Sustainable Hydrogenation of Nitroarenes to Anilines with Highly Active in-situ Generated Copper Nanoparticles

Article abstract of DOI:10.1002/cctc.202000150

Metal nanoparticles (NPs) are usually stabilized by a capping agent, a surfactant, or a support material, to maintain their integrity. However, these strategies can impact their intrinsic catalytic activity. Here, we demonstrate that the in-situ formation of copper NPs (Cu⁰NPs) upon the reduction of the earth-abundant Jacquesdietrichite mineral with ammonia borane (NH₃BH₃, AB) can provide an alternative solution for stability issues. During the formation of Cu⁰NPs, hydrogen gas is released from AB, and utilized for the reduction of nitroarenes to their corresponding anilines, at room temperature and under ambient pressure. After the nitroarene-to-aniline conversion is completed, regeneration of the mineral occurs upon the exposure of Cu⁰NPs to air. Thus, the hydrogenation reaction can be performed multiple times without the loss of the Cu⁰NPs’ activity. As a proof-of-concept, the hydrogenation of drug molecules “flutamide” and “nimesulide” was also performed and their corresponding amino-compounds were isolated in high selectivity and yield.

Full text of DOI:10.1002/cctc.202000150

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Accepted Article
Title: Sustainable Hydrogenation of Nitroarenes to Anilines with Highly
Active in-situ Generated Copper Nanoparticles
Authors: Fatma Pelin Kinik, Tu N. Nguyen, Mounir Mensi, Christopher
P. Ireland, Kyriakos C. Stylianou, and Berend Smit
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.202000150
Link to VoR: http://dx.doi.org/10.1002/cctc.202000150
A Journal of
www.chemcatchem.org

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
10.1002/cctc.202000150
ChemCatChem
FULL PAPER
Sustainable Hydrogenation of Nitroarenes to Anilines with
Highly Active in-situ Generated Copper Nanoparticles
F. Pelin Kinik[a], Tu N. Nguyen[a], Mounir Mensi[b], Christopher P. Ireland[a],
Kyriakos C. Stylianou*[a] and Berend Smit*[a]
[a]
[b]
Laboratory for Molecular Simulation (LSMO), Institute of Chemical Sciences and Engineering (ISIC), École Polytechnique Fédérale de Lausanne, CH-1950
Sion, Switzerland.
Institute of Chemical Sciences and Engineering (ISIC), École Polytechnique Fédérale de Lausanne, CH-1950 Sion, Switzerland.
e-mail: kyriakos.stylianou@epfl.ch and berend.smit@epfl.ch.
Abstract: Metal nanoparticles (NPs) are usually stabilized by a
capping agent, a surfactant, or a support material, to maintain their
integrity. However, these strategies can impact their intrinsic catalytic
activity. Here, we demonstrate that the in-situ formation of copper NPs
(Cu0NPs) upon the reduction of the earth-abundant Jacquesdietrichite
mineral with ammonia borane (NH3BH3, AB) can provide an
alternative solution for stability issues. During the formation of Cu0NPs,
hydrogen gas is released from AB, and utilized for the reduction of
nitroarenes to their corresponding anilines, at room temperature and
under ambient pressure. After the nitroarene-to-aniline conversion is
completed, regeneration of the mineral occurs upon the exposure of
Cu0NPs to air. Thus, the hydrogenation reaction can be performed
multiple times without the loss of the Cu0NPs’ activity. As a proof-of-
concept, the hydrogenation of drug molecules “flutamide” and
“nimesulide” was also performed and isolated their corresponding
amino-compounds in high selectivity and yield.
conditions, which might have a detrimental effect on the stability
of Cu0NPs over several catalytic cycles.[12] In case of using
supporting matrices, obtaining a uniform distribution of the
Cu0NPs is challenging, and the aggregation of NPs might still
occur before or during the catalytic cycles.[10] Similarly, creating
core-shell structures of Cu0NPs with other metals might not
prevent the catalytic deactivation.[13] Therefore, a fundamentally
different strategy is needed to address the stability challenges that
Cu0NP catalysts encounter. Our strategy is based on the notion
that any attempt to stabilize Cu0NP catalysts may change their
intrinsic catalytic activity. We therefore explore an approach, in
which we aim to develop materials that generate these
nanoparticles in-situ.
