Reactivity of cyclopropanic δ-oxo-α,β-unsaturated esters towards SmI2: 3-exo-trig cyclisation versus cyclopropane ring opening

Article abstract of DOI:10.1016/j.tet.2007.02.080

trans-(2′,2′-Diphenyl-bicyclopropyl-2-yl)-4,4-dimethyl-5-oxo-pent-2-enoic acid methyl ester 9, undergoes 3-exo-trig cyclisation in the presence of SmI₂ without competitive ring opening of Newcomb's bicyclopropylic probe next to the carbonyl gro

Full text of DOI:10.1016/j.tet.2007.02.080

Tetrahedron 63 (2007) 3728–3736
Reactivity of cyclopropanic d-oxo-a,b-unsaturated esters towards
SmI2: 3-exo-trig cyclisation versus cyclopropane ring opening
*
*
ꢀ
ꢀ
Chama Cammoun, Riadh Zriba, Sophie Bezzenine-Lafollee and Franc¸ois Guibe
´
Institut de Chimie Moleculaire et des Materiaux d’Orsay, Laboratoire de Catalyse Moleculaire, UMR-8182, Bat. 420,
´
´
^
´
Universite Paris-Sud, 91405 Orsay, France
Received 21 December 2006; revised 14 February 2007; accepted 20 February 2007
Available online 23 February 2007
Abstract—trans-(20,20-Diphenyl-bicyclopropyl-2-yl)-4,4-dimethyl-5-oxo-pent-2-enoic acid methyl ester 9, undergoes 3-exo-trig cyclisation
in the presence of SmI2 without competitive ring opening of Newcomb’s bicyclopropylic probe next to the carbonyl group. From this result, it
may be concluded that the absence of ring opening observed earlier in the case of 5-cyclopropyl-4,4-dimethyl-5-oxo-pent-2-enoate 7 is not
due to the potentially reversible character of this process. Meanwhile, as deduced from kinetic considerations based on data of the literature,
the absence of ring opening does not necessarily mean that formation of ketyl radicals is not involved in the 3-exo-trig cyclisations of d-oxo-
a,b-unsaturated esters.
Compounds 7 and 9 cyclise with total syn selectivity, leading ultimately to lactones. This syn selectivity contrasts with that of other alkylic
d-oxo-a,b-unsaturated esters.
Ó 2007 Elsevier Ltd. All rights reserved.
1. Introduction
diastereoselectivity (sometimes excellent) was shown to be
highly dependent on the structure of the molecule, that is,
inter alia, the nature of the Rd group at d-position and the
presence or not of a second carbalkoxy group attached to
the double bond (alkylidenemalonates vs monoesters).2,3
For several years now, we have been engaged in the study of
SmI2-induced radical 3-exo-trig cyclisations. We have thus
devised a new access to cyclopropanes and cyclopropanols
through cyclisation of, respectively, d-halogeno1 and g-di-
substituted-d-oxo-a,b-unsaturated esters.2 In the case of
oxo-esters, g-disubstitution is required to avoid migration
of the ethylenic bond from a,b- to b,g-position. When the
two substituents at g-position are identical, the reaction
may give two diastereoisomers: the trans-cyclopropanol re-
sulting from cyclisation according to the anti mode and the
cis-cyclopropanol resulting from cyclisation according to
the syn mode (Scheme 1, in which Rg¼Me for the sake of
clarity). For methyl and benzyl esters, the cis adducts spon-
taneously lactonise in the conditions of the reaction. Lacto-
nisation does not take place with tert-butyl esters. The
As already published,2a two possible mechanisms were con-
sidered for these reactions (Scheme 2). According to the first
one, the ketonic substrate 1 is first reduced by SmI2 to ketyl
radical 2. Intramolecular condensation of the ketyl anion
onto the activated double bond of the enoate moiety then
leads (potentially in a reversible way) to the cyclised radical
intermediate 3. Whatever the position of the equilibrium, the
electrophilic a-carbalkoxy radical is rapidly and irreversibly
reduced to carbanion 4 by a second molecule of SmI2 and the
whole reaction is therefore displaced towards cyclisation.
Protonation of the carbanion completes the reaction. This
O
CO2R
O
Z
HO
HO
R
Z
SmI2 2.2 equiv.
CO2R
O
+
or
CO2R
R
AH
R
THF, 0 °C to rt
R
Z
Z
Z = H, CO2R
AH = t-BuOH (4 equiv) or PhOH (2.5 equiv)
(R = t-Bu)
(R = Me, Bn)
Scheme 1.
Keywords: 3-exo-trig Cyclisations; Radical cyclisations; Ketyl radicals.
* Corresponding authors. Tel.: +33 1 69 15 47 36; fax: +33 1 69 15 46 80 (S.B.-L.); tel.: +33 1 69 15 52 59; fax: +33 1 69 15 46 80 (F.G.); e-mail addresses:
sbezzenine@icmo.u-psud.fr; fraguibe@icmo.u-psud
0040–4020/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2007.02.080

C. Cammoun et al. / Tetrahedron 63 (2007) 3728–3736
3729
Mechanism 1
O
O
O
O
H+
SmI2
cyclisation
products
SmI2
CO2R"
CO2R"
CO2R"
CO2R"
R
R
R
R
R' R'
R' R'
R'
R'
R'
R'
1
2
3
4
Mechanism 2
O
O
O
O
SmI2
SmI2
H+
CO2R"
CO2R"
cyclisation
products
CO2R"
CO2R"
R
R
R
R
R' R'
R'
R'
R' R'
R'
R'
1
5
6
4'
Scheme 2.
mechanism is akin to the one we previously proposed1 for
cyclisation of d-halogeno-a,b-unsaturated esters.
to be fast,4 with a ko constant at least equal to (and probably
greater than) 107 sꢀ1 at 25 ꢁC. We therefore investigated the
reaction of benzyl 5-cyclopropyl-4,4-dimethyl-5-oxo-pent-
2-enoate 7 to see if radical ring opening of the cyclopropyl
group at d-position would interfere with the cyclisation pro-
cess. If so, mechanism 1 would have been validated. In the
event, no product of ring opening of the cyclopropyl group
was detected and the cyclisation reaction was found to give
selectively lactone 8 in virtually quantitative yield (Fig. 1).2a
According to the second mechanism, the reaction starts with
reduction of the enoate moiety to radical anion 5. Reversible
intramolecular condensation of this radical anion onto the
carbonyl group gives the cyclised cyclopropanoxy radical
6. Radical 6 is further reduced, in a fast and irreversible
process, to cyclopropanoate 40 already encountered (but not
necessarily with the same diastereoisomeric composition) in
mechanism 1.
The absence of ring opening with substrate 7 does not nec-
essarily validate mechanism 2. At least three explanations
may be proposed:
In our previous work we also used the ring opening of cyclo-
propyl ketyl radical as a probe to try to have further insight
into the mechanism of the reactions. There are several ex-
amples, in the literature, of reductive ring opening of cyclo-
propyl ketones in the presence of SmI2.5 The rate of ring
opening of methyl cyclopropyl ketyl radical has not yet
been accurately determined. It has nevertheless been shown
(i) the reaction effectively goes according to mechanism 2;
(ii) the reaction goes through mechanism 1 but the intramo-
lecularcondensationoftheketylradicalontotheactivated
double bond is rapid enough to exclude any competitive
ring opening of the cyclopropyl group at d-position;
O
O
O
CO2Me
O
CO2Bn
Ph
Ph
7
8
9
Figure 1.
