Selective shape control of cerium oxide nanocrystals for photocatalytic and chemical sensing effect

Article abstract of DOI:10.1039/c9ra01519a

In this study, we report the precise shape control of crystalline cerium oxide, whose morphology changes between nanorods and nanoparticles in a short time. The proposed synthetic route of cerium oxide nanorods was highly dependent on the reaction time, and 10 min was determined to be the optimum synthetic condition. The cerium oxide nanorods were further converted into nanoparticles by the spontaneous assembly of cerium oxide nanoparticles into nanorods. The transmission electron microscopy results showed that the synthesized nanorods grew with high crystallinity along the ?110? direction. The cerium oxide nanorods have been proven to be very efficient electron mediators for use as excellent photocatalytic materials and highly sensitive chemical sensors. The chemical sensor fabricated on a carbon paper substrate showed the high sensitivity of 1.81 μA mM^-1 cm^-2 and the detection limit of 6.4 μM with the correlation coefficient of 0.950.

Full text of DOI:10.1039/c9ra01519a

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Selective shape control of cerium oxide
nanocrystals for photocatalytic and chemical
Cite this: RSC Adv., 2019, 9, 13829
sensing effect†
Nam-Woon Kim, Dong-Kyu Lee* and Hyunung Yu *a
ab
c
In this study, we report the precise shape control of crystalline cerium oxide, whose morphology changes
between nanorods and nanoparticles in a short time. The proposed synthetic route of cerium oxide
nanorods was highly dependent on the reaction time, and 10 min was determined to be the optimum
synthetic condition. The cerium oxide nanorods were further converted into nanoparticles by the
spontaneous assembly of cerium oxide nanoparticles into nanorods. The transmission electron
microscopy results showed that the synthesized nanorods grew with high crystallinity along the h110i
direction. The cerium oxide nanorods have been proven to be very efficient electron mediators for use
as excellent photocatalytic materials and highly sensitive chemical sensors. The chemical sensor
Received 28th February 2019
Accepted 18th April 2019
DOI: 10.1039/c9ra01519a
ꢀ1
ꢀ2
fabricated on a carbon paper substrate showed the high sensitivity of 1.81 mA mM cm
and the
rsc.li/rsc-advances
detection limit of 6.4 mM with the correlation coefficient of 0.950.
of applications. Therefore, electrochemical sensors that are
simple, fast-responding, and capable of detection are the
1
. Introduction
13
The rapid increase in organic pollutants as a result of agricul- alternative candidates attracting signicant attention. For the
ture and large-scale industrial development has caused serious fabrication of efficient electrochemical sensors, it is important
problems in the global scientic community. Due to their to recalibrate the material, size, structure, and properties of the
toxicities and persistence, organic pollutants esthetically electrodes. For application in the practical industry, it is
1
–4
contaminate, disturb, and eutrophicate aquatic organisms.
Ethanol, a colorless liquid among a variety of organic pollut- photocatalysts.
ants, is prone to be released into the environment when
Metal oxides such as ZnO and TiO
Dyes have also been found chemical sensors and organic contamination photocatalysts
in freshwater, industrial wastewater, and marine water because because their redox properties are related to the electron–hole
essential to manufacture simple and low-cost sensors and active
14,15
2
are widely used in
1,4,5
abundantly used in the industry.
6
3,4,14–25
they have high and stable solubility. Ethanol can cause oxida- pair on the metal oxide surface with analyte species.
To
tive damage to the brain, stomach, liver, and red blood cells improve the sensing and photocatalytic properties of nano-
when it enters the body, whereas dyes exhibit the adverse effects materials, it is necessary to reduce the particle size and increase
of toxicity and carcinogenicity and are hazardous to the envi- the active surface area of nanomaterials. Previous studies have
1,4
ronment and human health; the effective detection of toxic shown that decreasing the particle size improves conductivity,
26–28
ethanol and dyes in the environment is signicantly important, electrical, sensing, and catalytic properties.
In addition,
and thus, a variety of methods have been reported in this one-dimensional (1D) nanomaterials, such as nanowires,
7–12
regard.
Previously, various chromatographic and spectro- nanorods, and nanotubes, exhibit signicantly different phys-
scopic techniques have been used for the detection of toxic ical and chemical properties from the bulk state due to their
solvents. However, these conventional techniques are oen volume ratio and increased area for quantum connement.
29–33
complex, slow, and difficult to be implemented in a wide range They also form the building blocks for nanosized devices.
Cerium oxide is a good material candidate for its well-known
function in applications involving catalysts, sensors, optics,
Korea Research Institute of Standards and Science (KRISS), Daejeon 34113, Republic adsorption, electrochemistry, batteries, and energy storage.
6,34,35
a
of Korea. E-mail: peacewithu@kriss.re.kr
A variety of techniques have been used to synthesize nanoscale
cerium oxide using surfactants or templates, such as precipi-
tation, sonochemical, hydrothermal, reverse micelles, micro-
b
Korea Advanced Institute of Science and Technology (KAIST), Daejeon 34141, Republic
of Korea
c
Chungbuk National University (CBNU), Cheongju 28644, Republic of Korea. E-mail:
33,35–41
emulsion, and sol–gel processes.
Among these,
dklee@chungbuk.ac.kr
hydrothermal synthesis is advantageous for applications in
practical industries because simple processes can be easily
†
Electronic supplementary information (ESI) available. See DOI:
0.1039/c9ra01519a
1
This journal is © The Royal Society of Chemistry 2019
RSC Adv., 2019, 9, 13829–13837 | 13829

View Article Online
RSC Advances
Paper
scaled at a lower cost. Cerium oxide nanorods are efficient diffractometer (XRD, SmartLab, Rigaku Co.). The sample was
˚
ꢁ
ꢀ
1
during CO oxidation, and they are oen synthesized by the scanned with Cu-Ka radiation (l ¼ 1.5406 A) at a rate of 0.2 s
hydrothermal method with a high-pressure reactor (autoclave) to conrm the crystal structure, and the measured data were
at 100–200 C for 10–72 h.
