FULL PAPER

DOI:10.1002/ejic.201403172

Facile Synthesis of ZnO–Reduced Graphene Oxide

Nanocomposites for NO2 Gas Sensing Applications

Nagesh Kumar,[a] Arvind Kumar Srivastava,[b] Hari Shankar Patel,[c]

Bipin Kumar Gupta,*[d] and Ghanshyam Das Varma*[a]

Keywords: Nanocomposites / Sensors / Zinc / Graphene

Graphene and its exotic forms have been widely recognized to investigate the quality of the as-synthesized powder sam-
as exceptional materials for gas-sensing applications be- ples as well as the extent of graphitization of the samples.
cause of their extraordinary electrical conductivity and large Coil sensors with two Pt terminals and a heating element
surface area to volume ratios. Herein, chemically reduced have been designed to extensively monitor the effect of tem-
graphene oxide (rGO) and zinc oxide–reduced graphene ox- perature on the electrical and gas-sensing properties of the
ide (ZrGO) nanocomposite powders have been successfully rGO and ZrGO nanocomposite samples. The ZrGO nano-
synthesized through a simple hydrolysis method followed by composites possess better electrical and NO2 gas sensing
annealing in ambient N2 gas. The reduction of graphene ox- properties than the pristine rGO. The ZrGO nanocomposite
ide by hydrazine hydrate and the decoration of the graphene sensor exhibits a high response (ca. 32 %) for 50 ppm NO2 at
surface by ZnO nanoparticles have occurred during the syn- relatively low temperature (50 °C). Our results suggest that
thesis process. The prepared samples were characterized by the ZrGO nanocomposite material could be used to fabricate
various microscopic techniques to explore the surface mor- a new generation of low-power portable NO2 sensors.
phology and uniformity. Spectroscopic techniques were used

Introduction its high conductivity and metallic transport properties (vF

Graphene is a purely two-dimensional (2D) atomic layer = 106 m s–1) lead to very little Johnson noise.[7]
of sp2-hybridized carbon atoms in a honeycomb hexagonal Graphene oxide (GO), synthesized by the chemical oxid-
lattice; since the experimental breakthrough in its pro- ation of graphite, is intrinsically hydrophilic and can be dis-
duction in 2004, it has been considered as a material for persed in various solvents by sonication.[8] These disper-
various applications in diverse research domains that ex- sions of GO are widely accepted as precursor materials for
ploit its exceptional mechanical, electrical, chemical, bio- the large-scale syntheses of reduced graphene oxide (rGO)
logical, thermal, and optical properties.[1,2] Graphene exhib- films, rGO powder, and various rGO-based composite ma-
its unique electronic properties such as high charge-carrier terials at relatively low cost.[9–12] To date, various pristine
mobility (200,000 cm2 V–1 s–1), tunable band gap (0– rGO thin-film-based gas sensors have been fabricated by
2.5 eV),[3] ballistic transport,[4] and quantum Hall effect at various techniques including expensive nanolithography.
room temperature.[5] Along with these fascinating proper- However, pristine rGO sensors exhibit poor sensing re-
ties, the very high theoretical specific surface area sponses and relatively long recovery times (up to 2 h). To
(2630 m2 g–1) of graphene, the highest surface area to vol- improve their recovery times, these sensors are either heat-
ume ratio among known layered materials,[6] makes it a po- treated or exposed to UV radiation.[13] Moreover, the use
tential material for gas-sensing applications. Furthermore, of the nanolithographic technique to fabricate electrodes or
patterns leads to samples with contaminated surfaces,
[a] Department of Physics and Centre of Nanotechnology, Indian which greatly influence the charge transport and, hence,
Institute of Technology, Roorkee,
Roorkee 247667, India sensing performance of the device.[14,15] However, as sensing
E-mail: gdvarfph@[Link] is a surface phenomena, the decoration of a graphene sur-
[b] Indus Synchrotrons Utilization Division, Raja Ramanna face with metal nanoparticles (NPs) or metal oxide nano-
Centre for Advanced Technology,
Indore 452013, India structures (NSs) will affect the sensing performance of
[c] Laser Biomedical Applications and Instrumentation Division, graphene-based sensing devices. Several such graphene
Raja Ramanna Centre for Advanced Technology, nanohybrids with improved sensing performance have been
Indore 452013, India
[d] National Physical Laboratory (CSIR), reported. These studies revealed that diameter, concentra-
Dr K S Krishnan Road, New Delhi 110012, India tion, distribution of NPs or NSs, and temperature are the
E-mail: bipinbhu@[Link]
Supporting information for this article is available on the parameters that greatly influence the performance of the
WWW under [Link] gas sensors.[16–19]

Eur. J. Inorg. Chem. 2015, 1912–1923 1912 © 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 10990682c, 2015, 11, Downloaded from [Link] by Indian Institute Of Technology, Wiley Online Library on [18/06/2026]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link] FULL PAPER

Here, we report the synthesis of rGO and zinc oxide– XRD pattern of rGO suggests that the large-scale long-
reduced graphene oxide (ZrGO) nanocomposite powder range ordering in the GO sheets has been greatly disturbed
samples by a hydrolysis method in which 85 % hydrazine by the reduction process and that there is a poor ordering
hydrate was utilized as the reducing agent. The morpho- of the graphene sheets along their stacking direction.
logies of the samples have been probed in detail through
various microscopic techniques. The electrical and gas-sens-
ing properties of the rGO and ZrGO powder samples have
been investigated in a sensing chamber with a two-terminal
setup and a heating coil. We chose NO2 as the test gas for
our sensor because it is very toxic to the environment and
living beings. NO2 is mainly released in the atmosphere by
vehicles, manufacturing, food processing, and the petro-
leum and metal-refining industries. The exposure to various
concentrations of NO2 gas results in different adverse ef-
fects on human health. The odor threshold, that is, the con-
centration above which people can recognize NO2 in the
atmosphere by its pungent acidic odor, is 0.1 to 0.4 ppm
[U.S. National Institute for Occupation Safety and Health
(NIOSH), 1976]. According to the U.S. Occupation Safety
and Health Administration (OSHA), the permissible expo-
sure level (PEL) and short-term (15 min) exposure limit
(STEL) for NO2 are 1 and 2 ppm, respectively. The
NIOSH-recommended STEL and immediately dangerous
to life and health (IDLH) values for NO2 are 1 and 20 ppm,
respectively.[20] NO2 may initiate lung irritations, increase
susceptibility to infections, contribute to acid rain, and play Figure 1. XRD and spectroscopic studies of GO, rGO, and ZrGO
samples. (a) XRD patterns, (b) TGA profiles, (c) FTIR spectra,
an important role in the formation of ozone in the lower and (d) Raman spectra.
atmosphere.[21] We have observed that ZrGO possesses bet-
ter electrical and gas-sensing properties than rGO and in-
vestigated the sensing performance of ZrGO against dif- The XRD pattern of the ZrGO composite clearly dem-
ferent concentrations of NO2 gas at room temperature as onstrates the formation of ZnO nanoparticles through the
well as at higher temperatures. We conclude that this mate- characteristic sharp peaks that are representative of the
rial and sensing technique can be used to develop a smart (100), (002), (101), (102), (110), (103), and (112) planes of
portable inexpensive NO2 gas sensor with a fast response the wurtzite structure of ZnO. In ZrGO, the particle size
and short recovery times. was estimated to be ca. 21 nm by the Scherrer formula. A
small broad peak at 2θ ≈ 25° verifies the presence of rGO
in the ZrGO nanocomposite, and the dashed oval in Fig-
Results and Discussion ure 1 (a) highlights this peak.
The thermogravimetric analysis (TGA) plots of the three
Compositional and Morphological Studies
samples are shown in Figure 1 (b). The TGA measurements
The powder X-ray diffraction (XRD) patterns of GO, of the GO, rGO, and ZrGO samples were performed to
rGO, and ZrGO powder samples are shown in Figure 1 (a). investigate their thermal stability. The samples were heated
The XRD pattern of GO shows a characteristic intense under N2 flow (200 mL min–1) from room temperature to
peak (001) at 2θ = 10.442° with a layer-to-layer d-spacing 600 °C at a heating rate of 5 °C min–1. The GO seems ther-
of 0.8465 nm. After the reduction of GO to rGO with mally unstable as it starts to lose weight as the temperature
hydrazine hydrate, the peak at 2θ = 10.442° disappears and increases. The GO sample loses ca. 30 % of its mass below
a broad peak (002) with a d-spacing of 0.3585 nm appears 200 °C owing to the evaporation of absorbed water. How-
at 2θ = 13.5 to 35°. The higher d-spacing value of the GO ever, the maximum weight loss occurs at ca. 200 °C and is
sample suggests the presence of a single-molecule-thick attributed to the decomposition of the oxygen-containing
layer of water molecules attached through intermolecular functional groups in the GO layers and the formation of
hydrogen bonding as these intercalate between the sheets, CO, CO2, and steam.[24] The rGO loses ca. 13 % of its mass
along with other oxygen-containing functional groups such at ca. 100 °C, and this can be attributed to the loss of water
as epoxy, hydroxy, carbonyl, and carboxyl groups, bound in the form of steam. The other downward trends above
covalently to sp3-hybridized carbon atoms, which get 200 °C signify the removal of the remaining labile oxygen-
pushed out of the plane as a result. This results in a more containing functional groups.[26,27] The rGO is much more
open structure. Upon reduction with hydrazine hydrate, the thermally stable than GO owing to the removal of a larger
removal of these intercalated molecules causes a reduction number of oxygen-containing functional groups by the re-
in the d-spacing.[22–25] Moreover, the broad peak in the duction process. The rGO retains ca. 64 % of its weight to

