Hydrogenation of nitroarenes using defined iron-phosphine catalysts

Article abstract of DOI:10.1039/c3cc42983k

A novel iron-catalyzed hydrogenation of nitroarenes to the corresponding amines is reported. An in situ combination of Fe(BF₄) ₂·6H₂O and phosphine allows for highly selective hydrogenation of a broad range of aromatic and nitroarenes tolerating different functional groups.

Full text of DOI:10.1039/c3cc42983k

ChemComm
View Article Online
View Journal
COMMUNICATION
Hydrogenation of nitroarenes using defined
iron–phosphine catalysts†
Cite this: DOI: 10.1039/c3cc42983k
Gerrit Wienh o¨ fer, Mario Baseda-Kr u¨ ger, Carolin Ziebart, Felix A. Westerhaus,
Wolfgang Baumann, Ralf Jackstell, Kathrin Junge and Matthias Beller*
Received 11th June 2013,
Accepted 17th July 2013
DOI: 10.1039/c3cc42983k
www.rsc.org/chemcomm
6
7,8
9,11
6,8,10 11
A novel iron-catalyzed hydrogenation of nitroarenes to the corre- protocols primarily based on Au, Pt, Rh,
Pd,
have been developed in the last decades.
On the way to a more sustainable chemistry, the replacement of
of a broad range of aromatic and nitroarenes tolerating different precious metals by non-noble metals is a major goal. In this respect,
Ir and
8
,12
sponding amines is reported. An in situ combination of Fe(BF
4
)
2
Á
Ru
6H
2
O and phosphine allows for highly selective hydrogenation
functional groups.
iron as an abundant, non-toxic and cheap metal is an ideal candidate
for the development of environmentally benign processes. Significant
13
Aniline and its derivatives are important building blocks for the achievements have been realized in recent years in this area. In
production of dyes, pigments, agrochemicals, and pharma- contrast to precious metals, molecular-defined hydrogenation cata-
ceuticals produced annually on bulk industrial scale. In general, lysts based on iron have been scarcely studied. In fact, for the iron-
1
they are prepared via direct hydrogenation using heterogeneous catalyzed hydrogenation of nitroarenes only two protocols have been
catalysts. Applying more complex substrates these catalysts can developed. Knifton applied an Fe(CO) (PPh ) catalyst at 80 bar
3
3 2
12a
cause side-reactions such as dehalogenation processes in the hydrogen pressure for 9 hours at 125 1C to obtain moderate yields.
case of halonitroarenes or unselective reduction of other reducible On the other hand Chaudhari et al. used a system consisting of
moieties. To overcome these limitations, in organic synthesis on Fe(SO O and Na EDTA operating in situ at high temperature
)Á7H
the laboratory scale stoichiometric methods based on reducing (150 1C) in a biphasic medium.
agents such as Fe, Zn, Sn or Al and the sulphide reduction are still Based on our experience in the field of iron-catalyzed
4
2
2
14
2
15
popular. From ecological and economic points of view hydrogen reductions, we became interested in applying iron–phosphine
1
6
is the preferred reducing agent, because only water is formed complexes for this transformation. In this respect, recently we
as a by-product. Thus, numerous attempts have been made developed an iron–phosphine based catalyst for the mild and
to modify common heterogeneous catalysts to achieve high selective transfer hydrogenation of nitroarenes. Here, formic
1
7
3
chemoselectivities. Notably, significant progress was achieved acid is used as a stoichiometric reducing agent. Although the
by Corma and co-workers who developed a specific gold catalyst system is attractive for laboratory scale production, it produces
4
for a general reduction of nitroarenes. Most recently, we also carbon dioxide as a side product and is also corrosive because
developed a novel cobaltoxide-based catalyst which allows for of the high concentration of formic acid. In contrast, in
5
broad functional group tolerance.
industry the direct hydrogenation with molecular hydrogen is
Another approach to control selectivity for nitroarene reduction favoured as green technology. Hence, we wanted to adopt this
is working with organometallic complexes. The well-defined catalyst system for the direct hydrogenation of nitroarenes.
structure of the catalyst can be tuned easily to direct the
selectivity towards the desired product. Choosing the right with the ligand tris[(2-diphenylphosphino)ethyl]-phosphine
combination of the metal and the ligand can lead to significant [P(CH CH PPh ; L1] and the defined complex [FeF(L1)][BF ] for
Initially, we tested the in situ generated catalyst of Fe(BF
4
)
2
Á6H
2
O
2
2
2
)
3
4
improvements. Although it is clear that homogeneous catalyst the direct hydrogenation of nitrobenzene (Table 1, entries 7 and 8).
