Metal-free carbocatalyst for catalytic hydrogenation of N-containing unsaturated compounds

Article abstract of DOI:10.1016/j.jcat.2019.07.017

Eco-friendly carbocatalyst is extremely desirable in organic catalysis and wastewater treatment. Herein, we reported a one-pot synchronous hydrothermal process to synthesize a highly efficient N doped holey graphene (NHG) carbocatalyst, in which graphene

Full text of DOI:10.1016/j.jcat.2019.07.017

Journal of Catalysis 377 (2019) 199–208
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Metal-free carbocatalyst for catalytic hydrogenation of N-containing
unsaturated compounds
a
a
b
a
a
a
b
b
Zhaolin He , Jin Liu , Qijun Wang , Meng Zhao , Zhipan Wen , Jun Chen , Devaraj Manoj , Chuyi Xie ,
a,
⇑
a,
⇑
c
a
b
Jiangbo Xi , Junxia Yu , Chunyan Tang , Zhengwu Bai , Shuai Wang
a
School of Chemistry and Environmental Engineering, Wuhan Institute of Technology, Wuhan 430073, China
Key Laboratory of Material Chemistry for Energy Conversion and Storage, Ministry of Education, School of Chemistry and Chemical Engineering, Huazhong University of Science
and Technology, Wuhan 430074, China
b
c
School of Computer, Electronics and Information in Guangxi University, Nanning, 530004 Guangxi, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Eco-friendly carbocatalyst is extremely desirable in organic catalysis and wastewater treatment. Herein,
we reported a one-pot synchronous hydrothermal process to synthesize a highly efficient N doped holey
graphene (NHG) carbocatalyst, in which graphene oxide nanosheets are self-assembled, reduced, chem-
ically etched, and simultaneously doped with nitrogen to form a graphene hydrogel in the presence of
Received 14 May 2019
Revised 7 July 2019
Accepted 10 July 2019
2 2 4
H O and NH OH. Benefiting from the good hydrophilicity, unique hierarchical porous structure, large
2
ꢀ1
specific surface area (322.1 m g ), and high N content (9.77 at. %), the resultant NHG carbocatalyst exhi-
bits an excellent catalytic performance for the hydrogenation of N-containing unsaturated compounds. In
Keywords:
Carbocatalyst
Nitrogen doped holey graphene
Hydrogenation
Organic dye
ꢀ2
4
-nitrophenol reduction reaction, the NHG carbocatalyst delivers a turnover frequency of 3.32 ꢁ10
-
ꢀ1
min . This catalytic activity is superior to that of the commercial Pd/C (5.0 wt%), other previously
reported carbocatalysts, and many noble-metal-based catalysts. Furthermore, the molecular structure
analysis combined with density functional theory calculations elucidate metal-free catalytic mechanism
of hydrogenation reaction for unsaturated chromophore groups in N-containing organic dyes.
Ó 2019 Elsevier Inc. All rights reserved.
Density functional theory
1
. Introduction
As a major source of water pollution, toxic organic dyes from
opment [14]. In order to solve above-mentioned issues associated
with metal-based catalyst, exploring and developing novel metal-
free materials as alternatives to metal-based catalysts, is of grow-
ing scientific and technological importance [15–17]. Moreover,
metal-free catalysts can provide more beneficial effects as of
non-secondary contamination for environmental catalysis, espe-
cially in water/wastewater remediation [18]. In this regard, carbo-
catalysts have captured profound attention in metal-free catalysis
due to their wide availability, environmental acceptability, low-
cost, stability/recyclability, and sustainable properties [19,20].
Unfortunately, environmental catalysis process of carbocatalyst
have not been intensively investigated, and therefore related
research is still in the early stages of development [21,22].
As one of the most promising carbocatalysts, graphene-based
materials are endowed with excellent catalytic performance due
to their large specific surface area, high conductivity as well as
excellent corrosion resistant property. Particularly, their catalytic
activity can be dramatically enhanced by doping foreign atoms into
the graphene lattice and thus modulate the electronic and chemi-
cal properties [23,24]. Chen and coworkers reported the metal-free
catalytic reduction of 4-nitrophenol (4-NP) to 4-aminophenol
manufacturing process, such as textiles, foods, drugs and cosmetic
industry activities, have become a global concern due to their non-
biodegradation, environmental persistence and bio-accumulation,
thus leading to adverse ecological impact in living organisms
[
1–3]. Organic synthetic dyes have reactive functional groups such
as nitro- (O N@O), azo (AN@NA) and imine (AC@NA) moieties
along with heterocyclic and aromatic rings in their structures.
Therefore, it is necessary to cleave these unsaturated chromophore
groups of these toxic dyes by decolorization, and finally to convert
them into nontoxic forms [4–6]. In environmental catalysis indus-
try, most of the homogeneous and heterogeneous catalytic conver-
sions are mainly based on either noble [7,8], transition [9–11] and
rare-earth metals or their oxides [12,13]. However, these metal-
based catalysts have numerous drawbacks, such as scarcity, high
cost, deactivation, and secondary pollution in the disposal of waste
catalysts, which raised the questions in terms of sustainable devel-
⇑
Corresponding authors.
E-mail addresses: jbxi@wit.edu.cn (J. Xi), yujunxia_1979@163.com (J. Yu).
(
4-AP) mediated by N doped graphene (NG). Their theoretical
https://doi.org/10.1016/j.jcat.2019.07.017
021-9517/Ó 2019 Elsevier Inc. All rights reserved.
