One-pot reductive amination of carbonyl compounds with nitro compounds with CO/H2O as the hydrogen donor over non-noble cobalt catalyst

Article abstract of DOI:10.1016/j.jcat.2017.05.026

The one-pot reductive amination of carbonyl compounds with nitro compounds over heterogeneous non-noble metal catalysts was developed for the first time by transfer hydrogenation with CO/H₂O as the hydrogen donor. Nitrogen-doped carbon supported cobalt nanoparticles were observed to be active toward this reaction, affording structurally-diverse secondary amines with high yields. Kinetic studies revealed that the transfer hydrogenation of imines (C[dbnd]N bonds) was the rate-determining step. Reaction mechanism studies indicated that both nitrogen and cobalt nanoparticles were important for the transfer hydrogenation with CO/H₂O to generate the proton (N[sbnd]H⁺) and hydride (Co[sbnd]H^?) as the active species. Furthermore, the heterogeneous cobalt catalyst was highly stable without the loss of its catalytic activity during the recycling experiments.

Full text of DOI:10.1016/j.jcat.2017.05.026

Journal of Catalysis 352 (2017) 264–273
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
One-pot reductive amination of carbonyl compounds with nitro
compounds with CO/H O as the hydrogen donor over non-noble cobalt
2
catalyst
a
b
a
a
a,
⇑
Peng Zhou , Changlin Yu , Liang Jiang , Kangle Lv , Zehui Zhang
a
Key Laboratory of Catalysis and Materials Sciences of the Ministry of Education, South-Central University for Nationalities, Wuhan 430074, PR China
School of Metallurgy and Chemical Engineering, Jiangxi University of Science and Technology, 86 Hongqi Road, Ganzhou 341000, PR China
b
a r t i c l e i n f o
a b s t r a c t
Article history:
The one-pot reductive amination of carbonyl compounds with nitro compounds over heterogeneous non-
Received 28 March 2017
Revised 29 May 2017
Accepted 31 May 2017
2
noble metal catalysts was developed for the first time by transfer hydrogenation with CO/H O as the
hydrogen donor. Nitrogen-doped carbon supported cobalt nanoparticles were observed to be active
toward this reaction, affording structurally-diverse secondary amines with high yields. Kinetic studies
revealed that the transfer hydrogenation of imines (C@N bonds) was the rate-determining step.
Reaction mechanism studies indicated that both nitrogen and cobalt nanoparticles were important for
Keywords:
Nitrogen-doped carbon
Cobalt catalysts
Transfer hydrogenation
Reductive amination
Secondary amines
+
ꢀ
2
the transfer hydrogenation with CO/H O to generate the proton (NAH ) and hydride (CoAH ) as the
active species. Furthermore, the heterogeneous cobalt catalyst was highly stable without the loss of its
catalytic activity during the recycling experiments.
Ó 2017 Elsevier Inc. All rights reserved.
1
. Introduction
Synthesis of amines and its derivates represents one of the most
pounds has been less developed [9–15]. Methods for the one-pot
reductive amination are mainly performed over noble metal
catalysts with H
[9–15]. The use of H
2
as the hydrogen source (Scheme 1, method A)
as the reducing reagent is environmentally
active fields in synthesis chemistry. Nitro compounds as a type of
readily available starting materials have been extensively studied
for the synthesis of primary amines with fruitful results [1]. The
as-formed primary amines can be further upgraded into many
kinds of highly valuable chemicals such as secondary or tertiary
amines, imines, azo compounds and other nitrogen-containing
compounds, which are of great importance in the synthesis of
pharmaceuticals, dyes, pigments, agrochemicals, and materials
2
friendly, as water was the oxidation product. Some noble metal
catalysts were even reported to be active for the one-pot reductive
amination under mild conditions. For example, the PdAg bimetallic
nanocatalyst can promote the one-pot reductive amination at
room temperature and atmospheric hydrogen pressure with high
yields [13]. Nowadays, great effort has been developed to the
search of non-noble metal catalysts to replace noble metal cata-
lysts for the chemical transformations. To this context, heteroge-
neous carbon-supported iron oxide and cobalt oxide catalysts
were studied for this type reaction [14,15]. For example, the one-
pot reductive amination over the carbon-supported iron oxide cat-
alyst produced 45ꢁ91% yields of secondary amines within 30 h at
[
2–6]. From the viewpoints of green and sustainable chemistry, it
is highly attractive to perform the synthesis of their derivates
directly from nitro compounds by one-pot reaction strategies [7].
Compared with the methods by multiple steps, the one-pot
method demonstrates higher atom efficiency by avoiding tedious
purification steps and releasing fewer wastes.
2
170 °C and 70 bar H [15]. Obviously, these processes have poten-
Reductive amination, the coupling of carbonyl compounds with
primary amines in the presence of a reducing reagent, has been
considered to be one of the most efficient methods for the synthe-
sis of secondary amines [8]. Although the reductive amination of
carbonyl compounds with primary amines has enjoyed a huge suc-
cess, the one-pot reductive amination directly from nitro com-
tial safety issues regarding the handling of explosive H
conditions.
2
at harsh
So far, organic synthesis by transfer hydrogenation has
attracted great interest with CO/H O as the hydrogen donor via
2
water–gas shift reaction [16]. Apart from being well known as a
very useful C1 building block in the chemical industry, CO has also
been considered to be a useful reductant in organic synthesis
[
17,18]. Compared with H
2
and other hydrogen donors, CO is
⇑
Corresponding author.
E-mail address: zhangzehui@mail.scuec.edu.cn (Z. Zhang).
produced in multiton quantities as a by-product of the steel
http://dx.doi.org/10.1016/j.jcat.2017.05.026
021-9517/Ó 2017 Elsevier Inc. All rights reserved.
0

P. Zhou et al. / Journal of Catalysis 352 (2017) 264–273
265
heating rate of 3 °C/min from room temperature. The as-prepared
materials were denoted as the Co/N-C-T catalysts, where T indi-
cates the pyrolysis temperature. On the meanwhile, the Co/N-C-
6
00 catalyst was heated in the air at 250 °C for 4 h to get the
Co/N-C-600-Air catalyst, in which metallic Co nanoparticles were
oxidized to its oxidation state.
2.3. Catalyst characterization
Transmission electron microscope (TEM) images of the samples
2
were obtained on an FEI Tecnai G -20 instrument. The samples
were firstly dispersed in ethanol with an assist of ultrasonication
and dropped onto copper grids for observation. Micropore and
mesopore surface area and pore size measurements were
Scheme 1. Methods for the synthesis of secondary amines via reductive amination.
