SmI2-mediated 3-exo-trig cyclisation of δ-oxo-α, β-unsaturated esters to cyclopropanols and derivatives

Article abstract of DOI:10.1016/j.tet.2004.05.074

In the presence of samarium diiodide and a proton source, δ-oxo-γ,γ-disubstituted-α,β-unsaturated esters of general formula R-CO-C(R′,R′)-CHCH-CO₂Bn readily cyclise to trans-cyclopropanol products and/or lactones derived from the cis isomers. F

Full text of DOI:10.1016/j.tet.2004.05.074

Tetrahedron 60 (2004) 6931–6944
SmI2-Mediated 3-exo-trig cyclisation of d-oxo-a,b-unsaturated
esters to cyclopropanols and derivatives
*
Sophie Bezzenine-Lafollee, Francois Guibe, Helene Villar and Riadh Zriba
´
´
´ `
´ ´ ´ ˆ ´
Institut de Chimie Moleculaire et des Materiaux d’Orsay, Laboratoire de Catalyse Moleculaire, UMR-8075, Bat 420, Universite Paris-Sud,
91405 Orsay, France
Received 12 March 2004; revised 19 May 2004; accepted 21 May 2004
Abstract—In the presence of samarium diiodide and a proton source, d-oxo-g,g-disubstituted-a,b-unsaturated esters of general formula
R–CO–C(R0,R0)–CHvCH–CO2Bn readily cyclise to trans-cyclopropanol products and/or lactones derived from the cis isomers. For
R¼aryl, good stereoselectivities (ca 90%) in favor of the alcohols are generally obtained while a mixture of alcohols and lactones is obtained
with R¼alkyl or H. For R¼cyclopropyl, the lactone is exclusively obtained in more than 90% yield. A mechanistic rationalisation of these
variations of diastereoselectivity is proposed.
q 2004 Elsevier Ltd. All rights reserved.
1. Introduction
further fast radical reactions such as cyclisations2e,f (cascade
processes) or fragmentations (b-elimination of thiyl
group).4
In the past 20 to 30 years, radical cyclisations have been
extensively used, often with great success, for obtaining
carbocycles, especially five-membered and to a lesser extent
six-membered ring systems. Meanwhile, few radical
cyclisations have been described, which lead to cyclo-
propanes. This paucity of results is easily accounted for by
the fact that, contrary to 5-exo-trig cyclisations which are
usually fast and irreversible, 3-exo-trig radical cyclisations,
albeit kinetically feasible, are on the other hand usually
thermodynamically strongly disfavored. Indeed, in the
parent system, cyclisation of the homoallyl radical is
somewhat 104 times slower (kC¼1.0£104 s21 vs
k2C¼1.3£108 s21 at 25 8C)1 than reopening of the cyclo-
propylmethyl radical. As a result, only homoallylic systems
with very specific features have been successfully cyclised
to cyclopropane molecules under exclusive radical con-
ditions. Such specific features include structural constraints2
or the presence of groups able, once the cyclopropylcarbinyl
radical is formed, either to stabilise it3 or to involve it into
A few years ago, we reported that d-bromo and d-iodo-a,b-
unsaturated esters could be cyclised to cyclopropane
compounds in the presence of two equivalents of samarium
diiodide and a proton donor (typically tert-butanol).5 The
success of this procedure was attributed to the known ability
of SmI2 to promote radical-anionic tandem reactions, a
property that has extensively been used for various synthetic
purposes.6a,b In our case, the following mechanism has been
proposed5 (Scheme 1): the homoallylic radical initially
formed by monoelectronic reduction of the starting halide
cyclises to a-carbalkoxy-substituted cyclopropylcarbinyl
radical. Despite the presence of the carbalkoxy substituent,
kinetic measurements have shown that this equilibrated
process is still in favor of the open radical form, albeit to a
lesser extent than in the parent system (kc/k2c¼ca 10).7
This, however, is of no consequence as the displacement of
the overall reaction towards cyclisation is ensured by
Scheme 1.
Keywords: Samarium diiodide; Radical cyclisations; Cyclopropanols; Ketyl radicals.
*
Corresponding author. Tel.: þ33-1-69-15-52-59; fax: þ33-1-69-15-46-80; e-mail address: fraguibe@icmo.u-psud.fr
0040–4020/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2004.05.074

´
S. Bezzenine-Lafollee et al. / Tetrahedron 60 (2004) 6931–6944
6932
Figure 1.
Scheme 2.
subsequent and irreversible facile monoelectronic reduction
of the cyclopropylcarbinyl radical to the corresponding
enolate.† We later showed8 that considerable racemisation
was observed in the cyclisation of enantioenriched d-halo-
a,b-unsaturated esters bearing a substituent at the
g-position. Such racemisation was accounted for by
assuming that reopening of the cyclopropylcarbinyl radical
is probably faster that its reduction to enolate (k2c.kr2ed
[SmI2] in Scheme 1).
samarium of various functional groups in the substrate
molecule may however complicate the stereochemical
issue.11
A preliminary communication on the cyclisation of d-oxo-
a,b-unsaturated esters had already been issued12 essentially
dealing with aldehydic substrates. We present here a full
report of our work which includes aromatic and alkyl
ketones. For reasons that will be later specified some
cyclisations of aldehydic compounds were also reinvesti-
gated. Several results given in our preliminary communi-
cation were thus found erroneous and have been
consequently revised.
Notwithstanding the racemisation problem, the SmI2
mediated cyclisation of d-halo-a,b-unsaturated esters was
found to tolerate the presence of various substituents at the
b, g and d positions (a-position was not tested in this
respect). Unfortunately, the diastereoselectivities of these
reactions are low and mixtures of cis and trans substituted
cyclopropanes were usually obtained in comparable
amounts. We then turned our attention to the cyclisation
of d-oxo-a,b-unsaturated esters as a way to obtain
cyclopropanols through formation of the corresponding
ketyl radicals and their subsequent intramolecular addition
on the double bond. It was hoped that better diastereo-
selectivities would be attained in these reactions. Indeed,
many examples of related SmI2 promoted 5-exo-trig, 6-exo-
trig and even 4-exo-trig cyclisations of z- h- and e-oxo-
enoates respectively can be found in the literature. These
reactions often display very good stereoselectivities9 due to
the fact that in the transition state, the C–O bond of the ketyl
radical and the CvC bond conjugated to the carbalkoxy
group adopt preferentially, for stereoelectronic reasons, an
antiparallel relationship.10 Steric effects or coordination to
2. Results
2.1. Synthesis of substrates for cyclisation
In order to avoid any complication which could arise from
accidental migration of the double bond from a,b to b,g
position, we have limited our studies to g,g-disubstituted
d-keto enoates of general formula 1–4 (Fig. 1) Those
include aromatic substrates 1a–d and 2a, alkyl ketones
1e–g and 2b and aldehydic compounds 3 and 4. Most
ketonic substrates were synthesized by the two-step
procedure of Scheme 2. In the first step, acylation of
morpholino-enamines derived from isobutyraldehyde or
cyclohexane–carboxaldehyde with the appropriate acyl
chloride followed by hydrolytic work-up, as described by
Inukai and Yoshigawa,13 gave the corresponding b-keto-
aldehydes. In a second step, Wadsworth–Emmons
condensation under standard conditions with benzyl
dimethoxyphosphono-acetate gave the desired products.
Probably due to more severe steric crowding, acylation of
isobutyraldehyde morpholino-enamine with isobutyryl
chloride and with naphthoyl chloride gave very poor results.
†
An analogous radical–anionic tandem process has been proposed to
explain the formation of cyclopropyl ring in some nickel complex
catalysed electroreductive cyclisations: Ozaki, S.; Matsui, E; Waku, J.;
Ohmori, H. Tetrahedron Lett. 1997, 38, 2705–2708. See also Gassman,
P. G.; Lee, C. J. J. Am. Chem. Soc., 1989, 111, 739–740.

´
S. Bezzenine-Lafollee et al. / Tetrahedron 60 (2004) 6931–6944
6933
At this point of our investigation, we became aware of
another publication14 showing that acylation of pyrrolidino-
enamines takes place much more readily than that of
morpholino-enamines. Indeed, acylation of isobutyr-
aldehyde pyrrolidino-enamine with 1-naphthoyl chloride
gave after hydrolytic work-up the desired b-keto-aldehyde
in very satisfactory yield. As to 3-oxo-2,2,4-trimethyl-
pentanal, on the way to 1f, it was prepared by aldol
autocondensation of isobutyraldehyde in fair yield15
followed by oxidation with PCC.
different batches of SmI2 solution with good reproducibility.
Experimental procedures other than that described above—
especially inverse addition and reactions carried out in the
additional presence of four equivalents of HMPA—were
also sometimes investigated. Since no significant differ-
ences in the outcome of the reaction were observed, they are
not reported here.
In most cases only products of reductive cyclisation (Fig. 2)
were obtained in our experiments. However, uncyclised
products of direct reduction of the carbonyl (alcohol 9) or of
direct reduction of the carbon–carbon double bond
(saturated ester 10) were also found, in minor amounts, in
the reaction of methyl ketone 2b and isopropylketone 1f
respectively. In no cases could products of pinacol coupling
be detected. Products of reductive cyclisation were
cyclopropanols 11 in which the hydroxyl group and the
carbalkoxymethyl group are trans to each other relatively to
the cyclopropane ring and/or lactones 12 undoubtedly
originating from lactonisation of the preliminary formed
cis stereoisomeric cyclopropanol adducts. In the following
of the text, stereoisomers 11 will be referred to as ‘trans
cyclopropanols’. On NMR spectra, the methylene protons a
to carbonyl of lactones 12 display a very characteristic ABX
system made of one doublet and one doublet of doublets
(JAB¼ca 18 Hz, JAX¼ca 7 Hz, JBX¼0 Hz) while cyclo-
propanols 11 display the usual pair of doublets of doublets
(JBX–0 Hz).
As reported in our preliminary communication, the
aldehydic substrates 3 and 4 were prepared in three steps.
Morpholino-enamines derived from isobutyraldehyde or
cyclohexane–carboxaldehyde were first alkylated with
2-chloro-1,3-dithiane.16 Hydrolytic work-up gave dithiane
aldehydes 5 and 6 that were submitted to Wadsworth–
Emmons olefination. The latent aldehydic function was
finally unmasked by hydrolysis of the dithiane group in
water/acetone in the presence of methyl iodide and
collidine17 (Scheme 3).
2.2. Cyclisations: procedure and experimental results
SmI2-mediated cyclisations were conducted under inert
(argon) dry atmosphere. The procedure was generally as
follows: a 0.2 M solution of substrate in THF also contain-
ing 4 equiv. of tert-butanol was cooled to 0 8C. 2.2 equiv. of
a 0.1 M solution of SmI2 in THF was added dropwise
through a cannula over a period of approximately 2 min.
The reaction mixture was then stirred at room temperature
for several (usually four to twelve) hours until TLC analysis
on aliquots showed total consumption of the substrate. The
reaction mixture was then quenched with dilute aqueous
HCl. After standard work-up, the crude residue was first
analysed by NMR and then submitted to column chroma-
tography in order to isolate pure compounds. For a number
of substrates, the reaction was carried out twice with
The main results of our investigations are summarised in the
table. On the NMR spectra of the crude reaction products,
only two benzylic signals usually showed up, one
corresponding to the trans cyclopropanol adduct and the
other to benzyl alcohol. If we assume that all benzyl alcohol
comes from lactonisation of the primarily formed cis
cyclopropanol adduct, the anti/syn selectivity can also be
deduced from the relative intensity of the NMR benzylic
peaks. In the table, we have therefore reported both anti/syn
Scheme 3.
Figure 2.

