H2 Activation with Co Nanoparticles Encapsulated in N-Doped Carbon Nanotubes for Green Synthesis of Benzimidazoles

Article abstract of DOI:10.1002/cssc.202002344

Co nanoparticles (NPs) encapsulated in N-doped carbon nanotubes (Co@NC₉₀₀) are systematically investigated as a potential alternative to precious Pt-group catalysts for hydrogenative heterocyclization reactions. Co@NC₉₀₀ can efficiently catalyze hydrogenative coupling of 2-nitroaniline to benzaldehyde for synthesis of 2-phenyl-1H-benzo[d]imidazole with >99 % yield at ambient temperature in one step. The robust Co@NC₉₀₀ catalyst can be easily recovered by an external magnetic field after the reaction and readily recycled for at least six times without any evident decrease in activity. Kinetic experiments indicate that Co@NC₉₀₀-promoted hydrogenation is the rate-determining step with a total apparent activation energy of 41±1 kJ mol^?1. Theoretical investigations further reveal that Co@NC₉₀₀ can activate both H₂ and the nitro group of 2-nitroaniline. The observed energy barrier for H₂ dissociation is only 2.70 eV in the rate-determining step, owing to the presence of confined Co NPs in Co@NC₉₀₀. Potential industrial application of the earth-abundant and non-noble transition metal catalysts is also explored for green and efficient synthesis of heterocyclic compounds.

Full text of DOI:10.1002/cssc.202002344

Full Papers
doi.org/10.1002/cssc.202002344
ChemSusChem
1
2
3
4
5
6
7
8
9
H2 Activation with Co Nanoparticles Encapsulated in
N-Doped Carbon Nanotubes for Green Synthesis of
Benzimidazoles
Chuncheng Lin,[a] Weihao Wan,[a] Xueting Wei,[a] and Jinzhu Chen*[a]
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Co nanoparticles (NPs) encapsulated in N-doped carbon nano-
tubes (Co@NC900) are systematically investigated as a potential
alternative to precious Pt-group catalysts for hydrogenative
heterocyclization reactions. Co@NC900 can efficiently catalyze
hydrogenative coupling of 2-nitroaniline to benzaldehyde for
synthesis of 2-phenyl-1H-benzo[d]imidazole with >99% yield at
ambient temperature in one step. The robust Co@NC900 catalyst
can be easily recovered by an external magnetic field after the
reaction and readily recycled for at least six times without any
evident decrease in activity. Kinetic experiments indicate that
Co@NC900-promoted hydrogenation is the rate-determining
step with a total apparent activation energy of 41�1 kJmolÀ 1.
Theoretical investigations further reveal that Co@NC900 can
activate both H2 and the nitro group of 2-nitroaniline. The
observed energy barrier for H2 dissociation is only 2.70 eV in the
rate-determining step, owing to the presence of confined Co
NPs in Co@NC900. Potential industrial application of the earth-
abundant and non-noble transition metal catalysts is also
explored for green and efficient synthesis of heterocyclic
compounds.
Introduction
transfer hydrogenation of nitroarenes.[18–24] A synergistic effect
between the embedded transition metal NPs and the sur-
rounded N-doped carbon was suggested for the observed
catalytic performance of these encapsulated catalyst.[25–30]
Confined Co was investigated for hydrogenation of quinoline
and biomass-based aldehyde, ketone, and carboxyl, as well as
transfer hydrogenation of N-heterocycles.[22,24,31,32] Finally, em-
bedded bimetallic NiCo was examined with selective hydro-
genation of alkenes and alkynes.[23] In contrast, oxidative
reactions mainly focus on encapsulated Co catalysts for
oxidative dehydrogenation of N-heterocycles, aerobic oxidation
of alcohols, hydrocarbons, and p-cresols.[33–35]
Generally, the encapsulated structure effectively enhances
catalyst stability by protecting the confined metal-NPs within
the carbon shells from leaching, aggregation, poisoning and
harsh/corrosive reaction conditions. For example, an acid-
resistant catalyst of carbon nitride (CN) with underlying Ni was
investigated for nitroarene hydrogenation under strongly acidic
conditions.[36] In this nanocomposite catalyst, the inert outer
layer of CN protects active Ni from acid-corrosion; while, the
inner metallic Ni core activates the inert CN for hydrogenation.
Additionally, theoretical research on these encapsulated cata-
lysts demonstrates electron transfer from the confined metal to
the nearby carbon shells, which thereby redistributes electron
density, increases the density of state (DOS) near Fermi level,
lowers the local work function, promotes formation of active
sites, and reduces adsorption free energy of the reaction
substrate on the outer carbon surface.[1,37,38] Inspired by the
above results, a robust encapsulated catalyst was thus designed
in this research for syntheses of benzimidazole compounds
with widely pharmaceutical, medicinal and industrial interests.
As typically important bioactive heterocyclic compounds,
benzimidazoles have been extensively investigated for their
broad spectrum in biological and pharmacological activities in
Efforts to explore earth-abundant and non-noble transition
metal catalysts have always been going on in industry in order
to bring an efficient catalytic process into real applications.[1–3]
Recently, non-precious transition-metal-based catalysts, involv-
ing Fe, Co and Ni, were developed by encapsulating into
nitrogen-doped or nitrogen-containing carbon frameworks to
improve their stability and durability.[4–6] These encapsulated
catalysts have received a growing attention as potential
alternatives to precious Pt-group cathodes in various electro-
chemical reactions such as CO2 reduction, water splitting,
oxygen reduction and hydrogen evolution.[5,7–12] These confined
electro-catalysts demonstrate high activity, small overpotential
and long-term durability in strong acidic or basic electrolytes
and exhibit quite close electro-catalytic performance to com-
mercially available Pt/C catalyst.[9,13] Evidently, these embedded
catalysts are typically identified by their stability and durability
with high tolerance to reaction medium when compared with
surface-supported catalysts by traditional preparation
method.[14–17]
Notably, in sharp contrast, research on thermally induced
redox transformations is very limited by using these encapsu-
lated transition metal catalysts. In the cases of reductive
reactions, for instance, encapsulated Fe, Co, Ni and bimetallic
FeCo, NiCo nano-catalysts were reported for hydrogenation and
[a] C. Lin, W. Wan, X. Wei, Prof. J. Chen
Department of Chemistry, College of Chemistry and Materials Science
Jinan University, No. 855, East Xingye Avenue, Panyu District, Guangzhou
511443 (P. R. China)
.
E-mail: chenjz@jnu.edu.cn
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/cssc.202002344
ChemSusChem 2020, 13, 1–13
1
© 2020 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cssc.202002344
ChemSusChem
antibacterial, antifungal, antiviral, anti-inflammatory, anti-ulcer,
antihypertensive, antihistamine, antitumor, anticancer, and anti-
HIV.[39–44] Therefore, various synthetic methods were
developed.[45–48] However, these synthetic processes generally
suffer from various limitations such as the use of strong acidic
reaction medium, toxic solvents, high reaction temperature,
reactive reagent, strong dehydrating agent, oxidizing agent,
As an important and efficient application of this method,
commercially available fuberidazole (4l, Scheme 1c), a seed-
dressing agent against fusarium disease,[57–59] was quantitatively
obtained by Co@NC900 in one step under mild reaction
conditions.