2
performed with N adsorption/desorption isotherms at 77 K on a
V-Sorb 2800P instrument. Before measurements, the samples were
degassed at 100 °C for 12 h. X-ray powder diffraction (XRD) pat-
terns of samples were determined with a Bruker advanced D8
industry [19]. H
2
is a potential clear energy, and thus the use of CO
instead of H will also provide much more clean energy to our soci-
2
powder diffractometer (Cu Ka). All XRD patterns were collected
ety. Therefore, the use of CO not only resolves the concerns of envi-
ronment, but also makes the chemical process more economic and
more sustainable [20]. In addition, there is no explosive risk on the
use of CO for the chemical reactions. To the best of our knowledge,
there have been only two examples on the reductive amination
over either homogeneous or heterogeneous noble rhodium cata-
lysts (Scheme 1, Method B) [21,22]. For example, Chusov and List
reported the reductive amination of carbonyl compounds with
amines to generate the corresponding secondary amines with
o
with a scanning rate of 0.016 /s. X-ray photoelectron spectroscopy
XPS) was conducted on a Thermo VG scientific ESCA MultiLab-
000 spectrometer with monochromatized Al source
1486.6 eV) at constant analyzer pass energy of 25 eV. The binding
(
2
a
Ka
(
energy was estimated to be accurate within 0.2 eV. All binding
energies were corrected referencing to the C1s (284.6 eV) peak of
the contamination carbon as an internal standard. The cobalt con-
tent in the catalysts was quantitatively determined by inductively
coupled atomic emission spectrometer (ICP-AES) on an IRIS Intre-
pid II XSP instrument (Thermo Electron Corporation). Raman spec-
tra were measured on a confocal laser micro-Raman spectrometer
yields of 70 ꢁ 98% over homogeneous Rh
20 ꢁ 140 °C and 20 ꢁ 100 bar CO pressure in terahydrofuran
21]. In this work, CO was the real reducing reagent, and this cat-
2 4
(OAc) catalyst at
1
[
(
5
Thermo Fisher DXR) equipped with a diode laser of excitation of
32 nm (laser serial number: AJC1200566). Spectra were obtained
alytic system was obviously not suitable for the reductive amina-
tion of carbonyl compounds with nitro compounds. Thus, it is
highly desirable to perform the one-pot reductive amination over
heterogeneous non-noble metal catalysts with a high activity and
a high selectivity.
at a laser output power of 1 mW (532 nm), and a 0.2 s acquisition
time with 900 lines/mm grating (Grating serial number:
ꢀ
1
AJG1200531) in the wavenumber range of 50–3500 cm . Electron
paramagnetic resonance (EPR) spectra were recorded on a Bruker
EMX spectrometer. The EPR experiments were conducted with a
center field of 3507.815 G and a frequency of 9.83 GHz using an
Elexsys probe head with 15 mg of sample placed in a 4 mm tube.
Currently, nitrogen-doped carbon-supported metal catalysts
have gained considerable attention for advanced catalysis
[
23,24]. The incorporation of nitrogen in the carbon architecture
played multiple roles in the improvement of the catalytic activity
such as the stabilization of metal nanoparticles and the enhance-
ment of the electron transfer ability [25]. ZIF-67 (ZIF: zeolitic imi-
dazolate framework), which is composed of cobalt cation and 2-
methylimidazole ligand, is a promising precursor for the prepara-
tion of nitrogen-doped carbon supported cobalt catalysts [26].
Herein, nitrogen-doped carbon-supported cobalt catalysts by the
pyrolysis of ZIF-67 were prepared and studied for the one-pot
reductive amination of carbonyl compounds with nitro compounds
2.4. Catalytic transfer hydrogenation of nitrobenzene
In a typical run, the Co/N-C-600 catalyst (20 mg), nitrobenzene
(
1 mmol) and water (10 mL) were charged in a 50 mL autoclave,
which was equipped with a magnetic stirring bar and a tempera-
ture controller. After the removal of air with CO, the autoclave
was charged with 20 bar CO at room temperature. Then the auto-
clave was heated at 110 °C for 10 h with a magnetic stirring at
(Scheme 1). To the best of our knowledge, this process is
1
000 rpm. After cooling to room temperature, the organic products
unprecedented.
in the reaction mixture were extracted by ethyl acetate, and the
chemicals in the reaction mixture were quantified by gas chro-
matography (GC) with toluene as the internal standard.
2
. Experimental section
.1. Synthesis of ZIF-67
Co(NO O (99%, 0.45 g) in 3 mL of water was added to a
ꢂ6H
2
2.5. General procedure for the one-pot reductive amination
3
)
2
2
The reaction procedure for the one-pot reductive amination was
solution of 2-methylimidazole (99%, 5.5 g) in 20 mL of water. Then
the mixture was stirred at 25 °C for 6 h. The resulting purple pre-
cipitates were obtained via centrifugation and washed with water
and methanol for twice, respectively, and finally dried in a vacuum
oven.
almost the same as described above with some modifications.
Firstly, apart from the addition of nitrobenzene (1 mmol), the
Co/N-C-600 catalyst (20 mg) and water (10 mL) into the autoclave,
benzaldehyde (2 mmol) was also added. In addition, reactions
were performed at 170 °C and 40 bar CO. Toluene and phenyl ether
were used as double internal standards, in which phenyl ether was
used to quantify the imines and secondary amines. The products in
the reaction mixture were identified by comparison of their retain
times with the authentic chemicals, and further confirmed by
GC–MS (Agilent 7890A GC/5973 MS, HP-5 column).
2.2. Catalyst preparation
ZIF-67 (2.0 g) was pyrolyzed at three representative tempera-
tures (600, 800 and 900 °C) for 8 h in a nitrogen atmosphere at a

2
66
P. Zhou et al. / Journal of Catalysis 352 (2017) 264–273
2.6. Analytic methods
mesopores (Fig. S3). The textural parameters of the Co/N-C-T cata-
lysts are summarized in Table S2. The Co/N-C-600 and Co/N-C-800
catalysts have similar micropore and mesopore surface areas,
which are both larger than those of the Co/N-C-900 catalyst. In
addition, the Co/N-C-T catalysts have similar mesopore volume,
but the Co/N-C-800 catalyst has a little larger micropore volume
than the Co/N-C-600 and Co/N-C-900 catalysts.
Products analysis was performed on Agilent 7890A GC with
autosampler and a flame ionization detector. The products were
separated by a HP-5 capillary column (30 m ꢃ 530 m ꢃ 1.5 m).