systems will not be used in existing continuously driven pro- A reaction temperature of 120 1C and 30 bar pressure were chosen
cesses, they are interesting for the production of specialties and initially for hydrogen activation. After two hours, a minor conversion
synthetic applications. For this purpose various hydrogenation of 9% was observed, indicating some reactivity of the catalyst. We
assumed that due to the high reaction temperature the ligand is
partly decomposed. Unfortunately, at lower temperature no reactivity
is observed. Hence, we started to synthesize the thermally more stable
Leibniz-Institut f u¨ r Katalyse e.V., Albert Einstein Straße. 29a, 18059 Rostock,
Germany. E-mail: matthias.beller@catalysis.de; Fax: +49-381-1281-5000;
2 3
tris[(2-diphenylphosphino)phenyl]-phosphine ligand [P(PhPPh ) ; L2]
Tel: +49-381-1281-113
Electronic supplementary information (ESI) available. See DOI: 10.1039/c3cc42983k where the phosphorus atoms are bridged by phenylene moieties.
†
This journal is c The Royal Society of Chemistry 2013
Chem. Commun.

View Article Online
Communication
ChemComm
a
Table 1 Iron-catalyzed hydrogenation of nitrobenzene
aniline (A) and the dimerization via azobenzene (B). To eluci-
date the preferred pathway with our system, we performed
catalytic hydrogenations with N-phenylhydroxylamine as well
as diazobenzene. In both cases, full conversion was achieved
at a similar rate. Thus, both reaction pathways seem to be
possible in the present system.
b
b
Entry
Catalyst
—
Conv. (%)
Yield (%)
Select. (%)
a
Table 2 Iron-tetraphos-catalyzed reduction of nitroarenes
1
2
3
4
o1
o1
o1
o1
o1
o1
9
9
49
48
o1
o1
o1
o1
o1
o1
5
9
49
48
o1
o1
o1
o1
o1
o1
56
>99
>99
>99
4
Fe(BF )
2
Á6H
2
O
L2
20 mL TFA
c
5
Fe(BF
L2
4
)
2
Á6H
2
O
c
b
b,c
6
Entry
Substrate, R =
Cat. (mol%)
Conv. (%)
Yield (%)
c
7
Fe(BF
4
)
2
Á6H
2
O/L1
c
8
[FeF(L1)][BF
4
]
c
9
4
Fe(BF )
2
Á6H
2
O/L2
1
2
2
2
>99
>99
99
c
1
0
[FeF(L2)][BF
4
]
a
Reaction conditions: 0.5 mmol nitrobenzene, 1 mol% catalyst, 120 1C,
96 (93)
b
2
h, 30 bar H , 1.5 mL THF. Determined by GC using n-hexadecane as
2
c
an internal standard. Addition of 50 mol% (20 mL) TFA.
3
2
>99
99 (92)
To our delight, applying either the in situ system or the defined
complex [FeF(L2)][BF
8%, respectively, were obtained under the same reaction conditions
Table 1, entries 9 and 10). The similar reactivity of the in situ catalyst
and the defined complex showed that both systems formed the same
active species in solution. Further, we tested the different compo-
nents of the system to exclude simple iron-, acid- or ligand-catalyzed
reaction (Table 1, entries 1–6). In none of these reactions any reactivity
4
] significantly higher conversions of 49% and
4
5
5
4
>99
>99
98
4
(
99 (90)
6
7
8
9
4
4
4
3
2
>99
>99
>99
>99
>99
99
97
98
99
98
4
was observed proving the [FeF(L2)][BF ] to be the real pre-catalyst.
Next, the reaction parameters were optimized to increase the
reactivity of this system (ESI†). The selectivity of the catalyst in tertiary
alcohols such as tert-amylalcohol is superior, while the deployment of
tetrahydrofuran led to similar results. Notably, addition of acid is
required to obtain significant hydrogenation activity, which is differ-
ent from most noble metal-based hydrogenation catalysts. By testing
the influence of different acids, significant reactivity is obtained only
in the presence of strong acids such as trifluoroacetic acid (TFA).
Further, the effect of the concentration of TFA was examined. The
addition of one equivalent (with respect to the substrate) is most
favourable. Investigating the dependency of the conversion on the
1
1
0
1
4
5
5
>99
>99
>99
99 (92)
97 (91)
99
hydrogen pressure showed best results between 10–20 bar of 12
hydrogen at 120 1C. Finally, a catalyst concentration of 2 mol%
was chosen to ensure complete conversion of the substrate to give
d
1
3
a quantitative yield of aniline.
An important aspect in the industrial reduction of nitroarenes
is the accumulation of toxic and/or explosive intermediates
1
4
2
5
4
>99
>99
>99
99 (94)
99
1
8
(Scheme 1). Under our conditions we did not detect any
formation of such intermediates. Two pathways are considered 15
for the reduction of nitrobenzene; the direct reduction to
16
91
d
1
1
a
7
4
4
>99
>99
91
78
8
Reaction conditions: 120 1C, 0.5 mmol substrate, 20 bar H
t-AmOH, 40 mL TFA, 2 h. Determined by GC using n-hexadecane as an
internal standard. Isolated yields given in brackets. 80 mL TFA.
