Superhydrophobic nickel/carbon core-shell nanocomposites for the hydrogen transfer reactions of nitrobenzene and N-heterocycles

Article abstract of DOI:10.1039/c9gc04358f

In this work, catalytic hydrogen transfer as an effective, green, convenient and economical strategy is for the first time used to synthesize anilines and N-heterocyclic aromatic compounds from nitrobenzene and N-heterocycles in one step. Nevertheless, how to effectively reduce the possible effects of water on the catalyst by removal of the by-product water, and to further introduce water as the solvent based on green chemistry are still challenges. Since the structures and properties of carbon nanocomposites are easily modified by controllable construction, a one step pyrolysis process is used for controllable construction of micro/nano hierarchical carbon nanocomposites with core-shell structures and magnetic separation performance. Using various characterization methods and model reactions the relationship between the structure of Ni?NCFs (nickel-nitrogen-doped carbon frameworks) and catalytic performance was investigated, and the results show that there is a positive correlation between the catalytic performance and hydrophobicity of catalysts. Besides, the possible catalytically active sites, which are formed by the interaction of pyridinic N and graphitic N in the structure of nitrogen-doped graphene with the surfaces of Ni nanoparticles, should be pivotal to achieving the relatively high catalytic performance of materials. Due to its unique structure, the obtained Ni?NCF-700 catalyst with superhydrophobicity shows extraordinary performances toward the hydrogen transfer reaction of nitrobenzene and N-heterocycles in the aqueous state; meanwhile, it was also found that Ni?NCF-700 still retained its excellent catalytic activity and structural integrity after three cycles. Compared with traditional catalytic systems, our catalytic systems offer a highly effective, green and economical alternative for nitrobenzene and N-heterocycle transformation, and may open up a new avenue for simple construction of structure and activity defined carbon nanocomposite heterogeneous catalysts with superhydrophobicity.

Full text of DOI:10.1039/c9gc04358f

Green Chemistry
View Article Online
View Journal
PAPER
Superhydrophobic nickel/carbon core–shell
nanocomposites for the hydrogen transfer
Cite this: DOI: 10.1039/c9gc04358f
reactions of nitrobenzene and N-heterocycles†
Shaofeng Pang,
*
a,b Yujing Zhang,c Qiong Su,a,b Fangfang Liu,a,b Xin Xie,a,b
Zhiying Duan,a,b Feng Zhou,d Ping Zhanga,b and Yanbin Wang*a,b
In this work, catalytic hydrogen transfer as an effective, green, convenient and economical strategy is for
the first time used to synthesize anilines and N-heterocyclic aromatic compounds from nitrobenzene and
N-heterocycles in one step. Nevertheless, how to effectively reduce the possible effects of water on the
catalyst by removal of the by-product water, and to further introduce water as the solvent based on green
chemistry are still challenges. Since the structures and properties of carbon nanocomposites are easily
modified by controllable construction, a one step pyrolysis process is used for controllable construction
of micro/nano hierarchical carbon nanocomposites with core–shell structures and magnetic separation
performance. Using various characterization methods and model reactions the relationship between the
structure of Ni@NCFs (nickel–nitrogen-doped carbon frameworks) and catalytic performance was investi-
gated, and the results show that there is a positive correlation between the catalytic performance and
hydrophobicity of catalysts. Besides, the possible catalytically active sites, which are formed by the inter-
action of pyridinic N and graphitic N in the structure of nitrogen-doped graphene with the surfaces of Ni
nanoparticles, should be pivotal to achieving the relatively high catalytic performance of materials. Due to
its unique structure, the obtained Ni@NCF-700 catalyst with superhydrophobicity shows extraordinary
performances toward the hydrogen transfer reaction of nitrobenzene and N-heterocycles in the aqueous
state; meanwhile, it was also found that Ni@NCF-700 still retained its excellent catalytic activity and struc-
tural integrity after three cycles. Compared with traditional catalytic systems, our catalytic systems offer a
highly effective, green and economical alternative for nitrobenzene and N-heterocycle transformation,
and may open up a new avenue for simple construction of structure and activity defined carbon nano-
composite heterogeneous catalysts with superhydrophobicity.
Received 21st December 2019,
Accepted 15th February 2020
DOI: 10.1039/c9gc04358f
rsc.li/greenchem
lines and N-heterocyclic aromatic compounds, great progress
Introduction
has been made in the past decade.5–8 Among the different
Arylamines, especially anilines and N-heterocyclic aromatic methods, catalytic reduction of nitrobenzene to anilines or
compounds, have considerable importance in a variety of catalytic oxidation of N-heterocycles to quinolines and indoles,
industrial processes, including pharmaceuticals, dyes, pesti- which occur under relatively mild reaction conditions and
cides and so forth.1–4 In view of the catalytic synthesis of ani- with simple facilities compared to other methods, as impor-
tant routes for the synthesis of anilines or quinolines and
indoles, constitute fundamental processes in organic synthesis
(Scheme 1).9–15 However, both catalytic reduction and catalytic
oxidation require the addition of external reagents (i.g., reduc-
aChemical Engineering Institute, Northwest Minzu University, Lanzhou, 730030,
P. R. China
tants or oxidants), which often suffer from several drawbacks,
such as relatively low atomic economy and undesirable by-pro-
ducts. In particular, these two kinds of reactions are per-
formed in different catalytic systems, which will further
increase the cost of the whole reaction.13,16,17 Hence, how to
achieve the catalytic preparation of anilines, quinolines and
indoles with higher atom economy by the concept of green
chemistry continues to attract scientific interest.
bKey Laboratory of Environmental Friendly Composite Materials and Biomass in
Universities of Gansu Province, Northwest Minzu University, Lanzhou, 730030,
P. R. China
cCollege of Chemistry and Chemical Engineering, Northwest Normal University,
Lanzhou 730070, P. R. China
dDalian National Laboratory for Clean Energy, Dalian Institute of Chemical Physics,
Chinese Academy of Sciences, 457 Zhongshan Road, Dalian 116023, P. R. China
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
c9gc04358f
This journal is © The Royal Society of Chemistry 2020
Green Chem.

View Article Online
Paper
Green Chemistry
efficient catalysts for this reaction with the removal of the by-
product water and further introduction of water as the solvent
based on the 12 principles of green chemistry.31
The superhydrophobicity of a material surface is an impor-
tant property, particularly in catalysis; the special wettability
properties of catalysts played a key role in heterogeneous cata-
lytic reactions. To date, a number of construction protocols for
fabrication of catalytic materials with superhydrophobic sur-
faces have been explored. However, to the best of our knowl-
edge, most of these methods and techniques are costly and
time-consuming, and often involve complicated multistage
Scheme 1 Catalytic synthesis of anilines and N-heterocyclic aromatic
compounds.
Generally, there are two concepts described in green chem- processes.32–36 Thus, a low cost and simple method for the
istry: atom economy and chemical specificity.18 Catalytic
preparation of catalysts with superhydrophobicity is still
hydrogen transfer, as a much safer and more environmentally urgently desired.
benign process, adopts a privileged position in many green
chemistry processes due to the clean transformation and
As is well known, carbon nanocomposites as highly
efficient and green catalytic materials, with well-defined mole-
higher atom economy of the overall reaction, especially in the cular structures have been investigated for many important
methodological tool box of the fine chemical industry and chemical transformations.37–39 More importantly, the struc-
organic laboratories.19,20 Till now, a considerable variety of elec-
tures and performances of this class of materials are easy to
trophiles (alkanes, alcohols, and amines) have been obtained in modify by controllable construction based on the structural
good to excellent yields and water is generated as one of the by- design.40 Therefore, combining the structural design with con-
products by the catalytic hydrogen transfer reaction under hom-
trollable construction may be a low cost and simple strategy
ogenous or heterogeneous conditions.21–27 Recently, our group for preparing the superhydrophobic catalysts with both micro–
has carried out research on the preparation of biphenylamine nano hierarchical structure and catalytic active sites.
and triphenylamine by the reaction of nitrobenzene and cyclo-
Nevertheless, it is still a big challenge to fabricate both a desir-
hexanone via a catalytic hydrogen transfer process. The mechan- able overall structure and catalytically active sites in one step.
istic studies prove that this reaction starts with the transfer According to the survey, we also observed that superhydropho-
hydrogenation reaction between nitrobenzene and cyclohexa-
bic carbon nanocomposite catalytic materials and their use in
none to generate anilines.28 Motivated by this, therefore, it also the catalytic hydrogen transfer reaction have rarely been
should be interesting to build up a bridge between the hydro- studied. However, very recently, we described the preparation
genation of nitrobenzene and the oxidation of N-heterocycles
for the preparation of anilines and quinolines or indoles via a waste wheat straw via controllable construction, where their
of conductive superhydrophobic nanocarbon materials from
catalytic hydrogen transfer strategy (Scheme 2).