´
S. Bezzenine-Lafollee et al. / Tetrahedron 60 (2004) 6931–6944
6934
Scheme 4.
selectivities deduced on the one hand from the respective
amounts of trans cyclopropanol and lactone isolated by
chromatography and on the other hand from NMR of the
crude reaction mixture. A fair agreement between the two
values was generally observed.
with alkyl ketones. From a kinetic point of view, recent rate
constant measurements with SmI2 as the one electron
reducing agent have shown that the reaction is 104 more
rapid with acetophenone than with 2-butanone.18 On the
basis of these considerations, the following mechanism
(mechanism I, Scheme 4) may be proposed for cyclisation
of our arylic substrates. The keto group is first reduced most
probably in a reversible way18,19 to ketyl radical 17 which
then adds to the double bond to give the cyclopropylcarbinyl
radical 18.‡ Further reduction to enolate by a second
molecule of SmI2 followed by protonation completes the
reaction. Mechanism I is therefore akin to the one proposed
earlier by ourselves for the cyclisation of d-halo-a,b-
unsaturated esters (see Scheme 1). It also corresponds to
what is generally invoked for SmI2 promoted intermolecular
condensation of conjugated enoic esters with carbonyl
compounds20 and, as already mentioned, for SmI2 promoted
4-exo, 5-exo and 6-exo-trig cyclisations of e-,z- and h-oxo-
enoates.9 Therefore, in the present 3-exo-trig cyclisations,
the anti stereoselectivity observed with substrates 1a–c and
2a could likewise be the result of a preference for a trans
relationship of the C–O bond of the ketyl radical and the
carbon–carbon double bond of the enoate moiety in the
transition states of cyclisation (Fig. 3, A preferred to B).
However, an essential difference between 3-exo-trig
cyclisations and 5- and 6-exo-trig-cyclisations is that the
former are reversible while the latter are not.§ As a result, in
Aromatic ketonic substrates (entries 1–4) were found to
give usually in high yield and with complete or very high
stereoselectively trans cyclopropanols and only small
amounts of lactones. With the more sterically encumbered
1-naphthyl substrate 1d however, the proportion of lactones
becomes important (entry 5). Concerning the cyclisation of
alkyl ketonic compounds (entries 6–8), a mixture of lactone
and of trans-cyclopropanol in comparable amounts was
obtained not only with isopropyl ketone 1f but even with the
less sterically demanding methyl ketone 1e and 2b.
Surprisingly, exclusive formation of the lactone in close-
to-quantitative yield was observed with the cyclopropyl
compound 1g (entries 9a–c).
Given the results obtained above with non-aromatic
ketones, we decided to reinvestigate the cyclisation of
some aldehydes because our previous observations that
cyclisation takes places with exclusive anti selectivity
seemed now dubious. Indeed, in this reinvestigation, we
found that these substrates also lead to a mixture of trans
cyclopropanols and of lactones (entries 10, 11). We also
found that the yields of the reactions was somewhat poorer
than with ketonic compounds, but neither uncyclised
products of direct reduction, nor products of pinacol
coupling were detected.
‡
It is thus assumed that cyclisation takes place at the radical stage. A
reaction sequence involving first two-electron reduction of the carbonyl
group and then cyclisation cannot be totally ruled out. However, the fact
that the cyclisation reactions are carried out in the presence of a proton
donor renders improbable a cyclisation at an anionic stage. For other
arguments in disfavor of anionic cyclisations (including the case where
the initial two-electron reduction occurs at the carbon–carbon double
bond of the enoate moiety instead of carbonyl) see Ref. 5 and references
cited therein.
3. Discussion
In our opinion, the collection of results presented here is too
limited to allow for a complete understanding of the
mechanisms by which d-oxo a,b-unsaturated esters cyclise
in the presence of SmI2. We will therefore limit ourselves to
some broad comments and mechanistic proposals must be
considered more as working hypotheses for further
investigation rather than as definite statements.
§
4-exo-trig cyclisations are potentially reversible but the k2c constant for
ring opening of cyclobutyl methyl ketyl radical (2.5£104 s21 at 25 8C) is
at least three orders of magnitude smaller than that for cyclopropyl methyl
ketyl radical.23 Therefore, the probability for a rapid equilibration
between open and cyclised radicals in samarium promoted reductice
cyclisation of e-oxo-a,b-unsaturated esters 9a–f appears low. An isolated
example of reductive opening by SmI2/DMPU, of an a-ketocyclobutane
ring within a fused tricyclic system, has been reported (Comins, D. L.;
Zheng, X. J. Chem. Soc.Chem. Commun. 1994, 2681–2682). Probably, in
this rigid structure, the fragmentation process is greatly facilitated by
favorable orbital overlapping.
Reduction of ketones to ketyl radicals is notoriously
thermodynamically much more favorable with aryl than

´
S. Bezzenine-Lafollee et al. / Tetrahedron 60 (2004) 6931–6944
6935
Figure 3.
our case the argument based on stereoelectronic preferences
in the transition state of cyclisation holds true only if the
retrocyclisation process is slow compared to the irreversible
reduction of cyclopropylcarbinyl radical to enolate
(k2c!k2red [SmI2] in Scheme 4). In the reverse limit case
(k2c@ k2red [SmI2]), application of the Curtin–Hammet
principle21 leads to the conclusion that the anti/syn
selectivity depends now on the relative energy of the
transition states leading from the trans and cis cyclo-
propylcarbinyl radicals C and D to the corresponding
enolates. As pointed out in our preliminary communi-
cation,12 the trans (anti) selectivity could then be explained
by the development of strong repulsive electronic inter-
actions during the reduction of the cis radical D to
carbanion. At the present time of our investigation, it is
difficult to ascertain which one (if any) of these two
situations prevails. But the second one (fast equilibrium
between homoallylic and cyclopropylcarbinyl radicals
before further reduction to enolate) seems to us more in
keeping with our past results on the cyclisation of d-halo-
a,b-unsaturated esters8 (vide supra) as well as with the
relative ease of formation and stability of aromatic ketyl
radicals. It may be recalled that the ring opening of
(2-phenylcyclopropyl)methyl radical 13 (Fig. 4) is among
the fastest radical reactions calibrated to date with
k2c¼1.6£1011 s21 at 20 8C.22 To the best of our knowledge,
no kinetic data are available at the present time concerning
the ring opening of 2-phenylcyclopropyl methyl ketyl
radical 14. Meanwhile ring opening of (2-phenylcyclo-
propyl) phenyl ketyl radical 15 is at least 105 to 106 faster
than that of cyclopropyl phenyl ketyl radical 16
(k2c¼3£105 to 3£106 s21 vs # 2 s21 at 25 8C).23
The increase in the proportion of lactone in the cyclisation
of the 1-naphthyl substrate 1d must be the result of steric
effects that may be of two kinds: direct in favouring the
formation of the less crowded cis cyclopropanol adduct,
indirect in preventing coplanarity of the aromatic ring and
the ketonic group due to unfavorable steric interactions
between the peri-hydrogen of the naphthyl group with the
carbonyl group on one side and with the gem-dimethyl
group on the other side. Compound 1d would thus react
more like an alkyl ketonic compound than like an aryl
ketonic compound.24
A most intriguing result of our investigation concerns the
total selectivity observed in favor of the formation of
lactone in the case of the cyclopropylketone substrate 1g.
The cyclisation of this substrate in which a potential
cyclopropylcarbinyl-homoallyl type rearrangement is used
as a radical probe was undertaken in the hope that it could be
give helpful information as to the mechanism of cyclisation.
Other examples of utilization of such cyclopropyl radical
probes in SmI2 mediated reactions may be found in the
literature.9e–f,25
In a recent work,23 Tanko and co-workers have shown that
cyclopropyl-methyl ketyl radicals 19 rapidly open into
distonic radical anion 20 with a k2c constant at least equal to
and probably higher than 107 s21 (25 8C) (Scheme 5).
Several examples of ring opening of cyclopropylketones in
the presence of one electron reducing system such as Li in
26
25a,25c,27
NH3 or SmI2
or under photon induced electron
transfer28 may be found in the literature. In our case, we
may expect that monoelectronic reduction of the carbonyl
group is immediately followed by or takes place con-
comitantly with (vide infra) ring opening leading to distonic
radical 22 (Scheme 6). Probably due to a strong preference
for Z configuration of the enolate double-bond (the E-isomer
could cyclise in a 6-exo-trig fashion on the enoate moiety),
22 cannot undergo further chemical transformations other
than reduction to a homoallylic organosamarium species 23
by a second molecule of SmI2. Obviously this second
reduction does not take place under our conditions since we
do not observe the corresponding protonated adduct. In
other words, in the case of cyclopropyl substrate 1g the ketyl
Figure 4.
Scheme 5.

