Competing Diels-Alder intramolecular pathways from cross-conjugated trienes: Fused hydrindenones (bicyclo[4.3.0]nonenones) vs. bridged-ring (bicyclo[3.3.1]nonenones) adducts from a diene-transmissive precursor

Article abstract of DOI:10.1139/v05-055

D-Isoascorbic acid was converted to the tetraenone 12 to examine its behaviour in the intramolecular Diels-Alder reaction with pendant dienes in a diene-transmissive environment. These studies were directed toward the synthesis of the hydrindenone nucleus via a tether-controlled Diels-Alder reaction. It was anticipated that this would lead to a diene-transmissive strategy for steroids such as Desogestrel. In contrast to our previous studies with decalins, the dominant isomer (150 °C) was the bicyclo[3.3.1]nonadieneone adduct 23, and none of fused adduct 27 was detected. These products were accompanied by two minor hydrindenone adducts 24 and 25. The isomer ratio was dependent on the temperature of the reaction.

Full text of DOI:10.1139/v05-055

559
Competing Diels–Alder intramolecular pathways
from cross-conjugated trienes: Fused
hydrindenones (bicyclo[4.3.0]nonenones) vs.
bridged-ring (bicyclo[3.3.1]nonenones) adducts
from a diene-transmissive precursor1
Matthew D. Clay, Ditte Riber, and Alex G. Fallis
Abstract: D-Isoascorbic acid was converted to the tetraenone 12 to examine its behaviour in the intramolecular Diels–
Alder reaction with pendant dienes in a diene-transmissive environment. These studies were directed toward the synthe-
sis of the hydrindenone nucleus via a tether-controlled Diels–Alder reaction. It was anticipated that this would lead to a
diene-transmissive strategy for steroids such as Desogestrel®. In contrast to our previous studies with decalins, the
dominant isomer (150 °C) was the bicyclo[3.3.1]nonadieneone adduct 23, and none of fused adduct 27 was detected.
These products were accompanied by two minor hydrindenone adducts 24 and 25. The isomer ratio was dependent on
the temperature of the reaction.
Key words: Diels–Alder, hydrindane, pentadienylation, intramolecular cycloaddition, bicyclo[3.3.1]nonenone.
Résumé : L’acide D-isoascorbique a été transformé en tétraénone 12 dans le but d’étudier son comportement dans la
réaction intramoléculaire de Diels–Alder avec des diènes pendants, dans un environnement permettant une transmission.
Ces études ont été dirigées vers la synthèse du noyau hydrindénone via une réaction de Diels–Alder contrôlée par une
laisse. Il était prévu que cette méthode permettrait de développer une stratégie diénique permettant une transmission
pour réaliser la synthèse de stéroïdes tel le Desogestrel®. Par opposition aux études antérieures réalisées dans ce labo-
ratoire avec des décalines, l’isomère dominant (150 °C) est l’adduit bicyclo[3.3.1]nonadiénone 23 et on n’a pas détecté
la présence du produit condensé 27. Ces produits étaient accompagnés de deux adduits hydrindénones mineurs, 24 et
25. Le rapport des isomères varie avec la température de la réaction.
Mots clés : Diels–Alder, hydrindane, pentadiénylation, cycloaddition intramoléculaire, bicyclo[3.3.1]nonénone.
[Traduit par la Rédaction] Clay et al. 568
Introduction
pared with trans-acetonides. The cycloadditions directed by
cis-acetonides (1) occurred under milder conditions and al-
lowed for the introduction of an inside methyl group into the
decalin adducts (Scheme 1). Because of the unfavourable
trans-geometry of the diene induced by this substituent, its
introduction is often a challenge in related intermolecular
cycloadditions.
Despite decades of research, the total synthesis of steroid
nuclei by improved strategies continues to receive consider-
able attention. A short synthesis of unnatural steroids, such
as the oral contraceptive Desogestrel®, has proved particu-
larly challenging (1).
Previously we reported the enantioselective synthesis of
cis-decalins via intramolecular Diels–Alder reactions (2).
These strategies employed either planar or isopropylidene
tether control groups to bias the stereochemistry of the ad-
ducts. We discovered that cis-acetonides are preferred com-
We have also extended this tether control group protocol
to the stereoselective construction of highly oxygenated nor-
steroid and nor-triterpenoid skeletons via a diene-transmissive
Diels–Alder strategy (3). A key step in our approach was the
selective indium-mediated γ-pentadienylation of the appro-
Received 20 September 2004. Published on the NRC Research Press Web site at http://canjchem.nrc.ca on 14 June 2005.
This paper is dedicated to Professor Howard Alper. Presented with respect and gratitude for his contributions to organometallic
chemistry and to our friendship.
M.D. Clay, D. Riber,2 and A.G. Fallis.3 Department of Chemistry, University of Ottawa, 10 Marie Curie, Ottawa, ON K1N 6N5,
Canada.
1This article is part of a Special Issue dedicated to Professor Howard Alper.
2Present address: Department of Chemistry, University of Aarhus, DK-8000, Aarhus, Denmark.
3Corresponding author (e-mail: afallis@science.uottawa.ca).
Can. J. Chem. 83: 559–568 (2005)
doi: 10.1139/V05-055
© 2005 NRC Canada

560
Can. J. Chem. Vol. 83, 2005
Scheme 1. Cycloadditions controlled by cis- and trans-acetonides.
O
O
O
H
O
O
O
O
O
O
200 °C, 24 h
40 °C, 4 h
87%
Me
Me
H
H
sealed tube
82%
H
Me
MOMO
MOMO
MOMO
OMOM
OMOM
OMOM
1 cis
2a cis
2b trans
3 trans
Scheme 2. A diene-transmissive Diels–Alder approach to steroids.
O
OH
OH
O
O
O
O
O
O
OH
OMOM
O
MOMO
MOMO
4
5
6
O
O
H
H
O
MOMO
O
H
MOMO
O
H
O
H
H
n=1,2
O
O
H
O
H
MOMO
O
O
7
8
priate aldehyde-hemiacetal 4 derived from L-arabinose.
Elimination of water from the product 5, as illustrated in
Scheme 2, afforded the cross-conjugated triene 6. Sequential
cycloaddition reactions rapidly produced the tetracyclic nor-
steroid or nor-terpenoid core structures 8 (4).
The versatility of this protocol prompted us to investigate
the reverse synthetic strategy for steroids in which the
hydrindane portion would be constructed first by a related
tether-controlled cycloaddition reaction, followed by a sec-
ond intermolecular cyclization. The challenge of efficiently
and stereoselectively synthesizing the trans-fused hydrindane
system of steroids is well recognized. Previous successful
solutions have been based on an intramolecular ester enolate
alkylation (5) or on the Lewis acid assisted cyclization of a
suitably substituted dienone (6). Strategies involving an
intramolecular Diels–Alder cyclization as the key step have
also been employed (7).
To explore the feasibility of our protocol for the construc-
tion of the trans-fused hydrindane ring system of steroids, a
widely prescribed oral contraceptive, Desogestrel® (9), was
selected as a commercially important and challenging target.
The initial industrial synthesis of Desogestrel® required 24
steps from diosgenin, the cost of which was partially offset
by the potent activity of the drug (0.150 mg per pill) (8).