structures and hydrophobicity can be tuned efficiently by the
annealing temperature.41 Based on this, and according to the
Notably, on account of the fact that the hydrogen transfer
process is usually accompanied by the production of water as basic principles of the coordination of transition metals to
one of the by-products, the formation of water may affect the PVP (polyvinyl pyrrolidone),42 herein we present a simple and
performance of the catalytic materials.19,29,30 Thus, we con-
low-cost strategy to convert a mixture of PVP/nickel nitrate to
sider that the strong water repelling ability can promote the catalytic materials with a hierarchical structure composed of
desorption of water from the catalyst surface, effectively reduce
the possible effects of water on the catalyst, and increase the
micro-flakes and nano-globules in one facile step (Scheme 3),
namely, the target materials were formed by controllable
adhesion of the catalyst to less polar reactants (i.g., nitro- carbonization after mixing the PVP solution with nickel
benzene and N-heterocycles), so as to decrease the mass trans-
fer resistance on active sites, achieving relatively high catalytic
nitrate and removing water from the mixture, where the
release rate of gas in the pyrolysis process of PVP/nickel nitrate
activities and selectivities. As mentioned above, we propose is the key factor to determine the hydrophobicity of materials.
the idea of the construction of superhydrophobic materials as
Furthermore, the size of nano-globules with core–shell struc-
tures of the final solid carbons can be adjusted by changing
the annealing temperature in the initial stage. Notably, in our
previous work, nickel can be considered as an essential
element to generate the active sites of catalysts and exhibited
many advantages in the hydrogen transfer process because of
its great magnetism and basic attributes, such as reusability,
simplicity and certain stability.43,44 Thus, nickel nitrate is not
only used as the catalytically active component, but also as the
foaming agent for the preparation of carbon nanocomposite
catalytic materials with hydrophobicity. Crucially, according to
various characterization methods and model reactions used to
Scheme 2 One-step catalytic synthesis of anilines and quinolines or
indoles.
Green Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Green Chemistry
Paper
Scheme 3 Preparation of Ni@NCF-700 with a hierarchical structure through controllable carbonization.
investigate the relationship between the structure of Ni@NCFs Typical procedures for aniline and quinoline synthesis and
(nickel–nitrogen-doped carbon frameworks) and the catalytic catalyst recycling
performance, the excellent catalytic performance of the
0.5 mmol (61.5 mg) nitrobenzene, 0.75 mmol (99.9 mg)
samples can be assigned to the hydrophobicity of catalysts
1,2,3,4-tetrahydroquinoline, 50 mg catalyst and 2 mL H2O were
and the possible catalytically active sites, which are composed
added into a 15 mL pressure tube equipped with a magnetic
of Ni nanoparticles with pyridinic N and graphitic N in the
stirrer. Subsequently, the pressure tube was purged with N2 by
structure of the nitrogen-doped graphene shell. In particular, a
using a long needle for three minutes and then quickly sealed
positive correlation was found between the catalytic perform-
with a Teflon thread plug. The pressure tube was placed into
ance and hydrophobicity of catalysts, where Ni@NCF-700
an aluminium block, heated to 145 °C (reaction temperature)
shows superhydrophobicity with a WCA (water contact angle)
and stirred for 18 h. The pressure in the tube reached 0.35
of up to 160°, and excellent catalytic performance for the reac-
MPa during the reaction. After the reaction, the external stan-
tion of nitrobenzene and N-heterocycles to synthesize anilines
dard method was used for quantitative analysis. The pressure
and N-heterocyclic aromatic compounds in the aqueous
tube was cooled to room temperature, and 70 mg biphenyl as
the external standard and 10 mL ethanol as the solvent were
phase.
added for quantitative analysis by GC-FID (Agilent 6890A) with
a 29 m HP-5 column. For GC-MS analysis, the Agilent
GC-MS-HP5890 instrument was used. The magnetic stir bar
Experimental section
Typical procedure for Ni@NCF-700 preparation
covered with the catalyst was taken out directly, washed with
ethyl acetate and n-hexane, and used for the next run.
A mixture of 1.0 g of PVP (K30) and Ni(NO3)2·6H2O (1.08 g, Purification procedure
3.7 mmol) was added to deionized water (30 mL) and agitated
The crude products were subjected to silica gel column chrom-
until complete dissolution. Subsequently, the mixture was
transferred into a 150 mL balloon flask and kept at 90 °C in
vacuo to remove the water. The obtained solid material was
further dried at 80 °C overnight. Then, the sample was placed
in the feed pipe and pyrolyzed under a N2 flow (99.999%) at
700 °C for 1 h (heating rate: 4 °C min−1), after which the
reactor was cooled to room temperature under the N2 flow,
and 0.6 g Ni@NCF-700 was obtained. In contrast, different
atography (60, 0.053–0.038 mm supplied by Qingdao Haiyang
Chemical and Special Silica Gel Co, Ltd) and eluted with about
75 mL of petroleum ether, followed by petroleum ether/EtOAc
(500–200 : 1) to give the corresponding products in good yields.
Results and discussion
samples were prepared with the same procedure but at In order to understand the correlation between the structure
different annealing temperatures, and denoted as and catalytic performance of catalysts, the corresponding
Ni@NCF-600, Ni@NCF-800 and Ni@NCF-900, respectively. Ni@NCFs were preliminarily studied and investigated by scan-
The other catalysts (Ni@1/2-NCF-700, Ni@2-NCF-700 and ning electron microscopy (SEM) and water contact angle
NCF-700) were prepared with the same procedure but with (WCA) measurement, as shown in Fig. 1 and the insets. The
different amounts of nickel nitrate, such as Ni(NO3)2·6H2O structures of Ni@NCFs consist of micro-flakes which can be
(0.75 g, 1.85 mmol), (3 g, 7.4 mmol) and no nickel nitrate.
clearly observed. Besides, this kind of microstructure as a
This journal is © The Royal Society of Chemistry 2020
Green Chem.

View Article Online
Paper
Green Chemistry
Fig. 1 SEM images of Ni@NCFs at low magnification (original magn. × 15 000, bar = 10 µm). The insets are the corresponding wetting experiments
and water contact angle degree.
basic unit structure is made up of a 3D framework, but can be recognized only at high SEM magnifications. As seen in
borders of the 3D framework tend to be fuzzy with the increase Fig. 2, high SEM magnifications of Ni@NCF samples confirm
of annealing temperature, and even more blurry markedly over the differentiation of their nanostructure. By using various
700 °C. It is suggested that the annealing temperature is a criti- annealing temperatures from 600 °C to 900 °C, the mor-
cal factor to control the bifurcation process of the carbon skel- phology of the resulting Ni@NCFs (Fig. 2a–d) is dramatically
etons of micro-flakes. Meanwhile, it is noteworthy that the different because a number of nano-globules on the surface of
nano-globules gradually become visible with increased anneal- materials have gradually become big and more and more
ing temperature as shown in Fig. 1a–d. However, when the use apparent. Eventually, nano-globules almost completely covered
of nickel nitrate as a foaming agent was reduced, the cell walls the micro-flakes (Fig. 2d). However, it should be noted that the
of micro-flakes became thicker and stronger (Fig. 1e), while surfaces of micro-flakes of Ni@1/2-NCF-700 are covered with
when the amount of nickel nitrate was doubled, the micro- tiny nano-burrs, so that it is very difficult to analyze the par-
structure of Ni@2-NCF-700 (Fig. 1f) tightly re-aggregated and ticle size distribution (Fig. 2e). Besides, for Ni@2-NCF-700,
formed irregular mass. This phenomenon may be caused by only a few larger particles on micro-flakes were observed
the excessive foaming during the process of the core–shell for- clearly (Fig. 2f). In general, the size of nanostructures and the
mation, which might result in hard aggregation of metal par- average interval between nanostructures are also crucial
ticles initially encapsulated in the shell structures.45,46 In factors to achieve the superhydrophobicity of the
addition, the insets of Fig. 1a–f show the results of wetting materials.48,49 Thus, statistical analysis of the particle size dis-
experiments performed on a glass substrate using a water tribution (PSD) was carried out by measuring the number of
droplet of 1 μL. As expected, the water contact angle of nano-globules in the insets of Fig. 2. The average nano-globule
Ni@NCF-600 is 132°; the contact angle then continues to diameters of Ni@NCF-700, Ni@NCF-800 and Ni@NCF-900
increase remarkably to 160° for Ni@NCF-700, which indicates were 49.09 nm, 53.96 nm and 131.56 nm, respectively. A more
that Ni@NCF-700 has superhydrophobic properties. However, detailed analysis of the PSD reveals that with the increase of
when the annealing temperature rises, the contact angle of the annealing temperature, the proportion of nano-globule
Ni@NCF-800 and Ni@NCF-900 decreases obviously, which sizes within 25–75 nm decreased from 90.48% to 12%, as well
have a water contact angle of 147° and 140°, respectively. as the interval size within 0–150 nm decreased from 87%
Beyond expectation, Ni@1/2-NCF-700 and Ni@2-NCF-700 also to 21%. In contrast, the average interval of nano-globules
exhibited surface hydrophobicity, with a water contact angle of increased from 118.53 nm to 131.56 nm with increasing
134° and 136°, respectively.