ꢁ
42–46
These synthesis conditions are characterized by the Joint Committee on Power Diffraction
oen difficult to emulate in the industry due to cost and time Standards (JCPDS) database. The surface functional groups
factors. Therefore, it is necessary to develop a very simple were observed with a confocal Raman microscope (Raman,
process that can be tactfully used in industrial applications at DXR, Thermo Fisher Scientic). The granulation product was
lower costs.
excited with a 532 nm laser and the spectrum was measured
In this study, we investigated the morphology trans- with a 50ꢂ objective combined with a pinhole (diameter: 50
formation of cerium oxide nanocrystals with the reaction time. mm). The X-ray photoelectron spectra (XPS, K-alpha, Thermo VG
Cerium oxide grows into nanorods and then rapidly reassem- Scientic) were used to analyze the valence states of the cerium
bles into nanoparticles. Our cerium oxide nanorods exhibited oxide nanocrystals. Ultraviolet visible spectroscopy (UV/Vis
worthwhile photocatalytic behavior over organic dyes and yiel- spectrophotometer, S-3100, Scinco Co.) was used to determine
ded a much-enhanced sensing effect as an ethanol sensor.
the optical properties of cerium oxide nanocrystals with large
absorption in the ultraviolet (UV) region. The analytical method
involved the total reection of the powder samples using the
integrating sphere (Integrating Sphere, Diffuse Reector, Scinco
Co.). The bandgap of the cerium oxides was determined from
the measured data by the Kubelka–Munk function.
2. Experimental
2.1. Synthesis of cerium oxide nanocrystals
All the reagents are of the analytical grade and were used
without further purication. The synthesis of cerium oxide
47
2
.3. Photocatalytic experiment involving cerium oxide
nanorods is described in more detail in our earlier report.
Cerium chloride (CeCl $7H O, 99.9%, Sigma Aldrich Co., Ltd.),
ammonium chloride (NH Cl, $99.5%, Sigma Aldrich Co., Ltd.),
nanorods and nanoparticles
3
2
4
The photocatalytic decomposition of methyl orange was inves-
tigated by optical absorption spectroscopy. The photocatalytic
reaction was carried out by mixing 20 mL of methyl orange
solution (1.5 ꢂ 10 M, C H N NaO S, extra pure, Samchun
ammonia solution (NH OH, 28.0–30.0%, Junsei Chemical Co.,
4
Ltd.), and deionized water were used to prepare the cerium
oxide nanocrystals. Here, a 0.1 mol aliquot of CeCl $7H O and
ꢀ4
3
2
1
4
14
3
3
1
mol of NH
4
Cl were added to a round-bottomed ask con-
Co., Ltd.) and 10 mg of cerium oxide nanorods in a 50 mL quartz
beaker. The well-dispersed mixed solutions were irradiated
using a 365 nm UV lamp (ltered UV lamp, VL-4.LC 4 W, Vilber
Lourmat) and 2 mL samples were collected at 20 min intervals
during irradiation. The collected methyl orange solution was
separated from the cerium oxide nanorods via centrifugation.
The UV/Vis absorbance was measured for the solution only aer
the nanorods were fully removed. The cerium oxide nano-
particles prepared aer the synthesis time of 20 min were also
analyzed under the same conditions for comparison.
ꢁ
taining 20 mL of deionized water that boils at 100 C. Then,
12 mL of boiling deionized water was added to 8 mL of
ammonia solution under magnetic stirring. The diluent
ammonia solution was directly injected into the cerium
resource solution. It was conrmed that Ce(OH)3 sol was
formed as soon as the solution was injected at the initial time of
the reaction. Then, Ce(OH) sol rapidly converted into CeO
(0
3
2ꢀx
3
+
<
4
x < 0.5) via Ce(OH) formation at the basic condition as Ce is
4
+
known to be less stable than Ce . The synthesis of cerium oxide
nanoparticles varied depending on the reaction time of 1–
2
0 min. The resulting precipitates were collected by centrifu-
2
.4. Fabrication and characterization of ethanol chemical
ꢁ
gation (2000 rpm, 5 min), washed with 100 C deionized water,
and dried at 60 C for 24 h. The dried cerium oxide nano-
particles were calcined at 300 C for 3 h (5 C min ). The color
of the synthesized product was observed to be straw-yellow.
sensor
ꢁ
ꢁ
ꢁ
ꢀ1
Ethanol chemical sensors were fabricated by coating the cerium
oxide nanorods onto carbon electrodes (carbon paper, Alfa
Aesar Co., Ltd.; surface area: 1 cm ). The carbon electrodes were
2
cleaned with acetone, methanol, and isopropyl alcohol before
coating. The cerium oxide nanorod powder was dispersed in
2.2. Characterizations of cerium oxide nanocrystals
The morphology of the cerium oxide nanocrystals was observed a mixture of isopropyl alcohol, distilled water, and Naon
with an environmental scanning electron microscope (ESEM, (Naon D-521, Alfa Aesar Co., Ltd.) for coating. The prepared
ꢁ
Quanta 650 FEG, FEI Company) at an accelerating voltage of 30 mixed solution was coated onto the carbon electrode of 100 C
ꢁ
kV. The samples for ESEM imaging were prepared by dispersing by the drop-casting method before being dried at 70 C for 6 h.
the cerium oxide nanocrystals in distilled water and drying To measure the sensing properties, the electrochemical cell was
them on a silicon wafer. Their chemical composition and constructed using a simple two-electrode system (SP-150, Bio-
weight were measured by energy-dispersive spectroscopy (EDS, Logic). A carbon electrode coated with cerium oxide nanorods
ULTRA PLUS, Carl Zeiss NTS GmbH). A high-resolution trans- was used as the working electrode and a Pt wire was used as the
mission electron microscope (HRTEM, FEI Tecnai G2 F30 300 counter electrode. The current response was measured from 0.0
kV, FEI Company) was used to measure the intrinsic micro- to +1.2 V. The prepared sensor was evaluated by adding various
structure and intrinsic crystallographic properties. The crystal- concentrations (0.17 to 85 mM) of ethanol (CH
3 2
CH OH,
linity of the nanocrystals was determined using an X-ray $99.8%, Sigma Aldrich Co., Ltd.) to 20 mL phosphate buffer.
13830 | RSC Adv., 2019, 9, 13829–13837
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
RSC Advances
The phosphate buffer solution was prepared by mixing 0.2 M increased to 15 min, it is apparent that smaller nanoparticles
sodium hydrogen phosphate (Na HPO , $99.0%, Sigma Aldrich (15–25 nm) were attached to the rod surface. Finally, it was
2
4
Co., Ltd.) and 0.2 M sodium dihydrogen phosphate (NaH PO , conrmed that cerium oxide was present only in the form of
2
4
$
99.0%, Sigma Aldrich Co., Ltd.) solution in 200 mL deionized nanoparticles. It is assumed that metastable nanorods collapse
51
water. The prepared cerium oxide nanoparticles were also to convert nanoparticles with minimal surface free energy.
fabricated as a sensor and analyzed under the same conditions. Fig. S1† shows the EDS results obtained to analyze the stoi-
Selectivity analyses (9100 HPLC system, Young Lin) and sensor chiometric composition of the nanorods at a reaction time of
characteristics for other pollutants (acetone, CH
3
COCH
3
,
3
10 min. The precursor chemical species of Cl and NH were
$
99.9%, Sigma Aldrich Co., Ltd.) were performed for completely removed by washing, and pure Ce and O were
comparison.
monitored. From the EDS analysis, the composition ratio of the
cerium oxide nanorods is expressed as CeO1.8 (Ce: 32.38 atom%;
O: 58.17 atom%). This is evident as the straw-yellowish cerium
powder in the optical images, as shown in Fig. 1f.
3. Results and discussion
3
.1. Growth process and properties of synthetic cerium
Fig. 2 shows the HRTEM and fast Fourier transform (FFT)
patterns of the synthesized cerium oxide nanorods. In the
HRTEM image of Fig. 2a, the clear (110) and (001) lattice fringe
oxide nanocrystals
Cerium oxide nanocrystals were prepared by hydrothermal
synthesis using an alkali solution. Pure cerium oxide powder is
white; however, the color of the synthesized powder was straw-
yellow and had a composition ratio of CeO2ꢀx (0 < x < 0.5). Fig. 1
shows a panorama of the ESEM images of the cerium oxide
morphology with reaction time. Aer the initial nucleation,
cerium oxide formed nanorods by self-assembly before
becoming spherical particles. At a reaction time of 1 min, both
spherical nanoparticles (size: ꢃ13 nm) and nanorods (length:
ꢃ100 nm) were observed, as shown in Fig. 1a. Aer 5 min, long
nanorods were prominent with some small debris of nano-
particles, as shown in Fig. 1b. Aer 10 min, denite cerium
oxide nanorods had lengths between 400 and 800 nm with
diameters of 20–50 nm, and the small spherical particles almost
completely disappeared. It can be presumed that the small
particles with the large surface energy per unit mass are more
soluble than the larger particles, according to the Ostwald
ripening phenomenon. Hence, large crystals continuously grow
˚
˚
directions, with interplanar spacings of 3.82 A and 5.41 A,
respectively, show that the nanorods grew along the preferred
34
growth direction, namely, h110i. According to the FFT anal-
ysis, three kinds of lattice fringe directions, namely, (111), (002),
and (220), of a typical nanorod were obtained. Fig. 2b shows the
selected area electron diffraction (SAED) pattern recorded to
investigate the crystal structure of the cerium oxide nanorods.
The SAED pattern reveals several Debye–Scherrer diffraction
rings that can be indexed to the crystal planes of a cubic uorite
structure with cerium oxide nanorods.
Fig. 3 shows the diffraction peaks in the XRD pattern of
calcined cerium oxide, matching the reported values of the
cerium oxide cubic uorite structure with a lattice constant of
˚
a ¼ 5.411 A and space group Fm3m (JCPDS no.: 43-1002). The
peak with the highest intensity is the (111) plane direction at 2q
ꢁ
¼
28.5 . The crystallite size of cerium oxide was determined by
the Debye–Scherrer equation with the highest intensity (111)
peak along the reaction time, as summarized in Table S1.† It is
interesting that the crystallite size increased from 79.7 to
48–50
by consuming smaller crystals.
As a result, our cerium oxide
nanorods were fabricated aer growth by the Ostwald ripening
process. As shown in Fig. 1d and e, as the reaction time further
1
13.7 nm with the reaction time, regardless of its morphological
Fig. 1 SEM images of cerium oxide synthesized at different reaction times of (a) 1 min, (b) 5 min, (c) 10 min, (d) 15 min, and (e) 20 min; (f)
photographic image of cerium oxide nanorods and nanoparticles. Scale bar: 300 nm.