´
S. Bezzenine-Lafollee et al. / Tetrahedron 60 (2004) 6931–6944
6936
Scheme 6.
radical pathway is unproductive. We therefore propose for
cyclisation of the cyclopropyl substrate a mechanism
(mechanism II, Scheme 7) which starts with one electron
reduction of the enoate moiety to radical anion 24.
Subsequent radical 3-exo-trig cyclisation on the carbonyl
group leads to the cyclopropanoxy radical 25. Finally,
reduction of the cyclopropanoxy radical to dianion 26 could
be extremely fast, thus displacing the reaction towards
cyclisation. The reason why this cyclisation takes place with
syn selectivity, leading ultimately to lactone 12 will be
discussed later.
substitution induces stabilization of the open distonic form.
On another hand, no data concerning putative reversibility
of ring opening of alkyl cyclopropyl ketyl radical is
available at the present time. The second one is that one-
electron reduction of the carbonyl group and opening to
distonic radical 22 by C–C bond breaking should in some
way be concerted (Scheme 6, path b). Indeed, in the case of
a stepwise mechanism going through the formation of a
discrete ketyl radical 21 (path a), the existence of a further
equilibrium between 21 and distonic radical 22 is not
supposed to affect the relative proportions of ketyl radical
21 and radical anion 24. Therefore, the reversible formation
of 22 should not induce cyclopropyl ketone 1g to cyclise
according to mechanism II rather than mechanism I. Recent
electrochemical investigations by Tanko and co-workers,23
however, have led these authors to consider the stepwise
mechanism as more probable.
Before discussing mechanism II in some more details, it
should be clearly stated that the proposal of a switch from
mechanism I to mechanism II for cyclisation of cyclo-
propylketone 1g on the ground that the ketyl radical
pathway would now be unproductive relays on two
assumptions that would need confirmation. The first one is
that ring opening to distonic radical 22 is a reversible
process. Reversibility of cyclopropane ring opening in
the case of aryl cyclopropyl ketyl radical is well
established.23,29 This even includes23,29b, aryl cyclopropyl
ketones bearing an extra phenyl group at the 2-position of
cyclopropane (i.e., 15, Fig. 4) despite the fact that such
Concerning now the plausibility of mechanism II itself and
in support of it, a,b-unsaturated esters are known to be
easily reduced by SmI2 to radical anions. Those, in turn, can
be further reduced to saturated esters30 or give homo
coupling products in dependance of the exact reaction
conditions.31 Intramolecular coupling involving 3-exo and
Scheme 7.

´
S. Bezzenine-Lafollee et al. / Tetrahedron 60 (2004) 6931–6944
6937
Scheme 8. 33
6-exo-trig cyclisations of bis-enoates have also been
reported.31a On a different ground, additions of alkyl radical
onto carbonyl groups constitute a well-documented class of
reactions, even if their utilisation for synthetic purpose32 is
much more limited than radical additions on ethylenic
double bonds. This fact can be readily explained by
comparison of kinetic data33 concerning 5- and 6-exo-trig
cyclisations of ethylenic and aldehydic parent systems as
represented in Scheme 8. In both cases, reactions are fast but
in the case of carbonyl compounds, the process is less
favorable because they are reversible with a k2c cyclo-
reversion constant substantially higher than kc. To the best
of our knowledge, no kinetic data concerning either
cyclisation of 3-oxo-propyl radical to cyclopropanoxy
radical or cycloreversion of the latter species are available
in the literature. However it may be inferred that this
equilibrium, if it does take place, is largely displaced
towards the open form. This of course is not in favor of
mechanism II. It should nevertheless be noted that the
formation of cyclopropanoxy radical have been postulated
in other radical transformations for instance in the tin
induced ring expansion of a-halomethyl34 or other35
ketones or in some tin induced 1,2 group migrations within
radical species that are related to coenzyme B12 mediated
rearrangements.36,{ Samarium diiodide37 and more recently
zinc and indium powder mediated38 ring expansion of
a-halomethyl cyclic b-keto esters have also been reported
but in these cases it is difficult to decide whether these
reactions truly involve the formation of cyclopropanoxy
intermediates or follow an ionic pathway.
The main problem associated with mechanism II lays in the
difficulty to account for syn selectivity. We may suspect
that, as it is often the case, chelation of samarium plays a
crucial role. We tentatively propose mechanism II0 being
well aware of its speculative character. Mechanism II0, as
represented in the right upper part of Scheme 9 is a further
elaboration of mechanism II in which problems of
stereochemistry and of coordination to samarium are more
specifically addressed. Cyclisation of radical anion 27 under
chelation control leads to the cyclopropanoxy radical 28a
tautomeric with the a-carbalkoxy substituted radical form
28b. 28a,b may undergo cycloreversion back to 27 or to
ketyl radical 21 when R–cyclopropyl. When R is
cyclopropyl this last process is replaced by double
cycloreversion back to distonic radical 22 which do not
react further. Reduction to the dianionic species therefore
seems the only possible evolution for the cyclopropyl
substrate. After subsequent lactonisation–protonation in
situ and/or during work-up, lactone 12 is finally produced.
Alternatively, lactonisation may take place within 28a,b to
give 29 thus locking at this stage the molecule in the cis
configuration.
The obtention of mixtures of trans cyclopropanols and
lactones with aldehyde and alkyl ketone substrates could
signify that both mechanisms I and II0 take place
concurrently. If so, an additional supposition must however
be made for the sake of consistency. Since ketyl radical 21 is
the starting point of mechanism I, cycloreversion of 28-a,b
to 21 must not be too fast as compared to its further
reduction to 12 or its lactonisation to 29. Otherwise,
{
Moreover, as pointed out by Curran,25a the alkoxy radical is likely to exist
not as a free species, but coordinated to samarium(III) as represented in
Scheme 9, species 28a (vide infra). By further electron transfer from
samarium, 28a may also be seen25a as a carbo-alkoxy dianionic organic
entity coordinated to samarium(IV).

´
S. Bezzenine-Lafollee et al. / Tetrahedron 60 (2004) 6931–6944
6938
Scheme 9.
mechanism I (Schemes 4 and 9, left lower part) would
ultimately operate.k
radical anion. Radical 4-exo-trig cyclisations being rather
slow,39 a competitive way would then be preferred.
Protonation of the radical anion followed by one electon
reduction would give the ester enolate which would
finally add to the carbonyl group in a 5-exo-trig fashion.
Interestingly, if the reaction is conducted in the absence
of methanol, but only in THF with 6 equiv. of HMPA and
3 equiv. of tert-butanol, a 4-exo-trig cyclisation now takes
place but with a syn stereoselectivity opposite to that
observed with aldehydes in THF/MeOH. This switch of
reactivity upon change of alcohol cosolvent has, since
then, been confirmed on other related substrates by the
same authors.9f
The (partial) change of mechanism on going from aromatic
ketones to aldehydes and alkyl ketones would reflect the less
favorable formation of the ketyl radical. Interestingly an
analogous change of mechanism—but this time on going
from aldehydes to alkyl ketones—has been proposed by
Procter and co-workers in the samarium mediated cyclisa-
tion of e-oxo-a,b-unsaturated esters in THF/MeOH.9e
According to the authors, aldehydic compounds (Eq. 1,
Scheme 10) that lead to cyclobutanols with complete
anti-selectivity, cyclise by addition of the ketyl radical
onto the enoate moiety, i.e. according to mechanism I. In
the case of methylketone that lead to carbalkoxysubsti-
tuted cyclopentanols (Eq. 2), the reaction on the contrary
would start by the reduction of the enoate moiety to
4. Conclusion
In the presence of SmI2, d-oxo-g,g-disubstituted-a,b-
unsaturated esters readily undergo 3-exo-trig cyclisation.
High or total diastereoselectivities in favor of the formation
of ‘trans’ cyclopropanol in which the OH group and the
carbalkoxymethyl group are trans to each other are
generally observed with substrates bearing a terminal aryl
substituent. Total opposite stereoselectivity leading
ultimately to the lactone derived from ‘cis’ cyclopropanol
is observed when the terminal substituent is cyclopropyl.
k
On the contrary, as already discussed, ring opening of 28ab should be very
fast for aryl substrates. Therefore, even if initial reduction should occur at
the enoate moiety also for these substrates—for us an unlikely
hypothesis—a fast equilibration between radical 28 and 18 would very
likely take place rapidly. The syn/anti stereoselectivity would therefore
still be related to the relative energies of the transition states leading from
the radical species 28 and 18 to the dianionic species, or alternatively to
the relative energies of the transition states leading from 18 to dianionic
species on one hand and from 28 to 29 (lactonisation) on the other hand.

