Intramolecular Hetero-Michael addition of ss-hydroxyenones for the preparation of highly substituted tetrahydropyranones

Article abstract of DOI:10.1002/chem.200600415

Structurally diverse ss-hydroxyenones are shown to undergo nonoxidative 6-endo-trig ring closure to form highly substituted tetrahydropyranones. Amberlyst-15, Al(C1O₄) ₃·9H₂O and [Pd(MeCN)₄](BF ₄)

Full text of DOI:10.1002/chem.200600415

DOI: 10.1002/chem.200600415
Intramolecular Hetero-Michael Addition of b-Hydroxyenones for the
Preparation of Highly Substituted Tetrahydropyranones
Maud Reiter, Hazel Turner, and VØronique Gouverneur*[a]
Abstract: Structurally diverse b-hy-
droxyenones are shown to undergo
nonoxidative 6-endo-trig ring closure to
form highly substituted tetrahydropyra-
Lewis acid Al
tion proceeded with a similar level of
diastereocontrol; however, in contrast
A
to emerge from this newchemistry is
the possibility to access tri- and tetra-
substituted
2,6-anti-tetrahydropyra-
to [Pd
A
nones from anti-aldol precursors. These
compounds are particularly difficult to
access by using alternative methodolo-
gies. Two modes of activation were en-
visaged for the ring closure, involving
metal coordination to either the C=C
or C=O functional groups. Experimen-
tal results suggest that C=O coordina-
tion was the preferred mode of activa-
tion for reactions performed in the
nones. Amberlyst-15, Al
A
promote enolisation. The palladium-
mediated reaction was also found to be
compatible with an enantioenriched b-
hydroxyenone substrate, giving no loss
of enantiopurity upon ring closure. The
most distinctive synthetic development
and [Pd(MeCN)4](BF4)2 were found to
A
ACHTREUNG
be suitable catalysts for these intramo-
lecular conjugate additions, preventing
side reactions, such as dehydration or
retroaldolisation. The use of [Pd-
A
N
tive, as this palladium-mediated reac-
tion is under kinetic control and gener-
ates tri- and tetrasubstituted tetrahy-
dropyranones with high levels of dia-
stereocontrol. In the presence of the
Keywords:
Michael addition
catalysts
tetrahydropyranones
presence of Al
A
·
A
ACHTREUNG
Introduction
whereas for the Prins cyclisation, ring closure occurs with
À
C C bond formation only. The Maitland–Japp reaction
Tetrahydropyranones (THPs) are valuable intermediates for
the synthesis of a large variety of biologically active com-
pounds and there has been a long-standing interest in their
stereoselective construction.[1] Some of the most widely used
methods for their preparation are based on manipulations of
carbohydrates,[2] hetero-Diels–Alder (HDA) reactions,[3]
oxy-Cope/Prins cascades[4] and simple Prins cyclisations,[5]
variants of the Maitland–Japp reaction (Knoevenagel/Mi-
chael-addition cascades)[6] and Bu2BOTf-mediated aldol-
type cyclisations.[7] These methodologies rely on comple-
mentary retrosynthetic approaches. In the HDA reaction,
relies on an intramolecular conjugate Michael addition, for
À
which a C O bond is formed upon cyclisation (Scheme 1).
Perhaps the most impressive of these approaches is the
HDA reaction. This transformation allows for the diastereo-
and enantioselective formation of up to three stereocentres
in a single step, provided that a single geometrical diaste-
reomer of the starting diene is used. The groups of Jacobsen
and Hashimoto et al.[3] developed elegant catalytic asymmet-
ric cycloadditions, involving a range of carbonyl dienophiles
and Danishefsky dienes possessing only one oxygenated
substituent. This route provides a direct entry into all-syn
trisubstituted THPs with excellent enantio- and diastereo-
meric excesses. The Diels–Alder route is, however, more
commonly used for the preparation of dihydropyranones
(DHPs) rather than THPs. This is due to the limited
number of catalytic asymmetric routes available for the
more demanding cycloadditions involving less activated
dienes.[9] Mechanistically, only limited data are available for
the HDA reaction leading to THPs, whereas it is well estab-
lished that the formation of DHPs can proceed through
both concerted and stepwise pathways.
À
À
the C C and C O bonds are constructed simultaneously,
[a] Dr. M. Reiter, H. Turner, Dr. V. Gouverneur
University of Oxford, Chemistry Research Laboratory
12, Mansfield Road, Oxford OX1 3TA (UK)
Fax : (+44)1865-275-644
E-mail: veronique.gouverneur@chem.ox.ac.uk
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
7190
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 7190 – 7203
Please note: Minor changes have been made to this manuscript since its publication in Chemistry - A European Journal Early View. The Editor.

FULL PAPER
lysts was initially tested in rac-
emic series with the prepara-
tion of a representative trisub-
stituted
tetrahydropyranone
from the corresponding b-hy-
droxyenone. For the most ef-
fective catalysts, we conducted
a detailed study on the scope
and generality of these intra-
molecular Michael additions
with a variety of structurally
diverse aldol products. We
have also examined the stereo-
chemical outcome of these
nonoxidative cyclisations by
varying the relative stereo-
chemistry of diastereomerically
enriched b-hydroxyenones and
the substitution pattern of
À
their C C double bonds. The
synthesis of THPs featuring
from one up to four stereogen-
ic centres is reported. The
methodology was also chal-
lenged with the cyclisation of
Scheme 1. Representative approaches towards tetrahydropyranones. TBAF=tetrabutylammonium fluoride;
TFA=trifluoroacetic acid; de=diastereomeric excess.
Although a range of intermolecular additions of alcohols
onto a,b-unsaturated ketones have been reported in the lit-
erature, to the best of our knowledge, examples of endocy-
clic oxy-Michael additions to proximal a,b-unsaturated sys-
tems leading to THPs are rare.[8,9] This approach can be re-
garded as a variant of the stepwise HDA reaction, involving
the formation of a b-hydroxyenone, followed by a nonoxida-
tive ring closure.[10] In previous studies, we have shown that
PdCl2 in the presence of the co-oxidants CuCl and oxygen
catalyzed the intramolecular oxidative cyclisation of b-hy-
droxyenones to give DHPs.[11] In this paper, we report on
how to reroute the ring closure pathway towards a nonoxi-
dative cyclisation, with the result of preparing tetrahydro-
pyranones instead of dihydropyranones from the same b-hy-
droxyenone precursors (Scheme 2).
Inspired by the established palladium(ii)-mediated inter-
molecular hydroalkoxylation and amidation addition reac-
tions to enones,[12] we selected several palladium(ii)-based
catalysts to mediate the intramolecular nonoxidative ring
closure. We also tested Brønsted and Lewis acids, which
were shown to promote other nonoxidative intermolecular
Michael conjugate additions.[13] The efficiency of these cata-
an enantioenriched b-hydroxyenone substrate. Finally, we
discuss the mechanism of the most successful nonoxidative
ring closures conducted in the light of control experiments
and the observed stereoselectivities upon cyclisation.
Results and Discussion
Cyclisation versus competitive dehydration or retro-aldolisa-
tion: The groups of Kobayashi,[14] Spencer[13] and Banik
et al.[15] have recently published work on the intermolecular
aza- and oxy-Michael addition of weakly basic nitrogen-con-
taining nucleophiles, such as carbamates, oxazolidinones or
benzylic alcohols, to a,b-unsaturated ketones. They have
shown that these intermolecular Michael additions can be
catalyzed by a range of Brønsted and Lewis acids, as well as
PdII-catalysts. These catalysts are believed to enhance the
electrophilicity of the enone substrates, thereby allowing the
reaction to proceed under mild conditions. Despite the fact
that Spencer et al. had only limited success with the inter-
molecular Michael addition of oxygen nucleophiles, we set
out to study an entropically more favourable intramolecular
variant of this reaction by using b-hydroxyenones as the
starting materials. Upon cyclisation, these substrates could
lead to side reactions, such as competitive dehydration or
retroaldolisation processes. The most suitable catalyst for
the intermolecular Michael addition of alcohols to enones
might, therefore, differ from the catalyst of choice for the 6-
endo-trig cyclisation of b-hydroxyenones. We investigated in
detail the ring closure of the representative aldol 1a in the
presence of numerous Brønsted and Lewis acids, as well as
palladium(ii) complexes (Table 1). Several organic acids
Scheme 2. PdII-mediated oxidative and nonoxidative ring closure of b-hy-
droxyenones.
Chem. Eur. J. 2006, 12, 7190 – 7203
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7191

