Heterogenized cobalt oxide catalysts for nitroarene reduction by pyrolysis of molecularly defined complexes

Article abstract of DOI:10.1038/nchem.1645

Molecularly well-defined homogeneous catalysts are known for a wide variety of chemical transformations. The effect of small changes in molecular structure can be studied in detail and used to optimize many processes. However, many industrial processes require heterogeneous catalysts because of their stability, ease of separation and recyclability, but these are more difficult to control on a molecular level. Here, we describe the conversion of homogeneous cobalt complexes into heterogeneous cobalt oxide catalysts via immobilization and pyrolysis on activated carbon. The catalysts thus produced are useful for the industrially important reduction of nitroarenes to anilines. The ligand indirectly controls the selectivity and activity of the recyclable catalyst and catalyst optimization can be performed at the level of the solution-phase precursor before conversion into the active heterogeneous catalyst.

Full text of DOI:10.1038/nchem.1645

ARTICLES
PUBLISHED ONLINE: 12 MAY 2013 | DOI: 10.1038/NCHEM.1645
Heterogenized cobalt oxide catalysts for
nitroarene reduction by pyrolysis of molecularly
defined complexes
Felix A. Westerhaus, Rajenahally V. Jagadeesh, Gerrit Wienh o¨ fer, Marga-Martina Pohl, J o¨ rg Radnik,
Annette-Enrica Surkus, Jabor Rabeah, Kathrin Junge, Henrik Junge, Martin Nielsen,
Angelika Br u¨ ckner and Matthias Beller
*
Molecularly well-defined homogeneous catalysts are known for a wide variety of chemical transformations. The effect of
small changes in molecular structure can be studied in detail and used to optimize many processes. However, many
industrial processes require heterogeneous catalysts because of their stability, ease of separation and recyclability, but
these are more difficult to control on a molecular level. Here, we describe the conversion of homogeneous cobalt
complexes into heterogeneous cobalt oxide catalysts via immobilization and pyrolysis on activated carbon. The catalysts
thus produced are useful for the industrially important reduction of nitroarenes to anilines. The ligand indirectly controls
the selectivity and activity of the recyclable catalyst and catalyst optimization can be performed at the level of the
solution-phase precursor before conversion into the active heterogeneous catalyst.
atalysis is a key technology for achieving more benign pro- active and selective catalysts, especially those based on non-
cesses in the chemical, pharmaceutical and material indus- noble metals.
1
C
tries . Owing to their stability, ease of separation and
Several anilines important for both organic synthesis and indus-
recycling, heterogeneous catalysts represent an important tool box try are densely substituted with a variety of different functional
for controlling chemical reactivity. Today, the preparation of bulk groups. Hence, for a generally applicable catalyst system it is essen-
and impregnated catalysts as well as of mixed agglomerated tial to perform the reduction of the nitro group highly chemoselec-
catalysts relies mainly on classic methods, such as hydrothermal tively. Notably, such functional-group tolerance might also omit the
2
,3
transformations, precipitation, calcination and impregnation . use of protecting groups and streamline more-complex syntheses.
More specialized methods for catalyst preparation include resin- Especially with respect to selectivity, the reduction of 2NO in
2
bound complexes or complexes bound by a supported ionic liquid the presence of other reducible substituents, such as ketones, alde-
1
2,16–28
phase. In addition, complex organometallic precursors are used hydes, alkenes or alkynes, remains challenging
. Most
for the synthesis of heterogeneous materials, for example via common commercially available Raney–nickel, palladium or plati-
metal–organic chemical vapour deposition4 . Despite all these num catalysts are troubled by these issues. Nevertheless, Siegrist,
attempts, the use of defined non-volatile organometallic Blaser and co-workers achieved the selective reduction of substi-
complexes as precursors for stable heterogeneous catalysts has tuted nitroarenes using commercially available platinum catalysts
–6
been scarcely investigated7
–10
.
modified by special additives
29–33
. More recently, another important
Herein, we report a protocol to prepare novel nanoscale cobalt development came from the group of Corma, who disclosed a het-
oxide catalysts supported on a carbon–nitrogen surface via pyrolysis erogeneous gold-based catalyst that can hydrogenate nitro com-
of defined non-volatile organometallic amine complexes (Table 1). pounds selectively in the presence of a number of functional
3
4–36
Depending on the organic ligands used, modification of the groups
. Nevertheless, so far there exists no inexpensive non-
surface takes place and remarkable differences in catalyst activity noble metal catalyst system that allows for a general reduction of
are observed. Thereby, the ligand indirectly controls the activity all kinds of nitroarenes.
and selectivity of the given reaction (Table 1).
