Reductive Formylation of Nitroarenes using HCOOH over Bimetallic C?N Framework Derived from the Integration of MOF and COF

Article abstract of DOI:10.1002/cctc.202100412

CoZn embedded C?N framework is prepared by the carbonization of CoZn containing MOF integrated with COF porous architecture in Ar atmosphere. The graphitic nature of porous carbon is confirmed from Raman analysis. The porosity and nanostructure information are retrieved from N₂-sorption and transmission electron microscopic analysis, respectively. The incorporation of different metals and their oxidation states and types of nitrogen present in the C?N framework are confirmed from X-ray photoelectron spectroscopy. The basicity of the materials is determined from a CO₂-temperature programmed desorption. ZnCo embedded C?N framework exhibits excellent activity in the selective reductive formylation using HCOOH. For comparison, more than 15 materials are prepared, and their activities are compared. Several control experiments are performed to establish a structure-activity relation. The recycling experiment, hot-filtration test, and poisoning experiment demonstrate the metal embedded porous C?N framework‘s recyclability and stability. A reaction mechanism for the reductive N-formylation of nitroaromatics is presented based on structure-activity relationship, control reactions, and physicochemical characterizations. The development of interesting MOF-COF-derived metal nanoclusters embedded C?N framework for selective reductive formylation of nitroaromatics using formic acid will be highly attractive to catalysis researchers and industrialists.

Full text of DOI:10.1002/cctc.202100412

The European Society Journal for Catalysis
Supported by
Accepted Article
Title: Reductive Formylation of Nitroarenes using HCOOH over
Bimetallic C-N Framework Derived from the Integration of MOF
and COF
Authors: Rajendra Srivastava and Ashish Kumar Kar
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.202100412
Link to VoR: https://doi.org/10.1002/cctc.202100412
01/2020

10.1002/cctc.202100412
ChemCatChem
FULL PAPER
Reductive Formylation of Nitroarenes using HCOOH over
Bimetallic C-N Framework Derived from the Integration of MOF
and COF
Ashish Kumar Kar[a] and Rajendra Srivastava*[a]
[a]
Mr. Ashish Kumar Kar, Dr. Rajendra Srivastava
Catalysis Research Laboratory, Department of Chemistry,
Indian Institute of Technology Ropar,
Rupnagar-140001, Punjab, India.
E-mail: rajendra@iitrpr.ac.in
Supporting information for this article is given via a link at the end of the document. ((Please delete this text if not appropriate))
Abstract: CoZn embedded C-N framework is prepared by the
carbonization of CoZn containing MOF integrated with COF porous
architecture in Ar atmosphere. The graphitic nature of porous carbon
is confirmed from Raman analysis. The porosity and nanostructure
information are retrieved from N2-sorption and transmission electron
microscopic analysis, respectively. The incorporation of different
metals & their oxidation states and types of nitrogen present in the
C-N framework are confirmed from X-ray photoelectron
spectroscopy. The basicity of the materials is determined from a
CO2-temperature programmed desorption. ZnCo embedded C-N
framework exhibits excellent activity in the selective reductive
formylation using HCOOH. For comparison, more than 15 materials
are prepared, and their activities are compared. Several control
experiments are performed to establish a structure-activity relation.
The recycling experiment, hot-filtration test, and poisoning
experiment demonstrate the metal embedded porous C-N
framework's recyclability and stability. A reaction mechanism for the
reductive N-formylation of nitroaromatics is presented based on
structure-activity relationship, control reactions, and physicochemical
characterizations. The development of interesting MOF-COF-derived
metal nanoclusters embedded C-N framework for selective reductive
formylation of nitroaromatics using formic acid will be highly
attractive to catalysis researchers and industrialists.
be implemented across the world.[9-10] The significant bottlenecks
of this development are expensive experimental setup and low
efficiency. Therefore, it is beneficial to carry out chemical
synthesis without the direct use of hydrogen gas. Moreover, the
use of H2 gas requires a particular leak-proof reaction setup.
Thus, transfer hydrogenation is employed, which is a more user-
friendly, facile, economical, and sustainable method. In general,
transfer hydrogenation uses alcohol as a transfer hydrogen
source.[11-12] However, these alcohols (such as methanol) are
also produced using H2 as
a
reactant.13 Various other
hydrogenation sources such as hydrazine hydrate and sodium
borohydride are non-economical and non-eco-friendly with poor
reversibility of hydrogen.[14-15] Thus, it is essential to develop a
catalytic transfer hydrogenation process involving renewable
hydrogenation sources such as formic acid.[16-17] Formic acid is
produced by biomass.[16,18] Moreover, formic acid (FA) can also
be used as a formylating source.[17,19] Herein, FA is employed as
an H2 and formylating source for the reductive formylation of
nitrobenzene (NB) in a direct one-step process.
Aryl amines and N-Aryl formamides are important
intermediates and have broad application in polymers,
pharmaceuticals, agrochemicals, dyes, and fragrances.[20-22]
Moreover, their inherent property for being a potential candidate
to synthesize various biologically active scaffolds cannot be
ruled out. Typical routes for the synthesis of N-aryl formamides
include direct formylation of aniline and nitrobenzene's reductive
formylation.[23-24] The former is more facile and involves a
straightforward synthetic route for the synthesis of N-aryl
formamides. However, the later synthetic method involves a
more challenging pathway when it is carried out in one-pot. In
this work, the challenging reductive N-formylation of
nitrobenzene to N-aryl formamides is demonstrated via a
sustainable catalytic approach using formic acid as a hydrogen
and formylating source.
