DOI: 10.1002/anie.201005823
Iron Catalysis
A General and Selective Iron-Catalyzed Aminocarbonylation of
Alkynes: Synthesis of Acryl- and Cinnamides**
Katrin Marie Driller, Saisuree Prateeptongkum, Ralf Jackstell, and Matthias Beller*
Dedicated to Professor Karlheinz Drauz on the occasion of his 60th birthday
Carbonylation reactions rank among the most important
examples of industrially applied homogeneous catalytic
reactions. In addition to hydroformylation, the synthesis of
carboxylic acid derivatives from unsaturated hydrocarbons
Scheme 1. Iron-catalyzed double carbonylation to 3,4-bisaryl malei-
mides. Ar1 =(hetero)aryl, Ar2 =(hetero)aryl, aryl, DDQ=2,3-dichloro-
5,6-dicyano-1,4-benzoquinone, L=2-(di-tert-butylphosphino)-N-phenyl-
indole, TMEDA=N,N,N’,N’-tetramethethylenediamine.
has in particular attracted considerable industrial and aca-
demic interest.[1] Since the pioneering work of Reppe,[2]
various organometallic catalysts and synthetic procedures
have been explored for the carbonylation of alkynes.[3]
Initially, nickel-based catalysts prevailed in industry. In fact,
the carbonylation of acetylene to acrylic acid was one of the
first large-scale industrial applications of organometallic
complexes.[4]
iron-catalyzed method is known. Notably, the resulting acryl-
or cinnamides are present in the structure of numerous
natural products[16] which show considerable biological and
insecticidal activity.[17] In addition, acrylamides are employed
in a wide range of organic reactions,[18] including polymeri-
zations.[19]
After the important development by Drent et al. in the
1980s,[5] cationic palladium complexes became the catalysts of
choice for the carbonylation of alkynes to substituted acrylic
acid derivatives. Thus, Shell developed a palladium-catalyzed
process for the production of methyl methacrylate from
propyne.[6] Since then, mainly noble-metal catalysts have been
investigated for such transformations.[7] However, the high
cost, limited availability, and sometimes sensitivity as well as
toxicity of precious-metal complexes means there is increas-
ing interest to substitute them by more easily available
biorelevant metals. In this respect, homogeneous catalysis
with iron complexes offers a highly attractive replacement.
Without doubt, this area has become one of the “hot topics”
in organometallic catalysis.[8–10] Although a number of stoi-
chiometric iron-mediated carbonylation reactions are
known,[11] catalytic processes have been only scarcely inves-
tigated. One of the exceptions is our synthesis of 3-(hetero)-
aryl-4-arylsuccinimides and -maleimides. Here, the key step
was the selective double aminocarbonylation of an internal
alkyne in the presence of [Fe3(CO)12] (Scheme 1).[12] This
work awoke our interest in the selective monocarbonylation
of alkynes to give the corresponding a,b-unsaturated
amides.[13] To the best of our knowledge no such general
At the start of our investigations we studied the influence
of different parameters (ligands, pressure, temperature, and
solvents) on the reaction of phenylacetylene with carbon
monoxide and cyclohexylamine in the presence of various
iron salts and complexes. Selected results are shown in
Table 1. Iron(II) and iron(III) chloride were completely
inactive (Table 1, entries 1 and 2), whereas iron carbonyl
complexes in general showed high activity (Table 1, entries 3–
9). However, in addition to the desired monocarbonylation
product 1, the double carbonylation product N-cyclohexyl-2-
phenylsuccinimide was observed. The chemoselectivity for 1
is significantly increased by using [(cot)Fe(CO)3] (cot =
cyclooctatetraene) as the catalyst precursor (Table 1,
entry 8) compared to reactions with [Fe(CO)5], [Fe2(CO)9],
or [Fe3(CO)12] as catalyst precursors (Table 1, entries 3–5).
Interestingly, in the latter cases no significant differences were
observed, thus making a similar active species likely. To prove
that the observed catalyst activity does not depend on other
metal impurities, we investigated possible contaminations and
analyzed [Fe3(CO)12] by atomic absorption spectroscopy
(AAS) and inductively coupled plasma (ICP). However, all
other metals were below the detection limits of these methods
(see the Supporting Information).[20] Furthermore, catalytic
experiments were carried out using 5 mol% [Cr(CO)6],
[Mo(CO)6], and [W(CO)6], and with 100 ppm [Ru3CO12],
[Pd(dba)2] (dba = trans,trans-dibenzylideneacetone), [Ni-
(acac)2] (acac = acetylacetanoate), as well as six different
copper salts. In no case was the carbonylation product
obtained, which clearly proves that the reaction is catalyzed
by iron.
[*] K. M. Driller, S. Prateeptongkum, Dr. R. Jackstell, Prof. M. Beller
Leibniz-Institut fꢀr Katalyse e. V.
Albert-Einstein-Strasse 29a, 18059 Rostock (Germany)
Fax: (+49)381-1281-51113
E-mail: matthias.beller@catalysis.de
[**] This work was funded by the the Deutsche Forschungsgemeinschaft
(Graduiertenkolleg 1213 and Leibniz-price). We thank Dr. C. Fischer,
S. Buchholz, S. Schareina, A. Kammer, K. Fiedler, A. Lehmann, I.
Stahr, A. Koch, and Dr. W. Baumann for their excellent technical and
analytical support.
Next, we tested different solvents (diglyme, NMP, THF,
and cyclohexylamine) in the benchmark reaction (Table 1,
entries 5, 11–13). Performing the reaction in neat amine
Supporting information for this article is available on the WWW
Angew. Chem. Int. Ed. 2011, 50, 537 –541
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
537