The ethyl group at C13 of Desogestrel® is responsible for a
50-fold enhancement in activity compared with the methyl
analogue. Corey et al. (1a) have demonstrated that this mod-
est structural change presents a significant complication in
its synthesis. In his initial synthesis, Corey surmounted this
challenge by starting with the ethyl substituted Wieland–
Miescher ketone (1a). A more recent solution employs an
oxazaborolidinium-catalyzed Diels–Alder reaction as a key
step (1b).
Our strategy to Desogestrel® envisaged sequential Diels–
Alder cycloadditions in which the CD ring system was
initially constructed via a tether-controlled intramolecular
Diels–Alder reaction, followed by an intermolecular cyclo-
addition to provide the AB rings, as outlined retrosyntheti-
cally in Scheme 3. The AB rings of Desogestrel® (9) should
arise from a Diels–Alder reaction of 10 (exocyclic alkene or
ketone) with a suitable cyclohexanone derivative using
Northrup and MacMillan’s imine-catalyzed protocol (9). The
hydrindane portion 11 could be generated via a tether-
controlled intramolecular Diels–Alder reaction of triene 12.
An unknown variable in this approach was whether the
hydrindane portion could be generated in acceptable yield
and with the desired stereoselectivity, especially in view of
the large number of unsaturated centers in the tetraenone.
An earlier model study using an isopropylidene acetal gave
poor selectivity in the Diels–Alder reaction, so we chose to
protect the diol of 14 as an isopentylidene acetal. It was an-
ticipated that the additional steric bulk of the acetal would
help to differentiate the competing Diels–Alder transition
states and would improve the selectivity of the initial cyclo-
addition. The Diels–Alder precursor 12 can be synthesized
from 13, a product arising from an indium-mediated penta-
© 2005 NRC Canada

Clay et al.
561
Scheme 3. Retrosynthetic route to Desogestrel®.
HO
O
HO
O
X
H
H
O
O
H
H
O
O
H
H
X = CH or O
2
10
9
11
OH
H
OH
O
In
O
O
O
O
O
O
O
OH
HO
H
HO
OH
O
OH
13
O
15
14
12
Scheme 4. Synthesis of Diels–Alder precursors 12a and 12b.
O
O
4OH
(ref. 10)
HO
(a)
(b)
O
15
O
O
O
O
O
O
HO
16
17
14
(c)
(f)
(e)
(d)
O
O
O
O
O
O
O
OH
OH
O
OTIPS
OH
13
OH
OTIPS
20
19
18
(g)
(h)
(i)
O
O
O
O
O
O
R
R
O
OH
O
21
22a: R = Me
22b: R = Et
12a: R = Me
12b: R = Et
cis-diol 16 (Scheme 4) (10). Acetalization of 16 with 3-
pentanone and a catalytic amount of p-toluenesulfonic acid
using a Dean–Stark apparatus provided cis-isopentylidene
acetal 17 in 92% yield after distillation. Reduction of the
lactone moiety with diisobutylaluminum hydride gave lactol
14 in 87% yield as a 5:1 mixture of diastereomers, epimeric
at C4. Indium-mediated pentadienylation onto the aldehyde
dienylation of lactol 14. This enantiomerically pure lactol
can be easily prepared from D-isoascorbic acid (15).
Results and discussion
Our synthesis commenced with the known oxidative deg-
radation of D-isoascorbic acid to generate the enantiopure
© 2005 NRC Canada

562
Can. J. Chem. Vol. 83, 2005
Table 1. Summary of Diels–Alder reactions.
Entry
Conditions
Resultsa
1
2
Et2AlCl, CH2Cl2, –78 °C (2 h) (R = Me)
Starting material
Starting material
Et2AlCl, CH2Cl2, –78 °C (30 min) to 21 °C (4 h) (R = Me)
Et2AlCl, CH2Cl2, –78 °C (30 min) to 42 °C (4 h) (R = Me)
BF3OEt2, CH2Cl2, –78 °C (4 h) (R = Me)
3
Partial decomposition
Starting material
4
5
Complete decomposition
Starting material
BF3OEt2, CH2Cl2, –78 °C (30 min) to 0 °C (1 min) (R = Me)
HN(i-Pr)2, HClO4, H2O, 0 °C (16 h) (R = Et)
HN(i-Pr)2, HClO4, H2O, rt (24 h) (R = Et)
6
7
Starting material
8
Starting material
HN(i-Pr)2, HClO4, EtOH, 0 °C (6 h) to rt (16 h) (R = Et)
HN(i-Pr)2, HClO4, EtOH, 0 °C (5 min) to 65 °C (6 h) (R = Et)
p-TsOH, HO(CH2)2OH, PhH, 80 °C (20 h) (R = Et)
9
Starting material
3 isomers (23:21:56)b
10
11
12
13
Toluene, 75 °C (6 days) (R = Me)
Toluene, 150 °C, microwave (3 h) (R = Me)
Toluene, 150 °C, microwave (3 h) (R = Et)
3 isomers (28:24:48)b
3 isomers (41:20:39)b
3 isomers (53:17:30)b
aRatio determined by NMR integration.
bRatio of 23:24:25.
of the open-chain form of the lactol provided diol 13 in 82%
yield as a variable mixture of diastereomers. This was incon-
sequential as this secondary alcohol stereocenter was re-
moved later in the synthesis. Selective protection of the
primary alcohol as a triisopropylsilyl ether provided 18 in
93% yield (as a mixture of diastereomers).
Addition of iso-propenyllithium or sec-butenyllithium (gen-
erated from the corresponding bromides via lithium–halogen
exchange) to the aldehyde provided allyl alcohols 22a and
22b, respectively. It was subsequently discovered that addi-
tion of the Grignard reagent to the aldehyde was more effi-
cient on larger scales and afforded 22a and 22b in 58% and
53% yields, respectively. Oxidation of both compounds was
accomplished using the Dess–Martin periodinane–water
combination (described above) and provided Diels–Alder
precursors 12a and 12b in 78% and 72% yield, respectively.
We initially attempted the Diels–Alder reaction on the
simpler model system 12a (rather than the ethyl analogue
12b). However, once this material was consumed, subse-
quent reactions were conducted using 12b. The details of
these Diels–Alder studies are presented in Table 1. It was
initially believed that a Lewis acid mediated cycloaddition at
low temperature would provide the best selectivity. There-
fore, the cycloaddition was attempted with either diethyl-
aluminum chloride or boron trifluoride etherate as Lewis
acid catalysts. However, neither gave any trace of cyclo-
addition products at low temperatures, and only decomposi-
tion was observed at higher temperatures (entries 1–5). In
both cases, the decomposition products were not the free
diol from Lewis acid assisted cleavage of the isopentylidene
acetal. Apparently the increased bulk created by Lewis acid
association with the carbonyl inhibits the beneficial effect
anticipated from the electronic activation.
An attempt to induce the cycloaddition using a variation
of MacMillan’s imine-catalyzed procedure was also exam-
ined (9). Diisopropylamine was employed as the amine com-
ponent, in place of the more complex amine used by
MacMillan. Unfortunately, despite several attempts, no reac-
tion or decomposition occurred, even in the presence of
perchloric acid and refluxing ethanol (entries 6–9).
