Regioselective iron-catalyzed allylic amination

Article abstract of DOI:10.1002/anie.200602261

(Chemical Equation Presented) Iron age: An air- and water-resistant iron(-II) catalyst, in the presence of triphenylphosphane under buffered conditions, allows for the highly regioselective and stereoselective allylation of primary aromatic amines. A broa

Full text of DOI:10.1002/anie.200602261

Angewandte
Chemie
Homogeneous Catalysis
DOI: 10.1002/anie.200602261
Regioselective Iron-Catalyzed Allylic
Amination**
Bernd Plietker*
Catalysis in the presence of inexpensive and nontoxic metals
has gained increasing attention in chemical research within
the past years. In particular, iron complexes proved to be
efficient and selective catalysts for a variety of transforma-
tions.[1] With regard to the importance of allylic substitutions
in organic synthesis[2] the development of an iron-catalyzed
protocol appeared to be a useful extension of the existing
methodologies. Recently, we presented the first regioselec-
tive, salt-free, iron-catalyzed allylic alkylation of allyl carbo-
nates.[3] In this reaction a new C C bond is formed with
À
retention of the configuration at the carbon atom that was
substituted by the leaving group in the starting material.
Aiming to broaden the scope of the reaction and to gain
further information on stability and reactivity of the iron
catalyst [Bu4N][Fe(CO)3(NO)], we further developed this
reaction into an iron-catalyzed, allylic amination. Such a
reaction has not been developed until now.
Various transition metals are known to catalyze the allylic
amination.[4] The classical version in the presence of Pd
catalysts, however, exhibits a lack of regioselectivity. In
general, the product resulting from an attack of the nucleo-
phile at the sterically less encumbered allyl terminus is
formed predominantly. The preparation of higher-substituted,
branched amines can be realized in the presence of Ir[5] or
Ru catalysts.[6] However, in all of these protocols the sub-
stitution pattern in the p-allyl metal intermediate determines
the regioselective course of the reaction. The Rh-catalyzed
allylic amination on the other hand might be regarded as a
complementary procedure.[7] In this reaction both the regio-
and stereoselectivity are solely determined by the constitution
of the starting material.[8] The new C N bond is formed
À
selectively at the carbon atom that was substituted by the
leaving group in the starting material.
[*] Dr. B. Plietker
Organische Chemie II
FB Chemie
Universität Dortmund
Otto-Hahn-Strasse 6, 44221 Dortmund (Germany)
Fax: (+49)231-755-3884
E-mail: bernd.plietker@uni-dortmund.de
[**] Financial support from Prof. Dr. N. Krause, the Fonds der
Chemischen Industrie, the Deutsche Forschungsgemeinschaft, the
Dr.-Otto-Röhm-Gedächtnisstiftung, and the Fonds zur Förderung
des wissenschaftlichen Nachwuchses im Fachbereich Chemie der
Universität Dortmund is gratefully acknowledged.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2006, 45, 6053 –6056
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6053

