, P. Reece, S. Zhuiykov, S.
377
Sriram, M. Bhaskaran, R.B. Kaner, K. Kalantar-zadeh, Elevated temperature anodized Nb2O5 : A
378
photoanode material with exceptionally large photoconversion efficiencies, ACS Nano, 6 (2012)
379
4045-4053.
380
[32] M.K. Siddiki, S. Venkatesan, Q. Qiao, Nb2O5 as a new electron transport layer for double
381
junction polymer solar cells, Phys. Chem. Chem. Phys., 14 (2012) 4682-4686.
382
[33] G. Ramírez, S.E. Rodil, S. Muhl, D. Turcio-Ortega, J.J. Olaya, M. Rivera, E. Camps, L.
383
Escobar-Alarcón, Amorphous niobium oxide thin films, J.Non-Cryst. Solids, 356 (2010) 2714-
384
2721.
Ac ce p
te
d
M
an
us
cr
ip t
364
18
Page 18 of 37
[34] X. Fang, L. Hu, K. Huo, B. Gao, L. Zhao, M. Liao, P.K. Chu, Y. Bando, D. Golberg, New
386
ultraviolet photodetector based on individual Nb2O5 nanobelts, Adv. Funct. Mater., 21 (2011)
387
3907-3915.
388
[35] K. Yoshimura, T. Miki, S. Iwama, S. Tanemura, Characterization of niobium oxide
389
electrochromic thin films prepared by reactive d.c. magnetron sputtering, Thin Solid Films, 282
390
(1996) 235-238.
391
[36] R.A. Rani, A.S. Zoolfakar, J.Z. Ou, M.R. Field, M. Austin, K. Kalantar-zadeh, Nanoporous
392
Nb2O5 hydrogen gas sensor, Sens. Actuators B, 176 (2013) 149-156.
393
[37] T. Hyodo, J. Ohoka, Y. Shimizu, M. Egashira, Design of anodically oxidized Nb2O5 films as
394
a diode-type H2 sensing material, Sens. Actuators B, 117 (2006) 359-366.
395
[38] Z. Wang, Y. Hu, W. Wang, X. Zhang, B. Wang, H. Tian, Y. Wang, J. Guan, H. Gu, Fast and
396
highly-sensitive hydrogen sensing of Nb2O5 nanowires at room temperature, Int. J. Hydrogen
397
Energ., 37 (2012) 4526-4532.
398
[39] Y.D. Wang, L.F. Yang, Z.L. Zhou, Y.F. Li, X.H. Wu, Effects of calcining temperature on
399
lattice constants and gas-sensing properties of Nb2O5, Mater. Lett., 49 (2001) 277-281.
400
[40] H.G. Moon, H.W. Jang, J.-S. Kim, H.-H. Park, S.-J. Yoon, A route to high sensitivity and
401
rapid response Nb2O5-based gas sensors: TiO2 doping, surface embossing, and voltage
402
optimization, Sens. Actuators B, 153 (2011) 37-43.
403
[41] J. Choi, J.H. Lim, S. Rho, D. Jahng, J. Lee, K.J. Kim, Nanoporous niobium oxide for label-
404
free detection of DNA hybridization events, Talanta, 74 (2008) 1056-1059.
405
[42] H.L. Skriver, N.M. Rosengaard, Surface energy and work function of elemental metals,
406
Phys. Rev. B, 46 (1992) 7157-7168.
Ac ce p
te
d
M
an
us
cr
ip t
385
19
Page 19 of 37
[43] S. Ge, H. Jia, H. Zhao, Z. Zheng, L. Zhang, First observation of visible light photocatalytic
408
activity of carbon modified Nb2O5 nanostructures, J. Mater. Chem., 20 (2010) 3052-3058.
409
[44] H. Yamaura, Y. Abe, K. Ino, S. Ezawa, K. Sagata, K. Ikushima, S. Yamaguchi, H. Yahiro,
410
Carbon oxidation reaction over Pt/Spherical alumina beads catalysts prepared by sputtering
411
method, Top. Catal., 53 (2010) 648-653.
412
[45] C. Varenne, A. Ndiaye, J. Brunet, G. Monier, L. Spinelle, A. Pauly, L. Bideux, B. Lauron,
413
C. Robert-Goumet, Comparison of InP Schottky diodes based on Au or Pd sensing electrodes for
414
NO2 and O3 sensing, Solid State Electron., 72 (2012) 29-37.
415
[46] Y.H. Ng, M. Wang, H. Han, C.L.L. Chai, Organic polymer composites as robust, non-
416
covalent supports of metal salts, Chem. Commun., (2009) 5530-5532.
417
[47] N. Thi Giang Huong, P. Thi Van Anh, P. Thi Xuan, L. Thi Xuan Binh, T. Van Man, N. Thi
418
Phuong Thoa, Nano-Pt/C electrocatalysts: synthesis and activity for alcohol oxidation, Adv. Nat.
419
Sci. Nanosci. Nanotechnol., 4 (2013) 035008.
420
[48] X. Liang, C.-j. Liu, P. Kuai, Selective oxidation of glucose to gluconic acid over argon
421
plasma reduced Pd/Al2O3, Green Chem., 10 (2008) 1318-1322.
422
[49] A.F. Lee, C.J. Baddeley, C. Hardacre, R.M. Ormerod, R.M. Lambert, G. Schmid, H. West,
423
Structural and catalytic properties of novel Au/Pd bimetallic colloid particles - EXAFS, XRD,
424
and acetylene coupling, J. Phys. Chem-Us, 99 (1995) 6096-6102.
425
[50] A.C. Schmitz, A.T. Ping, M.A. Khan, Q. Chen, J.W. Yang, I. Adesida, Schottky barrier
426
properties of various metals on n-type GaN, Semicond. Sci. Tech., 11 (1996) 1464-1467.
427
[51] S.M Sze, K.K. Ng, Physics of Semiconductor Device, John Wiley & Sons, Inc Publication,
428
2006.
Ac ce p
te
d
M
an
us
cr
ip t
407
20
Page 20 of 37
[52] B.A. Sexton, K.D. Rendulic, A.E. Huges, Decomposition pathways of C1-C4 alcohols
430
adsorbed on platinum (111), Surf. Sci., 121 (1982) 181-198.
431
[53] A. Gazsi, A. Koós, T. Bánsági, F. Solymosi, Adsorption and decomposition of ethanol on
432
supported Au catalysts, Catal. Today, 160 (2011) 70-78.
433
[54] P.Y. Sheng, G.A. Bowmaker, H. Idriss, The reactions of ethanol over Au/CeO2, Appl. Catal.
434
A-Gen., 261 (2004) 171-181.
435
[55] A. Yee, S.J. Morrison, H. Idriss, The reactions of ethanol over M/CeO2 catalysts - Evidence
436
of carbon-carbon bond dissociation at low temperatures over Rh/CeO2, Catal. Today, 63 (2000)
437
327-335.
438
[56] J.L. Davis, M.A. Barteau, Decarbonylation and decomposition pathways of alcohols on
439
Pd(111), Surf. Sci., 187 (1987) 387-406.
440
[57] H. Idriss, Ethanol reactions over the surfaces of noble metal/cerium oxide catalysts,
441
Platinum Met. Rev., 48 (2004) 105-115.
442
[58] D.D. Yao, R. Abdul Rani, A.P. O'Mullane, K. Kalantar-zadeh, J.Z. Ou, High performance
443
electrochromic devices based on anodized nanoporous Nb2O5, J. Phys. Chem. C, (2013).
444
[59] C.C. Cheng, Y.Y. Tsai, K.W. Lin, H.I. Chen, W.H. Hsu, H.M. Chuang, C.Y. Chen, W.C.
445
Liu, Hydrogen sensing characteristics of Pd- and Pt-Al0.3Ga0.7As metal-semiconductor (MS)
446
Schottky diodes, Semicond. Sci. Tech., 19 (2004) 778-782.
447
[60] S. Joo, I. Muto, N. Hara, Hydrogen gas sensor using Pt- and Pd-added anodic TiO2 nanotube
448
films, J. Electrochem. Soc., 157 (2010) J221-J226.
449
[61] M. Armgarth, D. Soderberg, I. Lundstrom, Palladium and platinum gate metal-oxide-
450
semiconductor capacitors in hydrogen and oxygen mixtures, Appl. Phys. Lett., 41 (1982) 654-
451
655.
Ac ce p
te
d
M
an
us
cr
ip t
429
21
Page 21 of 37
[62] E. Bus, J.A. van Bokhoven, Hydrogen chemisorption on supported platinum, gold, and
453
platinum-gold-alloy catalysts, Phys. Chem. Chem. Phys., 9 (2007) 2894-2902.
454
[63] N. Yamamoto, S. Tonomura, T. Matsuoka, H. Tsubomura, A study on a palladium-titanium
455
oxide Schottky diode as a detector for gaseous components, Surf. Sci., 92 (1980) 400-406.