annealing temperature; these data of Ni@NCF-700 and
As described in Wenzel’s theory, the microstructure and Ni@NCF-800 basically obey a normal distribution. In particu-
nanostructure provide sufficient roughness beneficial to lar, based on these data and previous reports,41,49 it is clear
surface superhydrophobicity.47 However, the size of the fine that the morphology features of Ni@NCF-700 are more suit-
structure of materials is within the nanometer-scale, and thus able for exhibition of superhydrophobic surfaces. As men-
Green Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Green Chemistry
Paper
Fig. 2 SEM images of Ni@NCFs at high magnification (original magn. × 120 000, bar = 1 µm). The insets are the corresponding PSD histograms.
tioned above, it should be concluded that the amount of cules of water. Meanwhile, the corresponding conversions and
nickel nitrate used as a foaming agent and annealing tempera- yields were compared with different solvent systems, which
ture played key roles in the preparation of superhydrophobic included H2O and toluene. The detailed data are listed in
catalytic materials.
Table 1. As expected, when using water as a solvent, it was
In order to verify whether the performance of the catalysts found that the hydrophobicity of catalysts has obvious effects
is related to hydrophobicity, annealing temperature, nickel on the catalytic performance; with the increase of the hydro-
content or specific surface area, we have used the reaction of phobicity of the catalyst, the catalytic performance can be
nitrobenzene and 1,2,3,4-tetrahydroquinoline for the synthesis improved significantly, and this exhibits the correlation
of aniline and quinoline as a model reaction, and the perform- between the two. Ni@NCF-600 with a water contact angle of
ances of catalysts were compared with each other. Here in our 132° gives the nitrobenzene conversion at 70% with 68%
model reaction, in principle, 1,2,3,4-tetrahydroquinoline can aniline and 66% quinoline yields (Table 1, entry 1), while 99%
act as a hydrogen donor for nitrobenzene conversion in a nitrobenzene conversion with 97% aniline and 96% quinoline
process involving hydrogen transfer from tetrahydroquinoline, yields was obtained when superhydrophobic Ni@NCF-700 was
where 1.5 equivalents of 1,2,3,4-tetrahydroquinoline are used as the catalyst (Table 1, entry 2). But when using
needed to transform –NO2 into –NH2 and release two mole- Ni@NCF-800 (water contact angle of 147°) as a catalyst, not
Table 1 Optimization of the reaction conditionsa
Entry
Catalyst
Solvent
Con.d (%)
Con.e (%)
Y(a)f (%)
Y(b)f (%)
1
2
3
4
5
6
7
Ni@NCF-600
Ni@NCF-700
Ni@NCF-800
Ni@NCF-900
Ni@1/2-NCF-700
Ni@2-NCF-700
Ni@NCF-700
H2O or toluene
H2O or toluene
H2O or toluene
H2O or toluene
H2Ob
70b/96c
99b(97h)/98c
77b/96c
66b/74c
21b
71b/96c
99b(98h)/99c
79b/98c
69b/75c
22b
68b/95c
97b(95h)/95c
74b/94c
62b/70c
14b
66b/93c
96b(93h)/94c
71b/93c
60b/66c
11b
H2Ob
40b
43b
34b
30b
—
11
15
8g
7g
a 0.5 mmol of nitrobenzene, 0.75 mmol of 1,2,3,4-tetrahydroquinoline, 50 mg of the catalyst, 2 mL of solvent, 145 °C (reaction temperature), 18 h,
N2. b 2 mL of H2O as the solvent. c 2 mL of toluene as the solvent. d Nitrobenzene conversion determined by GC-MS. e 1,2,3,4-Tetrahydroquinoline
conversion determined by GC-MS. f Isolated yield. g Determined by GC-FID using biphenyl as the external standard material. h The catalyst was
used at the 3rd run.
This journal is © The Royal Society of Chemistry 2020
Green Chem.

View Article Online
Paper
Green Chemistry
only conversion but also the yield has a significant decrease the co-presence of the micro and mesoporous structures. This
(Table 1, entry 3). It is noteworthy that Ni@NCF-900 possessed type of structure could offer channels for mass transfer, in
good hydrophobicity with a water contact angle of 140°; which the surface structure can be penetrated by the less polar
however, only 62% aniline and 60% quinoline yields were molecules so that catalytically active sites can be reached
obtained (Table 1, entry 4), indicating that the catalytic per- easily.51,52
formance may not just be associated with its hydrophobicity.
To further recognize the morphology of Ni@NCFs with the
As mentioned above, it can be inferred that the annealing influence of different annealing temperatures and the amount
temperature of 900 °C may induce particular changes in the of nickel nitrate, the TEM images of Ni@NCFs are also
interior structure compared with a relatively lower annealing obtained as shown in Fig. S2.† It can be clearly observed that
temperature. In addition, it can be observed obviously that the smaller Ni nanoparticles are well dispersed on the surface
both Ni@1/2-NCF-700 and Ni@2-NCF-700 exhibit lower cata- of N-doped carbon supports (Fig. S2a and b†), indicating that
lytic performance, which illustrates that a moderate amount of the introduction of N/O-functional groups may effectively form
nickel in catalysts is also the key factor to achieve high catalytic a stronger metal–support interaction (SMSI) between the metal
performance. As control reactions, we also tested the catalytic nanoparticles and N-doped carbon supports. However, the
activity of the Ni powder (APS 0.25–0.55 micron), NiO obvious aggregation of Ni nanoparticles can be observed, and
(325 mesh) and NCF-700 under the same reaction conditions; the aggregation behavior is gradually intensified with the
Ni powder was used as the catalyst, and gave a relatively lower increasing of annealing temperature (Fig. S2c and d†).
yield (14% aniline and 12% quinoline). Similarly, a lower yield Besides, different amounts of nickel have an influence on the
(9% aniline and 8% quinoline) was also obtained when using morphologies of catalysts. Fig. S2e and f† show the granular
NiO as the catalyst. However, no reaction was observed when structure on the N-doped carbon supports with uneven sizes.
NCF-700 was used as the catalyst. Thus, the control experiment In consideration of the results of the model reaction of
revealed that Ni and NiO have a certain role in the success of different catalysts (Table 1), it can be imagined that the pres-
this reaction, but Ni, NiO and NCF-700 cannot be used as cata- ence of smaller Ni nanoparticles and their uniform distri-
lysts directly. Besides, when the whole catalytic system was bution also should be the key factors for the high activity.
subjected to solvent-free conditions, the yields of aniline and
In addition, for Ni@NCF-700, the nano-globule with a core–
quinoline were 8% and 7%, respectively. This phenomenon is shell structure was confirmed by HR-TEM, EDX and SAED
a result of solvent-free conditions, which is not a benefit for measurements. Apparently, in Fig. 3a, the HR-TEM image
the diffusion of reactant molecules in the whole catalytic revealed that the Ni nanoparticle was present in metallic crys-
system.50
tallite morphology completely covered by a nitrogen-doped gra-
Although the hydrophobicity of catalysts (such as phene shell with 3–5 layers which correspond to a thickness of
Ni@NCF-600, Ni@NCF-700 and Ni@NCF-800) is related to the 1.12–1.87 nm, and the crystal lattice plane distance of Ni (111)
catalytic performance in an aqueous solvent system, neverthe- (0.21 nm) was measured. Furthermore, the energy dispersive
less, an important question is whether or not the solvent X-ray spectroscopy (EDX) elemental line scans over a randomly
species influences the model reaction, or catalysts inherently selected nano-globule also indicated the formation of a struc-
have an appropriate catalytic performance which is not ture with a Ni core and N-doped graphene shell (Fig. 3b and d).
affected by the solvent species. Thus, toluene was chosen as These results were further confirmed by the ring-like selected-
the solvent for the model reaction. Interestingly, although the
BET specific surface area of Ni@NCF-700 is 263 m2 g−1
(Table S1†), much larger than those of Ni@NCF-600 (206 m2 g−1
,
Table S1†) and Ni@NCF-800 (215 m2 g−1, Table S1†), both
the conversion of nitrobenzene and the yields of aniline and
quinoline are excellent and around 95% if Ni@NCF-600,
Ni@NCF-700 or Ni@NCF-800 is used as a catalyst. Thus, the
catalytic performance of catalysts may not be due to their
higher BET surface area. Besides, in consideration of the
results of the model reaction with H2O as a solvent, two con-
clusions can be drawn: one is water as a solvent has an influ-
ence on the properties of the catalysts, so that the catalyst with
superhydrophobicity can avoid the effect and exhibit high
catalytic performance; the other is that the annealing temp-
erature has some influence on the overall structure of
catalysts, but may not have a great influence on the structure
of the catalytically active sites of Ni@NCF-600, Ni@NCF-700 or
Ni@NCF-800. Besides, as listed in Fig. S1,† the N2 adsorption–
desorption isotherms of Ni@NCF-600 to Ni@2-NCF-700 show
Fig. 3 HR-TEM (a), HAADF (b), the corresponding SAED pattern (c) and
a typical IV isotherm shape with a hysteresis loop, suggesting the line-scan EDS analysis (d).
