The tandem intermolecular Paterno-Buechi reaction: Formation of tetrahydrooxepins

Article abstract of DOI:10.1039/a802922i

The Paterno-Buechi reaction is the [2 + 2] photocycloaddition between carbonyl compounds and electron rich alkenes to generate oxetane products. By the introduction of substituted cyclopropyl rings to the alkene components, the utility of this reaction has been extended to facilitate the synthesis of substituted tetrahydrooxepins. It is proposed that initial addition of oxygen radicals to cyclopropyl enol ethers generates cyclopropylmethyl radicals which, when the cyclopropane ring bears appropriate radical-stabilising groups (e.g. phenyl, CO₂Et), undergo rapid fragmentation to form homoallylic 1,7-biradicals which then recombine to deliver the observed tetrahydrooxepin products. The importance of various radical-stabilising substituents on the efficiency of tetrahydrooxepin formation is examined.

Full text of DOI:10.1039/a802922i

View Article Online / Journal Homepage / Table of Contents for this issue
The tandem intermolecular Paternò–Büchi reaction: formation of
tetrahydrooxepins
Chee Yong Gana and John N. Lambert*,†,b
a Research School of Chemistry, Australian National University Canberra, ACT 0200, Australia
b School of Chemistry, The University of Melbourne, Grattan St., Parkville, Vic. 3052,
Australia
The Paternò–Büchi reaction is the [2 ؉ 2] photocycloaddition between carbonyl compounds and electron
rich alkenes to generate oxetane products. By the introduction of substituted cyclopropyl rings to the
alkene components, the utility of this reaction has been extended to facilitate the synthesis of substituted
tetrahydrooxepins. It is proposed that initial addition of oxygen radicals to cyclopropyl enol ethers
generates cyclopropylmethyl radicals which, when the cyclopropane ring bears appropriate radical-
stabilising groups (e.g. phenyl, CO2Et), undergo rapid fragmentation to form homoallylic 1,7-biradicals
which then recombine to deliver the observed tetrahydrooxepin products. The importance of various
radical-stabilising substituents on the efficiency of tetrahydrooxepin formation is examined.
R2
R3
Introduction
O
O
O
Intersystem
crossing
R1
(ISC)
hν
In 1909, Paternò and Chieffi discovered that when a mixture of
benzaldehyde and 2-methylbut-2-ene was irradiated, a novel
small ring heterocycle, namely an oxetane ring, was formed.1
The structure of this product was not confirmed until 1954
when Büchi and co-workers reinvestigated the reaction.2
Initially, the Paternò–Büchi reaction was not widely accepted
by synthetic organic chemists and it was not until the reviews of
both Hammond and Turro in 19633 and Searles in 19644 that
the usefulness of the Paternò–Büchi reaction in organic syn-
thesis was clearly demonstrated. This photochemical method
can now be used to overcome a number of the complexities
associated with other methods for oxetane formation and the
use of the Paternò–Büchi reaction for the formation of oxetane
rings is now well established.5,6
The Paternò–Büchi reaction can be employed in both intra-
and inter-molecular reactions and has been reported to give
products with high chemo-, regio- and diastereo-selectivity.7–10
The starting materials are generally readily available and the
yields ranges from medium to quantitative. It can also be
applied to other unsaturated systems other than alkenes, such
as dienes,11 allenes,12 acetylenes13 and ketenimines.12 Due to
their strain and basicity, oxetane rings have been shown to be
useful synthetic intermediates and some transformations of the
oxetane ring include hydrogenolysis,9 ring opening through
inter-10 and intra-molecular14 nucleophilic attack and ring
expansion.15
Ar
R1
Ar
Ar
R1
T1
S1
R1 = H, alkyl or aryl
R2
R3
R2
R3
R2
R3
O
O
O
Recombination
ISC
Ar
Ar
R1
R1
Ar
R1
S1
T1
Scheme 1
component of the Paternò–Büchi reaction, methylene cyclo-
propyl radicals could be generated as intermediates in these
reactions. Cyclopropyl ring fragmentation of these intermedi-
ates might then be followed by radical recombination to deliver
oxepane-type products (Scheme 2). For a 1,4-biradical (viz. 1)
O
O
O
hν
R
R
R1
R1
R
R1
1
The mechanism proposed for the formation of oxetanes
using the Paternò–Büchi reaction is outlined in Scheme 1.5,11,16
Upon irradiation with ultra-violet light (280–400 nm), the car-
bonyl group is excited in an n–π* transition from the ground
state (S0) to the singlet excited state (S1). For an aromatic car-
bonyl compound, the S1 biradical very efficiently undergoes an
intersystem crossing (ISC) to the triplet excited state (T1). The
excited oxygen radical then adds to the olefinic component,
generating a 1,4-biradical which, after intersystem crossing
back to a singlet state, recombines to generate the observed
oxetane products.
O
O
R
R
R1
R1
3
2
Scheme 2
to undergo ring opening, the alkene substrate must bear an
α-cyclopropyl ring moiety. Ring fragmentation of cyclopropyl
methyl radical 1 would then lead to a homoallylic 1,7-biradical
intermediate 2 which could cyclize to generate the oxepane
product 3. For the 1,7-biradicals (2) to cyclize, geometrical
isomerism of the alkene formed after cyclopropyl ring frag-
mentation must be taken into consideration and we expected
that while both geometrical isomers of 2 might be formed,
Given that the accepted mechanism for the Paternò–Büchi
reaction involves the intermediacy of both 1,2- and 1,4-
biradicals, we proposed that by incorporation of cyclopropyl
rings at appropriate locations on either the carbonyl or olefinic
† E-Mail: j.lambert@chemistry.unimelb.edu.au
J. Chem. Soc., Perkin Trans. 1, 1998
2363

View Article Online
only the (Z)-isomers would be productive in terms of oxepane
formation.17,18
propyl methyl ketone was commercially available]. In most
cases, the α,β-unsaturated ketones were reacted with a mixture
of trimethylsulfoxonium iodide and sodium hydride to give
appropriate cyclopropyl ketones.21 The tricyclic silyl enol ether
12 was prepared by heating ketone 1322 at reflux in 1,4-dioxane
in the presence of potassium hydride and triethylsilyl chloride.
Ester 8 and diphenylcyclopropyl silyl enol ether 9 were pre-
pared using ethyl diazoacetate as a cyclopropanation agent.
Thus, reaction of ethyl diazoacetate with methyl vinyl ketone
in the presence of a catalytic amount of palladium() acetate
gave the known23 cyclopropyl methyl ketone 14 in 51% yield
(Scheme 4). Conversion to the silyl enol ether was accomplished
using standard conditions to give enol ether 8 in 31% yield.
Oxepane formation via cyclopropyl ring fragmentation
has been observed previously by Nishida,17 Wagner18 and
Neckers,19 respectively, in a process that makes use of an inter-
molecular Paternò–Büchi reaction.17–19 Ring fragmentation was
achieved at high temperatures and the effect of substituents on
the alkene and carbonyl components were investigated. Since
this report, studies of the rates of cyclopropyl radical fragmen-
tation have shown that ring opening of cyclopropyl methyl
radicals is heavily influenced by the nature of the substituents
on the cyclopropyl ring.20 In terms of our own goals, the avail-
able data on cyclopropyl ring fragmentation rates supported the
notion that fragmentation–oxepane formation pathways might
be able to compete effectively with Paternò–Büchi-like oxetane
formation. In this article we report on the results of our studies
where we demonstrate that appropriately substituted Paternò–
Büchi reaction partners are able to serve as precursors to substi-
tuted oxepane products.
O
OEt
(a)
–
+
N
N
C
OEt
O
O
O
14
Scheme 4 Reagents and conditions: (a) Pd(OAc)2, 40 ЊC to room
temperature
Results and discussion
Synthesis of cyclopropyl silyl enol ethers
In the synthesis of diphenyl silyl enol ether 9, ethyl diazo-
acetate was heated in 1,1-diphenylethene to give the ester 15
which was not isolated but hydrolysed with potassium hydrox-
ide to provide the carboxylic acid 16 (60%) (Scheme 5).24 This
Initially we chose to investigate the generation of tetrahydro-
oxepin systems employing α-cyclopropyl silyl enol ethers in
tandem reactions with various aromatic ketones and aldehydes
(cf. Scheme 2). The various cyclopropyl silyl enol ethers that
were used in these studies are shown in Scheme 3. In order to
O
Ph
Ph
Ph
–
N
+
N
OEt
(a)
C
OEt
O
O
OSi(CH3)3
Ph
O
R
(a)
R
(b)
R
15
R1
R1
R1
(b)
4 R = H, R1 = H
5 R = H, R1 = CH3
6 R = CH3, R1 = CH3
7 R = H, R1 = Ph
8 R = H,
R1 = C(O)OCH2CH3
9 R = R1=Ph
Ph
Ph
Ph
(c)
OH
Ph
O
O
16
17
Scheme 5 Reagents and conditions: (a) 160 ЊC; (b) KOH (aq.) then
H3Oϩ; (c) CH3Li
O
n
O
n
OSi(CH3)3
carboxylic acid was then treated with methyllithium to give the
required diphenylcyclopropyl methyl ketone 17 (88%).25 Ketone
17 was then subjected to enolisation/trapping (LDA–TMSCl)
conditions as above to generate the cyclopropyl enol ether 9
(93%).
(a)
(b)
n
10 n = 1
11 n = 2
Photoreactions of monocyclic cyclopropyl silyl enol ethers
The unsubstituted cyclopropyl silyl enol ether 4, when irradi-
ated in the presence of either benzophenone or 4-cyano-
benzaldehyde delivered the 3-substituted oxetane products 19
(73%) and 20 (46%) respectively (Scheme 6). The regiochemical
O
OSiEt3
(c)
CO2CH3
CO2CH3
12
13
OSi(CH3)3
OSi(CH3)3
Scheme 3 Reagents and conditions: (a) (CH3)3S(O)I/NaH, DMSO, RT;
(b) LDA then (CH3)3SiCl, THF, Ϫ78 ЊC; (c) KH, Et3Cl, 1,4-dioxane,
reflux
hν, ketone or aldehyde
CH3CN, RT
O
R
R1
4
promote fragmentation and oxepane formation, all these com-
pounds carried the common feature of a cyclopropyl ring
adjacent to the silyl enol ether moiety. These ethers included
monocyclic 4–9, bicyclic 10,11 and tricyclic system 12 and the
varying substituents attached to the cyclopropyl ring were
installed to allow us to investigate the relative fragmentation
efficiency of primary, secondary, tertiary and benzylic radical
intermediates.