Eur. J. Inorg. Chem. 2015, 1912–1923 1913 © 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 10990682c, 2015, 11, Downloaded from [Link] by Indian Institute Of Technology, Wiley Online Library on [18/06/2026]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link] FULL PAPER

600 °C, and the weight loss above 600 °C is attributed to and 1152 cm–1 arise because of multiple phonon scattering
the degradation of the carbon skeleton. The ZrGO nano- processes in ZrGO, and the prominent feature at ν̃ =
composite is even more stable and shows much less weight 334 cm–1 is assigned to the second-order Raman scattering
loss than rGO. Thus, it can be safely concluded that the arising from zone-boundary phonons E2 (high)–E2 (low),
addition of ZnO nanoparticles increases the thermal sta- whereas the broad peak at ν̃ = 1152 cm–1 in the spectrum
bility of graphene oxide (GO). has contributions from the 2A1 (LO) and 2E1 (LO) modes
The FTIR transmittance spectra (KBr) of GO, rGO, and at the point of the Brillouin zone.[33] After the reduction
the ZrGO nanocomposite are shown in Figure 1 (c). The and decoration of the rGO layers with ZnO nanoparticles,
FTIR spectrum of the GO sample illustrates a wide band changes to the positions and intensity ratio of the G (IG)
at ν̃ = 3450 cm–1, which is attributed to the O–H stretching and D (ID) bands were observed. The position of the G
of the intercalated water trapped between the pristine band in the Raman spectra of graphene samples is highly
graphite sheets upon oxidation, and the peaks at ν̃ = 1055, sensitive to doping.[34] This is why the G band position in
1740, 1625, and 1232 cm–1 correspond to epoxy or alkoxy the Raman spectrum of the ZrGO sample shifts to higher
(C–O), carboxylic or carbonyl (C=O), skeletal vibrations of frequency relative to the G band position in the Raman
unreacted pristine graphite C=C bonds, and C–OH stretch- spectrum of the pristine rGO sample. This blueshift in the
ing peaks, respectively.[28] In the FTIR spectrum of rGO, G band is attributed to the electron-accepting behavior of
the C=O and C–OH peaks have mostly disappeared, ZnO in the composite[35] and, consequently, ZrGO exhibits
whereas the intensity of the C–O peak has decreased. This higher electric conductivity in comparison to that of the
provides evidence for the substantial reduction of the oxy- pristine rGO sample, which will be discussed later. The in-
gen moieties in GO under the action of hydrazine hydrate. tensity of the D peak is directly related to the number of
Two new peaks in the rGO spectrum at ν̃ = 2924 and sp2-hybridized in-plane carbon atoms; after reduction, the
2865 cm–1 represent asymmetric and symmetric stretching increase in the intensity of the D peak points to the forma-
CH2 vibrations, respectively.[28] The FTIR spectrum of tion of more sp2 domains. However, changes in the relative
ZrGO is mostly similar to that of rGO except for an ad- intensities of the D and G peaks in the Raman spectra of
ditional peak at ν̃ = 471 cm–1, which is attributed to the GO, rGO, and ZrGO help to verify the reduction of GO,[36]
Zn–O stretching mode.[29] and a slight increase in the ID/IG ratio reveals a decrease in
Raman spectroscopy is a standard nondestructive tech- number of sp2 domains. The ID/IG ratio and the Raman
nique that is extensively used to investigate the ordered/dis- excitation laser energy, El = 2.41 eV (λl = 514.5 nm), can be
ordered crystal structures of nanographitic materials such utilized to find out the in-plane crystallite size (La) of the
as carbon nanotubes, carbon fibers, and glassy carbon. The sample by using the model given in Equation (1).[37]
Raman spectra of GO, rGO, and ZrGO are shown in Fig-
ure 1 (d) and possess two intense peaks, which correspond La [nm] = (2.4 ⫻ 10–10)λl4(ID/IG)–1 (1)
to the well-known D and G bands. The existence of the
sharp D band (related to the order/disorder of the sample) By applying the above formula, we calculated the La val-
reveals the defective nature of the GO, rGO, and ZrGO ues of GO, rGO, and ZrGO to be 18.46, 16.97, and
crystal structures, whereas the sharp G band is representa- 16.42 nm, respectively, as shown in Table 1. We observed
tive of the stacking nature of the samples. The D band orig- that the ZrGO sample exhibits smaller in-plane crystallite
inates from the breathing mode of k-point phonons of A1g sizes (La) than the pure rGO sample, which is consistent
symmetry, whereas the G band arises from the first-order with previous results.[34]
Brillouin scattering of the E2g phonons of the planar sp2-
hybridized carbon atoms.[30] The chemically synthesized
Table 1. D and G band positions [cm–1], ID/IG ratios, and in-plane
GO shows a broad G band and an intense D band at ν̃ = crystallite sizes La [nm] of the GO, rGO, and ZrGO powder sam-
1597 and 1368 cm–1, respectively. In addition to the G and ples calculated from their Raman spectra.
D bands, the Raman spectrum of the ZrGO nanocomposite
Sample D band G band ID/IG (peak ID/IG (fit- La
possesses a few extra peaks that are the signatures of the position position intensity) ted area)
ZnO nanoparticles in the composite. The peak at ν̃ =
GO 1370.01 1599.31 0.9569 0.9221 18.24
441 cm–1 is due to the nonpolar Raman-active optical E2 rGO 1363.98 1593.42 0.9978 0.9912 16.97
(high) phonon mode, a characteristic branch of the wurtzite ZrGO 1363.98 1602.25 0.9989 1.0245 16.42
lattice structure of zinc oxide associated with the vibrations
of the O atoms.[31] The intensity of the peak is much re-
duced in comparison to that reported for pure zinc oxide, The chemical states of the elements present in the sam-
mainly because of the ZrGO interaction. Furthermore, the ples were investigated by X-ray photoelectron spectroscopy
peak at ν̃ = 585 cm–1 is attributed to the E1 longitudinal (XPS). The survey spectra of GO and rGO possess only
optical (LO) mode phonons arising from defects such as peaks of C and O, whereas that of ZrGO contains some
oxygen vacancies or interstitial zinc atoms in zinc oxides.[32] extra peaks of Zn, as shown in Figure 2 (a). The survey
The strong intensity of the E1 (LO) mode in the Raman spectrum of the rGO sample shows that the intensity of the
spectrum is attributed to the presence of oxygen-deficient O 1s peak decreases after the reduction process, whereas
ZnO nanoparticles in the composite. The peaks at ν̃ = 334 that of C 1s increases. The intensities of the C 1s and O 1s