O
O
O
k1c
k2-c
k2c
CO2R
CO2R
CO2R
k1-c
O
A
C
kcred
kored
SmI2
SmI2
O
O
CO2R
CO2R
O -
C -
Scheme 3.

3730
C. Cammoun et al. / Tetrahedron 63 (2007) 3728–3736
X
X
very fast
irreversible
Ph
X
k-c
Ph
Ph
R
R
Ph
Ph
Ph
R
10a X = R = H
10b X = O-; R = Me
11
12
Scheme 4.
O
O
O
HO
H
C(O)SEt
LiAlH4
(78%)
Ph
Ph
N(Me)OMe
Ph
Ph
H
Ph
Ph
SEt
LDA
(50%)
15
13
14
O
O
O
HO
H
Dess-Martin
periodinane
(EtO)2P(O)CH2CO2Me
(Et)3SiH,
9
Pd/C
Ph
Ph
H
Ph
Ph
H
NaH, THF
(57%)
(27%)
(53%)
17
16
Scheme 5.
(iii) the ketyl radical A (Scheme 3) does form and may open
to homoallylic distonic radical O. However, this reac-
tion is potentially reversible; if further reduction of
the homoallylic radical O by a second molecule of
SmI2 is slow enough compared to ring reformation,
then it may be conceived that the whole reaction is
shifted towards formation, through A and C, of ring
closed product Cꢀ.
opening may also be considered as irreversible. Its rate has
been estimated to be about 4–5 times that of the parent cyclo-
propylcarbinyl radical.7 To the best of our knowledge, there
is no data concerning the reactivity of the corresponding
ketyl radical 10b.
2. Results
2.1. Preparation of d-oxo-a,b-unsaturated ester 9
The relative paucity of kinetic values concerning the reac-
tions involving ketyl radicals,6 especially as compared to
those involving carbinyl radical for which much more data
are available, makes it difficult to choose among the three
above alternatives. Nevertheless, a simple consideration of
(even roughly) estimated values for kinetic constants im-
plied in the whole process is in disfavour of proposal (iii)
(see Section 3). But it should be noted that all these estima-
tions refer to free radical reactions and may very well be, as
already convincingly pointed out,5a unsuitable for samar-
ium-induced reactions, especially if, as it is the case here,
charged radical species are involved. Furthermore many
puzzling results have in the past been obtained in connection
with the use of ring opening of cyclopropyl ketyl radicals as
mechanistic probes in samarium promoted reactions.5a We
therefore wanted to get additional experimental convincing
evidence to decide whether the potentially reversible nature
of ring opening of ketyl radical A could affect or not the
SmI2-induced reaction of 7.
Diastereoisomerically pure 4,4-dimethyl-5-(trans-2-(2,2-
diphenyl-1R*-cyclopropyl)-1S*,2R*-cyclopropyl)-5-oxo-
prop-2-enoic acid methyl ester 9 was synthesised from
diastereoisomerically pure Weinreb amide 13 (Scheme 5)
whose synthesisy and thorough characterisation had already
been published by Newcomb and co-workers.7 Reduction of
13 with lithium aluminium hydride gave aldehyde 14. Com-
pound 14 was converted by aldol condensation with the lith-
ium salt of ethyl thioisobutyrate to b-hydroxythioester 15 as
a mixture of diastereoisomers in 50% overall yield. Com-
pound 15 was hydrogenolysed to b-hydroxyaldehyde 16
by Et3SiH in the presence of Pd/C8 in 53% yield. Oxidation
of 16 with Dess–Martin periodinane gave b-ketoaldehyde 17
in 27% yield. Compound 17 was finally converted to 9 in
57% yield through Wadsworth–Emmons olefination as
already published for similar substrates.2a
Given the rather low yield of most steps involved, homo-
logation of aldehyde 14 to b-ketoaldehyde 17 was achieved
only in 7% overall yield. We therefore investigated another
approach, as summarised in Scheme 6. Aldehyde 14 was
condensed with the dilithio salt of isobutyric acid to give
in 92% yield b-hydroxyacid 18, which was converted to allyl
To do so, we decided to carry out the SmI2-induced cycli-
sation process on compound 9 (4,4-dimethyl-5-(trans-2-
(2,2-diphenyl-1R*-cyclopropyl)-1S*,2R*-cyclopropyl)-5-
oxo-prop-2-enoic acid methyl ester) that incorporates
Newcomb’s 20,20-diphenyl-bicyclopropyl group7 at d-posi-
tion. Rearrangement of the 20,20-diphenyl-bicyclopropyl-2-
carbinyl radical 10a to diphenylcarbinyl radical 12 involves
two consecutive ring openings (Scheme 4). Due to the fast-
ness (and irreversibility) of ring opening of the intermediate
diphenylcyclopropylcarbinyl radical 11a, the first ring
y
Weinreb amide was prepared from 2,2-diphenyl-cyclopropane carbox-
aldehyde. In our work, this aldehyde was itself not prepared by CrO3–
pyridine oxidation of 2,2-diphenylcyclopropylcarbinol, which gives poor
yields17 but by palladium catalysed hydrostannolysis11 of 2,2-diphenyl-
cyclopropanecarbonyl chloride.

C. Cammoun et al. / Tetrahedron 63 (2007) 3728–3736
3731
O
O
HO
H
HO
H
Br
H
CO2H
Ph
Ph
O
Ph
Ph
O
14
Cs2CO3, DMF
(84%)
LDA
(92%)
18
19
1) Bu3SnH, Pd(PPh3)4 cat
2) ClCO-COCl
O
O
17
Dess-Martin
periodinane
Ph
Ph
O
3) Bu3SnH, Pd(PPh3)4 cat
(25% overall)
(84%)
20
Scheme 6.
ester 19 in 84% yield by treatment with caesium carbonate
and allyl bromide in DMF.9 b-Hydroxyester 19 was oxidised
to allyl b-ketoester 20 with Dess–Martin periodinane in 84%
yield. Compound 20 was finally converted to b-ketoalde-
hyde 17 through three consecutive treaments according to
a one-pot protocol, namely, hydrostannolysis to tributyltin
carboxylate by the Bu3SnH/cat Pd(PPh3)4 system,10 conver-
sion to acyl chloride with oxalyl chloride and hydrostannol-
ysis to aldehyde, again by the Bu3SnH/cat Pd(PPh3)4
system.11 The overall yield for this three step transformation
was only 25% but could in our opinion be optimised.
(n(CO)¼1772 cmꢀ1, identical with the one observed for
lactone 8) and the characteristic ABX system with JAB¼ca.
19 Hz, JAX¼6–7 Hz and JBX¼0 Hz formed by the protons
a and b to carbonyl and which, in the case of lactone 21,
is split into two due to the presence of the two diastereo-
isomeric forms in comparable amounts (Fig. 2).
3. Discussion
The absence of products from ring opening in the reaction of
9 with SmI2 clearly invalidates proposal (iii).