We next attempted the cycloaddition under thermal condi-
tions. Heating tetraene 12a in the microwave at 150 °C for
3 h gave a mixture of Diels–Alder adducts in 78% yield. The
individual isomers were separated by preparative HPLC, and
their structures and relative stereochemistry were determined
by NOESY and COSY NMR experiments. The significant
Mitsunobu-type dehydration of 18 with three equivalents
each of triphenylphosphine and diethyl azodicarboxylate
generated triene 19. Interestingly, the minor isomer of 18
was resistant to dehydration. Whereas reaction of the major
isomer (of the mixture) was complete within one day, dehy-
dration of the minor isomer required three days, and often
additional reagents (PPh3 and DEAD) were required. This is
likely because of steric hindrance about the reaction center
caused by the isopentylidene acetal and the bulky triiso-
propylsilyl protecting group. Experimentally, triene 19 was
isolated in 44% yield, and subsequent desilylation with tetra-
butylammonium fluoride provided 20 in 82% yield. The low
yield of the dehydration step was attributed to the instability
of 19, which readily decomposed during column chromatog-
raphy or upon storage in a refrigerator for short periods.
Thus a one-pot reaction sequence, in which 18 was first
dehydrated and then desilylated to 20 without isolation of
intermediate 19, was investigated. We were gratified to
achieve a higher yield of crude 20 from the one-pot proce-
dure (~71%, compared with 36% over the two separate steps).
However, 20 could not be easily separated from several im-
purities generated during the dehydration phase. Thus the
preferred experimental method to ensure purity was the orig-
inal two-step sequence.
Oxidation of the alcohol of trienol 20 also proved chal-
lenging. Oxidation occurred under a variety of conditions
(e.g., Collins, Corey–Schmidt, IBX, Dess–Martin, Swern),
but the product aldehyde slowly decomposed under the reac-
tion conditions. In accord with earlier observations (11), a
substantial rate enhancement occurred when water was
added to the Dess–Martin oxidation of 20. This technique
pushed the oxidation to completion before noticeable de-
composition of the product aldehyde had occurred and pro-
vided stable aldehyde 21 in an isolated yield of 74%.
© 2005 NRC Canada

Clay et al.
563
Fig. 1. Significant nOe interactions in Diels–Alder adducts 23, 24, and 25.
1.0
0.6
1.8
1.5
H
H
H
H
O
H
H
O
H
H
H
H
O
H
H
H
O
H
O
O
Me
H
Me
H
H
1.5
H
H
H
H
H
H
H
H
0.2
0.7
2.2
O
2.0
H
H
1.0
Me
O
H
O
H
1.6
1.6
0.5
23
24
25
nOe interactions are illustrated in Fig. 1. We were intrigued
to find that the major isomer was the unexpected
bicyclo[3.3.1]nonadieneone 23 rather than one of the antici-
pated hydrindenone isomers.
agents. n-BuLi solutions in hexanes from Aldrich Chemical
Company (Oakville, Ont.) were titrated prior to use against
diphenylacetic acid and used directly. Starting materials
were purchased from Aldrich Chemical Company (Oakville,
Ont.) and used without further purification unless otherwise
stated.
The highest ratio of the bicyclo[3.3.1]nonenone isomer 23
was obtained with the ethyl substrate 12b (53:17:30, 23:24:25)
upon heating in a microwave oven at 150 °C for 3 h in tolu-
ene (entry 13). In contrast, under identical microwave condi-
tions, the methyl substrate 12a afforded only a slight excess
of 23 (41:20:39) (entry 12). The ratio changed to favor for-
mation of hydrindenone 25 (28:24:48) when the temperature
was lowered to 75 °C. The reaction required 6 days at this
temperature for completion (entry 11). In an attempt to
change the electronic environment in the precursor 12b,
acetalization of the ketone of 12b was attempted using ethyl-
ene glycol and catalytic p-toluenesulfonic acid in refluxing
benzene (80 °C) for 20 h (entry 10). To our surprise, the
acetal did not form; rather, 12b underwent the Diels–Alder
reaction to generate 25 as the major isomer in a ratio of
23:21:56 (23:24:25). This ratio was similar to the thermal re-
action in toluene at 75 °C. The substantially shorter reaction
time (20 h vs. 6 days) indicated the rate influence from
ketone protonation by the p-toluenesulfonic acid.
Examination of the transition states in Fig. 2 suggests that
formation of the desired trans-isomer 27 was disfavored by
steric repulsion between the R group of the dienophile and
the “extra” vinyl group of the diene as illustrated in 26. This
hypothesis was supported by the fact that this Diels–Alder
precursor can rotate to avoid this interaction (28) and prefer-
entially generate the bicyclic isomer 23, with lesser amounts
of 24 and 25. It is also interesting that the cis-hydrindenone
24, from transition state 30, was formed in approximately
the same quantity under all conditions. It was anticipated
that the steric repulsion between the acetal ethyl group and
the R group of the dienophile would disfavor this adduct for-
mation. This is partly why the large isopentylidene acetal
was chosen instead of the smaller isopropenyl acetal used in
model studies.
Reactions were monitored by thin layer chromatography
(TLC) using commercial aluminum-backed silica gel sheets
coated with silica gel 60 F254 (E. Merck, Darmstad, Ger-
many). TLC spots were developed by heating the plate after
treatment with a 5% solution of ammonium molybdate in
10% aqueous sulphuric acid or a solution of KMnO4 in
aqueous potassium hydroxide. Room temperature corre-
sponds to 21 °C. Anhydrous magnesium sulfate (MgSO4)
was used to dry solutions in organic solvents. Excess sol-
vents were removed (concentrated) in vacuo at pressures ob-
tained by a water or air aspirator connected to a Büchi rotary
evaporator. Trace solvents were removed on a vacuum
pump. Product purification by flash chromatography was
performed with E. Merck Silica Gel 60 (230–400 mesh). Pe-
troleum ether refers to a mixture of hydrocarbons with a
boiling range of 30–60 °C.
Melting points were determined with a Thomas Hoover
Unit melting point apparatus and are uncorrected. Infrared
(IR) spectra were obtained as neat thin films or as a solution
of the sample in CDCl3 or CHCl3 in a sodium chloride solu-
tion cell. All IR spectra were recorded on a Bomem
Michelson 100 Fourier transform infrared spectrometer (FT-
1
IR). H NMR (500 MHz) and 13C NMR (125 MHz) spectra
were run on a Brucker AMX500 spectrometer. Chemical
shifts are reported downfield from tetramethylsilane (δ scale)
in ppm. 1H NMR data are reported as follows: chemical shift
(multiplicity (s = singlet, d = doublet, t = triplet, q = quar-
tet), coupling constants (Hz), and integration). Low-
resolution mass spectroscopy (MS), using electron impact
(EI), was performed on a V.G. Micromass 7070 HS mass
spectrometer with an electron beam energy of 70 eV. High-
resolution mass spectroscopy (HR-MS) was performed on a
Kratos Concept-11A mass spectrometer with an electron
beam energy of 70 eV. The purity of all title compounds was
General experimental
1
judged to be >95% as determined by a combination of H
NMR and 13C NMR analyses.
All non-aqueous reactions were performed under a posi-
tive pressure of dry nitrogen in flame-dried glassware using
dry solvents. Tetrahydrofuran and diethyl ether were dis-
tilled from sodium–benzophenone. Dichloromethane, tolu-
ene, triethylamine, and methyl vinyl ketone were distilled
from calcium hydride. Standard inert atmosphere techniques
were employed in handling air- and moisture-sensitive re-
3,4-Dihydroxy-dihydro-furan-2-one (16)
The title compound was prepared according to the litera-
ture procedure (10). mp 103–104 °C (lit. (10) value mp
97.5–99.5 °C). IR (neat, cm–1) ν: 3421, 3281, 2989, 2922,
1
2867, 1773, 1154, 1105, 1031, 970. H NMR (500 MHz,
© 2005 NRC Canada

564
Can. J. Chem. Vol. 83, 2005
Fig. 2. Transition states leading to various adducts.