Eur. J. Inorg. Chem. 2015, 1912–1923 1914 © 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 10990682c, 2015, 11, Downloaded from [Link] by Indian Institute Of Technology, Wiley Online Library on [18/06/2026]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link] FULL PAPER

Figure 2. (a) XPS survey spectra of GO, rGO, and ZrGO samples. (b) Magnified Zn 2p core level XPS spectrum of ZrGO. (c) Deconvol-
uted high-resolution XPS C 1s core level spectra of rGO and ZrGO. (d) Deconvoluted high-resolution XPS O 1s core level spectra of
rGO and ZrGO.

core levels in the XPS spectrum of ZrGO are lower than O=C–OH groups), the most-prominent central binding en-
those in the spectrum of rGO; this implies that there is a ergy (531.8 eV) peak is linked to singly bonded oxygen-con-
greater extent of reduction and a lower carbon content in taining functional groups such as alcohols, epoxides, and
the ZrGO composite. The possible reason for this is the ethers (C–OH), whereas the higher binding energy
partial decoration of the graphene sheets by ZnO nanopar- (533.4 eV) peak is attributed to the singly bonded oxygen
ticles. In Figure 2 (b), the high-resolution Zn 2p scan shows atoms in carboxy and ester functionalities and the presence
doublet spectral lines at binding energies of 1022.2 and of bound water in the sample.[40,41] The deconvoluted O 1s
1045.3 eV. These spectral lines are attributed to Zn 2p3/2 XPS spectrum of ZrGO possesses similar peaks to the those
and 2p1/2 electrons, respectively, with a spin–orbit splitting in the O1s XPS spectrum of rGO, but these peaks shift
(ΔE) of 23.1 eV. This confirms that the chemical state of to lower binding energy owing to the attachment of ZnO
the Zn2+ ions in the ZrGO sample remains uninfluenced by nanoparticles to the ZrGO hybrids (Figure 2, d). Moreover,
the presence of the rGO sheets. The distinct and sharp peak the intensities of the lower and higher binding energy peaks
at a binding energy of 1022.2 eV is associated with Zn spe- increase, whereas the intensity of the central binding energy
cies in the completely oxidized state.[38] The magnified core peak decreases compared with those of rGO.
level XPS C 1s spectrum of rGO and ZrGO samples were The TEM image in Figure 3a shows that the rGO sheet
deconvoluted into four Gaussian peaks that correspond to is continuous and consists of a few wrinkles (as indicated by
the following functional groups: nonoxygenated graphitic the arrows in Figure 3, a), which are expected for graphene
sp2 carbon atoms (C=C/C–C), epoxy and alkoxy (C–O), surfaces. The TEM image of the rGO nanosheet decorated
carbonyl (C=O), and carboxylate (O–C=O) groups, as with quasispherical ZnO nanoparticles of ca. 25 nm is
shown in Figure 2 (c).[24] The C–C peak for the ZrGO sam- shown in Figure 3 (b). The ZnO nanoparticles are agglom-
ple shifts slightly towards lower binding energy in compari- erated and randomly attached onto the surfaces of the
son to that of rGO; this implies that the ZnO in the ZrGO graphene sheets. The arrows in Figure 3 (b) indicate the
sample behaves as a p-type dopant,[39] and this result is con- graphene sheet and the ZnO nanoparticles anchored on it.
sistent with the Raman analysis. In Figure 2 (d), the O 1s The selected-area electron diffraction (SAED) pattern of
XPS spectrum of rGO is deconvoluted into three Gaussian the ZrGO nanocomposite (Figure 3, c) reveals that the ZnO
peaks, which correspond to different oxygen species in the nanoparticles are polycrystalline in nature with the wurtzite
sample. The lower binding energy (ca. 530 eV) peak is at- structure, consistent with the XRD results. The HRTEM
tributed to doubly bonded oxygen atoms (i.e. C=O and image of the ZrGO composite is shown in Figure 3 (d). The

Eur. J. Inorg. Chem. 2015, 1912–1923 1915 © 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 10990682c, 2015, 11, Downloaded from [Link] by Indian Institute Of Technology, Wiley Online Library on [18/06/2026]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link] FULL PAPER

Figure 3. (a) TEM micrograph of a continuous and wrinkled rGO nanosheet. (b) TEM image of an rGO nanosheet decorated with
quasispherical ZnO nanoparticles. (c) SAED pattern of ZrGO nanocomposite. (d) HRTEM image of ZrGO nanocomposite; the magnified
“A” region represents the HRTEM image of a rGO nanosheet, and the magnified “B” region represents the HRTEM image of an
individual ZnO nanoparticle attached to the graphene surface.

magnified “A” region represents the HRTEM image of the 0.787 nm, which corresponds to the (100) crystal plane of
rGO nanosheet; the nanosheet is composed of three layers ZnO.
and possesses lattice fringes with estimated interplanar The field-emission SEM (FESEM) images of the rGO
spacing of ca. 0.354 nm, which corresponds to the (002) and ZrGO powder samples are shown in Figure 4a and b,
plane of graphite. The magnified “B” region in Figure 3 (d) respectively. The energy-dispersive X-ray spectroscopy
is the HRTEM image of an individual ZnO nanoparticle (EDX) profile of a small physical area of the ZrGO powder
attached on the graphene surface. This image reveals that sample shown in Figure 4 (b) is presented in Figure 4 (c).
the lattice spacing between the two adjacent planes is ca. Here, the x axis represents the energy in keV, and the y axis

Figure 4. FESEM images of (a) rGO and (b) ZrGO powder samples. (c) EDX spectrum of a small physical area shown in image (b); the
inset shows the elemental composition of the ZrGO sample as determined by EDX analysis. (d)–(f) Uniform elemental distributions of
C, O, and Zn in the ZrGO sample detected by the X-ray mapping process.