0

2
00
Z. He et al. / Journal of Catalysis 377 (2019) 199–208
calculations further demonstrated that the C atoms next to the
doped N atoms can be activated, leading to a higher positive charge
density [25]. The drastic enhancement in catalytic behavior is attri-
butable to the electronegativity of doped heteroatoms, which
break carbon’s electroneutrality via induced effect, and conse-
quently create catalytic sites. In our previous work, the reduction
reaction of 4-NP catalyzed by N/P dual doped graphene (NPG)
was also examined. Attributed to the co-doping of N and P atoms,
the resultant NPG carbocatalyst exhibited excellent catalytic per-
formance for 4-NP reduction. In addition, the remarkable catalytic
activity of NPG can also be extended to the reduction of various
substituted nitrobenzenes [26] Generally, heteroatoms are usually
introduced into graphene frameworks through post-treatment of
graphene-base materials (eg. Graphene oxide (GO), exfoliated gra-
phene) with heteroatom-containing precursors [27,28]. In fact, the
homogeneous doping of heteroatoms in graphitic structure
remains a formidable challenge, because it is difficult to accurately
control the dopant content and doping sites in graphene. As a rule,
the amount of catalytic sites and the accessibility of the reactants
to active sites are essential for activity of the catalyst. Therefore,
the graphene materials with high doping content and porous struc-
ture are highly desirable, because they can offer high density of
exposed active surface and diffusion channel for efficient catalytic
processes.
for SEM characterization were prepared by drop-casting the sam-
ple suspensions (dispersed in ethanol, 1 mg/mL) on pre-cleaned
silicon wafers. For TEM, 5 lL of the sample suspensions (dispersed
in ethanol, 0.5 mg/mL) were drop-casted on a carbon-coated cop-
per grid. X-ray photoelectron spectroscopy (XPS) was performed
with an ESCALAB MKII spectrometer (VG Co., UK), using Mg K
a
ꢀ10
radiation (1253.6 eV) at a pressure of 2.0 ꢁ10
mbar. The peak
positions were internally referenced to the C 1s peak at 284.6 eV.
Raman spectrum was measured by a confocal laser micro-Raman
spectrometer (DXR, USA) equipped with a He–Ne laser of excita-
tion of 532 nm at a laser power of 0.6 mW. Nitrogen adsorption/
desorption isotherms were obtained at 77 K on an accelerated sur-
face area and porosimetry system (ASAP 2020, Micromeritics, USA)
to measure the surface area of the material using the Brunauer-
Emmett-Teller (BET) method. The UV–vis measurements of the
synthesized NHG carbocatalyst along with the time-dependent
kinetic spectra during catalysis were performed on a UV-2550
spectrophotometer (Shimadzu, Japan). High-performance liquid
chromatography (HPLC) analysis was performed on an Agilent-
1100 system with a Zorbax Eclipse XDB-C18 4.6 ꢁ150 mm column
(Agilent, USA). The MALDI–ultra-high resolution MS of reduced
product was performed using a solariX 7.0 T FTICR mass spectrom-
eter (Bruker Daltonics, Germany). The reduced product was deter-
1
mined by the integrals in H NMR spectrum which was measured
In this work, we describe the development of an efficient N
doped holey graphene (NHG) carbocatalyst, using GO as a starting
with a NMR 400 MHz spectrometer (Varian, USA). The sample
solution (20 mg/ml) of reduced product was prepared with deuter-
material, hydrogen peroxide (H
2
O
2
) as etching agent and ammo-
3
ated CDCl as the solvent.
nium hydroxide (NH OH) as nitrogen sources via a facile one-pot
4
hydrothermal synthesis route. Due to high N doping content and
porous structure, the as-prepared NHG demonstrates enhanced
catalytic activity and excellent dispersible property in water. In
aqueous solution, NHG is employed as an efficient carbocatalyst
and exhibits an excellent organocatalytic activity, high stability,
selectivity and good recyclability for the hydrogenation of organic
dyes and nitrobenzenes. Furthermore, the molecular structure
analysis combined with density functional theory (DFT) calcula-
tions elucidate metal-free catalytic mechanism of hydrogenation
reaction for unsaturated chromophore groups in N-containing
organic dyes.
2
.3. Synthesis of NHG carbocatalyst
GO aqueous solution was initially prepared by the modified
Hummers method [29]. As-prepared GO solution (30 mL, 8.0 wt%)
was diluted by 30 mL water by a sonication process. To the above
prepared mixture, 15 mL of NH
4 2 2
OH (28 wt%) and 7 mL of H O
(
0.3 wt%) was injected rapidly under stirring. Subsequently, the
mixture was transferred into a 100 mL Teflon-lined autoclave
and heated at 180 °C for 8 h. During this hydrothermal process,
GO nanosheets were reduced and self-assembled to form a mono-
lithic hydrogel. At the same time, reduce GO nanosheet was also
2
. Experimental section
etched by H
framework. The obtained NHG hydrogel was soaked in deionized
water and washed for several times to remove residual NH OH.
2 2
O , and was doped with N atoms in the graphene
2.1. Materials
4
Finally, the NHG hydrogel was immersed in ca. 100 mL water, fol-
lowed by pulverization via an ultrasonic disintegrator for a period
of 3 h, and a homogeneous NHG suspension can be obtained. The
concentration of the NHG suspension was measured via freeze-
drying before use. For comparison, NG and holey graphene (HG)
Ammonium hydroxide (NH
4
OH), hydrogen peroxide (H
2 2
O ),
Nitrobenzene,
4-nitrotoluene, 4-chloronitrobenzene,
4
4
-bromonitrobenzene, 4-nitrobenzoic acid, 4-nitrobenzonitrile,
-nitrophenol (99%) and sodium borohydride (96%) were pur-
chased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai,
China). Pd/C (5 wt%) were purchased from Aladdin Chemistry Co.,
Ltd. (China)., methylene blue (MB), methylene orange (MO),
rhodamine B (RhB) and Congo red (CR) were purchased from
Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China), and the
rest of the chemicals were procured from Sigma-Aldrich. Deionized
were also prepared by the same process without the adding H
and NH OH, respectively.