The temperature of the column was initially kept at 80 °C for
l
l
ꢀ1
3
min, and then increased at a rate of 20 °C min to 220 °C. Prod-
ucts were identified by the comparison of the retention time with
the authentic chemicals, and further confirmed by GC–MS (Agilent
Three peaks at 2h = 44.2, 51.5 and 76.0° were observed in the
XRD patterns of the Co/N-C-T catalyst (Fig. 4), which can be
indexed to the (111), (200) and (220) plans of metallic Co
nanoparticles (JCPDS No. 15–0806) [23]. The particle sizes of cobalt
nanoparticles were calculated to be 13, 26 and 38 nm for the
Co/N-C-600, Co/N-C-800 and Co/N-C-900 catalysts by the Scherrer
equation, respectively, which were in agreement with TEM results.
7
890A GC/5973 MS, HP-5 column). The amounts of products were
determined based on GC data using the internal standard method.
3
. Results and discussion
Characteristic peaks of Co
5-3103) at 2h = 31.4, 36.8, 44.7, 55.6, 59.2 and 65.2° were
observed in the XRD pattern of Co/N-C-600-Air (Fig. S4), suggesting
that metallic Co in Co/N-C-600 was oxidized to Co after the
3 4
O crystalline structure (JCPDS) No.
3
.1. Catalyst preparation and characterization
6
Fig. 1 illustrates the process of the fabrication of nitrogen-doped
3 4
O
carbon-supported cobalt catalysts. ZIF-67 was used as the precur-
sor for the synthesis of the Co/N-CꢀT catalysts, which was formed
by the coordination of Co2 with 2-methylimidazole ligand. The
XRD pattern of the as-made ZIF-67 (Fig. S1) was consistent with
its standard XRD pattern [27]. ZIF-67 was reported to start decom-
position at 500 °C in the inert atmosphere [27], and thus the pyrol-
ysis of ZIF-67 was performed at three representative temperatures
of 600, 800 and 900 °C to generate three kinds of Co/N-CꢀT
catalysts.
Cobalt nanoparticles were clearly observed in the TEM image of
the Co/N-C-600 catalyst with an average size of 14 nm (Fig. 2). It
was observed that cobalt nanoparticles tended to grow with the
increase in the pyrolysis temperature, and their average sizes were
calculated to be 14, 22 and 42 nm for the Co/N-C-600, Co/N-C-800
and Co/N-C-900 catalysts, respectively (Fig. S2). The weight per-
centage of cobalt in the Co/N-C-T catalysts increased with the
pyrolysis temperature from 34.8% at 600 °C to 40.7% at 900 °C
treatment in the air at 250 °C. The peak with 2h around 25.8°
was generally observed for the nitrogen-doped carbon materials,
which was assigned to a turbostratic ordering of the carbon and
nitrogen atoms in the graphite layers (JCPDS No.01-0646) [23].
However, this peak was not present in the XRD patterns of the
Co/N-C-T catalysts, because of the high Co loading [30]. Raman
spectra confirmed the structure of graphene of the Co/N-C-T cata-
lysts. The spectra of the Co/N-C-T catalysts show two characteristic
+
ꢀ1
D and G bands at 1352 and 1580 cm , corresponding to the disor-
2
dered carbon atoms and the sp hybridized graphitic carbon atoms,
respectively (Fig. S5) [31].
Fig. 5 shows the XPS spectra of Co 2p3/2 and N 1s of the Co/N-C-
6
00 catalyst. The Co 2p3/2 peak can be deconvoluted into two peaks
with the binding energies (BEs) at 780.3 and 778.3 eV, correspond-
ing to the oxidation state cobalt and metallic cobalt, respectively.
The presence of the oxidation state cobalt was due to the surface
oxidation of metallic cobalt nanoparticles during the storage in
the air [31]. The relative peak area of metallic cobalt for the
Co/N-C-600 catalyst was much weaker than the Co/N-C-800 and
Co/N-C-900 catalysts (Fig. S6), which may be due to the fact that
the smaller size of Co nanoparticles was easier to be oxidized
(Table S1). The weight percentage of nitrogen in the Co/N-C-T cat-
alysts decreased from 10.4 wt.% for the Co/N-C-600 catalyst to 0.8
wt.% for the Co/N-C-900 catalyst (Table S1). On the meanwhile, the
atom percentage of nitrogen decreased with the pyrolysis temper-
ature from 8.2% for the Co/N-C-600 catalyst to 1.8% for the Co/N-C-
[
32]. Typically, the N 1 s XPS spectrum of the Co/N-C-600 catalyst
can be fitted into three peaks at BEs about 398.5, 399.5 and
00.8 eV, corresponding to pyridinic N (N-1), pyrrolic N (N-2)
and graphitic N (N-3), respectively (Fig. 5) [31]. Similarly, the N
s of the Co/N-C-600-Air catalyst can also be fitted into the
9
00 catalyst (Table S1). Compared with Co/N-C-600, the cobalt
weight percentage decreased from 34.8% to 30.9% in the Co/N-C-
00-Air catalyst, which was due to the oxidation of metallic cobalt
to Co (Table S1). Similar decrease in nitrogen content in the
4
6
3 4
O
1
Co/N-C-600-Air catalyst was also observed (Table S1). The decrease
in nitrogen content should be one of the main reasons for the
growth of Co nanoparticles at higher pyrolysis temperature, as
the nitrogen atoms have the ability to stabilize metal nanoparticles
same three peaks, and the intensity was the same as the N 1 s
for the Co/N-C-600 catalyst, suggesting that the nitrogen structure
in the support was maintained during the oxidation process under
our conditions (Fig. S7). As far as the C 1 s XPS spectrum of the
Co/N-C-600 catalyst, four peaks including C@C (284.6 eV), CAN
[
28,29]. Nitrogen adsorption–desorption isotherm curves of the
Co/N-C-T catalysts are similar to the IV-type isotherm with a H4
hysteresis loop (Fig. 3), suggesting its dominating mesoporous
structure. However, they also exhibit a typical I-type isotherm with
equilibrium in the P/P range of 0–0.1, indicating that the Co/N-C-T
o
catalysts also have microporous structure. Pore size distribution
shows that all the Co/N-C-T catalysts have both micropores and
(
285.0 eV), C .•. .O (286.2 eV), and OAC@O (289.2 eV) were also fit-
ted (Fig. S8).
To better account for the interaction between the nitrogen-
doped supports, EPR was conducted for diagnosing the spin state
of unpaired electrons. Initially, we tried to get the nitrogen-
Fig. 1. Schematic illustration of the fabrication of the Co/N-C-600 catalyst.

P. Zhou et al. / Journal of Catalysis 352 (2017) 264–273
267
40
35
30
25
20
15
10
Average diameter = 14 nm
5
0
3
6
9
12
15
18
21
24
Diameter (nm)
Fig. 2. TEM image and the particle size distribution of the Co/N-C-600 catalyst.
