Nanodiamond@carbon nitride hybrid with loose porous carbon nitride layers as an efficient metal-free catalyst for direct dehydrogenation of ethylbenzene

Article abstract of DOI:10.1016/j.apcata.2018.12.016

This work presents a facile and low-cost approach to prepare nanodiamond@carbon nitride hybrid with loose porous carbon nitride layers close wrapped on well dispersed NDs (ND-CN) through the thermal polymerization of the mixture composed of hexamethylenet

Full text of DOI:10.1016/j.apcata.2018.12.016

AppliedCatalysisA,General571(2019)82–88
Contents lists available at ScienceDirect
Applied Catalysis A, General
journal homepage: www.elsevier.com/locate/apcata
Nanodiamond@carbon nitride hybrid with loose porous carbon nitride
layers as an efficient metal-free catalyst for direct dehydrogenation of
ethylbenzene
Guifang Ge, Zhongkui Zhao⁎
State Key Laboratory of Fine Chemicals, Department of Catalysis Chemistry and Engineering, School of Chemical Engineering, Dalian University of Technology, 2 Linggong
Road, Dalian, 116024, China
A R T I C L E I N F O
A B S T R A C T
Keywords:
Hybrid
Nanodiamond
Dispersant and precursor
Direct dehydrogenation
Defects and ketonic carbonyl groups
This work presents a facile and low-cost approach to prepare nanodiamond@carbon nitride hybrid with loose
porous carbon nitride layers close wrapped on well dispersed NDs (ND-CN) through the thermal polymerization
of the mixture composed of hexamethylenetetramine (HTM) and ammonium chloride adsorbed on dispersed
nanodiamonds (ND), in which hexamethylenetetramine serves as both carbon nitride precursor and dispersing
agent and the ammonium chloride acts as gas template. ND-CN shows 1.9 and 1.2 times much higher styrene rate
for oxygen and steam-free direct dehydrogenation of ethylbenzene than the pristine ND and the previously
reported carbon nitride layers encapsulated ND (H-ND) formed by the thermal polymerization of mixture
composed of HTM and ND, respectively, ascribed the more exposed active sites containing structural defects and
ketonic carbonyl groups and the not weakened synergistic effect of carbon nitride with ND that originating from
the enlarging surface area and pore volume owing to the characteristic concerning loose porous structure of
carbon nitride layers.
1. Introduction
nitride layer (H-ND) [16]. ND powder and hexamethylenetetramine
were milled to be well mixed, and then to be pyrolyzed. The H-ND
Nanocarbon materials have attracted continuous interest in many
fields including batteries, electrochemical energy storage and catalysis
[1–3], due to their good corrosion resistance, stable structure and ex-
cellent surface chemistry [4,5]. Among which, nanodiamond (ND) is a
distinctive material for its unique sp2-sp3 hybridized structure, extreme
hardness and chemical stability [6], which bring up its wide application
in drug delivery, electrocatalyst and metal-free catalysis [7–9]. And it
was demonstrated that ND can be employed for oxygen and steam-free
direct dehydrogenation (DDH) of ethylbenzene to styrene, Su and co-
workers have revealed the much superior catalytic activity and great
potential of ND regarding of direct dehydrogenation of ethylbenzene
than that of the commercial potassium-iron catalyst for industrial ap-
plication [10]. A number of nanostructured carbon materials such as
nanodiamond/CNT-SiC [11], nanodiamond/nitrogen-doped meso-
porous carbon [12], nitride‐doped carbon nanotube [13], and nano-
diamond/grapheme [14,15] have previously been shown to be effective
in enhancing the DDH of ethylbenzene in recent years. Our group
(Advanced Catalytic Material Research Group), DUT, have developed a
two-step method to synthesize a ND closely wrapped with carbon
displayed a higher and more stable activity than M-ND manufactured
from ND with melamine [17] for steam-free DDH of ethylbenzene to
styrene. The existence of synergistic effect between ND and carbon
nitride accounted for the good DDH performance. However, a large
amount of active sites of ND was also closely-wrapped and covered by
carbon nitride layer and that depressed catalytic performance. There-
fore, we envision that the ND-CN hybrid with loose and porous carbon
nitride layers closely encapsulating on ND surface may exhibit out-
standing catalytic performance in DDH of ethylbenzene owing to the
maintained strong synergistic effect of ND can carbon nitrides but not
decreasing the accessibility of active sites.
Recent studies have demonstrated that the surface structural defects
and ketonic carbonyl groups are active sites for CeH activation [18,19].