´
S. Bezzenine-Lafollee et al. / Tetrahedron 60 (2004) 6931–6944
6939
9e
Scheme 10.
Finally when the d-carbon bears an hydrogen atom
(aldehydic substrates) or an alkyl group, a mixture of
‘trans’ cyclopropanol and lactone is obtained.
b-ketoaldehydes had similarly been prepared from pyrro-
lidino-enamines instead of morpholino-enamines.
R0, R0¼Me, Me; R¼CH3: 2,2-Dimethyl-3-oxo-butanal.
Yield 66% (crude product); oil. 1H NMR (250 MHz,
CDCl3) d 9.60 (s, 1H); 2.09 (s, 3H); 1.34 (s, 6H).
‘Trans’ cyclopropanols are thought to arise from initial
formation of ketyl radicals followed by cyclisation. On the
contrary, the favored (but not necessarily exclusive) path-
way to lactone would start by one electron reduction of the
enoate moiety and the stereochemical issue of the reaction
would be the result of samarium chelation in the cyclisation
step.
R0, R0¼Me, Me; R¼i-C3H7: 3-Oxo-2,2,4-trimethyl-
1
pentanal. Yield (over two steps, see above): 38%; oil. H
NMR (200 MHz, CDCl3) d 9.61 (s, 1H); 2.92 (sept, J¼7 Hz,
1H); 1.32 (s, 6H); 1.04 (d, J¼7 Hz, 6H).
R0, R0¼Me, Me; R¼c-C3H7: 3-Cyclopropyl-2,2-
dimethyl-propanal. Purified by distillation, bp 60 8C,
5 Torr; yield 40%; oil. 1H NMR (250 MHz, CDCl3) d
9.58 (s, 1H); 2.07–1.9 (m, 1H); 1.37 (s, 6H); 1.05–0.92
(m, 2H); 0.92–0.80 (m, 2H). 13C NMR (63 MHz, CDCl3) d
208.9; 201.1; 60.3; 17.7; 10.7; 7.4. IR (CHCl3): 1729.5,
5. Experimental
5.1. General information
1H NMR spectra were recorded at 200 or 250 MHz and 13
C
1692 cm21
.
spectra at 63 MHz. Chemical shifts are quoted in ppm
relative to TMS. High resolution mass spectra (HRMS)
were obtained on a Finningan-MAT-95-S spectrometer.
Infra-red spectra were taken on a Perkin-Elmer ‘Spectrum
One’ model and in CHCl3 solution.
R0, R0¼Me, Me; R¼Ph: 2,2-Dimethyl-3-oxo-3-phenyl-
propanal. Purified by column chromatography (AcOEt/
cyclohexane 10:90); yield 50%; oil. H NMR (200 MHz,
CDCl3) d 9.75 (s, 1H); 7.82–7.25 (m, 5H); 1.45 (s, 6H). IR
1
(CHCl3): 1731.5, 1675.5 cm21
.
5.2. Preparation of starting compounds
R0, R0¼Me, Me; R¼2-furyl: 2,2-Dimethyl-3-(2-furyl)-3-
5.2.1. Preparation of b-keto-aldehydes RCOC(R0R0)-
CHO with R0, R05Me, Me or (CH2)5. For R¼Ph, 2-furyl,
2-thienyl, CH3, c-C3H7 and R0, R0¼Me, Me, the b-keto-
aldehydes were prepared by condensation of acyl chlorides
with the morpholino-enamine of isobutyraldehyde35 as
described by Inukai and Yoshizawa.13 The acetylation
procedure was followed for acylation with aliphatic acyl
chlorides and the benzoylation procedure was followed
for acylation with aromatic acyl chlorides. The b-ketoalde-
hyde CH3COC(RR0)CHO (R0R0¼(CH2)5) was prepared in
the same way by acetylation of the morpholino-enamine
of cyclohexane carboxaldehyde according to the first
procedure. i-C3H7COC(CH3)3CHO was prepared by
aldol autocondensation of isobutyraldehyde15 followed
by oxidation of the aldol product with pyridinium chloro-
chromate.40 1-naphthyl-COC(CH3)2CHO was prepared by
acylation of isobutyraldehyde pyrrolidino-enamine41 with
1-naphthoyl chloride according to Kuhlmey et al.14
Probably better yields would have been obtained if other
1
oxo-propanal. Yield 66% (crude product); oil. H NMR
(200 MHz, CDCl3) d 9.4 (s, 1H); 7.54 (d, J¼1.0 Hz, 1H);
3
7.17 (d, 3J¼4.5 Hz, 1H); 6.49 (dd, 3J¼4.5, 2.0 Hz, 1H); 1.4
(s, 6H). IR (CHCl3): 1732, 1664 cm21
.
R0, R0¼Me, Me; R¼2-thienyl: 2,2-Dimethyl-3-oxo-3-(2-
thienyl)-propanal. Yield 74% (crude product); oil. 1H NMR
(200 MHz, CDCl3) d 9.73 (s, 1H); 7.65 (d, 3J¼1.0 Hz, 1H);
7.63 (d, 3J¼5 Hz, 1H)); 7.11–7.07 (m, 1H); 1.49 (s, 6H). IR
(CHCl3): 1728, 1653 cm21
.
R0, R0¼Me, Me; R¼1-naphthyl: 2,2-Dimethyl-3-(1-
naphthyl)-3-oxo-propanal. Yield (crude product): 70%;
oil. H NMR (250 MHz, CDCl3) d 9.79 (s, 1H); 9.10 (d,
1
J¼8.5 Hz, 1H); 8.75 (d, J¼8.5 Hz, 1H), 8.59 (d, J¼8 Hz,
1H), 8.19–8.08, 7.96–7.84, 7.77–7.32 (three m, 3H); 1.50
(s, 6H). HRMS (EI) calcd for C15H14O2 226.0994, found
226.0990. IR (CHCl3): 1687, 1727.5 cm21
.