456
[64] M. Penza, C. Martucci, G. Cassano, NOx gas sensing characteristics of WO3 thin films
457
activated by noble metals (Pd, Pt, Au) layers, Sens. Actuators B, 50 (1998) 52-59.
cr
ip t
452
us
458
an
459
Ac ce p
te
d
M
460
22
Page 22 of 37
us
cr
ip t
460 461
462
Fig. 1. Three dimensional schematic of the Schottky-based Nb2O5 ethanol vapour sensor.
an
463 464
Ac ce p
te
d
M
465
23
Page 23 of 37
466
Ac ce p
te
d
M
an
us
cr
ip t
465
467
Fig. 2. SEM images of a Nb2O5 nanoporous film after the annealing process: (a) cross-
468
sectional view of the whole nanoporous structure, (b) top view of the nanoporous structure,
469
(c) a higher magnification SEM image of cross-sectional view of the nanoporous structure.
24
Page 24 of 37
ip t cr us an M
470
Fig. 3 (a) XPS survey spectra of Nb2O5 (inset: XPS spectrum of Nb3d) (b) Pt−Nb2O5 (inset:
472
XPS spectrum of Pt 4f peaks) (c) Pd−Nb2O5 (inset: XPS spectrum of Pd 3d peaks and (d)
473
Au−Nb2O5 (inset: XPS spectrum of Au 4f peaks) films.
Ac ce p
te
d
471
25
Page 25 of 37
ip t cr us an M d te Ac ce p 474 475
Fig. 4 XRD patterns of the (a) Pt−Nb2O5 (b) Pd−Nb2O5 and (d) Au−Nb2O5 samples.
476
Orthorhombic Nb2O5 (ICD 27-1003) are indicated by * while Pt, Pd and Au peaks are indicated
477
by °, Δ and ♦, respectively.
26
Page 26 of 37
ip t cr us an
Fig. 5 Forward−biased I−V characteristics of (i) Au−Nb2O5, (ii) Pd−Nb2O5 and (iii) Pt−Nb2O5 in
480
air at room temperature.
483
d te
482
Ac ce p
481
M
478 479
27
Page 27 of 37
ip t cr an
us 485
M
484
Fig. 6 I−V characteristics of Pd−Nb2O5 in air and 1 % ethanol at elevated temperatures.
Ac ce p
te
d
486
28
Page 28 of 37
ip t cr us an M d te Ac ce p
487 488
Fig. 7 (a) I−V characteristics of sensors in air and 1 % ethanol (b) dynamic response to 1%
489
ethanol biased at 2 µA of (i) Pt−Nb2O5 (ii) Pd−Nb2O5 and (iii) Au−Nb2O5 Schottky gas sensors
490
at 180 °C.
491 492
29
Page 29 of 37
ip t cr us an M
493
Fig. 8 (a) I−V characteristics of sensor in air and 1 % ethanol (b) dynamic responses to
495
1% ethanol biased at 2μA of (i) Pt−Nb2O5 at 200 °C and (ii) Au−Nb2O5 at 240 °C.
te
497
Ac ce p
496
d
494
30
Page 30 of 37
ip t cr us an M d te
498
Fig. 9 Energy band diagram illustrations of a metal and Nb2O5 contact for ΦM > χ (a) is before
500
establishing the contact and (b) their corresponding band bending after establishing a contact and
501
exposure to ethanol vapour. χ is the electron affinity of Nb2O5, EFM and EFS are the Fermi levels of
502
metal and Nb2O5, respectively, EVAC is the reference vacuum level, EV and EC are the valence and
503
conduction band and the ΦBAir and ΦBEth are the SBHs in air and ethanol.
Ac ce p
499
504 505 506
31
Page 31 of 37
ip t cr us an
507
Fig. 10 Voltage shift of the sensors at different ethanol concentrations obtained at 180 °C (i)
509
Pd−Nb2O5 (ii) Pt−Nb2O5 and (iii) Au−Nb2O5.
M
508
te Ac ce p
511
d
510
32
Page 32 of 37
ip t cr us an M d te Ac ce p 511 512
Fig. 11 Dynamic response to 1 % of different vapour/gases at 180 °C of (a) Pt−Nb2O5 (b)
513
Pd−Nb2O5 and Au−Nb2O5 gas sensor.
514
33
Page 33 of 37
ip t cr us an
515
Fig. 12 Voltage shift of all sensors at concentrations of 1 % exposure to different
517
vapour/gases.
M
516
d
518
te
519
520
Ac ce p
Table 1 SBH change under different vapour/ gases SBH change, ΔΦB (eV), under Metal work different gas or vapour Sensor Metal function C2H5OH H2 CH4 Au−Nb2O5
Au
5.4
~ 0.017
~ 0.020
~ 0.008
Pd−Nb2O5
Pd
5.6
~ 0.234
~ 0.391
~ 0.203
Pt−Nb2O5
Pt
5.7
~ 0.015
~ 0.273
~ 0.007
521 522 523
34
Page 34 of 37
523 524 Table 2 Response and recovery time of Nb2O5 Schottky gas sensors with different types of metal catalysts Pt
Au
ip t
Pd
38
98
57
0.12
34
166
42
0.25
28
152
30
0.5
23
154
1.0
17
240
58
227
283
46
324
351
25
267
16
256
25
259
25
290
15
310
M
Ac ce p
te
d
525
232
an
0.06
cr
Response Recovery Response Recovery Response Recovery time (s) time (s) time (s) time (s) time (s) time (s)
us
Ethanol cconcentration (vol %)
35
Page 35 of 37
Rosmalini Ab. Kadir received her degrees in bachelor of electrical engineering (2001) and Master of Science in microelectronic (2005) from the Universiti Malaya and Universiti Kebangsaan Malaysia, Malaysia. She is currently undertaking her Ph.D. program at RMIT University, Australia in memristor and gas sensor technology. Her research interest includes
ip t
fabrication and characterization of nanostructured materials. Rozina Abdul Rani received her bachelor of engineering (Electrical and Electronic) with
cr
Honours (2001) and Master of Science (2007) degree from Universiti Sains Malaysia, Malaysia. She is currently undertaking her candidature for a Ph.D. at RMIT University. Her
sensitized solar cells, batteries and chemical sensors.
us
research interests include synthesis and characterization of nanostructured metal oxides, dye-
an
Ahmad Sabirin Zoolfakar received his bachelor of engineering (Electrical) with Honours from Universiti Malaya, Malaysia (2001) and Master of Science (Engineering) (Microelectronic Systems and Telecommunications) from The University of Liverpool, UK
M
(2003). He just finished his Ph.D. at RMIT University. His research interests include aqueous chemical synthesis, gas vapour synthesis, nanotechnology and materials science.
d
Jian Zhen Ou received his Ph.D. from RMIT University, Australia in 2012. His research
te
interests include synthesis and characterization of nanostructured metal oxides, as well as the developments of metal oxides based high performance electronic and optical devices.
Ac ce p
Mahnaz Shafiei received her bachelor of science in electrical and electronics engineering from Amir Kabir University of Technology (Tehran Polytechnique), Iran in 1998. She completed her Ph.D. at RMIT University, Melbourne, Australia in 2011. She is currently a postdoctoral research fellow at QUT, Queensland, Australia. Her major area of research includes chemical sensors, nanotechnology and material science. Wojtek Wlodarski has worked in the areas of sensor technology and instrumentation for over 35 years in the USA, Canada, Holland, France, Poland, and currently in Australia. He has published 4 books and monographs, several book chapters, over 450 papers and holds 29 patents. He is a full professor at RMIT University, Melbourne, Australia, and heads the Sensor Technology laboratory at School of Electrical and Computer Engineering. His research interests include chemical, physical and biosensors, nanotechnology, materials science and instrumentation.
Page 36 of 37
Kourosh Kalantar-zadeh is a Professor at RMIT University, Australia. He received his Bachelor of Science. (1993) and Master of Science (1997) degrees from Sharif University of Technology, Iran, and Tehran University, Iran, respectively, and a Ph.D. at RMIT University, Australia (2001). His research interests include chemical and biochemical sensors,
ip t
nanotechnology, micro systems, materials sciences, electronic circuits, and microfluidics. He
cr
is the author of over 200 scientific manuscripts.