The conversion of formic acid to H2 and/or CO requires a
suitable metal center.[25-27] The metal should be stabilized in a
porous matrix to not leach out during the reaction. Further, the
adsorption of formic acid requires basic support.[17] In this
reaction, one should have a suitable N doping along with a high
surface area for the concomitant adsorption of formic acid and
reactant. Thus, a high surface area porous N-doped carbon
material is a key feature. In general, N-doped carbon materials
are built with pyrrolic, pyridinic, and graphitic N-units.[17,28] Higher
N contents are not always dominant factors for catalysis, but the
N environment is an important factor responsible for providing
optimum metal stability and reactivity.[28] Metal or metal oxide-
containing porous carbon or N-doped carbon materials can be
prepared using a metal-organic framework (MOF).[29-31] It is
Introduction
The current energy demands are satisfied through non-
renewable resources threatening to maintain the ecological
system suitable for human survival. Thus, it has become
imperative to develop an environment-benign energy conversion
process to meet global energy demand.[1-2] Hydrogen is one of
the most abundant elements on the earth. However, it is not
found in the troposphere in large concentrations in the gaseous
form. It is present in the combined state in various renewable
and non-renewable resources. Hydrogen is a key precursor for
the production of a large number of bulk and commodities
chemicals.[3-5] Moreover, hydrogen-based sustainable energy
conversion processes such as fuel cells can serve the energy
demands to a greater extent.[6] No doubt, hydrogen is a clean
energy source and one of the best options for zero-emission
energy carriers. However, in conventional practice, H2 is mainly
produced by the steam reforming process, by which a large
excess of CO2 (almost three times that of H2) is emitted, which is
a real threat to global warming.[7-8] The benign environmental
process for H2 production is photocatalytic water splitting or
water electrolysis that is under the growing stage and is yet to
1
This article is protected by copyright. All rights reserved.

10.1002/cctc.202100412
ChemCatChem
FULL PAPER
possible to synthesize metal-containing porous N doped carbon
materials by directly reacting carbon and nitrogen-bearing
sources and metal precursors.[17] However, MOF-derived metal-
containing N-doped porous carbon is more attractive and active
due to the pre-existing bonding of metal with MOF framework
that provides high stability, highly dispersed and small size metal
clusters, suitable for excellent catalysis.[29,30] Covalent organic
framework provides an ordered N-doped carbon framework with
better textural properties than other N-containing carbon sources
providing C-N framework.[32] It is possible to enhance the N-
doping with a suitable N environment and textural properties by
integrating MOF with other N-containing porous materials. Only
a handful of work in this direction has been done.[33-35] Thus, it is
interesting to develop metal nanoclusters containing N-doped
carbon materials by integrating MOF and other C-N frameworks.
This study demonstrates the catalytic reductive formylation of
nitrobenzene (NB) in a direct one-step process with formic acid
using the ZnCo-N doped porous carbon-nitrogen framework.
ZnCo co-doped porous carbon-nitrogen framework is prepared
by carbonizing Co-Zn zeolitic imidazole framework (ZIF) grown
over the porous covalent organic framework. Several mono-
metallic and bi-metallic C-N frameworks are also prepared and
investigated in this reaction. A suitable composition of Co and
Zn in the C-N framework is required to achieve high activity and
selectivity for the reductive formylated product. The development
of a tunable metal composition with a particular C-N framework,
exhibiting excellent activity for the challenging one-pot reductive
formylation of nitrobenzene, will be highly attractive for catalysis
and materials synthesis prospective.
species. Two more bimetallic catalysts were prepared to
understand the structural-activity relationship in the reductive N-
formylation of nitrobenzene. ZnCo-ZIF precursors were added to
melamine to prepare ZnCo-ZIF integrated with melamine. The
resultant material was carbonized to obtain ZnCo-N@C(ZIF-MEL)
catalyst. ZnCo-ZIF was carbonized to obtain ZnCo-N@C(ZIF). Zn-
ZIF and Co-ZIF were carbonized to obtain Zn-N@C(ZIF) and Co-
N@C(ZIF), respectively.
Physicochemical Characterization
The structure of COF was confirmed by FT-IR and Solid-state
13CP MAS NMR. The FT-IR spectra of terepthaldehyde,
melamine, and COF are plotted (Figure S1). Melamine exhibits
characteristics sharp peaks at 3468 cm-1, and 3419 cm-1
correspondings to NH2-stretching vibrations and peak at 1649
cm-1 correspond to NH2-deformation mode. Terepthaldehyde
exhibits characteristics peak at 2865 cm-1 and 1687 cm-1
corresponding to aldehydic C-H stretching and carbonyl group,
respectively. The above-mentioned FT-IR peaks are absent in
the COF material suggesting the complete condensation of –
NH2 of melamine and -CHO of terepthaldehyde. COF shows
triazine units' presence due to FT-IR peaks at 1543 cm-1 and
1479 cm-1 corresponding to C=N, which confirm the presence of
melamine units in COF structure. No peak at 1600 cm-1 is
observed, suggesting that the COF framework is built with
aminal moiety (HN-C-NH).[36] 13CP MAS NMR also confirmed the
formation of the COF framework (Figure S2). A peak at 166 ppm
corresponds to triazine units, peaks at 135 ppm and 129 ppm
correspond to aromatic carbons, and 54 ppm corresponds to
aminal carbon.[36] No peaks at 190 ppm and 160 ppm confirm
the absence of the free aldehyde group of terepthaldehyde and
imine bonds, respectively. Field-emission scanning electron
microscopy (FE-SEM) analysis shows that COF exhibits
aggregated morphology, whereas ZIF exhibits a rhomboid
dodecahedron morphology (Figure S3). The composite material
shows that the ZnCo-ZIF is uniformly dispersed with the COF
matrix.
All the synthesized C-N catalysts exhibit a broad XRD
peak in the range of 15-30° (Figure 1A). The diffraction maxima
of the broad reflection appear close to 25, which are marginally
deviated from the C (001) reflection known for a carbon
material.[37] This suggests that the synthesized catalysts are
graphitic.[37] Diffraction peaks at 43.3° and 51.3° are observed
for all the ZnCo-N@C catalysts indicating the presence of
metallic Co species.17 A weak diffraction peak around 36.5° is
observed for the ZnCo-N@C(ZIF), suggesting the presence of
Scheme 1. Schematic presentation for the synthesis of Zn-Co doped porous
carbon-nitrogen framework.
CoO in the ZnCo-N@C(ZIF)catalyst. Moreover,
a weaker
Results and Discussion
diffraction peak around 36.5° is observed in ZnCo-N@C(ZIF-MEL)
than ZnCo-N@C(ZIF) (Figure 1A). An exceptionally weak
diffraction peak could be observed around 36.5° in ZnCo-
N@C(ZIF-COF). Based on this investigation, it can be concluded
that ZIF-COF integrated carbonized framework provides stability
to Co and Zn species in a highly dispersed state and prevent
them from oxidation leading to produce metal oxide phase. The
XRD patterns of monometallic MOF, bi-metallic MOF, COF, and
other materials prepared in this study are provided in the
supporting information section (Figure S4).