2 2
O
4
2.4. Catalytic reduction of N-containing unsaturated compounds
ꢀ1
water (resistivity > 18
experiments.
X
ꢂcm ) was used for all syntheses and
In
2.7 mg/mL) was injected into 3 mL aqueous solution with 4-NP
20 mM) and NaBH (2 M). Then the reaction mixture was
a typical catalysis reaction, 1 mL of NHG suspension
(
(
4
2
.2. Instrumentation
thoroughly mixed under magnetic stirring at room temperature.
The reaction process was monitored by thin-layer chromatography
(TLC) in regular intervals. After completion of the reaction, the
reaction solution was analyzed with UV/Vis detection or HPLC
measurement after filtering off the catalyst. Decrease in extinction
intensity with time was measured at maximum absorbance values
(k max), the k max = 400 nm, 465 nm, 498 nm, 554 nm and 665 nm
for 4-NP, MO, CR, RhB and MB, respectively.
The morphology of the synthesized NHG catalyst was character-
ized using a FEI Nova NanoSEM 450 scanning electron microscope
SEM). Transmission electron microscopy (TEM) and high-
(
resolution transmission electron microscopy (HRTEM) images
were obtained using a TECNAI G2 20 U-Twin instrument (Nether-
lands) operated at an acceleration voltage of 200 kV. The samples

Z. He et al. / Journal of Catalysis 377 (2019) 199–208
201
2
.5. Computational methodology
a black monolithic graphene hydrogel (Fig. 1B). At the same time,
GO was also reduced, chemically etched to form pore openings
on conjugated carbon surface i.e., holey reduced graphene oxide
(RGO) and simultaneously doped with nitrogen. As can be seen
from the SEM image, the NHG hydrogel exhibit a 3D network with
highly interconnected sheets, which are interlocked together and
prevented them from restacking to maintain a highly macroporous
framework (Fig. 1C and 1D).
All the DFT calculations of the 5 organic dyes were carried out
by using Gaussian 09 program package [30]. The molecular struc-
tures of these 5 dye molecules were fully optimized in vacuum
by a standard 6-31G (d, p) set with B3LYP functional without con-
sidering the solvent. C, H, N, O, S and Na were described by C, H, N,
O, S and Na [31–33]. Vibrational frequencies were calculated for all
stationary points at the same level to verify whether the each was a
minimum (NIMAG = 0) on the potential energy surface.
The fine structural properties were further characterized by
TEM and dark field scanning TEM (DF-STEM) images. Fig. 2A and
2
B indicate the existence of small pores in the surface of graphene
nanosheets. From higher magnification TEM image, it further con-
firms the presence of tiny pores with 2–4 nm in size, and are dis-
tributed on the basal plane of graphene nanosheets (Fig. 2C).
There is an obvious contrast between the mesoporous surface in
NHG and intact ones of hydrothermally treated graphene without
3
. Results and discussion
3.1. Synthesis and characterizations of materials
The preparation involves a facile one-pot method by hydrother-
2
H O
2
(Fig. 2D), indicating the successful etching of graphene sheets
mal treatment of GO in the presence of NH
During this process, GO nanosheets were self-assembled to realize
4 2 2
OH and H O (Fig. 1A).
on the surface. The distributions of C, O and N in NHG are almost
Fig. 1. (A) Schematic illustration of the fabrication of the NHG. (B) Photograph of NHG hydrogel and aqueous dispersion; (C) The typical SEM image of the NHG hydrogel
bottom left) reveals a 3D porous configuration with interconnected pores ranging from sub-micrometers to several micrometers; and (D) High resolution SEM image of NHG.
(

2
02
Z. He et al. / Journal of Catalysis 377 (2019) 199–208
Fig. 2. (A) High-magnification bright-field TEM image of NHG; (B) DF-STEM image of NHG. (C) TEM images of NHG in comparison to (D) pore-free NG. (E) STEM image and
elemental mapping of C, O, and N for NHG.
overlap with each other according to the element mappings in
Fig. 2E, confirming the homogeneous doping of N atomic species
in NHG.
facilitates the doping of heteroatoms [34]. As depicted in Fig. 3B,
there are three components in the C 1s spectrum of the HNG.
The strong signal at 285.0 eV can be assigned to the graphite-like
2
To gain further insight of content and chemical status of ele-
ments in NHG, XPS analysis were carried out. As can be seen in
Fig. 3A, the peaks located at ca. 285, 400, and 532 eV correspond
sp carbon, indicating most of the C atoms in the HNG are arranged
in a conjugated honeycomb lattice even when the doped N is high
up to 9.77 at. %. The weak peaks at 285.7 and 287.4 eV reflect
different bonding structure of the C-N bonds, corresponding to
2
to sp hybridized carbon 1 s, dopant nitrogen 1 s, and oxygen 1 s,
2
3
respectively, which is in accordance with the elemental mapping
analysis. The atomic percentage of N in the HNG is calculated to
be 9.77 at. %, which is much higher than that in pore-free NG
the N-sp C and N-sp C bonds, respectively, and would originate
from either substitution of N atoms, defects or the edge of the gra-
phene nanosheets [35]. The N 1s peak can be deconvoluted into
four single peaks, indicating that N atoms are in the four different
bonding characters doped into the graphene framework (Fig. 3C).
(
6.13 at. %) as shown in Fig. 3A Inset. This high doping degree
should be attributed to the defects of the holey structure that

Z. He et al. / Journal of Catalysis 377 (2019) 199–208
203
Fig. 3. (A) XPS survey spectrum of NHG and NG, the according elemental analysis is shown in the inset. (B) High-resolution XPS spectrum of C 1s, and (C) High-resolution XPS
spectrum of the N 1s peak, (D) Raman spectra of NHG in comparison to N-free HG and pore-free and RGO.