2.2. SmI2-induced cyclisation of d-oxo-a,b-unsaturated
ester 9
According to Scheme 3, the rates of formation at a given
instant t of ring opened Oꢀ and ring closed Cꢀ products is
given by:
The reaction of 9 with SmI2/t-BuOH was conducted in the
following way: to d-oxo-a,b-unsaturated ester 9 in solution
in THF at 0 ꢁC was added dropwise over a period of a few
minutes 2.2 equiv of a 0.1 M solution of SmI2 in THF. Com-
plete discharge of the typical blue colouration of SmI2 in
THF was observed after 2 h at room temperature. After con-
ventional work-up and purification by flash chromatography
on silica, lactone 21 was isolated in 80% yield as a mixture
of two diastereoisomers (Scheme 7). No other product could
be characterised. The lactone is clearly identified by high
resolution mass spectrometry, infra-red spectroscopy
ꢀ
ꢀ
ꢁ
d½Oꢀꢂ
dt
¼ kroed½Oꢂt½SmI2ꢂt
ð1Þ
t
ꢁ
d½Cꢀꢂ
dt
¼ krced½Cꢂt½SmI2ꢂt
ð2Þ
t
If we assume an ideal case in which the substrate is put in
reaction with a large excess of SmI2 and if we use a steady
O
O
SmI2, 2.2 equiv.
t-BuOH 4 equiv.
O
CO2Me
Ph
Ph
(80%)
THF
Ph
Ph
0 °C to rt
9
21
Scheme 7.
200 MHz
250 MHz
2.80
2.70
2.60
Lactone 8
Figure 2. 1H NMR ABX splitting patterns displayed by protons a to the carbonyl group of lactones 8 and 21.
2.50
2.40
2.30
2.80
2.70
2.60
2.50
2.40
2.30
ppm
ppm
Lactone 21

3732
C. Cammoun et al. / Tetrahedron 63 (2007) 3728–3736
state approximation for intermediates O and C, the ratio F of
opened to cyclised products is given by Eq. 3:z
the ketyl radical. The cyclic voltammogram of
CH3CH]CH–CO2Et, performed under the same condi-
tions, exhibited a well-defined irreversible reduction peak
located at a less negative potential: Ep¼ꢀ2.48 V versus
SCE. When comparing the reduction peak potentials of
each compound, one sees that the ethyl crotonate is more
easily reduced than the aliphatic ketones. The cyclic voltam-
mogram performed on the aromatic ketone PhCOCH(CH3)2
exhibited a reversible reduction peak at Ep¼ꢀ2.16 V
(E0¼ꢀ2.07 V). In a first approach, when comparing peak
potentials Ep, one sees that only aromatic ketones are more
easily reduced than the a,b-unsaturated ester. However,
this assumption would be totally true if standard potentials
were compared. Indeed, the aliphatic ketones and ester
investigated above exhibited irreversible reduction peaks
whose peak potentials cannot be compared because they
depend on the rate of the chemical reaction which follows
the electron transfer. In other words, only standard potentials
E0 can be compared.
À
Á
kꢀ2 c kꢀ1 c þ krced½SmI2ꢂ
kroed½Oꢂ kroed
¼
À
Á
F ¼
ꢃ
ð3Þ
krced½Cꢂ krced
kc1 kc2 þ kroed½SmI2ꢂ
For the reversible character of ring opening of ketyl radical
to become of consequence on the partitioning between ring
closed and ring opened products, the value of k2c must be at
least of the same order of magnitude as the product kroed
[SmI2]. The value of kroed may be approximated to about
7ꢃ106 sꢀ1 at 25 ꢁC.12 The value of k2c is not known but, in
the carbinyl series, a value of 0.8ꢃ104 sꢀ1 has been found
for cyclisation of the free 2-butenyl radical.13 Even if, in
our reaction conditions, the actual value of SmI2 concentra-
tion, which is not constant, is on an average much lower than
0.1 M, reduction of the opened radical appears favoured
compared to cyclisation back to A. Thus both kinetic consid-
erations and experimental observations are concordant to ex-
clude any influence of the reversibility of ring opening of
ketyl radical on the course of the SmI2-induced cyclisation
of d-oxo-a,b-unsaturated esters bearing a cyclopropyl group
at d-position.
This is why the cyclic voltammetry of two related com-
pounds PhCH]CH–CO2Et and PhCOCH(CH3)2 has been
performed in DMF. Both compounds exhibited a reversible
reduction peak at E0¼ꢀ1.77 V and ꢀ2.01 V, respectively.14
These standard potentials can now be compared. It is thus es-
tablished that the cinnamic ester is more easily reduced than
the aromatic ketone, itself of course more easily reduced
than aliphatic ketones.
Concerning mechanisms 1 and 2, we have already amply
discussed the argument in favour of or against each of
them.2a Some more considerations can nevertheless be
made.
Therefore, one can assume that the part of the molecules 1,
which preferentially accepts the first electron transfer is the
a,b-unsaturated ester group. Meanwhile, caution should be
exercised when extrapolating such conclusion based on elec-
trochemistry to reduction by SmI2 since it has been shown
that samarium diiodide behaves not as an outer-sphere
but as an inner-sphere electron donating agent towards
ketones.15 Besides, cyclisation according to mechanism 2
leads to a supposedly high in energy cyclopropanoxy radical
C and could be on that ground disfavoured.
As suggested by one of the referees, a comparison of the re-
duction potentials of ketones on one hand and of a,b-unsat-
urated esters on the other hand could help to determine
which part of d-oxo-a,b-unsaturated esters 1 is more likely
to accept the first electron transfer. Cyclic voltammetry
was thus performed on some representative molecules, in
collaboration with Dr Anny Jutand’s group (UMR CNRS-
ꢀ
ENS-UMPC 8640, Ecole Normale Superieure, Paris). The
cyclic voltammogram of aliphatic ketones t-BuCOMe and
cyclopropylCOMe performed in THF (containing n-
Bu4NBF4, 0.3 M, as supporting electrolyte), at a gold elec-
trode and a scan rate of 0.5 V sꢀ1, exhibited a badly resolved
reduction wave, appearing as a shoulder close to the reduc-
tion of the solvent at ca. ꢀ2.7 V versus SCE for both com-
pounds. Thus, as observed by Tanko and co-workers,4 the
cyclopropyl group does not appear to substantially stabilise
Examination of the values of the different kinetic constants
involved in mechanism 1 (Scheme 3), when they are avail-
able or may be reasonably estimated, is also instructive.
