A silyl-mediated [3+2] photochemical cycloaddition

Article abstract of DOI:10.1139/V06-077

The generation of larger rings from intermediate cyclobutanes via a two-step [2+2] photocycloaddition - ring expansion, known as the de Mayo reaction, has been widely applied in synthesis. Herein a one-step synthesis of cyclopentanoids has been developed

Full text of DOI:10.1139/V06-077

1259
A silyl-mediated [3+2] photochemical
cycloaddition1
Michael G. Organ and Debasis Mallik
Abstract: The generation of larger rings from intermediate cyclobutanes via a two-step [2+2] photocycloaddition – ring
expansion, known as the de Mayo reaction, has been widely applied in synthesis. Herein a one-step synthesis of
cyclopentanoids has been developed based on the photochemical irradiation of an enone in the presence of an
allylsilane. The ability of the silyl moiety to stabilize the intermediate biradical is believed to be responsible for this
unique transformation where one normally expects to see the [2+2] cyclobutane adduct.
Key words: [2+2] photocycloaddition – ring expansion, allylsilanes, cyclopentanoids, photochemical.
Résumé : La génération de cycles plus grands à partir d’intermédiaires cyclobutanes, par le biais d’une séquence en
deux étapes comportant une photocycloaddition [2+2] suivie expansion de cycle, connue sous le nom de réaction de de
Mayo, est très souvent utilisée en synthèse. Dans ce travail, on décrit une synthèse en une étape de cyclopentanoïdes
qui repose sur l’irradiation photochimique d’une énone en présence d’un allylsilane. On croit que la possibilité pour la
portion silane de stabiliser l’intermédiaire biradicalaire est responsable de cette transformation unique au cours de la-
quelle on s’attend normalement à obtenir une addition cyclobutane [2+2].
Mots clés : photocycloaddition [2+2] – expansion de cycle, allylsilanes, cyclopentanoïdes, photochimie.
[Traduit par la Rédaction] Organ and Mallik 1262
Since the seminal work on the allylation reaction between
allyltrimethylsilane and α,β-unsaturated ketones was dis-
closed by Hosomi and Sakurai (1), there has been a great
deal published regarding the expansion of the scope and util-
ity of allylsilanes in such reactions (2). This methodology
has been expanded to include ring formation when the allyl-
silanes involved possess “bulky groups” on silicon to pro-
vide cyclobutane and cyclopentane products via Lewis acid
catalyzed [2+2] and [3+2] annulation, respectively (3–5).
The general addition process for allylsilanes onto unsatu-
rated substrates is illustrated in Scheme 1. Siliranium ion in-
termediate 3 is central to all possible products of initial
conjugate addition. In the case of enone substrates (X =
alkyl), if Y is large (e.g., isopropyl) the principal product
isolated is the five-membered ring adduct 6, which appears
to be both the thermodynamic and kinetic product (6, 7).
When Y is methyl, the allylation product 4, i.e., the Sakurai
product, is the major adduct isolated. In the case of unsatu-
rated ester substrates (e.g., X = OR), the formation of both
four- and five-membered rings is also with precedent (8, 9).
In these situations, the cyclobutane is typically the kinetic
product and the cyclopentane 6 forms from 5 over time in
the presence of the Lewis acid, presumably via intermediate 3.
Scheme 1.
–
ML
+
L
+
ML
R1
n-1
n-1
O
R
O
R
O
X
ML
n
-Y3SiL
SiY3
X
X
R
1
R1
R1
3
4
+
O
SiY3
SiY3
2
X
MLn-1
R
O
R1
X
SiY3
6
R
R1
5
syn and anti
The primary driving force behind nucleophilic reactions
involving allylsilanes is the presence of the silicon moiety,
which activates the olefin and stabilizes the intermediate
carbocation that is generated β to itself during the addition
process by strong hyperconjugative effects (see intermediate 3)
(10). Thus, the silyl group is electron-rich and strongly
σ-donating. The larger and more electron-rich the ligands
on silicon, the more nucleophilic, hence reactive, the
allylsilane becomes (11). Further, the larger the ligands
on silicon, the more the silyl moiety resists proto-
desilylation.