18
OSi(CH3)3
4
O
R
R1
19 R = R1 = Ph
20 R = H, R1 = p-C6H4CN
The mono- (4–7) and bi-cyclic ethers (10,11) were prepared
as shown in Scheme 3 from commercially available α,β-
unsaturated ketones [except for 4 (R, R1 = H), where cyclo-
Scheme 6
2364
J. Chem. Soc., Perkin Trans. 1, 1998

View Article Online
outcome of these reactions was predicted based on previously
reported experiments and revealed that initial addition of the
oxygen radical to the least substituted olefinic carbon generated
the 1,4-biradical 18 which then underwent recombination to
give the observed oxetane products 19 and 20.26
In a similar manner, reaction of mono- and di-methyl-
cyclopropyl silyl enol ethers 5 and 6 with benzophenone
delivered the 3-substituted oxetane compounds 21 and 22
respectively as inseparable mixtures of diastereoisomers
(Scheme 7). Thus, reactions of mono-, di- and un-substituted
distinct peak at δ 103, indicating the presence of an acetal
carbon, and supported the contention that the second oxygen
radical addition had occurred at the most substituted olefinic
position of 25. Two methylene carbon resonances appeared at
δ 31.9 and δ 71.1 as well as five aromatic quaternary carbon
signals. Heteronuclear (1H/13C) correlation spectroscopy
(HETCOSY) established that the two methine carbons of 26
had the same chemical shift at δ 50.3.
Oxepane 26 was apparently formed by addition of the benzo-
phenone oxygen-centred radical to the less substituted carbon
of alkene 7, giving rise to a cyclopropylmethyl radical 23. Rapid
fragmentation of the cyclopropyl ring to give a 1,7-biradical
intermediate 24 followed by radical recombination delivered
the putative intermediate tetrahydrooxepin derivative 25. A
second molecule of benzophenone then underwent photo-
addition to the more substituted carbon of enol ether 25 to
deliver the observed dioxabicyclo[5.2.0]nonane 26. Despite
extensive spectroscopic examination of 26, we are currently
unable to assign the relative stereochemistry of the oxetane ring
and the C5-phenyl group in this compound. Treatment of
compound 26 with tetrabutylammonium fluoride removed the
silyl group to give a tertiary alkoxide 27 which spontaneously
underwent oxetane ring-opening to give the ketone 28 (82%)
and benzophenone. The outcome of this reaction further
confirmed the structure of the photoaddition product 26.
This result indicated that when a sufficiently stable biradical
intermediate could be generated by cyclopropyl fragmentation
(e.g. 24), oxepane products may be generated from carbonyl
compounds and silyl enol ethers in a single step. In order to
investigate the scope of this new tandem radical reaction, silyl
enol ether 7 was further reacted with both benzaldehyde and
4-methoxybenzaldehyde. Reaction with benzaldehyde gave the
tetrahydrooxepin product 29 (27%) and oxepanone 32 (6%)
(Scheme 9). The two phenyl groups of 29 and 32 were expected
OSi(CH3)3
OSi(CH3)3
hν, benzophenone
R
R
O
CH3CN, RT
R1
R1
Ph
Ph
5 R = H, R1= CH3
6 R = R1 = CH3
21 R = H, R1 = CH3
22 R = R1 = CH3
Scheme 7
cyclopropyl silyl enol ethers 4–6 failed to deliver any ring-
opened oxepane products.
While all previous photochemical reactions had exclusively
delivered oxetane-bearing products, it was highly satisfying
to observe that Paternò–Büchi reaction of the mono-phenyl
substituted cyclopropyl enol ether 7 with benzophenone gave
a product whose structure was consistent with that of ring-
opened compound 26 (15%) (Scheme 8). The structure of 26
was identified by a number of spectroscopic techniques. The
100 MHz {1H}13C NMR spectrum of compound 26 displayed a
OSi(CH3)3
O
OSi(CH3)3
Ph
Ph
Ph
OSi(CH3)3
O
hν
Ph
Ph
7
Ph
23
Ph
7
hν, aldehyde
benzene
Ph
Ph
Ph
Ph
Ha
R1
Ha
Ph
OSi(CH3)3
OSi(CH3)3
Ph
R2
Ph
Hb
O
OSi(CH3)3
O
Ph
O
Ph
O
O
25
24
O
29 R1 = Ph, R2 = H
32
30 R1 = p-C6H4OCH3, R2 = H
31 R1 = H, R2 = p-C6H4OCH3
Ph
Ph
hν
Scheme 9
Ph Ph
Ph Ph
O
5
Ph
Ph
Ph
Ph
to reside on opposite faces of the oxepane ring and this predic-
tion was supported by the observation of a 10.2 Hz coupling
constant in the 400 MHz H NMR spectrum between Ha and
Hb for both 29 and 32. The most likely source of ketone 32 is by
proto-desilylation of 29 during work-up and isolation. Photo-
reaction of 4-methoxybenzaldehyde with silyl enol ether 7 gave
the diastereoisomeric tetrahydrooxepin products 30 (40%) and
31 (12%) (Scheme 9). The stereochemistry of both 30 and 31
was again assigned by looking at the coupling constants
[CH3(CH2)3]4NF
O
CH2Cl2, 0 °C–RT
OSi(CH3)3
O
1
–
O
O
Ph
Ph
26
27
H2O
Ph
Ph
O
1
between Ha and Hb in the 400 MHz H NMR spectra. The
O
coupling constants observed for 30 and 31 were 10.2 and 5.6 Hz
respectively indicating that 30 carried its vicinal aryl groups in a
trans-relationship while in 31, the corresponding groups were
cis-disposed.
Having established that phenyl-substituted cyclopropane 7
underwent fragmentation upon reaction with different aromatic
aldehydes and ketones, we sought to explore the scope of this
Ph
28
O
+
Ph
Ph
Scheme 8
J. Chem. Soc., Perkin Trans. 1, 1998
2365

View Article Online
OSi(CH3)3
reaction in terms of the nature of the radical stabilising groups
attached to the cyclopropyl ring. Specifically, we turned our
attention to the evaluation of an ester as a radical stabilising
group. Irradiation of benzaldehyde in the presence of the
cyclopropyl ester 8 delivered small yields of two diastereo-
isomeric tetrahydrooxepin products 33 (8%) and 34 (8%)
(Scheme 10). Again, the coupling constant between Ha and Hb
OSi(CH3)3
hν, RT, CH3CN
Hydrogen abstraction
H
O
Ph
OH
10
Ph
Ph
Ph
OSi(CH3)3
Radical
recombination
EtO
OSi(CH3)3
O
O
8
hν, benzaldehyde
CH3CN
Desilylation
4
O
O
OH
Ph
OH
Ph
Ha
Ph
Ph
Ha
EtO
EtO
OSi(CH3)3
35
36
OSi(CH3)3
Ph
Hb
O
O
Hb
33
Scheme 11
Ph
34
OSi(CH3)3
Scheme 10
4
was used for the assignment of the relative stereochemistry in
33 and 34 and a 10.5 Hz coupling for 33 and 2.7 Hz coupling
for 34 indicated that these were respectively trans- and cis-
diastereoisomers (ester group with respect to the phenyl group).
Enol ether 8 was also reacted with benzophenone, 4-cyano-
benzaldehyde and 4-methoxybenzaldehyde. Starting enol ether
was recovered from these latter reactions.
11
Benzophenone
hν, RT, CH3CN
Ph
Ph
Diphenyl cyclopropyl enol ether 9 failed to react with benz-
aldehyde, 4-cyanobenzaldehyde and 4-methoxybenzaldehyde
and the starting enol ether 9 was recovered from these reactions.
When enol ether 9 was irradiated in the presence of benzo-
phenone, various high molecular weight compounds were
observed by mass spectrometric examination. The 1H and
{1H}13C NMR spectra were complex and structural assignment
of the products of this reaction was not possible. The absence
of any identifiable products may be a reflection of the fact that
while cyclopropyl ring fragmentation in systems such as 9 may
be rapid, the recombination of the derived 1,7-biradical inter-
mediate may be slowed by unfavourable steric interactions in
the transition state leading to the seven-membered ring prod-
uct. Consequently, decomposition products tend to predomin-
ate over oxepane-containing materials.
Ph
Ph
O
OSi(CH3)3
OSi(CH3)3
O
7
7
37
39
Ph
O
Ph
OSi(CH3)3
OH
Ph
(H3C)3SiO
Ph
7
Ph
HO
6
Ph
38
40
Scheme 12
Photoreactions of bicyclic cyclopropyl silyl enol ethers
isolated by repeated chromatography and were identified as 37
(1.5%), 39 (2%), 38 (0.6%) and 40 (6%) (Scheme 12).
The five-membered ring enol ether 10 was photoreacted with
benzophenone in acetonitrile using similar conditions to those
described for the monocyclic systems. This reaction gave a
complex mixture of products which, after repeated chrom-
atography, yielded diphenylmethanol derivative 36 (9%). The
remainder of the material isolated from this reaction consisted
of either 1,1,2,2-tetraphenylethane-1,2-diol or unreacted start-
ing enol ether.
A proposed mechanism for the formation of 36 is shown in
Scheme 11 and involves initial allylic hydrogen atom abstrac-
tion by the photoexcited benzophenone triplet diradical. It is
proposed that exo-face recombination with the diphenyl-
methanol radical gives silyl enol ether 35 which upon proteo-
lytic desilylation delivers the ultimately observed product 36.
Similar processes have been previously observed by other work-
ers.27 It is interesting to note that in this bicyclic enol ether,
no oxetane or oxepane products were observed and that the
allylic hydrogen abstraction pathway is clearly the favoured
reaction pathway.