The k1c may be evaluated as follows. The kc constant for cyc-
lisation of 4-carbalkoxy-but-2-enyl radical is known,16 with
kc¼2ꢃ106 sꢀ1 at 25 ꢁC (from Arrhenius plot, kc¼1.6ꢃ
107 sꢀ1 at 80 ꢁC), a 200-fold increase if compared to the cyc-
lisation13,17 of the parent 3-butenyl radical (similarly, the
5-exo cyclisation of the 6-cyano-5-hexenyl radical is 1000
times faster than the cyclisation of 5-hexenyl radical at
room temperature18). Ketyl radicals are likely to be more
nucleophilic and therefore to cyclise more rapidly. We
have also (in our case) to take into account the gem-dimethyl
effect. In the case of 3-butenyl radicals, Newcomb was able
to evaluate that gem-disubstitution by methyl groups at the
2-position results in a 500-fold increase in the rate of cycli-
sation.17 The cyclisation constant k1c of Scheme 3 should
therefore be at least as high as 2ꢃ106ꢃ500 sꢀ1¼109 sꢀ1
(25 ꢁC). This value is to be compared with that of ring open-
ing of cyclopropyl ketyl radical of which we unfortunately
z
The cycloreversion of radical intermediate C to homoallylic dimethyl-
carbinyl radical D must also be considered. However, in all SmI2-induced
reactions of d-oxo-a,b-unsaturated esters we investigated, we never
detected any
O
k3-c
CO2R
CO2R
(i)
k3c
O
C
D
product derived from radical intermediates such as D. It may be recalled that
tertiary carbinyl radicals are hardly reduced to organosamarium in the pres-
ence of SmI2.23,24 It should be noted that the simple inclusion of equilibrium
i in Scheme 3 does not affect the above kinetic equations. Meanwhile, equi-
librium i should be taken into consideration when dealing with the
diastereoisomeric issue of the reaction, a fact that we overlooked in our
previous publication.2a
know only the lower limit4 (k2 ¼107 sꢀ1 at 25 ꢁC). Never-
ꢀc
theless, it is clear that the absence of ring opened products

C. Cammoun et al. / Tetrahedron 63 (2007) 3728–3736
3733
does not necessarily exclude mechanism 1 (i.e., proposal
4. Conclusion
(ii)). The lack of values for most kinetic constants involved
in mechanism 1 (to say nothing about mechanism 2) makes it
unfortunately delicate to draw more precise conclusions.x
trans-(20,20-Diphenyl-bicyclopropyl-2-yl)-4,4-dimethyl-5-
oxo-pent-2-enoic acid methyl ester 9, alike 5-cyclopropyl-
4,4-dimethyl-5-oxo-pent-2-enoate 7, cyclises in the pres-
ence of SmI2 to lactone 21 without competitive ring opening
of the bicyclopropylic probe at d-position. This absence of
ring opening rules out a possible effect of the reversibility
of ring opening of cyclopropyl ketyl radical on the issue of
these reactions. It shows that the cyclisation process is, what-
ever the mechanism (condensation of the ketyl radical onto
the activated ethylenic bond or condensation of the radical
anion derived from the enoate moiety to the carbonyl group),
a very fast process. However, it does not allow to choose be-
tween the two mechanisms. Indeed, based on kinetic consid-
erations, ketyl radicals if they form are more likely to evolve
through 3-exo-trig cyclisation onto the activated double
bond than through ring opening of the cyclopropyl group
at d-position.
The absence of ring opening in the 4-exo-trig cyclisation of
compound 22, reported by Procter and co-workers19 is more
intriguing. 4-exo-trig Cyclisations involving carbinyl radi-
cals are notoriously much slower than 3-exo-trig ones
(kc¼0.1–1 sꢀ1 at 60 ꢁC for ring closing of 4-pentenyl radical
against 0.8ꢃ104 sꢀ1 at 25 ꢁC for ring closing of 3-butenyl
radical13). It suggests that whatever the mechanism (mecha-
nism 1 or as proposed by the authors mechanism 2), the cyc-
lisations either of the radical anion onto the carbonyl group
or of the ketyl radical onto the activated double bond are fast
4-exo-trig processes (Scheme 8).
O
SmI2
O
Finally, the syn selectivity of cyclisation of d-cyclopropylic
d-oxo-a,b-unsaturated esters (as compared to other d-alkylic
substrates) remains unexplained.
O
O
THF-t-BuOH
0 °C
O
H
OH
22
5. Experimental
Scheme 8.
5.1. General information
Finally, the total diastereoselectivity observed in favour of
the formation of lactone with compound 9 incorporating
Newcomb’s diphenyl-bicyclopropyl group deserves some
comment. This diastereoselectivity is intriguing since ap-
proximately equimolecular amounts of trans-cyclopropanol
and of lactone are found in the cyclisation of d-oxo-a,b-un-
saturated esters bearing indifferently an H, Me or i-Pr group
at d-position.2a The present result thus confirms our earlier
observation concerning cyclopropanic compound 7. But it
also invalidates our tentative explanation based on a revers-
ible concerted ring opening of the cyclopropane group
during electron transfer from SmI2 to the carbonyl group.2a
1H NMR spectra were recorded at 200, 250, 360 or 400 MHz
and 13C spectra at 63 or 100 MHz. Chemical shifts are
quoted in parts per million relative to TMS. High resolution
mass spectra (HRMS) were obtained on a Finnigan-
MAT-95-S spectrometer. Infra-red spectra were taken on
a Perkin–Elmer ‘Spectrum One’ model and in CHCl3
solution.
As a rule, reactions were carried out under argon atmo-
sphere, using Schlenk tube, septum and syringe technics.
5.2. Preparation of starting compound 9 (4,4-dimethyl-
5-(trans-2-(2,2-diphenyl-1R*-cyclopropyl)-1S*,2R*-
cyclopropyl)-5-oxo-prop-2-enoic acid methyl ester)
If the k2c term is neglected against the product kored[SmI2], the opened to
cyclised products ratio F is now given by:
x
ꢀ
ꢁ
kꢀ2 c
kꢀ2 ckꢀ1 c
kꢀ2 c
kroed
kꢀ2 ckꢀ1 c
5.2.1.
2,2-Diphenylcyclopropanecarbaldehyde.
To
F ¼
þ
¼
þ
ꢃ
ð4Þ
kc1 kc1krced½SmI2ꢂ kc1
krced kc1kroed½SmI2ꢂ
a stirred solution of crude 2,2-diphenylcyclopropanecar-
bonyl chloride (obtained by reaction of 3 g (12.8 mmol) of
2,2-diphenylcyclopropanecarboxylic acid20 and oxalyl chlo-
ride) and Pd(PPh3)4 (296 mg, 0.26 mmol) in benzene
(40 mL) was slowly added with a syringe pump tributyltin
hydride (4.32 mL, 16 mmol). The reaction was further
stirred at room temperature for 1 h. After evaporation of benz-
ene, the residue was taken up in acetonitrile and partitioned
between hexane and acetonitrile using four separatory fun-
nels. The acetonitrile extracts were joined and evaporated
to give 4.3 g of crude product. This residue was purified
by column chromatography on 150 g of silica mixed with
15 g of potassium fluoride ground in a mortar.21 White
crystals were obtained (2.60 g, 91% yield, from carboxylic
acid).
