Nitrogen-doped carbon nanotubes encapsulate cobalt nanoparticles as efficient catalysts for aerobic and solvent-free selective oxidation of hydrocarbons

Article abstract of DOI:10.1039/c7gc00469a

The selective oxidation of hydrocarbons to the corresponding ketones with solvent-free and molecular oxygen as an oxidant is of great importance in academic and industrial fields in view of economy and environment. In this respect, we present the facile synthesis and characterization of excellent catalysts comprising cobalt nanoparticles encapsulated into graphitic nitrogen-doped carbon nanotubes (Co@GCNs) via one-pot pyrolysis of a chelate compound containing citric acid, melamine, and CoCl₂·6H₂O. The selective oxidation of ethylbenzene under molecular oxygen and solvent-free conditions is employed as a probe reaction to investigate the catalytic performance; the optimized catalyst shows the best conversion (68%) and selectivity for acetophenone (93%). Combination of the catalytic results of the control group and the different characterization methods demonstrates that high catalytic activity is due to the synergistic effect between metallic cobalt and nitrogen-doped carbon nanotubes. Moreover, the catalyst has high catalytic activity for the aerobic and solvent-free oxidation of other arylalkane substrates. The proposed mechanistic study illustrates that the reaction is a free radical reaction progressing through superoxide radical anions (^?O^2-).

Full text of DOI:10.1039/c7gc00469a

Green Chemistry
View Article Online
View Journal
PAPER
Nitrogen-doped carbon nanotubes encapsulate
cobalt nanoparticles as efficient catalysts for
aerobic and solvent-free selective oxidation of
hydrocarbons†
Cite this: DOI: 10.1039/c7gc00469a
a
a
b
a
a
Xiu Lin, Zhenzhen Nie, Liyun Zhang, Shuchuan Mei, Yuan Chen,
b
c
a
Bingsen Zhang,* Runliang Zhu and Zhigang Liu
*
The selective oxidation of hydrocarbons to the corresponding ketones with solvent-free and molecular
oxygen as an oxidant is of great importance in academic and industrial fields in view of economy and
environment. In this respect, we present the facile synthesis and characterization of excellent catalysts
comprising cobalt nanoparticles encapsulated into graphitic nitrogen-doped carbon nanotubes
(Co@GCNs) via one-pot pyrolysis of a chelate compound containing citric acid, melamine, and
CoCl ·6H O. The selective oxidation of ethylbenzene under molecular oxygen and solvent-free con-
2
2
ditions is employed as a probe reaction to investigate the catalytic performance; the optimized catalyst
shows the best conversion (68%) and selectivity for acetophenone (93%). Combination of the catalytic
results of the control group and the different characterization methods demonstrates that high catalytic
activity is due to the synergistic effect between metallic cobalt and nitrogen-doped carbon nanotubes.
Moreover, the catalyst has high catalytic activity for the aerobic and solvent-free oxidation of other aryl-
alkane substrates. The proposed mechanistic study illustrates that the reaction is a free radical reaction
progressing through superoxide radical anions ( O2−).
Received 13th February 2017,
Accepted 28th March 2017
DOI: 10.1039/c7gc00469a
•
rsc.li/greenchem
Co, Ni, Mn and Cr have attracted considerable attention, which
Introduction
have been designed and synthesized in order to replace noble
1
Selective oxidation of organic compounds, such as alcohols, metals. As far as we know, a series of green oxidation reactions
saturated or unsaturated hydrocarbons, to the corresponding of hydrocarbons have been reported. For example, Peng et al.
high value-added alcohols, aldehydes and ketones is of crucial have studied the oxidation of cumene and ethylbenzene under
8–11
importance in the synthesis of chemical intermediates or fine molecular oxygen conditions,
but the catalytic performance
1–4
chemicals both in laboratory and industry.
Accordingly, in still needs to be improved. In view of the catalytic performance,
the past few decades, a great deal of research have been devoted non-noble transition metal catalysts are usually inferior to
to designing and synthesizing effective catalysts for selective noble metal catalysts, thus some additives and oxidants, includ-
1
2–14
oxidation of organic compounds. Noble metal catalysts such as ing toxic metal oxides and peroxides,
are employed to
palladium, platinum, ruthenium, and gold show superior cata- enhance the catalytic activity and selectivity of the desired pro-
2
,5–7
lytic activity and selectivity for catalytic oxidation,
but the ducts. Moreover, liquid phase solvent is generally added to reac-
1
3–17
high price and limited reserves greatly restrict the further usage tion systems,
which lead to the presence of many difficul-
in industry. Therefore, non-noble transition metals such as Fe, ties in products’ separation, catalysts’ recovery and a sharp rise
in cost. Consequently, in terms of economy and green reaction
process, a catalytic system with no solvent use, molecular
oxygen as an oxidant, low cost, and abundant effective catalysts
will be preferred for heterogeneous catalysis. Meanwhile, it
provides great opportunities and challenges for commercial
large-scale application.
a
State Key Laboratory of Chemo/Biosensing and Chemometrics, School of Chemistry
and Chemical Engineering, Hunan University, Changsha, 410082, China.
E-mail: liuzhigang@hnu.edu.cn
b
Shenyang National Laboratory for Materials Science Institute of Metal Research,
Chinese Academy of Sciences, Shenyang, 110016, China. E-mail: bszhang@imr.ac.cn
c
Guangdong Provincial Key Laboratory of Mineral Physics and Material Research &
In recent years, carbon based materials have given rise to
much interest due to their desirable physical and chemical pro-
perties including low cost, abundance as well as controllable
Development, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences,
Guangzhou, 510640, China
†
Electronic supplementary information (ESI) available. See DOI: 10.1039/
17,18
c7gc00469a
characteristics and forms.