Recently, we have shown that the synthetic form of the
Jacquesdietrichite mineral with the chemical formula of
(Cu2[(BO)(OH)2](OH)3[14] generates Cu0NPs in-situ as an integral
part of the hydrolytic dehydrogenation of ammonia borane
(AB).[15] AB can be considered as an alternative high quality H2
source since it possesses a high H content of 19.6 wt% and is
robust, non-toxic and highly soluble in aqueous solutions. By the
hydrolysis of AB, hydrogen release occurs at room temperature
and under ambient pressure, eliminating the safety issues and the
poor solubility problem of H2 gas in water.[16] We demonstrated
that after the hydrogen generation from AB is complete, fresh and
highly crystalline mineral microcrystals are regenerated from
Cu0NPs in a quantitative yield. Therefore, AB hydrolysis was
performed for at least 10 cycles with negligible decrease of the
catalyst’s activity. This result is inspiring since the fresh Cu0NPs
can be generated from the synthetic mineral when needed,
preventing them from aggregation, oxidation, and deposition. This
is an attractive strategy of limiting the deactivation problems of
Cu0NPs and the difficulty in their handling and storage.
In this work, we show that our strategy can be generalized to
the hydrogenation of nitrobenzenes to anilines, by utilizing the
hydrogen generated from AB and using the in-situ generated
Cu0NPs. Anilines are key intermediates in the production of dyes,
pharmaceuticals, and fragrances. In industry, the hydrogenation
of nitrobenzene to aniline is conducted in the gas phase with
copper or nickel catalysts at high temperatures of 553–573 K and
pressures of H2 of 0.1–0.5 MPa.[17] However, in some cases of
nitroarene reduction under the conditions used in industry, Cu
catalysts are exposed to a comparatively fast deactivation, and
reactivation is required.[18] In the literature, other noble and non-
noble metal nanoparticles have been also investigated for the
hydrogenation of nitroarene compounds. Since copper is an
earth-abundant and one of the cheapest non-noble metals,
Cu0NPs have been an attractive alternative for this catalytic
process. In our study, we propose a strategy to protect Cu0NPs
Introduction
Recent advances in our understanding of heterogeneous
catalysis is that, although a catalyst keeps its stoichiometry, it can
change shape, size and structure during the reaction.[1] In fact,
one can argue that the most active catalysts can exhibit variety of
structural alterations during reactions due to their high surface
energy, leading to a lower energy of the transition state and hence
higher reaction rate.[2] Therefore, it can be deduced that the most
active catalysts encounter with the instability problems more
frequently.[2] Copper nanoparticles (Cu0NPs) are reported to be
among the most active catalysts for a wide range of catalytic
reactions, including water-gas shift,[3] hydrogenations,[4] organic
transformations such as cycloaddition[5] and cross coupling,[6] as
well as photocatalytic reactions.[7] Despite the versatility of
catalytic reactions with Cu0NPs, their low stability has been a
great challenge, and this is the critical drawback for industrial
applications.[8]
Cu0NPs can be deactivated due to many different
mechanisms, such as sintering and aggregation, oxidation,
deposition of reactants or products covering the surface, or
poisoning of the catalyst. Many strategies have been developed
to stabilize Cu0NPs such as the employment of capping agents
(e.g., polymers and surfactants)[9] and supports (e.g., oxides,
polymers and zeolites)[10] or the generation of core-shell
structures with other metals (e.g., Ag, Ni, and Pd).[11] These
strategies, however, often only address one of the issues
mentioned above and they are short term solutions. For example,
using capping agents can effectively prevent the aggregation of
Cu0NPs; however, they can also act as a “poison” and limit the
accessibility of reactants to the catalytically active sites. Moreover,
they can be detached from the NPs under some reaction
1
This article is protected by copyright. All rights reserved.