This journal is © The Royal Society of Chemistry 2019
RSC Adv., 2019, 9, 13829–13837 | 13831

View Article Online
RSC Advances
Paper
ꢁ
Fig. 2 (a) HRTEM and FFT images and (b) SAED pattern of cerium oxide nanorods synthesized at 100 C for 10 min.
ꢀ
1
acoustic (2TA) mode (239.4 and 258 cm ) and defect-induced
ꢀ
1
(
D) or oxygen vacancies mode (596.6 and 596.5 cm ). An
47
earlier study reported that the F2g peak of cerium oxide
exhibits a relatively low Raman frequency, which can result
from the stabilizing effect of the oxygen structure in cerium
oxide. As shown in Fig. 4, the intensities of the oxygen
displacement and vacancy bands are higher for the cerium
oxide nanorods than those of the cerium oxide nanoparticles.
These trends are further revealed by the broadening of the F2g
ꢀ1
52
mode around the 460 cm band. A higher abundance of
oxygen vacancies in the rod when compared with that of the
particles indicates a more reducible surface oxygen species,
which is the main factor for the photocatalytic behavior or
52
chemical sensing. Fig. S2† shows the valence states of Ce and
O ions in the cerium oxide nanocrystals for comparison with the
Fig. 3 XRD patterns of cerium oxide at different reaction times of (a) oxygen vacancy rate on the surface. Both Ce and O ions have
1
min, (b) 10 min (nanorod), (c) 15 min, and (d) 20 min (nanoparticle).
similar peak positions, as observed in earlier reports. The Ce
5/2 (882.1 eV) and Ce 3d3/2 (900.9 eV) peaks are assigned to the
3d
4
+
change, indicating particle growth and ripening. Here, we need Ce ion and Ce 3d5/2 (884.4 eV) and Ce 3d3/2 (903.0 eV) peaks
to identify the growth direction during the shaping of the
cerium oxide nanomaterial. In this characteristic peak, there
are three kinds of planes: a stable and neutral plane (111),
47
a stable plane (110), and a plane of high energy (001). The
intensity ratio between (200) and (220) in the cerium oxide
nanocrystal structure suggests the growth direction of the
lattice parameter. The intensity ratio between (200) and (220) of
the synthesized cerium oxide has a value between 0.42 and 0.58.
The cerium oxide nanorods, with a reaction time of 10 min,
show a ratio of approximately 0.58, which approaches the ideal
47
value of 0.6. The nanorod crystallites tended to effectively grow
along the (220) axis, which has also been observed for some
47,50
other 1D materials.
Cerium oxide was synthesized without
other phases by matching the XRD pattern conrmed with the
SAED pattern.
The crystallinity and structural information of cerium oxide
were studied using a confocal Raman microscope. As shown in
Fig. 4a and b, the symmetric stretching mode of the Ce–O
vibrational unit (uorite structure, F2g), which is the main peak
ꢀ1
of cerium oxide, was observed at 458.7 and 461.5 cm (F2g
mode). In addition, we can observe the second-order transverse 10 min (nanorod) and (b) 20 min (nanoparticle).
Fig. 4 Raman spectra of cerium oxide at different reaction times of (a)
13832 | RSC Adv., 2019, 9, 13829–13837
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
RSC Advances
3
+
are assigned to the Ce ion. Further, it can be observed that the power was set to the minimum. The degradation rate of the
4
+
3+
mixed valence state of Ce and Ce ions were present as methyl orange chemical species by cerium oxide nanorods was
3
+
4+
satellites, and the ratio of Ce peak area/Ce peak area is monitored by measuring the absorbance aer every 20 min.
53
summarized in Table S2.† As revealed by the Raman analysis, Fig. 6a shows that the maximum absorbance of methyl orange
the XPS results also show that the cerium oxide nanorods have dye at 461 nm gradually decreases with UV irradiation time.
more oxygen vacancies on the surface than the cerium oxide This phenomenon shows that the dye efficiently decomposed
nanoparticles.
on the surface of the synthesized cerium oxide nanorods under
The optical properties of the prepared cerium oxide were UV irradiation. Furthermore, the cerium oxide nanorods aer
investigated using a UV/Vis reectance spectrophotometer with the photocatalytic reaction were fully separated by a centrifuge
an integrating sphere, as shown in Fig. 5. Absorption in the UV to identify no absorbance persisting in the range from 600 to
region of cerium oxide occurs by the charge transfer transition 700 nm. Approximately 50% of the methyl orange decomposed
2
ꢀ
4+
between the O 2p and Ce 4f states of O and Ce , respec- aer 80 min of UV irradiation. Fig. 6b shows a plot of the irra-
51
tively. The synthesized cerium oxide showed a high absorption diation time vs. the relative concentration of the dye from the
rate of approximately 90% in the UV region, while the reec- initial state during photolysis. No considerable decomposition
tance was more than 90% in the visible-light region, yielding the can be observed when the methyl orange solution was exposed
straw-yellow color (Fig. 1f). As the reaction proceeded from 10 to to UV irradiation without the cerium oxide nanorods. In addi-
20 min, a notable red-shi (deep yellow) was observed due to tion, when only the cerium oxide nanorods were added to the
the quantum connement effect of excitons as the size of methyl orange solution in the absence of UV irradiation, the
cerium oxide nanoparticle (crystallite size: 113.7 nm) became absorption rate was 11.5% aer 80 min. When the cerium oxide
larger than that of the cerium oxide nanorod (crystallite size: nanorods were exposed to light above the bandgap, electron
8
2.3 nm), as listed in Table S1.† In addition, cerium oxide and hole pairs were generated on the cerium oxide surface.