2
ꢀ1
These four kinds of doped N atoms are located at ca. 398.7, 399.6,
00.4 and 401.6 eV, corresponding to pyridinic, amine, pyrrolic and
graphitic configurations, respectively [25]. The peak intensity for
(108.9 m g ) (Fig. 4C). In addition, these two samples possess a
mesoporous structure according to the pore-size distribution
curves (Fig. 4B and 4D). Meanwhile, the curves in Fig. 4B demon-
strate that HNG possess a wider pore-size distribution (2–4 nm)
compare with NG (centered at ca. 4 nm) (Fig. 4D). This observa-
tions are also confirmed by HRTEM images as shown in Fig. 2.
Therefore, the higher specific surface area and wider pore-size dis-
tribution of HNG are attributed to the holey structure coursed by
4
‘
‘
‘pyridinic”, ‘‘amine” and ‘‘pyrrolic” N is much higher than that of
‘graphitic” N, thus it clearly evidences N atoms should be mainly
substitutionally doped into the graphene lattice at the edge of gra-
phene nonosheets and inner boundary of holes in holey graphene,
where oxygen-containing functional groups are rich. This result is
in consistent with Dai group’s findings that N-doping depend on
the amounts of these oxygen functional groups at the defect and
edge sites of graphene using ammonia as nitrogen source [36].
Raman spectroscopy measurements have also been applied to
investigate the structural defects in HNG and control samples.
2 2
H O etching. These marcoporous structure and mesoporous gra-
phene nanosheets are expected to facilitate the exposure of active
sites, enhance the diffusion of reactants in organocatalysis process.
3.2. Evaluation of catalytic performance of synthesized NHG toward
the hydrogenation of organic dyes
ꢀ1
The typical D band located at around 1350 cm representing the
edges, defects and disordered carbon sites, and the G band cen-
ꢀ1
tered at 1580 cm
motion of pairs in sp -C atoms [37]. Therefore, in the Raman spec-
tra, a larger ratio of the D band and G band (I /I ) intensity for gra-
assigning to the in-plane bond-stretching
The catalytic performance of NHG was further evaluated by
investigating the hydrogenation reaction of five common organic
dyes (4-NP, MO, CR, RhB and MB) in aqueous solution, using NaBH4
as a reducing agent at room temperature (Fig. 5A and Fig. S1). Due
to the excellent hydrophilic nature, NHG can be well dispersed in
aqueous media (Fig. 1B), consequently leading to sufficient expo-
sure and easier access of active sites to the reactants during the
reaction. The catalytic decolorization processes of these dyes were
monitored by the color bleaching of dye solution after the addition
of an excess amount of NaBH4 and NHG carbocatalysts, as indi-
cated by the gradual decrease in the maximum absorbance values
with time in the UV–vis spectra [5,39,40]. As shown in Fig. 5B-F,
the adsorption peak at k max gradually decreases with respect to
reaction time, indicating the reduction of these 5 dyes occur in
the presence of NHG carbocatalyst (40 s, 9 min, 2 min, 3 min, and
2
D
G
phene materials usually indicates a higher content of disordered
carbon and defect level. As shown in Fig. 3D, the intensity ratio
D G
(I ) for HNG, HG, NG and RGO are calculated as 1.12, 1.08,
/I
1
.06 and 1.01, respectively, the higher intensity ratio observed at
HNG is attributed to the holey structure and N doping in graphene
framework. These intrinsic properties may further interrupted the
graphitic carbon configuration as well as electronic structure of
graphene and result in high dense of defective sites [38].
2
N adsorption-desorption isotherms and pore-size distribution
of NHG and pore-free NG are shown in Fig. 4. The as-prepared
HNG (freeze-dried) exhibits a specific surface area (SSA) of
2
ꢀ1
3
22.1 m g
(Fig. 4A), much higher than that of NG

2
04
Z. He et al. / Journal of Catalysis 377 (2019) 199–208
Fig. 4. (A) Nitrogen adsorption/desorption isotherms of the NHG and (B) DFT pore-size distribution of NHG. (C) Nitrogen adsorption/desorption isotherms of the pore-free NG
and (D) DFT pore-size distribution of NG.
1
1 min for 4-NP, MO, CR, RhB and MB, respectively). To clarify the
subsequently. Notably, a 92% conversion was maintained even in
the eighth run (Fig. S4), revealing an excellent durability of NHG
carbocatalyst.
catalytic effect of NHG, our extensive detection excludes the
presence of adsorbed dyes in the ethanol extraction solution of
used NHG carbocatalyst. No absorbed dye was extracted, further
confirming the complete decolorization. Herein, the reduction of
TOF ðminꢀ Þ¼
1
mmoles of 4 ꢀNP con
verted per minute
mg of catalyst
4
-NP is chosen as a model reaction to evaluate the catalytic activity
ðCalculation formula of TOFÞ
of NHG. The bright yellow colored aqueous mixture of 4-NP/NaBH
4
presents a maximum absorption band at 400 nm due to the
formation of dark 4-nitrophenolate ions [41]. The color of
3
.3. Evaluation of catalytic performance of NHG carbocatalyst toward
4
-nitrophenolate completely fade within 40 s (Fig. 5B). Addition-
2
nitrobenzenes (Ar-NO ) reduction
ally to the absorption band at 300 nm, the reaction solution pre-
sents an absorption band at 230 nm associated to the aromatic
group, corresponding to the formation of 4-AP [42]. A high excess
On the basis of the results, we further extended the NHG carbo-
catalyst to the reduction reactions of various substituted nitroben-
zenes to examine its common applicability. Due to the poor
solubility of nitrobenzene or substituted nitrobenzenes in water,
4
of NaBH was employed to ensure that its concentration remained
essentially constant during the whole reaction, which allowed the
assumption of pseudo-first-order kinetics with respect to the
substrates. Thus the catalytic efficiency can calculated by turnover
frequency (TOF), which is defined as the amount of 4-NP molecules
that 1 mg catalyst can convert into 4-AP per min (calculation
formula). As a result, the TOF of NHG is calculated to be
H
2
O/ethanol mixture was used as their solvent for the catalytic
process. The According to the data are summarized in Table 2,
nitrobenzenes can be reduced into anilines by NaBH in H O/etha-
nol mixture in the presence of NHG carbocatalyst. It should be
noted that the reaction efficiency in H O/ethanol mixture was
4
2
2
ꢀ
2
3
.32 ꢁ10 , which is superior to that of other carbocatalysts, com-
lower than that in water. As a result, the different molecular struc-
ture and reaction conditions should be ascribe to the different
behavior of reduction of 4-NP (40 s) compared to all others
nitrobenzenes systems (5–20 min).