1
1
1
1
60
40
20
00
carbonization [33]. As shown in Fig. 6, the N-C-900 sample demon-
strates the typical spin characteristics of unpaired electrons with a
g factor of 2.06. It is reported that the n-type dopant of carbon
materials by group V impurities proceeds via a substitutional
mechanism, which facilitates the formation of deep defect levels
Co/N-C-600
Co/N-C-800
Co/N-C-900
and could capture considerable electrons [34]. At the same time,
2
the incomplete localization of
p
bonds from the sp -type sites also
8
6
4
2
0
0
0
0
0
significantly affects the electronic properties of N-C-900. All these
features of N-C-900 are responsible for the presence of unpaired
electrons. The incorporation of cobalt species with the nitrogen-
doped carbon material for the Co/N-C-900 catalyst results in the
entire disappearance of the spin resonance peaks, which could be
ascribed to the electronic transfer inducing the reconjugation of
single electrons or the reaction between supports with precursor
solution during the deposition of cobalt particles [35]. Therefore,
EPR results confirmed that there was a strong interaction between
the cobalt nanoparticles and nitrogen atoms.
0
.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Relative pressure (P/Po
)
Fig. 3. Nitrogen adsorption–desorption isotherms of the catalysts at 77 K.
3.2. Reductive amination of benzaldehyde with nitrobenzene
The one-pot reductive amination of nitro compounds with car-
bonyl compounds is composed of three consecutive steps of the
transfer hydrogenation of nitro compounds, the second condensa-
tion step to generate imines and the final of transfer hydrogenation
of imines into secondary amines. The first step of the transfer
hydrogenation of nitro compounds depended on the success of
the whole reaction process. Therefore, the catalytic activity of the
Co/N-C-T catalysts was initially evaluated by the transfer hydro-
2
genation of nitrobenzene in CO/H O system. No conversion was
observed with the parent precursor of ZIF-67 (Table 1, Entry 1).
To our delight, the Co/N-C-T catalysts were active for the transfer
hydrogenation of nitrobenzene (Table 1, Entries 2ꢁ4), but the cat-
alytic activity of the Co/N-C-T catalysts decreased in an order of Co/
N-C-600, Co/N-C-800 and Co/N-C-900 with the same loading of
cobalt. Quantitative aniline yield was achieved over the Co/N-C-
6
00 catalyst after 10 h at 110 °C and 20 bar CO (Table 1, Entry 2),
while those were 67% and 59% for the Co/N-C-800 and Co/N-C-
900 catalysts (Table 1, Entries 3 and 4), respectively. As we used
the same cobalt weight for the Co/N-C-T catalysts, the difference
in the catalyst activity of the Co/N-C-T catalysts was not caused
by the content of cobalt. XRD and TEM images revealed that the
size of metallic cobalt nanoparticles increased in an order of
Co/N-C-600, Co/N-C-800 and Co/N-C-900 catalysts, due to the
increase in pyrolysis temperature. Generally speaking, metal
nanoparticles with a smaller average size exhibit a higher catalytic
activity [36], as metal nanoparticles with a smaller size provide
much more active sites. Therefore, the catalytic activity decreased
in an order of Co/N-C-600, Co/N-C-800 and Co/N-C-900 with the
same cobalt loading. In addition, it was noted that the nitrogen
content significantly decreased in an order of Co/N-C-600,
Fig. 4. XRD spectra of the catalysts. (a) Co/N-C-600; (b) Co/N-C-800; (c) Co/N-C-
9
00.
doped carbon material without cobalt by washing off the cobalt
nanoparticles from the Co/N-C-600 catalyst with 6 M H SO at
0 °C, but we failed to get it. As shown in Fig. S9(a), some cobalt
2
4
9
nanoparticles were observed to be embedded in the support for
the Co/N-C-600-H+ with a much smaller size as compared with
the Co/N-C-600 catalyst (Fig. S9b). Hence, we prepared ZIF-8 in a
same way as the ZIF-67 by the use of Zn(NO
of Co(NO . The pyrolysis of ZIF-8 at 900 °C could produce the
nitrogen-doped carbon material without Zn (abbreviated as N-C-
00) as shown in Fig. S9(c), since Zn would be evaporated during
3 2
) in the substitute
3 2
)
9

2
68
P. Zhou et al. / Journal of Catalysis 352 (2017) 264–273
2
2
2
2
2
1
1
4000
3000
2000
1000
0000
9000
Co 2p 3/2
19000
N1s
1
1
1
1
1
1
1
8000
7000
6000
5000
4000
3000
2000
7
80.3
N-1
7
85.9
N-3
N-2
778.3
8000
7
90 788 786 784 782 780 778 776 774
406
404
402
400
398
396
394
Binding Energy (eV)
Binding Energy (eV)
Fig. 5. XPS spectra of Co 2p 3/2 and N 1s of the Co/N-C-600 catalyst.
lysts, a physical mixture of commercial metallic cobalt (30 nm)
and activated carbon produced a much lower conversion than
the Co/N-C-600 catalyst at 90 °C (Table 1, Entries 5 vs 6). All of
these experiments suggested both nitrogen and Co nanoparticles
were of synergic importance in the high activity of the Co/N-C-
5
5
5
6
.0x10
.0x10
.0x10
N-C-900
Co/N-C-900
g = 2.06
4
2
6
00 catalyst. Interestingly, the reaction was also performed by
the use of the Co/N-C-600-Air catalyst, in which the active phase
of Co (Fig. S4), but the conversion of nitrobenzene greatly
0
.0
3 4
O
5
-
-
-
2.0x10
4.0x10
6.0x10
decreased (Table 1, Entries 6 vs 7). These results indicated that
metallic cobalt nanoparticles were the active sites for this reaction.
The Co/N-C-600 catalyst was even highly active under very mild
conditions (5 bar and 90 °C) with quantitative aniline yield after
5
5
1
4 h (Table 1, Entry 8), which demonstrated comparable or higher
3
425 3450 3475 3500 3525 3550 3575
X (G)
activity than the reported noble metal catalysts [37,38]. For exam-
ple, the homogeneous ruthenium bis(arylimino)acenaphthene
(
Ru/Ar-BIAN) catalyst produced similar results, but harsh reaction
conditions (150 °C and 50 bar CO) and 2 equiv. of triethylamine
Et N) as the additive were required (Table 1, Entry 9).
With the best catalyst of Co/N-C-600 in hand, the reductive
Fig. 6. EPR spectra of the as-prepared samples.
(
3
Co/N-C-800 and Co/N-C-900 with the increase in the pyrolysis
temperature (Table S1). On the one hand, the nitrogen atoms with
lone electron pairs played key roles in the stabilization of the
metallic cobalt nanoparticles. On the other hand, EPR results
amination of benzaldehyde with nitrobenzene was studied as a
model reaction. Firstly, the reaction was performed in water at
9
0 °C and 5 bar CO without the formation of N-benzylaniline. Apart
from aniline, the major product was the intermediate of
N-benzylideneaniline, which was formed by the condensation of
aniline with benzylaldhyde (Table 2, Entry 1). Similar results were
also observed at 110 °C (Table 2, Entry 2). These results suggested
that the transfer hydrogenation of the imine (C@N bond) by
CO/H O was much more difficult than the nitro group. Therefore,
2
the reactions were further performed at elevated temperatures.