If the following values are entered (25 ꢁC): kroed¼7ꢃ106 sꢀ1 Mꢀ1
,
k2c¼104 sꢀ1 (the rate constant value for cyclisation of the 3-butenyl radi-
cal),13,17 k2 ¼5ꢃ107 sꢀ1 (the lower limit value for ring opening of cyclo-
ꢀc
propyl ketyl radical4 multiplied by 5 due to the presence of the
diphenylcyclopropyl reporter group7), kc1¼109 sꢀ1 (see the text),
k1 ¼1.6ꢃ107 sꢀ1 (k value for ring opening of a-carbalkoxy substituted car-
ꢀc
binyl radical as determined by Bowry and Beckwith:16 k¼0.8ꢃ106 sꢀ1 from
Arrhenius plot, and re-evaluated (ꢃ20) according to Newcomb7), Eq. 4
becomes:
10ꢀ1
F ¼ 5 ꢃ 10ꢀ2
þ
ð5Þ
À
Á
krced=kroed ½SmI2ꢂ
It may be safely assumed that the reduction constant of the electrophilic a-
carbalkoxy radical C is higher than that for the opened primary alkyl radical
O (krced>kroed¼7ꢃ106 sꢀ1 Mꢀ1) even if it is difficult to define to which extent.
If enough confidence may be put in k values incorporated in Eq. 4 to give Eq.
5, ring opening of cyclopropyl groups at d-position should not interfere with
ring closing of the ketyl radical according to mechanism 1 if [SmI2] is not
too low.
1
Mp: 72–74 ꢁC. H NMR (250 MHz, CDCl3) d: 8.67 (d,
J¼6.9 Hz, 1H), 7.42–7.18 (m, 10H), 2.59–2.50 (m, 1H),

3734
C. Cammoun et al. / Tetrahedron 63 (2007) 3728–3736
2.26 (t, J¼5.1 Hz, 1H), 1.88 (dd, J¼8.2 Hz and J0¼5.1 Hz,
1H); 13C NMR (63 MHz, CDCl3) d: 200.7 (C]O), 144.1
(Cq), 139.6 (Cq), 130.3 (CHar), 129.1 (CHar), 128.8
(CHar), 127.6 (CHar), 127.1 (CHar), 41.1 (Cq), 36.9 (CH),
20.5 (CH2); IR (CHCl3): n(C]O) 1701 (s).
5.2.4. 2,2-Dimethyl-3-(trans-2-(2,2-diphenyl-1R*-cyclo-
propyl)-1S*,2R*-cyclopropyl)-3-hydroxy-propionalde-
hyde 16. Thioester 15 (600 mg, 1.52 mmol) was dissolved in
dry degassed dichloromethane (8 mL). 208 mg, (0.2 mmol)
of 10% palladium on charcoal was rapidly introduced, fol-
lowed by triethylsilane (1.56 mL) with a syringe. The reac-
tion was monitored by TLC. After 2 h stirring at room
temperature, the reaction mixture was filtered through
a pad of Celite. After evaporation of solvent, the residue
was purified by flash column chromatography on silica gel
(eluent heptane/AcOEt 80:20). The aldehyde was obtained
as white crystals (271 mg, 53% yield).
5.2.2. trans-2-(2,2-Diphenyl-1R*-cyclopropyl)-1S*,2R*-
cyclopropane-carboxaldehyde 14. LiAlH4 (196 mg,
5.17 mmol) was dissolved in dry diethyl oxide (30 mL). A
solution of Weinreb amide 13 (1.33 g, 4.13 mmol) in dry di-
ethyl oxide (10 mL) was added with a cannula. The reaction
mixture was further stirred for 2 h at ꢀ78 ꢁC and then
quenched with aqueous KHSO4 (1.25 g in 26 mL of water).
The organic layer was recuperated and the aqueous phase
was extracted three more times with 100 mL of diethyl ox-
ide. The organic phases were extracted successively with
1 N aqueous HCl (3ꢃ100 mL), aqueous saturated sodium
carbonate (2ꢃ40 mL) and brine. The organic phases were
joined, dried (MgSO4), then evaporated on a Rotovap. The
residue was purified by column chromatography on silica
(eluent heptane/AcOEt 80:20) to give 850 mg of pure alde-
hyde (78% yield) as a white solid.
Mp: 44–45 ꢁC. 1H NMR (250 MHz, CDCl3) d: 9.57 (s, 1H),
7.35–7.13 (m, 10H), 3.12 (d, J¼7.5 Hz, 1H), 1.73–1.61 (m,
3H), 1.25–1.12 (m, 7H), 1.08–0.69 (m, 2H), 0.18–0.07 (m,
1H); 13C NMR (63 MHz, CDCl3) d: 206.7 (C]O), 130.8
(CHar), 128.6 (CHar), 128.5 (CHar), 127.7 (CHar), 126.6
(CHar), 126.0 (CHar), 78.1, 51.4 (Cq), 35.8 (Cq), 31.3,
28.4, 21.3, 19.6 (CH2), 18.6, 17.5, 8.41 (CH2); HRMS (EI)
calcd for C23H26O2: 334.1933, found: 334. 1938; IR
(CHCl3): n(C]O) 1721 (s).
1
Mp: 50–51 ꢁC. H NMR (200 MHz, CDCl3) d: 8.77 (d,
5.2.5. 2,2-Dimethyl-3-(trans-2-(2,2-diphenyl-1R*-cyclo-
propyl)-1S*,2R*-cyclopropyl)-3-hydroxy-propionic acid
18. A 2.5 M solution of n-BuLi (8.85 mL, 22.1 mmol) was
added to a solution of DIEA (2.25 g, 22.3 mmol) in THF
(50 mL) at ꢀ40 ꢁC. The reaction mixture was brought to
0 ꢁC, stirred for 25 min at this temperature and cooled again
at ꢀ40 ꢁC. Isobutyric acid (980 mg, 11.1 mmol) was then
syringed into the reaction mixture. After heating at 50 ꢁC
for 2 h, the reaction mixture was once more cooled at
ꢀ40 ꢁC. Aldehyde 14 (2.43 g, 9.28 mmol) in solution in
THF (50 mL) was added dropwise and the reaction mixture
was maintained at ꢀ40 ꢁC for 2 h. Quenching with water
(50 mL) was followed by extraction with diethyl oxide
(3ꢃ70 mL). The aqueous phase was recuperated and cooled
in an ice/water bath. The acid was precipitated by acidifica-
tion with dilute aqueous HCl. The heterogeneous mixture
was extracted with diethyl oxide (3ꢃ70 mL). The ethereal
extracts were joined, dried (MgSO4) and evaporated to
give 3.0 g of acid 18 as a white solid. NMR showed the pres-
ence of the two diastereoisomeric forms in a close to 1:1
ratio (2.99 g, 92% yield).
J¼5.9 Hz, 1H), 7.35–7.12 (m, 10H), 1.89–1.80 (m, 1H),
1.49–1.41 (m, 1H), 1.37–1.25 (m, 2H), 1.16–1.08 (m, 1H),
1.03–0.85 (m, 2H); 13C NMR (63 MHz, CDCl3) d: 200.2
(C]O), 146.4 (Cq), 141.3 (Cq), 130.7 (CHar), 128.6
(CHar), 128.5 (CHar), 127.5 (CHar), 126.9 (CHar), 126.2
(CHar), 35.8 (Cq), 31.0 (CH), 28.1 (CH), 23.9 (CH), 19.8
(CH2), 14.4 (CH2); HRMS (EI) calcd for C19H18O:
262.1358, found: 262.1363; IR (CHCl3): n(C]O) 1701 (s).