mercial Pd/C (5.0 wt%), and many noble-metal-based catalysts
reported previously (Table 1). By comparison, NG and HG were also
used as catalysts for 4-NP reduction. 4-NP aqueous solutions
(
0.06 mmol) can be completely reduced within 80 s in the presence
of NG (Fig. S2), while HG just exhibited a low adsorption capacity
and no 4-AP was detected from UV/Vis analysis (Fig. S3). These
findings demonstrate that the enhanced catalytic activity should
originate from N doping and hierarchical porous structure of the
graphene frameworks. Furthermore, the recyclability of the NHG
carbocatalyst was also investigated by repeated uses of the recy-
cled catalyst in the same condition. In each cycle, NHG was recov-
ered from the reaction mixture by filtration via a disassembling
filter and subsequently washed with ethanol and deionized water
3
.4. Mechanism investigation of hydrogenation azo (AN@NA) and
imine (AC@NA) catalyzed by NHG carbocatalyst
Based on these considerations, the catalytic centers of NHG in
activation of hydrogenation should originate from the created
defective sites due to nitrogen atoms are incorporated into the por-
ous graphene framework. The existence of nanoholes in the basal
plane of NHG not only provides a ‘‘short-cut” for efficient mass
transport, but also provides more exposed active sites due to

Z. He et al. / Journal of Catalysis 377 (2019) 199–208
205
Fig. 5. (A) Photograph of the five common organic dyes (4-NP, MO, CR, RhB and MB) in aqueous solution; Time-dependent absorption spectra showing the hydrogenation of
-NP (B), MO (B), CR (C), RhB (D) and MB (F) using NHG as metal-free catalyst in the presence of NaBH , insets are structural formula of the corresponding dye molecules. The
4
4
inset of Fig. 5B is the photograph of the reduced 4-NP solution.
increased edges [53]. In addition, we explored its innovative appli-
cation in the catalytic hydrogenation reaction of azo (AN@NA),
imine (AC@NA). These two hydrogenation reactions catalyzed by
metal-free carbocatalyst has never been reported previously.
Therefore, little attention has been devoted to the metal-free cat-
alytic mechanism of hydrogenation reaction for azo and imine.
To gain further understood about the metal-free catalytic process,
the reaction mechanism is elucidated by molecular structure anal-
ysis and DFT calculations. The molecule structure of hydrogenation
product of MO was identified from mass and 1H NMR spectra. As
shown in Figs. S5 and S6, N, N-dimethyl-p-phenylenediamine
agreement with previously reported work [54]. Similar
phenomenon was also observed for the RhB because of the same
chromophore group as MB was contained in RhB molecular struc-
ture. Therefore, we failed to analyze the molecule structures of
hydrogenated MB and hydrogenated RhB.
In order to elucidate the catalytic effect on dyes hydrogenation
of NHG, DFT along with time dependent DFT calculations were per-
formed to investigate the density of states (DOS), charge density,
electrostatic potential distribution and UV–vis spectra. As can be
seen from Fig. 6A, compared with NG, NHG shows a significantly
increase of DOS near the Fermi level due to its holey structure. This
will bring NHG a superior electrical conductivity to NG, which can
facilitate the electron transfer and finally improve the hydrogena-
tion activity of NHG. The charge density mapping diagram shows
that N atom possess a high charge density (inset in Fig. 6A), which
play an important role to create catalytically active sites in their
neighboring C atoms. Therefore, the activated C sites bear more
8 12 2
(C H N ) was obtained after reduction, indicating the cleavage
of AN@NA bond in MO and resulted in formation of amino-
group. In addition, we also tried to obtain the products of hydro-
genated imine via extraction. Unfortunately, colorless Leuco MB
is very readily oxidized back to blue MB by oxygen in the process
of rotary evaporation even at room temperature, which is in

2
06
Z. He et al. / Journal of Catalysis 377 (2019) 199–208
Table 1
The comparison of catalytic activity of NHG and other metal-free carbocatalysts or noble metal based catalysts for 4-NP reduction.
Catalyst
Mass of catalyst (mg)
Amount of 4-NP (mmol)
Conversion time (min)
TOF (mmol 4-NP/(mg cat.ꢂmin)
Ref.