H
O
O
R
O
O
H
O
R
O
26
27
O
O
H
O
O
R
H
O
O R
23
28
O
O
R
O
H
H
O
O
12a: R = Me
12b: R = Et
O
O
R
R
O
O
29
25
H
H
R
O
O
O
O
R
O
O
24
30
D2O) δ: 4.77 (d, J = 4.9 Hz, 1H), 4.65 (dd, J = 4.9, 3.0 Hz,
1H), 4.59 (dd, J = 10.8, 3.0 Hz, 1H), 4.43 (d, J = 10.8 Hz,
1H). 13C NMR (125 MHz, D2O) δ: 178.2 (C), 72.5 (CH2),
69.2 (CH), 68.4 (CH). EI-MS m/z (%): 118 ([M]+, 3), 73
(26), 60 (91), 56 (100), 43 (56). HR-MS calcd. for C4H6O4
([M]+): 118.0242; found: 118.0268.
(m, 1H), 4.02–4.05 (m, 1H), 3.91 (d, J = 10.8 Hz, 1H),
3.54–3.58 (m, 1H), 1.41 (q, J = 7.5 Hz, 2H), 1.56 (q, J =
7.5 Hz, 2H), 0.75 (t, J = 7.5 Hz, 3H), 0.86 (t, J = 7.5 Hz,
3H). 13C NMR (125 MHz, CDCl3) δ: 174.0 (C), 117.7 (C),
75.6 (CH), 74.7 (CH), 70.3 (CH2), 29.4 (CH2), 29.0 (CH2),
7.9 (CH3), 7.0 (CH3). EI-MS m/z (%): 157 ([M+ – C2H5],
45), 57 (100), 29 (85). HR-MS calcd. for C7H11O4 ([M+ –
C2H5]): 157.0501; found: 157.0492.
2,2-Diethyl-dihydro-furo[3,4-d][1,3]dioxol-4-one (17)
Diol 16 (14.00 g, 0.1186 mol), 3-pentanone (125 mL,
1.186 mol), and p-toluenesulfonic acid (2.0 g, catalytic)
were dissolved in benzene (150 mL) and refluxed overnight
using a Dean–Stark apparatus. The solution was then cooled
to room temperature, and the majority of the solvent and ex-
cess 3-pentanone were removed by rotary evaporation at a
bath temperature of ~50 °C. The remaining brown-coloured
oil was distilled under aspirator pressure using a propane
burner. The title compound collected as a pale brown oil
(20.38 g, 92%) over a fluctuating (the aspirator pressure var-
ied during the distillation) temperature range of 160–200 °C.
IR (neat, cm–1) ν: 2976, 2946, 2880, 1785, 1465, 1183,
2,2-Diethyl-tetrahydro-furo[3,4-d][1,3]dioxol-4-ol (14)
Lactone 17 (19.0 g, 0.102 mol) was dissolved in CH2Cl2
(285 mL) and cooled to –78 °C. A 1.5 mol/L solution of
DIBAL-H in THF (81.6 mL, 0.122 mol) was added
dropwise over a period of 15 min, after which the solution
was stirred for 5 h at –78 °C. The reaction was quenched at
–78 °C with MeOH (16.0 mL) and brine (8.0 mL) and al-
lowed to warm to room temperature. Ether (420 mL) and
MgSO4 (121 g) were then added under vigorous stirring, and
stirring was continued overnight. The mixture was filtered
(suction), and the filter cake was washed generously with
ether (~1.0 L) in several portions until the presence of prod-
1
1107, 1069, 921. H NMR (500 MHz, C6D6) δ: 4.17–4.19
© 2005 NRC Canada

Clay et al.
565
uct was no longer detected in the filtrate (use of a soxlet ex-
tractor to extract the product from the filter cake was also
effective). The solvent was removed under reduced pressure
to yield a colorless oil identified as a 5:1 mixture of
diastereomers of the title compound (16.72 g, 87%). Major
diastereomer: IR (neat, cm–1) ν: 3425, 2975, 2943, 2883,
1H), 3.73 (dd, J = 10.3, 3.4 Hz, 1H), 3.46 (t, J = 7.8 Hz,
1H), 1.64 (dq, J = 7.5, 1.5 Hz, 2H), 1.58 (q, J = 7.5 Hz, 2H),
0.92–1.00 (m, 21H), 0.93 (t, J = 7.5 Hz, 3H), 0.87 (t, J =
δ
137.2 (CH), 117.6 (CH2), 115.6 (CH2), 112.5 (C), 78.6
(CH), 77.9 (CH), 72.8 (CH), 63.6 (CH), 51.9 (CH), 30.7
(CH2), 29.4 (CH2), 18.29 (CH3), 18.28 (CH3), 12.7 (CH),
12.4 (CH), 9.3 (CH3), 8.5 (CH3). EI-MS m/z (%): 383
([M+ – C2H5], 8), 187 (53), 131 (56), 67 (100). HR-MS
calcd. for C21H39O4Si ([M+ – C2H5]): 383.2618; found:
383.2628.
7.5 Hz, 3H). 13C NMR (125 MHz, C6D6) : 140.5 (CH),
1
1464, 1102, 1075, 929. H NMR (500 MHz, CDCl3) δ: 5.38
(s, 1H), 4.79 (dd, J = 6.0, 3.3 Hz, 1H), 4.53 (d, J = 6.0 Hz,
1H), 4.03–3.92 (m, 2H), 3.18 (1H, bs), 1.65 (q, J = 7.5 Hz,
2H), 1.54 (q, J = 7.5 Hz, 2H), 0.87 (t, J = 7.5 Hz, 3H), 0.84
(t, J = 7.5 Hz, 3H). 13C NMR (125 MHz, CDCl3) δ: 116.5
(C), 101.9 (CH), 85.3 (CH), 80.2 (CH), 72.1 (CH2), 29.3
(CH2), 28.8 (CH2), 8.3 (CH3), 7.3 (CH3). EI-MS m/z (%):
159 ([M+ – C2H5], 64), 85 (17), 57 (100), 29 (37). HR-MS
calcd. for C7H11O4 ([M+ – C2H5]): 159.0657; found:
159.0659.