Among the carbon based
This journal is © The Royal Society of Chemistry 2017
Green Chem.

View Article Online
Paper
Green Chemistry
materials, graphitic nitrogen-doped carbon nanotubes (NCNTs) melamine, and CoCl
2
·6H
2
O. The presence of cobalt nano-
are widely used in various research studies because of their high particles encapsulated into carbon nanotubes has been certi-
catalytic performance, providing huge opportunity for fied by HRTEM, and other characterization data are collected
potential application of heterogeneous catalysis.
functional theory (DFT) revealed that nitrogen doping can tune an excellent catalytic performance in selective oxidation of
a
19–22
Density from SEM, BET, Raman, XRD, and XPS. The Co@GCNs exhibit
23
the charge density, spin density, and energy bandgap.
hydrocarbons under solvent-free and molecular oxygen con-
Magnetic metal elements are extensively introduced in hetero- ditions, and significantly have higher catalytic activity and
geneous catalysis due to their unique electron properties, low selectivity compared to other similar catalysts. To the best of
24–26
cost, and easy recovery and separation.
The magnetic metal our knowledge, in terms of the catalytic oxidation with no
elements can also promote the formation of nitrogen-doped solvent use and molecular oxygen as the oxidant, this work
carbon nanotubes with nitrogen-containing precursors during presents higher catalytic oxidation performance compared to
19,27
the calcination.
Furthermore, metal nanoparticles encapsu- previous work.
lated in nitrogen-doped carbon nanotubes is a novel strategy for
improving the performance of the catalyst. For example, Asefa
and Bao groups have reported that the encapsulation of metal
nanoparticles into nitrogen-doped carbon nanotubes can effec-
Results and discussion
tively improve the catalytic activity of the Hydrogen Evolution Fig. 1 depicts a simple synthetic pathway of the Co@GCNs. In
2
8,29
Reaction (HER).
Deng et al. encapsulated 3d magnetic this experiment, citric acid not only is a carbon source, but
also acts as a chelating agent, which can chelate cobalt ions
metals into nitrogen-doped carbon nanotubes through
a
chemical vapor deposition (CVD) method and thought that the and positively charged melamine to form a homogeneous
desirable catalytic activity originated from the synergistic effects chelate precipitate in aqueous solution through the protonated
28
between metal and nitrogen. The encapsulation of metal nano- carboxyl group. Melamine is easily decomposed at a tempera-
particles is widely researched, however, as we know, the strategy ture beyond 200 °C, and layered graphitic carbon nitride will
is rarely applied to selective catalytic oxidation reactions.
be formed at a temperature up to 550 °C at a slower heating
3
0
Citric acid, an important organic acid compound, is widely rate. Furthermore, the layered graphitic carbon nitride will
used in the food, pharmaceutical, and chemical industries due be quickly decomposed once the temperature is higher than
to its great industrial value and low cost. More especially, citric 600 °C. During the pyrolysis, carbonized citric acid is con-
acid can act as a chelating agent for cationic compounds. densed to a carbon skeleton so as to form carbon nano-
2
7,31
Thus, citric acid can chelate compounds with positive charge tubes.
Therefore, the calcination temperature of catalysts
to form a novel precipitate of chelated composites. However, is varied from 500 °C to 1000 °C. The resulted samples are
although citric acid is common and low price, its use as a che- denoted as Co@GCNs-x, where x represents the calcination
lating agent and carbon source in the preparation of carbon temperature.
material catalysts is scarce.
The morphology of the materials was investigated by scan-
Herein, we report a facile preparation approach in which ning electron microscopy (SEM). As shown in Fig. 2, a bulk gra-
novel cobalt nanoparticles were encapsulated into graphitic phitic carbon nitride structure has been formed after heating
3
2
nitrogen-doped carbon nanotube hybrids (Co@GCNs) through at 500 °C (Fig. 2a). Co@GCNs-600 shows a rod like structure
pyrolysis of chelate compounds derived from one-pot method in which many small holes and long-channels are discovered
self-assembly of inexpensive original materials of citric acid, (Fig. 2b). What’s more, when the temperature is higher than
Fig. 1 Schematic illustration showing the general procedure for the synthesis of Co@GCNs.
Green Chem.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Green Chemistry
Paper
carbon nanotubes indicates that the diameter is mainly
located at 30–50 nm (Fig. S1†). It is interesting that white dots
are found at the end of nitrogen-doped carbon nanotubes. The
microstructure of Co@GCNs-800 was observed through high-
resolution transmission electron microscopy (HRTEM).
According to the HRTEM images (Fig. 3a–c), the nitrogen-
doped carbon nanotubes are formed on the carbonaceous sub-
strate presenting a bamboo-like architecture. The thickness of
the wall is 4–5 nm. The metal nanoparticles with an average
size of about 21 nm are encapsulated into nitrogen-doped
carbon nanotubes, and some metal particles are encapsulated
at the end of carbon nanotubes. So the white dots seen by
SEM are the metal particles located at the end of nitrogen-
doped carbon nanotubes. However, there are no carbon nano-
tubes formed in the Co/C-800 derived from only citric acid as a
carbon source (Fig. S2b†), and the average particle size is
ca. 81 nm, suggesting that the nitrogen-doped carbon nanotubes
can encapsulate metal particles and hinder the growth. The
lattice spacing of metal particles is measured to be 0.18 nm
1
5
coincided with that of cubic Co and the (111) plane (Fig. 3d).