6 ꢁ10ꢀ
2
0.67
1.33
1.10
2.53
10
120
21
80
3.32 ꢁ10
ꢀ2
a
a
a
NHG
NG
Pd/C (wt.5%)
N,P dual-doped graphene
S, N co-doped carbon nanotubes
N-doped graphene
N-doped graphene
AA/GO
2.7
2.0
2.0
1.0
0.0093
2.0
0.137
2.5
2.5
2
1.0
492
1
2
2
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
2
2
2
4
1
ꢀ2
6 ꢁ10
6 ꢁ10
6 ꢁ10
2 ꢁ10
5 ꢁ10
2.26 ꢁ10
ꢀ2
2.73 ꢁ10
ꢀ2
2.37 ꢁ10
[26]
[43]
[44]
[25]
[45]
[46]
[47]
[9]
[48]
[49]
[50]
[51]
[52]
ꢀ3
2.15 ꢁ10
ꢀ3
2.08 ꢁ10
ꢀ
4
ꢀ5
1.76 ꢁ10
6.05 ꢁ10
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
4
4
2
2
2
4
2
2
4
ꢀ6
5 ꢁ10
5 ꢁ10
4 ꢁ10
3 ꢁ10
1 ꢁ10
6 ꢁ10
3 ꢁ10
6 ꢁ10
2.5 ꢁ10
ꢀ6
pH-mediated graphene
60
3.33 ꢁ10
ꢀ2
3
Fe O
4
@P(MBAAm-co-MAA)@Ag
0.75
1.65
0.25
0.67
1.92
0.67
0.33
2.67 ꢁ10
ꢀ2
RGO@AC/Pd
CMF@PDA/Pd
NCT@Pd
Fe@NC@Pd
ASNTs@Pd
Pd-rGO-CNT
1.8 ꢁ10
ꢀ5
8.13 ꢁ10
ꢀ4
9 ꢁ10
ꢀ3
7.8 ꢁ10
ꢀ2
4.48 ꢁ10
3 ꢁ10ꢀ
3.03 ꢁ10
ꢀ4
5
aThis work.
Table 2
Catalytic reduction of nitrobenzene and substituted nitrobenzenes into amines by NaBH
4
catalyzed by NHG carbocatalysta.
Entry
1
Reactant
Product
T (min)
Yield (%)b
99.1
15
2
3
4
5
6
20
7
98.9
99.0
99.1
95.0
93.6
5
5
7
a
2 4
Standard reaction conditions: 0.06 mmol nitrobenzene or substituted nitrobenzenes, 2.7 mg NHG, 3 mL H O/ethanol = 1/9 (v/v), and 6 mmol (100 equiv.) NaBH at room
temperature.
b
HPLC content.
positive charge and thus enhance the H ions transfer during the
hydrogenation reaction of organic dyes [25]. We also examined
the cleave reaction process of unsaturated chromophore group
by calculating the electrostatic potential distributions of these 5
organic dyes. Fig. 6B–F show that the AN@NA or AC@NA bonds
areas (orange and red areas) have lower charge density than other
area in the molecules. As a result, these N-containing unsaturated
chromophore groups should be low electrostatic potential sites,
and thus can be considered as the potential reactive site during
the hydrogenation process. These consequences clearly indicate
that the partially negative charged N atoms in NHG are act as the
adsorb sites to interact with the low electrostatic potential sites
in organic dyes and result in corresponding hydrogenated prod-
ucts. Based on the experimental results and the theoretical calcula-
groups (low electrostatic potential sites), (2) dye molecules capture
the reductive H species as soon as they contacted with the active
sites, leading to the hydrogenation of N-containing unsaturated
chromophore groups, and (3) the hydrogenated products desorbed
from the active sites.
In addition, the frontier orbitals along with UV–vis spectra of
the hydrogenated products are also calculated in Fig. S7. As shown
in Fig. S7A–6E, the modulated UV–vis spectra match well with the
experimental results, which provide extra evidence for the correct-
ness of the structure of target products. The corresponding inset
HOMO and LUMO orbitals of five products indicates that charge
transfer can be ascribed to the inter-molecular charge transfer pro-
cess. This process may further result in the higher HOMO/LUMO
energy gap, and thus lead to the blue shift of maximum absorption
wavelength (<400 nm). Furthermore, the 1 s orbitals of N atoms in
NHG owns more partial density of states (PDOS) than that in NG
(Fig. S7F), which is consistent with the calculated total DOS in
Fig. 6A. The above calculation results demonstrate that graphene
with holey structure and nitrogen atoms doping can act as the effi-
cient material for the catalytic hydrogenation of organic dyes.
tions,
a reasonable explanation for the metal-free catalytic
mechanism can be proposed. The catalytic behavior may occur in
ꢀ
three steps: (1) in aqueous solution, BH
4
ions interact with the
activated C atoms in NHG and dye molecules diffused/adsorbed
onto NHG via electrostatic interaction between the partially nega-
tive charged N sites and N-containing unsaturated chromophore

Z. He et al. / Journal of Catalysis 377 (2019) 199–208
207
Fig. 6. (A) Total DOS (state per eV) of NG and NHG, inset is electron density mapping diagram of NHG; Electrostatic potential distribution of (B) 4-NP, (C) MO, (D) CR, (E) RhB
and (F) MB.
4
. Conclusion
Author contributions
In summary, we report a facile one-pot, and effective synthesis
J.B. Xi conceived the idea. Z.L. He, J. Liu, M. Zhao, Z.P. Wen, C.Y.
Xie and J. Chen carried out the materials synthesis and the chem-
ical catalysis. Q.J. Wang and C.Y. Tang performed the DFT calcula-
tion. J.B. Xi and Z.L. He performed the materials
characterizations. The manuscript was written through contribu-
tions of all authors. All authors have given approval to the final ver-
approach to synthesize NHG through hydrothermal treatment of
GO in the presence of ammonium hydroxide and hydrogen perox-
ide. The graphene oxide sheets can simultaneously be reduced,
self-assembled, chemically etched and nitrogen doped into mono-
lithic NHG hydrogel and then finally converted into NHG suspen-
sion. The metal-free carbocatalyst prepared with this method had
a 3D macroporous framework, hydrophilic property, large SSA
§
sion of the manuscript. Z.L. He, J. Liu, Q.J. Wang and M. Zhao
contributed equally.
2
ꢀ1
(
322.1 m g ) and hierarchical mesoporous structure, as well as
high content of doped N element (9.77 at. %). Owing to the excel-
lent properties, NHG exhibited a high activity toward the catalytic
hydrogenation of five different organic dyes in aqueous solution.