[2,2-Diethyl-5-(2-vinyl-buta-1,3-dienyl)-[1,3]dioxolan-4-
ylmethoxy]-triisopropyl-silane (19)
Alcohol 18 (21.99 g, 53.3 mmol) was dissolved in ben-
zene (1.20 L) and cooled to 0 °C, and PPh3 (41.92 g,
159.8 mmol) and DEAD (25.16 mL, 159.8 mmol) were suc-
cessively added. The solution was refluxed for two days and
the majority of the solvent was removed in vacuo to yield a
red oil. A slurry of silica gel (30:1, petroleum ether – ethyl
acetate) was then added, and the mixture was quickly fil-
tered through a thin (~7 cm) plug of silica gel to remove the
majority of reaction byproducts. The silica gel plug was
washed (30:1, petroleum ether – ethyl acetate) until no addi-
tional product was detected in the filtrate. The washes were
combined and concentrated, and petroleum ether was added,
and the triphenylphosphine oxide precipitate was filtered off
and discarded. Column chromatography (100:1, petroleum
ether – ethyl acetate) yielded the title compound as a color-
less oil (9.36 g, 44%). Owing to the instability of the prod-
uct in common NMR solvents, clean NMR spectra could not
be obtained. IR (neat, cm–1) ν: 3089, 2867, 2943, 2867,
1464, 1129, 1087, 939, 883. EI-MS m/z (%): 394 ([M]+, 19),
265 (100), 108 (84). HR-MS calcd. for C23H42O3Si ([M]+):
394.2903; found: 394.2925
1-(2,2-Diethyl-5-hydroxymethyl-[1,3]dioxolan-4-yl)-2-
vinyl-but-3-en-1-ol (13)
Lactol 14 (5.00 g, 26.56 mmol) was dissolved in DMF
(15 mL); the solution was cooled to 0 °C, and 5-bromo-1,3-
pentadiene (15.62 g, 106.3 mmol) was added. Indium
powder (100 mesh, 6.10 g, 53.13 mmol) was added to the
vigorously stirred solution in small portions over a period of
55 min. The dark green mixture was then warmed to room
temperature and stirred overnight. The reaction was poured
into a vigorously stirred solution of ether (1000 mL) and wa-
ter (3.2 mL) and stirred for a further 10 min. The mixture
was then dried, filtered, and concentrated. Column chroma-
tography (4:1, petroleum ether – ethyl acetate) yielded the ti-
tle compound as a waxy solid mixture of diastereomers
(5.55 g, 82%). Major diastereomer: mp 42.5–44.0 °C. IR
(neat, cm–1) ν: 3323, 3245, 2940, 2916, 2880, 1459, 1173,
1081, 920. 1H NMR (500 MHz, C6D6) δ: 6.05–6.14 (m, 2H),
5.06–5.26 (m, 4H), 4.15–4.23 (m, 2H), 3.92 (m, 1H), 3.69–
3.74 (m, 1H), 3.56–3.60 (m, 1H), 3.33–3.37 (m, 2H), 2.73
(bs, 1H), 1.57 (q, J = 7.5 Hz, 2H), 1.51 (q, J = 7.5 Hz, 2H),
0.86 (t, J = 7.5 Hz, 3H), 0.82 (t, J = 7.5 Hz, 3H). 13C NMR
(125 MHz, C6D6) δ: 139.6 (CH), 136.1 (CH), 118.4 (CH2),
116.4 (CH2), 112.4 (C), 78.0 (CH), 77.7 (CH), 72.7 (CH),
61.9 (CH2), 51.4 (CH), 30.5 (CH2), 29.4 (CH2), 9.2 (CH3),
8.4 (CH3). EI-MS m/z (%): 227 ([M+ – C2H5], 38), 209
(0.2), 159 (11), 87 (100), 57 (87). HR-MS calcd. for
C12H19O4 ([M+ – C2H5]): 227.1284; found: 227.1283.
[2,2-Diethyl-5-(2-vinyl-buta-1,3-dienyl)-[1,3]dioxolan-4-
yl]-methanol (20)
Triene 19 (13.71 g, 34.7 mmol) was dissolved in THF
(300 mL), cooled to 0 °C, and a 1.0 mol/L solution of TBAF
in THF (41.7 mL, 41.7 mmol) was added dropwise over a
period of 2 min. The reaction was stirred at 0 °C for 1 h.
Water was added, and the reaction was extracted with ether,
dried, filtered, and concentrated. Column chromatography
(9:1, petroleum ether – ethyl acetate) yielded the title com-
pound as a clear oil (6.79 g, 82%). IR (neat, cm–1) ν: 3431,
1-(2,2-Diethyl-5-triisopropylsilanyloxymethyl-
[1,3]dioxolan-4-yl)-2-vinyl-but-3-en-1-ol (18)
1
3090, 2975, 2941, 2882, 1604, 1173, 1077, 1041, 927. H
Diol 13 (19.01 g, 74.2 mmol) was dissolved in a mixture
of DMF (65 mL) and THF (200 mL) and cooled to 0 °C.
Imidazole (10.12 g, 148.5 mmol) and triisopropylsilyl chlo-
ride (23.83 mL, 111.4 mmol) were added sequentially, and
the solution was stirred overnight. Water was then added,
and the mixture was extracted with ether. The ether extracts
were washed once with saturated NH4Cl, dried, filtered, and
concentrated. Column chromatography (50:1, petroleum
ether – ethyl acetate) yielded the title compound as a color-
less oil (28.62 g, 93%). Major diastereomer: IR (neat, cm–1)
ν: 3467, 2965, 2940, 2867, 1638. 1H NMR (500 MHz,
C6D6) δ: 6.31–6.41 (m, 2H), 5.12–5.32 (m, 4H), 4.37–4.41
(m, 1H), 4.29–4.33 (m, 1H), 4.19 (ddd, J = 5.5, 2.5, 0.5 Hz,
1H), 4.00 (t, J = 10.2 Hz, 1H), 3.96 (dd, J = 3.3, 1.3 Hz,
NMR (500 MHz, C6D6) δ: 6.27 (dd, J = 17.4, 10.7 Hz, 1H),
6.23 (dd, J = 17.6, 11.2 Hz, 1H), 5.25 (dd, J = 17.4, 1.3 Hz,
1H), 5.16 (d, J = 17.5 Hz, 1H), 5.10 (d, J = 11.2 Hz, 1H),
5.02 (m, 1H), 4.96 (d, J = 10.7 Hz, 1H), 4.13 (dd, J = 11.5,
7.0 Hz, 1H), 3.47–3.57 (m, 2H), 2.16 (bs, 1H), 1.70 (q, J =
7.5 Hz, 2H), 1.58 (q, J = 7.5 Hz, 2H), 0.97 (t, J = 7.5 Hz,
3H), 0.86 (t, J = 7.5 Hz, 3H). 13C NMR (125 MHz, C6D6) δ:
141.2 (C), 138.1 (CH), 132.2 (CH), 128.0 (CH), 119.8
(CH2), 116.4 (CH2), 113.1 (C), 79.9 (CH), 74.5 (CH), 62.8
(CH2), 30.3 (CH2), 29.5 (CH2), 9.3 (CH3), 8.5 (CH3). EI-MS
m/z (%): 238 ([M]+, 4), 135 (38), 108 (38), 91 (78), 57
(100), 29 (88). HR-MS calcd. for C14H22O3 ([M]+):
238.1569; found: 238.1551.
© 2005 NRC Canada

566
Can. J. Chem. Vol. 83, 2005
m/z (%): 278 ([M]+, 0.1), 249 (5), 121 (52), 105 (85), 91
(65), 57 (100), 29 (63). HR-MS calcd. for C15H21O3 ([M+ –
C2H5]): 249.1491; found: 249.1488.
2,2-Diethyl-5-(2-vinyl-buta-1,3-dienyl)-[1,3]dioxolane-4-
carbaldehyde (21)
Triene-ol 20 (1.50 g, 6.29 mmol) was dissolved in CH2Cl2
(60 mL), and Dess–Martin reagent (5.45 g, 13.2 mmol) was
added. The solution was stirred for 30 min at room tempera-
ture, and then water (0.113 mL, 6.29 mmol) was added. A
white precipitate formed within a few minutes. After a fur-
ther 15 min of stirring the reaction was quenched with a 1:1
mixture of 2 mol/L Na2S2O3 and sat. NaHCO3, and the reac-
tion was extracted with CH2Cl2. The combined organics
were dried, filtered, and concentrated. Column chromatogra-
phy (12:1, petroleum ether – ethyl acetate) yielded the title
compound as a clear oil (1.11 g, 74%). IR (neat, cm–1) ν:
3093, 2976, 2942, 2884, 2812, 1736, 1464, 1172, 1075, 924.