Oxetanes 37 and 39 are clearly formed by conventional
Paternò–Büchi reactions, with 37 being formed by initial photo-
addition of the triplet oxygen radical to the less substituted
carbon of the alkene and 39 by addition of the oxygen radical
to the more hindered carbon of the alkene. It was not possible
to determine the relative stereochemistry of C7 in 37 and
39 unequivocally but we propose, based on analogy to the 5-
membered homologue 36, that the cyclopropyl and oxetane
rings of these compounds reside in a trans-relationship.
Compound 38 was presumably formed by allylic hydrogen
atom abstraction at C4 of 11 followed by exo-face addition of
the diphenylmethanol radical (Scheme 13). Another molecule
of benzophenone then underwent photoaddition with enol
ether 43 from the exo-face to form the 2-substituted oxetane
product 38. Once more, the mechanism for the formation of 40
is proposed to involve initial hydrogen abstraction at C4 of 11
by the photoexcited benzophenone triplet radical. The radical
then undergoes allylic rearrangement to give a more stable
α-oxy radical at C2. This is followed by radical recombination
from the exo-face to generate 40. The assignment of structure
Photoaddition of bicyclo[4.1.0]heptane enol ether 11 with
benzophenone gave a mixture of compounds which were again
2366
J. Chem. Soc., Perkin Trans. 1, 1998

View Article Online
OSiEt3
OSi(CH3)3
OSiEt3
hν, benzophenone
RT, CH3CN
3
4
Hydrogen abstraction
H
H
O
CO2CH3
12
O
CO2CH3
11
Ph
Ph
45
Ph
Ph
Cyclopropyl
fragmentation
OH
Ph
OSi(CH3)3
OH
OSiEt3
Ph
1) Aromatisation
2) Desilylation
(H3C)3SiO
4
Ph
Ph
2
CO2CH3
HO
CO2CH3
46
47
Radical
recombination
Scheme 14
formationally constrained systems indicate that apart from
carrying suitable radical stabilising substituents, cyclopropyl
enol ethers must be able to accommodate the stereoelectronic
requirements of rapid cyclopropyl ring fragmentation. These
studies have shown that the tandem Paternò–Büchi reactions of
simple silyl enol ethers with suitable ketones or aldehydes may
provide access to more complex oxygenated heterocycles.
OSi(CH3)3
Ph
OH
Ph
(H3C)3SiO
Ph
HO
Ph
40
43
O
Experimental
hν
[CH3(CH2)3]4NF
Proton (1H) and carbon (13C) nuclear magnetic resonance
(NMR) spectra were recorded using a Varian Unity 300
spectrometer or a JEOL GX-400 spectrometer, operating at 300
and 400 MHz for proton and 75 and 100 MHz for carbon
respectively. Deuterochloroform (CDCl3) was used as a solvent
and the residual peak of chloroform (CHCl3) was used as
internal reference (δ 7.26) for proton while the central peak
of CDCl3 (δ 77.0) was used as reference for carbon spectra.
Chemical shifts are quoted on the δ (ppm) scale followed by
multiplicity (s: singlet, d: doublet, t: triplet, q: quartet, m: multi-
plet), coupling constant(s) J (Hz), assignment (where possible)
and integrated intensity (for proton spectra). Distortionless
enhancement by polarisation transfer (DEPT), nuclear Over-
hauser effect (NOE) difference spectroscopy, cycled Overhauser
effect difference spectroscopy (cyclonoe), homonuclear (1H–1H)
correlation spectroscopy (COSY) and heteronuclear (1H–13C)
correlation spectroscopy (HETCOSY) techniques were used in
the assignment of NMR spectra.29
Ph
Ph
Ph
Ph
H7
O
Ph
OSi(CH3)3
OH
Ph
HO
J6,7 = 3.6 Hz
J6,5a = 3.6 Hz
J6,5b = 13.5 Hz
Ph
H6
HO
H5b
H5a
Ph
38
44
Scheme 13
40 was partly supported by subjection to reaction with tetra-
butylammonium fluoride to give the corresponding diol 44
(16%).
Photoreactions of a tricyclic cyclopropyl silyl enol ether
Irradiation of the tricyclic enol ether 12 in the presence of
benzophenone, followed by desilylation with tetrabutyl-
ammonium fluoride gave the tetrahydronaphthalene derivative
47 in 28% yield. This structure was confirmed by various
spectroscopic techniques including X-ray analysis.28 A mechan-
ism has been proposed for the formation of this compound
(Scheme 14). Once more, allylic hydrogen atom abstraction is
invoked, generating allylic radical 45. The cyclopropyl ring then
undergoes fragmentation to give the radical intermediate 46
which upon oxidation followed by desilylation affords the tetra-
hydronaphthalene product 47.
In summary, the photoreaction of monophenyl substituted
silyl enol ether 7 with benzophenone, benzaldehyde and 4-
methoxybenzaldehyde gives oxepane derivatives in a single step.
Other monocyclic systems give mainly 3-substituted oxetane
products. These results demonstrate that in order to generate a
seven-membered oxepane ring, a stabilised but unencumbered
radical, (e.g. 24) must be produced by fragmentation of the
cyclopropylmethyl radical intermediate. It is also observed
that the bicyclic and tricyclic systems predominantly undergo
H-abstraction at the allylic position in preference to cyclo-
addition/fragmentation reactions. The results from these con-
Melting point determinations were performed using a Kofler
hot-stage and are uncorrected. Microanalyses were performed
by the Microanalytical Unit of Research School of Chemistry,
The Australian National University, Canberra, Australia.
Electronic spectra were recorded in chloroform or aceto-
nitrile unless otherwise stated, using a Varian Super Scan 3
spectrophotometer and are listed as absorbance maxima (nm)
followed by extinction coefficient ε (cmϪ1 Ϫ1).
Infrared spectra were recorded as potassium bromide discs
for solids or as neat films between sodium chloride plates for
liquids using a Perkin-Elmer 983G spectrophotometer or a
Perkin-Elmer 1600 Series FTIR spectrophotometer.
High and low resolution mass spectra were recorded using a
VG Micromass 7070F instrument using positive ion electron
impact techniques at 70 eV. Ions of intensity greater than 20%
are quoted as an m/z value followed by intensity (%).
Analytical thin layer chromatography (TLC) was conducted
on aluminium-backed 0.2 mm thick silica gel 60 GF254 plates
supplied by Merck and the chromatograms were visualised
under a 254 nm UV lamp and/or with anisaldehyde–sulfuric
acid–ethanol (2:5:93) dip reagent. Preparative thin layer
chromatography (PLC) was carried out on 20 × 20 cm glass
J. Chem. Soc., Perkin Trans. 1, 1998
2367

View Article Online
plates coated with silica gel (Merck Kieselgel GF254) and eluted
with the solvent system indicated. The separated compounds
were located under 254 nm UV light and extracted using diethyl
ether or dichloromethane. Flash chromatography was carried
out according to the method of Still et al.30 using Merck
Kieselgel 60G No. 7731 and eluted with the solvent system
indicated.
Medium pressure liquid chromatography (MPLC) was con-
ducted using a Büchi 681 pump, a Büchi 648 fraction collector
and ‘Plastiglass’ columns packed with Merck Kieselgel 60G
No. 7731, also supplied by Büchi. Normal-phase high pressure
liquid chromatography (HPLC) was carried out with an EXSIL
100 5µ SiOH 250 × 10 mm column using HPLC quality or
redistilled solvents as indicated and an ISCO Model 2350 pump
connected to the column indicated. The peaks were detected
using an ERMA ERC-7512 refractive index detector connected
to a Spectra-Physics SP4270 integrator.
mmHg, 65%; δH(CDCl3, 400 MHz) 0.23 (s, 9H), 1.03 (m, 1H),
1.28 (m, 1H), 1.67 (m, 1H), 2.13 (m, 1H), 4.07 (s, 1H), 4.18 (s,
1H), 7.00–7.30 (m, 5H); δC(CDCl3, 100 MHz) 0.1, 14.1, 22.4,
27.9, 88.5, 125.5, 125.9, 128.3, 142.4, 157.5; νmax/cmϪ1 3062,
3029, 2959, 1645, 1498, 1316, 1288; m/z 233 (Mϩ ϩ 1, 39%),
᝽
145 (51), 143 (45), 117 (39), 115 (30), 91 (40), 73 (100) (HRMS:
C14H20OSi requires Mϩ , 232.1283. Found: Mϩ, 232.0964).
᝽
[2-(1-Trimethylsilyloxyvinyl)cyclopropane]carboxylic
acid
ethyl ester 8. Prepared from (2-acetylcyclopropane)carboxylic
acid ethyl ester 14. Oil, bp 75 ЊC/0.15 mmHg, 31%; δH(CDCl3,
300 MHz) 0.19 (s, 9H), 1.18 (m, 2H), 1.26 (t, J 7, 3H), 1.81 (m,
1H), 2.02 (m, 1H), 4.07 (d, J 1.2, 1H), 4.13 (q, J 7, 2H), 4.22
(d, J 1.2, 1H); δC(CDCl3, 75 MHz) Ϫ0.1, 12.9, 14.2, 19.4, 26.2,
60.5, 89.6, 155.3, 173.7; νmax/cmϪ1 2962, 1728, 1627, 1255, 1179;
m/z 228 (Mϩ , 74%), 213 (27), 199 (26), 185 (36), 183 (68), 156
᝽
(39), 155 (100), 141 (64), 111 (25), 110 (55), 103 (66), 82 (43),
75 (91), 73 (100), 57 (21), 55 (22).
[1-(2,2-Diphenylcyclopropyl)vinyloxy]trimethylsilane 9. Pre-
pared from 1-(2,2-diphenylcyclopropyl)ethanone 17. Oil,
183 ЊC/0.8 mmHg, 93%; δH(CDCl3, 300 MHz) Ϫ0.03 (s, 9H),
1.43 (dd, J 9.0, 4.6, 1H), 1.87 (dd, J 6.2, 4.6, 1H), 2.34 (dd,
J 9.0, 6.2, 1H), 3.98 (s, 1H), 4.18 (s, 1H), 7.13–7.50 (m, 10H);
δC(CDCl3, 75 MHz) Ϫ0.5, 18.5 (CH2), 32.0, 36.7, 89.5, 125.7,
126.1, 127.5, 127.9, 128.2, 130.2, 141.5, 146.9, 155.7; νmax/cmϪ1
Materials
Many reagents were available from the Aldrich Chemical Com-
pany and were used as supplied. Drying agents and other
inorganic salts were generally purchased from Ajax Chemicals.