With more than 4,000,000 tons produced per year, aniline and its Results and discussion
derivatives represent important feedstocks in the chemical industry For some time, we have been interested in developing molecularly
(
www.strategyr.com/Aniline_Market_Report.asp). They are used defined organometallic catalysts for the selective reduction of nitroar-
3
7
widely for the synthesis of dyes, agrochemicals, pharmaceuticals, enes . Based on this work, we started a programme to develop similar
polymers and various other fine chemicals1
,11,12
. The common heterogeneous systems. More specifically, we had the idea to prepare
route for their preparation is the reduction of the corresponding novel nanostructured catalysts via deposition of defined organometal-
nitro derivatives (which can be carried out using stoichiometric lic complexes onto commercially available inorganic supports with
reducing agents such as Fe, Zn, Sn, Al and sulfur compounds), elec- subsequent pyrolysis. According to this concept, ultimately the for-
trochemical methods or catalytic hydrogenation1
,11,13–15
. Obviously, mation and size of the metal particles should be controlled by the
the most environmentally benign and cost-effective reducing agent precatalyst and the surrounding ligands. Notably, the plethora of
is hydrogen. Despite numerous developments of hydrogenation cat- organic ligands available adds another dimension of possibilities
alysts, there still exists a significant interest in novel inexpensive, for the generation of new heterogeneous catalytic materials.
Leibniz-Institut f u¨ r Katalyse e.V. an der Universit a¨ t Rostock, Albert-Einstein Straße 29a, Rostock, 18059, Germany. *e-mail: matthias.beller@catalysis.de
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
2013 Macmillan Publishers Limited. All rights reserved.
1
©

NATURE CHEMISTRY DOI: 10.1038/NCHEM.1645
ARTICLES
The exact preparation of the catalyst material is crucial to gener-
ate highly active species. For example, optimization of the pyrolysis
step showed that the temperature gradient and the heating time at
the maximum temperature significantly influence the activity and
selectivity of the catalyst. The best system (see below) was obtained
by performing the calcinations at 800 8C for two hours.
Table 1 | Hydrogenation of nitrobenzene: comparison of
selected supported cobalt oxide catalysts.
NO2
NH2
Catalyst (1 mol%)
H2 (50 bar), 110 °C
THF/H2O, 4 h
1
2
All novel catalytic materials were tested in the model reaction at
110 8C under hydrogen (50 bar). The plain cobalt oxide/carbon cat-
alyst and materials in which the organic ligands were immobilized
on carbon material revealed no product at all (Table 1, entries 1–3).
On pyrolysis of cobalt nitrogen complexes on carbon the catalytic
activity is improved (Table 1, entries 4–6). Among the different
amine ligands tested, phenanthroline (L1) was found to give the
most-active system (Table 1, entry 4).
N
N
N
N
N
N
N
3
, L1, 1,10-phenanthroline
4, L2, 2,2':6',2"-terpyridine
6, L4, 2,2'-bipyridine
H
N
It is important that this catalyst exhibited superior reactivity even
in comparison with similar nitrogen ligands such as terpyridine
N
N
NH
N
(L2) or bis(benzimidazolyl)pyridine (L3) (Table 1, entries 4–6).
N
Hence, comparably small changes in the organic ligand cause
extensive loss of reactivity and selectivity. The model reaction
does not occur to any extent in the presence of the homogeneous
precatalyst (Table 1, entry 10), even when using a tenfold amount
of cobalt(II) acetate and phenanthroline. In addition, a related
iron-based catalyst showed insignificant formation of aniline
(Table 1, entry 9).
5
, L3, 2,6-bis(2-benzimidazolyl)pyridine
7, L5, pyridine
Entry
Catalyst
Conversion (%)*
Yield (%)*
†
†
†
1
Co/carbon
L1/carbon
Carbon
5
1
6
0
0
0
2
3
4
5
6
7
8
9
‡
§
ꢀ
Co-L1/carbon
.99
8
26
,1
,1
15
95/99
Co-L2/carbon
Co-L3/carbon
Co-L4/carbon
Co-L5/carbon
Fe-L1/carbon
5
19
0
Catalyst characterization. To understand the nature of the active
metal species, electron paramagnetic resonance (EPR)
measurements of the supported pyrolysed cobalt catalyst were
performed. The different EPR spectra were recorded in the
0
†
1
}
1
1
0
1
Co-L1
Ni-L1
4
0
0
}
,1
X-band at 88 K and 290 K on
microspectrometer equipped with an ER 4119HS-WI high-
sensitivity cavity and variable-temperature control unit
microwave power 6.64 mW, modulation frequency 100 kHz,
a
Bruker EMX CW-
L, ligand. Reaction conditions 110 8C, 4 h, 0.5 mmol nitrobenzene, 1 mol% catalyst (3 wt% Co–L/C),
0 bar hydrogen, 2 ml THF, 100 ml H O. *Determined by GC using n-hexadecane as an internal
5
2
a
†
‡
§
ꢀ
2
standard. Reaction time 16 h. Pyrolysed carbon. Metal/ligand ratio 1:2. Conducted in H O
(
}
(
3 ml). Homogeneous catalyst.
