Copper-catalyzed reaction cascade: Direct conversion of alkynes into N-sulfonylazetidin-2-imines

Article abstract of DOI:10.1002/anie.200503936

(Chemical Equation Presented) Selective conversion, take three: Densely functionalized azetidine derivatives are formed in an experimentally simple three-component catalytic procedure through the highly selective reaction of readily available terminal alkynes under mild conditions (see scheme). The azetidinimine products are remarkably stable to a wide range of reaction conditions and readily undergo further functionalization.

Full text of DOI:10.1002/anie.200503936

Angewandte
Chemie
Synthetic Methods
DOI: 10.1002/anie.200503936
Copper-Catalyzed Reaction Cascade: Direct
Conversion of Alkynes into N-Sulfonylazetidin-2-
imines**
Matthew Whiting and Valery V. Fokin*
As part of our ongoing research into the discovery and
development of highly efficient, versatile reactions,[1] and
inspired by the success of the copper-catalyzed 1,3-dipolar
cycloaddition reaction between terminal alkynes and azides,[2]
we became interested in the unusual reactivity of copper(i)
acetylides towards sulfonyl azides. This initially surprising
reactivity is illustrated by their conversion into N-sufonyl-
amidines when the reaction is conducted in the presence of
amines[3] and N-acylsulfonamides in the presence of water.[4]
Herein, we describe the direct, stereoselective conversion of
alkynes to N-sulfonylazetidin-2-imines by the initial reaction
of copper(i) acetylides with sulfonyl azides, followed, in situ,
by the formal [2+2] cycloaddition of a postulated N-sulfon-
ylketenimine intermediate with a range of imines.
We began our investigations by looking into the copper(i)-
catalyzed reaction of phenylacetylene with para-toluenesul-
fonyl azide. With the expectation of forming the 1-sulfonyl-4-
phenyl-1,2,3-triazole, the reaction was carried out in the
absence of a nucleophile, under conditions that were known
to promote dipolar cycloaddition between alkynes and alkyl
or aryl azides.[2a] We were therefore surprised to find that the
only isolated product was the cyclobutene derivative 1
(Scheme 1). The results of extensive NMR studies correlate
with the proposed structure.
Compound 1 could result from the dimerization of an
initially formed toluenesulfonylketenimine intermediate fol-
Scheme 1. Formation of ketenimine dimer 1. Tol=p-tolyl.
[*]Dr. M. Whiting, Prof. V. V. Fokin
Department of Chemistry
The Scripps Research Institute
10550 North Torrey Pines Road, La Jolla, CA 92037 (USA)
Fax: (+1)858-784-7562
E-mail: fokin@scripps.edu
[**]We thank the National Institutes of Health, the National Institute of
General Medical Sciences (GM28384), the Skaggs Institute for
Chemical Biology, and Pfizer, Inc. for financial support. We also
thank Prof. K. B. Sharpless for valuable advice and stimulating
discussions.
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, 3157 –3161
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3157

Communications
Table 1: N-sulfonyl azetidinimine synthesis: effect of ligands.
lowed by tautomerization, through a sequence proposed
below. Initial cycloaddition of the copper acetylide and
sulfonyl azide would give the (1-sulfonyl-4-phenyltriazol-5-
yl)copper intermediate 2. Direct elimination of dinitrogen, as
previously reported for a range of 5-lithiated-triazoles,[5]
would generate copper alkynamide 3. Protonation of this
species would then furnish the requisite sulfonylketenimine 4
with concomitant liberation of the copper catalyst. Alterna-
tively the cuprated triazole 2 could undergo ring-chain
isomerization, which is known to be rapid for electron-
deficient triazoles, especially those with an electron-with-
drawing group at N-1.[6] This route would give rise to the
cuprated diazoimine 5, which upon loss of dinitrogen and
protonation would again generate the N-sulfonylketenimine 4
and regenerate the copper catalyst (Scheme 2).