´
S. Bezzenine-Lafollee et al. / Tetrahedron 60 (2004) 6931–6944
6940
R0, R0¼(CH2)5; R¼CH3: 1-Acetyl-cyclohexanecarboxalde-
hyde: Purified by column chromatography (AcOEt/cyclo-
hexane 20:80); yield 30%; oil. 1H NMR (250 MHz, CDCl3)
d 9.43 (s, 1H); 2.09 (s, 3H); 2.08–1.26 (m, 10H). IR
Benzyl 4,4-dimethyl-5-oxo-hex-2-enoate (1e). Purified by
column chromatography (AcOEt/cyclohexane 20:80); yield
75%; oil. 1H NMR (250 MHz, CDCl3) d 7.32 (m, 5H); 7.05
(d, J¼16 Hz, 1H); 5.90 (d, J¼16 Hz; 1H); 5.14 (s, 2H);
2.07 (s, 3H); 1.23 (s, 6H). 13C NMR (63 MHz, CDCl3) d
208.8; 166.0; 135.6; 128.5; 128.3; 120.5; 66.4; 50.8; 25.9;
23.4. HRMS (EI) calcd for C15H18O3 246.1260, found
246.1256.
3
3
(CHCl3): 1731, 1699 cm21
.
R0, R0¼(CH2)5; R¼2-furyl: 1-(2-Furoyl)-cyclohexane-
carboxaldehyde: Purified by chromatography; yield 45%.
1H NMR (250 MHz, CDCl3) d 9.65 (s, 1H); 7.43 (d,
J¼1.5 Hz, 1H); 7.08 (d, J¼3.5 Hz, 1H); 6.38 (dd, J¼1.5,
3.5 Hz, 1H); 1.48–1.21 (m, 10H). 13C NMR (63 MHz,
CDCl3) d 200.7; 186.5; 151.2; 146.0; 118.9; 112.1; 49.5;
28.5; 25.5; 21.8.
Benzyl 5-oxo-4,4,6-trimethyl-hept-2-enoate (1f). Purified by
column chromatography (AcOEt/cyclohexane 15:85); yield
72%; oil. 1H NMR (250 MHz, CDCl3) d 7.41–7.22 (m, 5H);
7.11 (d, 3J¼16 Hz, 1H); 5.94 (d, 3J¼16 Hz, 1H); 3.01 (sept,
3
3J¼7 Hz, 1H); 1.26 (s, 6H); 1.04 (d, J¼7 Hz, 6H). 13C
5.2.2. Preparation of benzyl d-oxo-a,b-unsaturated
esters 1a–g and 2a,b. 1a–g and 2a,b were prepared by
Wadsworth–Emmons olefination of b-keto-aldehydes with
benzyl dimethoxyphosphonoacetate.42 The experimental
procedure was the same as that used by Nicolaou and
co-workers43 for Wadsworth–Emmons olefination of
3-oxo-2,2-dimethyl-l pentanal with tert-butyl diethoxy-
phosphonoacetate. Purification of crude products was
achieved by column chromatography on silica with appro-
priate mixtures of AcOEt and heptane or cyclohexane as the
eluents.
NMR (63 MHz, CDCl3) d 215.3; 166.4; 152.1; 136.1; 128.9;
128.6; 120.9; 66.8; 51.5; 36.0; 23.6; 20.3. IR (CHCl3):
1710.5, 1646 cm21
.
Benzyl 5-cyclopropyl-4-4,-dimethyl-5-oxo-pent-2-enoate
(1g). Purified by column chromatography (AcOEt/heptane
20:80); yield 65%; oil. 1H NMR (250 MHz, CDCl3) d 7.43–
3
3
7.12 (m, 5H); 7.16 (d, J¼16 Hz, 1H); 5.96 (d, J¼16 Hz,
1H); 5.17 (s, 2H); 2.08–1.95 (m, 1H); 1.30 (s, 6H); 1.05–
0.92 (m, 2H); 0.92–0.80 (m, 2H). 13C NMR (63 MHz,
CDCl3) d 210.9; 166.4; 152.6; 136.1; 128.8; 128.5; 128.4;
120.7; 66.6; 51.0; 23.7; 17.4; 12.0 HRMS (EI) calcd for
C17H20O3 166.0993, found 166.0991; IR (CHCl3) 1716.5,
Benzyl 4,4-dimethyl-5-oxo-5-phenyl-pent-2-enoate (1a).
Purified by column chromatography (AcOEt/cyclohexane
10:90); yield 77%; oil. 1H NMR (200 MHz, CDCl3) d 7.60–
1698, 1647 cm21
.
3
7.15 (m, 11H); 6 (d, J¼15 Hz, 1H); 5.15 (s, 2H); 1.41 (s,
Benzyl 3-[1-(2-furoyl) cyclohexyl]-prop-2-enoate (2a).
Purified by column chromatography (AcOEt/heptane
20:80); yield 27%; oil. 1H NMR (250 MHz, CDCl3) d
7.49 (d, J¼1.5 Hz, 1H); 7.31 (m, 6H); 7.16 (d, J¼3.5 Hz,
1H); 6.45 (dd, J¼1.5, 3.5 Hz, 1H); 5.90 (d, 3J¼16 Hz,
1H); 5.14 (s, 2H); 2.17–1.41 (m, 10H). 13C NMR
(63 MHz, CDCl3) d 189.3; 166.1; 156.4; 146.4; 135.7;
128.5; 128.1; 121.9; 118.8; 111.8; 65.9; 52.9; 33.4; 25.5;
22.4. HRMS (EI) calcd for C21H22O4 338.1518, found
338.1512.
6H). 13C NMR (63 MHz, CDCl3) d 202.8; 166.5; 153.8;
136.6; 136.1; 132.6; 129.5; 128.9; 128.8; 128.7; 128.6;
120.4; 66.9; 50.1; 26.3. IR (CHCl3): 1717, 1681.5,
1642 cm21
.
Benzyl 4,4-dimethyl-5-(2-furyl)-5-oxo-pent-2-enoate (1b).
Purified by column chromatography (AcOEt/cyclohexane
10:90); yield 63%; white solid mp 53 8C. 1H NMR
(250 MHz, CDCl3) d 7.47 (d, J¼2 Hz, 1H); 7.29 (m, 6H);
7.13 (d, J¼4 Hz, 1H); 6.42 (dd, J¼2, 4 Hz, 1H); 5.90 (d,
3J¼16 Hz, 1H); 5.12 (s, 2H); 1.40 (s, 3H). 13C NMR
(63 MHz, CDCl3) d 189.6; 166.2; 152.0; 146.0; 135.2;
128.5; 128.3; 120.1; 119.3; 112.0; 66.1; 48.8; 24.3. IR
Benzyl 3-(1-acetylcyclohexyl)-prop-2-enoate (2b). Purified
by column chromatography (AcOEt/cyclohexane 10:90);
1
yield 74%; white solid; mp 47 8C. H NMR (250 MHz,
3
(CHCl3): 1716.5, 1669, 1645 cm21
.
CDCl3) d 7.35 (m, 5H); 6.87 (d, J¼16 Hz, 1H); 5.89 (d,
3J¼16 Hz, 1H); 5.16 (s, 2H); 2.08 (s, 3H); 1.99–1.40 (m,
10H) 13C NMR (63 MHz, CDCl3) d 208.5; 166.3; 151.5;
136.1; 128.9; 128.7; 122.5; 66.8; 55.7; 32.9; 26.5; 25.9;
22.9. HRMS (EI) calcd for C18H22O3: 274.1574, found
Benzyl 4,4-dimethyl-5-oxo-5-(2-thienyl)-pent-2-enoate (1c).
Purified by column chromatography (AcOEt/cyclohexane
10:90); yield 80%; white solid mp: 69 8C. 1H NMR
(200 MHz, CDCl3) d 7.67 (d, J¼4 Hz, 1H); 7.6–7.05 (m,
274.1574. IR (CHCl3): 1708.5, 1643 cm21
.
8H); 6.08 (d, 3J¼16 Hz, 1H); 5.20 (s, 2H); 1.48 (s, 6H). 13
C
NMR (63 MHz, CDCl3) d 194.5; 167.1; 153.4; 142.4; 136.1;
133.9; 133.7; 128.9; 128.7; 128.6; 128.3; 120.7; 66.4; 49.9;
5.2.3. Preparation of 2-(1,3-dithian-2-yl)-2-methyl-
propanal 5 and 1-(1,3-dithian-2-yl) cyclohexane carbox-
aldehyde 6. 5 and 6 were prepared by alkylation of the
morpholino-enamines of isobutyraldehyde and cyclo-
hexanecarboxaldehyde, respectively, with 2-chloro-1,3-
dithiane. The procedure of Taylor and LaMattina was
followed.16
25.9. IR (CHCl3): 1717; 1694 (shoulder), 1646 cm21
.
Benzyl 4,4-dimethyl-5-(1-naphthyl)-5-oxo-pent-2-enoate
(1d). Purified by column chromatography (AcOEt/cyclo-
hexane 20:80); yield 48%; oil. 1H NMR (250 MHz, CDCl3)
d 7.97–7.28 (m, 13H); 6.03 (d, 3J¼16 Hz, 1H); 5.21 (s, 2H);
1.45 (s, 6H). 13C NMR (63 MHz, CDCl3) d 194.5; 167.1;
153.4; 136.1; 133.9; 133.7; 128.9; 128.7; 128.6; 128.3;
120.7; 66.4; 49.9; 25.9. HRMS (EI) calcd for C24H22O3
358.1569, found 358.1568. IR (CHCl3): 1717, 1693.5,
2-(1,3-dithian-2-yl)-2-methylpropanal (5). Purified by
vacuum distillation bp 56–57 8C, 0.1 Torr; yield 72% on a
40 mmol scale; white solid; mp: 50 8C. 1H NMR (250 MHz,
CDCl3) d 9.41 (s, 1H); 4.25 (s, 1H); 3.01–2.58 (m, 4H);
2.07–1.69 (m, 2H); 1.10 (s, 6H). 13C NMR (63 MHz,
1645 cm21
.

´
S. Bezzenine-Lafollee et al. / Tetrahedron 60 (2004) 6931–6944
6941
CDCl3) d 202.2; 54.9; 49.9; 30.9; 25.6; 19.3. HRMS (EI)
calcd for C8H14OS2: 190.0485, found 190.0487.
Benzyl 4,4-dimethyl-5-oxo-pent-2-enoate 3. Purified by
column chromatography (petroleum ether/diethyl ether
1
80:20); yield 45%; oil. NMR (250 MHz, CDCl3) d 9.33
3
1-(1,3-Dithian-2-yl)cyclohexanecarboxaldehyde (6). Puri-
fied by Ku¨gelrohr distillation (180 8C, 0.03 Torr); yield 33%
(s, 1H); 7.26 (m, 5H); 6.92 (d, J¼14.5 Hz, 1H); 5.84 (d,
3J¼14.5 Hz, 1H); 5.09 (s, 2H); 1.18 (s, 6H). 13C NMR
(63 MHz, CDCl3) d 200.7; 165.7; 149.2; 135.6; 128.5,
128.1; 121.5; 66.4; 49.1; 21.0. HRMS (EI) calcd for
C14H16O3: 232.1099, found 232.1100.
1
on a 20 mmole scale; oil. H NMR (250 MHz, CDCl3) d
9.52 (s, 1H); 4.16 (s, 1H); 3.20–2.50 (m, 4H); 2.03–1.17
(m, 12H). 13C NMR (63 MHz, CDCl3) d 203.9; 54.8; 52.7;
31.0; 28.6; 25.6; 24.5; 21.9. HRMS (EI) calcd for
C11H18OS2: 230.0795, found 230.0799.
Benzyl 3-(1-formyl-cyclohexyl)-prop-2-enoate 4. Purified
by column chromatography (petroleum ether/diethyl ether
1
5.2.4. Preparation of benzyl 4-(1,3-dithian-2-yl)-4-
methylpent-2-enoate 7 and benzyl 4-[1-(1,3-dithian-2-
yl)cyclohexyl]-but-2-enoate 8. Benzyl 4-(1,3-dithian-2-
yl)-4-methylpent-2-enoate 7: in a Schlenk tube and under
argon atmosphere, 0.22 g (5.5 mmol) of sodium hydride
60% in oil was washed twice with dry pentane and put in
suspension in 45 mL of anhydrous THF at 0 8C. To this
suspension 1.42 g (5.5 mmol) of benzyl dimethylphos-
phonoacetate was added dropwise. After cessation of
dihydrogen gas evolution a solution of 0.95 g (0.5 mmol)
of 2-(1,3-dithian-2-yl)-2-methyl-propanal 5 was slowly
added while the temperature was maintained near 0 8C.
The reaction mixture was then refluxed for 5 h. The reaction
mixture was cooled and water was first cautiously and then
more rapidly added. The organic products were extracted
with diethyl ether. The organic phase was dried over
MgSO4, filtrated and concentrated under vacuum to give
1.45 g (90% yield, oil) of crude benzyl 4-(1,3-dithian-2-yl)-
4-methylpent-2-enoate which was found pure by NMR.
This crude product was used as such in the following
step.
80:20); yield 74%. HNMR (250 MHz, CDCl3) d 9.32 (s,
1H); 7.33 (m, 5H); 6.77 (d, 3J¼14.5 Hz, 1H); 5.88 d,
3J¼14.5 Hz, 1H); 5.15 (s, 2H); 1.90–1.32 (m, 10H). 13C
NMR (63 MHz, CDCl3) d 200.8; 165.5; 148.4; 128.5; 128.3;
122.7; 66.4; 53.2; 30.5; 25.2; 22.0. HRMS (EI) calcd for
C17H20O3: 272.1412, found 272.1419.
5.3. SmI2 mediated cyclisations
5.3.1. Preparation of SmI2 solutions in THF. THF was
distilled over benzophenone sodium and under an argon
atmosphere. All experiments involving SmI2 were carried
out under an argon atmosphere using standard Schlenk
techniques. Diiodoethane, used for preparation of SmI2, was
purified as follows: commercial 1,2 diodoethane was
dissolved in diethyl ether and the ethereal solution was
washed twice with aqueous sodium thiosulfate to remove
any iodine. The organic phase was then dried on magnesium
sulfate and then concentrated under vacuum to a small
volume. The precipitated white solid was collected by
filtration and dried on a vacuum line. All these operations
were carried out in the dark.
Benzyl 4-(1,3-dithian-2-yl)-4-methylpent-2-enoate (7). 1H
NMR (250 MHz, CDCl3) d 7.35 (m, 5H); 7.04 (d,
0.1 M solutions of SmI2 in THF were prepared in the
following way: to 1.80 g (12 mmol) of samarium powder
(from Labelcomat Company) was slowly added through a
cannula and at room temperature a solution of 2.82 g of
freshly purified 1,2-diiodoethane in 100 mL of THF. A blue-
green coloration developed immediately and the reaction
was somewhat exothermic. Once the addition was com-
pleted (after ca 20 min), the suspension was stirred for 12 h,
upon which the reaction was considered as complete. The
0.1 M solutions of SmI2 were of a deep blue color.
They were kept as such in the presence of samarium
powder in excess and under argon atmosphere for no more
than 4 days.
3
3J¼14.5 Hz, 1H); 5.84 (d, J¼14.5 Hz, 1H); 5.14 (s, 2H);
4.05 (s, 1H); 2.86–2.75 (m, 4H); 2.07–1.76 (m, 2H); 1.23
(s, 6H). 13C NMR (63 MHz, CDCl3) d 166.2; 155.2; 135.8;
128.4; 128.0; 118.9; 66.1; 59.0; 41.5; 31.0; 25.6; 24.3; 19.3.
HRMS (EI) calcd for C17H22O2S2: 322.1061, found
322.1063.
Benzyl 4-[1-(1,3-dithian-2-yl)-cyclohexyl]-but-2-enoate 8
was similarly prepared from 1-(1,3-dithian-2-yl)cyclo-
hexanecarboxaldehyde 6. In this case, the crude product
was purified by column chromatography on silica (cyclo-
hexane/AcOEt 95:5).
Benzyl 4-[1-(1,3-dithian-2-yl)-cyclohexyl]-but-2-enoate
1
(8). Yield 95%; oil. H NMR (250 MHz, CDCl3) d 7.33
5.3.2. Cyclisation reactions. General procedure: to a
solution of 1 mmol of substrate to be cyclised and 4 mmol
of tert-butanol in 5–6 mL of THF at 0 8C were added
dropwise 22 mL (2.2 equiv.) of a 0.1 M solution of SmI2 in
THF. The reaction mixture was then stirred at room
temperature with monitoring by TLC or IR spectroscopy
on aliquots. Reaction were usually complete within 4–12 h.
After quenching with dilute aqueous HCl, the products were
extracted in diethyl ether. After drying (MgSO4) and
evaporation of ether, the residue was column chromato-
graphied on silica with appropriate mixtures of ethyl acetate
and heptane or cyclohexane as the eluents.
3
3
(m, 5H); 6.96 (d, J¼16 Hz, 1H); 5.90 (d, J¼16 Hz, 1H);
5.17 (s, 2H); 4.15 (s, 1H); 2.83 (dd, J¼8, 3.5 Hz, 4H); 2.03
(m, 4H); 1.77 (m, 2H); 1.50–1.28 (m, 6H). 13C NMR
(63 MHz, CDCl3) d 166.2; 153.4; 135.9; 128.5; 128.2;
128.1; 121.7; 66.2; 58.3; 44.3; 33.2; 31.3; 28.95; 25.9; 25.7;
21.9. HRMS (EI) calcd for C20H26O2S2: 362.1374, found
362.1367.
5.2.5. Preparation of benzyl 4,4-dimethyl-5-oxo-pent-2-
enoate 3 and benzyl 3-(1-formyl-cyclohexyl)-prop-2-
enoate 4. Heating 7 or 8 in acetone/water in the presence
of MeI and sym-collidine according to Redlich et al.17 gave
the corresponding aldehydes 3 or 4.
5.3.3. Physical and spectroscopic data for cyclisation
products. All products were obtained as colorless oils.