The elemental mappings show that N and O elements are
more along with the distribution of Co elements, it may be
due to the formation of some compounds among Co, N, and O
elements such as Co–N, Co–O and so on (Fig. 3e). Moreover,
the material without a metal exhibits thick graphite layers
Fig. 2 SEM images of (a) Co@GCNs-500, (b) Co@GCNs-600, (c)
Co@GCNs-700, (d) Co@GCNs-800, and (e, f) Co@GCNs-900.
(
Fig. S2a†), suggesting that metal nanoparticles can positively
promote the formation of nitrogen-doped carbon nanotubes.
The X-ray diffraction (XRD) patterns of the Co@GCNs cal-
6
00 °C, the materials begin to generate nitrogen-doped carbon cined at different temperatures are presented in Fig. 4. A weak
nanotubes. With the increase of temperature, more and more peak at 26° is attributed to the (002) plane of graphitic carbon.
carbon nanotubes are formed and the morphology becomes Besides, three pronounced peaks at 44.2°, 51.5°, and 75.9° are
more and more regular (Fig. 2c–f). The size distribution of ascribed to the (111), (220), and (220) reflections of cubic Co
Fig. 3 HRTEM images (a–d), HAADF image and elemental mappings (e) of Co@GCNs-800.
This journal is © The Royal Society of Chemistry 2017
Green Chem.

View Article Online
Paper
Green Chemistry
Fig. 4 XRD patterns of Co@GCNs-500, Co@GCNs-600, Co@GCNs-
7
00, Co@GCNs-800, and Co@GCNs-900.
1
5
(
fcc, Fm3m (225), JCPDS 15-0806), respectively. The weak
broad peak of graphitic carbon is found only in Co@GCNs-500
and Co@GCNs-600, but the others are found with narrow
peaks, which may be the existence of a lower crystallization
degree at low temperature. And the diffraction peak at 44.2° is
apparently more intense with the increase of temperature,
indicating that the crystallization of the metallic Co phase is
higher at elevated temperature. Raman spectroscopy was per-
formed to observe the electronic and structural nature of the
materials (Fig. S3†). Apparently, the D band locates at
Fig. 5 Nitrogen adsorption–desorption isotherms (a) and pore size dis-
350 cm− , which is derived from the disorder induced fea- tribution (b) of Co@GCNs-600, Co@GCNs-700, Co@GCNs-800, and
1
1
−
1
Co@GCNs-900.
tures caused by lattice defects, and the G band at 1587 cm
represents the in-plane vibrational modes of sp carbon atoms
2
3
3
in graphene. The relative ratio of the D band to G band is
more and more small along with the increase of temperature, the formation of nitrogen-doped carbon nanotubes at elevated
suggesting that a higher graphite degree is formed. Moreover, temperature. Interestingly, the pore size distribution based on
−
1
a
weak peak at ca. 670 cm
is assigned to cobalt the Barrett–Joyner–Halenda (BJH) formula demonstrates that
3
4
nanoparticles.
the carbon nitride composites and nitrogen-doped carbon
Nitrogen adsorption–desorption isotherms were obtained nanotubes have uniform mesopores of 3.7 nm (Fig. 5b and
to investigate the textural properties of Co@GCNs. The Table 2). Co/C-800 and MF-800 display a smaller specific
isotherms with a large hysteresis loop in the relative pressure surface area (Fig. S4a and b†), indicating that melamine and
(
0
P/P ) range of 0.4–1.0 show a type IV curve according to the cobalt play an important role in preparing nitrogen-doped
IUPAC nomenclature (Fig. 5a), corresponding to the character- carbon nanotubes. The specific surface area of Co@GCNs-800
istics of mesoporous materials with narrow pore size distri- after acid treatment is significant larger than Co@GCNs-800
bution. The large hysteresis loop suggests that a capillary (Fig. S4c†), which is due to the release of space occupied by
condensation phenomenon has occurred in the mesoporous the metal.
structure. The specific surface area and total pore volume of
To further investigate the elemental compositions of
2
−1
the materials are calculated to be as high as 232 m g and Co@GCN catalysts, X-ray photoelectron spectroscopy (XPS) was
3
−1
0
.37 cm
g
(Table 2), respectively. However, Co@GCNs- carried out. It can be seen from the full survey spectrum that
5
00 has a very small specific surface area, which should be due the main elements are C, N, O, and Co in the Co@GCNs
3
2
to the formation of dense graphitic carbon nitride.
(Fig. S5†). The high-resolution C 1s XPS spectrum of
Co@GCNs-600 shows a higher specific surface area, originat- Co@GCNs-800 reveals CvC, CvN & C–O, and C–N & C–O–C at
3
5,36
ing from the porous and multi-channel structure in the rod 284.1 eV, 285.5 eV, and 286.9 eV, respectively (Fig. S6a†).
solids as evidenced in Fig. 2b. For the other materials, higher The high-resolution O 1s spectrum of Co@GCNs-800 is de-
specific surface area and total pore volume are attributed to convoluted into three peaks with binding energies of 529.7 eV,
Green Chem.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Green Chemistry
Paper
5
31.3 eV, and 532.4 eV, which can be ascribed to C–OH and/or using a certain pressure of molecular oxygen as an oxidant.
C–O–C groups, CvO (aldehydes, ketones, lactones), and the The results of catalytic reactions are summarized in Table 2.
3
7,38
oxygen related to cobalt oxides, respectively (Fig. S6b†).