1H NMR (500 MHz, C6D6) δ: 9.45 (d, J = 3.5 Hz, 1H), 6.17
(dd, J = 17.4, 10.7 Hz, 1H), 6.16 (dd, J = 17.6, 11.2 Hz,
1H), 5.51 (d, J = 9.0 Hz, 1H), 5.22 (dd, J = 17.6, 1.9 Hz,
1H), 5.20 (dd, J = 17.4, 0.9 Hz, 1H), 5.12 (dd, J = 11.2,
1.7 Hz, 1H), 5.10 (t, J = 8.5 Hz, 1H), 4.94 (dd, J = 10.7,
0.9 Hz, 1H), 4.08 (dd, J = 8.0, 3.5 Hz, 1H), 1.75 (q, J =
7.5 Hz, 2H), 1.48 (q, J = 7.5 Hz, 2H), 0.98 (t, J = 7.5 Hz,
3H), 0.80 (t, J = 7.5 Hz, 3H). 13C NMR (125 MHz, C6D6) δ:
199.5 (C), 142.7 (C), 137.7 (CH), 132.0 (CH), 125.7 (CH),
120.6 (CH2), 117.2 (CH2), 115.7 (C), 83.5 (CH), 76.2 (CH),
30.2 (CH2), 29.5 (CH2), 9.1 (CH3), 8.5 (CH3). EI-MS m/z
(%): 207 ([M+ – C2H5], 11), 133 (31), 99 (100), 57 (95), 29
(53). HR-MS calcd. for C12H15O3 ([M+ – C2H5]): 207.1021;
found: 207.1016.
1-[2,2-Diethyl-5-(2-vinyl-buta-1,3-dienyl)-[1,3]dioxolan-4-
yl]-2-methyl-propenone (12a)
Allyl alcohol 22a (1.10 g, 3.96 mmol) was dissolved in
CH2Cl2 (60 mL), and Dess–Martin reagent (3.35 g,
8.12 mmol) was added. The solution was stirred for 30 min
at room temperature, and then water (0.071 mL, 3.96 mmol)
was added. A white precipitate formed within a few minutes.
After a further 15 min of stirring the reaction was quenched
with a solution of 2 mol/L Na2S2O3 and saturated NaHCO3
(1:1), and the reaction was extracted with CH2Cl2. The solu-
tion was dried, filtered, and concentrated. Column chroma-
tography (30:1, petroleum ether – ethyl acetate) yielded the
title compound as a clear oil (0.85 g, 78%). IR (neat, cm–1)
ν: 3091, 2974, 2942, 2883, 1693, 1463, 1173, 1071, 1056,
930. 1H NMR (500 MHz, C6D6) δ: 6.20 (dd, J = 17.6,
11.2 Hz, 1H), 6.12 (dd, J = 17.3, 10.7 Hz, 1H), 5.58 (d, J =
9.7 Hz, 1H), 5.33 (s, 1H), 5.10–5.23 (m, 6H), 4.88 (d, J =
10.7 Hz, 1H), 2.01 (qd, J = 7.5, 2.4 Hz, 2H), 1.69 (t, J =
0.6 Hz, 3H), 1.65 (q, J = 7.5 Hz, 2H), 1.13 (t, J = 7.5 Hz,
3H), 0.92 (t, J = 7.5 Hz, 3H). 13C NMR (125 MHz, C6D6) δ:
196.5 (C), 144.9 (C), 141.3 (C), 137.7 (CH), 132.0 (CH),
127.8 (CH), 124.5 (CH2), 119.8 (CH2), 117.0 (CH2), 115.3
(C), 80.3 (CH), 75.0 (CH), 30.5 (CH2), 29.6 (CH2), 18.3
(CH3), 9.3 (CH3), 8.9 (CH3). EI-MS m/z (%): 276 ([M]+,
0.2), 247 (11), 139 (50), 69 (100), 57 (44), 41 (58), 29 (33).
HR-MS calcd. for C15H19O3 ([M+ – C2H5]): 247.1334;
found: 247.1334.
1-[2,2-Diethyl-5-(2-vinyl-buta-1,3-dienyl)-[1,3]dioxolan-4-
yl]-2-methyl-prop-2-en-1-ol (22a)
2-Bromopropene (1.00 mL, 11.26 mmol) was dissolved in
THF (10 mL) and added dropwise to a flask containing
freshly washed (sequentially with 10% HCl, water, and ace-
tone) and dried magnesium turnings (0.27 g, 11.26 mmol) in
THF (10 mL). A small chip of iodine was added to initiate
the Grignard reaction. When the Grignard had fully formed
(as indicated by the complete consumption of Mg),
1.1 equiv. of it (13.3 mL, 7.49 mmol) was added dropwise to
a solution of aldehyde 21 (1.61 g, 6.83 mmol) in THF
(90 mL). The resultant pale yellow solution was stirred for
45 min. The reaction was quenched with saturated NH4Cl,
extracted with ether, dried, filtered, and concentrated. Col-
umn chromatography (30:1, petroleum ether – ethyl acetate)
yielded the title compound as a waxy solid mixture of two
diastereomers (1.10 g, 58%). IR (neat, both diastereomers,
cm–1) ν: 3422, 3089, 2973, 2941, 1648, 1460, 1172, 1073.
1-[2,2-Diethyl-5-(2-vinyl-buta-1,3-dienyl)-[1,3]dioxolan-4-
yl]-2-ethyl-prop-2-en-1-ol (22b)
The title compound was prepared by a method analogous
to that used to prepare the methyl analogue (22a), except
that iso-butenylmagnesium bromide was used as the Grig-
nard reagent. (53% yield). IR (neat, both diastereomers, cm–1)
ν: 3475, 3089, 2971, 2940, 2882, 1463, 1172, 1074, 1056,
928. Major diastereomer: 1H NMR (500 MHz, C6D6) δ: 6.33
(ddd, J = 17.3, 10.7, 0.6 Hz, 1H), 6.26 (dd, J = 17.5,
11.2 Hz, 1H), 6.10 (d, J = 9.1 Hz, 1H), 5.32 (dd, J = 17.4,
1.5 Hz, 1H), 5.19 (dd, J = 17.3, 1.9 Hz, 1H), 5.05–5.13 (m,
3H), 4.98 (dd, J = 10.7, 1.5 Hz, 1H), 4.89 (m, 1H), 4.21 (dd,
J = 7.3, 3.9 Hz, 1H), 4.10–4.13 (m, 1H), 2.08–2.16 (m, 1H),
1.93–2.01 (m, 1H), 1.72 (q, J = 7.5 Hz, 2H), 1.57 (q, J =
7.5 Hz, 2H), 0.97 (t, J = 7.5 Hz, 3H), 0.96 (t, J = 7.5 Hz,
3H), 0.83 (t, J = 7.5 Hz, 3H) (OH not detected). 13C NMR
(125 MHz, C6D6) δ: 151.2 (C), 140.9 (C), 138.2 (CH), 132.4
(CH), 129.2 (CH), 119.4 (CH2), 116.4 (CH2), 112.9 (CH2),
110.9 (C), 80.6 (CH), 75.0 (CH), 73.7 (CH), 30.1 (CH2),
28.9 (CH2), 25.7 (CH2), 12.7 (CH3), 9.3 (CH3), 8.6 (CH3).
