Catalytic formal homo-Nazarov cyclization

Article abstract of DOI:10.1021/ol802970g

The first catalytic method for the cyclization of vinyl-cyclopropyl ketones (formal homo-Nazarov reaction) is reported. Starting from activated cyclopropanes, heterocyclic, and carbocyclic compounds were obtained under mild conditions using Bronsted acid catalysts. Preliminary investigation of the reaction mechanism indicated a stepwise process.

Full text of DOI:10.1021/ol802970g

ORGANIC
LETTERS
2009
Vol. 11, No. 4
1023-1026
Catalytic Formal Homo-Nazarov
Cyclization
Filippo De Simone, Julien Andre`s, Riccardo Torosantucci, and Je´roˆme Waser*
Laboratory of Catalysis and Organic Synthesis Ecole Polytechnique Fe´de´rale de
Lausanne EPFL SB ISIC LCSO, BCH 4306, 1015 Lausanne, Switzerland
jerome.waser@epfl.ch
Received December 26, 2008
ABSTRACT
The first catalytic method for the cyclization of vinyl-cyclopropyl ketones (formal homo-Nazarov reaction) is reported. Starting from activated
cyclopropanes, heterocyclic, and carbocyclic compounds were obtained under mild conditions using Brønsted acid catalysts. Preliminary
investigation of the reaction mechanism indicated a stepwise process.
Carbocyclic and heterocyclic scaffolds occupy a privileged
position in both natural products and pharmaceuticals.1
Consequently, the development of cyclization and cycload-
dition reactions for the efficient formation of cyclic structures
is a very important goal in organic chemistry. In this respect,
the development of new, highly stereoselective catalytic
methods is crucial to allow a more efficient and environ-
mentally friendly access to polycyclic molecules.2
One classical approach toward the construction of cyclo-
pentenone rings is the Nazarov reaction, which is the
electrocyclic ring closure of a pentadienyl cation, followed
by proton transfer (A, Scheme 1).3 The potential of the
Scheme 1. Nazarov and Homo-Nazarov Cyclizations
(1) Clardy, J.; Walsh, C. Nature 2004, 432, 829.
(2) (a) Balskus, E. P.; Jacobsen, E. N. Science 2007, 317, 1736. (b)
Mohr, J. T.; Krout, M. R.; Stoltz, B. M. Nature 2008, 455, 323.
(3) (a) Nazarov, I. N.; Zaretskaya, I. I. IzV. Akad. Nauk. SSSR. Ser. Khim.
1941, 211. (b) Habermas, K. L.; Denmark, S. E.; Jones, T. K. Org. React.
(N. Y.) 1994, 45, 1–158. (c) Giese, S.; West, F. G. Tetrahedron 2000, 56,
10221. (d) Wang, Y.; Schill, B. D.; Arif, A. M.; West, F. G. Org. Lett.
2003, 5, 2747. (e) Aggarwal, V. K.; Beffield, A. J. Org. Lett. 2003, 5, 5075.
(f) Bee, C.; Leclerc, E.; Tius, M. A. Org. Lett. 2003, 5, 4927. (g) He, W.;
Sun, X. F.; Frontier, A. J. J. Am. Chem. Soc. 2003, 125, 14278. (h) Janka,
M.; He, W.; Frontier, A. J.; Eisenberg, R. J. Am. Chem. Soc. 2004, 126,
6864. (i) Malona, J. A.; Colbourne, J. M.; Frontier, A. J. Org. Lett. 2006,
8, 5661. (j) He, W.; Herrick, I. R.; Atesin, T. A.; Caruana, P. A.;
Kellenberger, C. A.; Frontier, A. J. J. Am. Chem. Soc. 2008, 130, 1003. (k)
Liang, G. X.; Gradl, S. N.; Trauner, D. Org. Lett. 2003, 5, 4931. (l) Liang,
G. X.; Trauner, D. J. Am. Chem. Soc. 2004, 126, 9544. (m) Walz, I.;
Bertogg, A.; Togni, A. Eur. J. Org. Chem. 2007, 2650. (n) Walz, I.; Togni,
A. Chem. Commun. 2008, 4315. (o) Rueping, M.; Ieawsuwan, W.;
Antonchick, A. P.; Nachtsheim, B. J. Angew. Chem., Int. Ed. 2007, 46,
2097. (p) Amere, M.; Blanchet, J.; Lasne, M. C.; Rouden, J. Tetrahedron
Lett. 2008, 49, 2541. For reviews, see: (q) Tius, M. A. Eur. J. Org. Chem.
2005, 2193. (r) Frontier, A. J.; Collison, C. Tetrahedron 2005, 61, 7577.
(s) Pellissier, H. Tetrahedron 2005, 61, 6479.
Nazarov cyclization was recognized at an early stage in
organic synthesis. Solutions to control the termination of the
reaction were devised several decades ago,3b but the necessity
of using a stoichiometric amount of strong Lewis or Brønsted
acids has limited the use of this reaction. However, in the
last five years, the first examples of catalytic Nazarov
10.1021/ol802970g CCC: $40.75
Published on Web 01/29/2009
2009 American Chemical Society

