Enhancement of propane combustion activity over CoO: Xcatalysts by introducing C2-C5diols

Article abstract of DOI:10.1039/d1nj00052g

A series of nanocrystalline CoOx catalysts were synthesized by a citric acid complexation strategy. The effects of C2-C5 diols (including ethylene glycol, 1,3-propanediol, 1,4-butanediol, and 1,5-pentadiol) on catalyst structure and propane combustion performance were investigated. The results showed that the structure of CoOx transformed from pure Co3O4 to mixed-phase CoO-Co3O4 with a larger surface area and smaller crystallite size after the introduction of diol compounds. Importantly, diol compounds effectively weaken the Co-O bond strength in nanocrystalline CoOx and enhance the activity of oxygen species, thus effectively degrading propane. Due to the weakest Co-O bond strength, excellent oxygen species activity, abundant oxygen vacancies, and the strongest redox ability, the catalyst prepared by adding 1,3-propanediol (Co-PDO) exhibited the best catalytic activity (T90 = 227 °C). Moreover, the increase in pore size, surface Co2+ and oxygen vacancy content of Co-PDO in the 40 h stability test can compensate for the loss of activity caused by the decrease in specific surface area, thereby maintaining stable catalytic activity.

Full text of DOI:10.1039/d1nj00052g

NJC
View Article Online
View Journal
PAPER
Enhancement of propane combustion activity
over CoO catalysts by introducing C –C diols†
x
2
5
Cite this: DOI: 10.1039/d1nj00052g
ab
ab
ab
ab
ab
Zhao Liu,
‡
Lijun Cheng,‡ Shiyun Zhangxue, Min Huang, Jia Zeng,
a
a
a
a
Shanliang Yuan, Qifei Bo, Biao Zhang and Yi Jiang*
A series of nanocrystalline CoO
effects of C –C diols (including ethylene glycol, 1,3-propanediol, 1,4-butanediol, and 1,5-pentadiol) on
catalyst structure and propane combustion performance were investigated. The results showed that the
structure of CoO transformed from pure Co to mixed-phase CoO–Co with a larger surface area
and smaller crystallite size after the introduction of diol compounds. Importantly, diol compounds
effectively weaken the Co–O bond strength in nanocrystalline CoO and enhance the activity of oxygen
x
catalysts were synthesized by a citric acid complexation strategy. The
2
5
x
3
O
4
3 4
O
x
species, thus effectively degrading propane. Due to the weakest Co–O bond strength, excellent oxygen
species activity, abundant oxygen vacancies, and the strongest redox ability, the catalyst prepared by
adding 1,3-propanediol (Co-PDO) exhibited the best catalytic activity (T90 = 227 1C). Moreover, the
Received 5th January 2021,
Accepted 14th April 2021
2+
increase in pore size, surface Co and oxygen vacancy content of Co-PDO in the 40 h stability test can
compensate for the loss of activity caused by the decrease in specific surface area, thereby maintaining
stable catalytic activity.
DOI: 10.1039/d1nj00052g
rsc.li/njc
1
4,15
16–18
VOCs can be improved through method control,
structure induction, morphology regulation,
doping,
1
. Introduction
19
20,21
defect con-
etc. The essence of catalytic combustion is the deep
atmospheric and environmental pollution, especially to the for- oxidation reaction between active oxygen and reactant molecules.
2
2,23
Volatile organic compounds (VOCs) contribute significantly to struction,
1,2
mation of photochemical smog, ozone, and greenhouse gases.
In view of this, the activation ability of adsorbed oxygen, the
Among them, short-chain alkanes represented by methane, ethane, mobility of surface lattice oxygen, the formation of oxygen
and propane are the most difficult VOC molecules to eliminate due vacancies, and the redox properties in CoOx will be directly
to the stability of C–C and C–H bonds. Catalytic combustion has related to the conversion activity of VOCs. Basically, the proper-
been recognized as an effective method to degrade VOCs, such as ties of the metal–oxygen bond (such as bond strength) will
3
,4
5,6
O-containing VOCs, Cl-containing VOCs, and especially short- fundamentally affect the properties related to oxygen species.
7
24
chain alkanes, due to its environmental friendliness and high Hou et al. found that high temperature would enhance the
efficiency. The conversion efficiency of VOCs in catalytic combus- Mn–O bond strength of the cryptomelane-type octahedral mole-
tion depends on the design and preparation of high-performance cular sieve (OMS-2) nanorods, thereby inhibiting the lattice oxygen
25
catalysts. Supported platinum group metal (PGM) (such as Pt, Pd, activity, which was not conducive to benzene oxidation. Rong
and Ru) catalysts are effective components for the degradation of et al. also demonstrated that the Mn–O bond strength of the (310)
8
–11
short-chain alkanes,
but their application is hindered due to planes on a-MnO is lower than that of the (100) and (110) planes,
2
their scarcity and high price. In view of this, the development of which will increase the lattice oxygen reactivity and promote the
2
2
cheap and resource-rich transition metal oxides (TMO) for the formaldehyde oxidation. Shen et al. found that calcination
degradation of short-chain alkanes is a promising alternative. temperature significantly affected the degradation activity of
Co is a promising TMO catalyst due to its strong activation Co for toluene, which was attributed to the influence of Co–O
3
O
4
3 4
O
1
2,13
ability and tunability.
Currently, the combustion efficiency of bond strength on the redox and surface oxygen species activity.
These works fully demonstrate the impact of metal–oxygen bond
a
National Engineering Laboratory for VOCs Pollution Control Material &
Technology Chengdu, Chengdu Institute of Organic Chemistry, Chinese Academy of
Sciences, Chengdu 610041, Sichuan, China
strength on the catalyst oxygen species activity and the combustion
performance of VOCs. To our knowledge, it is still a challenge to
enhance the degradation activity of CoO catalyst to short-chain
b
x
University of Chinese Academy of Sciences, Beijing 100049, China.
E-mail: yjiangcioc@163.com; Tel: 028-8522945
alkanes by weakening the metal–oxygen bond.
Previous work showed that citric acid can efficiently
†
‡
Electronic supplementary information (ESI) available. See DOI: 10.1039/d1nj00052g
These authors contributed equally to this work.
reduce the crystallite size of CoO
x
and increase the propane
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem.