ꢀ
inherently has a strong electron–phonon coupling effect; Hydroxyl radical (OHc) and superoxide radical anions (O
2
c )
further, when the nanoparticle was more likely to aggregate originated from the water in the oxygen in combination with the
than the rod, this coupling modied the effective mass of the electron–hole pairs in the redox reaction, and they played
54
carriers and scattering by the lattice, leading to a red-shi. The a critical role in the oxidative reaction of the dye for photode-
6
inset in Fig. 5 shows the calculated Kubelka–Munk function to composition. Fig. 6c shows a comparison of the photocatalytic
measure the bandgap of cerium oxides. The bandgap was performance between the cerium oxide nanorods and cerium
determined to be ꢃ2.79 eV for the nanorods and 2.66 eV for the oxide nanoparticles when irradiated with 365 nm UV rays for
nanoparticles, which is good for harvesting a wider range of 80 min. The nanorods show ꢃ11.9% higher photocatalytic
light.
activity than that of the nanoparticles, with less light absor-
bance in the UV region, as shown in Fig. 5. Meanwhile, the
cerium oxide nanorods predominantly expose the reactive (110)
and (100) planes, while the nanoparticles mainly expose the less
reactive (111) plane on the surface. A typical cerium oxide
3.2. Photocatalytic degradation of methyl orange dye using
cerium oxide nanorods and nanoparticles
The photocatalytic activity of the prepared cerium oxide was nanoparticle comprises a polyhedron with eight (111) and six
investigated by performing decomposition of methyl orange (100) planes, while the rod structure prefers to expose four (110)
under irradiation with 365 nm UV light, where the irradiation and two (100) planes. The oxygen is conned mainly on the
surface of the nanoparticle, but it is occupied both on the
surface and bulk so the nanorod provides a much higher oxygen
capacity to signicantly enhance the redox property. Therefore,
the higher reactive plane and well-dened active sites induce
44,55
additional catalytic behavior with less UV absorbance.
In
addition, the geometrical 1D structure of cerium oxide nano-
56
rods enables fast and long-distance electron transport. We
conclude that cerium oxide nanorods decompose methyl
orange more effectively than cerium oxide nanoparticles due to
shape and structural advantages, as shown in Fig. 6. The
stability of cerium oxide nanorods as a photocatalyst was also
conrmed, as shown in Fig. S3.† A cyclic degradation experi-
ment was performed using cerium oxide nanorods fully sepa-
rated by a centrifuge aer 80 min of UV irradiation. The moiety
of methyl orange was adsorbed on the surface of the cerium
oxide, which decomposed 4% less than the rst decomposition
rate. The cerium oxide nanorods exhibited good photocatalytic
performance even aer decomposition was repeated for three
times.
Fig. 5 UV/Vis diffuse reflectance spectra of cerium oxide at different
reaction times of (a) 10 min (nanorod) and (b) 20 min (nanoparticle).
Inset: UV/Vis diffuse reflectance spectra plot as the Kubelka–Munk
function.
This journal is © The Royal Society of Chemistry 2019
RSC Adv., 2019, 9, 13829–13837 | 13833

View Article Online
RSC Advances
Paper
Fig. 6 (a) Typical plot for the change in the absorption spectra on irradiation of an aqueous suspension of cerium oxide nanorod containing
ꢀ
4
ꢀ4
methyl orange (1.5 ꢂ 10 M) at various time intervals. (b) Plot for the degree of decomposition (C/C
o
) of methyl orange (1.5 ꢂ 10 M) with
respect to UV exposure time. (c) Comparison of photocatalytic properties of cerium oxide nanorods with the nanoparticles (UV irradiation time:
8
0 min). (d) Decomposition degree of the methyl orange comparison with cerium oxide nanorod and nanoparticle with respect to UV irradiation
time.
3.3. Ethanol chemical sensor using cerium oxide nanorods
produce CH CHO (reaction (1)) and C H (reaction (2)) as the
3 2 4
and nanoparticles
respective intermediates.
To monitor the sensing behavior, the synthesized cerium oxide
was coated onto carbon paper, functioning as a working elec-
trode. To analyze the sensing parameters of the electrochemical
sensors, the current–voltage (I–V) curves were determined for
a 0.1 M phosphate buffer solution (PBS, pH ¼ 7.0) at various
ethanol concentrations ranging from 0.17 to 85 mM. Fig. 7a
C
2
H
5
OH / CH
3
CHO + H
2
(1)
(2)
C H OH / C H + H O
2
5
2
4
2
In the second step, the resulting CH CHO and C H are
3
2
4
shows the electrochemical effect of the carbon paper substrate oxidized by oxygen molecules physically adsorbed on the
4,57
on the I–V curve. The current decreased aer the coating of surface to yield CO , CO, and H O. Fig. 7c shows the electrical
2
2
cerium oxide nanorods (red circle) onto the substrate as response of cerium oxide nanorod thin lms by adding 0.17–
compared to pure carbon paper (black circle). It is natural that 85 mM of various ethanol concentrations to a 0.1 M PBS solu-
the resistance increases by the nanomaterials mixed with the tion. Each experiment was conducted by preparing a 0.1 M PBS
4
conductive binder uniformly coated on the sensor surface.