As The catalytic properties of these catalysts for the aerobic oxi-
shown in Fig. 6a, two main peaks are located at 780 and 796 dation of ethylbenzene are determined in a high pressure reac-
eV corresponding to the Co 2p3/2 and Co 2p1/2 binding energy, tion kettle with 0.8 MPa O2 at 120 °C holding for 5 hours.
0
respectively. The high-resolution of Co 2p3/2 is fitted into Co , Among the various temperatures of Co@GCNs, Co@GCNs-500
3
+
2+
Co , Co , and satellites peaks. The peaks at 777.9 and 794.0 and Co@GCNs-600 show a similar and low conversion ca. 23%
0
eV are characteristic of Co in the Co 2p. It is worth noting that (Table 2, entries 4 and 5). However, compared with Co@GCNs-
the content of cobalt is lower when the temperature is higher 500 and Co@GCNs-600, the other catalysts have significantly
than 600 °C, it may be due to the formation of nitrogen-doped improved catalytic performance for the oxidation of ethyl-
carbon nanotubes that encapsulate metal nanoparticles and benzene. Co@GCNs-700, Co@GCNs-800, Co@GCNs-900, and
0
make them disperse more evenly (Table 2). The content of Co
Co@GCNs-1000 all display conversion higher than 60% and
is significantly higher than that of the catalysts synthesized at selectivity higher than 90% (Table 2, entries 6–9).
lower temperature along with the increase of temperature, and In order to thoroughly investigate the influence of catalysts,
the nitrogen-doped carbon nanotubes begin to form at elev- some control group experiments are performed under the
ated temperatures. Fig. 6b shows the N 1s spectra of catalysts, same conditions. Obviously, the catalytic activity of the blank
the N 1s can be deconvoluted into four peaks with the binding test without any catalyst is well controlled, and only 4.3% and
energies of 398.5, 400.3, 401.0 and 203.6 eV in agreement with 59.0% of ethylbenzene conversion and acetophenone selecti-
the generally given range of binding energies of about 398.6, vity are obtained (Table 2, entry 1), respectively. Furthermore,
4
00.5, 401.3, and 403–404 eV, which are attributed to pyridinic the pristine Co and CNTs are also examined and present rela-
N, pyrrolic N, graphitic N, and N-oxide species, respect- tively low catalytic activity (Table 2, entries 2 and 3). To study
3
9,40
ively.
Pyridinic N is the dominant phase in the Co@GCNs- the influence of metal species, cobalt has been substituted by
5
00 and Co@GCNs-600, but graphitic N becomes the domi- iron and nickel under the same conditions. It is amazing that
nant phase since the temperature is higher than 700 °C and iron and nickel catalysts exhibited poor effects in comparison
the pyrrolic N peak is disappearing, which is due to the conver- with cobalt catalysts (Table 2, entries 10 and 11). As the non-
sion of pyridinic and pyrrolic N to graphitic N and the for- cobalt metal catalysts show low catalytic activity, the metal-free
mation of nitrogen-doped carbon nanotubes.
catalyst prepared in the absence of a cobalt precursor shows a
The selective oxidation of ethylbenzene was chosen as the low catalytic activity (Table 2, entry 12). The Co@GCNs-800-AT,
model reaction and carried out under solvent-free conditions derived from the treatment of Co@GCNs-800 with sulfuric
Fig. 6 High-resolution Co 2p (a) and N 1s (b) XPS spectra of Co@GCNs-500, Co@GCNs-600, Co@GCNs-700, Co@GCNs-800, and Co@GCNs-900.
This journal is © The Royal Society of Chemistry 2017
Green Chem.

View Article Online
Paper
Green Chemistry
acid, shows some activity in the oxidation of ethylbenzene tions (Fig. S8†). The Co@GCNs-800 after using 6 times retains
Table 2, entry 13), which demonstrates that the cobalt species the carbon nanotube structure with encapsulation of metal
(
may be the active sites in reaction. The opinion that nitrogen- nanoparticles (Fig. S9†), suggesting that there is a stable nitro-
containing carbon materials hold higher catalytic activity and gen-doped carbon nanotube structure in this catalyst.
stability for some catalytic oxidation reactions than that of However, the catalytic activity almost is recovered to the first
non-nitrogen carbon materials has been reported by many level when the catalyst after the recycling 6 times is reheated
1
4,23,41,42
research groups.
a carbon resource possesses higher catalytic performance com- reduction of oxidized metallic cobalt at high temperature.
pared with other control groups, but it is still inferior to the Combination of the above characterization and the catalytic
The Co/C-800 with only citric acid as under the same conditions. This phenomenon is due to the
nitrogen-containing catalysts (Table 2, entry 14), demonstrat- reaction results suggests that metallic cobalt is the active site
ing that nitrogen species have an important role in enhancing for reactions. Nevertheless, there is a small amount of metallic
the conversion of ethylbenzene. These control experiments cobalt in Co@GCNs-600 (Table 1), but the catalytic activity of
indicate that the presence of a synergistic interaction between Co@GCNs-600 is not significantly improved compared with
cobalt and nitrogen-doped carbon nanotubes plays a major that of Co@GCNs-500 without metallic cobalt (Table 2, entries
role in affecting the catalytic activity of Co@GCNs-800 in the 4 and 5). Moreover, the pristine cobalt does not exhibit an
selective oxidation of ethylbenzene.