1
2
3
4
5
6
7
8
9
and noble metal catalysts.[49–54] For example, the most tradi- Results and Discussion
tional method for 2-phenyl-1H-benzo[d]imidazole (4a) synthesis
involves condensation of 1,2-phenylenediamine (2) with benzal-
dehyde (3a, Scheme 1a),[55] whereas 2 is industrially obtained
by 2-nitroaniline (1a) reduction.[56] Therefore, a directly hydro-
genative coupling between 1a and 3a in one-step is a very
promising and green route for 4a synthesis by using transition
metal catalysts.
Herein, we report that Co NPs encapsulated in N-doped
carbon nanotubes (Co@NC900), as a robust and magnetic
catalyst, can efficiently promote hydrogenative coupling of 1a
[or 1,2-dinitrobenzene (5a)] to 3a for 4a synthesis with the
yield exceeding 99% at room temperature (Scheme 1a). Addi-
tionally, Co@NC900 can readily catalyze reductive heterocycliza-
tion of a wide range of substrates 1 or 5 to various aldehydes 3
for synthesis of up to 34 different benzimidazoles (Scheme 1b).
In this research, the sample of Co NPs encapsulated in N-doped
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
carbon nanotubes (denoted as Co@NCT) was obtained by a
multi-step pyrolysis a mixture of triblock copolymer P123,
melamine and cobalt acetate at various temperature under N2
atmosphere, followed by a thoroughly leaching with 1 M HCl at
°
80 C for 12 h (see the Supporting Information, Figure S1). The
resulting magnetic Co@NCT sample was further examined with
inductively coupled plasma-atomic emission spectroscopy (ICP-
AES) until no soluble Co ions was observed before the
subsequent characterization and catalytic reaction. The sub-
script T in Co@NCT indicates the final pyrolysis temperature for
Co@NCT formation. For instance, Co@NC900 is obtained by
pyrolysis a mixture of P123, melamine and cobalt acetate at
°
900 C. For comparison purpose, N-doped carbon (NC900) was
prepared with similar synthetic procedure as Co@NC900 without
addition of cobalt acetate. To probe the effect of the confined
Co NPs on the catalytic performance of Co@NC900, Co catalysts
supported on the surface of Co@NC900 (Co/Co@NC900) and NC900
(Co/NC900) were also investigated in this research.
The microstructures of Co@NC900 was initially analyzed by
Scanning Electron Microscopy (SEM) and transmission electron
microscopy (TEM). The SEM image of Co@NC900 exhibits a
regular and tubular morphology (Figure S3a,b). According to
the TEM images, Co@NC900 adopts a typical architecture of
bamboo-like carbon nanotubes with diameter ranging from 30–
50 nm (Figure 1a,b). Moreover, the cobalt NPs, with the
diameter range of 15–25 nm, are encapsulated at the tip or in
the middle of the nanotubes. The high resolution TEM of
Co@NC900 further confirms that the Co NPs are confined by thin
layers of graphitic carbon with the carbon-wall thickness around
1–2 nm (Figure 1c).
The Co@NC900 inner core exhibits (200) lattice fringes with
an interspacing of 0.17 nm, which is in accordance with the Co
crystal structure, whereas the outer carbon-shell shows graphite
(002) fringes with an interspacing of 0.34 nm. Moreover, TEM
Energy-Dispersive X-ray Spectrometry (EDX) confirms the pres-
ence of C, N and Co elements in Co@NC900. The Co core of
Co@NC900 is encapsulated by an N-doped carbon layer, as
shown by EDX elemental mapping (Figure 1i–l). The TEM
images of both Co@NC800 (Figure 1d,e) and Co@NC700 (Fig-
ure 1f,g) show irregular and porous sponge-like structure with
confined Co NPs as cores of graphitic carbon. The lattice fringes
of Co crystal and graphite in both Co@NC800 and Co@NC700 are
similar with those of Co@NC900 sample. The formation of
porosity in both Co@NC800 and Co@NC700 is attributed to HCl
leaching with incompletely coated Co NPs as porogens.
Evidently, pyrolysis temperature plays a key role on the
architecture of the resulting Co@NCT series samples. The TEM
Scheme 1. a) Synthetic route to 4a. b) Co@NC900-promoted hydrogenative
coupling of 1 or 5 with 3 for the synthesis of 4. c) Synthesis of 4l.
ChemSusChem 2020, 13, 1–13
www.chemsuschem.org
2
© 2020 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cssc.202002344
ChemSusChem
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Figure 1. TEM images: a–c) Co@NC900; d,e) Co@NC800; f,g) Co@NC700; h) NC900. i–l) TEM-EDX of Co@NC900
.
image of NC900 exhibits thin-layer nanosheet structure (Fig-
ure 1h), demonstrating the promotion effect of Co NPs on the
formation of nanotube morphology. The TEM images of Co/
NC900 show supported Co NPs on the thin layers carbon-
nanosheet of NC900 support (Figure S3c,d). Similarly, TEM
analysis of Co/Co@NC900 confirms the loading of Co NPs on the
out surface of tubular Co@NC900 support (Figure S3e,f).
The N2 sorption isotherm of Co@NC900 shows a type-IV
isotherm with a specific Brunauer-Emmet-Teller (BET) surface
area of 648 m2 gÀ 1 and an average pore diameter of 3.5 nm
(Figure 2a,b and Table 1). The evident hysteresis loop in a wide
range from 0.4 to 1.0 P/P0 further confirms the existence of
mesoporous structure. In contrast to the tubular morphology of
Co@NC900, Co@NC800 has a porous sponge-like structure, thus
yielding an increased specific BET surface area to 708 m2 gÀ 1
(Table 1). Co@NC700 exhibits a decreased specific BET surface
area to 541 m2 gÀ 1 (Table 1). Generally, all Co@NCT serial samples
displayed a mesopore-dominated structure with a mean pore
diameter of 3.5–3.9 nm, demonstrating a porogen function of
the unconfined Co NPs. NC900 has a specific BET surface area of
around 384 m2 gÀ 1, which decreased to 297 m2 gÀ 1 upon loading
of Co NPs to give Co/NC900 (Table 1 and Figure S4a,b). Similarly,
supporting Co on Co@NC900 leads to a reduction in surface area
to 455 m2 gÀ 1 for the Co/Co@NC900 sample (Table 1 and Fig-
ure S4a,b).
The XRD diffraction patterns of Co@NC900 display one weak
°
peak at 44.2 corresponding to (111) plane of metallic cobalt
(Figure 2c), which indicates the presence of metallic cobalt
Table 1. Textural parameters of various sample investigated in this research.
[a]
[b]
[c]
[d]
[e]
[f]
Sample
SBET
Smicro
Smeso
D
[nm]
micro/Dmeso
Vtotal
V
micro/Vmeso
[m2 gÀ 1
]
[m2 gÀ 1
]
[m2 gÀ 1
]
[cm3 gÀ 1
]
[cm3 gÀ 1
]
Co@NC900
Co@NC800
Co@NC700
NC900
Co/Co@NC900
Co/NC900
648
708
541
384
455
297
74
87
57
98
110
73
574
621
484
285
345
224
1.0/3.5
-/3.5
1.0/3.9
1.0/3.5
1.0/3.5
1.2/3.7
1.23
1.62
1.05
2.00
1.85
0.77
0.04/1.19
0.12/1.50
0.03/1.02
0.08/1.92
0.03/1.82
–/0.76
[a] SBET is the specific surface area calculated by multipoint BET method. [b] Smicro is the specific surface area of micropores calculated by t-plot method. [c]
meso is the surface area of mesopores calculated by SBET–Smicro. [d] Dmicro and Dmeso are the micropore and mesopore diameters, respectively, estimated from the
S
local maximum of DFT pore size distribution obtained in the adsorption branch of the N2 isotherm. [e] Vtotal is the total specific pore volume determined by
using the adsorption branch of the N2 isotherm at P/P0 =0.99. [f] Vmicro is the specific pore volume of micropores determined by the t-plot method and Vmeso is
the specific pore volume of mesopores calculated by Vtotal–Vmicro
.
