Tailored synthesis of CoOX thin films for catalytic application

Article abstract of DOI:10.1039/c5ra20013j

Cobalt oxide thin films were systematically synthesized on an inert carrier by pulsed-spray evaporation chemical vapor deposition (PSE-CVD). The effect of substrate temperature on the structure, morphology and surface composition of the prepared films was investigated by XRD, FTIR, SEM and XPS spectroscopies. An in situ diffuse reflectance infrared Fourier transform spectroscopy (in situ DRIFTS) and a gas chromatograph (GC) were involved to identify the surface and gaseous species occurring in the total oxidation of propene as a representative of VOCs, respectively. The structural analysis indicated that the obtained thin films transformed from CoO to pure Co₃O₄ spinel as the temperature rose from 350 to 450°C. A homogeneous grain distribution was observed. For all samples, oxygen was mainly composed of lattice oxygen and adsorbed oxygen constituted a minor proportion. The catalytic tests showed that all the thin films exhibited competitive performances to those of noble metals. According to the observed adsorption peaks of propene at low-temperature and transformation of CoO-Co₃O₄ from the in situ DRIFTS spectra, a combined redox and L-H mechanism was proposed for the catalytic oxidation of propene over cobalt oxide films. The porous structure and adsorbed oxygen on the film surface may well contribute to the catalytic oxidation of propene.

Full text of DOI:10.1039/c5ra20013j

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Tailored synthesis of CoO thin films for catalytic
X
application†
Cite this: RSC Adv., 2015, 5, 97272
Shi-Bin Fan, Guan-Fu Pan, Jing Liang and Zhen-Yu Tian*a
ab
ab
a
Cobalt oxide thin films were systematically synthesized on an inert carrier by pulsed-spray evaporation
chemical vapor deposition (PSE-CVD). The effect of substrate temperature on the structure, morphology
and surface composition of the prepared films was investigated by XRD, FTIR, SEM and XPS spectroscopies.
An in situ diffuse reflectance infrared Fourier transform spectroscopy (in situ DRIFTS) and a gas
chromatograph (GC) were involved to identify the surface and gaseous species occurring in the total
oxidation of propene as a representative of VOCs, respectively. The structural analysis indicated that the
obtained thin films transformed from CoO to pure Co
O
3 4
spinel as the temperature rose from 350 to
ꢀ
4
50 C. A homogeneous grain distribution was observed. For all samples, oxygen was mainly composed of
lattice oxygen and adsorbed oxygen constituted a minor proportion. The catalytic tests showed that all the
thin films exhibited competitive performances to those of noble metals. According to the observed
Received 28th September 2015
Accepted 3rd November 2015
3 4
adsorption peaks of propene at low-temperature and transformation of CoO–Co O from the in situ
DRIFTS spectra, a combined redox and L–H mechanism was proposed for the catalytic oxidation of
propene over cobalt oxide films. The porous structure and adsorbed oxygen on the film surface may well
contribute to the catalytic oxidation of propene.
DOI: 10.1039/c5ra20013j
www.rsc.org/advances
spinel, other cobalt oxides which contain low-valence cobalt are
also expected to have promising catalytic activity. Although there
Introduction
Volatile organic compounds (VOCs) are one of the main air are some works focusing on the preparation of individual
1
8,24–27
28,29
pollutants originating mainly from transportation and the Co
O
3 4
or CoO
available, the systematic and controllable
1–3
combustion of different fuels. Due to their toxicity to human synthesis of CoO
health and involvement in the formation of photochemical still remains a challenging task. Moreover, in many of the above-
smog, many countries have imposed strict standards to reduce mentioned studies,
by a direct method for catalytic applications
X
4
5
18,27,28
the substrates or supports generally play
the industrial VOCs emissions. Several methods have been used signicant roles, which prevents the real evaluation of the cata-
6
in the past few decades, such as thermal incineration, biodeg- lytic performance of CoO
X
. Thus, it is desired to prepare CoO
X
on
7
8
9
radation, high-energy electron beam treatment, adsorption
inert substrates.
CoO can be synthesized through several methods, including
much attention as one of the most promising VOCs abatement pulsed laser deposition (Co
10
and absorption. Recently, catalytic oxidation has attracted
X
18
24
3 4
O
), sol–gel (Co
O
3 4
), spray
sputtering (Co O ), electrodeposition
3 4
25,27
26
technologies regarding its high efficiency and selectivity to pyrolysis (Co
O ),
3 4
11–14
28
29
30
2
CO .
Among the reported catalysts, transition metal oxides (CoO), hydrolysis (CoO), solvothermal (Co
3
O
4
and CoO) and
). However, with these methods,
Specically, the purity of the acquired samples was usually difficult to ensure
31
(
TMOs) exhibit great potential due to its low cost and efficiency excess impregnation (Co O
X Y
15,16
compared to the traditional noble metal catalysts.
cobalt oxides (CoO
X
) show interesting catalytic properties for the and some of them involved strict conditions or complex treat-
deep oxidation of VOCs. A lot of works have been contributed to ments. Compared to the above-mentioned techniques, pulsed-
the synthesis, characterization and catalytic application of spray evaporation chemical vapor deposition (PSE-CVD) is easy
17–19
20–23
Co O
and cobalt-based oxides.