N-benzylaniline was produced with a very low yield of 2.3% after
(Fig. 6) indicate that there has been a strong interaction between
the support and the cobalt nanoparticles, which was also positive
for the improvement of the catalyst activity by enhancement of
the electron transfer between the metal active sites and the sup-
port [25]. Furthermore, the nitrogen atoms in the Co/N-C-T cata-
+
lysts would serve as base sites to capture the
activation of CO/H
H
during
2
O to generate the active species of a proton
and a hydride (discussed later). To further illustration the role of
the nitrogen atoms in the catalytic activity of the Co/N-C-T cata-
1
0 h at 130 °C (Table 2, Entry 3). The transfer hydrogenation of
N-benzylideneaniline was sensitive to the reaction temperature.
Table 1
Catalyst screening for the reductive amination of benzaldehyde.a
Entry
Catalyst
Temperature (°C)
CO (bar)
Time (h)
Con. (%)
Sel. A (%)
1
2
3
4
5
6
7
8
9
ZIF-67
110
110
110
110
90
90
90
90
150
20
20
20
20
5
5
5
5
50
10
10
10
10
2
2
2
16
3
0
100
67
59
6.1
36.7
1.9
100
Yield 91%
–
Co/N-C-600
Co/N-C-800
Co/N-C-900
Co and AC
Co/N-C-600
Co/N-C-600-Air
Co/N-C-600
Ru/Ar-BIAN
>99
>99
>99
>99
>99
>99
>99
a
2
Reaction conditions: Nitrobenzene (1 mmol), H O (10 ml), Co weight (7.68 mg); Ref. [38].

P. Zhou et al. / Journal of Catalysis 352 (2017) 264–273
269
Table 2
a
Optimization of the reaction conditions for the reductive amination.
H
N
NH2
NO2
N
CO/H O
+
O
2
+
+
1
2
3
Entry
Temperature (°C)
CO (bar)
Solvent
Con. (%)
Sel. 1 (%)
Sel. 2 (%)
Sel. 3 (%)
1
2
3
4
5
6
7
8
9
90
5
5
5
5
5
5
10
20
30
40
H
2
H
2
H
2
H
2
H
2
H
2
H
2
H
2
H
2
H
2
O
O
O
O
O
O
O
O
O
O
100
100
100
100
100
100
100
100
100
100
34.6
12.7
8.4
0
0
–
–
–
–
–
65.4
87.3
89.7
76.0
65.7
43.0
49.2
22.1
2.4
0
0
2.3
24.0
34.3
53.9
51.8
78.9
97.6
99.4
110
130
150
170
190
170
170
170
170
1
0
0.6
a
2
Reaction conditions: Nitrobenzene (1 mmol), benzaldehyde (2 mmol), H O (10 ml), Co/N-C-600 (20 mg), reaction time (10 h).
The selectivity of N-benzylaniline continuously increased with the
increase in the reaction temperature (Table 2, Entries 3–6). How-
ever, the transfer hydrogenation of benzaldehyde (C@O) also
occurred at 190 °C (Table 2, Entry 6). So the optimal reaction tem-
perature was 170 °C for the one-pot reductive amination. In order
to improve the yield of N-benzylaniline, the reactions were per-
formed at higher CO pressures at 170 °C. The transfer hydrogena-
tion of N-benzylideneaniline to N-benzylaniline greatly enhanced
with the increase in CO pressure (Table 2, Entries 5, 7–10). The
higher selectivity of N-benzylaniline at higher CO pressure should
be due to the higher concentration of CO in water, which acceler-
ated the transfer hydrogenation reaction rate. Nearly quantitative
N-benzylaniline yield was achieved after 10 h at 170 °C and
other kinds of nitro compounds such as aliphatic and cyclic ones
were also active for the reductive amination with benzaldehyde
(Table 3, Entries 26–27). Of particular note is that the developed
catalytic system was even effective with the substrates that were
both non-aromatic ones (Table 3, Entries 28–31). The steric hin-
drance was found to show a great effect on the activity of the sub-
strates. For example, the ortho-nitrotoluene was less active than
the para- or meta-isomers (Table 3, Entries 19 vs 17 and 18), which
required a longer reaction time to get high yield (Table 3, Entry 20).
This phenomenon was much more obvious for the reductive ami-
nation of nitrobenzene with ketones. For example, the reductive
amination of cyclohexanone with nitrobenzene only gave the cor-
responding secondary amine with a yield of 32% after 10 h, and
aniline was the other main product (Table 3, Entry 32). These
results indicated that the difficulty in the reductive amination of
cyclohexanone with nitrobenzene was the condensation step,
due to the steric hindrance. Prolonging the reaction time to 30 h,
the corresponding secondary amine yield was enhanced to 85%.
Of particular note is that this catalytic system was highly tolerant
to other functional groups such as halogen (Entries 6, 7, 21 and 22),
ester (Entry 9), ketone (Entry 25), nitrile (Entries 10 and 23) and
phenol (Entry 24).
4
0 bar CO (Table 2, Entry 10).
3.3. Substrate scope
The general scope of the developed method was investigated.
Firstly, nitrobenzene was reductively aminated with various alde-
hydes. Despite the electronic properties of the substituted groups,
the reductive amination of nitrobenzene with substituted aromatic
aldehydes except 4-bromobenzaldehyde underwent smoothly to
give the secondary amines with good to excellent yields at
1
70 °C and 40 bar CO pressure (Table 3, Entries 1–10). The reduc-
3.4. Kinetic and mechanism studies
tive amination of 4-bromo-benzaldehyde with nitrobenzene pro-
duced a lower yield of the corresponding secondary amine, due
to the debromination of 4-bromo-benzaldehyde (Table 3, Entry
Fig. 7 recorded the products distribution at different time points
during the reaction process of the reductive amination of benzalde-
hyde with nitrobenzene. It was noted that the transfer hydrogena-
tion of nitrobenzene was very fast. Nitrobenzene conversion
reached 80% after 1 h, and full conversion was noted after 4 h.