Solvents and reagents for the reactions were purified according
to well established procedures.31 Tetrahydrofuran (THF) and
diethyl ether were purified by distillation from sodium benzo-
phenone ketyl. All irradiation experiments were conducted
using a medium pressure mercury lamp with a Pyrex filter in
spectroscopic grade solvents.
3026, 2964, 1600, 1496, 1447, 1253; m/z 308 (Mϩ , 52%), 219
᝽
(32), 218 (61), 217 (90), 203 (42), 202 (29), 194 (23), 193 (59),
192 (29), 191 (32), 189 (20), 183 (27), 179 (32), 178 (47), 165
(76), 141 (42), 118 (20), 117 (95), 115 (81), 91 (69), 77 (25), 75
(73), 74 (35), 73 (100).
Preparation of silyl enol ethers
(Bicyclo[3.1.0]hex-2-en-2-yloxy)trimethylsilane 10. Prepared
from bicyclo[3.1.0]hexan-2-one.35 Oil, 61%; δH(CDCl3, 300
MHz) 0.22 (s, 9H), 1.05 (m, 2H), 1.59 (m, 2H), 2.22 (d, J 12.0,
1H), 2.49 (dd, J 12.0, 5.1, 1H), 4.22 (s, 1H); δC(CDCl3, 75 MHz)
0.0, 13.9, 15.1, 22.7, 31.1, 97.6, 158.5; νmax/cmϪ1 2957, 2915,
1-Trimethylsilyloxy-1-cyclopropylethylene 4 was prepared as
follows: butyllithium (14.3 ml, 35.7 mmol, 2.5  solution in
hexane) was added to a solution of diisopropylamine (5.0 ml,
35.7 mmol) in dry tetrahydrofuran (35 ml) at 0 ЊC under a
nitrogen atmosphere. The solution was stirred for 15 min then
cooled to Ϫ78 ЊC, followed by dropwise addition of cyclo-
propyl methyl ketone (2.0 g, 23.8 mmol) in dry tetrahydrofuran
(0.4 ml). The reaction mixture was stirred for 30 min and
chlorotrimethylsilane (7.0 ml, 54.7 mmol) was then added drop-
wise. The resulting solution was then stirred for a further 1 h,
then warmed to 0 ЊC before pouring into a mixture of water (20
ml) and pentane (20 ml). The pentane extract was washed with
water (2 × 10 ml), dried (MgSO4), filtered and concentrated
under reduced pressure. Distillation (39 ЊC/12 mmHg, lit.,32 bp
38–40 ЊC/12 mmHg) gave enol ether 4 as a colourless oil (3.49 g,
94%); δH(CDCl3, 300 MHz) 0.19 (s, 9H), 0.55 (m, 4H), 1.40 (m,
1H), 3.99 (s, 1H), 4.13 (s, 1H); δC(CDCl3, 75 MHz) 0.0, 4.4,
15.5, 87.6, 159.1; νmax/cmϪ1 3093, 2960, 1648, 1443, 1253, 1082,
1654, 1394, 1251; m/z 169 (Mϩ ϩ 1, 58%), 75 (28), 73 (100).
᝽
(Bicyclo[4.1.0]hept-2-en-2-yloxy)trimethylsilane 11. Prepared
from bicyclo[4.1.0]heptan-2-one.36 Oil, bp 75 ЊC/4.5 mmHg,
86%; δH(CDCl3, 300 MHz) 0.18 (s, 9H), 0.60–0.78 (m, 2H), 1.09
(m, 1H), 1.35 (m, 1H), 1.52 (m, 1H), 1.65–1.99 (m, 3H), 4.58
(dt, J 6.9, 1.8, H3, 1H); δC(CDCl3, 75 MHz) 0.2, 8.9, 13.8, 14.8,
18.8, 19.1, 98.0, 152.3; νmax/cmϪ1 3004, 2957, 2917, 1395, 1251;
m/z 182 (Mϩ , 11%), 149 (20), 74 (41), 73 (100), 72 (24), 57 (32),
᝽
55 (30) (HRMS: C10H18OSi requires Mϩ , 182.1127. Found:
᝽
Mϩ, 182.1075).
(2-Acetylcyclopropane)carboxylic acid ethyl ester 14
A solution of ethyl diazoacetate (2 ml, 19.0 mmol) in benzene
(16 ml) was added to a mixture of methyl vinyl ketone (2 ml,
24 mmol) and palladium() acetate (42 mg) in benzene (10 ml)
at a rate of 7 mmol hϪ1 while the temperature of the reaction
mixture was maintained at 40 ЊC. After half of the ethyl diazo-
acetate was introduced, another portion of palladium()
acetate (42 mg) was added. Upon completion of addition (3 h),
the reaction mixture was cooled and washed with saturated
aqueous sodium carbonate (3 × 10 ml) and water (2 × 5 ml).
The organic layer was dried (MgSO4), filtered and concentrated
under reduced pressure. Distillation (80 ЊC/0.3 mmHg, lit.,23 bp
115–125 ЊC/15 mmHg) yielded the title compound 14 (1.52 g,
51%) as a colourless liquid; δH(CDCl3, 300 MHz) 1.24 (t, J 7,
3H), 1.39 (m, 2H), 2.23 (m, 1H), 2.27 (s, 3H), 2.45 (m, 1H), 4.12
(q, J 7, 2H); δC(CDCl3, 75 MHz) 14.1, 17.0, 24.1, 29.5, 30.7,
61.0, 171.9, 205.2; νmax/cmϪ1 3000, 1740, 1714, 1343, 1200; m/z
1011, 847; m/z 156 (Mϩ , 29%), 147 (40), 141 (76), 131 (21), 99
᝽
(43), 75 (73), 73 (100), 59 (24), 57 (52), 56 (36), 55 (75).
Silyl enol ethers 5–11 were prepared in a similar manner and
their physical properties are given below.
[1-(2-Methylcyclopropyl)vinyloxy]trimethylsilane 5. Prepared
from 1-(2-methylcyclopropyl)ethanone. Oil, bp 73 ЊC/760
mmHg, 79%; δH(CDCl3, 300 MHz) 0.18 (s, 9H), 0.32 (m, 1H),
0.75 (m, 1H), 1.00 (m, 1H), 1.04 (s, 3H), 1.10 (m, 1H), 3.96 (s,
1H), 4.08 (s, 1H); δC(CDCl3, 75 MHz) 0.1, 12.7, 12.9, 18.3, 24.3,
87.2, 150.0; νmax/cmϪ1 3078, 3002, 2957, 1645, 1403, 1253; m/z
171 (Mϩ ϩ 1, 33%), 91 (26), 73 (100), 69 (23), 57 (25), 55 (23).
᝽
[1-(2,2-Dimethylcyclopropyl)vinyloxy]trimethylsilane 6. Pre-
pared from 1-(2,2-dimethylcyclopropyl)ethanone.33 Oil, bp
85 ЊC/760 mmHg, 58%; δH(CDCl3, 300 MHz) 0.20 (s, 9H), 0.53
(d, J 7.0, 2H), 1.07 (s, 3H), 1.09 (s, 3H), 1.22 (t, J 7.0, 1H), 4.02
(s, 1H), 4.12 (s, 1H); δC(CDCl3, 75 MHz) 0.2, 18.5, 19.4, 27.1,
29.9, 90.2, 158.1, C2 not observed; νmax/cmϪ1 3068, 2957, 1634,
156 (Mϩ , 1%), 91 (29), 90 (20), 75 (24), 61 (25), 57 (100).
᝽
1-(2,2-Diphenylcyclopropyl)ethanone 17
1307, 1253; m/z 184 (Mϩ , 21%), 169 (26), 147 (23), 141 (75), 84
A solution of ethyl diazoacetate in 1,1-diphenylethylene (4 ml)
was added to 1,1-diphenylethylene (25.5 ml) at 160–165 ЊC over
a period of 6 h. After addition, the reaction mixture was cooled
and a solution of potassium hydroxide (1.6 g) in ethanol (20 ml)
was added. The reaction was heated at reflux for another 1 h.
᝽
(27), 79 (41), 75 (65), 73 (100), 57 (23), 55 (30) (HRMS:
C10H20OSi requires Mϩ , 184.1283. Found: Mϩ, 184.1281).
᝽
[1-(2-Phenylcyclopropyl)vinyloxy]trimethylsilane 7. Prepared
from 1-(2-phenylcyclopropyl)ethanone.34 Oil, bp 95 ЊC/0.7
2368
J. Chem. Soc., Perkin Trans. 1, 1998

View Article Online
Upon cooling, the reaction mixture was concentrated under
reduced pressure. The residue was dissolved in water (100 ml)
and 2  sodium hydroxide (20 ml). The aqueous layer was
extracted with diethyl ether (2 × 100 ml) and then acidified
using 2  hydrochloric acid. The solid formed was filtered and
recrystallised from methanol–water to give carboxylic acid 16
(3.71 g, 60%) as a white solid, mp 169–171 ЊC (lit.,24 mp 170–
171 ЊC); δH(CDCl3, 300 MHz) 1.66 (dd, J 8.1, 5.0, 1H), 2.11 (dd,
J 6.0, 5.0, 1H), 2.49 (dd, J 8.1, 6.0, 1H), 7.17–7.38 (m, 10H),
OH not observed; δC(CDCl3, 75 MHz) 20.7, 28.6, 41.1, 126.7,
127.1, 127.6, 128.4, 128.5, 129.6, 139.8, 144.6, 176.7; νmax/cmϪ1
δC(CDCl3, 75 MHz) 1.2, 1.7, 1.8, 17.8, 76.7, 80.2, 87.5, 124.9,
126.2, 126.3, 126.5, 127.5, 127.6, 142.9, 144.1; νmax/cmϪ1 3087,
3022, 2950, 1589, 1449, 1380; m/z 308 (Mϩ Ϫ CH2O, 14%), 156
᝽
(68), 141 (100), 77 (20), 75 (68), 73 (92); λmax(CH3CN)/nm 250.0,
224 (sh), 206 nm (log ε 5.41, 6.52, 6.61) (Found: C, 74.2; H, 7.8.