However, besides Co O
3 4
to control the quality and thickness of the lms on exible
3
4
substrates with low cost. In recent years, PSE-CVD has been
3
2
applied widely to the synthesis of TMOs, such as Co O ,
3
4
a
Institute of Engineering Thermophysics, Chinese Academy of Sciences, 11 Beisihuanxi
Road, Beijing 100190, China. E-mail: tianzhenyu@iet.cn; Fax: +86 10 82543184; Tel:
33
34
35
Mn O ,
Co Fe O
and Co Cr O .
It should be
3
4
3ꢁX
X
4
3ꢁX
X 4
mentioned that as a exible substrate the stainless steel gird
mesh is proved to have no or negligible catalytic effect and can
be used for most reactors shapes. Thus, PSE-CVD exhibits great
+
86 10 82543184
b
University of Chinese Academy of Sciences, Beijing 100049, China
Electronic supplementary information (ESI) available: Experimental conditions;
†
results of repeated catalytic tests with coated mesh, NCM and a blank system; potential to prepare CoO
thin lms on stainless steel gird mesh
X
standard IR spectrum for C
3 6
H ; in situ DRIFTS spectra for Mesh 350 and 450.
for real evaluation of the catalytic performance of CoO samples.
X
See DOI: 10.1039/c5ra20013j
97272 | RSC Adv., 2015, 5, 97272–97278
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
The present work is oriented to synthesize the thin lms of
CoOX by PSE-CVD in a tailored way. The effect of substrate using electrodeposition method. Before deposition, their
temperature on the phases of the lms was investigated. Several electrode needed special pretreatments such as polishing and
dedicated techniques were used to characterize the prepared sonication, and all solutions were prepared by doubly distilled
RSC Advances
In 2010, Heli and Yadegari reported the CoO preparation by
28
lms with respect to structure, chemical composition and water. However, their XRD results are too ambiguous to indicate
morphological properties. The catalytic performance of the the formation of pure CoO with good crystallization. Also,
18
grown CoO
X
was investigated for the total oxidation of propene as a large grain size (800 nm) was reported. Recently, Edla et al.
nanoparticles (NPs) by pulsed laser deposi-
a representative of VOCs in a xed-bed quartz reactor. Further- synthesized Co
3 4
O
more, to get an insight into the catalytic mechanism, a newly tion. These NPs needed heating in Ar atmosphere to remove
developed in situ diffuse reectance infrared Fourier transform possible moisture adsorption before CO oxidation tests. The
spectroscopy (DRIFTS) was employed. With this setup, surface sizes of Co O NPs were in the range of 18–500 nm which
3
4
adsorbed species can be detected during the catalytic test which increased with the increasing laser uence. These reported
provides information to understand the catalytic mechanism.
large grain sizes are expected to offer less surface area and thus
lead to limited catalytic activity.
Further analysis with FTIR spectroscopy also conrmed the
Results and discussion
Structure
3 4
transformation of the phases from CoO to pure Co O spinel. As
presented in Fig. 2, two obvious bands with maxima at about 657
ꢁ
1
The XRD patterns of the CoO thin lms prepared at tempera- and 553 cm can be observed and become stronger from Mesh
X
ꢀ
ture range of 350–450 C are displayed in Fig. 1. With the 350 to Mesh 450. It should be noted that there is a strong peak at
ꢀ
ꢁ1
substrate temperature increasing from 350 to 450 C, the ob- 580 cm for Mesh 350, while such peak becomes weak for Mesh
tained oxide transformed from CoO to pure Co
gradually. 400 and can be hardly recognized for Mesh 450 in view of relative
3 4
O
ꢀ
ꢀ
ꢀ
ꢀ
36,37
For Mesh 350, peaks at 36.50 , 42.40 , 61.52 and 73.70 can be intensity. According to the literature,
the absorption peaks at
ꢁ1
seen clearly, which are attributed to (111), (200), (220) and (311) 553 and 657 cm are assigned to the transversal optical mode
ꢁ
1
orientations of CoO (JCPDS no. 43-1004), respectively. For Mesh (TO) of Co
O
3 4
and the peak at 580 cm is due to the longitu-
ꢀ
4
3
50, well-dened diffraction peaks are observed at 2q of 19.00 , dinal optical mode (LO) of CoO. Therefore, from Fig. 2, it can be
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
1.27 , 36.85 , 38.54 , 44.81 , 55.66 , 59.36 and 65.24 , which deduced that a pure Co
O
3 4
was acquired at 450 C and the
, as
ꢀ
correspond to (111), (220), (311), (222), (400), (422), (511) and sample prepared at 400 C was a mixture of CoO and Co
O
3
4
3 4
(440) orientations of Co O
(JCPDS no. 42-1467). No character- for Mesh 350, CoO was the main composition synthesized at this
istic peaks of any other impurities were observed with Mesh temperature. Since exposure to the atmosphere, there are more
3
+
4
50, revealing the high purity of the synthesized thin lms. For Co on the surface than the bulk average which will strengthen
Mesh 400, both the peaks of CoO and Co
were observed, Co peaks in FTIR. So stronger peaks of Co were observed
which indicates the formation of a mixture of CoO and Co
spinel. Though weak peaks at 31.27 and 65.24 can also be
observed for Mesh 350, the three main peaks manifest that the
main composition of Mesh 350 is CoO. The crystallite sizes of
3
O
4
3
O
4
3 4
O
3
O
4
in Fig. 2 for Mesh 350 than Fig. 1.