Eur. J. Inorg. Chem. 2015, 1912–1923 1916 © 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 10990682c, 2015, 11, Downloaded from [Link] by Indian Institute Of Technology, Wiley Online Library on [18/06/2026]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link] FULL PAPER

represents the count in arbitrary units. The EDX profile

(Figure 4, c) exhibits strong C, O, and Zn peaks. The quali-
tative analysis of the C, O and Zn in the sample as deter-
mined by EDX analysis is shown in the inset of Figure 4
(c). The uniform elemental distributions of C, O, and Zn in
the sample as detected through the X-ray mapping process
are illustrated in Figure 4 (d–f). The AFM images of the
ZrGO powder sample (Figure S1, Supporting Information)
show that the ZrGO composite is formed by layer-by-layer
self-assembly, which is in good agreement with our SEM
analysis results.

Electrical Properties

The electrical behavior of the as-prepared rGO and

ZrGO powder samples was investigated in the temperature
range 300–373 K with a two-terminal arrangement (see Ex-
perimental Section), and the measured normalized resis- Figure 5. I–V curves for the (a) rGO and (b) ZrGO samples at
different temperatures. (c) Normalized resistance versus tempera-
tance value of rGO is consistent with the previously re- ture curves for the rGO and ZrGO samples. (d) Plots of ln G versus
ported theoretical and experimental results.[42,43] The nor- T–1/4 for the rGO and ZrGO samples.
malized resistance values for each sample at different tem-
peratures were determined by dividing the resistance values
rGO and ZrGO samples are shown in Figure 5 (d). As rGO
by the initial values.
is a highly disordered system with low carrier density, its
The observed resistance values for rGO and ZrGO are
temperature-dependent transport mechanism could be ana-
in the range ca. 0.6 to 0.3 MΩ and 0.3 MΩ to 18 kΩ, respec-
lyzed by the variable-range hopping (VRH) model for 3D
tively, within the temperature range 308–373 K. The ob-
transport behavior with Equation (2),[49] which provided a
tained resistivity values are based on measurements with
good fit with the experimental results, as shown in Figure 5
one sensor. However, the results were confirmed by several
(d).
reproducible statistical runs. The resistivity of the ZrGO
sample is several times lower than that of rGO. As equal G(T) = G1exp(–B/T1/4) (2)
amounts of GO and same synthesis technique were used
for the synthesis of the rGO and ZrGO samples, the lower Here, B is a constant given by B = 2.1{1/[kBα3D(εF)]}1/4, kB
resistivity of ZrGO is a consequence of the ZnO nanopar- is the Boltzmann constant, α is the localization length of
ticles tightly bound to the surface of the rGO sheets the charge carriers, and D(εF) is the electronic density of
through electrostatic attraction between the Zn2+ ions and states.
the negatively charged rGO sheets.[44] The I–V curves for
the rGO and ZrGO samples exhibit almost linear character-
Gas-Sensing Properties and Mechanism
istics, as shown in Figure 5 (a and b), respectively, and thus
indicate the Ohmic nature of the samples in the abovemen- The chemical reduction of GO by hydrazine hydrate can-
tioned temperature range. This linear behavior of the I–V not remove the oxygen-containing functional groups (carb-
curves demonstrates the metallic nature of the rGO and oxylic and epoxide) completely, and some of them still re-
ZrGO samples.[45] At normal temperature, the ZrGO sam- main attached to the basal plane, as confirmed by the XPS
ple behaves as a p-type material; therefore, we can assume results. These residual groups possibly increase the number
that the ZnO nanoparticles on the surface of rGO shift the of holes in the sample, as they are considered to have elec-
Fermi level downwards and, consequently, increase the rela- tron-accepting behavior.[50] This is the reason why rGO nor-
tive change in the resistivity, as shown in Figure 5 (c).[46] mally acts as a p-type semiconducting material. NO2 gas is
However, this has to be investigated further. The conductiv- a strong oxidizing gas and is known for its highly electron-
ity of rGO is lower than that of ZrGO, and the possible withdrawing nature. When rGO and ZrGO coil sensors
reason for this is that residual defects in rGO[47] prevent the were exposed to different concentrations of NO2 gas, the
transport of charge carriers, whereas the ZnO nanoparticles adsorbed electrophilic NO2 molecules on the surfaces of the
in the ZrGO sample provide an alternative path for charge rGO and ZrGO samples extract electrons from these sam-
transport by a hopping mechanism in the defective rGO. ple surfaces. Therefore, the exposure of both of these sens-
The conductivity of the rGO and ZrGO samples depends ing materials to NO2 gas increases the number of conduc-
on temperature and decreases with increasing temperature, tion holes in the samples and, hence, reduces their resis-
as shown in Figure 5 (c). However, this behavior highly de- tance. Thus, we can conclude that NO2 gas behaves as a p-
pends on carrier-phonon scattering with the temperature type dopant for both rGO and ZrGO. The changes in the
change.[48] Plots of ln G (conductance) versus T–1/4 for the resistance of the rGO and ZrGO sensors upon exposure to

Eur. J. Inorg. Chem. 2015, 1912–1923 1917 © 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 10990682c, 2015, 11, Downloaded from [Link] by Indian Institute Of Technology, Wiley Online Library on [18/06/2026]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link] FULL PAPER

Figure 6. (a) Percentage change in the resistance of rGO and ZrGO sensors for 50 ppm NO2 concentration at room temperature (298 K).
(b) Percentage change in the resistance of ZrGO sensor for different concentrations (ppm) of NO2 gas at room temperature (298 K). (c)
Power-law fit to the magnitude of percentage change in the resistance, that is, the percentage response (S) of the ZrGO sensor versus
NO2 concentration; the inset shows S versus NO2 concentration with error bars. (d) Response times of the sensor as a function of
different NO2 concentrations.

50 ppm of NO2 at room temperature are shown in Figure 6 we also performed several gas-sensitivity experiments with
(a). The percentage change in the sensor resistance is de- the as-synthesized ZnO nanoparticles that we used to form
fined by Equation (3). the ZrGO composite. We did not observe any noticeable
response for pristine ZnO at 50 °C, which would be re-
(Rg – Ra)/Ra ⫻ 100 = ΔR/Ra ⫻ 100 (3) quired for a comparative study with rGO and ZrGO. The
sensitivity of ZnO for NO2 gas is observed at ca. 200 °C and
Ra and Rg are the resistances of the sensor in air and in the
has been well established by several research groups.[51–53]
test gas, respectively.
Basically, ZnO is an n-type semiconductor, but its nature in
We define the percentage response (S) of the sensor as
a composite depends on the interaction of ZnO with graph-
the magnitude of the percentage change in the sensor resis-
ene.[54]
tance as shown in Equation (4).
In the ZrGO sample, ZnO behaves as a p-type dopant,
S = |ΔR/Ra| ⫻ 100 (4) as confirmed by Raman spectroscopy, XPS, and the better
electrical properties of ZrGO than those of the rGO sam-
The resistances of the rGO and ZrGO sensors decrease ple. However, further investigation is required to under-
significantly as the NO2 gas molecules are adsorbed on stand the mechanism of the rGO–ZnO interaction that re-
their surfaces (Figure 6, a). For the ZrGO sensor, the mag- sults in a better NO2 response for ZrGO than for rGO. The
nitude of the percentage change in the sensor resistance, percentage changes to the resistance of the ZrGO sensor
that is, S (14.3), the response time (96 s), and the recovery upon exposure to different concentrations (ppm) of NO2
time (1552 s; defined as the time it takes the sensor to reach gas at room temperature are shown in Figure 6 (b). The
90 % of the saturation value in a NO2 atmosphere and 10 % resistance of the ZrGO sensor has a rational dependence
of the base value in air, respectively) are better than the on the concentration of NO2 gas; it decreases as the concen-
values for S (10.8), response time (161 s), and recovery time tration of the test gas increases. The percentage changes in
(3034 s) for the rGO sensor. This implies that ZrGO is a the resistance of the sensor for each NO2 concentration
better sensing material than rGO. As the same synthesis were measured when the sensor had recovered its initial re-
parameters and an equal amount of GO were used to syn- sistance value (baseline value) after the sensing chamber
thesize the rGO and ZrGO samples, the better sensing per- had been degassed. This sensor exhibits noticeable changes
formance of ZrGO is attributed to the interaction of the in resistance for NO2 concentrations above 5 ppm at room
ZnO nanoparticles in this nanocomposite sample. However, temperature. Moreover, the curve for the magnitude of the