expected catalytic effect (Table 2, entry 2). Thus, this finding
In addition to the discussion of the conversion and selecti- demonstrates that the metallic cobalt is not the only activity
vity, the stability of Co@GCNs-800 is also investigated. As seen site for catalytic reactions. Furthermore, there is no significant
in Fig. S7,† the activity decreases obviously after using two change of metallic cobalt content when the temperature
times, and finally stabilizes at ca. 25%, this is attributed to the higher than 700 °C, indicating that although high temperature
reason that metallic cobalt is severely oxidized during reac- is conducive to the reduction of cobalt, the reduction effect is
Table 1 Physical characterization and chemical compositions of various Co@GCN catalysts
a
Content (at%)
BET (m g−1)
2
V
total (cm g−1)
3
D
(nm)
C
N
O
Co
Co
0 b
Sample
S
p
Co@GCNs-500
Co@GCNs-600
Co@GCNs-700
Co@GCNs-800
Co@GCNs-900
<10
296
137
232
226
—
—
63.19
68.67
89.72
91.58
93.17
29.84
19.11
3.70
3.62
2.54
4.00
7.72
3.71
3.30
3.01
2.97
4.5
2.87
1.50
1.28
—
0.20
0.30
0.37
0.34
3.7
3.7
3.7
3.7
8.06
21.19
26.97
22.57
a
b
0
The contents of various elements were measured by X-ray photoelectron spectroscopy. The relative content of Co in the Co 2p.
Table 2 Catalytic activity of different catalysts for the oxidation of ethylbenzene under O
2
as an oxidant and solvent-free conditions
a
Sel. [%]
a
Entry
Catalyst
Conv. [%]
1
2
3
4
5
6
7
8
9
1
1
1
1
1
—
Co
CNTs
4.3
22.2
12.7
22.5
23.7
65.4
68.1
62.3
62.4
16.2
12.4
12.5
30.6
49.7
59.0
80.9
87.4
81.0
77.8
92.3
93.2
91.3
90.6
72.8
83.0
78.9
82.3
84.7
34.5
17.0
12.0
17.6
21.5
7.7
6.8
8.7
9.3
24.8
17.0
18.0
16.9
14.8
6.5
2.1
0.6
1.4
0.7
—
Co@GCNs-500
Co@GCNs-600
Co@GCNs-700
Co@GCNs-800
Co@GCNs-900
Co@GCNs-1000
Fe/CN-800
Ni/CN-800
MF-800
—
—
0.1
2.4
—
3.1
0.8
0.5
0
1
2
3
4
Co@GCNs-800-AT
Co/C-800
a
Reaction conditions: ethylbenzene (10 mL), catalyst (26 mg), O
2
(0.8 MPa), 120 °C, and 5 h. Conversion and selectivity were determined by GC
(dual internal standard: 1,4-dichlorobenzene and bromobenzene).
Green Chem.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Green Chemistry
Paper
not proportional to the content of metallic cobalt. The intro- Table 3 Aerobic oxidation of arylalkanes without solvent catalyzed by
a
Co@GCNs-800
duction of nitrogen can facilitate the formation of nitrogen-
doped carbon nanotubes and is beneficial to the encapsula-
tion and dispersion of cobalt nanoparticles (Fig. 3), giving rise
to the improvement of catalytic activity to some degree
Entry Substrate
1
Product
Conv. [%] Sel. [%]
78.6
73.7
89.3
92.2
(Table 2, entry 14). According to the poor activity of pristine
cobalt, carbon nanotubes as well as Co@GCNs-600 (Fig. S10†),
it further demonstrates that the co-existence of metallic cobalt
and nitrogen-doped carbon nanotubes can exert an excellent
catalytic effect for the selective aerobic oxidation of ethyl-
benzene. The special structure of nitrogen-doped carbon nano-
tubes promotes the transfer of substrate molecules in contact
with active sites.
2
3
4
5
55.0
69.2
67.8
82.8
63.0
62.0
Since Co@GCNs-800 displays the best activity for the selec-
tive oxidation of ethylbenzene, it is deployed in the scope
study under the same conditions. As shown in Table 3, it is
seen that Co@GCNs-800 shows excellent activity for selectivity
aerobic oxidation of arylalkanes to ketones. Arylalkanes con-
taining electron donating and withdrawing functional groups
show excellent conversion and selectivity to the corresponding
ketones (Table 3, entries 1 and 2). The Co@GCNs-800 also
exhibits superior catalytic performance for selective oxidation
of n-propylbenzene, cumene, tetralin, indane, and diphenyl-
methane (Table 3, entries 3–7). Therefore, it is reasonable to
say that Co@GCNs-800 is a general catalyst for selective oxi-
dation of hydrocarbons under mild reaction conditions. In
addition, several reported works about ethylbenzene aerobic
6
7
70.7
78.2
87.6
95.0
4
3–45
oxidation are compared with this work.
Notably, the
conversion 68.1% and selectivity 93.2% of Co@GCNs-800
are significantly higher than those of Co-KZrSBA-15 and
x
Ce0.5Mn0.5O @500 (Table 3, entries 8 and 9). Our groups have
reported a catalyst of CoNCB-2, which exhibits excellent cata-
lytic activity (Table 3, entry 10). But Co@GCNs-800 still pos-
sesses higher conversion and selectivity under the same reac-
tion conditions to CoNCB-2.
To gain insights into the mechanism of Co@GCN catalyzed
2
oxidation of hydrocarbons without solvent and using O as an
oxidant, the oxidation of ethylbenzene in the presence of buty-
lated hydroxytoluene (BHT) which is a kind of radical scaven-
ger was tested (Table 3, entry 11). Notably, no acetophenone
b
8
29.4
20.3
60.0
—
71.2
87.0
86.0
—
c
9
(AcPO) is produced in the presence of BHT, indicating that the
d
10
reactions progress through radicals, e.g. the superoxide radical
•
−
2 2
anion ( O ). As shown in Fig. 7, in the first step, O gets elec-
•
−
trons to form O with the help of a catalyst, and ethylbenzene
2
is absorbed onto the surface of catalyst and activated to form
e
4
6,47
11
the phenylethyl radical (PE).