C21H26O2Si requires C, 74.5; H, 7.7%).
4-(3-Cyclopropyl-3-trimethylsilyloxyoxetan-2-yl)benzonitrile
20. A mixture of 4 (89 mg, 0.57 mmol), 4-cyanobenzaldehyde
(50 mg, 0.38 mmol) and sodium carbonate (5 mg) in benzene
(4.3 ml) was degassed and irradiated at 25 ЊC for 23 h using a
medium pressure mercury lamp. The reaction mixture was con-
centrated and the resulting residue was subjected to preparative
thin layer chromatography using light petroleum (bp 40–
60 ЊC)–diethyl ether (2:1). The major band was isolated to give
20 as a colourless oil (Rf 0.63) (50 mg, 46%); δH(CDCl3, 400
MHz) Ϫ0.16 (s, 9H), 0.50–0.72 (m, 4H), 1.36 (m, 1H), 4.43 (d,
J 5.0, 1H), 4.52 (d, J 5.0, 1H), 5.49 (s, 1H), 7.42 (d, J 6.0, 2H),
7.66 (d, J 6.0, 2H); δC(CDCl3, 100 MHz) 1.4, 1.9, 2.0, 18.7, 78.1,
80.0, 92.0, 110.0, 119.0, 127.1, 131.5, 144.1; νmax/cmϪ1 3083,
3576, 3026, 1703, 1600, 1495, 1447, 1254; m/z 238 (Mϩ , 40%),
᝽
193 (74), 192 (34), 191 (21), 178 (29), 165 (25), 115 (31), 91 (32),
58 (31).
A solution of 16 (2.0 g, 8.4 mmol) in diethyl ether (63 ml) was
added to a solution of methyllithium (18 ml, 1.4  in diethyl
ether) in diethyl ether (66 ml) over a period of 20 min under a
nitrogen atmosphere. The reaction mixture was stirred for a
further 20 min and poured into aqueous ammonium chloride
(2.1 g in 105 ml water). The organic layer was washed with
saturated aqueous sodium chloride until the washings were
pH neutral to litmus. It was then dried (MgSO4), filtered and
concentrated under reduced pressure. Distillation (170 ЊC/0.8
mmHg, lit.,25 bp 159–160 ЊC/2.5 mmHg) gave ketone 17 (1.98 g,
88%) as a colourless oil; δH(CDCl3, 300 MHz) 1.64 (dd, J 7.8,
4.5, 1H), 2.21 (s, 3H), 2.34 (dd, J 6.3, 4.5, 1H), 2.88 (dd, J 7.8,
6.3, 1H), 7.18–7.52 (m, 10H); δC(CDCl3, 75 MHz) 20.9, 31.1,
37.1, 42.6, 126.5, 126.9, 127.4, 128.3, 128.4, 129.8, 139.3, 144.9,
203.7; νmax/cmϪ1 3027, 1702, 1602, 1496, 1446, 1273; m/z 236
3009, 2959, 2229, 1610, 1504, 1378, 1253; m/z 287 (Mϩ , 3%),
᝽
157 (23), 156 (27), 141 (32), 130 (37), 91 (35), 75 (45), 73 (90),
58 (100), 51 (25); λmax(CH3CN)/nm 235 nm (log ε 7.23) (HRMS:
C16H21NO2Si requires Mϩ , 287.1341. Found: Mϩ, 287.1328).
᝽
[3-(2-Methylcyclopropyl)-2,2-diphenyloxetan-3-yloxy]-
trimethylsilane 21. A solution of 5 (200 mg, 1.18 mmol) and
benzophenone (236 mg, 1.29 mmol) in acetonitrile (12.3 ml)
was degassed and irradiated at 25 ЊC for 48 h using a medium
pressure mercury lamp. The reaction mixture was concentrated
and the resulting residue was subjected to preparative thin
layer chromatography using light petroleum (bp 40–60 ЊC)–
diethyl ether (8:1). The major band was isolated to give an
equimolar mixture of two inseparable diastereoisomeric
oxetanes 21 (Rf 0.73) as a white crystalline solid (104.4 mg,
25%), mp 65–68 ЊC; δH(CDCl3, 300 MHz) (Ϫ0.17, Ϫ0.07)
(s, 9H), (0.04, 0.25) (m, 1H), 0.51 (m, 2H), (0.62, 0.86) (m,
1H), (0.78, 0.80) (s, 3H), (4.01, 4.38) (d, J 6.3, 1H), (4.45, 4.61)
(d, J 6.3, 1H), 7.18–7.71 (m, 10H); δC(CDCl3 75 MHz) (1.1,
1.3), (8.8, 10.6), (9.2, 10.8), (17.5, 17.8), (26.4, 26.5), (75.1,
77.5), (80.5, 80.7), (97.4, 97.5), (124.7, 125.0), (126.1, 126.5),
(126.3, 126.4), (127.4, 127.5), (127.6, 127.7), (142.8, 143.0),
144.2; νmax/cmϪ1 3059, 2998, 2956, 1597, 1449, 1253; m/z 322
(Mϩ , 22%), 193 (85), 178 (35), 165 (43), 115 (100), 91 (35), 83
᝽
(24).
Methyl 2-triethylsilyloxytricyclo[4.4.0.01,5]dec-2-ene-5-carb-
oxylate 12
Potassium hydride (254 mg, 1.4 mmol, 20–22% dispersion in
mineral oil) was washed with freshly distilled light petroleum
(bp 40–60 ЊC) (5 × 5 ml) and dried under vacuum. 1,4-Dioxane
(5 ml), methyl 2-oxotricyclo[4.4.0.01,5]decane-5-carboxylate22
13 (200 mg, 0.96 mmol) and chlorotriethylsilane (218 µl, 1.3
mmol) were added and the resulting suspension was heated at
reflux for 24 h. The cooled mixture was quenched with satur-
ated aqueous sodium hydrogen carbonate (5 ml) and extracted
with diethyl ether (3 × 3 ml). The combined diethyl ether
extracts were dried (MgSO4), filtered and concentrated under
reduced pressure. Flash chromatography using light petroleum
(bp 60–80 ЊC)–diethyl ether (10:1) gave 12 (Rf 0.81) (84.3 mg,
27%) as a light yellow oil; δH(CDCl3, 400 MHz) 0.68 (q, J 10.6,
6H), 0.98 (t, J 10.6, 9H), 1.18–1.50 (m, 5H), 1.58–1.73 (m, 2H),
1.89 (m, 1H), 2.17 (m, 1H), 2.37 (dd, J 21.8, 2.8, 1H), 2.77 (dd,
J 21.8, 2.8, 1H), 3.70 (s, 3H), 4.21 (t, J 2.8, 1H); δC(CDCl3, 100
MHz) 4.7, 6.6, 18.0, 19.9, 20.9, 21.1, 33.3, 34.1, 35.5, 38.9, 51.4,
(Mϩ Ϫ CH2O, 38%), 293 (29), 255 (26), 171 (36), 170 (100),
᝽
156 (20), 155 (64), 142 (25), 141 (74), 114 (21), 105 (41), 80 (20),
77 (38), 75 (63), 74 (21), 73 (97); λmax(CH3CN)/nm 250, 220 (sh),
205 nm (log ε 5.09, 6.89, 6.95) (Found: C, 75.0; H, 8.3.
C22H28O2Si requires C, 75.0; H, 8.0%).
[3-(2,2-Dimethylcyclopropyl)-2,2-diphenyloxetan-3-yloxy]-
trimethylsilane 22. A solution of 6 (200 mg, 1.09 mmol) and
benzophenone (199 mg, 1.09 mmol) in acetonitrile (11.3 ml)
was degassed and irradiated at 25 ЊC for 24 h using a medium
pressure mercury lamp. The reaction mixture was then concen-
trated under reduced pressure and the resulting residue was
subjected to preparative thin layer chromatography using light
petroleum (bp 40–60 ЊC)–diethyl ether (8:1). The major prod-
uct was identified to be oxetane 22 (Rf 0.73) as a white solid (40
mg, 10%), mp 57–58 ЊC; δH(CDCl3, 300 MHz) (Ϫ0.20, Ϫ0.14)
(s, 9H), 0.28 (d, J 7.8, 1H), (0.42, 0.63) (s, 3H), (0.44, 0.88) (m,
1H), 0.98 (t, J 7.5, 1H), (1.02, 1.17) (s, 3H), (4.41, 4.47) (d, J 6.6,
1H), (4.50, 4.62) (d, J 6.6, 1H), 7.15–7.65 (m, 10H); δC(CDCl3,
75 MHz) (1.1, 1.6), (15.2, 17.2), (16.4, 18.0), (19.3, 19.6), (27.6,
28.8), (31.9, 32.3), (76.4, 79.6), (82.1, 82.2), (97.5, 98.1), (125.0,
126.4), (126.0, 126.2), (126.5, 126.6), 127, (127.4, 127.7), (127.4,
127.6), (142.5, 143.3), (143.6, 144.4); νmax/cmϪ1 3058, 2953,
95.0, 159.0, 172.4; νmax/cmϪ1 2929, 1224, 1108; m/z 322 (Mϩ
,
᝽
100%), 321 (21), 293 (51), 291 (29), 264 (21), 263 (80), 233 (33),
159 (52), 131 (24), 117 (21), 115 (37), 103 (27), 87 (77), 75 (27),
59 (69) (HRMS: C18H30O3Si requires Mϩ , 322.1964. Found:
᝽
Mϩ, 322.1966).
Photoreactions of cyclopropyl silyl enol ethers
3-Cyclopropyl-2,2-diphenyl-3-(trimethylsilyloxy)oxetane 19.
A solution of 4 (37 mg, 0.24 mmol) and benzophenone (52 mg,
0.28 mmol) in acetonitrile (2.5 ml) was degassed and irradiated
at Ϫ25 ЊC for 1 day using a medium pressure mercury lamp.