Communications
Although allylic substitutions are formally related to C or
amine with a high degree of regioselectivity in favor of the
ipso-substitution product (Table 1, entry 8). A transfer of
these results to the amination of less-substituted allyl
carbonates, however, proved to be difficult (Scheme 2).[11]
N nucleophiles, the direct nucleophilicity[9] of the nitrogen
atom causes significant mechanistic differences (Scheme 1).
Scheme 2. Iron-catalyzed allylic amination. Conditions: A) Fe cat.
(5 mol%), PPh3 (5 mol%), 1m, 36% yield; B) Fe cat. (5 mol%), PPh3
(5 mol%), Pip·HCl (30 mol%), 10m, 84% yield.
Scheme 1. Mechanistic dichotomy in allylic aminations. EWG=elec-
tron-withdrawing group.
Again, significant catalyst decomposition was observed.
This undesired side effect was retarded upon addition of a
catalytic amount of piperidinium chloride (Pip·HCl) as a
buffer. This addition in combination with an increase in the
substrate concentration led to a significant increase in the
conversion (Scheme 2), thus allowing for the regioselective
amination of primary and secondary allyl carbonates in good
yield (Table 2).
Similar to the C nucleophiles, electron-poor secondary
amines have to be deprotonated separately or in situ prior to
the allylic substitution and then treated with an allyl metal
species I (Scheme 1, path A). The low tendency of these
nucleophiles and the resulting product III to coordinate to the
metal center is advantageous for the reaction. Similar
processes often lead to catalyst inhibition. Electron-rich
primary or secondary amines on the other hand can coor-
dinate to a metal center but offer the possibility to act as a
direct nucleophile[9] in the reaction with allyl metal species
like I (Scheme 1, path B). Upon substitution of the metal
center, an ammonium cation II is formed which is deproton-
ated in the final step of the mechanism.
Table 2: Amination of allyl carbonates.[a]
Since no iron-catalyzed allyic amination has been de-
scribed in the literature so far,[10] the similarity of the allylic
amination using electron-poor secondary amines (Scheme 1,
path A) with the allylic alkylation motivated us to apply this
amine class in initial screening experiments (Table 1,
entries 1–4). However, the conversion using this nucleophile
class was very low. Moreover, a brown solid was formed
during the reaction, which indicated probable catalyst decom-
position.
Entry
1
Carbonate
Product
t [h][b]
Yield [%][c,d]
69 (97:3)
48 (5)
5
4
The reaction in the presence of primary amines on the
other hand (Scheme 1, path B) furnished the desired allyl
2
3
4
6 (2.5)
87 (98:2)
61 (97:3)
62 (96:4)
1
6
Table 1: Development of the allylic alkylation of amines.
48 (5)
24 (5)
7
9
8
[a]
Entry
R1
R2
A/B[a]
Conv. [%]
10
1
2
3
4
5
6
7
8
Ph
Ph
Ph
nBu
nPr
nBu
tBu
Ph
Ts
Ms
Ac
Ts
nPr
H
n.d.[b]
n.d.[b]
n.d.[b]
n.d.[b]
n.d.[b]
n.d.[b]
96:4
8
12
–
–
–
12
23
67
5
24 (5)
47 (97:3)
11
12
H
H
[a] All reactions were performed on a 1-mmol scale. [b] Amount of
catalyst in mol% is given in brackets. [c] Yield of isolated product.
[d] Regioselectivity of the crude product according to GC integration is
given in brackets.
97:3
[a] According to GC integration. [b] Not determined. Ts=p-toluenesul-
fonyl, Ms=methanesulfonyl, Ac=acetyl.
6054 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 6053 –6056