V. Gouverneur et al.
Table 1. Optimisation study for the intramolecular Michael addition (IMA) of 1a.
conditions, neither PdCl2 nor
Pd(OAc)2 promoted the cycli-
A
sation event (entry 15). Al-
though the addition of halide
ions was shown in some cases
to further promote the IMA
reaction, under the present
conditions their addition did
Entry
Catalyst (10 mol%)
Conditions
Ratio 2a/3a
Yield 2a [%]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
AcOH
MeCN, 18 h
MeCN, 18 h
MeCN, 18 h
MeCN, 18 h
MeCN, 18 h
CH2Cl2, 3 h
MeCN, 18 h
MeCN, 18 h
MeCN, 18 h
MeCN, 18 h
MeCN, 3 h
–
100:0
1:5
1:1
4:1
100:0
3:7
2:1
2:1
100:0
100:0
–
0
20
14
–
–
90
–
–
–
60
71
0
0
0
0
0
72
68
TFA[a]
not have a noticeable effect
G
(entry 16).[16] However, when
the reaction was conducted in
the presence of the two palla-
dium(ii) catalyst featuring ace-
NHTf2
pTSA
HBF4
Amberlyst-15 (A)[b]
Bi
Cu
ZrO
Fe(ClO4)2·H2O
Al(OCl4)3·9H2O (B)
ACHTREUNG
H
tonitrile
(MeCN)2] and [Pd
(BF4)2), 1a was successfully cy-
ligands
([PdCl2-
N
A
ACHTREUNG
R
AHCTREUNG
ACHTREUNG
clised into 2a in 68 and 72%
yields, respectively (entries 17–
18). The cationic palladium
species was clearly the best
catalyst for this reaction, as re-
flected by the shorter reaction
time (entry 18). The results
summarised in Table 1 re-
vealed that the most suitable
NEt3
KOtBu
NaH
CH2Cl2, 70 h
THF, 0.5 h
–
–
–
–
CH2Cl2, RT, 3 h
CH2Cl2, 18 h
CH2Cl2, 18 h
CH2Cl2, 1 h
CH2Cl2, 18 h
PdCl2 or Pd
PdCl2 or Pd
U
[Pd
(MeCN)4]
N
100:0
7:1
A
ACHTREUNG
[a] No reaction for 10 mol%, results shown for 200 mol%. [b] Amount used=2 equiv w/w.
were tested first (entries 1–6). Acetic and trifluoroacetic
acid gave no or very lowconversions, with mainly starting
material being recovered (entries 1–2). Upon treatment with
NHTf2, HBF4 or pTSA (pTSA=para-toluenesulfonic acid),
a competitive dehydration process took place and the cy-
clised product 2a was obtained in low yields (entry 3–5). We
then tested Amberlyst-15, a catalyst employed by Kalesse
et al.[8] for the formation of tetrahydropyrans and by Spenc-
er et al.[13] for intermolecular Michael additions to enones.
With this mild sulfonic acid resin, aldol 1a was cleanly con-
verted into the desired THP 2a with an isolated yield of
catalysts for the ring closure of b-hydroxyenones differed
significantly from the intermolecular series, in which the
Brønsted acid NHTf2 was superior to the palladium(ii) salts.
The most efficient catalysts for the IMA of the b-hydroxye-
none 1a were Amberlyst-15 (A), Al
A
A
N
ACHTREUNG
Scope and generality: The substrate tolerance of these four
catalysts was investigated next for the ring closure of a
series of achiral and racemic b-hydroxyenones featuring one
or no stereocentre (Table 2). For the palladium(ii)-mediated
cyclisations, we favoured the use of catalyst C, as this cata-
lyst gave better yields of the desired product within shorter
reaction times and with lower catalytic loading. b-Hydroxy-
enones 1a–g ring closed cleanly to give the corresponding
tetrahydropyranones 2a–g in good yields (54–91%, Table 2,
entries 1–7). Interestingly, for the hydroxyenones 1d–g, fea-
turing a b-hydrogen on the double bond, the nonoxidative
Michael conjugate addition was found to be the only reac-
tion pathway taking place in the presence of two palladi-
um(ii) catalysts C and D. No traces of the unwanted oxi-
dised dihydropyranone products were detected in the crude
reaction mixture. For substrate 1d, significant 1,5-stereoin-
duction was observed in the presence of Brønsted acid cata-
lyst A, leading to the preferential formation of the syn-die-
quatorial THP 2d. The two diastereomeric products were
separated cleanly by silica-gel chromatography. In contrast,
no significant level of stereoinduction was observed for the
PdII-mediated reaction, which led to the formation of syn-
and anti-2d in almost equal proportions (entry 4). When
using the highly active cationic palladium catalyst C, the b-
90% (entry 6). Various Lewis acids, including Cu
ZrO(ClO4)2·8H2O, were used next, but they predominantly
gave mixtures of cyclic and dehydrated products (entries 7–
9). The Lewis acid Fe(ClO4)2·H2O gave the desired THP 2a
in 60% yield (entry 10). The use of Al(ClO4)3·9H2O proved
(OTf)2 and
ACHTREUNG
AHCTREUNG
AHCTREUNG
to be even more suitable, as this reaction gave 2a within
three hours in a 71% isolated yield without any detectable
dehydration (entry 11). This Lewis acid was therefore re-
tained for further cyclisations. As anticipated, the use of
basic conditions was unsuccessful. In the presence of NEt3,
only starting material was recovered, NaH and KOtBu led
to significant side reactions, such as retroaldolisations (en-
tries 12–14). The efficiency of four palladium(ii) catalysts
was also examined, namely palladium(ii)diacetate Pd
and palladium(ii)dichloride PdCl2, bis(acetonitrile)palladiu-
m(ii)dichloride [Pd(MeCN)2]Cl2 and tetrakis(acetonitrile)-
palladium(ii)tetrafluoroborate [Pd(MeCN)4](BF4)2 (en-
(OAc)2
ACHTREUNG
A
ACHTREUNG
tries 15–18). When used in combination with CuCl and
oxygen, palladium dichloride was a suitable catalyst for the
oxidative cyclisation of b-hydroxyenones featuring at least
one b-hydrogen on the double bond.[11] Under nonoxidising
hydroxyenones featuring no double-bond substituent a to
G
7192
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 7190 – 7203

Intramolecular Hetero-Michael Addition
FULL PAPER
Table 2. Preparation of racemic tetrahydropyranones 2a–i and dihydropyranone 2j.
Entry
Starting material
Yield [%]
Product
Catalyst, t [h][a]
Yield [%]
dr[b]
A, 3
B, 3
C, 1
D, 18
90
71
72
68
–
–
–
–
1
96
A, 3
C, 1
D, 18
91
79
78
–
–
–
2
3
82
73
A, 3
C, 1
67
65
–
–
A, 3
D, 12
90
89
93:7 (99:1)
1:1 (99:1)[c]
4
5
6
74
84
55
C, 1
63
58
–
C, 12
1:1
A, 18
C, 18
54
65
4:1[d]
9:1[d]
7
8
9
96
98
97
C, 44
C, 44
55
54
–
–
10
74
C, 12
60
–
[a] A=Amberlyst-15, CH2Cl2, RT; B=Al
A
G
N
after purification. [c] Yield for the syn isomer is 47%. [d] Minor diastereomer is all-syn 2g.
the carbonyl, cyclised within one hour, but the a-methylated
b-hydroxyenones 1 f–g required overnight stirring (entries 6–
7). The THP 2 f was obtained as a mixture of diastereomers
in the presence of C, indicating that no 1,4-stereoinduction
had occurred (entry 6). Noteworthy, is that the b-hydroxye-
none substrate 1 f did not cyclize under oxidative conditions
(PdCl2, CuCl, O2).[11] For 1g, featuring a trisubstituted E
double bond, both catalysts A and C gave predominantly
the trisubstituted THP anti,anti-2g. NOE and X-ray analysis
confirmed that the major diastereomer featured all three
substituents in equatorial positions. The minor diastereomer
was identified as the all-syn 2g product. For aldols 1h–i, de-
rived from ketone acceptors, much longer reaction times (up
to 44 h) were required in comparison to aldols 1a–g derived
from aldehyde precursors. A similar trend was observed for
intermolecular Michael additions.[12] A competitive dehydra-
tion process also took upon ring closure of 1h–i, which re-
sulted in lower chemical yields for the formation of 2h–i
Chem. Eur. J. 2006, 12, 7190 – 7203
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7193

V. Gouverneur et al.
(54–55%, entries 8–9). Finally, we found that the palladium-
mediated methodology was not limited to b-hydroxyenones;
in the presence of catalyst C, the b-hydroxyynone 1j cy-
clised into the corresponding dihydropyranone 2j in 60%
yield (entry 10).
diastereoselectivities. The anti stereoisomers were obtained
in pure form upon silica-gel chromatography; however, this
was only possible for aldol products derived from benzalde-
hyde. With these more heavily substituted substrates in
hand, we performed the subsequent intramolecular Michael
additions. Three stereochemical issues were raised by these
cyclisations: the stereochemical integrity of the existing ste-
Oxidative versus nonoxidative cyclisation: We verified
whether we could obtain the oxidised or the nonoxidised
cyclic product derived from a given b-hydroxyenone sub-
strate by using the same catalyst, but fine-tuning the reac-
tion conditions (Scheme 3). Aldol 1d cyclised within one
A
tions, the level of diastereocontrol upon ring closure and the
directionality of the final protonolysis step. Catalysts A–C
were selected for this study to determine which catalyst
might be the most useful syn-
thetically. We also hoped to
shed some light on their re-
spective mode of action
(Table 3).
Brønsted acid mediated IMA:
In the first set of experiments,
we investigated the stereo-
chemical outcome of the Am-
berlyst-15-mediated IMA reac-
Scheme 3. Role of the catalyst in the ring closure. DME=dimethoxyethane.
hour in the presence of cation-
Table 3. Preparation of syn- and anti-b-hydroxyenones 1k–m.
ic [Pd
A
C
Entry
1
syn-b-Hydroxyenones
Yield [%], dr
Entry
anti-b-Hydroxyenones
Yield [%], dr
chloromethane under an argon
atmosphere to give 2d in 89%
yield as a mixture of two dia-
stereomers. No reaction, how-
ever, occurred in the presence
of PdCl2. When we submitted
1d to the same cationic cata-
lyst C (8 mol%) under oxidis-
1
1k
62, 99:1
4
72, 4:1 (99:1)[a]
2
3
1l
50, 99:1
67, 99:1
5
6
40, 9:1 (99:1)[a]
45, 3:1, (99:1)[a]
1m
ing
conditions
(CuCl,
Na2HPO4, O2 in DME at
508C),[11] the cyclisation reac-
[a] Number in brackets indicates dr for major diastereomer after purification.
tion required 48 h to reach
completion and the desired
oxidised DHP product 4d was formed in 30% yield, along
with products of decomposition. Interestingly, no product re-
sulting from an intramolecular conjugate addition was isolat-
ed. In contrast, in the presence of PdCl2 under oxidising
conditions, 4d was formed in 79% yield within 4 h of stir-
ring at 508C.[11]
tion of syn- and anti-enones 1k–m. Most cyclisations
reached completion after 5 h at RT, but anti-1k–m sub-
strates required an extended reaction time of 18 h (Table 4).
The Brønsted acid mediated cyclisation of syn-1k led to the
formation of all-syn-2k and anti,anti-2k in a total yield of
75% and in a ratio of 7:3 (Table 4, entry 1). The level of
control for the ring closure was excellent, with both isolated
diastereomers featuring an equatorial methyl group at the
newly formed stereocentre at the 6-position. The minor dia-
stereomer anti,anti-2k resulted from an epimerisation of the
stereocentre at the 3-position. This was possibly caused by a
lack of stereocontrol upon protonolysis, combined with an
enolisation process occurring either before or after the cycli-
sation event. For the cyclisation of syn-1l, an excellent level
of control was also observed, with all three diastereomers
featuring an equatorial methyl group at the 6-position. For
the major diastereomer, syn,anti,anti-2l, the relative syn-
stereochemistry of the starting aldol was faithfully translated
and the protonolysis step led to an equatorial methyl group
Intramolecular Michael addition of syn- and anti-a-methyl-
b-hydroxyenones: We then directed our attention to the cyc-
lisation of aldol products possessing two stereogenic centres.
The syn- and anti-b-hydroxyenones 1k–m were obtained fol-
lowing previously published procedures,[11] which were
adapted from the diastereoselective boron-mediated aldol
reaction developed by Paterson et al.[17] The syn-aldols 1k–l
were formed with excellent diastereocontrol by coupling
(E)-4-hexen-3-one, (E)-4-methylhex-4-en-3-one or (E)-4,6-
dimethylhept-4-en-3-one with benzaldehyde in the presence
of nBu2BOTf and iPr2EtN. The use of cHex2BCl and
EtMe2N preferentially gave anti-aldols 1k–l with moderate
7194
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 7190 – 7203