During this process, the met-
allic cobalt donates electrons to form positively charged cobalt
ions as evidenced in Fig. S8.† Then, the PE quickly reacts with
•
−
11
O
2
to form the 1-phenylethylperoxyl radical (PEPO). Next,
a
the PEPO is able to weaken the α-H bond of ethylbenzene,
causing the formation of 1-phenyl-ethylhydroperoxide (PEHP)
that is the main chain propagator and the most important
2
Reaction conditions: substrate (10 mL), catalyst (26 mg), O (0.8
MPa), 120 °C, 5 h; the conversion and selectivity were determined by
GC. Ref. 43: catalyst of Co-KZrSBA-15, reaction conditions: ethyl-
benzene (10 mL), catalyst (20 mg), O
b
c
2
(1 MPa), 120 °C, 6 h. Ref. 44: cata-
1
1
intermediate in this process and PE. In the fourth step, a lyst of Ce0.5Mn0.5
O @500, reaction conditions: ethylbenzene (1 mmol),
x
d
catalyst (30 mg), CH
lyst of CoNCB-2, reaction conditions: ethylbenzene (10 mL), catalyst
3 2
CN (5 mL), O (1 MPa), 120 °C, 6 h. Ref. 45: cata-
reaction the same as the first step will take place and cycle.
And in the fifth step, it is believed that the catalyst is favorable
e
2
(26 mg), O (0.8 MPa), 120 °C, 5 h. Reaction conditions are consistent
for the activation of the –OH bond of PEHP, leading to easier with this work beside adding butylated hydroxytoluene (BHT) (1 mL).
This journal is © The Royal Society of Chemistry 2017
Green Chem.

View Article Online
Paper
Green Chemistry
2
Fig. 7 Proposed mechanism for the ethylbenzene oxidation over the catalysts using O as an oxidant.
reaction with PEPO to generate AcPO and higher AcPO
percentage corresponding to the experimental results.
Simultaneously, the PEHP reacts with PE to form phenethyl
Experimental
Materials
4
3
alcohol (PEA) and 1-phenylethyloxy radical (PEO), however, the Citric acid, melamine, CoCl ·6H O, NiCl ·6H O, FeCl ·6H O,
2
2
2
2
3
2
reaction rate of the latter is much lower than the former result- Co, and all solvents were of commercially available analytical
ing in the low selectivity of PEA, in accordance with the experi- grade and used without further purification. The CNTs were
4
3,48
mental results.
Finally, a negligible benzaldehyde (BA) is purified with HNO in order to remove the impurities.
3
4
8
produced due to the β-C–C cleavage of PEO.
Catalyst preparation
Synthesis of Co@GCNs. A mixture including citric acid
1.8 g), melamine (1.2 g), CoCl ·6H O (0.6 g), and deionized
2 2
Conclusions
(
In summary, we have shown the facile synthesis of cobalt water (50 mL) was heated to 100 °C and stirred for 3 hours.
nanoparticles encapsulated into graphitic nitrogen-doped During this process, more and more precipitate was generated
carbon nanotube materials (Co@GCNs) via one-pot thermal and the color of the precipitate changed from red to white.
decomposition of chelate composites comprising citric acid, After cooling down to room temperature, the obtained white
2 2
melamine, and CoCl ·6H O. The Co@GCN composites show precipitate was filtered and washed several times with de-
excellent catalytic performance for selective oxidation of ethyl- ionized water, and then dried overnight at 80 °C in an oven.
benzene under solvent-free and molecular oxygen conditions. The dried solid was finely ground into powder and transferred
−
1
Via a series of controlled group experiments and characterization into a crucible, directly heated at a rate of 3 °C min to
results such as HRTEM, XRD, XPS, and so on, it is found that 800 °C, and kept for 2 h at this temperature under N . After
2
the catalysts hold higher catalytic activity when metallic cobalt that, the sample was cooled to ambient temperature, and the
and nitrogen-doped carbon nanotubes co-exist in the system. black product was collected and ground. The sample was
Thus, the superior catalytic activity is assigned to the synergistic denoted as Co@GCNs-800.
effect between metallic cobalt and nitrogen-doped carbon nano-
For comparison, the samples calcined at various tempera-
tubes. In addition, the catalyst also presents high activity for the tures were denoted as Co@GCNs-500, Co@GCNs-600,
oxidation of other arylalkanes under the same conditions. The Co@GCNs-700, Co@GCNs-900, and Co@GCNs-1000. In the
proposed mechanism of oxidation reactions is discussed and light of a similar preparation method, the samples in which
2 2 3 2
proved to be free radical reactions. This work may provide a metal salt was substituted by NiCl ·6H O and FeCl ·6H O were
simple, effective, economical, and environmentally friendly cata- denoted as Ni@CTNs-800 and Fe@CNTs-800, respectively. The
lytic system for the selective oxidation of hydrocarbons.
sample without a nitrogen source of melamine was named Co/
Green Chem.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Green Chemistry
Paper
C-800. The control sample without metal ions was called
metal-free (MF-800). And the Co@GCNs-800 treated with
Notes and references
0
.5 M H
2
SO
4
at 90 °C for 4 h was denoted as acid treated
1 A. E. Shilov and G. B. Shul’pin, Chem. Rev., 1997, 97, 2879–
(Co@GCNs-800-AT).