Surface defects not only endow the materials with more exposed active
edges but also provide a large number of boundaries thus reduce ag-
gregation. On the other hand, it is necessary to generate more defects
for the sake of higher catalytic performance. A variety of tactics in-
cluding alkali-assisted route [20,21], thermal polymerization with
precursors [22] and gas etching [23] were utilized to yield defects.
⁎ Corresponding author.
E-mail address: zkzhao@dlut.edu.cn (Z. Zhao).
https://doi.org/10.1016/j.apcata.2018.12.016
Received 7 September 2018; Received in revised form 13 December 2018; Accepted 14 December 2018
Availableonline14December2018
0926-860X/©2018ElsevierB.V.Allrightsreserved.

G. Ge, Z. Zhao
AppliedCatalysisA,General571(2019)82–88
Ammonium chloride has also been employed to introducing defects
[24]. NH3 and HCl pyrolyzed from ammonium chloride act as soft
template to produce porous structure and surface defects from NH3
solely [25].
energy = 50 eV). The Raman spectra were measured using a laser with
an excitation wavelength of 532 nm at room temperature on a Thermo
Scientific DXR Raman microscope. Nitrogen adsorption and desorption
isotherms were determined on a Micromeritics apparatus of model
ASAP-2050 system at −196 °C. The specific surface areas were calcu-
lated by the BET method and the pore size distributions were calculated
from adsorption branch of the isotherm by BJH model. Note: the cat-
alysts after 20 h reaction were characterized because we compared the
steady state catalytic performance (steady state styrene rate and se-
lectivity) of catalysts after the 20 h of time of stream.
Herein, we introduce ammonium chloride to produce ND@carbon
nitride with loose porous carbon nitride structure to release more active
sites that belong to ND, but not weaken the synergistic effect between
the ND and the encapsulated carbon nitrides on ND. Compared with the
approaches previously reported [16], besides the strong synergistic
effect of ND and carbon nitride layers owing to the carbon nitride layers
close wrapped on ND led by the unique molecular structure of HTM, in
the developed wet-mixing of ND with HTM, and HTM can also act as
dispersing agent to allow the ND well being dispersed. Moreover, the
introduced ammonium chloride serves as gas template to produce pores
and defect. As a consequence, nanodiamond@carbon nitride hybrid
with loose porous carbon nitride layers close wrapped on well dispersed
NDs (ND-CN) was successfully prepared. ND-CN shows 1.9 and 1.2
times much higher styrene rate for oxygen and steam-free direct de-
hydrogenation of ethylbenzene than the pristine ND and the previously
reported carbon nitride encapsulated ND (H-ND) formed by the thermal
polymerization of mixture composed of HTM and ND, respectively,
ascribed the increased active sites containing defects and ketonic car-
bonyl groups and the not weakened synergistic effect of carbon nitride
with ND that originating from the enlarging surface area and pore vo-
lume owing to the characteristic concerning loose porous structure of
carbon nitride layers.
2.3. Catalytic performance measurement
The oxidant- and steam-free direct dehydrogenation of ethylben-
zene was performed over the developed catalyst and the experimental
details are as follows: the reaction was performed at 550 °C for 20 h in a
stainless steel, fixed bed flow reactor (6 mm O.D.). 25 mg catalyst was
loaded at the centre of the reactor with two quartz wool plugs at its two
sides. The system was heated to 600 °C and kept for 30 min in Ar for
pretreating catalyst. After the system was cooled down to 550 °C and
kept for 10 min, the feed containing 2.8% ethylbenzene with 10 ml
min−1 of gas hourly space velocity (GHSV) and Ar as balance was then
fed into the reactor from a saturator kept at 40 °C. The effluent from the
reactor was condensed in two traps containing certain amount of
ethanol connected in a series. The condensed material was cooled ex-
ternally in an ice water bath. Quantitative analysis of the collected
reaction products (ethylbenzene, styrene, toluene, and benzene) was
performed on a FULI 9790 II GC equipped with HP-5 column, 30
m × 0.32 mm × 0.25 μm, and FID detector. The resulting carbon bal-
2. Experimental
ance was above 100
4% in all reactions. The selectivity of styrene is
2.1. Catalysts preparation
employed as the evaluation standard for the catalytic performance of
the fabricated catalysts. The styrene rate is calculated as the formed
styrene molar amount per g catalyst per hour, and the selectivity of
styrene is denoted as the percentage of the desired styrene to the total
products including the desired styrene and the by-products that con-
taining benzene and toluene. The variation of conversion and se-
lectivity with contact time was measured by changing the GHSV from 6
to 14 ml min-1 with a step of 2 ml min-1. The contact time is denoted as
the volume of catalyst being divided by GHSVEB, in which the GHSV
Commercially available nanodiamond from Beijing Grish Hitech Co.
