NiO@ZnO heterostructured nanotubes: Coelectrospinning fabrication, characterization, and highly enhanced gas sensing properties

Article abstract of DOI:10.1021/ic300749a

Novel NiO@ZnO heterostructured nanotubes (NTs) were fabricated by the coelectrospinning method, consisting of external hexagonal ZnO shell and internal cubic NiO NTs. They are carefully investigated by scanning electron microscopy, transmission electron microscopy, scanning transmission electron microscopy, energy-dispersive X-ray spectroscopy mapping, X-ray diffraction, and X-ray photoelectron spectroscopy techniques. A reasonable formation mechanism of the hierarchical NiO@ZnO NTs is proposed, which is discussed from the view of degradation temperature of different polymers and the amount of inorganic salts. They were then explored for fabrication of H₂S gas sensors. The gas sensing test reveals that compared with the pure ZnO, NiO, and the ZnO-NiO mixed gas sensors, hierarchical gas sensor exhibits highly improved sensing performances to dilute hydrogen sulfide (H₂S) gas. The response of the optimum NiO@ZnO NTs sensor to 50 ppm H₂S increases as high as 2.7-23.7 times compared to the other sensors, whereas the response and recovery times also become shorter considerably. These enhanced gas sensing properties are closely related to the change of nanostructure and activity of ZnO and NiO nanocrystals as well as combination of homo- and heterointerfaces in the optimum gas sensor, which are confirmed by a series of well-designed experiments.

Full text of DOI:10.1021/ic300749a

Article
pubs.acs.org/IC
NiO@ZnO Heterostructured Nanotubes: Coelectrospinning
Fabrication, Characterization, and Highly Enhanced Gas Sensing
Properties
Lin Xu, Ruifang Zheng, Shuhai Liu, Jian Song, Jiansheng Chen, Biao Dong, and Hongwei Song*
State Key Laboratory on Integrated Optoelectronics, College of Electronic Science and Engineering, Jilin University, Changchun,
130012, People’s Republic of China
S
* Supporting Information
ABSTRACT: Novel NiO@ZnO heterostructured nanotubes
(NTs) were fabricated by the coelectrospinning method,
consisting of external hexagonal ZnO shell and internal cubic
NiO NTs. They are carefully investigated by scanning electron
microscopy, transmission electron microscopy, scanning trans-
mission electron microscopy, energy-dispersive X-ray spec-
troscopy mapping, X-ray diffraction, and X-ray photoelectron
spectroscopy techniques. A reasonable formation mechanism
of the hierarchical NiO@ZnO NTs is proposed, which is
discussed from the view of degradation temperature of different polymers and the amount of inorganic salts. They were then
explored for fabrication of H2S gas sensors. The gas sensing test reveals that compared with the pure ZnO, NiO, and the ZnO−
NiO mixed gas sensors, hierarchical gas sensor exhibits highly improved sensing performances to dilute hydrogen sulfide (H2S)
gas. The response of the optimum NiO@ZnO NTs sensor to 50 ppm H2S increases as high as 2.7−23.7 times compared to the
other sensors, whereas the response and recovery times also become shorter considerably. These enhanced gas sensing properties
are closely related to the change of nanostructure and activity of ZnO and NiO nanocrystals as well as combination of homo- and
heterointerfaces in the optimum gas sensor, which are confirmed by a series of well-designed experiments.
uring the past decades, one-dimensional (1D) semi-
conductor oxides nanomaterials, such as nanowires,
surface to volume ratio for both the inner and outer layers, as
well as the large interface between the two layers. These
characteristics make it have potential application in various
electrical applications.
D
nanorods and naotubes (NTs) have captured extensive
attention in view of enhancing performance in electrical
nanodevices, such as gas sensor, biosensor, solar cell, etc.1As
a special branches of 1D structure, heterostructured nanoma-
terials have strong heterointeractions between the closely
packed interface nanounits, thus their properties cannot be
considered as a simple contacted of the individual components,
but more complex and more superior.2 Recently, Si et al.
prepared Fe2O3/ZnO heterostructured nanorods by a hydrol-
ysis method, by which considerably enhanced ethanol gas
sensing properties were obtained.2c Kim et al. prepared SnO2−
In2O3 heterostructured nanowires via a thermal evaporation
method, and the electronic conductivity of the individual
SnO2−In2O3 nanowires was improved by 2 orders of
magnitude compared to that of pure SnO2 nanowires.2d
Among various 1D synthesis techniques, coelectrospinning is
a simple and scalable method for generating ultralong 1D
heterostructure in an efficient, versatile, one-step synthesis.
Polymer, organic, inorganic, and hybrid heterostructured
materials have been fabricated by this facile method.3 However,
to the best of our knowledge, the heterostructured semi-
conductor oxide NTs have not been fabricated until now.