5.2.3. 2,2-Dimethyl-3-(trans-2-(2,2-diphenyl-1R*-cyclo-
propyl)-1S*,2R*-cyclopropyl)-3-hydroxy-thiopropionic
acid, S-ethyl ester 15. n-BuLi 2.48 M in hexanes (1.49 mL,
3.70 mmol) was added to a solution of DIEA (0.55 mL,
3.9 mmol) in dry THF (15 mL) at 0 ꢁC. The reaction mixture
was stirred at 0 ꢁC for 30 min, then cooled to ꢀ78 ꢁC. A
solution of thioisobutyric acic S-ethyl ester (510 mg,
3.9 mmol) in THF (5 mL) was added and the reaction mix-
ture was further stirred for 1.5 h at ꢀ78 ꢁC. A solution of al-
dehyde 14 (850 mg, 3.2 mmol) in THF (9 mL) was added
via a cannula. The reaction mixture was further stirred at
ꢀ78 ꢁC. After ca. 1 h at ꢀ78 ꢁC (the reaction was monitored
by TLC), the reaction mixture was quenched with aqueous
saturated NH4Cl (22 mL) followed by ethyl acetate
(70 mL). After decantation, the aqueous layer was extracted
with ethyl acetate (3ꢃ25 mL). The organic extracts were
joined, washed with brine, dried (MgSO4) and evaporated.
The residue was purified by flash column chromatography
on silica (eluent heptane/AcOEt 90:10) to give 642 mg of
pure 15 as an oil (51% yield).
1H NMR (250 MHz, CDCl3) d: 7.36–7.14 (m, 20H), 3.81–
3.68 (m, 2H), 3.10 (d, 1H, dia 1, J¼7.5 Hz), 2.68 (d, 1H,
dia 2, J¼10 Hz), 1.85–1.69 (m, 2H), 1.52–1.38 (m, 2H),
1.35 (s, 6H), 1.22 (s, 6H), 1.23–1.12 (m, 2H), 0.96–0.74
(m, 2H), 0.61–0.39 (m, 2H), 0.20–0.10 (m, 4H); IR
(CHCl3): n(C]O) 1748 (s).
5.2.6. 2,2-Dimethyl-3-(trans-2-(2,2-diphenyl-1R*-cyclo-
propyl)-1S*,2R*-cyclopropyl)-3-hydroxy-propionic acid
allyl ester 19. Acid 18 (2.96 g, 8.45 mmol) was dissolved
in dry DMF (22 mL). Caesium carbonate (2.0 g,
6.1 mmol) was rapidly introduced. The reaction mixture
was stirred for 10 min before adding allyl bromide
(1.05 g, 8.65 mmol). Stirring was continued for 20 h at
room temperature. The reaction mixture was evaporated
under vacuum (0.5 Torr) and the residue was taken up in
chloroform. To the heterogeneous mixture was added aque-
ous sodium bicarbonate. The organic phase was decanted
1H NMR (250 MHz, CDCl3) d: 7.28–7.11 (m, 10H), 3.19 (d,
J¼7.1 Hz, 1H), 2.86 (q, J¼7.5 Hz, 2H), 2.1 (s, 1H), 1.72–
1.64 (m, 1H), 1.38–1.13 (m, 10H), 0.87–0.66 (m, 3H),
0.19–0.05 (m, 1H); 13C NMR (63 MHz, CDCl3) d: 207.6
(C]O), 147.4 (Cq), 142.0 (Cq), 130.9 (CHar), 128.5
(CHar), 128.4 (CHar), 127.6 (CHar), 126.6 (CHar), 125.9
(CHar), 79.2, 54.9 (Cq), 35.7 (Cq), 31.2, 28.6, 23.4, 23.1
(CH2), 21.1 (CH2), 18.7, 17.2, 14.7, 8.34 (CH2); IR
(CHCl3): n(C]O) 1663 (s).

C. Cammoun et al. / Tetrahedron 63 (2007) 3728–3736
3735
and dried over MgSO4. After evaporation of chloroform, the
residue was purified by flash column chromatography on
silica (eluent heptane/AcOEt 80:20). A ca. 1:1 mixture of
the two diastereoisomers was collected (2.77 g, 84% yield).
5.2.8.2. From 2,2-dimethyl-3-(trans-2-(2,2-diphenyl-
1R*-cyclopropyl)-1S*,2R*-cyclopropyl)-3-oxo-propionic
acid allyl ester 20. In a first Schlenk tube and under argon
atmosphere, tributyltin hydride (0.86 mL, 3.2 mmol) diluted
in degassed benzene (1.2 mL) was added with a syringe
pump over 1 h to a solution of b-ketoester 20 (1.00 g,
2.57 mmol) and Pd(PPh3)4 (74 mg, 0.065 mmol) in degassed
benzene (8 mL). At the end of addition, complete conversion
of allyl ester to tributyltin carboxylate was checked by IR
1H NMR (250 MHz, CDCl3) d: 7.30–7.10 (m, 20H), 5.96–
5.82 (m, 2H), 5.27–5.17 (m, 4H), 4.61–4.54 (m, 4H),
3.16 (d, 1H, dia 1, J¼7 Hz), 2.83 (d, 1H, dia 2, J¼9.5 Hz),
1.71–1.64 (m, 2H), 1.32–1.13 (m, 18H), 0.97–0.68
(m, 4H), 0.38–0.24 (m, 4H), 0.17–0.03 (m, 2H); 13C NMR
(63 MHz, CDCl3) d: 177.1 (C]O), 147.2 (Cq), 141.9
(Cq), 132.4 (CH), 130.7 (CHar), 128.4 (CHar), 127.7
(CHar), 126.6 (CHar), 125.9 (CHar), 118.2 (CH2), 80.7,
78.9, 65.4 (CH2), 53.7 (Cq), 47.9 (Cq), 35.7 (Cq), 28.9,
22.4, 21.8, 20.9 (CH2), 19.0, 16.7, 11.8 (CH2), 8.19 (CH2);
IR (CHCl3): n(C]O) 1725 (s), n(C]C) 1648 (w).
spectroscopy on an aliquot (n(CO2All)¼1735 cmꢀ1
,
n(CO2SnBu3)¼1646 cmꢀ1 in CHCl3). The reaction mixture
was then transferred via a cannula and over a period of
10 min into a second Schlenk tube containing a solution of
oxalyl chloride in degassed benzene (8 mL). A vigorous
gas evolution was observed. After completion of addition,
the reaction was stirred for 15 min and concentrated under
vacuum (0.5 Torr). Benzene (8 mL) was then added and
the reaction mixture was reloaded with Pd(PPh3)4 (74 mg).
Tributyltin hydride (0.86 mL, 3.2 mmol) diluted in degassed
benzene (2 mL) was added with a syringe pump over 2 h.
The solvent was removed under vacuum and the residue
was partitioned between hexane and acetonitrile using four
separatory funnels. The acetonitrile extracts were joined
and evaporated. The residue was purified by column chroma-
tography on silica/KF21 10:1 (eluent heptane/AcOEt 95:5) to
give 213 mg of aldehyde 17 as a white solid.