View Article Online
Paper
NJC
2
6
ꢀ1
conversion activity. In this work, a series of nanocrystalline recorded from room temperature to 600 1C (10 1C min ) after
CoO
x 2
catalysts were prepared by the citric acid complexation baseline stabilization. The H consumption of the sample was
method. The effects of C –C diols (including ethylene glycol, calibrated by the H -TPR experiment of CuO. Oxygen temperature-
2
5
2
1
,3-propanediol, 1,4-butanediol, and 1,5-pentanediol) on the programmed desorption (O -TPD) and H -TPR have the same
2 2
physicochemical properties of nanocrystalline CoO , especially experimental device. In the experiment, 100 mg of the sample
x
the Co–O bond strength, were systematically studied by BET, was put into a U-tube and pretreated at 300 1C for 1 h under high
XRD, Raman, XPS, TEM/HRTEM, H
2
-TPR and O
2
2
-TPD, and the purity O . After the U-tube was cooled to room temperature, the
catalytic performance of the catalysts in the catalytic combustion
of propane was evaluated. The purpose of this work is to study the from room temperature to 600 1C (10 1C min ) after baseline
O
2
was switched to high purity He. TCD signals were recorded
ꢀ1
effects of diol compounds on the structure and surface micro- stabilization.
environment of CoO , and to establish the structure–activity
x
relationship between CoO and propane conversion.
2.3. Activity evaluation
x
The catalytic activity was evaluated on a CGK-5A fixed bed
reactor with an inner diameter of 8 mm. The raw gas consisted
of 0.3 vol% propane and balanced air, with a gas hourly space
2
. Experimental
ꢀ1
ꢀ1
2.1. Catalyst preparation
velocity (GHSV) of 30 000 ml g
h . The propane concen-
tration was measured online using an SC-8000 gas chromato-
graph (GC) equipped with a flame ionization detector (FID),
while the CO and CO2 concentrations in exhaust gas were
detected using an SC-3000 GC equipped with a thermal capture
detector (TCD). The conversion of propane is shown in eqn (1):
A citric acid complexation method, which was similar to that
19,26
reported in the literature, was employed.
process, 0.0411 mol each of ethylene glycol (EG), 1,3-propanediol
PDO), 1,4-butanediol (BDO), and 1,5-pentanediol (PTO) was added
In a typical synthesis
(
ꢀ
1
separately to 25 ml of 1 mol L cobalt nitrate solution under
stirring. Next, 0.0274 mol of citric acid was added to the cobalt salt
mixed solution. The mixed solution was continuously stirred and
x(C
3
H
8
) = (1 ꢀ (C
3
H
8
)out/(C
3
H
8
)in) ꢁ 100%
(1)
evaporated for 1.5 h at 80 1C and then dried directly at 120 1C for where x(C
h to obtain the corresponding precursor. In the following (C
3 8
H ) represents the conversion of propane, and
6
3
H
8
)
in and (C out are the volume concentration of propane
3 8
H )
calcination process, the precursor was calcined in an air atmo- before and after the reaction, respectively.
ꢀ
1
sphere at 300 1C for 1 h at a heating rate of 2 1C min to
27
decompose the organic framework. Furthermore, the catalyst 2.4. Kinetic data measurements
was obtained by heating to 400 1C for 2 h at the same heating rate.
The kinetic experiment is similar to those reported in the
The catalyst without diol addition was labeled as Co-Original,
while the catalyst with diol addition was labeled as Co-EG,
Co-PDO, Co-BDO, and Co-PTO, respectively.
13,26
literature,
by eliminating the effect of internal and external
diffusion. When the conversion of propane is less than 15%,
ꢀ1 ꢀ1
the reaction rate (r, mol s
g ) can be expressed by eqn (2):
2
.2. Catalyst characterization
The X-ray diffraction (XRD) pattern was obtained by using a
100 diffractometer (Shimadzu, Japan) with Cu Ka radiation
l = 0.1541 nm) operated at 30 mA and 40 kV (step size = 0.021,
r = F(C
3
H
8
) ꢁ x(C
3
H
8
)/W(Cat)
(2)
ꢀ1
where F(C H ) represents the flow rate of propane (mol s ) and
W(Cat) is the mass of the catalyst (g). The combustion process
of propane is irreversible, and the influence of reaction products
2 2
such as H O and CO on the reaction rate can be ignored.
Therefore, the reaction rate can also be expressed by the Arrhenius
equation (eqn (3)):
3
8
6
(
ꢀ1
scan rate = 41 min ). N
on a NOVA 2200E surface area analyzer (Quantachrome, USA) at
196 1C. The pore size distributions and specific surface areas
2
adsorption–desorption was completed
ꢀ
were determined by Barrett–Joyner–Halenda (BJH) and Brunauer–
Emmett–Teller (BET) methods, respectively. Raman spectra were
obtained using a Dxi Raman microscope spectrometer (Thermo
Fisher Scientific, USA) with an excitation wavelength of 532 nm
a
b
r ¼ A expðꢀEa=RTÞP
P
(3)
C3H8 O2
ꢀ
1
where A is the pre-exponential factor, E (kJ mol ) is the apparent
activation energy, R is the ideal gas constant (8.314 J mol
and T (K) is the reaction temperature. When the conversion rate of
propane is less than 15%, the composition of the raw gas can be
regarded as approximately constant. The relationship between ln r
and RT can be obtained by eqn (4):
a
(InGaN laser). Transmission electron microscopy (TEM/HRTEM)
ꢀ
1
ꢀ1
K ),
was performed by using a Tecnai G2 F20 electron microscope (FEI,
USA) with an acceleration voltage of 200 kV. X-ray photoelectron
spectroscopy (XPS) was performed by using an XSAM800 spectro-
meter (Kratos, USA) with Al Ka radiation. The electron binding
energy was calibrated based on the C 1s spectrum (284.8 eV).