Furthermore, the nitrobenzenes reduction reactions were imple-
mented with very high efficiency catalyzed by NHG. These hydro-
genation reactions were found to proceed at room temperature
and afforded the desired product in excellent yields. Typically,
the TOF for 4-NP reduction (3.32 ꢁ10 min ) was higher than
that of commercial Pd/C (5.0 wt%) and those gained by carbocat-
alysts which have been reported in literatures. Molecular struc-
ture analysis combined with theoretical calculations elucidated
the metal-free catalytic mechanism. Based on the obtained
results, it is envisioned that the NHG carbocatalyst has great
potential applications in organic catalysis and wastewater
treatment.
Declaration of Competing Interest
The authors declare no competing financial interest.
Acknowledgements
ꢀ2
ꢀ1
This research was financially supported by the Natural Science
Foundation of Hubei Province (Nos. 2016CFB263, 2016CFB168 and
2018CFB159) and the National Natural Science Foundation of
China (Nos. 21702155, 51772110 and 51708427). The authors
would like to acknowledge the Analytical and Testing Center of
Huazhong University of Science and Technology and the Wuhan
National Laboratory for Optoelectronics for SEM, TEM, XPS and
Raman measurements. The authors also would like to show great

2
08
Z. He et al. / Journal of Catalysis 377 (2019) 199–208
gratitude to Multi-function Computer Center of Guangxi University
for the computational resources.
[26] J.B. Xi, Q.J. Wang, J. Liu, L. Huan, Z.L. He, Y. Qiu, J. Zhang, C.Y. Tang, J. Xiao, S.
Wang, J. Catal. 359 (2018) 233–241.
[
[
[
27] Y.Y. Shao, S. Zhang, M.H. Engelhard, G.S. Li, G. Shao, Y. Wang, J. Liu, I.A. Aksay,
Y.H. Lin, J. Mater. Chem. 20 (2010) 7491–7496.
28] B.M. Wei, Z.W. Zhai, H.B. Wang, J.T. Zhang, C.Z. Xu, Y.L. Xu, L. He, D. Xie, J. Agric.
Food Chem. 66 (2018) 9080–9086.
Appendix A. Supplementary material
29] Jr W.S. Hummers, R.E. Offeman, J. Am. Chem. Soc. 8 (1958) 1339–1339.
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.jcat.2019.07.017.
[30] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
H. Nakatsuji, et al., Gaussian 09, Revision A.02, Gaussian, Inc, Wallingford, CT,
2009.
[
31] F. Weigenda, R. Ahlrichs, Phys. Chem. Chem. Phys. 7 (2005) 3297–3305.
References
[32] S.A. Carabineiro, R.M. Bellabarba, P.T. Gomes, S.I. Pascu, L.F. Veiros, C. Freire, L.
C.J. Pereira, R.T. Henriques, M.C. Oliveira, J.E. Warren, Inorg. Chem. 47 (2008)
8896–8911.
[33] J. Autschbach, T. Ziegler, S.J.A. van Gisbergen, E.J. Baerends, J. Chem. Phys. 116
(2002) 6930–6940.
[34] X.P. Wang, L.X. Lv, Z.H. Cheng, J. Gao, L.Y. Dong, C.G. Hu, L.T. Qu, Adv. Energy
Mater. 6 (2016) 150210.
[35] J.W. Jang, C.E. Lee, S.C. Lyu, T.J. Lee, C.J. Lee, Appl. Phys. Lett. 84 (2004) 2877–
2879.
[
1] A. Bafana, S. Devi, T. Chakrabarti, Environ. Rev. 19 (2011) 350–371.
[
2] Z.P. Wen, Y.L. Zhang, G. Cheng, Y.R. Wang, R. Chen, Chemosphere 218 (2019)
1
002–1013.
3] H. Wang, L.F. Liang, X.J. Cheng, Y.M. Luo, S. Sun, Photochem. Photobiol. 94
2018) 17–26.
4] Z. Ulma, E. Rahayuningsih, T.D. Wahyuningsih, IOP conference series: materials
science and engineering, IOP Publishing 299 (2018) 012075.
[
[
(
[5] P. Singh, S. Roy, A. Jaiswal, J. Phys. Chem. C 121 (2017) 22914–22925.
[6] Y.M. Slokar, A. Majcen Le Marechal, Dyes Pigm. 37 (1998) 335–356.
[7] J.B. Xi, H.Y. Sun, D. Wang, Z.Y. Zhang, X.M. Duan, J.W. Xiao, F. Xiao, L.M. Liu, S.
Wang, Appl. Catal. B- Environ. 225 (2018) 291–297.
[36] X.L. Li, H.L. Wang, J.T. Robinson, H. Sanchez, G. Diankov, H.J. Dai, J. Am. Chem.
Soc. 131 (2009) 15939–15944.
[37] J.B. Xi, Y. Zhang, Q.J. Wang, J. Xiao, K. Chi, X.M. Duan, J. Chen, C.Y. Tang, Y.M.
Sun, F. Xiao, S. Wang, Sensor Actuat. B-Chem. 273 (2018) 108–117.
[38] J.T. Xu, Y. Lin, J.W. Connell, L.M. Dai, Small 11 (2015) 6179–6185.
[39] Q.J. Wang, S. Zhou, S. Xiao, F.F. Wei, X.Z. Zhao, J.E. Qu, H.R. Wang, RSC Adv. 8
(2018) 14775–14786.
[40] S.L. Yang, C.Y. Cao, Y.B. Sun, P.P. Huang, F.F. Wei, W.G. Song, Angew. Chem. 127
(2015) 2699–2702.
[41] B. Naik, S. Hazra, V.S. Prasad, N.N. Ghosh, Catal. Commun. 12 (2011) 1104–
1108.
[42] A. Hernández-Gordillo, A.G. Romero, F. Tzompantzi, S. Oros-Ruiz, R.J. Gómez,
Photochem. Photobio. A: Chem. 257 (2013) 44–49.