Entry
Ligand[a]
t [h]
trans-6:cis-6:7[b]
1
2
3
4
6
7
8
9
2,6-lutidine
none
pyridine
2,6-di-tert-butylpyridine
neocuproine
Et3N
3
16
3
16
80:13:7
56:11:33
95:5:0
45:10:45
–
94:6:0
83:17:0[c]
42:58:0
no reaction
3
3
<2
TMEDA
TBTA[d]
[a]1 equiv of ligand used. [b]Determined from the 1H NMR spectra of
the crude reaction product. [c]The oxidatively coupled di-alkyne was also
observed. [d]0.1 equiv of ligand used. TMEDA =N,N,N’,N’-tetramethyl-
ethylenediamine,
amine.
The isolation of 1 was significant in itself, as there are very
few reports of ketenimine dimerization and all proceed across
at least one of the cumulenic C N bonds,[7] as opposed to both
=
TBTA=tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl]-
=
C C bonds observed herein. Of greater synthetic interest was
absence of an added ligand and furnished an
increased amount of triazole by-product
(entry 2). The use of pyridine completely
suppressed triazole formation and gave the
desired azetidinimine as a 95:5 ratio of
isomers in favor of the trans product
(entry 3). A number of other ligands were
screened, but all gave a reduction in the
selectivity or the rate of reaction (entries 4–
9). Interestingly TBTA,[10] a potent ligand
for the CuI-catalyzed azide/alkyne cyclo-
addition, not only increased the rate of
azetidinimine formation, but also reversed
the selectivity, giving a slight excess of the
cis isomer (entry 9). The observed depend-
ence of the reaction selectivity upon the
additive is not yet fully understood and is
the subject of continued investigation. It
does however imply that the amine, either
itself directly or as a ligand for copper, is
Scheme 2. Possible mechanistic pathways leading to 1.
the possibility of capturing the N-sulfonylketenimine inter-
mediates by alternative means. Ketenimines with N-alkyl or
-aryl substituents are considerably less reactive than their
oxygen analogues, ketenes, although they are still known to
undergo a range of useful transformations,[7a,8] particularly in
an intramolecular fashion.[7a] Introduction of an N-sulfonyl
substituent leads to a marked increase in reactivity. For
example, N-sulfonylketenimines have been shown by Ghosez
and co-workers to readily undergo intermolecular Staudinger-
type [2+2] cycloadditions with imines to furnish azetidini-
mines.[9]
We were pleased to find that carrying out the reaction in
the presence of N-benzylideneaniline gave the expected
azetidinimine product 6 as a 6:1 mixture of the trans and cis
isomers, along with a small amount of the 1,4-disubstituted
triazole 7 (Table 1, entry 1). The use of alternative ligands/
bases in the reaction gave greatly altered selectivities. For
example, the reaction proceeded much more slowly in the
involved in promotion of the [2+2] process.
Reducing the amount of pyridine resulted in a slight
reduction in the stereoselectivity, while increasing it to two
equivalents gave an even more selective reaction that
furnished the azetidinimine with greater than 95:5 selectivity
in favor of the trans isomer. With regard to the reaction
solvent, more polar solvents tended to give reduced trans:cis
selectivity, and less polar solvents generally required longer
reaction times. None of those tested, however, were as
efficient or resulted in as high a selectivity as acetonitrile.
Under the optimal conditions (1 equiv sulfonyl azide, 2 equiv
pyridine, 1 equiv alkyne, 1.2 equiv imine, 10 mol% CuI, RT,
16 h), 6 was obtained in 90% yield and in greater than 95%
purity by simple dilution of the reaction mixture with 1m HCl
and collection of the precipitated product by filtration
(Table 2, entry 1).