´
S. Bezzenine-Lafollee et al. / Tetrahedron 60 (2004) 6931–6944
6942
5.3.3.1. Cyclopropanols 11a: (R0, R05CH3, CH3).
R¼CH3: 1H NMR (250 MHz, CDCl3) d 7.41–7.24 (m, 5H);
5.11 (s, 2H); 2.41–2.14 (two close dd, ABX system
JAB¼16 Hz, JAX¼7.5 Hz, JBX¼7.5 Hz, (1þ1)H); 1.65–
1
R¼H: H NMR (250 MHz, CDCl3) d 7.32 (m, 5H); 5.09
(s, 2H); 2.89 (d, 3J¼3 Hz, 1H); 2.29 (d, J¼7.6 Hz, 2H); 1.13
(s, 3H); 0.93 (s, 3H); 0.91 (m, 1H). 13C NMR (63 MHz,
CDCl3) d 173.6; 128.6; 128.2; 66.3; 61.8; 32.6; 27.3; 19.6;
19.3. HRMS (EI) calcd for C14H18O3 234.1256, found
3
1.15 (m, 13H); 0.91 (t, J¼7.5 Hz, 1H). HRMS (EI) calcd
for C18H24O3 288.1725, found 288.1723. IR (CHCl3):
.
1731.5 cm21
234.1256. IR (CHCl3): 1732.5 cm21
.
1
R¼2-furyl: H NMR (250 MHz, CDCl3) d 7.36 (m, 6H);
7.30 (m, 1H); 6.23 (m, 1H); 5.12 (d, J¼6 Hz, 2H); 2.60–
2.47 and 2.40–2.32 (two dd, ABX system, JAB¼17 Hz,
JAX¼7.5 Hz, JBX¼7.5 Hz (1þ1)H); 1.80–1.12 (m, 11H).
13C NMR (63 MHz, CDCl3) d 173.1; 153.4; 142.0; 135.8;
128.5, 128.3; 127.6; 110.0; 108.7; 66.4; 59.7; 32.8; 31.5;
31.1; 31.0; 28.1; 26.3; 25.8; 25.1. HRMS (EI) calcd for
C21H24O4 288.1725, found 288.1723. IR (CHCl3):
R¼Me: 1H NMR (250 MHz, CDCl3) d 7.34 (m, 5H); 5.15 (s,
2H); 2.27 (d, 3J¼8 Hz, 2H); 1.10 (s, 3H); 1.07 (s, 3H); 1.03
(t, 3J¼8 Hz, 1H); 0.88 (s, 3H). 13C NMR (63 MHz, CDCl3)
d 173.3; 135.9; 128.5; 128.1; 66.3; 59.7; 30.6; 29.2; 26.9;
21.3; 17.0; 16.5. HRMS (EI) calcd for C15H20O3 248.1412,
found 248.1423.
R¼i-Pr: 1H NMR (250 MHz, CDCl3) d 7.36–7.24 (m, 5H);
5.10 (s, 2H); 2.55–2.40 and 2.28–2.10 (two dd, ABX
system JAB¼16.5 Hz, JAX¼6.5 Hz; JBX¼8.5 Hz, (1þ1)H);
1.62–1.51 (m, 1H); 1.19 (s, 3H); 1.12–1.01 (m, 1H); 1.00–
0.98 (m, 9H). HRMS (EI) calcd for C17H24O3: 276.1725,
1732 cm21
.
5.3.3.3. Lactones 12a: R0, R05CH3, CH3. R¼H: 1H
NMR (250 MHz, CDCl3) d 3.95 (d, 3J¼6.3 Hz, 1H); 2.78–
2.67 (dd) and 2.33 (d) (ABX system JAB¼21 Hz, JAX¼8 Hz,
JBX¼0 Hz, (1þ1)H); 1.02 (m, 7H). HRMS calcd for
C7H10O2 126.0681, found 126.0677. IR (CHCl3):
found 276.1723. IR (CHCl3): 1732 cm21
.
1775.5 cm21
.
1
R¼Ph: H NMR (250 MHz, CDCl3) d 7.4–7.15 (m, 10H);
5.12 (two d, AB system, JAB¼12 Hz, 2H); 2.58–2.41 and
2.18–2.05 (two dd, ABX system, JAB¼18 Hz, JAX¼7 Hz,
JBX¼6.5 Hz, (1þ1)H); 1.4 (s, 3H); 1.4–1.28 (m, 1H); 0.9 (s,
3H). HRMS (EI) calcd for C20H22O3 310.1569, found
1
R¼Me: H NMR (250 MHz, CDCl3) d 2.80 (dd) and 2.11
(d) (ABX system JAB¼21.0 Hz, JAX¼7 Hz, JBX¼0 Hz,
(1þ1)H); 1.23–0.99 (m, 10H). HRMS (EI) calcd for
C8H12O2 140.0837, found 140.0840. IR (CHCl3):
310.1572. IR (CHCl3): 1732 cm21
.
1772 cm21
.
1
R¼2-furyl: H NMR (250 MHz, CDCl3) d 7.36–7.31 (m,
6H); 6.31–6.22 (m, 2H); 5.15 (two d, AB system,
JAB¼12 Hz, 2H); 2.61–2.51 and 2.37–2.26 (two dd, ABX
system, JAB¼17.5 Hz, JAX¼7.5 Hz, JBX¼8 Hz, (1þ1)H);
1.44–134 (m, 1H); 1.37 (s, 3H); 0.95 (s, 3H). 13C NMR
(63 MHz, CDCl3) d 173.1; 153.5; 142.2; 135.9; 128.6;
128.3; 127.0; 110.1; 109.1; 66.5; 59.2; 31.6; 31.2; 26.2;
R¼i-Pr: 1H NMR (250 MHz, CDCl3) d 2.80–2.69 (dd) and
2.40 (d) (ABX system JAB¼19.0 Hz, JAX¼7.5 Hz,
JBX¼0 Hz, (1þ1)H); 1.85–1.71 (sept, 3J¼7 Hz, 1H);
1.16–0.99 (m, 13H). 13C NMR (63 MHz, CDCl3) d 178.5;
78.2; 31.4; 29.1; 25.5; 24.2; 22.4; 19.6; 18.6; 14.0; HRMS
(EI) calcd for C10H16O2 168.1150, found 168.1148. IR
(CHCl3): 1771.5 cm21
.
21.2; 17.9. IR (CHCl3): 1732.5 cm21
.
R¼c-Pr: 1H NMR (250 MHz, CDCl3) d 2.76–2.58 (dd) and
2.38 (d) (ABX system JAB¼19 Hz, JAX¼7.0 Hz, JBX¼0 Hz,
(1þ1)H); 1.44–1.43 (m, 1H); 1.17 (s, 3H); 1.03 (s, 3H), ca
1.02 (d, partially masked, 1H); 0.77–0.55 (m, 2H); 0.48–
0.23 (m, 2H). 13C NMR (63 MHz, CDCl3) d 177.9; 76.1;
31.2; 26.1; 22.5; 22.1; 13.9; 9.4; 6.0; 4.0. HRMS (EI) calcd
for C10H14O2 166.0993, found 166.0991. IR (CHCl3):
R¼2-thienyl: 1H NMR (400 MHz, CDCl3) d 7.29–7.20 (m,
6H); 6.87–6.72 (m, 2H); 5.08 (two d, AB system,
JAB¼12 Hz, 2H), 2.63–2.51 and 2.35–2.23 (two dd, ABX
system, JAB¼17.5 Hz, JAX¼7 Hz, JBX¼8 Hz, (1þ1)H);
1.34 (m, 1H); 1.32 (s, 3H); 0.92 (s, 3H). 13C NMR
(63 MHz, CDCl3) d 173.4; 142.7; 136.3; 129.0; 128.7;
128.0; 127.1; 126.6; 66.9; 60.9; 32.6; 32.1; 26.9; 21.8; 19.4.
1772 cm21
.
IR (CHCl3): 1732.5 cm21
.
1
R¼2-furyl: H NMR (250 MHz, CDCl3) d 7.43–7.35 (m,
1H); 6.45–6.34 (m, 2H); 3.04–2.93 (dd) and 2.54 (d) (ABX
system JAB¼19 Hz, JAX¼7.5 Hz, JBX¼0 Hz, (1þ1)H); 1.78
R¼1-naphthyl: 1H NMR (250 MHz, CDCl3) d 8.1–7.7,
7.65–7.1 (two m, 12H); 5.25–5.12 (two broad d (app. q),
AB system, 2H); 3.0–2.75 and 2.61–2.24 (two broad dd,
ABX system, JAB<16 Hz, JBX<6 Hz; JAX<8 Hz, (1þ1)H);
1.59–1.39 (m, 4H); 0.65 (broad s, 3H). HRMS (EI) calcd for
C24H24O3 360.1725, found 360.1710. IR (CHCl3):
3
(d, J¼7.5 Hz, 1H); 1.18 (s, 3H); 1.02 (s, 3H). 13C NMR
(63 MHz, CDCl3) d 177.1; 143.2; 111.2; 110.5; 67.7; 30.6;
.
27.5; 25.6; 22.8; 13.1. IR (CHCl3): 1783 cm21
1732 cm21
.
R¼2-thienyl: 1H NMR (250 MHz, CDCl3) d 7.42–7.31 (m,
1H); 7.13–6.98 (m, 2H); 3.07–2.96 (dd) and 2.59 (d) (ABX
system, JAB¼19 Hz, JAX¼7.5 Hz, JBX¼0 Hz, (1þ1)H);
5.3.3.2. Cyclopropanols 11b: R0, R05(CH2)5. R¼H:
1H NMR (250 MHz, CDCl3) d 7.30 (m, 5H), 5.11 (s, 2H);
3
3
2.98 (d, 1H, J¼3.0 Hz); 2.44–2.20 (two close dd, ABX
1.77 (d, J¼7.5 Hz, 1H); 1.17 (s, 3H); 1.02 (s, 3H). IR
system JAB¼17 Hz, JAX¼7.5 Hz, JBX¼8.0 Hz, (1þ1)H);
(CHCl3): 1773 cm21
.
1.65–1.23 (m, 10H); 0.88 (dt, 3Jd¼3, 3Jt¼7.5 Hz, 1H). 13
C
NMR (63 MHz, CDCl3) d 166.0; 135.7; 128.5; 128.3; 128.0;
66.4; 50.8; 39.7; 25.9; 25.2; 23.4. HRMS (EI) calcd for
C17H22O3 274.1574, found 274.1574. IR (CHCl3):
R¼1-naphthyl: 1H NMR (250 MHz, CDCl3) d 8.03 (d,
J¼8 Hz, 1H); 7.97–7.79, 7.64–7.42 (two m, 6H); 3.25–3.0
and 2.71 (dd and d, ABX system, JAB¼19 Hz, JAX¼7 Hz,
3
1732 cm21
.
JBX¼0 Hz, (1þ1)H); 1.92 (d, J¼7 Hz, 1H); 1.41 (s, 3H);