ꢀ
ꢀ
Morphology
the CoO
X
oxides were estimated to be 20 ꢂ 5 nm by applying To have an insight into the surface morphology and spatial
Scherrer's formula D ¼ 0.89l/b cos q to the most intense organization of the obtained CoO
thin lms, SEM was
X
diffraction peak, where l ¼ 0.154056 nm and where b and q employed here. As displayed in Fig. 3, the morphology of the
represent the full width at half maximum (FWHM) and
20
diffraction angle of the observed peak, respectively.
Fig. 2 Effect of the substrate temperature on the diffuse reflectance
Fig. 1 XRD patterns of the CoO
X
thin films.
FTIR spectra of the prepared samples.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 97272–97278 | 97273

View Article Online
RSC Advances
Paper
deposited thin lms could also differ with substrate tempera-
ꢀ
tures. For the sample prepared at 350 C (Fig. 3a), highly
dispersed tiny rods are distributed in the smooth amorphous
oxide matrix, which is the highly agglomeration of these rods.
Fig. 3b shows a loose spatial organization and a blocky structure
with amount of tiny folds. Especially, In fact, beneath these
large grains appeared at the surface, agglomeration of smaller
structures can also be recognized. Since the grain sizes calcu-
lated by XRD data are smaller, it can be deduced that these large
grains maybe supported by numerous tiny structures. As for the
ꢀ
Co O spinel obtained at 450 C (Fig. 3c), numerous small
3
4
grains with general agglomeration were observed. The micro-
morphology can be explained by the growth process which
contains nucleation, growth and coalescence process. For the
initial grown grains, they could get a bigger size by further
deposition and get close to each other. Then, by the coalescence
of small ones the larger grains would appear. This agglomera-
tion phenomenon was also observed in the pulsed laser depo-
Fig. 4 Entire XPS spectra of Mesh 400 before (a) and after etching (b).
18
sition of Co O . From the three micrographs in Fig. 3, nano-
3
4
2
+
Mesh 350 contains 91% of Co and the calculated formula is
CoO1.05 which indicates that the main composition of Mesh 350
structured particles were observed, which could be due to the
agglomeration phenomenon. The size of these particles on the
surface seems to be 30–40 nm on average, a little bigger than the
2
+
3+
is CoO. Mesh 400 has a similar amount of Co and Co and the
stoichiometric formula is CoO1.29. For Mesh 450, the ratio of Co/
38
values calculated by the XRD analysis. Barreca et al. had
employed cold-wall low-pressure CVD to prepare CoO previ-
3 4
O is 3 : 4.11, which is very close to Co O . These observations
X
further conrm the transformation of CoO to pure Co
temperature increases from 350 to 450 C.
3
O
4
as the
ously, but the substrate needed sputtering as pretreatment.
ꢀ
Rounded grains with diameter in the range of 340–580 nm were
38
The O 1s spectra of the synthesized lms were deconvoluted
into two peaks in the binding energy (BE) range of 528–532 eV in
Fig. 6. The lower peak is attributed to the lattice oxygen species
obtained. Spray pyrolysis had also been used to synthesize
Co O and got some overgrown clusters with the average size of
3
4
27
8
X
0 nm. Compared to these large grains, CoO with much
2
ꢁ
40,41
O
, and the other one is assigned to adsorbed oxygen.
As
smaller grain size was obtained in the present work. Moreover,
the tiny structure and loose spatial organization which always
accompanied with a high specic surface area are expected to
be benecial to the catalytic performances.
depicted in the tted O 1s spectra, all the samples were
composed of two kinds of oxygen. In each sample, the atomic
percentage of lattice oxygen was more than 90% and Mesh 400
possesses the most adsorbed oxygen (see Table 1). It is generally
recognized that with TMOs catalysts, the lattice oxygen plays an
important role in the catalytic oxidation process according to
the redox mechanism. However, among the three samples, the
amount of lattice oxygen has no much differences. Different
amount of adsorbed oxygen may lead to different performances
since adsorbed oxygen may create more active sites for catalytic
reactions.
Chemical composition
To explore the chemical composition of the prepared thin lms
on stainless steel, XPS was carried out on both bare and etched
surfaces of the CoO lms. As depicted in Fig. 4, on the bare
X
surfaces, much carbon was detected, which may be contami-
nation products originated from precursor decomposition and
ambient air. Aer 50 nm etching, carbon contamination is
negligible. Fig. 5 and 6 are the Co 2p and O 1s spectra aer
etching, respectively. To avoid Auger correction effect, only the
39
Co 2p1/2 spectra were tted. According to Gautier et al., the Co
2
+
2
p1/2 peaks at around 795.0 eV are composed of Co (796.4 eV)
3
+
and Co (795.2 eV). As shown in Fig. 5, all the samples are
composed of Co and Co . The tted results are in Table 1,
2
+
3+
X
Fig. 3 SEM images of CoO thin films coated on stainless steel mesh:
(
a) Mesh 350, (b) Mesh 400 and (c) Mesh 450.