ChemSusChem 2020, 13, 1–13
www.chemsuschem.org
3
© 2020 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cssc.202002344
ChemSusChem
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Figure 2. a) N2 adsorption-desorption isotherm of Co@NCT serial samples and NC900; b) the corresponding pore size distribution. c) XRD patterns of Co/
Co@NC900, Co/NC900, Co@NCT serial samples and Co. d) EDX spectra of Co@NC900
.
°
species in Co@NC900. Moreover, a broad peak at 26.2 for
ICP-AES measurement shows a surface Co content of only
0.14 wt% for Co@NC900
Co@NC900 is assigned to diffractions from the (002) planes of
graphite. Therefore, the XRD analysis conforms a Co@NC900
structure of encapsulated metallic Co NPs in N-doped carbon
nanotubes, which is in line with its TEM analysis (Figure 1a–c).
Both Co@NC800 and Co@NC700 samples show similar XRD
patterns with Co@NC900 (Figure 2c); however, the graphite
diffraction patterns significantly decreased due to a reduced
pyrolysis temperature for these two samples. In the cases of Co/
.
The surface elemental composition and chemical state of
Co@NC900 were determined by X-ray photoelectron spectro-
scopy (XPS). The survey XPS of Co@NC900 demonstrates the
presence of C (90.9 wt%), N (2.5 wt%), O (5.9 wt%) and Co
(0.6 wt%) on the sample surface (Figure 3a and Table S1). In the
case of high-resolution Co 2p XPS, the binding energies at
780.6 and 795.4 eV can be assigned to the CoÀ Nx moieties,
which supports the existence of the coordination between Co
and Nx species (Figure 3b).[32,60] Moreover, a nearby peak at
778.8 eV is indexed to the metallic Co species, which is
accompanied by two characteristic Co shake-up (satellites)
peaks at 785.2 and 802.5 eV.[61] The high-resolution N 1s XPS of
Co@NC900 are deconvoluted into four peaks (Figure 3c), corre-
sponding to graphitic-N (401.5 eV), pyrrolic-N (400.9 eV), CoÀ N
(399.7 eV), and pyridinic-N (398.7 eV).[62,63] Finally, the C 1s XPS
shows four different types of C species which involves C=C
(284.8 eV, 53.8%), CÀ O/C=N (285.5 eV, 17.8%), CÀ N/C=O
(286.6 eV, 15.0%) and OÀ C=O (290.1 eV, 13.4%), thus indicating
the existence of doped heteroatoms such as N and O.[61,62]
°
°
Co@NC900 and Co/NC900, two peaks at 26.2 and 44.2 are
respectively assigned to the (002) plane of graphite and (111)
plane of metallic cobalt (Figure 2c).
Energy-dispersive X-ray spectroscopy (EDX) of Co@NC900
was then performed to analyze its bulk elemental constituents.
The EDX of Co@NC900 shows a C peak at 0.28 KeV together with
an N tail peak (Figure 2d). Moreover, Co@NC900 exhibits three
Co peaks at 0.78 (CoLα1), 6.93 (CoLβ1) and 7.65 KeV (CoLγ1).[38]
The EDX investigation of Co@NC900 further reveals its composi-
tion of C (85.0 wt%), N (1.3 wt%), O (6.6 wt%) and Co
(7.1 wt%). A subsequent elemental analysis of Co@NC900 show
its composition of C (83.8 wt%), N (2.2 wt%), O (6.0 wt%) and H
(1.9 wt%), which is close to the DEX result. However, due to the
effective protection of Co NPs by the N doped carbon layer, our
The 0.6 wt% Co content in Co@NC900, observed by XPS
analysis, is significantly lower than calculated from EDX
(7.1 wt%) due to the sensitivity of XPS analysis to the surface
ChemSusChem 2020, 13, 1–13
www.chemsuschem.org
4
© 2020 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cssc.202002344
ChemSusChem
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Figure 3. XPS spectra of Co@NC900 and Co/Co@NC900: a) Full survey spectrum; b) Co 2p; c) N 1s; d) C 1s.
Table 2. Reductive heterocyclization of 1a with 3a for synthesis of 4a
with various catalysts.[a]
chemical composition. Presumably, the Co NPs confined in the
graphitic carbon layers of Co@NC900 can hardly be detected by
XPS analysis. However, our thermogravimetric analysis (TGA)
indicates Co content of 5.7 wt% based on the assumption that
all Co species are transferred into Co3O4 and all C elements are
burnt out (Figure S4c).[64] Therefore, the Co content of Co@NC900
determined by EDX is close to TGA result. The above Co content
analysis thus demonstrates well protection of graphitic carbon
thin layers to the encapsulated Co core in Co@NC900. The Co/
Co@NC900 shows almost the same XPS to Co@NC900 with,
however, significantly increased surface Co content (5.7 wt%,
Figure 3a,b and Table S1). Moreover, the XPS of Co@NC800 and
Co@NC700 are very close to that of Co@NC900 (Table S1 and
Figure S5).
Entry
Catalyst (mg)
1a conversion
[%]
4a yield[a]
[%]
1
2
/
/
/
/
/
NC900 (10)
3
Co@NC900 (20)
Co@NC900 (20)
Co@NC900 (10)
Co@NC800 (10)
Co@NC700 (10)
Co/NC900 (10)
Co/Co@NC900 (10)
>99
/
72
60
20
/
>99
/
71
57
18
/
4[b]
5
6
7
8
9
34
32
The obtained Co@NC900 was then examined as catalyst for
model reaction of 4a synthesis by reductive heterocyclization
of 1a with 3a under H2 atmosphere (Table 2). The blank
experiment reveals that 4a was unobserved without using
Co@NC900 (Table 2, entry 1). Moreover, NC900 cannot promote
the reductive heterocyclization (Table 2, entry 2). In sharp
contrast, Co@NC900 shows remarkable catalytic activity to the
reaction by producing quantitative yield of 4a (Table 2, entry 3).
As expected, Co@NC900 fails to yield 4a under N2 atmosphere
(Table 2, entry 4). Moreover, the reductive heterocyclization is
promoted by an increased loading amount of Co@NC900
[a] Reaction conditions: 1a (0.2 mmol), catalyst (10–20 mg), 3a (0.4 mmol),
H2 (1.0 MPa), ethyl acetate (EtOAc, 4 mL), RT, 20 h. [b] N2 (1.0 MPa) was
used instead of H2 (1.0 MPa).
(Table 2, entries 3 and 5), demonstrating an enhanced avail-
ability of catalytic active sites for the reaction. The Co@NCT
serial samples were prepared by pyrolysis a mixture of P123,
melamine and cobalt acetate under N2 atmosphere. Therefore,
the correlation between Co@NCT activity and its pyrolysis
°
temperature reveals that Co@NC900, obtained from 900 C-
ChemSusChem 2020, 13, 1–13
www.chemsuschem.org
5
© 2020 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cssc.202002344
ChemSusChem
pyrolysis, is the most active among the Co@NCT samples
(Table 2, entries 5–7).