modulation amplitude 1 G). The EPR spectrum of the active
catalyst contains a broad signal at g ¼ 2.12, the intensity of which
Catalyst synthesis and catalytic experiments. In a preliminary increases slightly with rising temperature (Fig. 2a). This is
catalyst screening, biorelevant and abundant metal–amine characteristic for antiferromagnetic Co O4 particles. The Neel
3
complexes based on iron, cobalt, manganese and nickel pyrolysed temperature, T (above which bulk Co O becomes paramagnetic),
N
3 4
on carbon were tested for the industrially important benchmark is below 40 K (refs 38,39). Thus, according to the Curie–Weiss
of hydrogenation of nitrobenzene. Among the different metals law, a decrease of the magnetic susceptibility and, consequently,
used, cobalt(II)–amine acetate complexes showed significant also of the EPR signal intensity would be expected on increasing
activity. This was surprising to us as cobalt catalysts are scarcely the temperature from 88 to 290 K, which is not the case in
used for such transformation2 . Therefore, we prepared a Fig. 2a. However, it was found that the temperature dependence
variety of catalysts from different cobalt precursors and nitrogen of the EPR signal intensity reflected very sensitively the onset
9–31
ligands (Table 1) by adsorbing a solution of the respective cobalt of antiferromagnetic ordering in Co O4 and also in other
3
complex onto commercially available carbon (Vulcan XC72R) and antiferromagnetic oxide materials already well above TN
subsequent pyrolysis under inert conditions (Fig. 1). In doing so, (ref. 40). In bulk Co O , the EPR intensity increased gradually
3
4
cobalt oxide particles were expected to form that interact with the up to 150 K and then remained constant up to 250 K before it
nitrogen-enriched carbon surface.
started to decrease.
2+
Co(OAc)2 * 4H2O
+
N
Vulcan XC72R
0 °C, 4 h
800 °C, 2 h
Ar
r.t., 30 min
–
Co(Phen)2(OAc)2
adsorbed on carbon
N
N
2
( OAc)
Co3O4
Co
6
EtOH
N
N
N
N
N
N
Schematic representation of
the active Co oxide–N/C catalyst
Figure 1 | Synthesis of an active cobalt oxide–nitrogen/carbon catalyst by pyrolysing a cobalt(II) acetate–phenanthroline complex on carbon.
r.t. ¼ room temperature.
2
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
©
2013 Macmillan Publishers Limited. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.1645
ARTICLES
a
×106
2
1
×106
0
g = 2.12
–
1×106
2×106
T = 88 K
T = 290 K
–
0
1,000 2,000 3,000 4,000 5,000 6,000 7,000
Bo (G)
b
0.5 µm
50 nm
20 nm
007
1
,500
,350
,200
,050
1
1
1
9
7
6
4
3
1
00
50
00
50
00
50
0
20 nm
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00
Energy (keV)
c
d
NCoN
Npyrrol
3.12 nm
Namm
20 nm
20 nm
410
405
400
395
390
Electron binding energy (eV)
Figure 2 | Catalyst characterization. a, EPR spectra of the active cobalt catalyst recorded at 88 and 290 K. b, HAADF images and EDXS (bottom right) of
Co O –L1/C. c, HAADF images of Co O –L2/C. d, XPS N1s spectrum of Co O –L1/C. The different N1s states are labelled as N for nitrogen bound to
3
4
3
4
3
4
CoN
cobalt, Npyrrol for pyrrolic N and Namm for nitrogen of quaternary amine species. a.u. ¼ arbitrary units.
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
3
©
2013 Macmillan Publishers Limited. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.1645
ARTICLES
The observed intensity behaviour in Fig. 2a is exactly in line with
the previous results and also with the fact that a considerable
fraction of rather large particles (20–80 nm) was observed by trans-
mission electron microscopy (TEM) (see below). The EPR signal at
g ¼ 2.12 is superimposed on a second very broad anisotropic signal,
the positive lobe of which is cut off at B ¼ 0. This suggests that the
magnetic properties are not exactly the same for each particle.
The reason may be different particle sizes (see below) and/or
replacement of oxygen by nitrogen in the coordination of cobalt
a
100
8
6
0
0
40
(as observed by X-ray photoelectron spectroscopy (XPS), see
below). The presence of metallic cobalt and/or cobalt oxide can
be ruled out because their magnetic properties are not in
line with the observed EPR intensity behaviour. Indeed, cobalt
is ferromagnetic but cobalt oxide is antiferromagnetic, with
2
0
Reaction time 12 h
Reaction time 3 h
0
0
1
2
3
4
5
6
7
8
9
10
T ≈ 293 K (ref. 41).
N
Number of runs
To understand in more detail the remarkable activity difference
of the carbon-supported Co O –L1 material compared to that of
the other catalysts, scanning transmission electron microscopy
b
3
4
1
00
80
60
100
80
60
40
20
0
investigations of Co O –L1/C and Co O –L2/C were conducted.