A brief survey of the scope with regard to the alkyne
component revealed that both alkyl and aryl alkynes gave the
3158
www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 3157 –3161

Angewandte
Chemie
Table 2: Synthesis of azetidinimines: scope with respect to the alkyne
component.
aromatic aldehydes appeared to be similar, and equally high
selectivities were observed (Table 3, entries 1 and 2). Incor-
poration of the readily cleavable benzyl (entry 3) or para-
methoxyphenyl (PMP, entry 4) groups on the nitrogen atom
led to slightly reduced selectivities, with PMP being the more
selective of the two. The use of more elaborate N-PMP imines
gave rise to highly functionalized products in good yield,
albeit with slightly reduced selectivities (entries 5 and 6). As
expected, the electron-deficient benzylidenebenzenesulfona-
mide (entry 7) did not participate in the reaction, presumably
because of the low nucleophilicity of its nitrogen atom. The
use of aldimines and ketimines bearing a protons gave rise to
alternative modes of reaction; however, imines derived from
ethyl glyoxylate gave the desired products with inverted
selectivity of < 5:95 trans:cis (entries 8 and 9).
Entry
R
Yield [%][a]
trans:cis[b]
1
2
3
4
5
6
7
8
9
10
Ph
90
90
79
69
81
62
60
55
55
48
>95:5
>95:5
>95:5
90:10
>95:5
92:8
>95:5
>95:5
>95:5
25:75
4-BrC6H4
2-(CF3)C6H4
4-(MeO)C6H4
3-Pyr
Bn
PhthCH2
CH2Cl(CH2)2
MeO2C(CH2)3
TMS
From our initial studies the reaction appears to be
amenable to the use of a wide range of sulfonyl azides
(Table 4). For example, the desired azetidinimines product
[a]Yield of the isolated isomeric mixture. [b]Determined from the
1H NMR spectra of the crude reaction product. Bn=benzyl, Phth=
phthalimido, TMS=trimethylsilyl.
Table 4: Synthesis of azetidinimines: scope with respect to the sulfonyl
azide component.
desired product, generally with high trans selectivity
(Table 2). The electron-rich (para-methoxyphenyl)acetylene
(entry 4) and 3-phenyl-1-propyne (entry 6) gave slightly
reduced selectivities, and in the case of trimethylsilylacetyl-
ene the selectivity was reversed to give a 3:1 cis:trans ratio of
products (entry 10), possibly as the silyl substituent promotes
a more concerted [2+2] reaction that leads to a greater
amount of the kinetic product. A wide range of functionality
was accepted; however incorporation of nucleophilic groups
tended to give mixtures of products arising from intra- or
intermolecular nucleophilic trapping of the ketenimine inter-
mediate.
Entry
R
Yield [%][a]
trans:cis[b]
1
2
3
4
4-(NO2)C6H4
4-BrC6H4
4-IC6H4
Me
63
77
76
68
>95:5
>95:5
>95:5
95:5
[a]Yield of isolated isomeric mixture. [b]Determined from the 1H NMR
spectra of the crude reaction product.
We have also examined the scope with regard to the imine
and the azide components (Tables 3 and 4, respectively). The
reactivity of imines derived from electron-rich or -poor
bearing a readily cleavable nitrobenzenesulfonyl (nosyl, Ns)
group on the imino nitrogen atom was obtained from p-
nitrobenzenesulfonyl azide with high selectivity (entry 1). The
use of para-bromobenzenesulfonyl azide and para-iodoben-
zenesulfonyl azide gave equally selective conversion into the
expected halogenated products (entries 2 and 3). The reaction
also proceeded well when alkylsulfonyl azides were used
(entry 4).
Table 3: Synthesis of azetidinimines: scope with respect to the imine
component.
Further support for the proposed mechanism was
obtained by conversion of the isolated 1-sulfonyltriazole 8
into the corresponding azetidinimine product 9 (the triazole 8
was isolated as a minor product (17%) from the reaction of
benzenesulfonyl azide with phenylacetylene and N-benzyl-
ideneaniline, carried out in the presence of one equivalent of
2,6-di-tert-butylpyridine.[11]) Although 8 was stable when
resubjected to the reaction conditions, metalation at the 5-
position using nBuLi in THF resulted in the immediate
extrusion of N2, even at À788C, presumably with formation of
the relatively stable metalated ketenimine 10.[12] Addition of
N-benzylideneaniline and one equivalent of anhydrous HCl
gave rise to the azetidinimine product 9 which could be
isolated in 20% yield as an 85:5 mixture of isomers favoring
the cis product (Scheme 3).