Table 3. Co@NC900-promoted benzimidazole syntheses from 1a/5a and
various aldehydes 3.[a]
1
2
Previously, Co-based catalysts were reported for hydro-
genation reactions.[65] Therefore, the presence of Co species in
Co@NC900 could be presumably responsible for the observed
activity to H2 activation. The active species may come from
either surface residual Co or encapsulated Co NPs within the
NC900 shell of Co@NC900. To address the role of Co on the
hydrogenation, the effect of surface residual Co on the
reductive heterocyclization was then investigated by exper-
imental methods. Two alternative strategies by catalyst-purifica-
tion and impurification methods were carried out.[66] For
purification, Co@NC900 was further treated with concentrated
HCl to remove surface Co species as much as possible. However,
the obtained purified Co@NC900 shows almost the same
catalytic activity with its precursor. This result thus demon-
strates that the reduced surface-Co species cannot decrease the
catalytic activity of Co@NC900. In the case of impurification
strategy, additional Co was loaded on the surface of Co@NC900
to give Co/Co@NC900 with surface Co loading level around
6 wt%. However, the resulting Co/Co@NC900 exhibits a signifi-
cantly reduced 4a yield to 32% when compared with a 71%
yield from Co@NC900 (Table 2, entries 5 and 9). This observation
thus reveals that the addition of Co species on the Co@NC900
surface cannot boost its hydrogenation activity. In addition to
Co@NC900 support, Co was also supported on NC900 to give a
6 wt% Co/NC900 sample for comparison purpose. However, the
obtained Co/NC900 shows a negligible activity towards the
reductive heterocyclization (Table 2, entries 5 and 8). Therefore,
the surface residual Co is thought to inhibit rather than
promote the hydrogenation activity of Co@NC900. The presence
of residual Co species on Co@NC900 might block true active sites
of the catalyst, and thus reduces its catalytic activity. Recent
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
[a] Reaction conditions: Co@NC900 (20 mg), 1a or 5a (0.20 mmol), 3
°
(0.40 mmol), EtOAc (4 mL), H2 (1.0 MPa), 60 C for conversion of 1a and
°
70 C for that of 5a, 20 h. Values in parentheses are yields obtained from
°
°
5a. [b] T=RT. [c] T=100 C. [d] T=130 C.
research suggests that
a synergetic effect between the
embedded transition metal NPs and the surrounded N-atoms is
responsible for a heterolytic dissociation of the hydrogen
molecule.[25–30] This mechanism is more energetically favorable
when compared to the homolytic H2 dissociation process
observed for traditional surface-supported NPs catalysts. This
phenomenon would also explain the results herein reported, a
better performance of Co@NC900 compared to Co/NC900. In
conclusion, Co@NC900 might contain negligible surface Co
species; however, the catalytic activity of Co@NC900 should be in
line with its intrinsic activity as proved by the above controlled
experiments from purification and impurification strategies.
Therefore, the encapsulated Co NPs within the NC900 shell of
Co@NC900 might actually contribute to the H2 activation. Its role
was subsequently investigated by density functional theory
(DFT) calculations (will be discussed below).
produce quantitative yields of the corresponding benzimida-
zoles (4b–e). In the case of halogen-substituted benzaldehydes,
both
p-fluorobenzaldehyde
and
p-chlorobenzaldehyde
smoothly undergo the heterocyclization, producing excellent
yields of the corresponding benzimidazoles 4f and 4g, whereas
p-bromobenzaldehyde affords moderate yield of 2-(4-bromo-
phenyl)-1H-benzo[d]imidazole (4h). As expected, the heterocyc-
lization of bulky 1-naphthaldehyde also leads to a moderate
yield of 2-(1-naphthalenyl)-1H-benzo[d]imidazole (4i), suggest-
ing the presence of steric hindrance effect of the aromatic
aldehyde. However, a quantitative 4i yield up to 97% is
observed with the increased reaction temperature. In addition,
2-heteroaromatic-substituted benzimidazoles, bearing 2-pyridyl
(4j), 2-thienyl (4k), or 2-furyl (4l,m) substituents, are readily
prepared in excellent yields from the corresponding hetero-
aromatic aldehydes. Transition metal-based homogeneous
catalysts were previously reported to undergo deactivation with
heteroaromatic compounds containing sulfur atoms (4k).[67]
Notably, both furfural and 5-hydroxymethylfurfural are biomass-
based aromatic aldehydes and are applicable to the developed
catalytic system (4l,m). Moreover, the commercially available
Therefore, 4a yield of >99% with complete 1a conversion
were observed with Co@NC900 under very mild conditions by
using 1.0 MPaH2 at ambient temperature. Encouraged by the
above result, the 2-arylbenzimidazole scope was further inves-
tigated by using various aromatic aldehydes 3 with Co@NC900
.
Generally, the method tolerates a wide range of aldehydes for
the heterocyclization (Table 3). The substituted benzaldehydes
with electron-donating groups, including methyl and methoxy,
ChemSusChem 2020, 13, 1–13
www.chemsuschem.org
6
© 2020 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cssc.202002344
ChemSusChem
fungicide fuberidazole (4l) was obtained in quantitative yields
from 1a and furfural. The above results thus indicate that
Co@NC900 can efficiently promote heterocyclization of 1a with a
wide range of aromatic aldehydes 3 for various benzimidazole
syntheses.
1
2
3
In addition to 1a, 5a is smoothly transferred into
benzimidazoles 4 in excellent to good yields by Co@NC900
-
4
promoted reductive heterocyclization with various aldehydes 3
under H2 atmosphere (Table 3). Under such conditions,
Co@NC900 may initially promote hydrogenation of 5a to give 1a
as an in situ-formed “intermediate”, as a very close yields of 4
are observed with the two precursors of 5a and 1a in the
heterocyclization reaction (Table 3). Finally, various substituted
1a, including methyl (1b–f), tert-butyl (1g), methoxy (1h) and
fluoro groups (1i), smoothly undergo the heterocyclization
giving excellent yields of 4 (4n–s, Table 4), further extending
the benzimidazole scope. Notably, steric hindrance effect was
observed with tert-butyl substituted 2-nitroaniline (1g) in the
heterocyclization (4q, Table 4).
Scheme 2a thus shows a proposed mechanism for trans-
formation of 5a into 4a according to our experiments and
previous reports. 5a is initially hydrogenated to 1a by
Co@NC900. The subsequent path a (Scheme 2a) demonstrates
that successive hydrogenation of 1a leads to formation of 2,
followed by a condensation of 2 with 3a to give amino-imine 6.
In contrast, path b (Scheme 2a) shows formation of 6 through
preliminary condensation of 1a with 3a and a subsequent
hydrogenation of nitro-imine 7. Path a may be a prevailing
route due to a higher pKa of amino group in 2 than in the case
of 1a, which can facilitate the subsequent condensation
process. Intramolecular cyclization of 6 and the following
dehydrogenation of 8 yield 4a with aromatization as a potential
driving force. The oxidative dehydrogenation of 8 can be
promoted by Co@NC900 catalyst (Scheme 2a, path c) and/or 1a/
5a-oxidation (Scheme 2a, path d).[23]
5
6
7
8
[a]
Table 4. Synthesis of benzimidazoles from 1 and 3a using Co@NC900
.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
To verify the proposed mechanism, the Co@NC900-promoted
transformation of 1a into 4a was further investigated with
[a] Reaction conditions: Co@NC900 (20 mg), 1 (0.20 mmol), 3a (0.40 mmol),
°
°
EtOAc (4 mL), H2 (1.0 MPa), 60 C, 20 h. [b] T=100 C.