Hydrogen temperature-programmed reduction (H -TPR) was mea-
sured on a self-assembled device. In each test, 15 mg of the sample
2
ln r = ꢀE /RT + C
(4)
a
was placed in a U-tube and pretreated at 300 1C for 1 h under high where C is a constant. By calculating the r of propane at
purity N . After the U-tube was cooled to room temperature, the N different temperatures, E can be inferred from the slope of
was switched to a 5% H –95% N mixed gas. TCD signals were the ln r and 1/RT curves.
2
2
a
2
2
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
NJC
Paper
3. Results and discussion
3.1. Crystal phase, texture properties, morphology, and crystal
plane
2
The N adsorption–desorption isotherm of the catalysts is
shown in Fig. 1; all catalysts exhibit typical IV shape isotherms
and obvious hysteresis loops, indicating the presence of meso-
porous structures. Among them, Co-Original has H3 type
hysteresis loops, while Co-EG, Co-PDO, Co-BDO, and Co-PTO
all present H4 type hysteresis loops, indicating that the addition
of diol compounds significantly affects the pore structure of the
catalyst. Specifically, Co-Original has hysteresis loops when P/P 4
a
0.75, while the catalyst added with diol compounds has hysteresis
loops when P/P 4 0.45. A wider hysteresis loop indicates less
a
2
8
pore-blockage and more efficient mass transfer. Meanwhile, the
change of the specific surface area and pore structure of
the catalyst is affected by the hysteresis loop. Table 1 summarizes
the texture properties of the as-prepared catalysts. Based on the
larger hysteresis loop, the specific surface area and pore volume
of Co-EG, Co-PDO, Co-BDO, and Co-PTO are significantly
improved compared to Co-Original. The results indicate that
Fig. 2 XRD patterns of the as-prepared catalysts.
diols, the diffraction peak intensity of Co O in the Co-EG,
3
4
Co-PDO, Co-BDO, and Co-PTO catalysts is significantly weakened,
indicating the poor crystallinity. The calculated results of the
Scherrer formula (Table 1) further confirm that the crystalline size
of the catalyst decreased by about 6 nm after the addition of diol
compounds.
2 5
the introduction of C –C diols during the citric acid complexa-
tion process can effectively optimize the specific surface area
and pore volume of the catalyst, thereby effectively exposing
Co particles.
The XRD patterns of the as-prepared catalysts are shown in
Fig. 2. Co-Original exhibits diffraction peaks at 2y = 31.31, 36.81,
Moreover, the Co-EG, Co-PDO, Co-BDO, and Co-PTO catalysts
not only display the Co
2-1474) diffraction peaks at 2y = 42.31 and 61.81, corresponding to
200) and (220) planes. The gel network structure formed by the
3 4
O diffraction peaks, but also CoO (JCPDS
7
(
3
8.51, 44.81, 55.71, 59.41, and 65.21, which are attributable to
spinel Co (JCPDS 43-1003), corresponding to (220), (311),
222), (400), (422), (511) and (440) planes. After adding C –C
interaction of diol compounds and citric acid plays a reducing role
in the process of calcination and decomposition, which may be the
3 4
O
(
2
5
3 4
reason for the partial conversion of Co O to CoO. Interestingly,
the formation of the CoO phase leads to irregular shifts of Co
3 4
O
diffraction peaks, which is conducive to the generation of lattice
defects.
The Raman spectra of the as-prepared catalysts are shown in
Fig. 3a. All catalysts show 5 Raman peaks in the range of 100–
ꢀ
1
9
00 cm , with different peak positions and peak intensities. The
variation of peak intensity in the five catalysts is attributed to the
influence of crystallinity and the orientation effect in the Raman
2
9
test. The symmetrical Raman peaks of Co-Original at 197, 481,
ꢀ
1
1
2
3
5
22, 618 and 688 cm correspond to F2g, E
g
, F2g, F2g, and A1g
30
stretching vibrations in the cubic spinel Co
3 4
O , respectively.
According to XRD, the addition of diol compounds caused the
catalyst to produce the CoO phase, but there was no obvious
difference from the Raman peak of Co-Original. According to
Fig. 1
2
N adsorption–desorption isotherms of the as-prepared catalysts.
relevant reports, both CoO and Co
O
3 4
g
have F2g, E , and A1g
3
1,32
Raman active modes.
The Raman peak of CoO is not clearly
discriminated from the entire spectrum, which may be due to the
fact that the signal of CoO is relatively weak and superimposable
Table 1 Texture properties and crystalline size of the as-prepared
catalysts
33
3 4
to that of Co O .
Compared with Co-Original, the Raman peak
-
2
S
BET
V
p
ꢁ 10
D
(nm)
BJH
Crystalline
size (nm)
2
-1
3
ꢀ1
a
of Co-EG, Co-PDO, Co-BDO, and Co-PTO moves to the lower
Sample
(m ꢂg )
(cm ꢂg
)
wavenumber direction, indicating that the Co–O bond was
Co-Original
Co-EG
Co-PDO
Co-BDO
Co-PTO
31.4
77.8
87.8
101.2
67.4
5.6
7.6
7.2
8.6
5.8
4.0
3.6
3.6
3.6
3.6
17.5
11.9
11.8
11.1
11.4
22,34
distorted. According to Hooke’s law,
the Co–O bond force
constant (k), which represents the Co–O bond strength, can be
calculated by using the following equation:
pffiffiffiffiffiffiffiffi
a
o ¼ 1=2pc k=m
(5)
3 4
The crystalline size of Co O was calculated based on the (311) plane.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem.

View Article Online
Paper
NJC
Fig. 3 (a) Raman spectra and (b) Co–O bond force constant k of the as-prepared catalysts.
ꢀ
1
where o is the Raman shift (cm ), c is the light velocity, and m is
TEM images, particle size distributions, and lattice fringes
the effective mass. As shown in Fig. 3b, the Co–O force constant of typical catalysts Co-Original and Co-PDO are shown in Fig. 4.