Ac ce p
te
d
M
an
us
Page 37 of 37
S0925-4005(14)00494-8 http://dx.doi.org/doi:10.1016/j.snb.2014.04.083 SNB 16849
To appear in:
Sensors and Actuators B
Received date: Revised date: Accepted date:
27-12-2013 9-4-2014 24-4-2014
Please cite this article as: R.A. Kadir, R.A. Rani, A.S. Zoolfakar, J.Z. Ou, M. Shafiei, W. Wlodarski, K. Kalantar-zadeh, Nb2 O5 Schottky based ethanol vapour sensors: effect of metallic catalysts, Sensors and Actuators B: Chemical (2014), http://dx.doi.org/10.1016/j.snb.2014.04.083 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Nb2O5 Schottky based ethanol vapour sensors: effect
2
of metallic catalysts
3
Rosmalini Ab Kadir1,*, Rozina Abdul Rani1, Ahmad Sabirin Zoolfakar1, Jian Zhen Ou1,
4
Mahnaz Shafiei2, Wojtek Wlodarski1 and Kourosh Kalantar-zadeh1,*
cr
ip t
1
1
6
8
School of Electrical and Computer Engineering, RMIT University, Melbourne, VIC, Australia
2
Chemistry, Physics and Mechanical Engineering School, Queensland University of Technology,
an
7
us
5
Brisbane, QLD, Australia
10
M
9
Keywords: Niobium oxide, anodization, ethanol vapour, gas sensor, catalytic effect
14
te
13
Ac ce p
12
d
11
1
Page 1 of 37
ABSTRACT: The aim of this paper is to compare the performances of the highly porous Nb2O5
15
Schottky based sensors formed using different catalytic metals for ethanol vapour sensing. The
16
fabricated sensors consist of a fairly ordered nano-vein like porous Nb2O5 prepared via an
17
elevated temperature anodization method. Subsequently, Pt, Pd and Au were sputtered as both
18
Schottky contacts and catalysts for the comparative studies. These metals are chosen as they have
19
large work functions in comparison to the electron affinity of the anodized Nb2O5. It is
20
demonstrated that the device based on Pd/Nb2O5 Schottky contact has the highest sensitivity
21
amongst the developed sensors. The sensing behaviors were studied in terms of the Schottky
22
barrier height variations and properties of the metal catalysts.
an
us
cr
ip t
14
23 1. Introduction
25
Ethanol vapour sensors are in always in demand due to their significant applications in food,
26
biomedical, transportation and chemical industries as well as health and safety [1, 2]. Ethanol is
27
also commonly used as a fuel or a fuel additive. It is either utilized or appears as a by-product of
28
many chemical processes and is a common solvent in many industries. Ethanol is employed in
29
alcoholic beverages and is also a famous antiseptic. Obviously, sensing ethanol is of importance
30
for all the aforementioned applications [3-5].
Ac ce p
te
d
M
24
31
There are many types of ethanol vapour sensors based on different transducers such as
32
conductometric, surface acoustic wave, electrochemical, optical and Schottky diode based [6-11].
33
Amongst these transducers, Schottky diodes with nanostructured metal oxide sensing elements
34
offer simplicity of fabrication; reliable robustness and high sensitivity towards ethanol vapour
35
[12]. A Schottky diode based sensor generally consists of a metal-semiconductor contact. The
36
most common choices of metals are catalytic type noble metals including gold (Au), platinum
37
(Pt) and palladium (Pd) [13-17]. They catalyse the target gas molecules, while regain their
2
Page 2 of 37
metallic properties after the interaction. Additionally with their relatively large work functions,
39
these metals are likely to form Schottky contacts at their interface with most of the metal oxide
40
semiconductors [18]. Metal catalysts are employed in the chemical industry for numerous
41
chemical reactions. They operate by interacting with the target molecules, lowering the activation
42
barriers and hence increasing the rate of reactions. The metal catalysts also enhance the
43
sensitivity and reduce response and recovery time of the sensor. Au, Pt and Pd in the Schottky
44
diode based sensors enable catalytic decomposition of ethanol, which otherwise would not be
45
possible on the oxide surface. After the adsorption of ethanol, the local electronic properties of
46
the metal/metal oxide semiconductor changes that results in a change in the Schottky contact
47
behaviour and this alteration is used as the sensing parameter [19, 20].
an
us
cr
ip t
38
Metal oxides that have been widely investigated in Schottky diode based ethanol sensors
49
include SnO2, TiO2, ZnO, WO3, MoO3 and Ga2O3 [21-29]. Niobium oxide (Nb2O5) is an important
50
but yet much less studied metal oxide for Schottky diode based ethanol sensing. It is a n-type
51
semiconductor with a ~3.5 eV band gap that have shown to offer unique properties for
52
developing highly efficient solar-cells, electrochromic devices and sensors [30-35]. Devices
53
based on Nb2O5 have been reported for sensing of various gases and chemical components [36-
54
41]. Numerous methods have been established for fabricating Nb2O5 based gas sensing devices.
55
For example, Hyodo et al. and Rozina et al. have used Nb2O5 prepared via an anodization method
56
[36, 37], while Nb2O5 nanowires were synthesized by Wang et al. through thermal oxidation
57
process for hydrogen gas sensing [38]. Nb2O5 thin films doped with TiO2 have also been
58
developed via sputtering method for carbon monoxide gas sensing [40]. There are also
59
investigations regarding the effect of the calcination temperature on gas sensing properties of
60
Nb2O5 [39].
Ac ce p
te
d
M
48
3
Page 3 of 37
In this work, the Nb2O5 Schottky diode based sensors, integrated with different catalytic
62
metals, are developed and investigated. Three noble metals of Pt, Pd, and Au with different work
63
functions of 5.7, 5.6 and 5.4 eV respectively, are employed [42]. The effects of Pt, Pd and Au on
64
the ethanol sensing performances of the Schottky diode based sensors are fully studied and
65
discussed.
ip t
61
2. Experimental
us
67
cr
66
68 2.1. Fabrication
70
The Schottky diode based sensor devices were fabricated using niobium foil (99.9% purity,
71
Sigma Aldrich) of 0.25 mm thickness. The foil was cut into pieces of 1.0 cm × 1.0 cm, which
72
were cleaned with acetone, followed by isopropanol and rinsed with deionized water then dried
73
in a stream of high purity nitrogen gas. Nb2O5 film synthesized via an anodization process that
74
was conducted at 50 °C in the electrolyte of NH4F/ethylene glycol, containing a small amount of
75
water using a process that was based on our previous work [36]. Films were then annealed at 440
76
°C for 30 min at the ramp up/down of 2 °C/min for crystallization. The final stage of the
77
fabrication was the formation of the Schottky contacts. For this study, Pt, Pd and Au layers with
78
the thickness of ~20 nm each were deposited on the annealed Nb2O5 film using a GATAN
79
PECSTM sputter coating system. A shadow mask with a 2.5 mm diameter hole was utilized to
80
obtain a circular pad of catalytic contact. Sensor with Pt, Pd and Au contacts will be addressed as
81
Pt−Nb2O5, Pd−Nb2O5 and Au−Nb2O5, respectively, throughout this paper.
82
2.2. Structural characterizations.
Ac ce p
te
d
M
an
69
83
The morphologies of Nb2O5 films were studied using scanning electron microscopy (SEM -
84
FEI Nova NanoSEM), while the crystallographic properties of the films were studied via X-ray
4
Page 4 of 37
diffraction (XRD) analysis using a Bruker D8 DISCOVER microdiffractometer fitted with a
86
general area detector diffraction system (GADDS) that accommodates Cu-Kα radiation (l =
87
1.54178 Å). Analysis of the X-ray photoelectron spectrometry (XPS) was conducted using
88
Thermo Scientific K-alpha X-ray Photoelectron.
89
d
M
an
us
cr
ip t
85
2.3. Measurement setup
91
The gas sensing measurements devices were conducted in a LINKAM (Scientific Instruments)
92
customized gas testing chamber. A computerized mass flow control (MFC) multi-channel gas
93
calibration system was used, which allowed the sensors to be exposed to ethanol with different
94
concentrations by diluting it using 99.99 % dry synthetic air. This is carried out by adjusting the
95
flow rates of each gas channel, producing a specified concentration of the analyte vapour species
96
flowing at a total constant flow rate of 200 sccm to the LINKAM chamber. Electrical
97
connections to the Schottky gas sensors were achieved by physically connecting the probes to the
98
metal pads (top electrode) and Nb (bottom electrode) as schematically shown in Fig. 1. The
99
current−voltage (I−V) measurements were carried out using Keithley 2606 current source meter
100
as they were sequentially heated from room temperature to 270 °C and exposed to a mixture of
Ac ce p
te
90
5
Page 5 of 37
ethanol vapour and air. The sensors were mounted on heaters, which controlled their operational
102
temperatures. The measurements started with the Pd−Nb2O5 sensor, and it was observed that the
103
largest voltage shift occurred at 180 °C. Due to this observation and for comparison purposes, the
104
dynamic measurements of other sensors were also obtained at 180 °C. The dynamic responses
105
were measured as a change in the voltage magnitude utilizing an Agilent 34410A digital
106
multimeter to record the voltage alterations under the exposure to ethanol vapour with different
107
concentrations in air.