solution containing a fresh cerium oxide nanorod electrode and
Fig. 7b shows the current change in the electrochemical fresh ethanol. The current of the working electrode increased
response as a function of the working potential when exposed to with the concentration of ethanol. Thus, the conductivity of the
ethanol for the fabricated chemical sensor. There appears to be cerium oxide nanorod electrode increased according to the
no current response in the pure PBS solution, but the electrical concentration of the target chemical (EtOH). Fig. 7d shows
response signicantly increased with the addition of ethanol. a calibration curve plot determined from the data shown in
This phenomenon can be explained by the rapid electron Fig. 7c. The modied cerium oxide nanorod chemical sensor
ꢀ1
ꢀ2
exchange and excellent electrocatalytic oxidation properties of shows high sensitivity of 1.81 mA mM cm , detection limit of
the cerium oxide nanorods, which are suitable for ethanol- 6.4 mM, and correlation coefficient (R) of 0.950. The calibration
sensing applications. Generally, ethanol is subjected to a two- curve (Fig. 7d) shows a low concentration range up to 8.5 mM
step process in the I–V analysis to determine its conductivity. with good linear correlation of 95%. The prepared cerium oxide
The rst step involves dehydrogenation, or dehydration, to nanorod chemical sensor electrode exhibits an increase in
13834 | RSC Adv., 2019, 9, 13829–13837
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
RSC Advances
Fig. 7 I–V characteristics of cerium oxide: (a) comparison of the current with and without cerium-oxide-nanorod-coated carbon paper. (b)
Current test with ethanol (17 mM) in the cerium oxide nanorod. (c) Plot of the sensing behavior by a series of ethanol concentrations in the cerium
oxide nanorod. (d) Calibration plot to deduce ethanol sensitivity of the nanorod. (e) Plot of the sensing behavior with different concentrations of
ethanol in the cerium oxide nanoparticle. (f) Calibration plot to deduce ethanol sensitivity of the nanoparticle.
current with increasing ethanol content up to 8.5 mM, followed cerium oxide nanoparticles increased. Fig. 7f shows the cali-
by saturation. This can be attributed to the low availability of bration curve plotted using the data shown in Fig. 7e. The
free active sites because the concentration of ethanol is higher fabricated cerium oxide nanoparticle chemical sensor shows
ꢀ1
ꢀ2
than the active surface of the cerium oxide nanorod electrode sensitivity of 0.87 mA mM cm , detection limit of 14.7 mM,
for the adsorption of ethanol. Oxidation reactions using cerium and R of 0.93. The calibration curve shown in Fig. 7f has
58
oxide involve the participation of lattice oxygen species. Fig. 7e a concentration range up to 17 mM and R of 0.93. Earlier
42
shows the electrical response of cerium oxide nanoparticles reports have shown that cerium oxide nanorods have better
measured for comparison with the cerium oxide nanorods used CO oxidation performance with lower specic surface areas
as chemical sensors. For comparison, the experiment was also than the nanoparticles. This originates from the different
conducted under the same conditions as the nanorods using reactivities along the crystal facets. The cerium oxide nanorods
0.1 M PBS solution. By increasing the concentration of ethanol prefer to grow along the (110) and (100) planar surfaces of high
from 0.17 mM to 85 mM, the current at the working electrode of activity, while cerium oxide nanoparticles grow along the (111)
This journal is © The Royal Society of Chemistry 2019
RSC Adv., 2019, 9, 13829–13837 | 13835

View Article Online
RSC Advances
Paper
necessitating the use of a high-pressure reactor. The cerium
oxide nanorods dramatically changed in shape into nano-
particles in only 10 min. The synthesized cerium oxides are
present in a composition of CeO2 (0 < x < 0.5) and exhibit
ꢀx
bandgaps of 2.79 eV and 2.66 eV for the nanorod and nano-
particle, respectively. The photocatalytic properties of cerium
oxide were monitored by the decomposition of methyl orange.
The results show that by using cerium oxide nanorods, 50% of
the methyl orange dye was degraded in 80 min by simple
exposure to UV light at 365 nm. The production of chemical
sensors to detect ethanol was successfully accomplished using
a substrate made of carbon paper. The fabricated cerium oxide
nanorod chemical sensor exhibits high sensitivity of 1.81 mA
mM cm , detection limit of 6.4 mM, and R of 0.950. The
cerium oxide nanorods were approximately twice as sensitive as
cerium oxide nanoparticles. This indicates that the morphology
control of cerium oxide is critical in chemical sensor applica-
tions. We have developed a novel synthesis method to fabricate
cerium oxide nanorods that can be benecial for mass
production at lower costs in industrial applications. The
synthesized cerium oxide nanorods are suitable for the appli-
cation of photocatalysts for organic dye decomposition and are
effective and sensitive chemical sensors.
Fig. 8 Proposed mechanism of ethanol sensing in the presence of
cerium oxide nanorods.
ꢀ
1
ꢀ2
42
plane, which is less reactive. Many theoretical calculations
have shown the oxidation reactivity follows the order: (100) >
44,45
(110) > (111).
Therefore, a chemical sensor performance
accompanied by an oxidation reaction is highly dependent on
59,60
the crystal facet exposed to the particle surface.
It has also
been reported that a reduced cerium oxide (CeO2 ) structure
ꢀx
exhibits high photocatalytic efficiency due to reoxidation
61
resulting from the unstable oxygen vacancies. In summary,
our synthetic cerium oxide nanorods show high sensitivity as
compared to cerium oxide nanoparticles. Moreover, the conse-
quent chemical sensors fabricated using cerium oxide nano-
rods have higher sensitivity than those proposed in earlier
Conflicts of interest
There are no conicts to declare.
62
reports (Table S3†) using cerium oxide nanoparticles and are
successfully investigated as electrodes. Fig. S4† shows the
sensing properties of cerium oxide nanorods for other pollut-
ants such as acetone, which is an excellent chemical sensor for
ethanol. For comparing the sensing performances, acetone was
also added in the same volume ratio as ethanol under the same
conditions as those used in the ethanol experiment. The cerium
Acknowledgements
This work was supported by the Korea Research Institute of
Standards and Science under the project “Development of
hybrid microscope for multiscale 3D imaging (grant number
19011160)”. We also thank M. W. Pin for the TEM measurement
ꢀ1
ꢀ2
and analysis.
oxide nanorod sensor showed sensitivity of 1.99 mA mM cm
,
detection limit of 11.68 mM, and R of 0.81. To determine the
contamination selectivity of the cerium oxide nanorod chemical
sensor, we operated the chemical sensor with the same volume
of ethanol and acetone in a 0.1 M PBS solution. It was conrmed
that the sensitivity of cerium oxide nanorods toward acetone is
higher than that toward ethanol, as shown in Fig. S5.† The
different sensing behaviors suggest that the fabricated cerium
oxide nanorod chemical sensor can detect other pollutants in
water with good selectivity.