The reaction mixture was concentrated and the resulting
residue was subjected to preparative thin layer chromatography
using light petroleum (bp 60–80 ЊC)–diethyl ether (4:1) to give
oxetane 19 as a white crystalline solid (Rf 0.7) (59 mg, 73%),
mp 67–68 ЊC. The product 19 was further recrystallised from
light petroleum (bp 40–60 ЊC) at Ϫ25 ЊC for microanalysis;
δH(CDCl3, 300 MHz) Ϫ0.15 (s, 9H), 0.20–0.58 (m, 4H), 1.05 (m,
1H), 4.31 (d, J 6.2, 1H), 4.56 (d, J 6.2, 1H), 7.11–7.62 (m, 10H);
1654, 1489, 1448; m/z 336 (Mϩ Ϫ CH2O, 9%), 184 (35), 169
᝽
(26), 141 (100), 105 (20), 74 (24), 73 (61); λmax(CH3CN)/nm
250, 220 (sh), 205 nm (log ε 5.25, 6.68, 6.76) (Found: C, 75.2;
H, 8.6. C22H28O2Si requires C, 75.4; H, 8.3%).
(4,4,5,8,8-Pentaphenyl-3,9-dioxabicyclo[5.2.0]nonan-1-yloxy)-
trimethylsilane 26. A solution of 7 (300 mg, 1.29 mmol) and
benzophenone (282 mg, 1.55 mmol) in acetonitrile (13.5 ml)
J. Chem. Soc., Perkin Trans. 1, 1998
2369

View Article Online
was degassed and irradiated at 25 ЊC for 2.5 days using a
medium pressure mercury lamp. The reaction mixture was con-
centrated. The resulting residue was subjected twice to prepar-
ative thin layer chromatography using light petroleum (60–
80 ЊC)–ethyl acetate (9:1). The high Rf component was isolated
to give a colourless oil 26 (Rf 0.59) (119 mg, 15%); δH(CDCl3,
300 MHz) 0.08 (s, 9H), 1.83 (m, 1H), 1.96 (t, J 13.6, 1H), 3.23
(dd, J 12.5, 6.2, 1H), 3.74 (d, J 12.9, 1H), 3.80 (d, J 12.9, 1H),
4.28 (d, J 15.4, 1H), 6.77–7.41 (m, 25H); δC(CDCl3, 75 MHz)
1.5, 31.9, 50.3, 71.1, 83.4, 86.6, 103.5, 125.1, 125.4, 125.6, 125.7,
126.5, 126.6, 127.1, 127.2, 127.6, 127.7, 128.3, 130.8, 139.8,
142.8, 144.1, 146.7, 146.9; νmax/cmϪ1 3058, 3024, 2954, 1599,
For 31: δH(CDCl3, 400 MHz) 0.24 (s, 9H), 2.17 (ddd, J 15.9,
7.7, 2.8, 1H), 2.84 (m, 1H), 3.58 (m, 1H), 3.71 (s, 3H), 4.36 (m,
2H), 4.93 (d, J 5.6, 1H), 5.08 (ddd, J 7.6, 4.1, 1.6, 1H), 6.61–7.20
(m, 9H); δC(CDCl3, 75 MHz) 0.3, 27.1, 50.8, 55.1, 71.5, 84.5,
106.5, 112.7, 125.9, 127.6, 127.8, 128.9, 132.4, 142.0, 151.8,
158.3; νmax/cmϪ1 3030, 3001, 2958, 1666, 1612, 1513, 1453, 1251;
m/z 368 (Mϩ , 3%), 233 (22), 232 (100), 231 (29), 142 (26), 141
᝽
(71), 75 (20), 73 (39); λmax(CH3CN)/nm 279, 273, 219.5 (sh), 208
(log ε 6.16, 6.23, 7.15, 7.33) (HRMS: C22H28O3Si requires Mϩ
,
᝽
368.1808. Found: Mϩ, 368.1811).
(2R*,3S*)-2-Phenyl-6-trimethylsilyloxy-2,3,4,7-tetrahydro-
oxepine-3-carboxylic acid ethyl ester 33 and (2R*,3R*)-2-phenyl-
6-trimethylsilyloxy-2,3,4,7-tetrahydrooxepine-3-carboxylic acid
ethyl ester 34. A solution of 8 (50 mg, 0.22 mmol), benzalde-
hyde (38 µl, 0.33 mmol) and sodium carbonate (5 mg) in ben-
zene (2.3 ml) were degassed and irradiated at 25 ЊC for 24.5 h
using a medium pressure mercury lamp. The reaction mixture
was concentrated under reduced pressure and the resulting resi-
due was purified using HPLC [light petroleum (bp 40–60 ЊC)–
diethyl ether elution, 6:1, 2 ml minϪ1] to afford a colourless oil
33 (tR 10.0 min) (6 mg, 8%) and enol ether 34 as a colourless oil
(tR 11.0 min) (6 mg, 8%).
For 33: δH(CDCl3, 300 MHz) 0.13 (s, 9H), 1.17 (t, J 7.2, 3H),
2.16 (m, 1H), 2.80 (m, 2H), 3.54 (d, J 10.5, 1H), 3.78 (m, 2H),
4.07 (q, J 7.2, 2H), 4.96 (dm, J 6.6, 1H), 7.21–7.67 (m, 5H);
δC(CDCl3, 75 MHz) 0.1, 13.9, 24.9, 53.8, 60.4, 65.7, 80.1, 96.1,
126.5, 127.3, 128.0, (3 × C) not observed; νmax/cmϪ1 2984, 1729,
1675, 1445, 1253.
1493, 1447, 1264; m/z 596 (Mϩ , 0.78%), 399 (29), 313 (32), 297
᝽
(52), 282 (24), 270 (27), 256 (20), 255 (21), 232 (69), 207 (48),
193 (50), 167 (68), 115 (22), 105 (30), 91 (24), 75 (33), 73 (100),
58 (38); λmax(CH3CN)/nm 258 (log ε 6.24) (HRMS: C40H40O3Si
requires Mϩ , 596.2747. Found: Mϩ, 596.2750).
᝽
(6R*,7S*)-(6,7-Diphenyl-2,5,6,7-tetrahydrooxepin-3-yloxy)-
trimethylsilane 29 and (6R*,7S*)-6,7-diphenyloxepan-3-one 32.
A solution of 7 (98 mg, 0.43 mmol), benzaldehyde (29 µl, 0.29
mmol) and sodium carbonate (5 mg) in benzene (3.0 ml) were
degassed and irradiated at 25 ЊC for 17.5 h using a medium
pressure mercury lamp. The reaction mixture was concentrated
under reduced pressure and the residue was purified using
HPLC [light petroleum (40–60 ЊC)–diethyl ether elution, 10:1,
3.0 ml minϪ1] to afford a colourless oil 29 (tR 6.0 min) (26 mg,
27%) and ketone 32 as a white crystalline solid (tR 24.7 min)
(5 mg, 6%), mp 99–100 ЊC.
For 29: δH(CDCl3, 300 MHz) 0.25 (s, 9H), 2.38 (m, 1H), 2.88
(m, 1H), 3.22 (dt, J 10.5, 5.2, 1H), 4.13 (d, J 16.0, 1H), 4.42 (d,
J 16.0, 1H) 4.86 (d, J 10.2, 1H), 5.17 (t, J 6.7, 1H), 7.05–7.21 (m,
10H); δC(CDCl3, 75 MHz) 0.3, 28.8, 53.6, 71.9, 86.9, 104.9,
125.9, 126.3, 126.6, 126.9, 127.3, 127.6, 127.8, 127.9, 128.2,
128.9, 141.5, 143.7, 151.0; νmax/cmϪ1 3062, 3031, 2959, 1666,
For 34: δH(CDCl3, 300 MHz) 0.19 (s, 9H), 1.26 (t, J 7.1, 3H),
238 (m, 3H), 4.14 (q, J 7.1, 2H), 4.21 (m, 2H), 4.67 (d, J 2.9,
1H), 4.91 (m, 1H), 7.20–7.63 (m, 5H); δC(CDCl3, 75 MHz) 0.5,
14.2, 20.8, 34.0, 60.2, 69.7, 82.9, 110.2, 125.5, 128.0, 128.4,
(3 × C) not observed; νmax/cmϪ1 2960, 1737, 1692, 1253.
[1-Hydroxy-1,1-(4-diphenyl)methyl]bicyclo[3.1.0]hexan-2-one
36. A solution of 10 (60 mg, 0.36 mmol) and benzophenone (78
mg, 0.43 mmol) in acetonitrile (5 ml) was degassed and irradi-
ated at 25 ЊC for 30 h using a medium pressure mercury lamp.
The reaction mixture was concentrated under reduced pressure
and the resulting residue was subjected to preparative thin layer
chromatography using light petroleum (bp 60–80 ЊC)–diethyl
ether (4:1) as eluent. The major band was isolated to give a
white crystalline solid 36 (tR 0.03) (9 mg, 9%), mp 180–181 ЊC;
δH(CDCl3, 300 MHz) 0.96 (q, J 4.8, 1H), 1.15 (m, 1H), 1.84 (m,
1H), 1.92 (m, 1H), 2.17 (d, J 7.6, 1H), 2.22 (d, J 8.4, 1H), 2.27
(s, 1H), 3.49 (dd, J 8.2, 1.4, 1H), 7.18–7.56 (m, 10H); νmax/cmϪ1
1494, 1452, 1382, 1253; m/z 338 (Mϩ , 6%), 233 (22), 232 (100),
᝽
180 (22), 142 (24), 141 (61), 75 (22), 73 (52); λmax(CH3CN)/nm
257 (log ε 5.96) (HRMS: C21H26O2Si requires Mϩ , 388.1702.
᝽
Found: Mϩ, 388.1709).