The schematic presentation for the synthesis of the Zn-Co co-
doped porous carbon-nitrogen framework is presented in
Scheme 1. First, a covalent organic framework (COF) was
prepared by condensation of melamine and terepthaldehyde.
Then ZnCo-ZIF precursors (Zn & Co precursors and 2-methyl
imidazole) were added to the as-synthesized COF to prepare
ZnCo-ZIF integrated with the COF framework. The resultant
material was carbonized under a specific condition described in
the experimental section to obtain ZnCo-N@C(ZIF-COF) catalyst.
During the carbonization process, the ZIF framework was
integrated with the COF framework and resulted in the formation
of an N-doped porous carbon framework embedding metal
Raman analysis provides information about the nature of
carbon present in the catalyst. The Peaks at 1358 cm-1 and 1589
cm-1 in the Raman spectrum of various ZnCo-N@C catalysts
2
This article is protected by copyright. All rights reserved.

10.1002/cctc.202100412
ChemCatChem
FULL PAPER
correspond to the D-band and G-band of graphitic carbon
(Figure 1B).[31] The ratio of ID/IG defines the surface defect
degree of the carbon framework. The ID/IG values obtained for
the ZnCo-N@C(ZIF-MEL), ZnCo-N@C(ZIF-COF), and ZnCo-N@C(ZIF)
are 1.05, 1.01, and 0.99, respectively. Vibration peaks observed
in the range of 300-500 cm-1 correspond to the Zn/Co-O/N
species present in the ZnCo-N@C(ZIF-COF) catalyst (Figure 1B).[17]
The graphitic nature of the various materials prepared in this
study suggests that they are conducting and facilitating the
electron transfer between the reactant and catalyst surface,
promoting the decomposition of HCOOH to CO/H2.
ZnCo-N@C(ZIF)
ZnCo-N@C(ZIF-COF)
ZnCo-N@C(ZIF-MEL)
601.3
684.6
549.5
2.28
4.36
3.60
0.91
2.1
391.8
409.7
349.1
1.78
TEM and high-resolution TEM (HRTETM) images show
aggregated porous carbon nanostructure present in ZnCo-
N@C(ZIF-COF) (Figure 1D-F). HRTEM image shows regions with
lattice fringes corresponding to graphitic carbon present in the
material, consistent with the Raman analysis. Cobalt or Zn
nanoparticles are not visible in the TEM images suggesting that
the metal nanoclusters are present in a highly dispersed state in
the material. EDAX analysis and elemental mapping indicate the
uniform distribution of Co and Zn metals in a porous C-N
framework (Figure S5). The low concentration of Zn compared
to Co in the elemental mapping confirms the Zn’s evaporation
during the carbonization step.
Co
D
G
ZnCo-N@C(ZIF-COF)
ZnCo-N@C(ZIF-MEL)
CoOx
Co/Zn-Nx/Ox
ZnCo-N@C(ZIF-COF)
A)
B)
ZnCo-N@C(ZIF-MEL)
ZnCo-N@C(ZIF)
ZnCo-N@C(ZIF)
10
20
40
50
60
70
500
1000
2000
2500
)
3000
230 (degree)
Rama1n50s0hift (cm-1
Co 2
Co 2pp33/2/2
N 1sPyrrolic N
Pyridinic N
1000
+2
Co+O3 x
Co
CoNx
Co
0.08
ZnCo-N@C
ZnCO-N@C(ZIF-COF)
ZnCO-N@C
ZnCo-N@C(ZIF
ZnCO-N@C
ZnCo-N@C
)
(ZIF-COF
C)
D)
(Z IF-M-M
E
E
L
L
)
)
0.06
0.04
0.02
0.00
(A)
800
600
400
200
0
(A)
(ZIF)
(ZIF)
Shakeup
Graphitic N
Co0
5
10
15
20
25
30
Pore Width (nm)
ZnCo-N@C(ZIF-COF)
ZnCo-N@C(ZIF-MEL)
ZnCo-N@C(ZIF)
100 nm
0.2
0.4
0.6
0.8
1.0
Relative pressure (P/P
)
0
(B)
(C)
(B)
(C)
E)
F)
50 nm
10 nm
Figure 1. (A) Powder XRD patterns, (B) Raman spectra, (C) N2-sorption
isotherms of various ZnCo-N@C catalyst. (D-F) TEM images of ZnCo-
406 404 402 400 398 396 394
792 788 784 780 776 772
N@C(ZIF-COF)
.
Binding energy (eV)
Binding energy (eV)
The N2-sorption analysis was carried out to determine the
textural properties of the ZnCo-N@C catalysts. All three
catalysts exhibit a type-IV isotherm and H3 hysteresis loop
(Figure 1C). A steady increase in the N2-adsorption volume is
observed in the range of 0 to 0.8 (P/P0), followed by a steep
increase in the N2-adsorption volume representing interparticle
porosity. The inset of Figure 1C shows the pore size distribution
obtained from the BJH method. The surface area of the
materials follows the order ZnCo-N@C(ZIF-COF) > ZnCo-N@C(ZIF)
Figure 2. The high resolution XPS spectra of N 1s and Co 2p present in (A)
ZnCo-N@C(ZIF-COF), (B)ZnCo-N@C(ZIF-MEL), and (C) ZnCo-N@C(ZIF)
.
The surface elemental compositions and their oxidation
states were evaluated from XPS analysis. The surface survey
analysis confirms Zn, Co, N, O, and C in the synthesized
catalyst (Figure S6). The details of the surface elemental
compositions are presented in Table S1. The Zn contents in the
materials are smaller than the input value. This is due to the
>
ZnCo-N@C(ZIF-MEL), indicating that MOF-COF entangled
material provides better surface area (Table 1).
volatile nature of Zn, which escapes from the materials during
37]
the carbonization process.[35,
The surface Zn has more
Table 1. Textural properties of various ZnCo-N@C catalysts.
escaping tendency than the Zn incorporated in the C-N
framework. The high-resolution XPS spectra of Zn 2p show two
deconvoluted peaks at 1044.7 eV and 1021.5 eV corresponding
to the Zn 2p1/2 and Zn 2p3/2 of Zn-Nx species (Figure S7).[38] The
Catalyst
SBET
(m2/g)
Pore
diameter
(nm)
Pore
Sext
(m2/g)
volume
(cm3/g)
3
This article is protected by copyright. All rights reserved.