Entry
R1
R2
Yield [%][a]
trans:cis[b]
1
2
3
4
5
6
7
8
9
4-FC6H4
4-(MeO)C6H4
Ph
Ph
Ph
Bn
87
79
80
77
73
65
>95:5
>95:5
89:11
92:8
86:14
77:23
–
Ph
4-(MeO)C6H4
4-(MeO)C6H4
4-(MeO)C6H4
SO2Ph
4-(MeO2C)C6H4
4-(TMSC C)C6H4
Ph
EtO2C
EtO2C
ꢀ
ca. 5[b]
53
63
Ph
<5:95
<5:95
4-(MeO)C6H4
[a]Yield of isolated isomeric mixture. [b]Determined from the 1H NMR
spectra of the crude reaction product.
Angew. Chem. Int. Ed. 2006, 45, 3157 –3161
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3159

Communications
treatment with ceric ammonium nitrate
(CAN) (Scheme 7).[17]
In summary, we have developed an
experimentally simple catalytic procedure
for the highly selective conversion of
alkynes into N-sulfonylazetidin-2-imines
under mild conditions. This three compo-
Scheme 3. Conversion of 1-sulfonyl-4-phenyltriazole 8 into azetidinimine 9.
The sulfonyl azetidinimne products described above
can be viewed as b-lactam analogues, and as such they
have potential applications as therapeutic agents.[13]
They are highly stable under acidic conditions: the
starting material was quantitatively recovered after
stirring in HCl (2m, dioxane/water) overnight. Moder-
ately basic conditions (K2CO3 in dioxane/water) did not
cause any degredation either. However, prolonged
exposure to KOH under similar conditions led to
various hydrolysis products.
The stability of azetidinimines to a wide range of
reaction conditions makes them valuable synthetic
intermediates. Several examples of their selective func-
tionalization (see below) illustrate their synthetic utility.
Thus, deprotonation of 6 with KH followed by quench-
ing with allyl bromide furnished 11 in 92% yield as a
single diastereomer, presumably that with the cis
configuration (Scheme 4).
Scheme 6. Further functionalization of 15.
Scheme 7. Cleavage of the N-1-PMP group.
Scheme 4. Acid/base stability and allylation of 6.
nent process is thought to proceed through the initial reaction
of in situ generated copper(i) acetylides with sulfonyl azides
to give transient (1-sulfonyltriazolyl)copper intermediates
which, upon extrusion of dinitrogen, generate N-sulfonylke-
tenimines. The azetidin-
imine products are remarkably stable to a wide range of
reaction conditions and readily undergo further functionali-
zation. This newly discovered reaction sequence rapidly
constructs densely functionalized azetidine derivatives from
readily available terminal alkynes in just one simple step and
should prove useful for exploring the utility of these four-
membered heterocycles.
Cleavage of the p-Ns group[14] from 12 furnished the NH
imine 13 which reacts readily with activated carbonyl com-
pounds. For example, carbamate 14 was obtained in 66%
yield over the two steps (Scheme 5).
Received: November 7, 2005
Published online: March 29, 2006
Scheme 5. Cleavage of the p-nitrobenzenesulfonyl (Ns) group and
formation of a carbamate.
Keywords: alkynes · azides · cycloaddition ·
.
homogeneous catalysis · small ring systems
Aryl halide functionality can be easily converted into a
range of alternative products. As an example, iodide 15 was
converted into azide 16 by copper-catalyzed azidation,[15] or to
the biaryl derivative 17 by using a Suzuki coupling with a
boronic acid (Scheme 6).[16] 1H-Azetidinimines, such as 19,
are easily accessible from N-1-PMP derivatives (18) by
[1] H. C. Kolb, M. G. Finn, K. B. Sharpless, Angew. Chem. 2001, 113,
2056 – 2075; Angew. Chem. Int. Ed. 2001, 40, 2004 – 2021.
