Hierarchical Co3O4 nanostructures in-situ grown on 3D nickel foam towards toluene oxidation

Article abstract of DOI:10.1016/j.mcat.2018.05.006

Herein, a facile strategy for the in-situ growth of Co-based oxide precursors with unique hierarchical architectures oriented aslant or perpendicular to the Ni surfaces was reported. Toluene was chosen as probe molecule to evaluate their catalytic performance over the synthesized integrated Co-based oxide catalysts. The morphology of integrated Co-based oxide catalysts contained intercrossed frizzy nanosheets (NS), nanowires (NW), columnar nanoclusters (NC) and coral-like microspheres. The structure-performance relationship (SPR) is further confirmed via using H₂-TPR, XPS, Raman spectrum. The Co₃O₄ NC on Ni foam exhibited the excellent catalytic activity for toluene oxidation (T₁₀₀ = 270 °C, 50 °C lower than that of Co₂AlO₄ arrays with coral-like microspheres). The excellent catalytic performance of Co₃O₄ NC catalyst might be resulted from its lower-temperature reducibility, lattice distortion of spinel structure and abundant surface-adsorbed oxygen. The findings of this work provide a new perspective on the design of the high-efficiency integrated catalysts for VOCs combustion.

Full text of DOI:10.1016/j.mcat.2018.05.006

Molecular Catalysis 454 (2018) 12–20
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.elsevier.com/locate/mcat
Hierarchical Co O nanostructures in-situ grown on 3D nickel foam towards
3
4
toluene oxidation
Qi Zhanga,1, Shengpeng Moa,1, Bingxu Chen , Weixia Zhang , Chunlei Huang , Daiqi Ye
a
a
a
a,b,c,⁎
a
School of Environment and Energy, South China University of Technology, Guangzhou, 510006, PR China
Guangdong Provincial Key Laboratory of Atmospheric Environment and Pollution Control (SCUT), Guangzhou, 510006, PR China
b
c
Guangdong Provincial Engineering and Technology Research Centre for Environmental Risk Prevention and Emergency Disposal, South China University of Technology,
Guangzhou Higher Education Mega Centre, Guangzhou, 510006, PR China
A R T I C L E I N F O
A B S T R A C T
Keywords:
Catalytic oxidation
Integrated catalysts
Co O
3 4
Ni foam
Herein, a facile strategy for the in-situ growth of Co-based oxide precursors with unique hierarchical archi-
tectures oriented aslant or perpendicular to the Ni surfaces was reported. Toluene was chosen as probe molecule
to evaluate their catalytic performance over the synthesized integrated Co-based oxide catalysts. The mor-
phology of integrated Co-based oxide catalysts contained intercrossed frizzy nanosheets (NS), nanowires (NW),
columnar nanoclusters (NC) and coral-like microspheres. The structure-performance relationship (SPR) is further
Structure-performance relationship
confirmed via using H
activity for toluene oxidation (T100 = 270 °C, 50 °C lower than that of Co
spheres). The excellent catalytic performance of Co NC catalyst might be resulted from its lower-temperature
2
-TPR, XPS, Raman spectrum. The Co
3
O
4
NC on Ni foam exhibited the excellent catalytic
2
AlO arrays with coral-like micro-
4
3 4
O
reducibility, lattice distortion of spinel structure and abundant surface-adsorbed oxygen. The findings of this
work provide a new perspective on the design of the high-efficiency integrated catalysts for VOCs combustion.
1. Introduction
Among these transition metal oxide catalysts, tricobalt tetraoxide
(
3 4
Co O ) has attracted much attention due to its low cost-efficiency,
With the growing daily global consumption of fossil fuels, the
environmental benignity, and relatively high catalytic activity in nu-
amount of Volatile organic compounds (VOCs) in urban areas has
reached much higher levels, thus air pollutants emission from trans-
portation and industrial plants are emerging as the challenges for at-
mosphere and humanity [1,2]. Development of highly effective, eco-
nomic, non-secondary pollution conversion technologies can
completely remove hazardous VOCs. Among the various strategies
proposed to degrade VOCs, catalytic oxidation (250–500 °C) is believed
to be the most effective and economically feasible route for the abate-
ment of VOCs due to low-energy consumption and less harmful reaction
by-products [3,4]. The key issue of the catalytic oxidation of VOCs is
rational design of high-performance catalysts with low cost and high
efficiency. Although it is widely recognized that supported noble metals
merous reactions, e.g., the carbon monoxide (CO) [16–18], nitric oxide
(NO) [19,20], hydrocarbon (C
[21–23]. For many heterogeneous reactions, it is well known that ra-
tional design and synthesize of Co nanomaterials with preferred
x y
H ) oxidation as well as soot combustion
3 4
O
porous structure, sizes and morphologies (e.g. nanorods, nanowires,
nanobelts and nanocubes) is efficient ways of increasing catalyst ac-
tivity and selectivity [10,16,21,24,25]. For instance, Xie et al. [16]
3 4
reported that the Co O nanorods with predominantly exposed their
(110) planes calcined at 450 °C could catalyse CO oxidation at tem-
−
1
−1
peratures as low as −77 °C at a SV of 15,000 mL h
[10] prepared a series of mesoporous Co
treatment, and the acid-treated Co catalysts using lower HNO
3
g
cat . Li et al.
3 4
O catalysts by the acid
3
O
4
(
i.e. Pt, Pd, Au and Ag) are effective catalysts to remove VOCs at low
concentrations exhibited the optimum catalytic toluene activity
(T90 = 225 °C) and excellent catalytic durability. Our group previously
temperature [5–8], their high cost and susceptibility to poisoning have
limited their wide and abundant industrial application. Alternatively,
synthesized three-dimensional (3D) hierarchical Co
with different morphology and various exposed crystal planes, it could
be found that 3D hierarchical cube-stacked Co microspheres
3 4
O nanocatalysts
these transition metal oxide catalysts ((MnO
x 3 4 2 3
, Co O , CuO, Fe O ,
CeO ) have been considered as the alternative to expensive noble metal
2
3 4
O
for the VOCs oxidation due to relatively low cost, strong oxygen sto-
rage/release ability [9–15].
achieved a toluene conversion of 90% (T90) of approximately 248 °C
[25]. These researches suggested that the design and synthesis of
⁎
Corresponding author at: School of Environment and Energy, South China University of Technology, Guangzhou, 510006, PR China.
E-mail addresses: cedqye@scut.edu.cn, 1106389373@qq.com (D. Ye).
Shengpeng Mo and Qi Zhang are co-first authors.
1
https://doi.org/10.1016/j.mcat.2018.05.006
Received 25 March 2018; Received in revised form 27 April 2018; Accepted 7 May 2018
2468-8231/ © 2018 Published by Elsevier B.V.