Green Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Green Chemistry
Paper
area electron diffraction (SAED) patterns (Fig. 3c), indicat- materials prepared by the in situ doping process with high
ing that the core–shell structures are polycrystalline. The annealing temperature.56,57 Most notable for Ni@NCF-900, the
diffraction rings from inside to outside can also be seen intensities of the peaks due to the C–HX and C–O bonds are
clearly, and the corresponding pattern can be classified as the enhanced, which demonstrated that the high annealing temp-
(111), (200) and (220) planes of Ni in local regions, erature leads to an increase of the defect degree of the
respectively.
materials and simultaneously induces a strong IR activity.58,59
More information can be gained from the XRD characteriz- Notably, for NCF-700, the mixed stretching mode of the CvN
ation results. From Fig. S3,† sharp diffraction peaks of and CvC groups appears at 1621 cm−1. However, the absorp-
Ni@NCFs centered at 44.507°, 51.846° and 76.370° can be tion bands of the CvN and CvC groups in Ni@NCFs have
indexed to the Ni (111), Ni (200) and Ni (220) planes, which been shifted to
a lower wavenumber and around at
indicated that the PVP pyrolytical product as a carbon source 1548 cm−1 60–62
,
especially for Ni@NCF-900, this shift is most
can effectively reduce nickel oxide during the annealing obvious. It can be inferred that mutual interactions between
process. However, the diffraction peak (002) of carbon almost these functional groups in materials and the surface of the
cannot be observed easily, which may be attributed to the fact nickel nanoparticles may take place.54,63 Additionally, a broad
that there are only small graphite domains present in the peak appeared between 1000 cm−1 and 1500 cm−1 which can
samples.53
be attributed to the overlap of bands, including the C–N and
In addition, the FT-IR spectrum was used to investigate the N–CH3 bonds at 1250 and 1372 cm−1, the bands of the CvC
types of functional groups present in the catalysts, and further stretching (1460 cm−1), and the asymmetric and symmetric C–
to understand the possible interaction between the nickel O (1116 cm−1) stretching in the C–O–C group (1053 cm−1),
nanoparticles and functional groups. The spectra of NCF-700 etc.63 In particular, it was difficult to further observe the
and Ni@NCFs were stacked and are displayed in Fig. 4. All the stretching vibration of NiO at 678 cm−1 54
This result indicates
.
samples containing broad peaks with varying intensities that the nickel species in catalysts is in the form of metallic
around 3436 cm−1 can be assigned to the O–H stretching nickel, which also corroborates with the XRD results.
vibration due to the presence of moisture in the samples.54
To further reveal the structure of the support and shell,
The absorption peaks at 2924 and 2840 cm−1 are consistent which are composed of N-doped graphene, and the role of
with the C–HX stretching vibrations, such as the asymmetric nitrogen doping in modifying the structure of carbon
and symmetric CH, CH2 or CH3 stretching.55 The peak at materials, a detailed Raman study on Ni@NCFs was carried
2348 cm−1 ascribed to the C–O bonds and the features at out. As shown in Fig. 5, evidently, there are two intense peaks
1743 cm−1 are consistent with the vibrational modes of multi-
walled carbon nanotubes, which usually existed in carbon
Fig. 4 FT-IR spectra of Ni@NCFs and NCF-700.
Fig. 5 Raman spectra of Ni@NCFs.
This journal is © The Royal Society of Chemistry 2020
Green Chem.

View Article Online
Paper
Green Chemistry
Table 2 EA analysis results of different catalystsa
of the Raman spectra for Ni@NCFs. The first one is observed
at approximately 1340 cm−1 for a well-defined D band, which
can be attributed to the multiple overlapping peaks due to the
disorder and a large number of obvious defects of the carbon
crystallites, including the bands of C–N and CvN vibration at
around 1280 cm−1 and 1400 cm−1, and the D band of
graphene.64,65 The second peak at around 1588 cm−1 is
assigned to the G band, which might be caused by the doubly
degenerate zone center E2g mode.41 Furthermore, for
Ni@NCF-600, Ni@NCF-700, Ni@NCF-800 and Ni@NCF-900,
the ID /IG ratio (peak area) is 2.2, 2.9, 3.4 and 4.6, respectively,
which showed that graphitization of the sample is decreased
Entry
Catalyst
N
C
H
O
Nib
1
2
3
4
Ni@NCF-600
Ni@NCF-700
Ni@NCF-800
Ni@NCF-900
4.09
2.83
1.01
0.41
40.83
45.42
46.77
46.87
1.11
0.59
0.39
0.13
4.96
1.03
0.76
0.47
49.01
50.13
51.07
52.12
a Elemental analysis (N, C, O and H, wt%) of the samples was carried
out on a Vario EL microanalyzer. b The loadings of all catalyst samples
were determined by ICP-AES.
but the defect degree is increased as a result of the increase of temperature rises to 900 °C, Ni@NCF-900 is obtained with
annealing temperature. These results are in good accord with ∼0.41 wt% nitrogen, ∼46.87 wt% carbon, ∼0.13 wt% hydrogen,
the FT-IR results above. In addition, the ID/IG ratios (peak and ∼0.47 wt% oxygen (Table 2, entry 4). In consideration of
area) for Ni@1/2-NCF-700 and Ni@2-NCF-700 are 2.5 and 3.1, the results of the model reaction (Table 1), it can be imagined
respectively; these results suggested that although the that a suitable amount of the nitrogen functional group and
reduction mechanisms may be the same, there can be a poss- the surface of Ni nanoparticles may form the appropriate cata-
ible effect for the defect degree of the sample as the amount of lytically active sites, so as to achieve a high catalytic perform-
metallic oxide increases. Notably,
a broad 2D band of ance from the catalyst. However, nitrogen-doped materials are
Ni@NCFs was observed clearly at around 2800 cm−1, revealing a complex system because of the coexistence of various nitro-
the typical Raman characteristics for few layers of graphene gen and oxygen functional groups, and which type of function-
sheets with abundant defects.66 These results are consistent ality is responsible for the electronic interaction and further
with the literature and thus explain why Ni@NCF-600, Ni@1/ form catalytic activity sites are still undefined. Thus, XPS tests
2-NCF-700 and Ni@2-NCF-700 are without an obvious micro– have been used to confirm the relationship between the chemi-
nano hierarchical structure but possess good hydrophobicity; cal environment and different annealing temperatures of the
for example, Koratkar et al. have reported that with the increas- samples.