reactions using milder Lewis3c-n or Brønsted3o,p acids were
reported, together with the first examples of asymmetric
induction.3e,k,l,n,o
Nazarov cyclization: stoichiometric SnCl4.6b-e Using these
conditions, complete polymerization of the sensitive substrate
was observed.8
When considering these recent successes in the Nazarov
reaction, we wondered if similar concepts could be successful
in other cyclization reactions to access larger ring systems.
A viable approach to access homologous rings via electro-
cyclic reactions is the substitution of a double bond by a
cyclopropyl group, as exemplified by the divinylcyclopropyl
rearrangement.4 Intra- and intermolecular ring-opening of
cyclopropyl ketones and diesters have been examined ex-
tensively.5 The reaction of vinyl-cyclopropyl ketones has
been less studied (B, Scheme 1).6 Tsuge has reported the
cyclization of vinyl-cyclopropyl ketones using an excess of
polyphosphoric acid at 80 °C, but this reaction was not
general and several other products were obtained beside the
desired cyclohexenones.6a More work has been done on
the related aryl-cyclopropyl ketones, first by Murphy for the
synthesis of tetralones using an excess of SnCl4 as
reagent.6b-d During completion of our work, Yadav also
demonstrated that diverse polycyclic heterocycles could be
accessed using 3 equivalents SnCl4 at 80 °C, but no vinyl-
cyclopropyl ketones were reported.6e Up to now, the harsh
conditions needed have limited the use of the homo-Nazarov
reaction in organic synthesis. Herein, we report the first
example of a catalytic formal homo-Nazarov process for non
aromatic substrates which lead to the formation of valuable
polycyclic cyclohexenones at room temperature as well as
preliminary experiments to probe the reaction mechanism.
Inspired by recent progress in the catalytic Nazarov
reaction,3j we decided to examine dihydropyran-derived
substrate 2a (Scheme 2). Substrate 2a was synthesized from
As most Lewis acid led to extensive polymerization, we
then turned toward Brønsted acid catalysts. The pKa value
of the catalyst had a strong influence on the outcome of the
reaction: sulfuric and toluenesulfonic acids were optimal.
Stronger acids led to decomposition of the starting material
and no full conversion could be achieved with weaker acids.
Examination of solvent effects showed that the reaction was
faster in noncoordinating solvents, like dichloromethane, but
polymerization was also difficult to suppress. Acetonitrile
finally offered the best compromise, with sufficient reactivity
but less pronounced polymerization. The cyclization of 2a
in acetonitrile with 20 mol % toluenesulfonic acid at room
temperature led to the formation of the desired cyclohex-
enone 3a in 70% isolated yield (Scheme 2).
The scope of the reaction was examined next (Table 1).
Variation of the aromatic substituent on the cyclopropane
confirmed the importance of its electron-donating ability;
whereas no reaction was observed with a simple phenyl
group (entry 2), a quantitative yield was observed with a
3,4- or 2,4- dimethoxyphenyl group (entries 3 and 4). This
result is noteworthy, as electron-rich aromatic substituents
are well represented in bioactive natural products9 and are
easily oxidized to the corresponding carboxylic acids.10
A
furan group was also tolerated at this position, although the
yield was moderate due to partial polymerization (entry 5).
Finally, the influence of a methyl group R to the ketone
was examined. Interestingly, a strong accelerating effect was
observed and cyclohexenone 3f was obtained in quantitative
yield after only 15 min (entry 6). A plausible explanation
would be a faster ring-opening of the cyclopropane ring due
to sterical strain release and the higher stability of the formed
enol intermediate.11 Importantly, this accelerating effect on
the formal homo-Nazarov cyclization has never been reported
before.
Scheme 2. Synthesis and Cyclization of Model Substrate 2a
We then examined variation of the electron-rich side of
the ketone. A dihydrofuran group proved to be more prone
to polymerization and the desired product was isolated only
in low yield with a 4-methoxyphenyl group on the cyclo-
propane (entry 7). The stronger stabilizing effect of the 2,4-
dimethoxyphenyl substituent allowed the isolation of the
desired 5-6 ring system in quantitative yield (entry 8).
Replacing the dihydropyran group with an electron-rich
N-methylindole heterocycle lead to an efficient cyclization
in quantitative yield (entry 9), but only polymerization was
observed with a benzofuran ring (entry 10). Interestingly,
similar results were obtained in the related Nazarov
cyclization.3i
Weinreb amide 1 via Corey-Chaykovsky cyclopropanation7
followed by addition of a lithiated nucleophile to afford 2a
in good yield (Scheme 2).
With our model substrate in hand, we began our studies
by examining the most frequently used procedure for homo-
(4) Piers, E. In ComprehensiVe Organic Synthesis; Trost, B. M., Ed.;
Pergamon Press: New York, 1991; Vol. 5, pp 971-998.
(5) For a few selected examples, see: (a) Stork, G.; Marx, M. J. Am.
Chem. Soc. 1969, 91, 2371. (b) Grieco, P. A.; Finkelhor, R. S. Tetrahedron
Lett. 1974, 527. (c) Pohlhaus, P. D.; Sanders, S. D.; Parsons, A. T.; Li, W.;
Johnson, J. S. J. Am. Chem. Soc. 2008, 130, 8642.
(8) Oligomerization, then polymerization was apparent in 1H NMR via
formation of broad signals in several regions of the spectra, see Supporting
Information (Figure S5).
(6) (a) Tsuge, O.; Kanemasa, S.; Otsuka, T.; Suzuki, T. Bull. Chem.
Soc. Jpn. 1988, 61, 2897. (b) Murphy, W. S.; Wattanasin, S. Tetrahedron
Lett. 1980, 21, 1887. (c) Murphy, W. S.; Wattanasin, S. J. Chem. Soc. Perkin
Trans. 1 1981, 2920. (d) Murphy, W. S.; Wattanasin, S. J. Chem. Soc.,
Perkin Trans. 1 1982, 1029. (e) Yadav, V. K.; Kumar, N. V. Chem.
Commun. 2008, 3774.
(9) For example in Podophyllotoxin natural products and their deriva-
tives: Bohlin, L.; Rosen, B. Drug DiscoV. Today 1996, 1, 343.
(10) (a) Carlsen, P. H. J.; Katsuki, T.; Martin, V. S.; Sharpless, K. B. J.
Org. Chem. 1981, 46, 3936. (b) Voight, E. A.; Rein, C.; Burke, S. D. J.
Org. Chem. 2002, 67, 8489.
(11) As an alternative explanation, a higher fraction of the more stable
enol tautomer could be envisaged to favor cyclization.
(7) Rodriques, K. E. Tetrahedron Lett. 1991, 32, 1275.
1024
Org. Lett., Vol. 11, No. 4, 2009