3
4
3 4
To image the complete size spectra of cobalt oxide particles, a
high-angle annular dark field (HAADF) at a spherical aberration
(Cs)-corrected microscope was used. With a conventional TEM it
was not possible to image the smallest particles because of the
weak contrast.
4
2
0
0
0
As shown in Fig. 2b,c, for the different materials the cobalt oxide
particle size varies significantly. To our surprise, in the active cata-
lyst system Co O –L1/C the particle sizes had a wide size distri-
3
4
bution with a fraction of particles of 2–10 nm, and particles and
agglomerates in the range 20–80 nm. Occasionally, even larger
structures up to 800 nm were present. As shown by energy-disper-
sive X-ray spectroscopy (EDXS) analysis, both the very small par-
ticles and the larger ones contained cobalt. Also, we observed
some cloudy structures with weak contrast even in the HAADF
0
50
100
150
200
250
300
Time (min)
Figure 3 | Catalyst recycling and kinetic data of the hydrogenation of
nitrobenzene catalysed by cobalt oxide. a, Recycling experiments over ten
runs with different reaction times (three and 12 hours). b, Kinetic profile of
the reduction of nitrobenzene to aniline using the cobalt catalyst. Reaction
conditions: 110 8C, 0.5 mmol nitrobenzene, 1 mol% catalyst (3 wt% Co-L1/C),
(
Fig. 2b) that contained Co O as well. However, when the less-reac-
3 4
tive L2–Co(II) acetate complex adsorbed onto carbon was analysed
Fig. 2c), the size of the well-defined particles rarely exceeded the
(
range 2–10 nm. In the latter case, larger elements were not
present. In addition, cloudy cobalt-containing structures with
weak contrast were absent. Although in most catalytic reactions
smaller nanoparticles exhibit higher activity, apparently in this cat-
alytic reaction the best performance is achieved with larger particles,
50 bar hydrogen, 2 ml THF, 100 ml H O. Conversion and yields were
2
determined by gas chromatography (GC) using n-hexadecane as an internal
standard. The error bars were calculated using Origin 8.
although we cannot rule out the activity of the observed cloudy
cobalt structures4
2–44
.
was evaluated after three hours an initial loss of activity was
To obtain further insight into the structure of the catalyst and observed. After the fifth run the activity started to level off and
especially the role of nitrogen from the organic ligand, XPS investi- yields remained at around 30% until the tenth run (Fig. 3a). The
gations on the bonding of nitrogen and cobalt were carried out. kinetic profile of the reaction reveals that nitrobenzene was
Interestingly, three distinct peaks are observed in the N1s spectra converted smoothly into aniline without a significant formation of
of Co O –L1/C with an electron-binding energy of 399.0 eV, dangerous intermediates, for example hydroxylamine, being
3
4
400.8 eV and 402.3 eV (Fig. 2d). The lowest binding-energy peak detected (Fig. 3b). Interestingly, the reaction rate of the
can be attributed to a pyridine-type nitrogen, which is bound to a hydrogenation is dependent on the amount of water present.
45
metal ion . The electron-binding energy of 400.8 eV is character- When we used dry organic solvents, for example tetrahydrofuran
istic of a pyrrole-type nitrogen that contributes two electrons to (THF), the reaction times had to be prolonged to achieve full
the carbon matrix. It is bound to a hydrogen atom. Such types of conversion. This stimulated us to perform the reaction simply ‘on
nitrogen are found after the carbonization of nitrogen-containing water’ without any additional organic solvent present. To our
46
organic materials . Finally, the small peak at 402.3 eV is typical delight, the catalyst activity was even higher in pure water
for quaternary amine species (NR4 ) . The ratio between all compared to that in THF48,49.
þ
47
cobalt atoms and all nitrogen atoms in the near-surface region
Next, the general scope of our supported cobalt catalyst was
was 1:4.7. Deconvolution revealed that around 64% of all nitrogen demonstrated in the reduction of more than 30 functionalized
atoms are bound to the metal ions and therefore, statistically, nitroarenes (some of these compounds are toxic, for example
three nitrogen atoms are bound to one cobalt atom.
entry 20 in Table 2, and should be handled with appropriate
caution). As shown in Table 2, a variety of industrially relevant ani-
Catalytic experiments. As shown in Fig. 3 the catalyst is recycled up lines such as chloro- and fluoroanilines were prepared in good to
to ten times with little loss of activity, which shows the potential excellent yields (83–99% (Table 2, entries 3–10)). Even sterically
applicability of this system. Selectivity and yield of aniline demanding nitroarenes (Table 2, entries 14 and 15) and substrates
remained constant through all cycles when the reaction was run bearing more-labile bromide substituents or sulfur were hydrogen-
for 12 hours. However, when the activity of the reused catalyst ated smoothly (Table 2, entries 9 and 13).