Scheme 2. a) Proposed mechanism and b) controlled experiments for formation of 4a.
ChemSusChem 2020, 13, 1–13
www.chemsuschem.org
7
© 2020 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cssc.202002344
ChemSusChem
time-dependent gas chromatography-mass spectrometry (GC-
MS) analysis. 4a (m/z 194.1) was detected to increase with
reaction time (Figure 4 and Figure S6), suggesting a final
product character. Whereas, three species of 2 (m/z 108.1), 6
([6-H]À , m/z 195.1) and 8 ([8-H]À , m/z 195.1) were observed to
be formed initially with very low abundances and later
consumed in the interval of time (Figure 4 and Figure S6), thus
indicating an intermediate behavior. The above results accord-
ingly suggest path a in Scheme 2a for transformation of 1a into
4a. Low intensities of the intermediates 2, 6, and 8 demonstrate
their fast conversions upon in situ formation without further
accumulation (Figure 4). Therefore, hydrogenation of 1a was
presumably the rate-determining step for 4a formation (Sche-
me 2a). While, condensation of 2 with 3a, cyclization of 6, and
dehydrogenation of 8 may smoothly proceed with fast reaction
rates under the investigated conditions (Scheme 2a).
Both paths a and b for transformation of 1a into 6
(Scheme 2a) were further investigated and compared based on
reaction kinetics. Path a involves a tandem reaction of hydro-
genation of 1a to 2 and condensation of 2 with 3a to yield 6.
The kinetics of the hydrogenation of 1a promoted by Co@NC900
shows a linear correlation of the initial 1a hydrogenation rate
with Co@NC900 loading level (Figure 5a) and H2 pressure (Fig-
ure 5b) at various reaction temperatures. Therefore, first-order
kinetics of 1a hydrogenation rate with respect to the Co@NC900
loading amount and H2 pressure are respectively suggested
based on the proportionately positive influence of catalyst
concentration and H2 pressure on the hydrogenation of 1a.
However, a zero-order kinetic dependence of the initial 1a
hydrogenation rate on 1a concentration is observed (Figure 5c),
suggesting a quick adsorption of 1a on the Co@NC900 surface.
Generally, the kinetic behaviors of Co@NC900-promoted hydro-
genation of 1a are very close to our recently reported
nitrobenzene hydrogenation with phosphorus-doped carbon
nanotubes as a metal-free catalyst.[68] Therefore, a pseudo-first-
order kinetics is applied to hydrogenation of 1a with respect to
H2 pressure if considering Co@NC900 loading level as a constant
in the reaction. This assumption is further proven by a linear
fitting of ln(1-X) vs reaction time (Figure 5d). The resulting
Arrhenius plot of (ln kobs) vs (1/T×103) indicates a total apparent
Ea of 41�1 kJmolÀ 1 for hydrogenation of 1a (Figure 5g).
1
2
3
4
5
6
7
8
9
In the case of 2 condensation with 3a, the reaction follows
a second-order kinetics,[69] showing proportionally linear de-
pendence of initial condensation rate on both concentrations of
2 and 3a. A linear kinetic profile of 1/(1-X) vs reaction time
further confirms a second-order kinetics of the condensation
reaction (Figure 5e). The total apparent Ea is 38�3 kJmolÀ 1 for
the condensation of 2 with 3a based on the Arrhenius equation
as shown in Figure 5g. For the condensation of 1a and 3a by
path b (Scheme 2a), the total apparent Ea is 65�2 kJmolÀ 1
(Figure 5f,g). Therefore, our kinetic experiments reveal total
apparent Ea values of 41�1 kJmolÀ 1 for hydrogenation of 1a
and 38�3 kJmolÀ 1 for 2!3a condensation in path a, which are
all significantly lower than 65�2 kJmolÀ 1 for the 1a!3a
condensation step in path b. Thus, path a should be a
predominant route for the transformation of 1a into 6 with
hydrogenation of 1a as the rate-determining step.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Finally, control experiments were performed to probe the
direct heterocyclization of
2 and 3a for 4a formation
(Scheme 2b). Treatment 2 with 3a in the absence of Co@NC900
leads to quantitative formation of 6 with negligible 4a yields
under N2 or air atmosphere at ambient temperature (Sche-
me 2b, paths a and b). In contrast, quantitative 4a was
observed when Co@NC900 was added into the above reaction
systems (Scheme 2b, paths c-e), thus indicating the promotion
effect of Co@NC900 for aromatization of 8 through oxidative
dehydrogenation process (Scheme 2a, path c).
Our above controlled experiments thus indicate that 5a!
4a transformation follows an initially successive hydrogenation
of 5a to 2 by Co@NC900 with 1a as the “intermediate”
(Scheme 2a, path a). A subsequent heterocyclization of 2 with
3a for 4a formation goes through a tandem reaction of
condensation of 2 with 3a, intramolecular cyclization of 6, and
a final dehydrogenation of 8 by aromatization (Scheme 2a,
paths c and d). Co@NC900-promoted hydrogenation is the rate-
determining step for transformation of 5a into 4a, whereas
both catalyst Co@NC900 and weak oxidants such as 1a and 5a
may accelerate the dehydrogenation process (Scheme 2a, paths
c and d).
Finally, the reusability of Co@NC900 demonstrates that the
catalyst was readily recycled for six times at least without loss
of its catalytic activity (Figure 5h). Notably, the recovered
Co@NC900 after the recycling shows almost the same morphol-
ogy with the fresh one, demonstrating the stability of Co@NC900
(Figure S3g,h).
To probe the effect of encapsulated Co NPs in Co@NC900 on
its catalytic performance, Co4 cluster encapsulated in a single-
walled N-doped carbon nanotube (Co4@N-CNT) is designed as a
model catalyst of Co@NC900 for DFT investigation. Although the
sizes of the Co4 cluster and N-CNT considered in the calculation
are much smaller and simpler than that of the experimentally
observed Co@NC900, this reduced geometry can capture the
Figure 4. Time-dependent GC-MS analysis of 4a formation. Reaction con-
ditions: Co@NC900 (10 mg), 1a (0.2 mmol), 3a (0.4 mmol), H2 (1.0 MPa), EtOAc
°
(4.0 mL), 80 C.
ChemSusChem 2020, 13, 1–13
www.chemsuschem.org
8
© 2020 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cssc.202002344
ChemSusChem
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Figure 5. a–c) Effects of initial Co@NC900 loading (a), H2 pressure (b), and 1a concentration (c) on the initial 1a hydrogenation rate. d) Kinetic plots between ln
(1-X) and reaction time for hydrogenation of 1a. e,f) Kinetic profiles between 1/(1-X) and reaction time for 2!3a condensation (e), and 1a!3a condensation
(f). g) Arrhenius profiles between observed rate constant kobs and reaction temperature for 1a hydrogenation, 2!3a condensation, and 1a!3a condensation.
h) Reusability of Co@NC900. Reaction conditions: a) 1a (0.04 mmol), Co@NC900 (5–10 mg), EtOAc (4.0 mL), H2 (1.0 MPa); b) 1a (0.04 mmol), Co@NC900 (5 mg),
EtOAc (4.0 mL), H2 (1.0–3.0 MPa); c) Co@NC900 (5 mg), H2 (1.0 MPa), EtOAc (4.0 mL); d) 1a (0.04 mmol), Co@NC900 (5 mg), EtOAc (4.0 mL), H2 (1.0 MPa); e) 3a
(0.04 mmol), 2 (0.04 mmol), EtOAc (4.0 mL); f) 3a (0.04 mmol), - 1a (0.04 mmol), EtOAc (4.0 mL); h) Co@NC900 (10 mg), 1a (0.20 mmol), 3a (0.40 mmol), EtOAc
(4.0 mL), H2 (1.0 MPa), RT, 20 h.