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
10.1002/cctc.202000150
ChemCatChem
FULL PAPER
during the hydrogenation of nitrobenzenes to their corresponding
anilines from the most common catalyst deactivation problems
such as agglomeration and oxidation, without using any other
metal, capping agent or support material.
hydrogenation cycle can be attributed to the oxidation of the in-
situ generated Cu0NPs by their exposure to air, in combination
with the presence of BO2− ions, which is the product resulting from
the hydrolysis of ammonia borane in the reaction mixture.
-
Of particular interest of the nitro compounds are two drug
molecules, flutamide (a nonsteroidal antiandrogen drug which is
used primarily to treat prostate cancer)[19] and nimesulide (a
nonsteroidal anti-inflammatory drug for pain medication and fever
reducing),[20a] where the hydrogenation can be carried out to make
these drug molecules more compatible with the body.[20] For the
hydrogenation of these molecules, Jagadeesh et al.[21] developed
an earth-abundant catalyst, Fe2O3, which performs the reaction at
120 ºC with a hydrogen pressure of 50 bar in 15 hours. In
pharmaceutical industry, hydrogen gas is often used with a
flammable organic solvent, which increases the safety risk within
the manufacturing facility. Moreover, H2 gas has a limited
solubility in solvents (especially in water) and therefore there is a
need for high pressure for the utilization in hydrogenation
reactions.[22] Therefore, one would expect that the introduction of
highly active Cu0NPs, and the use of AB as the hydrogen source,
would allow these reactions to be performed at milder conditions
efficiently.
Therefore, the accumulation of BO2 ions in the reaction solution
over repeated cycles can be eliminated. Since the material is
regenerated in the pure and fresh form, the generation of new and
active Cu0NPs can catalyze the hydrogenation reaction without
any loss of their catalytic activity.
Exposed to air, in solution
Synthetic mineral
In-situ generation of Cu0NPs
NH3BH3
H2O/CH3OH
Herein, we present that the in-situ generation of Cu0NPs from
the synthetic mineral in the presence of AB was successfully used
to catalyze the reduction of nitroarene compounds to their
corresponding anilines by utilizing the hydrogen generated from
AB, with good selectivity and yields. The catalytic activity and the
integrity of the Cu0NPs remained almost unaffected for at least 5
cycles of reaction, which is due to the reformation of fresh mineral
at the end of each cycle. The reusability of the mineral for the
generation of fresh Cu0NPs in each cycle prevents the catalyst
poisoning and the use of stabilizing agents. The efficient reduction
of a wide range of nitro substrates, including drug molecules
flutamide and nimesulide, illustrates the versatility of our strategy.
Scheme 1. Schematic illustration of in-situ generation of Cu0NPs from the
synthetic mineral using ammonia borane (AB) as the reducing agent, with the
simultaneous conversion of nitroarenes to anilines by in-situ hydrogenation.
Fresh mineral can be regenerated at the end of the reaction upon the exposure
of reaction solution to air, resulting in the formation of fresh Cu0NPs for the next
cycle. Color code for mineral: oxygen in red, copper in green, boron in blue.