2932.
2
L. Kesavan, R. Tiruvalam, M. H. Ab Rahim, M. I. bin
Saiman, D. I. Enache, R. L. Jenkins, N. Dimitratos,
J. A. Lopez-Sanchez, S. H. Taylor, D. W. Knight, C. J. Kiely
and G. J. Hutchings, Science, 2011, 331, 195–199.
J. Zhang, X. Liu, R. Blume, A. H. Zhang, R. Schlögl and
D. S. Su, Science, 2008, 322, 73–77.
D. Su, J. Wang, H. Jin, Y. Gong, M. Li, Z. Pang and Y. Wang,
J. Mater. Chem. A, 2015, 3, 11756–11761.
D. Deng, Y. Yang, Y. Gong, Y. Li, X. Xu and Y. Wang, Green
Chem., 2013, 15, 2525–2531.
Catalytic reaction
The selective oxidation of hydrocarbons was carried out in a
5
0 mL stainless steel autoclave lined with Teflon. Typically,
3
4
5
6
7
substrates (10 mL) and catalyst (26 mg) were added into the
reactor and filled with molecular oxygen (0.8 MPa). The reac-
tion was kept at 120 °C for 5 h with continuous magnetic stir-
ring at 900 rpm. After reaction, the catalyst was separated and
recovered via centrifugation, and the dual internal standard of
1
,4-dichlorobenzene and bromobenzene as well as the as-
R. P. Doherty, J.-M. Krafft, C. Méthivier, S. Casale, H. Remita,
C. Louis and C. Thomas, J. Catal., 2012, 287, 102–113.
T. Maegawa, A. Akashi, K. Yaguchi, Y. Iwasaki,
M. Shigetsura, Y. Monguchi and H. Sajiki, Chem. – Eur. J.,
obtained clarified reaction mixture were added into the absol-
ute alcohol solvent. Finally, the products were analyzed quanti-
tatively by gas chromatography (GC) with a HP-5 ms capillary
column (30 m, DF = 0.25 mm, 0.25 mm i.d.). The recovered
catalyst via separation was washed three times with absolute
alcohol, and then dried in a vacuum. The recyclability of the
catalysts was investigated for the selective oxidation ability
maintaining the same reaction conditions as described above.
The conversion is the average results of three experiments, and
error was controlled within 1%.
2
009, 15, 6953–6963.
S. Liao, Y. Chi, H. Yu, H. Wang and F. Peng, ChemCatChem,
014, 6, 555–560.
S. Liao, F. Peng, H. Yu and H. Wang, Appl. Catal., A, 2014,
78, 1–8.
0 J. Luo, H. Yu, H. Wang and F. Peng, Catal. Commun., 2014,
1, 77–81.
8
9
2
4
1
1
5
1 J. Luo, F. Peng, H. Yu, H. Wang and W. Zheng,
ChemCatChem, 2013, 5, 1578–1586.
Characterization
Field emission scanning electronic microscopy (SEM) was per- 12 M. Jafarpour, A. Rezaeifard, V. Yasinzadeh and H. Kargar,
formed on a JSM-6700F microscope to visualize the mor-
RSC Adv., 2015, 5, 38460–38469.
phology and size of the catalysts. High-resolution transmission 13 A. Chen, Y. Yu, R. Wang, Y. Yu, W. Zang, P. Tang and
electron microscopy (HRTEM) was carried out on a Tecnai G2
D. Ma, Nanoscale, 2015, 7, 14684–14690.
F20 S-TWIN instrument operated at 120 kV and a Philips 14 S. Yang, L. Peng, P. Huang, X. Wang, Y. Sun, C. Cao and
CM200 FEG instrument operated at 200 kV. Brunauer–
W. Song, Angew. Chem., Int. Ed., 2016, 128, 4084–4088.
Emmett–Teller (BET) specific surface areas (SBET) and the total 15 C. Bai, A. Li, X. Yao, H. Liu and Y. Li, Green Chem., 2016,
pore volume (Vtotal) were measured by N adsorption–desorp-
18, 1061–1069.
tion analysis at 77 K and a relative pressure P/P of 0.98 in a 16 W. Zhong, H. Liu, C. Bai, S. Liao and Y. Li, ACS Catal.,
2
0
NOVA 1000e from Quantachrome Instruments. And the pore
size (D ) distribution was calculated by the Barrett–Joyner– 17 L. L. Zhang and X. S. Zhao, Chem. Soc. Rev., 2009, 38, 2520–
Halenda (BJH) formula. The X-ray diffraction (XRD) pattern
2531.
was collected on a Japan XRD-6100 diffractometer with the 2θ 18 W. J. Lee, U. N. Maiti, J. M. Lee, J. Lim, T. H. Han and
scan range between 10° and 80° using Ni-filtered Cu Kα radi-
S. O. Kim, Chem. Commun., 2014, 50, 6818–6830.
ation (50 kV, 10 mA). Raman spectra were obtained using a 19 C. H. Choi, S. H. Park and S. I. Woo, ACS Nano, 2012, 6,
33 nm laser on a LabRAM Aramis micro Raman spectrometer
7084–7091.
at room temperature. The surface composition was determined 20 P. Su, H. Xiao, J. Zhao, Y. Yao, Z. Shao, C. Li and Q. Yang,
by X-ray photoelectron spectroscopy (XPS) using a PHI 5000C
Chem. Sci., 2013, 4, 2941–2946.
