Taking advantage of the ambivalent reactivity of ynamides in gold catalysis: A rare case of alkyne dimerization

Article abstract of DOI:10.1002/anie.201100327

A gold mine of results: A series of ynamides have been dimerized in the presence of a gold(I) complex. This unprecedented transformation involves the formation of a key keteniminium intermediate that reacts to form a variety of cyclic and acyclic products. The substitution pattern of the ynamide determines which product is formed (see scheme; EWG=electron-withdrawing group, Ts=p-toluenesulfonyl). Copyright

Full text of DOI:10.1002/anie.201100327

Communications
DOI: 10.1002/anie.201100327
Gold Catalysis
Taking Advantage of the Ambivalent Reactivity of Ynamides in Gold
Catalysis: A Rare Case of Alkyne Dimerization**
Søren Kramer, Yann Odabachian, Jacob Overgaard, Mario Rottlꢀnder, Fabien Gagosz,* and
Troels Skrydstrup*
Ynamide derivatives are easily accessible and relatively stable
compounds which have proved to be useful buildings blocks
for the synthesis of a large variety of nitrogen-containing
molecules.[1] Owing to the polarization of the triple bond,
ynamides can be regioselectively activated by an electrophile
to induce the addition of a variety of oxygen-, nitrogen-, or
carbon-containing nucleophiles at the position a to the
nitrogen atom (Scheme 1). Gold salts or gold complexes,
products appear to be dependent on the electronic nature of
the substituents located on the nitrogen atom and at the
alkyne terminus.
During the course of our investigations on the gold-
catalyzed addition of external nucleophiles onto ynamide 1a,
we noticed the formation of a side product that could also be
formed and isolated in the absence of a nucleophile. Its
structure was assigned on the basis of 1H NMR spectroscopic
analysis, and was confirmed as cyclopentadiene 2a.[5] Differ-
ent catalysts and experimental conditions were screened to
optimize this transformation (Table 1).[6] Compound 2a was
obtained in an excellent 85% yield when 1a was treated with
Table 1: Optimization of the catalytic system with ynamide 1a.[a]
Entry
Catalyst (mol%)
T [8C]
t [h]
Yield [%][b]
1
2
3
4
[(Ph3P)AuNTf2] (2)
[(Ph3P)AuNTf2] (2)
HNTf2 (20)
20
40
40
40
0.5
0.5
2
85
98
0[c]
4[d]
Scheme 1. Gold-catalyzed dimerization of ynamides. EWG=electron-
withdrawing group.
AgOTf (10)
24
which are strong alkynophilic species,[2] are particularly
efficient catalysts for such transformations, as reported in a
series of recent articles.[3,4] Following our previous work in
this field,[4] we report herein an unprecedented mode of
reactivity where, in the presence of a gold(I) complex,
ynamides can play the role of both the electrophilic and
nucleophilic partner to produce various complex, dimerized
products. The formation and the nature of the dimeric
[a] Reactions were carried out under argon and at a concentration of
0.1m. [b] Yield of the isolated product. [c] Degradation occurred. [d] Yield
was determined by 1H NMR spectroscopy. Bn=benzyl, Tf=trifluorome-
thanesulfonyl, Ts=p-toluenesulfonyl.
2 mol% of [(Ph3P)AuNTf2][7] in dichloromethane at 208C for
0.5 hour (Table 1, entry 1). The transformation was nearly
quantitative (98%) when the reaction was conducted at reflux
(Table 1, entry 2). Control experiments showed that the
Brønsted acid HNTf2 did not catalyse this transformation
(Table 1, entry 3), while AgOTf was almost as inefficient
(Table 1, entry 4).
The formation of cyclopentadiene 2a is remarkable. To
the best of our knowledge, no example of such a [3+2] cyclo-
dimerization of an alkyne has been reported in the literature.
This unprecedented transformation is not only performed
under mild experimental conditions with a low loading of gold
catalyst, but is also rapid and efficient.
The scope of the reaction was next investigated by
screening various ynamides bearing alkyl substituents at the
terminus of the alkyne. As shown in Table 2, the trans-
formation could be applied to a series of ynamides 1a–h. The
corresponding cyclopentadienes 2a–h were rapidly formed
(0.5–3 h) in good to excellent yields (73–98%). The benzyl
moiety (Table 2, entries 1, 2, and 6–9), as well as other
[*] S. Kramer, Dr. J. Overgaard, Prof. T. Skrydstrup
Department of Chemistry, Aarhus University
Langelandsgade 140, 8000 Aarhus C (Denmark)
E-mail: ts@chem.au.dk
Y. Odabachian, Dr. F. Gagosz
Dꢀpartement de Chimie, UMR 7652 CNRS
Ecole Polytechnique, 91128 Palaiseau (France)
E-mail: gagosz@dcso.polytechnique.fr
Dr. M. Rottlꢁnder
Division of Medicinal Chemistry
H. Lundbeck A/S, Ottiliavej 9, 2500 Valby (Denmark)
[**] We are deeply appreciative of generous financial support from the
Carlsberg Foundation, the Danish National Research Foundation,
H. Lundbeck A/S, the OChem Graduate School, the Ecole
Polytechnique and Aarhus University.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201100327.