5.2.7. 2,2-Dimethyl-3-(trans-2-(2,2-diphenyl-1R*-cyclo-
propyl)-1S*,2R*-cyclopropyl)-3-oxo-propionic acid allyl
ester 20. A solution of Dess–Martin periodinane (4.63 g,
9.97 mmol) in dry dichloromethane (50 mL) was added
through a cannula to a solution of b-hydroxyester 19 in
dry dichloromethane (50 mL). The reaction mixture, which
became progressively cloudy, was stirred for 2 h at room
temperature. After dilution in diethyl oxide, the organic
phase was washed with a 1:1 mixture of aqueous 10%
Na2S3O3 and aqueous 5% NaHCO3 and with brine. The or-
ganic phase was dried (MgSO4) and evaporated. The residue
was purified by flash column chromatography on silica (elu-
ent heptane/AcOEt 80:20) to give 2.25 g (84% yield) of
a white solid.
Mp: 74–75 ꢁC. 1H NMR (250 MHz, CDCl3) d: 9.57 (s, 1H),
7.27–7.10 (m, 10H), 1.99–1.94 (m, 1H), 1.53–1.48 (m, 1H),
1.30–1.19 (m, 8H), 1.05–0.98 (m, 1H), 0.94–0.85 (m, 1H),
0.76–0.70 (m, 1H); 13C NMR (63 MHz, CDCl3) d: 207.4
(C]O), 201.2 (C]O), 146.5 (Cq), 141.2 (Cq), 130.4
(CHar), 128.5 (CHar), 127.7 (CHar), 126.7 (CHar), 126.1
(CHar), 60.4 (CH2), 35.9 (CH2), 28.4, 27.3, 25.9, 19.4, 19.3,
18.6, 17.7; HRMS (EI) calcd for C23H24O2: 332.1776,
found: 332. 1781; IR (CHCl3): n(C]O) 1729 (s), 1698 (s).
1
Mp: 45–48 ꢁC. H NMR (250 MHz, CDCl3) d: 7.30–7.10
(m, 10H), 5.94–5.83 (m, 1H), 5.33–5.17 (m, 2H), 4.71–
4.57 (m, 2H), 1.92–1.86 (m, 1H), 1.59–1.54 (m, 1H), 1.43
(s, 3H), 1.37 (s, 3H), 1.23–1.20 (m, 2H), 0.96–0.85 (m,
2H), 0.65–0.58 (m, 1H); 13C NMR (63 MHz, CDCl3) d:
206.6 (C]O), 173.8 (C]O), 146.8 (Cq), 141.4 (Cq),
131.9 (CH), 130.7 (CHar), 128.6 (CHar), 128.5 (CHar),
127.5 (CHar), 126.8 (CHar), 126.1 (CHar), 118.8 (CH2),
66.1 (CH2), 55.9 (Cq), 36.1 (Cq), 28.4, 26.5, 25.7, 22.2,
18.6 (CH2), 18.2 (CH2); HRMS (EI) calcd for C26H28O3:
388.2038, found: 388. 2042; IR (CHCl3): n(C]O) 1735
(s), 1697 (s), n(C]C) 1648 (w).
5.2.9. 4,4-Dimethyl-5-(trans-2-(2,2-diphenyl-1R*-cyclo-
propyl)-1S*,2R*-cyclopropyl)-5-oxo-prop-2-enoic acid
methyl ester 9. Compound 9 was prepared by Wads-
worth–Emmons olefination of b-keto-aldehyde 17 with
methyl diethoxyphosphonoacetate. The experimental proce-
dure was the same as that used by Nicolaou and co-workers22
for Wadsworth–Emmons olefination of 2,2-dimethyl-3-oxo-
pentanal with tert-butyl diethoxyphosphonoacetate. The res-
idue was purified by flash column chromatography on silica
(eluent heptane/AcOEt 90:10). Starting from 120 mg of
aldehyde, 80 mg of 9 was obtained as an oily compound
(57% yield).
5.2.8. 2,2-Dimethyl-3-(trans-2-(2,2-diphenyl-1R*-cyclo-
propyl)-1S*,2R*-cyclopropyl)-3-oxo-propionaldehyde
17.
5.2.8.1. By oxidation of 2,2-dimethyl-3-(trans-2-(2,2-
diphenyl-1R*-cyclopropyl)-1S*,2R*-cyclopropyl)-3-hy-
droxy-propionaldehyde 16. A solution of b-hydroxyalde-
hyde 16 (270 mg, 0.81 mmol) in dried dichloromethane
(1 mL) was added, through a cannula, to a solution of
Dess–Martin periodinane (540 mg, 1.28 mmol) in dichloro-
methane (2 mL). The reaction mixture was stirred at room
temperature for 45 min, diluted with diethyl oxide (5 mL)
and washed with a 1: 1 mixture of aqueous 10% Na2S3O3
and aqueous 5% NaHCO3 and with brine. The aqueous
washings were re-extracted once with diethyl ether. The or-
ganic phases were joined, dried over MgSO4, evaporated and
the residue was purified by flash column chromatography on
silica (eluent heptane/AcOEt 90:10). A white solid was ob-
tained (72 mg, 27% yield).
1H NMR (200 MHz, CDCl3) d: 7.28–7.10 (m, 10H), 5.96 (d,
1H, J¼14.4 Hz), 5.28 (d, 1H, J¼1 Hz), 3.74 (s, 3H), 2.04–
1.95 (m, 1H), 1.69–1.44 (m, 1H), 1.30–1.25 (m, 8H),
1.01–0.84 (m, 2H), 0.73–0.64 (m, 1H); HRMS (EI) calcd
for C26H28O3: 388.2038, found: 388.2044; IR (CHCl3):
n(C]O) 1719 (s), 1694 (s), n(C]C) 1647 (m).
5.3. Reaction of 5-(20,20-diphenyl-bicyclopropyl-2-yl)-
4,4-dimethyl-5-oxo-prop-2-enoic acid methyl ester 9
with SmI2
In a Schlenk tube and under argon atmosphere, tert-butanol
(80 mL, 0.82 mmol) was added with a syringe to a solution

3736
C. Cammoun et al. / Tetrahedron 63 (2007) 3728–3736
McKie, J. A. J. Org. Chem. 1991, 56, 4112–4120; (c)
Batey, R. A.; Harling, J. D.; Motherwell, W. B. Tetrahedron
1996, 42, 11421–11444; (d) Sheikh, S. E.; Kausch, N.; Lex,
€
J.; Neudorfl, J.-M.; Schmalz, H.-G. Synlett 2006, 1527–
1530.
of 9 (80 mg, 0.205 mmol) in anhydrous degassed THF
(1 mL). The reaction mixture was cooled to 0 ꢁC and
a 0.1 M THF solution of SmI2 (4.5 mL) was added dropwise
over a period of a few minutes. The reaction mixture was
stirred at room temperature. After ca. 2 h discharge of the
blue colour of SmI2 in THF was observed. The reaction mix-
ture was diluted with diethyl oxide and quenched with 1 N
aqueous HCl. The organic phase was recuperated while
the aqueous phase was extracted twice with diethyl oxide.
Each ethereal phase was in turn washed with aqueous
Na2S2O3 and brine. The organic phases were joined, dried
and evaporated. Purification was achieved by flash column
chromatography on silica gel (eluent heptane/AcOEt) to
give 60 mg of an oily product, which was identified as lac-
tone 21 (yield 81%).