electronic constraints which limit the number of structures
that can be synthesized. The use of substrates lacking an
electron-donating heteroatom on the double-bond is highly
desirable. Based on the seminal work of Denmark on silyl
group-directed Nazarov reactions,3b,12 we decided to use an
allyl silane group to enhance the nucleophilicity of the double
bond and favorize cyclization (Scheme 3).
Table 1. Scope of the Formal Homo-Nazarov Cyclization
Scheme 3. Cyclization for Carbocycles Synthesis
Gratifyingly, submitting vinyl-cyclopropyl ketone 11 to
the optimized reaction conditions led to the formation of
bicyclic ketone 12 in 30 min and 55% yield. The regiose-
lectivity of the double bond formation was completely
controlled by the elimination of the silyl group. Interestingly,
only one diastereoisomer of 12 was isolated. The structure
of 12 was tentatively assigned by NMR experiments (COSY,
HSQC, NOESY).13 This preliminary result held promises
for the application of the method in the synthesis of
carbocyclic compounds.
The strong influence of electron-donating groups and acid
strength on the reaction rate led us to propose a tentative
stepwise mechanism for the reaction with cyclopropane
opening as rate-limiting (Scheme 4).
Scheme 4. Postulated Mechanism
In order to further support this mechanism, the following
experiments were performed (Scheme 5): (1) The reaction
a Reaction conditions: 0.4 mmol substrate in 8 mL CH3CN with 20
mol % TsOH at 23 °C.
(12) Denmark, S. E.; Klix, R. C. Tetrahedron 1988, 44, 4043
.
(13) The obtained 2D NMR data strongly support the proposed structure
assignment for 12. Further confirmation of the structure will be attempted
by X-rays analysis of the corresponding thiosemicarbazone, a procedure
developed by Denmark.12
To further increase the versatility of the formal homo-
Nazarov process, it would be important to diminish the strong
Org. Lett., Vol. 11, No. 4, 2009
1025