ChemSusChem 2020, 13, 1–13
www.chemsuschem.org
9
© 2020 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cssc.202002344
ChemSusChem
essential character of the electronic structure in Co@NC900
Additionally, a single-walled N-doped carbon nanotube (N-CNT)
is considered for comparison purpose.
.
1
2
3
Our control experiments confirmed a key step of Co@NC900
-
4
promoted hydrogenation of 1a for the transformation of 1a
into 4a (Scheme 2a). Therefore, the initial step of hydro-
genation of 1a to 2-nitrosoaniline (10) via 2-(dihydroxyamino)
aniline (9) as intermediate was further investigated over the
two model catalysts Co4@N-CNT and N-CNT for comparison
(Figure 6). Figure 6 shows the step-by-step reaction pathway in
term of energy diagram based on DFT calculations. The whole
hydrogenation process is assumed to proceed through dissoci-
ation of H2 molecule, adding of hydrogen atom to nitro group
of 1a, and water-removal.
We initially calculated the activation and dissociation of H2
molecule on the surface of the two catalyst models (Figure 6).
H2 is weakly adsorbed on the N-CNT with a negligible
adsorption energy of 0.01 eV [N-CNT··(H2), IM1]. The subsequent
HÀ H bond cleavage on N-CNT is kinetically unfeasible, corre-
sponding to a total endogenic reaction by 2.61 eV [N-CNT··(2H),
IM2] with an extremely high energy barrier of 3.54 eV (TS1). In
sharp contrast, Co4@N-CNT exhibits a significantly decreased
energy barrier to 2.70 eV (TS1) for the HÀ H bond dissociation
with a corresponding endogenic energy only 1.46 eV [Co4@N-
CNT··(2H), IM2]. These results can be further interpreted in view
of orbital overlapping. Figure 7 thus depicts the DOS of H 1s
and N 2p in Cat··(2H) state, the overlap between H 1s and N 2p
increases in Co4@N-CNT··(2H) if compared with N-CNT··(2H),
indicating a stronger NÀ H bond within Co4@N-CNT··(2H) (Fig-
ure 6, IM2) than in the case of N-CNT··(2H) (Figure 6, IM2).[70–72]
In addition, the increased DOS near Fermi level for Co4@N-
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Figure 7. Comparison of projected DOS for H 1s and its bonded N 2p
between Co4@N-CNT··(2H) and N-CNT··(2H). Dashed lines represent the Fermi
Level. Yellow and blue regions show charge increase and decrease,
respectively.
CNT··(2H) is beneficial to its subsequent hydrogenation reaction
(Figure 7).[37]
After H2 dissociation on the catalyst, adsorption of 1a on
the Cat··(2H) adduct is a prerequisite for subsequent hydro-
genation of the nitro group. The adsorption energies of 1a are
À 1.67 eV on Co4@N-CNT··(2H) and À 1.05 eV on N-CNT··(2H)
(IM3), respectively (Figure 6). Subsequent stepwise hydrogena-
tions of 1a in both adducts of Co4@N-CNT··(2H)··1a and N-
CNT··(2H)··1a are endothermic from IM3 to IM5 (Figure 6). Nitro
group (À NO2) in Cat··(2H)··1a is gradually hydrogenated to form
À N(OH)2, thus yielding Cat··9 adduct. Notably, a successive
Figure 6. The structures involved in hydrogenation of 1a to 10 over the two model catalysts. Energies are given in eV.
ChemSusChem 2020, 13, 1–13
www.chemsuschem.org
10
© 2020 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cssc.202002344
ChemSusChem
proton transfer of À N(OH)2 to form À N(O)OH2 is observed in N-
CNT··9 (from IM5 to TS2) before 9 dehydration (Figure 6 and
Figure S7). While, in the case of Co4@N-CNT··9, the -N(OH)2
group forms a hydrogen bond between OH in À N(OH)2 and N
atom on the carbon shell of Co4@N-CNT (from IM5 to TS2)
before 9 dehydration (Figure 6 and Figure S7).
Finally, the dehydration of Cat··9 adduct shows a lower
energy barrier of 0.83 eV for Co4@N-CNT than that of N-CNT
(1.46 eV) from TS2 to FS. The whole reaction from IS to FS is
exothermic by 2.29 eV for N-CNT and 2.58 eV for Co4@N-CNT
(Figure 6). The above DFT analyses thus suggest that Co4@N-
CNT can activate both H2 and nitro group of 1a more easily
Synthesis of benzimidazole 4a
1
2
3
4
5
6
7
8
9
The hydrogenative coupling was carried out in a Teflon-lined
stainless-steel autoclave (25 mL) with a magnetic stirrer. Co@NC900
(20 mg), 1a (0.2 mmol), 3a (0.4 mmol) and ethyl acetate (4 mL)
were successively added into the reactor. H2 was slowly purged into
the autoclave to replace the air inside the reactor for three times.
H2 (1.0 MPa) was then slowly loaded into the reactor. The reaction
was carried out at ambient temperature for 20 h. After the reaction,
the mixture in the reactor was then examined by Gas Chromatog-
raphy (GC).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
than N-CNT, which will evidently contribute a lot for the Acknowledgements
hydrogenation of 1a. Moreover, the rate-determining step for
transformation of 1a into 10 is H2 dissociation with energy
barrier of 2.70 eV on Co4@N-CNT and 3.54 eV on N-CNT catalyst,
respectively.
We acknowledge the financial support from National Natural
Science Foundation of China (U1810111, 22075104), Natural
Science Foundation of Guangdong Province, China
(2018B030311010), and Youth Science and Technology Innovation
Talent of Guangdong TeZhi Plan (2019TQ05L111).
Conclusion
In summary, Co@CN900/H2 was developed as a novel, green, Conflict of Interest
mild, and efficient catalytic system for a wide variety of
benzimidazole syntheses by hydrogenative coupling of 2-nitro-
aniline or 1,2-dinitrobenzene substrates with aldehydes. In
addition, robust and recyclable Co@CN900, which is magnetic in
nature, proved a promising catalyst for the current method-
ology for benzimidazole synthesis. Both the mechanism and the
kinetics of the hydrogenative coupling were systematically
demonstrated by experimental and theoretical investigations.
This research thus extends sustainable applications of the
encapsulated transition metal catalysts in thermally induced
organic transformations.
The authors declare no conflict of interest.
Keywords: cobalt · heterogeneous catalysis · hydrogenation ·
kinetics · reaction mechanisms.
[1] J. Deng, D. Deng, X. Bao, Adv. Mater. 2017, 29, 1606967.
[2] W. Xia, J. Tang, J. Li, S. Zhang, K. C. W. Wu, J. He, Y. Yamauchi, Angew.
Chem. Int. Ed. 2019, 58, 13354–13359; Angew. Chem. 2019, 131, 13488–
13493.