5090
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 5090 –5094

Table 2: Dimerization of alkyl-substituted ynamides.
of the unreactive Au-1j species by
coordination of the gold(I) complex
to the more electron-rich ynamide
1j. When 1a was treated with 1i,
only the homodimer 2a was pro-
duced by the selective interaction of
the more favored Au-1a gold com-
plex (versus Au-1i) with the more
nucleophilic ynamide 1a (versus
1i). Finally, no selectivity was
obtained when 1a was treated with
1h, and a mixture of homodimers
and cross-dimers was obtained.[8]
Entry
Substrate
R2
t [h]
Product
Yield [%][a]
EWG R1
R3
1
2
3
Ts
Ts
Ts
Bn
Bn
n-C4H9
n-C5H11
n-C3H7
n-C5H11
H
H
H
1a
1b
1c
0.5
0.5
1
2a
2b
2c
98
90
91
4
Ts
n-C5H11
H
1d
2
2d
76
5
6[b]
Ts
Ts
n-C4H9
H
H
1e
1 f
3
2e
2 f[d]
73
83
Bn
0.5
7[b]
8
Ts
Mbs
Nos
Bn
Bn
Bn
H
H
H
1g
1h
1i
0.5
0.5
24
2g
2h
2i
87
91
0
A
surprising but interesting
divergence in reactivity was
observed when using tosylynamides
1k–n that possess cycloalkyl groups
at the terminus of the alkyne in the
gold(I)-catalyzed reaction. In these
cases, no cyclopentadiene forma-
tion was observed, and instead
structurally complex tricyclic com-
pounds 2k–n were formed in an
efficient (79–96% yield) and dia-
stereoselective manner (Table 2,
entries 11–14).[9]
n-C5H11
n-C5H11
9
10
1j
24
2j
0
2k
n=1
2l
n=1
2m
n=1
2n
11
12
Ts
Ts
Bn
-C5H10-[e]
-C5H10-[e]
1k
1l
18
5
96
80
n-C4H9
13[c]
14[b]
Ts
Ts
-C5H10-[e]
-C6H12-[e]
1m 24
1n 24
85
79
Bn
n=2
[a] Yield of the isolated product. [b] 5 mol% of catalyst was used. [c] 4 mol% of catalyst was used.
2
3
À
[d] Unstable compound. [e] R R =cycloalkyl. Mbs=para-methoxybenzenesulfonyl, Nos=para-nitro-
benzenesulfonyl, Phth=phthalimido.
substituents were tolerated on the nitrogen atom (Table 2,
entries 3–5). The reaction was also compatible with a range of
functional groups, such as a halide (Table 2, entries 4 and 6),
an acetate (Table 2, entry 5), and a phthalimido group
(Table 2, entry 7). The nature of the electron-withdrawing
group was shown to strongly affect the efficiency of the
reaction. While the dimerization worked equally as well with
a para-methoxybenzenesulfonyl group (91%, 0.5 h; Table 2,
entry 8), no cyclopentadiene was obtained with the more
electron-withdrawing
para-nitrobenzenesulfonyl
group
(Table 2, entry 9). Also, the reaction could not be performed
with a more electron-rich ynamide, which possesses an
oxazolidinone moiety (Table 2, entry 10). These results high-
light the necessity to have a good balance in the electronic
properties of the ynamide, which should play both the roles of
the electrophile and the nucleophile in the dimerization
process (Scheme 2).
Scheme 2. Influence of the electronic properties of the ynamide.
Ynamides containing a tosyl or a methoxybenzenesul-
fonyl group seem to favor the dimerization because the gold-
complexed ynamide is electrophilic enough and the ynamide
is nucleophilic enough for the reaction to proceed. The
unsuccessful homodimerizations of substrates 1i and 1j
(Table 2, entries 9 and 10) could be respectively explained
by the comparatively weak nucleophilic character of 1i and
the comparatively weak electrophilic character of Au-1j.