ing number of graphene layers, the contact angle of water
Typical XPS survey scans of the catalysts are shown in
should increase, due to the fact that the adsorption energy Fig. S4,† which indicate that the samples consisted of C, N,
decreases monotonically with the increasing number of gra- O, and Ni as the main elements. The high resolution C 1s
phene layers and causes an increase in the water contact spectrum can be fitted by two peaks centered at 284.7 and
angle.67
285.5 eV, where the peak at 284.7 eV is assigned to the C–C
Normally, using nitrogen-containing organic matter as the of hybridized graphite-like carbon atoms, which indicates
precursor, through the in situ doping process one can achieve that most of the C atoms in the N-doped graphene sheets are
N-doped carbon materials with a homogeneous elemental dis- arranged in a conjugated honeycomb lattice, and the peak at
tribution at controlled temperature, and they always contain N, 285.5 eV is assigned to the formation of the C–N (Fig. 6).53,71
C, H and O as the main elements.68,69 In particular, a stronger In addition, the shifting with the already existing peaks of
metal–support interaction (SMSI) can occur between the nitro- the C 1s spectra is very hard to observe directly with the
gen or oxygen functional groups of N-doped carbon materials increase of annealing temperature. Likewise, the C–O func-
and metal nanoparticles, which can strongly influence the tional groups (such as CvO, C–O and O–CvO) are difficult
catalytic performance in specific reactions.28,70 Thus, to preli- to observe in the higher-binding energy region of the C 1s
minarily reveal the relationship between the amount of N, C, spectra.72
H and O in catalysts and the annealing temperature, elemental
However, the O 1s spectrum of samples can be clearly
analyses of Ni@NCFs with different annealing temperatures observed and deconvoluted into three peaks centered at 531.4,
are conducted after completing the carbonization process. The 532.4, and 533.8 eV (Fig. 7, left). Based on the reports about
elemental analysis of Ni@NCF-600 showed the presence of the O 1s spectra of atomic oxygen and referring to the XPS C
∼4.09 wt% nitrogen, ∼40.83 wt% carbon, ∼1.11 wt% hydrogen, 1s spectra mentioned above, the O species is assigned to the
and 4.96 wt% oxygen (Table 2, entry 1). Nitrogen content, as dissociative atomic oxygen, the oxygen atom association and
very important information, is one of the characteristics of the absorbed water, respectively (Fig. 7a–e).41,73,74 Moreover, these
signal for the content of nitrogen functional groups in results of the O 1s spectra are in good agreement with our pre-
N-doped carbon materials. However, the nitrogen, hydrogen vious report of carbon-based superhydrophobic materials and
and oxygen content obviously reduced with the increasing also with DFT-calculations.75 Notably, in contrast to the O 1s
annealing temperature during the carbonization process. The spectra of NCF-700, no obvious O 1s binding energy shift can
elemental analysis showed that Ni@NCF-700 contains nitrogen be seen in other samples when introducing Ni, indicating
(∼2.83 wt%), carbon (∼45.42 wt%), hydrogen (∼0.59 wt%) and again that the surface of samples, especially the shell struc-
oxygen (∼1.03 wt%) (Table 2, entry 2). When the annealing tures, may not contain oxygen functional groups. Thus, as
Green Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Green Chemistry
Paper
centered at 398.6, 400.5, and 401.6 eV, corresponding to pyridi-
nic N, pyrrolic N, and graphitic N, respectively (Fig. 7a1).28,76
Generally, the pyridinic N and pyrrolic N exist on the edges or
defect sites in the N-doped graphene sheets, and the graphitic
N is within the plane of N-doped graphene; these N species are
formed through substituting the nitrogen atom for the carbon
atom.77 Besides, according to the literature, the graphitic N
can interact with metal species and plays a key role in dehydro-
genation and oxygen reduction due to the stronger π bonds of
aromatic rings.57,70,78–80 Thus, as expected, for Ni@NCF-600
and Ni@NCF-700 (Fig. 7b1 and c1), a noticeable shift of the
peaks of the pyridinic N and graphitic N from 398.6 eV and
401.6 eV to 398.3 eV and 401.2 eV occurred after introduction
of Ni nanoparticles; these results suggested that the Ni species
wrapped into the N-doped graphene shells can effectively
increase the electron density of N species, conversely, which
also caused the electron-deficient chemical state of Ni nano-
particles. Besides, compared with the N 1s XPS spectra of
Ni@NCF-600 and Ni@NCF-700, there is an obvious shift of
peaks in the spectra of not only the pyridinic N and graphitic
N but also pyrrolic N as the annealing temperature rises from
800 °C to 900 °C, such as Ni@NCF-800 and Ni@NCF-900
(Fig. 7d1 and e1), which strongly implies two facts, one is that
the electronic interaction occurred between pyrrolic N and Ni
nanoparticles, and the other is that a higher annealing temp-
erature can effectively make N species to interact with Ni nano-
particles more strongly. However, based on the results of the
model reaction in water or toluene (Table 1), it can be
assumed that the interaction between pyridinic N, graphitic N
and the surface of the Ni core might form “the structure of cat-
alytically active sites”, which are consistent with the Ni 2p XPS
spectra of the samples (Fig. 8).
Fig. 6 C 1s spectra of Ni@NCFs and NCF-700.
mentioned above, we can infer that a carbon-based hydro-
phobic surface with a special structure can trap more air
because of not only the trenches of the nano-globules but also
the pores on the surface of samples. The N 1s high-resolution
XPS spectra of NCF-700 can be deconvoluted into three peaks
Fig. 7 O 1s (left) and N 1s (right) spectra of Ni@NCFs and NCF-700.
This journal is © The Royal Society of Chemistry 2020
Green Chem.

View Article Online
Paper
Green Chemistry
The Ni 2p spectra of Ni@NCF-600 and Ni@NCF-700 are and 879.2 eV (Ni 2p1/2) (peak-3), which can be observed
deconvoluted into two groups of peaks (Fig. 8a and b), where obviously in Ni@NCF-800 and Ni@NCF-900 (Fig. 8c and d),
two obvious peaks centered at 852.8 and 870.2 eV were indicating that the local electronic structure of Ni nano-
observed in the Ni 2p XPS spectra (peak-1), corresponding to particles can be significantly affected by the interaction with
the Ni 2p3/2 and 2p1/2, respectively; this mightily proves that not only the pyridinic N and graphitic N but also the pyrrolic
the Ni nanoparticles maintain the metallic state as the N, which is in good agreement with the results of the N 1s
predominant species in the two samples,53,81 which is in good XPS spectra. Interestingly, the Ni 2p spectra of Ni@NCF-800
accord with the HR-TEM, EDX, SAED and XRD results above. and Ni@NCF-900 are almost identical, including the peak-1,
Combined with the results of the N 1s XPS spectra of the two peak-2 and peak-3. However, the catalytic performances of
samples (Ni@NCF-600 and Ni@NCF-700) and the model reac- Ni@NCF-800 and Ni@NCF-900 in model reactions with
tion, the peaks at 856.3 eV (Ni 2p3/2) and 873.7 eV (Ni 2p1/2) toluene as the solvent are quite different. This phenomenon
can be assigned to the Ni species which interacted with can be explained by the results of TEM analysis; when the
the pyridinic N and graphitic N (peak-2), and play the impor- annealing process is conducted at high temperature (e.g.,
tant role of catalytically active sites in the model reaction. 900 °C), Ni nanoparticles become larger as a result of agglom-
Moreover, further increasing the annealing temperature to eration, correspondingly, which could reduce the catalytic
800 °C and 900 °C results in new peaks at 861.3 eV (Ni 2p3/2) performance.
The above results and discussions collectively indicated
that the Ni@NCF precursor is directly annealed at high temp-
erature under a nitrogen atmosphere, which can result in a
superhydrophobic catalyst with a hierarchical structure com-
posed of micro-flakes and nano-globules (i.g., Ni@NCF-700),
where Ni nanoparticles can be encapsulated in an N-doped
graphene shell and formed nano-globules with a core–shell
structure. During the annealing process, Ni2+ cations in the
precursor might be completely reduced to Ni nanoparticles by
graphene layers, meanwhile pyridinic N and graphitic N in the
structure of the nitrogen-doped shell could interact with the
surface of Ni nanoparticles for the formation of “the structure
of catalytically active sites”, so an illustration of the catalyst
local structures is given in Scheme 4.
Since Ni@NCF-700 displayed the best activity for the model
reaction, it was used in the hydrogen transfer reaction with
respect to 1,2,3,4-tetrahydroquinoline with different nitro-
benzene compounds under the optimized conditions. As
shown in Scheme 5, first, the hydrogen transfer reaction of
1,2,3,4-tetrahydroquinoline with different nitrobenzene com-
pounds was tested (Scheme 5). Notably, substituents on nitro-
benzene at the 3 or 4 positions are well tolerated, such as
p-toluidine (3b), p-anisidine (3c), 3-ethylaniline (3e), 4-butyla-
Fig. 8 Ni 2p spectra of Ni@NCFs.
Scheme 4 Pictorial description of Ni@NCFs.