10.1002/cctc.202100412
ChemCatChem
FULL PAPER
high-resolution XPS spectra of the Co 2p3/2 exhibits four peaks
at the binding energy of 778.2 eV for Co(0), 779.4 eV for CoOx,
781.6 eV for CoNx, and a shakeup peak at 786.1 eV (Figure
2).[17] A low area deconvoluted profile with a peak maximum at
778.2 eV indicates that the low concentration of metallic Co(0)
present in ZnCo-N@C(ZIF-COF) and ZnCo-N@C(ZIF-MEL) catalysts.
Co nanoclusters are bonded and stabilized through various
types of “N” species present in the material. The coordination
between “N” species and Co results in a partial electron transfer
from Co to N atoms, which create a positive charge on Co and
form Co-Nx species occurring at a higher binding energy of
781.6.[39-41] Furthermore, the surface Co nanoclusters are highly
entries 6-7, Table 2). However, the selectivity for formanilide (1b,
compare entries 6-7, Table 2) is more in Zn-N@C(ZIF-8). Further,
upon integrating MOF and COF, the resultant monometallic
carbonized catalysts exhibit better NB conversion (Compare
entries 6,7 with 9,10, Table 2).
Table 2. Reductive formylation of nitrobenzene using HCOOH over various
catalysts prepared in this study.
O
NO2
NH2
+
NH
Catalyst
HCOOH
(1A)
(1B)
sensitive to the aerobic atmosphere and form
a
low
concentration of CoOx phase, which is observed at 779.4 eV in
the deconvolution spectrum.[42] The deconvoluted N 1s XPS
spectra suggest the occurrence of pyridinic N, pyrrolic N, and
graphitic N at 398.2 eV, 399.5 eV, and 401.2 eV, respectively
(Figure 2).[17] An additional peak appears at 404.1 eV in ZnCo-
N@C(ZIF-MEL) suggests the presence of pyridine N-oxide.[17] The
presence of different nitrogen environments in all three catalysts
suggests the successful entrapment of metallic (Co and Zn)
constituents in the materials. Moreover, the XPS spectrum of
C1s indicates the presence of C=C in all three catalysts (Figure
S8). The carbonized structure of ZIF-COF contains a suitable N
environment (pyrrolic N, pyridine N) for stabilizing different Co
species. However, some Co species (especially surface Co
species) could not be prevented from oxidation in aerobic
conditions. Hence the occurrence of CoOx can be observed in
the XRD and XPS analysis.
Product
Selectivity (%)
NB
Conv.
(%)
Entry
Catalyst
1A
1B
1
2
None
0
ND
ND
ZIF-67
3
48
29
20
79
11
55
20
9
52
71
80
21
89
45
80
91
47
86
89
91
90
76
91
62
94
86
85
-
3
ZIF-8
2
4
ZnCo-ZIF
4
5
COF
1
6
aZn-N@C(ZIF-8)
26
52
71
34
69
98
96
94
71
41
69
84
89
79
65
30
7
aCo-N@C(ZIF-67)
Evaluation of catalytic activity
8
aZnCo-N@C(ZIF)
The one-pot direct N-formylation of the -NO2 of NB to the
corresponding N-formyl aniline (AN) using formic acid as the
reducing and formylating source is presented here. The
activation of formic acid takes place via its adsorption on the
suitable metal sites. The basic nature of C-N framework is ideal
for formic acid adsorption.[17] Besides, the high surface area and
porosity are essential for the adsorption of reactant molecules
and entrapment of metal species in a highly dispersed state. Zn
is known to facilitate the adsorption and decomposition of formic
9
aZn-N@C(ZIF-8-COF)
aCo-N@C(ZIF-67-COF)
aZnCo-N@C(ZIF-COF)
aZn1.5Co0.5-N@C(ZIF-COF)
aZn1.75Co0.25-N@C(ZIF-COF)
aZn1.9Co0.1-N@C(ZIF-COF)
Zn1.75Co0.25-N@C(ZIF-COF)973K
Zn1.75Co0.25-N@C(ZIF-COF)1023K
Zn1.75Co0.25-N@C(ZIF-COF)1073K
Zn1.75Co0.25-N@C(ZIF-COF)1173K
aZn1.75Co0.25-N@C(ZIF-MEL)
aZn1.75Co0.25-N@C(ZIF)
a,bZn1.75Co0.25-N@C(ZIF-COF)
10
11
12
13
14
15
16
17
18
19
20
21
53
14
10
9
acid.[43, 44] In contrast, the Co is known for forming H2 + CO2
[26,45-46]
10
24
9
or/and CO + H2O from the formic acid.
Thus, the
bimetallic constituent of the catalyst is advantageous for utilizing
the formic acid in the one-pot reductive formylation of NB. The
reductive formylation depends on the composition of H2 + CO2
and CO
+ H2O, whose generation depends on several
experimental factors, including the catalyst’s chemical
compositions. Thus, it is highly interesting to develop a catalyst
to generate H2 and CO in a suitable ratio to form the desired N-
formylated product. Herein, various ZnCo-N@C catalysts are
employed for the reductive formylation of the NB and its
derivatives to its corresponding N-formyl AN derivatives. NB is
chosen as a model substrate to find the best catalyst and
optimized reaction conditions. No NB conversion is observed in
the absence of a catalyst (Entry 1, Table 2). There is no reduced
or formylated product obtained when the reaction is carried out
in the absence of the formic acid but in the presence of the
catalyst. Very low NB conversion is observed using ZIF-67, ZIF-
8, ZnCo-ZIF, and COF catalysts (Entries 2-5, Table 2). All
carbonized catalysts are active for this reaction. Co-N@C(ZIF-67)
exhibits higher NB conversion than Zn-N@C(ZIF-8) (Compare
38
6
14
15
100
Reaction condition: Nitrobenzene (1 mmol), HCOOH (12 equiv.), catalyst (30
mg), THF (7 mL), temperature (413 K), time (8 h). aCarbonized at 1123 K, bH2
pressure = 5 bar.