Q. Zhang et al.
Molecular Catalysis 454 (2018) 12–20
catalysts contribute to improving the catalytic activity, whereas the
above researches focus on the study of powder catalysts.
Taking into accounting the practical application of powder cata-
lysts, how to achieve large-scale preparation and practical application
2.1.3. Synthesis of Co3O4 NC on Ni foam
The cleaned Ni foam was put against the 50.0 mL Teflon-lined au-
2
toclave which contained a homogeneous solution of Co(NO ·6H O
(4 mmol), urea (8 mmol), Polyvinylpyrrolidone (PVP-K30, 0.3 g) and
40 mL water. The autoclave maintained at 95 °C for 8 h and then raised
the temperature to 110 °C for 5 h in an electric oven. Finally, the as-
grown precursor was put into a quartz tube and then annealed at 350 °C
3
)
2
3 4
of such Co O nanomaterials with tunable morphologies is of vital
importance and remains a challenging goal. The conventional powder
catalysts are usually introduced onto the surface or mixed into the
channel walls of a structured support (e.g. ceramic or metallic substrate)
by wash-coating, wet-impregnation or extrusion, which are always in-
homogeneous and easy to aggregate, and have a negative effect on the
practical application of the monolithic catalysts [26–28]. Therefore,
some efforts have been devoted to grow nanoarray-based catalysts on
channeled monolithic substrates to promote catalytic activity
3 4
for 3 h. The as-prepared catalyst is denoted as Co O NC.
2.1.4. Synthesis of Co2AlO4 on Ni foam
The cleaned Ni foam was put against the 50.0 mL Teflon-lined au-
toclave which contained a homogeneous solution of Co(NO ·6H
(2 mmol), Al(NO ·9H O (1 mmol), urea (8 mmol) and 40 mL water.
3
)
2
2
O
3
)
3
2
[
26,29,30]. As compared to ceramic monoliths, nickel (Ni) foam has
The autoclave was sealed and maintained at 95 °C for 12 h in an electric
oven. Finally, the as-grown precursor was put into a quartz tube and
then annealed at 350 °C for 3 h. The as-prepared catalyst is denoted as
distinguished features such as low cost, a three-dimensional (3D) re-
ticular configuration, rich accessible electroactive sites, high thermal
conductivity and desired mechanical strength, and high surface area to
volume, which has been attracting sustained attention in the applica-
tions of lithium-ion batteries [31], energy storage [32,33], and catalysis
2 4
Co AlO .
2.1.5. Synthesis of WI-Ni foam and WI-honeycomb substrate
[
19]. Zhao et al. [32] fabricated successfully MnCo
2
O
4
@Ni(OH)
2
mul-
Firstly, a solution of Co(NO ) ·6H O (4 mmol) dissolved in 100 mL
3
2
2
ticomponent core-shell nanoflowers supported on the 3D macroporous
nickel foam substrate for the first time by a facile and cost-effective
method, which exhibits a significantly enhanced specific capacitance
of deionized water was produced. Secondly, the Ni foam and honey-
comb substrate were put against the nitrate salt solution, respectively.
And then NH OH (water solution 25%) was added dropwise to the
4
−
1
−1
(
2154 F g
at 5 A g ). Liu et al. [19] developed a hierarchical
core-shell nanowire arrays in situ grown on the nickel
cat-
alyst. Therefore, these works inspired us to construct well-defined
Co nanoarray-based catalysts on porous nickel (Ni) foam for the
nitrate salt solution, keeping at a constant pH value of 9.0. After 6 h at
MnO
2
@NiCo
2
O
4
room temperature, the samples were collected. Finally, the as-grown
precursors were put into a quartz tube and then annealed at 350 °C for
3 h. The as-prepared catalysts are denoted as WI-Ni foam and WI-hon-
eycomb substrate.
surface serving as a promising candidate for the monolith de-NO
x
3 4
O
application of toluene oxidation.
In this work, we try to consider that a series of Co-based oxide ar-
rays were synthesized via in-situ growth on porous nickel foam with
different morphology, and further investigated their catalytic perfor-
mances for toluene oxidation. The phase compositions, surface ele-
mental states and catalytic performances of catalysts were investigated
by numerous techniques to better understand the impact of the mor-
phological transformation on their physicochemical properties, thus
further confirming the structure-activity relationship. To the best of our
knowledge, rarely previous work has reported the catalytic behaviors of
2.2. Materials characterizations
X-ray powder diffraction (XRD) patterns of the monolithic struc-
tured samples were recorded on a Panalytical X'Pert PRO system with
Cu-Kα (λ = 1.5406 Å) radiation at a scan rate of 10 min
range of scattering angle 2θ of 5–90°, operated at 40kv and 40 mA. The
size and morphology of samples were characterized using field-emis-
sion scanning electron microscopy (FESEM, JEOL JSM-6700F) with an
acceleration voltage of 15 kV, 10 mM. The microstructures of samples
were obtained using transmission electron microscopy (TEM,
JEOLJEM-2010F) with an accelerating voltage of 200 kV. The reduction
behavior of the samples was studied by the temperature–programmed
−1
within the
3 4
well-defined Co O array catalysts on porous nickel (Ni) foam for VOCs
oxidation.
2
reduction of hydrogen (H –TPR) with Automated Catalyst
2. Experimental section
Characterization System (Autochem 2920, MICROMERITICS) equipped
with thermal conductivity detector (TCD). The monolithic structured
2
2
.1. Synthesis of catalysts
Co
3 4
O array sample (1 cm × 3 cm × 1.6 mm) was placed in a quartz
−1
reactor under a gas flow (10% H
2
/Ar, 25 mL min ) with a constant
−1
.1.1. Synthesis of Co3O4 NS on Ni foam
The cleaned Ni foam (approximately 5 cm × 7 cm, thickness 1.6 mm
rate of 10 °C min up to 800 °C. The Raman spectra of the monolithic
samples were conducted on a Renishaw RM2000 Raman Spectrometer
(laser wavelength = 532 nm). Surface species of the catalysts were
characterized by X–ray photoelectron spectroscopy (XPS) using an
XLESCALAB 250Xi electron spectrometer from VG Scientific with
monochromatic Al Ka radiation, and the binding energies of elements
were calibrated based on the C 1s peaks at 284.6 eV.
and pore density 110 ppi) was put against the 50.0 mL Teflon-lined
autoclave which contained a homogeneous solution of Co(NO ·6H
4 mmol), urea (8 mmol) and a 40 mL mixed solvent of ethanol/ethanol
ethanol:Vmethanol = 40:40). The autoclave was sealed and maintained
3
)
2
2
O
(
(
V
at 95 °C for 12 h to synthesize the precursor in an electric oven. The
sample was rinsed several times and dried in air at 80 °C. Finally, the Ni
foam with the as-grown precursor was annealed at 350 °C for 3 h. The
2.3. Catalytic activity measurements
3 4
as-prepared catalyst is denoted as Co O NS.