The as-synthesized synthetic mineral and in-situ generated
Cu0NPs were characterized by powder X-ray diffraction (PXRD)
(Figure 1a and 1c), showing that the experimental PXRD patterns
of both materials matched well with the simulated ones. The SEM
and TEM images obtained before and after the reaction confirm
the formation of Cu0NPs, as our synthetic mineral consists of
cuboid shaped single crystals (Figure 1b and S1a) while Cu0NPs
with a size ranging between 20-40 nm are the only present
compound after the reaction is complete (Figure 1d and S1b). The
X-ray photoelectron spectroscopy (XPS) data was collected on
as-synthesized mineral and in-situ generated Cu0NPs after the
hydrogenation reaction (Figure 1e and 1f). Cu 2p XPS spectrum
revealed that the synthetic mineral consisted of mostly Cu(II) (≥
95%) ions. The XPS spectrum of in-situ generated Cu0NPs shows
two components with binding energies at 932.5 and 934.5 eV
corresponding to the Cu 2p3/2 core level, which can be assigned
to Cu(0) and/or Cu(I) (932.5 eV), and Cu(II) (934.5 eV),
respectively.[24] Further information about the valence states of
copper were obtained by analyzing the Cu LMM Auger spectrum
of in-situ generated Cu0NPs obtained from XPS analysis (Figure
S2). LMM spectra shows the presence of both, Cu(II) and/or Cu(I),
as well as Cu(0). Together with the Cu 2p data analysis, it is
therefore clear that Cu(0) and Cu(II) are present. It is finally not
possible to exclude Cu(I). In order to give more clarity to the
findings of XPS, Auger electron spectroscopy was performed on
the sample (Figure S3). The findings from both XPS and Auger
electron microscopy analyses are in agreement with the PXRD
pattern of the bulk material, showing that the majority of the
sample consists of Cu(0). The observation of Cu(I) and Cu(II) is
expected because of the surface oxidation of Cu0NPs due to the
exposure of sample to air during the sample preparation.
Results and Discussion
Our strategy of in-situ generation of Cu0NPs is illustrated in
Scheme 1. The synthetic mineral generates Cu0NPs in water in
the presence of a reduction agent, ammonia borane (AB), which
is the source for hydrogen. The hydrolytic dehydrogenation of AB
is catalyzed by in-situ generated Cu0NPs and hydrogen is
released into the reaction solution. The solution also contains the
reactants for the hydrogenation reaction, and the conversion of
nitroarene to aniline takes place in parallel using the in-situ
generated hydrogen from AB. In our study, the formation of by-
products arising from ammonia borane hydrolysis was not
investigated since the unravelling of the reaction mechanism as
well as process optimization and design (e.g., regeneration of
spent ammonia borane) is hindered. The composition of B-
containing by-products is controversial, which could be attributed
to different assignments of 11B NMR and/or FTIR spectra for the
B-containing by-products.[23] Since undesired by-products such as
borazine (B3N3H6) can be liberated in the case of thermolysis of
ammonia borane, hydrolysis of ammonia borane was performed
for the release of hydrogen:
NH3BH3 + 2H2O à NH4+ + BO2- + 3H2
As we interpreted in our previous study,[15] the re-formation of
synthetic mineral (Cu2[(BO)(OH)2](OH)3) at the end of each
2
This article is protected by copyright. All rights reserved.

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
10.1002/cctc.202000150
ChemCatChem
FULL PAPER
while for 1-bromo-4-nitrobenzene, we obtained 40% yield after 2
hours also with 100% selectivity. The presence of the electron-
withdrawing Cl- and F- groups does not affect the reaction and the
aniline products were obtained in quantitative yields (Table 1,
entries 2-3). In case of Br- analogue (Table 1, entry 4), the yield is
40%, which might probably be due to the lower electronegativity of
Br compared to Cl and F; therefore, the nitro group is less electron
deficient. The reduction of the nitro group is not affected by the
presence of another nitro group, as shown for the case of 1,3-
dinitrobenzene. In fact, both nitro groups were efficiently reduced
to amines with 100% yield. Dell’Anna et al.[28] reported similar
findings, where the hydrogenation rate of 1-bromo-4-nitrobenzene
is slower (61% yield in 6 hours) than that of 1,3-dinitrobenzene
(96% yield in 6 hours). In that study, mono-amino products were
observed only in negligible quantities during the reaction. Similarly,
in our case, no mono-amino product was observed at the end of 1-
hour reaction as indicated by NMR analysis. Therefore, it can be
interpreted that the reduction of the intermediate product is fast,
whereas electronegativity difference makes the reaction rate
slower in the case of 1-bromo-4-nitrobenzene.