The tentative rationale presented in Scheme 2 is sup-
ported by the results obtained when a series of cross-
dimerization reactions were attempted with substrates 1a
and 1h–j (Scheme 3). No dimer was obtained when 1j was
treated with 1i or 1a, as a result of the preferential formation
Scheme 3. Investigations into cross-dimerization.
Angew. Chem. Int. Ed. 2011, 50, 5090 –5094
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5091

Communications
A mechanistic proposal that accounts for the divergent
character of the aryl group could lower the nucleophilicity
of the ynamide 8, thus disfavoring the dimerization process.
The N-benzyl tosylamide moiety in 8 was therefore replaced
by the less electron-withdrawing oxazolidine to compensate
for the negative electronic effect of the aryl group. After
optimization of the reaction conditions,[6] we were pleased to
see that the corresponding ynamide 9a could be converted
into a mixture of three products in the presence of [(John-
Phos)AuSbF6][15] (Table 3, entry 1).
formation of the dimeric products 2a–h and 2k–n is presented
in Scheme 4. The nucleophilic addition of the ynamide 3 to
the gold-activated ynamide could lead to the formation of the
Table 3: Dimerization of aryl-substituted ynamides.
Scheme 4. Mechanistic proposal for dimerization of alkyl-substituted
ynamides.
Entry Substrate
Catalyst
Yield [%][a]
Combined 10 11 12
keteniminium intermediate 4.[10] A subsequent 1,5-hydride
shift[11] followed by a metalla-Nazarov reaction[3d,12] would
lead to gold carbene 5. The cyclopentadiene can then be
obtained by a 1,2-hydride shift/gold catalyst regeneration
analogous to that proposed for the gold-catalyzed cyclo-
isomerization of allenenes.[13] Alternatively, gold carbene 5
could lead to the formation of a fused tricyclic compound by
1
2
3
4
5
9a (R=H)
9a (R=H)
9b (R=Me)
9c (R=OMe) [(JohnPhos)AuSbF6]
9d (R=F) [(JohnPhos)AuSbF6]
[(JohnPhos)AuSbF6]
[(IPr)AuSbF6]
[(JohnPhos)AuSbF6]
75
88
82
57
76
55
9
11
35 24 29
26 39 17
9
32 16
12 28 36
[a] Yield of the isolated product. IPr=1,3-bis(2,6-diisopropylphenyl)imi-
dazol-2-ylidene.
À
insertion of the gold carbene into a C H bond of the adjacent
spirocycloalkyl fragment. While such a process has already
been reported to operate with seven- and eight-membered
cycloalkyl rings, no example is known with a six-membered
ring because the elimination of the gold fragment was shown
to be more favored in this case.[12a,14]
Naphthalene 10a was isolated as the major product (55%
yield) along with minor amounts of azulene 11a (9% yield)
and enyne 12a (11% yield; Table 3, entry 1).[16] The use of
[(IPr)AuSbF6] as the catalyst (5 mol%) led to an improve-
ment of the combined yield (88%) with increased formation
of 11a and 12a (Table 3, entry 2).[17] Notably, the reaction
required a higher temperature (toluene, 1008C), a longer
reaction time (24 h), and a higher catalyst loading (5 mol%)
than for the cyclopentadiene formation. The transformation
was also attempted with aryl-substituted ynamides (Table 3,
entry 3–5) using [(JohnPhos)AuSbF6] as the catalyst.[18] The
presence of a methyl or a methoxy group on the phenyl ring
favored the formation of the azulene compound (11), while a
preference for the formation of the enyne (12) was observed
in the presence of an electron-withdrawing fluorine atom.
A mechanistic proposal for the formation of the three
products is shown in Scheme 6. The naphthalenes 10 and the
To extend the scope of the dimerization process, we next
turned our attention to the reactivity of aryl-substituted
ynamides of type 6 (Scheme 5). It was hypothesized that the
resulting keteniminium intermediate 7 could be trapped by
the nucleophilic pendant aryl group. Tosylynamides such as 8
proved to be unreactive under the reaction conditions used.
This behavior strongly contrasts with the high reactivity of
tosylynamides 1a–g in the formation of cyclopentadienes 2a–
g (Table 2). We reasoned that the electron-withdrawing
Scheme 5. Proposed dimerization of aryl-substituted ynamides.
5092 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 5090 –5094

Keywords: dimerization · gold · homogeneous catalysis ·
.
synthetic methods · ynamides
[1] For recent reviews on ynamides, see: a) G. Evano, A. Coste, K.
Jouvin, Angew. Chem. 2010, 122, 2902; Angew. Chem. Int. Ed.
2010, 49, 2840; b) K. A. DeKorver, H. Li, A. G. Lohse, R.