but this result is difficult to interpret due to fast proton
exchange between substrate and catalyst. (3) Attempts were
made to trap the proposed intermediates (enol and carboca-
tion) of the catalytic cycle.14 With substrate 2a, all nucleo-
philic (water, allyl silane, butyl vinyl ether) and electrophilic
(benzaldehyde, ethyl glyoxalate, acetic anhydride, ethyl
acrylate) trapping agents tested so far were not successful.
With cyclohexene derivative 15, however, alcohol 16 was
obtained in 31% yield when the reaction was conducted in
the presence of water.
Scheme 5. Mechanism Investigation: (1) Van’t Hoff Plot and
Reaction Order, (2) Deuterium, and (3) Trapping Experiments
All the data collected so far are in agreement with a rate-
determining cyclopropane opening, followed by a fast
cyclization. In the case were the cyclization is too slow (as
with 15), polymerization can occur instead of the desired
process. We speculate that key for catalysis is the fast
tautomerization of the enol intermediates, which contrasts
with the strong binding of stoichiometric reagents like SnCl4,
which prevent catalytic turnover.
In summary, we have reported the first catalytic formal
homo-Nazarov process. We have demonstrated that principles
successful in the corresponding Nazarov reaction could also
be applied to vinyl-cyclopropyl ketones, which allow the first
high-yielding cyclization reaction for this class of substrates
under mild conditions. First investigations of the reaction
mechanism seem to indicate a stepwise mechanism with a
rate-limiting cyclopropane ring opening: consequently, the
reaction is mechanistically different from the classical
Nazarov cyclization. Our future work will focus on the
development of asymmetric variations as well as on applica-
tions in the synthesis of natural products and their analogs.
1
kinetic was followed via H NMR spectroscopy, and the
reaction was found to be first order in tosic acid for substrate
2a. (2) The use of stoichiometric deuterated tosic acid
resulted in a mixture of nondeuterated, mono and bis-
deuterated products at the R position to the ketone. A control
experiment showed that no deuterium exchange was observed
for the isolated cyclization product in the presence of
deuterated tosic acid. A possible explanation for this surpris-
ing results would be the intramolecular attack of the oxygen
atom of intermediates Ia or Ib to form an oxonium
intermediate IIIa. From IIIa, proton-deuterium exchange
should be easy via dihydrofuran IIIb. Dihydrofuran products
have indeed been isolated in the related stoichiometric
reaction of aryl vinyl ketones.6c Alternatively, proton ex-
change could be more rapid on cyclized intermediate IIb
and IIa. Proton lost form IIb, or proton lost followed by
tautomerization from IIa would then lead to the cyclohex-
enone product. Additionally, a weak kinetic isotope effect
(1.15) was observed using 40 mol % deuterated tosic acid,
Acknowledgment. The EPFL is acknowledged for finan-
cial support. We thank Prof. K. Gademann, Dr. T. Woods
and Dr. H. Jessen from the Chemical Synthesis Laboratory
at the EPFL for proofreading this manuscript.
Supporting Information Available: Experimental pro-
cedures, spectroscopic information for new compounds and
kinetic data. This material is available free of charge via the
Internet at http://pubs.acs.org.
OL802970G
(14) To gain stronger evidence for the cationic mechanism, we will study
the influence of the cyclopropane stereochemistry on the reaction outcome,
for example using 11 with a cis-substituted cyclopropane.
1026
Org. Lett., Vol. 11, No. 4, 2009