Green Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Green Chemistry
Paper
Scheme 6 Results of the hydrogen transfer reaction of nitrobenzene
and N-heterocycles. Conditions (n = 0 or 1): 1 (0.5 mmol), 2 (1.5 or
0.75 mmol), 2 mL of H2O, 50 mg of catalyst, 145 °C (reaction tempera-
ture), 18 h, N2. All the numbers are isolated yields.
noline (4f) and 5-nitro-1H-indole (4j) were isolated with yields
of 12% and 15%, respectively, even after a prolonged reaction
time. It should be noted that very small amounts of by-pro-
ducts were observed if 7-nitro-1,2,3,4-tetrahydroquinoline was
used as the starting material, such as quinolin-7-amine and
1,2,3,4-tetrahydroquinolin-7-amine. As well as indolin-5-amine
and 1H-indol-5-amine were observed when 5-nitroindoline was
used as the starting material. These results indicated that this
Scheme 5 Results of the hydrogen transfer reaction of 1,2,3,4-tetrahy-
droquinoline and different nitrobenzene compounds.Conditions:
1
(0.5 mmol), 2 (0.75 mmol), 2 mL of H2O, 50 mg of catalyst, 145 °C (reac-
tion temperature), 18 h, N2. All the numbers are isolated yields.
niline (3f) and 3,5-dimethylaniline (3g) which were produced hydrogen transfer reaction was accompanied by the intra-mole-
in 96%, 93%, 83%, 93% and 76% yields, respectively. However, cular redox-oxidative reaction.
more sterically hindered functional groups, such as 2-ethyl
Finally, the reaction also proceeded successfully with other
(3d), 2,6-dimethyl (3h), and 2-carbonyl (3o), were converted to structurally and electronically diverse nitrobenzene com-
the corresponding quinolines in 21%–62% yields. Electron- pounds and N-heterocycles (Scheme 7). Clearly, excellent
withdrawing groups had a certain influence on the activity of results were obtained and the yields of p-toluidine (3b) and
this catalytic system. Thus, 4-chloroaniline (3i), 4-bromoani- 7-methylquinoline (4b) reached 94% and 92%, respectively.
line (3j), 1-naphthylamine (3m) and butyl 4-aminobenzoate Besides, both p-anisidine (3c) and 7-methylquinoline (4b) were
(3p) were obtained with 63%, 62%, 64% and 20% yields. obtained in yields up to 92%. Other nitrobenzene compounds
Particularly, a product containing more electron-withdrawing and N-heterocycles with different substituents were also
groups such as 1-(2-aminophenyl)-ethanone (3o) was isolated tested, respectively, which gave 50–76% moderate yields
in only 21% yield. Meanwhile, it should be pointed out that towards the corresponding products. However, it should be
the reaction does not well tolerate the presence of reducible or mentioned that the steric hindrance effect and the electronic
hydrolysable functional groups (i.g., carbanyl or ester groups) effect of substituents had a substantial effect on the reaction.
in the catalytic system, where the production of 3o and 3p is The yields of 3c and 4c were 71% and 70%, respectively, when
accompanied by the production of by-products, such as 1-(2- 4-nitroanisole and 6-methoxy-1,2,3,4-tetrahydroquinoline were
aminophenyl)ethanol, 1-(2-nitrophenyl)ethanol, 4-aminoben- used as the starting materials, while only 33% 3h and 35% 4h
zoic acid or p-nitrobenzoic acid, etc., which was detected by was obtained from the reaction of 2,6-dimethyl-1-nitrobenzene
GC-MS.
and indolin-5-ol. In addition, if nitrobenzene compounds and
Next, we tested the catalyst by preparing aniline and N-heterocycles containing strong electron-poor groups, i.e., F,
N-heterocyclic aromatic compounds from nitrobenzene and were used as the starting materials, almost no desired pro-
N-heterocycles. As shown in Scheme 6, the catalyst was also ducts were detectable. Nevertheless, this work offers a syntheti-
efficient for catalytic hydrogen transfer of nitrobenzene and cally powerful and interesting method for catalytic synthesis of
similar N-heterocycles under 145 °C in water. Gratifyingly, anilines and N-heterocyclic aromatic compounds.
7-methylquinoline (4b) and 2-methyl-1H-indole (4k) were iso-
Heterogeneous catalysts are used in more than 70% of the
lated in yields up to 94% and 87%, respectively. The yield of industrially relevant processes because catalysts are easily sep-
the other products is about 60–70%. However, products con- arated and recycled. Compared with other separation pro-
taining more electron-withdrawing groups such as 7-nitroqui- cesses, the magnetic separation process can effectively reduce
This journal is © The Royal Society of Chemistry 2020
Green Chem.

View Article Online
Paper
Green Chemistry
Scheme 8 Digital photos before the reaction started (a), reaction in
progress (b) and after the reaction (c).
as is known, for a heterogeneous catalyst, the aggregation and
leaching of the active metal will be difficult to avoid during the
reaction, which often results in deactivation of the catalysts.
Thus, the reusability of Ni@NCF-700 in the model reaction
was tested. For this purpose, the active material was reused
directly 3 times. To our delight, no obvious deactivation was
observed. The yield of the desired products was up to 93%
(Table 1, entry 2). It is worth noting that the micro–nano hier-
archical structure and core–shell structure of Ni@NCF-700
remained intact after the catalyst was used three times,
demonstrating the stability of the Ni@NCF-700 architecture
(the corresponding morphology is shown in Fig. S5†). All these
results suggest that Ni@NCF-700 may have potential appli-
cations in the field of fine chemistry and the pharmaceutical
industry.
In general, the electrical state of the active hydrogen species
(H*) can be polar (hydride and proton, Hδ− and Hδ+) and non-
polar (radical, Hδ0).84 If H* exists in the electronic neutral
state, hydrogen radicals must be involved in the reaction, and
the model reaction in the presence of butylated hydroxyto-
luene (BHT, 0.1 mmol) was investigated. Obviously, in the
presence of this radical scavenger, no effect on the catalytic
activity was observed, indicating a non-radical hydrogen
species-involved reaction path. Furthermore, in order to
Scheme 7 Results of the hydrogen transfer reaction of nitrobenzene
compounds and N-heterocycles. Conditions (n = 0 or 1): 1 (0.5 mmol), 2
(1.5 or 0.75 mmol), 2 mL of H2O, 50 mg of catalyst, 145 °C (reaction
temperature), 18 h, N2. All the numbers are isolated yields.
the production cost;82,83 thus, it has good application pro- conduct further discussion on the catalytic mechanism, it was
spects in industrial catalysis. In order to confirm that imperative to clarify the reaction process. The reaction of nitro-
Ni@NCF-700 shows good magnetic separation performance, benzene and 1,2,3,4-tetrahydroquinoline was traced by GC-FID
the phenomena in the reaction process were carefully to determine the quantities of the intermediates at different
observed. When Ni@NCF-700 as a catalyst was added to the reaction times. The reaction kinetics study has shown that,
solvent (water as the solvent), it floats on the surface because two hours after the reaction, 0.7 μmol of nitrosobenzene and
of its superhydrophobicity. However, when a magnetic stir bar 0.4 μmol of 1,2-dihydroquinoline were detected. Next, the
was added to the reaction system, it was obvious that the cata- content of nitrosobenzene and 1,2-dihydroquinoline decreased
lyst was rapidly coated on the surface of the magnetic stir bar at 6 h, i.e., 0.34 μmol of nitrosobenzene and 0.21 μmol of 1,2-
due to the magnetic properties of the catalyst (see Scheme 8a). dihydroquinoline can be observed inside the reaction mixture.
During the reaction process, the catalyst wrapped on the As the reaction continued, notably, both nitrosobenzene and
surface was dispersed in the solution due to the centrifugal 1,2-dihydroquinoline cannot be observed obviously at 14 h.
force generated by the high-speed rotation of the magnetic stir This result shows that the quantities of nitrosobenzene and
bar (see Scheme 8b). Finally, when the reaction was completed, 1,2-dihydroquinoline decreased remarkably at 14 h, so they fall
the catalyst was re-coated on the surface of the magnetic stir below the detection limit of GC-FID and cannot be observed,
bar because the magnetic stir bar stopped rotating at a high or no longer exist in the reaction system. In light of the afore-
speed (see Scheme 8c), and then the magnetic stir bar covered mentioned results, it can be concluded that both nitroso-
with the catalyst could be taken out together and recycled benzene and 1,2-dihydroquinoline may be the reaction inter-
directly after washing with ethyl acetate and n-hexane. Besides, mediates, as well as that the radical reaction is not involved in
Green Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Green Chemistry
Paper
Conclusions
In summary, we have developed carbon nanocomposite cata-
lytic materials with a unique hydrophobic nature from PVP/
nickel nitrate via a simple, green and catalyst-free process by
controllable construction, in the absence of any expensive cata-
lyst or complex equipment. Their structures and hydrophobi-
city can be easily modulated by the annealing temperature. In
particular, for the unique Ni@NCF-700, the superhydrophobic
N-doped graphene shell can behave as a “solid ligand” to cover
the Ni core and shows high affinity towards organic substrates
in the aqueous phase and excellent catalytic performance in
the hydrogen transfer reaction for the synthesis of anilines
and N-heterocyclic aromatic compounds in one step. In this
catalytic system, the water repelling property of Ni@NCF-700
can remove the by-product water, so as to effectively reduce the
possible effects of water on the catalyst, furthermore, which
can also further introduce water as a solvent based on the 12
principles of green chemistry. Crucially, a positive correlation
was found between the catalytic performance and hydrophobi-
city of catalysts. Given the above, it is believed that this work
Scheme 9 Proposed reaction mechanism.
this reaction process, which is also consistent with the publi- should open up a new avenue for green catalytic synthesis of
cations reported by Fukuzumi S, et al.85 Besides, based on the anilines and N-heterocyclic aromatic compounds, and struc-
previous reports of catalytic hydrogen transfer, we assumed ture and activity defined carbon nanocomposite hetero-
that the reduction of the nitro group occurred via the direct geneous catalysts with superhydrophobicity for fine chemical
pathway, i.e., the nitro group was first hydrogenated to nitroso synthesis.
and then further to hydroxylamine, and finally to amine;
especially, the premise of hydrogenation reactions was to gene-
rate the hydride and proton on the catalyst surface.84,86 Conflicts of interest
Meanwhile, 1,2,3,4-tetrahydroquinoline was dehydrogenated to
There are no conflicts to declare.