Eur. J. Inorg. Chem. 2015, 1912–1923 1918 © 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 10990682c, 2015, 11, Downloaded from [Link] by Indian Institute Of Technology, Wiley Online Library on [18/06/2026]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link] FULL PAPER

change in the sensor resistance, S, versus NO2 concentra- peratures are shown in Figure 7 (a), in which “1” and “2”
tion (C) shows an excellent power-law (S = ACα, A is a are the points at which gas is injected into the sensing
constant) dependence (Figure 6, c); this relationship arises chamber and removed from it, respectively. To investigate
because of the adsorption or interaction of NO2 with the the effect of temperature on the performance of the ZrGO
sensor surface as well as the change of surface potential. sensor, a graph is plotted of percentage response versus
From this fit, the calculated value of α is 0.58, which is in working temperature (Figure 7, b). The ZrGO sensor exhib-
agreement with the prediction of a rational value (1 or 1/2) its the highest response (ca. 32 %) at 50 °C; this value is ca.
for the power-law exponent.[55,56] The inset of Figure 6 (c) 2.2 times higher than that at room temperature, and the
shows the percentage response versus NO2 concentrations increase in response might be due to the higher rate of ad-
graph with error bars, which represent the standard devia- sorption on the surface of the ZrGO sample at 50 °C. As
tion at a fixed concentration of NO2, and the error bars are the temperature increases above 50 °C, the percentage re-
small in all cases. This implies a low experimental variabil- sponse decreases, but, concurrently, an improvement in the
ity. The response time curve of the sensor for exposure to response and recovery times of the sensor were observed.
different concentrations of NO2 (Figure 6, d) reveals that The variations of the response and recovery times of the
the response time of the sensor decreases as the gas concen- sensor with respect to temperature for 50 ppm NO2 concen-
tration increases. This is because the NO2 molecules take tration are shown in Figure 7 (c). The response and recov-
longer to equilibrate on the surface of the sensing material ery times of the ZrGO sensor are in the ranges 94–31 s and
at lower concentrations. However, at room temperature, the 26–3 min, respectively.
recovery times (not shown here) of this sensor for all con- Above 50 °C, the sensor resistance gradually increases
centrations are much larger than the response time, as a (Figure 5, c), and this increase in resistance may be due to
consequence of the strong chemisorption of NO2 molecules surface oxidation of the reduced graphene oxide sheets in
on the surface of the sensing material (ZrGO). The strong the ZrGO sample. Moreover, at higher temperatures, the
chemisorption may occur owing to the molecular interac- increased desorption of the NO2 molecules from the sensing
tions of NO2 molecules with higher-energy binding sites, surface will improve the recovery times of the sensor.[62] To
such as structural defects and residual oxygen functional demonstrate repeatability, the ZrGO sensor was exposed to
groups.[57] The recovery time of this sensor can be improved 50 ppm NO2 for five successive cycles at 50 °C, and the re-
either by exposure to UV light or heat-treatment through sults are shown in Figure 7 (d). We observed that the per-
external sources, as reported previously for other graphene- centage response of the sensor decreases after the initial few
based sensors.[58–61] The percentage changes in the sensor cycles, possibly because of the incomplete desorption of
resistance for 50 ppm NO2 concentration at different tem- NO2 molecules from the sensing surface.[63] These remain-

Figure 7. (a) Percentage change in the resistance of the ZrGO sensor as a function of time for 50 ppm NO2 at different temperatures. (b)
Percentage response (S) curve of the sensor for 50 ppm NO2 concentration as a function of working temperature. (c) Variation of response
and recovery times of the sensor with respect to temperature for 50 ppm NO2 concentration. (d) Percentage response of the sensors for
50 ppm NO2 for five successive test cycles at 50 °C.

Eur. J. Inorg. Chem. 2015, 1912–1923 1919 © 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 10990682c, 2015, 11, Downloaded from [Link] by Indian Institute Of Technology, Wiley Online Library on [18/06/2026]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link] FULL PAPER