Then, the condensation of aniline with benzaldehyde generated
the intermediate of N-benzylideneaniline. The condensation step
could occur without catalysts, but it was not very fast, which
was thought to be greatly retarded in the presence of a large excess
of water [40]. As shown in Fig. 7, the transfer hydrogenation of the
intermediate of N-benzylideneaniline was much slower than the
transfer hydrogenation of nitrobenzene over the Co/N-C-600 cata-
lyst. The yield of N-benzylaniline gradually increased during the
reaction process, and it reached 97.6% after 10 h. The data revealed
that the transfer hydrogenation of the imines (C@N bonds) was the
rate-determining step for the one-pot reductive amination. Fur-
thermore, the kinetics of the transfer hydrogenation of nitroben-
zene and N-benzylideneaniline were studied separately. Fig. 8
8
). This phenomenon was also observed by other researchers
[
39]. For example, the hydrogenation of 4-bromobenzonitrile over
cobalt complex produced a lower amine yield of 30%, while the
fluoro- and chloro- counterparts produced higher yields beyond
9
0% [39]. In addition, other kinds of aldehydes were also studied.
The reductive amination of the fused-ring aromatic aldehyde
2-naphthaldehyde) produced the secondary amine with a high
(
yield of 85% (Table 3, Entry 11). It should be pointed that the reduc-
tive amination using aliphatic and heterocyclic aldehydes is more
challenging. To our delight, the Co/N-C-600 catalyst was also active
for these transformations with excellent yields (Table 3, Entries
1
2–15).
The reductive amination of benzaldehyde with different nitro
compounds was also studied. Again, it was observed that nitroare-
nes with different substituted groups successfully reacted with
benzaldehyde to generate the corresponding secondary amines
with high yields (Table 3, Entries 16–25). Besides nitroarenes,
t 0
shows a plot of ln(C /C ) versus time for the transfer hydrogenation

2
70
P. Zhou et al. / Journal of Catalysis 352 (2017) 264–273
Table 3
Substrate scope of the developed catalytic system.a
O
NHR
R-NO2
+
R'
R''
R'
R''
Entry
1
Nitro compounds
Carbonyl compounds
Time (h)
10
Con. (%)
100
Sel. A (%)
96
NO
2
NO
2
NO
2
NO
2
Me
CHO
CHO
2
3
4
10
10
10
100
100
100
89
91
99
MeO
CHO
5
10
100
90
NO
2
6
7
8
10
10
10
100
100
100
92
91
55
NO
2
NO
2
NO
2
F
CHO
CHO
Cl
9
10
100
85
1
1
0
1
10
10
100
100
86
85
1
2
10
100
92
1
1
1
3
4
5
10
10
10
100
100
100
96
95
96
1
1
1
6
7
8
10
10
10
100
100
100
95
94
90
1
2
9
0
10
22
100
100
50
93
2
2
2
2
1
2
3
4
10
10
10
10
100
100
100
100
89
87
96
83
2
5
10
100
93
2
2
2
6
7
8
10
10
10
100
100
100
95
91
96
2
3
9
0
10
10
100
100
98
98
3
3
3
1
2
3
10
10
30
100
100
100
89
32
85
a
2
Reaction conditions: nitro compound (1 mmol), carbonyl compound (2 mmol), Co/N-C-600 (20 mg), H O (10 mL), CO (40 bar) and 170 °C.
ꢀ1
of nitrobenzene or N-benzylideneaniline at 70 °C and 40 bar CO.
0.0026 min for the transfer hydrogenation of nitrobenzene. The
kinetic studies once again confirmed that the transfer hydrogena-
tion of the imines determined the whole reaction rate of the one-
The rate constant of the transfer hydrogenation of
ꢀ1
N-benzylideneaniline was 0.0008 min
,
while that was

P. Zhou et al. / Journal of Catalysis 352 (2017) 264–273
271
Nitrobenzene
1
00
N-benzylideneaniline
N-benzylaniline
8
6
4
2
0
0
0
0
0
Aniline
0
100 200 300 400 500 600
Time (min)
Fig. 7. Time course of the products distribution during the reductive amination.
Reaction conditions: Nitrobenzene (1 mmol), benzaldehyde (2 mmol), H O (10 mL),
2
Co/N-C-600 (20 mg), 170 °C and 30 bar CO.
0
.0
-1
N-benzylideneaniline, k = 0.0008min
-
-
-
-
0.1
0.2
0.3
Scheme 2. Plausible reaction mechanism for the one-pot reductive amination by
transfer hydrogenation with CO/H
2
O over the Co/N-C-600 catalyst.
5
0
H O
-
1
2
Nitrobenzene, k=0.0026 min
45
D O
2
4
3
3
0
5
0
0.4
-
KIE = 2.3
50
0
50 100 150 200 250 300 350 400
Time (min)
25
2
1
1
0
5
0
5
0
Fig. 8. Kinetic studies of the transfer hydrogenation of nitrobenzene and N-
benzylideneaniline. Reaction conditions: Nitrobenzene or N-benzylideneaniline
(
2
1 mmol), H O (10 ml), Co/N-C-600 (20 mg), 70 °C and 40 bar CO.
pot reductive amination. In addition, the transfer hydrogenation of
N-benzylideneaniline was also performed over the three kinds of
the representative Co/N-C-600, Co/N-C-800 and Co/N-C-900 cata-
lysts, and the results are shown in Table S3. The results demon-
strated that the catalytic activity toward the imines (C@N bonds)
decreased in an order of Co/N-C-600, Co/N-C-800 and Co/N-C-
0
1
2
3
4
Time (h)
Fig. 9. Isotopic effect of hydrogenation of nitrobenzene with H
2 2
O and D O.
9
00, which was similar as those observed in Table 1 for the trans-
formation of nitrobenzene.
As discussed above, the key steps of the one-pot reductive ami-
nation are the two kinds of transfer hydrogenation reactions (nitro
groups and imines). To simplify the study of the mechanism, the
hydrogenation of nitrobenzene was used as the model reaction.