2
, 1.5 mL
b
Scheme 1 Possible reaction pathways for the catalytic hydrogenation of
c
d
nitrobenzene.
Chem. Commun.
This journal is c The Royal Society of Chemistry 2013

View Article Online
ChemComm
Communication
the fluoride substituent does not play a significant role in the
catalytic cycle. To support this assumption, we varied the iron source
and tested the different in situ generated catalyst systems. Although
Fe(BF )Á6H O, which forms species 1 with the ligand L2, serves as the
4
2
best metal precursor, also the fluoride-free Fe(acac) and Fe(OAc)
2
2
led to active catalyst systems (ESI†). It should be noted that these iron
precursors cannot be used in related transfer hydrogenations.
In summary, we have developed a molecularly-defined iron-based
catalyst for the hydrogenation of nitroarenes. In the presence of the
phosphine ligand L2 the resulting stable iron complex is able to
activate hydrogen. After optimization, a variety of functionalized nitro-
arenes as well as heteroaromatic nitro compounds were successfully
hydrogenated to their corresponding amines while other reducible
Scheme 2 Proposed simplified catalytic cycle for the iron-catalyzed hydrogenation
of nitrobenzene.
Next, we investigated the substrate scope of our iron-based cata- moieties such as vinyl or acetyl groups remained unaffected.
lyst system (Table 2). Nitrotoluenes and related derivatives including
Notes and references
bulky ortho-substituted nitroarenes were hydrogenated giving quan-
titative yields (Table 2, entries 2–4). Halide-containing substrates
1
N. Ono, The Nitro Group in Organic Synthesis, Wiley-VCH, New York, 2001.
were also fully converted albeit a higher catalyst loading is required
Table 2, entries 5–8). More demanding nitroarenes bearing other
2 P. F. Vogt and J. J. Gerulis, Ullmann’s Encyclopedia of Industrial
Chemistry, Wiley-VCH, Weinheim, Germany, 2000.
(
3
(a) H. U. Blaser, H. Steiner and M. Studer, ChemCatChem, 2009,
, 210; (b) H. U. Blaser, U. Siegrist, H. Steiner and M. Studer, in
reducible moieties were studied, too (Table 2, entries 9–15). Notably,
substrates with C–C-double bonds, esters, and ketone functionalities
can be smoothly hydrogenated with excellent selectivity leaving the
additional substituent unaffected. Further, we tested different hetero-
aromatic nitro compounds due to the importance of the corre-
sponding anilines as valuable intermediates for pharmaceuticals
1
Fine Chemicals through Heterogeneous Catalysis, ed. R. A. Sheldon
and H. van Bekkum, Wiley-VCH, Weinheim, 2001, p. 389.
(a) A. Corma and P. Serna, Science, 2006, 313, 332; (b) A. Corma, P. Serna,
P. Concepci o´ n and J. Calvino, J. Am. Chem. Soc., 2008, 130, 8748.
4
5 F. A. Westerhaus, R. V. Jagadeesh, G. Wienh o¨ fer, M.-M. Pohl,
J. Radnik, A.-E. Surkus, J. Rabeah, K. Junge, H. Junge, M. Nielsen,
A. Br u¨ ckner and M. Beller, Nat. Chem., 2013, 6, 537.
19
and agrochemicals. Gratifyingly, 6-nitrobenzothiazol and 8-nitro-
quinoline were fully reduced to the corresponding amines. In order to
achieve complete conversion, an additional amount of TFA was
necessary to block the heterocyclic nitrogen atom from coordinating
to the catalyst (Table 2, entries 17 and 18). To get more insight into
6
7
8
A. Corma, C. Gon ´a lez-Arellano, M. Iglesias and F. S ´a nchez,
Appl. Catal., A, 2009, 356, 99.
A. I. Zakhariev, V. V. Ivanova, M. L. Khidekel, E. G. Chepaikin and
D. M. Shopov, React. Kinet. Catal. Lett., 1978, 8, 195.
Z. Yu, S. Liao, Y. Xu, B. Yang and D. Yu, J. Mol. Catal. A: Chem., 1997,
120, 247.
1
31
9 E. G. Chepaikin, M. L. Khidekel, V. V. Ivanova, A. I. Zakhariev and
M. Shopov, J. Mol. Catal., 1980, 10, 115.
the catalytic cycle, H and P NMR experiments were performed
under the standard reaction conditions. First, the reaction solution
10 S. Xu, J. Shi and S. Cao, J. Mol. Catal. A: Chem., 2000, 160, 287.
without the substrate was investigated at room temperature under 11 S. G. Harsy, Tetrahedron, 1990, 46, 7403.