4
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
2013 Macmillan Publishers Limited. All rights reserved.
©

NATURE CHEMISTRY DOI: 10.1038/NCHEM.1645
ARTICLES
Table2 | Cobalt-catalysed hydrogenation of substituted nitroarenes.
Entry Substrate
Time (h) Yield (%)*
Entry Substrate
Time (h) Yield (%)* Entry Substrate
Time (h) Yield (%)*
NO2
†
NO2
H
N
1
4
91/93
10
6
83
19
4
78
F
8
Cl
NO2
25
17
NO2
†
NO2
O N
NO2
Me
†,‡,§
2
2
4
99/99
11
6
93
78
20
12
83
F3C
1
8
26
9
NO2
NO2
NO2
NO2
NO2
NO2
†
3
4
5
6
6
6
92
95
12
13
14
6
6
21
22
22
12
12
12
85
F
O
S
O
O
1
2
9
0
27
28
10
†
NO2
‡
85/83
53
Cl
Cl
N
1
1
1
2
NO2
†
95/93†
12
96
58
NO2
N
Br
29
30
21
NO2
Cl
ꢀ
NO2
NO2
†
NO2
†
S
N
6
7
4
6
99/99
15
16
12
4
97/99
24
25
12
12
75
22
13
Cl
NO2
97
99
73
N
NO2
Cl
31
23
14
Cl
Cl
NO2
†
§
NO2
NO2
8
9
6
6
93/96 /98
17
18
12
91
26
27
12
4
73
89
H N
2
1
Cl
O
3
2
15
NO2
§
NO2
CF3
†
NO2
94/98
4/12
78/88
MeO
Br
Br
24
16
O
33
Reaction conditions: 110 8C, 0.5 mmol nitroarene, 1 mol% catalyst (3 wt% Co-L1/C), 50 bar hydrogen, 2 ml THF, 100
m
l H
2
O; .99% conversion observed in all cases. *Determined by GC using n-hexadecane as
†
‡
§
ꢀ
an internal standard. Conducted in H
2
O (3 ml). Reduction selectively yields the diamine. Isolated yield (5 mmol scale). Isolated yield (50 mmol scale), 0.2 mol% catalyst, 48 h, 10 ml THF, 1 ml H
2
O.
Moreover, both electron-deficient substituents, such as trifluoro- to reduce nitro groups in the presence of alkenes, but also alkynes
methyl, and electron-rich groups, for example methoxy and amino, with full conversion and without any detectable concurrent
were tolerated well. In addition to anilines, heteroaromatic amines reduction of the unsaturated unit (88–96% (Table 3, entries 2–4)).
are valuable building blocks for the preparation of a variety of Compared to well-known noble metal catalysts (Supplementary
50
agrochemicals and pharmaceuticals . Therefore, five different nitro- Table S6), clearly our novel catalyst performs in a superior
heteroarenes were subjected to the standard hydrogenation con- manner. For example, applying nitrosubstituted aryl iodides in
ditions. Without further optimization all the substrates were reduced general facile dehalogenation is observed with current commercially
in moderate-to-good yields (53–85% (Table 2, entries 21–25)).
available catalysts based on precious metal, but using the cobalt-
Clearly, the most-challenging substrates are those that bear other based catalyst the reaction proceeds efficiently in 92% yield
easily reducible moieties (Table 3). In this respect, the supported (Table 3, entry 7).
gold catalyst from Corma and Serna showed impressive chemoselec-
To our delight, substrates that contained sensitive aldehyde or
34
tivity, even in the presence of olefinic bonds . However, even with keto substituents were also converted smoothly and selectively
the gold catalyst no examples with demanding substituents, such as into the corresponding anilines (79–97% (Table 3, entries 5 and 6)).
alkynes, iodides or heteroarenes, were reported. In this respect, we Furthermore, carboxylic acid derivatives, such as nitriles, were
were pleased to find that the novel cobalt catalyst is not only able tolerated well.
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
2013 Macmillan Publishers Limited. All rights reserved.
5
©

NATURE CHEMISTRY DOI: 10.1038/NCHEM.1645
ARTICLES
Received 5 April 2012; accepted 22 March 2013;
published online 12 May 2013
Table 3 | Selective cobalt-catalysed hydrogenation of
nitroarenes that bear easy-to-reduce functional groups.
Entry
Substrate
Time (h)
Yield (%)*
References
1.
Vogt, P. F. & Gerulis, J. J. in Ullmann’s Encyclopedia of Industrial Chemistry.
http://dx.doi.org/10.1002/14356007.a02_037 (Wiley-VCH, 2000).
NO2
1
4
83
2.
Ertl, G., Kn o¨ zinger H. & Weitkamp, J. Handbook of Heterogeneous Catalysis
(
Wiley-VCH, 1997).