EI-MS (both diastereomers) m/z (%): 263 ([M+ – C2H5], 3),
149 (34), 121 (54), 105 (74), 91 (61), 57 (100), 29 (79). HR-
MS calcd. for C16H23O3 ([M+ – C2H5]): 263.1647; found:
263.1664.
1
Major diastereomer: H NMR (500 MHz, C6D6) δ: 6.36 (dd,
J = 17.3, 10.7 Hz, 1H), 6.29 (dd, J = 17.6, 11.2 Hz, 1H),
6.13 (d, J = 9.1 Hz, 1H), 5.34 (dd, J = 17.3, 1.5 Hz, 1H),
5.21 (dd, J = 17.7, 1.8 Hz, 1H), 5.14 (ddd, J = 11.2, 1.9,
0.7 Hz, 1H), 5.09 (dd, J = 9.1, 7.3 Hz, 1H), 5.01 (dd, J =
10.7, 1.5 Hz, 1H), 5.01 (quintet, J = 0.9 Hz, 1H), 4.87 (t, J =
1.5 Hz, 1H), 4.10 (dd, J = 7.9, 3.1 Hz, 1H), 2.58 (d, J =
6.9 Hz, 1H), 1.74 (q, J = 7.5 Hz, 2H), 1.73 (s, 3H), 1.58 (q,
J = 7.5 Hz, 2H), 0.90 (t, J = 7.5 Hz, 3H), 0.85 (t, J = 7.5 Hz,
3H), (OH not detected). 13C NMR (125 MHz, C6D6) δ:
146.0 (C), 141.0 (C), 138.2 (CH), 132.4 (CH), 129.1 (CH),
119.5 (CH2), 116.5 (CH2), 113.2 (CH2), 112.9 (C), 80.5
(CH), 75.0 (CH), 74.1 (CH), 30.0 (CH2), 28.8 (CH2), 19.2
(CH3), 9.3 (CH3), 8.6 (CH3). EI-MS (both diastereomers)
1-[2,2-Diethyl-5-(2-vinyl-buta-1,3-dienyl)-[1,3]dioxolan-4-
yl]-2-ethyl-propenone (12b)
The title compound was prepared by a method analogous
to that used to prepare the methyl analogue (12a) (72%
© 2005 NRC Canada

Clay et al.
567
yield). IR (neat, cm–1) ν: 3091, 2973, 2940, 2882, 1692,
(CH), 79.7 (CH), 48.7 (C), 46.6 (CH), 30.0 (CH2), 29.8
(CH2), 29.2 (CH2), 22.0 (CH2), 22.0 (CH3), 8.6 (CH3), 8.2
(CH3). EI-MS m/z (%): 276 ([M]+, 2), 247 (44), 191 (8), 162
(12), 145 (34), 119 (100), 105 (24), 57 (50), 43 (32). HR-
MS calcd. for C17H24O3 ([M]+): 276.1725; found: 276.1748.
1
1463, 1173, 1078, 930. H NMR (500 MHz, C6D6) δ: 6.20
(dd, J = 17.6, 11.2 Hz, 1H), 6.12 (ddd, J = 17.4, 10.7,
0.6 Hz, 1H), 5.59 (d, J = 9.6 Hz, 1H), 5.40 (s, 1H), 5.10–
5.25 (m, 6H), 4.88 (dd, J = 10.6, 1.5 Hz, 1H), 2.19 (qq, J =
7.5, 0.8 Hz, 2H), 2.01 (dq, J = 7.5, 2.4 Hz, 2H), 1.65 (dq,
J = 7.5, 1.1 Hz, 2H), 1.13 (t, J = 7.5 Hz, 3H), 0.92 (t, J =
7.5 Hz, 3H), 0.84 (t, J = 7.5 Hz, 3H). 13C NMR (125 MHz,
C6D6) δ: 196.2 (C), 150.5 (C), 141.2 (C), 137.8 (CH), 132.1
(CH), 127.9 (CH), 123.2 (CH2), 119.9 (CH2), 116.9 (CH2),
115.2 (C), 80.2 (CH), 75.1 (CH), 30.5 (CH2), 29.7 (CH2),
24.9 (CH2), 13.0 (CH3), 9.3 (CH3), 8.9 (CH3). EI-MS m/z
(%): 290 ([M]+, 0.1), 261 (6), 153 (28), 83 (100), 55 (73), 29
(62). HR-MS calcd. for C16H21O3 ([M+ – C2H5]): 261.1491;
found: 261.1508.
(3aS,3bR,7aR,8aS)-2,2-Diethyl-7a-methyl-4-vinyl-7,7a-
dihydro-3aH-indeno[2,1-d][1,3]dioxol-8(3bH,6H,8aH)-one
(25)
1
Tr = 72.0 min. H NMR (500 MHz, C6D6) δ: 6.37 (ddd,
J = 17.8, 11.5, 0.8 Hz, 1H), 5.93 (d, J = 17.8 Hz, 1H), 5.39
(d, J = 3.3 Hz, 1H), 5.06 (d, J = 11.5 Hz, 1H), 4.59 (ddd, J =
7.6, 7.6, 1.4 Hz, 1H), 4.49 (d, J = 7.6 Hz, 1H), 2.67 (br dd,
J = 8.3, 2.9 Hz, 1H), 1.84–1.74 (m, 2H), 1.63 (qd, J = 7.6,
2.9 Hz, 2H), 1.60–1.57 (m, 1H), 1.51–1.45 (m, 1H), 1.46 (q,
J = 7.6 Hz, 2H), 0.92 (t, J = 7.6 Hz, 3H), 0.81 (t, J = 7.6 Hz,
3H), 0.58 (s, 3H). 13C NMR (125 MHz, C6D6) δ: 208.6 (C),
137.5 (CH), 135.4 (C), 127.9 (CH), 118.5 (C), 113.6 (CH2),
79.9 (CH), 77.4 (CH), 51.8 (C), 50.8 (CH), 29.5 (CH2), 29.3
(CH2), 28.3 (CH2), 23.4 (CH2), 15.3 (CH3), 8.7 (CH3), 7.8
(CH3). EI-MS m/z (%): 276 ([M]+, 0.1), 247 (38), 181 (30),
145 (16), 119 (46), 99 (14), 43 (100). HR-MS calcd. for
C15H19O3 ([M+ – C2H5]): 247.1334; found: 247.1342.
Diels–Alder adducts 23, 24, 25
Diels–Alder precursor 12a (0.051 g, 0.185 mmol) was
dissolved in toluene (20 mL), and the solution was purged
with argon for 20 min. The solution was then placed in the
microwave oven (program: 300 W power, ramp temperature
to 150 °C in 15 min, then hold at 150 °C for 3 h). After al-
lowing the reaction to cool to rt, the toluene was removed
under reduced pressure. Column chromatography (20:1, pe-
troleum ether – ethyl acetate) afforded a 41:20:39 mixture of
Diels–Alder adducts (0.039 g, 78%) (Table 1, entry 12). A
pure sample of each isomer was obtained by exhaustive pre-
parative HPLC (2% hexanes in ethyl acetate; Waters Nova-
Pak 19 × 300 mm silica (6 µm) column).