References
1
2
3
4
5
6
M. Faisal, S. B. Khan, M. M. Rahman, A. Jamal and A. Umar,
Mater. Lett., 2011, 65, 1400–1403.
S. B. Khana, M. M. Rahmana, E. S. Jangb, K. Akhtarc and
H. Hanb, Talanta, 2011, 84, 1005–1010.
M. Faisal, S. B. Khan, M. M. Rahmana, A. Jamala, A. M. Asiri
and M. M. Abdullaha, Chem. Eng. J., 2011, 173, 178–184.
S. B. Khan, M. Faisal, M. M. Rahman and A. Jamal, Sci. Total
Environ., 2011, 409, 2987–2992.
A schematic mechanism of ethanol sensing with cerium
oxide nanorods is shown in Fig. 8. As compared to the other
forms of cerium oxide, the enhanced catalysis of rods is strongly
inuenced by the crystal plane and geometrical structure. The
structural sensitivity of rods can be explained by the unique
surface structure of the (110) and (100) planes, which increases
L. Senf, F. Stapf, B. Diisedau and H. Frank, J. Therm. Anal.,
1988, 33, 371–374.
S. B. Khan, M. Faisal, M. M. Rahman, K. Akhtar, A. M. Asiri,
A. Khan and K. A. Alamry, Int. J. Electrochem. Sci., 2013, 8,
7284–7297.
44
the oxygen diffusion rates through the 1D cerium oxide axis.
7
8
N. Rastakhiz, A. Kariminik, V. S. Nejad and S. Roodsaz, Int. J.
Electrochem. Sci., 2010, 5, 1203–1212.
Y. Mo, Y. Okawa, T. Nakai, M. Tajima and K. Natukawa, Thin
Solid Films, 2002, 416, 248–253.
4
. Conclusion
In summary, we successfully controlled the shape of cerium
oxide nanocrystals by controlling only the reaction time without
9 X. H. Wu, Y. D. Wang, Y. F. Li and Z. L. Zhou, Mater. Chem.
Phys., 2002, 77, 588–593.
13836 | RSC Adv., 2019, 9, 13829–13837
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
RSC Advances
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
0 A. L. Prudenf and D. F. Ollis, Environ. Sci. Technol., 1983, 17, 36 J. Qian, Z. Chen, C. Liu, X. Lu, F. Wang and M. Wang, Mater.
28–631. Sci. Semicond. Process., 2014, 25, 27–33.
1 H. D. Burrows, M. L. Canle, J. A. Santaballa and S. Steenken, 37 T. Feng, X. Wang and G. Feng, Mater. Lett., 2013, 100, 36–39.
6
J. Photochem. Photobiol., B, 2002, 67, 71–108.
2 I. K. Konstantinou and T. A. Albanis, Appl. Catal., B, 2003, 42,
38 C. R. Michel and A. H. M. Preciado, Sens. Actuators, B, 2014,
197, 177–184.
3
19–335.
39 S. K. Jha, C. N. Kumar, R. P. Raj, N. S. Jha and S. Mohan,
Electrochim. Acta, 2014, 120, 308–313.
40 K. Singh, A. A. Ibrahim, A. Umar, A. Kumar, G. R. Chaudhary,
S. Singh and S. K. Mehta, Sens. Actuators, B, 2014, 202, 1044–
1050.
3 M. M. Rahman, A. Jamal, S. B. Khan and M. Faisal, J. Phys.
Chem. C, 2011, 115, 9503–9510.
4 M. M. Rahman, A. Jamal, S. B. Khan and M. Faisal, ACS Appl.
Mater. Interfaces, 2011, 3, 1346–1351.
5 M. M. Rahman, S. B. Khan, A. Jamal, M. Faisal and 41 K. Anupriya, E. Vivek and B. Subramanian, J. Alloys Compd.,
A. M. Asiri, Sens. Transducers J., 2011, 134, 32–44. 2014, 590, 406–410.
6 S. Wang, F. He, P. Dong, Z. Tai, C. Zhao, Y. Wang and F. Liu, 42 K. Zhou, X. Wang, X. Sun, Q. Peng and Y. Li, J. Catal., 2005,
J. Mater. Res., 2018, 33, 3917–3927. 229, 206–212.
7 S. Ge, L. Bao, Q. Shao, Q. Zhang and Q. Liu, J. Nanosci. 43 P. Stelmachowski, K. Ciura, P. Indyka and A. Kotarba, Mater.
Nanotechnol., 2017, 17, 833–836. Chem. Phys., 2017, 201, 139–146.
8 Y. Li, X. Deng, J. Tian, Z. Liang and H. Cui, Appl. Mater. 44 H. Tana, M. Zhang, J. Li, H. Li, Y. Li and W. Shen, Catal.
Today, 2018, 13, 217–227. Today, 2009, 148, 179–183.
9 Y. Li, Y. Xue, J. Tian, X. Song, X. Zhang, X. Wang and H. Cui, 45 X. S. Huang, H. Sun, L. C. Wang, Y. M. Liu, K. N. Fan and
Sol. Energy Mater. Sol. Cells, 2017, 168, 100–111. Y. Cao, Appl. Catal., B, 2009, 90, 224–232.
0 B. Yin and C. Liu, J. Nanosci. Nanotechnol., 2018, 18, 4771– 46 C. Sun, H. Li, H. Zhang, Z. Wang and L. Chen,
779. Nanotechnology, 2005, 16, 1454–1463.
1 X. Hu, H. Zhao, J. Tian, J. Gao, Y. Li and H. Cui, Sol. Energy 47 N. W. Kim, H. J. Maeng, D. K. Lee and H. Yu, J. Nanosci.
Mater. Sol. Cells, 2017, 172, 108–116. Nanotechnol., 2014, 14, 8834–8838.
2 F. Liu, G. Huang, X. Wang, X. Xie, G. Xu, G. Lu, X. He, J. Tian 48 P. W. Voorhees, J. Stat. Phys., 1985, 38, 231–252.
4
and H. Cui, Sens. Actuators, B, 2018, 277, 144–151.
3 X. Xie, X. Wang, J. Tian, J. Liu, H. Cao, X. Song, N. Wei and
H. Cui, Ceram. Int., 2017, 43, 5053–5060.
4 X. Xie, X. Wang, J. Tian, X. Song, N. Wei and H. Cui, Ceram.
Int., 2017, 43, 1121–1128.
49 D. Zhang, H. Fu, L. Shi, C. Pan, Q. Li, Y. Chu and W. Yu,
Inorg. Chem., 2007, 46, 2446–2451.
50 C. Pan, D. Zhang, L. Shi and J. Fang, Eur. J. Inorg. Chem.,
2008, 15, 2429–2436.
51 M. Y. Cui, J. X. He, N. P. Lu, Y. Y. Zheng, W. J. Dong,
W. H. Tang, B. Y. Chen and C. R. Li, Mater. Chem. Phys.,
2010, 121, 314–319.
5 J. Li, H. Cui, X. Song, N. Wei and J. Tian, Appl. Surf. Sci., 2017,
396, 1539–1545.
6 F. Meshkani and M. Rezaei, Powder Technol., 2010, 199, 144– 52 M. Kovacevic, B. L. Mojet, J. G. van Ommen and L. Lefferts,
48. Catal. Lett., 2016, 146, 770–777.
7 O. Tunusoglua, R. M. Espı, U. Akbey and M. M. Demir, 53 K. I. Maslakov, Y. A. Teterin, A. J. Popel, A. Y. Teterin,
1
¨
´ ¨
Colloids Surf., A, 2012, 395, 10–17.
8 T. Sreethawong, S. Ngamsinlapasathian and S. Yoshikawa,
Mater. Lett., 2012, 78, 135–138.
K. E. Ivanov, S. N. Kalmykov, V. G. Petrov, P. K. Petrov and
I. Farnan, Appl. Surf. Sci., 2018, 448, 154–162.
54 X. Lu, X. Li, F. Chen, C. Ni and Z. Chen, J. Alloys Compd.,
2009, 476, 958–962.
9 B. Gates, B. Mayers, B. Cattle and Y. Xia, Adv. Funct. Mater.,
2
002, 12, 219–227.
0 T. Li, S. Yang, L. Huang, B. Gu and Y. Du, Nanotechnology,
004, 15, 1479–1482.
1 Y. Huang, X. Duan, Q. Wei and C. M. Lieber, Science, 2001,
91, 630–633.
55 W. Lei, T. Zhang, L. Gu, P. Liu, J. Rodriguez, G. Liu and
M. Liu, ACS Catal., 2015, 5, 4385–4393.
56 Z. R. Tang, Y. Zhangab and Y. J. Xu, RSC Adv., 2011, 1, 1772–
1777.
57 S. B. Khan, M. M. Rahman, K. Akhtar, A. M. Asiri, J. Seo,
H. Han and K. Alamry, Int. J. Electrochem. Sci., 2012, 7,
4030–4038.
2
2
2 N. S. Arul, D. Mangalaraj, P. C. Chen, N. Ponpandian and
C. Viswanathan, Mater. Lett., 2011, 65, 2635–2638.
3 N. S. Arul, D. Mangalaraj, P. C. Chen, N. Ponpandian, 58 A. Trovarelli, Catal. Rev.: Sci. Eng., 1996, 38, 439–520.
P. Meena and Y. Masuda, J. Sol-Gel Sci. Technol., 2012, 64, 59 M. Kong, Y. Li, X. Chen, T. Tian, P. Fang, F. Zheng and
515–523.
X. Zhao, J. Am. Chem. Soc., 2011, 133, 16414–16417.
3
3
4 L. Liao, H. X. Mai, Q. Yuan, H. B. Lu, J. C. Li, C. Liu, 60 L. Ren, Y. Li, J. Hou, J. Bai, M. Mao, M. Zeng, X. Zhao and
C. H. Yan, Z. X. Shen and T. Yu, J. Phys. Chem. C, 2008,
12, 9061–9065.
5 A. Umar, R. Kumar, M. S. Akhtar, G. Kumar and S. H. Kim, J.
Colloid Interface Sci., 2015, 454, 61–68.
N. Li, Appl. Catal., B, 2016, 181, 625–634.
61 Y. Li, Q. Sun, M. Kong, W. Shi, J. Huang, J. Tang and X. Zhao,
J. Phys. Chem. C, 2011, 115, 14050–14057.
1
62 A. Khan, A. A. P. Khan, M. M. Rahman, A. M. Asiri and
K. A. Alamry, RSC Adv., 2015, 5, 105169–105178.
This journal is © The Royal Society of Chemistry 2019
RSC Adv., 2019, 9, 13829–13837 | 13837