X
Fig. 5 XPS of Co 2p of the prepared CoO samples.
97274 | RSC Adv., 2015, 5, 97272–97278
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
reaction process over NCM with a considerable amount which
can be attributed to the partial oxidation. In contrast, no traces
of CO or C H were detected with any coated meshes, as indi-
2
4
cated in Fig. 7b (Mesh 350 and Mesh 450 have overlapped with
Mesh 400, so only Mesh 400 was given as a representative),
demonstrating that CoO
formation of CO and C
With an Arrhenius expression for evaluation of the light-off
proles in the region where less than 15% of C was con-
X
is an effective catalyst to prevent the
2
4
H .
3
H
6
verted, apparent activation energies (Eappa) were calculated
following ref. 17. Fig. 8 compares the Eappa values obtained in
X
Fig. 6 O 1s signals of CoO thin films.
this work. The reaction with Mesh 400 shows relatively low E
appa
Table 1 Results of fitted Co 2p1/2 and O 1s of different samples
ꢁ1
(
78.1 kJ mol ) which indicates its high reactivity at low
ꢁ
1
temperatures. It increases to 107.1 and 132.6 kJ mol with
Mesh 450 and 350, respectively. The non-catalyzed reaction
Stoichiometric
formula
Atomic percentage
(%)
Proportion
ꢁ
1
presents an Eappa of 95.8 kJ mol which is surprisingly lower
than Mesh 350 and 450. Above all, Eappa is a parameter which
focuses only on the initiation of the chemical reactions. Only
with a low Eappa does not mean high-activity and can't promise
a low temperature for complete conversion of the fuel mole-
cules. The results indicate that the activity of catalysts increased
greatly as temperature increases while NCM exhibits inert
activity. For the initiation processes of propene oxidation,
crystal defects formed in the preparation of Mesh 350 and 450
3
+
2+
Sample
Co : Co
CoO
X
O
Lattice
O
Adsorbed
Mesh 350
Mesh 400
Mesh 450
0.09 : 0.91
0.58 : 0.42
0.74 : 0.26
CoO1.05
CoO1.29
CoO1.37
92.1
91.3
92.7
7.9
8.7
7.3
Catalytic performance
Fig. 7 compares the outlet proles of C
oxidation of C over three CoO
H
3 6
2 4
, CO and C H in the samples could make them less active. On the other hand, the
H
3 6
X
samples and with NCM as oxygen vacant sites, which normally play key roles in the redox
ꢀ
reference. For the three coated meshes, the conversion of C
H
3 6
mechanism, tend to increase from 350 to 400 C and then
ꢀ
ꢀ
becomes measurable at around 200 C and complete conversion decrease from 400 to 450 C, enabling Mesh 350 and 450 to
ꢀ
is reached within 400 C. Mesh 350 and 400 exhibit slight better exhibit high Eappa values. This is supported by the lowest Eappa
performance than Mesh 450 for about 25 C with the complete and best performance for sample Mesh 400. Moreover, the
ꢀ
conversion temperature. For the reaction with NCM, the adsorbed oxygen on the NCM could benet for the initiation of
ꢀ
consumption of C
H
3 6
was found to be negligible up to 500 C propene oxidation, which leads to a low Eappa with NCM.
ꢀ
and 700 C was needed to attain its complete conversion to CO
2
.
Compared with Mesh 350 and Mesh 450, Mesh 400 has a loose
Table 2 summarizes the comparison of the catalytic performance spatial organization and so higher specic surface area to
of different catalysts by T90. It is obvious that CoO samples absorb more oxygen and fuel, this behavior may promote the
X
prepared in this work are quite competitive to many other TMOs activation process. Moreover, from XPS results, Mesh 400 owns
and noble metal catalysts. It is worth mentioning that these the most adsorbed oxygen, which can provide more active sites
catalytic tests were repeated three times with the same sample in the catalytic process. For Mesh 350, though with a higher
and the results have no apparent difference (see ESI Fig. S1†), Eappa, the complete conversion temperature is lower than Mesh
ꢀ
demonstrating that the prepared CoO
The investigation of C
X
has good reusability.
450 by 25 C, which can be due to its higher mobility of lattice
2+
42
3
H
6
oxidation over NCM showed no oxygen according to its better reducibility with more Co .
signicant difference compared with a blank system (see ESI
Fig. S2†), suggesting that the mesh has a negligible catalytic
effect and can be used as an inert carrier for the objective
Mechanism
evaluation of deposited catalyst. In each case, CO was observed In order to reveal the possible catalytic mechanism over the
2
to be the nal product. CO and C H were detected during the prepared CoO , in situ DRIFTS was performed. To identify the
2
4
X
characteristic peaks of C H , pure gas of C H was measured
3
6
3
6

rst using an IR chamber and the observed spectrum is showed
in ESI Fig. S3.† For this in situ DRIFTS test, the spectra with
Mesh 400 are presented in Fig. 9 as a representative and the le
two spectra of Mesh 350 and 450 could be found in ESI Fig. S4
and S5,† respectively. Fig. 9 depicts the IR spectra at different
ꢁ
1
temperatures. Among these absorption peaks, 913 cm
is
assigned to the CH wagging vibration of adsorbed C H on the
2
3
6
52
ꢁ1
deposited lm; 669 cm comes from the bending vibration of
ꢁ1
CO
2
; 557 and 657 cm are characteristic absorption peaks of
ꢁ
1
Fig. 7 Outlet profiles of C
H
3 6
(a), CO and C
2
H
4
(b).