Conclusions
Twenty years ago we commented that the then-current lit-
erature indicated that, for the synthesis of bicycle[4.3.0]nonenones
from intramolecular Diels–Alder reactions, the following ap-
plied: “The ring fusion is determined by a complex interplay
of conformational, steric, and electronic effects that may
vary independently” (12). Clearly this still applies! We have
discovered that, contrary to our expectations based on studies
in the decalin series, these relatively flat tetraenes follow a
novel pathway in which the trans-adduct bicycle[4.3.0]nonenone
was not generated but instead the bicyclo[3.3.1]nonenone
skeleton was formed preferentially at 150 °C in toluene in a
microwave oven. At lower temperatures, the fused-ring
hydrindenone products dominated, but the bridged-ring
product was still produced. Indeed, once we understand
more about nuances that control these competing reactions it
may be possible to direct the product to either the fused- or
bridged-ring pathway from related tetraenes. A contributing
factor to begin to achieve this control is to examine the size
and nature of the R group. Unfortunately, the adduct mixture
from the synthesis made further study impractical, despite
the attraction of generating the steroid framework from a
second cycloaddition. Possibly future investigations, with
simpler and (or) sterically more demanding dienophile–
diene combinations, may permit the stereoselective synthesis
of functionalized hydrindanes in this manner.
(1R,2S,6S,8R)-4,4-Diethyl-8-methyl-11-vinyl-3,5-dioxa-
tricyclo[6.3.1.02,6]dodec-10-en-7-one (23)
Tr = 64.5 min. 1H NMR (500 MHz, C6D6) δ: 6.08 (dd, J =
17.7, 10.9 Hz, 1H), 5.26 (dd, J = 4.3, 4.3 Hz, 1H), 5.16 (d,
J = 17.7 Hz, 1H), 4.87 (d, J = 10.9 Hz, 1H), 4.49 (dt, J =
6.3, 2.2 Hz, 1H), 3.95 (d, J = 6.3 Hz, 1H), 3.00 (d, J =
2.2 Hz, 1H), 2.37 (dt, J = 13.2, 1.9 Hz, 1H), 1.96 (dd, J =
19.6, 4.6 Hz, 1H), 1.71 (q, J = 7.6 Hz, 2H), 1.65 (dd, J =
19.6, 1.1 Hz, 1H), 1.50 (qd, J = 7.6, 2.4 Hz, 1H), 1.22 (ddd,
J = 13.2, 3.6, 1.8 Hz, 1H), 1.10 (s, 3H), 0.98 (t, J = 7.6 Hz,
3H), 0.81 (t, J = 7.6 Hz, 1H). 13C NMR (125 MHz, C6D6) δ:
210.8 (C), 137.6 (CH), 136.9 (C), 130.2 (CH), 114.5 (C),
111.3 (CH2), 78.8 (CH), 77.4 (CH), 43.1 (C), 38.2 (CH2),
32.6 (CH), 30.7 (CH2), 29.8 (CH2), 28.5 (CH2), 25.0 (CH3),
8.7 (CH3), 8.6 (CH3). EI-MS m/z (%): 276 ([M]+, 1), 247
(77), 191 (19), 145 (38), 119 (54), 105 (27), 91 (15), 57
(100). HR-MS calcd. for C15H19O3 ([M+
247.1334; found: 247.1314.
– C2H5]):
(3aS,3bR,7aS,8aS)-2,2-Diethyl-7a-methyl-4-vinyl-7,7a-
dihydro-3aH-indeno[2,1-d][1,3]dioxol-8(3bH,6H,8aH)-one
(24)
Tr = 70.0 min. 1H NMR (500 MHz, C6D6) δ: 6.26 (dd, J =
17.7, 11.1 Hz, 1H), 5.66 (d, J = 17.7 Hz, 1H), 5.49 (t, J =
4.3 Hz, 1H), 5.06 (d, J = 11.1 Hz, 1H), 4.29 (dd, J = 6.9,
2.8 Hz, 1H), 4.23 (dd, J = 6.9, 0.5 Hz, 1H), 2.80 (br s, 1H),
1.65 (q, J = 7.5 Hz, 2H), 1.70–1.53 (m, 2H), 1.47 (q, J =
7.5 Hz, 2H), 1.22 (ddd, J = 13.0, 8.4, 5.5 Hz, 1H), 1.10 (s,
3H), 0.97–0.93 (m, 1H), 0.93 (t, J = 7.5 Hz, 3H), 0.81 (t, J =
7.5 Hz, 3H). 13C NMR (125 MHz, C6D6) δ: 215.6 (C), 138.7
(CH), 136.0 (C), 130.0 (CH), 116.8 (C), 112.8 (CH2), 80.7
Acknowledgements
We are grateful to the Natural Sciences and Engineering
Research Council of Canada (NSERC) for financial support
of this research. M. Clay thanks NSERC, the Ontario Gradu-
ate Scholarship Program, and the University of Ottawa for
Graduate Scholarships. D. Riber thanks the University of
Aarhus for a visiting fellowship. Preliminary experiments by
A. Melekhov and S. Graham are also acknowledged. We
© 2005 NRC Canada

568
Can. J. Chem. Vol. 83, 2005
4. S. Woo, N. Squires, and A.G. Fallis. Org. Lett. 1, 573 (1999).
5. D. Kim and Y.K. Lee. Tetrahedron Lett. 32, 6885 (1991).
6. B.B. Snider and T.C. Kirk. J. Am. Chem. Soc. 105, 2364
(1983).
thank K. Fagnou for use of his chiral columns and the refer-
ees for their suggestions.
7. (a) S.A. Bal and P. Helquist. Tetrahedron Lett. 22, 3933
(1981); (b) M.E. Jung and K.M. Halweg. Tetrahedron Lett. 22,
3929 (1981); (c) J. Shiina and S. Nishiyama. Tetrahedron, 59,
6039 (2003).
8. J. Endrikat, M. Cronin, C. Gerlinger, A. Ruebig, W. Schmidt,
and B. Dusterberg. Contraception, 64, 201 (2001).
9. A.B. Northrup and D.W.C. MacMillan. J. Am. Chem. Soc.
124, 2458 (2002).
References
1. (a) E.J. Corey and A.X. Huang. J. Am. Chem. Soc. 121, 710
(1999); (b) Q.-Y. Hu, P.D. Rege, and E.J. Corey. J. Am. Chem.
Soc. 126, 5984 (2004).
2. (a) C. Martin, N. Macintosh, N. Lamb, and A.G. Fallis. Org.
Lett. 3, 1021 (2001); (b) A. Melekhov, P. Forgione, S.
Legoupy, and A.G. Fallis. Org. Lett. 2, 2793 (2000); T. Wong,
P.D. Wilson, S. Woo, and A.G. Fallis. Tetrahedron Lett. 38,
7045 (1997).
10. N. Cohen, B.L. Banner, A.J. Laurenzano, and L. Carozza. Org.
Synth. 63, 127 (1984).
3. S. Woo, S. Legoupy, S. Parra, and A.G. Fallis. Org. Lett. 1,
1013 (1999).
11. S.D. Meyer and S.L. Schreiber. J. Org. Chem. 59, 7549 (1994).
12. A.G. Fallis. Can. J. Chem. 62, 183 (1984).
© 2005 NRC Canada