1
(
k) of CoO (F ) and CoO (A ) decreases in the sequence of Both Co-Original and Co-PDO are nanocrystalline assembled
4 2g 6 1g
Co-Original 4 Co-EG 4 Co-BDO4 Co-PTO 4 Co-PDO. The results from an irregular polyhedron, and the Co-PDO particles have a
reveal that the introduction of diol weakened the Co–O bond higher dispersion degree. After the addition of PDO, the
strength of CoO
oxygen in CoO
4
and CoO
and CoO
6
, thus increasing the reactivity of lattice average particle size of Co-PDO (11.69 nm) is smaller than that
6
. Moreover, Co–O bond distortion also of Co-original (13.89 nm), which is in line with the average
4
indicates that the addition of diol compounds causes structural crystalline size measured by XRD. Lattice fringes with a spacing
defects and lattice distortion, which leads to the increase of oxygen of 0.243 nm and 0.286 nm are exposed on the Co-Original
22
vacancies. These oxygen vacancies can convert molecular oxygen surface, which is assigned to (311) and (220) planes of Co O .
3
4
to reactive oxygen species and replenish the depleted lattice Compared with Co-Original, Co-PDO exposes the (220) planes
oxygen, thus increasing the degradation rate of propane. of Co , with higher visibility and clarity.
3 4
O
Fig. 4 TEM images, particle size distributions and lattice fringes of (a–c) Co-Original and (d–f) Co-PDO catalysts.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
NJC
Paper
3
.2. Surface composition
of Oads/Olatt on the catalyst surface is Co-PDO 4 Co-PTO 4
Co-BDO 4 Co-EG 4 Co-Original. Generally, the mobility of
adsorbed oxygen on the surface is better than that of lattice
oxygen, and thus the adsorbed oxygen is more active than the
lattice oxygen in the corresponding VOC oxidation process. In
short, the catalyst prepared by adding diol compounds has
more surface adsorbed oxygen species and will exhibit better
activity in the combustion of propane.
The surface element composition and chemical state of the
as-prepared catalysts were analyzed by XPS technology. The Co
2
7
p XPS spectra (Fig. 4a) display two main peaks at about
80.5 eV and 795.5 eV, which correspond to the characteristic
4
2
peaks of Co 2p3/2 and Co 2p1/2, respectively. Meanwhile, two
satellite peaks at about 786.0 eV and 803.0 eV are typical
characteristics of Co species. Moreover, the satellite features
of the Co 2p spectrum can distinguish the Co oxidation state.
2
+
35
2
+
3.3. Oxygen species and redox properties
Compared with Co-Original, the satellite peaks (Co species) of
Co-EG, Co-PDO, Co-BDO, and Co-PTO have a larger peak area The H -TPR curves of different catalysts are shown in Fig. 6.
2
and stronger peak intensity, which are well-matched with the After peak fitting, Co-Original shows two partially overlapping
3
6
satellite peaks of CoO. These results further confirm that reduction peaks (a peak and b peak) at 336 1C and 390 1C,
3
+
2+
diol compounds can promote the formation of mixed-phase which are attributed to the reduction of Co - Co and
2
+
43,44
CoO–Co O . The Co 2p spectra were de-convoluted into two Co - Co in Co O , respectively.
Compared with Co-
peaks located at a BE of about 779.2 eV and 781.2 eV, belonging Original, the catalyst prepared by adding diol compounds not
3
4
3/2
3 4
3
+
2+ 37–39
to the surface Co and Co .
The corresponding XPS only formed a and b peaks but also produced a broad peak
parameters are listed in Table 2. The order of the molar ratio (g peak) near 460 1C, which can be attributed to the reduction
2
+
3+
2+
19
of Co /Co on the catalyst surface is as follows: Co-PDO 4 of Co - Co in CoO. Since all the catalysts are calcined at
Co-PTO4 Co-BDO4 Co-EG4 Co-Original. Since the increase 400 1C, the reduction properties of the catalyst within 400 1C are
2
+
in Co content caused by CoO has little effect on oxygen discussed emphatically. Compared with Co-Original, the a and b
vacancies, the content of oxygen vacancies cannot be simply peaks of the catalyst prepared by adding diol compounds move to
2
+
3+
judged by the molar ratio of Co /Co .
the low-temperature direction, indicating stronger redox ability.
Table 3 summarizes the reduction peak temperature and H2
In the O 1s XPS spectra (Fig. 5b), the peaks near 529.0 eV,
5
30.5 eV, and 532.5 eV after de-convolution are assigned to the consumption of the catalysts. The H2 consumption of the a
2
ꢀ
lattice oxygen species Olatt (O ), surface adsorbed oxygen peak of the catalyst prepared by adding diol compounds is
species Oads (O
to adsorb oxygen.
2
ꢀ
2ꢀ
ꢀ
3+
2
, O , O ), and molecular water or hydroxyl significantly reduced, which is due to the reduction of Co ions
4
0,41
From Table 2, the order of the molar ratio content caused by the formation of CoO. Among them, Co-PDO
Table 2 XPS parameters of the as-prepared catalysts
Binding energy (eV)
Surface element content (mol%)
2+
3+
2+
3+
Catalyst
Co
Co
O
OH
ꢀ
/CO32ꢀ
O
ads
O
latt
Co
O
Co /Co
Oads/Olatt
Co-Original
Co-EG
Co-PDO
Co-BDO
Co-PTO
781.0
781.3
781.1
781.4
781.4
779.2
779.3
779.1
779.5
779.4
532.3
532.4
532.5
532.5
532.6
530.5
530.8
530.9
530.8
530.9
528.9
529.0
529.3
529.3
529.4
28.6
28.8
27.8
36.9
31.5
71.4
71.2
72.2
63.1
68.5
1.01
1.14
1.28
1.20
1.22
1.19
1.48
1.63
1.54
1.55
Fig. 5 (a) Co 2p and (b) O 1s XPS spectra of the as-prepared catalysts.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem.

View Article Online
Paper
NJC
(below 350 1C), a2 region (350 1C–500 1C), and a3 region (above
500 1C) are attributed to the desorption of surface adsorbed
oxygen, surface lattice oxygen, and bulk lattice oxygen,
4
5
respectively. Since all the catalysts were calcined at 400 1C,
the oxygen desorption properties of the catalysts in the a1 and
a2 regions are discussed emphatically. Compared with
Co-Original, the peak position of the catalyst prepared by
adding diol compounds shifts to a low temperature to varying
degrees in the a1 and a2 regions, showing better surface
adsorbed oxygen and surface lattice oxygen migration ability.