The as-prepared Co
3 4
O catalysts (1 cm × 4 cm × 1.6 mm, about
0.1450 g) buckling into a cylinder were evaluated in a fixed-bed quartz
2
.1.2. Synthesis of Co3O4 NW on Ni foam
The cleaned Ni foam was put against the 50.0 mL Teflon-lined au-
toclave which contained a homogeneous solution of Co(NO ·6H
4 mmol), urea (8 mmol), polyethylene glycol (PEG-6000) (0.3 g) and
tubular micro-reactor (φ = 6 mm) with quartz wool packed at both
−1
ends of the catalyst bed (gas hourly space velocity = 20,000 h ). The
reactant gas composed of 1000 ppm toluene balanced with air (20 vol.
3
)
2
2
O
(
4
2 2
% O + balance N ) was purged into the reactor at a flow rate of
−
1
0 mL water. The autoclave was sealed and maintained at 95 °C for 12 h
100 mL min . After reacted at the final temperature for 1 h, the con-
centrations of effluent gases were analyzed on-line by a gas chroma-
tograph (Shimadzu GC-2014) equipped with two flame ionization de-
tector (FID).
in an electric oven. Finally, the as-grown precursor was put into a
quartz tube and then annealed at 350 °C for 3 h. The as-prepared cat-
alyst is denoted as Co O NW.
3 4
13

Q. Zhang et al.
Molecular Catalysis 454 (2018) 12–20
Fig. 1. FESEM images of as-synthesized (a–b) Co
3
O
4
NS, (c–d) Co
3
O
4
NW, (e–f) Co
3
O
4
NC, (g–h) Co
2
AlO
4
array samples on Ni foam, and (i) FESEM image of Ni foam
substrate.
3
. Results and discussion
area over the Co-based samples were about 3.1–3.5 mg on a circular
disk of Ni foam (Table 1). The conditions for hydrothermal synthesis
varied between each method, the mass of Co2+ ions was excessive in
the reaction bath, therefore the mass loadings of the Co-based samples
3.1. Morphology
Fig. 1(a–h) shows field-emission scanning electron microscopy
FESEM) images of the as-prepared Co AlO and Co arrays on the Ni
3 4
owned the same amounts of powder Co O .
(
2
4
3 4
O
foam with nanosheet (NS), nanowire (NW) and nanocluster (NC). The
top-view FESEM image of pristine Ni foam substrate in Fig. 1i shows
that Ni foam has an interconnected porous framework, providing ef-
fective accessible channels for the support and the growth center of
powder catalysts with strong adhesion. As shown in Fig. 1a and b, it can
3.2. XRD analysis
Fig. 2 shows the XRD patterns of as-prepared Co
3 4
O NS, NW, NC and
Co
2
AlO samples. There were three sharp diffraction peaks at 44.8°,
4
be clearly seen that the as-synthesized Co
3
O
4
NS displayed frizzy na-
52.1°, and 76.6° in each spectrum, corresponding to the (111), (200)
and (220) planes of pure Ni metal substrate (JCPDS card no. 65-2865),
respectively. The positions of diffraction peaks of all samples were ra-
noflake structure with a slightly curved lateral dimension of approxi-
mately 1.0–2.0 μm, almost no ab-faces and slightly intercrossed with
each other, and it grew uniformly vertical to the Ni foam forming a
highly dense film with strong adhesion, covering the entire surface of
the Ni foam. The FESEM images (Fig. 1c and d) of Co O NW sample
3 4
indicated that nanowire arrays vertically grown on Ni foam have ap-
2 4
ther similar. For Co AlO sample, besides the peaks originating from
the Ni substrate, there were six diffraction peaks at 19.1°, 31.4°, 36.9°,
38.3°, 59.2°, and 65.1°, respectively, which belonged to the (111),
(220), (311), (222), (511), and (440) planes and was well matched with
peared, and the width dimensions were approximately 200–300 nm.
spinel oxide phases (Co
JCPDS card no. 44-0160). The diffraction peaks of another three Co
array samples were perfectly assigned to pure face-centered cubic phase
of Co (JCPDS card no. 42-1467). It was rather difficult to distinguish
between Co AlO and Co spinels via XRD due to their almost
identical diffraction positions. Both Co
tetrahedral or octahedral lattice sites, thus Co
2 4 2 4
AlO , JCPDS card no. 38-0814; CoAl O ,
From the SEM images of Co
3
O
4
NC in Fig. 1(e and f), it was observed
3 4
O
that the thin nanowires with interconnected with each other were
uniformly distributed on the Ni foam ligament surface, forming a series
of columnar nanoclusters. Meanwhile, it is seen from Fig. 1(g and h)
3 4
O
2
4
3 4
O
3+
3+
that the morphology of Co
2
AlO
4
sample presented hierarchical coral-
and Al
cations occupy
3
+
like microspheres and/or pompom-like frizzy nanoflakes with a dia-
meter of 1.5–3.0 μm that resembled a large quantity of densely packed
and interlaced nanoplatelets. On average, the mass loading of per unit
can be replaced by
3+
Al
2 4
to from stable Co AlO spinel phase [23].
Table 1
Catalytic activities and surface elemental compositions of the Co-based oxide catalysts.
Sample
Mass Loadings of per unit area
S
BET/m2 g−1
V
pore/cm3 g−1
T10/°C
T50/°C
T100/°C
Co3+/Co2+
OII/O
I
Co
Co
Co
Co
2
3
3
3
AlO
4
3.482 mg
3.193 mg
3.223 mg
3.526 mg
27.49
32.68
31.38
48.88
0.131
0.140
0.151
0.187
290
283
265
247
313
295
275
264
320
300
280
270
1.08
1.19
1.37
1.69
0.584
0.543
0.653
0.871
O
4
O
4
O
4
NS
NW
NC
14

Q. Zhang et al.
Molecular Catalysis 454 (2018) 12–20
between (400) and (220) crystal face, which demonstrates that the
Co
Co
3
O
O
4
NS dominantly exposed {100} surfaces. Form the TEM image of
NW (Fig. 4c), the morphology showed a bunch of nanowires, and
3
4
the width dimensions were approximately 200 nm in agreement with in
Fig. 1(c and d), and the surface lattice spacing in Fig. 4d was measured
to be 0.467 nm corresponding to the (110) crystal plane of Co
dicating the nanowire was along [110] direction. The morphology of
Co NC was composed of nanowires to form nanoclusters with flat-
3 4
O , in-
3 4
O
tened surfaces, and the individual nanowires were also composed of a
number of tiny particles (Fig. 4e). The width and length dimensions
were approximately 200 nm and 3.0 μm, respectively. From high-re-
3 4
solution TEM (HRTEM) image of Co O NC in Fig. 4f, there were two
lattice fringes about 0.243 nm and 0.284 nm, which were derived from
the (311) and (220) crystal faces of pure Co , respectively. The thin
3 4
O
nanowires of columnar nanoclusters can preferentially expose {110}
surfaces. What is more, it can be seen that several ledges on the surface
Fig. 2. XRD patterns of as-synthesized Co-based array samples on Ni foam.