ing molecules on the sensing surface would reduce the of both rGO- and ZrGO-based sensors increases. However,
number of available adsorption sites for the next sensing at higher humidity levels, the increase in the resistance of
cycle, and a state of saturation is reached after a few cycles; the ZrGO sample is slightly lower than that for rGO; there-
therefore, the number of available adsorption sites becomes fore, the swelling effect is smaller for ZrGO than for rGO,
almost constant for each subsequent cycle. The saturation that is, at high RH values, the contribution of ionic conduc-
value of the percentage response of the ZrGO sensor after tivity in the ZrGO sample is slightly greater than the contri-
a few cycles is ca. 27 %. bution of ionic conductivity in rGO.
To explore the effect of relative humidity on the resis- It is interesting to explore the effect of humidity on the
tance of the fabricated NO2 gas sensors, we introduced dif- ZrGO nanocomposite because, in this case, we observed
ferent levels of relative humidity (RH) into the test chamber that the stability of the nanocomposite system is higher
to measure the variation in resistance of the fabricated de- than those of pristine ZnO and rGO for an optimum con-
vice at 50 °C. The graphs of relative humidity versus the centration of ZnO in ZrGO nanocomposite. The observed
electrical resistance of the samples are presented in Fig- results have good consistency with the TGA data. Only a
ure 8. The obtained results reveal that the pristine ZnO gas slight variation in the resistance was observed above 75 %
sensor exhibits a noticeable variation in resistance at 60 % RH; this stability to humidity is highly desirable for gas-
relative humidity. For rGO, the resistance varies above 70 % sensing applications. The stability as well as the better sens-
relative humidity. We observed that several layers of water ing results may be attributed to the optimum concentration
molecules will be physically adsorbed on the chemisorbed of ZnO molecules in the nanocomposite system, which was
layer at the higher humidity levels (60 % or more).[64] The obtained by introducing ZnCl2 aqueous solution (20 mL,
physisorbed layers of water molecules accelerate the transfer 10 mm). We tried several concentrations of ZnCl2 solution
of protons (H+) or hydronium ions (H3O+) according to the ranging from 5–15 mm to aqueous solutions of GO
well-known Grotthuss transport mechanism, H2O + (20 mL). At concentrations of ZnCl2 solution higher than
H3O+ 씮 H3O+ + H2O.[65] In this mechanism, a proton is 10 mm, the nanocomposite system behaves like a pristine
attached to a water molecule to form a hydronium ion, ZnO surface owing to the deposition of excess ZnO mo-
which releases another proton to a second water molecule lecules on the rGO nanosheet surfaces. Similarly, at lower
without rotating. The second water molecule accepts this concentrations of ZnCl2 (⬍10 mm), the nanocomposite sys-
proton and releases the third proton and so on. This proton tem behaves like rGO surfaces as the rGO nanosheets are
transfer process goes on through the physisorbed water lay- not completely decorated by ZnO molecules. Particularly,
ers in such a way that the mean lifetime of any particular at the optimum concentration, all of the available delocal-
hydronium ion is estimated to be ca. 2 ps or less.[66] Thus, at ized π electrons are conjugated and the in situ formed ZnO
higher RH values, the physisorbed layers exhibit liquidlike molecules cover all possible surfaces of the rGO nanosheet
behavior and H+ can move freely; as a result, the resistance and, as a result, it exhibits excellent sensing properties.
of the ZnO sample decreases. Further, at higher humidity In future, we would like to investigate the effect of other
levels, the water molecules physisorbed on the surfaces of gases such as O3, CO, H2, and NO on the gas-sensing prop-
the rGO and ZrGO samples enter into the rGO layers and erties of the ZrGO sensor. We are doing separate studies in
lead to an increase in the interlayer distance between the this direction, and the results will be reported very soon.
rGO sheets. This phenomenon is termed the “swelling ef-
fect”[67] and reduces the connectivity among rGO sheets in
pristine rGO and ZrGO samples; as a result, the resistance Conclusions
Chemically reduced graphene oxide (rGO) and ZrGO
nanocomposite samples were synthesized by hydrolysis, and
morphological, structural, and spectroscopic studies were
performed by TEM, FESEM, EDX, XRD, TGA, Raman
spectroscopy, FTIR spectroscopy, and XPS, which evi-
denced the successful synthesis of ZrGO composites and
the homogeneous distribution of ZnO nanoparticles on the
surfaces of the rGO sheets. The electrical and gas-sensing
properties of the rGO and ZrGO nanocomposite samples
were extensively monitored, and the results show that the
ZrGO composite possesses better electrical conductivity
and NO2 gas sensing properties than those of the rGO sam-
ple. The enhancements in the electrical and gas-sensing
properties of the nanocomposite are attributed to the inter-
action of the ZnO nanoparticles with the rGO sheet. The
effects of temperature on the response and recovery times
Figure 8. The resistance versus relative humidity curves for ZnO, of the ZrGO nanocomposite at 50 ppm NO2 were analyzed
rGO, and ZrGO nanocomposite at 50 °C. extensively. The ZrGO nanocomposite sensor exhibits the

Eur. J. Inorg. Chem. 2015, 1912–1923 1920 © 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 10990682c, 2015, 11, Downloaded from [Link] by Indian Institute Of Technology, Wiley Online Library on [18/06/2026]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link] FULL PAPER

highest response (ca. 32 %) for 50 ppm NO2 at 50 °C. Ad- tuning the heating voltage, and the temperature was recorded with
ditionally, the effects of relative humidity on the resistances a K-type thermocouple. The powder sample of the sensing material
of the ZnO, rGO, and ZrGO samples were investigated, and was gently ground with a small amount of ethanol in an agate
the results show that the resistance of ZrGO depends very mortar to form a paste, which was then coated between the gap of
two Pt electrodes with the help of a drawing brush, and this system
little on the relative humidity. We believe that the present
was dried at 60 °C for 5 h in a vacuum oven. The sensing assembly,
work could be extended to the synthesis of various compos- containing the sensing material coated on the Pt electrodes, heating
ites of graphene with other metal oxides and could be help- coil, and temperature sensor, was placed in a stainless steel housing
ful for investigations of their sensing properties for various of volume 500 cm3. The whole sensing setup is shown in Figure 9.
gases at different temperatures.

Experimental Section
Synthesis of GO: The starting material, GO, was synthesized by an
improved method reported by Marcano et al.[68] The obtained solid
GO sample of distinct brownish color was dried in a vacuum oven
at 45 °C for 3 d to remove traces of moisture before further work
was undertaken.
Synthesis of Graphene ZnO Nanocomposite: To synthesize the
rGO–ZnO nanocomposite (ZrGO), GO (40 mg) was first dispersed
in Millipore water (18 MΩ cm, 80 mL) through 4 h of intense ultra-
sonication with a bath sonicator (Wensar, 150 W, 40 ⫾ 3 kHz).
Henceforth, the pH was adjusted to 10 by the addition of 28 % Figure 9. (a) Schematic diagram of sensing setup. (b) Digital image
ammonia solution. To this solution, ZnCl2 solution (20 mL, of coil sensor with heating coil and thermocouple. (c) Schematic
10 mm) was added with vigorous stirring by a magnetic stirrer. Af- diagram of coil sensor with heating coil.
ter the addition of 85 % hydrazine hydrate (25 μL), the solution in
a 250 mL round-bottomed flask was heated under reflux at 90 °C Electrical and Gas-Sensing Properties: The electrical and gas-sens-
with a silicon oil bath for 5 h. The mixture was then filtered ing properties of the rGO and ZrGO powder samples were investi-
through a nylon membrane filter (47 mm, 0.2 μm pore size) gated with the coil sensing setup (Figure 9), which works as a two-
mounted on a vacuum filter assembly. To remove unreacted chlor- terminal device. Before the investigations of the electrical and gas-
ide ions, the residue was washed with deionized water (⫻ 3) and sensing properties, the sensors were heated at 100 °C for 3 h to
ethanol (⫻ 3). The obtained product was vacuum-dried at 45 °C for avoid the influence of moisture and other impurities from the sens-
12 h. Finally, the material was annealed under a nitrogen atmo- ing material. To determine the gas-sensing performance of the
sphere at 500 °C for 5 h. A sample of reduced graphene oxide rGO- and ZrGO-based gas sensors, different quantities of the test
(rGO) was also synthesized for comparison by following the same gas (NO2) from a 1000 ppm canister were inserted into the sensing
method. chamber through the insertion window with a plastic syringe. Be-
fore the gas was injected into the chamber, the plastic syringe was
Characterization: XRD (Bruker d-8 Advanced diffractometer, Cu-
calibrated by the standard method based on the relationship: ca-
Kα1 radiation, λ = 1.5406 Å) was utilized to perform structural
pacity of syringe used ⫻ concentration of canister (ppm) = capacity
analysis of the as-synthesized GO, rGO, and ZrGO powder sam- of sensing container ⫻ required concentration (ppm). Once the sen-
ples. The SEM measurements were performed with a Quanta 200 sor achieved saturation after the insertion of the test gas into the
FEG instrument with an accelerating voltage of 20 kV. The Raman chamber, the gas was removed from the chamber by the removal
analysis of the samples was performed with a WI Tec alpha 300 of the rubber gasket and the simultaneous operation of the vacuum
SR instrument with excitation with a 488 nm Ar ion laser, and an pump. The 1000 ppm NO2 gas canister was purchased from Chem-
Andor 420 BR DD Peltier-cooled (–72 °C) CCD detector was used.
tron Science Laboratories Pvt. Ltd. Mumbai, India. The relative
The XPS spectra were recorded with a PHI 5000 Versa Probe II
humidity (RH) was ca. 57 % before the start of the experiment at
XPS system (ULVAC–PHI, Inc., Japan). The XPS measurements
room temperature (25 °C), and the temperature was kept in the
were performed with a monochromatic Al-Kα source (hν =
range 25–140 °C for further tests. A Keithley 617 programmable
1486.6 eV), a hemispherical analyzer, and a multichannel detector. electrometer was used to monitor the variations in electric current.
The typical pressure in the analysis chamber during the measure- For humidity measurements, two-probe arrangements were used to
ments was ca. 10–10 Torr. The elemental compositions were deter-
monitor the variation in electrical resistance at different RH values
mined from the spectra acquired at a pass energy of 117.4 eV. High-
at a constant temperature of 50 °C. A hygrometer was used to mon-
resolution spectra were obtained with an analyzer pass energy of
itor the humidity changes. The humidity measurement setup pos-
29.35 eV with steps of 0.125 eV and 50 ms time per step. An Ar+
sesses two channels, one of which is used to allow the humid air
ion gun was used for surface cleaning and depth profiling.
into the chamber, and the other is used to remove the humid air.
Fabrication of Sensing Devices: To fabricate sensing devices with Supporting Information (see footnote on the first page of this arti-
ZrGO and rGO powder samples, two Pt wires of diameter 0.3 mm cle): 2D and 3D AFM images of the ZrGO powder sample.
were tightly wrapped ca. 1 mm apart around a quartz tube of
13 mm outer diameter, as shown in Figure 9. These two Pt wires
act as the terminals of the device for electrical measurements. A Acknowledgments
ceramic coil heater, which was fabricated with 28 gauge Nicrom
wire (35 cm long), was inserted into this small quartz tube. The The authors are thankful to Prof. O. N. Srivastava (B. H. U.) and
working temperature of the ceramic coil heater was controlled by Prof. Neeraj Khare (IIT Delhi) for helpful discussions. The authors