2
ꢀ
ꢀ
2ꢀ
Generally, surface adsorbed oxygen (O , O , O ) and surface
2
2
ꢀ
lattice oxygen (O ) with strong mobility are easier to desorb
from transition metal oxides and participate in the oxidation
reaction as surface active oxygen species, thus having stronger
reactivity. Importantly, Co-PDO has strong surface adsorbed
oxygen and surface lattice oxygen activity, which is due to its
weak Co–O bond strength.
Fig. 6
2
H -TPR profiles of the as-prepared catalysts.
3.4. Catalytic performance
Table 3
2
H -TPR parameters of the as-prepared catalysts
The catalytic performance of catalysts was evaluated by propane
combustion experiments. Previous work showed that propane
ꢀ1
2
H consumption (mmol g )
Catalyst
a (T/1C)
b (T/1C)
g (T/1C)
was completely oxidized on CoO
carbon and CO formation . Fig. 8a shows the activity curve of
without associated hydro-
x
1
9
Co-Original
Co-EG
Co-PDO
Co-BDO
Co-PTO
6.91(336)
1.61(310)
0.81(297)
1.63(305)
1.49(306)
5.34(390)
5.21(346)
5.53(339)
7.31(351)
6.76(345)
—
2.82(456) propane conversion over the catalysts. The Co-Original catalyst
3.60(476)
has the worst catalytic activity, with T90 = 250 1C. Compared
1.73(471)
2.78(469)
with Co-Original, the activity of the catalyst prepared by adding
diol compounds has increased to varying degrees. The order of
activity of the catalysts from large to small is Co-PDO4 Co-BDO
E Co-PTO4 Co-EG4 Co-Original. Among them, Co-PDO has
the best propane combustion activity, with T90 = 227 1C, which
is 23 1C lower than that of Co-Original.
The Arrhenius relationship curve between ln r and 1/RT is
a
shown in Fig. 8b, and the corresponding activation energy (E )
is obtained from the slope of the fitted line. The activation
energy sequence of the catalysts is as follows: Co-Original 4
Co-EG 4 Co-BDO 4 Co-PTO 4 Co-PDO. Co-PDO has the
ꢀ
1
has the least H consumption (0.81 mmol g ). In addition,
2
Co-PDO has the lowest a and b peak temperatures, indicating
that the surface and bulk oxygen species have the strongest
oxidation activity. In this case, the easiest reduction of Co-PDO
may be due to the weakest Co–O bond strength and more
oxygen vacancies that promote the activity of oxygen species.
The effect of C –C diols on the oxygen species of the catalyst
2 5
was studied by O -TPD, and the patterns are shown in Fig. 7. All
catalysts produced several different desorption peaks, corresponding
to different oxygen species. The desorption peaks in the a1 region
2
ꢀ1
lowest activation energy (33.2 kJ mol ), which can reduce the
energy barrier of the propane oxidation reaction, thereby making
the reaction easier to initiate.
The stability test of Co-PDO at 227 1C is shown in Fig. 8c. In
the 40 h continuous operation, the conversion of propane was
maintained at about 91%, indicating that Co-PDO has stable
catalytic performance.
Moreover, Fig. 8d also exhibits the catalytic stability of
Co-PDO in the presence of water vapor. For water resistance
tests, a water vapor saturator is connected between the feed gas
and the fixed-bed reactor and placed in a water bath to control
the temperature. About 3.0% of water vapor is introduced into
the feed gas through bubbling water at 25 1C. During the 40 h
continuous test, no significant change in propane conversion
was caused by introducing 3.0 vol% water vapor and removing
water vapor, indicating that Co-PDO has stable water resistance.
Generally, the addition of water vapor will compete with the
adsorption of C
of adsorbed C
3
H
8
and O
2
, causing a reduction in the number
3
8
H , O , and active sites, thereby increasing the
2
Fig. 7
O
2
-TPD profiles of the as-prepared catalysts.
propane conversion temperature. The stability of Co-PDO in the
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
NJC
Paper
Fig. 8 (a) Catalytic activity of propane combustion; (b) Arrhenius plots of the relationship between ln r and 1/RT; (c) stability test of Co-PDO at 227 1C;
(
d) effect of 3.0 vol% water vapor on the catalytic stability of Co-PDO at 227 1C. Test condition: 0.3 vol% propane balanced by air, total flow rate =
00 mL min , GHSV = 30 000 ml g h .
ꢀ1
ꢀ1 ꢀ1
1
presence of water vapor may be due to its stronger propane catalysts, the specific reaction rate roughly demonstrated that
adsorption capacity than water vapor, thus maintaining stable Co-PDO has better catalytic activity. These results indicate that
propane conversion activity.
Co-PDO is a promising material for propane combustion.
As shown in Table 4, the propane conversion activity of the
catalyst prepared here was compared with other materials
previously reported in the literature. The specific reaction rate
3
.5. Crystal phase, texture properties and surface composition
after 40 h reaction
is determined by the material characteristics and test conditions. XRD, N adsorption–desorption, and XPS techniques were used
2
Compared with Co O , two-component transition metal oxides, to investigate the crystal phase, texture, and surface composition
3
4
three-component transition metal oxides, and noble metal of Co-PDO after 40 h reaction. The XRD patterns of Co-PDO and
Table 4 Comparison with catalysts reported in the literature
ꢀ1
ꢀ1
ꢀ8
ꢀ1 ꢀ1
Catalysts
Co-PDO
Test conditions
GHSV (ml g
h
)
T
90 (1C)
r (ꢁ10 mol g
s
)
Ref.
a
b
b
b
b
0.3% C
0.1% C
0.1% C
3
3
3
H
H
H
8
8
8
, 21% O
, 10% O
, 21% O
2
2
2
30 000
240 000
40 000
120 000
120 000
20 000
64 800
80 000
36 000
227
252
238
241
310
325
290
250
296
37.91, 83.79
This work
a
Co
3
Co
3
Co
3
O
O
O
4
4
4
27.70 , 69.25
17.56 , 31.42
24.26 , 76.25
46
47
48
49
50
51
52
53
a
a
0.25% C
3
H
8
, 10% O
2
c
CeCoO
MnZrO
MnCeNbO
Pt/AlF
x
0.2% C
0.2% C
0.1% C
0.2% C
0.2% C
3
3
3
3
3
H
H
H
H
H
8
8
8
8
8
, 5% O
, 8% O
, 17.85% O
, 2% O
, 10% O
2
44.8
a
d
x
2
1.35 , 32.52
x
2
—
b
3
2
2.07b
Pd–CeO
2
2
8.30
a
b
c
Specific reaction rate normalized to catalyst mass at 200 1C. Specific reaction rate normalized to catalyst mass at 220 1C. Specific reaction rate
d
normalized to catalyst mass at 250 1C. Specific reaction rate normalized to catalyst mass at 300 1C.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem.