reaction conditions for the one-pot reductive aminations. During
the process, the pressure of the autoclave was observed to be con-
stant, and H was also not detected by GC. Thus, the transfer hydro-
2
genation reactions over the Co/N-C-600 catalyst should undergo by
the in situ CO-induced active hydrogen species. As proposed in
Scheme 2, CO should be firstly absorbed on the surface of the Co
As far as the transfer hydrogenation reactions with CO/H
might envisage that the present Co-catalyzed reaction should pro-
ceed via water–gas shift reaction (CO + H + H ) to pro-
O ! CO
duce H , which was the real reducing agent. As listed in Fig. S10,
the hydrogenation of nitrobenzene at 5 bar H produced much
2
O, one
0
nanoparticles via the electron interaction between the Co d orbital
2
2
2
and the C„O 2
p
⁄ antibonding orbital [41]. The electronegative
2
nitrogen atoms neighbor to Co sites acted as Lewis base to promote
the cleavage of HAOAH bond to generate proton (NAH ). In addi-
2
+
higher conversion than that with 5 bar CO at 90 °C. Hence, it seems
that the transfer hydrogenation of nitrobenzene occurred via
tion, the deuterium-labeling studies further confirm the proposed
mechanism (Fig. 9). A larger kinetic isotope effect (KIE) of 2.3
was observed using D O for the reduction of nitrobenzene, which
2
indicated that the reaction involves the cleavage of HAOAH, and
this step was also reported to be the first step for water–gas-
shift reaction [42]. On the meanwhile, the adsorbed CO can interact
water–gas-shift reaction with H
ever, the treatment of Co/N-C-600 catalyst in water at 5 bar CO and
0 °C did not produce any H by gas chromatography (GC). Fur-
2
as the real reducing agent. How-
9
2
thermore, the same experiment was also performed at a high tem-
perature of 170 °C and 40 bar CO for 10 h, which were the same

2
72
P. Zhou et al. / Journal of Catalysis 352 (2017) 264–273
with hydroxyl groups to form formate species, which was also
observed by other researchers for water–gas-shift reaction over
heterogeneous noble metal catalysts [42]. Then, one CO molecule
2
dently indicated that the Co/N-C-600 catalyst was highly stable,
due to the fact that the strong interaction between cobalt nanopar-
ticles and nitrogen atoms.
was released from the catalyst surface to generate cobalt-hydride
ꢀ
species (CoAH ). Indeed, CO
2
was detected in the gas mixture after
3.6. Conclusion
reaction. Although we cannot give the direct evidence on the for-
mation of formate species, we tried to perform the reaction by
To summarize: an unprecedented method was developed for
the synthesis of secondary amines via the one-pot reductive ami-
nation of carbonyl compounds with nitro compounds by transfer
the use of ammonium formate (HCOONH
4
). Because the Co/N-C-
as the hydro-
6
00 catalyst can react with acid, we used HCOONH
4
gen donor for the one-pot reductive amination of nitrobenzene and
benzaldehyde at 170 °C. As expected, 41.8% yield of the corre-
sponding secondary amine was produced after 10 h. These results
confirmed that the Co/N-C-600 catalyst was capable to promote
the reduction reactions by transformation hydrogenation with for-
mate species. Overall, this step resembles the process of the trans-
fer hydrogenation with formic acid as the hydrogen donor, in
which formate was firstly absorbed on the surface of metal cata-
hydrogenation with CO/H O over the Co/N-C-600 catalyst. Both
nitrogen and Co nanoparticles were of synergetic importance
2
toward this reaction. The Co/N-C-600 catalyst with the highest
nitrogen content and the smallest size of Co nanoparticles showed
the highest catalytic activity. By applying the Co/N-C-600 catalyst,
the one-pot reduction amination of carbonyl compounds with
nitro compounds affording the structurally diverse secondary ami-
nes with good to excellent amines yields (85 ꢁ 99%). Kinetic stud-
ies revealed that the transfer hydrogenation of imines was the
rate-determining step. A plausible mechanism was proposed based
on some controlling experiments, which involved the generation of
ꢀ
lysts. In our catalytic systems, the combination of CoAH with
+
NAH to generate H
2
was not observed, while this phenomenon
was observed over the noble metal catalysts with high activity
+
ꢀ
+
ꢀ
[
43]. The as-formed proton (NAH ) and a hydride (CoAH ) were
the active species for the hydrogenation of nitro groups and imines
C@N bonds) (Scheme 2), which was in accordance with the gen-
a proton (NAH ) and a hydride (CoAH ) as the active species. In
addition, the Co/N-C-600 catalyst was highly stable without the
loss of its catalytic activity. To the best our knowledge, there has
been no other a heterogeneous non-noble metal catalyst for the
(
+
ꢀ
eral accepted concept that a proton (H ) and a hydride (H ) were
used for the reduction in polar bonds such as C@O and C@N bonds
one-pot reductive amination by CO/H O. Thus, this method pro-
2
[
44]. The proposed mechanism was also supported by the results in
vides great potential for industrial application in the production
Table 1, which indicated that both nitrogen atoms and Co nanopar-
ticles were both important for the transfer hydrogenation with
CO/H O. The Co/N-C-600 catalyst showed the highest catalytic
2
of secondary amines via the one-pot reductive amination by trans-
fer hydrogenation.
activity, as it contains the highest nitrogen content and the small-
est size of Co nanoparticles.
Acknowledgments
This work was supported by the Chenguang Youth Science and
Technology Project of Wuhan City (No. 2014070404010212).
3.5. Recycling of the Co/N-C-600 catalyst
Finally, the stability of the Co/N-C-600 catalyst was also inves-
Appendix A. Supplementary material
tigated. After reaction, the Co/N-C-600 catalyst could be easily col-
lected from the reaction mixture with an external magnet. Then, it
was washed with ethanol to remove the absorbed organic chemi-
cals, and dried in a vacuum oven at 50 °C. The spent catalyst was
reused for the next run, and the reaction was performed under
the same reaction conditions. As depicted in Fig. 10,
N-benzylaniline yield remained almost the same during the recy-
cling runs. On the meanwhile, ICP analysis revealed that the cobalt
content in the reaction solution was negligible. These results evi-
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.jcat.2017.05.026.
References
[
[
1] N. Ono, The Nitro Group in Organic Synthesis, Wiley, New York, NY, 2001.
2] X.N. Guo, C.H. Hao, G.Q. Jin, H.Y. Zhu, X.Y. Guo, Angew. Chem. Int. Ed. 53 (2014)
1973–1977.
[3] C.W. Cheung, X.L. Hu, Nat. Commun. 7 (2016) 12494.
[
[
[
4] J. Feng, S. Handa, F. Gallou, B.H. Lipshutz, Angew. Chem. Int. Ed. 55 (2016)
8979–8983.
5] H. Yang, S.J. Bradley, A. Chan, G.I.N. Waterhouse, T. Nann, P.E. Kruger, S.G.
Telfer, J. Am. Chem. Soc. 138 (2016) (1881), 11872-11881.
6] B. Zhang, X.W. Guo, H.J. Liang, H.B. Ge, X.M. Gu, S. Chen, H.M. Yang, Y. Qin, ACS
Catal. 6 (2016) 6560–6566.
1
00
[
[
[
7] P. Ruiz-Castillo, S.L. Buchwald, Chem. Rev. 116 (2016) 12564–12649.
8] E.W. Baxter, A.B. Reitz, Org. React. 59 (2004) 1–714.
9] L.S. Li, Z.Q. Niu, S.F. Cai, Y. Zhi, H. Li, H.P. Rong, L.C. Liu, L. Liu, W. He, Y. Li, Chem.
Commun. 49 (2013) 6843–6845.
8
6
4
2
0
0
0
0
0
[
[
[
[
[
10] L. Hu, X. Cao, D. Ge, H. Hong, Z. Guo, L. Chen, X. Sun, J. Tang, J. Zheng, J. Lu, H.