[43] F. Wang, S.Y. Song, K. Li, J.Q. Li, J. Pan, S. Yao, X. Ge, J. Feng, X. Wang, H.J. Zhang,
Adv. Mater. 28 (2016) 10679–10683.
[
8] R.F. Nie, M. Miao, W.C. Du, J.J. Shi, Y.C. Liu, Z.Y. Hou, Appl. Catal. B- Environ. 180
(
2016) 607–613.
[
9] Y.J. Sun, J.Z. Jiang, Y. Liu, S.L. Wu, J. Zou, Appl. Surf. Sci. 430 (2018) 362–370.
[
[
10] F.X. Chen, S.L. Xie, X.L. Huang, X.H. Qiu, J. Hazard. Mater. 322 (2017) 152–162.
11] J. Ke, J. Liu, H.Q. Sun, H.Y. Zhang, X.G. Duan, P. Liang, X.Y. Li, M.O. Tadé, S.M. Liu,
S.B. Wang, Appl. Catal. B- Environ. 200 (2017) 47–55.
[
12] V.C. Bhethanabotla, D.R. Russell, J.N. Kuhn, Appl. Catal. B- Environ. 202 (2017)
1
56–164.
[
13] F. He, J.Q. Luo, S.T. Liu, Chem. Eng. J. 294 (2016) 362–370.
[
14] L.M. Dai, Y.H. Xue, L.T. Qu, H.J. Choi, J.B. Baek, Chem. Rev. 115 (2015) 4823–
4
892.
[
15] L. Miao, D.Z. Zhu, M.X. Liu, H. Duan, Z.W. Wang, Y.K. Lv, W. Xiong, Q.J. Zhu, L.C.
Li, X.L. Chai, L.H. Gan, Chem. Eng. J. 347 (2018) 233–242.
16] J. Yu, Y.Y. Zhang, H. Li, Q.J. Wan, Y.W. Li, N.J. Yang, Carbon 129 (2018) 301–339.
17] M. Liu, Q. Guan, S.T. Liu, Ionics 24 (2018) 2783–2793.
18] X.G. Duan, H.Q. Sun, Y.X. Wang, J. Kang, S.B. Wang, ACS Catal. 5 (2015) 553–
[44] F. Yang, C. Chi, C.X. Wang, Y. Wang, Y.F. Li, Green Chem. 18 (2016) 4254–4262.
[45] H.W. Hu, J.H. Xin, H. Hu, X.W. Wang, Nano Res. 8 (2015) 3992–4006.
[46] H.W. Hu, X.W. Wang, D.G. Miao, Y.F. Wang, C.L. Lai, Y.J. Guo, W.Y. Wang, J.H.
Xin, H. Hu, Chem. Commun. 51 (2015) 16699–16702.
[47] W. Zhou, Y. Zhou, Y. Liang, X.H. Feng, H. Zhou, RSC Adv. 5 (2015) 50505–50511.
[48] J.B. Xi, J.W. Xiao, F. Xiao, Y.X. Jin, Y. Dong, J. Feng, S. Wang, Sci. Rep. 6 (2016)
21904.
[49] X.M. Duan, M.C. Xiao, S. Liang, Z.Y. Zhang, Y. Zeng, J.B. Xi, S. Wang, Carbon 30
(2018) 806–813.
[50] X.M. Duan, J. Liu, J.F. Hao, L.M. Wu, B.J. He, Y. Qiu, J. Zhang, Z.L. He, J.B. Xi, S.
Wang, Carbon 130 (2018) 806–813.
[51] J. Liu, J.F. Hao, C.C. Hu, B.J. He, J.B. Xi, J.W. Xiao, S. Wang, Z.W. Bai, J. Phsy. Chem.
C 122 (2018) 2696–2703.
[52] T. Sun, Z.Y. Zhang, J.W. Xiao, C. Chen, F. Xiao, S. Wang, Y.Q. Liu, Sci. Rep. 3
(2013) 2527.
[53] F. Hu, M. Patel, F.X. Luo, C. Flach, R. Mendelsohn, E. Garfunkel, H.X. He, M.
Szostak, J. Am. Chem. Soc. 137 (2015) 14473–14480.
[54] S.K. Lee, A. Mills, Chem. Commun. 18 (2003) 2366–2367.
[
[
[
5
59.
19] D.R. Dreyer, H.P. Jia, C.W. Bielawski, Angew. Chem. Int. Ed. 49 (2010) 6813–
816.
[
[
[
[
[
[
[
6
20] C.L. Su, M. Acik, K. Takai, J. Lu, S.J. Hao, Y. Zheng, P.P. Wu, Q.L. Bao, T. Enoki, Y.J.
Chabal, K.P. Loh, Nat. Commun. 3 (2012) 1298.
21] H.Q. Sun, S.Z. Liu, G.L. Zhou, H.M. Ang, M.O. Tadé, S.B. Wang, A.C.S. Appl, Mater.
Interfaces 4 (2012) 5466–5471.
22] X.G. Duan, H.Q. Sun, S.B. Wang, Metal-Free carbocatalysis in advanced
oxidation reactions, Accounts Chem. Res. 51 (2018) 678–687.
23] B. Frank, J. Zhang, R. Blume, R. Schlogl, D.S. Su, Angew. Chem. Int. Ed. 48 (2009)
6
913–6917.
24] P.W. Gong, J.Y. Du, D.D. Wang, B.B. Cao, M. Tian, Y.H. Wang, L. Sun, S.J. Ji, Z. Liu,
J. Mater. Chem. B 6 (2018) 2769–2777.
25] X.K. Kong, Z.Y. Sun, M. Chen, C.L. Chen, Q.W. Chen, Energy Environ. Sci. 6
(
2013) 3260–3266.