View Article Online
Paper
NJC
2
Fig. 9 (a) XRD patterns, (b) N adsorption–desorption isotherms, (c) Co 2p and (d) O 1s XPS spectra of Co-PDO and Co-PDO-40 h.
Table 5 Physical and chemical parameters of Co-PDO and Co-PDO-40 h
Surface element content (mol%)
Pore
size (nm)
Crystalline
size (nm)
2
ꢀ1
2+
3+
Catalyst
S
BET (m g
)
Co
O
Co /Co
Oads/Olatt
Co-PDO
Co-PDO-40 h
87.8
23.2
3.6
13.1
11.8
13.2
27.8
31.2
72.2
68.8
1.28
1.15
1.63
1.01
Co-PDO-40 h are shown in Fig. 9a. Compared with Co-PDO, the
111) and (220) planes of CoO disappeared in Co-PDO-40 h, which are shown in Fig. 9c and d. Compared with Co-PDO, the O
was attributed to the partial oxidation of CoO to Co O after 40 h species content of Co-PDO-40 h decreased from 72.2% to
The Co 2p and O 1s XPS spectra of Co-PDO and Co-PDO-40 h
(
3
4
reaction. Besides, the crystalline size (Table 5) of the catalyst 68.8%, while the Co species content increased from 27.8% to
2
+
3+
slightly increased from 11.8 nm to 13.2 nm after the reaction, 31.2%. Moreover, the Co /Co and Oads/Olatt ratio on the catalyst
revealing its stable anti-sintering performance. surface both decreased to varying degrees (Table 5), which was
Fig. 9b shows the N adsorption–desorption isotherms of due to the partial oxidation of CoO and the consumption of
Co-PDO and Co-PDO-40 h. After a 40 h reaction, the hysteresis surface adsorbed oxygen. After calculation, it was found that the
2
2
+
loop of the catalyst moved in the direction of high P/P
a
,
surface content of Co increased from 15.6% to 16.7%, which
5
4
indicating a larger pore size. As shown in Table 5, compared can form abundant oxygen vacancies, accelerate the adsorption,
with Co-PDO, the specific surface area of the catalyst decreased migration, and activation of gas-phase oxygen, and provide
from 87.8 m g to 23.2 m g , which would lead to a serious enough active oxygen species for the oxidation of propane.
change in the dispersion of the active phase, resulting in a Moreover, under low-temperature conditions, the oxidation
2
ꢀ1
2
ꢀ1
decline in catalytic activity. However, the pore size of the catalyst of propane is in the kinetic control stage, and the reaction rate
increased from 3.6 nm to 13.1 nm, which promoted the rapid mainly depends on the interaction between propane molecules
diffusion and migration of reactant and product molecules.
and oxygen molecules on the surface of the catalyst, so the
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
NJC
Paper
specific surface area will have a significant impact on the activity of compounds led to the formation of CoO, which promoted the
the catalyst; however, under high-temperature conditions, the dispersion of Co (N adsorption–desorption) and decreased
oxidation of propane is in the stage of material transfer limitation, the crystalline size (XRD and TEM). CoO and the high specific
O
3 4
2
2
+
3+
and the reaction molecules and product molecules gather in the surface area effectively increase the Co /Co and Oads/Olatt
catalyst pores and cannot diffuse rapidly. At this stage, the pore ratio (XPS) on the catalyst surface. Interestingly, the interaction
size and surface species distribution have a significant impact on of CoO and Co
the activity of the catalyst. (Raman), thereby increasing the activity of surface oxygen
Therefore, the catalytic activity was mainly affected by the species (H -TPR, O -TPD). Especially, Co-PDO has the best
pore structure and surface species during the stability test. The propane conversion activity (T90 = 227 1C), which can be
3 4
O reduces the strength of the Co–O bond
2
2
2
+
increase of pore size, surface Co , and oxygen vacancy content attributed to the weakest Co–O bond strength, strong oxygen
can compensate for the activity loss caused by the decrease of species activity, abundant oxygen vacancies, and excellent redox
specific surface area, thus maintaining stable catalytic activity. ability. Moreover, Co-PDO showed stable catalytic activity in a
4
0 h continuous test, which was attributed to the excellent anti-
3
.6. Structure–activity relationship
2+
sintering ability and more Co exposure.
48,55
Relevant literature
shows that the combustion of propane on
cobalt oxide follows the Mars–van-Krevelen (MVK) mechanism,
which is based on in situ diffuse reflectance infrared spectroscopy
4. Conclusions
(
in situ DRIFT) experiments. The oxidation process of propane on
In conclusion, in the citric acid complexation method, C
diols (ethylene glycol, 1,3-propanediol, 1,4-butanediol, and
,5-pentanediol) significantly affected the phase composition,
2 5
–C
cobalt oxide is shown in Fig. 10. During the reaction, propane is
first adsorbed on the surface of the catalyst and reacts with the
lattice oxygen connected to Co to generate Co and oxygen
1
3+
2+
structure, and surface microenvironment of nanocrystalline
CoO . After introducing diol compounds, the structure of CoO
56
vacancies. Then, the generated oxygen vacancies adsorb gas-
phase oxygen molecules to form lattice oxygen, which is activated
to generate reactive oxygen species and participate in the reaction
cycle. The MVK mechanism reveals that the lattice oxygen species
on the catalyst surface plays an important role in the catalytic
combustion of propane. In fact, the MVK mechanism clarifies the
relationship between lattice oxygen, reactive oxygen species, and
oxygen vacancies. Oxygen vacancies can adsorb and capture
gaseous oxygen molecules to form lattice oxygen. After the
lattice oxygen is activated, reactive oxygen species are formed
to participate in the oxidation reaction and generate oxygen
vacancies, forming a reaction cycle. Therefore, the Co-PDO
catalyst in this manuscript has a weaker Co–O bond, abundant
oxygen vacancies, and strong lattice oxygen activity, which
accelerates the adsorption and activation of gas-phase oxygen,
and can form abundant reactive lattice oxygen species to
participate in the oxidation of propane.