[2] a) V. V. Rostovtsev, L. G. Green, V. V. Fokin, K. B. Sharpless,
Angew. Chem. 2002, 114, 2708 – 2711; Angew. Chem. Int. Ed.
3160
www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 3157 –3161

Angewandte
Chemie
2002, 41, 2596 – 2599; b) C. W. Tornøe, C. Christensen, M.
Meldal, J. Org. Chem. 2002, 67, 3057 – 3064.
[3] I. Bae, H. Han, S. Chang, J. Am. Chem. Soc. 2005, 127, 2038 –
2039.
[4] a) S. H. Cho, E. J. Yoo, I. Bae, S. Chang, J. Am. Chem. Soc. 2005,
127, 16046 – 16047; b) M. Cassidy, J. Raushel, V. V. Fokin,
Angew. Chem. 2006, 118, 3226; Angew. Chem. Int. Ed. 2006,
45, 3154.
[5] a) R. Raap, Can. J. Chem. 1971, 49, 1792 – 1798; b) T. L. Gil-
christ, Adv. Heterocycl. Chem. 1987, 41, 41 – 74.
[6] W.-Q. Fan, A. R. Katritzky in Comprehensive Heterocyclic
Chemistry II, Vol. 4 (Eds: A. R. Katritzky, C. W. Rees, E. F. V.
Scriven), Pergamon, Oxford, 1996, chap. 1.
[7] a) M. Alajarin, A. Vidal, F. Tovar, Targets Heterocycl. Syst. 2000,
4, 293 – 326, and references therein; b) E. Langhals, R. Huisgen,
K. Polborn, Chem. Eur. J. 2004, 10, 4353 – 4357.
[8] For leading references see: a) G. R. Krow, Angew. Chem. 1971,
83, 455; Angew. Chem. Int. Ed. Engl. 1971, 10, 435 – 449; b) G.
Barbaro, P. Giorgianni, D. Giacomini, Tetrahedron 1993, 49,
4293 – 4306; c) A. H. Mermerian, G. C. Fu, Angew. Chem. 2005,
117, 971 – 974; Angew. Chem. Int. Ed. 2005, 44, 949 – 952.
[9] A. Van Camp, D. Goosens, M. Moya-Portuguez, J. Marchand-
Brynaert, L. Ghosez, Tetrahedron Lett. 1980, 21, 3081 – 3084.
[10] T. R. Chan, R. Hilgraf, K. B. Sharpless, V. V. Fokin, Org. Lett.
2004, 6, 2853 – 2855.
[11] The major product, N-(1,3,4-triphenyl-azetidin-2-ylidene)-ben-
zenesulfonamide, was obtained in 71% yield as a 90:10 mixture
of isomers in favor of the trans product. The regiochemistry of
the triazole by-product was confirmed by unambiguous synthesis
of the 1,5-regioisomer and comparison of their spectral data. See
the Supporting Information for further details.
[12] K. Sung, J. Chem. Soc. Perkin Trans. 2 1999, 1169 – 1173.
[13] a) Comprehensive Heterocyclic Chemistry II, Vol. 1 (Eds: A. R.
Katritzky, C. W. Rees, E. F. V. Scriven), Pergamon, Oxford,
1996, chap. 18–20; b) Chemistry and Biology of b-Lactam Anti-
biotics, Vol. 1–3 (Eds: R. B. Morin, M. Goldman), Academic
Press, New York, 1982.
[14] T. Fukuyama, C.-K. Jow, M. Cheung, Tetrahedron Lett. 1995, 36,
6373 – 6374.
[15] a) W. Zhu, D. Ma, Chem. Commun. 2004, 888; b) J. Andersen, U.
Madsen, F. Björkling, X. Liang, Synlett 2005, 2209 – 2213.
[16] N. Miyaura, A. Suzuki, Chem. Rev. 1995, 95, 2457 – 2483.
[17] D. R. Kronenthal, C. Y. Han, M. K. Taylor, J. Org. Chem. 1982,
47, 2765 – 2768.
Angew. Chem. Int. Ed. 2006, 45, 3157 –3161
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3161