Hayashi, Z. Lu, Y. Zhang, R. P. Hsung, Chem. Rev. 2010, 110,
5064.
[2] For recent reviews on gold catalysis, see: a) A. Fꢀrstner, Chem.
Soc. Rev. 2009, 38, 3208; b) V. Michelet, P. Y. Toullec, J. P. GenÞt,
Angew. Chem. 2008, 120, 4338; Angew. Chem. Int. Ed. 2008, 47,
4268; c) E. Jimꢁnez-Nfflæez, A. M. Echavarren, Chem. Rev. 2008,
108, 3326; d) Z. Li, C. Brower, C. He, Chem. Rev. 2008, 108,
3239; e) A. Arcadi, Chem. Rev. 2008, 108, 3266; f) D. J. Gorin,
F. D. Toste, Chem. Rev. 2008, 108, 3351; g) A. S. K. Hashmi,
Chem. Rev. 2007, 107, 3180; h) A. Fꢀrstner, P. W. Davies, Angew.
Chem. 2007, 119, 3478; Angew. Chem. Int. Ed. 2007, 46, 3410.
[3] For selected gold-catalyzed transformations of ynamides, see:
a) S. Couty, C. Meyer, J. Cossy, Tetrahedron 2009, 65, 1809; b) A.
Buzas, F. Istrate, X. F. Le Goff, Y. Odabachian, F. Gagosz, J.
Organomet. Chem. 2009, 694, 515; c) A. S. K. Hashmi, M.
Rudolph, J. W. Bats, W. Frey, F. Rominger, T. Oeser, Chem.
Eur. J. 2008, 14, 6672; d) G.-Y. Lin, C.-W. Li, S.-H. Hung, R.-S.
Liu, Org. Lett. 2008, 10, 5059; e) F. Istrate, A. Buzas, I.
Dias Jurberg, Y. Odabachian, F. Gagosz, Org. Lett. 2008, 10,
925; f) A. S. K. Hashmi, R. Salathꢁ, W. Frey, Synlett 2007, 1763;
g) S. Couty, C. Meyer, J. Cossy, Angew. Chem. 2006, 118, 6878;
Angew. Chem. Int. Ed. 2006, 45, 6726; h) P. W. Davies, A.
Cremonesi, N. Martin, Chem. Commun. 2011, 47, 379.
[4] For recent contributions from our research group, see: a) S.
Kramer, J. L. H. Madsen, M. Rottländer, T. Skrydstrup, Org.
Lett. 2010, 12, 2758; b) S. Kramer, K. Dooleweerdt, A. T.
Lindhardt, M. Rottländer, T. Skrydstrup, Org. Lett. 2009, 11,
4208.
Scheme 6. Mechanistic proposal for the dimerization of aryl-substi-
tuted ynamides.
azulenes 11 might be produced by the interception of the
keteniminium species 13 by the pendant aromatic ring.
Compound 10 might be formed by the direct formation of
À
the C C bond at the ortho position of the aromatic ring
(path A) or through
a
spirocyclization/rearrangement
sequence (path B). The spiro intermediate 14 might be in
equilibrium with the fused tricyclic cationic species 15, which
could give 16 (path C). Species 16 could then produce azulene
11 after the elimination of the cationic gold fragment. The
spirocyclization pathway leading to 11 should be favored by
the presence of aryl substituents that help stabilize the
cationic charge on intermediate 14 (e.g. R = Me, OMe).
Enyne 12 might be formed by the alternative nucleophilic
addition of the pendant oxazolidinone moiety to ketenimin-
ium 13 (path D). The resulting four-membered cyclic species
17 might then rearrange to produce an allenyliminum
intermediate 18,[19] which would finally produce 12 by the
elimination of the cationic gold fragment. The formation of
enyne 12 corresponds to a formal stereoselective insertion of
[5] The cyclopentadiene core structure of compound 2a was
confirmed by X-ray crystallographic analysis (see the Supporting
Information). CCDC 811770 (2a) contains the supplementary
crystallographic data for this paper. These data can be obtained
free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
[6] See the Supporting Information for details about the optimiza-
tion of the catalytic system.
[7] N. Mꢁzailles, L. Ricard, F. Gagosz, Org. Lett. 2005, 7, 4133.
[8] Attempts to cross-dimerize ynamide 9a with 1a were also
unsuccessful.
À
the alkyne functionality into the C N bond of the ynamide.
This process should be more favored when the nucleophilicity
of the aryl group is lowered (e.g. R = F).