It is known that radicals show similar stabilization as do
cations when positioned β to silicon, but little has been done
with this synthetically. We were interested in trying to ex-
ploit this feature of silicon in the development of new cycli-
zation reactions. The photochemical excitation of enones in
Received 14 October 2005. Accepted 12 April 2006.
Published on the NRC Research Press Web site at
http://canjchem.nrc.ca on 20 July 2006.
M.G. Organ2 and D. Mallik. Department of Chemistry,
York University, 4700 Keele Street, Toronto,ON M3J 1P3,
Canada.
1This article is part of a Special Issue dedicated to Dr. Alfred
Bader.
2Corresponding author (e-mail: organ@yorku.ca).
Can. J. Chem. 84: 1259–1262 (2006)
doi:10.1139/V06-077
© 2006 NRC Canada

1260
Can. J. Chem. Vol. 84, 2006
Scheme 2.
O
SiR3
path a
SiR3
hv
O
O
b
9
SiR
a
3
O
7
8
path b
SiR3
10
the presence of ground-state alkenes is a reliable and well-
established method for the production of radicals. If the
alkene was in fact an allylsilane, then the silyl component
could become directly involved during the photoaddition it-
self, thus giving rise to new ringed structures directly upon
irradiation. We envisioned that this methodology could be
used for the production of five-membered rings and this is
demonstrated in our approach to the angular triquinane
framework shown in Scheme 2. In this case, it is an
intramolecular photoaddition that would lead, presumably,
to the intermediate biradical 8, which would be stabilized by
hyperconjugation. This would place radical character at both
position a and b. Closure at a would lead to cyclobutane 9,
whereas b would provide cyclopentane 10, which posseses
the triquinane ring structure in one overall transformation
from 7.
Scheme 3. Reagents and conditions: (a) DHP, PPTS (cat.),
CH2Cl2, rt, 82%; (b) (1i) n-BuLi, THF, –78 °C, (1ii)
PhMe2SiCH2I (13), reflux; (2) H2SO4 (cat.), MeOH, rt, 99%;
(c) H2 (1 atm, 1 atm = 101.325 kPa), Ni(AcAc)2, 4H2O (cat.),
NaBH4 (cat.), Et3N, EtOH, rt, 75%; (d) I2, PPh3, Et3N, MeCN,
rt, 76%; (e) (i) t-BuLi (2.2 equiv.), Et2O, –78 °C, (ii) 4,4-
dimethylcyclopent-2-en-1-one; ( f ) PDC, CH2Cl2, rt, 99%.
b
a
OH
11
OTHP
PhMe2Si
HO
I
14
12
HO
c
d
PhMe2Si
PhMe2Si
Our synthetic approach to photoprecursor 18 began with
4-pentyn-1-ol (11, Scheme 3). Protection of the alcohol fol-
lowed by alkyne deprotonation and electrophilic capture with
dimethylphenylsilylmethyl iodide provided the corresponding
allylsilane. Deprotection (14) and controlled hydrogenation
with P2 catalyst provided the requisite cis-olefin geometry in
15, which we determined was necessary to deliver the re-
quired stereochemistry in the triquinane structure through
the ring-forming steps later on (see Scheme 4). Conversion
of the alcohol to the corresponding iodide (16) provided the
substrate for metal–halogen exchange and 1,2-addition to 4,4-
dimethylcyclopent-2-en-1-one, yielding 17. Finally, chromium-
mediated [1,3] rearrangement and concomitant oxidation
provided photoprecursor 18 cleanly and in good recovery.