(China) were treated in acid solution, and then obtained oxidized na-
nodiamond (O-ND). ND-CN was synthesized as follows: 0.25 g HTM and
NH4Cl were dissolved in 20 ml deionized H2O. Next, O-ND was added
into the mixture, and then treated by further ultrasonication. The water
was removed under the reduced pressure, subsequently placed in a
quartz tube filled with flowing N2 at 750 °C. The preparation method of
Carbon nitride (CN) is same as above, but no ND is introduced. H-ND
was obtained by a method described in a previous report [16].
multiplies ethylbenzene content to obtain GHSVEB
.
3. Results and discussion
2.2. Catalysts characterization
3.1. Morphology and structure feature of samples
X-ray diffraction (XRD) profiles were collected from 10 to 80° at a
step width of 0.02° using Rigaku Automatic X-ray Diffractometer (D/
Max 2400) equipped with a CuKa source (λ = 1.5406 Å). Field emis-
sion scanning electron microscope (FESEM) experiments were per-
formed on JEOL JSM-5600LV SEM/EDX instrument. Transmission
electron microscopy (TEM) images were obtained by using Tecnai F30
HRTEM instrument (FEI Corp.) at an acceleration voltage of 300 kV.
The XPS spectra were carried out on an ESCALAB 250 XPS system with
a monochromatized Al Ka X-ray source (15 kV, 150 W, 500 μm, pass
Fig. 1 presents the preparation process of the ND-CN hybrid fea-
turing the NDs dispersed in loose porous carbon nitride structure.
Firstly, the commercial ND was well dispersed by ultrasonic in solution
of ammonium chloride and hexamethylenetetramine, then the followed
thermal polymerization of hexamethylenetetramine in the presence of
ammonium chloride. The morphology of the ND (a), H-ND (c) and ND-
CN (d) was investigated by transmission electron microscopy (TEM)
Fig. 1. Proposed process for the preparation of ND-CN material.
83

G. Ge, Z. Zhao
AppliedCatalysisA,General571(2019)82–88
Fig. 2. (a) TEM images of ND. (b) FE-SEM image of CN. (c,d) TEM images of H-ND (c) and ND-CN (d). Inset in Fig. 2d: FE-SEM image of ND-CN.
and the CN (b) by field-emission scanning electron microscopy
(FESEM). As shown in Fig. 2, the pristine ND is composed of particles
less than 10 nm that aggregate together. Carbon nitride obtained from
hexamethylenetetramine and ammonium chloride features plat piece
morphology with fluffy structure, while ND is flat on carbon nitride
piece to form ND-CN. Compared with the pristine ND, the ND ag-
gregation is in a looser status in H-ND and ND-CN. From the previous
report [16], the H-ND features unconsolidated carbon-nitride layer
close-wrapped nanodiamond hybrid (Fig. 2c). Compared to H-ND, the
as-synthesized DN-CN hybrid exhibits looser carbon nitride layers
coated on nanodiamond (Fig. 2d and the inset in Fig. 2d), ascribed to
the bubbling NH3 and HCl gas pyrolyzed from ammonium chloride. As
is presented in our previous report [16], different from the melamine,
the condensation of hexamethylenetetramine takes place through the
CeN splitting. The formed units can be linked with CH2, and therefore
the flexible carbon nitride structure can be formed, which is different
from the rigid structure from melamine. The flexible carbon nitride
Fig. 3. XRD patterns of ND, H-ND, ND-CN and CN.
units can be further polymerized to form the CNx layers, and therefore
the CNx-layer-wrapped ND nanohybrid can be formed, which can be
identified by FESEM, HRTEM, BET, XRD, and Raman analyses.
deaggregating effect of gas collision pyrolyzed from ammonium
chloride.
All four samples exhibit a diffraction peak at about 25.3° (Fig. 3),
corresponding to the typical graphite (002) of hexagonal carbon ma-
terial. The widened (002) peak indicates their amorphous and dis-
ordered stacking structure of interlayer, thus demonstrating that the CN
is composed of nitrogen containing loose carbon nitride layers [16].