1
2 (a) J. F. Knifton, J. Org. Chem., 1976, 41, 1200; (b) A. Toti, P. Frediani,
A. Salvini, L. Rosi and C. Giolli, J. Organomet. Chem., 2005, 690,
641; (c) A. A. Deshmukh, A. K. Prashar, A. K. Kinage, R. Kumar and
argon atmosphere. No NMR signal is detectable as the [FeF(L2)][BF
complex is paramagnetic. Pressurising the reaction solution with
0 bar of hydrogen resulted in partial formation of the diamagnetic
iron hydride complex [FeH(H )(L2)][BF ] which is characterized by its
4
]-
3
2
R. Meijboom, Ind. Eng. Chem. Res., 2010, 49, 12180.
1
3 For selected reviews see: (a) C. Bolm, J. Legros, J. Le Paih and L. Zani,
Chem. Rev., 2004, 104, 6217; (b) B. Plietker, Iron Catalysis in Organic
Chemistry, Wiley-VCH, Weinheim, 2008; (c) W. M. Czaplik, M. Mayer
and A. J. von Wangelin, Angew. Chem., Int. Ed., 2009, 48, 607;
2
4
31
1
H NMR signal at À9.18 ppm and two P NMR signals at 142.9 ppm
(
quartet) and 87.6 ppm (doublet) (ESI†). Raising the temperature to
(
4
d) K. Junge, K. Schr ¨o der and M. Beller, Chem. Commun., 2011,
7, 4849.
80 1C increased the formation of this hydrogenated complex. At a
temperature of 120 1C the signals of the complex could be detected
1
1
4 R. M. Deshpande, A. N. Mahajan, M. M. Diwakar, P. S. Ozarde and
R. V. Chaudhari, J. Org. Chem., 2004, 69, 4835.
31
only in the P NMR spectra due to the reduced intensity of the
5 For recent examples see: (a) N. S. Shaikh, S. Enthaler, K. Junge and
M. Beller, Angew. Chem., Int. Ed., 2008, 47, 2497; (b) K. Junge, B. Wendt,
N. Shaikh and M. Beller, Chem. Commun., 2010, 46, 1769; (c) S. Zhou,
S. Fleischer, K. Junge, S. Das, D. Addis and M. Beller, Angew. Chem.,
Int. Ed., 2010, 49, 8121; (d) S. Zhou, S. Fleischer, K. Junge and M. Beller,
Angew. Chem., Int. Ed., 2011, 50, 5120.
signals. The addition of nitrobenzene at 60 1C resulted in an
1
31
immediate disappearance of the H and P NMR signals of the
complex [FeH(H )(L2)][BF ]. After two hours at 120 1C nitrobenzene
was fully converted to aniline. As a result at 60 1C the complex
FeH(H )(L2)][BF ] was detected again by NMR.
Thus, we propose a simplified catalytic cycle (Scheme 2). The
2
4
[
2
4
1
6 (a) C. Federsel, A. Boddien, R. Jackstell, R. Jennerjahn, P. J. Dyson,
R. Scopeletti, G. Laurenczy and M. Beller, Angew. Chem., Int. Ed.,
2010, 49, 9777; (b) A. Boddien, D. Mellmann, F. G ¨a rtner, R. Jackstell,
H. Junge, P. J. Dyson, G. Laurenczy, R. Ludwig and M. Beller, Science,
+
defined pre-catalyst [FeF(L2)] (1) is hydrogenated in two steps.
Firstly, fluoride is replaced by a hydride while hydrogen fluoride
2011, 333, 1733.
+
is released and the complex [FeH(L2)] (2) is formed as part 17 G. Wienh ¨o fer, I. Sorribes, A. Boddien, F. A. Westerhaus, K. Junge,
H. Junge, R. Llusar and M. Beller, J. Am. Chem. Soc., 2011, 133,
of the catalytic cycle. Subsequently, 2 is hydrogenated to give
12875.
+
20
[
2
FeH(H )(L2)] (3). Complex 3 reduces nitroarenes to the corre-
1
8 (a) M. Freifelder, Practical Catalytical Hydrogenation, Wiley-Interscience,
New York, 1971; (b) P. Baumeister, H. U. Blaser and M. Studer,
Catal. Lett., 1997, 49, 219.
sponding nitrosoarenes. In parallel, complex 2 is formed, which is
hydrogenated again to regenerate 3. Two further consecutive hydro-
genations led to the formation of anilines. In contrast to our previous
1
2
9 C. M. Marson, Chem. Soc. Rev., 2011, 40, 5514.
0 C. Bianchini, M. Perruzini and F. Zanobini, J. Organomet. Chem.,
1988, 354, C19.
1
7
report on the mechanism of the transfer hydrogenation,
This journal is c The Royal Society of Chemistry 2013
Chem. Commun.