The optimized photochemical step is outlined in
Scheme 4. Initially, we attempted the reaction under a vari-
ety of conditions. In cyclohexane solvent, all other parame-
ters being the same as those in Scheme 4, the Norrish type I
cleavage product (i.e., the aldehyde) was the principal one
isolated. All reactions attempted using a Hanovia 450 W Hg
medium pressure lamp failed to provide any identifiable cyclo-
adducts. However, we did observe isomerization of the
allylsilane double bond geometry, which could happen by
one of at least two possible mechanisms. The reaction could
proceed to the biradical intermediate (19), but under these
reaction conditions, reversion to starting material is faster
than closure to product, thus scrambling olefin geometry.
15
16
SiMe2Ph
O
e
f
SiMe2Ph
17
HO
18
Conversely, and more likely, the double bond could be
getting excited, either directly or by triplet transfer, and re-
laxation occurs with olefin scrambling. We reasoned that if
we could eliminate light of wavelengths less than 300 nm, at
least direct excitation of the allylsilane would be eliminated
with certainty. This meant that Pyrex with a 280 nm cutoff
was not a suitable filter, so we opted to use uranium (canary)
glass that possesses a 314 nm cutoff (12). As mentioned,
attempts with the Hg lamp failed, but prolonged irradiation
with a low power Rayonet reactor equipped with 350 nm
lamps led to two products (20 and 21) in equal amounts
(yield = 60%) that were purified by preparative HPLC.
Assuming that the intramolecular addition followed the
“rule of 5” (13), leading to the linearly fused and not
crossfused photoadduct, the trajectory of the olefin to the
enone could follow an “endo-like” or “exo-like” arrangement
© 2006 NRC Canada

Organ and Mallik
1261
Scheme 4.
H
H
SiMe2Ph
H
O
350 nm hv
O
exo-like
CH2Cl2, 0 °C, 18 h
Canary glass filter
H
H
H
SiMe2Ph
PhMe2Si
18
O
O
O
H
H
endo-like
SiMe2Ph
H
path a
H
20
O
δ
H
δ
SiMe2Ph
21
SiMe2Ph
path b
H
19
(see Scheme 4). The endo approach would lead to a highly
congested transition state, which is exacerbated by the pres-
ence of the gem-dimethyl moiety that we installed in the
photoprecursor, which is present in the general structure of
all angular triquinanes and is therefore necessary. Further to
this mechanistic assessment, the structures of 20 and 21
were assigned on the basis of 1D and 2D proton NMR spec-
tra, including COSY and nOe that are in the Supporting
information.3
unsaturation at the C-9 site, which is common in the struc-
ture of triquinanes.
Acknowledgments
We thank the Ontario Research and Development Chal-
lenge Fund (ORDCF) and the Natural Sciences and Engi-
neering Research Council of Canada (NSERC) for
financially supporting our program.
There are a few examples in the literature pertaining to
the use of olefins that are substituted with structural moi-
eties, which can lead to cycloadducts other than the expected
[2+2] photoproducts from the biradical intermediates that
are presumably formed on the reaction pathway (14). In-
deed, vinylcyclopropane and 1,6-hexadiene derivatives have
been used to directly produce six- and seven-membered
rings during irradiation with suitable enones. These systems
were used as mechanistic probes to determine the initial site
of bond formation during the addition process itself. Inter-
estingly, when compound 18 was treated with a variety of
Lewis acids (i.e., Me2AlCl, EtAlCl2, and TiCl4), decomposi-
tion of the substrate (18) was observed, indicating that the
ring expansion is a unique photochemical outcome. To our
knowledge, no photoaddition has been conducted that di-
rectly leads to the formation of a cyclopentane ring. Further,
the silyl moiety is a latent alcohol that can be used to mod-
ify the photoadduct to deliver the methyl group and
References
1. A. Hosomi and H. Sakurai. J. Am. Chem. Soc. 99, 1673 (1977).
2. For general reviews on the uses of silicon in organic synthesis,
see: (a) I. Fleming. Chem. Soc. Rev. 10, 83 (1981); (b) Z.N.
Parnes and G.I. Bolestova. Synthesis, 991 (1984); (c) T.A.