´
S. Bezzenine-Lafollee et al. / Tetrahedron 60 (2004) 6931–6944
6943
Table 1.
Entry
Starting compound
R0R0
Cyclopropanol 11 (%)a
Lactone 12 (%)a
anti/syn selectivity
(S)
R
No.
(A)b
(B)c
95/5
1
Ph
MeMe
MeMe
1a
1b
85
—
100/0
2a
2b
3
Run 1
Run 2
73
82
77
13
14
8
85/15
85/15
91/9
87/13
84/16
88/12
(CH2)5
MeMe
2a
1c
4
60
10
86/14
85/15
5a
5b
MeMe
1d
Run 1
Run 2
30
20
49
47
38/62
30/70
28/72
26/74
6a
6b
7
CH3
MeMe
1e
Run 1
Run 2
60
55
52
22
23
73/27
70/30
63/37
71/29
66/33
64/66
(CH2)5
MeMe
2b
1f
30d
8a
8b
i-C3H7
c-C3H7
Run 1
Run 2
58
58
35e
30e
62/38
66/34
61/39
70/30
9a
9b
9c
10
11a
11b
MeMe
1g
Run 1
Run 2
Run 3
—
—
—
37
30
26
92
95
90
30
21
32
0/100
0/100
0/100
55/45
59/61
44/66
0/100
0/100
0/100
53/47
54/46
50/50
H
MeMe
(CH2)5
3
4
Run 1
Run 2
a
Yields of pure compounds after separation by chromatography.
Calculated from S¼yield of 11/yield of 12.
b
c
Deduced from NMR on crude reaction mixtures by integration of benzylic H peaks of 11 and of benzylic alcohol thus assuming that all benzyl alcohol is
produced by lactonisation.
d
e
10% of alcohol of direct reduction 9 was also obtained.
8–10% of saturated ester 10 was also obtained.
0.79 (s, 3H). 13C NMR (63 MHz, CDCl3) d 177.6; 134,7;
134,1; 133.7; 130.2; 128.7; 128.6; 127.2; 126.6; 125.3; 74.6;
31.4; 31.3; 15.6; 13.7. HRMS (EI) calcd for C17H16O2
References and notes
1. Zhang, W.; Dowd, P. Tetrahedron Lett. 1994, 35, 4531–4535,
and references cited therein.
252.1147, found 252.1150. IR (CHCl3): 1776 cm21
.
0
0
3
1
5.3.3.4. Lactones 12b: R , R 5(CH2)5. R¼H: H NMR
2. (a) Warner, C. R.; Strunk, R. J.; Kuivila, H. G. J. Org. Chem.
1966, 31, 3381–3384. (b) Srikrishna, A.; Sharma, G. V. R. J.;
Hamamalini, P. J. Chem. Soc. Chem. Commun. 1990,
1681–1683. (c) Srikrishna, A.; Sharma, G. V. R. J. Chem.
Soc., Perkin Trans. 1 1997, 177–181. (d) Denis, R. C.;
Rancourt, J.; Ghiro, E.; Boutonnet, F.; Gravel, D. Tetrahedron
Lett. 1993, 34, 2091–2094. (e) Journet, M.; Malacria, M.
J. Org. Chem. 1994, 59, 718–719. (f) Devin, P.; Fensterbank,
L.; Malacria, M. J. Org. Chem. 1998, 63, 6764–6765.
3. silylvinyl group: (a) Weng, W.-W.; Luh, T.-Y. J. Org. Chem.
1993, 58, 5574–5575. carbonyl group: (b) Zhang, W.; Dowd,
P. Tetrahedron Lett. 1992, 33, 7307–7310. aromatic sub-
stituents: (c) Halgren, T. A.; Howden, M. E. H.; Roberts, J. D.
J. J. Am. Chem. Soc. 1967, 89, 3051–3052. (d) Sikrishna, A.;
Danieldoss, S. J. J. Org. Chem. 1997, 62, 7863–7865.
4. (a) Cekovic, Z.; Saicic, R. Tetrahedron Lett. 1990, 31,
6085–6088. (b) Gravel, D.; Denis, R. C. Tetrahedron Lett.
1994, 26, 4531–4534. (c) Ferjancic, Z.; Saicic, R. N.;
Cekovic, Z. Tetrahedron Lett. 1997, 38, 4165–4168.
(200 MHz, CDCl3) d 3.95 (d, J¼6.2 Hz, 1H); 2.80–2.60
(dd) and 2.32 (d) (ABX system, JAB¼19.0 Hz, JAX¼7.5 Hz;
JBX¼0 Hz, (1þ1) H); 1.54–1.16 (m, 11H). HRMS (EI)
calcd for C10H14O2 166.0994, found 166.0985. IR (CHCl3):
1775.5 cm21
.
R¼CH3: 1H NMR (250 MHz, CDCl3) d 2.87–2.76 (dd) and
2.40 (d) (ABX system, JAB¼19.0 Hz, JAX¼7.0 Hz; JBX¼0),
3
(1þ1) H); 1.79–1.15 (m, 13H); 1.06 (d, J¼7 Hz, 1H).
HRMS (EI) calcd for C11H16O2: 180.1150, found 180. 1147.
.
IR (CHCl3): 1770 cm21
1
R¼2-furyl: H NMR (250 MHz, CDCl3) d 7.39 (m, 1H);
7.31 (m, 1H); 6.40 (m, 1H); 2.91 (dd) and 2.50 (d, ABX
system, JAB¼19.0 Hz, JAX¼7.5 Hz, JBX¼0 Hz, (1þ1) H);
1.72 (d, J¼7.5 Hz, 1H); 1.65–1.12 (m, 10H). 13C NMR
(63 MHz, CDCl3) d 178.0; 150.6; 143.1; 110.9; 110.4; 56.8;
32.6; 30.2; 26.0; 25.1; 24.7; 24.5; 23.6 (Table 1).