Angewandte
Chemie
À
In all cases the new C N bond is formed with very high
reason for the observed high regioselectivity. If a stereospe-
selectivity at the carbon atom that was substituted by the
leaving group in the starting material. However, both reaction
time and yield strongly depend on the substitution pattern of
the allyl carbonate: the more highly substituted the double
bond the slower the reaction proceeds. Whereas monosub-
stituted olefins are efficiently aminated, the introduction of
one more olefinic substituent leads to a significant increase in
the reaction time (2 vs. 7; Table 2, entry 3).[12] Furthermore,
tertiary allyl carbonates are significantly more reactive
compared to primary or secondary carbonates (1, 2, and 5;
Table 2). Both steric and electronic effects might account for
this finding. Gratifyingly, the longer reaction times had no
impact on the regioselectivity of the reactions.
cific double SN2’-anti mechanism takes place, the use of
enantiomerically pure allyl carbonates should furnish the
corresponding amines with retention of the configuration.
Indeed, the allylic amination of (S)-9 occurrs with almost
complete chirality transfer and retention of the configuration
to give amine (S)-10 (Table 4, entries 1–3). This result
Table 4: Chirality transfer in allylic aminations.
Entry
R
Substrate Cond.[b] Product[c] ee [%][a] Yield [%][d]
(ee [%][a])
Table 3: Allylation of primary anilines.[a]
1
2
3
Ph
(S)-9
(94)
A
B
C
(S)-10
86
83
84
–
15
62
4
5
6
H
(S)-2
(92)
A
B
C
(S)-3
87
82
80
–
36
84
[a] According to HPLC using Chiracel columns. [b] Conditions:
A) 1 mmol substrate, 2 mmol aniline, 250 mL DMF, 808C, 5 h;
B) 1 mmol substrate, 2 mmol aniline, 10 mol% [Bu4N][Fe(CO)3(NO)],
10 mol% PPh3, 250 mL DMF, 808C, 5 h; C) 1 mmol substrate, 2 mmol
aniline, 5 mol% [Bu4N][Fe(CO)3(NO)], 5 mol% PPh3, 30 mol% Pip·HCl,
250 mL DMF, 808C, 5 h. [c] The absolute configuration was assigned by
comparison to the literature. [d] Yield of isolated product.
Entry
Product
t [h]
Yield [%][b,c]
1
2
3
4
5
6
13 (X=m-CH3)
14 (X=p-CH3)
15 (X=m-OCH3)
16 (X=p-OCH3)
17 (X=m-Cl)
6
6
12
12
15
6
85 (97:3)
87 (98:2)
87 (97:3)
82 (98:2)
71 (96:4)
89 (97:3)
18 (X=3,5-CH3)
corresponds nicely with the observations made in the allylic
alkylation of malonic acid derivatives.[3] Interestingly, only a
small decrease in the enantiopurity was observed in the allylic
amination of (S)-2 (Table 4, entries 4–6). We were pleased to
find that the additive had no influence on the chirality
transfer.
7
8
6
6
67 (97:3)
93 (97:3)
19
20
This result is remarkable since the chirality of the a-
carbon atom in the in situ generated s-allyl iron complex V is
À
conserved only through coordination to the C C double
À
[a] All reactions were performed on a 1-mmol scale. [b] Yield of isolated
product. [c] Regioselectivity of the crude product according to GC
integration is given in brackets.
bond, thus preventing fast rotation around the C C single
bond, which would result in a loss of stereochemical
information (Scheme 3). Furthermore, with regard to the
observed strong substituent effects, a stereospecific ionization
by a SN2mechanism appears improbable.
Various aniline derivatives are regioselectively allylated
under the optimized reaction conditions (Table 3). Chloride,
ether, or ester functionalities are tolerated. Even the presence
of strongly coordinating functional groups such as the
oxazolidinone in 19 does not lead to a reduction of the
catalytic activity. However, the reaction time depends
strongly on the nucleophilicity of the nitrogen atom. Elec-
tron-poor anilines are allylated more slowly, albeit with the
same high degree of regioselectivity, compared to the
electron-rich derivatives. Whereas para- or meta-substituted
anilines are efficiently transformed into the corresponding
products, a substituent in the ortho position inhibits the
reaction.
Herein, our results on the first iron-catalyzed allylic
amination of allyl carbonates are summarized. As compared
to the already established Rh-catalyzed procedure, this
protocol is distinguished in particular by the use of the
As for the iron-catalyzed allylic alkylation, we envisioned
a mechanism involving a s-allyl metal intermediate as the
Scheme 3. Stereoselectivity in allylic aminations.
Angew. Chem. Int. Ed. 2006, 45, 6053 –6056
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6055