Actually, heterostructured NTs are very favorable for electrical
devices, because it should not only have excellent electrical
property like the other 1D nanomaterials but also have a large
Herein, we present the first preparation of the heterostruc-
tured semiconductor oxide NTs consisting of inner NiO NTs
and outer ZnO shell through coelectrospinning method and its
enhanced gas sensing characteristics to H2S. As is well-known,
ZnO is a n-type semiconductor oxide with a wide bandgap
(3.37 eV) and large exciton binding energy (60 meV),4 while
NiO is a p-type one with a bandgap of 4.2 eV,5 and they have
been both recognized as excellent gas sensing materials.6
Moreover, ZnO and NiO can easily form p−n heterojunction
and generate an electrical barrier between crystal grains,7 which
can show good sensitivity to reducing gases.8
EXPERIMENTAL SECTION
■
Sample Preparation. The NiO@ZnO heterostructured NTs were
fabricated by coelectrospinning followed by calcination, which
consisted of following three steps: (a) Preparation of the spinnable
precursor outer and inner sol solutions. In a typical procedure, 0.4 g
(named as sample ZN1) or 1 g (named as sample ZN2) of zinc acetate
with two crystal water (Zn(Ac)2) were dissolved in a mixed solvent
composed of 8 mL N,N-dimethylformamide (DMF) and 2 mL
Received: April 12, 2012
Published: June 29, 2012
© 2012 American Chemical Society
7733
dx.doi.org/10.1021/ic300749a | Inorg. Chem. 2012, 51, 7733−7740

Inorganic Chemistry
Article
ethanol at room temperature. After 15 min, 2 g of poly(vinyl
pyrrolidone) (PVP, Mw = 130 000) was added and followed by
vigorous stirring for 3 h to obtain the transparent outer sol solutions.
In a familiar way, 0.8 g of nickel acetate with four crystal water
(Ni(Ac)2) was dissolved in 10 mL of DMF with magnetic stirring for
15 min at room temperature. Subsequently, 1.5 g of poly(acrylonitrile)
(PAN, Mw = 150 000) was added into the above solution, followed by
vigorous stirring for 12 h to obtain the colorless inner precursor
solution. (b) Coelectrospinning the obtained solutions. The schematic
ning fibers were annealed in a tube furnace with a rising rate of 1 °C/
min from room temperature to 700 °C and kept for 3 h, then was self-
cooled down to room temperature again, forming the final
heterostructured NTs. In addition, to understand the formation
mechanism of the as-fabricated NTs nanostructure, we also fabricated
some other samples with different amount of Zn(Ac)2 or Ni(Ac)2 and
the pure ZnO and NiO for contrast and the preparation details are
listed in Supporting Information.
Characterization and Measurement. The surface morphology
of the as-prepared samples was inspected using a JEOL JSM-7500F
scanning electron microscopy (SEM) at an accelerating voltage of 15
kV with gold sputtered on samples. The average diameter of samples
was calculated by selecting at least 100 fibers randomly from the SEM
micrographs. Transmission electron microscopy (TEM) and scanning
transmission electron microscopy (STEM) images were recorded on a
JEM-2010 transmission electron microscope under a working voltage
of 200 kV equipped with energy dispersive X-ray (EDX) spectrometer.
X-ray diffraction (XRD) patterns of the samples were conducted on a
Rigaku D/max 2550 X-ray diffractometer, using a monochromatized
Cu target radiation resource (λ = 1.5045 Å). X-ray photoelectron
spectroscopy (XPS) experiments were measured using an ESCAlab
250 Analytical XPL Spectrometer with a monochromatic Al Kα
source. All the binding energies were referenced to the C1s peak at
284.7 eV of the surface adventitious carbon. The fitted peaks in the
XPS spectra were deconvoluted using the XPS Peak 4.1 software.
The gas sensing properties were measured on a WS-30A system
(Weisheng Instruments Co. Zhengzhou, China). The gas sensors were
fabricated by coating the as-prepared NiO@ZnO NTs ethanol colloids
on the ceramic tube (the coating thickness is about 0.25 mm) on
which a pair of gold electrodes were previously printed. A small spring-
like Ni−Cr alloy was inserted into the ceramic tube to provide the
operating temperature. The gas sensors were dried and aged for 10
days before the first measurement. When testing, a given amount of
target gases together with dry synthetic air were injected into the
chamber (about 2.5 L in volume) to obtain desired concentrations.
The response is defined as Rair/Rgas for n-type gas sensors and Rgas/Rair
for p-type gas sensors, where Rair and Rgas are the resistance for sensors
in air and in target gas. The response and recovery times are defined as
the time required reaching 90% of the final equilibrium value.
Figure 1. Schematic diagram of the experimental coelectrospinning
setup. The insets are the actual photo of the coaxial dual spinnerets
(the top one) and the SEM image of the precursor nanofibers (the
below one).
diagram of the coelectrospinning setup is illustrated in Figure 1. It
consists of three major components: a high-voltage power supply, a
syringe pump, and a coaxial spinneret setup. The actual coaxial
spinneret is shown in the top inset of Figure 1. The stainless syringe
needles marked as # 1 (1.4 mm in diameter) and # 2 (0.7 in diameter)
constitute the key part of the coaxial spinneret which connected with
outer and inner solutions respectively. The stainless syringe needles
marked as # 3 used to transport the outer solution into the outer
plastic syringe. The two different solutions were delivered
independently through the concentric nozzles with the flow rates
controlled by two separate syringe pumps. The feeding rates of outer
and inner solution are 0.6 and 0.2 mL/h, respectively. In
coelectrospinning process, the collection distance between coaxial
spinneret tip and collector is 15 cm and the applied steady-voltage is
15 kV. The obtained coelectrospinning nanowires (see the inset of
Figure 1) were collected and dried 12 h at room temperature under
vacuum. (c) The thermal treatment. In this process, the coelectrospin-
RESULTS AND DISCUSSION
■
Morphologies and Structural Properties of Hierarch-
ical NiO@ZnO NTs. The NiO@ZnO heterostructured NTs
(samples ZN1 and ZN2) as well as the pure ZnO and NiO
Figure 2. (a, b) SEM, (c, d) TEM, and (e, f) STEM images of core−shell ZN1 and ZN2 NTs, respectively. The insets of panels a and b show the
corresponding close-ups of the NTs sections.