Co
3
O
and 582 cm is attributed to CoO as depicted in Fig. 2.
4
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 97272–97278 | 97275

View Article Online
RSC Advances
Paper
Table 2 Comparison of T90 of different catalysts
ꢁ1
a
ꢀ
Catalyst
Weight (mg)
Inlet composition
Flow rate (mL min
)
T
90 ( C)
Ref.
Mesh 350
Mesh 400
Mesh 450
13
13
13
12
12
41.5
12
12
20
12
20
12
200
200
4/200
8.2/200
—
1% C
1% C
1% C
1% C
2% C
3
3
3
3
3
H
H
H
H
H
6
6
6
6
6
/10% O
/10% O
/10% O
/10% O
/20% O
2
2
2
2
2
in Ar
in Ar
in Ar
in Ar
in Ar
15
15
15
15
15
500
15
15
15
15
15
15
100
120
75
30
15
347
348
369
380
385
356
356
301
375
403
382
384
500
351
410
412
648
This work
This work
This work
32
43
17
43
44
45
33
46
47
48
49
50
51
Co
3
Co
3
Co
3
O
O
O
4
4
4
0.4% C
3
H
6
/3.6% O
2
in Ar
Co2.66Mn0.34
CuO
O
4
2% C
1% C
1% C
1% C
1% C
1% C
3
3
3
3
3
3
H
H
H
H
H
H
6
6
6
6
6
6
/20% O
/10% O
/10% O
/10% O
/10% O
/10% O
2
2
2
2
2
2
in Ar
in Ar
in Ar
in Ar
in Ar
in Ar
in N
Cu
Mn
2
O
3
O
4
Co2.1Fe0.9
O
4
CuCo
La1.7Sr0.3CuO
Co0.3Ce
2
O
4
4
S
0.2
0.1% C
0.1% C
3
H
H
6
/5% O
/9% O
2
2
3
O
6.4
3
6
2
in He
Au/Al
Au/Al
NCM
2
O
O
3
0.15% C
0.4% C
1% C
3
H
6
/4% O
2
in He
in He
2
3
3
H
6
/3.6% O
2
3
H
6
/10% O
2
in Ar
This work
a
T90 refers to the temperature at which 90% of the fuel is converted.
and Co can then be reoxidized to form CoO
X
by the replenished
56,57
oxygen in the gas phase.
tion involving chemisorbed species O
3 4
the total oxidation of C to CO over Co O . It's worth noting
Moreover, the electrophilic oxida-
ꢁ
ꢁ
2
and O will promote
57
3
H
6
2
that the catalysts take part in the reaction during the propene
oxidation process and the chemical composition of the active
surface is always changing as a consequence of sustainable
oxidation–reduction, but the compositions of the catalysts
43
would keep the same before and aer the catalytic reaction.
This phenomena were also conrmed by running the catalytic
Fig. 8 The apparent activation energy (Eappa) of NCM and coated tests several times and the same results were obtained. As dis-
ꢀ
meshes prepared at 350, 400 and 450 C during catalytic processes of
ꢀ
played in Fig. 9, with temperature increased from 50 to 400 C,
C
3 6
H .
ꢁ1
X
the three CoO adsorption bands at 657, 582 and 557 cm
ꢁ
1
exhibit red shi of 11, 9 and 13 cm , respectively. The red shi
is due to the change of vibration energy level as the temperature
increases, which further indicates that more bulk oxygen is
ꢀ
As can be seen from Fig. 9, spectra measured below 200 C had
ꢁ1
strong adsorption peak of C
3 6
H (913 cm ). With temperature
increasing, adsorption peak of C H vanishes gradually and the
3
1
6
ꢁ
peak attribute to CO (669 cm ) become stronger. This indi-
2
cates that during the catalytic process, C
surface of the CoO catalyst rstly, and then the reaction rate of
adsorbed C is accelerated as temperature increases and
more CO is generated. With a higher C consumption rate
and so CO production rate, the adsorption peak of C
becomes weaker and the peak of CO becomes stronger. With in
situ DRIFTS spectra, adsorption of C H on CoO thin lms was
3 6
H is adsorbed on the
X
3
H
6
2
3 6
H
2
3 6
H
2
3
6
X
observed, which suggests that the catalytic reaction of C H on
3
6
CoO
X
is preceded with the Langmuir–Hinshelwood (L–H)
mechanism.
Normally, redox mechanism is accepted for the catalytic
oxidation of light hydrocarbons over TMOs as the metal ions
can be circulated by reduction to lower ionic states and recovery
5
3–55
3+
to higher ionic states via re-oxidization.