6. The kinetics and thermodynamics of cyclopropyl ketyl radical
ring opening have been extensively studied by Tanko and
Tanner, but mainly for ketyl radicals issued from aryl cyclo-
propyl ketones whose study is facilitated by their thermo-
dynamic stability. See in particular: (a) Ref. 4; (b) Chahma,
M.; Li, X.; Phillips, J. P.; Schwartz, P.; Brammer, L. E.;
Wang, Y.; Tanko, J. M. J. Phys. Chem. A 2005, 109, 3372–
3382; (c) Tanko, J. M.; Gillmore, J. G.; Friedline, R.;
Chahma, M. J. Org. Chem. 2005, 70, 4170–4173; (d) Tanko,
J. M.; Phillips, J. P. J. Am. Chem. Soc. 1999, 121, 6078–
6079; (e) Tanko, J. M.; Drumright, R. E. J. Am. Chem. Soc.
1992, 114, 1844–1854; (f) Tanner, D. P.; Chen, J. J.; Luelo,
C.; Peters, P. M. J. Am. Chem. Soc. 1992, 114, 713–717.
7. Horner, J. M.; Tanaka, N.; Newcomb, M. J. Am. Chem. Soc.
1998, 120, 10379–10390.
The reaction was repeated twice with virtually identical
results.
Mixture of two diastereoisomeric forms in comparable
1
8. Fukuyama, T.; Lin, S.-C.; Li, L. J. Am. Chem. Soc. 1990, 112,
7050–7051.
ꢀ
9. Kumar, U.; Kato, T.; Frechet, J. M. J. Am. Chem. Soc. 1992,
amounts: H NMR (250 MHz, CDCl3) d: 7.41–7.09 (m,
10H+10H), 2.67–2.53 (two overlapping dd, 2Jw20 Hz,
3Jw7 Hz, 1H+1H), 2.40–2.32 (two overlapping d, 2Jw
20 Hz, 1H+1H), 1.52–1.36 (m, 1H+1H), 1.25–0.68 (m,
11H+11H), 0.67–0.65 (m, 1H+1H), 0.43–0.38 (m,
1H+1H); 13C NMR (63 MHz, CDCl3) d: CO: 177.6, 117.5,
Cq ar: 147.3, 147.1, 141.9, 141.8, CH ar: 131.2, 130.6,
128.7, 128.5, 128.4, 127.9, 127.7, 126.7, 126.6, 126.1,
125.9, Cq: 75.4, 74.3, 36.2, 35.8, 26.1, 25.5, CH2: 30.82,
30.80, 18.80, 18.60, 11.49, 10.01, CH and CH3: 29.4, 29.0,
23.7, 22.4, 22.3, 22.1, 18.1, 17.8, 17.3, 17.2, 13.9, 13.6;
HRMS calcd for C25H26O2: 358.1927, found: 358.1915; IR
(CHCl3): n(C]O) 1772.
114, 6630–6639.
10. Dangles, O.; Guibe, F.; Balavoine, G.; Lavielle, S.; Marquet, A.
ꢀ
J. Org. Chem. 1987, 52, 4984–4993.
ꢀ
11. Four, P.; Guibe, F. J. Org. Chem. 1981, 46, 4439–4445.
12. This is the value recommended by Curran and co-workers5a
for reduction of primary radicals by SmI2 under Barbier
conditions (i.e., in the presence of ketonic compounds, which
obviously is our case since we deal with intramolecular
reactions).
13. Park, S.-U.; Varick, T. R.; Newcomb, M. Tetrahedron Lett.
1990, 31, 2975–2978 and references cited therein.
14. Avalue of ꢀ1.789 V was previously obtained by Utley and co-
€ €
workers: Fussing, I.; Gullu, M.; Hammerich, O.; Hussain, A.;
Acknowledgements
Nielsen, M. F.; Utley, J. H. P. J. Chem. Soc., Perkin Trans. 2
1996, 649–658.
We thank Dr. A. Jutand (UMR CNRS-ENS-UMPC 8640,
15. (a) Enemaerke, R. J.; Daasbjerg, K.; Skrydstrup, T. Chem.
Commun. 1999, 343–344; (b) Enemaerke, R. J.; Hertz, T.;
Skrydstrup, T.; Daasbjerg, K. Chem.—Eur. J. 2000, 6, 3747–
3753.
ꢀ
Ecole Normale Superieure, Paris) for carrying out the cyclic
voltammetry study and Dr. A. Jutand and Dr. M. Mellah
ꢀ
(ICMMO, UMR CNRS 8182, Universite Paris-Sud Orsay)
for very helpful discussions.
16. Beckwith, A. L. J.; Bowry, V. W. J. Am. Chem. Soc. 1994, 116,
2710–2716.
References and notes
17. Newcomb, M.; Glenn, A. G.; Williams, W. G. J. Org. Chem.
1989, 54, 2675–2681.
18. Newcomb, M.; Varick, T. R.; Ha, C.; Manek, M. B.; Yue, X.
J. Am. Chem. Soc. 1992, 114, 8158–8163.
19. Hutton, T. K.; Muir, K. W.; Procter, D. J. Org. Lett. 2003, 5,
4811–4814.
1. (a) David, H.; Bonin, M.; Doisneau, G.; Guillerez, M.-G.;
ꢀ
Guibe, F. Tetrahedron Lett. 1999, 40, 8557–8561; (b) Villar,
H.; Guibe, F.; Aroulanda, C.; Lesot, P. Tetrahedron:
ꢀ
Asymmetry 2002, 13, 1465–1475.
20. Walborsky, H. M.; Barash, L.; Young, A. E.; Impastato, F. J.
J. Am. Chem. Soc. 1961, 83, 2517–2525.
ꢀ
2. (a) Bezzenine-Lafollee, S.; Guibe, F.; Villar, H.; Zriba, R.
Tetrahedron 2004, 60, 6931–6944; (b) Zriba, R.; Bezzenine-
ꢀ
21. Harrowven, D. C.; Guy, I. L. Chem. Commun. 2004, 1968–
1969.
ꢀ
ꢀ
Lafollee,S.;Guibe, F.;Guillerez, M.-G. Synlett2005, 2362–2366.
3. Cyclic d-oxo-a,b-unsaturated esters also display cyclisation
stereoselectivities different from those of the linear substrates
22. 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.
23. Curran, D. P.; Fevig, T. L.; Jasperse, C. P.; Totleben, M. J.
Synlett 1992, 943–961.
ꢀ
represented in Scheme 1 (Zriba, R; Bezzenine-Lafollee, S.;
Guibe, F.; Guillot, R.; Magnier-Bouvier, C., in preparation).
ꢀ
4. Stevenson, J. P.; Jackson, W. F.; Tanko, J. M. J. Am. Chem. Soc.
2002, 124, 4271–4281.
24. Inanaga, J.; Ishikawa, M.; Yamaguchi, M. Chem. Lett. 1987,
1485–1486.
5. See for instance: (a) Curran, D. P.; Gu, X.; Zhang, W.; Dowd, P.
Tetrahedron 1997, 53, 9023–9042; (b) Molander, G. A.;