7734
dx.doi.org/10.1021/ic300749a | Inorg. Chem. 2012, 51, 7733−7740

Inorganic Chemistry
Article
nanofibers were fabricated though coelectrospinning method
followed by calcination. The SEM images of hierarchical
samples ZN1 and ZN2 are shown in Figure 2a,b, respectively. It
can be observed that these randomly oriented NTs all have
uniform and long continuous surface. In detail, for sample ZN1,
the inset of Figure 2a clearly displays the cross section of the
obtained NTs with a rough surface, the wall thicknesses are
determined to be ∼55 nm and outer diameters are ∼210 nm.
Moreover, there are some nanoparticles attached randomly on
the surface of the NTs. The morphology of sample ZN2 seems
somewhat different. It can be seen that sample ZN2 yields
smoother surface with outer diameters ∼310 nm. The NTs
structure can also be affirmed by the cross-section SEM images
from inset of Figure 2b and the wall thicknesses are determined
to be ∼100 nm. TEM and STEM images of sample ZN1 and
ZN2 can give more information about the structure character-
istics. As shown in Figure 2c,e, the sample ZN1 displays
heteroarchitectures and there are some irregular shape
implanted on the wall of NTs. In addition, the wall surface
where not covered by nanoparticles show a porous structure
agglomerated with very small nanocrystals. Note that the
porous substructure are formed because of the calcination
aimed at removing the polymer, besides, the high annealing
temperature is helpful of creating spacings among the
nanograins.9 For sample ZN2 which contains more zinc salt
in the outer electrospinning solution, the TEM image (Figure
2d) can further prove its tubular structure. What’s important,
from STEM image (Figure 2f) the hierarchy structure can be
distinguished, which is caused by the differences of reflection
and transmission ability of different materials of outer shell and
inner NTs.
elements are uniform and discontinuous. This demonstrates
that Zn elements mainly dispersed on the outer wall of Ni NTs
and aggregated to form nanoparticles. Differently, the
distribution of Zn elements in sample ZN2 seems homoge-
neous and continuous, and shows the hollow tubular structure.
That is to say, sample ZN2 yields more completed NiO@ZnO
heterostructure with both uniform and continued inner NiO
NTs and outer ZnO shell. Moreover, the distribution of ZnO
can be adjusted by changing the zinc salts content in precursor
outer solution as shown in below.
The XRD patterns of hierarchical ZN1 and ZN2 NTs
compared with pure ZnO and NiO are shown in Figure 4,
Figure 4. XRD patterns of hierarchical samples ZN1, ZN2, pure ZnO,
and NiO.
which provide further insight into the crystallinity of the
products. It can be seen that all the samples are well
crystallized. Compared with pure ZnO and NiO, the XRD
pattern taken from the samples ZN1 and ZN2 can be both
indexed to the mixed of cubic NiO (JCPDS card: 78−0643)
and hexagonal ZnO (JCPDS card: 36−1451) with separate
phases and no obvious peak shift or any trace of other phases
can be detected. Besides, because of the different initial
Zn(Ac)2 content in samples ZN1 and ZN2, the main XRD
peaks of cubic NiO and hexagonal ZnO in each nanocomposite
are different. Moreover, the relative intensity of the dominant
diffraction peak of ZnO is always higher than that of NiO, even
in the sample whose molecular weights of ZnO and NiO are
almost the same (sample ZN2), indicating the good crystallinity
of ZnO in hierarchical sample ZN2.
To further investigate the heteroarchitectures, we conducted
the EDX mappings to determine the specific distribution of Ni
and Zn elements in hierarchical samples ZN1 and ZN2. As
shown in Figure 3, it can be clearly seen that the distribution of
Ni elements exhibit a hollow structure and the outer diameters
are both ∼210 nm corresponding to samples ZN1 and ZN2,
furthermore, the distribution of Ni elements in two samples are
homogeneous. While, the distribution of Zn elements exhibit
larger diameter than Ni elements. For sample ZN1, the Zn
XPS Spectra: Surface States and Interface Interaction.
To further illuminate the surface composition and chemical
state of the elements existing in hierarchical NiO@ZnO NTs
samples, we studied XPS spectra of samples ZN1 and ZN2 as
well as pure ZnO and NiO. As shown in Figure 5a, the
complete spectra of samples ZN1 and ZN2 confirm the
presence of Zn, Ni, O, and C atoms. The peak at 284.7 eV is
attributed to the adventitious carbon based contaminant
adsorbed on the surfaces of the NTs.10 As is shown, when
the content of different components changes in the studied
ZN1 and ZN2 samples, the intensities of the ZnO and NiO
characteristic peaks also change in the same trend.
The O 1s XPS spectra of various samples are enlarged in
Figure 5b. Accordingly, The XPS spectra of O1s core level
electrons measured from pure ZnO and NiO both display there
peaks. The binding energies of 529.4 in NiO and 530.2 in ZnO
correspond to the lattice oxygen in crystalline NiO and ZnO,
respectively. The O 1s peak at 531.9 eV is assigned to the
deficient oxygen and the peak at 532.7 eV belongs to the
chemisorbed or dissociated oxygen or OH species on the
Figure 3. The EDX elemental mapping images of Ni and Zn of sample
(a−c) ZN1 and (d−f) ZN2.
7735
dx.doi.org/10.1021/ic300749a | Inorg. Chem. 2012, 51, 7733−7740

Inorganic Chemistry
Article
ZN1. The shift of Zn 2p peaks to lower binding energy is about
0.9 eV and Ni 2p peaks to higher binding energy is about 0.7
eV compared with the corresponding value for pure ZnO or
NiO samples respectively, which are comparable or even larger
than the binding energy shifts in nanocomposite ZnO−NiO
nanofibers (0.9 eV for Zn 2p and 0.5 eV for Ni 2p).14 This
phenomenon can be explained by the strong interaction
between ZnO and NiO nanocrystals in sample ZN2,15 It will
lead to the increased surface activity of ZnO nanocrystals and
the decreased surface activity of NiO nanocrystals.