(b)
(a)
Synthetic mineral
(experimental)
Jacquesdietrichite
PDF 00-056-0054
2 µm
10
20
30
40
50
60
2q (deg)
(d)
(c)
Cu0NPs
(experimental)
Cu
PDF 00-004-0836
1 µm
10 20 30 40 50 60 70 80
2q (deg)
(e)
(f)
Spectrum
Background
Cu(I)
Spectrum
Background
Cu(II)
Table 1. Hydrogenation of various nitroarenes to anilines by the simultaneous
in-situ formation of Cu0NPs and dehydrogenation of NH3BH3
Cu(II)
Cu(0)
AB, catalyst
CH3OH/H2O (v:v = 1:9), RT
R-NO2
R-NH2
Entry
Substrate
Product
Yield Selectivity
Time
(%)
(%)
940 938 936 934 932 930 928
Binding energy (eV)
940 938 936 934 932 930 928
Binding energy (eV)
1
>90
>99
<1 h
Figure 1. (a) Comparison of the simulated PXRD pattern of Jacquesdietrichite
with the experimental PXRD pattern of our synthetic mineral. (b) SEM image of
our synthetic mineral. (c) Comparison of the simulated PXRD pattern of Cu0NPs,
with the PXRD pattern of in-situ generated Cu0NPs collected after the
hydrogenation reaction. (d) SEM image of in-situ generated Cu0NPs collected
after the hydrogenation reaction. Cu 2p XPS spectra of (e) as-synthesized
mineral, (f) in-situ generated Cu0NPs.
2
>99
>99
<1 h
To demonstrate the applicability of our strategy in
hydrogenation reactions, we first investigated the catalytic activity
of Cu0NPs toward the reduction of different nitroarenes as shown
in Table 1. These nitroarenes represent different classes and each
of these classes is used for the production of different
intermediates. The simplest reaction is the reduction of
nitrobenzene to aniline (Table 1, entry 1). The halogenated anilines
(Table 1, entries 2, 3 and 4) are commonly used intermediates in
the manufacture of herbicides, drugs and laboratory reagents[25]
and m-phenylenediamine (Table 1, entry 5) is used in the
manufacture of azo dyes.[26] 3-aminopyridine is an important
building blocks in pharmaceuticals and agrochemicals (Table 1,
entry 6).[27] All reactions were carried out at room temperature
using the optimized conditions (Figure S4). After combining all
constituents of the reaction, the solution was sampled after 1 or 2
hours depending on the completion of the reaction. The yield and
the selectivity of the products were determined by UV-Vis
spectroscopy (Figure S5-10) and 1H NMR analysis (Figure S11-
16). In the absence of the catalyst, no conversion was observed.
As shown in Table 1, except from 1-bromo-4-nitrobenzene and 3-
nitropyridine, all investigated nitroarene compounds were
converted within 1 hour with >90% yield and >99% selectivity. For
3-nitropyridine, we achieved a similar yield at the end of 2 hours
3
4
>99
~40
>99
>99
<1 h
<2 h
5
6
>99
>99
<1 h
~83
(100)
<1 h
(<2 h)
>99
Reaction conditions: 0.1 mmol nitro compound, 0.583 mmol AB, 12 mg catalyst,
1 ml MeOH, 9 ml H2O, 25 °C.
As the stability and recyclability of the catalyst are highly crucial for
practical applications, the integrity and catalytic performance of in-
situ generated Cu0NPs were investigated. Figure 2 shows the
recyclability of the catalyst during five consecutive runs. Within the
3
This article is protected by copyright. All rights reserved.