Eur. J. Inorg. Chem. 2015, 1912–1923 1921 © 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 10990682c, 2015, 11, Downloaded from [Link] by Indian Institute Of Technology, Wiley Online Library on [18/06/2026]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link] FULL PAPER

gratefully acknowledge the Government of India, BRNS (DAE), [31] V. Pachauri, C. Subramaniam, T. Pradeep, Chem. Phys. Lett.
for financial assistance, and N. K. acknowledges the financial sup- 2006, 423, 240–246.
port from the Council of Scientific and Industrial Research (CSIR), [32] X.-Y. Ye, Y.-M. Zhou, Y.-Q. Sun, J. Chen, Z.-Q. Wang, J.
New Delhi through a Senior Research Fellowship (2012–2013). Nanopart. Res. 2009, 11, 1159–1166.
[33] R. Cusco, E. A.-Llado, J. Ibanez, L. Artus, Phys. Rev. B 2007,
75, 165202.
[1] A. K. Geim, Science 2009, 324, 15304. [34] L. S. Panchakarla, K. S. Subrahmanyam, S. K. Saha, A. Gov-
[2] A. K. Geim, K. S. Novoselov, Nat. Mater. 2007, 6, 183–191. indaraj, H. R. Krishnamurthy, U. V. Waghmare, C. N. R. Rao,
[3] E. V. Castro, K. S. Novoselov, S. V. Morozov, N. M. R. Peres, Adv. Mater. 2009, 21, 4726–4730.
J. M. B. L. dos Santos, J. Nilsson, F. Guinea, A. K. Geim,
[35] Y. Y. Hui, G. Tai, Z. Sun, Z. Xu, N. Wang, F. Yan, S. P. Lau,
A. H. C. Neto, Phys. Rev. Lett. 2007, 99, 216802.
Nanoscale 2012, 4, 3118–3122.
[4] X. Du, I. Skachko, A. Barker, E. Y. Andre, Nat. Nanotechnol.
[36] T. N. Narayanan, Z. Liu, P. R. Lakshmy, W. Gao, Y. Nagaoka,
2008, 3, 491.
[5] K. S. Novoselov, Z. Jiang, Y. Zhang, S. V. Morozov, H. L. D. S. Kumar, J. Lou, R. Vajtai, P. M. Ajayan, Carbon 2012, 50,
Stormer, U. Zeitler, J. C. Maan, G. S. Boebinger, P. Kim, A. K. 1338–1345.
Geim, Science 2007, 315, 1379. [37] B. Jayasena, S. Subbiah, Nanoscale Res. Lett. 2011, 6, 1–7.
[6] M. D. Stoller, S. Park, Y. Zhu, J. An, R. S. Ruoff, Nano Lett. [38] A. Prakash, S. K. Misra, D. Bahadur, Nanotechnology 2013,
2008, 8, 3498–3502. 24, 095705.
[7] M. I. Katsnelson, K. S. Novoselov, A. K. Geim, Nat. Phys. [39] Q. Su, S. Pang, V. Alijani, C. Li, X. Feng, K. Mullen, Adv.
2006, 2, 620–625. Mater. 2009, 21, 3191–3195.
[8] J. I. Paredes, S. Villar-Rodil, A. Martínez-Alonso, J. M. D. Tas- [40] L. Al-Mashat, K. Shin, K. Kalanter-Zadeh, J. D. Plessis, S. H.
con, Langmuir 2008, 24, 10560–10564. Han, R. W. Kojima, R. B. Kaner, D. Li, X. Gou, S. J. Ippolito,
[9] N. T. Hu, Y. Y. Wang, J. Chai, R. G. Gao, Z. Yang, E. S. W. W. Wlodarski, J. Phys. Chem. C 2010, 114, 16168–16173.
Kong, Y. F. Zhang, Sens. Actuators B 2012, 163, 107–114. [41] R. Rozada, J. I. Paredes, S. V.-Rodil, A. Martínez-Alonso,
[10] J.-C. Lin, B.-R. Huang, T.-C. Lin, Appl. Surf. Sci. 2014, 289, J. M. D. Tascón, Nano Res. 2013, 6, 216–233.
384–387. [42] F. Vasko, V. Ryzhii, Phys. Rev. B 2007, 76, 233404.
[11] Jaleh, A. Jabbari, Appl. Surf. Sci. 2014, 320, 339–347. [43] Q. Shao, G. Liu, D. Teweldebrhan, A. A. Balandin, Appl. Phys.
[12] Q. Xiang, J. Yu, M. Jaroniec, Chem. Soc. Rev. 2012, 41, 782– Lett. 2008, 92, 202108.
796. [44] S. Stankovich, R. D. Piner, X. Q. Chen, N. Q. Wu, S. T.
[13] W. Yuan, G. Shi, J. Mater. Chem. A 2013, 1, 10078–10091. Nguyen, R. S. Ruoff, J. Mater. Chem. 2006, 16, 155–158.
[14] Y. P. Dan, Y. Lu, N. J. Kybert, Z. T. Luo, A. T. C. Johnson, [45] O. Akhavan, E. Ghaderi, Carbon 2012, 50, 1853–1860.
Nano Lett. 2009, 9, 1472–1475.
[46] G. Giovannetti, P. A. Khomyakov, G. Brocks, V. M. Karpan, J.
[15] G. H. Lu, S. Park, K. H. Yu, R. S. Ruoff, L. E. Ocola, D. Ro-
van den Brink, P. J. Kelly, Phys. Rev. Lett. 2008, 101, 026803.
senmann, J. H. Chen, Acs Nano 2011, 5, 1154–1164.
[47] V. Lopez, R. S. Sundaram, C. G.-Navarro, D. Olea, M. Burgh-
[16] M. Gautam, A. H. Jayatissa, Solid-State Electron. 2012, 78,
159–165. ard, J. G.-Herrero, F. Zamora, K. Kern, Adv. Mater. 2009, 21,
[17] G. Singh, A. Choudhary, D. Haranath, A. G. Joshi, N. Singh, 4683–4686.
S. Singh, R. Pasricha, Carbon 2012, 50, 385–394. [48] A. Seeger, K. Clausecker, Phys. Status Solidi B 1971, 46, 137–