ND, H-ND and ND-CN have three diffraction peaks located at 43.8°,
75.5° and 91° belong to (111), (022) and (113) planes of ND respec-
tively [17]. Compared with the pristine ND and H-ND, the further
weakened and widened diffraction peaks of developed ND-CN hybrid
sample demonstrate promoted ND dispersity by the introduction of CN
from hexamethylenetetramine and ammonium chloride and
In order to study the subtle structural variation of the samples,
Raman spectra were performed (Fig. 4). It is well known that D band
(A1g mode, disordered graphitic layer and structural defects) is cen-
tered at 1310 cm−1, and G band (E2g mode, symmetry-allowed gra-
phite) located at 1591 cm−1 [26], So there is exposed graphene in all
four samples. In comparison with ND, a shift of D band and of G band
can be observed in H-ND, indicating the strong interaction between CN
and ND, or the appearance of closely wrapped ND by carbon nitride
layer, which means poor efficiency of surface carbonyl on ND [15,16].
On the contrary, no visible shift on ND-CN indicates fully exposure of
84

G. Ge, Z. Zhao
AppliedCatalysisA,General571(2019)82–88
Table 1
Textural properties of the catalysts.
a
b
b
Samples
SBET (m2 g−1
)
Vtotal (cm3 g−1
)
Vmicro (cm3 g−1
)
APSc (nm)
ND
CN
ND-CN
H-ND
230.0
306.2
343.2
269.0
0.68
0.42
0.49
0.39
0.08
0.15
0.12
0.11
9.86
5.50
5.75
5.80
a
The total surface area (Stotal
)
was obtained from multipoint
Brunauer–Emmett–Teller (BET) plots and V–t plots, respectively.
b
The total pore volume (Vtotal) was determined at P/P0 = 0.99, and the
micropore volume (Vmicro) was calculated from the H-K plot.
c
Average pore size (APS).
ammonium chloride and hexamethylenetetramine, surface of which is
306.2 m2 g-1) was introduced into the system, the enlargement of sur-
face area may not only come from dispersion effect of CN, but also from
deaggregation of the aggregated ND by the collision of formed gases
from ammonium chloride. On the other hand, the gases introduced by
ammonium chloride blew hexamethylenetetramine derived polymers
into many large bubbles and obtain crinkly carbon nitride, and, surface
defect from gas corrosion promoted ND dispersion [28]. Compared with
ND-CN, the widen pore size distribution and large pore size for ND and
H-ND are contributed from accumulation pores among the aggregated
NDs, so although the pore volumes of ND-CN decrease from 0.68 to
0.49 cm3 g-1, ND is in a decentralized state in ND-CN sample. Mean-
while, a shift to smaller pore size can be observed for ND-CN’s in
comparison with that of ND, the smaller pores could be attributed to
gases etch of ND and CN, which issues in increasing porosity and pro-
moting dispersity. Moreover, the unique textural properties of CN-CN
also imply the loose porous carbon nitride layers on ND.
XPS was employed to analyze the loading content and surface state
of the four catalysts after reaction [3,29]. All four samples show
dominant C peaks (284.6 eV), N peaks (400 eV) and O peaks (533 eV).
The concentrations of C, N, and O are calculated and listed in Table 2.
The developed ND-CN catalyst shows much different characteristic in
XPS peak from the previously reported H-ND. The N/C ratios of ND, CN,
ND-CN and H-ND are 0.013, 0.075, 0.027 and 0.073. The synergistic
effect between the ketonic carbonyl groups on the ND and the nitrogen
species comes from the electron transfer from N to C]O. The in-
troduced electron-rich N atom can increase the nucleophilicity of C = O
C]O active sites, which subsequently improve the catalytic activity of
active sites for CeH activation [13–18]. The small amount of N in the
commercially available ND comes from the preparation process of de-
tonation method. The increased N content of ND-CN compared to the
pristine ND would facilitate the direct dehydrogenation of ethylbenzene
to styrene. The common issue is that CN has the largest N content
among the samples. Interestingly, ND-CN hybrid has a lower N content
than the previously reported H-ND, ascribed to the gas template role of
the added ammonium chloride in the pyrolysis process. Moreover, In
the C 1 s spectrum (Fig. 6b), the high resolution C 1 s spectrum can be
resolved into five peaks, C]C (284.6 eV) corresponding to the graphitic
structure, CeN (285.2 eV) attributed from defects on the carbon, CeC/
CeO (286 eV), C]O/C = N (N-sp2 C) (287 eV) and carboxyl O] CeO
(288.6 eV) [16,30]. For the CN material, the peak of C]O/C]N
Fig. 4. Raman spectra of ND, H-ND, ND-CN and CN.