Formation Mechanism of Hierarchical NiO@ZnO NTs.
Herein, we discuss the formation mechanism of the as-
fabricated hierarchical NiO@ZnO NTs from the view of
reaction temperature and the ratio of inorganic salts to
polymers. Our previous studies about electrospinning NTs
structure indicate that the ratio of inorganic salts to polymer
and slowly annealing temperature are key factors for the
formation of NTs.16 In this case, due to the incompatibility of
PVP and PAN, the initial hierarchical solid core−shell structure
has already been formed in precursor nanofibers.3 When at a
slow rising rate (1 °C/min) in the annealing process, the
electrospun solid gel fibers could be homogeneously heated
from the surface to the center.17 According to the literatures,
the degradation temperatures of Zn(Ac)2 in PVP and Ni(AC)2
in PAN are both ∼250 °C,18 the degradation temperature range
of PAN is 280−320 °C,19 and that of PVP is 343−361 °C.17,18a
This means that small ZnO and NiO nanoparticles would be
formed (∼250 °C) before the removing of PAN and PVP, and
the PAN core would decompose (280−320 °C) earlier than the
PVP shell (343−361 °C). Moreover, it was previously observed
that in the mixture of PVP and PAN nanofibers, the
degradation temperature of PVP (389 °C) and PAN (360
°C) both increase than pure PVP and PAN nanofibers because
of the reduced interactions between ions and polymers.19b,20 In
the present coelectrospinning, we have reason to consider that
the degradation temperature of PVP and PAN on or near the
interface of outer and inner layers would be higher than the
internal, because the boundary of PVP and PAN is not so
distinct.
Figure 5. (a) Survey, (b) O 1s, (c) Zn 2p, and (d) Ni 2p high-
resolution XPS spectra of samples ZN1, ZN2, pure ZnO, and NiO.
samples surface.11 The samples ZN1 and ZN2 can both be well
dissolved by four components, locating at ∼529.6, 530.4, 531.9,
and 532.7 eV, respectively. Note that because the different
chemical environment of O in composite oxides, each
characteristic peak has some shift in the two samples.
Moreover, defective oxygen occupies the biggest proportion
in both samples, suggesting a large number of surface oxygen
vacancies. The atomic ratio percentages of deficient oxygen of
O 1s are calculated as 30.4%, 33.8%, 39.1, and 43.7% for NiO,
ZnO, ZN1, and ZN2 respectively. Generally, when the ZnO
and NiO nanocrystals attach continuously, their corresponding
Fermi energy levels would be adjusted to the same value. The
electrons transferred from ZnO to NiO, while the hole
transferred from NiO to ZnO due to the carrier diffusion
caused by the concentration gradient near the p-n junction.
This diffusion process leads to the higher valence concentration
of ZnO. Besides, it is obvious that the relative intensity of lattice
oxygen of NiO in sample ZN2 is much lower than that in
sample ZN1 though they have the same amount of NiO. This
can be explained by NiO being coated by external ZnO more
effectively in sample ZN2.
On the basis of the above analysis, the formation process of
hierarchical NiO@ZnO NTs is proposed, as shown in Figure 6.
The characteristic spin−orbit split X-ray photoelectron
spectroscopy (XPS) data of Zn 2p core level region of
hierarchical ZN1, ZN2 and pure ZnO samples are given in
Figure 5c. The Zn 2p spectrum in all the samples show two
symmetric peaks, the peak centered at 1022.1 eV corresponds
to the Zn 2p3/2 and another one centered at 1045.1 eV is
assigned to Zn 2p1/2, indicating a normal state of Zn 2p in the
pure ZnO NTs.12 Compared to pure ZnO, the Zn 2p core level
region of hierarchical sample ZN1 has no significant shift. For
sample ZN2, the Zn 2p peaks shift to the low binding energy
obviously after introducing of NiO, but with the same splitting
between Zn 2p1/2 and Zn 2p3/2 (23 eV). Ni 2p electron core
level XPS spectra of hierarchical ZN1, ZN2 and pure NiO
samples are shown in Figure 5d. For the pure NiO, the Ni 2p
signal could be deconvoluted into five peaks.13 The peaks at
854.5, 856.2, and 861.2 eV were attributed to Ni 2p3/2, and the
peaks at 872.7 and 879.1 eV to Ni 2p1/2. The Ni 2p peaks of
sample ZN1 and ZN2 can also be deconvoluted into five ones,
and the spectrum of sample ZN2 also exhibits overall shift to
higher binding energy in contrast to pure NiO as well as sample
Figure 6. Illustrative diagram of the possible formation process of
heterostructured NTs through slow heat treatment procedure.
When the temperature gradually increased from room temper-
ature to 250 °C, Zn(Ac)2 and Ni(Ac)2 first begun to
decompose to small ZnO and NiO nanocrystals in outer PVP
and inner PAN templates, respectively (see Figure 6). As the
annealing temperature was gradually elevated to 280−320 °C,
the core PAN started to break down from inner to the interface,
while the external PVP shell still existed. At this time, the PVP
shell and the interface PAN played a role as template for
distribution of internal NiO nanoparticles and made these NiO
nanoparticles gathered along the inner wall of the shell. When
7736
dx.doi.org/10.1021/ic300749a | Inorg. Chem. 2012, 51, 7733−7740

Inorganic Chemistry
Article
the amount of NiO was appropriate, the internal NiO
nanoparticles formed tubular structure and gathered closely
to the external shell. Note that this process can be proved by
Figure 3 together with Figure 7a, when only electrospun the
the corresponding prepared condition and the SEM image are
shown in Supporting Information.