3 4
of Co O NC sample were observed (Fig. 4f), demonstrating the pre-
sence of abundant surface defect sites. Additionally, the active sites
resulted from exposed surfaces or growth orientation of crystal struc-
tures [25]. According to previous studies [34–36], the (100) and (111)
2+
planes contain only Co
plane presents 5 Co2+ and 4 Co in one unit cell. Therefore, the Co
nanowires exposed {110} facets would show higher catalytic activity
cation, whereas the highly reactive (110)
3+
3 4
O
for toluene oxidation, especially Co
fect sites.
3 4
O NC with abundant surface de-
3.5. Temperature programmed reduction (TPR)
Low-temperature reducibility of catalysts is an important factor
influencing its surface oxygen reactivity, surface reaction sites and re-
lated catalytic performances in Mars-van Krevelen (redox) mechanism.
2
Herein, hydrogen temperature programmed reduction (H -TPR) ex-
periments were carried out to determine different forms and reduction
behaviors in monolithic samples, as shown in Fig. 5. It can be seen that
the Co
the maximum rates of H
spectively, which was ascribed to a stepwise process via the sequence of
3
O
4
NS presented two overlapping H
2
consumption peaks with
Fig. 3. N
samples on Ni foam.
2
adsorption–desorption isotherms of as-synthesized Co-based array
2
consumption located at 271 and 308 °C, re-
Co → Co → Co0 [23].The low-temperature reduction behaviors of
Co NW and NC catalysts were similar to that of Co NS in terms of
their stepwise process despite the different H consumptions. The
Co NW sample was reduced in two main temperature regions, and
the maximum rate of H consumption for the reduction of the Co
phase occurred at 284 and 305 °C, respectively. For Co NC sample,
two overlapping H consumption peaks at 286 and 315 °C were ob-
served, and there were the reduction peak area of Co
the case of Co AlO sample, there were three asymmetrical and/or
3+
2+
3.3. Surface area analysis
3
O
4
3 4
O
2
Fig. 3 displays the N
2
adsorption–desorption isotherms for as-syn-
3 4
O
thesized Co-based samples. Before the tests, the samples were degassed
under vacuum at 150 °C for 6 h. The isotherms are consistent with type
IV isotherms with a H2-type hysteresis loop in the relative pressure
2
3 4
O
3 4
O
2
3+
ranges (p/p
o
), which are associated with capillary condensation taking
reduction. In
place in mesopores. The BET data of Co-based samples are listed in
Table 1. The surface area and pore volume of the uncoated Ni foam are
2
4
separate reduction peaks in a temperature range 100–700 °C. The first
two peaks in the low temperature range of 250–350 °C were attributed
2
−1
3 −1
2
4.85 m g and 0.064 cm g , respectively. Compared to the pure Ni
to the reduction of Co3+ into Co , and the further reduction of Co
2+
2+
foam, it can be found that there are slightly increased in the surface
2
−1
2 −1
areas of Co-based catalysts (27.49 m g
pore volumes increased slightly in the ranges of 0.131–0.187 cm g
It is noticeable that the Co NC possessed largest surface area and
pore volume among the Co-based catalysts.
to 48.88 m g ), and the
to metallic cobalt, respectively. The third broad peak in the low tem-
perature range of 400–700 °C was associated with the further reduction
of large cobalt-containing spinel particles, in which Co2+ and Co ions
in spinel phases can interact with neighboring groups through the
3
−1
.
3+
3 4
O
CoeO bonds polarized by Al3+ (with the formation of CoAl
O
or
-TPR results, the first re-
NC revealed a largest peak area than those of
other array samples, which was generally recognized as a sign of the
2
4
Co
2 4 2
AlO phases) [37]. According to the H
3.4. Microstructure
duction peak of Co O
3 4
3+
2+
To further reveal microstructures of Co
3
O
4
NS, NW and NC samples,
improved reducibility of Co
tage of surface Co3+ species. And the exposed {110} facets of Co
O
3 4
to Co
caused by the exposed percen-
transmission electron microscopy (TEM) measurement was carried out,
with the results shown in Fig. 4. Both Co array samples scraped off
from the Ni foam can maintain the primary shapes. As shown in Fig. 4a,
the Co NS was composed of a randomly aggregated structure of
ultrathin frizzy nanosheets with flattened surfaces. The HRTEM image
Fig. 4b) of Co NS revealed that there were two lattice fringes about
.203 nm and 0.286 nm, which were derived from the (400) and (220)
crystal faces of pure Co , respectively. There is an angle of 45°
NC
3
O
4
were benefited for better low-temperature reducibility, in agreement
with the TEM results. Nevertheless, the hindered reduction of cobalt
species on the Co AlO sample was ascribed to the formation of cobalt
2 4
3
O
4
aluminium oxide phases, suggesting that a negative effect on the low-
(
0
3
O
4
temperature reducibility was induced by Al3+ species, which was in
good agreement with the XRD result of Co
implied the different distribution proportions of Co
2 4
AlO sample. These results
3+
2+
3
O
4
and Co
in
15

Q. Zhang et al.
Molecular Catalysis 454 (2018) 12–20
Fig. 4. TEM images of (a) Co
3
O
4
NS, (c) Co
3
O
4
NW and (e) Co
3
O
4
NC samples; (b, d and f) the corresponding to HRTEM images in (a, c and e), respectively.
2
Fig. 5. H -TPR profiles of as-prepared Co
3
O
4
NS, Co
3
O
4
NW, Co
O
3 4
NC and
3 4 3
Fig. 6. Raman spectroscopy profiles of as-prepared Co O NS, Co O4 NW,
Co AlO samples.