N
3
4
5
3. Baerns M. Basic Principles in Applied Catalysis (Springer, 2004).
4
.
Jones, A. C. & Hitchman, M. L. Chemical Vapour Deposition: Precursors,
Processes and Applications 1–36 (The Royal Society of Chemistry, 2009).
5. Serp, P. & Kalck, P. Chemical vapor deposition methods for the controlled
NO2
2
3
4
5
6
6
4
4
91
preparation of supported catalytic materials. Chem. Rev. 102, 3085–3128 (2002).
George, S. M. Atomic layer deposition: an overview. Chem. Rev. 110,
111–131 (2010).
Lef e´ vre, M., Proietti, E., Jaouen, F. & Dodelet, J-P. Iron-based catalysts with
improved oxygen reduction activity in polymer electrolyte fuel cells. Science 324,
71–74 (2009).
Lara, P. et al. Ruthenium nanoparticles stabilized by N-heterocyclic carbenes:
ligand location and influence on reactivity. Angew. Chem. Int. Ed. 50,
12080–12084 (2011).
6
.
.
3
7
NO2
†
96/97
8
9
1
1
.
.
Ph
3
6
Bronger, R. et al. Multi-site coordination N-phosphanylamidine ligands as
stabilizers for the synthesis of ruthenium nanoparticles. New J. Chem. 35,
2653–2660 (2011).
0. Guerrero, M. et al. Design of new N,O hybrid pyrazole derived ligands and
their use as stabilizers for the synthesis of Pd-nanoparticles. Langmiur 26,
15532–15540 (2010).
1. Downing, R. S., Kunkeler, P. J. & van Bekkum, H. Catalytic syntheses of aromatic
amines. Catal. Today 37, 121–136 (1997).
12. Ono, N. The Nitro Group in Organic Synthesis (Wiley-VCH, 2001).
NO2
‡
88
37
NO2
97
79
1
3. Suchy, M., Winternitz, P. & Zeller, M. Heterocyclic compounds. WO patent
1/00278 (1991).
O
9
1
4. Butera, J. & Bagli, J. n-Heteroaralkyl-substituted 1-aryloxy-2-propanolamine
and propylamine derivatives processing class III antiarrythmic activity.
WO patent 91/09023 (1991).
15. Kolvar, R. F. & Armond, E. F. Ethynyl-substituted aromatic ortho diamines and
method of synthesis. US patent 3,975,444 (1976).
6. Reis, P. M. & Royo, B. Chemoselective hydrogenation of nitroarenes and
deoxygenation of pyridine N-oxides with H catalyzed by MoO Cl . Tetrahedron
Lett. 50, 949–952 (2009).
7. Liu, L., Qiao, B., Chen, Z., Zhang, J. & Deng, Y. Novel chemoselective
hydrogenation of aromatic nitro compounds over ferric hydroxide supported
H
8
3
NO2
6
7
4
6
1
2
2
2
O
39
1
I
NO2
†
92/99
nanocluster gold in the presence of CO and H O. Chem. Commun.
2
653–655 (2009).
1
8. Cao, S., Xu, S. & Xu, S. Hydrogenation of nitroaromatics containing a carbonyl
group catalyzed by the palladium complex of MgO-supported melamino-
formaldehyde polymer. Polym. Adv. Technol. 10, 43–47 (1999).
40
19. Boix, C. & Poliakoff, M. Selective reductions of nitroarenes to anilines
using metallic zinc in near-critical water. J. Chem. Soc. Perkin Trans. 1,
Reaction conditions: 110 8C, 0.5 mmol nitroarene, 1 mol% catalyst (3 wt% Co-L1/C), 50 bar
hydrogen, 2 ml THF, 100 l H O; .99% conversion observed in all cases. *Determined by GC
using n-hexadecane as an internal standard. Isolated yield (5 mmol scale). 2 mol% cobalt catalyst.
1
487–1490 (1999).
m
2
†
‡
20. Cardenas-Lizana, F. et al. Pd-promoted selective gas phase hydrogenation of
p-chloronitrobenzene over alumina supported Au. J. Catal. 262, 235–243 (2009).
21. Cardenas-Lizana, F., Gomez-Quero S. & Keane, M. A. Exclusive production of
chloroaniline from chloronitrobenzene over Au/TiO and Au/Al O .
ChemSusChem 1, 215–221 (2008).
2
2 3
Summary and conclusion
We have developed novel cobalt-based catalysts for the hydrogen- 22. Ichikawa, S., Tada, M., Iwasawab, Y. & Ikariya, T. The role of carbon dioxide in
chemoselective hydrogenation of halonitroaromatics over supported noble metal
catalysts in supercritical carbon dioxide. Chem. Commun. 924–926 (2005).
