Angewandte
Chemie
DOI: 10.1002/anie.201004263
Iron-Catalyzed CO Hydrogenation
2
A Well-Defined Iron Catalyst for the Reduction of Bicarbonates and
Carbon Dioxide to Formates, Alkyl Formates, and Formamides**
Christopher Federsel, Albert Boddien, Ralf Jackstell, Reiko Jennerjahn, Paul J. Dyson,
Rosario Scopelliti, Gabor Laurenczy,* and Matthias Beller*
Carbon dioxide is the primary carbon source in the atmos-
phere and provides the basis for all organic matter on earth.
carbonates and carbon dioxide constitutes a challenging and
highly attractive goal.
Even though CO is abundant, cheap, and relatively nontoxic
Despite the recent achievements in iron-catalyzed hydro-
2
[
12]
compared to the alternative C source carbon monoxide, it is
genations and transfer hydrogenations, there are only two
examples of the homogeneous catalyzed reduction of carbon
dioxide. Evans and Newell reported a catalyst system derived
1
the latter species that is used as a raw material in many bulk-
scale chemical processes. One promising approach to over-
[
1]
À
[13]
coming the low reactivity of CO2 is its activation by catalytic
from [HFe (CO) ] for the synthesis of methyl formate.
3 11
[
2]
hydrogenation to form formic acid or its derivatives. Until
However, the product is only formed with a very low TOF
[
3,4]
À1
now, noble-metal catalysts based on rhodium,
ruthe-
(0.06 h ) and turnover number (up to 6). Jessop and co-
[5–7]
[8,9]
nium,
and iridium
were mostly used for this trans-
workers demonstrated by high-throughput screening that the
combination of Fe(OAc) or FeCl with bidentate phosphines
formation. Graf and Leitner had already achieved significant
turnover numbers (TONs) of 3400 with rhodium phosphine
2
3
created active catalysts for the reduction of carbon dioxide.
Although formic acid was formed under optimized conditions
[
3]
complexes in the early 1990s. Jessop, Ikariya, and Noyori
II
were able to increase the catalyst efficiency by using a Ru
(FeCl /1,2-bisdicyclohexylphosphinoethane)
with
an
3
[
5,6]
complex in supercritical CO2 (scCO ).
Very recently,
improved catalyst turnover number of 113, the reaction
proceeded only in the presence 0.5 equivalents of the
expensive base 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) at
2
III
Nozaki and co-workers used a pincer-type Ir catalyst for
the hydrogenation of CO which gave the highest TON
2
[9]
[14]
reported so far.
100 bar total pressure. Moreover, a large excess of carbon
Much less work on the biologically relevant reduction of
carbonates and bicarbonates has been reported, and the
reported activities of hydrogenation catalysts are significantly
dioxide was used in both reports, and consequently no yields
based on CO or amine are given.
2
Therefore, the need still exists to develop an efficient iron
lower compared to the reaction of CO . For example, Joꢀ and
catalyst for the hydrogenation of CO and carbonates. Based
2
2
co-workers obtained
a
TON of 108 with [{RuCl2-
on our previous studies on the ruthenium-catalyzed hydro-
[
15]
(
mTPPMS) } ] (mTPPMS = meta-monosulfonated triphenyl-
genation of carbonate and bicarbonate, we describe herein
for the first time an active iron catalyst system which can be
used for the reduction of both carbon dioxide and bicarbon-
ates to give formates, alkyl formates, and formamides. Central
to the success of this work is the combination of the iron
source Fe(BF ) ·6H O (1) and the tetradentate ligand
2
2
[10]
phosphine) in aqueous solution.
An important long-standing goal in chemistry is the
development of bio-inspired catalysis and the replacement of
noble metal based catalysts, that is, ruthenium, iridium, and
rhodium, with nonprecious metals, such as iron, zinc, and
4
2
2
[11]
manganese.
In this respect, an iron-based reduction of
P(CH CH PPh ) (PP ) which form defined iron hydride
2 2 2 3 3
complexes [FeH(PP )]BF4 (2) and [FeH(H )(PP )]BF (6)
3
2
3
4
under the reaction conditions.
In our initial approach we tested different iron precursors
and various nitrogen- and phosphine-containing ligands for
the hydrogenation of sodium bicarbonate to sodium formate.
Selected results of this investigation are shown in Table 1. To
the best of our knowledge, the use of any homogeneous iron
catalyst in this reaction has not been described before.
Commercially available mono-, bi- as well as tridentate
ligands—for example, different derivatives of triphos [1,1-
bis(2-diphenylphosphinoethyl)phenylphosphine (triphos 1),
[
*] C. Federsel, A. Boddien, Dr. R. Jackstell, R. Jennerjahn,
Prof. Dr. M. Beller
Leibniz-Institut fꢀr Katalyse e. V.
an der Universitꢁt Rostock
Albert Einstein Strasse 29a, 18059 Rostock (Germany)
Fax: (+49)381-1281-5000
E-mail: matthias.beller@catalysis.de
Homepage: http://www.catalysis.de
A. Boddien, Prof. Dr. P. J. Dyson, Dr. R. Scopelliti,
Prof. Dr. G. Laurenczy
Institut des Science et Ingꢂnierie Chimiques
Ecole Polytechnique Fꢂdꢂrale de Lausanne (EPFL)
Lausanne, 1015 (Switzerland)
1
(
,1,1-tris(diphenylphosphino)methane (triphos 2), 1,1,1-tris-
diphenylphosphinomethyl)ethane (triphos 3)], bisdiphenyl-
phosphinomethane (dppm), 1,2-bisdiphenylphosphinoethane
dppe), triphenylphosphine, 4,5-bis(diphenylphosphino)-9,9-
(
[
**] This work was supported by the state of Mecklenburg-Vorpommern,
the BMBF, and the DFG (Leibniz prize). We thank the EPFL and the
Swiss National Science Foundation for financial support.
dimethylxanthene (xantphos), tris[2-(dimethylamino)ethyl]-
amine (Me TREN), and tris(2-aminoethyl)amine (TAEA)—
6
showed no activity at all (Table 1, entries 1–5). Also, no
activity was observed when blank experiments were per-
Angew. Chem. Int. Ed. 2010, 49, 9777 –9780
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
9777