Blumenkopf and L.E. Overman. Chem. Rev. 86, 857 (1986);
(d) D. Schinzer. Synthesis, 263 (1988); (e) I. Fleming, J.
Dunogues, and R. Smithers. Org. React. (N.Y.), 31, 57 (1989);
( f ) T.H. Chan, D. Wang, P. Pellon, S. Lamothe, Z.Y.
Wie, L.H. Li, and L.M. Chen. In Frontiers of organosilicon
chemistry. The Royal Society of Chemistry, Cambridge. 1991.
p. 344; (g) E.W. Colvin, M.A. Loreto, M. Montieth, and I.
Tommasini. In Frontiers of organosilicon chemistry. The Royal
Society of Chemistry, Cambridge. 1991. p. 356.
3. For approaches to four-membered rings using allylsilanes, see:
(a) G.D. Hartman and T.G. Traylor. Tetrahedron Lett. 16, 939
(1975); (b) M. Ochiai, M. Arimoto, and E. Fujita. J. Chem.
3 Supplementary data for this article are available on the journal Web site (http://canjchem.nrc.ca) or may be purchased from the Depository
of Unpublished Data, Document Delivery, CISTI, National Research Council Canada, Ottawa, ON K1A 0R6, Canada. DUD 5050. For more
information on obtaining material refer to http://cisti-icist.nrc-cnrc.gc.ca/irm/unpub_e.shtml.
© 2006 NRC Canada

1262
Soc. Chem. Commun. 460 (1981); (c) L.A. Paquette, R.S.
Can. J. Chem. Vol. 84, 2006
8. H.-J. Knölker, G. Baum, and R. Graf. Angew. Chem. Int. Ed.
Engl. 33, 1612 (1994).
9. M.G. Organ, V. Dragan, M. Miller, R.D.J. Froese, and J.D.
Valpey, and G.D. Annis. J. Org. Chem. 49, 1317 (1984); (d) T.
Fujiwara, A. Suda, and T. Takeda. Chem. Lett. 1631 (1992);
(e) M. Hojo, K. Tomita, Y. Hirohara, and A. Hosomi. Tetrahe-
dron Lett. 34, 8123 (1993); ( f ) T. Akiyama and M. Kirino.
Chem. Lett. 723 (1995).
Goddard. J. Org. Chem. 65, 3666 (2000).
10. For discussions regarding silyl-stabilized β-carbocations, see:
(a) W. Hanstein, H.J. Berwin, and T.G. Traylor. J. Am. Chem.
Soc. 92, 829 (1970); (b) C. Eaborn. J. Chem. Soc. Chem.
Commun. 1255 (1972); (c) I. Fleming. In Frontier orbitals and
organic chemical reactions. Wiley, London. 1976. p. 81;
(d) S.G. Wierschke, J. Chandrasekhar, and W.L. Jorgensen. J.
Am. Chem. Soc. 107, 1496 (1985); (e) J.B. Lambert, G.T.
Wang, R.B. Finzel, and D.H. Teramura. J. Am. Chem. Soc.
109, 7838 (1987); ( f ) J.B. Lambert. Tetrahedron, 46, 2677
(1990); (g) J.B. Lambert and E.C. Chelius. J. Am. Chem. Soc.
112, 8120 (1990); (h) A.J. Green, Y.L. Kuan, and J.M. White.
J. Org. Chem. 60, 2734 (1995); (i) V. Gabelica and A.J.
Kresge. J. Am. Chem. Soc. 118, 3838 (1996); ( j) V.Y. Chan,
C.I. Clark, J. Giordano, A.J. Green, A. Karalis, and J.M.
White. J. Org. Chem. 61, 5227 (1996).
11. For examinations of the β effect in relation to different groups
on silicon, see: (a) M.A. Brook, M.A. Hadi, and A. Neuy. J.
Chem. Soc. Chem. Commun. 957 (1989); (b) M.A. Brook and
C. Henry. Tetrahedron, 52, 861 (1996); (c) H. Mayr and G.
Hagen. J. Chem. Soc. Chem. Commun. 91 (1989); (d) H.