3. Knifton, J. F. Homogenous catalyzed reduction of nitro-compounds for
selective and sequential hydrogenation of nitroaromatics. J. Org. Chem. 41,
1200–1206 (1976).
ation of nitroarenes into the corresponding anilines. The most-
active catalyst is prepared via adsorption of a cobalt–phenanthroline
complex onto a commercially available carbon support and sub-
sequent pyrolysis. Depending on the amine ligands used, different
2
cobalt particles form on a carbon–nitrogen surface. Although the 24. Joshi, R. & Chudasama, U. Hydrogenation and oxidation reactions involving
ruthenium supported catalysts. Ind. Eng. Chem. Res. 49, 2543–2547 (2010).
control of the active metal centre by organic ligands is well estab-
lished in homogeneous catalysis51,52, such concepts are less investi-
25. Deshmukh, A. A., Prashar, A. K., Kinage, A. K., Kumar, R. & Meijboom, R.
Ru(II)– phenanthroline complex as catalyst for chemoselective hydrogenation of
nitro-aryls in a green process. Ind. Eng. Chem. Res. 49, 12180–12184 (2010).
gated in heterogeneous catalysis. In contrast to known examples of
‘
ship-in-the-bottle’ surface organometallic or immobilized homo- 26. Savoia, D., Trombini, C., Umani-Ronchi, A. & Verardo, G. Active metals from
geneous catalysts, in our approach the overall catalyst performance
potassium–graphite. Palladium–graphite as catalyst in the hydrogenation of
is controlled indirectly by the ligand5
3–55
nitro-compounds, alkenes, and alkynes. J. Chem. Soc. Chem. Comm.
.
540–541 (1981).
The optimal catalyst system constitutes one of the most general,
functional-group tolerant, inexpensive and environmentally benign
27. Tafesh, A. M. & Beller, M. First selective reduction of aromatic nitro compounds
using water soluble catalysts. Tetrahedron Lett. 36, 9305–9308 (1995).
catalysts known to date for the industrially important reduction of 28. Onopchenko, A., Sabourin, E. T. & Selwitz, C. M. Selective catalytic-
nitroarenes. It tolerates numerous other reducible functional
groups and can be reused several times without loss of activity.
Advantageously, during catalysis protective gas techniques or dry
solvents are not necessary and reactions can be performed at an
increased rate on water.
hydrogenation of aromatic nitro-groups in the presence of acetylenes – synthesis
of (3-aminophenyl)acetylene via hydrogenation of dimethycarbinol substituted
(
3-nitrophenyl)acetylene over heterogeneous metallic ruthenium catalysts.
J. Org. Chem. 44, 1233–1236 (1979).
9. Siegrist, U., Baumeister, P. & Blaser, H-U. in Catalysis of Organic Reactions (ed.
Herkes, F.) 207–219 (Chemical Industries Series 75, Marcel Dekker, 1998).
2
6
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
2013 Macmillan Publishers Limited. All rights reserved.
©

NATURE CHEMISTRY DOI: 10.1038/NCHEM.1645
ARTICLES
3
3
3
0. Raja, R. et al. Highly efficient catalysts for the hydrogenation of nitro-substituted
aromatics. Chem. Commun. 2026–2028 (2005).
1. Braden, R., Knupfer H. & Hartung, S. Process for the preparation of unsaturated
amino compounds. US patent 4,002,673 (1977).
2. Blaser, H-U., Siegrist, U., Steiner, H. & Studer, M. in Aromatic Nitro Compounds:
Fine Chemicals through Heterogeneous Catalysis (eds Sheldon, R. A. &
van Bekkum, H.) 389–406 (Wiley-VCH, 2001).
47. Gr u¨ nert, W. et al. A new facility for inert transfer of reactive samples to XPS
equipment. J. Electron. Spectrosc. Relat. Phenom. 40, 187–192 (1986).
48. Simon, M. O. & Li, C. J. Green chemistry oriented organic synthesis in water.
Chem. Soc. Rev. 4, 1415–1427 (2012).
49. de Cienfuegos, L. A., Robles, R., Miguel, D., Justicia, J. & Cuerva, J. M. Reduction
reactions in green solvents: water, supercritical carbon dioxide, and ionic liquids.
ChemSusChem 8, 1035–1048 (2011).
3
3
3
3. Blaser, H-U., Steiner, H. & Studer, M. Selective catalytic hydrogenation of
functionalized nitroarenes: an update. ChemCatChem 1, 210–221 (2009).
4. Corma, A. & Serna, P. Chemoselective hydrogenation of nitro compounds with
supported gold catalysts. Science 313, 332–334 (2006).
5. Corma, A., Gon a´ lez-Arellano, C., Iglesias, M. & S a´ nchez, F. Gold complexes as
catalysts: chemoselective hydrogenation of nitroarenes. Appl. Catal. A 356,
50. Marson, C. M. New and unusual scaffolds in medicinal chemistry. Chem. Soc.
Rev. 40, 5514–5533 (2011).