2
4
3 4 2 4
Co O NC and Co AlO samples.
monolithic samples, and the morphological transformation tended to
induce more surface Co3+ species playing an significant role on the
oxidation of toluene, since the oxidation of VOCs proceed via a Mars-
van Krevelen (redox) mechanism.
the Raman spectra of as-prepared monolithic Co-based samples, and
five Raman-active modes (A1g + E + 3F2g) were observed. In the case
g
of the Co AlO sample, five Raman bands locate at 196, 523, 625, 484
2
4
1
−
and 695 cm , were assigned to the 3F2g, Eg and A1g Raman active
modes of Co nanocrystals, respectively [38]. The weak peaks lo-
cated at approximately 196, 523, 625 cm
characteristics of tetrahedral sites (CoO ), whereas the peak located at
can be attributed to the characteristics of the symmetric
Co eO (octahedral CoO sites) stretching vibration. The Co NC
3 4
O
−
1
can be attributed to the
3.6. Structural defects
4
−
1
6
95 cm
Raman spectroscopy is an intriguing tool to analyze the structural
properties (lattice distortion) of Co-based oxide arrays. Fig. 6 displays
3
+
6
3 4
O
16

Q. Zhang et al.
Molecular Catalysis 454 (2018) 12–20
ratio of the Co
3
O
4
NC sample was obviously highest than those
(
1.08–1.37) of the other Co-based samples, indicating that relatively
3+
more Co
2 4
species presented. Whereas the Co AlO sample possessed
lower Co3+/Co (1.08) atomic ratio due to the formation of CoAl
2+
2 4
O
molar ratio
phases on the bulk. The increased Co3+/Co
2+
or Co
2
AlO
4
on the surface acted as the active site for toluene oxidation can actively
take part in the catalytic process, and result in a charge imbalance
which can be confirmed by the fine-scanned O 1s XPS spectra.
As for the O 1s XPS spectra of Co-based samples, the signals for each
catalyst can be decomposed into two components, lattice oxygen spe-
2
–
cies (O
I
) (∼529.3 eV for O ), highly oxidative oxygen species (OII
)
−
−
(
∼531.6 eV for O
2
, O or OH) [11,23]. Obviously, the Co
3
O
4
NC
sample exhibited the relatively higher area of OII peak, further in-
dicating the presence of oxygen vacancies on the surface of Co NC.
The surface OII/O atomic ratio increased in the following order: Co
NC > Co NW > Co AlO > Co NS. The increased concentra-
3 4
O
I
3 4
O
3
O
4
2
4
3 4
O
tion of surface or chemisorbed oxygen (OII) would benefit the catalytic
toluene performance of Co-based oxide catalysts. The result reveals that
3
+
2+
the exposed {110} facets also were benefited for surface Co /Co
and OII/O
results.
I 3 4
atomic ratio of Co O NC, confirmed by the Co 2p and O 1s
3.8. Catalytic activity measurement
Fig. 8a shows the toluene conversion curves of the as-prepared in-
tegrated Co-based oxide catalysts (1 cm × 4 cm × 1.6 mm) with in-
creasing reaction temperature under a gas hourly space velocity
−1
(
GHSV) of 20,000 h , all the as-prepared catalysts displayed positive
toluene conversion with the only CO and water products, and other
2
intermediate oxidation products were not monitored on line by a gas
chromatograph. To conveniently compare the catalytic activities of the
integrated Co-based oxide catalysts, the catalytic activities of these
catalysts were compared based on the values of T10, T50 and T100 (the
temperatures corresponding to the 10, 50 and 100% conversion of
catalytic toluene combustion) as a reference, which were summarized
Fig. 7. (a) Co 2p and (b) O1s XPS profiles of as-prepared Co
3
O
4
NS, Co
3
O
4
NW,
3 4
in Table 1. The Ni foam and honeycomb substrate loading Co O were
Co
3
O
4
NC and Co
2
AlO
4
samples.
synthesized by wet-impregnation method, the as-prepared catalysts are
denoted as WI-Ni foam and WI-honeycomb substrate, respectively. The
WI-Ni foam and WI-honeycomb substrate catalysts showed the com-
pletely catalytic activities of toluene oxidation at 320 and > 350 °C,
respectively. The Co AlO catalyst prepared by the hydrothermal
2 4
synthesis also showed the relatively lower catalytic activity, its T10, T50
displayed the five Raman peaks located at 188, 516, 619, 478 and
−1
683 cm , respectively. In comparison to the Co
3
O
4
NS and Co
2 4
AlO
sample, the peak positions of five active modes of Co
lower wave numbers by 5–15 cm , suggesting that polyhedral dis-
tortion occurring in the spinel lattice has influenced on the original
coordinative environment of octahedral sites, in which Co and Co
cations are situated at tetrahedral and octahedral sites in the cubic
lattice of Co [38]. This red-shift phenomenon indicated the idea of
the more crystal defects beneficial for giving rise to surface oxygen
vacancies [23,39], in addition, and the presence of crystal defects have
been confirmed by the HRTEM result in Fig. 4f.
3 4
O
NC shifted to
−
1
and T100 were achieved at 290, 313 and 320 °C, respectively. Never-
2
+
3+
theless, the Co O4 NC catalyst displayed the best catalytic activity for
3
toluene oxidation, with the complete toluene conversion at 270 °C. And
the T50 and T100 values for toluene oxidation over the Co
9 and 50 °C lower than those achieved over the Co AlO
spectively. Specifically, it can be deduced that the following activity
3
O
4
NC were
3 4
O
4
2
4
catalyst, re-
trend of four catalysts was observed in the order: Co
NW > Co NS > Co AlO . The Co NW, Co
3
O
4
NC > Co
3
O
4
3
O
4
2
4
3
O
4
3
O
4
NS and Co AlO
2
4
catalysts had very lower activity (< 20%) when the temperature was
even as high as 270 °C, whereas at the same temperature, the toluene
3.7. Surface element composition
conversion of Co
Fig. 8b shows the reaction rate of all Co-based oxide catalysts at 270 °C.
The reaction rate of the Co NC at 270 °C was about
, which was clearly greater than those of
other samples. Fig. 8c shows the catalytic performances of the Co
NC catalysts with different calcination temperature (350, 450 and
550 °C). It is obvious that the Co NC(350) catalyst obtained at
350 °C exhibits the highest catalytic activity among all Co NC cat-
alysts, achieving complete toluene conversion at about 270 °C, and
whose T100 is 20 °C lower than that achieved over the Co NC(550).
The effects of GHSV on the catalytic activities of the Co NC sample
3 4
O NC can completely catalyse toluene degradation.