4
This article is protected by copyright. All rights reserved.

10.1002/cctc.202100412
ChemCatChem
FULL PAPER
However, the formanilide selectivity is higher in the case of Zn
catalyst (Compare entries 6,7 with 9,10, Table 2). Thus, the Zn
and Co ratio's influence is varied to obtain higher NB conversion
and formanilide selectivity. Since Zn metal drives the selectivity,
therefore higher input Zn content is selected. The input ratio of
reaction temperature, the formanilide selectivity increases. This
suggests that in water, up to 413 K, decarboxylation leading to
H2 + CO2 is favoured over dehydration leading to H2O + CO.
However, above 433 K, dehydration is facilitated, and more
selectivity for formanilide is obtained (Figure S11). Thus, this
result clearly shows that extent of decarboxylation and
dehydration depends on temperature and solvent.
Zn:Co (in wt%) is varied, and the material with Zn1.75Co0.25
-
N@C(ZIF-COF) exhibits the best selectivity towards formanilide
(compare entries 11-14, Table 2). With other compositions,
either lower selectivity of the formanilide or the lower NB is
observed. The influence of carbonization temperature on the
catalytic activity is investigated. Better NB conversion and
formanilide selectivity are obtained for the material treated as
1123 K (compare entries 13, 15-18, Table 2). Zn1.75Co0.25
-
N@C(ZIF-MEL) and Zn1.75Co0.25-N@C(ZIF) exhibit inferior activity
than Zn1.75Co0.25-N@C(ZIF-COF), carbonized at 1123 K (compare
entries 13, 19, and 20, Table 2). When the hydrogenation of the
NB is carried out in the presence of the H2 gas, the aniline yield
is only 30 % (Entry 20, Table 2). This suggests the in-situ
generated CO+H2 from the formic acid is responsible for this NB
reductive formylation reaction. After finding the best catalyst
Zn1.75Co0.25-N@C(ZIF-COF) (hereafter denoted as ZnCo-N@C(ZIF-
COF)), factors influencing the product yields are investigated. The
amount of catalyst is varied in the range of 20-40 mg. The best
activity is achieved using 30 mg of catalyst for 1 mmol of
nitrobenzene (Figure 3A). The reaction duration also plays an
important role because more aniline is formed at the beginning
of the reaction, subsequently reaction with CO to form the
formanilide (Figure 3B). After 8 h of the reaction, the best activity
is achieved. Another important parameter is the temperature
because the mode of HCOOH decomposition extensively
depends on the reaction temperature, which subsequently
affects the formanilide selectivity. The HCOOH at elevated
temperature (>393 K) preferentially decomposes under the
dehydration (HCOOH CO+H2O) pathway. Hence the chance of
forming the formylated product via an indirect pathway is more
(Discussed later in the mechanism section).[45,48] Thus, the
reactions are performed at four different reaction temperatures
in the range of 393 K to 423 K (Figure 3C). With an increase in
the temperature, formanilide selectivity increases. The best
activity and formanilide selectivity are obtained at 413 K. An
almost similar result is obtained when the reaction temperature
is raised to 423 K. Careful inspection of data reveal that above
393 K, the rate of CO formation is increased when compared to
H2 formation. This is the reason; higher formanilide selectivity is
obtained at 413 K. The activation energy obtained for this
reaction is calculated to be 25.6 kJ/mole (The details can be
found in the Supporting information section, Figure S9). The NB
conversion and the formanilide product are directly proportional
to the concentration of HCOOH (Figure 3D). The influence of
HCOOH:NB molar ratio is investigated. When the HCOOH
concentration varies from 6 equivalents to 14 equivalents, the
NB conversion increases linearly. Using 12 equivalents of
HCOOH with respect to NB, the best result is obtained. Thus 1
mmol of NB is reacted with 12 mmol of HCOOH at 413 K for 8 h
using 30 mg of ZnCo-N@C(ZIF-COF) to obtain the best activity. The
influence of various solvents suggests that the reaction
proceeds well in THF and toluene (Figure S10). Higher
formanilide selectivity is obtained in toluene, which is consistent
with a previous report, which suggests that higher CO is
produced in toluene.[17,45] Reaction proceeds even good in H2O
medium. However, under the optimized reaction condition, low
formanilide selectivity is observed. With an increase in the
Figure 3. Influence of (A) catalyst amount, (B) reaction time, (C) reaction
temperature and (D) formic acid concentration, in the reductive formylation of
nitrobenzene over ZnCo-N@C(ZIF-COF)
.
The reaction proceeds very well using substituted nitroarenes.
A good to excellent conversion and N-formylated product
selectivity of the substituted nitroarenes is observed (Table 3). p-
amino nitro benzene leads to producing di N-formylated
compound as a major product. No dehalogenation or de-
methoxylation, or dehydroxylation is observed, suggesting that
selective reduction and reductive formylation take place in the
presence of several functional groups. This suggests the present
catalyst tolerates a wide range of functional groups and
indicates the general applicability of this catalyst in reductive N-
formylation of nitro compounds.
Table 3. Reductive N-formylation of nitroaromatics using HCOOH over ZnCo-
N@C(ZIF-COF)
.
Selectivity (%)
1a
Conv.
(%)
Entry
Reactant
1b
1
92
(9)
(91)
2
89
(88)
(12)
5
This article is protected by copyright. All rights reserved.

10.1002/cctc.202100412
ChemCatChem
FULL PAPER
TEM. The catalytic poisoning test confirms that the catalytic
activity is mainly due to the presence of Co-Nx and Co
nanoclusters. One more control experiment was performed to
overrule the possibility of CO poisoning. First, the reaction was
carried out in the presence of 1 bar CO pressure, and then the
recovered catalyst was subjected to a fresh catalytic run.
Reaction conducted using recovered catalyst exhibits similar
activity (NB conversion = 91 % and formanilide selectivity =
93 %) to that of the original catalyst suggesting that CO
poisoning did not occur.