surface carbonyl, meanwhile fluffy carbon nitride prepared from am-
monium chloride and hexamethylenetetramine can efficiently separate
the aggregated ND. ID/IG (intensity ratio of D band to G band) on ND,
CN, ND-CN and H-ND was collected. due to the bubbling of NH3 and
HCl gases generated from ammonium chloride, the ID/IG for ND-CN
(1.1780) is remarkably higher than three other samples (0.9462 for ND,
1.0767 for CN, and 1.1004 for H-ND) and this suggests more defects
and edge plane exposure in ND-CN [27], originating from the role of
ammonium chloride regarding of gas template. However, the ID/IG of
CN is lower than ND-CN, although it has the highest ammonium
chloride content, ascribed to the intensified gas template effect of am-
monium chloride by the supporting effect of the existing MD in the
mixture during pyrolysis. This is also the reason for the formation of
loose carbon nitride layers in the ND-CN hybrid. Corresponded to the
TEM and XRD data, Raman spectra further confirm positive actor of
NH4Cl in producing carbon nitride with more defects and edge plane
exposure. That is to say, the much different carbon nitride layers on ND-
CN owing to the presence of ammonium chloride leads to more defects
on NC-CN in comparison with H-ND and ND.
BET experiment was performed to further explore the specific sur-
face area and pore size of ND, CN, H-ND and ND-CN (Fig. 5). The N2
adsorption/desorption isotherms for all catalysts exhibit a typical IV
curve, implying the presence of mesopores in all four samples. Due to
the separation effect of CN (pyrolyzed from hexamethylenetetramine
alone), slight improvement in surface area occurred (230.0 m2 g−1 for
ND, 269.0 m2 g-1 for H-ND, Table 1). However, the surface area of ND-
CN is 343.2 m2 g-1, since only little amount of CN (prepared from
Table 2
XPS analysis of the samples.
Samples
N
(%)
N-1
(%)
N-2
(%)
N-3
(%)
N-4
(%)
O
(%)
C = O
(%)
O = C-O
(%)
C-O
(%)
ND
CN
ND-CN
H-ND
1.2
6.4
2.4
6.3
–
71.7
23.7
36.6
25.0
28.3
8.8
18.8
9.8
6.9
7.9
7.1
7.5
63.6
29.6
68.4
45.6
35.0
57.7
24.0
31.3
1.4
12.7
7.6
32.6
35.8
53.6
34.9
8.8
11.6
23.1
Fig. 5. Nitrogen adsorption-desorption isotherms. Inset: BJH mesopore size
distribution of ND, CN, ND-CN and H-ND.
N-1 (pyridinic N), N-2 (pyrrolic N), N-3 (graphitic N) and N-4 (oxidized N).
85

G. Ge, Z. Zhao
AppliedCatalysisA,General571(2019)82–88
Fig. 6. XPS spectra of the as-prepared catalysts. (a) Survey spectra, (b) C 1 s peak, (c) N 1 s peak, and (d) O 1 s peak.
(287 eV) is weaker than that of the others, ascribed to the less C]O
confirmed by the following O 1 s spectra (Fig. 6d). Compared with ND
and H-ND, the C]O/C = N peak of ND-CN shows higher intensity and
the CeN peaks shows smaller intensity which could be directly linked
with the introduced by the surface defect transformation in the com-
posite. From Fig. 6c, the N 1 s spectra of the catalysts can be fitted into
four peaks at 398.0, 399.5, 400.8 and 402.8 eV, respectively, corre-
sponding to the pyridinic N (CeN]C in six-atom rings at the edge of
grapheme), pyrrolic N (N in the heterocyclic ring of the pentad that is
attached to the phenolic or carbonyl group on the adjacent carbon atom
of the ring), graphitic N (sp2-hybridized N bonded to three C atoms in
grapheme layer’s central or valley position), and oxidized N respec-
tively [31,32]. There are only two peaks observed at 398.0 eV (pyr-
idinic N) and 403 eV (oxidized N) for ND, due to the absent of carbon
nitride in ND. As summarized in Table 2, the pyridinic N of ND-CN is
calculated to be 35.8 at% after deconvolution (Table 2), which is lower
than that of H-ND (53.6 at%) and CN (32.6 at%), but the content of
pyrrolic N on ND-CN is higher than H-ND. These results indicate the
surface of ND-CN exit more defects under the double blast attack of
hexamethylenetetramine and ammonium chloride. As to the chemical
bond types of oxygen, the detailed information were obtained in
Fig. 6d, the deconvoluted peaks located at 530.8, 532.5 and 533.8 eV in
the O 1 s spectra are attributed to ketonic carbonyl group (C]O),
O]CeO (oxygen atom in ester, anhydrides and carboxylic acid) and
CeO in phenol/ether, respectively [33,34]. Total surface oxygen con-
tents are 6.9at%, 7.9 at%, 7.1 at% and 7.5 at% for ND, CN, ND-CN and
H-ND, respectively. The higher O content on the developed ND-CN than
ND may be resulted from either the surface carbon nitride layers or the
increased exposure of O on the ND surface. The presence of O species of
CN may from oxidation of surface defects and edges by air [35,36].