In this work, Co
2
+
2+
and Co in the CoO thin lms could be reduced to Co and
X
Co by releasing active oxygen. As temperature increases,
adsorbed C H will react with trapped or released oxygen. Co
3 6
ꢀ
X 3 6
Fig. 9 In situ DRIFTS spectra of CoO prepared at 400 C during C H
oxidation at different temperatures.
2
+
97276 | RSC Adv., 2015, 5, 97272–97278
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
45
activated. From the random changes of the relative intensity of setup can be found elsewhere and only a brief description is
CoO and Co O , a sustainable transformation process between given here. A feed mixture of 1 vol% C H and 10 vol% O in the
3
4
3
6
2
3
+
2+
Co and Co can be proposed, which supports the redox balance of argon was introduced into a tubular reactor at a total
mechanism in the oxidation of C H over CoO .
ꢁ
1
ow rate of 15 mL min , corresponding to weight hourly space
velocity (WHSV) of about 69 000 mL gcat
of the reactor was raised with a ramp of 5 C min using
a digital electrical furnace. The exhaust gases were analyzed
using a gas chromatograph (Agilent, GC3000).
3
6
X
ꢁ
1
ꢁ1
h . The temperature
ꢀ
ꢁ1
Experimental
Synthesis of CoO lms
X
The CoOX thin lms were synthesized in a PSE-CVD system
To explore the catalytic mechanism, in situ DRIFTS tests were
combined with cold-wall stagnation point-ow CVD reactor and carried out. The stainless steel mesh with deposition was placed
waste collector. The experimental setup can be found in our in the temperature-controlled chamber. C
previous work and typical deposition conditions are shown in ratio of 1%/10%/89% and a total ow rate of 15 mL min were
H
/O
/Ar with the
3
6
2
45
ꢁ1
ESI Table S1,† respectively. Cobalt acetylacetonate (Co(acac)
2
)
introduced into the chamber. The temperature of the chamber
ꢀ
ꢁ1
was dissolved in ethanol with a concentration of 2.5 mM as the was raised with a ramp of 5 C min from room temperature to
ꢀ
feedstock. The frequency and width of the injector are 4 Hz and 400 C during the tests. The background spectrum was deduc-
1
.2 ms. The liquid feedstock, with a feeding rate of 1.0 mL ted by measuring the spectrum of a non-coated mesh (NCM) in
ꢁ
1
min , was injected as a ne spray into a 30 cm long evapora- the in situ chamber under the same gas inlet conditions. By
ꢀ
tion chamber kept at 220 C. The resulting vapor was trans- online monitoring the transform of the characteristic bands of
ported to the deposition chamber, with N /O ow rates of 0.25/ CoO
and absorption of fuel gases, possible catalytic mecha-
2
2
X
0
.75 standard liter per minute (SLM), respectively. Flow rates of nism can be proposed.
gases were controlled by MKS mass ow controllers. The total
pressure in the reactor during deposition was kept at 3.0 kPa. To
meet the requirements of different characterization techniques,
Conclusions
the deposition was performed on several types of substrates The one-step PSE-CVD method was applied to prepare CoO
X
without any treatments, including bare glass, stainless steel and thin lms from Co(acac)
mesh grid of stainless steel. Mesh 350, Mesh 400 and Mesh 450 350 to 450 C. The obtained samples were characterized in
2
by adjusting the temperature from
ꢀ
are dened as the deposited lms on relevant substrate at 350, terms of structure, morphology, surface composition and cata-
ꢀ
4
00 and 450 C, respectively. The substrates were heated using lytic properties with XRD, FTIR, SEM, XPS and in situ DRIFTS.
a at resistive heater by a heating controller (HT60, Horst). The Structure studies demonstrate that with an increase of substrate
ꢀ
waste vapor was collected by a liquid nitrogen trap and the rest temperature from 350 to 450 C, the obtained oxides varied
gases were pumped away. The weight of the deposited lms was from CoO to pure Co
3 4
O spinel gradually. Adsorption of oxygen
determined gravimetrically by measuring the weight change of species is more with Mesh 400 than 350 and 450 which may
the substrates with a microbalance (Sartorius).
benet the catalytic performances. Catalytic tests indicate that
CoO exhibit a competitive catalytic activity compared to sup-
X
ported noble metals and many other TMOs in the deep oxida-
tion of C H . Mesh 350 and 400 showed slightly better
performance over Mesh 450 due to its more low-valence cobalt,
porous structure and more adsorbed oxygen. With increasing
Characterization
3
6
The phases of the deposited lms obtained at different
substrate temperatures were identied by X-ray diffraction
(
0
4
XRD) using a Bruker D8 Focus equipped with a Cu Ka (l ¼
temperature, adsorption peaks of C and CO
3
H
6
2
were detected
.154056 nm) radiation source and operated at 40 kV and
and surface conversion of CoO–Co O was also observed during
3 4
0 mA. By referring to the Joint Committee on Powder Diffrac-
in situ DRIFTS tests. These observations indicate that the cata-
lytic reaction was preceded with redox and L–H mechanism.
tion Standards (JCPDS) database cards, the crystalline phases
were identied. The morphology of the deposited lms was
examined using SEM (QUANTA FEG 250). The characteristic
vibrations of the prepared thin lm catalysts were ascertained
using a diffuse reectance infrared Fourier transform spec-
troscopy (DRIFTS, Bruker VERTEX 70) with the spectral reso-
Acknowledgements
ZYT thanks the nancial support from the Recruitment
Program of Global Youth Experts. The authors thank Dr Jun-Jie
Weng and Mr Bing-Yin Wang for help discussions.