For 32: δH(CDCl3, 300 MHz) 2.00–2.20 (m, 2H), 2.63 (ddd,
J 12.5, 7.5, 1.6, 1H), 3.24 (m, 2H), 4.22 (d, J 18.5, 1H), 4.37 (d,
J 10.2, 1H), 4.45 (d, J 18.5, 1H), 6.88–7.15 (m, 10H); δC(CDCl3,
75 MHz) 31.1, 41.9, 55.4, 77.7, 92.2, 126.5, 126.6, 127.5, 127.8,
127.9, 128.3, 141.6, 141.7, 218.5; νmax/cmϪ1 3062, 3031, 2925,
1716, 1603, 1453, 1331, 1252; m/z 266 (Mϩ , 2.6%), 160 (100),
᝽
117 (29), 104 (85), 91 (24); λmax(CH3CN)/nm 255.5 (log ε 6.98)
(HRMS: C18H18O2 requires Mϩ , 266.1307. Found: Mϩ,
3400, 3031, 2915, 2894, 1704; m/z 279 (Mϩ , 3%), 205 (27), 183
᝽
᝽
266.1304).
(100), 105 (90), 77 (57); λmax(CH3CN)/nm 210, 205 (log ε 6.55,
[(6R*,7S*)-7-(4-Methoxyphenyl)-6-phenyl-2,5,6,7-tetrahydro-
oxepin-3-yloxy]trimethylsilane 30 and [(6R*,7R*)-7-(4-
methoxyphenyl)-6-phenyl-2,5,6,7-tetrahydrooxepin-3-yloxy]-
trimethylsilane 31. A solution of 7 (72 mg, 0.31 mmol), 4-
methoxybenzaldehyde (36 µl, 0.29 mmol) and sodium carbon-
ate (5 mg) in benzene (3 ml) were degassed and irradiated at
25 ЊC for 24 h using a medium pressure mercury lamp. The
reaction mixture was concentrated under reduced pressure and
the resulting residue was purified using HPLC [light petroleum
(bp 40–60 ЊC)–diethyl ether elution, 5:1, 2.5 ml minϪ1] to afford
a colourless oil 30 (tR 7.6 min) (43 mg, 40%) and enol ether 31
as a colourless oil (tR 7.8 min) (13 mg, 12%).
6.68) (HRMS: C19H18O2 requires Mϩ , 278.1307. Found: Mϩ,
᝽
278.1309).
6,7,7-Triphenyloxepan-3-one 28
To a solution of 26 (23 mg, 0.04 mmol) and 3 Å molecular
sieves (10 mg) in dry dichloromethane (1 ml) at 0 ЊC was added
tetrabutylammonium fluoride (44 µl, 1  in THF). The reaction
mixture was warmed to room temperature and stirred for 30
min. The resulting solution was diluted with water (1 ml) and
extracted with dichloromethane (3 × 1 ml). The organic layer
was dried (MgSO4), filtered and concentrated under reduced
pressure. The residue was purified using HPLC [light petroleum
(bp 40–60 ЊC)–diethyl ether elution, 6:1, 2.0 ml minϪ1] to afford
a white crystalline solid 28 (tR 17.0 min) (11 mg, 82%), mp
164–165 ЊC and benzophenone (tR 11.0 min) (5 mg, 70%);
δH(CDCl3, 300 MHz) 2.30 (m, 3H), 3.02 (m, 1H), 4.10 (d,
J 17.9, 1H), 4.18 (d, J 17.9, 1H), 4.35 (dd, J 6.3, 3.0, 1H), 7.00–
7.56 (m, 15H); δC(CDCl3, 75 MHz) 26.7, 37.9, 49.5, 71.7, 87.5,
126.2, 126.3, 126.8, 127.2, 127.6, 127.7, 128.3, 128.6, 130.3,
140.9, 142.4, 145.5, 213.7; νmax/cmϪ1 3056, 2928, 1712, 1598,
For 30: δH(CDCl3, 300 MHz) 0.24 (s, 9H), 2.42 (m, 1H), 2.82
(m, 1H), 3.33 (dt, J 10.2, 5.1, 1H), 3.74 (s, 3H), 4.08 (d, J 16.2,
1H), 4.38 (d, J 16.2, 1H), 4.68 (d, J 10.2, 1H), 5.16 (t, J 6.7, 1H),
6.72–7.21 (m, 9H); δC(CDCl3, 75 MHz) 0.4, 28.9, 53.3, 55.2,
71.2, 86.3, 105.1, 113.4, 126.3, 127.9, 128.2, 128.3, 133.6, 143.9,
151.1, 158.8; νmax/cmϪ1 3030, 2958, 1666, 1612, 1513, 1453,
1251; m/z 368 (Mϩ , 3%), 233 (44), 232 (100), 231 (57), 217 (20),
᝽
210 (29), 209 (33), 143 (32), 142 (53), 141 (78), 121 (21), 115
(20), 75 (37), 73 (55); λmax(CH3CN)/nm 280, 270, 221 (log ε 6.30,
1493, 1265; m/z 342 (Mϩ , 0.4%), 162 (23), 161 (100), 117 (22),
᝽
6.35, 7.18) (HRMS: C22H28O3Si requires Mϩ , 368.1808.
105 (30), 104 (27); λmax(CH3CN)/nm 257 (log ε 6.11) (HRMS:
᝽
Found: Mϩ, 368.1804).
C24H22O2 requires Mϩ , 342.1697. Found: Mϩ, 342.1650).
᝽
2370
J. Chem. Soc., Perkin Trans. 1, 1998

View Article Online
Photochemistry of (bicyclo[4.1.0]hept-2-en-2-yloxy)trimethyl-
silane 11 with benzophenone. Formation of (9,9-diphenyl-8-
oxatricyclo[5.2.0.02,4]nonan-1-yloxy)trimethylsilane 37, (8,8-
diphenyl-9-oxatricyclo[5.2.0.02,4]nonan-1-yloxy)trimethylsilane
39, (8,8-diphenyl-1-trimethylsilyloxy-9-oxatricyclo[5.2.0.02,4]-
nonan-6-yloxy)diphenylmethanol 38 and (2-trimethyl silyloxy-
bicyclo[4.1.0]hept-3-en-2-yl)diphenylmethanol 40
7H), 3.25 (br s, 1H), 5.52 (m, 1H), 5.63 (m, 1H), 7.20–7.75 (m,
10H); νmax/cmϪ1 3545, 3469, 3015, 2897, 1493, 1447, 1346; m/z
274 (Mϩ Ϫ H2O, 7%), 256 (22), 184 (22), 183 (84), 169 (58);
᝽
λmax(CH3CN)/nm 257.5, 218 (sh) nm (log ε 5.88, 6.94) (HRMS:
C20H20O2 requires (Mϩ Ϫ H2O), 274.1358. Found: 274.1355).
᝽
4-Hydroxy-5,6,7,8-tetrahydronaphthalene-1-carboxylic acid
methyl ester 47
A solution of 11 (1.05 g, 5.8 mmol) and benzophenone (1.26 g,
6.9 mmol) in acetonitrile (60 ml) were degassed and irradiated
at 25 ЊC for 68 h using a medium pressure mercury lamp. The
reaction mixture was concentrated under reduced pressure and
the resulting residue was subjected to preparative thin layer
chromatography using light petroleum (bp 60–80 ЊC)–ethyl
acetate (9:0.5). A number of products were isolated and were
identified to be oxetane 37 (Rf 0.55) (31 mg, 1.5%) obtained
as a colourless oil, 39 (Rf 0.48) (42 mg, 2%), a white crystalline
solid 38 (Rf 0.40) (18 mg, 0.6%), mp 126–128 ЊC and a white
crystalline solid 40 (Rf 0.35) (136 mg, 6%), mp 62–64 ЊC.
For 37: δH(CDCl3, 300 MHz) Ϫ0.05 (s, 9H), 0.35 (q, J 5.4,
1H), 0.46 (m, 1H), 0.82 (m, 1H), 1.03 (td, J 8.6, 5.4, 1H), 1.42
(m, 1H), 1.63 (m, 1H), 1.81 (dm, J 15.6, 1H), 2.07 (tt, J 13.6,
4.5, 1H), 4.52 (s, 1H), 7.13–7.70 (m, 10H); δC(CDCl3, 75 MHz)
1.9, 4.5, 13.6, 16.0, 18.1, 23.0, 82.1, 92.5, 125.1, 126.2, 126.4,
126.9, 127.3, 127.6, 143.3, 143.7, 1 × C not observed; νmax/cmϪ1
A solution of 12 (30 mg, 0.093 mmol) and benzophenone (19
mg, 0.10 mmol) in pentane (2.5 ml) were degassed and irradi-
ated at 25 ЊC for 24 h using a medium pressure mercury lamp.
The reaction mixture was concentrated under reduced pressure
and the residue was purified using HPLC [9:1 light petroleum
(bp 60–80 ЊC)–ethyl acetate elution, 2.0 ml minϪ1] to afford an
oil (tR 9.3 min) (14.9 mg). To a solution of this oil (14.9 mg) and
3 Å molecular sieves (30 mg) in dry dichloromethane (0.5 ml)
at 0 ЊC was added tetrabutylammonium fluoride (49 µl, 1  in
THF). The reaction mixture was warmed to room temperature
and stirred for 30 min. The resulting solution was diluted with
water (1 ml) and extracted with dichloromethane (3 × 2 ml).
The organic layer was dried (MgSO4), filtered and concentrated
under reduced pressure. The residue was purified using prepar-
ative thin layer chromatography [light petroleum (bp 60–
80 ЊC)–diethyl ether, 4:1] to afford a white crystalline solid 47
(Rf 0.17) (3 mg, 27%), mp 137–139 ЊC; δH(CDCl3, 400 MHz)
0.80–1.88 (m, 6H), 2.65 (t, J 6.2, 1H), 3.09 (t, J 6.2, 1H), 3.84
(s, 3H), 5.05 (s, 1H), 6.63 (d, J 8.8, 1H), 7.70 (d, J 8.8, 1H);
δC(CDCl3, 100 MHz) 21.8, 22.6, 23.3, 28.4, 51.6, 111.4, 127.7,
127.8, 129.8, 141.9, 156.6, 167.9; νmax/cmϪ1 3324, 2920, 1724,
3056, 3020, 2926, 1447, 1357, 1250, 1159; m/z 364 (Mϩ , 4%),
᝽
183 (35), 182 (100), 167 (30), 149 (23), 105 (42), 77 (29), 75 (27),
73 (92), 69 (34), 57 (35), 55 (35); λmax(CH3CN)/nm 257, 220
(log ε 5.78, 6.98) (HRMS: C23H28O2Si requires Mϩ , 364.1855.