Intramolecular Hetero-Michael Addition
FULL PAPER
Table 4. Diastereomeric distribution for Amberlyst-15-mediated IMA.
Entry
1
b-Hydroxyenone
THPs
dr
Yield [%]
75
–
7:3
2
10:2:3
90
3
4
5
–
–
–
93:7
84
75
70
–
–
12:1:1
12:1:1
[a] syn or anti annotation assigned anticlockwise according to Cahn–Ingold–Prelog priorities.
at the 5-position. For the minor anti,syn,anti-diastereomer,
an enolisation process had taken place at the 3-position,
which meant that the syn relationship of the starting materi-
al was no longer present in the cyclic products. For the third
anti,anti,syn-diastereomer, the protonolysis step had induced
an axial methyl group at the 5-position. The acid-mediated
ring closure of anti-1k led to two diastereomers in a 93:7
ratio. The major diastereomer, anti,syn-2k, featured an axial
methyl group at the 6-position, whereas the minor product
was the all-equatorial anti,anti-2k (entry 3). Both diaste-
stirred at room temperature for 3 h (Table 5). The Lewis
acid mediated ring closure of syn-1k occurred in 76% yield,
with an excellent level of diastereocontrol. Indeed, both dia-
stereomers formed featured an equatorial 6-methyl group.
The anti,anti-2k byproduct was likely the result of an enoli-
sation process occurring either before and/or after cyclisa-
tion (Table 5, entry 1). Compound anti-1k cyclised in the
presence of the Lewis acid to give the corresponding THPs
anti,syn-2k and anti,anti-2k in 70% overall yield and in a
96:4 ratio. For both diastereomers, the relative stereochem-
istry of the starting material was faithfully translated into
the cyclic products, with the main diastereomer featuring an
axial 6-methyl group and the minor diastereomer an equato-
rial 6-methyl group. A point to note is that the same sense
of diastereocontrol upon ring closure was obtained for both
the Brønsted and Lewis acid mediated reactions. We also at-
tempted the ring closure of syn-1k in the presence of the
Lewis acid TMSOTf and iPr2NEt.[17] To prevent decomposi-
tion and dehydration, this IMA reaction had to be conduct-
ed at lowtemperature and submolar concentrations. The
major all-syn 2k THP product was formed after prolonged
reaction times in a lower chemical yield (50%) and with a
diastereomeric ratio of 91:9. Subsequent cyclisations were
therefore not performed under these conditions.
A
tography. Upon cyclisation of aldol anti-1l, the resulting re-
action mixture was composed of an inseparable mixture of
diastereomers formed in a 12:1:1 ratio. Only the major dia-
stereomer could be identified irrefutably as the all equatori-
al THP anti,syn,anti-2l. Similar results were obtained with
aldol anti-1m. The stereochemistry of all THPs was unam-
biguously assigned on the basis of their NMR coupling con-
stants and NOE experiments.[SI]
Lewis acid mediated IMA: The level of stereochemical in-
tegrity and control upon ring closure obtained for the Al-
ACHTREUNG
Chem. Eur. J. 2006, 12, 7190 – 7203
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7195

V. Gouverneur et al.
Table 5. Diastereomeric distribution for Lewis acid mediated IMA.
Entry
1
Starting material
THPs
Yield [%]
dr
76
70
95:5
2
96:4
PdII-mediated IMA: The PdII-mediated nonoxidative cycli-
sation was carried out on b-hydroxyenone substrates syn-
and anti-1k–m by using catalyst C. Reactions reached com-
pletion after 3 h of stirring for syn- and anti-1k but required
overnight stirring for syn- and anti-1l–m (Table 6). All cyclic
products 2k–m were formed in good overall yields, ranging
from 60 to 78%. The palladium-mediated ring closure of
syn-1k occurred with a high level of diastereocontrol, with
the major diastereomer syn,syn-2k and minor diastereomer
syn,anti-2k formed in a 96:4 ratio (Table 6, entry 1). The rel-
ative syn-stereochemistry of the starting material was re-
tained for both diastereomers with the major diastereomer
featuring an equatorial 6-methyl group and the minor dia-
stereomer an axial 6-methyl group. The same sense and
level of control was observed for the ring closure of syn-
aldols 1l–m, with the major diastereomers possessing an
equatorial 6-methyl group and the minor diastereomer syn,
syn,anti-2l–m featuring an axial 6-methyl group (entries 2–
3). The all-syn tetrasubstituted 2l–m THP products were the
result of the protonolysis step placing the 5-methyl group
axially, instead of its more common equatorial position. In
contrast to the Brønsted and Lewis acid mediated IMA re-
actions, the relative syn stereochemistry of the starting
aldols was translated faithfully into each of the diastereo-
meric cyclic products, indicating that no enolisation process
occurred for these palladium-mediated reactions. In the anti
series, the level and sense of diastereocontrol for the ring
closure of anti-1k were similar to the Lewis or Brønsted
acid mediated reactions. The major diastereomer anti,syn-
2k, featuring an axial 6-methyl group, and the minor all-
equatorial anti,anti-2k diastereomer were formed in a ratio
of 97:3 (entry 4). The cyclisations of anti-1l–m gave a mix-
ture of three diastereomers formed in equal proportions
with an overall yield of 63 and 70%, respectively. Only the
all-equatorial anti,syn,anti-2l–m could be isolated in pure
form (entries 5–6). The level of control for the ring closure
event of the anti-enone precursors containing a trisubstitut-
ed alkene group was lower than for that of anti-1k. The dia-
stereomeric products, featuring an axial 6-methyl group,
were formed in a 2:1 ratio with respect to anti,syn,anti-2l–m,
featuring an equatorial 6-methyl group.
Stereochemical stability: To account for the stereochemical
outcome of the various ring closures, it was necessary to de-
termine whether these cyclisations were under kinetic or
thermodynamic control. We therefore tested the stereo-
chemical stability of the 2k–l THPs under the reaction con-
ditions upon extended reaction times in the presence of cat-
alysts A–C (Table 7). We found that syn,syn-2k fully epimer-
ised in the presence of A or B to give the all-equatorial
THP anti,anti-2k (entries 1–2). This suggests that for these
two catalysts enolisation proceeded easily at room tempera-
ture to give the thermodynamic product. In contrast, we
found that syn,syn-2k was perfectly stable when stirred in
the presence of catalyst C after three days at room tempera-
ture (entry 3). This indicates that this catalyst was not able
to coordinate to the carbonyl group to such an extent that
enolisation occurred and therefore epimerisation of the ste-
A
Similarly, tetrasubstituted syn,anti,anti-2l was converted into
the all-equatorial anti,syn,anti-2l (dr=99:1; dr=diastereo-
meric ratio) when resubjected to the reaction conditions in
the presence of A (entry 4), but was found to be stable in
the presence of catalyst C (entry 5). When anti,syn-2k was
treated for an extended reaction time in the presence of cat-
alysts B and C, unchanged starting material was recovered
(entry 7–8). However, in the presence of A, epimerisation at
the C6 carbon occurred after six days (entry 6). As the ki-
netic product was converted into the more stable all-equato-
rial product, this result indicates that retrocyclisation can
occur under these conditions. Similarly, the mixture of the
three diastereomeric tetrahydropyranones 2l remained un-
changed in the presence of catalyst C, but in the presence of
A, it equilibrated after three days at room temperature to
exclusively give the more stable all-equatorial anti,syn,anti-
2l (dr=99:1). These experiments allowed us to conclude
that the palladium and aluminium-mediated cyclisations are
both irreversible under the given reaction conditions. The
7196
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 7190 – 7203

Intramolecular Hetero-Michael Addition
FULL PAPER
Table 6. Diastereomeric distribution for PdII-mediated IMA.
Entry
1
Starting material
THPs
dr
Yield [%]
78
–
96:4
2
87:10:3
68
3
4
5
6
88:10:2
97:3
60
75
63
70
–
1:1:1
1:1:1
palladium-catalyzed ring clo-
sures were sufficiently mild to
prevent enolisation and epi-
merisation at the 3- and 5-posi-
tions, which was not the case
Table 7. Stereochemical stability to reaction conditions.
for Al(ClO4)3·9H2O. The use
G
of Amberlyst-15, however, was
probably narrower in scope, as
this catalyst promoted both
enolisation and retrocyclisa-
tions, consequently leading to
the formation of the most
stable product upon extended
reaction times.
Entry
1
Conditions[a]
I
Product
Major
diastereomer
Conditions[a]
II
dr
Major
diastereomer
distribution[b]
1
2
3
4
5
6
7
8
syn-1k
A, 5 h, RT
B, 3 h, RT
C, 3 h, RT
A, 5 h, RT
C, 18 h, RT
A, 5 h, RT
B, 3 h, RT
C, 3 h, RT
A, 18 h, RT
C, 18 h, RT
7:3
95:5
syn,syn
syn,syn
syn,syn
syn,anti,anti
syn,anti,anti
anti,syn
anti,syn
anti,syn
A, 3 d, RT
B, 2 d, RT
C, 3 d, RT
A, 3 d, RT
C, 2 d, RT
A, 6 d, RT
B, 2 d, RT
C, 2 d, RT
A, 3 d, RT
C, 3 d, RT
99:1
99:1
99:1
99:1
99:1
99:1
99:1
99:1
99:1
1:1:1
anti,anti
anti,anti
syn,syn
anti,syn,anti
syn,anti,anti
anti,anti
anti,syn
anti,syn
96:4 (99:1)
10:2:3 (9:1)
87:10:3 (99:1)
93:7 (99:1)
96:4 (99:1)
97:3 (99:1)
1:1:1
syn-1l
anti-1k
Intramolecular Michael addi-
tion of a representative enan-
tiopure b-hydroxyenone: To
further probe the synthetic
scope of these IMA, we stud-
9
10
anti-1l
anti,syn,anti
anti,syn,anti
anti,syn,anti
anti,syn,anti
1:1:1
[a] A=Amberlyst-15, CH2Cl2, RT; B=Al
[b] Number in brackets indicates dr with respect to major diastereomer after purification. Subsequent stability
tests were carried out on diastereomerically pure cyclic compounds.
A
A
Chem. Eur. J. 2006, 12, 7190 – 7203
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7197