3
90
(16)
(84)
4
93
The nitrogen-rich polymer framework present in the COF
template empowers the ZnCo-ZIF framework to architect a well-
ordered porous architecture with high surface area and
bimetallic constituent’s embedded in the nitrogen-containing
graphitic carbon framework upon the carbonization process. The
carbonization strategy is adopted here to prepare metal
nanoclusters embedded porous N-doped graphitic carbon
catalyst. Using conventional carbon/nitrogen precursors, during
the carbonization process, the framework shrinkage and metal
agglomeration are commonly observed. Hence the resulted
graphitic carbon-nitride lacks porosity and exhibits low surface
area. Hard template such as silica nanoparticles or mesoporous
silica has been used as a sacrificial template to overcome this
problem. However, the removal of sacrificial silica template is
cumbersome and hazardous process due to the involvement of
a strong base or HF.[17] Herein, COF is used as a soft-template
and C-N source that entangled with ZnCo-ZIF to produce a
porous framework, which upon carbonizing produces porous C-
N framework with entrapment of Co and Zn nanoclusters. Such
advantageous features are not obtained when melamine is used
as C, N source. Thus, the ZnCo-N@C(ZIF-COF) superior catalytic
activity is due to the large numbers of dispersed active sites (Zn-
Nx/Co-Nx) throughout the porous graphitic C-N framework. The
formic acid adsorption also depends on the basicity of the
catalyst. To determine the basicity of the catalyst, CO2-
temperature-programmed desorption (CO2-TPD) analysis was
conducted (Figure S15). The amount of desorbed CO2 depends
on the basicity and surface area of the material. The CO2
desorbed amount follows the order ZnCo-N@C(ZIF-COF) > ZnCo-
N@C(ZIF-MEL) > ZnCo-N@C(ZIF). CO2-TPD profiles show that CO2
desorbs at three different temperatures. The physisorbed CO2
desorption occurs below 373 K, whereas the desorption in the
range of 373 K - 453 K and 623 K - 723 K attributes weak and
strong basic sites, respectively. The strong basic strength
(89)
(11)
5
84
(79)
(72)
(21)
(28)
6
78
Reaction condition: Nitroarene (1 mmol), HCOOH (12 equiv.), Catalyst (30
mg), THF (7 mL), temperature (413 K), time (8 h).
The hot-filtration test was carried to substantiate the
heterogeneity of the ZnCo-N@C(ZIF-COF) catalyst (Figure
S12A).[17] The catalyst was removed after 4 h, and the reaction
was continued for another 4 h in the absence of catalyst. No
significant increase in NB conversion was observed, which
signifies the truly heterogeneous nature of the reaction. The
ZnCo-N@C(ZIF-COF) was recycled five times in the N-formylation
of NB (Figure S12B). After each run, the catalyst was filtered,
washed with ethanol, and dried in a vacuum, and then used in
the next cycle. No significant loss in the catalytic activity was
observed after five cycles. The Co content for the spent catalyst
is compared with the fresh catalyst using elemental analysis
(MP-AES). A negligible loss of the cobalt content (2.5 ppm) is
observed for the recovered catalyst. The spent catalyst was
subjected to PXRD, Raman analysis, and TEM measurement
(Figure S13). The obtained results demonstrate that the spent
catalyst is stable and remains undisturbed after the recycling
experiments. The ID/IG value did not change significantly
compared to the fresh catalyst, indicating that the graphitic
nature is preserved during recycling experiments.
KSCN poisoning test was conducted to justifying the role of
the metallic constituents present in ZnCo-N@C(ZIF-COF) in the N-
formylation of NB. After 4 h of the reaction, KSCN was
introduced in the reaction mixture. The progress of the reaction
was seized because SCN- masks and deactivates the metallic
constituents present in the catalyst (Figure S14). KSCN (SCN-)
strongly interacts with the small metallic Co nanoclustures and
Co-Nx species. SCN- masks the active sites by coordinating with
metal centres in the catalyst and deactivate the catalyst.
Whereas the CoOx and larger Co nanoparticles are reported to
exhibit no effect with the KSCN poisoning test.[46,48] This
experiment suggests that the small nanoclusters are present in
the catalyst, which further confirms the results obtained from
follows the order ZnCo-N@C(ZIF-COF)
>
ZnCo-N@C(ZIF-MEL)>
ZnCo-N @C(ZIF). Moreover, the suitable nitrogen environment
(pyrrolic group) present in the catalyst provides stability and
reactivity to Zn-Nx/Co-Nx active sites, which is responsible for the
reductive N-formylation. The XPS result shows the pyrrolic N
content follows the order ZnCo-N@C(ZIF-COF) > ZnCo-N@C(ZIF-
> ZnCo-N@C(ZIF). Being an acidic molecule, HCOOH
MEL)
(HCOOH H2, CO, CO2) efficiently adsorbs on the large surface
area containing basic catalysts.
The NB adsorption on the catalyst surface is also an
important step in the heterogeneous catalytic reaction. The NB
is a UV-active molecule. Hence the extent of NB adsorption on
the different catalysts was estimated using UV-Vis spectroscopy
(Figure S16). Initially, 100 ppm concentration of an ethanolic
solution of NB was taken with 20 mg of the catalyst. After 1 h of
equilibration, the liquid part was withdrawn and evaluated under
UV-Vis spectroscopy. The NB concentration in the solution was
decreased (as shown in Figure S16), which confirms the
6
This article is protected by copyright. All rights reserved.

10.1002/cctc.202100412
ChemCatChem
FULL PAPER
adsorption of NB at the catalyst surface. Thus, the NB
adsorption, optimum basicity, N-environment & contents, and
effective entrapment of highly dispersed metal nanoclusters are
CO2) and dehydration pathway (CO + H2O) at an optimum
temperature in a suitable solvent medium in optimum ratio to
produce formanilide by indirect pathway. Both paths are
responsible for forming the desired reductive N-formylated
product. Zn-Nx sites were active sites for HCOOH adsorption,
and Co nanoclusters & Co-Nx were active sites for CO+H2
generation for reductive N-formylation. The doping of Zn and Co
in the form of tiny nanoclusters on a high surface area porous C-
N framework will be interesting to materials scientists, and its
selective reductive formylation of nitrobenzene will be attractive
for catalysis researchers.
responsible for the excellent activity of ZnCo-N@C(ZIF-COF)
.
After evaluating the catalytic data, control reactions, and
physicochemical properties, a plausible mechanism for the
reductive N-formylation of the NB is proposed. HCOOH and NB
adsorb in close proximity at the surface of ZnCo-N@C(ZIF-COF)
catalyst. The Zn-Nx sites and the C-N framework facilitate the
HCOOH adsorption process by Zn-formate mechanism pathway
or direct pathway.[23] This leads to the decarboxylation of
HCOOH and the formation of Co-H active species, which reacts
with the adsorbed NB to form adsorbed aniline via a transfer
hydrogenation approach. The lone pair of nitrogen of adsorbed
aniline reacts with adsorbed HCOOH, followed by dehydration to
form formanilide via a direct pathway. At the same time, the
dehydration/decarbonylation of HCOOH produces adsorbed CO
species by an indirect pathway at Co sites. Subsequently, the
anilineads and COads react and produce formanilide and
regenerate the active catalytic sites (Scheme 2).
Experimental Section
Catalyst preparation
Initially, the COF was prepared using a reported method.32 A specific
amount of COF was dispersed in methanol. Then, Zn(NO3)2.6H2O (0.550
g) and Co(NO3)2.6H2O (0.184 g) were dissolved in the dispersed solution
and stirred for 2 h. After this, a 70 mL methanolic solution of 2-methyl
imidazole was added and stirred for 24 h at ambient conditions. Then it
was centrifuged and washed with methanol several times and dried
under an oven at 343 K. The synthesized COF-ZIF composite material
was carbonized at the desired temperature with a heating ramp of 5 °K
min−1 in an Ar atmosphere for 2 h and designated as ZnCo-N@C(ZIF-COF)
.
For comparison, another ZnCo-N@C was prepared by taking melamine
instead of COF as support. The obtained ZnCo-N@C was designated as
ZnCo-N@C(ZIF-MEL). The support free ZnCo bimetallic ZIF was also
prepared and carbonized to give ZnCo-N@C(ZIF). The details of
instruments used for the catalyst characterization are provided in the
Supporting Information Section.
Procedure of the catalytic reactions
Scheme 2. Plausible mechanism of the reductive N-formylation of
nitrobenzene over ZnCo-N@C(ZIF-COF)
.
The procedures of the catalytic reactions discussed in this study are
provided in the supporting information section.
Conclusion
In summary, ZnCo-MOF and COF entangled material was
synthesized and carbonized to prepare ZnCo embedded N-
doped porous carbon material. ZnCo embedded N doped
porous carbon exhibited a very good surface area of 684 m2/g.
Raman and TEM analysis confirmed that the material is graphitic.
TEM analysis confirmed that the tiny nanoclusters were highly
dispersed, and metal nanoparticles were not visible in the
HRTEM image. Elemental mapping and XPS analysis confirmed
that Zn's large content was evaporated during the carbonization
process, and low content of Zn was embedded in the C-N
framework. XPS analysis confirmed that Co is present in Co(0),
CoOx, and CoNx states, and pyrrolic, pyridinic, and graphitic N
were present in the C-N framework. The catalyst exhibited
excellent activity in the selective reductive N-formylation of
nitroaromatics using formic acid. The high surface area porous
C-N framework was suitable for the adsorption of nitrobenzene.
The high surface area and basic C-N framework were ideal for
the adsorption of HCOOH. Moreover, Zn sites present in the
catalyst facilitated the adsorption of HCOOH. By the direct
formylation route, adsorbed nitrobenzene and formic acid
reacted to produce formanilide. Embedded Co metals
decomposed HCOOH to H2 and CO via decarboxylation (H2 +
Acknowledgments
Authors are grateful to SERB New Delhi (CRG/2020/000028)
and the Director IIT Ropar for the Mid-carrier Excellency in the
Research through grant number 9-35/2009/IITRPR/3121. AKK
thanks MHRD for Senior Research Fellowship.
Conflicts of interest
The authors declare no competing financial interest.
Keywords: Selective reductive formylation • N-Formylation •
MOF • COF • C-N framework.
Supporting Information
The details for the catalytic reactions, activation enenergy determination, and
some catalytic activity data and physicochemical characterization are provided
in the Supporting Information section. Supporting information also includes
7
This article is protected by copyright. All rights reserved.

10.1002/cctc.202100412
ChemCatChem
FULL PAPER
material characterization, and steps involved in catalytic reactions, XRD, FE-
SEM, catalyst recyclability, optimization of reaction parameters, adsorption
experiments data, and CO2-TPD data.
[32] M. G.Schwab, B. Fassbender, H. W. Spiess, A. Thomas, X.Feng, K.
Mullen, J. Am. Chem. Soc., 2009,131, 7216–7217.
[33] G. L. Zhuang, Y. F.Gao, X.Zhou, X. Y.Tao, J. M.Luo, Y. J.Gao, Y. L. Yan,
P. Y.Gao, X.Zhong and J. G. Wang, Chem. Eng. J.,2017,330, 1255–1264.
[34] X. Xu, S. Zhang, J. Tang, L. Pan, M. Eguchi, J. Na, Y.Yamauchi, Energy
Chem., 2020, 2, 100043–100076
[1] M. M. Khin, A. S. Nair, V. J. Babu, R. Murugan, S. Ramakrishna, Energy
Environ. Sci.,2012, 5, 8075–8109.
[2] W. F. Hoelderich, Catal. Today, 2000, 62, 115–130.
[3] B. Johnston, M. C. Mayo, A.Khare, Technovation, 2005, 25, 569–585.
[4] M. D. Porosoff, B. Yan, J. G.Chen, Energy Environ. Sci., 2016, 9, 62–73.
[5] A. K.Kar, S. P. Kaur, T. J. D. Kumar, R. Srivastava, Catal. Sci. Technol.,
2020, 10, 6892–6901.
[35] J. Tan, X. He, F. Yin, B. Chen, X. Liang, G. Li, H. Yin, Int J Hydrogen
Energy., 2020, 45, 15453–15464.
[36] X. Zhao, N.Yan, RSC Adv., 2015, 5, 69955–69961.
[37] K. Sun, S.Chen, J.Zhang, G. P.Lu, C.Cai, ChemCatChem, 2019,
11,1264–1271.
[6] I. Staffell, D.Scamman, A. V.Abad, P.Balcombe, P. E.Dodds, P. Ekins,
N.Shahd and K. R. Warda, Energy Environ. Sci., 2019, 12, 463–491.
[7] C. S.Martavaltzi, E. P. Pampaka, E. S. Korkakaki, A. A. Lemonidou, Energy
Fuels., 2010, 24, 2589–2595.
[38] J. Xue, Y. Li, J. Hu, J. Mater. Chem. A, 2020, 8, 7145–7157.
[39] Y. Zhang, P. Cao, H. Y. Zhang, G. Yin, J. Jhao, Catal. Commun., 2019,
129, 105747.
[40] X. Lan, B. Ali, Y. Yang, T. Yang, ACS Appl. Mater. Interfaces., 2020, 12,
3624−3630.
[8] P. Parthasarathy, K. S. Narayanan, Renewable Energy., 2014, 66, 570–
579.
[41] Y. Wu, Z. Chen, W. C. Cheong, C. Zhang, L. Zheng, W. Yan, R. Yu, C.
Chen, Y. Li, Chem. Sci., 2019, 10, 5345–5352.
[9] Z.Wang, R. R. Roberts, G. F.Naterer, K. S. Gabriel, Int. J Hydrogen
Energy., 2012,37,16287–16301.
[42] Y. Su, Y. Zhu, H. Jiang, J. Shen, X. Yang, W. Zou, J. Chen, C. Li,
Nanoscale., 2014, 6, 15080–15089.
[10] Y. Wang, S.Zhang, Energy Sources Part B: Econ, Plan Policy., 2017, 12,
1022–1029.
[43] E.Jeroro, J. M. Vohs, Catal Lett., 2009,130, 271–277.
[44] D. A. Bulushev, M.Zacharska, S.Beloshapkin, Y. Guo, I. Yuranov, Appl.
Catal. A., 2018, 561, 96–103.
[11] M. J. Gilkey, B.Xu, ACS Catal., 2016, 6, 1420−1436.
[12] D. Wang, D. Astruc, Chem. Rev., 2015, 115, 6621−6686.
[13] K. Fang, D.Li, M.Linm M. Xiang, W.Wei, Y. Sun, Catal. Today, 2009,147,
133–138.
[45] F. Chen, B. Sahoo, C. Kreyenschulte, H. Lund, M. Zeng, L. He, K. Junge,
M.Beller, Chem. Sci., 2017, 8, 6239–6246.
[14] M.Mahmudzadeh, H.Yari, B.Ramezanzadeh, M. Mahdavian, J. Hazard.
Mater., 2019, 371, 609–624.
[46] C. Tang, A. E. Surkus, F.Chen, M. M. Pohl, G. Agostini, M. Schneider, H.
Junge, M. Beller, Angew. Chem., 2017, 129, 16843-16847.
[47] N. Ishito, K. Hara, K. Nakajima, A. Fukuoka, J Energy Chem., 2016, 25,
306–310.
[15] Q. L. Zhu, Q. Xu, Energy Environ. Sci., 2015, 8, 478–512.
[16] F. Valentini, V.Kozell, C. Petrucci, A.Marrocchi, Y.Gu, D.Gelman, L.
Vaccaro, Energy Environ. Sci., 2019, 12, 2646−2664.
[17] A. K.Kar, R. Srivastava, ACS Sustainable Chem. Eng., 2019, 7,
13136−13147.
[48] M. Li, S. Chen, Q. Jiang, Q. Chen, X. Wang, Y. Yan, J. Liu, C. Lv, W.
Ding, and X.Guo, ACS Catal., 2021, 11, 3026−3039.
[18] J. Reichert, J. Albert, ACS Sustainable Chem. Eng., 2017, 5, 7383−7392.
[19] P. Zhou, Z.Zhang,L. Jiang, C.Yu, K.Lv, J.Sun, S. A. Wang, Appl.Catal., B.,
2017, 210, 522–532.
[20] B. C.Chen, M. S.Bednarz, R. Zhao, J. E. Sundeen, P.Chen, Z.Shen, A.
P.Skoumbourdis, J. C. Barrish, Tetrahedron Lett., 2000, 41, 5453−5456.
[21] D. N. Sawant, Y. S.Wagh, K. D.Bhatte, B. M. Bhanage, J. Org. Chem.,
2011, 76, 5489–5494.
[22] B. B.Lohray, S.Baskaran, B. S.Rao, B. Y.Reddy, I. N. Rao, Tetrahedron
Lett.,1999, 40, 4855–4856.
[23] M. Nasrollahzadeh, N. Motahharifar, M.Sajjadi, A. M.Aghbolagh, M.
Shokouhimehr, R. S. Varma, Green Chem., 2019, 21, 5144–5167.
[24] L.Mámani, M.Sheykhan, A. Heydari, M. Faraji, Y. Yamini, Appl.Catal. A.
2010,377, 64–69.
[25] M. G.Mura, L. D. Luca, G. Giacomelli, A. Porcheddu, Adv. Synth. Catal.,
2012, 354, 3180–3186.
[26] J. J. Sims, C. A. O.Hamou, R.Reocreux, C.Michel, J. B.Giorgi, J. Phys.
Chem. C., 2018,122, 20279−20288.
[27] H. S.Inglis, D. Taylor, J. Chem. Soc. A.,1969, 2985-2987.
[28 M. Li, F.Xu, H. Li, Y. Wang, Catal. Sci. Technol., 2016, 6, 3670–3693.
[29] K.Shen, X.Chen, J.Chen, Y. Li, ACS Catal., 2016, 6, 5887−5903.
[30] W. Yang, X.Li, Y.Li, R.Zhu, H.Pang, Adv. Mater., 2019, 31, 1804740–
1804775.
[31] A. K. Kar, R. Srivastava, Inorg. Chem. Front., 2019, 6, 576–589.
8
This article is protected by copyright. All rights reserved.

10.1002/cctc.202100412
ChemCatChem
FULL PAPER
Insert graphic for Table of Contents here. ((Please ensure your graphic is in one of following formats))
Insert text for Table of Contents here. ((Reactants adsorption, optimum basicity, N-environment & contents, and effective entrapment of highly dispersed metal
nanoclusters are responsible for the excellent activity of ZnCo-N@C(ZIF-COF) in the reductive N-formylation of nitroaromatics to corresponding formanilide as
potential precursors.))
9
This article is protected by copyright. All rights reserved.