51. Cornils, B. & Herrmann, W. A. Applied Homogeneous Catalysis with
Organometallic Compounds (Wiley-VCH, 2002).
52. Colman, J. P., Hegedus, L. S., Norton, J. R. & Finke, R. G. Principles and
Applications of Organotransition Metal Chemistry (University Science
Books, 1987).
9
9–102 (2009).
3
6. Corma, A., Serna, P., Concepci o´ n, P. & Calvino, J. Transforming nonselective
into chemoselective metal catalysts for the hydrogenation of substituted
nitroaromatics. J. Am. Chem. Soc. 130, 8748–8753 (2008).
53. Corma, A. & Garcia, H. Engineering metal organic frameworks for
heterogeneous catalysis. Chem. Rev. 8, 4606–4655 (2010).
54. Dioos, B., Vankelecom, I. & Jacobs, P. Aspects of immobilisation of catalysts on
polymeric supports. Adv. Synth. Catal. 348, 1413–1446 (2006).
55. Collins, A. & Horvath, I. Heterogenization of homogeneous catalytic systems.
Catal. Sci. Tech. 1, 912–919 (2011).
3
7. Wienh o¨ fer, G. et al. General and selective iron-catalyzed transfer hydrogenation
of nitroarenes without base. J. Am. Chem. Soc. 133, 12875–12879 (2011).
8. Angelov, S., Zhecheva, E., Stoyanova, R. & Atanasov, M. Bulk defects in Co O ,
pure and slightly doped with lithium, revealed by EPR of the tetrahedral CO2
ions. J. Phys. Chem. Solids 51, 1157–1161 (1990).
3
3
4
þ
Acknowledgements
This research was supported by Federal Ministry of Education and Research (BMBF) and
the state of Mecklenburg-Vorpommern.
3
4
9. Dutta, P., Seehra, M. S. Thota, S. & Kumar, J. A comparative study of the
magnetic properties of bulk and nanocrystalline Co O . J. Phys. Condens. Matter
3
4
2
0, 015218 (2008).
0. Br u¨ ckner, A., Martin, A., Steinfeldt, N., Wolf, G-U. & L u¨ cke, B. Investigation of
vanadium phosphorus oxide catalysts (VPO) during toluene ammoxidation:
new mechanistic insights by in situ EPR. J. Chem. Soc. Faraday Trans. 92,
Author contributions
M.B., F.A.W. and R.V.J. planned and developed the project, F.A.W. performed and designed
the experiments and co-wrote the paper, R.V.J. developed, designed and prepared the
catalysts and was involved in writing the paper, G.W. conducted catalytic experiments and
performed the analytics, M.N. was involved in discussions and contributed to the writing of
the paper, K.J. participated in discussions and contributed to the writing of the paper, H.J.
was involved in the development of the catalyst, A.E.S. performed electrochemical studies of
the catalyst and was involved in the catalyst development, J. Radnik did the XPS
experiments, M.M.P. performed the TEM experiments, J. Rabeah and A.B. were responsible
for EPR experiments, M.B. directed the project, participated in the design of catalytic
experiments and in the design of the catalyst, and co-wrote the paper.
4
257–4263 (1996).
4
1. Wdowik, U. D. & Legut, D. CoO under pressure from first principles. J. Phys.
Chem. Solids 69, 1698–1703 (2008).
2. Papaefthimiou, V. et al. Nontrivial redox behavior of nanosized cobalt: new
insights from ambient pressure X-ray photoelectron and absorption
spectroscopies. ACS Nano 5, 2182–2190 (2011).
4
4
4
4
4
3. Herranz, T., Deng, X. Y. Cabot, A., Guo, J. G. & Salmeron, M. Influence of the
cobalt particle size in the Co hydrogenation reaction studied by in situ X-ray
absorption spectroscopy. J. Phys. Chem. B 113, 10721–10727 (2009).
4. Bezemer, G. L. et al. Cobalt particle size effects in the Fischer–Tropsch reaction
studied with carbon nanofiber supported catalysts. J. Am. Chem. Soc. 128,
Additional information
Supplementary information and chemical compound information are available in the
online version of the paper. Reprints and permissions information is available online
at www.nature.com/reprints. Correspondence and requests for materials should be
addressed to M.B.
3
956–3964 (2006).
5. Casanovas, J., Ricart, J. M., Rubio, J., Illas, F. & Jim e´ nez-Mateos, J. M. Origin of
the large N 1s binding energy in X-ray photoelectron spectra of calcined
carbonaceous materials. J. Am. Chem. Soc. 118, 8071–8076 (1996).
6. Pels, J. R., Kapteijn, F., Moulijn, J. A., Zhu, Q. & Thomas, K. M. Evolution of
nitrogen functionalities in carbonaceous material during pyrolysis. Carbon 33,
Competing financial interests
The authors declare no competing financial interests.
1
641–1653 (1995).
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
2013 Macmillan Publishers Limited. All rights reserved.
7
©