x
x
changed from pure Co O to mixed-phase CoO–Co O , with a
3
4
3 4
larger surface area and smaller crystallite size. Importantly, the
addition of diol compounds weakens the Co–O bond strength
and increases the activity of oxygen species. Among them, the
catalyst prepared by adding 1,3-propanediol (Co-PDO) exhibited
the best propane conversion activity (T90 = 227 1C), which was
attributed to the weakest Co–O bond strength, excellent oxygen
species activity, abundant oxygen vacancies, and the strongest
redox ability. Furthermore, the increase in pore size, surface
2+
Co and oxygen vacancy content of Co-PDO in the 40 h stability
test can compensate for the loss of activity caused by the
decrease in specific surface area, thereby maintaining stable
catalytic activity. This low-cost, highly active, and stable nano-
crystalline CoO is a potential target for replacing platinum
x
group metal (PGM)-based catalysts.
Overall, during the synthesis of CoO
the addition of C –C diols affected the phase composition,
structure, and surface microenvironment of the catalyst. XRD,
x
by citric acid complexation,
2
5
Author contributions
XPS, and H -TPR technology confirmed that the addition of diol Zhao Liu and Shiyun Zhangxue: Methodology, Investigation,
2
Writing – original draft. Min Huang, Jia Zeng, and Lijun Cheng:
Formal analysis. Shanliang Yuan and Qifei Bo: Investigation.
Jiang Yi and Zhang Biao: Supervision, Formal analysis.
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
This work was supported by the Sichuan Province Science and
Technology Achievement Transfer and Transformation Project
Fig. 10 Schematic of propane oxidation on cobalt oxides.
(2021ZHCG0019).
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem.

View Article Online
Paper
NJC
References
25 S. Rong, P. Zhang, F. Liu and Y. Yang, ACS Catal., 2018, 8,
3435–3446.
1
W. Zhang, J. D ´ı ez-Ram ´ı rez, P. Anguita, C. Descorme, J. L.
2
2
2
2
3
3
6 Z. Liu, L. Cheng, J. Zeng, X. Hu, S. Zhangxue, S. Yuan, Q. Bo,
B. Zhang and Y. Jiang, J. Solid State Chem., 2020, 292, 121712.
7 S. Prasad, A. Vijayalakshmi and N. S. Gajbhiye, J. Therm.
Anal. Calorim., 1998, 52, 595–607.
8 W. Wang, H. Qi, H. Long, X. Wang and H. Ru, J. Mater.
Chem. A, 2014, 2, 5363–5370.
9 Z. Wang, W. Wang, L. Zhang and D. Jiang, Catal. Sci.
Technol., 2015, 6, 3845–3853.
0 Y. Lou, J. Ma, X. Cao, L. Wang, Q. Dai, Z. Zhao, Y. Cai,
W. Zhan, Y. Guo and P. Hu, ACS Catal., 2014, 4, 4143–4152.
1 Q. Wang, Z. Xu, S. Wang, Z. Wang, J. Jia, H. Li, Y. Cao,
Y. Chen, Y. Qin and F. Cui, Sep. Purif. Technol., 2021,
Valverde and A. Giroir-Fendler, Appl. Catal., B, 2020, 273,
118894.
2
3
4
5
6
7
C. He, J. Cheng, X. Zhang, M. Douthwaite, S. Pattisson and
Z. Hao, Chem. Rev., 2019, 119, 4471–4568.
J. H. Miller, A. Joshi, X. Li and A. Bhan, J. Catal., 2020, 389,
714–720.
H. Dib, R. El Khawaja, G. Rochard, C. Poupin, S. Siffert and
R. Cousin, Catalysts, 2020, 10, 870.
J. Zhu, W. Zhang, Q. Qi, H. Zhang, Y. Zhang, D. Sun and
P. Liang, Sci. Rep., 2019, 9, 12162.
W. Deng, Q. Tang, S. Huang, L. Zhang, Z. Jia and L. Guo,
Appl. Catal., B, 2020, 278, 119336.
T. P. O. Mkhwanazi, M. D. Farahani, A. S. Mahomed,
S. Singh and H. B. Friedrich, Appl. Catal., B, 2020, 275,
263, 118369.
3
3
3
3
2 Y. Wang, S. Guo, X. Xin, Y. Zhang, B. Wang, S. Tang and
X. Li, Appl. Surf. Sci., 2021, 549, 149108.
3 J.-S. Gwag and Y.-K. Sohn, Bull. Korean Chem. Soc., 2012, 33,
119118.
8
9
Y. R. Liu, X. Li, W. M. Liao, A. P. Jia, Y. J. Wang, M. F. Luo
and J. Q. Lu, ACS Catal., 2019, 9, 1472–1481.
C. Rao, C. Peng, H. Peng, L. Zhang, W. Liu, X. Wang,
N. Zhang and P. Wu, ACS Appl. Mater. Interfaces, 2018, 10,
505–510.
4 Y. J. Wei, L. Y. Yan, C. Z. Wang, X. G. Xu and G. Chen,
J. Phys. Chem. B, 2004, 108, 18547–18551.
5 X. X. Liu, J. B. Zang, L. Chen, L. B. Chen, X. Chen, P. Wu,
S. Y. Zhou and Y. H. Wang, J. Mater. Chem. A, 2017, 5,
9
220–9224.
0 K. A. Ledwa, M. Pawlyta and L. K ˛e pi n´ ski, Appl. Catal., B,
018, 230, 135–144.
1 X. Zou, Z. Ma, J. Deng, J. Zhong, Y. He and J. Liu, Chem. Eng.
J., 2020, 396, 125275.
1
1
1
1
1
1
1
1
5865–5872.