1,2-dihydroquinoline, and then sequentially dehydrogenated
to quinoline. Just as the basic ligands of transition metal com-
plexes promoted the heterolytic cleavage of H2 to give metal
hydride species,87 the mechanism involved a cooperation
effect between Ni nanoparticles and the N-doped graphene
shell as basic sites at the interface of the N incorporated
N-doped graphene shell and Ni particles Thus, a possible reac-
tion mechanism is proposed (Scheme 9). First, the dehydro-
genation of 1,2,3,4-tetrahydroquinoline to 1,2-dihydroquino-
line occurs at catalytically active sites, which promoted 1,2,3,4-
tetrahydroquinoline to give Ni–H-species in the metal hydride
and split-over H+ species in N–H+. Namely, the coordination of
1,2,3,4-tetrahydroquinoline to the Ni core with the assistance
of the N-doped graphene shell affords Ni–H-species, followed
by β-H elimination on N-doped graphene to give 1,2-dihydro-
quinoline and N–H+ species, thus more H+/H− pairs were
obtained. Then the catalytic dehydrogenation of 1,2-dihydro-
quinoline to quinoline occurred, which was similar to the
dehydrogenation of 1,2,3,4-tetrahydroquinoline to 1,2-dihydro-
quinoline. Meanwhile, the catalytic hydrogenation of nitro-
benzene to nitrosobenzene by the H+/H− pairs can then regen-
erate the catalyst. In the same way, the H+/H− pairs will react
with nitrosobenzene and generate hydroxylamine. Finally, the
reaction occurred between the H+/H− pairs and hydroxylamine
quickly to give aniline as a product, and the active catalyst
would be regenerated for the next cycle.
Acknowledgements
This research was supported by the National Natural Science
Foundation of China (21968032 and 51563022), the
Fundamental Research Funds for the Central Universities
(31920180043) and the Fundamental Research Funds for
Central Universities-Innovation Team Cultivation Project
(31920190012 and 31920190015). The Scientific Research
Foundation of Northwest Minzu University (xbmuyjrc 201705).
Northwest Minzu University’s Double First-class and
Characteristic Development Guide Special Funds-Chemistry
Key Disciplines in Gansu Province (11080316). The Natural
Science Foundation of Gansu Province, China (2017GS10863).
References
1 G. Vernin and C. Parkanyi, Chemistry of heterocyclic com-
pounds in flavours and aromas, Ellis Horwood Chichester,
1982.
2 Z. Rappoport, The chemistry of anilines, John Wiley & Sons,
2007.
3 C. H. McAteer, R. Murugan and Y. V. S. Rao, Adv.
Heterocycl. Chem., 2017, 121, 173–205.
This journal is © The Royal Society of Chemistry 2020
Green Chem.

View Article Online
Paper
Green Chemistry
4 P. V. Thanikachalam, R. K. Maurya, V. Garg and V. Monga, 30 R. Labes, C. Mateos, C. Battilocchio, Y. Chen, P. Dingwall,
Eur. J. Med. Chem., 2019, 180, 562–612.
5 J. Barluenga, F. Rodriguez and F. J. Fananas, Chem. – Asian
J., 2009, 4, 1036–1048.
6 R. Sharma, P. Kour and A. Kumar, J. Chem. Sci., 2018, 130,
73.
7 F. Shi and X. Cui, Catalytic amination for N-alkyl amine syn-
thesis, Academic Press, 2018.
8 J. Song, Z. F. Huang, L. Pan, K. Li, X. Zhang, L. Wang and
J. J. Zou, Appl. Catal., B, 2018, 227, 386–408.
G. R. Cumming, J. A. Rincón, M. J. Nieves-Remacha and
S. V. Ley, Green Chem., 2019, 21, 59–63.
31 P. Anastas and N. Eghbali, Chem. Soc. Rev., 2010, 39, 301–
312.
32 Q. Sun, Z. Dai, X. Meng and F. S. Xiao, Chem. Soc. Rev.,
2015, 44, 6018–6034.
33 J. Jeevahan, M. Chandrasekaran, G. B. Joseph, R. Durairaj
and G. Mageshwaran, J. Coat. Technol. Res., 2018, 15, 231–
250.
9 E. A. Gelder, S. D. Jackson and C. M. Lok, Chem. Commun., 34 T. Li, J. Wang, F. Wang, L. Zhang, Y. Jiang, H. Arandiyan
2005, 522–524. and H. Li, ChemCatChem, 2019, 11, 1576–1586.
10 E. Vitaku, D. T. Smith and J. T. Njardarson, J. Med. Chem., 35 P. Sudarsanam, E. Peeters, E. V. Makshina, V. I. Parvulescu
2014, 57, 10257–10274. and B. F. Sels, Chem. Soc. Rev., 2019, 48, 2366–2421.
11 Y. Zhang, S. Pang, Z. Wei, H. Jiao, X. Dai, H. Wang and 36 Y. Wu, J. Feng, H. Gao, X. Feng and L. Jiang, Adv. Mater.,
F. Shi, Nat. Commun., 2018, 9, 1465.
2019, 31, 1800718.
12 L.
Z.
Sassykova,
K. Tashmukhambetova,
Y.
Aubakirov,
N.
S.
Sendilvelan, 37 P. Serp and J. L. Figueiredo, Carbon materials for catalysis,
Zhakirova
and
Wiley Online Library, 2009.
M. Faizullaeva, Orient. J. Chem., 2019, 35, 22–38.
13 Z. Wei, F. Shao and J. Wang, Chin. J. Catal., 2019, 40, 980–
1002.
38 D. S. Su, S. Perathoner and G. Centi, Catal. Today, 2012,
186, 1–6.
39 C. Hu, J. Qu, Y. Xiao, S. Zhao, H. Chen and L. Dai, ACS
Cent. Sci., 2019, 5, 389–408.
14 Y. Wu, Z. Chen, W. C. Cheong, C. Zhang, L. Zheng,
W. Yan, R. Yu, C. Chen and Y. Li, Chem. Sci., 2019, 10, 40 E. Llobet, Advanced Nanomaterials for Inexpensive Gas
5345–5352. Microsensors, Elsevier, 2020, pp. 55–84.
15 Q. Liu, S. Tadrent, C. Proust, F. Gomez, A. Khelfa, D. Luart 41 Y. Wang, D. Zhang, J. Deng, F. Zhou, Z. Duan, Q. Su and
and C. Len, Chem. Eng. Sci., 2020, 211, 115275. S. Pang, ACS Sustainable Chem. Eng., 2019, 7, 3883–3894.
16 A. B. Charette, Handbook of Reagents for Organic Synthesis: 42 C. Hu, Y. Mu, S. Bai, J. Yang, L. Gao, S. D. Cheng, S. B. Mi
Reagents for Heteroarene Functionalization, John Wiley &
Sons, 2015.
17 H. Goksu, H. Sert, B. Kilbas and F. Sen, Curr. Org. Chem.,
2017, 21, 794–820.
and J. Qiu, Carbon, 2019, 153, 609–616.
43 X. Dai, X. Cui, H. Yuan, Y. Deng and F. Shi, RSC Adv., 2015,
5, 7970–7975.
44 S. Pang and F. Shi, Tetrahedron Lett., 2016, 57, 5872–5876.
18 C. J. Li and B. M. Trost, Proc. Natl. Acad. Sci. U. S. A., 2008, 45 R. Ghosh Chaudhuri and S. Paria, Chem. Rev., 2011, 112,
105, 13197–13202. 2373–2433.