108 109
3. Results and discussion
110 111
3.1
112
The SEM images of the cross-sectional and top views of nanoporous Nb2O5 after the annodization
113
and annealing processes are presented in Fig. 2. The thickness of the film in this figure is ~1 μm.
114
From Fig. 2(b) it is observed that Nb2O5 films are made of fairly organized pores, with inner
115
diameters ranging from 30 to 50 nm. Fig. 2(c) shows the higher magnification cross-sectional
116
images of the nanoporous Nb2O5 film. It is observed that the film is made of highly packed vein-
117
like nanostructured networks.
Ac ce p
te
d
M
Morphology and structural properties
an
us
cr
ip t
101
118
The XPS analysis was carried out in order to investigate the elemental composition of the
119
sensing films. As depicted in Fig 3(a-d), XPS survey spectra of the sensors show the presence of
120
Nb, O, Pt, Pd and Au elements with very little carbon as a contaminant. Inset of Fig. 3(a), shows
121
the high-resolution spectrum for the Nb region that exhibits the Nb3d3/2 peaks at 210.08 eV and
122
Nb3d5/2 at 207.28 eV. It is observed that the peaks obtained were in agreement with previous
123
reports [43], indicating that the film consists of stoichiometry Nb2O5. The inset of Fig. 3(b)
124
shows the high resolution XPS spectrum focusing on the Pt signature of Pt−Nb2O5 film. The
6
Page 6 of 37
peaks can be resolved into Pt4f peak with binding energies of 71.5 ± 0.2 eV for Pt4f7/2 and 74.4 ±
126
0.2 eV for Pt 4f5/2, respectively. These binding energies correspond to the Pt metallic state [44].
127
As can be seen in the inset of Fig. 3(c), two intense peaks centred at 341.3 ± 2 for Pd3d3/2 and
128
335.4 ± 2 for Pd3d5/2 are attributed to Pd the metallic state [45, 46]. Inset of Fig. 3(d) shows an
129
XPS spectrum of Au related area of the Au−Nb2O5 film. The Au 4f7/2 peak appears at a binding
130
energy of 83.98 eV ± 2 and the Au 4f5/2 peak appears at a binding energy of 87.71 eV± 2
131
indicating the metallic Au.
us
cr
ip t
125
Fig. 4 shows the XRD patterns of Pt−Nb2O5, Pd−Nb2O5 and Au−Nb2O5 films. All of them
133
display the Nb2O5 peaks which can be indexed to the orthorhombic phase of Nb2O5 (ICCD-27-
134
1003) depicted by the peaks of 22.6°, 28.3°, 36.6°, 42.4°, 46.2°, 49.7°, 55.1°, 58.3° and 63.1°
135
[31]. Fig. 4(a) shows the XRD pattern of Pt−Nb2O5 where the Pt peaks can be observed at 46.55°
136
and 67.85° [47]. Fig. 4(b) indicates the presence of Pd at peaks of 46.62° and 68.10°,
137
respectively, for the Pd−Nb2O5 sample [48]. Fig. 4(c) reveals the presence of three peaks located
138
at angles of 38.2°, 44.3° and 64.6°, which correspond to Au [49].
139
Ac ce p
te
d
M
an
132
140
3.2
Effect of metal catalyst on ethanol sensing
141
When the metal is brought into an intimate contact with the Nb2O5 semiconducting film, an
142
electron barrier is established at the interface. In equilibrium and in the absence of any gas
143
molecules, if the metal work function is larger than Nb2O5 electron affinity, a Schottky contact is
144
formed with a built in voltage that depends on the barrier height. For Pt, Pd and Au with
145
relatively large work functions, this barrier height is much more pronounced, which gives rise to
146
a Schottky behaviour for the contact. Nb2O5 conduction band edge is ~3.5 eV with reference to
147
vacuum. For the band gap of 3.5 eV, and when Nb2O5 is not doped, the Fermi level falls is at
7
Page 7 of 37
~5.25 eV. This means that Au makes a near ideal ohmic contact with intrinsic Nb2O5. In reality,
149
the anodized Nb2O5 is n-doped [36], which results in the electron affinity shifts closer to the
150
conduction band edge value. This means that it is likely that even the contact with Au is
151
Schottky. Fig. 5 demonstrates the I−V curves at room temperature for Pt−Nb2O5, Pd−Nb2O5 and
152
Au−Nb2O5 devices. The contact with Au is the closest to ohmic behaviour, while the Pt metal
153
establishes a Schottky contact with the largest barrier height.
cr
ip t
148
As a reference, the I−V curves of the Pd−Nb2O5 device was obtained at different operational
155
temperatures ranging from room temperature (23 °C) to 210 °C and then exposed to 1% ethanol
156
in ambient air for 10 mins. The current was set to 2 μA as at this value the voltage changes were
157
within the accepted range of the multimeter’s operation. Fig. 6 shows the I−V characteristics of
158
this device before and after the exposure to ethanol vapour. It is observed that the highest voltage
159
change occurred at the temperature of 180 °C. Based on this, as explained in the experimental
160
section and for direct comparison, the measurements of other sensors were also conducted at
161
180 °C.
te
d
M
an
us
154
Figs. 7(a) and (b) show the I−V characteristics of the sensors and dynamic response of the
163
sensors after the exposure to 1% ethanol concentration, respectively. From Figs. 7(a-i), it is
164
observed that the sensor with Pt contact gives the lowest forward current value at all voltages in
165
comparison to Pd and Au contacts (Figs. 7(a-ii,iii)). This is owing to the high Schottky barrier
166
height (SBH) for Pt−Nb2O5 device. The current is mainly governed by the SBH. The high work
167
function of the Pt contact metal leads to a high SBH and consequently a small number of free
168
charge carriers’ injection. Our results are in agreement with the work presented by Schmitz et al.
169
who observed that Pt on GaN resulted in the largest SBH as compared to Pd and Au [50].
Ac ce p
162
8
Page 8 of 37
The responses of the sensors biased at 2 μA, while being exposed to 1% ethanol vapour, are
171
shown in Fig. 7(b). It is observed that Pd−Nb2O5 showed the largest voltage shift followed by
172
Pt−Nb2O5 and Au−Nb2O5. The Pd−Nb2O5 sensor exhibited voltage shift of 2.64 V (at the baseline
173
of 14.0 V result in the response relative change of 19.3%), while Pt−Nb2O5 and Au−Nb2O5
174
sensors exhibited voltage shifts of 1.87 V (baseline of 19.9 V and the relative change of 10.1%)
175
and 1.07 V (baseline of 11.5 V and the relative change of 8.7%), respectively.
cr
ip t
170
Further measurements were conducted in the range of room temperature to 270 °C for
177
Pt−Nb2O5 and Au−Nb2O5 sensors in order to determine their optimum operating temperatures. It is
178
found that the optimum operating temperature of Pt−Nb2O5 was 200 °C, while it was 240 °C for
179
Au−Nb2O5. Fig. 8 shows the I−V characteristics and dynamic responses of the Pt–Nb2O5 and Au–
180
Nb2O5 sensors to 1 % of ethanol vapour at their respective optimum operating temperatures.
181
According to the presented results, the voltage shift of Pt–Nb2O5 and Au–Nb2O5 is smaller as
182
compared to Pd–Nb2O5 sensor. As such, it can be concluded that Pd–Nb2O5 sensor exhibits the
183
lowest operating temperature with the highest voltage shift. The voltage shift of each sensor is
184
attributed to the reduction of SBH, which can be calculated using the equations that are presented
185
as follows.
187
an
M
d
te
Ac ce p
186
us
176
According to thermionic emission theory for Schottky diodes, the dependence of forward current on the applied voltage is given by the expression [51]:
⎧ ⎛ qV ⎞ ⎫ I = I s ⎨exp ⎜ ⎟ − 1⎬ ⎝ kT ⎠ ⎭ ⎩
(1)
188
where q is the charge of one electron, V is the applied voltage, T is the absolute temperature in
189
Kelvin, k is the Boltzmann’s constant and Is is the saturation current as defined by:
9
Page 9 of 37
⎛ qΦB ⎞ I s = AA**T 2 exp⎜ − ⎟ ⎝ kT ⎠
(2)
in which A is the diode area, A** is the Richardson constant and ΦB is the zero bias Schottky
191
barrier height in eV. The SBH can be extracted from equation (2) as:
kT ⎛ AA**T 2 ⎞ ⎟ ln⎜ q ⎜⎝ I s ⎟⎠
(3)
cr
ΦB =
ip t
190
Here, Is can be obtained by the extrapolation of the current density at the zero voltage. Finally, the
193
change in Schottky barrier height can be determined from the following equation:
an
ΔΦB = (ΦBEth − ΦBAir )
us
192
(4)
where ΦBAir and ΦBEth are the SBHs in air and ethanol containing ambiences. Using equations (1)
195
to (4) the SBH differences of the sensors with Pt, Pd and Au metallic catalysts towards 1%
196
ethanol exposure were calculated and presented in Table 1. As can be seen, the Pd contact gives
197
the largest change in the barrier height, while Au based device resulted in the smallest change.