2
3
3
3
3
6 K. Dai, N. Zhang, L. Zhang, L. Yin, Y. Zhao and B. Zhang,
Chem. Eng. J., 2021, 414, 128804.
7 Z. Zhu, G. Lu, Z. Zhang, Y. Guo, Y. Guo and Y. Wang, ACS
Catal., 2013, 3, 1154–1164.
8 J. Marco, J. Gancedo, M. Gracia, J. Gautier, E. Rios and
F. Berry, J. Solid State Chem., 2000, 153, 74–81.
2 J. Zhao, Z. Tang, F. Dong and J. Zhang, Mol. Catal., 2019,
4
63, 77–86.
3 L. Cheng, Z. Liu, S. Yuan, M. Wei and Y. Jiang, Chem. Pap.,
019, 74, 1449–1457.
4 J. Gao, S. Gao, J. Wei, H. Zhao and J. Zhang, Catalysts, 2019,
, 994.
5 W. Ling, H. Zhao, Z. Tang and F. Zha, Catal. Surv. Asia, 2020,
4, 278–290.
2
´
9 T. Garcia, S. Agouram, J. F. Sanchez-Royo, R. Murillo, A. M.
Mastral, A. Aranda, I. Vazquez, A. Dejoz and B. Solsona,
Appl. Catal., A, 2010, 386, 16–27.
´
9
40 S. Rousseau, S. Loridant, P. Delichere, A. Boreave, J. Deloume
2
and P. Vernoux, Appl. Catal., B, 2009, 88, 438–447.
6 J. Zeng, H. Xie, G. Zhang, X. Cheng, G. Zhou and Y. Jiang, 41 Z. Chen, S. Wang, W. Liu, X. Gao, D. Gao, M. Wang and
Ceram. Int., 2020, 46, 21542–21550.
S. Wang, Appl. Catal., A, 2016, 525, 94–102.
7 B. Sellers-Anton, E. Bailon-Garcia, A. Cardenas-Arenas, 42 W. Tang, W. Xiao, S. Wang, Z. Ren, J. Ding and P.-X. Gao,
A. Davo-Quinonero, D. Lozano-Castello and A. Bueno-
Lopez, Environ. Sci. Technol., 2020, 54, 2439–2447.
8 T. Tabakova, L. Ilieva, P. Petrova, A. M. Venezia,
Appl. Catal., B, 2018, 226, 585–595.
43 W. Zhang, F. Wu, J. Li and Z. You, Appl. Surf. Sci., 2017, 411,
136–143.
1
Y. Karakirova, L. F. Liotta and G. Avdeev, Appl. Sci., 2020, 44 Q. Liu, L. C. Wang, M. Chen, Y. Cao, H. Y. He and K. N. Fan,
0, 1088.
J. Catal., 2009, 263, 104–113.
9 Z. Liu, L. Cheng, J. Zeng, X. Hu, S. Zhangxue, S. Yuan, Q. Bo, 45 L. Liotta, M. Ousmane, G. Di Carlo, G. Pantaleo, G. Deganello,
1
1
2
2
2
B. Zhang and Y. Jiang, Chem. Phys., 2021, 540, 110984.
0 Y.-C. Tsai, N. Nhat Huy, J. Lee, Y.-F. Lin and K.-Y. A. Lin,
Chem. Eng. J., 2020, 395, 124939.
G. Marc `ı , L. Retailleau and A. Giroir-Fendler, Appl. Catal., A,
2008, 347, 81–88.
46 L. Ma, Y. Geng, X. Chen, N. Yan, J. Li and J. W. Schwank,
Chem. Eng. J., 2020, 402, 125911.
1 T. Klein and G. Kickelbick, Dalton Trans., 2020, 49,
9
820–9834.
47 I. Driouch, W. Zhang, M. Heitz, J. L. Valverde and A. Giroir-
Fendler, Catal., 2020, 10, 900.
2 Y. Shen, J. Deng, S. Impeng, S. Li, T. Yan, J. Zhang,
L. Shi and D. Zhang, Environ. Sci. Technol., 2020, 54, 48 W. Tang, J. Weng, X. Lu, L. Wen, A. Suburamanian,
0342–10350.
C.-Y. Nam and P.-X. Gao, Appl. Catal., B, 2019, 256, 117859.
3 R. Camposeco, S. Castillo, M. Hinojosa-Reyes, N. Nava and 49 W. Zhu, X. Chen, J. Jin, X. Di, C. Liang and Z. Liu, Chin.
R. Zanella, Surf. Coat. Technol., 2020, 384, 125305.
J. Catal., 2020, 41, 679–690.
4 J. Hou, Y. Li, L. Liu, L. Ren and X. Zhao, J. Mater. Chem. A, 50 K. Zeng, X. Li, C. Wang, Z. Wang, P. Guo, J. Yu, C. Zhang and
1
2
2
2013, 1, 6736.
X. S. Zhao, J. Colloid Interface Sci., 2020, 572, 281–296.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
NJC
Paper
5
5
5
1 Y. Wang, D. Li, K. Li and R. Farrauto, Appl. Catal., B, 2020, 54 Q. Wang, L. Chen, S. Guan, X. Zhang, B. Wang, X. Cao,
2
67, 118363.
Z. Yu, Y. He, D. G. Evans, J. Feng and D. Li, ACS Catal., 2018,
8, 3104–3115.
55 Z. Ren, Z. Wu, W. Song, W. Xiao, Y. Guo, J. Ding, S. L. Suib
and P.-X. Gao, Appl. Catal., B, 2016, 180, 150–160.
2 X. Li, Y.-R. Liu, W.-M. Liao, A.-P. Jia, Y.-J. Wang, J.-Q. Lu and
M.-F. Luo, Appl. Surf. Sci., 2019, 475, 524–531.
3 T. Dong, W. Liu, M. Ma, H. Peng, S. Yang, J. Tao,
C. He, L. Wang, P. Wu and T. An, Chem. Eng. J., 2020, 56 X. Wang, Y. Liu, T. Zhang, Y. Luo, Z. Lan, K. Zhang, J. Zuo,
93, 124717. L. Jiang and R. Wang, ACS Catal., 2017, 7, 1626–1636.
3
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem.