Reductive C?N Coupling of Nitroarenes: Heterogenization of MoO3 Catalyst by Confinement in Silica

Article abstract of DOI:10.1002/cssc.202101203

The construction of C?N bonds with nitroaromatics and boronic acids using highly efficient and recyclable catalysts remains a challenge. In this study, nanoporous MoO₃ confined in silica serves as an efficient heterogeneous catalyst for C?N cross-coupling of nitroaromatics with aryl or alkyl boronic acids to deliver N-arylamines and with desirable multiple reusability. Experimental results suggest that silica not only heterogenizes the Mo species in the confined mesoporous microenvironment but also significantly reduces the reaction induction period and regulates the chemical efficiency of the targeted product. The well-shaped MoO₃@m?SiO₂ catalyst exhibits improved catalytic performance both in yield and turnover number, in contrast with homogeneous Mo catalysts, commercial Pd/C, or MoO₃ nanoparticles. This approach offers a new avenue for the heterogeneous catalytic synthesis of valuable bioactive molecules.

Full text of DOI:10.1002/cssc.202101203

Full Papers
doi.org/10.1002/cssc.202101203
ChemSusChem
Reductive CÀ N Coupling of Nitroarenes: Heterogenization
of MoO3 Catalyst by Confinement in Silica
Fu Yang,[a] Xuexue Dong,[a] Yang Shen,[a] Mengting Liu,[a] Hu Zhou,[a] Xuyu Wang,[a] Lulu Li,[a]
Aihua Yuan,[a] and Heng Song*[a]
The construction of CÀ N bonds with nitroaromatics and boronic
acids using highly efficient and recyclable catalysts remains a
challenge. In this study, nanoporous MoO3 confined in silica
serves as an efficient heterogeneous catalyst for CÀ N cross-
coupling of nitroaromatics with aryl or alkyl boronic acids to
deliver N-arylamines and with desirable multiple reusability.
Experimental results suggest that silica not only heterogenizes
the Mo species in the confined mesoporous microenvironment
but also significantly reduces the reaction induction period and
regulates the chemical efficiency of the targeted product. The
well-shaped MoO3@mÀ SiO2 catalyst exhibits improved catalytic
performance both in yield and turnover number, in contrast
with homogeneous Mo catalysts, commercial Pd/C, or MoO3
nanoparticles. This approach offers a new avenue for the
heterogeneous catalytic synthesis of valuable bioactive mole-
cules.
Introduction
Transition metal-catalyzed CÀ N coupling has shaped the dominant
protocols to afford secondary arylamines, which are of great
importance in materials, agrochemical, and pharmaceutical
industries.[1–3] The prevalent means for preparation of (hetero)aryl
amines, such as Buchwald-Hartwig,[4] Ullmann,[5] and Chan-Lam[6]
coupling reactions, involve homogeneous catalysis (Figure 1A).
The aforementioned well-known strategies utilize amines as the
nitrogen source. Generally, amines derivatives, especially for
anilines, are derived from nitro compounds by means of
stoichiometric reductions. Therefore, the direct implication of
nitroarenes rather than amines as starting materials for construc-
tion of (hetero)arylamines not only enables reducing energy and
saving time, but also avoiding the risks with respect to trans-
portation, storage, and handling of dangerous and toxic
intermediates.[7] Although nitroaromatics are efficiently employed
as substrates in intramolecular CÀ N bond-forming reductive
cyclization to afford N-heterocyclic compounds.[8–10] The intermo-
lecular CÀ N bond formation using nitroarene and another partner
in one single operational step has proved to be more challenging
because so many active intermediates such as nitroarenes,
azoarenes, and hydroxylamines may be produced during the
reducing process. Therefore, the determining factor is the
selectively transformation of the nitro compound into the
corresponding intermediate while avoiding overreduction or other
side reactions. In light of the above factors, there are few examples
in intermolecular reductive CÀ N coupling of nitro(hetero)arenes
with other partners.[11–16] Notable exceptions include the work of
Figure 1. (A) Transition metal-catalyzed CÀ N coupling (Buchwald-Hartwig,
Chan-Lam, Ullman). (B) Reductive coupling of nitrobenzene with phenybor-
onic acids. (C) Light-promoted CÀ N coupling of aryl halides with nitroarenes.
(D) Heterogeneous molybdenum-catalyzed CÀ N coupling
Sanz and co-workers,[17] who has demonstrated homogeneous
catalysis for amination of boronic acids with nitro compounds to
yield (hetero)arylamines, in conjunction with the use of affordable
PPh3 as reductant and a MoVI catalyst. Radosevich and co-workers
described a unique main-group PIII/PIV =O catalyzed coupling of
nitroarenes with boronic acid partners. Through DFT calculation,
nitrosoarenes were evidenced as the key intermediates in the
reaction (Figure 1B).[18,19] Very recently, Xue and co-workers
[a] Dr. F. Yang, X. Dong, Y. Shen, M. Liu, H. Zhou, Dr. X. Wang, Dr. L. Li,
Prof. A. Yuan, Dr. H. Song
School of Environmental and Chemical Engineering
Jiangsu University of Science and Technology
Zhenjiang 212003, Jiangsu (P. R. China)
E-mail: hengsong@just.edu.cn
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/cssc.202101203
ChemSusChem 2021, 14, 1–10
1
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cssc.202101203
ChemSusChem
disclosed a Ni-catalyzed and light-promoted CÀ N coupling of aryl
Enabling the high catalytic conversion still relies on the
high-density exposed active centers located on the catalyst
carrier, whereas the chemical efficiency modulation of reaction
demands appropriate functional catalyst support to optimize
the active phase state, thereby triggering a regulated reaction
route by efficient metal-support interplay. In this contribution,
MoO3-functionalized nanoporous silica catalyst was developed
with a guest@host configuration, denoted as MoO3@mÀ SiO2.
Our self-assembly strategy demonstrated Mo5+ cation coordi-
nating with dodecyl amine (DDA) could contribute to the
fabrication of a dispersed MoO3 cluster-confined functionalized
mesoporous nanocatalyst. The obtained catalyst was character-
ized through corresponding techniques to identify the phys-
icochemical properties including the special mesoporous
structure and MoO3 status in the confined mesopore space.
Catalyzed by the resultant MoO3@mÀ SiO2, we demonstrate the
efficient coupling of various nitroarenes with boronic acid
partners to afford various biarylamine products (Figure 1D) with
desirable recyclability, besides, we provided corresponding
mechanism identification for the improved catalytic properties
based on the developed MoO3@mÀ SiO2. This study provides the
first example for enabling the reductive CÀ N coupling of
nitroarenes and boronic acid using heterogeneous transition
metal catalysts even beyond the homogeneous counterparts.
[20]
°
halides with nitroarenes under 390 nm LEDs at 70 C (Figure 1C).
Nevertheless, utilization of heterogeneous catalysis for CÀ N
coupling with nitroarenes as nitrogen source remains scarce.
However, heterogeneous catalysis for the amination of boronic
acids with nitroarenes affords remarkable advantages in
recycling and reuse of catalysts and provides possibilities for
industrial applications, bulk chemicals. Moreover, the facile
separation and purification in the heterogeneous systems to
approach the desired products are much more beneficial to
consider the economy and green chemistry.
Despite the noble metal catalysts afford more advantages in
the reaction activity, the extremely high cost and low abundance
still hamper their wide applications. Exploring the rational transi-
tional metal catalysts seems preferable for organic reactions. In
previously reported cases, homogeneous molybdenum catalyst
decorated with various well-designed ligands has performed their
rational catalytic activity in this reaction.[17] Molybdenum oxides
with the high valence of MoVI and variable valence state have
attracted great attention for organic synthesis,[21] photocatalysis,[22]
and electrochemistry.[23] A unique advantage of heterogeneous
molybdenum oxides is that they can withstand a considerable
amount of oxygen vacancies and a certain metallicity degree.[24,25]
Moreover, many oxygen-deficient oxides (MoO<3) have stable
crystal phases, not just surface-reduced MoO3, which is an
attractive characteristic for new catalytic applications.[26] MoO3
serves as an effective catalytic active phase, owing to the possible Results and Discussion
presence of mixed valences including Mo6+, Mo5+, and Mo4+
,
which would be very beneficial for the redox cycle during the
catalytic process.[27] The bulk-like catalytic active phase fails to
provide sufficient catalytic reactive centers for the reaction,
generally rendering poor catalytic activity during the reaction.[28]
Tailoring the active size of catalytic substance into nanometric
even subnanometric level is normally promising for creating more
intrinsic activity sites, which could produce a great promotive
effect in catalytic transformations.[29] However, nanometric MoO3
species without host protection might easily loss their intrinsic
activity due to the rapid poison effect in the complicated reaction
conditions. The encapsulation of highly active MoO3 species in the
host materials might provide the possibility for stabilizing them as
the enduring active phase. So far, many attempts have denoted to
create a well-defined guest@host configuration embracing the
densely-dispersed active phase species in the flourishing hole of
porous materials.[30] However, the difficulty for the subject mainly
focused on developing an efficient constructed approach to reach
the rational functionalization of host materials with refined active
species and avoid the block of pore structure and wide
aggregation of metal oxides. Mesoporous silica provides a greater
superiority for the encapsulation of metal oxides to reach the
above-mentioned expectation compared to microporous zeolite
catalyst thanks to the suitable pore size and volume which might
be more applicable for the special reaction. However, the
traditional post-impregnation approach to attain the expected
confined catalyst is undesirable compared to the in situ assembly
process owing to the uncontrolled active phase growth in host
materials under a high temperature-driven annealing process.[31]
The resulted final catalyst was labeled as MoO3@mÀ SiO2. The
synthetic process is shown schematically in Figure 2. Capture of
DDA micelle for Mo5+ ions was regarded as the vital step to
form the MoO3 functionalized nanocatalyst. Specifically, the
Mo5+-functionalized DDA micelle, formed through coordination
interactions, would match with the hydrolyzed negatively
charged SiÀ O oligomers based on the counter-ions medium.
However, note that high-concentration of Mo5+ would be easily
2À
transformed to the MoO4 under the high pH environment,
thereby failing to contribute to the self-assembly of mesopo-
rous silica. As a result, the highest Mo introduction content with
the formation of Mo5+/DDA was controlled as a 0.1 molar ratio.
Upon reaching higher Mo introduction content, the mesopo-
rous silica would not be obtained in this case.
The morphology and appearance were ascertained by SEM
and TEM techniques. Specifically, the typical morphology of the
Figure 2. Synthesis of the MoOx-modified porous silica.
ChemSusChem 2021, 14, 1–10
www.chemsuschem.org
2
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cssc.202101203
ChemSusChem
sample was observed and defined as nanosphere with a large
number of external voids (Figure 3a). Moreover, the porous
state of the catalyst can be identified without any observation
of bulk-like MoO3 particles in the TEM image (Figure 3b),
suggesting no formation of aggregated Mo oxides on the
catalyst. But note the uniform distribution of Mo on the catalyst
can be demonstrated by the elemental mapping information
(Figure 3c). Correspondingly, the MoO3 species should be
confined into the mesoporous microenvironment as the highly-
dispersed phase in a guest@host configuration. Afterward, N2
adsorption-desorption measurements were conducted to ascer-
tain the porous structural characters.
The initial mesoporous state was identified and exhibited
typical type-IV adsorption behavior for MoO3@mÀ SiO2 (Figure 4a).
Nonlocal density functional theory (NLDFT) pore size distribution
results derived from adsorption isotherms demonstrated the
nearly microporous pore size state (1.6 nm) of the catalyst after
implanting MoO3 species into the original mesopore (Figure 4b).
The specific surface area and pore volume were calculated to be
634 cm2 gÀ 1 and 0.488 cm3 gÀ 1, respectively, which indicates a
certain weakening of the structural properties of pure mesoporous
silica (868 m2 gÀ 1, 0.78 cm3 gÀ 1), but still highlight the good porous
structure of MoO3@mÀ SiO2 and properties that could prove
greatly advantageous for the reaction process. Low-angle XRD
patterns were used for identifying the periodicity of mesoporous
structure (see the Supporting Information, Figure S1). Only weak
diffraction ascribed to the 100 mesophase peak can be observed,
suggesting certain retention of the ordered mesoporous structure
after introducing MoO3.[32] Apart from that, wide-angle XRD
patterns extracted from the MoO3@mÀ SiO2 sample demonstrated
the dominant amorphous state of the catalyst. In addition to the
°
broad diffraction peak at 20–30 , no visible diffraction of Mo
oxides phase was identified, which further confirms the high
dispersion and refined size of Mo oxide species in the mesopores
owing to the strong metal-carrier interaction and confinement
effects.[33]
Subsequently, the FTIR and UV/Vis spectra of samples were
ascertained to collect the moiety information of metal-support
and coordination state of Mo element. In the FTIR spectra
(Figure S2a), a representative absorption band ascribed to silica
indicated the generation of the silica carrier. Moreover, a typical
Mo=O moiety was detected in the high-loading Mo-functional-
ized MoO3@mÀ SiO2, confirming the immobilization of Mo
oxides in silica. The LMCT band at around 251 nm indicates the
presence of a significant fraction of isolated surface molybdates
(Figure S2b).[34] Diffuse-reflectance UV/Vis spectra further show
the UV absorption at 312 nm assigned to the charge transfer
from oxygen ligands to MoVI, which usually corresponding to
octahedrally coordinated MoVI in oxides as low-oligomeric
surface MoO3 supported on silica.[34] We propose the possible
conformations of Mo oxides on the surface of silica in Figure 5.
In general, Mo oxides can be divided into isolated or
aggregated species, both of which are present in the obtained
catalyst. However, whatever low or high loading of Mo in silica,
the isolated or aggregated Mo species would be formed.
Regarding the XPS analysis (Figure 6), Mo in silica could be
identified by the right of the XPS survey (Figure 6a). Mo3d3/2
and Mo3d5/2 core levels, with the orbital binding energies are
Figure 3. (a) SEM image and (b) TEM image of Mo–MS. (c) Corresponding
elemental mapping images.
Figure 4. N2 adsorption-desorption isotherms (a) and NLDFT pore size distribution (b) of MoO3@mÀ SiO2.
ChemSusChem 2021, 14, 1–10
www.chemsuschem.org
3
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cssc.202101203
ChemSusChem
Figure 5. Monomeric (a,b) and dimeric (c,d) surface structures interacting with silica.
Figure 6. X-ray photoelectron spectroscopy element survey and Mo3d core level of MoO3@mÀ SiO2.
235.5 eV and 232.6 eV (Figure 6b), respectively, and the absorp-
tion peak at 232.7 eV is attributed to the highest state of Mo.
Mo in MoO3 has the valence state MoVI.[35] On the other hand,
the observed Mo3d3/2 and Mo3d5/2 core-levels still afford certain
differences to pure MoO3 phase, which gives rise to an intense
doublet at EB (Mo3d5/2)=232.8 eV and EB (Mo3d3/2)=235.9 eV,
indicating the formed MoO3 interacts with the silica support by
strong chemical bonding.[36] Moreover, the O1s core-level with
relatively symmetrical peak shape indicate the dominant
composition of SiÀ OÀ Si configuration with only a little different
OÀ Mo in low loading (Figure S3).
we examined the reductive CÀ N bond formation between nitro
(hetero)arenes and boronic acids and found that porous
MoO3@mÀ SiO2 is capable of catalyzing this challenging trans-
formation. Under the optimized reaction conditions using
MoO3@mÀ SiO2 (2 mol% based on Mo atoms) as the catalyst
and 3 equivalents of PPh3 as the terminal reductant in toluene
solvent. The reductive CÀ N coupling between 1-chloro-4-nitro-
°
benzene and phenylboronic acid proceeded well at 100 C for
24 h, affording the diarylamine (3a) product in 83% isolated
yield. Aniline derivatives that often form as byproducts with
such coupling catalysis were not detected in this reaction.
With optimized conditions in hand (Table S1), the scope of
both nitro compound and boronic acid partners was extensively
studied. As shown in Table 1, a series of nitroarenes were
subjected to reaction with boronic acids, and the corresponding
amines were obtained in moderate to excellent yields. Various
halide groups at the para, ortho, or meta position on the
aromatic ring do not affect this reaction. Halides are well
tolerated in different positions with excellent yields (Table 1,
entries 1–4). Dimethyl, an electron-donating group (EDG), was
also tested and remained unscathed (Table 1, entry 5). Notable
performance is accomplished with nitroarenes bearing strong
electron-withdrawing groups (EWG), such as nitrile (Table 1,
entry 6) or ester (entry 8). Although being a reductive process,
even aldehyde (Table 1, entry 7) is not reduced, as opposed to
the classic reductive amination that requires the protection of
Catalytic activity assessment
We have previously reported that the partial reduced inter-
mediate, nitroso-compound derived from nitroaromatics, is
active material for CÀ N coupling reactions with olefins
partners.[37] Boronic acids are widely used building blocks in
organic chemistry as they exhibit good functional group
compatibility and synthetic versatility and are easily available
from commercial sources.[38] With respect to these features,
implement of a direct amination of boronic acids with nitro
compounds is an attractive project. Herein, we hypothesized
that the nitroso intermediate may allow the direct electrophilic
amination of boronic acids to afford secondary amines. Initially,
ChemSusChem 2021, 14, 1–10
www.chemsuschem.org
4
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cssc.202101203
ChemSusChem
Table 1. Substrate scope.[a]
Entry
Substrate
1
Product
Yield [%]
2
1
2
83
82
3
66
4
5
71
67
6
78
7
8
65
86
9
80
60
75
84
57
10
11
12
13
ChemSusChem 2021, 14, 1–10
www.chemsuschem.org
5
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cssc.202101203
ChemSusChem
Table 1. continued
Entry
Substrate
1
Product
Yield [%]
64
2
14
[a] Reaction conditions: catalyst (2 mol% based on Mo atoms), nitroarene (0.5 mmol), R2B(OH)2 (0.75 mmol, 1.5 equiv), PPh3 (3 equiv.), in toluene (4 mL) for
24 h under N2 atmosphere. Isolated yields were given.
these functional groups.[39] Nitroalkane such as nitromethane or
nitropropane has been adopted as substrate to couple with
phenylboronic acid in the catalytic conditions, but no desired
modified with well-designed ligands. We suggest that the well-
dispersed MoO3 interacted with silica afford the open reaction
microenvironment which could contact the reactant molecule
without the steric hindrance of modification ligand. This provides
a certain superiority for the reaction to exclude the effect of
surface ligand, thus contributing to higher catalytic activity and
efficiency. To our delight, the as-prepared MoO3@mÀ SiO2 catalyst
exhibited a much better catalytic performance in reductive CÀ N
coupling of nitroarenes with boronic acids than unmodified
commercial MoO3 NPs (Table 1, entry 5).
1
product was observed, judging from H NMR and TLC analysis.
The results are consistent with previously reported that the
nitroalkane is inherently less reactive than ArÀ NO2, due to a
larger local frontier orbital energy gap (ΔΔE=1.0 eV).[40]
Next, a series of different boronic acids was tested. Notable
performance is accomplished with boronic acids bearing strong
electron-donating groups (EWGs), such as methoxy (Table 1,
entry 9) or methyl (entry 11) moieties. At the same time, the
phenoxy group is also tolerated (Table 1, entry 10). Moreover,
m-substituted boronic acids proved to effectively promote the
transformation (Table 1, entries 11 and 12). To our delight, the
reaction is not just limited to C(sp2)À N bond formation; indeed,
the amination of nitrobenzene with C(sp3) boronic acid reagents
including strained cycloalkyl- and secondary alkyl-boronic acids
provide serviceable yields of the desired C(sp3)À N cross-
coupling products (3m and 3n).
As compared with homogeneous MoO2Cl2(dmf)2/bpy (dmf =
N,N-dimethylformamide; bpy = 2,2’-bipyridine) catalysis (Table 2),
the heterogeneous MoO3@mÀ SiO2 catalysis dispersed by silica
displays an improved catalytic performance on the basis of TON
value, the TON of which is about twice that of Sanz’s homogenous
catalyst. The result reveals that Mo sited interacted with silica
could be further optimized for the reductive coupling of nitro-
aromatic component compared to the complexes that of Mo
With the same reaction conditions, MoO3@mÀ SiO2 exhibited
both much higher conversion and higher yield (83% to
°
coupling products) at 100 C. In contrast, only 17% of coupling
product 3b was observed with commercial MoO3 NPs as
catalyst (Figure 7). With regard to Pd/C catalyst, formation of
cross-coupling product was not observed. Overall, the present
transformation allows the heterogeneous catalytic reductive
amination of boric acids with nitroarenes to afford secondary
amines, which is complementary to the previously described
homogeneous catalysis.
Compared with commercial MoO3 NPs, the catalytic per-
formance of MoO3@mÀ SiO2 improved dramatically in terms of
Table 2. Turnover numbers of various catalysts.[a]
Entry
R
Catalyst
TON
1
2
3
4
5
6
Cl
F
CN
CHO
F
MoO3@mÀ SiO2 vs. MoO2Cl2(dmf)2/bpy
MoO3@mÀ SiO2 vs. MoO2Cl2(dmf)2/bpy
MoO3@mÀ SiO2 vs. MoO2Cl2(dmf)2/bpy
MoO3@mÀ SiO2 vs. MoO2Cl2(dmf)2/bpy
MoO3@mÀ SiO2 vs. MoO3 NPs
42 vs. 18
41 vs. 14
39 vs. 16
33 vs. 19
41 vs. 8
Figure 7. Catalytic performance of commercial MoO3 NPs and MoO3@mÀ SiO2
for CÀ N coupling. Conditions: catalyst (2 mol% based on Mo atoms),
nitroarene (0.5 mmol), PhB(OH)2 (0.75 mmol, 1.5 equiv), PPh3 (3 equiv), in
toluene (4 mL) for 24 h under N2 atmosphere. Isolated yields were given.
F
MoO3@mÀ SiO2 vs. Pd/C
41 vs. –
[a] With the exception of the catalysts, reaction conditions were as given
Table 1.
ChemSusChem 2021, 14, 1–10
www.chemsuschem.org
6
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cssc.202101203
ChemSusChem
yield and TON. The yield of coupling amines increased from
17% for commercial MoO3 NPs to 83% for MoO3@mÀ SiO2, while
the TON increased from 16 to 42 (Figure 7). The results again
evident that the emerged silica interacted with MoO3 contrib-
utes to the enhanced activity conversion and efficiency and the
appearance of silica as ligand interacted with Mo could
optimize the reaction pathways toward the promotive produc-
tion of targeted coupling products. The results consisted with
previously reported work that CuÀ OÀ SiOx interfaces boost the
catalysis of copper.[41]
To identify the difference of reaction course in the light of
MoO3@mÀ SiO2 and commercial MoO3 NPs, the reaction progress
of the catalytic coupling of 1b and phenylboronic acid 2a was
monitored via 19F NMR analysis utilizing 4-fluorotoluene as an
internal standard. The reaction curves in Figure 8a indicate that
the forming rate of secondary amine decreased with the
concentration of starting material reduced. As expected, the part
of consumed nitroarene was quantitatively converted into the
corresponding coupling products other than forming byproducts
such as aniline. Substrate 1b was converted to diarylamine 3b in
82% yield in the presence of MoO3@mÀ SiO2 catalyst and without
discernible intermediates, consistent with the observations from
1H NMR spectroscopy. As to the commercial MoO3 NPs catalyst
(Figure 8b), the production of 3b was not observed even after 4 h
under reaction conditions judging from 19F NMR. In the contrast,
the MoO3@mÀ SiO2 catalysis exhibited dramatic reactivity in the
starting period of the reaction and the induction period of
MoO3@mÀ SiO2 catalysis was dramatically reduced compared with
that of commercial MoO3 NPs (Figure 8b). This phenomenon
implies that the silica platform not only regulates the chemical
efficiency of the targeted product through metal-support interplay
but also stabilizes catalyst via adjusting the steric hindrance of the
active metal site.
Mechanism
A proposed mechanism is depicted in Scheme 1 according to the
result reported by Sanz et al.[17] It is known that the molybdenum
(VI) oxide could be easily reduced by PPh3 to generate MoIV
intermediate Int1 with a vacuum coordinated site. We propose
that the coordination of nitroarene PhÀ NO2 to the molybdenum
(IV) center of Int1 leads to the formation of a MoIV-substrate (Int2)
intermediate. After deoxygenation of Ph-NO2 by MoIV results in the
formation of nitroso compound PhÀ N=O, and in conjunction with
the regeneration of catalyst MoO3@mÀ SiO2. In transition state TS1,
the PhÀ N=O interacted both with PPh3 and boronic acid.
Subsequently, CÀ N coupling occurred after 1,2-migration of
nucleophilic R group and electron transfer from NÀ O to PÀ O.
Eventually, a secondary amine product was provided by hydrolysis.
To identify the origin of regulated catalytic selectivity, the silica
interacted with Mo site was suggested as the possible ligand to
regulate the electronic state of active Mo and trigger a faster
catalytic reaction process.
Stability assessment
Recycling stability of heterogeneous catalysts the prerequisite
criteria for practical applications was tested with MoO3@mÀ SiO2
as catalyst (Figure 9a). No obvious loss of activity and selectivity
was observed during the five recycling processes. Moreover, we
further characterized the spent MoO3@mÀ SiO2 catalyst using
the XPS technique, as observed in Figure 9b, the Mo3d core
level of spent catalyst hardly exhibit any variation of binding
energy after multiple uses, which confirms the great stability
and durability of the synthesized MoO3@mÀ SiO2 catalyst during
the reductive CÀ N coupling of nitroarenes with phenylboronic
acid. This study might provide the possibility for industrial
application by virtue of the low-cost transition metal Mo-based
catalyst, which affords potential important value (see Figures S4
and S5 for more details).
Figure 8. (a) The reaction profile of MoO3@mÀ SiO2 catalyst in the cascade reductive coupling as a function of time. (b) The reaction profile of commercial
MoO3 NPs catalyst in the cascade reductive coupling as a function of time; Reaction conditions were the same as in Table 1.
ChemSusChem 2021, 14, 1–10
www.chemsuschem.org
7
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cssc.202101203
ChemSusChem
Scheme 1. Proposed mechanism for coupling of nitroarene with boronic acid in Mo site interacted with silica support.
Figure 9. (a) Recycling stability of MoO3@mÀ SiO2, the reaction conditions were the same as those in Table 1. (b) High-resolution Mo3d core-level XPS spectra
of the fresh and spent MoO3@mÀ SiO2 samples.
Conclusion
from ligands during the reaction process. This work provides
some insight on exploiting heterogeneous supports in the
reductive CÀ N coupling of nitroarenes by heterogeneous
transition metal catalysis.[42]
In summary, we have developed a heterogeneous catalyst
composed of MoO3 confined in mesoporous silica (denoted as
MoO3@mÀ SiO2). The resultant catalyst exhibited excellent
activity in the reductive CÀ N coupling of nitroarenes and
phenylboronic acid, which exceeded those of homogeneous Experimental Section
counterparts and commercial MoO3. The developed catalyst
also showed high tolerance for reductive CÀ N coupling of
various substituted nitroarenes and phenylboronic acid with
high catalytic activity. By comparing the TONs of MoO3@mÀ SiO2
and its homogeneous counterparts, we identified that the
catalytic activity enhancement of Mo active sites originates
from the interaction with silica, which replaces the ligands in
the homogeneous catalyst, rather than from the increased
exposure of active sites. The heterogeneous Mo sites on the
surface of silica could eliminate the effect of steric hindrance
Synthesis of molecular sieve-encapsulated MoO3
The synthesis of the heterogeneous MoO3@mÀ SiO2 catalyst was
designed to use Mo5+ and proceeded as follows. A solution of
dodecyl amine (1.85 g, 0.01 mol) in a mixture of deionized water
(62.5 mL), and ethanol (25 mL) was prepared in advance.
MoCl5 ·4H2O (1 mmol) was introduced and the reaction mixture
turned from an emulsion to a translucent blue solution. The
siliceous precursor tetraethyl orthosilicate (10.7 mL) was added
under the robust stirring and allowed to fully hydrolyze during 4 h
ChemSusChem 2021, 14, 1–10
www.chemsuschem.org
8
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cssc.202101203
ChemSusChem
°
Rubio-Presa, S. Suárez-Pantiga, M. R. Pedrosa, R. Sanz, Adv. Synth. Catal.
2018, 360, 2216–2220.
stirring. The resulted mixture was further aged for 18 h at 45 C. The
obtained product was collected after centrifugation and washing
with water and ethanol for twice and then annealed for 6 h at
[11] J. Gui, C.-M. Pan, Y. Jin, T. Qin, J. C. Lo, B. J. Lee, S. H. Spergel, M. E.
Mertzman, W. J. Pitts, T. E. La Cruz, M. A. Schmidt, N. Darvatkar, S. R.
Natarajan, P. S. Baran, Science 2015, 348, 886–891.
[12] C. W. Cheung, X. Hu, Nat. Commun. 2016, 7, 12494.
[13] C. W. Cheung, M. L. Ploeger, X. Hu, Chem. Sci. 2018, 9, 655–659.
[14] C. W. Cheung, J.-A. Ma, X. Hu, J. Am. Chem. Soc. 2018, 140, 6789–6792.
[15] M. L. Ploeger, A. Darù, J. N. Harvey, X. Hu, ACS Catal. 2020, 10, 2845–
2854.
[16] C. W. Cheung, M. L. Ploeger, X. Hu, Nat. Commun. 2017, 8, 14878.
[17] S. Suárez-Pantiga, R. Hernández-Ruiz, C. Virumbrales, M. R. Pedrosa, R.
Sanz, Angew. Chem. Int. Ed. 2019, 58, 2129–2133; Angew. Chem. 2019,
131, 2151–2155.
°
550 C in air.
General procedure for the synthesis of arylamines
An oven-dried 10 mL vial was charged with the nitro compound
(0.5 mmol, 1.0 equiv), toluene (4.0 mL), MoO3@mÀ SiO2 (40 mg,
0.01 mmol, 2 mol% Mo atoms). After stirring for 2 min, boronic acid
(0.75 mmol, 1.5 equiv) was added, followed by triphenylphosphine
°
(1.5 mmol, 3 equiv). The obtained suspension was stirred at 100 C for
[18] G. Li, T. V. Nykaza, J. C. Cooper, A. Ramirez, M. R. Luzung, A. T.
24 h in a preheated oil bath under N2 atmosphere. The reaction
mixture was then filtered to remove insoluble matter and the filtrate
was concentrated under reduced pressure to afford the crude (hetero)
arylamine, which was purified by flash column chromatography on
silica gel unless otherwise noted.
Radosevich, J. Am. Chem. Soc. 2020, 142, 6786–6799.
[19] T. V. Nykaza, J. C. Cooper, A. G. Li, N. Mahieu, A. Ramirez, M. R. Luzung,
A. T. Radosevich, J. Am. Chem. Soc. 2018, 140, 15200–15205.
[20] G. Li, L. Yang, J.-J. Liu, W. Zhang, R. Cao, C. Wang, Z. Zhang, J. Xiao, D.
Xue, Angew. Chem. Int. Ed. 2021, 60, 5230–5234; Angew. Chem. 2021,
133, 5290–5294.
[21] a) A. P. Amrute, S. Sahoo, A. Bordoloi, Y. K. Hwang, J.-S. Hwang, S. B.
Halligudi, Catal. Commun. 2009, 10, 1404–1409; b) A. R. Mouat, T. L.
Lohr, E. C. Wegener, J. T. Miller, M. Delferro, P. C. Stair, T. J. Marks, ACS
Catal. 2016, 6, 6762–6769; c) J. Li, S. Liu, T. L. Lohr, T. J. Marks,
ChemCatChem 2019, 11, 4139–4146.
[22] Z. Xie, Y. Feng, F. Wang, D. Chen, Q. Zhang, Y. Zeng, W. Lv, G. Liu, Appl.
Catal. B 2018, 229, 96–104.
[23] Z. Chen, D. Cummins, B. N. Reinecke, E. Clark, M. K. Sunkara, T. F.
Jaramillo, Nano Lett. 2011, 11, 4168–4175.
[24] Q. Liu, Y. Wu, J. Zhang, K. Chen, C. Huang, H. Chen, X. Qiu, Appl. Surf. Sci.
2019, 490, 395–402.
[25] G. Song, W. Gong, S. Cong, Z. Zhao, Angew. Chem. Int. Ed. 2021, 60,
5505–5511; Angew. Chem. 2021, 133, 5565–5571.
[26] S. Cao, J. Li, Y. Lin, T. Pan, G. Du, J. Zhang, L. Yang, X. Chen, L. Lu, N. Min,
Solar RRL 2019, 3, 1900274.
[27] K. Murugappan, C. Mukarakate, S. Budhi, M. Shetty, M. R. Nimlos, Y.
Román-Leshkov, Green Chem. 2016, 18, 5548–5557.
Acknowledgements
The authors acknowledge the project funding by the National
Natural Science Foundation (21908085), Natural Science Founda-
tion of Jiangsu Province (BK20190961), Postdoctoral Research
Foundation of Jiangsu Province (2020Z291), Natural Science
Research Project of Jiangsu Province Colleges and Universities
(20KJB150014), Scientific Research Foundation of Jiangsu Univer-
sity of Science & Technology (1112921902, 1112931909) and Youth
Science and Technology Innovation Project of Jiangsu University
of Science & Technology (1112922001).
[28] J. Wang, D. Zhao, P. Yang, L. Liu, S. Xiang, H. Chen, X. Ma, Mater. Des.
2016, 98, 41–46.
[29] F. Yang, S. Ding, H. Song, N. Yan, Sci. China Mater. 2020, 63, 982–992.
[30] C. Gao, F. Lyu, Y. Yin, Chem. Rev. 2021, 121, 834–881.
[31] S. Gao, F. Yang, X. Liu, X. Hu, C. Song, S. Zhou, Y. Kong, Mol. Catal. 2019,
478, 110568–110578.
Conflict of Interest
The authors declare no conflict of interest.
[32] F. Yang, S. Gao, Y. Ding, S. Tang, H. Chen, J. Chen, J. Liu, Z. Yang, X. Hu,
A. Yuan, New J. Chem. 2019, 43, 19020–19034.
[33] S. Yang, G. Liang, A. Gu, H. Mao, Appl. Surf. Sci. 2013, 285, 721–726.
[34] A. Uchagawkar, A. Ramanathan, Y. Hu, B. Subramaniam, Catal. Today
2020, 343, 215–225.
Keywords: CÀ N
coupling
·
confinement
effects
·
heterogeneous catalysis · molybdenum · nitroaromatics
[35] T. G. Novak, J. Kim, A. P. Tiwari, J. Kim, S. Lee, J. Lee, S. Jeon, ACS
Sustainable Chem. Eng. 2020, 8, 11276–11282.
[1] M. Liang, J. Chen, Chem. Soc. Rev. 2013, 42, 3453–3488.
[2] J. Wang, K. Liu, L. Ma, X. Zhan, Chem. Rev. 2016, 116, 14675–14725.
[3] C. Torborg, M. Beller, Adv. Synth. Catal. 2009, 351, 3027–3043.
[4] J. F. Hartwig, Angew. Chem. Int. Ed. 1998, 37, 2046–2067; Angew. Chem.
1998, 110, 2154–2177.
[5] J. Hassan, M. Sévignon, C. Gozzi, E. Schulz, M. Lemaire, Chem. Rev. 2002,
102, 1359–1470.
[36] L. Zheng, Y. Xu, D. Jin, Y. Xie, Chem. Mater. 2009, 21, 5681–5690.
[37] H. Song, Z. Yang, C.-H. Tung, W. Wang, ACS Catal. 2020, 10, 276–281.
[38] V. Coeffard, X. Moreau, C. Thomassigny, C. Greck, Angew. Chem. Int. Ed.
2013, 52, 5684–5686; Angew. Chem. 2013, 125, 5794–5796.
[39] M. Rauser, C. Ascheberg, M. Niggemann, Angew. Chem. Int. Ed. 2017, 56,
11570–11574; Angew. Chem. 2017, 129, 11728–11732.
[40] G. Li, Z. Qin, A. T. Radosevich, J. Am. Chem. Soc. 2020, 142, 16205–16210.
[41] C. Xu, G. Chen, Y. Zhao, P. Liu, X. Duan, L. Gu, G. Fu, Y. Yuan, N. Zheng,
Nat. Commun. 2018, 9, 3367.
[6] J. C. Vantourout, H. N. Miras, A. Isidro-Llobet, S. Sproules, A. J. B. Watson,
J. Am. Chem. Soc. 2017, 139, 4769–4779.
[7] J. Xiao, Y. He, F. Ye, S. Zhu, Chem 2018, 4, 1645–1657.
[8] a) J. I. G. Cadogan, M. Cameron-Wood, R. K. Mackie, R. J. G. Searle, J.
Chem. Soc. 1965, 4831–4837; b) R. Sanz, J. Escribano, M. R. Pedrosa, R.
Aguado, F. J. Arnáiz, Adv. Synth. Catal. 2007, 349, 713–718.
[9] a) H. Gao, Q.-L. Xu, M. Yousufuddin, D. H. Ess, L. Kgrti, Angew. Chem. Int.
Ed. 2014, 53, 2701–2705; Angew. Chem. 2014, 126, 2739–2743; b) R.
Rubio-Presa, M. a. R. Pedrosa, M. A. Fernández-Rodríguez, F. J. Arnáiz, R.
Sanz, Org. Lett. 2017, 19, 5470–5473.
[42] F. Zhang, J. Zhang, B. Zhang, L. Zheng, X. Cheng, Q. Wan, B. Han, J.
Zhang, Green Chem. 2020, 22, 5995–6000.
Manuscript received: June 9, 2021
Revised manuscript received: July 6, 2021
Accepted manuscript online: July 7, 2021
Version of record online: ■■■, ■■■■
[10] a) S. Tong, Z. Xu, M. Mamboury, Q. Wang, J. Zhu, Angew. Chem. Int. Ed.
2015, 54, 11809–11812; Angew. Chem. 2015, 127, 11975–11978; b) R.
ChemSusChem 2021, 14, 1–10
www.chemsuschem.org
9
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

FULL PAPERS
Mo’ better catalysts: Efficient
catalytic construction of CÀ N bonds
with nitroaromatics and boronic
acids remains a challenging aim.
Nanoporous MoO3 confined in silica
serves as an efficient heterogeneous
catalyst for CÀ N cross-coupling of ni-
troaromatics with aryl or alkyl
boronic acids to deliver N-arylamines.
Moreover, the catalyst exhibits
improved yield and turnover number
in comparison to homogeneous Mo
catalysts.
Dr. F. Yang, X. Dong, Y. Shen, M. Liu,
H. Zhou, Dr. X. Wang, Dr. L. Li, Prof. A.
Yuan, Dr. H. Song*
1 – 10
Reductive CÀ N Coupling of Nitroar-
enes: Heterogenization of MoO3
Catalyst by Confinement in Silica