199
d
te
It has been reported that at a temperature range of ~55 to ~200 °C, ethanol molecules start to dissociate into ethoxides on Pd, Pt and Au as [52-57]:
Ac ce p
198
M
194
C2H5OH (g) • CH3CH2O (a) + H (a)
(5)
200
in which (g) and (a) stand for the gaseous and adsorbed species, respectively. Ethoxides may be
201
further oxidized to acetaldehyde:
C2H5O (a) • CH3CHO (g) + H (a)
(6)
202
It is suggested that the H atoms are further broken down to H+ and an electron charge (e−) on
203
the catalytic metal layer. The accumulated absorbed H+ ions further spills through the catalytic
204
metal and intercalate with Nb2O5 to form HxNb2O5 [58]. As the system is maintained at an
10
Page 10 of 37
205
elevated temperature, it is also possible that HxNb2O5 breaks down into reduced niobium oxide
206
and produces H2O [36]. The process can be described as: 2HxNb2O5 • xH2O + 2Nb2O(5− x/2)
(7)
207
The released electrons and the formation of HxNb2O5 or Nb2O(5−
208
depletion region in the semiconductor and cause an effective reduction in the metal-
209
semiconductor barrier height, which allows more free charge carriers to flow. Exposing the
210
device back to air (oxygen environment with no ethanol molecules) allows the surface of the
211
reduced film to revert back to the original fully oxidized state as follows:
ip t
reduce the length of
an
Nb2O(5− x/2)+ (x/4)O2 • Nb2O5
us
cr
x/2)
(8)
This means that the original SBH values are regained and the sensors’ outputs return to their
213
original baselines.
M
212
The changes of SBH of Schottky-based gas sensors with a catalytic metal of larger work
215
function than the electron affinity of the semiconductor is schematically shown in Fig. 9. The
216
original SBH is determined by the work-function of catalytic metal and the electron affinity of
217
Nb2O5. When the metal is brought to an intimate contact with Nb2O5, electrons diffuse from the
218
metal to Nb2O5 in order to achieve a constant Fermi level throughout the system, forming a
219
thermal equilibrium. This results in the upward band bending within Nb2O5 at the interface with
220
metal. Consequently, further electrons in the metal see a barrier (SBH) against their migration
221
into the semiconductor. Upon the exposure to ethanol and its interaction with Nb2O5 near the
222
contact area, the SBH is reduced. When the sensor is exposed to air, oxygen molecules are
223
adsorbed at the metal-Nb2O5 interface, result in the SBH to increase.
Ac ce p
te
d
214
224
A plot of measured voltage shifts of the sensors equipped with different catalysts, under
225
different ethanol concentrations (ranging from 0.06 to 1 %) is shown in Fig. 10. At a low
226
concentration (0.06 and 0.12%), the Pt−Nb2O5 shows a higher voltage shift as compared to
11
Page 11 of 37
Pd−Nb2O5 and Au−Nb2O5. From the plots, it is clearly seen that the Pd−Nb2O5 device exhibited the
228
largest sensitivity towards ethanol at ethanol concentrations above 0.25%. It has been suggested
229
that the change in barrier height induced by hydrogen adsorption is proportional to the hydrogen
230
coverage of the surface [59]. It has also been reported that Pt is effective to increase the response
231
of the adsorption at low concentrations of hydrogen [60]. Ethanol and hydrogen breakdowns onto
232
a catalytic metal on a metal oxide semiconductor have a certain similarity as they both produce H
233
atoms in the process. Due to this, we concluded that they show a reasonably similar catalytic
234
behaviour towards ethanol. These explain why the barrier height changes of the Pd• Nb2O5 sensor
235
were higher than that of the Pt• Nb2O5 sensor at high concentrations of ethanol vapour.
an
us
cr
ip t
227
The response times were measured as time taken needs to reach 90% of its maximum response
237
and on the other hand the recovery time was defined as time that sensor needs to recover to its
238
90% of its baseline and are presented in Table 2. Generally, it is observed that Pd−Nb2O5 exhibit
239
faster response as compared to Pt−Nb2O5 and Au−Nb2O5. Response and recovery times for all
240
sensors display a strong dependence on the ethanol concentration. An increase in ethanol
241
concentration provides more ethanol atoms to be adsorbed on the oxide surface per unit time,
242
thereby favouring the fast electron kinetics. In addition, the increase of ethanol concentration can
243
increase the probability of collision between the atoms and then increase the reaction ratio [29].
244
On the other hand, the recovery times for all sensors were longer due to the slow desorption
245
kinetics of the ethanol vapour from the interface. As can be seen, the device with the Pd catalytic
246
metal had the shortest response, while the response for the Au−Nb2O5 was the slowest.
Ac ce p
te
d
M
236
247
For the Pt and Au samples the recovery times at different concentrations are very close.
248
However, for the Pd sample the recovery time is much smaller at low concentrations than that of
249
the higher concentrations. As Pd is more active, it reacts three times faster than Pt with hydrogen
250
containing gas species [61]. Such faster interaction is likely to translate into shorter responses at
12
Page 12 of 37
251
lower concentrations of ethanol gas for the Pd-Nb2O5 based sensor. Obviously as the gas
252
molecules interaction with Pt and Au are not as fast, there is a minimal difference in the recovery
253
times between the low and high ethanol concentrations for Pt- and Au-Nb2O5 based sensors. As mentioned earlier, ethanol and hydrogen gas molecules breaking downs onto a catalytic
255
metal show certain similarities. It is then logical to compare our observations with previous
256
reports comparing the behaviors of Pt, Pd and Au on metal oxide semiconductors in response to
257
hydrogen. Armgath et al. have reported that the hydrogen solubility in Pd is about three orders
258
higher than the hydrogen solubility in Pt [61]. Due to this, it is suggested that similarly the
259
response of ethanol on the Pd device should be larger than that of the Pt device, which describes
260
why the Pd−Nb2O5 sensor response is almost twice as large as Pt−Nb2O5 device. It is believed that
261
the relatively small shift of the Au−Nb2O5 sensor as compared to the Pd−Nb2O5 and Pt−Nb2O5 is
262
mainly due to the mechanism of the hydrogen adsorption onto Au. It has been reported that the
263
H–Au interaction is weaker than the H–Pt interaction [62]. On Au, a higher hydrogen pressure is
264
required to reach the same fractional coverage as on Pt. Moreover on Au, the activation energy
265
for the dissociation of hydrogen is higher than on Pt. These results are also in agreement with
266
those reported by Yamamoto et al. They reported that their TiO2 Schottky diode based gas
267
sensors sensitivity to hydrogen decreased in the order of Pd > Pt > Au [63]. Additionally, Penza
268
et al. have investigated the effect of catalyst on WO3 sensors and concluded that devices with the
269
Pd catalyst provided the highest sensitivity towards hydrogen as compared to Pt and Au [64].
Ac ce p
te
d
M
an
us
cr
ip t
254
270
In order to investigate the selectivity of the sensors, further experiments were conducted. All
271
sensors were exposed to other gases including hydrogen and methane of the same concentration
272
(1% vol). Fig. 11 shows the dynamic responses, while Fig. 12 shows the voltage shifts of each
273
sensor under exposure of 1 % target gases at 180 °C. It is found that all the voltage shifts under
274
hydrogen gas exposure is slightly higher than that of to ethanol vapour. However, the exposures
13
Page 13 of 37
275
to methane gas resulted in a relatively smaller voltage shifts. Thus, the sensors exhibited a
276
stronger selectivity to ethanol and hydrogen and were less sensitive to methane gas. The Schottky
277
barrier height change of the three sensors under different target vapour/gases is shown in Table 1.
ip t
278 4. Conclusions
280
This work presented the investigations of the properties of Schottky diode based devices made of
281
metal catalysts and porous Nb2O5 towards ethanol sensing. The effects of Pd, Pt and Au metal
282
catalysts on the barrier height of the contacts and sensing properies of the devices were
283
comprehensively studied. Based on the presented results, it was concluded that sensor with the
284
Pd catalyst offered the best performance. They provided the highest sensitivity. Additionally,
285
short reponse and recovery were achieved with Pd catalyst as well. The better performance of the
286
Pd−Nb2O5 gas sensor was attributed to the properties of Pd catalyst, allowing more efficient
287
breakdown of ethanol molecules and better solubility of the H atoms, which resulted in the
288
highest Schottky barrier change. The presented investigation provide an in-depth vision
289
regarding the performance of various catalytic metals in forming Schottky contacts with Nb2O5
290
and their performance as ethanol vapour sensors.
us
an
M
d
te
Ac ce p
291
cr
279
292
Acknowledgement
293
The authors acknowledge the facilities, and the scientific and technical assistance, of the
294
Australian Microscopy & Microanalysis Research Facility at the RMIT Microscopy &
295
Microanalysis Facility (RMMF), at RMIT University.
296
14
Page 14 of 37
296 297
References
298 [1] G. Wen, Y. Zhang, S. Shuang, C. Dong, M.M.F. Choi, Application of a biosensor for
300
monitoring of ethanol, Biosens. Bioelectron., 23 (2007) 121-129.
301
[2] E. Comini, G. Faglia, G. Sberveglieri, Y.X. Li, W. Wlodarski, M.K. Ghantasala, Sensitivity
302
enhancement towards ethanol and methanol of TiO2 films doped with Pt and Nb, Sens. Actuators
303
B, 64 (2000) 169-174.
304
[3] C. Lamy, A. Lima, V. LeRhun, F. Delime, C. Coutanceau, J.M. Leger, Recent advances in
305
the development of direct alcohol fuel cells (DAFC), J. Power Sources, 105 (2002) 283-296.
306
[4] Y. Lin, S. Tanaka, Ethanol fermentation from biomass resources: current state and prospects,
307
Applied Microbiol. Biot., 69 (2006) 627-642.
308
[5] G. Kampf, F.A. Pitten, P. Heeg, B. Christiansen, Efficacy of two ethanol-based skin
309
antiseptics on the forehead at shorter application times, BMC Microbiol., 7 (2007) 85.
310
[6] M.Z. Ahmad, A. Wisitsoraat, A.S. Zoolfakar, R. Ab Kadir, W. Wlodarski, Investigation of
311
RF sputtered tungsten trioxide nanorod thin film gas sensors prepared with a glancing angle
312
deposition method toward reductive and oxidative analytes, Sens. Actuators B 183 (2013) 364-
313
371.
314
[7] A.S. Zoolfakar, M.Z. Ahmad, R.A. Rani, J.Z. Ou, S. Balendhran, S. Zhuiykov, K. Latham,
315
W. Wlodarski, K. Kalantar-Zadeh, Nanostructured copper oxides as ethanol vapour sensors,
316
Sens. Actuators B, 185 (2013) 620-627.
317
[8] S.J. Ippolito, A. Ponzoni, K. Kalantar-Zadeh, W. Wlodarski, E. Comini, G. Faglia, G.
318
Sberveglieri, Layered WO3/ZnO/36° LiTaO3 SAW gas sensor sensitive towards ethanol vapour
319
and humidity, Sens. Actuators B, 117 (2006) 442-450.
Ac ce p
te
d
M
an
us
cr
ip t
299
15
Page 15 of 37
[9] S.J. Ippolito, A. Ponzoni, K. Kalantar-zadeh, W. Wlodarski, E. Comini, G. Faglia, G.
321
Sberveglieri, Ethanol sensor based on layered WO3/ZnO/36° LiTaO3 SAW devices, in:
322
Transducers '05, Digest of Technical Papers, Vols 1 and 2, 2005, pp. 1915-1918.
323
[10] C. Elosua, I. Vidondo, F.J. Arregui, C. Bariain, A. Luquin, M. Laguna, I.R. Matias, Lossy
324
mode resonance optical fiber sensor to detect organic vapors, Sens. Actuators B, 187 (2013) 65-
325
71.
326
[11] Y. Zhao, Z.C. Feng, Y. Liang, SnO2 gas sensor films deposited by pulsed laser ablation,
327
Sens. Actuators B, 56 (1999) 224-227.
328
[12] K. Potje-Kamloth, Semiconductor junction gas sensors, Chem. Rev., 108 (2008) 367-399.
329
[13] L. Wang, H. Dou, Z. Lou, T. Zhang, Encapsuled nanoreactors (Au@SnO2): a new sensing
330
material for chemical sensors, Nanoscale, 5 (2013) 2686-2691.
331
[14] A. Lajn, H.v. Wenckstern, Z. Zhang, C. Czekalla, G. Biehne, J. Lenzner, H. Hochmuth, M.
332
Lorenz, M. Grundmann, S. Wickert, C. Vogt, R. Denecke, Properties of reactively sputtered Ag,
333
Au, Pd, and Pt Schottky contacts on n-type ZnO, J. Vac. Sci. Technol. B., 27 (2009) 1769-1773.
334
[15] J. Mizsei, P. Sipilä, V. Lantto, Structural studies of sputtered noble metal catalysts on oxide
335
surfaces, Sens. Actuators B, 47 (1998) 139-144.
336
[16] I.T. Weber, A. Valentini, L.F.D. Probst, E. Longo, E.R. Leite, Influence of noble metals on
337
the structural and catalytic properties of Ce-doped SnO2 systems, Sens. Actuators B 97 (2004)
338
31-38.
339
[17] L. Wang, Z. Lou, R. Wang, T. Fei, T. Zhang, Ring-like PdO-NiO with lamellar structure for
340
gas sensor application, Journal of Materials Chemistry, 22 (2012) 12453-12456.
341
[18] Y. Hu, J. Zhou, P.-H. Yeh, Z. Li, T.-Y. Wei, Z.L. Wang, Supersensitive, fast-response
342
nanowire sensors by using Schottky contacts, Adv. Mater., 22 (2010) 3327-3332.
Ac ce p
te
d
M
an
us
cr
ip t
320
16
Page 16 of 37
[19] M. Calatayud, A. Markovits, M. Menetrey, B. Mguig, C. Minot, Adsorption on perfect and
344
reduced surfaces of metal oxides, Catal. Today, 85 (2003) 125-143.
345
[20] K. Hiehata, A. Sasahara, H. Onishi, Local work function analysis of Pt/TiO2 photocatalyst
346
by a Kelvin probe force microscope, Nanotechnology, 18 (2007).
347
[21] K.D. Schierbaum, U.K. Kirner, J.F. Geiger, W. Gopel, Schottky-barrier and conductivity gas
348
sensors based upon Pd/SnO2 and Pt/TiO2, Sens. Actuators B 4(1991) 87-94.
349
[22] T.-Y. Wei, P.-H. Yeh, S.-Y. Lu, Z. Lin-Wang, Gigantic enhancement in sensitivity using
350
Schottky contacted nanowire nanosensor, J. Am. Chem. Soc., 131 (2009) 17690-17695.
351
[23] S. Kandasamy, A. Trinchi, W. Wlodarski, E. Comini, G. Sberveglieri, Hydrogen and
352
hydrocarbon gas sensing performance of Pt/WO3/SiC MROSiC devices, Sens. Actuators B 111
353
(2005) 111-116.
354
[24] M. Shafiei, A.Z. Sadek, J. Yu, K. Latham, M. Breedon, D. McCulloch, K. Kalantar-zadeh,
355
W. Wlodarski, A hydrogen gas sensor based on Pt/nanostructured WO3/SiC Schottky diode,
356
Sens. Lett., 9 (2011) 11-15.
357
[25] M. Shafiei, J. Yu, M. Breedon, N. Motta, Q. Wu, Z. Hu, L. Qian, K. Kalantar-zadeh, W.
358
Wlodarski, Hydrogen gas sensors based on thermally evaporated nanostructured MoO3 Schottky
359
diode: A comparative study, in: Proceedings of IEEE Sensors Conference, Limerick, 2011, pp.
360
8-11.
361
[26] M. Shafiei, J. Yu, M. Breedon, A. Moafi, K. Kalantar-zadeh, W. Wlodarski, R.B. Kaner, K.
362
Galatsis, Pt/MoO3 nano-flower/SiC Schottky diode based hydrogen gas sensor, in: 9th IEEE
363
Conf. Sens. Kona, 2010, pp. 354-357.
Ac ce p
te
d
M
an
us
cr
ip t
343
17
Page 17 of 37
[27] S.J. Ippolito, S. Kandasamy, K. Kalantar-zadeh, W. Wlodarski, Hydrogen sensing
365
characteristics of WO3 thin film conductometric sensors activated by Pt and Au catalysts, Sens.
366
Actuators B 108 (2005) 154-158.
367
[28] J. Yu, Enhancement of electric field properties of Pt/nanoplatelet MoO3/SiC Schottky diode,
368
J. Phys. D. Appl. Phys., 43 (2010) 025103.
369
[29] J. Yu, S.J. Ippolito, M. Shafiei, D. Dhawan, W. Wlodarski, K. Kalantar-zadeh, Reverse
370
biased Pt/nanostructured MoO3/SiC Schottky diode based hydrogen gas sensors, Appl. Phys.
371
Lett., 94 (2009) 013504.
372
[30] R.A. Rani, A.S. Zoolfakar, J.Z. Ou, R. Ab. Kadir, H. Nili, K. Latham, S. Sriram, M.
373
Bhaskaran, S. Zhuiykov, R.B. Kaner, K. Kalantar-zadeh, Reduced impurity-driven defect states
374
in anodized nanoporous Nb2O5: the possibility of improving performance of photoanodes, Chem.
375
Commun., 49 (2013) 6349-6351.
376
[31] J.Z. Ou, R.A. Rani, M.-H. Ham, M.R. Field, Y. Zhang, H. Zheng, P. Reece, S. Zhuiykov, S.
377
Sriram, M. Bhaskaran, R.B. Kaner, K. Kalantar-zadeh, Elevated temperature anodized Nb2O5 : A
378
photoanode material with exceptionally large photoconversion efficiencies, ACS Nano, 6 (2012)
379
4045-4053.
380
[32] M.K. Siddiki, S. Venkatesan, Q. Qiao, Nb2O5 as a new electron transport layer for double
381
junction polymer solar cells, Phys. Chem. Chem. Phys., 14 (2012) 4682-4686.
382
[33] G. Ramírez, S.E. Rodil, S. Muhl, D. Turcio-Ortega, J.J. Olaya, M. Rivera, E. Camps, L.
383
Escobar-Alarcón, Amorphous niobium oxide thin films, J.Non-Cryst. Solids, 356 (2010) 2714-
384
2721.
Ac ce p
te
d
M
an
us
cr
ip t
364
18
Page 18 of 37
[34] X. Fang, L. Hu, K. Huo, B. Gao, L. Zhao, M. Liao, P.K. Chu, Y. Bando, D. Golberg, New
386
ultraviolet photodetector based on individual Nb2O5 nanobelts, Adv. Funct. Mater., 21 (2011)
387
3907-3915.
388
[35] K. Yoshimura, T. Miki, S. Iwama, S. Tanemura, Characterization of niobium oxide
389
electrochromic thin films prepared by reactive d.c. magnetron sputtering, Thin Solid Films, 282
390
(1996) 235-238.
391
[36] R.A. Rani, A.S. Zoolfakar, J.Z. Ou, M.R. Field, M. Austin, K. Kalantar-zadeh, Nanoporous
392
Nb2O5 hydrogen gas sensor, Sens. Actuators B, 176 (2013) 149-156.
393
[37] T. Hyodo, J. Ohoka, Y. Shimizu, M. Egashira, Design of anodically oxidized Nb2O5 films as
394
a diode-type H2 sensing material, Sens. Actuators B, 117 (2006) 359-366.
395
[38] Z. Wang, Y. Hu, W. Wang, X. Zhang, B. Wang, H. Tian, Y. Wang, J. Guan, H. Gu, Fast and
396
highly-sensitive hydrogen sensing of Nb2O5 nanowires at room temperature, Int. J. Hydrogen
397
Energ., 37 (2012) 4526-4532.
398
[39] Y.D. Wang, L.F. Yang, Z.L. Zhou, Y.F. Li, X.H. Wu, Effects of calcining temperature on
399
lattice constants and gas-sensing properties of Nb2O5, Mater. Lett., 49 (2001) 277-281.
400
[40] H.G. Moon, H.W. Jang, J.-S. Kim, H.-H. Park, S.-J. Yoon, A route to high sensitivity and
401
rapid response Nb2O5-based gas sensors: TiO2 doping, surface embossing, and voltage
402
optimization, Sens. Actuators B, 153 (2011) 37-43.
403
[41] J. Choi, J.H. Lim, S. Rho, D. Jahng, J. Lee, K.J. Kim, Nanoporous niobium oxide for label-
404
free detection of DNA hybridization events, Talanta, 74 (2008) 1056-1059.
405
[42] H.L. Skriver, N.M. Rosengaard, Surface energy and work function of elemental metals,
406
Phys. Rev. B, 46 (1992) 7157-7168.
Ac ce p
te
d
M
an
us
cr
ip t
385
19
Page 19 of 37
[43] S. Ge, H. Jia, H. Zhao, Z. Zheng, L. Zhang, First observation of visible light photocatalytic
408
activity of carbon modified Nb2O5 nanostructures, J. Mater. Chem., 20 (2010) 3052-3058.
409
[44] H. Yamaura, Y. Abe, K. Ino, S. Ezawa, K. Sagata, K. Ikushima, S. Yamaguchi, H. Yahiro,
410
Carbon oxidation reaction over Pt/Spherical alumina beads catalysts prepared by sputtering
411
method, Top. Catal., 53 (2010) 648-653.
412
[45] C. Varenne, A. Ndiaye, J. Brunet, G. Monier, L. Spinelle, A. Pauly, L. Bideux, B. Lauron,
413
C. Robert-Goumet, Comparison of InP Schottky diodes based on Au or Pd sensing electrodes for
414
NO2 and O3 sensing, Solid State Electron., 72 (2012) 29-37.
415
[46] Y.H. Ng, M. Wang, H. Han, C.L.L. Chai, Organic polymer composites as robust, non-
416
covalent supports of metal salts, Chem. Commun., (2009) 5530-5532.
417
[47] N. Thi Giang Huong, P. Thi Van Anh, P. Thi Xuan, L. Thi Xuan Binh, T. Van Man, N. Thi
418
Phuong Thoa, Nano-Pt/C electrocatalysts: synthesis and activity for alcohol oxidation, Adv. Nat.
419
Sci. Nanosci. Nanotechnol., 4 (2013) 035008.
420
[48] X. Liang, C.-j. Liu, P. Kuai, Selective oxidation of glucose to gluconic acid over argon
421
plasma reduced Pd/Al2O3, Green Chem., 10 (2008) 1318-1322.
422
[49] A.F. Lee, C.J. Baddeley, C. Hardacre, R.M. Ormerod, R.M. Lambert, G. Schmid, H. West,
423
Structural and catalytic properties of novel Au/Pd bimetallic colloid particles - EXAFS, XRD,
424
and acetylene coupling, J. Phys. Chem-Us, 99 (1995) 6096-6102.
425
[50] A.C. Schmitz, A.T. Ping, M.A. Khan, Q. Chen, J.W. Yang, I. Adesida, Schottky barrier
426
properties of various metals on n-type GaN, Semicond. Sci. Tech., 11 (1996) 1464-1467.
427
[51] S.M Sze, K.K. Ng, Physics of Semiconductor Device, John Wiley & Sons, Inc Publication,
428
2006.
Ac ce p
te
d
M
an
us
cr
ip t
407
20
Page 20 of 37
[52] B.A. Sexton, K.D. Rendulic, A.E. Huges, Decomposition pathways of C1-C4 alcohols
430
adsorbed on platinum (111), Surf. Sci., 121 (1982) 181-198.
431
[53] A. Gazsi, A. Koós, T. Bánsági, F. Solymosi, Adsorption and decomposition of ethanol on
432
supported Au catalysts, Catal. Today, 160 (2011) 70-78.
433
[54] P.Y. Sheng, G.A. Bowmaker, H. Idriss, The reactions of ethanol over Au/CeO2, Appl. Catal.
434
A-Gen., 261 (2004) 171-181.
435
[55] A. Yee, S.J. Morrison, H. Idriss, The reactions of ethanol over M/CeO2 catalysts - Evidence
436
of carbon-carbon bond dissociation at low temperatures over Rh/CeO2, Catal. Today, 63 (2000)
437
327-335.
438
[56] J.L. Davis, M.A. Barteau, Decarbonylation and decomposition pathways of alcohols on
439
Pd(111), Surf. Sci., 187 (1987) 387-406.
440
[57] H. Idriss, Ethanol reactions over the surfaces of noble metal/cerium oxide catalysts,
441
Platinum Met. Rev., 48 (2004) 105-115.
442
[58] D.D. Yao, R. Abdul Rani, A.P. O'Mullane, K. Kalantar-zadeh, J.Z. Ou, High performance
443
electrochromic devices based on anodized nanoporous Nb2O5, J. Phys. Chem. C, (2013).
444
[59] C.C. Cheng, Y.Y. Tsai, K.W. Lin, H.I. Chen, W.H. Hsu, H.M. Chuang, C.Y. Chen, W.C.
445
Liu, Hydrogen sensing characteristics of Pd- and Pt-Al0.3Ga0.7As metal-semiconductor (MS)
446
Schottky diodes, Semicond. Sci. Tech., 19 (2004) 778-782.
447
[60] S. Joo, I. Muto, N. Hara, Hydrogen gas sensor using Pt- and Pd-added anodic TiO2 nanotube
448
films, J. Electrochem. Soc., 157 (2010) J221-J226.
449
[61] M. Armgarth, D. Soderberg, I. Lundstrom, Palladium and platinum gate metal-oxide-
450
semiconductor capacitors in hydrogen and oxygen mixtures, Appl. Phys. Lett., 41 (1982) 654-
451
655.