᝽
Found: Mϩ, 364.1941).
1581, 1431, 1260; m/z 206 (Mϩ , 53%), 180 (31), 175 (56), 174
For 39: δH(CDCl3, 300 MHz) 0.13 (s, 9H), 0.46 (m, 2H), 0.72
(td, J 8.4, 5.1, 1H), 1.50 (m, 2H), 1.66 (m, 2H), 1.87 (m, 1H),
3.12 (br s, 1H), 7.12–7.54 (m, 10H); δC(CDCl3, 75 MHz) 1.7,
8.8, 14.3, 25.8, 27.2, 37.1, 44.0, 88.0, 108.2, 126.0, 126.1, 126.2,
126.4, 127.6, 127.9, 145.6, 147.7; νmax/cmϪ1 3063, 3020, 2956,
᝽
(100), 173 (31), 149 (40), 148 (21), 147 (47), 146 (31), 145 (28),
131 (21), 121 (23), 115 (33), 107 (30), 105 (39), 91 (57), 79 (30),
77 (46), 71 (27), 69 (25), 65 (27), 57 (59), 55 (34), 51 (26);
λmax(CH3CN)/nm 252, 219 (sh), 207 (log ε 6.26, 6.33, 6.40)
1446, 1311, 1249, 1150; m/z 364 (Mϩ , 0.93%), 182 (100), 167
(HRMS: C12H14O3 requires Mϩ , 206.0943. Found: Mϩ,
᝽
᝽
(32), 73 (85); λmax(CH3CN)/nm 218 (sh), 204 nm (log ε 6.74,
206.0946).
6.87) (HRMS: C23H28O2Si requires Mϩ , 364.1858. Found: Mϩ,
᝽
364.1828).
Acknowledgements
For 38: δH(CDCl3, 300 MHz) Ϫ0.05 (s, 9H), 0.62 (m, 2H),
0.95–1.30 (m, 4H), 1.65 (m, 1H), 2.95 (m, 1H), 3.68 (d, J 3.6,
1H), 6.73–7.71 (m, 20H); δC(CDCl3, 75 MHz) 1.6, 5.0, 13.4,
18.6, 23.6, 38.9, 53.0, 79.8, 83.2, 101.3, 124.6, 125.6, 126.0,
126.2, 126.8, 127.3, 127.4, 127.6, 128.0, 128.3, 128.6, 144.5,
145.6, 147.2, 148.9; νmax/cmϪ1 3412, 2359, 1641, 1446, 1250; m/z
We thank the Research School of Chemistry, Australian
National University for postgraduate and tuition fee scholar-
ships for C. Y. G., T. Culnane and Craig Morton for their
assistance in the acquisition of NMR spectra and the School of
Chemistry, University of Melbourne for financial support.
We also thank A. L. J. Beckwith and M. G. Banwell for their
helpful advice and suggestions.
528 (Mϩ Ϫ H2O, 0.54%), 183 (44), 182 (29), 181 (78), 167 (25),
᝽
105 (100), 77 (40), 75 (46), 73 (71); λmax(CH3CN)/nm 253.5,
219 (log ε 6.48, 7.34) (HRMS: C36H38O3Si requires Mϩ
TMS Ϫ OH Ϫ Ph, 379.1698. Found: 379.1691).
Ϫ
᝽
References
For 40: δH(CDCl3, 400 MHz) 0.09 (s, 9H), 0.45–2.20 (m, 7H),
5.50 (m, 1H), 5.58 (m, 1H), 7.10–7.80 (m, 10H); δC(CDCl3, 100
MHz) 2.3, 7.3, 11.9, 20.1, 24.3, 81.1, 83.7, 126.4, 126.9, 127.0,
128.7, 129.0, 144.6, 144.9, 2 × CH not observed; νmax/cmϪ1
1 E. Paterno and G. Chieffi, Gazz. Chim. Ital., 1909, 39, 341.
2 G. Büchi, C. G. Inman and E. S. Lipinsky, J. Am. Chem. Soc., 1954,
76, 4327.
3 G. S. Hammond and N. J. Turro, Science, 1963, 142, 1541.
4 J. S. Searles, ‘Oxetanes’, in Heterocyclic Compounds with Three- and
Four-Membered Rings, Part 2, Interscience, 1964, p. 984.
5 J. A. Porco and S. L. Schreiber, ‘The Paternò–Büchi Reaction’,
in Comprehensive Organic Synthesis, ed. B. M. Trost, Pergamon,
New York, 1991, p. 151.
3557, 3021, 2950, 1642, 1250, 1114; m/z 364 (Mϩ , 8%), 361
᝽
(28), 183 (22), 182 (28), 181 (100); λmax(CH3CN)/nm 258, 218 (sh)
(log ε 5.87, 6.75) (Found: C, 75.5; H, 7.7. C23H28O2Si requires C,
75.8; H, 7.7%).
2-(1-Hydroxy-1,1-diphenylmethyl)bicyclo[4.1.0]hept-3-en-2-ol
44
6 M. Bartok and K. L. Lang, ‘Oxiranes’, in Small Ring Heterocycles
(Part 3), ed. A. Hassner, Wiley, New York, 1964, p. 89.
7 D. S. Watt, S. Vasudevan, P. Brock and H. Morita, J. Org. Chem.,
1994, 59, 4677.
8 T. Bach and K. Jödicke, Chem. Ber., 1993, 126, 2457.
9 T. Bach, Tetrahedron Lett., 1994, 35, 5845.
10 H. A. J. Carless, J. Beanland and S. Mwesigye-Kibende, Tetrahedron
Lett., 1987, 28, 5933.
11 G. I. Jones, ‘Synthetic Applications of the Paternò–Büchi Reaction’,
in Organic Photochemistry, ed. A. Padwa, Marcel Dekker, New
York, 1981, p. 2.
To a solution of 40 (23 mg, 0.06 mmol) and 3 Å molecular
sieves (10 mg) in dry dichloromethane (1 ml) at 0 ЊC was added
tetrabutylammonium fluoride (72 µl, 1  in THF). The reaction
mixture was warmed to room temperature and stirred for 30
min. The resulting solution was diluted with water and
extracted with dichloromethane. The organic layer was dried
(MgSO4), filtered and concentrated under reduced pressure.
The residue was purified using HPLC [light petroleum (bp 40–
60 ЊC)–diethyl ether elution, 6:1, 2.0 ml minϪ1] to afford the
title compound 44 as a white crystalline solid (tR 30.0 min) (13
mg, 16%), mp 99–103 ЊC; δH(CDCl3, 300 MHz) 0.25–2.38 (m,
12 W. Horspool and D. Armesto, Organic Photochemistry: A Compre-
hensive Treatment, Prentice Hall, London, 1992.
13 G. Büchi, J. T. Kofron, E. Koller and D. Rosenthal, J. Am. Chem.
Soc., 1956, 78, 876.
J. Chem. Soc., Perkin Trans. 1, 1998
2371

View Article Online
14 M. Yamaguchi and I. Hirao, Tetrahedron Lett., 1984, 25, 4549.
15 M. Sevrin and A. Krief, Tetrahedron Lett., 1980, 21, 585.
16 D. R. Arnold, ‘The Photocycloaddition of Carbonyl Compounds to
Unsaturated Systems: The Synthesis of Oxetanes’, in Advances in
Photochemistry, ed. W. A. Noyes, G. S. Hammond and J. N. Pitts,
Interscience, New York, 1968, p. 301.
28 J. N. Lambert, C. Y. Gan and R. W. Gable, Acta Crystallogr., 1997,
C53, 1122.
29 A. E. Derome, Modern NMR Techniques for Chemistry Research,
Pergammon, London, 1987.
30 W. C. Still, M. Kahn and A. Mitra, J. Org. Chem., 1978, 43, 2923.
31 D. D. Perrin and W. L. F. Armarego, Purification of Laboratory
Chemicals, Pergamon, Sydney, 1986.
32 P. A. Grieco and Y. Ohfune, J. Org. Chem., 1978, 43, 2720.
33 R. M. Roberts, R. G. Landolt, R. N. Greene and E. W. Heyer,
J. Am. Chem. Soc., 1967, 89, 1404.
34 J. Baluenga, P. L. J. Bernad, J. M. Concellon, A. Pinera-Nicholas
and S. Garcia-Granda, J. Org. Chem., 1997, 62, 6870.
35 B. K. Carpenter, R. H. Newman-Evans and R. J. Simon, J. Org.
Chem., 1990, 55, 695.
17 S. Nishida, N. Shimizu, M. Ishikawa and K. Ishikura, J. Am. Chem.
Soc., 1974, 96, 6456.
18 P. J. Wagner and K. L. Cheng, J. Am. Chem. Soc., 1994, 116, 7945.
19 D. C. Neckers and S. Hu, J. Org. Chem., 1997, 62, 755.
20 D. C. Nonhebel, Chem. Soc. Rev., 1993, 347.
21 E. J. Corey and M. Chaykovsky, J. Am. Chem. Soc., 1965, 87, 1353.
22 H. Hagiwara, F. Abe and H. Uda, J. Chem. Soc., Perkin Trans. 1,
1993, 2651.
23 M. W. Majchrzak and A. Kotelko, Synthesis, 1983, 469.
24 H. M. Walborsky and F. M. Hornyak, J. Am. Chem. Soc., 1955, 77,
6026.
36 H. O. House, A. V. Prabhu, J. M. Wilkins and L. F. Lee, J. Org.
Chem., 1976, 41, 3067.
25 F. J. Impastato and H. M. Walborsky, J. Am. Chem. Soc., 1962, 84,
4838.
Paper 8/02922I
Received 2nd April 1998
Accepted 29th May 1998
26 T. Bach, Liebigs Ann./Recueil, 1997, 1627.
27 A. G. Griesbeck and S. Stadtmüller, J. Am. Chem. Soc., 1991, 113,
6923.
2372
J. Chem. Soc., Perkin Trans. 1, 1998