Furthermore, the surface carbonyl group is the active species for direct
dehydrogenation reaction [19], and the percentage contents of O con-
cerning C]O group in the surface O are 63.6 at%, 29.6 at%, 68.4 at%
and 45.6 at% for ND, CN, ND-CN and H-ND. The higher carbonyl group
content of ND-CN can be observed in comparison with H-ND, which
might attribute from the increased accessibility of surface O on ND
owing to the more loose and porous carbon nitride layers and/or in-
creased surface defect of ND-CN nanohybrid to form C]O led by the
gas template role of ammonium chloride. The gas release from am-
monium chloride destroys the integrity of the graphene basal plane and
creates more edges [37].
The aforementioned results and analysis demonstrate the formation
of nanodiamond@carbon nitride hybrid with loose porous carbon ni-
tride layers close wrapped on well dispersed NDs through the thermal
polymerization of the mixture composed of hexamethylenetetramine
and ammonium chloride adsorbed on dispersed nanodiamonds, in
which hexamethylenetetramine serves as both carbon nitride precursor
and dispersing agent and the ammonium chloride acts as gas template.
In comparison with the previously reported H-ND with carbon nitride
layers close wrapped on ND formed by the thermal polymerization of a
mixture composed of HTM and ND, the as-prepared ND-CN in this work
demonstrates more exposed surface ketonic carbonyl groups and
structural defects that serve as active sites for the direct dehydrogena-
tion of ethylbenzene, owing to the more loose and porous carbon ni-
trides led by the gas template role of the added ammonium chloride. As
a consequence, ND-CN may show much superior catalytic performance
in the oxygen-/steam-free direct dehydrogenation.
86

G. Ge, Z. Zhao
AppliedCatalysisA,General571(2019)82–88
Fig. 7. Steady state styrene rate for direct dehydrogenation of ethylbenzene to
styrene over the ND, CN, ND-CN, ND-CN-m and H-ND catalyst (20 h of time on
stream). Reaction conditions: Catalyst (25 mg), 550 °C, 2.8% ethylbenzene in
Ar, 10 ml min−1 GHSV.
Fig. 9. Variation of conversion and selectivity with contact time over the de-
veloped ND-CN catalyst for direct dehydrogenation of ethylbenzene to styrene.
Reaction conditions: Catalyst (25 mg), 550 °C, 2.8% ethylbenzene in Ar, variety
of GHSV values (6, 8, 10, 12, and 14 ml min−1).
through catalytic cracking of ethylbenzene by the surface phenolic
hydroxyl group possibly owing to the decreased phenolic hydroxyl
group along with the time on stream at 550 °C of high temperature.
Fig. 9 presents the variation of conversion and selectivity with contact
time. From Fig. 9, the extending contact time leads to a gradual increase
in conversion. However, with the increase in contact time, only slight
decreased selectivity can be observed if the contact time is not longer
than 60.6 s. The further prolonged contact time lead to a dramatic de-
crease in selectivity. Therefore, the 60.6 s of appropriate contact time is
required to obtain relatively high conversion with high selectivity.
4. Conclusions
In a summary, we have developed a simple, inexpensive and en-
vironmentally-friendly approach for synthesizing nanodiamond@
carbon nitride hybrid with loose porous carbon nitride layers close
wrapped on well dispersed NDs (ND-CN) through the thermal poly-
merization of the mixture composed of hexamethylenetetramine (HTM)
and ammonium chloride adsorbed on dispersed nanodiamonds (ND), in
which hexamethylenetetramine serves as both carbon nitride precursor
and dispersing agent and the ammonium chloride acts as gas template.
ND-CN shows 1.9 and 1.2 times much higher styrene rate for oxygen
and steam-free direct dehydrogenation of ethylbenzene than ND and
the previously reported carbon nitride layers encapsulated ND (H-ND)
formed by the thermal polymerization of mixture composed of HTM
and ND, respectively, ascribed the increased active sites containing
defects and ketonic carbonyl groups and the not weakened synergistic
effect of carbon nitride with ND that originating from the enlarging
surface area and pore volume owing to the characteristic concerning
loose porous structure of carbon nitride layers. This work presents a
new way for designing and fabricating loose and porous carbon nitride-
layer close-wrapped nanocarbon hybrid catalysts with excellent cata-
lytic performance in diverse reactions.