V. Gouverneur et al.
ied the cyclisation of the enantioenriched aldol precursor
syn-1k (Scheme 4). This compound was prepared in 94% ee
(ee=enantiomeric excess) and dr 9:1 by using the chiral
boron enolate derived from (+)-Ipc2BCl.[11,17] The aldol syn-
(2R,3R)-1k underwent ring closure with the preferential for-
or enol.[19] Amberlyst-15, the oxophilic Lewis acid Al-
AHCTREUNG
A
ACHTREUNG
mechanistic pathway. The alternative mechanism, usually
more characteristic of palladium-mediated cyclisations in-
volving neutral catalysts of weaker Lewis acidity is initiated
by direct coordination of the olefinic double bond to the
metal catalyst leading to an activated p-complex. This com-
plex is subsequently subjected to a regioselective 6-endo-trig
ring closure. The resulting s-alkyl metal intermediate then
undergoes protonolysis to give a tetrahydropyranone prod-
uct with concomitant regeneration of the palladium(ii) cata-
lytic species.[11] It is not clear whether the protonolysis in-
À
volves direct s-C Pd bond cleavage or transits through a
[13]
À
Pd O enolate intermediate. This mechanistic pathway is
À
only possible when the protonolysis of the s Pd bond is
faster than b-hydride elimination. This competing pathway
leads to a dihydropyranone, a product never observed in our
crude mixtures. With [Pd
N
N
AHCTREUNG
probably suppressed due to the presence of acetonitrile li-
gands for these catalysts. These strongly coordinating ligands
assure the unavailability of a vacant or weakly coordinated
site cis to the Pd-alkyl bond essential for the b-hydride elim-
ination process to take place. It is also possible that the car-
bonyl group further stabilizes the s-alkyl palladium inter-
Scheme 4. Asymmetric synthesis of (2R,3R,6R)-2-phenyl-3,6-dimethylte-
trahydropyran-4-one 2k.
mation of the all-syn 2k prod-
uct. The enantiomeric excess
of the all-syn (2R,3R,6R)-2k
THP was determined to be
94% by HPLC analysis using
chiral stationary phases. It
therefore mirrored exactly the
enantiopurity of the starting
material. This information al-
lowed us to conclude that the
chemistry outlined herein will
be directly applicable to the
preparation of enantioenriched
targets. The absolute configu-
Scheme 5. C=C versus C=O coordination pathways.
ration of (2R,3R,6R)-2k was
assigned by comparison with
published aD values.[3]
mediate towards elimination by further coordination with
the cationic metal.
Mechanism of the nonoxidative ring closures: The mechanis-
tic pathway of the nonoxidative cyclisations reported herein
involves initial activation of the b-hydroxyenone by coordi-
nation to the different catalysts. Two modes of activation
can operate for these enones with complexation of the cata-
lyst to either the C=O or C=C functional groups
(Scheme 5).[18] Both modes of coordination enhance the
electrophilicity of the double bond, thereby facilitating nu-
cleophilic attack by the hydroxy group. The C=O mode of
activation involves the reversible formation of a transient
carbonyl-metal complex or protonated ketone, followed by
ring closure and protonolysis of the resulting metal enolate
We have several factual pieces of evidence to support the
hypothesis that the cationic palladium catalyst [Pd-
A
N
AHCTREUNG
gest similar mechanistic pathways. An additional indication
arises from the difference of reactivity of aldol 1 f when
using the cationic palladium catalyst [Pd
N
N
under reaction conditions favouring the nonoxidative ring
closure pathway or using the neutral catalyst PdCl2 under
aerobic conditions leading to the oxidised dihydropyranone.
Indeed, aldol 1 f was successfully converted to the tetrahy-
7198
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 7190 – 7203

Intramolecular Hetero-Michael Addition
FULL PAPER
dropyranone 2 f by using [Pd
A
(BF4)2, but did not
liver the cyclised products as efficiently as the palladium(ii)
catalyst. This was due to the occurrence of competing dehy-
dration, epimerisation and retrocyclisation processes. Fur-
ther experiments and theoretical calculations are ongoing in
our laboratories to establish more clearly the exact mecha-
nism for these novel transformations.
lead to the desired dihydropyranone 4 f in the presence of
PdCl2. The lack of reactivity with PdCl2 indicates that the
presence of the methyl group on the nonterminal double
bond carbon hampered activation of the substrate by C=C
coordination. Also, the ring closure would involve the for-
À
mation of a highly unfavourable quaternary s-Pd C centre.
The successful cyclisation observed in the presence of [Pd-
Stereochemical outcome of the nonoxidative ring closures:
The overall stereochemical outcome for the intramolecular
Michael additions reported herein are the result of two dis-
tinct steps, a cyclisation event and a protonolysis which
define the stereochemistry at C6 and C5, respectively. The
cyclisations initiated with Amberlyst-15 are reversible and
susceptible to enolisation. They gave the all-equatorial prod-
ucts after extended reaction times, as anticipated for a ther-
modynamically controlled reaction. This discussion will
hence be solely focused on the kinetically controlled Al-
and Pd-mediated cyclisations and will attempt to rationalize
the stereochemical outcome of the most successful ring clo-
sures. The most distinctive result is the impact of the relative
stereochemistry of the starting aldol on the stereochemistry
A
N
activates the enone by the carbonyl group forming a palladi-
um-enolate intermediate, which could subsequently undergo
protonolysis (Scheme 6). Finally, literature data from the
of the cyclisation process catalyzed by Al
A
[Pd(MeCN)4](BF4)2. Indeed, we found that the syn-aldols
A
ACHTREUNG
led to 2,6-syn products with an equatorial substituent at the
6-position. The anti-aldols favoured the formation of 2,6-anti
tetrahydropyranones featuring an axial substituent at this
same position. For all aldol precursors, the preferential reac-
tive conformation of the enone group is assumed to be s-
trans, as this conformer enables the electrophilic centre to
be proximal to the nucleophile, thereby enabling ring clo-
sure.[22] For syn-aldol 1k, activated upon carbonyl complexa-
tion by the metal, the major diastereomer results from an
attack of the hydroxy group on the Cb-Si face of the enone.
The model proposed to account for this result places the
methyl and phenyl groups positioned on the stereogenic
centres in a preferential staggered gauche or synclinal con-
formation (Scheme 7). The attack of the hydroxy group on
the opposite Cb-Re face generated unfavourable steric inter-
actions between the methyl group positioned a to the car-
bonyl and the b-hydrogen placed on the terminal double
bond carbon. The sense and level of diastereocontrol for the
cyclisation event is very similar for syn-aldol 1k featuring an
E-trisubstituted alkene (Scheme 8). This suggests that the
presence of the additional methyl group on the double bond
Scheme 6. Difference in reactivity for substrate 1 f.
group of Oi et al.[20] suggested a similar C=O mode of acti-
vation for the hetero-Diels–Alder reaction of unactivated
dienes and aldehydes in the presence of [Pd
They showed that, in contrast, the milder neutral catalyst
[PdCl2(MeCN)2] was unable to promote these cycloaddi-
G
A
ACHTREUNG
tions, therefore supporting the hypothesis that this neutral
catalyst triggers the nonoxidative cyclisations reported
herein by C=C activation. For the palladium(ii)-mediated
ring closures, we also considered a third coordination mech-
anism involving intramolecular suprafacial addition of an
alkoxy-palladium intermediate.[21] However, this reaction
mechanism was discarded due to the highly strained nature
of the transition state resulting from an endocyclic ring clo-
sure.
The results of Spencer et al.[13] for the intermolecular Mi-
chael addition suggest that the mechanism of cationic palla-
dium(ii)-mediated intermolecu-
lar hetero-Michael additions
might take place according to
a reaction pathway that in-
volves catalytic activation of
the carbonyl group by in situ
formed Brønsted acids. Based
on our preliminary screening
results, we believe that this is
probably not the case for the
corresponding intramolecular
oxy-Michael additions, because
the Brønsted acids failed to de-
Scheme 7. Cyclisation of syn-aldol 1k.
Chem. Eur. J. 2006, 12, 7190 – 7203
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7199

V. Gouverneur et al.
Scheme 10. Cyclisation of anti-aldol 1l.
Scheme 8. Cyclisation of syn-aldol 1l.
has little influence on the stereochemical outcome of the
cyclisation. For this substrate, it is likely that the Al- or Pd-
based catalyst coordinates to the carbonyl lone pair anti to
the double bond. The ring closure led to a mixture of three
diastereomers in a ratio of 87:10:3, due to additional compli-
cations resulting from the directionality of the protonolysis
step. The two major diastereomers have their 6-methyl
group equatorial but are epimers at C5.
For the anti-aldol 1k, the methyl and phenyl groups
placed on the stereogenic centres are also believed to adopt
a preferential staggered synclinal conformation, nowplacing
the methyl group axially. Upon cyclisation, the predominant
stereoisomer is the result of an attack of the hydroxy group
on the Cb-Re face of the enone, as this attack minimizes
steric interactions with the methylated stereogenic centre
(Scheme 9). This attack leads to the preferential formation
Conclusion
In this report, several issues were under consideration. First
was the critical issue as to whether or not an intramolecular
Michael addition of b-hydroxyenones was feasible as an
entry to the corresponding tetrahydropyranones. Secondly,
the possibility of preserving existing stereocentres in the
starting enones and controlling simultaneously the diaster-
eoselectivity of the Michael addition had to be determined
for the formation of tetrahydropyranones possessing up to
four stereocentres. Several catalysts were investigated and
this work has clearly identified Al
A
[Pd(MeCN)4](BF4)2 as the most effective catalysts for these
G
ACHTREUNG
transformations. They undergo clean ring closure with no
detectable side products resulting from dehydration or ret-
roaldolisation. We also found that both catalysts display a
similar level and sense of dia-
stereocontrol for the ring clo-
sure of both syn- and anti-
aldol precursors. An important
difference between the two
catalysts is the stereochemical
stability of the cyclised product
under the reaction conditions:
no enolisation occurred for
[Pd
A
ACHTREUNG
presence of Al
AHCTREUNG
keto-enolisation process took
place. The enolisation led to
epimerisation of the centres a
Scheme 9. Cyclisation of anti-aldol 1k.
to the carbonyl in favour of
of tetrahydropyranone anti,syn-2k featuring an axial 6-
methyl group. The minor diastereomer resulting from an
attack on the opposite face of the enone led to the thermo-
dynamically more stable product with all substituents equa-
torial. The level of diastereocontrol for the ring closure
process drops significantly when an additional methyl group
is present on the nonterminal double bond carbon, as illus-
trated with the cyclisation of aldol anti-1l. This is likely the
result of conflicting steric interactions, that is, whether the
attack of the hydroxy group takes place on one or the other
face of the enone (Scheme 10). Rigorous theoretical studies
are needed to confirm the above proposed models to ex-
plain the origin of stereoinduction.
the thermodynamically more stable product. The data also
suggest that when enantiopure aldols are used as precursors
for the cyclisation process, no racemisation occurs. The high
level diasterocontrol upon ring closure, the mild reaction
conditions and the lack of additives and preactivation of the
nucleophile are some of the attractive features of the dica-
tionic-palladium-mediated transformation. Two potential
mechanisms are outlined, involving either C=O or C=C co-
ordination by the catalyst. A detailed mechanistic investiga-
tion remains to be carried out, but the present data have al-
lowed us to hypothesise that carbonyl coordination is likely
to take place for the cationic palladium catalyst.
7200
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 7190 – 7203

Intramolecular Hetero-Michael Addition
FULL PAPER
(1.02 mL, 10 mmol) in anhydrous CH2Cl2 (14 mL). Purification by
column chromatography (EtOAc/hexane 1:5) gave syn-1l (0.45 g,
2 mmol) as a colourless oil in 50% yield. dr=99:1; 1H NMR (CDCl3):
d=1.07 (d, J=7.3 Hz, 3H), 1.78 (s, 3H), 1.88 (d, J=6.8 Hz, 3H), 3.44
(qd, J=7.1, 3.1 Hz, 1H), 3.89 (brs, 1H), 5.05 (d, J=3.1 Hz, 1H), 6.77 (q,
J=6.8 Hz, 1H), 7.24–7.37 ppm (m, 5H); 13C NMR (CDCl3): d=11.0,
11.5, 15.0, 45.2, 73.4, 126.0, 127.2, 128.1, 137.2, 139.1, 142.0, 207.4 ppm;
IR (film): n˜ =1644 cmÀ1; HRMS: m/z: calcd for C14H18O2: 218.1307;
found: 218.1313.
The traditional routes towards trisubstituted THPs, based
on HDA, acid-mediated or Prins cyclisations, provide 2,6-
syn-substituted THP products. The novel palladium(ii)-medi-
ated stepwise route reported herein allows for the formation
of 2,6-syn THPs, when employing syn-aldol precursors and
of 2,6-anti THPs when using anti-aldols. Stereochemically
stable tetrasubstituted THPs were also formed with high di-
astereoselectivies when starting from a syn-aldol featuring a
trisubstituted E-alkene functional group.
anti-1-Hydroxy-2,4-dimethyl-1-phenylhex-4-en-3-one (anti-1l): Colour-
less oil; 40% yield; dr=9:1; 1H NMR (CDCl3): d=1.05 (d, J=7.1 Hz,
3H), 1.82 (s, 3H), 1.91 (d, J=6.8 Hz, 3H), 3.12 (d, J=5.0 Hz, 1H), 3.63
(q, J=7.1 Hz, 1H), 4.91 (dd, J=7.3, 5.0 Hz, 1H), 6.82 (q, J=6.8 Hz, 1H),
7.30–7.42 ppm (m, 5H); 13C NMR (CDCl3): d=11.3, 12.2, 16.5, 46.4, 77.1,
126.8, 127.9, 128.6, 138.5, 138.8, 142.8, 206.4 ppm; IR (film): n˜ =
Experimental Section
1644 cmÀ1
.
General Methods: The infrared spectra were recorded on a Paragon 1000
FTIR spectrometer and only peaks of interest are reported. The 1H and
13C NMR spectra were recorded on Bruker DQX400 and DPX400 spec-
trometers at 400 and 100 MHz, respectively; the chemical shifts are re-
ported in ppm downfield relative to CDCl3 (d=7.26 ppm) for 1H spectra
and relative to the central CDCl3 (d=77 ppm) for 13C NMR spectra.
Coupling constants in 1H NMR measurements are in Hz. Mass spectra
were recorded on Micromass GCT in Chemical Ionisation (NH3, CI+) or
on a AutoSpec-oaTof instruments (CI+). The enantiomeric excess (ee) of
the products was determined by a Waters HPLC using Chiralcel OD or
OJ-H chiral columns with ethanol/hexane as eluent.
syn-1-Hydroxy-2,4-dimethyl-1-phenylhept-4-en-3-one (syn-1m): Colour-
less oil; 67% yield; dr=99:1; 1H NMR (CDCl3): d=1.07 (t, J=7.6 Hz,
3H), 1.08 (d, J=7.6 Hz, 3H), 1.77 (s, 3H), 2.27 (quin, J=7.6 Hz, 2H),
3.45 (qd, J=7.1, 3.0 Hz, 1H), 3.83 (brs, 1H), 5.06 (d, J=3.0 Hz, 1H),
6.62 (t, J=7.4 Hz, 1H), 7.24–7.37 ppm (m, 5H); 13C NMR (CDCl3): d=
11.2, 11.6, 13.0, 22.6, 45.2, 73.5, 126.0, 127.2, 128.2, 135.6, 142.0, 146.0,
207.7 ppm; IR (film): n˜ =1652 cmÀ1; HRMS: m/z: calcd for C15H20O2:
232.1463; found: 232.1464.
anti-1-Hydroxy-2,4-dimethyl-1-phenylhept-4-en-3-one (anti-1m): Colour-
less oil; 42% yield; dr=3:1; 1H NMR (CDCl3): d=1.02 (d, J=7.3 Hz,
3H), 1.05 (t, J=7.6 Hz, 3H), 1.77 (s, 3H), 2.24 (quin, J=7.4 Hz, 2H),
3.05 (brs, 1H), 3.59 (quin, J=7.2 Hz, 1H), 4.87 (d, J=7.6 Hz, 1H), 6.60
(t, J=7.4 Hz, 1H), 7.26–7.37 ppm (m, 5H); 13C NMR (CDCl3): d=11.2,
13.0, 16.4, 22.5, 46.2, 76.8, 126.5, 127.7, 128.4, 136.5, 142.6, 145.6,
206.5 ppm; IR (film): n˜ =1653 cmÀ1; HRMS: m/z: calcd for C15H20O2:
232.1463; found: 232.1463.
Materials: Compounds 1a–b, d–f, k and 2b,e have been previously pub-
lished.[4,9,11,24] The remaining b-hydroxyenones were synthesised accord-
ing to adapted versions of our previously reported aldol procedures.[9,11]
X-ray crystallography data was obtained for compounds 2g and anti,anti-
2k. CCDC-602788 and -602789 contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of charge
from the Cambridge Data Centre via www.ccdc.cam.ac.uk/data_request/
cif.
Typical procedure for the catalysed intramolecular Michael addition: The
substrate dissolved in dry CH2Cl2 or MeCN (2 mLmmolÀ1), respectively,
was added to a solution of the catalyst (8 or 10 mol%) in anhydrous
CH2Cl2 (2 mLmmolÀ1, for PdII and Amberlyst-15) or anhydrous MeCN
(2 mLmmolÀ1, for Lewis acids) under an argon atmosphere. The reaction
was left to stir until completion at RT, as followed by TLC. The reaction
mixture was then diluted with Et2O and filtered through a pad of silica
gel. The solvent was removed in vacuo to give the crude product, which
was further purified by column chromatography on silica.
1-(4-Bromophenyl)-1-hydroxy-5-methylhex-4-en-3-one
(1c):
Yellow
solid; 73% yield; 1H NMR (CDCl3): d=1.91 (s, 3H), 2.18 (s, 3H), 2.78
(d, J=5.1 Hz, 2H), 3.88 (d, J=3.0 Hz, 1H), 5.12 (s, 1H), 6.03 (s, 1H),
7.25 (d, J=8.8 Hz, 2H), 7.47 ppm (d, 2H); 13C NMR (CDCl3): d=21.1,
27.9 51.9, 69.6, 123.5, 127.4, 128.3, 131.3, 131.5, 142.1, 158.0, 142.0, 157.1,
200.3 ppm; IR (film): n˜ =1681cmÀ1
.
1-Hydroxy-4-methyl-1-phenyl-hex-4-en-3-one (1g):[23] Colourless oil;
96% yield; 1H NMR (CDCl3): d=1. 80 (s, 3H), 1.86 (d, J=7.2 Hz, 3H),
3.04 (d, J=8.5 Hz, 1H), 3.05 (d, J=3.7 Hz, 1H), 5.19 (dd, J=8.5, 3.7 Hz,
1H), 6.76 (q, J=6.8 Hz, 1H), 7.26–7.41 ppm (m, 5H); 13C NMR (CDCl3):
d=10.8, 14.9, 45.8, 70.3, 125.7, 127.5, 128.5, 138.4, 139.2, 143.2,
2,2-Dimethyl-6-phenyl-tetrahydropyranone-4-one (2a): Colourless oil;
1H NMR (CDCl3): d=1.32 (s, 3H), 1.48 (s, 3H), 2.39 (dd, J=13.6,
2.0 Hz, 1H), 2.52 (dd, J=13.6, 10.6 Hz, 1H), 2.54–2.60 (m, 2H), 4.91 (dd,
J=10.6, 3.5 Hz, 1H), 7.29–7.43 ppm (m, 5H); 13C NMR (CDCl3) d=24.0,
31.1, 49.5, 53.2, 73.0, 76.0, 125.8, 128.0, 128.6, 141.4, 207.5 ppm; IR (film):
201.6 ppm; IR (film): n˜ =1656cmÀ1
.
n˜ =1716 cmÀ1
; HRMS: m/z: calcd for C13H17O2: 205.1229; found:
7-Benzyloxy-6-hydroxy-2,6-dimethylhept-2-en-4-one (1h): Colourless oil;
1
205.1233.
98% yield; H NMR (CDCl3): d=1.26 (s, 3H), 1.90 (s, 3H), 2.15 (s, 3H),
2.54 (d, J=16.4 Hz, 1H), 2.82 (d, J=16.4 Hz, 1H), 3.35 (d, J=9.1 Hz,
1H), 3.42 (d, J=9.1 Hz, 1H), 4.54 (s, 2H), 7.28–7.37 ppm (m, 5H);
13C NMR (CDCl3): d=20.9, 25.2, 27.8, 50.5, 72.0, 73.4, 127.4, 127.6, 128.4,
6-(4-Bromophenyl)-2,2-dimethyl-tetrahydropyran-4-one (2c): Yellow
solid; H NMR (CDCl3): d=1.30 (s, 3H), 1.47 (s, 3H), 2.42 (m, 2H), 2.54
1
(d, J=14.2 Hz, 2H), 4.87 (dd, J=11.4, 3.0 Hz, 1H), 7.28 (d, J=8.6 Hz,
2H), 7.51 ppm (d, J=8.6 Hz, 2H); 13C NMR (CDCl3): d=24.0, 31.0, 49.3,
53.0, 72.3, 75.8, 121.7, 127.5, 131.7, 140.4, 206.9 ppm; IR (film): n˜ =
138.3, 157.0, 201.9 ppm; IR (film): n˜ =1676, 1615 cmÀ1
.
1-(1-Hydroxycyclohexyl)-4-pent-3-en-2-one (1i):[4] Colourless oil; 97%
yield; 1H NMR (CDCl3): d=1.20–1.78 (m, 10H), 1.89 (s, 3H), 2.15 (s,
3H), 2.56 (s, 2H), 4.07 (s, 1H), 6.04 ppm (q, J=1.26 Hz, 1H); 13C NMR
(CDCl3): d=20.9, 22.0, 25.8, 27.8, 37.7, 52.9, 71.1, 124.8, 157.0,
1717 cmÀ1
300.0542.
; HRMS: m/z: calcd for C13H19NO2Br: 300.0599; found:
2-Methyl-6-phenylethyl-tetrahydropyran-4-one (2d):[24] syn-2d: 1H NMR
(CDCl3): d=1.37 (d, J=5.8 Hz, 3H), 1.81 (dddd, J=13.7, 9.2, 7.5, 4.4 Hz,
1H), 2.14 (dtd, J=13.9, 8.8, 5.4 Hz, 1H), 2.20–2.40 (m, 4H), 2.74 (ddd,
J=14.1, 8.8, 7.5 Hz, 1H), 2.83 (ddd, J=14.0, 8.2, 5.2 Hz, 1H), 3.52–3.59
(m, 1H), 3.71 (dqd, J=12.1, 5.8, 2.5 Hz, 1H), 7.17–7.33 ppm (m, 5H);
13C NMR (CDCl3): d=22.1, 31.5, 37.9, 47.5, 49.4, 73.1, 75.8, 126.0, 128.4,
128.5, 141.5, 207.4 ppm; anti-2d: 1H NMR (CDCl3): d=1.28 (d, J=
6.4 Hz, 3H), 1.74 (dddd, J=13.9, 9.2, 7.2, 4.4 Hz, 1H), 1.99 (dtd, J=13.9,
9.2, 5.4 Hz, 1H), 2.23–2.31 (m, 2H), 2.52–2.59 (m, 2H), 2.68 (ddd, J=
14.1, 8.8, 7.5 Hz, 1H), 2.74 (ddd, J=14.0, 8.2, 5.2 Hz, 1H), 4.12 (dtd, J=
14.0, 4.7, 1.3 Hz, 1H), 4.32 (dqd, J=11.3, 6.5, 1.3 Hz, 1H), 7.17–7.33 ppm
(m, 5H); 13C NMR (CDCl3): d=20.7, 31.6, 35.9, 46.9, 48.4, 68.2, 71.1,
202.4 ppm; IR (film): n˜ =1675, 1614cmÀ1
; HRMS: m/z: calcd for
C12H19O: 179.1436; found: 179.1429.
7-Hydroxy-9-phenylnon-3-yn-5-one (1j): Colourless oil; 74% yield;
1H NMR (CDCl3): d=1.22 (t, J=7.4 Hz, 3H), 1.72 (m, 1H), 1.84 (m,
1H), 2.39 (q, J=7.4 Hz, 2H), 2.67–2.86 (m, 4H), 4.13 (m, 1H), 7.18–
7.32 ppm (m, 5H); 13C NMR (CDCl3): d=12.6, 12.7, 31.7, 37.9, 52.2, 66.8,
80.3, 96.7, 125.9, 128.4, 128.5, 141.7, 187.7 ppm; IR (film): n˜ =2210, 1667,
1603 cmÀ1; HRMS: m/z: calcd for C15H19O2: 231.1385; found: 231.1386.
syn-1-Hydroxy-2,4-dimethyl-1-phenylhex-4-en-3-one (syn-1l): From di-
isopropylethylamine (1.1 mL, 7.34 mmol), di-n-butylborontriflate (5 mL,
5 mmol), (E)-4-methylhex-4-en-3-one (0.45 g 4 mmol) and benzaldehyde
Chem. Eur. J. 2006, 12, 7190 – 7203
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7201

V. Gouverneur et al.
126.0, 128.4, 128.5, 141.3, 207.5 ppm; IR (film): n˜ =1723cmÀ1; HRMS:
m/z: calcd for C14H19O2: 219.1385; found: 219.1385.
212.8 ppm; IR (film) n˜ =1714 cmÀ1; HRMS: m/z: calcd for C14H19O2:
219.1385; found: 219.1388.
anti,syn,anti-2-Phenyl-3,5,6-trimethyl-tetrahydropyran-4-one
(anti,syn,
5-Methyl-2-phenylethyl-tetrahydropyran-4-one (2 f): Colourless oil; 58%
yield; dr=1:1; 1H NMR (CDCl3): d=0.97 (d, J=6.8 Hz, 3H), 1.78–2.04
(m, 2H), 2.36–2.40 (m, 4H), 2.63 (m, 1H), 2.72 (ddd, J=14.0, 6.8, 2.0 Hz,
1H), 2.81 (ddd, J=14.1, 9.3, 5.1 Hz, 1H), 3.26 (dd, J=11.4 Hz, 1H),
3.53–3.60 (m, 1H), 4.24 (dd, J=11.4, 7.0 Hz, 1H), 7.17–7.31 ppm (m,
5H); 13C NMR (CDCl3): d=8.9, 31.7, 38.0, 45.4, 48.2, 73.1, 78.1, 126.0,
128.3, 128.4, 141.4, 208.4 ppm; IR (film1): n˜ =1715cmÀ1; HRMS: m/z:
calcd for C14H22NO2: 236.1651; found: 236.1648.
anti-2l): Colourless oil; 1H NMR (CDCl3): d=0.80 (d, J=6.6 Hz, 3H),
1.07 (d, J=6.6 Hz, 3H), 1.41 (d, J=6.0 Hz, 3H), 2.50 (dq, J=10.0,
6.6 Hz, 1H), 2.71 (dq, J=10.4, 6.6 Hz, 1H), 3.51 (dq, J=10.1, 6.0 Hz,
1H), 4.17 (d, J=10.4 Hz, 1H), 7.30–7.40 ppm (m, 5H); 13C NMR
(CDCl3): d=9.7, 9.8, 20.7, 51.0, 51.9, 80.0, 86.1, 127.2, 128.3, 128.6, 139.9,
209.7 ppm; IR (film): n˜ =1713 cmÀ1; HRMS: m/z: calcd for C14H19O2:
219.1385; found: 219.1389.
(Æ)-syn,anti,anti-3,5-Dimethyl-6-ethyl-2-phenyl-tetrahydropyran-4-one
(syn,anti,anti-2m): Colourless oil; 1H NMR (CDCl3): d=0.91 (d, J=
7.3 Hz, 3H), 1.04 (d, J=6.6 Hz, 3H), 1.13 (t, J=7.6 Hz, 3H), 1.70 (qd,
J=7.3, 7.0 Hz, 1H), 1.91 (qd, J=7.3, 3.0 Hz, 1H), 2.65 (dq, J=10.3,
6.5 Hz, 1H), 2.74 (qd, J=7.3, 2.8 Hz, 1H), 3.34 (ddd, J=10.3, 7.0, 3.0 Hz,
1H), 4.80 (d, J=2.8 Hz, 1H), 7.26–7.40 ppm (m, 5H); 13C NMR (CDCl3):
d=9.3, 9.4, 11.5, 26.9, 44.8, 51.0, 79.8, 83.8, 125.4, 127.1, 128.2, 139.2,
213.4 ppm; IR (film): n˜ =1714 cmÀ1; HRMS: m/z: calcd for C15H21O2:
233.1542; found: 233.1601.
syn,anti-2,3-Dimehthyl-6-methyl-tetrahydropyran-4-one (2g): Colourless
oil, 1H NMR (CDCl3): d=1.07 (d, J=6.5 Hz, 3H), 1.46 (d, J=6.0 Hz,
3H), 4.68 (t, J=7.2 Hz, 1H), 7.30–7.41 ppm (m, 5H); 13C NMR (CDCl3):
d=9.5, 20.6, 49.7, 51.7, 79.0, 79.5, 125.7, 128.0, 128.6, 140.9, 208.3 ppm;
IR (film): n˜ =1715 cmÀ1; HRMS: m/z: calcd for C13H17O2: 205.1229;
found: 205.1230.
2-(Benzyloxymethyl)-2,6,6-trimethyl-tetrahydropyran-4-one (2h): Colour-
1
less oil, 55% yield; H NMR (CDCl3): d=1.28 (s, 3H), 1.32 (s, 3H), 1.33
(s, 3H), 2.38 (d, J=16.2 Hz, 1H), 2.39 (d, J=16.4 Hz, 1H), 2.52 (d, J=
16.2 Hz, 1H), 2.70 (d, J=16.4 Hz, 1H), 3.26 (d, J=9.3 Hz, 1H), 3.39 (d,
J=9.3 Hz, 1H), 4.54 (d, J=12.1 Hz, 1H), 4.60 (d, J=12.1 Hz, 1H), 7.28–
7.37 ppm (m, 5H); 13C NMR (CDCl3): d=27.0, 30.5, 31.6, 46.4, 51.3, 73.4,
74.5, 76.2, 78.2, 127.4, 127.6, 128.4, 138.2, 208.4 ppm; IR (film): n˜ =
1718 cmÀ1; HRMS: m/z: calcd for C16H23O3: 263.1647; found: 263.1649.
anti,syn,anti-3,5-Dimethyl-6-ethyl-2-phenyl-tetrahydropyran-4-one (anti,
syn,anti-2m): Colourless oil; 1H NMR (CDCl3): d=0.81 (d, J=6.8 Hz,
3H), 1.02 (t, J=7.3 Hz, 3H), 1.04 (d, J=6.6 Hz, 3H), 1.64 (qd, J=7.3,
7.0 Hz, 1H), 1.85 (qd, J=7.3, 3.0 Hz, 1H), 2.59–2.70 (m, 2H), 3.37 (ddd,
J=10.1, 6.8, 3.0 Hz, 1H), 4.15 (d, J=10.1 Hz, 1H), 7.31–7.40 ppm (m,
5H); 13C NMR (CDCl3): d=8.8, 9.3, 9.8, 26.5, 49.2, 51.6, 84.0, 86.0, 127.1,
128.2, 128.4, 140.2, 210.3 ppm; IR (film): n˜ =1713 cmÀ1; HRMS: m/z:
calcd for C15H21O2: 233.1542; found: 233.1601.
2,2-Dimethyl-1-oxa-spiro[5.5]undecan-4-one (2i): Yellowoil; 54% yield;
G
1H NMR (CDCl3): d=1.31 (s, 6H), 1.32–1.49 (m, 6H), 1.66–1.74 (m,
4H), 2.42 (s, 2H), 2.43 ppm (s, 2H); 13C NMR (CDCl3): d=22.4, 25.4,
AHCTREUNG
31.5, 39.9, 49.8, 51.9, 74.6, 76.1, 209.0 ppm; IR (film): n˜ =1720 cmÀ1
HRMS: m/z: calcd for C12H21O2: 197.1542; found: 197.1536.
;
A
ACHTREUNG
ourless oil; 15% yield (2 steps); dr=9:1 and ee=94%, as assigned by
chiral HPLC with [a]2D5 =+27.1 (c=1.02 in CH2Cl2).
6-Ethyl-2-phenylethyl-2,3-dihydropyran-4-one (2j):[11] Pale yellowoil;
60% yield; 1H NMR (CDCl3): d=1.26 (t, J=7.6 Hz, 3H), 1.95 (m, 1H),
2.17 (m, 1H), 2.31 (q, J=7.6 Hz, 2H), 2.38 (dd, J=16.7, 4.0 Hz, 1H),
2.47 (dd, J=16.9, 12.8 Hz, 1H), 2.80 (m, 2H), 4.34 (ddd, J=12.8, 8.3,
4.0 Hz, 1H) 5.34 (s, 1H), 7.19–7.35 ppm (m, 5H); 13C NMR (CDCl3): d=
10.6, 28.0, 31.1, 36.0, 41.0, 78.0, 103.2, 126.2, 128.4, 128.6, 140.8, 178.8,
193.0 ppm; IR (film): n˜ =1666 cmÀ1; HRMS: m/z: calcd for C15H19O2:
231.1385; found: 231.1392.
Acknowledgements
This study was supported in part by the EPSRC (GR/S68095/01), the
Royal Society (Research Grant RSRG 23219B), the European Commun-
ity (COST Action D25), the Leverhulme Trust (F/08 341/B) and the Brit-
ish Council (Chevening Scholarship, M.R.). We also thank Dr. A.R.
Cowley and Dr Barbara Odell (University of Oxford) for providing us
with the X-ray and NOE data respectively. We also thank Dr. Charles
Baker-Glenn and Dr. Sandrine Ropp for preliminary efforts on this proj-
ect. Finally, Dr. John Brown and Professors M. GagnØ and L. Overman
are gratefully acknowledged for spending time with us to discuss the
mechanisms suggested herein.
syn,syn-3,6-Dimethyl-2-phenyl-tetrahydropyran-4-one (syn,syn-2k): Col-
ourless oil; 1H NMR (CDCl3): d=0.91 (d, J=7.3 Hz, 3H), 1.45 (d, J=
6.0 Hz, 3H), 2.33 (dd, J=14.4, 2.8 Hz, 1H), 2.55 (dd, J=14.6, 11.4 Hz,
1H), 2.66 (qdd, J=7.1, 2.7, 1.3 Hz, 1H), 3.91 (dqd, J=11.6, 6.0, 2.8 Hz,
1H), 4.83 (d, J=2.7 Hz, 1H), 7.25–7.41 ppm (m, 5H); 13C NMR (CDCl3):
d=11.3, 22.1, 45.5, 50.7, 73.7, 80.0, 125.5, 127.3, 128.3, 138.7, 211.5 ppm;
IR (film): n˜ =1715 cmÀ1; HRMS: m/z: calcd for C13H17O2: 205.1229;
found 205.1233. Use of (+)-IPc2BCl gave (2R,3R,6R)-2k with dr=99:1
and ee=94%.
anti,anti-3,6-Dimethyl-2-phenyl-tetrahydropyran-4-one
(anti,anti-2k):
Colourless oil; 1H NMR (CDCl3): d=0.81 (dd, J=6.6, 1.5 Hz, 3H), 1.39
(dd, J=6.3, 1.7 Hz, 3H), 2.50–2.53 (m, 2H), 2.63 (dq, J=10.4, 6.6 Hz,
1H), 3.93 (ddd, J=14.1, 11.9, 6.0 Hz, 1H), 4.18 (dd, J=10.4, 1.7 Hz, 1H),
7.31–7.40 ppm (m, 5H); 13C NMR (CDCl3): d=9.4, 22.3, 49.8, 51.2, 74.2,
[1] a) R. Jansen, V. Wray, H. Irschik, H. Reichenbach, G. Hçfle, Tetra-
hedron Lett. 1985, 26, 6031–6034; b) T. L. B. Bolvin, Tetrahedron
1987, 43, 3309–3362; c) D. Schummer, K. Gerth, H. Reichenbach,
G. Hçfle, Liebigs Ann. 1995, 685–688; d) D. A. Evans, P. H. Carter,
E. M. Carreira, A. B. Charette, J. A. Prunet, J. A. M. Lautens,
Angew. Chem. 1998, 110, 2526–2530; Angew. Chem. Int. Ed. 1998,
37, 2354–2359; e) U. Bhatt, M. Christmann, M. Quitschalle, E.
Claus, M. Kalesse, J. Org. Chem. 2001, 66, 1885–1893; f) I. Paterson,
C. De Savi, M. Tudge, Org. Lett. 2001, 3, 3149–3152; g) F. P.
Marmsater, F. G. West, Chem. Eur. J. 2002, 8, 4346–4353.
85.9, 127.2, 128.4, 128.6, 139.8, 208.3 ppm; IR (film): n˜ =1715 cmÀ1
HRMS: m/z: calcd for C13H17O2: 205.1229; found: 205.1228.
;
anti,syn-3,6-Dimethyl-2-phenyl-tetrahydropyran-4-one
(anti,syn-2k):
Colourless oil; 1H NMR (CDCl3): d=0.94 (d, J=6.8 Hz, 3H), 1.32 (d, J=
6.6 Hz, 3H), 2.41 (dd, J=13.9, 3.5 Hz, 1H), 2.77 (dq, J=8.8, 6.8 Hz, 1H),
2.88 (dd, J=13.9, 6.1 Hz, 1H), 4.57 (d, J=8.6 Hz, 1H), 4.57 (m, 1H),
7.31–7.41 ppm (m, 5H); 13C NMR (CDCl3): d=10.9, 19.3, 47.1, 50.7, 70.4,
[2] S. Hanessian, Total Synthesis of Natural Products: The “Chiron” Ap-
proach (Ed. J. E. Baldwin), Pergamon, Oxford, UK, 1983.
79.7, 127.3, 128.3, 128.6, 137.0, 211.0 ppm; IR (film): n˜ =1716 cmÀ1
HRMS: m/z: calcd for C13H17O2: 205.1229; found: 205.1229.
;
[3] a) D. L. Boger, S. M. Weinreb, Hetero Diels–Alder Methodology in
Organic Synthesis, Academic Press, San Diego, CA, 1987; b) M. Jo-
hannsen, K. A. Jorgensen, J. Org. Chem. 1995, 60, 5757; c) S. E.
Schaus, J. Branalt, E. N. Jacobsen J. Org. Chem. 1998, 63, 403;
d) A. G. Dossetter, T. F. Jamison, E. N. Jacobsen, Angew. Chem.
1999, 111, 2549–2552; Angew. Chem. Int. Ed. 1999, 38, 2398–2400;
e) C. F. Thompson, T. F. Jamison, E. N. Jacobsen, J. Am. Chem. Soc.
syn,anti,anti-3,5,6-Trimethyl-2-phenyl-tetrahydropyran-4-one
(syn,an-
1
ti,anti-2l): Colourless oil; H NMR (CDCl3): d=0.93 (d, J=7.1 Hz, 3H),
1.06 (d, J=6.8 Hz, 3H), 1.48 (d, J=6.1 Hz, 3H), 2.55 (dq, J=10.1,
6.6 Hz, 1H), 2.71 (qd, J=7.1, 2.2 Hz, 1H), 3.51 (dq, J=10.4, 6.1 Hz, 1H),
4.83 (d, J=2.8 Hz, 1H), 7.25–7.45 ppm (m, 5H); 13C NMR (CDCl3): d=
9.5, 11.5, 20.6, 47.2, 51.0, 79.5, 80.1, 125.4, 127.2, 128.2, 129.7, 138.8,
7202
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 7190 – 7203

Intramolecular Hetero-Michael Addition
FULL PAPER
2001, 123, 9974–9983; f) I. Paterson, C. A. Luckhurst, Tetrahedron
Lett. 2003, 44, 3749–3754; g) M. Anada, T. Washio, N. Shimada, S.
Kitagaki, M. Nakajima, M. Shiro, S. Hashimoto, Angew. Chem.
2004, 116, 2719–2722; Angew. Chem. Int. Ed. 2004, 43, 2665–2668;
h) E. Ruijter, H. Schueltingerkemper, L A. Wessjohann, J. Org.
Chem. 2005, 70, 2820–2823.
[14] S. Kobayashi, K. Kakumoto, M. Sugiura, Org. Lett. 2002, 4, 1319–
1322.
[15] N. Srivastava, B. K. Banik, J. Org. Chem. 2003, 68, 2109–2114.
[16] a) Z. Wang, Z. Zhang, X. Lu, Organometallics 2000, 19, 775–780;
b) G. Liu, X. Lu, Tetrahedron Lett. 2003, 44, 127–130.
[17] I. Paterson, S. Osborne, Tetrahedron Lett. 1990, 31, 2213–2216.
[18] For the palladium-mediated cyclisation, we cannot rule out a dual
mode of activation of the substrate with simultaneous activation of
both the C=O and C=C bonds. a) S. Ogoshi, T. Yoshida, T. Nishida,
M. Morita, J. Am. Chem. Soc. 2001, 123, 1994; b) M. Morita, K.
Inoue, T. Yoshida, S. Ogoshi, H. Kurosawa, J. Organomet. Chem.
2004, 689, 894–898.
[19] a) T. Pei, R. A. Widenhoefer, Chem. Commun. 2002, 650–651; b) H.
Qian and R. A. Widenhoefer, J. Am. Chem. Soc. 2003, 125, 2056–
2057; c) D. Yang, J-H. Li, Q. Gao, Y.-L. Yan, Org. Lett. 2003, 5,
2869–2871; d) H. Qian, T. Pei, R. A. Widenhoefer, Organometallics
2005, 24, 287–301.
[20] a) S. Oi, K. Kashiwagi, E. Terada, K. Ohuchi, Y. Inoue, Tetrahedron
Lett. 1996, 37, 6351–6354; b) S. Oi, K. Kashiwagi, Y. Inoue, Tetrahe-
dron Lett. 1998, 39, 6253–6256.
[21] a) T. Hayashi, K. Yamasaki, M. Mimura, Y. Uozumi, J. Am. Chem.
Soc. 2004, 126, 3036–3037, and references therein; b) J. P. Wolfe,
M. A. Rossi, J. Am. Chem. Soc. 2004, 126, 1620–1621; c) M. B. Hay,
A. R. Hardin, J. P. Wolfe, J. Org. Chem. 2005, 70, 3099–3107;
d) M. B. Bertrand, J. P. Wolfe, Tetrahedron 2005, 61, 6447–6459;
e) J. S. Nakhla, J. W. Kampf, J. P. Wolfe, J. Am. Chem. Soc. 2006,
128, 2893–2901.
[22] a) D. J. Raber, N. K. Raber, J. Chandrassekhar, P. Von RaguØ
Schleyer, Inorg. Chem. 1984, 23, 4076–4080; b) M. T. Reetz, M.
Hꢁllman, W. Massa, S. Berger, P. Bademacher, P. Heymanns, J. Am.
Chem. Soc. 1986, 108, 2405–2408; c) D. J. Nelson, J. Org. Chem.
1986, 51, 3185–3186; d) R. J. Loncharich, T. R. Schwartz, K. N.
Houk, J. Am. Chem. Soc. 1987, 109, 14–23; e) T. J. Lepage, K. B.
Wiberg, J. Am. Chem. Soc. 1988, 110, 6642–6650; f) J. Corey, C. J.
Helal, Tetrahedron Lett. 1995, 36, 9153–9156; g) A. B. Northrup,
D. W. C. Millan, J. Am. Chem. Soc. 2002, 124, 2458–2460; h) J. M.
Hawkins, M. Nambu, S. Loren, Org. Lett. 2003, 5, 4293–4295;
i) S. E. Denmark, B. R. Henke, E. Weber, J. Am. Chem. Soc. 1987,
109, 2512–2514; j) S. E. Denmark, N. G. Almstead, J. Am. Chem.
Soc. 1993, 115, 3133–3139; k) S. Fukuzumi, T. Okamoto, M. Fujita,
J. Otera, Chem. Commun. 1996, 393–394; l) A. R. Chamberlin, S. H.
Reich, J. Am. Chem. Soc. 1985, 107, 1440–1441.
[4] a) J. J. Jaber, K. Mitsui, S. D. Rychnovsky, J. Org. Chem. 2001, 66,
4679–4686; b) J. E. Dalgard, S. D. Rychnovsky, J. Am. Chem. Soc.
2004, 126, 15662–15663; c) R. Jasti, C. D. Anderson, S. D, Rych-
novsky, J. Am. Chem. Soc. 2005, 127, 9939–9945.
[5] a) B. B. Snider, Comprehensive Organic Synthesis, Vol. 2 (Ed. B. M.
Trost), Pergamon Press, Oxford, UK, 1991, 527–579; b) R. C.
Crosby, J. R. Harding, C. D. King, G. D. Parker, C. L. Willis, Org.
Lett. 2002, 4, 3407–3410; c) K. N. Cossey, R. L. Funk, J. Am. Chem.
Soc. 2004, 126, 12216–12217.
[6] a) F. R. Japp, W. Maitland, J. Chem. Soc. 1904, 85, 1473–1475; P. A.
Clarke, Paul A. , W. H. C. Martin, Org. Lett. 2002, 4, 4527–4529;
b) G. Sabitha, G. S. K. K. Reedy, M. Rajkumar, J. S. Yadav, K. V. S.
Ramakrishna, A. C. Kunwar, Tetrahedron Lett. 2003, 44, 7455–7457;
c) P. A. Clarke; W. H. C. Martin, J. M. Hargreaves, C. Wilson, A. J.
Blake, Org. Biomol. Chem. 2005, 3, 3551–3563; d) P. A. Clarke,
W. H. C. Martin, Tetrahedron 2005, 61, 5433–5438; e) P. A. Clarke;
W. H. C. Martin, J. M. Hargreaves, C. Wilson, A. J. Blake, Chem.
Commun. 2005, 8, 1061–1063.
[7] S. Das, L.-S. Li, S. C. Sinha, Org. Lett. 2004, 6, 123–126.
[8] a) C. Prandi, P. Venturello J. Org. Chem. 1994, 59, 3494–3496; b) M.
Christmann, U. Bhatt, M. Quitschalle, E. Claus, M. Kalesse, Angew.
Chem. 2000, 112, 4535–4538; Angew. Chem. Int. Ed. 2000, 39, 4364–
4366; c) D. R. Williams, D. C. Ihle, S. V. Plummer, Org. Lett. 2001, 3,
1383–1386; d) F. Hilli, J. M. White, M. A. Rizzacasa, Org. Lett.
2004, 6, 1289–1292.
[9] a) B. W. Greatrex, M. C. Kimber, D. K. Taylor, G. Fallon, E. R. T.
Tiekink, J. Org. Chem. 2003, 68, 4239–4246; b) T. D. Avery, D.
Caiazza, J. A. Culbert, D. K. Taylor, E. R. T. Tiekink, J. Org. Chem.
2005, 70, 8344–8351.
[10] a) V. Gouverneur, M. Reiter, Chem. Eur. J. 2005, 11, 5806–5815;
b) C. Baker-Glenn, N. Hodnett, M. Reiter, S. Ropp, R. Ancliff, V.
Gouverneur, J. Am. Chem. Soc. 2005, 127, 1481–1486.
[11] a) M. Reiter, S. Ropp, V. Gouverneur, Org. Lett. 2004, 6, 91–94;
b) M. Reiter, H. Turner, R. Mills-Webb, V. Gouverneur, J. Org.
Chem. 2005, 70, 8478–8485.
[12] a) T. Hosokawa, T. Shinohara, Y. Ooka, S. I. -Murahashi, Chem.
Lett. 1989, 2001–2004; b) S. Ganguly, R. D. Roundhill, Organome-
tallics 1993, 12, 4825–4832; c) K. J. Miller, T. T. Kitagawa, M. M.
Abu-Omar, Organometallics 2001, 20, 4403–4412.
[23] B. C. Hong, S. F. Chin, Synth. Commun. 1997, 27, 1191–1197.
[24] N. Petasis, S. P. Lu, Tetrahedron Lett. 1996, 37, 141–144.
[13] a) M. J. Gaunt, J. B. Spencer, Org. Lett. 2001, 3, 25–28; b) T. C Wab-
nitz, J. B. Spencer, Tetrahedron Lett. 2002, 43, 3891–3894; c) T. C.
Wabnitz, J. B. Spencer, Org. Lett. 2003, 5, 2141–2144; d) T. C Wab-
nitz, J. Q. Yu, J. B. Spencer, Chem. Eur. J. 2004, 10, 484–493.
Received: March 24, 2006
Published online: July 3, 2006
Chem. Eur. J. 2006, 12, 7190 – 7203
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7203