´
S. Bezzenine-Lafollee et al. / Tetrahedron 60 (2004) 6931–6944
6944
´
5. David, H.; Bonin, M.; Doisneau, G.; Guillerez, M. G.; Guibe,
F. Tetrahedron Lett. 1999, 40, 8557–8561.
21. Seman, J. I. Chem. Rev. 1983, 83–134.
22. (a) Choi, S.-Y.; Toy, P. H.; Newcomb, M. J. Org. Chem. 1998,
63, 8609–8613. (b) Newcomb, M.; Johnson, C. C.; Manek,
M. B.; Varick, T. R. J. Am. Chem. Soc. 1992, 114,
10915–10921.
6. Reviews: (a) Molander, G. A.; Harris, C. R. Chem. Rev. 1996,
96, 307–338. (b) Molander, G. A.; Harris, C. R. Tetrahedron
1998, 54, 3321–3354. for other recent reviews on SmI2
mediated reactions see: (c) Krief, A.; Laval, A.-M. Chem. Rev.
1999, 99, 745–777. (d) Steel, P. G. J. Chem. Soc, Perkin
Trans. 1 2001, 2727–2751. (e) Kagan, H. B. Tetrahedron
2003, 59, 10351–10372.
23. Stevenson, J. P.; Jackson, W. F.; Tanko, J. M. J. Am. Chem.
Soc. 2002, 124, 4271–4281.
24. For a quite similar indirect steric effect of the naphthyl group
(as compared to phenyl) on the course of radical reactions see
Ref. 3d.
7. (a) Horner, J. H.; Tanaka, N.; Newcomb, M. J. Am. Chem. Soc.
1988, 120, 10379–10390. (b) Beckwith, A. L. J.; Bowry, V. W.
J. Am. Chem. Soc. 1994, 116, 2710–2716.
´
8. Villar, H.; Guibe, F.; Aroulanda, C.; Lesot, P. Tetrahedron:
Asymmetry 2002, 13, 1465–1475.
25. (a) Curran, D. P.; Gu, X.; Zhang, W.; Dowd, P. Tetrahedron
1997, 53, 9023–9042. (b) Collin, J.; Namy, J.-L.; Kagan, H. B.
New J. Chem. 1986, 10, 229–232. (c) Molander, G. A.;
McKie, J. A. J. Org. Chem. 1991, 56, 4112–4120. (d) Tamiya,
H.; Goto, K.; Matsuda, F. Org. Lett. 2004, 6, 545–547.
(e) McKerlie, F.; Procter, D. J.; Wynne, G. Chem. Commun.
2002, 584–585.
9. Selected references: 4-exo-trig cyclisations: (a) Weinges, K.;
Schmidbauer, S. B.; Schick, H. Chem. Ber. 1994, 127,
1305–1309. (b) Johnston, D.; McCusker, C. M.; Procter,
D. J. Tetrahedron Lett. 1999, 40, 4913–4916. (c) Johnston, D.;
McCusker, C. M.; Muir, K.; Procter, D. J. J. Chem. Soc. Perkin
Trans. 1 2000, 681–695. (d) Johnston, D.; Francon, N.;
Edmons, D. J.; Procter, D. J. Org. Lett. 2001, 3, 2001–2004.
(e) Hutton, T. K.; Muir, K.; Procter, D. J. Org. Lett. 2002, 4,
2345–2347. (f) Hutton, T. K.; Muir, K. W.; Procter, D. J. Org.
Lett. 2003, 5, 4811–4814. 5-exo-trig cyclisations: (g) Enholm,
E. J.; Trivellas, A. Tetrahedron Lett. 1989, 30, 1063–1066.
(h) Enholm, E. J.; Trivellas, A. J. Am. Chem. Soc. 1989, 111,
6463–6465. (i) Molander, G. A.; Kenny, C. J. Am. Chem. Soc.
1989, 111, 8236–8246. (j) Kan, T.; Nara, S.; Ozawa, T.;
Shirahama, H.; Matsuda, F. Angew. Chem., Int. Ed. 2000, 39,
355–357, 6-exo-trig cyclisations: (k). Kito, M.; Sakai, T.;
Yamada, K.; Matsuda, F.; Shirahama, H. Synlett 1993,
158–163. (l) Kito, M.; Sakai, T.; Haruta, N.; Shirahama, H.;
Matsuda, F. Synlett 1996, 1057–1060.
26. Dauben, W. G.; Wolf, R. E. J. Org. Chem. 1970, 35, 374–379.
27. Batey, R. A.; Harling, J. D.; Motherwell, W. B. Tetrahedron
1996, 52, 11421–11444.
28. Pandey, B.; Rao, A. T.; Dalvi, P. V.; Kumar, P. Tetrahedron
1994, 50, 3843–3848.
29. (a) Tanko, J. M.; Drumright, R. E. J. Am. Chem. Soc. 1992,
114, 1844–1854. (b) Tanner, D. P.; Chen, J. J.; Luelo, C.;
Peters, P. M. J. Am. Chem. Soc. 1992, 114, 713–717.
30. Inanaga, J.; Sakai, S.; Handa, Y.; Yamaguchi, M.; Yokohama,
Y. Chem. Lett. 1991, 2117–2118.
31. (a) Inanaga, J.; Handa, Y.; Tabuchi, T.; Otsuva, K.;
Yamaguchi, M.; Hanamoto, T. Tetrahedron Lett. 1991, 32,
6557–6558. (b) Cabrera, A.; Alper, H. Tetrahedron Lett.
1992, 33, 5007–5008.
32. See for instance: Tsai, Y.-M.; Cherng, C.-D. Tetrahedron Lett.
1991, 32, 3515–3518, and references cited therein.
33. Beckwith, A. L. J.; Hay, B. P. J. Am. Chem. Soc. 1989, 111,
2674–2681.
10. See Ref. 6a p 315 and references cited therein.
11. See for instance: Hori, N. H.; Matsukura, H.; Matsuo, G.;
Nakata, T. Tetrahedron 2002, 58, 1853–1864, and Refs. 9g
and 9k.
34. Dowd, P.; Zhang, W. Chem. Rev. 1993, 2091–2115.
35. Rodriguez, J. R.; Mascarenas, J. L. Org. Lett. 2001, 3,
1181–1183.
´
12. Villar, H.; Guibe, F. Tetrahedron Lett. 2002, 43, 9517–9520.
13. Inukai, T.; Yoshigawa, R. J. Org. Chem. 1967, 32, 404–407.
14. Kuhlmey, S.-R.; Horst, A.; Rieth, K.; Opitz, G. Liebigs Ann.
Chem. 1979, 617–627.
36. Halpern, J.; Wollowitz, S. J. Am. Chem. Soc. 1988, 110,
3112–3120.
37. (a) Hasegawa, E.; Kitazume, T.; Suzuki, K.; Tosaka, E.
Tetrahedron Lett. 1998, 39, 4059–4062. (b) Chung, S. H.;
Cho, M. S.; Choi, J. Y.; Kwon, D. W.; Kim, Y. H. Synlett 2001,
1266–1268.
15. Saunders, R. H.; Murray, M. J.; Cleveland, F. F. J. Am. Chem.
Soc. 1943, 65, 1714–1717.
16. Taylor, E. C.; LaMattina, J. L. Tetrahedron Lett. 1977,
2077–2080.
17. Redlich, H.; Bruns, W.; Francke, W.; Schurig, V.; Payne, T. L.;
´
Vite, J.-P. Tetrahedron 1987, 43, 2029–2034.
18. Prasad, E.; Flowers, R. A., II. J. Am. Chem. Soc. 2002, 124,
6357–6361.
38. Sugi, M.; Sakuma, D.; Togo, H. J. Org. Chem. 2003, 68,
7629–7633.
39. Park, S.-U.; Varick, T. R.; Newcomb, M. Tetrahedron Lett.
1990, 31, 2975–2978.
40. Corey, E. J.; Suggs, W. J. Tetrahedron Lett. 1975, 2647–2650.
41. Benzing, E. Angew. Chem. 1959, 71, 521.
42. Bradley, J. C.; Bu¨chi, G. J. Org. Chem. 1976, 41, 699–701.
43. Nicolaou, K. C.; He, Y.; Vourloumis, D.; Vallberg, H.;
Roschangar, F.; Sarabia, F.; Ninkovic, S.; Yang, Z.; Trujillo,
J. I. J. Am. Chem. Soc. 1997, 119, 7962–7973.
19. Curran, D. P.; Fevig, T. L.; Jasperse, C. P.; Totleben, M. J.
Synlett 1992, 943–961.
20. (a) Fukuzawa, S.; Nakanishi, A.; Fujinami, T.; Sakai, S.
J. Chem. Soc., Chem. Commun. 1986, 624–625, see also: (b)
Otsubo, K.; Inanaga, J.; Yamaguchi, M. Tetrahedron Lett.
1986, 27, 5763–5764.