Gu, Chem. Eur. J. 17 (2011) 14283–14287.
11] M. Pintado-Sierra, A.M. Rasero-Almansa, A. Corma, M. Iglesias, F. Sanchez, J.
Catal. 299 (2013) 137–145.
12] E.A. Artiukha, A.L. Nuzhdin, G.A. Bukhtiyarova, S.Y. Zaytsev, P.E. Plyusnin, Y.V.
Shubin, V.I. Bukhtiyarov, Catal, Sci. Technol. 5 (2015) 4741–4745.
13] L. Li, Z. Niu, S. Cai, Y. Zhi, H. Li, H. Rong, L. Liu, L. Liu, W. He, Y. Li, Chem.
Commun. 49 (2013) 6843–6845.
14] T. Stemmler, F.A. Westerhaus, A.-E. Surkus, M.-M. Pohl, K. Junge, M. Beller,
Green Chem. 16 (2014) 4535–4540.
[15] R.V. Jagadeesh, T. Stemmler, A.E. Surkus, H. Junge, K. Junge, M. Beller, Nat.
Protoc. 10 (2015) 548–557.
1
2
3
4
5
6
[
[
[
16] X.-F. Wu, H. Neumann, M. Beller, Chem. Soc. Rev. 40 (2011) 4986–5009.
17] D. Chusov, B. List, Angew. Chem. Int. Ed. 53 (2014) 5199–5201.
18] P.N. Kolesnikov, N.Z. Yagafarov, D.L. Usanov, V.I. Maleev, D. Chusov, Org. Lett.
17 (2015) 173–175.
Recycling number
Fig. 10. Recycling of the Co/N-C-600 catalyst toward the reductive amination of
benzaldehyde with nitrobenzene. Reaction conditions: Nitrobenzene (1 mmol),
[19] B.P. Bhardwaj, Steel and Iron Handbook, NPCS, New Delhi, 2014.
benzaldehyde (2 mmol), H
2
O (10 mL), Co/N-C-600 (20 mg), CO (40 bar) and 170 °C.
[20] A. Ambrosi, S.E. Denmark, Angew. Chem. Int. Ed. 55 (2016) 12164–12189.

P. Zhou et al. / Journal of Catalysis 352 (2017) 264–273
273
[
[
[
21] D. Chusov, B. List, Angew. Chem., Int. Ed. 53 (2014) 5199–5201.
22] JW. Park, YK. Chung, ACS Catal. 5 (2015) 4846–4850.
23] Z.S. Wu, L. Chen, J.Z. Liu, K. Parvez, H.W. Liang, J. Shu, H. Sachdev, R. Graf, X.L.
Feng, K. Mullen, Adv. Mater. 26 (2014) 1450–1455.
[33] P. Zhang, F. Sun, Z. Xiang, Z. Shen, J. Yun, D. Cao, Energy Environ. Sci. 7 (2014)
442–450.
[34] G. Wang, L.-T. Jia, Y. Zhu, B. Hou, D.-B. Li, Y.-H. Sun, RSC Adv. 2 (2012) 11249–
11252.
[
24] A. Aijaz, J. Masa, C. Rosler, W. Xia, P. Weide, A.J.R. Botz, R.A. Fischer, W.
Schuhmann, M. Muhler, Angew. Chem. Int. Ed. 55 (2016) 4087–4091.
25] F.A. Westerhaus, R.V. Jagadeesh, G. Wienhöfer, M.M. Pohl, J. Radnik, A.E.
Surkus, J. Rabeah, K. Junge, H. Junge, M. Nielsen, A. Brückner, M. Beller, Nat.
Chem. 5 (2013) 537–543.
[35] Y. Yang, L. Jia, B. Hou, D. Li, J. Wang, Y.J. Sun, Phys. Chem. C 118 (2014) 268–
277.
[36] A. Bumajdad, M. Madkour, Phys. Chem. Chem. Phys. 16 (2014) 7146–7158.
[37] L. He, L.C. Wang, H. Sun, J. Ni, Y. Cao, H.Y. He, K.N. Fan, Angew. Chem. Int. Ed. 48
(2009) 9538–9541.
[
[
26] H. Furukawa, K.E. Cordova, M. O’Keeffe, O.M. Yaghi, Science 341 (2013)
[38] M. Viganò, F. Ragaini, M.G. Buonomenna, R. Lariccia, A. Caselli, E. Gallo, S.
Cenini, J.C. Jansen, E. Drioli, ChemCatChem. 2 (2010) 1150–1164.
[39] A. Mukherjee, D. Srimani, S. Chakraborty, Y. Ben-David, D. Milstein, J. Am.
Chem. Soc. 137 (2015) 8888–8891.
[40] R.I. Storer, D.E. Carrera, Y. Ni, D.W.C. MacMillan, J. Am. Chem. Soc. 128 (2006)
84–86.
[41] J. Tan, J. Cui, X. Cui, T. Deng, X. Li, Y. Zhu, Y. Li, ACS Catal. 5 (2015) 7379–7384.
[42] C.M. Kalamaras, P. Panagiotopoulou, D.I. Kondarides, A.M. Efstathiou, J. Catal.
264 (2009) 117–129.
[43] Q. Liu, X. Yang, Y. Huang, S. Xu, X. Su, X. Pan, J. Xu, A. Wang, C. Liang, X. Wang,
T. Zhang, Energy Environ. Sci. 8 (2015) 3204–3207.
[44] A.A. Mikhailine, M.I. Maishan, A.J. Lough, R.H. Morris, J. Am. Chem. Soc. 134
(2012) 12266–12280.
1
230444.
[
[
27] W. Zhong, H.L. Liu, C.H. Bai, S.J. Liao, Y.W. Li, ACS Catal. 5 (2015) 1850–1856.
28] T.Y. Yang, H.J. Ling, J.F. Lamonier, M. Jaroniec, J. Huang, M.J. Monteiro, J. Liu,
NPG Asia Mater. 8 (2016) e240.
[
[
[
[
29] X. Xu, Y. Li, Y. Gong, P. Zhang, H. Li, Y. Wang, J. Am. Chem. Soc. 134 (2012)
1
6987–16990.
30] J.F. Blandez, A. Primo, A.M. Asiri, M. Álvaro, H. García, Angew. Chem. Int. Ed. 53
2014) 12581–12586.
31] Y. Hou, Z.H. Wen, S.M. Cui, S.Q. Ci, S. Mao, J.H. Chen, Adv. Funct. Mater. 25
2015) 872–882.
32] W.N. Wang, Y. Itoh, I.W. Lenggoro, K. Okuyama, Mater. Sci. Eng. B 111 (2004)
9–76.
(
(
6