[18] S. M. J. Khadem, Y. Abdi, S. Darbari, F. Ostovari, Curr. Appl. 150.
Phys. 2014, 14, 1498–1503. [49] J. R. Hauptmann, T. Li, S. Petersen, J. Nygard, P. Hedegard,
[19] A. S. M. I. Uddin, G.-S. Chung, Sens. Actuators B 2014, 205, T. Bjornholm, B. W. Laursen, K. Norgaard, Phys. Chem.
338–344. Chem. Phys. 2012, 14, 14277–14281.
[20] Potential Health Effects from Release of Nitrogen Dioxide [50] S. Gilje, S. Han, M. Wang, K. L. Wang, R. B. Kaner, Nano
(NO2): [Link] Lett. 2007, 7, 3394–3398.
information/NEPA/cfodocs/[Link].55084. [51] B. Shouli, L. Xin, L. Dianqing, C. Song, L. Ruixian, C. Aifan,
[Link]/[Link]. Sens. Actuators B 2011, 153, 110–116.
[21] L. Talazac, J. Brunet, V. Battut, J. P. Blanc, A. Pauly, J. P. Ger- [52] N. Barsan, D. Koziej, U. Weimar, Sens. Actuators B 2007, 121,
main, S. Pellier, C. Soulier, Sens. Actuators B 2001, 76, 258– 18–35.
264. [53] H.-U. Lee, K. Ahn, S.-J. Lee, J.-P. Kim, H.-G. Kim, S.-Y. Jeong,
[22] C. Hontoria-Lucas, A. J. Lopez-Peinado, J. D. D. Lopez-Gon- C.-R. Cho, Appl. Phys. Lett. 2011, 98, 193114.
zalez, M. L. Rojas-Cervantes, R. M. Martin-Aranda, Carbon [54] W. Geng, X. Zhao, H. Liu, X. Yao, J. Phys. Chem. C 2013,
1995, 33, 1585–1592. 117, 10536–10544.
[23] D. A. Dikin, S. Stankovich, E. J. Zimney, R. D. Piner, G. H. B. [55] N. S. Ramgir, M. Ghosh, P. Veerender, N. Datta, M. Kaur,
Dommett, G. Evmenenko, S. T. Nguyen, R. S. Ruoff, Nature D. K. Aswal, S. K. Gupta, Sens. Actuators B 2011, 156, 875–
2007, 448, 457–460. 880.
[24] S. Stankovich, D. A. Dikin, R. D. Piner, K. A. Kohlhaas, A. [56] N. Yamazoe, K. Shimanoe, Sens. Actuators B 2008, 128, 566–
Kleinhammes, Y. Jia, Y. Wu, S. T. Nguyen, R. S. Ruoff, Carbon 573.
2007, 45, 1558–1565. [57] J. T. Robinson, F. K. Perkins, E. S. Snow, Z. Q. Wei, P. E. Shee-
[25] Z. Fan, K. Wang, T. Wei, J. Yan, L. Song, B. Shao, Carbon
han, Nano Lett. 2008, 8, 3137–3140.
2010, 48, 1686–1689.
[58] D. R. Patil, L. A. Patil, IEEE Sens. J. 2007, 7, 434–439.
[26] N. Hu, R. Gao, Y. Wang, Y. Wang, J. Chai, Z. Yang, E. S.-W.
Kong, Y. Zhang, J. Nanosci. Nanotechnol. 2012, 12, 99–104. [59] I. S. Kang, H. M. So, G. S. Bang, J. H. Kwak, J. O. Lee, C. W.
[27] G. Wang, Z. Yang, X. Li, C. Li, Carbon 2005, 43, 2564–2570. Ahn, Appl. Phys. Lett. 2012, 101, 123504.
[28] S. Wang, P.-J. Chia, L.-L. Chua, L.-H. Zhao, R.-Q. Png, S. [60] F. Yavari, Z. Chen, A. V. Thomas, W. Ren, H. M. Cheng, N.
Sivaramakrishnan, M. Zhou, R. G.-S. Goh, R. H. Friend, Koratkar, Sci. Rep. 2011, 1, 166.
A. T.-S. Wee, P. K.-H. Ho, Adv. Mater. 2008, 20, 3440–3446. [61] G. Lu, L. E. Ocola, J. Chen, Appl. Phys. Lett. 2009, 94, 083111.
[29] R. Y. Hong, J. H. Li, L. L. Chen, D. Q. Liu, H. Z. Li, Y. Zheng, [62] P. Qi, O. Vermesh, M. Grecu, A. Javey, Q. Wang, H. Dai, S.
J. Ding, Powder Technol. 2009, 189, 426–432. Peng, K. J. Cho, Nano Lett. 2003, 3, 347–351.
[30] M. A. Pimenta, G. Dresselhaus, M. S. Dresselhaus, L. G. Can- [63] M. G. Chung, D. H. Kim, H. M. Lee, T. Kim, J. H. Choi, D.-
cado, Z. A. Jorio, R. Saito, Phys. Chem. Chem. Phys. 2007, 9, k. Seo, J.-B. Yoo, S.-H. Hong, T. J. Kang, Y. H. Kim, Sens.
1276–1290. Actuators B 2012, 166, 172–176.

Eur. J. Inorg. Chem. 2015, 1912–1923 1922 © 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 10990682c, 2015, 11, Downloaded from [Link] by Indian Institute Of Technology, Wiley Online Library on [18/06/2026]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
[Link] FULL PAPER

[64] Z. Wang, Y. Xiao, X. Cui, P. Cheng, B. Wang, Y. Gao, X. Li, [68] D. C. Marcano, D. V. Kosynkin, J. M. Berlin, A. Sinitskii, Z.
T. Yang, T. Zhang, G. Lu, ACS Appl. Mater. Interfaces 2014, Sun, A. Slesarev, L. B. Alemany, W. Lu, J. M. Tour, ACS Nano
6, 3888–3895. 2010, 4, 4806–4814.
[65] N. Agmon, Chem. Phys. Lett. 1995, 244, 456–462. Received: December 9, 2014
[66] F. M. Ernsberger, J. Am. Ceram. Soc. 1983, 66, 747–750. Published Online: March 6, 2015
[67] S. Park, J. An, J. W. Suk, R. S. Ruoff, Small 2010, 6, 210–212.

Eur. J. Inorg. Chem. 2015, 1912–1923 1923 © 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim