Regio- and stereoselectivities in Diels-Alder cyclodimerizations of orthoquinonoid cyclohexa-2,4-dienones

Article abstract of DOI:10.1016/j.tet.2007.03.035

The [4+2] cyclodimerization of cyclohexa-2,4-dienone derivatives of the orthoquinone monoketal and orthoquinol types has been the topic of numerous investigations over the last 50 years in the aim of rationalizing the extraordinary level of regio-, site-,

Full text of DOI:10.1016/j.tet.2007.03.035

Tetrahedron 63 (2007) 6493–6505
Regio- and stereoselectivities in Diels–Alder cyclodimerizations
of orthoquinonoid cyclohexa-2,4-dienones
y
*
€
ꢀ
ꢀ
ꢀ ꢀ
Julien Gagnepain, Raphael Mereau, Delphine Dejugnac, Jean-Michel Leger, Frederic Castet,
*
ꢀ
Denis Deffieux, Laurent Pouysegu and Stephane Quideau
ꢀ
´
Universite de Bordeaux, Institut des Sciences Moleculaires (UMR 5255 CNRS), 351 cours de la Liberation, 33405 Talence, and Institut
´
´
´
Europeen de Chimie et Biologie, 2 rue Robert Escarpit, 33607 Pessac, France
Received 9 February 2007; revised 1 March 2007; accepted 5 March 2007
Available online 12 March 2007
Abstract—The [4+2] cyclodimerization of cyclohexa-2,4-dienone derivatives of the orthoquinone monoketal and orthoquinol types has been
the topic of numerous investigations over the last 50 years in the aim of rationalizing the extraordinary level of regio-, site-, and stereoselec-
tivities observed in these processes. In particular, the double diastereo-p-facial differentiation expressed in cyclodimerizations of chiral or-
thoquinols (i.e., 6-alkyl-6-hydroxycyclohexa-2,4-dienones) is an important aspect of these transformations, for they relate to the construction
of several natural products. The experimental and theoretical results that are described in this article offer a comprehensive understanding of
the factors controlling these site-specific regio- and diastereoselectivities. Our interpretation of these results relies on a combination of Wood-
ward–Hoffmann and Salem–Houk secondary orbital interactions and Cieplak-type hyperconjugative effects in bispericyclic C2-symmetric
transition states.
Ó 2007 Elsevier Ltd. All rights reserved.
Nu
1. Introduction
R
R
Nu
R
2
1
O
OH
O
1
6
[O]
Nu
Arene compounds are attractive starting materials for the
rapid preparation of synthetically useful cyclohexadiene
derivatives by various methods that are essentially based
on oxidative, reductive, and transition metal-mediated dear-
omatization approaches.1 Available in a variety of substitu-
tion patterns, arenes can be chosen to access selectively a
particular type of cyclohexadiene derivatives by using the
chemical method that is most appropriate to their reactivity
and functionality. For example, electron-rich arenes gener-
ally express a reactivity well suited for dearomatization by
oxidative means and, among this class of arenes, phenols
are ideally functionalized to furnish cyclohexadienones.2
Moreover, phenols of type 1 bearing electron-releasing sub-
stituents at their ortho-position(s) can be easily oxidized into
electron-deficient intermediates of type I, or equivalents, and
hence be regioselectively dearomatized into cyclohexa-2,4-
dienones of type 2 in the presence of a suitable nucleophile
(Scheme 1).
4
2
Z
Z
Z
1
2
I
R = an alkyl, aryl or alkoxy group
Z = various substituents
orthoquinols:
Nu = an oxygen-based nucleophile
R = a carbon-based group
orthoquinone monoketals:
Nu = R = oxygen-based groups
Scheme 1.
During the last eight years, we have investigated the chem-
istry of such cyclohexa-2,4-dienones bearing one or two
oxygen-based substituent(s) at their 6-position.3 These
compounds are referred to as orthoquinols or orthoquinone
monoketals. What makes these compounds useful building
blocks for organic synthesis is their oxygenated six-mem-
bered ring structure composed of a five-carbon conjugated
dienone unit linked to a sixth carbon center that is tetra-
hedral, quaternary, and possibly chiral (Scheme 1). When
considering such structural features and the large number
of possible transformations that can be envisaged on these
compounds in regio- and stereoselective manners, one can
easily understand why reports on their chemistry are so
abundant in the literature.3,4 Despite this plethora of infor-
mation already available, orthoquinols or orthoquinone
monoketals continue to stimulate the interest of synthetic
Keywords: Phenol dearomatization; Orthoquinone monoketals; Orthoqui-
nols; Cyclohexa-2,4-dienones; Cycloaddition; Diels–Alder reaction; Bis-
pericyclism; Anodic oxidation; Hypervalent iodine.
* Corresponding authors. Tel.: +33 540 00 30 10; fax: +33 540 00 22 15;
e-mail: s.quideau@iecb.u-bordeaux.fr
y
ꢀ
Present address: Laboratoire de Pharmacochimie, 146 rue Leo Saignat,
33076 Bordeaux Cedex, France.
0040–4020/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2007.03.035

6494
J. Gagnepain et al. / Tetrahedron 63 (2007) 6493–6505
organic chemists, as evidenced by numerous reports recently
published on the applications of their chemistry to the syn-
thesis of natural products.5 Indeed, many natural products of
various biosynthetic origins do feature orthoquinol units, and
many others can be made by taking advantage of their reac-
tivity in synthetic routes that often constitute biomimetic
approaches to these natural products.3a,c,4a,5j
a large halogen atom (i.e., bromine or iodine) at the car-
bon-4 position,8 or an acetoxy group at the tetrahedral car-
bon-6 position9 efficiently reduces its propensity toward
[4+2] cyclodimerization and hence enables the exploitation
of other facets of the reactivity of these cyclohexa-2,4-dien-
one derivatives. However, it is their Diels–Alder cyclodime-
rization that has been the topic of most studies over the years,
since Wessely’s pioneering work 50 years ago.9 The reasons
of this continuous interest in the investigation of this reaction
are its extraordinary level of selectivity and its probable im-
plication in the construction of several natural products. A
series of examples of such natural products (3a–f) that can
potentially be made in one chemical step from dimerizing
orthoquinol intermediates (i.e., 2a–f), which can themselves
be, respectively, derived from the phenolic precursors
1a–f by hydroxylative phenol dearomatization (HPD), are
shown in Scheme 2. There is little doubt that this sequence
of events corresponds to the final steps by which these
terpenes and vertinoid polyketides are biosynthetically
produced.10
The most characteristic aspect of the chemical reactivity of
orthoquinols and orthoquinone monoketals is unarguably
their ability to dimerize easily through [4+2] cycloaddition
reactions. This propensity toward Diels–Alder processes is
evidently due to the reactivity of their dienone unit that
can behave either as a diene or as a dienophile. This dimeri-
zation often occurs spontaneously at ambient temperature,
and can only be blocked, or at least retarded, if the cyclo-
hexa-2,4-dienone core of the monomer bears some substitu-
ents at some specific positions. For example, the presence of
a small alkyl or alkoxy group at its carbon-5 position,6 a large
electron-releasing group at the carbon-2 or -4 position,7
HO
HO
5
O
O
OH
5'
[4+2]
HPD
5
4
2'
4
OH
carvacrol (1a)
O
(+)-3a
2a
BzO
HO
BzO
2'
4
O
O
OH
O
OH
5'
BzO
5
OH
2b
2-hydroxybenzyl
benzoate (1b)
OBz
(–)-grandifloracin (3b)
HO
O
OH
HO
O
OH
O
2c
(–)-hydroxycuparene (1c)
(+)-aquaticol (3c)
HO
HO
O
OH
O
OH
O
(+)-bacchopetiolone (3d)
2d
1-hydroxy- -curcumene (1d)
O
OH
OH
HO
O
OH
O
2e
ferruginol (1e)
(+)-maytenone (3e)
O
HO
O
HO
OH
O
OH
O
O
OH
O
OH
O
OH
HO
sorbicillinol (2f)
sorbicillin (1f)
(+)-bisorbicillinol (3f)
Scheme 2. Examples of natural [4+2] cyclodimers of orthoquinols; HPD¼hydroxylative phenol dearomatization.

J. Gagnepain et al. / Tetrahedron 63 (2007) 6493–6505
6495
These constructions of natural products by cyclodimeriza-
2. Results and discussion
tion of orthoquinols would follow an endo-selective Diels–
Alder process during which the orthoquinol acting as the
dienophile would exclusively react through its D-4,5 bond.
To the best of our knowledge, no natural dimer resulting
from an exo-process and/or from an alternative participation
of the D-2,3 bond of the dienophilic orthoquinol has been
isolated. These selectivities already provide food for
thought, but the most intriguing aspects of these [4+2]
cycloadditions remain the facts that they express an even
more specific regioselectivity and an extraordinary level of
double diastereofacial selectivity. Indeed, these natural
cyclodimers result from a back-to-back combination during
which the orthoquinol monomers connect to each other
by joining their carbon-5 atoms and approach one another
with their carbon-6 hydroxy groups oriented towards
each other. Enzymatic control is certainly not at play in
these transformations, for such cyclodimerizations have
been performed in the course of the synthesis of several
natural products with the same selectivities.5k,10b,11,12 Nu-
merous model compound studies have been undertaken dur-
ing the last 50 years in the aim of providing a sound rationale
for the unique site-specific regioselectivity and the dia-
stereofacial selectivity always observed in these reactions,
but these investigations felt short of delivering a general
explanation.7,13,14
The first set of experiments that stirred our interest in the
Diels–Alder dimerization of orthoquinone monoketals and
orthoquinol variants had to do with the oxidative dearomati-
zation of phenol 1g using l3-iodane reagents.15 Depending
upon the reaction conditions used, two different products
were obtained. On one hand, using [bis(trifluoroacetoxy)iodo]-
benzene (BTI) in a MeOH–CH3CN (2:1) solvent mixture
according to the method originally conceived by Tamura and
Kita to access paraquinone monoketals,16 the only observed
product, isolated in 55% yield, was the cyclodimer 3g
(Scheme 3 and Table 1).15 This dimer resulted from a spon-
taneous [4+2] endo-cyclodimerization of the orthoquinone
monoketal 2g that was generated in situ by the BTI-mediated
oxidative methoxylation of 1g. On the other hand, using
(diacetoxyiodo)benzene (DIB) in a CH2Cl2–AcOH (3:1)
solvent mixture according to our own method,17 the only
observed product was the orthoquinol acetate 2g0. This or-
thoquinol resulted from the DIB-mediated oxidative acetoxy-
lation of 1g and was stable enough to be isolated as such in
95% yield (Scheme 3).15 These initial observations led us to
reason why the cyclodimerization was involving bond for-
mations during which the carbon-5 atoms of each orthoqui-
nol monomer 2g get exclusively connected to each other.
Furthermore, the fact that the substitution of a methoxy
group by an acetoxy group on the carbon-6 ketal function
of 2g was sufficient to block so efficiently the cyclodimeri-
zation process raised additional questions. This acetoxy
group blocking effect is well known,9,14a,18 but it still awaits
an explanation. Liao and co-workers suggested that one of
the two carbon-6 methoxy groups of dienophilic orthoqui-
none dimethyl ketals such as 2g participates, thanks to its
oxygen lone pairs, in secondary orbital interactions (SOIs)
in the endo-selective transition states (TSs), with the conse-
quence of enhancing the propensity of such monoketals
to self-dimerize.7 A downward modulation of this orbital
control in analogues such as 2g0 bearing instead an
Our work on oxidative dearomatization of phenols into
orthoquinone monoketals and orthoquinol variants led us
to revisit these [4+2] cyclodimerizations, and we recently
disclosed our own thoughts on these selectivity issues in
the context of the total synthesis of (+)-aquaticol (3c) from
(ꢀ)-hydroxycuparene (1c) (Scheme 2).11 We wish to de-
scribe herein a comprehensive view of the experimental
and theoretical results that we gathered to delineate further
the different reactivity and structural features controlling
the outcome of these Diels–Alder dimerizations in a more
general context.
R
R
R
O
MeO OAc
O
MeO
MeO6
*
O
O
[4+2]
*
O
R
X
O
OAc
X
OMe
OAc
OMe
AcO
O
SOI ?
DIB
OH
CO2Me
3g'
OMe
2g' 95%
2
2g'
BTI
endo-selective
approach
R
CO2Me
R
2'
R
4
O
1g
MeO OMe
O
4
MeO
O
MeO
5
R
[4+2]
O
MeO
2'
O
5'
5
SOI !
OMe
OMe
MeO
MeO
5'
CO2Me
3g 55%
2
2g
OMe
2g
O
OMe
OMe
R = CO2Me
OMe
R
MeO
O
5
[4+2]
2'
5
O
X
4
R
MeO
2'
O
5'
R
4
R
MeO
3g''
OMe
2
2g
5'
OMe
Scheme 3.

6496
J. Gagnepain et al. / Tetrahedron 63 (2007) 6493–6505
Table 1. [4+2] Cyclodimerization of orthoquinone monoketals and intra-
molecular [2+2] photoannelation of their [4+2] dimers
resulted from the same site selectivity than that observed for
3g (Scheme 3), as well as for all of so far reported cases
of orthoquinone monoketal and orthoquinol dimerizations,
including those leading to the natural products shown in
Scheme 2. This selectivity of connection by which the two
monomers join their two C5 atoms together and the car-
bon-4 atom of the dienophilic partner to the carbon-2 atom
of the dienic one would thus be independent of the nature
of the substituents on the cyclohexadienone core of the sys-
tem, as long as these substituents do not block the dimeriza-
tion process (vide supra). However, this connectivity first
appears counterintuitive when considering the probable po-
larization of the cyclohexa-2,4-dienone system. The more
electron-deficient carbon-5 atom at the end of the dienone
unit could be expected to join with the more electron-rich
carbon-2 atom adjacent to the carbonyl function. These
simple-minded considerations evidently do not allow any
prediction of site selectivity in these Diels–Alder reactions.
Cyclodimers resulting from such a combination, such as
3g00 from 2g (Scheme 3), have never been observed. As we
previously discussed,13 calculations of atomic coefficients
of either dienic or dienophilic HOMOs and LUMOs that
have been performed on unperturbed orthoquinone monoke-
tal reactants by using semiempirical (PM3) or ab initio RHF
(3-21G*) procedures are no better to provide any clear-cut
rationale. Suggestions were made on the possibility of hav-
ing SOIs controlling these cyclodimerizations at the transi-
tion state level,7,13 and this possibility is in fact backed-up
by an important feature of the inherent reactivity of these
cyclodimers. They readily undergo intramolecular [2+2]
cycloaddition reactions upon irradiation to furnish C2-sym-
metric pentacyclododecanedione derivatives.22 Cyclodimers
3g–l were indeed converted into such cage compounds 4g–l
in high yields by irradiating them at room temperature using
a medium-pressure mercury lamp (Table 1).13
Orthoquinone
monoketala
[4+2] Cyclodimer
[2+2] Cageb
R
R
MeO OMe
3' R
3
3' R
4'
2
2
4'
O
3
5
O
O
O
O
MeO
4
5'
MeO
5
OMe
OMe
MeO
MeO
R
MeO
OMe
2g:
4g 55%
4h 78%
4i 96%
4j 94%
3g
R = CO2Me
55%
78%
92%
100%
2h:
3h
3i
R = CH2OTBDPS
2i:
2j:
R = H
R = Br
3j
R
O
R
R
R
O
5
O
6
*
O
O
O
O
O
O
O
O
O
5'
5 O
O
O
O
O
R
O
O
O
4k
3k
3l
100%
100%
78%
78%
2k:
2l:
2m:
R = H
R = Pr
R = Br
4l
83%
3m
—
t-Bu
R
Br
O
O
O
O
O
O
O
—
R
Br
O
t-Bu
O
Br
t-Bu
2n
(–)-3n 54%
a
These monoketals were generated as racemates, except 2n.
Compounds 3m/n were not submitted to irradiation.
b
electron-withdrawing acetoxy group at the same position
could perhaps be put forward to propose a beginning of an
explanation of their resistance toward dimerization (Scheme
3). What is clearly known from work carried out on many or-
thoquinol acetates is that, once the acetate is hydrolyzed, the
resulting 6-hydroxylated product then readily self-dimeri-
zes.9,10b,14a,18 This experimental approach has, inter alia,
been followed by Corey and Nicolaou in their biomimetic
synthesis of bisorbicillinoid natural products, such as bisor-
bicillinol (3f), from sorbicillinol (2f) (Scheme 2).5k,12,19 This
recovery of the capability of dimerizing spontaneously when
going from 6-acetoxylated to 6-hydroxylated compounds is
in line with the aforementioned modulation of SOIs, to
which could perhaps be added a reduced steric effect.20
But then another question comes to mind! Why would the
acetoxy group of the dienophilic component of the Diels–
Alder cyclodimerization systematically orient itself toward
the dienic component of the reaction? Needless to say that
we were still intrigued by these reactivity differences, site
selectivity, and apparent facial recognition expressed by
these orthoquinone monoketals as well as their orthoquinol
variants in their [4+2] transformations.
This facility by which these [4+2] cyclodimers further react
upon irradiation is a key element of the understanding of the
factors controlling the manner by which their monomers
combine with each other during the [4+2] cycloaddition
process. Thus, it constitutes a sound chemical evidence of
the endo-selectivity of the process,22 since exo-cyclodimers
could not have the orbitals of the two reacting carbon–
carbon double bonds properly lined up for the [2+2] cyclo-
butanation. One can thus argue with confidence that the
endo-mode of [4+2] cycloaddition is under the influence
of p-orbital interactions taking place between the two C3’s
of each monomer, and between C2 of one monomer and
C4 of the other monomer (see Table 1). These SOIs are
then allowed to engage in intramolecular bond formations
during the [2+2] photoannelation process. Consequently,
they certainly play an important role in the kinetic selection
of the [4+2] TSs, but their establishment still does not ex-
plain the site-specific regioselectivity of the reaction. The
exclusive participation of the D-4,5 bond as a dienophilic
component can be simply explained by noting the loss of
enone conjugation energy that would impede the reaction
TS if the D-2,3 bond was instead involved. However, it is
still not clear why endo-dimers, such as 3g00 for which C4
and C5 of the preferred dienophilic double bond would
instead, respectively, connect with C5 and C2 of the dienic
partner, are not observed. The p-orbitals of their remaining
carbon–carbon double bonds should still be able, at least
2.1. Site selectivity
We produced several analogues of 3g in the course of our in-
vestigation on oxidative phenol dearomatization methods
based on the use of hypervalent iodine reagent as well as
on anodic oxidation.13,21 A selection of these endo-cyclo-
dimers 3h–n is displayed in Table 1. They were obtained in
good to excellent isolated yields, and all of them exclusively

J. Gagnepain et al. / Tetrahedron 63 (2007) 6493–6505
6497
Table 2. Activation energies in kcal/mola
out using different levels of theory (see Section 4). Activa-
tion energies of these different reactions are shown in
Table 2. The most striking geometrical characteristic of TSs
leading to observed dimers is that they all display a C2-axis
of symmetry. This can be viewed, for example, on the repre-
sentation of the TS leading to 3i from 2i (Fig. 1). Interest-
ingly, the alternative combination that would result by
connecting instead C5 of one monomer with C2 and C4
with C5 of the other monomer led to a TS (Fig. 1) that is
about 5 kcal/mol higher in energy than the TS leading to
3i, whatever the level of calculation is (Table 2). This
alternative combination is obtained by flipping over one
monomer and leads to a TS that is not C2-symmetric any-
more. The loss of the C2-symmetry is also at the origin of
significant changes in bond length alternation between the
reactive centers of the two monomers. However, the bond
length values (Fig. 1) do not clearly evidence a better effi-
ciency of the overall p-conjugation in the cyclohexadienone
moiety of TS-3i as compared to that of TS-3i0. Nevertheless,
the stabilization of the C2-symmetric structure TS-3i likely
originates from a better overlap of the p electronic clouds
of the two partners, hence maximizing SOIs (vide infra).
Geometry
6-31G(d)
6-31G(d)
RR/RSb
6-31G(d)
6-311+G(d,p)
RR/RSb
6-31G(d)
Wave function
Reaction
MP2/6-31G(d)
RR/RSb
2g/3g
2g0/3g0
2i/3i
27.05
31.38
2.97
41.12/44.44
26.76
32.19
26.18/25.64
38.93/42.46
26.22/30.33
38.93/51.22
23.67/32.00
45.30/48.96
30.70
35.85
28.96/28.65
43.21/46.96
29.23/33.44
43.82/55.16
28.54/35.66
20.88/22.87
8.03
13.25
7.35/7.14
22.92/25.95
5.10/9.98
20.72/34.01
5.59/14.85
2i0/3i0
2k/3k
2o/3o
2p/3p
2q/3q
2r/3r
a
Calculations were performed using the B3LYP functional, unless other-
wise noted.
For reactions involving chiral monomers, TSs of both RR* and RS*
combinations were calculated.
b
O
MeO OAc
Me OAc
MeO OH
*
O
O
O
O
O
*
*
*
2p
2o
2q
2r
We thought of finding an explanation for this preference of
bond formation between the two C5 centers of each mono-
mer by looking at the atomic coefficients of the perturbed
HOMO and LUMO at the TS level, but this analysis, like
the one performed on the unperturbed HOMO and LUMO
of the orthoquinone monoketal reactant 2i, did not unveil any
diagnostic information. A comparative examination of the
above asynchronous two TSs reveals that TS-3i is globally
much more compact than TS-3i0, which rather adopts a V-
shape (Fig. 1). The distances between the atoms involved in
to some extent, to interact with each other (e.g., see 3g00 vs 3g
in Scheme 3).
We thus decided to perform a search of a selection of the TSs
involved in these [4+2] cycloadditions, including those lead-
ing to dimers 3g, 3i, and 3k. In order to gain further insight
into these reactions, TSs of dimers not observed experimen-
tally, such as 3g0, 3g00 and those resulting from a different
facial selectivity (vide infra), were also calculated, as well
as those of other cyclodimerization systems derived from
model monomers 2o–r. These calculations were carried
˚
0
the first forming bond are about the same (dC5 –C5¼2.036 A
˚
0
00
vs dC5 –C2¼1.986 A), but the distance between C4 and C5 in
˚
TS-3i is rather long (i.e., 3.574 A) for enabling efficient
C2-symmetric TS-3i
C1-symmetric TS-3i'
MeO
5
O
OMe
O
MeO
MeO
2 x 2i
MeO
MeO
2
5'
2'
5'
2'
OMe
OMe
4'
4'
5
2
vs.
4
2 x 2i
O
O
4
5'
d'
5'
1.986
4'
2.036
a
2
a = 1.500
b = 1.413
c = 1.437
d = 1.351
d'
c'
5
3.574
4'
a
3.145
b
c'
d
a = a' = 1.459
b = b' = 1.372
c = c' = 1.418
d = d' = 1.402
3'
5
c
d
c
2
b
3'
a' = 1.459
b' = 1.390
c' = 1.396
d' = 1.413
a'
b'
3
b'
a'
4
2.691
3.145
4
2'
3
2'
OMe
4'
O
MeO
5'
2'
O
MeO
4
O
2
O
MeO
5'
5
OMe
MeO
5
MeO
3i
3i' not observed
MeO
0
˚
Figure 1. B3LYP/6-31G(d) TS structures with interatomic distances (A) of endo-dimers 3i and 3i in which C4 and C5 of the preferred dienophilic double bond
connect with C2 and C5 (TS-3i) and with C5 and C2 (TS-3i0) of the dienic partner.

6498
J. Gagnepain et al. / Tetrahedron 63 (2007) 6493–6505
overlap of the corresponding p-orbitals. The second shortest
2.2. The ‘acetate effect’
˚
0
distance in this TS is between C2 and C3 (i.e., 2.691 A), but
such a connection would imply the participation of the D-2,3
of the monomer acting as a dienophile and a consequent loss
of enone conjugation energy (vide supra). It then remains to
explain why TS-3i is more stable than TS-3i0 and to identify
plausible factors capable of thus cementing TS-3i in its more
compact C2-symmetric geometry. A relevant explanation
has been put forward early on to differentiate dimers of
cyclohexa-2,4-dienones on the basis of the principle of the
lowest dipole moment.23 Dimers such as 3i display a lower
dipole moment than those analogous to 3i0. This argument
holds at the TS level (TS-3i02.35 D vs TS-3i003.67 D),
but the C2-symmetric structure of TS-3i versus the C1-sym-
metric V-shaped structure of TS-3i0 allows another explana-
tion based on the resulting better alignment of the TS-3i
interacting p-orbitals already invoked to rationalize the
endo-selectivity of the process (vide supra).
Observations on the resistance of C6-acetoxylated deriva-
tives of orthoquinone monoketals and orthoquinols toward
cyclodimerization led us to analyze this phenomenon in
greater details. As mentioned above, a SOI-based stereo-
electronic explanation of this blocking effect would imply
that the faces of the monomers orienting themselves toward
each other are those bearing the acetoxy group (e.g., see
2g0 in Scheme 3). We will come back to this facial selectivity
issue in the next section, but first, we wanted to exploit this
‘acetate effect’ in cyclic variants of the acyl/methyl C6-ketal
function of the monomers. We thought that we could prepare
non-dimerizing O-spirolactonic orthoquinone monoketals
with the possibility of accessing them in a stereoselective
manner by equipping the carbon center a to the lactonic car-
bonyl group with different substituents.21b This enterprise
failed simply because the O-spirolactonic cyclohexa-2,4-di-
enone system does in fact readily dimerize, as observed for
the conversion of 2k–m into 3k–m (Table 1). The reasons of
such a difference of behavior between these cyclic and acy-
clic monoketals were not obvious, but computational model-
ing of these systems gave us some clues. In agreement with
our experimental observations, calculations performed on
the dimerization of 2k and 2o indicate a much lower energy
barrier for the dimerization of the O-spirolactone 2k (Table
2). However, a closer examination of the corresponding TS
structures does not reveal any significant difference neither
in the distances between the reactive centers of the two part-
ners nor in the orientations of the C6–O bond of the carboxy
units of the two systems (Fig. 3).
0
These interactions between C3 and C3 and between C2 and
0
0
C4 or between C4 and C2, respectively, correspond to the
Woodward–Hoffmann (WH) SOI, classically invoked to ex-
plain the preference of endo- over exo-TSs in Diels–Alder
reactions, and to the Salem–Houk (SH) SOI.24 In fact, in
all C2-symmetric TSs leading to observed dimers, such as
TS-3i, it is not possible to differentiate the diene from the
0
0
dienophile, the C2 /C4 and C4 /C2 orbitals interactions are
equivalent (Fig. 1). One will lead to the second bond forma-
tion in the Diels–Alder process, the other will play its stabi-
lizing role as a SH SOI, or vice versa (Figs. 1 and 2). In other
words, the [4+2] and the [2+4] cycloaddition reaction path-
ways are equivalent. Such TSs are said to be bispericyclic, as
recently discussed by Caramella for butadiene and cyclopen-
tadiene models.25 All known dimerizations of orthoquinone
monoketals and orthoquinols, including those leading to
natural products (Scheme 2), follow this theoretical bisperi-
cyclism principle and their respective TSs are all further
stabilized by this combination of WH and SH SOIs, which
are allowed to engage in [2+2] bond-forming events upon ir-
radiation to give rise to the observed cage compounds (see
Tables 1 and 3). Remarkably, TSs of [4+2]/[2+4] cyclodime-
rizations of orthoquinone monoketals or orthoquinols can
be viewed as the ‘fingerprints’ of the [2+2] products of the
photoannelation of the Diels–Alder products to which they
lead to (Fig. 2).
It is the comparison of the geometrical structures of the un-
perturbed reactants 2k and 2o that turned out to be more in-
formative (Fig. 3). In the acyclic ketal 2o, the carbonyl group
of the acetate points towards the center of the cyclohexadi-
enone core. A systematic conformational analysis confirmed
that this acetate orientation is quite optimal, and corresponds
to a stabilization of approximately 5 kcal/mol with respect to
structures in which the acetate adopts an external position.
This important stabilization of acetate reactants such as
2o, which cannot exist for geometrically constrained cyclic
ketals such as the O-spirolactone 2k, could thus be the origin
of their higher activation energies and, hence, it constitutes
a valid explanation of the blocking ‘acetate effect’ always
TS
R
R
R'
R'
R
R'
5
4
5'
2'
O
2
[4+2] or [2+4]
O
4'
3'
SH
SOIs
e.g., in TS-3i, R, R' = OMe
dC4'-C2 = dC2'-C4 = 3.145 Å
WH
O
2
3
d
C3'-C3 = 3.165 Å
R, R' = alkoxy, acyloxy, alkyl,
aryl or hydroxy groups
C2
R
R
R'
R'
R
R'
R
R'
5
5'
2'
[2+2]
2
O
O
4'
4
O
O
3
4
3
3'
Figure 2.

J. Gagnepain et al. / Tetrahedron 63 (2007) 6493–6505
6499
C2-symmetric TS-3k
C2-symmetric TS-3o
O
O
O
O
O
O
O
O
5
O
5
O
5'
2'
5'
vs.
O
O
O
O
4'
4'
2
2
2 x 2k
2 x 2o
4
4
2'
O
O
-44.9°
-45.7°
-163.6°
-162.3°
2.103
2.156
5'
4'
5
5'
4'
5
4
4
3
3.158
3.158
3.162
3.162
3'
2'
3'
3
2
2'
2
2k
2o
˚
Figure 3. B3LYP/6-31G(d) structures of TS-3k (top left) and TS-3o (top right) with dihedral angles and interatomic distances (A), and of their respective
unperturbed reactants 2k (bottom left) and 2o (bottom right).
1
observed in attempted dimerizations of orthoquinol and or-
thoquinone ketal acetates, such as 2g0 (Scheme 3), under ki-
netic conditions.9,14a,18,20 This ‘acetate effect’ was similarly
predicted by comparing the activation energies related to the
dimerization of the model orthoquinol acetates 2p and 2q
(Table 2).
the limits of sensibility of both H and 13C NMR analyses,
that the racemate 3k was the only dimer formed.
The orthoquinols 2a and 2s/t that we generated also sponta-
neously dimerized to furnish exclusively the endo-products
3a and 3s/t according to the same double diastereofacial se-
lectivity (Scheme 2 and Table 3). The structures of 3a and 3s
(and 3s0, vide infra) were again further confirmed by per-
forming [2+2] photoannelation into the cage compounds
4a, 4s, and 4s0 (see Section 5). The dimer 3t was refractory
to photoannelation, probably because of the steric demand
of the two tert-butyl groups. Interestingly, the mixture of
racemic orthoquinols 2a and 2s that we generated by a
l5-iodane-mediated hydroxylative phenol dearomatization
(HPD) reaction of a mixture of the phenolic terpenoids car-
vacrol (1a) and thymol led to a racemic mixture of 3a, 3s,
and 3s0 (see Section 5). This latter compound, isolated in
a yield of 9%, resulted from [4+2] cycloaddition between
2a behaving as a diene and 2s behaving as a dienophile. For-
mation of the alternative compound that could have resulted
from a switch of the dienic/dienophilic behavior of 2a and
2s was not observed, but that of 3s0 again results from a
combination of 2a and 2s during which they approach one
another with their C6-hydroxy groups oriented towards
each other.
2.3. Facial selectivity
The most striking aspect of the manner by which O-spirolac-
tonic orthoquinone monoketals 2k–m dimerize is that they
do so by orienting their lactonic C6–O bonds towards each
other (Table 1 and Fig. 3). The chirality of the ketal C6-cen-
ter thus strongly influences the outcome of the dimerization.
Not only the two faces of the dienophile are differentiated,
but also those of the monomer behaving as a diene. Starting
from racemic O-spirolactone dienones such as 2k–m, only
two enantiomers of eight possible endo-cyclodimers are
made. The two starting ketal enantiomers recognize each
other and do not cross-react. All of the known analogous di-
mers generated under kinetic control also resulted only from
chiral orthoquinone monoketals or orthoquinol variants
displaying the same configuration at their stereogenic C6-
center. This enantiospecific pairing was confirmed by our
calculations of the [4+2] TS structures of the dimerization
of 2g0 and 2o–r (Table 2). The calculated RR combinations
were systematically lower in energy than their RS alterna-
tives. The only exception was for the dimerization of 2k,
the TSs calculation of which indicated similar activation en-
ergies for both combinations (Table 2). This is in sharp con-
trast with our experimental observation, which indicated, to
To the best of our knowledge, the only reported case of di-
merization showing a deviation from this double diastereo-
facial selection concerns the racemate of an orthoquinol
acetate of type 2q (Table 2), which, when forced to dimerize
upon heating, gave rise to a cyclodimer derived from a

6500
J. Gagnepain et al. / Tetrahedron 63 (2007) 6493–6505
Table 3. [4+2] Cyclodimerization of orthoquinols and intramolecular [2+2]
photoannelation of their [4+2] dimers
R'
O
R
R
5
5'
6
R'
Orthoquinola
[4+2] Cyclodimer
[2+2] Cage
6'
HO
O
O
*
Figure 4. Proposed contribution to the stabilization of cyclohexa-2,4-di-
enone cyclodimerization TSs by double ‘Cieplak–Fallis’ hyperconjugation;
R, R0¼carbon-, oxygen-, sulfur-, or halogen-based groups with R0 involved
in the most electron-donating s-bond.
O
O
O
O
O
HO
HO
HO
2a
HO
3a 48%
4a 81%
relying on (i) a Cieplak-type hyperconjugative stabilization
of the reaction TSs by the most electron-donating s-bond of
the two C6-substituents connecting bonds being antiperipla-
nar to the s*-orbital of the incipient bond with the dieno-
phile, as first proposed by Macaulay and Fallis,30 or on (ii)
the orbital mixing rule, which stipulates that, as a result of
an out-of-phase combination of the p HOMO of the diene
with the n-orbital of the oxygen atom at C6, the out-of-phase
mixing of this n-orbital with the s-orbitals of the diene reac-
tion centers should favor an approach of the dienophile on
the face bearing the C6-oxygen atom.31 These suggestions
are of course still open to further discussions, but they are
unarguably fed by considerations that are much more ad-
vanced than those simply based on steric effects. These
effects have generally been invoked to explain the double
diastereofacial selectivity observed in the dimerization of
orthoquinols (i.e., Me and OH groups at C6) and their
spirooxirane versions.14,32 However, natural bond orbital
analyses we carried out on TSs of the conversion of ortho-
quinol 2c into (+)-aquaticol (3c)11 (Scheme 2) leads us to
favor an explanation based on a double ‘Cieplak–Fallis’
hyperconjugation as the determining factor in this stereo-
selectivity (Fig. 4). In all stereochemically defined cases
of orthoquinone monoketal and orthoquinol kinetically-
controlled dimerizations, the most electron-donating s-
bond of the two allylic C6-substituents connecting bonds is
ideally oriented to be quasi antiperiplanar to the s*-orbital
HO
O
*
O
i-Pr
O
i-Pr
O
HO
HO
HO
HO
4s 90%
2s
3s 50%
O
O
O
O
2a+2s
HO
HO
HO
HO
3a
21%
12%
4s' 64%
3s' 9%
+
3s
Ph
HO
O
R
O
Ph
—
O
Ph
HO
HO
(+)-3t 66%
2t
a
These orthoquinols were generated as racemates, except 2t.
cross-reaction between its two enantiomers.14a,18 The enantio-
mer that behaved as the dienophile in this cyclodimerization
approached its partner from its face bearing the methyl group.
Apparently, the resulting dimer is thermodynamically more
stable, probably because the bulkier acetoxy group is then
oriented anti to the ethylene bridge of the bicyclo[2.2.2]oc-
tene unit. However, the TS leading to this alternative dimer
must be less stable, since its formation is never observed to
any significant extent when the reaction is maintained under
kinetic control. The fact that the enantiomer behaving as the
diene conserves the orientation of its bulkier C6-acetoxy
group toward the Diels–Alder reaction centers has been at-
tributed to an electrostatic preference for orientation of the
most polarizable substituent at C6 (i.e., the acetoxy group)
anti to the ethylene bridge.14a Such a sterically-independent
preference for anti- over syn-products has been rationalized,
for Diels–Alder reactions involving cyclopentadiene deriva-
tives, in terms of attractive interactions between the diene
and the dienophile due to dipole–dipole, dipole-induced di-
pole, and London dispersion forces.26 The same anti-prefer-
ence is also observed when orthoquinols, their acetates or
their alkyl ethers react with dienophiles other than themsel-
ves.6b,20,27 This issue of p-facial selectivity of Diels–Alder
cycloadditions involving dienes bearing heteroatomic allylic
substituents has fueled numerous studies over the years;
electronic and steric forces or a combination of both have
all been proposed as the dominating factors depending on
the type of compounds under investigation.28 More recently,
Liao revisited this issue of p-facial selectivity in Diels–
Alder reactions of orthoquinol methyl ethers with various
dienophiles,29 and outlined two possible explanations
0
of the C5–C5 incipient bond in TS structures (e.g., see TS-3k
and TS-3o in Fig. 3). This is not only the case with carbon/
oxygen-based C6-substituents, but also with carbon/
halogen- and sulfur/oxygen-based C6-substituents.33 The
C2-symmetry of all of the calculated TSs leading to products
under kinetic conditions merits one final comment, as it al-
lows hyperconjugation from each antiperiplanar C6-allylic
0
bond of each monomer to affect the same C5–C5 incipient
bond (Fig. 4). This is in contrast to the non-observed sites’
connection alternative, such as that in the C1-symmetric
TS-3i0, in which hyperconjugation from the same C6-allylic
bonds are at best spread over two different incipient bonds
0
0
(i.e., C5 –C2 and C4 –C5, see Fig. 1). Hence, we suggest
that these hyperconjugative effects can also play a determin-
ing role in controlling not only the p-facial selectivity but
also the regioselectivity of these cyclodimerizations, even
in the case of achiral orthoquinone monoketals, such as 2i,
as well as other cyclohexa-2,4-dienones.
3. Conclusion
The interpretation of the experimental and theoretical results
described therein constitutes a comprehensive understand-
ing of the factors that control the regio- and stereoselectiv-
ities at play in kinetically-controlled [4+2] dimerizations

J. Gagnepain et al. / Tetrahedron 63 (2007) 6493–6505
6501
of orthoquinonoid cyclohexa-2,4-dienones. The dimeriza-
tion blocking ‘acetate effect’ has been rationalized, and
sterically-independent explanations of both the site-specific
regioselectivity and double diastereo-p-facial selectivity of
the process are proposed on the basis of the establishment
of asynchronous, bispericyclic, and C2-symmetric reaction
transition states. These transition states would be held to-
gether through a combination of both Woodward–Hoffmann
and Salem–Houk secondary orbital interactions. Their re-
sulting geometry enables hyperconjugation between the
s-bond linking the antiperiplanar allylic C6-substituent of
each monomer and the first-formed bond. These stabilizing
hyperconjugative effects could thus also contribute to the
site-specific regioselectivity observed in the dimerization
of orthoquinonoid and other cyclohexa-2,4-dienones. In
the same line of thought, the double diastereo-p-facial selec-
tivity observed with chiral orthoquinonoid cyclohexa-2,4-di-
enones could then be due to a preference for Cieplak-type
hyperconjugative effects, since the s-bond linking the anti-
periplanar allylic C6-substituent is always the most electron-
donating bond. The various aspects of the hypothesis
discussed in this article obviously still need to be exposed
to other in-depth investigations and further theoretical re-
finements before being confirmed or not, but it is, at this
stage, unarguably attractive to admit that these aspects all
appear to converge toward a single all-embracing explana-
tion of the regio-, site-, endo-, and p-facial selectivities of
this [4+2] cyclodimerization process involved in the biosyn-
thetic elaboration of many natural products.
(SIBX),36 and l3-iodanes, i.e., [bis(trifluoroacetoxy)iodo]-
benzene (BTI) and (diacetoxyiodo)benzene (DIB), were
furnished by SIMAFEX and were used as received. Evapo-
rations were conducted under reduced pressure at tempera-
tures less than 30 ꢁC unless otherwise noted. Column
chromatography was carried out under positive pressure us-
ing 40–63 mm silica gel (Merck) and the indicated solvents.
Melting points were measured in open capillary tubes and
are uncorrected. NMR spectra of samples in the indicated
solvent were run at 250, 300 or 400 MHz and calibrated
using residual solvent as an internal standard. Carbon multi-
plicities were determined by either DEPT135 or J-Mod
experiments. Electron impact (EIMS, 50–70 eV), chemical
ionization (CIMS), electrospray (ESIMS), and liquid
secondary ion mass spectrometry (LSIMS) low- and/or
high-resolution (HRMS) mass spectrometric analyses were
obtained from either the mass spectrometry laboratory at
ꢀ
the Centre d’Etude Structurale et d’Analyse des Molecules
Organiques (CESAMO), Universite Bordeaux 1, or the Cen-
ꢀ
ꢀ
tre Regional de Mesures Physiques de l’Ouest (CRMPO),
Universite Rennes 1, France. Elemental analyses were
ꢀ
carried out at the Service Central d’Analyses du CNRS, Ver-
naison. The X-ray crystallographic analyses were performed
ꢀ
at the Institut Europeen de Chimie et Biologie (IECB), and
X-ray diffraction data were collected on a crystal sealed
in a Lindemann-glass using Cu Ka radiation on a CAD4
diffractometer.
5.1.1. 3,3,10,10-Tetramethoxy-6,12-dimethoxycarbonyl-
tricyclo[6.2.2.02,7]dodeca-5,11-diene-4,9-dione (3g).15 To
a stirred solution of methyl 3-hydroxy-4-methoxybenzoate
(322 mg, 1.77 mmol) and K2CO3 (488 mg, 3.34 mmol) in
dry MeOH (10 mL) at ꢀ42 ꢁC was added a solution of
BTI (761 mg, 1.77 equiv) in CH3CN (5 mL). The reaction
mixture immediately became bright yellow, and after 2 h,
TLC monitoring [hexanes–EtOAc (1:1)] indicated complete
consumption of the starting material. The mixture was then
allowed to warm up to room temperature, and was kept under
stirring for 30 min, after which it was poured into water
(10 mL), extracted with Et2O (3ꢂ10 mL), washed with brine
(10 mL), dried over Na2SO4, filtered, and evaporated. The
resulting brown oil was purified by column chromatography,
eluting with hexanes–EtOAc (4:1/1:1), to furnish the
cyclodimer 3g as an amorphous yellowish solid. Further
crystallization from pentanes–CH2Cl2 afforded pure 3g as
amber needles (207 mg, 55%): mp 166–168 ꢁC; IR (KBr)
4. Computational
Geometry optimizations were carried out in vacuum using
the density functional theory (DFT) with the three-parameter
hybrid functional B3LYP34 and the 6-31G(d) basis set. Ther-
mal corrections were calculated from the unscaled harmonic
vibrational frequencies using standard temperature and pres-
sure conditions. Every transition state was characterized by
a single imaginary frequency in the diagonalized mass-
weighted Hessian matrix. In several significant cases, intrin-
sic reaction coordinates’ (IRC) calculations were performed
to determine an unambiguous path connecting transition
structures with reactants and products. Electronic energies
were further refined by using the larger 6-311+G(d,p) basis
set, as well as by using the second order-Møller–Plesset
(MP2) level of theory within the 6-31G(d) basis set. These
procedures are referred to as B3LYP/6-311+G(d,p)//B3LYP/
6-31G(d) and MP2/6-31G(d)//B3LYP/6-31G(d), respec-
tively. All calculations were performed using Gaussian03.35
1
1720 cmꢀ1; H NMR (CDCl3, 300 MHz) d 2.94 (s, 3H),
3.10 (s, 3H), 3.19 (br d, J¼6.9 Hz, 1H), 3.25 (dd, J¼1.2,
8.3 Hz, 1H), 3.30 (s, 3H), 3.36 (s, 3H), 3.56 (s, 3H), 3.73
(ddd, J¼1.3, 2.8, 8.3 Hz, 1H), 3.75 (s, 3H), 3.86 (dd, J¼
1.9, 2.8 Hz, 1H), 6.64 (d, J¼1.2 Hz, 1H), 6.96 (dd, J¼1.8,
6.9 Hz, 1H); 13C NMR (CDCl3, 75.5 MHz) d 200.1, 193.6,
164.7, 163.2, 144.2, 140.5, 132.6, 132.2, 98.4, 94.2, 52.7,
51.9, 51.5, 50.4, 50.1, 49.8, 49.0, 40.9, 38.2, 38.1; EIMS
m/z (rel intensity) 424 (M+, 8), 396 (64), 381 (14), 336
(17); Anal. Calcd for C20H24O10: C, 56.60; H, 5.70; O,
37.70. Found: C, 56.39; H, 5.83; CCDC 636463.
5. Experimental
5.1. General
All moisture and oxygen sensitive reactions were carried out
in flame-dried glassware under N2. Tetrahydrofuran (THF)
was purified immediately before use either by distillation
from sodium/benzophenone under N2, or by filtration
through alumina under N2. Acetonitrile (CH3CN), methanol
(MeOH), and 2,2,2-trifluoroethanol (CF3CH2OH) were used
as received. l5-Iodane, i.e., stabilized o-iodoxybenzoic acid
5.1.2. 7R,11-Dibromo-3R,10R-bis(20-tert-butyl-10,40-di-
oxaspirocyclopentane)-tricyclo[6.2.2.02,7]dodeca-5,11-
diene-4,9-dione (3n). A stirred solution of (+)-2-[(R)-
2-hydroxy-3,3-dimethylbutoxy]-4-bromophenol
(50 mg,
0.173 mmol) in CF3CH2OH (6 mL) was cooled at ꢀ35 ꢁC,

6502
J. Gagnepain et al. / Tetrahedron 63 (2007) 6493–6505
and was treated dropwise, over 5 min, with a solution of DIB
(57 mg, 0.176 mmol) in CF3CH2OH (1 mL). The reaction
mixture immediately became pale yellow and then slowly
changed to yellow-pink. After 20 min, TLC monitoring
[cyclohexane–acetone (2:1)] indicated complete consump-
tion of the starting material. Powdered NaHCO3 was added
in one portion to the reaction mixture, which was kept under
stirring at ꢀ35 ꢁC for 15 min. After CF3CH2OH removal in
vacuo, the residue was taken up with CCl4 (3ꢂ15 mL), fil-
tered, and evaporated. Further drying under high vacuum al-
lowed complete removal of the iodobenzene by-product and
afforded an orange oily mixture of the corresponding ortho-
quinone monoketal 2n and its cyclodimer 3n. Proton NMR
monitoring of the mixture in a CDCl3 solution (ca. 0.3 M) in-
dicated that quasi complete [4+2] cyclodimerization of 2n
into 3n was achieved after 15 days at room temperature.
Subsequent purification by column chromatography, eluting
with cyclohexane–acetone (5:1), gave pure cyclodimer 3n as
a pale orange oil (27 mg, 54%): [a]2D0 ꢀ161.7 (c 0.35,
1H NMR (CDCl3, 300 MHz) d 0.92 (s, 9H), 0.95 (s, 9H),
3.45–3.49 (m, 2H), 3.66–3.98 (m, 4H), 4.29 (dd, J¼7.3,
14.2 Hz, 2H), 4.39 (br t, J¼6.7 Hz, 1H), 6.04 (d,
J¼10.2 Hz, 1H), 6.19 (dd, J¼2.4, 6.8 Hz, 1H), 6.49 (d,
J¼10.2 Hz, 1H); 13C NMR (CDCl3, 75.5 MHz) d 200.6,
189.6, 146.5, 127.9, 127.3, 125.1, 102.4, 100.5, 85.9, 85.5,
67.7, 66.2, 63.2, 55.8, 53.9, 53.8, 33.4, 32.8, 25.3, 25.2;
CIMS m/z (rel intensity) 594 (MNH+4, 39), 592 (MNH4+,
77), 590 (MNH+4, 39), 577 (MH+, 24), 575 (MH+, 46),
573 (MH+, 24), 467 (100); HRMS (ESI) calcd for
C24H30O679Br2Na 595.0307, found 595.0308.
one portion. The resulting suspension was stirred at room
temperature for 24 h, after which TFA (780 mL, 10 mmol)
was added and the mixture was further stirred for 12 h.
The reaction mixture was then diluted with CH2Cl2
(100 mL) and H2O (50 mL). An aqueous 1 M solution of
NaOH was added slowly and continuously with vigorous
shaking until all solid material dissolved. The aqueous phase
was extracted with CH2Cl2 (3ꢂ20 mL). The combined or-
ganic extracts were washed with aqueous 1 M NaOH
(40 mL), H2O (50 mL), and brine (2ꢂ50 mL), then shaken
vigorously with a saturated aqueous solution of Na2S2O4
(100 mL), washed again with brine (50 mL), dried over
Na2SO4, filtered, and evaporated at room temperature to
give a residue, which was then purified by column chroma-
tography.
5.2.1. 3,10-Dihydroxy-6,12-di-iso-propyl-3,10-dimethyl-
tricyclo[6.2.2.02,7]dodeca-5,11-diene-4,9-dione (3a).
SIBX-mediated oxidation of carvacrol (1a, 1.5 g, 1 mmol)
was carried out according to the general procedure to give,
after purification by column chromatography, eluting with
hexanes–Et2O (1:1), 2,3-dihydroxy-4-iso-propyl-1-methyl-
benzene (5) as a brown oil (398 mg, 24%) and the cyclo-
dimer 3a as a yellow solid (930 mg). The latter was
recrystallized twice at low temperature from hexanes–
CH2Cl2 to afford pure 3a as white needles (795 mg, 48%).
Pure 3a was also obtained as colorless prisms from slow
evaporation of CHCl3.
CHCl3); IR (NaCl) 1748, 1708, 1194, 1107, 1008 cmꢀ1
;
Compound 5:37 IR (NaCl) 3453, 2963, 1467 cmꢀ1; 1H NMR
(CDCl3, 300 MHz) d 1.26 (d, J¼6.8 Hz, 6H), 2.24 (s, 3H),
3.17 (sept, J¼6.8 Hz, 1H), 5.31 (br s, 2H), 6.69 (d,
J¼7.9 Hz, 1H), 6.73 (d, J¼8.1 Hz, 1H); 13C NMR (CDCl3,
75.5 MHz) d 141.5, 141.0, 132.5, 121.9, 121.3, 117.4,
27.1, 22.6, 15.3; EIMS m/z (rel intensity) 166 (M+, 92),
151 (100), 133 (56), 105 (60).
5.1.3. (3R,10R)-6,12-di-tert-Butyl-3,10-dihydroxy-3,10-
diphenyltricyclo[6.2.2.02,7]dodeca-5,11-diene-4,9-dione
(3t). To a stirred suspension of acidic ion-exchange resin
AmberliteÒ IR120-H+ (ca. 200 mg) in dry THF (4 mL)
was added dropwise a solution of 2R,9-di-tert-butyl-(6R)-
phenyl-1,4-dioxaspiro[4.(5R)]deca-7,9-dien-6-ol (69 mg,
Compound 3a: mp 136 ꢁC (lit.10b 138–139 ꢁC); IR (KBr)
0.202 mmol) in dry THF (2 mL), in the presence of 4 A mo-
3451, 2967, 1724, 1676, 1157 cmꢀ1 1H NMR (CDCl3,
;
˚
lecular sieves. After stirring overnight at room temperature,
TLC monitoring [hexanes–Et2O (4:1)] indicated complete
consumption of the starting material. The reaction mixture
was then filtered and evaporated to furnish a pale yellow
oily residue, which was purified by column chromatography,
eluting with hexanes–Et2O (4:1), to give the cyclodimer 3t
as a colorless syrup (32.4 mg, 66%): [a]2D0 +187.9 (c 0.78,
300 MHz) d 0.84 (d, J¼6.8 Hz, 3H), 0.88 (d, J¼6.6 Hz,
3H), 1.11 (d, J¼7.0 Hz, 3H), 1.12 (d, J¼6.6 Hz, 3H), 1.22
(s, 3H), 1.24 (s, 3H), 1.83 (sept, J¼6.8 Hz, 1H), 2.48 (sept,
J¼6.8 Hz, 1H), 3.12 (dd, J¼2.5, 8.7 Hz, 1H), 3.13–3.17
(m, 1H), 3.23 (dd, J¼1.9, 8.7 Hz, 1H), 3.35 (dq, J¼2.4,
6.8 Hz, 1H), 5.84 (d, J¼6.8 Hz, 1H), 5.96 (s, 1H); 13C
NMR (CDCl3, 100 MHz) d 212.3, 201.9, 166.5, 145.6,
126.1, 119.9, 73.4, 72.8, 55.8, 44.6, 41.9, 40.8, 33.2, 32.8,
32.2, 25.8, 22.9, 20.7, 20.0, 19.2; EIMS m/z (rel intensity)
332 (M+, 31), 315 (16), 289 (18), 271 (40), 243 (48), 166
(42), 149 (61), 125 (100).
CHCl3); IR (NaCl) 3426, 1716, 1669, 1597 cmꢀ1 1H
;
NMR (CDCl3, 300 MHz) d 0.88 (s, 9H), 1.16 (s, 9H), 2.88
(s, 1H), 3.68 (s, 1H), 3.72 (d, J¼9.0 Hz, 1H), 3.78 (dd,
J¼3.0, 6.8 Hz, 1H), 3.99 (dd, J¼3.0, 9.0 Hz, 1H), 4.93 (s,
1H), 5.80 (dd, J¼1.9, 6.8 Hz, 1H), 5.88 (s, 1H), 7.16–7.29
(m, 10H); 13C NMR (CDCl3, 62.9 MHz) d 199.1, 166.2,
145.3, 144.4, 143.3, 129.3, 128.1, 128.0, 127.7, 126.7,
126.6, 122.6, 54.7, 47.1, 40.4, 40.0, 37.9, 34.4, 29.6, 28.5;
ESIMS m/z (rel intensity) 507 (MNa+, 100), 485 (MH+, 2),
439 (4), 265 (10); HRMS (ESI) calcd for C32H36O4Na
507.2511, found 507.2495.
5.2.2. 3,10-Dihydroxy-3,10-di-iso-propyl-6,12-dimethyl-
tricyclo[6.2.2.02,7]dodeca-5,11-diene-4,9-dione (3s).
SIBX-mediated oxidation of thymol (1.5 g, 1 mmol) was
carried out according to the general procedure to give, after
purification by column chromatography, eluting with hex-
anes–Et2O (1:1), 2,3-dihydroxy-4-iso-propyl-1-methylbenz-
ene (5) as a brown oil (229 mg, 14%) and the cyclodimer
3s as a yellow solid (970 mg). The latter was recrystallized
twice at low temperature from hexanes–CH2Cl2 to afford
pure 3s as white needles (830 mg, 50%). Pure 3s was also
obtained as colorless prisms from slow evaporation of
CHCl3: mp 159.5 ꢁC (lit.38 mp 168–170 ꢁC); IR (KBr)
5.2. General procedure for SIBX-mediated oxidation
reaction of 2-alkylphenols
To a solution of 2-alkylphenol (10 mmol) in dry THF
(25 mL) was added SIBX (6.875 g, 11 mmol) as a solid in

J. Gagnepain et al. / Tetrahedron 63 (2007) 6493–6505
6503
3481, 2972, 2927, 1719, 1681, 1448, 1361 cmꢀ1; 1H NMR
(CDCl3, 400 MHz) d 0.56 (d, J¼6.6 Hz, 3H), 0.81 (d,
J¼6.8 Hz, 3H), 0.84 (d, J¼7.1 Hz, 3H), 0.95 (d, J¼6.6 Hz,
3H), 1.50–1.65 (m, 1H), 1.60 (s, 3H), 1.76 (sept,
J¼6.8 Hz, 1H), 1.95 (s, 3H), 2.36 (br s, 1H), 3.08 (d,
J¼7.3 Hz, 1H), 3.16 (s, 1H), 3.26 (d, J¼8.3 Hz, 1H), 3.29
(d, J¼6.6 Hz, 1H), 3.79 (s, 1H), 5.82 (d, J¼6.1 Hz, 1H),
5.97 (s, 1H); 13C NMR (CDCl3, 100 MHz) d 214.8, 201.8,
155.9, 135.7, 126.6, 125.3, 78.1, 77.8, 57.2, 47.1, 41.8,
37.2, 37.1, 32.5, 22.1, 21.3, 16.7, 16.6, 16.3, 16.0; ESIMS
(MeOH) m/z 333 [MH+].
50.6, 48.9, 45.8, 38.5, 36.1; EIMS m/z (rel intensity) 424
(M+, 2), 409 (2), 393 (17), 365 (17), 306 (7), 212 (6), 174
(100); HRMS (ESI) calcd for C20H24O10Na 447.1267, found
447.1286.
5.3.2. 4,11-Bis(30,30-dimethyl-10,40-dioxaspirocyclopen-
tan-2-one)pentacyclo[6.4.0.02,7.03,10.06,9]dodeca-5,12-di-
one (4k). Photoannelation of 3,10-bis(30,30-dimethyl-10,40-
dioxaspirocyclopentan-2-one)tricyclo[6.2.2.02,7]dodeca-
5,11-diene-4,9-dione21b (3k, 9 mg, 0.023 mmol) in CH2Cl2
(2 mL) was carried out over 20 h according to the general
procedure. The reaction mixture was then evaporated to fur-
nish 4k as a white solid (9 mg, quantitative yield). Recrystal-
lization from hexanes–CH2Cl2 afforded pure 4k as colorless
prisms: mp 310–313 ꢁC; IR (NaCl) 2921, 1806, 1738,
5.2.3. 3,10-Dihydroxy-3,12-di-iso-propyl-6,10-dimethyl-
tricyclo[6.2.2.02,7]dodeca-5,11-diene-4,9-dione (3s0).
SIBX-mediated oxidation of a 1:1 mixture of carvacrol
(1a, 150 mg, 1 mmol) and thymol (150 mg, 1 mmol) was
carried out according to the general procedure to give, after
purification by column chromatography, eluting with hex-
anes–acetone (10:1), three fractions: one black and two yel-
low solids. Submission of the black solid residue (90 mg) to
column chromatography, eluting with hexanes–Et2O (4:1),
afforded pure catechol 5 as a brown oil (40 mg, 24%) and
pure thymol-based cyclodimer 3s as a white solid (40 mg,
12%). Recrystallization at low temperature from hexanes–
CH2Cl2 of the two yellow solids identified as the carvacrol-
based cyclodimer 3a and the cyclodimer 3s0, respectively,
furnished pure cyclodimers 3a (68 mg, 21%) and 3s0
(31 mg, 9%) as white needles.
1
1177 cmꢀ1; H NMR (CDCl3, 300 MHz) d 1.49 (s, 6H),
1.52 (s, 6H), 2.86 (s, 2H), 3.36 (s, 6H); 13C NMR (CDCl3,
75.5 MHz) d 201.8, 174.0, 98.7, 78.8, 45.0, 41.2, 35.9,
35.6, 27.0, 25.0; CIMS (NH3) 406 (MNH+4); HRMS (ESI)
calcd for C20H21O8 389.1236, found 389.1219; CCDC
637238.
5.3.3. 4,11-Dihydroxy-7,8-di-iso-propyl-4,11-dimethyl-
pentacyclo[6.4.0.02,7.03,10.06,9]dodeca-5,12-dione (4a).
Photoannelation of cyclodimer 3a (68 mg, 0.2 mmol) in
MeOH (5 mL) was carried out over 24 h according to the
general procedure. Purification by column chromatography,
eluting with hexanes–acetone (3:1), gave 4a as a colorless
solid (55 mg, 81%). Further recrystallization from hex-
anes–CH2Cl2 afforded pure 4a as colorless prisms: mp
238–239 ꢁC (lit.39 mp 229 ꢁC); IR (KBr) 3469, 2974,
Compound 3s0: mp 158–159 ꢁC; IR (KBr) 3451, 1724, 1676,
1157 cmꢀ1
;
1H NMR (CDCl3, 400 MHz) d 0.59 (d,
1
J¼6.6 Hz, 3H), 0.87 (d, J¼6.8 Hz, 3H), 0.90 (d, J¼6.6 Hz,
3H), 0.98 (d, J¼6.6 Hz, 3H), 1.22 (s, 3H), 1.58 (sept,
J¼6.6 Hz, 1H), 1.80–2.00 (m, 1H), 1.97 (s, 3H), 2.38 (br
s, 1H), 3.09 (d, J¼8.1 Hz, 1H), 3.25 (br s, 2H), 3.40 (d,
J¼8.6 Hz, 1H), 3.86 (br s, 1H), 5.82 (d, J¼6.6 Hz, 1H),
5.95 (s, 1H); 13C NMR (CDCl3, 100 MHz) d 213.0, 201.6,
156.4, 145.5, 126.0, 125.2, 76.7, 72.9, 55.0, 44.3, 43.9,
37.8, 36.8, 33.1, 25.9, 22.4, 20.4, 19.4, 16.7, 15.9; ESIMS
(MeOH) m/z 333 (MH+); Anal. Calcd for C20H28O4: C,
72.26; H, 8.49; O, 19.25. Found: C, 71.89; H, 8.51; CCDC
636462.
1702, 1131 cmꢀ1; H NMR (CDCl3, 300 MHz) d 0.77 (s,
6H), 1.06 (s, 6H), 1.32 (s, 6H), 2.04 (s, 2H), 2.75 (s, 2H),
2.94 (s, 4H), 2.90–3.30 (m, 2H); 13C NMR (CDCl3,
75.5 MHz) d 215.9, 73.3, 50.7, 50.2, 40.6, 36.2, 30.0,
23.2, 19.1, 18.3; EIMS m/z (rel intensity) 332 (M+, 10),
289 (30), 271 (32), 243 (63), 229 (29), 201 (58), 191 (32),
166 (56), 159 (84), 149 (66), 123 (100); HRMS (ESI)
calcd for C20H28O4Na 355.1885, found 355.1891; CCDC
636464.
5.3.4. 4,11-Dihydroxy-4,11-di-iso-propyl-7,8-dimethyl-
pentacyclo[6.4.0.02,7.03,10.06,9]dodeca-5,12-dione (4s).
Photoannelation of cyclodimer 3s (100 mg, 0.3 mmol) in
MeOH (5 mL) was carried out over 24 h according to the
general procedure. Purification by column chromatography,
eluting with hexanes–Et2O (1:1), furnished 4s as a white
solid (90 mg, 90%). Further recrystallization from hex-
anes–CH2Cl2 gave pure 4s: mp 150 ꢁC (lit.39 mp 155 ꢁC);
5.3. General procedure for [2D2] photoannelation
reaction of cyclodimers
A Pyrex tube containing a solution of cyclodimer in either
dry MeOH or dry CH2Cl2 (ca. 0.04 M) was irradiated at
room temperature using a medium-pressure mercury lamp.
The reaction mixture was then evaporated and purified by
column chromatography.
1
IR (KBr) 3507, 2974, 1710, 1154, 1017 cmꢀ1; H NMR
(CDCl3, 300 MHz) d 0.79 (s, 6H), 0.96 (s, 6H), 1.16 (s,
6H), 1.66 (s, 2H), 2.56 (s, 2H), 2.85 (s, 4H), 3.10 (br s,
2H); 13C NMR (CDCl3, 75.5 MHz) d 215.0, 79.1, 49.2,
44.4, 40.5, 39.6, 29.0, 16.6, 16.2, 15.7; EIMS m/z (rel inten-
sity) 332 (M+, 6), 314 (5), 296 (8), 271 (8), 243 (30), 225
(18), 173 (29), 166 (41), 151 (67), 135 (38), 123 (35), 105
(62), 91 (65), 71 (100); HRMS (ESI) calcd for
C20H28O4Na 355.1885, found 355.1881.
5.3.1. 4,4,11,11-Tetramethoxy-7,8-dimethoxycarbonyl-
pentacyclo[6.4.0.02,7.03,10.06,9]dodeca-5,12-dione (4g).
Photoannelation of cyclodimer 3g (40 mg, 0.09 mmol) in
MeOH (3 mL) was carried out over 10 h according to the
general procedure. Purification by column chromatography,
eluting with cyclohexane–acetone (2:1), afforded 4g as
a pale yellow oil (22 mg, 55%): IR (NaCl) 2957, 2850,
1735, 1437, 1282, 1252, 1223, 1121, 1079, 1050 cmꢀ1; 1H
NMR (CDCl3, 300 MHz) d 2.89 (d, J¼3.9 Hz, 2H), 3.21
(s, 6H), 3.35 (s, 6H), 3.68 (d, J¼6.8 Hz, 2H), 3.72 (s, 8H);
13C NMR (CDCl3, 75.5 MHz) d 200.7, 169.4, 95.5, 52.5,
5.3.5. 4,11-Dihydroxy-4,8-di-iso-propyl-7,11-dimethyl-
pentacyclo[6.4.0.02,7.03,10.06,9]dodeca-5,12-dione (4s0).
Photoannelation of cyclodimer 3s0 (25 mg, 0.075 mmol) in
MeOH (2 mL) was carried out over 24 h according to the

6504
J. Gagnepain et al. / Tetrahedron 63 (2007) 6493–6505
Liao, C.-C. Angew. Chem., Int. Ed. 2002, 41, 4090–4093; (e)
Hsu, D.-S.; Hsu, P.-Y.; Liao, C.-C. Org. Lett. 2001, 3, 263–
265; (f) Drutu, I.; Njardarson, J. T.; Wood, J. L. Org. Lett.
2002, 4, 493–496; (g) Njardarson, J. T.; McDonald, I. M.;
Spiegel, D. A.; Inoue, M.; Wood, J. L. Org. Lett. 2001, 3,
2435–2438; (h) Hoshino, O.; Suzuki, M.; Ogasawara, H.
Tetrahedron 2001, 57, 265–271 and references cited therein;
(i) Singh, V.; Samanta, B.; Kane, V. V. Tetrahedron 2000,
56, 7785–7795; (j) Nicolaou, K. C.; Li, J. Angew. Chem.,
Int. Ed. 2001, 40, 4264–4268; (k) Nicolaou, K. C.;
Vassilikogiannakis, G.; Simonsen, K. B.; Baran, P. S.;
Zhong, Y.-L.; Vidali, V. P.; Pitsinos, E. N.; Couladouros,
E. A. J. Am. Chem. Soc. 2000, 122, 3071–3079; (l) Cox, C.;
Danishefsky, S. J. Org. Lett. 2000, 2, 3493–3496.
general procedure. Purification by column chromatography,
eluting with hexanes–acetone (3:1), gave 4s0 as a colorless
amorphous solid (16 mg, 64%). Further recrystallization
from hexanes–CH2Cl2 afforded pure 4s0 as colorless crys-
tals: mp 173–175 ꢁC; IR (KBr) 3325, 2967, 1710,
1
1131 cmꢀ1; H NMR (CDCl3, 400 MHz) d 0.81 (s, 3H),
0.87 (s, 3H), 1.00 (s, 6H), 1.29 (s, 6H), 1.71 (s, 2H), 1.83
(s, 1H), 2.51 (s, 1H), 2.79 (s, 2H), 2.92 (s, 1H), 2.99 (s,
1H); 13C NMR (CDCl3, 100 MHz) d 215.7, 215.1, 78.9,
73.6, 51.0, 49.6, 47.2, 41.7, 41.4, 41.1, 40.7, 39.4, 30.2,
28.9, 23.1, 18.6, 18.2, 18.1, 16.4, 15.8; EIMS m/z (rel inten-
sity) 332 (M+, 6), 289 (7), 271 (20), 243 (45), 229 (21), 201
(25), 173 (44), 149 (87), 137 (52), 125 (56), 119 (64), 71
(100); HRMS (EIMS) calcd for C20H28O4 332.1987, found
332.1972.
6. (a) Andersson, G. Acta Chem. Scand. B 1976, 30, 64–70; (b)
Andersson, G. Acta Chem. Scand. B 1976, 30, 403–406.
7. Liao, C.-C.; Chu, C.-S.; Lee, T.-H.; Rao, P. D.; Ko, S.; Song,
L.-D.; Shiao, H.-C. J. Org. Chem. 1999, 64, 4102–4110.
8. (a) Andersson, G.; Berntsson, P. Acta Chem. Scand. B 1975, 29,
948–952; (b) Lai, C.-H.; Shen, Y.-L.; Liao, C.-C. Synlett 1997,
1351–1352.
Acknowledgements
This work was supported by the Institut Universitaire de
ꢁ
France (IUF) and the Ministere de la Recherche. The authors
also wish to thank the Ministere de la Recherche for Julien
ꢁ
9. Metlesics, W.; Wessely, W. Monatsh. Chem. 1957, 88, 108–
117.
Gagnepain and Delphine Dejugnac’s research assistantships,
Anne-Marie Lamidey for her contribution to the synthesis
part of this work, and Katell Bathany for high-resolution
mass spectrometric analyses.
10. (a) Carman, R. M.; Lambert, L. K.; Robinson, W. T.; Van
Dongen, J. M. A. M. Aust. J. Chem. 1986, 39, 1843–1850;
(b) Carman, R. M.; Owsia, S.; Van Dongen, J. M. A. M.
Aust. J. Chem. 1987, 40, 333–340; (c) Liao, Y.-H.; Xu,
L.-Z.; Yang, S.-H.; Dai, J.; Zhen, Y.-S.; Zhu, M.; Sun, N.-J.
Phytochemistry 1997, 45, 729–732; (d) Su, B.-N.; Zhu,
Q.-X.; Jia, Z.-J. Tetrahedron Lett. 1999, 40, 357–358; (e)
Zdero, C.; Bohlmann, F.; Niemeyer, H. M. Phytochemistry
1991, 30, 1597–1601; (f) Falshaw, C. P.; King, T. V.
J. Chem. Soc., Perkin Trans. 1 1983, 1749–1752; (g)
Clarkson, C.; Staerk, D.; Hansen, S. H.; Smith, P. J.;
Jaroszewski, J. W. J. Nat. Prod. 2006, 69, 1280–1288; (h)
Abe, N.; Arakawa, T.; Yamamoto, K.; Hirota, A. Biosci.
Biotechnol. Biochem. 2002, 66, 2090–2099; (i) Abe, N.;
Sugimoto, O.; Arakawa, T.; Tanji, K.-i.; Hirota, A. Biosci.
Biotechnol. Biochem. 2001, 65, 2271–2279; (j) Abe, N.;
Sugimoto, O.; Tanji, K.-i.; Hirota, A. J. Am. Chem. Soc.
2000, 122, 12606–12607.
Supplementary data
X-ray data on structures 3g, 3k, 3s0, 4a, 4k, and 4l are
available from the Cambridge Crystallographic Data Center
(CCDC 636463, 178536, 636462, 636464, 637238,
245748). Supplementary data associated with this article
can be found in the online version, at doi:10.1016/j.tet.
2007.03.035.
References and notes
1. Modern Arene Chemistry; Astruc, D., Ed.; Wiley-VCH:
Weinheim, 2002.
11. Gagnepain, J.; Castet, F.; Quideau, S. Angew. Chem., Int. Ed.
2007, 46, 1533–1535.
12. Nicolaou, K. C.; Simonsen, K. B.; Vassilikogiannakis, G.;
Baran, P. S.; Vidali, V. P.; Pitsinos, E. N.; Couladouros, E. A.
Angew. Chem., Int. Ed. 1999, 38, 3555–3559.
ꢀ
13. Deffieux, D.; Fabre, I.; Titz, A.; Leger, J.-M.; Quideau, S.
J. Org. Chem. 2004, 69, 8731–8738.
14. (a) Holmberg, K. Acta Chem. Scand. B 1974, 28, 857–865; (b)
Adler, E.; Holmberg, K. Acta Chem. Scand. B 1974, 28, 549–
554; (c) Adler, E.; Holmberg, K. Acta Chem. Scand. B 1974,
28, 465–472.
ꢀ
2. (a) Quideau, S.; Pouysegu, L.; Deffieux, D. Curr. Org. Chem.
2004, 8, 113–148; (b) Rodriguez, S.; Wipf, P. Synthesis 2004,
2767–2783; (c) Oxidative Activation of Aromatic Rings: An
Efficient Strategy for Arene Functionalization; Quideau, S.,
Feldman, K. S., Eds.; Elsevier Science/Pergamon: Oxford,
2001; Tetrahedron Symposium-in-Print Nꢁ85; (d) Swenton,
J. S. The Chemistry of Quinonoid Compounds; Patai, S.,
Rappoport, Z., Eds.; John Wiley: New York, NY, 1988;
pp 899–962.
3. (a) Quideau, S. Modern Arene Chemistry; Astruc, D., Ed.;
Wiley-VCH: Weinheim, 2002; pp 539–573; (b) Quideau, S.;
ꢀ
15. Quideau, S.; Looney, M. A.; Pouysegu, L.; Ham, S.; Birney,
ꢀ
Pouysegu, L.; Oxoby, M.; Looney, M. Tetrahedron 2001, 57,
319–329; (c) Quideau, S.; Pouysegu, L. Org. Prep. Proced.
Int. 1999, 31, 617–680.
D. M. Tetrahedron Lett. 1999, 40, 615–618.
16. (a) Tamura, Y.; Yakura, T.; Haruta, J.-i.; Kita, Y. J. Org. Chem.
1987, 52, 3927–3930; (b) Pelter, A.; Elgendy, S. Tetrahedron
Lett. 1988, 29, 677–680.
ꢀ
4. (a) Magdziak, D.; Meek, S. J.; Pettus, T. R. R. Chem. Rev. 2004,
104, 1383–1429; (b) Liao, C.-C.; Peddinti, R. K. Acc. Chem.
Res. 2002, 35, 856–866; (c) Singh, V. Acc. Chem. Res. 1999,
32, 324–333.
5. (a) Marsini, M. A.; Gowin, K. M.; Pettus, T. R. R. Org. Lett.
2006, 8, 3481–3483; (b) Cook, S. P.; Danishefsky, S. J. Org.
Lett. 2006, 8, 5693–5695; (c) Singh, V.; Vedantham, P.; Sahu,
P. K. Tetrahedron Lett. 2002, 43, 519–522; (d) Yen, C.-F.;
ꢀ
17. Quideau, S.; Pouysegu, L.; Looney, M. A. J. Org. Chem. 1998,
63, 9597–9600.
18. Adler, E.; Dahlen, J.; Westin, G. Acta Chem. Scand. 1960, 14,
1580–1596.
19. Barnes-Seeman, D.; Corey, E. J. Org. Lett. 1999, 1, 1503–1504.
20. Auksi, H.; Yates, P. Can. J. Chem. 1981, 59, 2510–2517.

J. Gagnepain et al. / Tetrahedron 63 (2007) 6493–6505
6505
ꢀ
21. (a) Quideau, S.; Pouysegu, L.; Deffieux, D.; Ozanne, A.;
30. Macaulay, J. B.; Fallis, A. G. J. Am. Chem. Soc. 1990, 112,
1136–1144.
31. Ishida, M.; Beniya, Y.; Inagaki, S.; Kato, S. J. Am. Chem. Soc.
1990, 112, 8980–8982.
Gagnepain, J.; Fabre, I.; Oxoby, M. ARKIVOC 2003, 6, 106–
119; (b) Deffieux, D.; Fabre, I.; Courseille, C.; Quideau, S.
J. Org. Chem. 2002, 67, 4458–4465.
€
22. (a) Becker, H.-D.; Ruge, B.; Westlof, T. Tetrahedron Lett.
32. Adler, E.; Brasen, S.; Miyake, H. Acta Chem. Scand. 1971, 25,
2055–2069.
33. (a) Kende, A. S.; MacGregor, P. J. Am. Chem. Soc. 1961, 83,
4197–4204; (b) Jung, M. E.; Jachiet, D.; Khan, S. I.; Kim, C.
Tetrahedron Lett. 1995, 36, 361–364.
1975, 16, 253–256; (b) Becker, H.-D.; Konar, A. Tetrahedron
Lett. 1972, 13, 5177–5180; (c) Becker, H.-D. Liebigs Ann.
Chem. 1973, 10, 1675–1685; (d) Iwakuma, T.; Yonemitsu,
O.; Kanamaru, N.; Kimura, K.; Witkop, B. Angew. Chem.,
Int. Ed. Engl. 1973, 12, 72–73; (e) Yamamura, S.; Terada, Y.;
Chen, Y.-P.; Hsu, H.-Y.; Hirata, Y. Tetrahedron Lett. 1975,
1903–1904; (f) Antus, S.; Nogradi, M.; Baitz-gacs, E.;
Radics, L.; Becker, H.-D.; Karlsson, B.; Pilotti, A.-M.
Tetrahedron 1978, 34, 2573–2577.
34. (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652; (b) Lee,
C.; Yang, W.; Parr, R. G. Phys. Rev. 1988, 37, 785–789.
35. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A. J.;
Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.;
Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennuci, B.; Cossi,
M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.;
Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.;
Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.;
Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.;
Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski,
J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels,
A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck,
A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui,
Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov,
B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng,
C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.;
Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople,
J. A. Gaussian 03 (Revison B.04); Gaussian: Pittsburgh, PA,
2003.
23. Brown, T. L.; Curtin, D. Y.; Fraser, R. R. J. Am. Chem. Soc.
1958, 80, 4339–4341.
24. (a) Salem, L. J. Am. Chem. Soc. 1958, 90, 553–566; (b)
Birney, D. M.; Houk, K. N. J. Am. Chem. Soc. 1990, 112,
4127–4133.
25. (a) Caramella, P.; Quadrelli, P.; Toma, L. J. Am. Chem. Soc.
2002, 124, 1130–1131; (b) Quadrelli, P.; Romano, S.; Toma,
L.; Caramella, P. Tetrahedron Lett. 2002, 43, 8785–8789; (c)
Quadrelli, P.; Romano, S.; Toma, L.; Caramella, P. J. Org.
Chem. 2003, 68, 6035–6038.
26. (a) Williamson, K. L.; Hsu, Y.-F. L. J. Am. Chem. Soc. 1969, 91,
7385–7389; (b) Williamson, K. L.; Hsu, Y.-F. L.; Lacko, R.;
Youn, C. H. J. Am. Chem. Soc. 1969, 91, 6129–6138.
27. (a) Adler, E.; Andersson, G.; Edman, E. Acta Chem. Scand. B
1975, 29, 909–920; (b) Adler, E.; Holmberg, K.; Ryrfors, L.-O.
Acta Chem. Scand. B 1974, 28, 888–894.
28. (a) Ohkata, K.; Tamura, Y.; Shetuni, B. B.; Takagi, R.;
Miyanaga, W.; Kojima, S.; Paquette, L. A. J. Am. Chem. Soc.
2004, 126, 16783–16792; (b) Paquette, L. A.; Branan, B. M.;
Rogers, R. D.; Bond, A. H.; Lange, H.; Gleiter, R. J. Am.
Chem. Soc. 1995, 117, 5992–6001 and references cited therein;
(c) Datta, S. C.; Franck, R. W.; Tripathy, R.; Quigley, G. J.;
Huang, L.; Chen, S.; Sihaed, A. J. Am. Chem. Soc. 1990,
112, 8472–8478 and references cited therein.
ꢀ
36. Ozanne, A.; Pouysegu, L.; Depernet, D.; Franc¸ois, B.;
Quideau, S. Org. Lett. 2003, 5, 2903–2906.
37. Schwarz, K.; Ernst, H.; Ternes, W. J. Sci. Food Agric. 1996, 70,
217–223.
38. Falshaw, C. P.; Franklinos, A. J. Chem. Soc., Perkin Trans. 1
1984, 95–100.
29. Hou, H.-F.; Pedditi, R. K.; Liao, C.-C. Org. Lett. 2002, 4,
2477–2480.
39. Paknikar, S. K.; Patel, J.; Ramanamma, C. V. Proc. 11th Int.
Congr. Frag. Flav. Ess. Oils 1989, 5, 105–111.