ꢁ
1
ꢁ1
lution of 4 cm in the range of 400–4000 cm
.
Catalytic test
Notes and references
3 6
Small unsaturated hydrocarbons, such as propene (C H ), is an
important class of VOCs and quite difficult to be oxidized. In this
1 J. J. Schauer, M. J. Kleeman, G. R. Cass and B. R. T. Simoneit,
Environ. Sci. Technol., 2001, 35, 1716–1728.
work, the catalytic behavior of CoO was investigated toward the
X
oxidation of C H at atmospheric pressure in a xed-bed quartz
2 C. J. Cai, F. H. Geng, X. X. Tie, Q. O. Yu and J. L. An, Atmos.
Environ., 2010, 44, 5005–5014.
3 P. R. Veres, P. Faber, F. Drewnick, J. Lelieveld and
J. Williams, Atmos. Environ., 2013, 77, 1052–1059.
3
6
reactor with 13 mg catalyst grown on meshes of stainless steel.
To exclude the effect of background, a contrast test using mesh
grid without any deposition was performed. The experimental
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 97272–97278 | 97277

View Article Online
RSC Advances
Paper
4
S. Batterman, F. C. Su, S. Li, B. Mukherjee, C. R. Jia and 32 P. Mountapmbeme Kouotou, Z. Y. Tian, U. Mundloch,
H. E. I. H. R. Committee, Res. Rep.–Health Eff. Inst., 2014,
–63.
N. Bahlawane and K. Kohse-H ¨o inghaus, RSC Adv., 2012, 2,
10809–10812.
3
5
6
7
8
9
T. Y. Chang and M. J. Suzio, J. Air Waste Manage. Assoc., 1995, 33 Z. Y. Tian, P. Mountapmbeme Kouotou, N. Bahlawane and
4
5, 20–28.
US, EPA, Air pollution control technology fact sheet, EPA-452/F-
3-022.
G. Chen, K. A. Strevett and B. A. Vanegas, Biodegradation,
001, 12, 433–442.
P. H. Tchoua Ngamou, J. Phys. Chem. C, 2013, 117, 6218–
6224.
34 N. Bahlawane, P. H. Tchoua Ngamou, V. Vannier, T. Kottke,
J. Heberle and K. Kohse-H ¨o inghaus, Phys. Chem. Chem.
Phys., 2009, 11, 9224–9232.
0
2
W. J. Cooper, M. G. Nickelsen, R. V. Green and S. P. Mezyk, 35 V. Vannier, M. Schenk, K. Kohse-H ¨o inghaus and
Radiat. Phys. Chem., 2002, 65, 571–577. N. Bahlawane, J. Mater. Sci., 2012, 47, 1348–1353.
A. M. Mastral, T. Garcia, M. S. Call ´e n, J. M. L ´o pez, 36 M. Lenglet, J. Lopitaux, L. Terrier, P. Chartier, J. F. Koenig,
M. V. Navarro, R. Murillo and J. Galb ´a n, Environ. Sci.
Technol., 2002, 36, 1821–1826.
0 H. L. Huang and W. M. Lee, J. Environ. Eng., 2002, 128, 60–67.
E. Nkeng and G. Poillerat, J. Phys. IV, 1993, 3, 477–483.
37 B. Lefez, P. Nkeng, J. Lopitaux and G. Poillerat, Mater. Res.
Bull., 1996, 31, 1263–1267.
1
1
1 E. N. Ndifor, T. Garcia, B. Solsona and S. H. Taylor, Appl. 38 D. Barreca, C. Massignan, S. Daolio, M. Fabrizio,
Catal., B, 2007, 76, 248–256.
2 K. Everaert and J. Baeyens, J. Hazard. Mater., 2004, 109, 113–
C. Piccirillo, L. Armelao and E. Tondello, Chem. Mater.,
2001, 13, 588–593.
1
1
1
1
39.
39 J. L. Gautier, E. Rios, M. Gracia, J. F. Marco and
J. R. Gancedo, Thin Solid Films, 1997, 311, 51–57.
40 G. Tyuliev and S. Angelov, Appl. Surf. Sci., 1988, 32, 381–391.
3 P. Papaehimiou, T. Ioannides and X. E. Verykios, Appl.
Catal., B, 1997, 13, 175–184.
4 S. Minic `o , S. Scir `e , C. Crisafulli, R. Maggiore and 41 S. Valeri, A. Borghi, G. C. Gazzadi and A. di Bona, Surf. Sci.,
S. Galvagno, Appl. Catal., B, 2000, 28, 245–251. 1999, 423, 346–356.
5 Z. Gu and K. L. Hohn, Ind. Eng. Chem. Res., 2004, 43, 30–35. 42 P. Mountapmbeme Kouotou, Z. Y. Tian, H. Vieker, A. Beyer,
1
1
6 L. F. Liotta, M. Ousmane, G. Di Carlo, G. Pantaleo,
A. Golzhauser and K. Kohse-H ¨o inghaus, J. Mater. Chem. A,
2013, 1, 10495–10504.