Communications
[2] a) Review on asymmetric allylic substitutions: B. M. Trost, M. L.
Crawley, Chem. Rev. 2003, 103, 2921; b) I. D. G. Watson, A. K.
Yudin, J. Am. Chem. Soc. 2005, 127, 17516, and references
therein; c) J. W. Faller, J. C. Wilt, Org. Lett. 2005, 7, 1239.
[3] B. Plietker, Angew. Chem. 2006, 118, 1497; Angew. Chem. Int.
Ed. 2006, 45, 1469.
[4] Review on allylic aminations: M. Johannsen, K. A. Jørgensen,
Chem. Rev. 1998, 98, 1689.
[5] a) A. Leitner, S. Shekhar, M. J. Pouy, J. F. Hartwig, J. Am. Chem.
Soc. 2005, 127, 15506; b) C. Welter, A. Dahnz, B. Brunner, S.
Streiff, P. Dübon, G. Helmchen, Org. Lett. 2005, 7, 1239, and
references therein.
[6] a) R. Hermatschweiler, I. Fernandez, F. Breher, P. S. Pregosin,
L. F. Veiros, M. J. Calhorda, Angew. Chem. 2005, 117, 4471;
Angew. Chem. Int. Ed. 2005, 44, 4397; b) R. Hermatschweiler, I.
Fernandez, P. S. Pregosin, E. J. Watson, A. Albinati, S. Rizzato,
L. F. Veiros, M. J. Calhorda, Organometallics 2005, 24, 1809.
[7] a) Review on rhodium-catalyzed allylic substitutions: D. K.
Leahy, P. A. Evans in Modern Rhodium-Catalyzed Organic
Reactions (Ed.: P. A. Evans), Wiley-VCH, Weinheim, 2005,
p. 191; b) P. A. Evans, J. E. Robinson, K. K. Moffett, Org. Lett.
2001, 3, 3269; c) K. Fagnou, M. Lautens, Org. Lett. 2000, 2, 2319;
d) P. A. Evans, J. E. Robinson, J. D. Nelson, J. Am. Chem. Soc.
1999, 121, 6761.
[8] P. A. Evans, J. D. Nelson, J. Am. Chem. Soc. 1998, 120, 5581.
[9] “Direct nucleophiles” are distinguished by the fact that the
nucleophilic center does not have to be activated prior to the
reaction (through, for example, deprotonation).
[10] Oxidative, iron-catalyzed aminations of unactivated olefins:
R. S. Srivastava, K. M. Nicholas, J. Am. Chem. Soc. 1997, 119,
3302, and references therein.
[11] For a common reaction between iron carbonyl complexes and
nitrogen bases, see: a) W. Hieber, N. Kahlen, Chem. Ber. 1958,
91, 2223; b) W. Hieber, N. Kahlen, Chem. Ber. 1958, 91, 2234.
[12] A significant influence of the substitution pattern was already
observed in the iron-catalyzed allylic alkylation;[3] Evans and
Robinson reported on similar results in Rh-catalyzed allylic
alkylations.[8] The electronic influence of various substituents on
conversion and yield in the stoichiometric amination of p-allyl
iron complexes has been described: S. Nakanishi, K. Okamoto,
H. Yamaguchi, T. Takata, Synthesis 1998, 1735.
[13] The only broadly applicable procedure for the allylic substitu-
tion of higher-substituted allyl carbonates by the mechanism
involving a s-allyl metal fragment to be reported so far was
published by Martin et al.: B. L. Ashfeld, K. A. Miller, S. F.
Martin, Org. Lett. 2004, 6, 1321.
inexpensive
and
nontoxic
iron
catalyst
[Bu4N]
[Fe(CO)3(NO)], the simple working procedure, and the
high stereo- and regioselectivity, especially in the amination
of higher-substituted allyl carbonates.[13]
Various allyl carbonates are converted under the de-
scribed conditions into essentially regio- and stereoisomeri-
cally pure allyl amines. The use of secondary amines as
N nucleophiles proved to be problematic as a result of catalyst
decomposition. However, this side reaction is retarded in the
presence of catalytic amounts of piperidinium chloride as a
buffer. The observations and results obtained within the
development of the iron-catalyzed allylic amination allowed
for important insights into the stability and reactivity of the
iron catalysts. Based on these results, future work will
concentrate on the development of CO-free iron(ÀII) com-
plexes, which might possess higher stability toward basic
amines.
Experimental Section
General procedure for the allylic amination: In a 2-mLWheaton vial
equipped with a stirring bar and a Mininert valve, a mixture of [Bu4N]
[Fe(CO)3(NO)] (20.6 mg, 0.05 mmol), PPh3 (13.2mg, 0.05 mmol),
and Pip·HCl (37.1 mg, 0.3 mmol) in DMF (250 mL) under argon was
stirred for 15 min at 808C. After cooling to ambient temperature, the
yellow-brown suspension was stirred further for 15 min before
addition of the aniline derivative (2mmol) and the allyl carbonate
(1 mmol). The closed system was heated at 808C until full conversion
of the carbonate (TLC analysis). Purification was performed directly
by column chromatography (SiO2, isohexane/ethyl acetate) of the
cooled reaction mixture. The products were obtained as air- and in
some cases light-sensitive, colorless oils.
Received: June 6, 2006
Published online: August 9, 2006
Keywords: allylic amination · arylamines · homogeneous
.
catalysis · iron · regioselectivity
[1] a) C. Bolm, J. Legros, J. Le Paih, L. Zani, Chem. Rev. 2004, 104,
6217; b) A. Fürstner, R. Martin, Chem. Lett. 2005, 34, 624.
6056 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 6053 –6056