[3] J. Wang, G. Han, L. Wang, L. Du, G. Chen, Y. Gao, Y. Ma, C. Du, X. Cheng,
P. Zuo, G. Yin, Small 2018, 14, 1704282.
[4] H. Tang, S. Cai, S. Xie, Z. Wang, Y. Tong, M. Pan, X. Lu, Adv. Sci. 2016, 3,
1500265.
Experimental Section
[5] H. Tan, J. Tang, J. Henzie, Y. Li, X. Xu, T. Chen, Z. Wang, J. Wang, Y. Ide,
Y. Bando, Y. Yamauchi, ACS Nano 2018, 12, 5674–5683.
[6] Z. Liang, X. Fan, H. Lei, J. Qi, Y. Li, J. Gao, M. Huo, H. Yuan, W. Zhang, H.
Lin, H. Zheng, R. Cao, Angew. Chem. Int. Ed. 2018, 57, 13187–13191;
Angew. Chem. 2018, 130, 13371–13375.
[7] X. Cui, P. Ren, D. Deng, J. Deng, X. Bao, Energy Environ. Sci. 2016, 9, 123–
129.
[8] D. Lyu, Y. Du, S. Huang, B. Y. Mollamahale, X. Zhang, S. W. Hasan, F. Yu,
S. Wang, Z. Q. Tian, P. K. Shen, ACS Appl. Mater. Interfaces 2019, 11,
39809–39819.
Preparation of catalysts
Co@NC900: Co@NC900 was prepared from a mixture of P123, cobalt
acetate tetrahydrate [Co(CH3COO)2·4H2O] and melamine based on
literature method with slight modifications.[73] In brief, melamine
(2.25 g), P123 (1.5 g) and cobalt acetate tetrahydrate (1.0 g) were
dissolved in an ultrapure water (80 mL). The resulting mixture was
stirred at room temperature for 2 h, and further stirred for 0.5 h at
[9] J. Li, Y. Kang, D. Liu, Z. Lei, P. Liu, ACS Appl. Mater. Interfaces 2020, 12,
°
80 C. The solvent was then slowly removed by vacuum-rotary
5717–5729.
°
evaporation at 80 C. The obtained powder was transferred into an
[10] H. B. Yang, S. F. Hung, S. Liu, K. Yuan, S. Miao, L. Zhang, X. Huang, H. Y.
Wang, W. Cai, R. Chen, J. Gao, X. Yang, W. Chen, Y. Huang, H. M. Chen,
C. M. Li, T. Zhang, B. Liu, Nat. Energy 2018, 3, 140–147.
[11] H. Zhang, T. Wang, J. Wang, H. Liu, T. D. Dao, M. Li, G. Liu, X. Meng, K.
Chang, L. Shi, T. Nagao, J. Ye, Adv. Mater. 2016, 28, 3703–3710.
[12] J. Wu, R. M. Yadav, M. Liu, P. P. Sharma, C. S. Tiwary, L. Ma, X. Zou, X. D.
Zhou, B. I. Yakobson, J. Lou, P. M. Ajayan, ACS Nano 2015, 9, 5364–5371.
[13] Y. Hou, S. Cui, Z. Wen, X. Guo, X. Feng, J. Chen, Small 2015, 11, 5940–
5948.
[14] J. L. Fiorio, R. V. Goncalves, E. Teixeira-Neto, M. A. Ortuño, N. López, L. M.
Rossi, ACS Catal. 2018, 8, 3516–3524.
[15] W. Li, J. Artz, C. Broicher, K. Junge, H. Hartmann, A. Besmehn, R.
Palkovits, M. Beller, Catal. Sci. Technol. 2019, 9, 157–162.
[16] R. L. Oliveira, M. C. Ben Ghorbel, S. Praetz, D. Meiling, C. Schlesiger, R.
Schomäcker, A Thomas, ACS Sustainable Chem. Eng. 2020, 8, 11171–
11182.
°
°
alundum boat and heated at 180 C for 2 h, 240 C for 2 h and
À 1
°
°
900 C for 1 h, respectively, with a heating rate of 2 C min under
nitrogen atmosphere. The obtained black sample was further
leached in HCl (1.0 M) for 12 h at 80 C, thoroughly washed with
°
deioned water and dried under vacuum to give Co@NC900 (about
700 mg).
Co@NC700 and Co@NC800: Co@NC700 and Co@NC800 were prepared
following the above synthetic procedure with the final calcination
°
temperature of 700 and 800 C, respectively.
NC900: NC900 was synthesized with similar synthetic procedure of
Co@NC900 without addition of Co(CH3COO)2·4H2O.
ChemSusChem 2020, 13, 1–13
www.chemsuschem.org
11
© 2020 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cssc.202002344
ChemSusChem
[17] R. L. Oliveira, J. Kerstien, R. Schomäcker, A. Thomas, Catal. Sci. Technol.
2020, 10, 1385–1394.
[18] R. V. Jagadeesh, K. Natte, H. Junge, M. Beller, ACS Catal. 2015, 5, 1526–
[47] C. A. Wang, Y. F. Han, K. Nie, Y. W. Li, Mater. Chem. Front. 2019, 3, 1909–
1917.
[48] X. Sang, X. Hu, R. Tao, Y. Zhang, H. Zhu, D. Wang, ChemPlusChem 2020,
1
2
3
4
5
6
7
8
9
1529.
85, 123–129.
[19] L. Hu, R. Zhang, L. Wei, F. Zhang, Q. Chen, Nanoscale 2015, 7, 450–454.
[20] X. Li, C. Zeng, J. Jiang, L. Ai, J. Mater. Chem. 2016, 4, 7476–7482.
[21] H. Huang, X. Wang, Y. Sheng, C. Chen, X. Zou, X. Shang, X. Lu, RSC Adv.
2018, 8, 8898–8909.
[22] D. Xu, H. Zhao, Z. Dong, J. Ma, ChemCatChem 2019, 11, 5475–5486.
[23] T. Ni, S. Zhang, F. Cao, Y. Ma, Inorg. Chem. Front. 2017, 4, 2005–2011.
[24] L. Liu, F. Gao, P. Concepción, A. Corma, J. Catal. 2017, 350, 218–225.
[25] D. Formenti, C. Topf, K. Junge, F. Ragaini, M. Beller, Catal. Sci. Technol.
2016, 6, 4473–4477.
[49] A. Kumar, R. A. Maurya, P. Ahmad, J. Comb. Chem. 2009, 11, 198–201.
[50] K. Bahrami, M. M. Khodaei, F. Naali, J. Org. Chem. 2008, 73, 6835–6837.
[51] X. Diao, Y. Wang, Y. Jiang, D. Ma, J. Org. Chem. 2009, 74, 7974–7977.
[52] Y. Shiraishi, Y. Sugano, S. Tanaka, T. Hirai, Angew. Chem. Int. Ed. 2010,
49, 1656–1660; Angew. Chem. 2010, 122, 1700–1704.
[53] L. Hao, Y. Zhao, B. Yu, H. Zhang, H. Xu, Z. Liu, Green Chem. 2014, 16,
3039–3044.
[54] J. C. Lewis, A. M. Berman, R. G. Bergman, J. A. Ellman, J. Am. Chem.,
2008, 130, 2493–2500.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
[26] L. Liu, P. Concepcion, A. Corma, J. Catal. 2016, 340, 1–9.
[27] D. Formenti, F. Ferretti, C. Topf, A. E. Surkus, M. M. Pohl, J. Radnik, M.
Schneider, K. Junge, M. Beller, F. Ragaini, J. Catal. 2017, 351, 79–89.