G. Deganello, A. Boreave and A. Giroir-Fendler, Catal. Lett.,
2
009, 127, 270–276.
43 Z. Y. Tian, P. H. Tchoua Ngamou, V. Vannier, K. Kohse-
H ¨o inghaus and N. Bahlawane, Appl. Catal., B, 2012, 117–
118, 125–134.
1
1
1
2
2
2
7 Z. Y. Tian, N. Bahlawane, F. Qi and K. Kohse-H ¨o inghaus,
Catal. Commun., 2009, 11, 118–122.
8 R. Edla, N. Patel, Z. El Koura, R. Fernandes, N. Bazzanella 44 Z. Y. Tian, H. J. Herrenbr u¨ ck, P. Mountapmbeme Kouotou,
and A. Miotello, Appl. Surf. Sci., 2014, 302, 105–108.
9 K. Shojaee, B. S. Haynes and A. Montoya, Appl. Surf. Sci.,
H. Vieker, A. Beyer, A. G ¨o lzh ¨a user and K. Kohse-
H ¨o inghaus, Surf. Coat. Technol., 2013, 230, 33–38.
45 G. F. Pan, S. B. Fan, J. Liang, Y. X. Liu and Z. Y. Tian, RSC
Adv., 2015, 5, 42477–42481.
46 Z. Y. Tian, P. Mountapmbeme Kouotou, A. El Kasmi,
P. H. Tchoua Ngamou, K. Kohse-H ¨o inghaus, H. Vieker,
A. Beyer and A. G ¨o lzh ¨a user, Proc. Combust. Inst., 2015, 35,
2207–2214.
2014, 316, 355–365.
0 Z. Y. Tian, N. Bahlawane, V. Vannier and K. Kohse-
H ¨o inghaus, Proc. Combust. Inst., 2013, 34, 2261–2268.
1 A. Goyal, S. Bansal, V. Kumar, J. Singh and S. Singhal, Appl.
Surf. Sci., 2015, 324, 877–889.
2 A. Jamal, M. M. Rahman, S. B. Khan, M. Faisal, K. Akhtar,
M. A. Rub, A. M. Asiri and A. O. Al-Youbi, Appl. Surf. Sci., 47 Z. Y. Tian, H. Vieker, P. Mountapmbeme Kouotou and
012, 261, 52–58. A. Beyer, Faraday Discuss., 2015, 177, 249–262.
3 D. H. Shang, Q. Zhong and W. Cai, Appl. Surf. Sci., 2015, 325, 48 M. Machida, K.-i. Ochiai, K. Ito and K. Ikeue, J. Catal., 2006,
11–216. 238, 58–66.
4 K. J. Kim and Y. R. Park, Solid State Commun., 2003, 127, 25–28. 49 L. F. Liotta, M. Ousmane, G. Di Carlo, G. Pantaleo,
2
2
2
2
2
5 P. S. Patil, L. D. Kadam and C. D. Lokhande, Thin Solid Films,
996, 272, 29–32.
G. Deganello, G. Marc `ı , L. Retailleau and A. Giroir-Fendler,
Appl. Catal., A, 2008, 347, 81–88.
1
2
2
2
2
6 H. Yamamoto, S. Tanaka and K. Hirao, J. Appl. Phys., 2003, 50 S. Ivanova, C. Petit and V. Pitchon, Gold Bull., 2006, 39, 3–8.
9
3, 4158–4162.
51 A. C. Gluhoi, N. Bogdanchikova and B. E. Nieuwenhuys, J.
Catal., 2005, 229, 154–162.
52 P. R. Griffiths and J. A. de Haseth, Fourier transform infrared
spectrometry, John Wiley & Sons, 2007.
7 V. R. Shinde, S. B. Mahadik, T. P. Gujar and C. D. Lokhande,
Appl. Surf. Sci., 2006, 252, 7487–7492.
8 H. Heli and H. Yadegari, Electrochim. Acta, 2010, 55, 2139–
2148.
53 M. Labanowska, ChemPhysChem, 2001, 2, 712–731.
9 L. Poul, S. Ammar, N. Jouini, F. Fi ´e vet and F. Villain, Solid 54 I. Aso, M. Nakao, N. Yamazoe and T. Seiyama, J. Catal., 1979,
State Sci., 2001, 3, 31–42.
57, 287–295.
3
3
0 K. Deori and S. Deka, CrystEngComm, 2013, 15, 8465–8474.
1 G. L. Zhou, H. M. Xie, B. G. Gui, G. Z. Zhang and X. X. Zheng,
Catal. Commun., 2012, 19, 42–45.
55 M. Konsolakis, M. Sgourakis and S. A. Carabineiro, Appl.
Surf. Sci., 2015, 341, 48–54.
56 D. Duprez, Catal. Today, 2006, 112, 17–22.
57 J. Haber and W. Turek, J. Catal., 2000, 190, 320–326.
97278 | RSC Adv., 2015, 5, 97272–97278
This journal is © The Royal Society of Chemistry 2015