19 G. Brieger and T. J. Nestrick, Chem. Rev., 1974, 74, 567–580. 46 D. Zhou, S. W. Bennett and A. A. Keller, PLoS One, 2012, 7,
20 R. A. Johnstone, A. H. Wilby and I. D. Entwistle, Chem.
Rev., 1985, 85, 129–170.
e37363.
47 R. N. Wenzel, Ind. Eng. Chem. Res., 1936, 28, 988–994.
21 S. Bähn, S. Imm, L. Neubert, M. Zhang, H. Neumann and 48 X. M. Li, D. Reinhoudt and M. Crego-Calama, Chem. Soc.
M. Beller, ChemCatChem, 2011, 3, 1853–1864. Rev., 2007, 36, 1350–1368.
22 J. Choi, A. H. R. MacArthur, M. Brookhart and 49 G. Wen, Z. Guo and W. Liu, Nanoscale, 2017, 9, 3338–
A. S. Goldman, Chem. Rev., 2011, 111, 1761–1779. 3366.
23 A. Corma, J. Navas and M. J. Sabater, Chem. Rev., 2018, 118, 50 R. A. Sheldon and H. Van Bekkum, Fine chemicals through
1410–1459. heterogeneous catalysis, John Wiley & Sons, 2008.
24 P. B. Vásquez, T. Tabanelli, E. Monti, S. Albonetti, 51 G. Carja, R. Nakamura, T. Aida and H. Niiyama,
D. Bonincontro, N. Dimitratos and F. Cavani, ACS
Sustainable Chem. Eng., 2019, 7, 8317–8330.
25 J. Wang, R. Nie, L. Xu, X. Lyu and X. Lu, Green Chem., 2019,
21, 314–320.
Microporous Mesoporous Mater., 2001, 47, 275–284.
52 G. Zou, H. Hou, J. Hu and X. Ji, Batteries Supercaps, 2019, 2,
712–722.
53 Y. Xu, W. Tu, B. Zhang, S. Yin, Y. Huang, M. Kraft and
R. Xu, Adv. Mater., 2017, 29, 1605957.
26 M. Xiao, X. Yue, R. Xu, W. Tang, D. Xue, C. Li, M. Lei,
J. Xiao and C. Wang, Angew. Chem., Int. Ed., 2019, 58, 54 R. Kumar, R. K. Singh, P. K. Dubey, D. P. Singh,
10528–10536.
R. M. Yadav and R. S. Tiwari, Adv. Mater. Interfaces, 2015, 2,
1500191.
55 A. Misra, P. K. Tyagi, M. K. Singh and D. Misra, Diamond
Relat. Mater., 2006, 15, 385–388.
56 A. Dillon, T. Gennett, J. Alleman, K. Jones, P. Parilla and
M. Heben, Proc. 2000 DOE/NREL Hydrogen program
review, 2000, pp. 8–10.
27 C. Yao, T. Dahmen, A. Gansäuer and J. Norton, Science,
2019, 364, 764–767.
28 S. Pang, Y. Zhang, Y. Huang, H. Yuan and F. Shi, Catal. Sci.
Technol., 2017, 7, 2170–2182.
29 M. Wang, F. Wang, J. Ma, C. Chen, S. Shi and J. Xu, Chem.
Commun., 2013, 49, 6623–6625.
Green Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Green Chemistry
Paper
57 X. Cui, Y. Li, S. Bachmann, M. Scalone, A. E. Surkus, 72 D. Yang, A. Velamakanni, G. Bozoklu, S. Park, M. Stoller,
K. Junge, C. Topf and M. Beller, J. Am. Chem. Soc., 2015,
137, 10652–10658.
R. D. Piner, S. Stankovich, I. Jung, D. A. Field and
C. A. Ventrice Jr., Carbon, 2009, 47, 145–152.
58 J. Kaufman, S. Metin and D. Saperstein, Phys. Rev. B: 73 R. Joyner and M. Roberts, Chem. Phys. Lett., 1979, 60, 459–
Condens. Matter Mater. Phys., 1989, 39, 13053. 462.
59 Y. Yap, S. Kida, T. Aoyama, Y. Mori and T. Sasaki, Appl. 74 J. J. Pireaux, M. Liehr, P. Thiry, J. P. Delrue and
Phys. Lett., 1998, 73, 915–917. R. Caudano, Surf. Sci., 1984, 141, 221–232.
60 Y. Jia, B. Xiao and K. Thomas, Langmuir, 2002, 18, 470–478. 75 A. Boronin, S. Koscheev, K. Murzakhmetov, V. Avdeev and
61 D. P. Dick, H. Knicker, L. G. Ávila, A. V. Inda Jr., E. Giasson
and C. A. Bissani, Org. Geochem., 2006, 37, 1537–1545.
62 Z. Jiang, J. Xie, D. Jiang, X. Wei and M. Chen,
CrystEngComm, 2013, 15, 560–569.
63 G. I. Danmaliki, T. A. Saleh and A. A. Shamsuddeen, Chem.
Eng. J., 2017, 313, 993–1003.
G. Zhidomirov, Appl. Surf. Sci., 2000, 165, 9–14.
76 W. Yang, J. Li, Q. Fu, L. Zhang, X. Zhu and Q. Liao,
Bioresour. Technol., 2017, 241, 325–331.
77 W. Qiu, J. Jiao, J. Xia, H. Zhong and L. Chen, Chem. – Eur.
J., 2015, 21, 4359–4367.
78 G. Mestl, N. I. Maksimova, N. Keller, V. V. Roddatis and
R. Schlögl, Angew. Chem., Int. Ed., 2001, 40, 2066–2068.
64 O. Beyssac, B. Goffé, J. P. Petitet, E. Froigneux, M. Moreau
and J. N. Rouzaud, Spectrochim. Acta, Part A, 2003, 59, 79 V. Strelko, N. Kartel, I. Dukhno, V. Kuts, R. Clarkson and
2267–2276. B. Odintsov, Surf. Sci., 2004, 548, 281–290.
65 I. Y. Jeon, H. J. Choi, M. J. Ju, I. T. Choi, K. Lim, J. Ko, 80 C. Chen, J. Zhang, B. Zhang, C. Yu, F. Peng and D. Su,
H. K. Kim, J. C. Kim, J. J. Lee and D. Shin, Sci. Rep., 2013,
3, 2260.
66 S. Y. Lu, M. Jin, Y. Zhang, Y. B. Niu, J. C. Gao and C. M. Li,
Adv. Energy Mater., 2018, 8, 1702545.
Chem. Commun., 2013, 49, 8151–8153.
81 L. P. Mei, R. Wang, P. Song, J. J. Feng, Z. G. Wang,
J. R. Chen and A. J. Wang, New J. Chem., 2016, 40, 2315–
2320.
67 J. Rafiee, X. Mi, H. Gullapalli, A. V. Thomas, F. Yavari, 82 A. Van Velsen, G. Van der Vos, R. Boersma and J. De
Y. Shi, P. M. Ajayan and N. A. Koratkar, Nat. Mater., 2012,
11, 217.
Reuver, Water Sci. Technol., 1991, 24, 195–203.
83 W. Knight and M. Sodhi, CIRP Ann., 2000, 49, 83–86.
68 W. J. Jiang, J. S. Hu, X. Zhang, Y. Jiang, B. B. Yu, Z. D. Wei 84 C. Zhang, J. Lu, M. Li, Y. Wang, Z. Zhang, H. Chen and
and L. J. Wan, J. Mater. Chem. A, 2014, 2, 10154–10160. F. Wang, Green Chem., 2016, 18, 2435–2442.
69 X. Xu, H. Li and Y. Wang, ChemCatChem, 2014, 6, 3328– 85 S. Fukuzumi and M. Chiba, J. Chem. Soc., Perkin Trans. 2,
3332. 1991, 1393–1398.
70 W. Shi, B. Zhang, Y. Lin, Q. Wang, Q. Zhang and D. S. Su, 86 Y. Sheng, X. Wang, Z. Xing, X. Chen, X. Zou and X. Lu, ACS
ACS Catal., 2016, 6, 7844–7854. Sustainable Chem. Eng., 2019, 9, 8908–8916.
71 P. Chen, L. K. Wang, G. Wang, M. R. Gao, J. Ge, W. J. Yuan, 87 A. Noujima, T. Mitsudome, T. Mizugaki, K. Jitsukawa and
Y.-H. Shen, A. J. Xie and S. H. Yu, Energy Environ. Sci., 2014,
7, 4095–4103.
K. Kaneda, Angew. Chem., Int. Ed., 2011, 50, 2986–
2989.
This journal is © The Royal Society of Chemistry 2020
Green Chem.