[28] X. Sun, A. I. Olivos-Suarez, L. Oar-Arteta, E. Rozhko, D. Osadchii, A.
Bavykina, F. Kapteijn, J. Gascon, ChemCatChem 2017, 9, 1854–1862.
[29] R. Gao, L. Pan, Z. Li, X. Zhang, L. Wang, J. J. Zou, Chin. J. Catal. 2018, 39,
664–672.
[55] D. Yang, D. Fokas, J. Li, L. Yu, C. M. Baldino, Synthesis 2005, 2005, 47–56.
[56] P. L. Beaulieu, B. Hache, E. V. Moos, Synthesis 2003, 2003, 1683–1692.
[57] B. Krzyzinska, M. Glazek, A. Maczynska, Acta Agrobot. 2005, 58, 37–42.
[58] T. H. Thomas, Ann. Appl. Biol. 1974, 76, 237–241.
[59] F. G. Sánchez, A. N. Díaz, A. G. Pareja, J. Liq. Chromatogr. Relat. Technol.
1995, 18, 2543–2558.
[60] S. Liu, Z. Wang, S. Zhou, F. Yu, M. Yu, C. Y. Chiang, W. Zhou, J. Zhao, J.
Qiu, Adv. Mater. 2017, 29, 1700874.
[61] J. Kang, H. Zhang, X. Duan, H. Sun, X. Tan, S. Liu, S. Wang, Chem. Eng. J.
2019, 362, 251–261.
[30] D. Formenti, F. Ferretti, F. K. Scharnagl, M. Beller, Chem. Rev. 2018, 119,
2611–2680.
[31] J. Li, G. Liu, X. Long, G. Gao, J. Wu, F. Li, J. Catal. 2017, 355, 53–62.
[32] W. Gong, Y. Lin, C. Chen, M. Al-Mamun, H. S. Lu, G. Wang, H. Zhang, H.
Zhao, Adv. Mater. 2019, 31, 1808341.
[33] H. Su, K. X. Zhang, B. Zhang, H. H. Wang, Q. Y. Yu, X. H. Li, M. Antonietti,
J. S. Chen, J. Am. Chem. Soc. 2017, 139, 811–818.
[34] X. Lin, Z. Nie, L. Zhang, S. Mei, Y. Chen, B. Zhang, R. Zhu, Z. Liu, Green
Chem. 2017, 19, 2164–2173.
[35] C. Liang, X. Li, D. Su, Q. Ma, J. Mao, Z. Chen, Y. Wang, J. Yao, H. Li, Mol
Catal. 2018, 453, 121–131.
[36] T. Fu, M. Wang, W. Cai, Y. Cui, F. Gao, L. Peng, W. Chen, W. Ding, ACS
Catal. 2014, 4, 2536–2543.
[37] D. Deng, L. Yu, X. Chen, G. Wang, L. Jin, X. Pan, J. Deng, G. Sun, X. Bao,
Angew. Chem. Int. Ed. 2013, 52, 371–375; Angew. Chem. 2013, 125, 389–
393.
[38] W. Zhou, J. Zhou, Y. Zhou, J. Lu, K. Zhou, L. Yang, Z. Tang, L. Li, S. Chen,
Chem. Mater. 2015, 27, 2026–2032.
[39] A. A. R. El, H. Y. Aboul-Enein, Mini-Rev. Med. Chem., 2013, 13, 399–407.
[40] B. Narasimhan, D. Sharma, P. Kumar, Med. Chem. Res. 2012, 21, 269–283.
[41] P. S. Hameed, A. Raichurkar, P. Madhavapeddi, S. Menasinakai, S.
Sharma, P. Kaur, R. Nandishaiah, V. Panduga, J. Reddy, V. K. Sambanda-
murthy, D. Sriram, ACS Med. Chem. Lett. 2014, 5, 820–825.
[42] B. Fang, C. H. Zhou, X. C. Rao, Eur. J. Med. Chem. 2010, 45, 4388–4398.
[43] M. M. Sirim, V. S. Krishna, D. Sriram, O. U. Tan, Eur. J. Med. Chem. 2020,
188, 112010.
[62] Z. Chen, R. Wu, Y. Liu, Y. Ha, Y. Guo, D. Sun, M. Liu, F. Fang, Adv. Mater.
2018, 30, 1802011.
[63] Q. Lin, X. Bu, A. Kong, C. Mao, F. Bu, P. Feng, Adv. Mater. 2015, 27, 3431–
3436.
[64] Y. Liu, H. Jiang, Y. Zhu, X. Yang, C. Li, J. Mater. Chem. 2016, 4, 1694–
1701.
[65] Y. Zhang, P. Cao, H. Y. Zhang, G. Yin, J. Zhao, Catal. Commun. 2019, 129,
105747.
[66] A. Primo, F. Neatu, M. Florea, V. Parvulescu, H. Garcia, Nat. Commun.
2014, 5, 5291.
[67] D. S. Wang, Q. A. Chen, S. M. Lu, Y. G. Zhou, Chem. Rev. 2012, 112, 2557–
2590.
[68] X. Chen, Q. Shen, Z. Li, W. Wan, J. Chen, J. Zhang, ACS Appl. Mater.
Interfaces 2020, 12, 654–666.
[69] H. Namli, O. Turhan, Vib. Spectrosc. 2007, 43, 274–283.
[70] J. Zeng, W. Zhang, Y. Yang, D. Li, X. Yu, Q. Gao, ACS Appl. Mater.
Interfaces 2019, 11, 33074–33081.
[71] M. Xing, L. Guo, Z. Hao, Catal. Lett. 2018, 148, 2117–2126.
[72] P. Bechthold, M. E. Pronsato, C. Pistonesi, Appl. Surf. Sci. 2015, 347, 291–
298.
[73] W. Yang, L. Chen, X. Liu, J. Jia, S. Guo, Nanoscale 2017, 9, 1738–1744.
[44] S. N. A. Bukhari, G. Lauro, I. Jantan, C. F. Chee, M. W. Amjad, G. Bifulco,
H. Sher, I. Abdullah, N. A. Rahman, Future Med. Chem. 2016, 8, 1953–
1967.
[45] L. M. Dudd, E. Venardou, E. Garcia-Verdugo, P. Licence, A. J. Blake, C.
Wilson, M. Poliakoff, Green Chem. 2003, 5, 187–192.
[46] J. C. Lewis, J. Y. Wu, R. G. Bergman, J. A. Ellman, Angew. Chem. Int. Ed.
2006, 45, 1589–1591; Angew. Chem. 2006, 118, 1619–1621.
Manuscript received: October 4, 2020
Revised manuscript received: November 1, 2020
Version of record online: ■■■, ■■■■
ChemSusChem 2020, 13, 1–13
www.chemsuschem.org
12
© 2020 Wiley-VCH GmbH
��
These are not the final page numbers!

FULL PAPERS
1
2
3
4
5
6
7
8
9
Seek and you confined: Co nanopar-
ticles encapsulated in N-doped
carbon nanotubes can efficiently
catalyze hydrogenative coupling for
the synthesis of benzimidazoles. The
robust catalyst can be easily
recovered by an external magnetic
field and readily recycled. Potential
industrial application of the earth-
abundant metal catalysts is thus
demonstrated for green synthesis of
heterocyclic compounds.
C. Lin, W. Wan, X. Wei, Prof. J. Chen*
1 – 13
H2 Activation with Co Nanoparticles
Encapsulated in N-Doped Carbon
Nanotubes for Green Synthesis of
Benzimidazoles
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57