Mayr, J. Bartl, and G. Hagen. Angew. Chem. Int. Ed. Engl. 31,
1613 (1992); (e) H. Mayr and M. Patz. Angew. Chem. Int. Ed.
Engl. 33, 938 (1994); ( f ) H. Mayr and G. Gorath. J. Am.
Chem. Soc. 117, 7862 (1995).
4. For approaches to five-membered ring carbocycles using
allylsilanes, see: (a) R.L. Danheiser, B.R. Dixon, and R.W.
Gleason. J. Org. Chem. 57, 6094 (1992); (b) M. Hojo, K.
Tomita, Y. Hirohara, and A. Hosomi. Tetrahedron Lett. 34,
8123 (1993); (c) R.L. Danheiser, T. Takahashi, B. Bertók, and
B.R. Dixon. Tetrahedron Lett. 34, 3845 (1993); (d) H.J.
Knölker and R. Graf. Tetrahedron Lett. 34, 4765 (1993); (e) S.
Kiyooka, Y. Shiomi, H. Kira, Y. Kaneko, and Tanimori. J. Org.
Chem. 59, 1958 (1994); ( f ) G.P. Brengel, C. Rithner, and A.I.
Meyers. J. Org. Chem. 59, 5144 (1994); (g) Y. Takahashi, K.
Tanino, and I. Kuwajima. Tetrahedron Lett. 37, 5943 (1996);
(h) T. Akiyama, E. Hoshi, and S. Fujiyoshi. J. Chem. Soc.
Perkin Trans. 1, 2121 (1998); (i) T. Akiyama and M.
Yamanaka. Tetrahedron Lett. 39, 7885 (1998).
5. For approaches to substituted furans using allylsilanes, see:
(a) J.S. Panek and M. Yang. J. Am. Chem. Soc. 113, 9868
(1991); (b) J.S. Panek and R. Beresis. J. Org. Chem. 58, 809
(1993); (c) J.S. Panek and N.F. Jain. J. Org. Chem. 58, 2345
(1993); (d) T. Akiyama, T. Yasusa, K. Ishikawa, and S. Ozaki.
Tetrahedron Lett. 35, 8401 (1994).
6. (a) H.J. Knölker, P.G. Jones, and J.B. Pannek. Synlett, 429
(1990); (b) H.J. Knölker, P.G. Jones, J.B. Pannek, and A.
Weinkauf. Synlett, 147 (1991); (c) H.J. Knölker, N. Foitzik, R.
Graf, and J.B. Pannek. Tetrahedron, 49, 9955 (1993); (d) H.J.
Knölker, N. Foitzik, H. Goesmann, and R. Graf. Angew.
Chem. Int. Ed. Engl. 32, 1081 (1993).
12. For photocyclization of conjugated enones using a uranium fil-
ter, see: (a) M.T. Crimmins and S.W. Mascarella. J. Am.
Chem. Soc. 108, 3435 (1987); (b) M.T. Crimmins and J.A.
DeLoach. J. Am. Chem. Soc. 108, 800 (1986).
7. For an example detailing the formation of a (trimethyl-
silyl)methyl-substituted cyclobutane adduct, in addition to
some of the trimethylsilyl-substituted cyclopentanoids, from
the reaction of naphthaquinone and allyltrimethylsilane in the
presence of the mild Lewis acid, Me2AlCl, see: W.S. Murphy
and D. Neville. Tetrahedron Lett. 38, 7933 (1997).
13. A.C. Weedon. In Synthetic organic chemistry. Edited by W.M.
Horspool. Plenum Press, New York. 1984.
14. (a) A. Rudolf and A.C. Weedon. Can. J. Chem. 68, 1590
(1990); (b) D. Becker, N. Haddad, and Y. Sahali. Tetrahedron
Lett. 30, 2661 (1989); (c) D. Becker, N. Haddad, and Y.
Sahali. Tetrahedron Lett. 30, 4429 (1989).
© 2006 NRC Canada