Fig. 8. The catalytic stability of the developed ND-CN catalyst for direct de-
hydrogenation of ethylbenzene to styrene under steam and oxidant-free con-
ditions. Reaction conditions: Catalyst (25 mg), 550 °C, 2.8% ethylbenzene in Ar,
10 ml min−1 GHSV.
3.2. Catalytic properties of the samples
The catalytic performance expressed in terms of styrene products
per gram of catalyst per hours (mmolST g−1 h−1) on the steady-state
and the selectivity to styrene after 20 h reaction over ND, carbon ni-
tride, ND-CN and H-ND were evaluated for the DDH reaction (Fig. 7).
The weight of the catalyst in the reaction is 25 mg in all catalytic tests.
ND-CN presents 5.61 mmol g−1 h−1 of styrene rate (1.9 times higher
than ND) with 99.3% selectivity, superior to the mixture of ND and CN
(3.11 mmolST g−1 h−1). The steady-state styrene rate of H-ND is
4.83 mmol g-1 −1 and the styrene selectivity is 98.8%, which is higher
h
than ND (2.99 mmolST g−1 h−1) but lower than ND-CN. For compar-
ison, although NG-CN has a lower N content than H-ND, the developed
ND-CN shows 1.2 times much higher styrene rate for oxygen and steam-
free direct dehydrogenation of ethylbenzene than the previously re-
ported carbon nitride encapsulated ND (H-ND) formed by the thermal
polymerization of mixture composed of HTM and ND, respectively,
ascribed the increased active sites containing defects and ketonic car-
bonyl groups and the not weakened synergistic effect of carbon nitride
with ND that originating from the enlarging surface area and pore vo-
lume owing to the characteristic concerning loose porous structure of
carbon nitride layers. Fig. 8 shows the long term stability of ND and ND-
CN for DDH of ethylbenzene to styrene at 550 °C. From Fig. 7b, the
results indicate that the conversion at the initial period, and then
maintains a relatively stable value. It can be found that the selectivity
gradually increases along with the time on stream, resulting from the
suppressed formation of side-products including benzene and toluene
Acknowledgments
This work was financially supported by the National Natural Science
Foundation of China (21676046 and U1610104) and the Chinese
Ministry of Education via the Program for New Century Excellent
Talents in Universities (NCET-12-0079).
References
[1] W. Lei, Y.-P. Deng, G. Li, Z.P. Cano, X. Wang, D. Luo, Y. Liu, D. Wang, Z. Chen, ACS
Catal. 8 (2018) 2464–2472.
[2] B.Y. Guan, S.L. Zhang, X.W. Lou, Angew. Chem. Int. Ed. 57 (2018) 6176–6180.
87

G. Ge, Z. Zhao
AppliedCatalysisA,General571(2019)82–88
[3] N.P. Wickramaratne, J. Xu, M. Wang, L. Zhu, L. Dai, M. Jaroniec, Chem. Mater. 26
(2014) 2820–2828.
[21] H. Yu, R. Shi, Y. Zhao, T. Bian, Y. Zhao, C. Zhou, G.I. Waterhouse, L.Z. Wu,
C.H. Tung, T. Zhang, Adv. Mater. 29 (2017) 1605148.
[4] C. Liu, X. Huang, J. Wang, H. Song, Y. Yang, Y. Liu, J. Li, L. Wang, C. Yu, Adv.
Funct. Mater. 28 (2018) 1705253.
[5] R. Yuge, F. Nihey, K. Toyama, M. Yudasaka, Adv. Mater. 28 (2016) 7174–7177.
[6] V.N. Mochalin, O. Shenderova, D. Ho, Y. Gogotsi, Nat. Nanotechnol. 7 (2012)
11–23.
[7] D.A. Simpson, E. Morrisroe, J.M. McCoey, A.H. Lombard, D.C. Mendis, F. Treussart,
L.T. Hall, S. Petrou, L.C. Hollenberg, ACS Nano 11 (2017) 12077–12086.
[8] Y. Liu, Y. Zhang, K. Cheng, X. Quan, X. Fan, Y. Su, S. Chen, H. Zhao, Y. Zhang,
H. Yu, Angew. Chem. Int. Ed. 56 (2017) 15607–15611.
[22] P.P. Sharma, J. Wu, R.M. Yadav, M. Liu, C.J. Wright, C.S. Tiwary, B.I. Yakobson,
J. Lou, P.M. Ajayan, X.D. Zhou, Angew. Chem. Int Ed.54 (2015) 13701–13705.
[23] Q. Liang, Z. Li, Z.H. Huang, F. Kang, Q.H. Yang, Adv. Funct. Mater. 25 (2015)
6885–6892.