1
[9] The structures of 2k–n were assigned on the basis of H NMR
In conclusion, we have discovered a range of new
reactions that involve ynamides, thus showing that these
compounds can be dimerized in the presence of a catalytic
amount of an electrophilic gold(I) complex. The efficiency of
the transformation was directly dependent on the electronic
properties of the ynamide, which acts both as the electrophile
and the nucleophile in the catalytic process. Depending on the
substitution pattern of the ynamide, the common ketenimin-
ium intermediate can form various dimeric compounds by
either a 1,5-hydride shift, an arylation, or a 1,3-oxazolidinone
shift. Further studies on this new process are in progress and
will be reported in due course.
spectroscopic analysis, and confirmed by X-ray crystallographic
analysis of 2n. CCDC 811773 (2n) contains the supplementary
crystallographic data for this paper. These data can be obtained
free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
[10] Gold-catalyzed alkyne–alkyne coupling is a rare process. For an
example of intramolecular alkyne–alkyne couplings leading to
enynes, see: a) Y. Odabachian, X. F. Le Goff, F. Gagosz, Chem.
Eur. J. 2009, 15, 8966; for the cyclization of diaryldiynes, see:
b) J.-J. Lian, P.-C. Chen, Y.-P. Lin, H.-C. Ting, R.-S. Liu, J. Am.
Chem. Soc. 2006, 128, 11372; c) T. Shibata, R. Fujiwara, D.
Takano, Synlett 2005, 2062.
[11] For examples of gold(I)-catalyzed 1,5-hydride shifts, see: a) B.
Bolte, Y. Odabachian, F. Gagosz, J. Am. Chem. Soc. 2010, 132,
7294; b) Y. Odabachian, I. Dias Jurberg, F. Gagosz, J. Am.
Chem. Soc. 2010, 132, 3543; c) Y. Harrak, A. Simonneau, M.
Malacria, V. Gandon, L. Fensterbank, Chem. Commun. 2010, 46,
865; d) E. Jimꢁnez-Nfflæez, M. Raducan, T. Lauterbarch, K.
Molawi, C. R. Solorio, A. M. Echavarren, Angew. Chem. 2009,
Received: January 14, 2011
Published online: April 19, 2011
Angew. Chem. Int. Ed. 2011, 50, 5090 –5094
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5093

Communications
121, 6268; Angew. Chem. Int. Ed. 2009, 48, 6152; e) S. Bhunia,
[16] The structures of 11a–d and 12a–d were assigned on the basis of
1H NMR spectroscopic analysis, and confirmed by X-ray crys-
tallographic analysis of 11b and 12d. CCDC 811771 (11b) and
811772 (12d) contain the supplementary crystallographic data
for this paper. These data can be obtained free of charge from
The Cambridge Crystallographic Data Centre via www.ccdc.
cam.ac.uk/data_request/cif.
[17] Despite intensive investigation, we did not manage to find a
catalytic system that resulted in the selective formation of one of
the products.
R.-S. Liu, J. Am. Chem. Soc. 2008, 130, 16488.
[12] a) G. Lemi›re, V. Gandon, K. Cariou, A. Hours, T. Fukuyama,
A.-L. Dhimane, L. Fensterbank, M. Malacria, J. Am. Chem. Soc.
2009, 131, 2993; b) C.-C. Lin, T.-M. Teng, C.-C. Tsai, H.-Y. Liao,
R.-S. Liu, J. Am. Chem. Soc. 2008, 130, 16417; c) G. Lemi›re, V.
Gandon, K. Cariou, T. Fukuyama, A.-L. Dhimane, L. Fenster-
bank, M. Malacria, Org. Lett. 2007, 9, 2207; d) L. Zhang, S.
Wang, J. Am. Chem. Soc. 2006, 128, 1442.
[13] J. H. Lee, F. D. Toste, Angew. Chem. 2007, 119, 930; Angew.
Chem. Int. Ed. 2007, 46, 912.
[14] Y. Horino, T. Yamamoto, K. Ueda, S. Kuroda, F. D. Toste, J. Am.
Chem. Soc. 2009, 131, 2809.
[18] Lower combined yields were obtained when [(IPr)AuSbF6] was
used as a catalyst for the dimerization of ynamides 9b–d.
[19] R. P. Hsung, C. A. Zificsak, L.-L. Wei, C. J. Douglas, H. Xiong,
J. A. Mulder, Org. Lett. 1999, 1, 1237.
[15] E. Herrero-Gómez, C. Nieto-Oberhuber, S. López, J. Benet-
Buchholz, A. M. Echavarren, Angew. Chem. 2006, 118, 5581;
Angew. Chem. Int. Ed. 2006, 45, 5455.
5094 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 5090 –5094