X-ray photoelectron spectra (XPS) measurements were performed to
probe the surface chemical states and adsorbed oxygen species of as-
prepared Co-based catalysts, as shown in Fig. 7. To get further insight
into the surface properties, the Co 2p3/2 spectra could be further de-
convoluted into two components peaks at binding energy
3 4
O
−4
−1 −1
53.38 × 10 mol gcat
h
3 4
O
(
BE) = 779.9 ± 0.2 and 781.1 ± 0.2 eV, and two weak shake-up sa-
3 4
O
tellite peaks [23,40]. The peaks located at 779.9 and 781.1 eV could be
3 4
O
3
+
2+
assigned to Co
and Co oxidation states, respectively. The other
peaks at higher binding energy of 784.8 ± 0.2 and 788.2 ± 0.2 eV
3 4
O
O
3 4
3
+
2+
were associated with the shake-up satellite of Co
and Co , respec-
3
+
2+
tively. The surface Co /Co molar ratios were determined based on
for toluene oxidation were further evaluated following the slight de-
crease with the rise in GHSV value, as shown in Fig. 8d. Under the
the area fitted curve peaks on Co 2p3/2 XPS spectra and are listed in
3
+
2+
−1
Table 1. It could be found that the surface Co /Co
(1.69) atomic
condition of GHSV = 10,000 h , the corresponding temperature for
17

Q. Zhang et al.
Molecular Catalysis 454 (2018) 12–20
Fig. 9. Toluene conversion as a function of on-stream time over the Co
3 4
O NC
catalyst under the condition of toluene concentration = 1000 ppm and
−
1
GHSV = 20,000 h : (a) the stability tests with dry conditions at 272 and
65 °C, and (b) the stability tests with or without moisture at 272 °C.
2
100% conversion was about 250 °C. These results suggested that the
morphology of arrays in-situ grown on Ni foam can improve the cata-
lytic activity for toluene oxidation.
Since the best-performing Co
lytic reactivity among all catalysts, the Co
3
O
4
NC catalyst owned the best cata-
NC catalyst was selected
3 4
O
as a model catalyst for long-term durability tests under dry and moist
conditions, as shown in Fig. 9. The toluene conversions of 100% and
61% maintained constant without any deactivation under dry condi-
tions at 272 and 265 °C during the activity test for 36 h. For practical
applications of VOCs oxidation catalysts, the inevitable presence of
moisture in an ambient environment has been known to exert a great
influence on the catalytic activity of a catalyst for VOCs oxidation.
2
When 1.5 vol.% H O was introduced into the stream at 272 °C, the to-
luene conversion decreased from 100% to 90% during the activity test
for 12 h, whereas the toluene conversion decreased sharply from 100%
to 70% with the addition of higher concentration (5 vol.%) of moisture.
Once the moisture was stopped, the activity was rapidly increased back
to its original level (100% toluene conversion). This indicates that
moisture indeed caused a sudden deactivation of toluene catalytic
combustion in our current study, in which the surface of the catalyst
was inhibited by moisture and there were the competitive adsorption of
water and target molecules on the surface active sites [10,41]. Most
Fig. 8. (a) Toluene conversion versus reaction temperature, (b) the reaction rate
3 4
importantly, the Co O NC catalyst exhibited the good regeneration
at 270 °C over the Co-based oxide samples under the condition of
ability and stability. Additionally, Table 2 summarizes the catalytic
activities for VOCs oxidation over monolithic catalysts previously re-
ported by other researchers [4,42–51]. Despite the different reaction
−
1
GHSV = 20,000 h , (c) toluene conversion curves over the Co
calcined at different temperatures (350, 450 and 550 °C), and (d) the effects of
different GHSV on the catalytic activities of the Co NC catalyst.
3 4
O NC catalysts
3 4
O
conditions, the as-prepared Co
3
O
4
NC catalyst with aligned
18

Q. Zhang et al.
Molecular Catalysis 454 (2018) 12–20
Table 2
Catalytic testing conditions (catalyst usage, substrate, SV, pollutant type and concentration) and catalytic activities (T50, T100) of as-prepared Co
other reported monolithic catalysts.
O
3 4
NC catalyst and
Ref. No.
Catalyst
Substrate
SV
pollutant type
concentration
T
50 (°C)
T100 (°C)
−
1
1
Fe-Mn mixed oxides
Mn-Ce-Cu mixed oxide
Cordierite
Cordierite
Cordierite
Cordierite
FeCrAl
10,000 h
10,000 h
Toluene
o-xylene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Benzene
CO
1000 ppm
1000 ppm
1000 ppm
175
260
290
195
305
205
260
210
250
248
375
264
400
> 350
315
250
374
240
270
> 400
330
> 260
400
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[4]
−
−1
−1
−1
−1
−1
−1
Cu0.5Co0.5/Al
Pt-Pd/Al
Cu0.5Co0.5/Al
2
O
3
56,000 mL h
g
g
g
cat
cat
cat
−1
3
2
O
3
20,000 h
56,000 mL h
6 g/m
2
O
3
1000 ppm
−1
3
0
.4 wt.% Pd/FeCrAl
FeCrAl
10,000 h
4 g/m
Co-Mn oxides
Ni-Mn oxides
CuCoAl-MMO
activated carbon
Cordierite
Al film
66,000 mL h
1.0 vol.%
1000 ppm
500 ppm
1.0 vol.%
1.0 vol.%
1000 ppm
−
1
1
10,000 h
–
–
–
Co
TiO
Co
3
O
4
Co foil
[50]
[51]
This study
2
/(La,Sr)MnO
3
FeCr20Al
Ni foam
5
foil
CO
Toluene
−
O
3 4
nanoclusters
20,000 h
270
nanoclusters in this work presented an outstanding activity for toluene
oxidation. Thus, the macropore walls of Ni foam substrates are bene-
surface defect sites on the surface of Co
HRTEM. And the peak positions of five active modes of Co
shifted to lower wave numbers by 5–15 cm , its red-shift phenomenon
implied the idea of polyhedral distortion in the spinel lattice giving rise
to surface oxygen vacancies. The presence of surface oxygen vacancies
has been confirmed by the O 1s XPS result. Hence, we could propose
that the morphological transformation of Co
the difference of surface adsorbed oxygen species, Co
redox behavior of the catalyst, bringing a significant improvement of
catalytic toluene combustion.
3
O
4
NC sample are observed by
NC
3 4
O
−
1
ficial to the formation of a 3D macroporous network of Co
3
O
4
, which
not only exhibits good mass transfer properties toward reactant mole-
cules, but increases their contact with surface sites.
It has been accepted that the catalytic activity of a transition metal
oxide for VOCs oxidation can be greatly influenced by several factors,
such as surface area, morphology, surface active species, structure de-
fects and reducibility of catalyst. Moreover, it is widely accepted that
the Mars-Van Krevelen (MVK) mechanism has been widely used to il-
lustrate the VOCs combustion mechanism on the surface of catalysts
3 4
O arrays tended to induce
3+
species and
[
49]. According to the catalytic and characterization analyses, the
Co NC composed of a series of columnar nanoclusters showed the
highest catalytic activity for toluene oxidation, and highest reaction
4. Conclusions
3 4
O
In conclusion, advanced monolithic Co-based oxide catalysts have
been successfully synthesized through a generic hydrothermal synthesis
route to in situ grow highly ordered Co O arrays onto Ni substrates.