[24] D. Zhang, Y. Guo, Z. Zhao, Appl. Catal. B Environ. 226 (2018) 1–9.
[25] C. Liu, Y. Zhang, F. Dong, A. Reshak, L. Ye, N. Pinna, C. Zeng, T. Zhang, H. Huang,
Appl. Catal. B Environ. 203 (2017) 65–474.
[26] Y.P. Zhu, Y. Liu, Y.P. Liu, T.Z. Ren, T. Chen, Z.Y. Yuan, ChemCatChem 7 (2015)
2903–2909.
[9] T. Liu, S. Ali, B. Li, D.S. Su, ACS Catal. 7 (2017) 3779–3785.
[10] J. Zhang, D.S. Su, R. Blume, R. Schlögl, R. Wang, X. Yang, A. Gajović, Angew. Chem.
Int. Ed. 49 (2010) 8640–8644.
[11] H. Liu, J. Diao, Q. Wang, S. Gu, T. Chen, C. Miao, W. Yang, D. Su, Chem. Commun.
50 (2014) 7810–7812.
[27] X. Gu, W. Qi, S. Wu, Z. Sun, X. Xu, D. Su, Catal. Sci. Technol. 4 (2014) 1730–1733.
[28] Z. Li, J. Liu, C. Xia, F. Li, ACS Catal. 3 (2013) 2440–2448.
[29] Y. Chen, Z. Xiao, Y. Liu, L.-Z. Fan, J. Mater. Chem. A 5 (2017) 24178–24184.
[30] Z. Zhao, Y. Dai, G. Ge, X. Guo, G. Wang, Phys. Chem. Chem. Phys. 17 (2015)
18895–18899.
[12] Y. Liu, H. Ba, J. Luo, K.-H. Wu, J.-M. Nhut, D.S. Su, C. Pham-Huu, Catal. Today 301
(2018) 38–47.
[31] X. Gao, Z. Chen, Y. Yao, M. Zhou, Y. Liu, J. Wang, W.D. Wu, X.D. Chen, Z. Wu,
D. Zhao, Adv. Funct. Mater. 26 (2016) 6649–6661.
[13] Z. Zhao, Y. Dai, G. Ge, X. Guo, G. Wang, Green Chem. 17 (2015) 3723–3727.
[14] T.T. Thanh, H. Ba, L. Truong-Phuoc, J.-M. Nhut, O. Ersen, D. Begin, I. Janowska,
D.L. Nguyen, P. Granger, C. Pham-Huu, J. Mater. Chem. A 2 (2014) 11349–11357.
[15] Z. Zhao, Y. Dai, G. Ge, Q. Mao, Z. Rong, G. Wang, ChemCatChem 7 (2015)
1070–1077.
[32] F. Su, C.K. Poh, J.S. Chen, G. Xu, D. Wang, Q. Li, J. Lin, X.W. Lou, Energy Environ.
Sci. 4 (2011) 717–724.
[33] J. Diao, Z. Feng, R. Huang, H. Liu, S.B.A. Hamid, D. Su, ChemSusChem 9 (2016)
662–666.
[34] Z. Ma, H. Zhang, Z. Yang, G. Ji, B. Yu, X. Liu, Z. Liu, Green Chem. 18 (2016)
[16] Z. Zhao, W. Li, Y. Dai, G. Ge, X. Guo, G. Wang, ACS Sustainable Chem. Eng. 3 (2015)
1976–1982.
3355–3364.
[35] X. Sun, R. Wang, D. Su, Chin. J. Catal. 34 (2013) 508–523.
[36] Z. Zhao, Y. Dai, J. Lin, G. Wang, Chem. Mater. 26 (2014) 3151–3161.
[37] X. Duan, Z. Ao, H. Sun, S. Indrawirawan, Y. Wang, J. Kang, F. Liang, Z.H. Zhu,
S. Wang, ACS Appl. Mater. Interf. 7 (2015) 4169–4178.
[17] Z. Zhao, Y. Dai, J. Mater. Chem. A 2 (2014) 13442–13451.
[18] Z. Zhao, G. Ge, W. Li, X. Guo, G. Wang, Chin. J. Catal. 37 (2016) 644–670.
[19] R. Wang, X. Sun, B. Zhang, X. Sun, D. Su, Chem. Eur. J. 20 (2014) 6324–6331.
[20] G. Ge, X. Guo, C. Song, Z. Zhao, ACS Appl. Mater. Interf. 10 (2018) 18746–18753.
88