ESCA system (PerkinElmer) with an Al Kα (1486.6 eV) X-ray 21 J. Shui, M. Wang, F. Du and L. Dai, Sci. Adv., 2015, 1,
2015, 5, 1850–1856.
p
6
source, and the C 1s line (284.4 eV) was used as the reference
e1400129.
to correct the binding energies (BE).
22 J. Wu, R. M. Yadav, M. Liu, P. P. Sharma, C. S. Tiwary,
L. Ma, X. Zou, X.-D. Zhou, B. I. Yakobson, J. Lou and
P. M. Ajayan, ACS Nano, 2015, 9, 5364–5371.
2
3 S. Wang, L. Zhang, Z. Xia, A. Roy, D. W. Chang, J.-B. Baek
and L. Dai, Angew. Chem., Int. Ed., 2012, 51, 4209–4212.
Acknowledgements
The authors thank the State Key Laboratory of Heavy Oil 24 T. Cheng, H. Yu, F. Peng, H. Wang, B. Zhang and D. Su,
Processing in China (no. SKCHOP201504) and the Key Catal. Sci. Technol., 2016, 6, 1007–1015.
Laboratory of Mineralogy and Metallogeny in Chinese 25 Y. Chen, S. Zhao and Z. Liu, Phys. Chem. Chem. Phys., 2015,
Academy of Sciences (no. KLMM20150103).
17, 14012–14020.
This journal is © The Royal Society of Chemistry 2017
Green Chem.

View Article Online
Paper
Green Chemistry
26 J.-M. Yan, X. B. Zhang, T. Akita, M. Haruta and Q. Xu, 37 H. T. Chung, J. H. Won and P. Zelenay, Nat. Commun.,
J. Am. Chem. Soc., 2010, 132, 5326–5327. 2013, 4, 1922–1926.
27 J. Wang, Z. Wei, Y. Gong, S. Wang, D. Su, C. Han, H. Li and 38 H.-W. Liang, W. Wei, Z.-S. Wu, X. Feng and K. Müllen,
Y. Wang, Chem. Commun., 2015, 51, 12859–12862. J. Am. Chem. Soc., 2013, 135, 16002–16005.
28 J. Deng, P. Ren, D. Deng, L. Yu, F. Yang and X. Bao, Energy 39 L. He, F. Weniger, H. Neumann and M. Beller, Angew.
Environ. Sci., 2014, 7, 1919–1923. Chem., Int. Ed., 2016, 55, 12582–12594.
29 X. Zou, X. Huang, A. Goswami, R. Silva, B. R. Sathe, 40 Z. Ma, H. Zhang, Z. Yang, G. Ji, B. Yu, X. Liu and Z. Liu,
E. Mikmeková and T. Asefa, Angew. Chem., Int. Ed., 2014,
26, 4461–4465.
0 L. Liu, Y. Qi, J. Lu, S. Lin, W. An, Y. Liang and W. Cui, Appl.
Catal., B, 2016, 183, 133–141.
1 X.-H. Li, S. Kurasch, U. Kaiser and M. Antonietti, Angew.
Chem., Int. Ed., 2012, 51, 9689–9692.
2 X. Lin, S. Zhao, Y. Chen, L. Fu, R. Zhu and Z. Liu, J. Mol.
Catal. A: Chem., 2016, 420, 11–17.
Green Chem., 2016, 18, 1976–1982.
41 L. Chen, B. Huang, X. Qiu, X. Wang, R. Luque and Y. Li,
Chem. Sci., 2016, 7, 228–233.
42 H. Liu, L. Chang, C. Bai, L. Chen, R. Luque and Y. Li,
Angew. Chem., Int. Ed., 2016, 55, 5019–5023.
43 S. Shi, C. Chen, M. Wang, J. Ma, J. Gao and J. Xu, Catal. Sci.
Technol., 2014, 4, 3606–3610.
44 P. Zhang, H. Lu, Y. Zhou, L. Zhang, Z. Wu, S. Yang, H. Shi,
Q. Zhu, Y. Chen and S. Dai, Nat. Commun., 2015, 6, 8446–
8455.
1
3
3
3
3
3 N. L. Torad, M. Hu, S. Ishihara, H. Sukegawa, A. A. Belik,
M. Imura, K. Ariga, Y. Sakka and Y. Yamauchi, Small, 2014,
1
0, 2096–2107.
45 Y. Chen, L. Fu, Z. Liu and Y. Wang, ChemCatChem, 2016, 8,
1782–1787.
3
4 M. Galaburda, V. Bogatyrov, O. Oranska, V. Gun’ko,
J. Skubiszewska-Zięba and I. Urubkov, J. Therm. Anal. 46 X. Cui, Y. Li, S. Bachmann, M. Scalone, A.-E. Surkus,
Calorim., 2015, 122, 553–561.
5 H. Wang, Y. Zhao, F. Cheng, Z. Tao and J. Chen, Catal. Sci.
Technol., 2016, 6, 3443–3448.
6 Y. Su, Y. Zhu, H. Jiang, J. Shen, X. Yang, W. Zou, J. Chen
and C. Li, Nanoscale, 2014, 6, 15080–15089.
K. Junge, C. Topf and M. Beller, J. Am. Chem. Soc., 2015,
137, 10652–10658.
47 D. Bégin, G. Ulrich, J. Amadou, D. S. Su, C. Pham-Huu and
R. Ziessel, J. Mol. Catal. A: Chem., 2009, 302, 119–123.
48 U. Neuenschwander and I. Hermans, J. Catal., 2012, 287, 1–4.
3
3
Green Chem.
This journal is © The Royal Society of Chemistry 2017