3 4
The morphology of monolithic Co-based oxide catalysts contained in-
tercrossed frizzy nanosheets (NS), nanowires (NW), nanoclusters (NC)
and coral-like microspheres. The catalytic oxidation of toluene as the
model reaction was further used to evaluate their catalytic perfor-
mances of monolithic Co-based oxide catalysts. Based on the catalytic
−4
−1 −1
rate (53.38 × 10 mol gcat
Co NS and Co AlO catalysts had very lower activity when the
temperature was even as high as 270 °C, whereas at the same tem-
perature, the toluene conversion of Co NC can completely catalyse
toluene degradation. This catalytic test results suggested the mor-
phology resulted from exposed surfaces or growth orientation of crystal
structures may play a key role in toluene oxidation. Furthermore, the
h
3 4
) at 270 °C. However, the Co O NW,
3
O
4
2
4
3 4
O
Co
And the exposed facets determine the surface Co
species and low-temperature reduction behaviors. The XPS confirmed
3
O
4
NC possessed largest surface area among the Co-based catalysts.
3 4
and characterization analyses, it can be found that the Co O NC
3
+
species, oxygen
composed of a series of columnar nanoclusters exhibited excellent
catalytic performance for toluene oxidation, and highest reaction rate
3
+
−4
−1 −1
that Co
3
O
4
NC catalyst possessed the more surface Co
species, in-
= 0.871). It
(53.38 × 10 mol gcat
was associated with the surface adsorbed oxygen species (OII
= 0.871), more surface Co3+ species, high low-temperature re-
ducibility and the structural defects. Most importantly, the Co NC
h
) at 270 °C. The catalytic performance
ducing the formation of abundant oxygen vacancies (OII/O
I
/
is known that oxygen vacancies can actively take part in the oxidation
process (the adsorption, activation and migration of oxygen species)
and greatly contribute to the catalyst activity. This redox mechanism
supposes that the catalytic VOCs oxidation consists of two successive
steps [41,52]: (1) the adsorption of reactants with the surface adsorbed
O
I
3 4
O
catalyst exhibited the good regeneration ability and stability. Thus, the
Ni foam substrates are beneficial to the formation of a 3D macroporous
network of Co O , which is significant for the development and prac-
3 4
oxygen species to transfer it to CO
2
and water; (2) the oxygen molecules
tical use of integrated catalysts.
(O
2
) are activated to become the surface adsorbed oxygen species and
3
+
2+
the redox couple Co /Co simultaneously take place. Therefore, the
abundant surface oxygen vacancies may be benefited for the increased
concentration of surface adsorbed oxygen and increased reaction rate
for VOCs oxidation.
Acknowledgments
This research described above was financially supported by the
Science and Technology Planning Project of Guangdong Province China
(No. 2015B0202236002), the National Natural Science Foundation of
China (No. 51108187, 51672273, B5151050), and the Guangdong
Natural Science Foundation (No. 2016A030311003).
According to MVK redox mechanism, excellent low-temperature
reducibility would offer higher performance of the catalyst. According
2 3 4
to the H -TPR results, Co O
NC reveals a largest peak area (Co3+ to
2
+
Co ) than those of other array samples, which caused by the exposed
percentage of surface Co3 species. Nevertheless, the hindered reduc-
+
References
tion of cobalt species on the Co
mation of cobalt aluminium oxide phases. Thus, both Co
Co AlO samples displayed the significant difference of catalytic per-
formance for toluene oxidation, leading to a low-temperature reduction
behavior confirmed by H -TPR investigation. Furthermore, the struc-
2
AlO
4
sample was ascribed to the for-
3
O
4
NC and
[1] Anders Hellman, Simon Klacar, H. Grönbeck, J. Am. Chem. Soc. 46 (2009)
1
6636–16637.
2
4
[
[
2] M.H. Castaño, R. Molina, S. Moreno, Appl. Catal. A Gen. 492 (2015) 48–59.
3] P. Li, C. He, J. Cheng, C.Y. Ma, B.J. Dou, Z.P. Hao, Appl. Catal. B Evirnon. 101
(2011) 570–579.
2
[
4] S. Li, H. Wang, W. Li, X. Wu, W. Tang, Y. Chen, Appl. Catal. B Evirnon. 166–167
tural defects of a catalyst are also beneficial for greater amounts of
surface oxygen vacancies. It could be seen that a presence of abundant
(2015) 260–269.
[
5] B. Liu, Y. Liu, C. Li, W. Hu, P. Jing, Q. Wang, J. Zhang, Appl. Catal. B Evirnon. 127
19

Q. Zhang et al.
Molecular Catalysis 454 (2018) 12–20
(2012) 47–58.
124–133.
[
6] C. Zhang, F. Liu, Y. Zhai, H. Ariga, N. Yi, Y. Liu, K. Asakura, M. Flytzani-
Stephanopoulos, H. He, Angew. Chem. Int. Ed. Engl. 51 (2012) 9628–9632.
7] S. Zhao, K. Li, S. Jiang, J. Li, Appl. Catal. B Evirnon. 181 (2016) 236–248.
8] C. Zhang, Y. Li, Y. Wang, H. He, Environ. Sci. Technol. 48 (2014) 5816–5822.
9] S.A. Hosseini, A. Niaei, D. Salari, S.R. Nabavi, Ceram. Int. 38 (2012) 1655–1661.
[29] Z. Ren, V. Botu, S. Wang, Y. Meng, W. Song, Y. Guo, R. Ramprasad, S.L. Suib,
P.X. Gao, Angew. Chem. Int. Ed. 53 (2014) 7223–7227.
[30] M. Zhou, L. Cai, M. Bajdich, M. García-Melchor, H. Li, J. He, J. Wilcox, W. Wu,
A. Vojvodic, X. Zheng, ACS Catal. 5 (2015) 4485–4491.
[31] L. Peng, Y. Feng, Y. Bai, H.-J. Qiu, Y. Wang, J. Mater. Chem. A 3 (2015) 8825–8831.
[32] Y. Zhao, L. Hu, S. Zhao, L. Wu, Adv. Funct. Mater. 26 (2016) 4085–4093.
[33] Y. Lv, A. Liu, H. Che, J. Mu, Z. Guo, X. Zhang, Y. Bai, Z. Zhang, G. Wang, Z. Pei,
Chem. Eng. J. 336 (2018) 64–73.
[34] K. Wang, Y. Cao, J. Hu, Y. Li, J. Xie, D. Jia, ACS Appl. Mater. Inter. 9 (2017)
16128–16137.
[
[
[
[
10] G. Li, C. Zhang, Z. Wang, H. Huang, H. Peng, X. Li, Appl. Catal. A Gen. 550 (2018)
6
7–76.
[
[
[
11] S. Mo, S. Li, J. Li, S. peng, J. Chen, Y. Chen, Catal. Commun. 87 (2016) 102–105.
12] S.C. Kim, W.G. Shim, Appl. Catal. B Evirnon. 98 (2010) 180–185.
13] C.Y. Ma, Z. Mu, J.J. Li, Y.G. Jin, J. Cheng, G.Q. Lu, Z.P. Hao, S.Z. Qiao, J. Am.
Chem. Soc. 132 (2010) 2608–2613.
[35] S. Mo, S. Li, H. Xiao, H. He, Y. Xue, M. Zhang, Q. Ren, B. Chen, Y. Chen, D. Ye, Catal.
Sci. Technol. 8 (2018) 1663–1676.
[
14] W. Tang, X. Wu, D. Li, Z. Wang, G. Liu, H. Liu, Y. Chen, J. Mater. Chem. A 2 (2014)
2
544.
15] F. Hu, J. Chen, S. Zhao, K. Li, W. Si, H. Song, J. Li, Appl. Catal. A Gen. 540 (2017)
7–67.
[36] S. Mo, S. Li, Q. Ren, M. Zhang, Y. Sun, B. Wang, Z. Feng, Q. Zhang, Y. Chen, D. Ye,
[
Nanoscale (2018) 7746–7758.
5
[37] R. Xie, G. Fan, L. Yang, F. Li, Catal. Sci. Technol. 5 (2015) 540–548.
[38] G. Cheng, T. Kou, J. Zhang, C. Si, H. Gao, Z. Zhang, Nano Energy 38 (2017)
155–166.
[39] S. Zhao, F. Hu, J. Li, ACS Catal. 6 (2016) 3433–3441.
[40] L. Xu, Q. Jiang, Z. Xiao, X. Li, J. Huo, S. Wang, L. Dai, Angew. Chem. Int. Ed. 55
(2016) 5277–5281.
[
[
16] X. Xie, Y. Li, Z.Q. Liu, M. Haruta, W. Shen, Nature 458 (2009) 746–749.
17] Y. Guo, Z. Ren, W. Xiao, C. Liu, H. Sharma, H. Gao, A. Mhadeshwar, P.-X. Gao, Nano
Energy 2 (2013) 873–881.
18] Y. Lou, J. Ma, X. Cao, L. Wang, Q. Dai, Z. Zhao, Y. Cai, W. Zhan, Y. Guo, P. Hu,
G. Lu, Y. Guo, ACS Catal. 4 (2014) 4143–4152.
[
[
19] Y. Liu, J. Xu, H. Li, S. Cai, H. Hu, C. Fang, L. Shi, D. Zhang, J. Mater. Chem. A 3
[41] S. Mo, S. Li, W. Li, J. Li, J. Chen, Y. Chen, J. Mater. Chem. A 4 (2016) 8113–8122.
[42] W.J. Ma, Q. Huang, Y. Xu, Y.W. Chen, S.M. Zhu, S.B. Shen, Ceram. Int. 39 (2013)
277–281.
[43] X. Zhang, D. Wu, Ceram. Int. 42 (2016) 16563–16570.
[44] F. Zhao, G. Zhang, P. Zeng, X. Yang, S. Ji, Chin. J. Catal. 32 (2011) 821–826.
[45] L. Jiang, N. Yang, J. Zhu, C. Song, Catal. Today 216 (2013) 71–75.
[46] Z. Fu-Zhen, Z. Peng-Hui, J. Sheng-Fu, Y. Xiao, L. Cheng-Yue, Acta Phys. Chim. Sin.
26 (2010) 3285–3290.
[47] Y. Li, Y. Li, Q. Yu, L. Yu, Catal. Commun. 29 (2012) 127–131.
[48] G. Zhou, X. He, S. Liu, H. Xie, M. Fu, J. Ind. Eng. Chem. 21 (2015) 932–941.
[49] Q. Huang, Z.-Y. Zhang, W.-J. Ma, Y.-W. Chen, S.-M. Zhu, S.-B. Shen, J. Ind. Eng.
Chem. 18 (2012) 757–762.
[50] Y. Sun, P. Lv, J.Y. Yang, L. He, J.C. Nie, X. Liu, Y. Li, Chem. Commun. 47 (2011)
11279–11281.
(2015) 11543–11553.
[
[
20] S. Cai, J. Liu, K. Zha, H. Li, L. Shi, D. Zhang, Nanoscale 9 (2017) 5648–5657.
21] Y. Sun, J. Liu, J. Song, S. Huang, N. Yang, J. Zhang, Y. Sun, Y. Zhu, ChemCatChem 8
(2016) 540–545.
[
[
[
[
[
22] H. Wu, G. Pantaleo, G. Di Carlo, S. Guo, G. Marcì, P. Concepción, A.M. Venezia,
L.F. Liotta, Catal. Sci. Technol. 5 (2015) 1888–1901.
23] S. Mo, S. Li, J. Li, Y. Deng, S. Peng, J. Chen, Y. Chen, Nanoscale 8 (2016)
1
5763–15773.
24] V. Iablokov, R. Barbosa, G. Pollefeyt, I. Van Driessche, S. Chenakin, N. Kruse, ACS
Catal. 5 (2015) 5714–5718.
25] Q. Ren, S. Mo, R. Peng, Z. Feng, M. Zhang, L. Chen, M. Fu, J. Wu, D. Ye, J. Mater.
Chem. A 6 (2018) 498–509.
26] S.Y. Chen, W. Song, H.J. Lin, S. Wang, S. Biswas, M. Mollahosseini, C.H. Kuo,
P.X. Gao, S.L. Suib, ACS Appl. Mater. Inter. 8 (2016) 7834–7842.
27] P.-Y. Peng, Aerosol Air Qual. Res. 15 (2015).
[51] Y. Guo, Z. Zhang, Z. Ren, H. Gao, P.-X. Gao, Catal. Today 184 (2012) 178–183.
[52] M.S. Kamal, S.A. Razzak, M.M. Hossain, Atmos. Environ. 140 (2016) 117–134.
[
[
28] L.E. Gómez, I.S. Tiscornia, A.V. Boix, E.E. Miró, Appl. Catal. A Gen. 401 (2011)
20