First, the responses of the hierarchical NiO@ZnO NTs and
different reference gas sensors to 50 ppm H2S as a function of
the working temperature were studied, as shown in Figure 8.
Figure 8. Response of heterostructured ZN1 and ZN2 NTs gas
sensors, the mixed ZN-mix NTs gas sensor and pure ZnO and NiO
gas sensors to 50 ppm H2S as a function of operating temperature.
Because of the different amount and contact form of n-type
ZnO and p-type NiO in hierarchical samples ZN1, ZN2, and
nanocomposite sample ZN-mix, ZN1 and ZN-mix gas sensors
show a typical gas sensing characteristics of p-type semi-
conductor and the response is calculated by Rgas/Rair, while
ZN2 gas sensor shows a typical n-type gas sensing character-
istics of semiconductor and the response is calculated by Rair/
Rgas. As can be seen, excepted NiO, the responses for the other
gas sensors all have the same change trend. The responses first
increase with temperature and approach at the maxima at 215
°C, then gradually decrease with further increasing the working
temperature. The response of different gas sensors at 215 °C
are response(ZN2) = 474 > response(ZnO) = 200 >
response(ZN1) = 152 > response(ZN-mix) = 98 > response-
(NiO) = 20.8. The response of ZN2 gas sensor is 2.7−23.7
times higher than the other gas sensors.
Figure 7. SEM images of (a) pure NiO, (b) pure ZnO, (c) sample
ZN3, and (d) sample ZN4.
inner solution (contained 0.8 g Ni(Ac)2), tubular structure NiO
could not be formed (see Figure 7a), while it was formed in
coelectrospinning case (see Figure 3) and when only
electrospun the outer solution (contained 1.0 g Zn(Ac)2), the
NTs structure was formed (see Figure 7b). These phenomena
indicate the template effect of external shell to internal NTs.
When the annealing temperature was further increased to 343−
361 °C, the outer PVP started to degrade and form the final
product. Because the PVP on the interface had higher
degradation temperature, the ZnO nanoparticles distributed
alone the internal NTs. As the amount of ZnO nanoparticles
were lower, the ZnO were too little to cover the inner NiO
NTs completely (sample ZN1). While the amount of ZnO
nanoparticles were suitable, the ZnO covered on the inner NiO
NTs uniformly and continuously (sample ZN2). When the
amount of ZnO nanoparticles were not suitable enough, a
mixture nanostructure of ZN1 and ZN2 was formed, as shown
in Figure 7c. This further demonstrates that when the amount
of NiO was appropriate, the external amount of ZnO
nanoparticles could not affect the inner NTs structure and
they just distributed on the inner NTs wall in different forms.
That is, the internal NiO NTs had been already formed before
the distribution of ZnO nanoparticles. In addition, when the
amount of internal NiO decreased, no continuous NTs were
formed during the decomposition of PAN (Figure 7d), the final
morphology did not show the hierarchical structure, which
further proved the template effect of internal NiO NTs to the
outer shell.
To further confirm the versatility of the as-synthesized gas
sensors, the selective test toward to 50 ppm H2S, H2, NH3 and
200 ppm acetone and methanol were conducted at 215 °C. As
demonstrated in Figure 9, the sensors show excellent selectivity
Figure 9. Selective test toward 50 ppm H2S, NO2, and NH3, 200 ppm
ethanol, and acetone gases of different gas sensors at 215 °C.
to H2S, whereas they have almost no response to the other
typical interference gases at the same temperature. This
indicates that the present sensors have quite excellent selectivity
to the H2S, especially the ZN2 one. The prominent selectivity
can be assigned to theirs particular gas sensing mechanism,
which will be discussed below.
Figure10a shows response transients of different gas sensors
exposed to 50 ppm H2S gas. The ZN2 sensor exhibits the
fastest switching dynamics to H2S at the optimal temperature.
Gas Sensing Properties of Hierarchical NiO@ZnO NTs.
In this work, the as-synthesized hierarchical NiO@ZnO NTs
samples were fabricated to gas sensors and the gas sensing
properties toward to H2S are carefully investigated. Here, in
order to better understanding the gas sensing properties of
hierarchical NiO@ZnO NTs, we also fabricated the composite
NTs composed of the same ZnO and NiO content as sample
ZN2, but ZnO and NiO are mixed uniformly (sample ZN-mix),
7737
dx.doi.org/10.1021/ic300749a | Inorg. Chem. 2012, 51, 7733−7740

Inorganic Chemistry
Article
Figure 10. (a) Response transients of different gas sensors to 20 ppm H2S at 215 °C. (b) Dynamic process of ZN2 gas sensor to different H2S
concentrations (2−80 ppm) at 215 °C, the inset shows the corresponding sensor response curves (1−150 ppm) and the red line is the fit one.
Its response time is ∼50 s and recovery time is ∼124 s. For the
ZnO gas sensor, although the response time is comparable to
ZN2, the recovery time of the sensor is too long (∼1254s) and
cannot return to its original state in time. And the other gas
sensors (ZN1, ZN-mix, and NiO gas sensors) show more
slower dynamic processes, the response and recovery times are
∼82 and 191 s for ZN1, ∼132 and 204 s for ZN-mix, and ∼112
and 200 s for NiO gas sensors. Figure 10b shows the dynamic
response of ZN2 hierarchical NTs gas sensor exposed to
different concentrations of H2S around 215 °C, which indicates
it exhibits rapid switching dynamics to H2S at the optimal
temperature and the excellent reproducible performance. The
inset of Figure 10b shows the corresponding relationship
between the sensor responses and gas concentrations (1−150
ppm) of H2S gases. The response to 150 ppm H2S is as high as
1891. The response and target gas concentration (Cgas) can be
approximated fitted by an empirically equation, response = 1
values to be 133, 270, and 211 meV, respectively, for the ZN2,
ZnO, and NiO gas sensors. As is well-known, the lower barrier
heights are, the easier of the response and recovery reactions to
take place. That is to say, response and recovery reactions are
easier to happen in ZN2 gas sensor than in the pure sensors. It
should be mentioned that the recovery time can be hardly
measured for ZN1 and ZN-mix gas sensor for all the working
temperature, indicating a huge ΔErec values for those two
samples.
Table 1 lists the performance of H2S gas sensor based on
various nanostructures that contained ZnO or NiO in the
Table 1. Comparison of the H2S Gas Sensors Contained
ZnO or NiO in Literatures to the ZN2 NT Sensor in This
Work
gas conc.
(ppm)
T
(°C)
samples
response
ref
+αCgas with β = 1.36 0.10.21 Here, β was determined to be
β
ZnO-TiO2 nanocrystals
10 000
3.5
250
260
230
2.5
1.4
23a
23b
23c
(1.36
0.10) for ZN2 gas sensors, which was much higher
than empirical value (0.5 or 1).21 Moreover, the on- and off-
responses could be repeated after continuous measurement
cycles without observing significant variation in the functional
response, illustrating the good reversibility and stability of the
hierarchical structure-based gas sensor.
ZnO-WO3 film
Cu doped ZnO
nanofibers
10
18.7
ZnO nanofibers
ZnO nanorods
ZnO nanorods
ZnO dendrites
NiO nanoparticles
ZN2 NTs
10
50
230
500
150
30
1.57 23c
29
4000
17.3
23d
23e
23f
20
The dependence of response and recovery times to 50 ppm
H2S of ZN2 gas sensor compared with pure ZnO and NiO gas
sensors on working temperature was also studied. Figure 11
100
100
50
150
215
9
23g
474
This work
literatures and ZN2 hierarchical NTs in this study. Comparing
our results with these literatures, it can be found that ZN2
hierarchical NTs gas sensor exhibits much better response to
most of the other sensors except the ZnO nanorods gas sensor
reported by ref 23e. It should be also noted that although the
response of the sensor listed in ref 23e is super high at 150 °C,
it cannot return to its original state unless the operating
temperature is increased to 450 °C. The present ZN2 sensor
yields not only high response but also favorable fast response
and recovery times.
Generally, there exist two possible mechanisms for semi-
conductor oxide-based H2S gas sensors: adsorption−oxida-
tion−desorption process and the sulfuration−desulfuration
process.23c,e In this experiment, the relative lower working
temperature, shorter dynamic process, higher response to low
concentrate H2S gas, and excellent selectivity are all character-
istics of the sulfuration−desulfuration process.22,23d,e The
possible reason that responsible for the enhanced gas sensing
properties of the ZN2 sensor for H2S can be mainly attributed
to the variation of nanostructure and surface activity of ZnO
and NiO nanocrystals in different gas sensors. For ZN2 sample,
Figure 11. Logarithm of (a) the response time constant and (b)
recovery time constants (both in second) at 215 °C to 50 ppm H2S
versus the reverse of temperature. The dots are experimental data and
the lines are the corresponding linear fitting functions.
shows in the form of the logarithm of time constants versus the
reverse of absolute temperature. Using the well-known thermal
activation function, we can be calculate the response (ΔEres)
and recovery (ΔErec) barrier heights through the temperature-
dependence of response and recovery dynamics.22 Accordingly,
ΔEres values are determined to be 98, 107, and 138 meV, ΔErec
7738
dx.doi.org/10.1021/ic300749a | Inorg. Chem. 2012, 51, 7733−7740

Inorganic Chemistry
Article
because of the particular continuous nanostructure of both
inner NiO and outer ZnO which has suitable contacted area,
appropriate p−n junctions are formed and created strong
interaction as proved by XPS data. For sample ZN1, as a result
of the discontinuous distribution of ZnO nanocrystals on the
NiO NTs, the contact areas of NiO and ZnO are too small to
make significant interaction. Although for the sample ZN-mix,
which also has obvious interaction,14 the corresponding sensor
shows much weaker gas sensing properties because of the
different distributed ways of ZnO and NiO nanocrystals in
nanostructure and bring the sensing properties of n-type ZnO
and p-type NiO cancel each other out. In other words, from the
above analysis, we can get the conclusion that the nanostructure
like ZN2, which contained a suitable contact area and
distribution way of NiO and ZnO is an optimal one. From
the view of surface activity of ZnO and NiO nanocrystals in
different gas sensors, the ZN2 sensor shows a typical gas
sensing characteristics of n-type semiconductor, that is to say,
ZnO are the dominant influence factor to its gas sensing
properties. As previously discussed in XPS spectra, the activity
of ZnO in the ZN2 gas sensor increases because of the strong
interaction between ZnO and NiO nanocrystals, and thus the
sulfuration−desulfuration process can take place easier, which
results in the improved gas sensing properties to H2S; this can
be further proved by the lower ΔEres and ΔErec values of ZN2
gas sensor (Figure 11). ZN1 and ZN-mix gas sensors show the
p-type gas sensing behavior, that is, NiO is the main effect
factor to their gas sensing properties. On the one hand, the
tendency of sulfuration reaction of NiO is weaker than ZnO
due to the Gibbs free energy (ΔG) value of these two
sulfuration reactions (NiO + H2S = NiS + H2O; ZnO + H2S =
ZnS + H2O) are −61.6 and −75.3 kJ mol−1 respectively. On the
other hand, the activity of NiO in ZN1 and ZN-mix gas sensors
only has little change or even much more stable (as proved by
XPS data), thus the gas sensing properties are no better than
ZN2 gas sensor and even lower than the pure ZnO gas sensor.
In addition, in nanofiber gas sensors the contact way of
nanofiber networking also influence the gas sensing properties.
In hierarchical ZN2 samples, combination of homo- (between
the NTs) and hetero- (between the ZnO and NiO) interfaces
are formed, which may build the potential barriers in the
junctions and improved the corresponding gas properties.24
However, in the other sensors, there are only the
homointerfaces (ZnO and NiO gas sensors) or heterointerfaces
(ZN-mix gas sensors) formed at the junctions.
that the suitable nanostructure, increased activity of ZnO
nanocrystals and combination of homo- and heterointerfaces
endowed the NiO@ZnO NT gas sensor with highly enhanced
gas sensing properties.
ASSOCIATED CONTENT
■
S
* Supporting Information
The detailed procedure of each reference sample and SEM
images of sample ZN-mix. This material is available free of
charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*Phone and fax: +86-431-85155129. E-mail: hwsong2005@
yahoo.com.cn.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors are thankful for the financial support of the
National Science Fund for Distinguished Young Scholars of
China (Grant 60925018), the National Natural Science
Foundation of China (Grants 51002062, 61177042, and
11174111), the Frontier Science and Interdisciplinary In-
novation Projects of Jilin University (Grants 421030951419).
REFERENCES
■
(1) (a) Law, M.; Greene, L. E.; Johnson, J. C.; Saykally, R.; Yang, P.
D. Nat. Mater. 2005, 4, 455−459. (b) Sun, L. L.; Zhao, D. X.; Zhang,
Z. Z.; Li, B. H.; Shen, D. Z. J. Mater. Chem. 2011, 21, 9674−9681.
(c) Chen, J.; Xu, L.; Li, W. Y.; Gou, X. L. Adv. Mater. 2005, 17, 582−
586. (d)
(2) (a) Lauhon, L. J.; Gudiksen, M. S.; Wang, D.; Lieber, C. M.
Nature 2002, 420, 57−61. (b) Hwang, Y. J.; Boukai, A.; Yang, P. D.
Nano Lett. 2009, 9, 410−415. (c) Si, S. F.; Li, C. H.; Wang, X.; Peng,
Q.; Li, Y. D. Sens. Actuators, B 2006, 119, 52−56. (d) Kim, D. W.;
Hwang, I. S.; Kwon, S. J.; Kang, H. Y.; Park, K. S.; Choi, Y. J.; Choi, K.
J.; Park, J. G. Nano Lett. 2007, 7, 3041−3045. (e) Hou, Z. Y.; Li, C. X.;
Ma, P. G.; Li, G. G.; Cheng, Z. Y.; Peng, C.; Yang, D. M.; Yang, P. P.;
Lin, J. Adv. Funct. Mater. 2011, 21, 2356−2365. (f) Hou, Z. Y.; Li, G.
G.; Lian, H. Z.; Lin, J. J. Mater. Chem. 2012, 22, 5254−5276.
(3) (a) Chen, M. L.; Dong, M. D.; Havelund, R.; Regina, V. R.;
Meyer, R. L.; Besenbacher, F.; Kingshott, P. Chem. Mater. 2010, 22,
4214−4221. (b) Magalhaes, W.; Cao, X. D.; Lucia, L. A. Langmuir
2009, 25, 13250−13257. (c) Gu, Y. X.; Chen, D. R.; Jiao, X. L.; Liu, F.
F. J. Mater. Chem. 2007, 17, 1769−1776. (d) Kurban, Z.; Lovell, A.;
Bennington, S. M.; Jenkins, D. W. K.; Ryan, K. R.; Jones, M. O.;
Skipper, N. T.; David, W. I. F. J. Phys. Chem. C 2010, 114, 21201−
21213.
(4) Gu, Z. J.; Paranthaman, M. P.; Xu, J.; Pan, Z. W. ACS Nano 2009,
3, 273−278.
(5) Moreira, I. R.; Illas, F.; Martin, R. L. Phys. Rev. B 2002, 65,
155102.
(6) (a) Li, J.; Fan, H. Q.; Jia, X. H. J. Phys. Chem. C 2010, 114,
14684−14691. (b) Wu, H.; Lin, D. D.; Zhang, R.; Pan, W. Chem.
Mater. 2007, 19, 1895−1897.
(7) Ohta, H.; Hirano, M.; Nakahara, K.; Maruta, H.; Tanabe, T.;
Kamiya, M.; Kamiya, T.; Hosono, H. Appl. Phys. Lett. 2003, 5, 1029−
1031.
CONCLUSIONS
■
In summary, novel hierarchy structure n-type ZnO and p-type
NiO NTs were fabricated using a facile coelectrospinning
technique. The hierarchical NiO@ZnO NTs possess a
hexagonal ZnO phase in the outer shell and a cubic NiO
phase in the inner NTs. Through controlling the ratio of ZnO
to NiO, we modulated the morphology of NTs, from the
external ZnO uncompleted coverage to entirely covered state
on the internal NiO NTs. The formation mechanism of ZnO-
NiO NTs was discussed. Moreover, the XPS spectra indicate
the strong interaction of ZnO and NiO on the interface. The
H2S gas sensing properties based on the hierarchy NiO@ZnO
NTs was studied. The results demonstrate that the different
ratio and distributed of ZnO to NiO have great effect on the gas
sensing behavior. ZN2 gas sensor is the outstanding one, it
shows higher response, excellent selectivity and faster dynamic
process, and lower barrier height. The present work suggests
(8) Wang, Z. J.; Li, Z. Y.; Sun, J. H.; Zhang, H. N.; Wang, W.; Zheng,
W.; Wang, C. J. Phys. Chem. C 2010, 114, 6100−6105.
(9) Chen, W. S.; Huang, D. A.; Chen, H. C.; Shie, T. Y.; Hsieh, C.
H.; Liao, J. D.; Kuo, C. S. Cryst. Growth Des. 2009, 9, 4070−4077.
(10) Zhang, Y. W.; Si, R.; Liao, C. S.; Yan., C. H. J. Phys. Chem. B
2003, 107, 10159−10167.
7739
dx.doi.org/10.1021/ic300749a | Inorg. Chem. 2012, 51, 7733−7740

Inorganic Chemistry
Article
(11) Erdem, B.; Hunsicker, R. A.; Simmons, G. W.; Sudol, E. D.;
Dimonie, V. L.; El-Aasser, M. S. Langmuir 2001, 17, 2664−2669.
(12) Ge, L.; Jing, X. Y.; Wang, J.; Wang, J.; Jamil, S.; Liu, Q.; Liu, F.;
Zhang, M. J. Mater. Chem. 2011, 21, 10750−10754.
(13) Ding, Y.; Wang, Y.; Su, L.; Zhang, H.; Lei, Y. J. Mater. Chem.
2010, 20, 9918−9926.
(14) Zhang, Z. Y.; Shao, C. L.; Li, X. H.; Wang, C. H.; Zhang, M. Y.;
Liu, Y. C. ACS Appl. Mater. Inter. 2010, 2, 2915−2923.
(15) Lin, D. D.; Wu, H.; Zhang, R.; Pan, W. Chem. Mater. 2009, 21,
3479−3484.
(16) Xu, L.; Song, H. W.; Dong, B.; Wang, Y.; Bai, X.; Wang, G. L.;
Liu, Q. J. Phys. Chem. C 2009, 113, 9609−9615.
(17) Mou, F. Z.; Guan, J. G.; Shi, W. D.; Sun, Z. G.; Wang, S. H.
Langmuir 2010, 26, 15580−15585.
(18) (a) Teh, P. F.; Sharma, Y.; Pramana, S. S.; Srinivasan, M. J.
Mater. Chem. 2011, 21, 14999−15008. (b) Ji, L. W.; Lin, Z.;
Alcoutlabi, M.; Toprakci, O.; Yao, Y. F.; Xu, G. J.; Li, S. L.; Zhang, X.
W. RSC Adv. 2012, 2, 192−198.
(19) (a) Kim, C.; Ngoc, B.; Yang, K. S.; Kojima, M.; Kim, Y. A.; Kim,
Y. J.; Endo, M.; Yang, S. C. Adv. Mater. 2007, 19, 2341−2346. (b) Ji, L.
W.; Lin, Z.; Li, Y.; Li, S. L.; Liang, Y. Z.; Toprakci, O.; Shi, Q.; Zhang,
X. W. Polymer 2010, 51, 4368−4374.
(20) Zhang, Z. Y.; Li, X. H.; Wang, C. H.; Wei, L. M.; Liu, Y. C.;
Shao, C. L. J. Phys. Chem. C 2009, 113, 19397−19403.
(21) Wang, J. X.; Sun, X.; Huang, W. H.; Lee, Y. C.; Tan, O. K.; Yu,
M. B.; Lo, G. Q.; Kwong, D. L. Appl. Phys. A: Mater. Sci. Process. 2007,
88, 611−615.
(22) Xu, L.; Dong, B.; Wang, Y.; Bai, X.; Liu, Q.; Song, H. W. Sens.
Actuat. B 2010, 147, 531−538.
(23) (a) Bodade, A. B.; Bende, A. M.; Chaudhari, G. N. Vacuum
2008, 82, 588−593. (b) Stankova, M.; Vilanova, X.; Calderer, J.;
Llobet, E.; Brezmes, J.; Gracia, I.; Cane, C.; Correig, X. Sens. Actuators,
B 2006, 113, 241−248. (c) Zhao, M. G.; Wang, X. C.; Ning, L. L.; Jia,
J. F.; Li, X. J.; Cao, L. L. Sens. Actuators, B 2011, 156, 588−592.
(d) Kim, J.; Yong, K. J. Phys. Chem. C 2011, 115, 7218−7224.
(e) Wang, D.; Chu, X. F.; Gong, M. L. Nanotechnology 2007, 18,
185601. (f) Zhang, N.; Yu, K.; Li, Q.; Zhu, Z. Q.; Wan, Q. J. Appl.
Phys. 2008, 103, 104305. (g) Du, Y.; Wang, W.; Li, X. L.; Zhao, J.; Ma,
J. M.; Liu, Y. P.; Lu, G. Y. Mater. Lett. 2012, 68, 168−170.
(24) Choi, S. W.; Park, J. Y.; Kim, S. S. Nanotechnology 2009, 20,
465603.
7740
dx.doi.org/10.1021/ic300749a | Inorg. Chem. 2012, 51, 7733−7740