Ac ce p
te
d
M
an
us
cr
ip t
429
21
Page 21 of 37
[62] E. Bus, J.A. van Bokhoven, Hydrogen chemisorption on supported platinum, gold, and
453
platinum-gold-alloy catalysts, Phys. Chem. Chem. Phys., 9 (2007) 2894-2902.
454
[63] N. Yamamoto, S. Tonomura, T. Matsuoka, H. Tsubomura, A study on a palladium-titanium
455
oxide Schottky diode as a detector for gaseous components, Surf. Sci., 92 (1980) 400-406.
456
[64] M. Penza, C. Martucci, G. Cassano, NOx gas sensing characteristics of WO3 thin films
457
activated by noble metals (Pd, Pt, Au) layers, Sens. Actuators B, 50 (1998) 52-59.
cr
ip t
452
us
458
an
459
Ac ce p
te
d
M
460
22
Page 22 of 37
us
cr
ip t
460 461
462
Fig. 1. Three dimensional schematic of the Schottky-based Nb2O5 ethanol vapour sensor.
an
463 464
Ac ce p
te
d
M
465
23
Page 23 of 37
466
Ac ce p
te
d
M
an
us
cr
ip t
465
467
Fig. 2. SEM images of a Nb2O5 nanoporous film after the annealing process: (a) cross-
468
sectional view of the whole nanoporous structure, (b) top view of the nanoporous structure,
469
(c) a higher magnification SEM image of cross-sectional view of the nanoporous structure.
24
Page 24 of 37
ip t cr us an M
470
Fig. 3 (a) XPS survey spectra of Nb2O5 (inset: XPS spectrum of Nb3d) (b) Pt−Nb2O5 (inset:
472
XPS spectrum of Pt 4f peaks) (c) Pd−Nb2O5 (inset: XPS spectrum of Pd 3d peaks and (d)
473
Au−Nb2O5 (inset: XPS spectrum of Au 4f peaks) films.
Ac ce p
te
d
471
25
Page 25 of 37
ip t cr us an M d te Ac ce p 474 475
Fig. 4 XRD patterns of the (a) Pt−Nb2O5 (b) Pd−Nb2O5 and (d) Au−Nb2O5 samples.
476
Orthorhombic Nb2O5 (ICD 27-1003) are indicated by * while Pt, Pd and Au peaks are indicated
477
by °, Δ and ♦, respectively.
26
Page 26 of 37
ip t cr us an
Fig. 5 Forward−biased I−V characteristics of (i) Au−Nb2O5, (ii) Pd−Nb2O5 and (iii) Pt−Nb2O5 in
480
air at room temperature.
483
d te
482
Ac ce p
481
M
478 479
27
Page 27 of 37
ip t cr an
us 485
M
484
Fig. 6 I−V characteristics of Pd−Nb2O5 in air and 1 % ethanol at elevated temperatures.
Ac ce p
te
d
486
28
Page 28 of 37
ip t cr us an M d te Ac ce p
487 488
Fig. 7 (a) I−V characteristics of sensors in air and 1 % ethanol (b) dynamic response to 1%
489
ethanol biased at 2 µA of (i) Pt−Nb2O5 (ii) Pd−Nb2O5 and (iii) Au−Nb2O5 Schottky gas sensors
490
at 180 °C.
491 492
29
Page 29 of 37
ip t cr us an M
493
Fig. 8 (a) I−V characteristics of sensor in air and 1 % ethanol (b) dynamic responses to
495
1% ethanol biased at 2μA of (i) Pt−Nb2O5 at 200 °C and (ii) Au−Nb2O5 at 240 °C.
te
497
Ac ce p
496
d
494
30
Page 30 of 37
ip t cr us an M d te
498
Fig. 9 Energy band diagram illustrations of a metal and Nb2O5 contact for ΦM > χ (a) is before
500
establishing the contact and (b) their corresponding band bending after establishing a contact and
501
exposure to ethanol vapour. χ is the electron affinity of Nb2O5, EFM and EFS are the Fermi levels of
502
metal and Nb2O5, respectively, EVAC is the reference vacuum level, EV and EC are the valence and
503
conduction band and the ΦBAir and ΦBEth are the SBHs in air and ethanol.
Ac ce p
499
504 505 506
31
Page 31 of 37
ip t cr us an
507
Fig. 10 Voltage shift of the sensors at different ethanol concentrations obtained at 180 °C (i)
509
Pd−Nb2O5 (ii) Pt−Nb2O5 and (iii) Au−Nb2O5.
M
508
te Ac ce p
511
d
510
32
Page 32 of 37
ip t cr us an M d te Ac ce p 511 512
Fig. 11 Dynamic response to 1 % of different vapour/gases at 180 °C of (a) Pt−Nb2O5 (b)
513
Pd−Nb2O5 and Au−Nb2O5 gas sensor.
514
33
Page 33 of 37
ip t cr us an
515
Fig. 12 Voltage shift of all sensors at concentrations of 1 % exposure to different
517
vapour/gases.
M
516
d
518
te
519
520
Ac ce p
Table 1 SBH change under different vapour/ gases SBH change, ΔΦB (eV), under Metal work different gas or vapour Sensor Metal function C2H5OH H2 CH4 Au−Nb2O5
Au
5.4
~ 0.017
~ 0.020
~ 0.008
Pd−Nb2O5
Pd
5.6
~ 0.234
~ 0.391
~ 0.203
Pt−Nb2O5
Pt
5.7
~ 0.015
~ 0.273
~ 0.007
521 522 523
34
Page 34 of 37
523 524 Table 2 Response and recovery time of Nb2O5 Schottky gas sensors with different types of metal catalysts Pt
Au
ip t
Pd
38
98
57
0.12
34
166
42
0.25
28
152
30
0.5
23
154
1.0
17
240
58
227
283
46
324
351
25
267
16
256
25
259
25
290
15
310
M
Ac ce p
te
d
525
232
an
0.06
cr
Response Recovery Response Recovery Response Recovery time (s) time (s) time (s) time (s) time (s) time (s)
us
Ethanol cconcentration (vol %)
35
Page 35 of 37
Rosmalini Ab. Kadir received her degrees in bachelor of electrical engineering (2001) and Master of Science in microelectronic (2005) from the Universiti Malaya and Universiti Kebangsaan Malaysia, Malaysia. She is currently undertaking her Ph.D. program at RMIT University, Australia in memristor and gas sensor technology. Her research interest includes
ip t
fabrication and characterization of nanostructured materials. Rozina Abdul Rani received her bachelor of engineering (Electrical and Electronic) with
cr
Honours (2001) and Master of Science (2007) degree from Universiti Sains Malaysia, Malaysia. She is currently undertaking her candidature for a Ph.D. at RMIT University. Her
sensitized solar cells, batteries and chemical sensors.
us
research interests include synthesis and characterization of nanostructured metal oxides, dye-
an
Ahmad Sabirin Zoolfakar received his bachelor of engineering (Electrical) with Honours from Universiti Malaya, Malaysia (2001) and Master of Science (Engineering) (Microelectronic Systems and Telecommunications) from The University of Liverpool, UK
M
(2003). He just finished his Ph.D. at RMIT University. His research interests include aqueous chemical synthesis, gas vapour synthesis, nanotechnology and materials science.
d
Jian Zhen Ou received his Ph.D. from RMIT University, Australia in 2012. His research
te
interests include synthesis and characterization of nanostructured metal oxides, as well as the developments of metal oxides based high performance electronic and optical devices.
Ac ce p
Mahnaz Shafiei received her bachelor of science in electrical and electronics engineering from Amir Kabir University of Technology (Tehran Polytechnique), Iran in 1998. She completed her Ph.D. at RMIT University, Melbourne, Australia in 2011. She is currently a postdoctoral research fellow at QUT, Queensland, Australia. Her major area of research includes chemical sensors, nanotechnology and material science. Wojtek Wlodarski has worked in the areas of sensor technology and instrumentation for over 35 years in the USA, Canada, Holland, France, Poland, and currently in Australia. He has published 4 books and monographs, several book chapters, over 450 papers and holds 29 patents. He is a full professor at RMIT University, Melbourne, Australia, and heads the Sensor Technology laboratory at School of Electrical and Computer Engineering. His research interests include chemical, physical and biosensors, nanotechnology, materials science and instrumentation.
Page 36 of 37
Kourosh Kalantar-zadeh is a Professor at RMIT University, Australia. He received his Bachelor of Science. (1993) and Master of Science (1997) degrees from Sharif University of Technology, Iran, and Tehran University, Iran, respectively, and a Ph.D. at RMIT University, Australia (2001). His research interests include chemical and biochemical sensors,
ip t
nanotechnology, micro systems, materials sciences, electronic circuits, and microfluidics. He
cr
is the author of over 200 scientific manuscripts.
Ac ce p
te
d
M
an
us
Page 37 of 37